Rilem Pro

PROCEEDINGS
Draft
III
03-05 August 2015 So Paulo, Brazil
Organization
Institucional Support
Sponsors
III Progress of Recycling in the Built Environment

03-05 August 2015 So Paulo - Brazil
Editors
Isabel Martins
Sergio C. Angulo
Carina Ulsen
Vanderley M. John
Organizing Committee
Anette Mueller
Bruno Damineli
Carina Ulsen
Engracia Maria Bartuciotti
Enric Vzquez Ramonich
Isabel Martins
Juliana Garcia
Marco Quattrone
Mirko Landmann
Srgio Cirelli Angulo
Valdecir Quarcioni
Vanderley Moacyr John
Germany
Brazil
Brazil
Brazil
Spain
Portugal
Brazil
Brazil
Germany
Brazil
Brazil
Brazil
Scientific Committee
Anette Mueller
Bogdan Cazacliu
Bruno Damineli
Carina Ulsen
Carlos H. Sampaio
Chi Sun Poon
Enric Vzquez Ramonich
Francisco Agrela
Giacomo Moriconi
Henrique Kahn
Isabel Martins
Jianzhuang Xiao
Jorge de Brito
Marco Quattrone
Mirko Landmann
Monica Leite
Paola Villoria
Pilar Alaejos Gutirrez
Srgio Cirelli Angulo
Takafumi Noguchi
Vanderley Moacyr John
Valeria Corinaldesi
Valdecir Quarcioni
Yury Villagrn Zaccardi
Germany
France
Brazil
Brazil
Brazil
Hong Kong
Spain
Spain
Italy
Brazil
Portugal
China
Portugal
Brazil
Germany
Brazil
Spain
Spain
Brazil
Japan
Brazil
Italy
Brazil
Argentina

Contents
MECHANICAL PROPERTIES OF CONCRETE WITH RECYCLED AGGREGATES
AND WASTE GLASS
Viviana Letelier, Rodrigo Osses, Ester Tarela, Giacomo Moriconi
INFLUENCE OF THE USE OF RECYCLED PRECAST DEBRIS IN THE

MECHANICAL PROPERTIES OF CONCRETE
Viviana Letelier, Ester Tarela, Rodrigo Osses, Juan Pablo Crdenas, Giacomo Moriconi
GRAVITY CONCENTRATION OF CONSTRUCTION AND DEMOLITION WASTE

MATERIALS ON AIR JIGS
Carlos Hoffmann Sampaio, Bogdan Grigore Cazacliu, Grson Luis Miltzarek, Florian Huchet,
Laurdan le Guen, Carlos Otvio Petter, Rgis Paranhos, Weslei Monteiro Ambrs, Marcos Leandro
Silva Oliveira
GRAVITY CONCENTRATION OF CONSTRUCTION AND DEMOLITION WASTE

MATERIALS ON WATER JIGS
Carlos Hoffmann Sampaio, Bogdan Grigore Cazacliu, Alejandra Liendo, Carlos Otvio Petter, Rgis
Paranhos, Grson Luis Miltzarek, Weslei Monteiro Ambrs, Marcos Leandro Silva Oliveira
SELECTIVE COMMINUTION OF CONCRETE WASTE AND RECOVERY OF THE

CONCRETE CONSTITUENTS IN A GRINDING AND DRYING PROCESS
Andr Btz, Holger Wulfert, Paul Erwerth, Andreas Schiffers, Andreas Jungmann, Dirk van Mechelen
RESEARCH AND MODIFICATION ON RECYCLED AGGREGATES

Jianzhuang Xiao, Guanzhen Zhang
OUTLINE OF AIJ RECOMMENDATION FOR MIX DESIGN, PRODUCTION AND

CONSTRUCTION PRACTICE OF CONCRETE WITH RECYCLED CONCRETE
AGGREGATE
Takafumi Noguchi
CHANGES OF THE MATERIAL PROPERTIES BY SELECTIVE CARBONATION

OF RECYCLED AGGREGATES FROM CONCRETE
Marko Seidemann, Anette Mller, Horst-Michael Ludwig
THE ECONOMIC RISK ANALYSIS TECHNIQUE APPLIED TO DIFFERENT

SCENARIOS OF CONSTRUCTION AND DEMOLITION WASTE RECYCLING
PLANT.
Renato Aurlio Petter,Rgis Sebben Paranhos, Raul Oliveira Neto, Carlos Otvio Petter, Carlos
Hoffmann Sampaio
RATIONAL MIX DESIGN APPROACH FOR RECYCLED AGGREGATE

CONCRETE
Marco Pepe, Romildo Dias Toledo Filho, Eduardus A.B. Koenders, Enzo Martinelli

PUTTING RECYCLED CONCRETE INTO PRACTICE: LESSONS LEARNED

FROM BELGIAN PILOT & DEMONSTRATION PROJECTS
Jeroen Vrijders, Bram Dooms, Petra Van Itterbeeck
CHEMICAL-MINERALOGICAL SORTING OF MINERAL CONSTRUCTION AND

DEMOLITION WASTES FOR A BETTER RECOVERY
Mirko Landmann, Anette Mller
USE OF FINE RECYCLED AGGREGATE AS SUPPLEMENTARY CEMENTITIOUS

MATERIAL
Maria Eva Sosa, Claudio J. Zega, Gabriela.B Coelho Dos Santos, Angel A. Di Maio
PROPERTIES OF LIGHTWEIGHT CONCRETE MANUFACTURED WITH
RECYCLED MIXED AGGREGATES
Francisco Agrela, Manuel G. Beltrn, Julia Rosales, Manuel Cabrera
IMPROVING THE QUALITY OF RECYCLED AGGREGATES BY SORTING IN

AIR JIG
Carlos Hoffmann Sampaio, Weslei Monteiro Ambrs, Leonardo Miranda, Gerson Luis Miltzarek,
Mrcio Andr Kronbauer
THE INFLUENCE OF USING A SUPER PLASTICIZER IN RECYCLED GYPSUM

PLASTER
Sayonara Maria de Moraes Pinheiro, Gladis Camarini
TREATMENT OF RESIDUAL DIATOMITE EARTH FOR USE AS POZZOLAN IN

CEMENTITIOUS MIXTURES
Julia Graciele Vieira, Leila Aparecida De Castro Motta, Welles Oliveira, Csar Afonso Ponciano
RELEASE OF DANGEROUS SUBSTANCES FROM CONSTRUCTION AND

DEMOLITION RECYCLED MATERIALS USED IN ROAD PAVEMENTS LABORATORY AND FIELD LEACHING TESTS
Isabel Milagre Martins, Antnio Jos Roque, Ana Cristina Freire, Jos Manuel Neves, Maria de Lurdes
Antunes
RECYCLED FINE AGGREGATE IN NEW CONCRETE

Amnon Katz, Daniele Kulisch
STUDY ON THE RECYCLING TECHNIQUE OF PVC BUILDING WASTES

Kohei Yamaguchi, Akio Koyama
A STUDY ON THE ACTUAL STATE FROM GENERATION OF CONSTRUCTION

WASTES TO TREATMENT AND RECYCLING
Kenichi Imagawa, Akio Koyama
RECYCLED CONCRETE AGGREGATES FROM DIFFERENT SOURCES:

CHARACTERIZATION AND USE IN STRUCTURAL CONCRETE
Guilherme Chagas Cordeiro, Gabriela Valvassori Maioli, Amanda Pereira Vieira, Mnica Batista
Leite

EVALUATION OF MORTAR BASED ON ELECTRIC MEASUREMENTS: EFFECT

OF PRE-WETTING CDW RECYCLED AGGREGATE
Janaide Cavalcante Rocha, Tain Stefani, Malik Cheriaf
RECOVERY OF PACKAGING GLASS REFUSED BY MECHANICAL

BIOLOGICAL TREATMENT PLANTS- CASE STUDY
Nilmara Russo Braz dos Santos Dias, Maria Teresa da Cruz Carvalho
ALTERNATIVES FOR THE PRODUCTION OF STRUCTURAL CONCRETE WITH

FULL RECYCLED AGGREGATES
Fernando Martirena, Taimi Castano, Daylin Puentes, Sandra Lichtblau, Stefan Linsel
TECHNICAL AND ENVIRONMENTAL STUDY OF CEMENT TREATED

RECYCLED MATERIALS MADE WITH RECYCLED AGGREGATES AND
BIOMASS BOTTOM ASH
Manuel Cabrera, Francisco Agrela, Julia Rosales, Cristina Dorado, Francisco Gonzlez Gallardo,
Adela Prez Galvn
SECONDARY BUILDING MATERIALS FROM CONSTRUCTION AND

DEMOLITION WASTE : ENVIRONMENTAL BENEFITS AND COSTS
Karin Weimann, Christian Adam
FATIGUE FAILURE OF RECYCLED AGGREGATE CONCRETE UNDER

COMPRESSIVE INCREASING AMPLITUDE TEST
Carlos Thomas, JA Polanco, J Setien, Csar Medina
EFFECT OF SUPERPLASTICIZER ON MORTARS WITH RECYCLED FINE

AGGREGATES
Gabriela A. Valencia, Winnie F. Santos, Srgio C. Angulo, Rafael G. Pileggi
REHYDRATION OF CEMENT FINES: A TG /CALORIMETRY STUDY
Srgio C. Angulo, Mrio S. Guilge, Valdecir A. Quarcioni, Raphael
Baldusco, Maria A. Cincotto
PRELIMINARY CHARACTERIZATION OF EXCAVATED SOILS

FROM A MINERAL CDW SO PAULO CITY LANDFILL
Karen Kataguiri, Sergio Cirelli Angulo, Maria Eugenia Boscov
RECYCLED AGGREGATE CHARACTERIZATION METHODS
Anette Mueller
DESIGN OF A SEALING BIOMASONRY SINCE CRUDE LAND: THE ADOBE
BRICK CASE PROJECT
Daniel Augusto de Moura Pereira, Alusio Braz Melo

MECHANICAL PROPERTIES OF CONCRETE WITH RECYCLED

AGGREGATES AND WASTE GLASS
Viviana Letelier (1), Ester Tarela (1), Rodrigo Osses (1), Juan Pablo Cardenas (1), and
Giacomo Moriconi (2)
(1) Departamento de ObrasCiviles, Universidad de la Frontera, Temuco, Chile
(2) Department of Scienze e Ingegneria della Materia, dellAmbiente ed Urbanistica
(SIAMU) Universit Politecnica delle Marche, Italy
Abstract
The present investigation analyzes the compressive and flexural strength of concrete with
recycled materials. The combined influence of the use of coarse aggregates from precast
debris and waste glass as a cement replacement is studied. The combination of both recycled
materials allows to increase the amount of recycled coarse aggregates used reducing the loss
in the mechanical performance of the concrete, and enhances the environmental value of the
final material. The recycled aggregates are obtained from debris of prefabricated concrete
pipes with a compressive strength of 20 MPa. The waste glass has been obtained from glass
disposal facilities. Four variables are considered in this analysis: the percentage of recycled
course aggregates, the amount of mortar adhered to their surface, the amount of cement
replaced by recycled glass powder and the maximum size o of its particles. The final goal is to
determine which of the variables has a more significant effect in the final product and which
combination is the most efficient to optimize the performance of the concrete maximizing its
environmental value. Significant results have been obtained that show that the calibration of
the parameters may not be trivial, and the importance of combining different recycled
materials.
Keywords: Recycled aggregates; Recycled concrete; Concrete mechanical properties;
Compressive strength.
1.
INTRODUCTION
The effect of the use of recycled aggregates (RA) in concrete has been widely studied
([1][2][3][4]). The differences found in the mechanical properties due to the substitution of
natural aggregates for recycled ones have been attributed to the old mortar adhered to the
surfaces of the RA. Two interfaces have to be considered, the old interface, between the old
mortar and the RA, and the new one, between the RA and the new cement mixture ([1]). The
quality of these interfaces, given by the quantity and quality of the old adhered mortar, is one
of the main parameters influencing the mechanical behavior of recycled concrete ([5][6][7]).

A popular alternative to control the loss of strength due to RA is to limit the amount of
substituted aggregates. Various authors ([8][9][10][11][12]) have proven that the use of
relatively low percentages of RA doesnt modify significantly the properties of the material
and, nowadays, several European technic regulations allow the use of up to 30% of RA in
structural concrete.
Other way to minimize the effect of the use of RA is to reduce the amount of adhered
mortar to their surfaces. Eliminating the adhered mortar improves the quality of the contact
surface between the RA and the new cementious mixture, enhancing the strength of the
specimen([2][4][13]).
On the other hand, due to its high silica content, the use of glass powder (GP) as a
supplementary cementious material has been widely investigated. When only small particles
are considered (<300m) the GPpresents pozzolanic propertieswhich increase when the
particle size decreases ([14][15][16][17]).
As a measure to reduce the environmental impact generated by concrete manufacturing,
studies have been performed combining the use of RA and different industrial waste as
cement replacement. Kou et al. [18]studied the behavior of RAC adding silica fumes,
metakaolin fly ash and ground granulated blast slag.akir[19]studied the use of silica fume
and ground granulated clast slag. Kim et al. [20]analyze the use of RA and fly ash, concluding
that fly ash replacement. Beltran et al. [21]analyzed the mechanical properties of concretes
with RA and biomass bottom ash. Nassarandand Soroushian[22]studied concretes with 50%
of RA, using a 20% of GP as cement replacement, prove that the mechanical properties in the
long term are similar to those of traditional concrete.
The goal of this investigation is to analyze the mechanical properties of concretes
containing different percentages of GP (10-20-30%), modifying the particle maximum size
(75-150-425m) and using different amount of RA (20-30-40%) that have suffered different
levels of abrasion to reduce the old mortar adhered to its surface (100-200-300 rev). To
minimize the amount of test needed Taguchis statistical method is applied, combining the
variables reducing the experimental tests to 9 specimens plus a control concrete elaborated
with natural aggregates and no cement replacement. Compressive strengths and flexural
strengths are analyzed and compared, to find the optimum combination of parameters and
determine which one is the most significant.
2.
MATERIALS
2.1
Cement and glass powder
Pozzolanic cement, equivalent to ASTM type P cement is used. Different percentages of
this cement will be replaced byGP classified by the maximum size of its particles. The glass
powder is obtained from municipal recycling containers in the region.
2.2
Natural and recycled aggregates
Natural and recycled aggregates were used to prepare the concrete mixtures. The recycled
aggregates were obtained from precast concrete debris and their nominal sizes are 9.5 mm and
6.3 mm.
The Los Angeles abrasion machine is used to reduce the amount of adhered mortar to the
surface of the RA. Using different revolution settings, different abrasion levels are obtained
and different degrees of mortar reduction can be used. The effect of 100 rev, 200 rev and 300
rev is analyzed.

3.
EXPERIMENTAL TESTS
To reduce the number of tests needed, due to the great amount of variables and levels
considered, Taguchis method [23]is applied. Taguchis method is a combination of statistical
and engineering methods for variable optimization in any process reducing de amount of
experimental tests needed. Combining the three levels considered for each of the four
variables the minimum amount of tests needed to evaluate their influence in the results is 9
series.
The levels of the percentage of RA considered were 20%, 30% and 40%. These values
were chosen in base to several European technic regulations that nowadays allow the
use of RA.
The levels of the percentage of GP replacing cement considered were 10%, 20% and
30%.
The abrasion levels applied to reduce the adhered mortar to the surface of the RA were
100 rev, 200 rev and 300 rev, based on previous experimental tests.
The maximum sizes of the recycled GP particles considered were 75 m, 150 m and
425 m.
These levels were combined following Taguchis method and 9 series were stablished to
design the specimens needed (Table 2).
All the specimens were designed following FauryJoisel[24]method dosed for a
compressive strength of 30MPa after 28 day of curing with a water/cement rate of 0.42.
To calculate the final amount of water included in each sample the aggregates were presaturated up to 80%. This previous saturation has been proven to seal the pores, limiting water
interchanges and avoiding workability issues in fresh concrete and the water/cement rate
([25][26]).
The final values used in the dosing and the results obtained in Abrams cone and air
content are shown in table 4.
All the samples were elaborated under laboratory conditions and unmolded after 242
hours. Later, they were submerged in water with lime at 233C during the appropriate curing
time.
Table 1: Concrete sample dosing and tests
H1
H2
H3
H4
H5
H6
H7
H8
H9
HC
NA
2,36-9,5
(mm)
Kg
975,56
975,56
975,56
822,37
822,37
822,37
704,89
704,89
704,89
1174,81
RA
4,75-9,5 (mm)
Kg
199,25 (20%)
199,25 (20%)
199,25 (20%)
352,44 (30%)
352,44 (30%)
352,44 (30%)
469,92 (40%)
469,92 (40%)
469,92 (40%)
(0%)
rev
100
200
300
300
100
200
200
300
100
-
Cement
GP
Kg
343,80
305,60
267,40
343,80
305,60
267,40
343,80
305,60
267,40
382,00
Kg
38,20 (10%)
76,40 (20%)
114,60 (30%)
38,20 (10%)
76,40 (20%)
114,60 (30%)
38,20 (10%)
76,40 (20%)
114,60 (30%)
(0%)
m
75
150
425
150
425
75
425
75
150
-
Sand
Water
Kg
591,31
591,31
591,31
591,31
591,31
591,31
591,31
591,31
591,31
591,31
Lt
205,68
205,44
205,55
211,74
211,86
211,56
216,31
216,55
216,72
197,30

The mechanical properties of the samples were determined for each dosage by compressive
and flexural tests. The compressive tests were performed following the NCh 1037 and the
NCh 1017, using three cubic 15x15x15 cm test specimens after 28 and 90 days of curing. The
flexural tests were performed following the NCh 1038 using three prismatic 15x15x50 cm test
specimens after 28 days of curing. In each case the mean value of the three samples tested is
considered
4.
RESULTS AND DISCUSSION
4.1
Compressive strength
ANOVA is used to analyze the results of the Taguchi method. The influence of each
variable in the compressive strength is analyzed using Figure 1.
20
Recycled Aggregates (%)

30
75
40
Compressive Strength (MPa)
Glass Size (mm)

225
300
375
30
30
29
28
27
Abrasion 28 days
26
29
28
27
26
% Glass Powder 28 days
Glass Size 28 days
% Recycled Aggregates 28 days

25
25
100
A)
150
200
Abrasion (rev)
300
10
20
Glass Powder (%)
30
B)
Figure 1: Taguchi analysis for compressive strength
The effect of the percentage of RA and the abrasion reducing the amount of mortar is
shown in Figure 1 A. The maximum strength value is obtained for a 30% replacement of RA,
overpassing the value obtained for a 20% replacement in an 11%. This behavior is similar to
the one obtained by Soares et al. [27]who also obtained maximum strength values for a 30%
replacement. On the other hand, the mechanic abrasion shows no significant effect, lower than
1%, in the strength values.
Figure 1 B shows the effects of the amount of GP and the maximum size of its particles.
The increase in the amount of GP doesnt reduce the compressive strength significantly.
When percentage grows from 10% to 30% the strength reduction provoked is only of a 5.3%.
As expected, an increase in the particle size of the GP produces a significant decrease in the
strength values, up to a 12% when the size rises from 75 m to 425 m. Lee and Choi
[3]establish that GP increases the compressive strength when very small particles are used and
Federico &Chidiac[28]conclude that pozzolanic properties are only observed for GP particles
under 300 m.
Table 2presents the ANOVA results of Taguchi method. After 28 days of curing the more
significant parameters were the maximum size of the GP and the percentage of RA. The
amount of mortar, measured through the abrasion level, shows almost no significance in the
final results.

Table 2: ANOVA results for compressive strength.

Parameter df Sum of squares (S) Variance (V) Significance (%)
% RA
2
13,371
6,685
33,949
% GP
2
3,837
1,918
9,743
GPmaxsize 2
22,058
11,029
56,003
Abrasion
2
0,119
0,059
0,304
4.2
Flexural strength
Results given by Taguchi analysis for the flexural strength after 28 days are shown in
Figure 2.
20
Recycled Aggregates(%)
30
40
75
3,5
150
Glass Size (mm)

225
300
375
3,4
Flexural Strength (MPa)
% Glass Powder 28 days
3,3
3,2
3,1
3
2,9
2,8
Abrasion 28 days
2,7
% Recycled Aggregates 28 days
2,6
2,5
2,5
100
A)
Glass Size 28 days

3,5
200
Abrasion (rev)
300
10
20
Glass Powder (%)
30
B)
Figure 2: Taguhi analysis for flexural strength
In Figure 2 A the effect of the percentage of RA and the abrasion level in the flexural
strength is shown. The flexural strength is achieves its maximum value for a RA replacement
of 30% with a gaining rate of a 3% from the value obtained for a 20% replacement. This
behavior is similar to that of the compressive strength. Nevertheless, reducing the amount of
adhered mortar through mechanical abrasion increases the flexural strength. Samples over a
10% stronger are obtained when the abrasion changes from 100 rev to 300 rev. A similar
behavior is obtained by Pepe et al. (2014) who uses a similar methodology to reduce the
amount of adhered mortar.
In Figure 2 B the effect of the maximum particle size of the GP and the size of its
particlesis shown. Increasing the amount of GP from 10% to 30% only provokes a total
decrease in the flexural strength of a 7.8%, showing a low significance in the results. On the
other hand, as the size of the particles of the GP increases the flexural strength decreases as
expected, presenting a global loss of 23% when the size increases from 75 m to 425 m.
ANOVA results are shown in Table 3. After 28 curing days, the most significant variable is
the maximum size of the GP. On the other hand, the one that has the least influence in the
flexural strength is the percentage of RA.

Table 3: ANOVA results for flexural strength.

Parameter
% RA
% GP
GP maxsize
Abrasion
5.
df
2
2
2
2
Sum of squares (S)

0,267
0,534
4,316
0,536
Variance (V)
0,133
0,267
2,158
0,268
Significance (%)
3,634
8,024
70,292
8,061
CONCLUSIONS
The ANOVA results show that the GP maximum size is the most relevant parameter of the
four analysed here. Both, for the compressive and the flexural strength the best performance is
obtained when the smallest particles are used. Either way, the amount of GP replacement has
low significance in the final results.
Following these conclusions an important amount of GP can be reused in recycled
structural concrete as long as the particle size is controlled.
On the other hand, the percentage of RA used in the dosage is also significant when the
compressive strength is considered, finding an optimum performance when a 30% of the
natural aggregates are replaced. The relevance decreases significantly when the flexural
strength is considered. The abrasion level shows very low effects in the final compressive and
flexural strengths results.
Comparing the performances obtained, the optimum combination of the four parameters,
for the materials and the variables tested here is RA=30%; GP max size = 75mm; GP=10%
and Abrasion level=300rev. Though, the variability in the significance of the parameters may
allow considerable changes in the GP percentage and the abrasion level.
ACKNOWLEDGEMENTS
This analysis is part of the Project FONDEF CA13I10261 Desarrollo y diseo de un ecobloque de hormignen base a materialesreciclados, con menorimpactoambiental, menorcosto
y mayor eficienciatrmica" (Design and development of a concrete eco-block with recycled
materials with low environmental impact and high thermal efficiency), funded by National
Commission of Scientific-Technic Investigation - CONICYT de Chile.
REFERENCES
[1] Tam, V. W. Y., Gao, X. F., Tam, C. M., and Ng, K. M.,Physio-chemical reactions in recycled
aggregate concreteJournal of Hazardous Materials163(2008) 823-828.
[2] Ismail, S., andRamli, M.,Engineering properties of treated recycled concrete aggregate (RCA)
for structural applicationsConstruction and Building Materials44(2013) 464-476.
[3] Lee, G. C. and Choi, H. B.Study on interfacial transition zone properties of recycled aggregate
by micro-hardness testConstruction and Building Materials40 (2013) 455-460.
[4] Ismail, S. andRamli, M.,Mechanical strength and drying shrinkage properties of concrete
containing treated coarse recycled concrete aggregatesConstruction and Building
Materials68(2014) 726-739.
[5] Seo, D. S., and Choi, H. B.Effects of the old cement mortar attached to the recycled aggregate
surface on the bond characteristics between aggregate and cement mortarConstruction and
Building Materials59(2014) 72-77.

[6] Gonzalez, A., andEtxeberria, M. Experimental analysis of properties of high performance

recycled aggregate concreteConstruction and Building Materials52(2014) 227-235.
[7] Pedro, D., de Brito, J. and Evangelista, L.Influence of the use of recycled concrete aggregates
from different sources on structural concreteConstruction and Building Materials71(2014) 141151.
[8] Tam, V. W. Y., Wang, K. and Tam, C. M.Assessing relationships among properties of
demolished concrete, recycled aggregate and recycled aggregate concrete using regression
analysisJournal of Hazardous Materials152(2)(2008) 703-714.
[9] Padmini, A. K., Ramamurthy, K. and Mathews, M. S.Influence of parent concrete on the
properties of recycled aggregate concreteConstruction and Building Materials23(2)(2009) 829836.
[10] Li X.Recycling and reuse of waste concrete in China: Part II. Structural behaviour of recycled
aggregate concrete and engineering applications Resources, Conservation and
Recycling53(3)(2009) 107-112.
[11] Kwan, W. H., Ramli, M., Kam, K.J. and Sulieman, M. Z.Influence of the amount of recycled
coarse aggregate in concrete design and durability propertiesConstruction and Building
Materials26(1)(2012) 565573.
[12] Mas, B., Cladera, A., delOlmo, T. and Pitarch, F.Influence of the amount of mixed recycled
aggregates on the properties of concrete for non-structural useConstruction and Building
Materials 27(1)(2012) 612622.
[13] Pepe, M., Toledo Filho, R. D.,Koenders, E. A. B. and Martinelli, E.Alternative processing
procedures for recycled aggregates in structural concreteConstruction and Building
Materials69(2014) 124-132.
[14] Carsana, M., Frassoni, M. and Bertolini, L.Comparison of ground waste glass with other
supplementary cementitious materialsCement and Concrete Composites45(2014) 39-45.
[15] Khmiri, A.,Chaabouni, M. and Samet, B. Chemical behaviour of ground waste glass when used
as partial cement replacement in mortarsConstruction and Building Materials44(2013) 74-80.
[16] Kim, J., Yi, C. andZi, G. Waste glass sludge as a partial cement replacement in
mortarConstruction and Building Materials75(2015) 242-246.
[17] Mirzahosseini, M. and Riding, K. A.Influence of different particle sizes on reactivity of finely
ground glass as supplementary cementitious material (SCM)Cement and Concrete
Composites56(2015) 95-105.
[18] Kou, S., Poon, C. and Agrela, F. Comparisons of natural and recycled aggregate concretes
prepared with the addition of different mineral admixturesCement and Concrete
Composites33(8)(2011) 788-795.
[19] akr, . Experimental analysis of properties of recycled coarse aggregate (RCA) concrete with
mineral additivesConstruction and Building Materials68(2014) 17-25.
[20] Kim, K., Shin, M. and Cha, S.Combined effects of recycled aggregate and fly ash towards
concrete sustainabilityConstruction and Building Materials48(2013) 499-507.
[21] Beltrn, M. G., Agrela, F.,Barbudo, A.,Ayuso, J. and Ramrez, A.Mechanical and durability
properties of concretes manufactured with biomass bottom ash and recycled coarse
aggregatesConstruction and Building Materials72(2014) 231-238.
[22] Nassar R. and Soroushian P. Strength and durability of recycled aggregate concrete containing
milled glass as partial replacement for cementConstruction and Building Materials29(2012) 368377.
[23] Taguchi, G.The System of Experimental Design: Engineering Methods to Optimize Quality and
Minimize Costs1stEdn (UNIPUB/Kraus International Publications, White Plains, N.Y. (1987).
[24] Faury, J. (1958). Le Bton. Editions Dunod, Paris, Fr.
[25] Etxeberria, M., Mari, A., and Vsquez, E. Recycled aggregate concrete as structural
materialMaterials and Structures40 (2007) 529-541.

[26] Agrela, F., Snchez de Juan, M., Ayuso, J., Geraldes, V.L., and Jimnez J.R. Limiting properties
in the characterization of mixed recycled aggregates for use in the manufacture of
concreteConstruction and Building Materials, 25(10) (2011) 3950-3955.
[27] Soares, D.,de Brito,J., Ferreira, J. and Pacheco, J.Use of coarse recycled aggregates from precast
concrete rejects: Mechanical and durability performanceConstruction and Building
Materials71(2014) 263-272.
[28] Federico, L. M. andChidiac, S. E. Waste glass as a supplementary cementations material in
concrete- Critical review of treatment methodsCement & Concrete Composites31 (2009) 606610.
[29] Instruccin Espaola del Hormign Estructural EHE-08, 2008
[30] ACI 318-11, (2011). Building Code Requirements for Structural Concrete and Commentary, ACI
Committee 318, 503.

INFLUENCE OF THE USE OF RECYCLED PRECAST DEBRIS IN THE

MECHANICAL PROPERTIES OF CONCRETE
Viviana Letelier (1), Ester Tarela (1), Rodrigo Osses (1), Juan Pablo Crdenas (1) and
Giacomo Moriconi (2)
(1) Departamento de ObrasCiviles, Universidad de la Frontera, Temuco, Chile
(2) Department of Scienze e Ingegneria della Materia, dellAmbiente ed Urbanistica
(SIAMU) Universit Politecnica delle Marche, Italy
Abstract
In this investigation, concrete slabs from a building demolition are used, not only to
replace natural coarse aggregates but also as recycled cement substitution, in the aim of
reusing the most rejected material as possible optimizing its behavior. The effect of four
parameters related to both replacements is quantified and compared to find the most efficient
combination. The variables analyzed are: the percentage of recycled coarse aggregates
replacing the natural ones, the percentage of recycled cement replacing the natural one, the
maximum size of the recycled cement particles, and the temperature of rehydration of the
recycled cement. The goal is to establish the participation rate of these parameters in the
results, obtaining not only what combination returns the specimen with best mechanical
behavior, but also which of the variables is the one that plays the most important role in the
variation of the properties of the final material. The variable with the highest participation rate
is the amount of recycled coarse aggregates. As expected, the material suffers a significant
decrease in its compressive strength as the percentage of recycled aggregates increases. The
size of the replaced cement particles also plays an important role in this result.
Keywords: Recycled aggregates; Recycled concrete; Concrete mechanical properties;
Compressive strength.
1.
INTRODUCTION
The mechanical properties of concretes containing recycled aggregates (RA) have been
widely investigated ([1][2][3][4]). It has been proven ([5][6][7][8][9]) that the use of
relatively low percentages of RA doesnt modify significantly the properties of the material.
Nowadays, several European technic regulations allow the use of up to 30% of RA in
structural concrete.
Many authors claim that the differences found in the mechanical properties due to the
substitution of natural aggregates for recycled ones are due to the low quality old mortar
adhered to the surfaces of the RA. Because of this old mortar a second weak interface has to
be considered, the old interface, between the old mortar and the RA [1]. The mechanical

properties will depend on the quality of these interfaces, given by the quantity and quality of
the old adhered mortar ([10][11][12]). Based on these assumptions, the effects of the use of
RA can be reduced by eliminating or minimizing the adhered mortar, improving the quality of
the contact surfaces. This has been tested using either chemical or mechanical abrasion
procedures ([2][4][13]).
The aim to reduce the environmental cost of concrete manufacturing has led many authors
to combine the use of RA as coarse natural aggregates replacement with the use of different
industrial waste as cement replacements. Kou et al. [14] studied the behavior of RAC adding
silica fumes, metakaolin fly ash and ground granulated blast slag. akir [15] studied the use
of silica fume and ground granulated clast slag. Kim et al. [16] analyze the use of RA and fly
ash. Beltran et al. [17] analyzed the mechanical properties of concretes with RA and biomass
bottom ash. Nassarand and Soroushian [18] studied concretes with 50% of RA, using a 20%
of glass powder as cement replacement and prove that the mechanical properties in the long
term are similar to those of traditional concrete.
The reduction of the adhered mortar through mechanical abrasion produces fine particle
waste that can be used as recycled cement (RC). Contradictory results have been obtained in
previous studies on RC that may be attributed to the difficultness of recycling cementious
materials, and that their characteristics are complex and variable [19].
RC treated with high temperatures (400C-800C) can gain hydration capacity [20],
limiting the negative effects of the replace of cement.
This investigation analyzes the mechanical properties of concretes containing RA and RC.
The RA will be put to 300rev mechanical abrasion to reduce the adhered mortar from their
surfaces. Four parameters related to both recycled materials are studied to compare their
influence in the final product: the percentage of RA (20-30-40%) and of RC (5-10-15%), the
maximum size of the RC particles (75-150-300m) and the rehydration temperature of the RC
(400-650-900C). To reduce de amount of specimens needed to analyze all the possible
combinations Taguchis statistical method is applied, reducing the amount of tests needed to
9. Three specimens are performed for each of them. Compressive and flexural strengths are
studied and compared through ANOVA method to find the significance of each of the
variables and find the optimum combination.
2.
MATERIALS
2.1
Natural and recycled aggregates
Natural and recycled aggregates were used to prepare the concrete mixtures. The recycled
aggregates were obtained from concrete slabs from building demolition and their nominal
sizes are 9.5 mm and 6.3 mm.
The recycled coarse aggregates from the slabs are subjected to mechanical abrasion (300
rev.), using a Los Angeles abrasion machine. This process reduces the amount of adhered
mortar to the surface of the aggregates and has been proven to improve its mechanical
performance when replacing natural coarse aggregates.
2.2
Natural and recycled cement
Pozzolanic cement, equivalent to ASTM type P cement is used. Different percentages of

this cement will be replaced by RC.

The mechanical abrasion suffered by RA to reduce the amount of adhered mortar produces
fine material that is usually rejected. The recycled cement replacement is obtained from the
leftovers of this abrasion process, maximizing the reuse of the concrete debris. Then the
particles are sieved using different sized grids and thermally treated with a controlled heating
selected temperature.
3.
EXPERIMENTAL TESTS
Three levels for each of the four parameters are considered.

Percentage of RA: 20%, 30% and 40%. These values were chosen in base to several
European technical regulations that nowadays allow the use of RA.
Percentage of RC: 5%, 10% and 15%.
Maximum size of the RC particles: 75 m, 150 m and 300m.
Rehydration RC temperature: 400C, 650C and 900C
To analyze all the possible combinations 34 different dosages would be needed. Applying
Taguchis method [21], a statistical method for variable optimization in any process, reduces
the amount of tests needed.
Using a set of so called orthogonal arrays the number of different dosages required can be
minimized while giving full information. In this case we consider 4 different independent
variables with 3 level values each, so L9 orthogonal array is used, limiting the number of
needed experimental tests to 9. The method assumes that the variables do not interact,
meaning that the effect of the variation in one variable does not depend on the level settings of
the other independent variables. The orthogonal arrays considered here are displayed in Table
3. The results will be analyzed applying ANOVA.
All the specimens were designed following Faury Joisel [22] method dosed for a
compressive strength of 30MPa after 28 days of curing with a water/cement rate of 0.42, and
using natural sand.
To calculate the final amount of water included in each sample the aggregates were presaturated up to 80%. This previous saturation has been proven to seal the pores, limiting water
interchanges and avoiding workability issues in fresh concrete and the water/cement rate
[23][24].
The final values used in the dosing and the results obtained in Abrams cone and air
content are shown in table 4.
All the samples were elaborated under laboratory conditions and unmolded after 242
hours. Later, they were submerged in water with lime at 233C during the appropriate curing
time.
Table 1: Concrete sample dosing and tests
H1
H2
H3
H4
H5
H6
NA
2,3619(mm)
Kg
818,9
818,9
818,9
716,5
716,5
716,5
RA
Cement
6,3-19(mm)
Kg
204,7 (20%)
204,7 (20%)
204,7 (20%)
307,1 (30%)
307,1 (30%)
307,1 (30%)
Kg
363,2
344,1
325,0
363,2
344,1
325,0
RC
Kg
19,1 (5%)
32,2 (10%)
57,3 (15%)
19,1 (5%)
32,2 (10%)
57,3 (15%)
m
300
150
75
150
75
300
C
400
650
900
900
400
650
Sand
Water
Kg
733,8
733,8
733,8
733,8
733,8
733,8
Lt
162,6
162,6
162,6
162,6
162,6
162,6

H7 614,2
H8 614,2
H9 614,2
HC 1023,3
409,4 (40%) 363,2

409,4 (40%) 344,1
409,4 (40%) 325,0
(0%)
382,00
19,1 (5%)
32,2 (10%)
57,3 (15%)
(0%)
75
300
150
-
650
900
400
-
733,8
733,8
733,8
733,8
162,6
162,6
162,6
162,6
The mechanical properties of the samples were determined for each dosage by compressive
and flexural tests. The compressive tests were performed following the NCh 1037 and the
NCh 1017, using three cubic 15x15x15 cm test specimens after 28 and 90 days of curing. The
flexural tests were performed following the NCh 1038 using three prismatic 15x15x50 cm test
specimens after 28 days of curing. In each case the mean value of the three samples tested is
considered.
4.
4.1
To analyze the results of the Taguchi method ANOVA method is applied. The influence of
each variable in the compressive strength is analyzed using Figure 1.
Temperature(C)
Recycled Aggregates(%)
20
30
400
30,0
800
29,5
29,5
29,0
% Recycled Cement
28,5
% Recycled Aggregate
28,0
27,5
27,0
26,5
26,0
29,0
28,5
Maximum size recycled cement
28,0
Recycled cement temperature treatment
27,5
27,0
26,5
26,0
25,5
25,5
25,0
25,0
5
10
Cement Recycled (%)
A)
600
30,0
40
15
75
150
225
Maximum size(m)
300
B)
Figure 1: Taguchi analysis for compressive strength
The effect of the percentage of RA and percentage of RC is shown in Figure 1 A. The

overall trend of the results is a decrease in the compressive strength values when the amount
of RA increases. When a 40% of natural aggregates are replaced the compressive strength of
the new material presents a decrease of the 7% from the values obtained when only a 20% had
been replaced. Kwan et al [8] and Gonzalez and Etxeberria [11] attribute this trend to the low
quality of the adhered mortar to the RA when a high percentage has been substituted. The
more amount of RA is used the chance of weak union areas is bigger.
On the other hand, no damaging effects in the final results are observed from changing the
amount of RC. The maximum variation in the compressive strength is only of a 2%. These
results are similar to those obtained by Yu and Shui [19]. They use similar percentages
(between 5% and 15%) concluding that no significant changes appear in the final compressive
strength.
Figure 1 B shows the effects of the amount of the maximum size of the RC particles and
the rehydration temperature on the compressive strength. A significant increase in the quality
of the final concrete is observed when smaller particles of RC are used, showing a rise of the

7% in the compressive strength when the minimum size particles are used (75m) when
compared to the one obtained when the 300 m RC particles are replaced. Lee et al. [25] use
glass powder as cement replacement, concluding that adding very small particles improves the
compressive strength of the specimen.
The effect is not as important when the rehydration temperature is analyzed. A difference
of only a 1% is observed between the three different levels of the parameter. Even though
Shui et al [20] obtain significant variations in the compressive strength when the rehydration
temperature increases, it must be taken into account that they replace 100% of the
cementitious material, applying the increase of temperature to the whole replacement. In this
study only the 5-15% replaced percentage suffers the rehydration process, minimizing its
effect.
Table 2 presents the ANOVA results of Taguchi method. After 28 days of curing the more
significant parameters in the final compressive strength value were the maximum size of the
RC and the percentage of RA. The rehydration temperature shows almost no significance in
the final results.
Table 2: ANOVA results for compressive strength.
Parameter Sum of squares (S) Significance (%)
% RA
13,371
33,949
% RC
3,837
9,743
RC max size
22,058
56,003
Temperature
0,119
0,304
4.2
Flexural strength
To analyze the results of the Taguchi method ANOVA method is applied. The influence of
each variable in the flexural strength is analyzed using Figure 2.
Temperature(C)
Recycled Aggregate (%)

30
40
400
5,4
5,4
5,2
5,2
20
4,8
Recycled Cement
Recycled Agreggate
4,6
4,4
800
4,8
4,6
Maximum size recycled cement
4,4
Recycled cement temperature treatment

4,2
4,2
4
5
10
15
Recycled cement(%)
A)
600
75
150
225
300
Maximum size(m)
B)
Figure 2: Taguhi analysis for flexural strength
In Figure 2 A the effect of the percentage of RA and RC in the flexural strength after 28
days of curing is shown. As it happened with the compressive strength, increasing the amount
of RA leads to a decrease in the specimens flexural strength. When the RA replacement
increases to a 40% the decrease in the flexural strength from the value obtained with a 20%

replacement is of 7%. Silva et al. [26] present information on studies about flexural behavior
of recycled concrete. They conclude that the decrease in the flexural strength of the sample
with the amount of RA will depend on the quality of the aggregates. In this analysis RA are
obtained from a very low quality concrete what induces the obtained loss in the flexural
strength.
On the other hand, increasing the amount of RC from a 5% to 15% increases the flexural
strength of the material up to a 25%. Few authors analyze deeply this property when RC is
used.
In Figure 2 B the effect of the maximum particle size of the RC and its rehydration
temperature is shown. The maximum flexural strength is achieved for a medium maximum
size of the RC particles. A total increase of the flexural strength of 20% is obtained for a
maximum size of the particles of 150 m. Meanwhile, the effect of the rehydration
temperature is insignificant with a total variation of a 1% in the final value.
ANOVA results are shown in Table 3. After 28 curing days, the most significant variable
is the maximum size of the RC. On the other hand, the one that has the least influence in the
flexural strength is the rehydration temperature.
Table 3: ANOVA results for flexural strength.
Parameter
% RA
% RC
RC max size
Temperature
5.
Sum of squares (S)

1,529
0,429
2,434
0,195
Significance (%)
26,48
7,441
42,148
3,389
CONCLUSIONS
The ANOVA results show that the RC maximum size is the most relevant parameter of the
four analysed here. Both for the compressive and the flexural strength a good performance is
obtained when the small (<150m) particles are used. Meanwhile, the rehydration
temperature of the RC replacement has low significance in the final results, leading to the
conclusion that there is no need to increase the rehydration temperature over 400C.
The amount of cement replaced by de reused powder shows a low significance, but still
perceptible. The best results are obtained for a 15% replacement in both analyses.
Following these conclusions an important amount of cement can be recycled as long as the
particle size is small enough with no need of increasing the rehydration temperature.
On the other hand, the effect of the RA in the final performance of the samples is medium.
A loss of the strengths, both compressive and flexural, takes place when the amount of the RA
increases, due to the low quality of the aggregates used.
Comparing the results obtained, the optimum combination of the four parameters, for the
materials and variables tested here is RA=20%; RC=15%; RC max size=150mm;
Temperature=400C.
ACKNOWLEDGEMENTS
This analysis is part of the Project DIUFRO DI14-0097 Evaluacin de metodologa para
la eliminacin de mortero por abrasin en ridos reciclados (Evaluation of the methodology
to reduce adhered mortar using abrasion in recycled aggregates), funded by Universidad de la
Frontera, Temuco, Chile.

REFERENCES
[1] Tam, V. W. Y., Gao, X. F., Tam, C. M., and Ng, K. M.,Physio-chemical reactions in recycled
aggregate concrete, Journal of Hazardous Materials 163 (2008) 823-828.
[2] Ismail, S., and Ramli, M., Engineering properties of treated recycled concrete aggregate (RCA)
for structural applications, Construction and Building Materials 44 (2013) 464-476.
[3] Lee, G. C. and Choi, H. B., Study on interfacial transition zone properties of recycled aggregate
by micro-hardness test Construction and Building Materials 40 (2013) 455-460.
[4] Ismail, S. and Ramli, M., Mechanical strength and drying shrinkage properties of concrete
containing treated coarse recycled concrete aggregates Construction and Building Materials 68
(2014) 726-739.
[5] Tam, V. W. Y., Wang, K. and Tam, C. M., Assessing relationships among properties of
demolished concrete, recycled aggregate and recycled aggregate concrete using regression
analysis Journal of Hazardous Materials152(2)(2008) 703-714.
[6] Padmini, A. K., Ramamurthy, K. and Mathews, M. S., Influence of parent concrete on the
properties of recycled aggregate concrete Construction and Building Materials 23 (2) (2009)
829-836.
[7] Li X., Recycling and reuse of waste concrete in China: Part II. Structural behaviour of recycled
aggregate concrete and engineering applications Resources, Conservation and Recycling 53 (3)
(2009) 107-112.
[8] Kwan, W. H., Ramli, M., Kam, K.J. and Sulieman, M. Z., Influence of the amount of recycled
coarse aggregate in concrete design and durability properties Construction and Building
Materials 26 (1) (2012) 565573.
[9] Mas, B., Cladera, A., del Olmo, T. and Pitarch, F., Influence of the amount of mixed recycled
aggregates on the properties of concrete for non-structural use Construction and Building
Materials 27 (1) (2012) 612622.
[10] Seo, D. S., and Choi, H. B., Effects of the old cement mortar attached to the recycled aggregate
surface on the bond characteristics between aggregate and cement mortar Construction and
Building Materials 59 (2014) 72-77.
[11] Gonzalez, A., and Etxeberria, M. Experimental analysis of properties of high performance
recycled aggregate concrete Construction and Building Materials 52 (2014) 227-235.
[12] Pedro, D., de Brito, J. and Evangelista, L., Influence of the use of recycled concrete aggregates
from different sources on structural concrete Construction and Building Materials 71 (2014)
141-151.
[13] Pepe, M., Toledo Filho, R. D., Koenders, E. A. B. and Martinelli, E., Alternative processing
procedures for recycled aggregates in structural concrete Construction and Building Materials 69
(2014) 124-132.
[14] Kou, S., Poon, C. and Agrela, F. Comparisons of natural and recycled aggregate concretes
prepared with the addition of different mineral admixtures Cement and Concrete Composites 33
(8) (2011) 788-795.
[15] akr, . Experimental analysis of properties of recycled coarse aggregate (RCA) concrete with
mineral additives Construction and Building Materials 68 (2014) 17-25.
[16] Kim, K., Shin, M. and Cha, S. Combined effects of recycled aggregate and fly ash towards
concrete sustainability Construction and Building Materials 48 (2013) 499-507.
[17] Beltrn, M. G., Agrela, F., Barbudo, A., Ayuso, J. and Ramrez, A. Mechanical and durability
properties of concretes manufactured with biomass bottom ash and recycled coarse aggregates
Construction and Building Materials 72 (2014) 231-238.
[18] Nassar R. and Soroushian P. Strength and durability of recycled aggregate concrete containing
milled glass as partial replacement for cement Construction and Building Materials 29 (2012)
368-377.

[19] Yu, R. and Shui, Z. Influence of agglomeration of a recycled cement additive on the hydration
and microstructure development of cement based materials Construction and Building Materials
49 (2013) 841-851.
[20] Shui, Z., Xuan, D., Wan, H. and Cao, B., Rehydration reactivity of recycled mortar from
concrete waste experienced to thermal treatment Construction and Building Materials 22 (2008)
1723-1729.
[21] Taguchi, G. The System of Experimental Design: Engineering Methods to Optimize Quality and
Minimize Costs1stEdn (UNIPUB/Kraus International Publications, White Plains, N.Y.,1987).
[22] J., Faury (1958). Le Bton. Editions Dunod, Paris, Fr.
[23] Etxeberria, M., Mari, A., and Vsquez, E., Recycled aggregate concrete as structural material
Materials and Structures 40 (2007) 529-541.
[24] Agrela, F., Snchez de Juan, M., Ayuso, J., Geraldes, V.L., and Jimnez J.R., Limiting properties
in the characterization of mixed recycled aggregates for use in the manufacture of concrete
Construction and Building Materials 25(10) (2011) 3950-3955.
[25] Lee G., Poon C. S., Wong Y. L., Ling, T. C., Effects of recycled fine glass aggregates on the
properties of drymixed concrete blocks Construction and Building Materials 38 (2013) 638643.
[26] Silva, R. V., de Brito, J. and Dhir, R. K., Tensile strength behaviour of the recycled aggregate
concrete Construction and Building Materials 83 (2015) 108-118.
[27] Instruccin Espaola del Hormign Estructural EHE-08, 2008
[28] ACI 318-11, (2011). Building Code Requirements for Structural Concrete and Commentary, ACI
Committee 318, 503.

GRAVITY CONCENTRATION OF CONSTRUCTION AND

DEMOLITION WASTE MATERIALS ON AIR JIGS
Carlos H. Sampaio (1), Bogdan Cazacliu (2), Gerson Miltzarek (1), Florian Huchet (2),
Laurdan le Guen (2), Carlos O. Petter (1), Rgis Paranhos (3), Weslei M. Ambrs (1),
Marcos S. Oliveira (4)
(1) UFRGS, Department of Metallurgy, Mineral Processing Laboratory, Brazil
(2) LUNAM, University of Nantes, IFSTTAR, Aggregates and Material Processing
Laboratory, France
(3) UNIPAMPA, Planning and Mineral Processing Laboratory, LATRAM, Brazil
(4) Odebrecht Latinvest, Peru
Abstract
This work deals with the separation of concrete, brick and gypsum particles from a CDW
recycled aggregate mixture. Construction and Demolition Waste materials can be handled as
an ore treatment problem. Efficient sorting processes of low quality CDW recycled aggregate
could allow the reuse of concrete, brick and gypsum particles. The processes also improve the
remaining mixed aggregates for recycling in unbound sub-base materials, by increasing their
self-cementing properties and by reducing the sulphate content through the removal of
gypsum. All tests were carried out in the size range between 4 and 20 mm using a laboratory
air jig. The aim of the work is to concentrate a high amount of gypsum (light material) and
concrete (heavy material) particles. Concentrates with concrete contents higher than 90% and
gypsum contents significantly lower than 1% were possible to be reached and can be an
alternative in aggregates for the concrete market. Indeed, gypsum reduction in concrete
concentrates was about 25 times. This level of reduction could be satisfactory in sorting real
demolition products. On the other hand, contents of over 70% of gypsum concentrates were
obtained, increasing the reuse and recycling abilities.
Keywords: jigging, gravity concentration, air jig, C&DW materials
1.
INTRODUCTION
The basic composition of Construction and Demolition Waste (C&DW) is strongly

dependent, among others, on geographical location, type of the construction commercial or
residential [1], or on construction method, conventional or precast [2]. The approximate mean
percentage of concrete and ceramics in demolition waste is estimated at 70% [3]. The use of
these inert residues is economically feasible [4] but unfortunately, when recycled, the
employment of inert waste materials as recycled aggregates is still limited to building

materials in public works (embankments, pavements, etc.), or as material for the recovery of
degraded areas in quarries. There are also examples of better uses for some of these residues
as they have a higher content of concrete and lower content of contaminants like gypsum.
They are recovered and used as Recycled Concrete Aggregates (RCA), mainly the coarse
RCA fraction (i.e. 4/20 mm), for new concrete mix-design. This can provide a net
environmental benefit for the concrete recycling method [5]. A more complete reuse is
generally hampered by the lack of suitable recycling plants [6].
The success of the off-site construction waste sorting program could be encouraged by
government incentive policies [7,8] or by improving the design decisions [9]. For instance the
current European and French goal to be achieved by 2020, requires the recovery of 70% by
weight of all C&DW from construction and public works [10]. The utilization of a waste
management policy on construction sites potentially enables a cost reduction of about 30-40%
[11,12]. However, only few building industry players implement the selective demolition and
disassembly [13], which is considered time and labor demanding [14]. In fact, the difficulties
in implementing in real practice an environmental management system in accordance with
ISO Standard 14001 are discussed by Rodriguez [15]. Consequently, improved mechanical
sorting systems should be considered to complete the efficiency of the production of high
quality recycled aggregates.
Some researchers have suggested the use of water jigs [16,17] or air jigs [11,12,18] for the
density separation of recycled aggregates. Other complementary methods are suggested, like
the spirals [11] or the sensor based sorting [18]. A list of available ideas to treat recycled
aggregates is also given by Schnellert and Mueller [19], who consider jigging and optical
sorting as having a good sorting performance. None of these devices was deeply investigated
in the literature for the use in RCA sorting. However, several authors are in agreement about
the jigging utilization as a promising method to treat C&DW, characterized by different
particles densities [20-21].
This paper evaluate the potential of a laboratory air jig equipment to separate concrete,
brick and gypsum particles aiming C&DW industrial applications.
2.
EXPERIMENTAL METHOD
The air jig tests were conducted with three types of materials - concrete (C25/30), solid
clay bricks and gypsum from solid gypsum blocks all particles having a size of 4/20 mm, a
range considered as typical for the coarse recycled aggregates size. This particle size is largely
used for recycling aggregates and it is a size range appropriate for jigging [22].
For each material the skeletal and envelope density [23] were measured. To obtain the
skeletal density a helium pycnometer (multipycnometer quantachrome) was used. The
envelope density was calculated by weighting and water volume displacement after surface
impermeabilization.
The bulk density of each individual material was determined by pouring the dry aggregate
in a box of known volume and the standard deviation was obtained after 10 repetitions. The
size distribution and density measures of the materials used in tests is given in Table 1.
An AllMineral Company air jig, modell AllAir S 500, was used to perform the tests
(Figure 2.a). The AllMineral jig is composed of two simultaneous inputs of air flow at the
bottom of the equipment. The first airflow is responsible for the expansion of the particles
bed, while the second flow vibrates the particle bed. With the entry of the two airflows, the
particle bed is expanded and vibrates. This movement enables the stratification of the particle
bed, which presents increasing density from the top to the bottom of the equipment.

Table 1: Particle size distribution, skeletal and envelope density of the materials.
Material
Fraction (mm)
15.9/19.1
9.5/15.9
6.35/9.5
4.75/6.35
Skeletal Density (g/cm3)
Envelope Density (g/cm3)
Bulk Density standard deviation (g/cm3)
(a)
Concrete
Mass (%)
3.9
38.6
28.8
28.7
2.67
2.39
1.67 0.037
Brick
Mass (%)
12.6
43.2
31.0
13.2
2.59
2.26
0.84 0.042
Gypsum
Mass (%)
3.6
46.4
38.1
11.9
2.30
1.86
0.61 0.046
(b)
Figure 2: AllMineral jig (a) modell AllAir S 500, used in the tests. (b) assembly of the
jig chambers.
All jigging tests were carried out in batch mode. The equipment simulates stratification
that happens inside a continuous jig. An industrial jig promotes along the equipment several
expansions and contraction of the particles bed, which can be simulated in the laboratory jig.
It was possible to set some jigging parameters during the tests, they are: the frequency of the
jigging pulsation, which is measured in pulsations per min.; the expansion of the particles,
which is measure by percentage of the fan capacity (motor generates 6,000 Pa) ; and the time
of each test, which may vary widely.
The tests were conducted with variation of operational parameters to avaliate the influence
of each one (frequency, expansion and time) on jigging performance. For that, samples of
about 39 kg of mixed particles were first obtained from 53% of concrete particles, 27% of
brick particles and 20% of gypsum particles, in mass. The mixed particles sample is poured in
the assembly of the boxes (Figure 3-a). The amount of each material was chosen to fill
completely one chamber after a hypothetical perfect separation of the three particles species
(Figure 3-a).
Once the particles stratified after jigging (Figure 3-c), the boxes were removed separately.
The particles inside each box were separated by hand and weighed to detemine the percentage
of each material in each box. The box close to the bottom of the air jig was named the Inferior
Layer; the box in the middle, Middle Layer; and the box on the top, Superior Layer.

(a)
(b)
(c)
Figure 3: Jigging bed. (a) Feed, before mixing; (b) Feed, after mixing, before sorting;
(c) After sorting, stratified particle bed.
3.
Table 2 presents the percentage in mass of each material type (concrete, brick and gypsum)
in the three layers as a function of frequency, jigging time and expansion ratio used in the
tests.
Table 2: Results of the jigging tests. f = frequency; Exp = expansion ratio; t = jigging time.
Inferior layer
Middle layer
Superior layer
Operational
Conditions Concrete Brick Gypsum Concrete Brick Gypsum Concrete Brick Gypsum
f = 160 rpm; Exp = 70%
t = 30 s
86.5% 10.8% 2.5%
40.1% 45.3% 14.5% 8.6% 30.1% 61.1%
t = 60 s
88.6% 9.1%
2.2%
32.5% 51.8% 15.6% 6.6% 33.9% 59.4%
t = 90 s
85.9% 11.6% 2.4%
30.7% 55.1% 14.1% 9.2% 26.6% 64.1%
t = 120 s
92.2% 7.0%
0.7%
21.6% 60.5% 17.7% 3.0% 20.3% 76.5%
t = 150 s
90.5% 8.4%
1.0%
30.4% 58.5% 11.0% 3.4% 18.1% 78.4%
t = 180 s
91.1% 7.6%
1.2%
26.7% 60.2% 13.0% 7.7% 29.3% 62.9%
Exp = 70%; t = 120 s
f = 150 rpm 90.8% 7.8%
1.4%
35.0% 52.5% 12.3% 6.6% 24.9% 68.3%
f = 170 rpm 89.4% 9.5%
1.0%
28.1% 56.0% 15.8% 2.8% 14.9% 82.2%
f = 160 rpm; t = 120 s
Exp = 60%
83.4% 12.9% 3.6%
46.4% 42.1% 11.3% 4.4% 36.9% 58.6%
Exp = 80%
89.1% 10.1% 0.7%
23.8% 59.8% 16.2% 4.7% 21.6% 73.6%
It can be observed that the mixture is stratified with the concrete mainly filling the Inferior
layer (about 80-90%), brick the Middle layer (40-60%) and gypsum the Superior layer (about
60-80%). All these concentrations are significantly higher than the mean composition in the
corresponding constituents.
The brick in the Middle layer is characterized by lower concentration. This is explained by
the presence of non-segregated brick particles in both Inferior and Superior layers. For other
hand, the low amount of gypsum in the Inferior layer as well as the high amount of gypsum in
the Superior layer gives good perspective for future industrial application.
All the elements of the stratification are mostly effective at the shortest tested jigging time,
i.e. 30 seconds. Indeed, stratification by gravity generally occurs quickly [24].
The stratification processes along the jigging time is better evidenced by the evolution of
the mass ratios in Inferior, Middle and Superior layers (Figure 4). On average, the mass ratios
diminish 25% each time the jigging time doubles, during stratification process progress and

follow power laws. A faster evolution is noticeable in each layer characterized by a larger
power coefficient. As for example, in the middle layer, the concrete/brick couple is faster than
for gypsum/brick. The evolution in this layer may suggests that the bulk densities has more
pronounced effect on stratification than the individual particle densities.
(a)
(b)
(c)
Figure 4: Evolution of the different mass ratios in the Inferior, Middle and Superior layers
along the jigging time.
The effect of the jigging frequency was analyzed by testing three frequencys 150 , 160
and 170 rpm - and can be well observed in the Superior layer (Table 2), where the proportion
of gypsum increases with high frequencies, while both concrete and brick decrease. Indeed,
up to a given level, it is expected that the increase in the frequency makes easier the
movement of particles up and downwards. However, the behavior in the Middle and Inferior
layers seems different from the Superior layer. One can observe, in these layers, that
stratification is better at the intermediate frequency value (160 rpm) than at the higher
frequency (170 rpm). This concerns all components of the mixture: concrete, brick and
gypsum. Even if it is difficult to conclude, given the inherent fluctuation of the concentrations
during the process, frequencies over 160 rpm seems to increase the re-mixture between the
concrete and brick particles. This stratification is driven by the difference in bulk densities.
The stratification between brick and gypsum, driven by the difference in particle densities, is
amplified at higher frequency than that between brick and concrete.
The influence of the jigging expansion ratio was analyzed by testing three ratios, 60%,
70% and 80%, keeping the other jigging parameters constants: same mixture composition as
in the previous tests, jigging time 120 s, frequency 160 rpm. It can be observed that an
increase in the expansion ratio from 60% to 70% produces significant improvement of the
stratification. The further increase of the expansion ratio to 80% does not produce benefits in
stratification. It seems rather that air pressures over 70% increases the re-mixture. This can be
observed on all the types of particles in all the layers. Indeed, expansion of the particle layer is
carried out through the pressure of inlet air. With high air pressures a very turbulent system is
reached, which potentially increase re-mixtures of particles.
4.
CONCLUSIONS
The main conclusions of the paper are presented below.

Demolition materials can be handled as a simple ore treatment problem, with reasonable
separation possibilities. This way, efficient sorting processes of low quality CDW recycled
aggregate could allow the reuse of concrete, brick and gypsum particles. The processes also

improve the remaining mixed aggregates for recycling in unbound sub-base materials, by
increasing their self-cementing properties and by reducing the sulphate content.
It is possible to separate gypsum from concrete and brick particles, with size range 4 to 20
mm, in air jigs, due to the difference of bulk densities and particle densities of the materials.
Concentrates (sink products inferior jig chamber) with concrete contents higher than 90%
and gypsum contents lower than 1% were possible to be reached.
Gypsum reduction in concrete concentrates was about 25 times. This level of reduction
could be satisfactory in sorting real construction and demolition waste aggregates.
Concentrates with lower densities presents over 70% of gypsum particles.
All materials (concrete, bricks and gypsum) used in this work were originated by
comminuting individual samples. No middlings (particles with different constituents) were
used in the tests. With real demolition materials there will be the presence of middlings,
which provides new difficulties in physical separation.
Encouraging results were reached in this work, despite problems described in the paper.
The results show that the accuracy of the concentration could be improved for lower gypsum
contents than tested in the present work, as expected in real demolition products.
Although the tests performed in this work have been made with an air jig, which presents a
Tromp imperfection higher than water jigs, it can be observed that the results of gypsum
concentration in superior layer (superior jig chamber), or concrete concentration in inferior
layer (inferior jig chamber), are quite similar to those presented in the literature.
ACKNOWLEDGEMENTS
The Capes-Cofecub Re-Maud project is acknowledged.
REFERENCES
[1] Camp, Dresser & McKee Inc., 1998. Quantity and Composition Study of Construction and
Demolition Debris in Wisconsin. Wisconsin Recycling Market Development Board.
[2] Lachimpadi, S.K., Pereira, J.J., Taha, M.R., Mohktar, M., 2012. Construction waste minimization
comparing conventional and precast construction (Mixed system and IBS) methods in high-rise
buildings: A Malaysia case study. Resource, Conservation and Recycling, 68, 96-103.
[3] Oikonomou, Nik.D., 2005. Recycled Concrete Aggregates. Cement and Concrete Composites 27,
no. 2, 31518.
[4] Begum, R.A., Siwar C., Pereira, J.J., Jaafar, A.H., 2006. A benefitcost analysis on the economic
feasibility of construction waste minimisation: The case of Malaysia. Resources, Conservation
and Recycling, 48 (1), 86-98.
[5] Tam,V.W.Y., 2008. Economic comparison of concrete recycling: A case study approach.
Resources, Conservation and Recycling, 52 (5), 821-828.
[6] Bianchini, G., Marrocchino, E., Tassinari, R., Vaccaro, C., 2005. Recycling of construction and
demolition waste materials: a chemicalmineralogical appraisal, Waste Management, 25 (2), 149159.
[7] Lu, W., Yuan, H., 2012. Off-site sorting of construction waste: What can we learn from Hong
Kong? Resources, Conservation and Recycling, 69, 100-108.
[8] Wang, J., Yuan, H., Kang, X., Lu, W., 2010. Critical success factors for on-site sorting of
construction waste: A China study. Resources, Conservation and Recycling, 54, 931-936.
[9] Osmani, M., Glass, J., Price, A.D.F., 2008. Architects perspectives on construction waste
reduction by design, Waste Management, 28 (7), 1147-1158.
[10] Commissariat Gnral au Dveloppement Durable, 2011. Waste Managed by the Building
Companies: Quantities and Management Methods in 2008. Chiffre et statistiques nr. 231, juillet

2011. Ministre du lEcologie du Dveloppement Durable du Transports e du Logement, France,

p.5 (in French).
[11] Coelho, A., Brito, J., 2013. Economic viability analysis of a construction and demolition waste
recycling plant in Portugal part I: location, materials, technology and economic analysis. Journal
of Cleaner Production, 39, 338-352.
[12] Coelho, A., Brito J. 2013 B. Environmental Analysis of a Construction and Demolition Waste
Recycling Plant in Portugal Part I: Energy Consumption and CO2 Emissions. Waste
Management 33 (5), 125867.
[13] Saez, P.V., Merino, M.R., Gonzlez, A.S.A., Amores, C.P., 2013. Best practice measures
assessment for construction and demolition waste management. Building constructions,
Resources, Conservation and Recycling, 75, 52-62.
[14] Poon, C.S., Yu, A.T.W., Ng, L.H., 2001. On-site sorting of construction and demolition waste in
Hong Kong, Resources, Conservation and Recycling, 32 (2), 157-172.
[15] Rodrguez, G., Alegre, F.J., Martnez G., 2007. The contribution of environmental management
systems to the management of construction and demolition waste: the case of Autonomous
Community of Madrid (Spain). Resources, Conservation and Recycling, 50 (3), 33449.
[16] Kohler, G., Kurkowski, H. Optomising the use of RCA. Sustainabel Construction: Use of recycled
concrete aggregate. Proceedings of the Int. Symposium held at Department of Trade and Industry,
Conference Centre, U.K., 11-12 Nov. 1998.
[17] Burke, T.T., Cohen, M.D., Scholer, C.F., 1992. Synthesis Study on Use of Concrete Recycled
from Pavement and Building Rubble in the Indiana Highway System, Joint Highway Research
Project.
[18] Mulder, E., Tako, P.R.J., Feenstra, L., 2007. Closed Cycle Construction: An Integrated Process
for the Separation and Reuse of C&D Waste. Waste Management, 10, 140815.
[19] Schnellert, T., Mueller, A., 2010. Separation techniques for CDW Best practice. www.abwrecycling.de/de/barcelona.html.
[20] Hendriks, C.W., Xing, W., 2004. Quality improvement of granular wastes by separation
techniques. In: Vzquez, E., Hendriks, C.W., Janssen, G.M.T. (Eds.). Int. RILEM Conference on
the use of recycled materials in building and structures. RILEM publications, SARL, 142-149.
[21] Mller A., Schnellert T. Kehr K., 2010. Gypsum reduced, AT Mineral Processing, 51:54-69.
[22] Sampaio, C.H., Tavares, L.M.M. 2005. Beneficiamento Gravimtrico: Uma introduo aos
processos de concentrao mineral e reciclagem de materiais por densidade. First edition, Porto
Alegre: Ed. UFRGS, v.1, 603p.
[23] Webb, P.A., 2001. Volume and Density Determinations for Particle Technologists. Micromeritics
Instrument Corporation, Norcross, Georgia.
[24] Tavares, L.M., 1999. Monte Carlo Simulations on the potential energy theory of jigging, Coal
Preparation, 20, 71-83.

GRAVITY CONCENTRATION OF CONSTRUCTION AND

DEMOLITION WASTE MATERIALS ON WATER JIGS
Carlos H. Sampaio (1), Bogdan Cazacliu (2), Alejandra Liendo (3), Carlos O. Petter (1),
Rgis Paranhos (3), Gerson Miltzarek (1), Weslei M. Ambrs (1), Marcos S. Oliveira (4)
(1) UFRGS, Department of Metallurgy, Mineral Processing Laboratory, Brazil
(2) LUNAM, University of Nantes, IFSTTAR, Aggregates and Material Processing
Laboratory, France
(3) UNIPAMPA, Planning and Mineral Processing Laboratory, LATRAM, Brazil
(4) Odebrecht Latinvest, Peru
Abstract
The paper describes separation of concrete, brick, ceramic and gypsum particles on a water
jig, AllMineral Company, modell AllJig S 400/600X400. The particles used in the tests
were obtained from comminuted samples of solid materials, without the presence of
contaminants. Not real samples of demolitions were used. The particle size distribution to
perform the tests was 20 to 2 mm. The tests were carried out with two and three different
particles in each test. During the tests, high contents of gypsum or concrete in concentrates
were reached. Separation of ceramic from brick particles does not occur in an efficient way.
Keywords: jigging, gravity concentration, water jig, recycled concrete aggregates
1.
INTRODUCTION
The term Construction and Demolition Waste (C&D) is used to define solid waste
generated in construction, renovations and demolition of buildings [1]. A significant amount
of C&D waste material is generated annually all over the world. Only USA generate about
350 million tons per year [2]. The quick growth of cities increased the generation of C&D
waste.
The C&D waste is a mixture of different solids. Their use, without any treatment, is very
restricted. It is generally used in landfills, causing environmental problems [3,4]. Part of the
C&D waste consists of inert materials that can be reused in civil engineering as granulate [5].
For that purpose, their separation by species or group of materials with similar properties is
required.
A large part of the debris generated in the civil construction are made of red bricks,
concretes, gypsums and ceramics. These materials show different characteristics that can be
used for their separation. A characteristic that differentiates them is the density and gravity
concentration processes can be used to beneficiate these materials.

Mineral processing techniques based on density differences of the particles are known and
largely used in Europe since the Middle Ages [6]. Nowadays, millions of tons of different
ores are concentrated by this technique, such as, iron ore, coal, alluvial minerals, etc.
Among the different equipments of gravity concentration, Several works were published of
gypsum concentration with the use of jigs, for recycling or simply for the removal of the final
concentrate [7,8,9]. Jigging is a separation process, which consists of repeated expansion
(dilatation) and contraction (compression) of a particle bed, by the use of a medium, usually
water or air. The result is the stratification of the bed with increasing densities of the particle
from the top to the base.
Jigs were and remain widely used mainly because of their low costs. Besides presenting
low operational costs, jigs are robust, have high capacity, are easy to operate and beneficiate
relatively large particle distribution, which simplifies mineral processing circuits. In
comparison with other beneficiation processes, jigs present great capacity to absorb large
fluctuations of ore contents, feed rates and solid percentages.
This study aims to determine the possibility of treating particles of concrete, brick, ceramic
and gypsum on water jigs.
2.
MATERIALS AND METHODS
All jigging tests were performed with the following materials: concrete particles (type
30MPa at 28 days), ceramic particles, brick particles and gypsum particles (Figure 1). The
particles were obtained from comminuted samples of solid materials, without the presence of
contaminants. No real samples of demolitions were used. The particle size distribution to
perform the tests was 20 to 2 mm (Table 1). The material was initially comminuted to a top
size of 20 mm and then separated the size 20 to 2 mm. This particle size was chosen due the
separation efficiency, which is appropriated for jigs [10].
(a)
(b)
(c)
(d)
Figure 1 : Particles of concrete (a), ceramic (b), brick (c) and gypsum (d) used in stratification
tests.

Table 1 Size distribution of the particles.

Material
Concrete
Ceramic
Brick
Gypsum
Size (mm)
Mass (%)
Mass (%)
Mass (%)
4x20
72.4
Mass (%)
94.3
70.3
66.2
2x4
10.9
2.8
10.9
9.3
0.1x2
15.5
2.4
16.2
23.5
<0.1
1.2
0.5
2.6
1.0
Total
100
100
100
100
For each material the skeletal and envelope density were measured. To obtain the skeletal
density a helium pycnometer (multipycnometer Quantachrome) was used. The envelope
density was calculated by weighting and water volume displacement after surface
waterproofing. The bulk density were determined fulfilling a jig chamber with each different
material. The mass was divided by the jig chamber volume (Table 2).
Table 2 Skeletal, envelope and bulk density of the materials.
Material
Concrete
Ceramic
Brick
Gypsum
Skeletal Density Envelope Density Skeletal Density in

(g/cm3)
(g/cm3)
Water (g/cm3)
2.67
2.39
2.49
2.56
2.32
2.41
2.59
2.26
2.39
2.30
1.86
2.05
Bulk Density
(g/cm3)
0.988
0.813
0.675
0.475
Table 3 presents Concentration Criterion made for the different particle separation
considering Envelope Densities, Skeletal Densities and Bulk Densities. Concentration
Criterion with bulk density was carried out with a simple relation between particle densities.
Table 3 Concentration Criterion for the different particles.
Concentration Criterion
Concrete / Ceramic
Concrete / Brick
Concrete / Gypsum
Ceramic / Brick
Ceramic / Gypsum
Brick / Gypsum
Envelope Skeletal Bulk

Density in water Density
1.05
1.06
1.22
1.10
1.07
1.46
1.62
1.42
2.08
1.05
1.01
1.20
1.53
1.34
1.71
1.47
1.32
1.42
A water jig of AllMineral Company, modell AllJig S 400/600X400, was used to perform
the tests (see Figure 2-a). The optimization of the best jigging parameters were determined
and were fixed for all tests: Expansion, 40 mm; frequency 70 pulses per min; height of water
above the particle layer, 30 mm; and test duration, 180 s for tests with 2 different particles and
240 s for tests with 3 different particles.
The following dried mass of particles was used in tests: 7,900 g of concrete, 6,500 g of
ceramics, 5,400 g of bricks and 3,800 g of gypsum. The amount of each materials (concrete,

ceramic, brick or gypsum) was chosen to completely fulfill each box (see figure 2-b). Before
starting the tests, the particles (concrete, ceramic, brick and gypsum) were completely mixed.
(a)
(b)
Figure 2 : (a) Water jig used in the tests ; (b) Material placed in the jig before the test.
After the test the particles are stratified in layers (Figure 3-a) according to the density.
Nearest the bottom of the jig, particles with higher density are preferentially concentrates.
Above of the first layer, particles with lower density are deposited.
After stratification and water removal, the the particles layers are removed one by one
(Figure 3-b), from the top to the bottom. With the material of each chamber, concrete,
ceramic, gypsum and brick particles are separated by hand (due to their different colors),
dried and weighed to know the materials percentage in each layer.
(a)
(b)
Figure 3 : Material stratified (a) and jig chambers (b).

For the tests with two different particles, it was called Inferior Layer the chamber close to
the bottom of the jig and Superior Layer the chamber on the top. For the tests with three
different particles, it was called Inferior Layer the chamber close to the bottom of the jig;
Middle Layer the chamber in the middle; and Superior Layer the chamber on the top.
3.
3.1
Jigging studies with 2 different particles
Table 4 presents the results for separation in water jig of all mixtures used for tests with
two different materials. It may be said that the separation between concrete and gypsum is
quite simple and efficient to be performed. In the superior layer there is almost 97% of
gypsum particles, while in the inferior layer more than 96% consists of concrete particles.

The displaced material, 3.1% of concrete particles in the superior layer and 3.6% of
gypsum particles in inferior layer, is probably more related to the separation of each chamber
(removal of the chambers from each other) than to the separation efficiency of the equipment.
It is expected in industrial processes the completely removal of all gypsum particles. The
envelope density difference between the two types of materials (concrete = 2.39 g/cm3 and
gypsum = 1.86 g/cm3) provides a Concentration Criterion (C.C.) of 1.62 (envelope density)
and 2.08 (bulk density), which are reasonable values for a good separation.
Table 4 Results of separation in water jigs of binary mixtures of particles.
Mixture
Superior layer (% mass)
Inferior layer (% mass)
Gypsum Concrete Ceramic Brick Gypsum Concrete Ceramic Brick
Concrete/Gypsum 96.9%
3.1%
3.6%
96.4%
Ceramic/Gypsum 96.0%
4.0%
7.2%
92.8%
Brick/Gypsum
64.5%
35.5%
9.2%
90.8%
Concrete/Ceramic
13.6%
86.4%
81.6%
18.4%
Ceramic/Brick
Concrete/Brick
31.0%
44.4% 55.6%
69.0%
80.2%
67.3% 32.7%
19.8%
In the separation of ceramic and gypsum particles, as the ceramic envelope density (2.32
g/cm3) is very similar to the concrete envelope density (2.39 g/cm3), the C.C. of the mixture
Ceramics / Gypsum is 1.53, slightly below the mixture Concrete / Gypsum.
The small difference in C.C. is confirmed by the masses of ceramic and gypsum found in
the superior and inferior layers. The superior layer is constituted by 96% of gypsum particles
and 4% of ceramic particles. Values slightly worse if compared with Concrete / Gypsum
mixtures. Since the envelope densities of ceramic and concrete particles are similar, the
results are nearly identical. One can expect that in mixtures of particles of concrete, gypsum
and ceramics, almost a complete removal of gypsum particles is possible. In the same way, it
is also possible to recover gypsum concentrates to be recycled with high contents (more than
95%), with mixtures of gypsum and concrete/ceramic particles.
Concerning the separation of brick and gypsum particles, the C.C. for treating this
materials are 1.47, which is slightly lower than the separation of Ceramic / Gypsum (1.53). In
this case, the separation has a slightly lower efficiency. The inferior layer consists of 90.8% of
brick particles and 9.2% of gypsum particles. Even so, it is possible to obtain a dense product
(inferior layer) with more than 90% of bricks particles.
In the case of separation of concrete and brick particles, C.C. is 1.10 (envelope densities).
Nevertheless, the inferior layer has 80.2% of concrete particles. The presented separation
efficiency is much better than expected. For C.C. just over 1, with this particle size range, the
separation efficiencies should be very low.
The bulk density of concrete and brick particles are 0.988 g/cm3 and 0.675 g/cm3,
respectively, and the C.C. of these particles is 1.46. Same explanation can be given for
mixtures of concrete and ceramics (bulk density of 0.813 g/cm3) particles, where C.C. is 1.22
(C.C. of bulk densities).
For the separation of concrete and ceramic particles, it can be noticed that the values of
particles separation of Concrete / Brick and Concrete / Ceramic are almost the same. The
inferior layer of the mixture Concrete / Brick presents 80.2% of concrete particles (19.8% of

brick particles), and the inferior layer of the mixture Concrete / Ceramic presents 81.6%
(18.4% of brick particles).
In relation to the separation of ceramic and brick particles, as already mentioned, the two
materials present close envelope densities (Table 2) and similar shape (Figure 1 - photos b
and c), making not possible their separation in water jigs. Even the bulk densities of the
particles are close (0.813 g/cm3 and 0.675 g/cm3 for ceramic and brick particles respectively)
and C.C. of bulk densities are 1.20.
The superior layer has 55.6% of bricks particles (present slightly lower density than
ceramic particles) and 44.4% of ceramic particles. If the two materials have the same shape
and the same envelope density, after separation, the layers would result in 50% of each one.
3.2
Jigging studies with 3 different particles
The stratification with 3 different particles were also tested. The results are shown in
Figure 4. It can be seen that the removal of gypsum particles is an easy task. In all
combinations, the amount of gypsum particles in the superior layer was always above 80%:
89.6% in mixture of concrete, bricks and gypsum particles; 89.6% in the mixture of ceramic,
brick and gypsum particles; and 81.8% in mixture of concrete, ceramic and gypsum particles.
From this data, with the use of just 1 or 2 jigs in series, it is expected concentrates with high
gypsum contents, as well as high mass recoveries.
Concrete/Ceramic/Gypsum
100
80
60
40
20
0
Gypsum
Brick
Concrete
Mass fraction
Mass fraction
Concrete/Brick/Gypsum
Inferior layer
Middle layer
0.1
8
91.9
5.1
82.1
12.8
Superior
layer
89.6
10
0.4
100
80
60
40
20
0
Gypsum
Ceramic
Concrete
Gypsum
Brick
Ceramic
Mass fraction
Mass fraction
100
80
60
40
20
0
Middle layer
0.1
8
91.9
5.1
82.1
12.8
Middle layer
0.1
18.9
81
0.5
78.6
20.9
Superior
layer
81.8
15.9
2.3
Concrete/Ceramic/Brick
Ceramic/Brick/Gypsum
Inferior layer
Inferior layer
Superior
layer
89.6
10
0.4
100
80
60
40
20
0
Brick
Ceramic
Concrete
Inferior layer
Middle layer
4.9
17.1
78
31
47
22
Superior
layer
57.3
31.2
11.5
Figure 4 : Separation in water jig of ternary mixtures of Concrete, Brick, Ceramic and
Gypsum particles.

Similar separation behavior between ceramic and brick particles for binary mixture is also
observed in this test, where similar amounts of ceramic and brick particles occurs in the
superior and middle layers. On the other hand, the production of a concrete concentrate will
be an easy task. Even in mixtures of concrete, ceramic and brick particles, where the envelope
densities are similar but different shapes (commented above), the amount of concrete particles
in the inferior layer is almost 80%.
4.
CONCLUSIONS
The main conclusions of this work are the following:

One can expect that in mixtures of particles of concrete, brick, ceramic and gypsum almost
a complete concentration of gypsum particles is possible.
Concentration of gypsum particles is an easy task in water jigs.
In the separation of concrete and brick/ceramic particles, the Concentration Criterion is
around 1.10, making separation very difficult. Even with a low C.C., separation occurs due to
the difference of bulk densities of the particles.
With the exception of the mixture of ceramic and brick particles, it is possible the
separation of other mixtures of particles with reasonable efficiency.
Production of concrete concentrates will be an easy task. Even in mixtures of concrete,
ceramic and brick particles, where the envelope densities are similar, the amount of concrete
particles in the inferior layer is almost 80%.
Separation of ceramic particles from brick particles does not occur in an efficient way, due
to their similar envelope and bulk densities and shapes.
REFERENCES
[1] Wang, J., Yuan, H., Kang, X., Lu, W., 2010. Critical success factors for on-site sorting of
construction waste: A China study. Resources, Conservation and Recycling, 54, 931-936.
[2] Lachimpadi, S.K., Pereira, J.J., Taha, M.R., Mohktar, M., 2012. Construction waste minimization
comparing conventional and precast construction (Mixed system and IBS) methods in high-rise
buildings: A Malaysia case study. Resource, Conservation and Recycling, 68, 96-103.
[3] Cazacliu, B., Sampaio, C.H., Miltzarek, G., Petter, C., Guen, L., Paranhos, R., Huchet, F.,
Kirchheim, A.P., 2014. The potential of using air jigging to sort recycled aggregates, Journal of
Cleaner Production, 66, 46-53.
[4] Coelho, A., Brito, J., 2013. Economic viability analysis of a construction and demolition waste
recycling plant in Portugal part I: location, materials, technology and economic analysis. Journal
of Cleaner Production, 39, 338-352.
[5] Medina, C., Banfill, P.F.G., Snches de Rojas, M.I., Fras, M., 2013. Construction and Building
Materials. Rheological and calorimetric behavior of cements blended with containing ceramic
sanitary ware and construction/demolition waste, 40, 822-831.
[6] Agricola, G., 1556. De Re Metallica, (Translated by H.C. Hoover and L.H. Hoover), Dover
Publications (1950).
[7] Mller, A., 2010. Gips reduziert. AT Mineral Processing, 51, 54-69.
[8] Mller, A., 2010 A. Gips im Griff. AT Mineral Processing, 51, 34-43.
[9] Mller, A., 2012. Bauschutt ohne Gips. Steinbruch und Sandgrube, 11/2012, 40-45.
[10] Sampaio, C.H., Tavares, L.M.M. 2005. Beneficiamento Gravimtrico: Uma introduo aos
processos de concentrao mineral e reciclagem de materiais por densidade. First edition, Porto
Alegre: Editora UFRGS, v.1, 603p.

SELECTIVE COMMINUTION OF CONCRETE WASTE AND

RECOVERY OF THE CONCRETE CONSTITUENTS IN A GRINDINGDRYING PROCESS
Andr Btz (1), Holger Wulfert (2), Paul Erwerth (3) Andreas Schiffers (4), Andreas
Jungmann (5), Dirk van Mechelen (6)
(1) Loesche GmbH, Dsseldorf Germany
(4) Cala Aufbereitungstechnik GmbH & Co. KG, Aachen Germany
(5) Cala Aufbereitungstechnik GmbH & Co. KG, Aachen Germany
(6) Aggregate Consultants bvba, Willebroek Belgium
Abstract
This article describes a new process for the separate recovery of the constituents of
concrete waste by means of selective comminution in combination with dry separation by
density. The process is based on a Loesche roller mill, which liberates the gravel and sand
fractions in a non-destructive manner. Post-separation of the coarse fraction using an air
jigging machine and/or a dry fluidized bed separation ultimately produces a recyclate of such
a good quality that it can be used to completely replace the natural rock fraction. It can be
demonstrated that under equal concrete manufacturing conditions the obtained recyclates
(recycled gravel, sand and filler) are all just as good as, if not superior to, the natural rock
fraction. Moreover, utilisation of filler consisting mainly of ground hardened cement paste
fraction promises residual reactivity. This process also enables the alternate grinding of very
different primary and secondary raw materials, such as steel slag, granulated blast furnace
slag, clinker and concrete in a single plant, operated in campaigns.
1.
FUNCTIONING OF SELECTIVE COMMINUTION IN A LOESCHE MILL
The fundamental idea is to use a Loesche mill, a grinding unit that is normally employed
for fine comminution, for reducing a material consisting of several components into its
individual components, without involving an undue amount of comminution work. The
multicomponent material in this case is concrete, which essentially consists of three solid
components gravel, sand and hardened cement paste.
From the numerous measurement criteria that can be applied for characterising the strength
of solid components, the authors decided to use the Vickers hardness of the individual
components for the purpose of describing the comminution resistance. To enable this, a series
of raw material tests was conducted at the Institute for Mineral Processing Machines at the
University Bergakademie Freiberg. Vickers hardness was determined from polished sections

prepared from relevant samples. By way of example, Figure 1 shows the great difference
between the strength values of the examined individual concrete components.
Concrete Components
Determination of Vickers hardness
Figure 1 Vickers hardness values of the individual concrete components

Given such differences like in this case, it is appropriate to operate the comminution
process in a manner that exclusively disintegrates the hardened cement paste/cement paste
matrix and preferably leaves the other components in their original particle size. However,
owing to the low strength of the hardened cement paste matrix, a gentle attritional pressure
crushing is more appropriate for this purpose. The disintegration and separation process is as
schematically represented in Figure 2:
Figure 2 Process sectors in the Loesche mill
The feed material (1) is deposited onto the centre of the grinding table (2) and then carried
under the grinding roller (3) by centrifugal force. A bed of material is formed between the
grinding roller and the grinding table, the height of which is regulated by the dam ring around
the outer edge of the grinding track. Due to appropriate setting of the grinding table rotation
speed and the grinding force the material in the bed is subjected to a shearing, frictional

compressive stress. The only difference to a standard design Loesche mill for the cement
industry is the special design of the hydraulic pressure system of the grinding rollers. A
substantial flow of material is discharged through the vane ring (4), even against the air
stream. This is the first point of material separation/classification, as lumpy and heavy
material falls into the discharge system (8) due to the force of gravity, despite the flow
strength of the rising stream of air. The air stream (5) transports the finer and lighter material
upward to the second classification stage, the dynamic separator (6). This separator collects
the adequately fine-ground filler from the coarser material, the fine sand and the remaining
mortar/hardened cement paste aggregates. These rejects pass through the separator cone (7)
and are returned to the grinding process.
2.
TESTS IN THE LOESCHE PILOT PLANT
2.1
Execution
All the tests were conducted in the Loesche pilot plant using a test mill of the type LM3,6
(Fig 3) which is normally used for trials in order to obtain the fundamental data for designing
machines, processes and plants for the cement industry.
Figure 3 Pilot plant mill LM 3.6

Detailed testing was performed with two types of concrete, the first with a quality of C
20/25 (30 MPa) subsequently called concrete 1 and the second with a quality of C 55 (60
MPa) subsequently called concrete 2.
The compositions of the concretes specially produced for the tests are shown in Table 1.

Table 1 Composition of the tested concrete qualities and of the hardened cement pastes
Concrete formulation
Material strength (28d)
Concrete (total)
Gravel
Sand
Concrete granulate
Hardened cement paste
True density
Hardened cement paste (total)
Cement
Slag
Limestone meal
Water
Concrete chemistry
MPa
kN/m2
M.-%
M.-%
M.-%
M.-%
M.-%
1
30
30 000
100
43.9
26.5
5.0
24.6
2
60
60 000
100
40.6
25.7
5.5
28.2
kg/dm3
M.-%
M.-%
M.-%
M.-%
M.-%
M.-%
2.13
100
19.2
44.8
0.0
35.8
0.2
2.23
100
35.9
35.9
0.0
27.7
0.5
For evaluation of the tests, the density of the concrete aggregate was defined as max. 2.5 t/
m, while the densities shown in Table 1 (2.13 and 2.23 t/m) were taken as a basis for the
filler. The evaluation was performed exclusively for the produced gravel (2-8 mm) and the
filler derived from hardened cement paste. In addition to the mass balancing of the test results
into the products sand, gravel and filler, the coarse products discharged from the mill were
also subjected to a float/sink analysis at a density of 2.5 t/m.
2.2
Separation Results
For both concrete types the achieved yield rate of filler product exceeded 85 %.
Moreover, the produced filler consisted to over 90 % of ground hardened cement
paste.
Parallel to that, the coarse grain fraction of > 2 mm contains a particle size mixture
consisting to 60-70 % of gravel with almost totally clean surfaces that has
obviously retained its natural round form. By contrast, the product with a density of
< 2.5 t/m is a mixture of agglomerates consisting of coalesced and of unground
mortar/hardened cement paste particles (Figure 6). In this case, the yield of gravel
is approx. 90 %.
If the focus is placed on obtaining a well-cleaned gravel product with minimal
adhesions and a minimal amount of coalesced components, the grinding circuit can
be set such that the coarse grained product has a purity of 93-95 %. In this case, the
yield of hardened cement paste in the filler product rises to over 90 %, although the
then produced filler product only has a hardened cement paste purity of 60-70 %.
Subsequent cleaning of the coarse grained product (sand and gravel) by means of
highly-selective density separation enables qualitative improvement of the fine and
coarse aggregates. Dry separation systems, such as those successfully employed for
coal and slag separation, are suitable for this post-cleaning stage. [1, 2]

Figure 6 Particles from the coarse ground product, embedded in resin for microscopic
examination. Left: light material (-2.5 t/m, right: heavy material (+ 2.5 t/m)
2.3
Usability results
Just the optical impression of the two products is enough to show that the concrete
recyclate produced by the Loesche process has greater resemblance to a sample of
conventional natural gravel than to a typical recycled concrete rubble product.
Figure 7 Granulates Left: 2/8 mm after the Loesche process, right: 4/30 mm, conventional
2.2.1 Usability of the gravel fraction
Concrete test blocks were produced using different mixtures of aggregates, but keeping a
constant water-to-cement ratio and constant amount of added binder. The following table
compares three comparable formulations and their results, which have been selected from the
numerous test series.
These formulations are mixtures of the following aggregate products:
Natural 2/8 mm gravel and natural 0/2 sand: Designation: NA

Recycled aggregates from a standard process, grain-size curve adapted to NA: RC
Recycled aggregate from the Loesche process, grain-size curve adapted to NA:
RCLM
Table 2 Comparison of the concrete properties after alternative usage of aggregates

Natural aggregates
Recycled aggregates
Recycled aggregates from Loesche mill
Cement CEM I 52.5 R
Water/Cement
Plasticiser/cement
Slump test 10
Slump test 60
Compressive strength 28 days
Compressive strength 56 days
M.-%
M.-%
M.-%
kg/m3
%
mm
mm
MPa
MPa
100
0
0
180
0.37
0.8
256
232
63
74
50
50
0
180
0.37
0.8
255
185
55
63
50
0
50
180
0.37
0.8
259
225
74
82

The concrete produced from natural gravel/RC material shows a reduction in workability,
indicated by the slump data, as well as lower compressive strengths than the concrete
produced exclusively from natural material (strength reduction between 10 and 15 %). The
characteristics of the formulation using natural aggregate and recyclate from the Loesche
process was astonishing, because the produced concrete had comparable workability and also
an increase in compressive strength of over 10 %. Possible explanations for the significantly
higher compressive strength of concrete containing recycled aggregate from the Loesche mill
are that due to the attritional stressing of the gravel particles their surfaces were roughened,
resulting in improved bonding capability of the CSH phases compared to the smooth surface
of primary gravel particles, or that the remaining coating of fine hardened cement paste on the
gravel particles promotes an improved adhesion of the new hardened cement paste.
2.2.2 Usability of the hardened cement paste fraction of the filler (RCLM)
For the scenario of reduced addition of binding agent (POZ), orientation tests had the aim
of indicating the degree to which a RCLM filler develops a reactivity that contributes to the
strength of the concrete. Tests with reference cement CEM I 42.5 R examined the influence of
the strength development in standard prisms in accordance with DIN EN 196 for two samples
of filler (hardened cement paste from the recycling tests of concrete qualities 1 & 2). The
individual mixtures were as follows:
30 weight-% of filler concrete 1 + 70 weight-% of reference cement

30 weight-% of filler concrete 2 + 70 weight-% of reference cement
Reference cement as initial binding agent (CEM I 42.5 R) (28-day value: 48.7
MPa)
30 weight-% of quartz sand (0/2 mm fraction) + 70 weight-% of reference
cement)
The quartz sand was employed as inert material with the purpose of being able to compare
the increase in strength due to the RCLM filler.
The results of the compressive strength test after 28 days are shown in Figure 8.
30 weight-%
30 M.-%RC-filler
Fller
30 MPa
37,4
20,0
10,0
0,0
33,1
30,2
70 M.-% Referenzzement
30 M.-% Quarzsand
30,0
40,0
Beton-Recovery / Festigkeiten
30 M.-% RC-30-1
50,0
60 MPa
48,7
30 M.-% RC-60-2
60,0
Druckfestigkeit (N/mm2)
Compressive Strength after 28 days (N/mm)
Lagerdauer: 28 d
FIB-Weimar / 06.01.2015
Druck-RC-2

Figure 8 Increase in compressive strength (MPa) after 28 days due to the usage of
reclaimed filler compared to the inert additive of 30 weight-% of quartz sand
The increase in compressive strength due to the 30 % addition of RCLM filler, i.e. filler with a
content of more than 70 % ground hardened cement paste, proves that this filler possesses
substantial binding agent reactivity.
In the case of the highest value (RCLM filler from concrete 1), it can be seen that the
replacement of 30 % Portland Cement by 30 % RCLM filler can compensate more than half
the loss of strength. Further test work is currently being carried out to answer the question
regarding the cause of the strength increase resulting from the use of recycled gravels.
3.
PLANT CONCEPTS
3.1
Concrete processing
Employing jigging machines [3, 4], the plant concept for producing aggregate materials
from waste concrete combines disintegration in a Loesche mill with efficient density
separation as it is shown in Figure 9. The only difference to the mentioned references is that
in this case the sorting is done by dry density separation processes.
Figure 9 Process for concrete rubble with the Loesche system

After the grinding, the overflow product (rejects) of the mill is separated from the sand
and gravel fraction by screening. Next, density separation on an air jigging machine [1]

separates the portion of finished gravel from the coalesced components that are not yet
completely disintegrated.
Analogue to the processing of the gravel fraction, the sand fraction is subjected to density
separation in a fluidised bed [2], which selectively separates the heavy material (sand
fraction) from the mortar fractions and not yet completely disintegrated hardened cement
paste conglomerates, i.e. the light material.
3.2
One mill for concrete processing AND for other multiple grinding & recycling
solutions
It becomes more and more popular that tailor made fillers/binders are produced.
Worldwide the production of composite cements (CEM V) is growing. A lower footprint in
CO2, less transport of raw materials availability of suitable raw materials in the
neighbourhood of the place where cement/binders are needed opens the possibility to
manufacture the proper binder nearby. One key word in this context is urban mining.
A lot of thermal/metallurgical processes produce a residue which can be used as a
component for a binder type CEM V or similar.
The authors are aware that not all ashes and slag types can be used fine ground as binders,
but some can.
Example: GBFS, BOF-slag, gypsum, limestone and concrete:
It is possible to use a standard Loesche mill not only for processing waste concrete, but also
for the production of further materials for the cement industry. Due to the flexible operational
mode of the Loesche mill and to the mills ability to change over from one type of feed
material to another within just a few minutes, it would be practicable to use one and the same
mill for alternately grinding concrete rubble, granulated blast furnace slag and possibly also
BOF-slag, limestone, gypsum or even cement clinker
Such a plant consists of 4 process stages:
(1)
(2)
(3)
(4)
The feed system for granulated blast furnace slag, gypsum, limestone or clinker
The central grinding plant with hot gas producer and filter system
Feed and processing system of the waste concrete or BOF slag
Treatment of the coarse mill discharge material by means of a dry density
separation process

Figure 10 Flow sheet of a Loesche-system co-grinding plant

All the process variants and partial processes have been tried and proven on an industrial
scale.
Sections (3) and (4) can be equipped with dry density separation systems in a fluidized bed
of an air jigging machine as shown here. Depending on the raw material to be treated
additional dry sorting can be incorporated such as low, medium and high intensity magnetic
sorting system, or with eddy current separators (e.g. municipal waste incineration ashes) as
well as sensor sorting systems. By that efficient metal (steel and others) recovery as well as
selective recovery of the concrete components can be realized in combination with a final fine
ground filler product.
4.
SUMMARY
Defined concrete made from river sand and gravel has been selectively comminuted by a
Loesche mill generating 3 recycled products, gravel (35-40%), sand (30-35%) and filler (2530%) from mortar, mass balance in brackets). For the additional cleaning of the gravel
fraction dry density separation (air jigging) was tested. The recycled aggregate has been used
successfully in second life application for concrete and the filler as well has been used to
identify the remaining hydraulicity. The filler (d50 30-40m, 10%> 90-100 m) from this
specific concrete turned out being able to replace a certain amount of cement, due to a
remaining reactivity. The complete process is shown including dry density separation and also
the alternative set-up to combine a layout of such a concrete treatment plant with grinding of
other materials such as conventional cement materials or different types of slag from iron and
steel making.
This process has been developed and tested for defined concretes with defined aggregate
used. It must be kept in mind that the reproducibility of this process for other concrete mix
has to be tested individually.

REFERENCES
[1] Horn, Andreas, et al.: Trockene Aufbereitung von Stein- und Braunkohlen mit der allairLuftsetzmaschine; Surface Mining-Braunkohle & Other Minerals, 55 (2003) No.2
(April/June), S. 185-189
[2] Jungmann, Andreas et al.: Dry Processing and High Quality Application for Steel Slag;
World of Metallurgy Erzmetall 63, 2010, No. 1, S. 5-10
[3] Jungmann, Andreas, et al.: Bauschuttaufbereitung mit alljig-Setzmaschinen in Europa und
USA; Aufbereitungstechnik, 38, 1997, Heft 10, S. 543-549
[4] Jungmann, Andreas, et al.: Gravity separation of rubble a contribution to the closed cycle
of raw material usage; Proceedings of the XX. IMPC Aachen, September 1997, Vol. 5.

RESEARCH AND MODIFICATION ON RECYCLED AGGREGATES

Jianzhuang Xiao (1), Guanzhen Zhang (2)
(1) Department of Structural Engineering, Tongji University, P.R. China
(2) Department of Structural Engineering, Tongji University, P.R. China
Abstract
Recycled concrete containing recycled aggregates not only consumes larger amount of water,
but also has lower mechanical properties and durability when compared with normal concrete.
Coupled with the strength degrees and source of parent concrete, all these shortcomings limit
recycled aggregates further application. Modification of recycled aggregates has become a
prominent technology for solving such problem. This paper presents current modification
methods, mainly include: (1) physical technology, its basic idea is to remove the waste cement
paste adhered to recycled aggregates; (2) chemical technology, carried out by immersing recycled
aggregates in different kinds of chemical grout; (3) carbonation technology, when newly collected
crushed aggregates are put into atmosphere with high CO2 concentration, the adhesive cement
paste will achieve higher properties after carbonation reaction. At last, we also consider the
modified gradation of recycled aggregates. The comparisons and evaluations of these methods
are summarized in detail, conclusions are put forward and several problems concerning
modification need to be considered for further research and application are proposed and
discussed.
Keywords: Recycled aggregates, strengthening, modification, carbonation, gradation.
1.
INTRODUCTION
The demolition of buildings will produce more than 200 million solid wastes in China
every year, and 80% are waste concrete, these wastes are usually buried or piled up, which
will occupy more arable lands or even produce pollution to the lands. Besides, China is also a
big country in concrete production and consumption, sands and aggregates consumed in
concrete had surpassed 10 billion tons in 2013. Therefore, its urgent to develop the recirculation and re-utilization of building resources [1]. Aggregates in concrete have relatively
high strength and wont have great change in performance even after parent concretes crush
or demolition, so they have the potential feasibility for further reuse [2]. However, crushing or
demolition processes will bring negative effects to the properties of recycled aggregates,
especially the adhesive coarse and hardened cement pastes, which include numerous microcracks or pores. The complex compositions of recycled concrete are shown in Figure 1. When
compared with natural aggregates, these detrimental factors will make recycled aggregates
apparent density and bulk density lower while porosity, silt content, water absorption,
crushing index value higher [3].

Moreover, when considering the strength degrees and sources of parent concrete, recycled
aggregates properties will have more variations. All these defects confine them to be used in
narrow fields such as non-bearing structures with lower strength. Therefore, in order to
improve the circulation rates of recycled aggregates, its necessary to improve their properties
by developing the research on strengthening and modification. Researchers all over the world
have conducted majority experiments on modification of recycled aggregates, such
modifications can be summarized as physical technology and chemical technology. In
addition, we also introduce the carbonation and gradation of recycled aggregates.
Natural aggregate
Interfacial zone between

aggregate and adhering
cement paste
Adhering cement paste

New cement paste

adhering cement paste
and new cement paste
Figure 1: Schematic diagram of recycled concrete

2.
PHYSICALTECHNOLOGY
2.1 Vertical Eccentric Grinding Method

Vertical eccentric grinding equipment [4] invented in Japan includes outer cylinder wall,
inner eccentric wheel and driving device. The inner-diameter of the outer cylinder and inner
eccentric wheel is 72mm and 66mm, respectively. Put the recycled aggregates that need to be
modified into the cavity between inner cylinder wall and outer cylinder wall, under the high
speed rotation of inner eccentric wheel, cement paste adhered to recycled aggregates will be
removed, so are the recycled aggregates protuberant edges.
Intlet
Mechanical
grinding
Intlet
Steel balls
Inner
cylinder
Partition
Power
Extrusion
grinding
Outlet
Figure 2: Vertical eccentric grinding
Outlet
Figure 3: Ball mill of abrasive grinding
2.2 Abrasive Grinding Method

For grinding the preliminary crushed concrete blocks (diameter 5~40mm), the use of ball
mill or rod mill to remove adhesive hardened cement pastes can produce high quality recycled
aggregates, this idea can be described as abrasive grinding method[4]. The ball mill usually has
dry treatment or wet treatment, while the rod mill is usually designed for wet treatment. Take
the example of ball mill, its working principle is shown in Figure 3. This device is provided
with several partitions, between the partitions there is a lot of steel balls. The interactive
collisions between steel balls and recycled aggregates or interactive collisions simply between
recycled aggregates will remove the adhesive cement paste.

2.4 Heating and Rubbing Method

Japans Mitsubishi Corporation developed this method [4] that has the following principle:
o
the preliminary crushed concrete blocks were heated to 300~400 C to make cement pastes
brittle under dehydration. The heated concrete was rubbed in two devices, during the process
the cement pastes were removed and collected as fine powder. When the heating temperature
was higher, the grinding process was easier. But if the heating temperature surpasses 500,
qualities of recycled aggregates would deteriorate, and the total energy consumed in heating
and rubbing would also expand significantly.
Figure 4: Heating and rubbing method

2.5 Particle Shaping Method
The particle shaping equipment consists of host system, dust removal system, electric
control system, lubricating system and pressure sealing system. Recycled aggregates are
introduced from the upper inlet and then are divided into two streams; one part falls into the
impeller cavity and is projected with high speed; another falls along the aggregates
distribution system and its aggregates then collide with that projected from the impeller. After
several times collisions, recycled aggregates will be crushed and modified [5].
Aggregates
inlet
Dust wiper
The airflow
direction
Induced
draft fan
The air
inlet
Exhaust port
Modified aggregates
Fine powder
Figure 5: Particle shaping equipment
Figure 6: Pilot microwave heating system
2.6 Microwave-assisted Method

This novel method [6] may be effectively used to partially remove the adhering cement
paste by developing high temperature gradients and high thermal stresses within the mortar,
especially at the interfacial zone with the gravels. Furthermore, recycled aggregates exposed
to microwaves can significantly increase other properties. According to experimental studies
[6]
, concrete prepared with microwave-treated recycled aggregates can significantly improve
the mechanical properties such as the compressive and flexural strengths and modulus of
elasticity.
3.
CHEMICAL TECHNOLOGY
Chemical technology is mainly about immersing recycled aggregates into specific

chemical grout which will cohere or fill the micro-cracks or micro-pores of recycled

aggregates. This technology achieves the purpose of modification by changing the chemical
composition of recycled aggregates surface and makes them denser or improves their
strength [7]. The available chemical grout can be listed as follows.
3.1
Mixed Cement Grout
Neat cement grout: the main component is cement without any other additives.
Cement grout with additive Kim powder: Kim powder is a kind of high efficient waterproof
agent; it can penetrate into the recycled aggregates for certain depths.
Cement grout with additive silica fume: the main component of silica fume is a very fine
powder called SiO2, which can fill in the pores easily.
Cement grout with additive fly ash: fly ash will react with cement hydration products such
as Ca(OH)2 to form cementitious hydration products and improve the bonding properties.
Cement grout with additive diatomite: Diatomite is a natural nano-material and has
excellent micro powder filling effect and strengthening effect.
3.2
Polymer Emulsion
Polymer emulsion solution can be used to improve the properties of recycled aggregates,
especially recycled fine aggregates. The tests of mortar blocks made of recycled fine
aggregates after the modification of polymer emulsion solution showed that: the flexural
strength improved obviously, while the compressive strength has little improvement [8].
Table 1 [9] summarized the properties of recycled aggregates by different chemical grouts
treatment. We can see that after being treated, recycled aggregates physical properties had
certain improvement. The strength of recycled concrete which was made of modified recycled
aggregates also increased, especially recycled concrete containing polymer emulsion had the
highest improvement growth.
Table 1: Results of recycled aggregates and recycled concrete
Physical properties of RA
Chemical grouts
Mechanical properties of
recycled concrete
28d
Growth
compressive
rate of
strength
strength
(MPa)
/%
30.78
0
Water
content /%
Water
absorption
rates /%
Apparent
density
/(kg/m3)
Crushing
index
value /%
untreated
2.58
6.77
2470
16.73
Neat cement grout

Cement grout with additive
diatomite
Cement grout with additive
silica fume
Polymer emulsion solution
3.37
6.93
2580
13.24
31.10
1.04
2.71
7.13
2523
12.11
36.39
18.23
2.79
7.43
2534
12.80
35.72
16.05
2.63
6.34
2500
11.80
37.58
22.09
3.3
Acid Solution
Spray proper amount of acid solution on the surface of recycled aggregates, the acid will
have chemical reaction with cement hydration products, this will lead to the regeneration of
the particles adhered on cement paste and obtain the purpose of modification [10]. The coarse
recycled aggregates modification degree was studied by glacial acetic acid solution with a
concentration of 5% and hydrochloric acid solution with a concentration of 3%. The
experimental results are shown in Table 4. The results show that: after modification of coarse
recycled aggregates, the strength and initial elastic modulus of corresponding recycled

concrete improved, and even exceed the value of natural aggregate concrete. Besides,
recycled aggregate concretes Poisson's ratio will reduce, showing a creep reduction.
Table 2: Experimental results of glacial acetic acid and hydrochloric acid
Aggregate types
Natural aggregates
Recycled aggregates
Acid solution
28d compressive
strength (MPa)
initial elastic modulus (103MPa)
Poisson's
ratio
cube
prism
Test value
Standard value
26.3
17.8
26.8
27.0
0.22
24.3
16.5
25.3
27.0
0.24
Recycled aggregates
glacial acetic acid
32.1
23.3
33.0
29.5
0.18
Recycled aggregates
hydrochloric acid
27.3
20.0
28.3
27.0
0.19
3.5 Sodium Silicate

Sodium silicate solution will change the bond performance and density of the recycled
aggregates, and can reduce the recycled aggregates water absorption rate effectively [11]. The
modification of recycled aggregates using sodium silicate solution was performed with
different sodium silicate concentration and duration, after these processes, recycled
aggregates were used to prepare recycled concrete, and then test the mechanical properties.
The following results showed that, with increasing the concentration of sodium silicate, the
compressive strength of recycled concrete with different curing age increased first and then
decreased, which, implied the concentration was not the higher the better, so was the soaking
time. The optimal selection of sodium silicate solution was a concentration of 5% and the best
soaking time should be 1h.
Figure 7: Compressive strength of recycled concrete treated by sodium silicate solution

4.
CARBONATION TECHNOLOGY
Based on recent studies, there is also an innovative method for the modification of recycled
aggregates: carbonation technology. As we all known, concrete has the ability of carbonation,
in fact, its almost the cement mortar that contribute to the carbonation. When concrete is
newly crushed, it will carbonate faster due to much larger exposed surface. Recycled
aggregates contain adequate hydration products, such as calcium hydroxide (Ca(OH)2) and
calcium silicate hydrate (3CaO2SiO23H2O or C-S-H). These components are in equilibrium
and stabilized with high pH and Ca2+ in the pore solution. If the atmosphere contains
substantial amount of carbon dioxide (CO2), the CO2 will diffuse into the pores or cracks in
the cement paste. Thus the CO2will dissolve in capillary pore water and form CO32- that in
turn will react with the Ca2+ in the pore water. Then Ca(OH)2 will continue to dissolve and
CaCO3 will precipitate, the process will last until all of the Ca(OH)2 is consumed, the similar

process will be replaced by C-S-H. In this transformation, water is needed while its amount
seldom changes. The carbonation mechanism is shown in Figure 9.
Carbonation will contribute to volume change, normally because the precipitation of
CaCO3 can fill the empty spaces in the capillary pores or cracks, thus leading to the reduction
of porosity of the cement paste. Meanwhile, both the densification and strength will be
increased. This is really a theoretically feasible technology for recycled aggregates
modification.
solid and liquid, Ca(OH)2
dissove here
2+
Ca and OH diffuse
in this zone
Solid phase
(cement mortar)
Carbonation
reaction zone
Liquid phase
CO2
Gaseous phase Pore
CO2
Liquid phase
Solid phase
(cement mortar)
Resultants of carbonization
reaction adhere to the
surface of cement mortar
Figure 8: Carbonation mechanism in capillary pore Figure 9: Results of concretes strength

The enumerated test results [12] chose recycled mortar aggregates (RMA) to illustrate the
relevant carbonation modification. RMA1 and RMA2with different water cement ratio, and
their carbonated counterparts were CI-RMA1 and CI-RMA2, respectively. Carbonation test
was conducted in a steel airtight container with CO2 concentration higher than 99%. After
carbonation, each type of RMA was used to prepare recycled concrete, and then the
subsequent durability properties would be tested.
Figure 9 showed that, although the compressive strength of concrete prepared with RMA,
carbonated or not, was still lower than that of control concrete at all tested ages, both the CIRMA1 and CI-RMA2 improved the compressive strength of recycled concrete significantly.
At 90 days, the compressive strength of concrete made with CI-RMA2 was only 1% lower
than that of control concrete. Besides, there is also an improvement in the resistance to
chloride ion penetration for the concrete prepared with CI-RMA, and the drying shrinkage
was decreased. Generally, carbonation is applicable for the modification of recycled
aggregates.
Table 3: Physical properties of different aggregates
Property
Density (Kg/m3)
Water absorption (%)
10% Fine value (KN)
5.
Particle
size (mm)
20
10
20
10
14
Natural granite
2620
2620
0.89
0.87
156
Aggregate type
RMA1
RMA2
CI-RMA1
2326
2355
2345
2326
2355
2351
11.82
9.30
7.32
12.25
10.81
7.57
96
116
108
CI-RMA2
2371
2379
4.84
4.95
134
THE GRADATION MODIFICATION OF RECYCLED AGGREGATES
The particle size proportion of recycled aggregates, which is different with that of natural
aggregates, will influence the property of concrete. By regulating the gradation of recycled
aggregates we will get better recycled concrete. Research shows that the modification of
recycled aggregates gradation will improve the bulk density and reduce the crushing index.
Besides, the gradation curve will become reasonable.

For recycled aggregates with particle size of 5-16mm, 16-31.5mm and 0-5mm [13]. After
recycled aggregates gradation modification, the results of bulk density and crushing index of
recycled coarse aggregates are shown in Table 4. Compared with primary gradation, modified
gradation improved the recycled coarse aggregates: both the loose bulk density increased and
the compacted bulk density increased, while the crushing index decreased.
Table 4: Gradation modification results of recycled aggregates
Aggregate types
5-16mm recycled
coarse aggregates
16-31.5mm recycled
coarse aggregates
0-5mm recycled fine
aggregates
Primary gradation
Modified gradation
Primary gradation
Modified gradation
Primary gradation
Modified gradation
Loose bulk
density(kg/m3)
1281
1305
1326
1355
1335
1339
Compacted bulk
density(kg/m3)
1372
1421
1390
1425
1343
1346
Crushing
index/%
14.0
12.7
12.4
11.5
18.7
17.1
Recycled concrete made of recycled aggregates with primary gradation was compared with
that made of recycled aggregates with modified gradation; 3 curing age were considered: 7d,
14d and 28d; the water cement ratio of all recycled concrete is 0.4. According to recent data,
recycled concrete achieves higher compressive strength when the replacement ratio of
recycled fine aggregates is 30%. Therefore, we add another group, 50+S, which means the
replacement ratio of recycled coarse aggregates and recycled fine aggregates is 50% and 30%,
respectively. Moreover, the fine aggregates in other groups are all natural sands. Finally,
according to the replacement ratio of recycled coarse aggregates, we have 8 groups: 0, 30%,
40%, 50%, 50%+S, 60%, 70% and 100%. The results of compressive strength are shown in
Figure 10 and Figure 11.
Figure 10: Compressive strength before

Figure 11: Compressive strength after
gradation modification
gradation modification
From Figure 10 to Figure 11, we can see that on 7d and 14d before and after gradation
modification, the average value of compressive strength seldom change, for example, the
compressive strength of 7d and 14d is 23MPa and 28.2MPa, respectively. Before gradation
modification, the compressive strength of 28d is 34MPa; while after gradation modification,
the compressive strength of 28d is 35.3MPa, which increased 4%.
6.
CONCLUSIONS
Physical technology: with the help of mechanical forces to remove the adhering cement
pastes, recycled aggregates will have higher purity of gravels. But this technology
consumes much energy and cost and demands complicated equipment which tends to
wear serious, so physical technology seems not suitable for long-term performance.

However, the current recycled aggregates modification is mainly about physical

technology.
Chemical technology: although it has the advantage of convenient processes, when
considering the possible subsequent chemical reaction and the influence on the ultimate
concrete strength or durability, this technology is still in evaluation and discussion
without been applied to engineering practice.
Carbonation technology: it has simple operation and environment friendly advantages,
but considering the cost of industrial carbon dioxide collection and carbonation room or
condition controls, there are lots of subsequent research work to be done.
For the modification of recycled aggregates, we should not confine in physical or
mechanical properties or any other short-term researches. When taking account of
recycled concretes long-term serviceability, its indispensable for us to conduct deep
researches on strength stability and durability. Besides, the available equipment,
recycling utilization rates and costs of recycled aggregates demolition and production
should be evaluated.
ACKNOWLEDGEMENTS
The authors would like to express deep gratitude to the financial support of the National
Natural Science Foundation of China (51325802).
REFERENCES
[1] Shima H, Tateyashiki H, Matsuhashi R, et al. 'An advanced concrete recyc1ing technology and its
applicability assessment through input-output analysis', Journal of Advanced Concrete
Technology, 3(1) (2005) 53-67.
[2] R.V. Silva, J. de Brito, R.K. Dhir. 'Properties and composition of recycled aggregates from
construction and demolition waste suitable for concrete production', Construction and Building
Materials, 65(2014) 201-217.
[3] GU Rong-jun, YUAN Jiang, GENG Ou, GAO Fei-jian. 'Basic properties of recycled aggregate',
Concrete, 9(2009) 85-86.
[4] LI Qiu-yi, LI Yun-xia, ZHU Chong-ji, TIAN Li. 'Strengthening technique of recycled concrete
aggregate', Concrete, 1 (2006) 74-77.
[5] Zhang Jianbo, Wu Yongsheng. 'Study on the Strength of Recycled Aggregates', China Resources
Comprehensive Utilization, 28(11) (2010) 58-60.
[6] A. Akbarnezhada, K.C.G. Onga, M.H. Zhang. 'Microwave-assisted beneficiation of recycled
concrete aggregates', Construction and Building Materials, 25(2011) 3469-3479.
[7] Li J, Xiao H, Zhou Y. 'Inuence of coating recycled aggregate surface with pozzolanic powder on
properties of recycled aggregate concrete'. Constr Build Mater, 23(2009) 87-91.
[8] Shi-Cong Kou, Chi-Sun Poon. 'Properties of concrete prepared with PVA-impregnated recycled
concrete aggregates', Cement and Concrete Composites, 32 (2010) 649-654.
[9] Du Ting, Li Huiqiang. 'Mechanical properties of intensified recycled aggregate concrete', China
concrete and cement products, 2(2003) 19-20.
[10] Qu Zhizhong, 'New trends of reinforced concrete demolition and reclamation technique',
Architecture Technology, 32 (2) (2001) 102-104.
[11] Cheng Haili, Wang Yinfang. 'Experimental Study on the recycled concrete aggregates
strengthened by Sodium silicate solution', New Building Materials, 12 (2004) 12-14.
[12] Kou Shi-Cong, Zhan Bao-jian, Poon Chi-Sun. 'Use of a CO2 curing step to improve the properties
of concrete prepared with recycled aggregates', Cement & Concrete Composites, 45 (2014) 22-28.
[13] Xiao Jianzhuang, Lin Zhuangbin, Zhu Jun. 'Effects of Recycled Aggregates Gradation on
Compressive Strength of Concrete', Journal of Sichuan University (Engineering Science Edition),
4 (2014) 154-160.

OUTLINE OF AIJ RECOMMENDATION FOR MIX DESIGN,

PRODUCTION AND CONSTRUCTION PRACTICE OF CONCRETE
WITH RECYCLED CONCRETE AGGREGATE
Takafumi Noguchi
Department of Architecture, The University of Tokyo, Japan
Abstract
Architectural Institute of Japan (AIJ) established and published "Recommendation for Mix
Design, Production and Construction Practice of Concrete with Recycled Concrete Aggregate" in
October 2014 based on the Japan Industrial Standards for recycled aggregate and recycled aggregate
concrete. The recommendation shows the standard specifications for the investigation of
demolished reinforced concrete structures which become the resource of recycled aggregate and
the production and quality control of recycled aggregate. It also recommends the appropriate type
of recycled aggregate concrete according to the condition of the part or the component of a
building, and prescribes the mix proportioning, production technique, transportation method,
execution technique, quality control method for recycled aggregate concrete. In the appendices,
lists of recycled aggregates and recycled aggregate concretes certified by the Japanese government
are shown, and typical application cases of these recycled aggregate concretes to buildings are
outlined.
Keywords: recommendation, concrete, aggregate, production, construction, inspection
1.
INTRODUCTION
In view of the large environmental loads of reinforced concrete buildings throughout their
lifecycle, the Architectural Institute of Japan (AIJ) published the Recommendation for
Environment-friendly Construction of Reinforced Concrete Structures in 2008 with the aim of
working life extension, energy saving, resource conservation, resource circulation, etc., of
reinforced concrete buildings. However, these Recommendation do not include methods and
procedures of utilizing recycled aggregate (RA) for concrete. A subcommittee for studying
recycled aggregate concrete (RAC) was therefore organized in AIJ, and after four years of
activities, Recommendation for Design, Production, and Placement of Concrete Containing
Recycled Aggregate (hereafter referred to as the Recommendation for RAC) were established
and published in October 2014.
These Recommendation for RAC show desirable methods of investigating demolished
reinforced concrete buildings, which are the sources of the materials for RA, and methods of
production and quality control of RA. The Recommendation also propose the types of RAC
recommended for respective segments and members of buildings, describing desirable methods
of proportioning, production, transportation, placement, and quality control. Also, the appendix

includes lists of RAs and RACs that received JIS accreditation and Ministry certification, with
representative applications being illustrated. The contents of the appendix are given in Table 1.
Table 1: Contents of the Recommendation for RAC
Chapter 1
Chapter 2
Chapter 3
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Chapter 10
Chapter 11
Appendix 1
Appendix 2
Appendix 3
Appendix 4
Appendix 5
Appendix 6
2.
General
Types and applicable members of RAC
Qualities of RAC
Materials
Production of RA
Proportioning
Order placement, production, and acceptance of RAC
Transportation, deposition, consolidation, and curing of RAC
Quality control/inspection
RAC M for structural members subjected to the effect of drying
RAC L for reinforced concrete members
RA and RAC that received JIS accreditation and Ministry certification
Application examples of closed-circuit concrete recycling technology
Placement examples of concrete made using RA H produced by equipment with an eccentric rotor
Application examples of RAC H and M to buildings
Application examples of RAC by aggregate replacement
Methods of reutilizing by-product fine powder
BACKGROUND OF ESTABLISHING THE RECOMMENDATION
Japans total material input in fiscal 2010 was 1.61 billion tons, with about one-third of it,
or 543 million tons, being accumulated in the form of buildings and civil structures [1]. Since
the production of ready-mixed concrete (RMC) in that year came to 85 million m3 (196 million
tons) [2], about 12% of the total material input is found to be used for concrete production. In
contrast, the production of steel (crude steel) was about 110 million tons in the year, while
timber (including timber for ply wood and woodchips) was about 72 million m3 (equivalent to
36 million tons with a density of 0.5). In this light, concrete is found to be the construction
material that is used in the largest quantity in Japan.
In recent years, construction waste has accounted for approximately 20% of the total amount
of industrial waste and 11% of the total amount of waste including general waste. Above all,
concrete rubbles amounted to nearly 50% of construction waste in 2008. The concrete-related
industry is therefore regarded as the key industry for reducing industrial waste. In addition, a
vast number of buildings constructed in the high economic growth period of 1960s through
1970s are going to outlive their usefulness. It follows that the generation of construction waste,
mainly concrete rubbles, is unmistakably going to increase, making it essential to recycle
concrete to form a resource-recycling society.
In this context, the Ministry of Environment and the Ministry of Land, Infrastructure,
Transport and Tourism have implemented a variety of measures and policies given in Table 2
to reduce the amounts of waste generation and final disposal and increase the amounts of reuse
and recycling with the aim of forming a resource-recycling society. As a result, the amount of
final disposal has drastically decreased and the ratio of recycling of each type of waste was
substantially improved, with that of concrete rubbles exceeding 98%. However, most recycled

concrete rubbles are neither used as concrete aggregate nor a cement material but as road
bedding. Nevertheless, in view of the recent and expected decline in the demand for road
bedding materials due to the decline in road construction, concrete rubbles should be used not
only as a road bedding material but also as RA for concrete.
As to RA and RAC, active research has been conducted on technical development to produce
concrete aggregate from concrete rubbles and on the use of concrete made using RA since the
1980s. The quality standard for RA has thus shifted to higher levels and grew in diversity as
given in Table 3 in an effort to utilize the right RA in the right parts of buildings.
Table 2: Measures/policies for formation of resource-recycling society
Year
Measures/policies
1970
Waste Disposal and Public Cleansing Act
1991
Act for Promotion of Use of Recycled Resources
1992
Act on the Proportion of Effective Utilization of Resources
1994
Action Plan on Measures for Construction By-products, Recycling Plan 21
1997
Construction Recycling Promotion Plan 97
2000
Basic Act for the Promotion of the Recycling-Oriented Society
2000
Construction Material Recycling Act
2000
Act concerning the Promotion of Procurement of Eco-Friendly Goods and Services by the State and
Other Entities (Act on Promoting Green Purchasing)
2002
Construction Recycling Promotion Plan 2002
2002
Construction Recycling Guidelines
Table 3: Transition of quality standards for RA

Year
1977
Associated General Constructors of Japan

Usage Standard for RA and RAC (Draft)
Ministry of Construction
1994 Temporary Standard for Recycling of
Concrete By-products for Different Uses
1999
2000
2005
2006
2007
3.
Coarse aggregate
Density
WA (%)
(g/cm3)
Standard-producing organization
Name of standard
Fine aggregate
Density
WA (%)
(g/cm3)
2.2
2.0
13
Type 1
Type 2
10
Type 3
2.5
3.0
2.5
3.5
10
2.5
2.3
3.0
5.0
2.5
2.2
3.5
7.0
7.0
13.0
Building Center of Japan

Quality Standard on RA for Architectural Structures
Japan Concrete Institute
TR A 0006 (Concrete Containing RA)
JIS A 5021 (Class H)
Japan Concrete Institute
JIS A 5022 (Class M)
RA for Concrete
JIS A 5023 (Class L)
TYPES, QUALITIES, AND APPLICABLE MEMBERS OF RAC
Recycled aggregates class H (high quality), class M (medium quality), and class L (low
quality), as well as concrete made using these aggregates were standardized in 2005 2007 as

JIS A 5021 (Recycled aggregate for concrete-Class H), JIS A 5022 (Recycled concrete using
recycled aggregate Class M), and JIS A 5023 (Recycled concrete using recycled aggregate
Class L) according to the quality of RA [3]. The production of recycled coarse aggregate class
H (or M) may produce recycled fine aggregate class M (or L). From the aspect of the effective
use of resources, it is desirable to use both these aggregates in combination. When using more
than one type of RA in combination, the resulting RAC is to be categorized as one made using
aggregate of the lower quality.
Meanwhile, RA may be combined with normal aggregate to meet the performance
requirements for the segment/member to which RAC is applied. In such a case, four
combinations of recycled and normal aggregates are possible: (1) recycled fine aggregate and
recycled coarse aggregate; (2) recycled coarse aggregate and normal fine aggregate; (3)
recycled fine aggregate and normal coarse aggregate; and (4) RA blended with normal
aggregate. The Recommendation for RAC classify concrete made using these aggregates into
two types: Type I RAC made using normal aggregate for all of the fine aggregate and Type II
RAC made using RA for all or part of the fine aggregate.
Also, the quality of RA is strongly affected by the original concrete. It is therefore necessary
to change the method of controlling the quality of RAC depending on whether or not the original
concrete is identified. For this reason, the Recommendation for RAC classify RA into two
types: specific type (concrete made using RA solely from specific concrete rubbles) and
unspecific type (concrete made using RA made from unspecific concrete rubbles). The specific
type is advantageous in that not only the products are of stable quality but also the acceptance
inspection of RA becomes unnecessary due to consistent responsibility throughout the
processes of demolition, production of RA, and production of RAC. Also, the specific type
generally reduces the transfer of waste, not only saving resources but also leading to
conservation of energy and reduction in CO2 emissions. Moreover, it is easy to identify the
original concrete when a structure made of specific type RAC is to be demolished, benefitting
re-recycling. In the case of unspecific type, the quality of RA and RAC is assured by third
parties through JIS accreditation or Ministry authorization, but the frequency of inspection has
to be increased, as the manufacturers of RA and RAC are not deeply involved in demolition.
As stated above, RAC is classified as given in Table 4 according to the type of RA used,
combination of aggregate, and whether or not the original concrete is identified.
Table 4: Types of RAC
Type
Coarse aggregate
Fine aggregate
RAC H Type 1
All or part of coarse aggregate is recycled coarse aggregate H
(Specific or unspecific)
Normal fine aggregate (NFA)
All or part of coarse aggregate is recycled coarse aggregate. H

RAC H Type 2
(Specific or unspecific) Normal coarse aggregate (NCA)
All or part of fine aggregate

is recycled fine aggregate H
RAC M Type 1
All or part of coarse aggregate is recycled coarse aggregate M
NFA
All or part of coarse aggregate is recycled coarse aggregate M

RAC M Type 2
(Specific or unspecific) NCA or recycled coarse aggregate H

is recycled fine aggregate M
RAC L Type 1
All or part of coarse aggregate is recycled coarse aggregate L
NFA
All or part of coarse aggregate is recycled coarse aggregate L

RAC L Type 2
(Specific or unspecific) NCA, recycled coarse aggregate H or recycled coarse aggregate M is recycled fine aggregate L

The Recommendation for RAC stipulate the quality items and specified values to be satisfied
for each RAC type. For RAC class H, the Recommendation specify the slump, compressive
strength (design strength and durability design strength), Youngs modulus, drying shrinkage
ratio, air-dried unit weight, and durability (resistance to chloride attack, resistance to alkalisilica reaction (ASR), resistance to chemical erosion, and resistance to frost damage). For RAC
class M, the Recommendation specify the slump, compressive strength, Youngs modulus, airdried unit weight, and durability. For RAC class L, the compressive strength and slump are
specified.
Table 5 gives the specified values of design strength and durability design strength for each
RAC type. The same values of durability design strength as normal concrete are specified for
RAC classes H and M based on test results of carbonation resistance. However, RAC class M
has not reportedly been used for buildings of a long design service-life. Therefore, the
durability design strength of RAC class M for the long grade is required to be established after
confirmation by testing.
As to air-dried unit weight, 2.30 t/m3, 2.25 t/m3, and 2.20 t/m3 are specified as a standard for
RAC classes H, M, and L, respectively, based on the survey results shown in Fig. 1 [4].
RAC class H is the only class that can be used without particular consideration for members
under the effect of drying. However, when it is used for a structure of a design service-life of
the long grade, its drying shrinkage is required to be not more than 8 10-4, and the allowable
crack width of in-situ concrete is required to be 0.3 mm.
Table 5: Specified values of Compressive Strength of RAC
Design strength for durability (N/mm2)
Grade of design service-life
Short
Standard
Long
18
24
30
RAC type
Design strength (N/mm2)
18, 21, 24, 27, 30, 33, 36
18, 21, 24, 27, 30
18
24
18 (standard)
Figure 1: Air-dried unit weight of RAC [4]

The carbonation resistance and chloride ion permeability of RAC classes H and M are
reported to be equivalent to those of normal concrete, with little differences due to aggregate
quality. For this reason, it is judged that the cover depths of reinforced concrete buildings made
of RAC necessary in terms of durability may be similar to those specified for normal concrete.
The cover depths are therefore specified as given in Table 6. Note that, when applying RAC
class L to plain concrete members having crack-arresting re6inforcement, etc., the cover depth
should be established according to the above-mentioned policy.
Performance requirements for reinforced concrete buildings, as well as the type and degree
of deteriorative external forces on reinforced concrete buildings, vary from one segment to
another. It follows that quality requirements and control methods for RAC should vary from
one segment to another accordingly. For this reason, the Recommendation for RAC provide
different requirements depending on whether or not the concrete requires special consideration,
taking account of the compatibility between the qualities of the RAC to be used and the
members to which the concrete is to be applied. That is, Chapters 3-9 provide Recommendation
for using RAC where no special consideration is required, such as when using RAC class H;
when using RAC class M for unexposed areas unaffected by drying; and when using RAC class
L for plain concrete. On the other hand, Chapters 10 and 11 provide Recommendation for using
RAC where special consideration is required, such as when using RAC class M for structural
members subjected to the effect of drying and when using RAC class L for members having
steel reinforcement. Table 7 gives the RAC types related to applicable members in the
Recommendation for RAC.
Table 6: Cover depth of RAC
Type of member
Column, beam, bearing wall

Floor slab, roof slab
Non-structural
Members required to have durability
member
comparable to structural members
Column, beam, wall, and floor directly in contact with soil and
stem wall of continuous footing
Foundation
Structural member
Grade of design service-life

Short
Standard/long
Indoor/
In-door
Out-door
outdoor
30
30
40
20
20
30
20
20
30
40
60
Table 7: RAC types related to applicable members

RAC
type
H
M
L
Structural member
Non-structural member
With
Without reinforcements
reinforcements
Applicable range with no need for special consideration
RAC M used for structural
Applicable range with no need for special consideration
members affected by drying
Applicable range with no
RAC L used for reinforced concrete members
need for special consideration
Affected by drying
Not affected by drying

4.
PRODUCTION OF RECYCLED AGGREGATE
As shown in Figure 2, a standard production process of RA consists of the investigation of

the original concrete and original aggregate, acceptance/storage of concrete rubbles,
production/storage of RA, quality control/inspection of RA, and shipment/delivery of RA. The
Recommendation for RAC provide requirements for each of these steps.
The Recommendation for RAC require that original concrete suitable for the type of RA be
selected as a standard. Confirmation of the following items is required when investigating the
original aggregate and original concrete of a demolished building:
The original concrete involves no aggregate-related defects, such as ASR.
The original concrete does not contain a large amount of chlorides.
No deleterious amount of impurities gain access to the original concrete.
The original concrete is free from the possibility of including waste subject to special
control, such as asbestos.
The following items are required regarding the acceptance/storage of concrete rubbles:
Check the manifest or other documents to confirm the source of the original concrete and
the absence of waste subject to special control.
Confirm that concrete rubbles do not contain a large amount of impurities and are not
contaminated.
Confirm that the grading and moisture condition of concrete rubbles are suitable for
producing RA.
Store concrete rubbles with care to not allow access of foreign matter.
When producing RA by specifying the original concrete or original aggregate, store the
concrete rubbles separately to avoid blending with other concrete rubbles.
As to the production of RA, the following items are required:
Use equipment for RA production that is proven to produce RA that satisfies the quality
requirements.
Set the operation conditions of equipment for RA production so that the resulting RA
would satisfy the performance requirements.
Besides, the types and characteristics of major equipment for RA production are categorized
as given in Table 8. Figure 3 shows general methods for producing recycled road subbase
materials, low-quality recycled aggregate for concrete, and high-quality recycled aggregate for
concrete. Single toggle-type jaw crushers are generally used for the primary crushing of
demolished concrete into pieces 40 to 50mm in size regardless of the ultimate quality of
recycled aggregate. While the material is carried to the next process on a belt conveyor, foreign
particles such as wood/plastic chips and reinforcing steel/nails are removed manually and with
a magnetic separator, respectively. The materials then undergo various treatments according to
their uses. Impact crushers are used for secondary and tertiary crushing when producing lowor middle-quality recycled aggregate for concrete. While the quality of recycled aggregate
produced by using such equipment is improved as the number of treatment processes increases,
the recovery percentage of recycled aggregate decreases with increased amounts of powder
byproducts as the aggregate itself is crushed. Other equipment in practical use for producing
low- or middle-quality recycled aggregate for concrete includes self-propelled or vehiclemounted jaw crushers and impact crushers that save the energy normally expended to haul the
demolished concrete. Special equipment is therefore necessary for efficient production of highquality recycled aggregate for concrete in order to minimize the adhering cement paste.
Efficient equipments such as scrubbers and heating machine for producing high-quality
recycled aggregate for concrete were developed and put into practical use around 2000 in Japan.

Investigation of
original concrete
and aggregate
Acceptance/storage
of concrete rubbles
Production/storage
of RA
QC/inspection of
RA
Shipment/ deliver
of RA
Figure 2: Production process of RA

Table 8: RA production equipment
Use
Coarse
crushing
Type
Jaw crusher
(Figure 4a)
Hammer crusher
Impact crusher
(Figure 4b)
Cone crusher
(Figure 4c)
Impact breaker
(Figure 4d)
Rod (ball) mill
(Figure 4e)
Feature
Composed of a fixed jaw and a reciprocating swing jaw. Concrete rubbles of
a human head size are crushed between these jaws.
Concrete rubbles are crushed by the impact of hammers and strips of plates
Crushing
fitted to the rotor shaft, which rotates at high speed. Impact plates attached to
the external cylinder also crush concrete rubbles.
Composed of a gyrating mantle in the center and an enclosing concave.
Concrete rubbles are squeezed and ground in the tapered space in between.
Concrete rubbles are discharged from a high speed rotor by centrifugal force
against the outer wall, where the mortar phase with a lower strength is broken.
Grinding
Rods or balls are used as a medium to crush/grind concrete rubbles by the
rotation of the drum.
Concrete crushed beforehand into 50 mm rubbles is discharged into an
Scrubber
eccentric tubular vertical mill or lateral twin cones to be scrubbed by one
(Figures 4f & 4g)
another to remove cement paste adhering to aggregate surfaces.
Sieve
The sieve size is selected to achieve the target grading.
Materials are sorted by the air flow velocity. Fine powder can be separated
Separation
Air classifier
by combining an air classifier and a crusher or a sieve, such as an air flow
through a mill.
Heating machine,
Concrete rubbles are heated at 300C or less. The paste strength is reduced
Pre-treatment e.g., kiln
by dehydration of cement hydrates to facilitate grinding.
(Figure 4h)
Water containing RA is vertically pulsated to separate RA into layers of
PostFlotator, e.g., jig
different apparent densities. RA can therefore be sorted by the difference in
treatment
(Figure 4i)
the apparent density.
Demolished concrete rubbles

Jaw crusher
Impact crusher
Vibratory sieves
Vibratory sieves
Cone crusher
Road bottoming,
Backfilling
Vibratory sieves
L-class
L-class
recycled
recycled
coarse
fine
aggregate aggregate
Road bottoming
L-class recycled
aggregate
Heating tower
Aggregate scrubber
Vibratory sieves
M- or HM- or HPowder
class
class
recycled
recycled fine
coarse
aggregate
aggregate
M- or H-class recycled aggregate
Figure 3: General methods for concrete recycling

Concrete
Driving blade
Chain
Concretecurtain Impact plate
Impact plate
Rotor
Hitting plate
(b) Impact crusher
Fixed blade
(a) Jaw crusher

Concrete
Mantle
Concrete
Conecape
Outlet High-speed rotation

(d)
Impact
breaker
(c) Cone crusher
(e) Rod mill

Concrete rubbles
Eccentric
tubular mill
Scrubbing
External cylinder
Input
Cylinder hollow
Middle cone
Ejection cone
Motor
Recovery
Transmission gear
(f) Scrubber (Eccentric tubular mill)
Rotary blade Outlet

(g) Scrubber (Screw mill)
(h) Rotary kiln
(i) Flotator
Figure 4: RA production equipments

5.
MATERIALS, PROPORTIONING, PRODUCTION, AND PLACEMENT OF
RAC
The requirements for the materials of RAC (cement, aggregate, supplementary cementitious
materials, chemical admixtures, and mixing water), excepting RA, are in accordance with
Japanese Architectural Standard Specification for Reinforced Concrete Work (JASS 5).
Recycled aggregate classes H, M, and L are required to conform to JIS A 5021, Appendix A of
JIS A 5022, and Appendix A of JIS 5023, respectively, and also the following are particularly
required:
The maximum size of recycled coarse aggregate should satisfy the values respectively
specified for the original coarse aggregates and areas of use in Table 9.
Aggregate made by blending RA and normal aggregate should be handled as the same
type of aggregate as the RA before blending.
Aggregate made by blending different types of RAs should be handled as the RA of the
lower quality.
In order that RA be judged as innocuous regarding alkali-silica reactivity, the conditions
given in Table 10 should be satisfied.
Table 9: Maximum size of recycled coarse aggregate (mm)
Location
When the original coarse aggregate is gravel
When the original coarse aggregate is crushed

rock or BF slag
Column, beam,
slab, wall
2025
20
Foundation
202540
202540
Table 10: Conditions for RA to be judged innocuous regarding alkali-silica reactivity

Type
Recycled coarse aggregate H
Identification of:
All of the original coarse
aggregate
All of the original coarse aggregate or

recycled coarse aggregate H
All of the original coarse and

fine aggregate
All of the original coarse and fine aggregate

or recycled fine aggregate H
or recycled coarse aggregate M
or recycled fine aggregate M
or recycled coarse aggregate L
or recycled fine aggregate L
Recycled fine aggregate H

Recycled coarse aggregate M
Recycled fine aggregate M
Recycled coarse aggregate L
Recycled fine aggregate L
Judgment as being innocuous by testing of:
Recycled aggregate concrete may basically be proportioned in accordance with JASS 5. The
equations for calculating the design strength and proportioning strength, as well as the
correction values for in-situ strength, are therefore specified similarly to JASS 5. However, the
standard deviation is basically required to be 3.0 N/mm2, even if the design strength is 27
N/mm2 or less. These were established in consideration of the following: the tolerance for the
air content of RAC class M (standard products) and RAC class L (products made to
specifications) are as large as 2.0%; no requirement for air content is provided for RAC class L

(standard products) and RAC class L (chloride-regulated products); the use of a mixing truck
is permitted for RAC class M; and the use of both a mixing truck and continuous mixer is
permitted for RAC class L). The maximum water-cement ratio (W/C) of RAC and its standard
air content under freezing and thawing action are specified as given in Tables 11 and 12,
respectively.
Table 11: Maximum W/C of RAC
Type
RAC H
RAC M
RAC M (product resistant to frost damage)
RAC L
Maximum W/C (%)

60
60
50
65
Table 12: Standard air content of RAC

Condition
Not subjected to severe freezing and thawing action
Subjected to severe freezing and thawing action
RAC H
RAC M (product resistant to

frost damage)
4.5%
5.5%
5.5%
Selection of a plant for producing RAC, and the order placement, production/transportation,
and acceptance of RAC are basically required to be in accordance with JASS 5. The
Recommendation for RAC stipulate the requirements for a production plant for RAC as
follows:
The stockyard for accepting RA should be furnished with partitions to avoid blending
with other aggregate and cleaned and well-maintained.
The plant should have several types of aggregate silos and storage bins with more than
sufficient equipment.
The plant should have pre-wetting equipment to stabilize the surface moisture of RA
before use.
Also, RAC is required to be produced using a fixed mixer and transported by an agitating
truck as a standard, but production using a mixing truck is permitted for RAC class M, and
production using a mixing truck and a continuous fixed mixer is permitted for RAC class L.
When producing RAC class L using a mixing truck and continuous fixed mixer, however, the
contractor is required to establish necessary matters related to the storage, batching, mixing,
and transportation of concrete materials and obtain the approval of the supervisor before
execution.
Conveyance within the job site, placement, consolidation, and curing of RAC may also be
carried out basically in accordance with JASS 5. However, when using a supplementary
cementitious material to suppress ASR for RAC classes M and L, it is required that an
appropriate curing method and period be established based on tests or reliable data.
6.
QUALITY CONTROL AND INSPECTION
The items, methods, and frequency of testing/inspection for quality control (QC), points of
taking samples and specimens, objects of testing/inspection, and judgment criteria for test
results are required to be in accordance with JASS 5. Additionally, when using RAC classes H,

M, and L, JIS A 5021, JIS A 5022, and JIS A 5023, respectively, are also required to be
observed. The items of QC/inspection are specified for each phase of using RAC as given in
Table 13.
As to the QC organization, if the QC of RA is carried out at a RMC plant and the contractor
can conduct the QC of RAC by giving instructions to and consulting with the RMC plant, such
as when the RMC plant is JIS-accredited regarding RAC, then QC/inspection can be carried
out by an organization similar to one for normal QC. On the other hand, when it is necessary to
identify the original aggregate and original concrete or when tolerances are specified for such
qualities as oven-dry density and fineness modulus, the contractor should directly give
instructions to and consult with the demolisher and RA manufacturer. In such a case, the quality
control organization should include the demolisher and RA manufacturer.
Table 13: Items and times of QC/inspection at each stage of RAC
Demolition of
structure
Concrete placing
After placing
RMC plant
During placing
Acceptance
Shipment
During
production
Trial mixing
RA manufacturer
Production of concrete
Acceptance of
RA
Shipment
Treatment
Acceptance of
conc. rubbles
Discharge
Demolition
contractor or RA
manufacturer
Original concrete
RA
Qualities to be confirmed
Location, use, and condition
of structure
Waste subject to special
control
Adherent finishing material
Type, nominal strength, and

air content of original
concrete
Type and source of original
aggregate
Impurity content
Oven-dry density
During
demolition
Responsible party for quality
Before
demolition
Stage
Production of RA
Contractor
Concrete
Water absorption
Grading/FM
Chloride content
Alkali-silica reactivity
Resistance freezing and

thawing, resistance to frost
damage
Slump
Air content
Drying shrinkage
In-situ concrete strength
Finish
Cover depth
: Exclude factors adversely affecting the quality by visual observation
: Routine control (visual observation or testing)
: Exclude factors adversely affecting the quality by establishing a standard
: Confirm by testing
: Confirm by testing as required

: Confirm by records and documents
7.
RAC REQUIRING SPECIAL CONSIDERATION
7.1
RAC class M (products with special consideration)
The application range of RAC class M, which is prone to greater drying shrinkage than
general concrete, is basically limited to members unaffected by drying, and the
Recommendation for RAC provide requirements within these limits in Chapters 3-9. From the
aspect of resource recycling, however, expanding the limits of application for RAC is of
significance. The quality of concrete can be improved by limiting the materials and reducing
the unit water content and W/C. For this reason, Chapter 10 is provided for RAC class M
(products with special consideration), which can be used for structural members under the effect
of drying by tailoring the materials and mixture proportions. This chapter contains the methods
of proportioning, production, placing, and QC. There have been few examples of applying RAC
class M to structural members subjected to drying, with information relating to durability on
the level of full-scale structures being insufficient. The durability design strength of RAC class
M (products with special consideration) is thus required to be increased by 3 N/mm2 from the
values given in Table 5. In other words, 21 and 27 N/mm2 are required as a standard for short
and standard design service-life grades, respectively. It is also required that the drying
shrinkage ratio of RAC class M (products with special consideration) be examined beforehand
by reliable data or testing to confirm there is no problem in using it for structural members
subjected to the effect of drying. Other requirements include the following:
Recycled aggregate class M to be used should be limited to coarse aggregate, and no
recycled fine aggregate class M should be used.
Recycled coarse aggregate class M should be one that is produced from identified original
concrete.
RAC class M (products with special consideration) should be produced at a RMC plant.
7.2
RAC class L (products with special consideration)
Due to its durability being lower than general concrete, the scope of application of RAC class
L is basically limited to plain concrete. Similarly to Chapter 10, however, the Recommendation
for RAC specify RAC class L (products with special consideration) from the aspect of resource
circulation. This is a RAC that can be used for reinforced concrete members by adjusting
proportioning, such as blending with normal aggregate. Chapter 11 describes the methods of
proportioning, producing, placing, and conducting QC of RAC class L (products with special
consideration) as follows:
The design strength should be 18, 21, 24, 27, and 30 N/mm2.
The design service-life grade of the object building should be short or standard, and
the durability design strength of RAC class L (products with special consideration) should
be 18 and 24 N/mm2, respectively.
The drying shrinkage ratio should have proven no problem before applying to members
subjected to the effect of drying.
The resistance to frost damage should be confirmed by testing.
The upper limit of the impurity content of the RAC class L should be the same as those
of RAC class H and RAC class M.
RAC class L that was produced from an identified concrete satisfying the following
requirements should be used:
(a) Compressive strength of not less than 18 N/mm2.

8.
(b) Total chloride ion content of not more than 0.30 kg/m3.
(c) No ASR.
RAC class L should be used by blending with normal aggregate, and the upper limits of
the contents (replacement ratios) of class L recycled coarse and fine aggregates should be
50% and 30%, respectively, when used singly, and 30% and 15%, respectively, when
used in combination with each other.
No continuous fixed mixer should be used for producing RAC class L (products with
special consideration).
CONCLUSIONS
JIS for RA and concrete made using RA were established in 2005-2007, and RA class H was
adopted as aggregate for normal concrete in the 2009 revision to JIS A 5308 (Ready-mixed
concrete). These have cleared away the institutional obstacles for applying RAC to buildings
and civil structures, raising expectations for a dramatic increase in the use of RAC. In reality,
however, the use of RAC as aggregate for concrete remained as small as 112,000 tons in 2013
[5], accounting for only 0.1% of the total amount of aggregate for RMC. There is no sign of
RA or RAC being in circulation on the market. One reason for this is Article 37 of the Building
Standard Act, which does not allow the use of RAC for major structural segments of buildings.
Effective utilization of fine powder as a byproduct of RA production is also a key to the spread
of RAC.
It is strongly expected that the establishment of the present Recommendation for RAC will
promote understanding of the technical aspect (quality and control) of RAC. If it also
accelerates preparation for logistic systems, then the use of RAC for buildings will truly begin,
expediting the construction of a resource-recycling society.
ACKNOWLEDGEMENTS
The author acknowledges the great contribution of AIJ committee members, Prof. Koyama,
Mr. Asami, Prof. Dosho, Dr. Kawanishi, Dr. Kitagaki, Dr. Kuroda, Prof. Morohashi, Mr.
Shirokuni, Dr. Tamura, Dr. Tanano, Mr. Tateyashiki, Mr. Tochigi and Dr. Yanagihashi to
establishment of the Recommendation for Mix Design, Production and Construction Practice
of Concrete with Recycled Concrete Aggregate.
REFERENCES
[1] Ministry of the Environment: Annual Report on the Environment, the Sound Material-Cycle
Society, and the Biodiversity in Japan 2014 (2014) 10.
[2] Sakai, K. and Noguchi, T., 'The Sustainable Use of Concrete' (CRC Press, 2012).
[3] Noguchi, T., Koyama, A. and Suzuki, Y., 'Japan Industrial Standards on Recycled Aggregate and
Recycled Aggregate Concrete', Concrete Journal 45 (7) (2007) 5-12.
[4] Shin, S., Park, W., Oh, D. and Noguchi, T., 'Relationship between Compressive Strength and
Various Mechanical Properties of Recycled Aggregate Concrete (Part 1: Relationship with Watercement Ratio and Density)', in 'Materials and Construction', Summaries of Technical Papers of
Annual Meeting, Architectural Institute of Japan, 2011, 229-230 (in Japanese).

CHANGES OF THE MATERIAL PROPERTIES BY SELECTIVE

CARBONATION OF RECYCLED AGGREGATES FROM CONCRETE
Dipl.-Ing. Marko Seidemann (1), Prof. Dr.-Ing. habil. Anette Mller (2), Prof. Dr.-Ing.
Horst-Michael Ludwig (1)
(1) Bauhaus-University Weimar (Germany), Department of Civil Engineering, F.A.
Finger-Institute for Building Materials Science, Germany
(2) Bauhaus-University Weimar (Germany), Department of Civil Engineering, Germany
Abstract
The aim of this work is to bind a maximum of carbon dioxide in concrete recyclates
permanently by means of a selective carbonation. This process is meant to change the
properties of the recyclate in such a way that it comes to a decrease in porosity and to an
increase in strength. The carbonation process is a well-known procedure. As diffusion-steered
process it runs, however, very slowly and takes years. The technical challenge of this work is
to accelerate the carbonation reaction in a reactor and thus making it technically usable. A test
device was developed for the directed carbonation. It consists of a tubular reactor and several
measuring units. The test specimens are continuously subjected to a preset gas mixture in the
tubular reactor. The gas flows evenly through the reactor. The first tests were carried out on
cement stone at varying CO2-concentrations over different time intervals. The results show a
correlation of the samples weight increase and their density after staying in the reactor. The
CO2 absorption was confirmed by means of thermal analysis after the carbonation. In further
tests a carbonation of concrete recyclates was carried out. The treated and the untreated
concrete recyclates as well as the natural aggregate were taken to produce concretes. Then
they were characterized as to their fresh concrete properties as well as their hardened concrete
properties and both compared and evaluated to one another.
Keywords: selective carbonation, recycled concrete, carbon dioxide
1.
INTRODUCTION
Concrete is by far the most deployed building material in the industry. This enormous usage
of concrete consequently leads to an increase of crushed concrete in the total waste balance.
Since sustainability becomes more and more important in our society and since we are facing
a shortage of all resources, the high-quality recycling of demolition waste is gaining
increasing significance. For environmental and economic reasons such recycling is strongly
required.
There already exist recycling processes for old concrete which aim at separating the used
concrete into cement stone and the original aggregate. However, by means of the recycling
techniques available on the market, a complete separation is not possible. The resulting

recycled aggregate consists of primary aggregate and adherent cement stone. As a result it
comes to an increase of the capillary pore share for the secondary aggregate in comparison to
the primary aggregate. That is why the deployment of concrete recyclates as aggregate results
in a deterioration of the fresh and hardened concrete properties.
By means of a selective carbonation of the concrete recyclates it should be possible to achieve
a decrease in porosity while the raw density as well as the strength increase. Furthermore, a
maximum of CO2 should permanently be bound in the concrete recyclates. So, a property
improvement of recycled concrete can be combined with the positive effect of a permanent
bonding of CO2. At the moment is the selective carbonation of the concrete recyclates in
practice already in force.
The course of carbonation within the old concrete is a process in which CO2 intrudes into the
concrete's capillary pore system by means of diffusion. CO2 and calcium hydroxide Ca(OH)2
dissolve in the pore water and react to calcium carbonate CaCO3 which settles in the
concrete's capillary pores. Due to the formation of CaCO3 it comes to a decrease of calcium
ions in the pore water. As a result, further Ca(OH)2 can dissolve and the reaction can go on
until all the Ca(OH)2 is used up. The CaCO3-crystals in the pores continue their growth which
leads to a structure densification. The carbonation of the calcium hydroxide in the Portland
cement stone leads to a total volume increase of approx. 11 Vol.-% [1]. The total porosity of
the concrete decreases and it comes to a strength increase of the concrete recyclates. Both
leads to a quality improvement of the old concrete. After a successful carbonation the old
concrete can be used as aggregate for new concrete.
2.
EXPERIMENTAL SET-UP
A test device was developed for the directed carbonation [Fig. 1]. It consists of a tubular
reactor and several measuring units. The test specimens are continuously subjected to a preset gas mixture in the tubular reactor. The gas flows evenly through the reactor. The CO2
concentration of the reaction gas can be varied between 0 Vol. - % CO2 and 100 Vol. - %
CO2. The gas concentration is taken by means of the CO2 measuring unit located in the
inlet. A salt solution in the tubular reactor ensures a continuous relative humidity of 70%. The
monitoring of the air humidity and the temperature inside the reactor is done by means of a
measuring unit.
Fig. 1: Schematic illustration of the experimental set-up

3.
TEST EXECUTION AND RESULTS
3.1
Selective carbonation of cement stone pastes
The first tests were carried out on cement stone blocks. The cement stone blocks had a size of
4 x 4 x 2 cm. The experiments were carried out over varying time intervals at a constant CO2
concentration. The results shed light on the carbonation depth and the relative weight increase
as a function of the w/c ratio of the cement paste prisms. The relative weight increase
indicates the carbonation level. As rule of thumb it can be said that at complete carbonation it
comes to a weight increase of 30 % as far as cement pastes are concerned.
In further test series investigations were carried out on samples having a grain size of 2/4 mm
and a w/c ratio of 0.5. The experiments were carried out at 20, 40, 60, 80 and 100 vol. -% CO2
at varying weathering times.
The figures 2 to 3 show a part of the test series results with a CO2 concentration of 20 vol. -%.
Figure 2 shows the correlation between the weight increase and density of the samples after
their staying in the reactor. A steady increase in density can be observed in dependence of the
weight increase. This suggests that a CO2 absorption with CaCO3 formation must have taken
place. After the weathering a decrease of the calcium hydroxide content and an increase of the
calcium carbonate content could be determined by means of thermogravimetry (Fig. 3). For
example, the calcium carbonate content in the initial sample was only at 7 %. After the
treatment in the carbonation reactor the calcium carbonate content doubled.
25
2,65
2,60
density
2,55
2,50
2,45
15
2,40
2,35
10
2,30
2,25
2,20
2,15
2,10
0
12
carbonation duration [h]
Fig. 2: Increase in weight and density after carbonation
24
48
density [g/cm]
weight increase [M.%]
20
weight increase

Fig. 3: DTA/TG before and after carbonation

The pore volume and the pore size distribution for the grain size 2/4 were determined by
means of mercury intrusion porosimetry. Table 1 and Fig. 4 show the gained results.
Table 1: Porosity before and after carbonation
sample
untreated
treated
porosity [%]
14.6
10.2
Fig. 4: Distribution of pore radii before and after carbonation

The results gained by means of mercury intrusion porosimetry show a decrease in the total
pore volume. The porosity of the cement stone matrix is reduced due to subjecting it to CO2
from 14.6 % to 10.2 %. The structure of the cement stone becomes denser by the carbonation.
Fig. 4 shows that the decrease in porosity occurred for the most part in the pore radii range of
8 nm to 1 m. In this range capillary pores exist. That means that, due to the treatment, a
reduction of the capillary porosity occurred. This fact has a pronounced effect on the
compressive strength of the cement stone since the capillary porosity is an essential parameter
for the compressive strength [2].
3.2
Concrete Tests
In order to get a better judgement on the effect of the carbonated cement stone matrix on the
respective fresh or hardened concrete investigations, several concretes were produced. The
concrete composition, i.e. aggregate, cement and water was consciously chosen to be simple
in order to reduce additional influences to a minimum. 5 concretes were produced that differ
in the composition of their aggregates. Table 2 lists the concretes with the different
aggregates. All concretes were produced using a Portland cement CEM I 42.5 N and a w/cratio of 0.5. The deployed aggregate was put together according to the particle size
distribution curve A/B16 [3]. Only the grain groups 2/8 and 8/16 were replaced by RCmaterial (45 Vol.-% respectively 100 Vol.-%) according to the DAfStb-guideline [4]. The
secondary and the carbonated secondary aggregate were inserted in the grain range 2 to 16
mm of the particle size distribution curve A/B16 in such a way that it came to an even
distribution following the primary aggregate.
Table 2: Designation and meaning of the concretes
Designation
RFB
RB/45
RB/100
RBC/45
RBC/100
meaning
Reference concrete with natural aggregate
Recycled concrete with 45 Vol.-% recycled aggregates from concrete
2 mm
Recycled concrete with 100 Vol.-% recycled aggregates from concrete
2 mm
Recycled concrete with 45 Vol.-% carbonated recycled aggregates from
concrete 2 mm
Recycled concrete with 100 Vol.-% carbonated recycled aggregates from
concrete 2 mm
3.2.1 Fresh Concrete Investigations

By determining the slump directly after the concrete production and then again after 30, 60
und 90 min the consistency course of the produced fresh concretes recorded. The slump
directly after production was similar for all 5 concretes. With time progressing the setting of
the concretes begins. Fig. 5 it can be observed that the concrete with the carbonated recyclate
shows a similar behaviour as the reference concrete as far as setting is concerned.
For all mixtures it can be said that there were no big deviations since the w/c-ratio, the cement
content and the core humidity of the deployed aggregates were set equal for all concretes.

Fig. 5: consistency behavior

3.2.2 Hardened Concrete Investigations
The evaluation of the strength development for the 5 concretes was carried out by taking their
compressive strengths after 7, 14, 28 and 56 days. The results are given in Figs. 6 9.
Fig. 6: compressive strengths after 7 d
Fig 7: compressive strengths after 14 d
The Figs. 6 to 9 illustrate that with the hydration time progressing an increase in compressive
strength can be observed for all concretes. When one compares the concretes with secondary
and carbonated secondary aggregate, one can see over the different hydration levels that it

comes to a slight but nevertheless significant increase in strength as far as concretes with
carbonated secondary aggregate are concerned. Only in case of early strength after 7 days no
strength increase can be observed yet when carbonated material with a recyclate share of 45
Vol.-% was used. The higher the content of recycled aggregate in the concrete, the more
pronounced is the strength increasing effect of carbonation.
3.
CONCLUSION
The test results show that a structure densification in the Portland cement stone by means of a
directed carbonation in a lab reactor is possible. It comes to a bonding of CO2 and an
improvement of the material properties. By means of a directed carbonation of the cement
stone matrix it comes to an increase in saturated density and a decrease in porosity.
Measurements show that the CO2-absorption by the cement stone can be significantly
accelerated under the chosen test conditions. The time scale for the carbonation of buildings
which is usually measured in years is shifted to the hour scale. The set CO2-concentration of
20 Vol.-% used in the tests is comparable to real concentrations in power plant exhaust fumes.
In the next step concretes were produced under usage of concrete recyclates that had
undergone the CO2-treatment, then these concretes were investigated. In these investigations
concretes with treated recyclates were compared and evaluated with concretes that had been
produced with untreated recyclates or natural aggregates. After the compressive strength tests,
tests on the concrete durability were carried out.
Looking at the determined fresh concrete parameters no significant differences could be found
regarding the usage of carbonated secondary aggregates. Due to the deployment of carbonated
recyclates it comes to an increase in the compressive strength in comparison to concretes with
non-carbonated recycled aggregates. In case of a carbonated recyclate share of 45 Vol.-% the
strength increase after 56 days is at 1.8 % compared to concretes with untreated recyclates.
When one increases the recyclate share to 100 Vol.-% it comes to a strength increase of 3.3 %
after 56 days.
Summarising it could be shown that a directed carbonation of adhering cement stone at the
secondary aggregates there is a denser material structure and thereby the hardened concrete
properties can be improved. This result puts the demolition material concrete once again into
the focus of the recycling and resource management industries and can contribute to close the
cycle of materials.
ACKNOWLEDGEMENT
The authors thank the German Federal Foundation for Environment (DBU) for the support in
the context of this research project.
LITERATURE
[1] Stark, J. and Wicht, B., 'Dauerhaftigkeit von Beton' (Springer-Verlag Berlin Heidelberg,
2013)
[2] Locher, F. W. And Wicht, B., 'Zement Grundlagen der Herstellung und Verwendung'
(Verlag Bau + Technik GmbH, 2000) 293-297

[3] DIN 1045-2, ' Tragwerke aus Beton, Stahlbeton und Spannbeton - Teil 2: Beton Festlegung, Eigenschaften, Herstellung und Konformitt - Anwendungsregeln zu DIN EN
206-1' (Beuth Verlag, Ausgabe August 2008)
[4] DAfStb-Richtlinie, ' Beton nach DIN EN 206-1 und DIN 1045-2 mit rezyklierten
Gesteinskrnungen nach DIN EN 12620 - Teil 1: Anforderungen an den Beton fr die
Bemessung nach DIN EN 1992-1-1; Deutscher Ausschuss fr Stahlbeton' (Beuth Verlag,
Ausgabe September 2010)

THE ECONOMIC RISK ANALYSIS TECHNIQUE APPLIED TO

DIFFERENT SCENARIOS OF CONSTRUCTION AND DEMOLITION
WASTE RECYCLING PLANTS.
Renato Petter (1); Carlos Hoffmann Sampaio (1); Regis Paranhos (2); Raul Oliveira Neto
(2); Carlos Petter (1)
(1) LAPROM - CT Universidade Federal do Rio Grande do Sul, Brazil
(2) Universidade Federal do Pampa, Brazil
Abstract
The economic risk analysis technique through Monte Carlo simulation is a fundamental tool
in economic studies of investment alternatives. The decision maker will have as a parameter
the probability of success (and consequently the risk of failure) on the recovery of invested
capital. The technique proposed in this paper will work against the most commonly used
economic analysis that is deterministic, in which the input variables that feed the discounted
cash flow are fix numbers. No room for uncertainty in the result, the numbers obtained in
response, such as rate of return (IRR) and net present value (NPV) are also fix numbers used
in decisions to invest or not, to move forward in a study preliminary or abandon the study,
give preference to equipment A over B, and so on. Regarding the issue of reuse of
construction and demolition waste (CDW), the risk analysis is very timely, because besides
the uncertainty due to imperfect knowledge of the limits of contamination of recyclable raw
material, the parameters involved in preparing the cash flow are, at most, variables. In this
sense, a number of factors such as the variability of the raw material source of the recycled
product, transport distances variations, due to arrangement of raw material collection and
discharges points, or the quality obtained according to the proposed separation technique
leads to multitude possibilities of variation. These contour conditions enhance the usefulness
of probabilistic economic analysis. The decision maker will have not a single value result, but
a probability distribution of values and, consequently, the likelihood of success or failure. The
manager will have also a clear indication of which variables deserves more care and have
more impact on the success (or failure) of the economic fortune.
Key words: risk analysis, economic evaluation, Monte Carlo.
1.
INTRODUCTION
The main objective of the paper through the application of economic risk analysis tools
discounted cash flow followed by Monte Carlo simulation is to propose an economic analysis
process to take decisions on C&D waste recycling. The focus is the comparison of conventional
mechanized plant against more advanced technologically options.
Technical and economic feasibility of a construction and demolition waste treatment plant is
not different from the analysis developed for other business such as mining, industry, commerce

etc. To visualize the business we have to know what would be the revenue sources, and if we
have available market.
Potential revenue sources for C&D waste treatment plant
a) Charge the waste recycling and final deposition (levy).
b) Sell mix of products (aggregates, sand, compost, brick fines, metal scrap, recycled
wood, plastic etc.).
1.1. Value chain
To make the value chain work (Porter, 1990), the local legislation should require partial or
total C & D waste recycling, in this way prohibiting direct landfills.
Margin is the difference of revenue of all stakeholders (builders, government, services
suppliers, and recycling plant owners) less than all costs involved - inspectors, project,
management, demolition, loading and dumping & transport, recycling, waste disposal. Figure
below summarizes this value chain.
Figure 1 - Value Chain recycling C&D waste.
Governamental Admnistration
Demolition &
Loading/Transport
Overhead
Recycling Project Management
Load/
Demolition Transport
&
Dump
Sell
Recycling
Products
Operations Management
Recycling
Margin
Final Waste
Disposal
Costs
Revenue
1.2. Mix of materials

Very important is to define approximated composition of C&D waste, which depends on
building culture, geographic location, weather, seismic activity zone, technological level,
countries social and economic conditions. The table below shows an estimated Brazilian debris
composition (Santos, 2009).
Table 1 - Brazilian debris composition
Components
Concrete
Brick
Sand
Soil/dust/mud
Rock
Asphalt
Metals
Wood
Paper/organic
material
Others
Road
Excavations
Demolition
Others
Cleaning
*Average
48%
6,10%
0,30%
9,60%
48,90%
32,50%
54,30%
6,30%
1,40%
11,90%
11,40%
1,60%
3,40%
1,60%
17,50%
12%
3,30%
16,10%
23,10%
1%
6,10%
2,70%
18,40%
5%
1,70%
30,50%
23,90%
0,10%
4,40%
3,50%
31%
5%
4%
26%
21%
6%
3%
2%
1,60%
0,90%
2,70%
0,90%
3,50%
2,00%
2%
1%
4,60%
16,80%
7,00%
23,60%
0,10%
0,50%
1,10%
1,00%

2.
METHODOLOGY
Sequence to improve evaluation: First step is to select a list of options. Second step is to
develop time series forecasts for production and prices. Third step is to prepare economic
models containing, production schedule, cash costs, capital expenditure, sustaining capital, tax
calculations, for each option. Fourth step is to run Monte Carlo simulation using model outputs
and practical interval of variability. Last step is to analyze Montecarlo simulation results to
understand probabilities of success or failure for the different options.
Case Study Options:
Conventional Plant (crushing, air separation, manual sorting, screening final products).
Plant Througput 100 kt /300 kt/ 600 kt per year waste treatment;
Air Dry Jig Plant (crushing, air separation, manual sorting, Jigging, screening final
products). Plant Througput 100 kt /300 kt/ 600 kt per year waste treatment;
Air Dry Jig Plant combined with Automatic sorting (crushing, air separation, manual
sorting, Jigging and Sensor based automatic, screening final products). Plant Througput
100 kt /300 kt/ 600 kt per year waste treatment.
For capital expenditure (CAPEX) calculation, Mular factorated capital estimates was used
(Mular, 1990):
Equation 1-Mular Factor
Total Fixed Capital cost
==> min==>Factor multiplier = (Capex_equipment+0,93*Capex Equipment)

==> max==>Factor multiplier = (Capex_equipment+3,04*Capex Equipment)
2.1. Conventional Plant option

The suggested standard flowsheet includes the necessary processes to separate products from
the mix of materials (concrete, metal, wood, plastic, brick and paper). For different plant sizes,
it is necessary a mass balance, equipment dimensioning, capital and operating cost calculations.
Figure below shows conventional plant:
Figure 2-Flowsheet Conventional Plant

2.1.1. Conventional Plant - Capital and Operating Costs

Flowsheet and mass balance allow defining equipment size and, Infomine database utilized
to estimate equipment price and operating costs a summarized table shows upper, average and
interval for capex value and average opex.
Table 2- Coventional Plant Capex & Opex
100 ktpy
Upper
Lower
Capex
kUS$
10436
5923
Variable Opex (US$/t)

18,06
Fix Opex (kUS$)

1301
Total Opex
(US$/t)
31,06
300 ktpy
Upper
Lower
Capex
kUS$
14145
8550
Upper
Lower
Capex
kUS$
19792
12702

8,51
Fix Opex (kUS$)

1680
Total Opex
(US$/t)
14,11
600 ktpy
5,89
Fix Opex (kUS$)

2021
Total Opex
(US$/t)
9,26
2.2. Dry Air Jig option

Front of treatment plant is similar to the conventional plant option, after trommel is added a
primary and secondary Jigs to reduce inpurities in the final products, allowing compete with
natural gravel.
Figure 3-Flowsheet for Air Dry Jig Plant:

2.2.1. Dry Air Jig option Capex and Opex

Dry Air Jigging capital and operating costs obtained from suppliers combined with Infomine
database used to estimate capex and opex.
Table 3 - Dry Air Jig option Capex and OpexPlant Capex & Opex
100 ktpy
Upper
Lower
Capex
kUS$
22556
11713

19,99
Fix Opex (kUS$)

1287
Total Opex (US$/t)

32,86
300 ktpy
Upper
Lower
Upper
Lower
Capex
kUS$
26265
14340
Capex
kUS$
37972
21387

9,41
Variable Opex
(US$/t)
6,52
Fix Opex (kUS$)

1659
600 ktpy
Fix Opex
(kUS$)
1987
Total Opex (US$/t)

14,94
Total Opex (US$/t)

9,83
2.3. Combined Plant option - Air Dry Jigging and Automatic Sorting
As dry Air jigs, have low efficiency to treat materials with similar densities, a combination
of Dry Air Jig, and sensor based automatic sorting is being proposed to improve quality of final
products. The optical sorting will treat the lighter pre-concentrate from dry air jig, in order to
separate gypsum, wood chips from brick rubble, concrete middlings (cement rock) using NIR,
color or X-Ray sensors.
Figure 4- Flowsheet Air Dry Jigging and Automatic Sorting

2.3.1. Combined Plant - Dry Air Jigging and Automatic sorting Capex and Opex
Sensor based sorting capital and operational costs considers data mentioned in I.Vegas at al
(Iigo Vegas, 2015). /Upgrading the quality of mixed recycled aggregates from construction
and demolition waste by using near-infrared sorting technology.
Table 4 - Dry Air Jigging and Automatic sorting Capex and Opex
100 ktpy
Upper
Lower
Capex
kUS$
18033
9552
Upper
Lower
Capex
kUS$
21467
12048
Upper
Lower
Capex
kUS$
28377
16803

21,25
Fix Opex (kUS$)

1287
Total Opex
(US$/t)
34,12
300 ktpy
9,54
Fix Opex (kUS$)

1659
Total Opex
(US$/t)
15,07
600 ktpy
2.

6,62
Fix Opex (kUS$)

1987
Total Opex
(US$/t)
9,93
DETERMINISTIC AND SIMULATION ANALYSIS

2.1. Deterministic analysis (Discounted Cashflow)
Discounted cashflow considering 10% discount rate, twenty years life of plant and following
product prices and recoveries for the different plant options.
Table 5 - Recycling Plants Prices and recoveries
Components/Products
Screened Soil
Aggregate
Sand Recycled
Crushed Brick Chips
Metal Scrap(Rebars)
Revenue Recycling
Service
Comp.
26%
37%
13%
5%
3%
Conventional
Plant
Price
Rec.
(US$/t)
%
3,3
90%
5,33
50%
6
90%
5,33
90%
133
70%
30
Air Jigging
Price
Rec.
(US$/t)
%
3,3
90%
8,40
95%
6
90%
5,33
90%
133
70%
30
Combined Air
Jigging + Sorter
Price
Rec.
(US$/t)
%
3,3
90%
7,87
85%
6
90%
12
90%
133
70%
30

Table 6 - Summary of Discounted cashflow analysis

NPV 10% [US$ 000]
IRR %
Plant
100 kt 300 kt 600 kt
Plant
100 kt
26431 74997
-1%
Conventional -4712
Conventional
-12559 24632 69408
-7%
Air Jig
Air Jig
Sorter & Air
Sorter & Air
-11645 21566 69913
-13%
Jigg
Jigg
300 kt
42%
27%
600 kt
71%
42%
28%
52%
2.2. Monte Carlo simulation

The economic risk analysis technique through Monte Carlo simulation was provided to
evaluate the three investment alternatives. Instead the deterministic above approach, in which
the input variables that feed the discounted cash flow are fix numbers and the numbers obtained
in response, such as rate of return (IRR) and net present value (NPV) are also fix numbers a
probability distribution of input values was provided (table 7).
Table 7 - Simulations Input Values
Simulated Variables
Plant Throughput
Prices
Costs
Capex calculated according to Mular Mular (1990)
Price to Recycle (Levy)
Debris composition
Screened Soil
Aggregate
Sand Recycled
Crushed Brick Chips
Metal Scrap
Debris composition normalized to avoid sum > 100%
Interval of Variation
Distribution
min
max
triangular
70%
110%
triangular
70%
130%
triangular
85%
115%
uniform
Min
Max.
triangular
20
40
uniform
uniform
uniform
uniform
uniform
11,90% 48,90%
9,9%
61,6%
3,2%
25,9%
0,3%
12,0%
0,5%
6,1%
Table 8 - Summary table of Simulation Results

Interval - 90% Confidence
Plant
Conventional
Air Jigging
Combined
Size
100 kt/y
300 kt/y
600 kt/y
100 kt/y
300 kt/y
600 kt/y
100 kt/y
300 kt/y
600 kt/y
NPV min & max

-12 to 0.475 MUS$
8.5 to 39 MUS$
50 to 101 MUS$
-20 to -6 MUS$
8.4 to 34 MUS$
38 to 90 MUS$
-19 to -6 MUS$
1.2 to 26 MUS$
39 to 87 MUS$
IRR min & max

-12% to 12%
20% to 61%
49% to 101%
-16% to 2%
16% to 38%
26% to 58%
-17% to 2%
10% to 29%
32% to 69%
Probabilty of
losses
93%
3%
0%
100%
0.08%
0%
100%
3.5%
0%

Figure 5 - Exemple of Simulation cumulative distribution curve and histogram
3. CONCLUSIONS
Production scale is a strong parameter (as usual) in terms of economicity of

plant units. From a throughput of 300 kt/year, the three strategies (conventional, with jigs,
with jigs and sorting) are almost successful to have no economic losses. Only the option
jigs has a very small chance (2% to have a negative cash-flow).
The second one, the service to be paid by the demolishers that was considered
to be between US$ 20 and US$ 40 per ton delivered at the recycling plant, is a key point.
The conventional plant that has the best economical result in the deterministic evaluation
(IRR of 71% and NPV@10% of 74 M US$) becomes negative of 29,5M US$ if there is
no payment for the recycling service.
Besides the throughput and payment for recycling, the simulation process
shows that investment in new technologies is potentially positive but the conventional
approach still more economical. However, the recovery simulation data used for the two
technological strategies, if not conservative, were not too optimistic in terms of final
product recovery. It gives room for economic improvement if the pilot tests to be done at
UFRGS in 2015 with actual CDW are successful.
Metal Revenue corresponds to 51% of the total recycling products revenue in

the Best Option (Conventional 600 kt/y). Jigging and Sorting technologies are focusing
only in 49% of Revenue without consider Services, with services becomes only 9% .
4. ACKNOWLEDGEMENTS
Monte Carlo simulation performed using software Risk Simulation of ROV
5. REFERENCES
[1] PORTER, M. (1990). Vantagem competitiva. Rio de Janeiro: Campus.
[2] Santos, A. L. (2009). Diagnstico ambiental da gesto e destinao dos resduos da construo e
demolio (RCD): anlise das construtoras associadas ao sinduscon/rn e empresas coletoras
atuantes no Municpio de Parnami. Natal: UFRN/ENGENHARIA DE PRODUO.
[3] Mular. (1990). Factored Estimate. Em L. Noakes, Cost Estimation Handbook (pp. 5-6). Victoria:
The Australasian Institute of Mining & Metallurgy.
[4] Iigo Vegas, K. B. (2014). Upgrading the quality of mixed recycled aggregates from construction
an demolition by usuing near infrared sorting technology. Construction and Building Materials 75
(2015) 121128.

RATIONAL MIX DESIGN APPROACH FOR RECYCLED

AGGREGATE CONCRETE
Marco Pepe (1), Romildo D. Toledo Filho (2), Eduardus A.B. Koenders (3), Enzo Martinelli (1)
(1) Department of Civil Engineering, University of Salerno, Italy
(2) Department of Civil Engineering, COPPE, Federal University of Rio de Janeiro, Brazil
(3) Faculty of Civil and Environmental Engineering, Institute for Construction and Building
Materials, Technical University of Darmstadt, Germany
Abstract
Being able to predict the mechanical properties of Recycled Aggregate Concretes (RACs)
is an essential requirement for Recycled Concrete Aggregates (RCAs) to be considered as a
feasible and sustainable alternative to the natural ones in structural concrete. In fact, since
RCAs show quite different engineering properties compared with natural aggregates, mix
design rules generally adopted for ordinary concrete can also not be applied in the case of
RACs. The research presented herein is intended at providing a contribution for better
understanding and making predictable the effects of RCAs on the resulting mechanical
properties of RACs. A series of analytical formulations, suitable for a rational mix design of
structural concrete made with RCAs, are proposed. These formulations take into account the
physical properties of the constituents and the concrete mixture composition for predicting the
time evolution of the concrete compressive strength. The formulations are calibrated on the
results received from experimental tests carried out on both recycled aggregates and concrete
mixtures, considering the variation of several aspects and parameters, such as the processing
procedures, the source of origin of the RCAs, their replacement ratio, the water-to-cement
ratio and, finally, the initial moisture condition of the coarse aggregates.
Keywords: Recycled Aggregate Concrete; Mix Design; Compressive strength; Porosity; Hydration
1.
INTRODUCTION
The construction sector is among the most resource and energy demanding industrial
activity. Therefore, the design of integrated recycling processes aimed at reducing waste to
produce new materials is of particular interest for reducing its environmental footprint and
enhancing its sustainability. In fact, hardened concrete can be crushed and reused in the
production of green concretes in which ordinary coarse aggregates are partially or totally
replaced by Recycled Concrete Aggregates (RCAs); this practice also results in reducing the
so-called Construction and Demolition Wastes (CDW) [1]. For considering RCAs as a
credible sustainable alternative of natural aggregates (NAs), the compressive strength
evolution of the resulting concrete (generally referred to as Recycled Aggregate Concrete,
RAC) has to be predictable as in the case of ordinary structural concrete. However, the
definition of accurate relationships capable of predicting the relevant properties of concrete
made out with these constituents is still considered an open issue. The main reason of this is
the lack of knowledge on the actual composition and the resulting engineering properties of
RCAs [2]. In fact, RCAs can be intended as a two phase composite consisting of the
original natural coarse aggregates and the Attached Mortar (AM), made of sand, hydration

products and fractions of un-hydrated cement. The Attached Mortar content is also the part
that is responsible for the increased adsorption capacity of RCAs due to its relatively higher
porosity in comparison to NAs.
The rational mix design approach proposed in this study for designing, controlling and
predicting the compressive strength of RAC is mainly based on identification of a key
parameter characterising the RCA properties and, consequently, the resulting concrete
performance. As a matter of the fact, porosity can be recognised as the main parameter
controlling the physical and mechanical properties of RCAs and the water transportation
phenomena occurring between cement and aggregates during the setting and hardened phases
of RAC [3]. In fact, the role of free water inside the RAC mixture becomes fundamental as it
determines the time evolution of the cement paste strength and the bond strength between
aggregates and cement (ITZs). In addition, the presence of a porous medium (i.e. RCAs) in
RAC results in a possible weak point for the resulting concrete composite. Based on this
approach, the authors proposes a series of simplified formulations taking into account the
porosity of RCA and how it affect the resulting properties for both RCAs and RACs.
2.
EXPERIMENTAL RESULTS AND MODELLING TECNIQUES
This section describes materials, techniques, methods and theoretical formulations used for
both investigating the physical properties of RCAs and unveiling their influence on the
resulting concrete performance [3].
2.1
Experimental characterisation of RCAs
The RCAs under investigation were collected from different sources, such as demolished
concrete structures and crushed concrete samples already tested in laboratory. Three different
size ranges were considered for the coarse aggregates: class 1, 2 and 3, i.e., nominal diameter
ranging between 4.75 to 9.5 mm, 9.5 to 19 mm and 19 to 31 mm, respectively. Moreover,
with the aim of better understanding the influence of the alternative processing procedure on
the aggregate properties, an autogenous cleaning process was conceived and performed [4].
Several experimental tests were performed both on single RCA particles and on
representative samples taken from fractions, for determining the relevant physical properties
of the RCAs: evaluation of the AM content by performing CT scan analysis [5], water
absorption capacity at 24 hours and particle density, determination of open porosity by
performing Mercury Intrusion Porosimetry [6]. An example of the CT scan technique,
performed on RCAs for characterising the internal morphology of the aggregates and
determining the actual volume percentage of the phases present inside the RCA particles, is
presented in Figure 1 where the clearer areas represent the old natural aggregates, the darker
part is the AM, and the porosity is represented by the black spots.
Figure 1: Mesoscopic view of a recycled concrete coarse aggregate.

The main experimental results performed on RCAs are summarised in Figure 2. As it can
be seen, the higher the amount of AM, the higher the porosity and, consequently, the lower
the particle density.
2800
Particle density [kg/m3]
AM = 1.4%
AM = 0.0%
2600
AM = 24.0%
AM = 48 5%
2400
AM = 25.0%
AM = 40.8%
2200
AM = 51.6%
AM = 55.9%
AM = 50.0%
2000
0
6
Porosity [%]
12
Figure 2: Physical properties for RCAs, Experimental data.

Finally, based on the analysis of the experimental results obtained from this study, some
analytical correlations between the actual AM content and the physical properties of RCAs,
such as the water absorption (that is strictly related to porosity) and particle density are
proposed in the next section.
2.2
Experimental characterisation of RACs and hydration process modelling
The experimental activities carried out on RACs were aimed at understanding the role
played by recycled aggregates in affecting the relevant properties of structural concrete. In
particular, several experimental tests were performed with the aim of investigating the
influence of relevant parameters, such as processing procedures for RCAs, initial moisture
condition of aggregates, aggregate replacement ratio and nominal water-to-cement ratio, on
the resulting concrete performance. Table 1 describes one of the experimental campaigns
performed for analysing the role of the aggregate replacement ratio, the initial moisture
conditions of coarse aggregates (i.e., oven dried or saturated with dry surface) and the
nominal water-to-cement ratio.
Table 1:Experimental campaign for RAC.
Mix
w/c RAC
Moisture condition
0.50_OC_Dry
0.50
0%
Oven Dried
0.50_OC_Sat
0.50
0%
Saturated with dry surface
0.50_RAC50_Dry 0.50 50%
Oven Dried
0.50_RAC50_Sat
0.50 50%
Saturated with dry surface
0.50_RAC100_Dry 0.50 100%
Oven Dried
0.50_RAC100_Sat 0.50 100% Saturated with dry surface
0.60_RAC100_Dry 0.40 100%
Oven Dried
0.40_RAC100_Dry 0.60 100%
Oven Dried
w/ceff
0.50
0.50
0.46
0.50
0.39
0.50
0.32
0.40
0.47
0.60
Rcm,28days [MPa]
49.12
47.00
47.55
40.88
38.57
31.66
42.39
41.15
36.60
22.59

For each mixture, cubic compressive strength tests were conducted after 1, 3 and 28 days
of curing in a water bath under isothermal conditions at a temperature 202C. Moreover, the
time development of temperature was measured in the centre of the cubes, prepared for all
concrete mixtures and cured in semi-adiabatic conditions [7]. The main compressive strength
results for this experimental campaign are reported in Table 1: they clearly show that the use
of RCAs in concrete, as well as their initial moisture condition, influences the resulting
concrete performance. In fact, when comparing the results of the concrete mixes characterised
by the same nominal water-to-cement ratio and same initial moisture condition it turns out
that with enhancing aggregate replacement ratio, the 28 days compressive strength is
decreasing. This effect becomes even more pronounced whenever employing saturated initial
moisture conditions. This result can be explained by the higher porosity of recycled
aggregates. In fact, when dry conditions are adopted, the aggregates tend to absorb part of the
mixing water and, then, reduce the actual water-to-cement ratio. Based on the above
considerations and the complete analysis of the results [3], it is clear that the initial moisture
condition and the aggregate porosity tend to modify the water-to-cement ratio and, for this
reason, an effective w/c ratio taking into account both the initial moisture condition and the
aggregate porosity can be defined as follows:
w w add
n p P
w
=
+
i i

c
c
c eff
i =1 c
(1)
where w/c is the nominal water-to-cement ratio, wadd is the extra water added to the mix
corresponding to (partially or totally) soaking the aggregates, pi and Pi represent the
absorption capacity and the weight in the mixture of the i-th aggregate fraction, and is a
parameter taking into account the initial moisture condition of the aggregates and is zero in
saturated conditions and 0.5 in dried ones [3].
Moreover, the authors recently proposed a theoretical model for predicting the time
evolution of the degree of hydration from temperature measurements performed on concrete
cubic samples cured under semi-adiabatic boundary conditions [7]. Therefore, the proposed
model enables calculation of the time evolution of the degree of hydration based on semiadiabatic temperature measurements and, based on this, simulate hydration reactions for
different thermal boundary conditions.
Once the time evolution of the degree of hydration has been calculated, the experimental
results obtained in compressive tests at a certain curing age and the corresponding values of
the degree of hydration were correlated. In fact, the following conceptual formula is available
in the literature for correlating degree of hydration and compressive strength Rc [8]:
0
R c = R c,max
(2)
1 0
where 0 is the minimum degree of hydration representing the moment that the strength in the
mixture is starting to build up (i.e. end of setting phase), and Rc,max the maximum cubic
compressive strength, ideally corresponding to a degree of hydration equal to 1. Then, the
correlation between the mechanical properties and the degree of hydration may show the
impact of RCAs on the resulting RAC properties. Particularly, it may express the influence of
the residual anhydrous cement fractions, the porosity and the moisture conditions of the
attached mortar on the overall hydration process that controls the formation of the concrete
microstructure and associated strength.

3.
PHYSICAL PROPERTIES OF RCAS: THE RULE OF ATTACHED MORTAR
Based upon the experimental results achieved for RCAs, a possible linear correlation is
proposed for describing the variation of the porosity p as a function of the Attached Mortar
(AM) content:
d
p = [p NA (1 AM ) + p AM (AM )] 0
(3)
d
where d is the diameter of the aggregate expressed millimetres and d0 is a reference diameter
of 20 mm. This formulation can be interpreted as a liner combination of the two phases NA
and AM. In fact, pNA represents the porosity of a natural aggregate inside RCA, while pAM
represents the porosity of the attached mortar residing to it. The calibration of equation (3) led
to approximate values for pNA and pAM equal to 0.45% and 15%, respectively.
Moreover, the particle density represents another fundamental parameter to be determined
for the mix design of RAC mixtures. The results reported in the previous section highlight the
particle density and complies with the general trend that its value tends to decrease with an
enhancing open porosity. As a first order approximation, the following correlation can be
proposed for the relationship between the particle density and the open porosity:
= 0 (1 p )
(4)
where 0, representing the particle density of a fictive aggregate in which the open porosity is
zero, is equal to 2700 kg/m3 and is a constant parameter that is taken equal to 2.
Finally, as already mentioned, an autogenous cleaning process was conceived and
performed on some employed particles. In particular, the autogenous cleaning implemented at
the laboratory scale led to interesting results in terms of enhancement of properties of crushed
concrete particles, especially in terms of reduction of the attached mortar content on RCA
surfaces and, consequently, on their water absorption capacity [4]. Consequently, an
analytical expression is proposed for evaluating the actual open porosity, p, as a function of
the autogenous cleaning time t, expressed in minutes:
at
p(t ) = 1
pt =0
(5)
b+t
where pt=0 is the initial open porosity of the aggregates without autogenous cleaning, a is a
constant equal to 0.6 for class 1 and 0.2 for the class 2 aggregates, while b=2 in both cases.
4.
A RATIONAL MIX DESIGN APPROACH FOR RAC
The first step in formulating a model capable of predicting the mechanical strength
development for RAC is based on taking into account both the concrete mixture composition
and the main parameters characterising RCAs.
As a matter of the fact, the first possible correlation can be elaborated between 0 and
Rc,max (calibrated for each produced mixture as described in section 2 [7]) and the effective
water-to-cement ratio defined by equation (1), which takes into account the amount of free
water available into the concrete mixture. However, the concrete strength is also affected by
the aggregate replacement ratio which is not explicitly considered in the definition of (w/c)eff.
Therefore, a further parameter is introduced that accounts for the average porosity of the
coarse aggregates employed in the mixture, and is defined as follows:
n
A MIX = p i Vi
i =1
(6)

where pi is the open porosity characterising the i-th fraction of the coarse aggregates and Vi is
the corresponding fraction volume.
This parameter can be useful for describing the reduction in compressive strength due to
the replacement of NAs by RCAs. Particularly, it can physically explain the decrease of 0
and Rc,max that characterises the evolution of the compressive strength of RACs with respect to
0,NAT and Rc,max,NAT, which represents the compressive strength of a reference concrete made
of only natural aggregates (and keeping the grain size distribution constant). Therefore, it is
possible to define the ratios of r* and * for expressing the reduction in strength and its rate
development in terms of Rc,max and 0, respectively. Moreover, the following analytical
formulations are introduced for describing the existing relation between the above described
parameters r* and * and the ANAT-to-AMIX ratio:
A
A
a NAT
a R NAT
R c, max
0
A MIX
A MIX
r* =
=
* =
=
(7)
(8)
R c, max, NAT b + A NAT
0, NAT b + A NAT
R
A MIX
A MIX
where, ANAT is the corresponding AMIX value for an ordinary concrete mixture and aR, bR, a
and b are constants that should, in principle, be calibrated on experimental results. Based on
tested mixtures, the values of aR and bR can be approximated to be 1.06 and 0.06, respectively,
while a and b are equal to 1.05 and 0.05.
As a result, if the individual pairs of Rc,max and 0 are divided by the corresponding values
of r* and * and, then, represented with respect to the effective water-to-cement ratio. The
analysis of these results [3] shows that all the points (w/ceff-Rc,max/r*) can be aligned on a
curve that can be analytically expressed by the well know Abrams law and, particularly, it
assumes the expression (9). Similarly, a possible linear correlation existing between w/ceff and
0/* can be expressed by the analytical formulation (10).
AR
w
r*
R c,max = w/c
B *
(9) 0 = A
(10)
eff
BR
c eff
In formula (9) the values of AR and BR can be assumed to be equal to 265 and 9.5,
respectively, while, in equation (10) A is equal to 1.62 and B is equal to 0.28.
The accuracy of the proposed rational method can be estimated by analysing Figure 3
where both the experimental data and the model prediction results are reported in terms of
cubic compressive strength of RACs at 28 days.
Compressive strenght at 28 days [MPa]
60
Experimental data
50
Model prediction
40
30
20
10
0
Figure 3: Rational mix design approach calibration.

200
Rc,max [MPa]
4.1
Design abaci and model potential
The novel approach methodology is based on defining the parameter AMIX that represents
the average porosity of the concrete mixture made with recycled concrete aggregates. Once
defined the composition of the concrete mixture the parameter AMIX is evaluable by applying
the formula (6) and, consequently, the ratio ANAT/AMIX is defined.
Figure 4 shows the trend of Rc,max and 0 values related to both the ANAT/AMIX parameter
(representing the amount of recycled concrete aggregates or, particularly, the amount of the
AM inside the resulting concrete mixture) and the effective water to cement ratio (that takes
into account of the initial moisture condition of coarse RCAs).
1.0
ANAT/AMIX
ANAT/AMIX
160
120
1.00
0.50
0.30
0.10
0.05
0.8
1.00
0.50
0.30
0.10
0.05
0.6
80
0.4
40
0.2
0.0
0
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.2
0.3
0.4
0.5
0.6
0.7
0.8
w/c eff
w/c eff
Rc,max (w/c)eff (a)
a0 (w/c)eff (b)
Figure 4: Influence of ANAT/AMIX on Rc,max and 0.
The curves reported in Figure 4, described by the proposed formulas (9) and (10), have the
meaning of mix design abaci for RAC. In fact, once defined the Rc,max and 0 parameters, the
time evolution of the compressive strength could be easily determined by considering the
numerical model proposed by the authors for estimating the time evolution of the degree of
hydration relation. Alternatively, the degree of hydration at 28 days can be determined by
means of the following relationship [9]:
w
1.031
c eff
max =
(11)
w
0.194 +
c eff
Finally, the above described rational method was applied to some experimental results
available in literature [10]. The comparison of this analysis are reported in Figure 5 and shows
that even if the proposed formulation should be better calibrated with further experimental
data, the introduction of the parameter AMIX is able to explain the changes of the mechanical
performance of RACs compared with ordinary mixtures.
50
40
30
20
Proposed Model
10
Younis & Pilakoutas (2013)
Compressive strenght [MPa]
60
60
60
50
40
30
20
Proposed Model
10
20
40
60
80
Aggregatate replacement ratio [%]
Aggregate type 1 (a)
100
40
30
20
Proposed Model
10
50
20
40
80
60
Aggregate type 2 (b)
100
20
40
60
80
Aggregate type 3 (c)
Figure 5: rational mix design method potential.
100

6.
CONCLUSIONS
This paper proposed a rational approach to assess the compressive strength development of
concretes made with recycled concrete aggregates (RCAs). The following conclusions can be
drawn:
- the AM in RCAs results in higher porosity than in NAs and and influences properties,
such as water absorption capacity and particle density;
- due to the higher porosity of RCAs, the resulting concrete performance is significantly
affected by the following parameters: the initial moisture condition of RCAs, aggregates
replacement ratio and the nominal water-to-cement ratio;
- a physically based definition of effective water-to-cement ratio is proposed for taking into
account the effect of the aforementioned parameters;
- the time evolution of the compressive strength of RACs can be estimated by considering
the parameters characterising the recycled aggregates, i.e. attached mortar that is included
in the proposed parameter AMIX.
Finally, it may be worth highlighting that, in spite of the promising conceptual results
emerging from the proposed analyses, a wider set of experimental data is needed to confirm
its general applicability. This would contribute to the formulation of a quantitative method to
predict the mechanical behaviour of Recycled Aggregate Concretes.
ACKNOWLEDGEMENTS
The present study is part of the activities carried out by the Authors within the "EnCoRe"
Project (FP7-PEOPLE-2011-IRSES n 295283; www.encore-fp7.unisa.it) funded by the
European Union within the Seventh Framework Programme.
REFERENCES
[1] McNeil K. and Kang T.H.K. Recycled concrete aggregates: A review, International Journal of
Concrete Structures and Materials 7(1) (2013) 61-9.
[2] Behera M., Bhattacharyya S.K., Minocha A.K., Deoliya R. and Maiti S., Recycled aggregate
from C&D waste & its use in concreteA breakthrough towards sustainability in construction
sector: A review, Constr Build Mater 68 (2014) 501-16.
[3] Pepe M. A conceptual model to design Recycled Aggregate concrete for structural applications,
Ph.D. thesis, University of Salerno, Italy (2015) pp.210.
[4] Pepe M., Toledo Filho R.D., Koenders E.A.B. and Martinelli E., Alternative processing
procedures for recycled aggregates in structural concrete, Constr Build Mater 69 (2014) 124-32.
[5] ASTM E1570, Standard Practice for Computed Tomographic (CT) Examination. American
Society for Testing and Materials, USA (2011).
[6] ASTM D4404, Standard Test Method for Determination of Pore and Pore Distribution of Soil and
Rock by Mercury Intrusion Porosimetry, American Society for Testing and Materials, USA
(2010).
[7] Martinelli E., Koenders E.A.B. and Caggiano A., A numerical recipe for modelling hydration
and heat flow in hardening concrete, Cement and Concrete Composites 40 (2013) 48-58.
[8] Lokhorst S.J., Deformation behaviour of concrete influenced by hydration related changes of the
microstructure Internal Report nr. 5-99-05, Delft University of Technology, Netherland (1999)
pp.178.
[9] Hansen T.C., Physical structure of hardened cement paste. A classical approach, Mater Struct 19
(6) (1986) 423-36.
[10] Younis K.H. and Pilakoutas K., Strength prediction model and methods for improving recycled
aggregate concrete, Constr Build Mater 49 (2013) 688-701.

PUTTING RECYCLED CONCRETE INTO PRACTICE: LESSONS

LEARNED FROM BELGIAN PILOT & DEMONSTRATION PROJECTS
Jeroen Vrijders (1), Bram Dooms (1), Petra Van Itterbeeck (1)
(1) Belgian Building Research Institute www.bbri.be, Brussels, Belgium
Abstract
In order to stimulate the use of recycled aggregates in concrete in Belgium, 15 pilot
projects were selected to demonstrate the possibilities in several road and building
applications. Project teams (designers, contractors, ) were guided throughout the design
and the preparation phase of actual construction projects on real scale. The execution on site
was documented and evaluated afterwards. This paper presents the lessons learned from these
projects. On the one hand, a manual or a step-by-step-plan was elaborated to assure good
quality and good execution. 3 main steps are followed in each project: First, a smart selection
of applications is made, by setting the ambitions (choice of recycled materials, binders,
replacement level, ) and matching the objectives to the technical possibilities and the
prescriptions of the European and Belgian standards. In a second phase, the project is
prepared by selecting high quality aggregates and initial type testing on the concrete
compositions. In the final stage, several executional aspects (workability, finishing, curing,
) are described. On the other hand, the pilot projects gave opportunity to evaluate the
performance of concrete produced with recycled aggregates in industrial circumstances. This
allows for comparison with available scientific data from laboratory research. The concrete
compositions in the different projects are evaluated on workability, compressive strength and
- depending on the application domain - also on resistance to carbonation and frost-thawcycles (with or without de-icing salts). Where possible, the results were compared to those
obtained with a reference concrete, used on the same construction site, made without recycled
aggregate. The pilot and demonstration projects are all well-documented and will be
monitored throughout the years to come, in order to evaluate their durability in real life
circumstances.
Keywords: Pilot projects, recycled aggregates, concrete, practical experience
1.
INTRODUCTION
The recycling of construction & demolition waste in Belgium is considered to be a grown

up, well-developed economic activity in an established market. However, only an estimated
amount of 1% of the available recycled aggregates is currently used as aggregate for structural
concrete [1]. This equals about 200,000 tons on 18 million tons each year. Given the mature
market circumstances, the recycling sector is looking for new applications with higher added
value, for example in closed-loop concrete. At the same time several clients, architects and
designers are interested in using greener building products in their projects, but they are

lacking knowledge and/or confidence with regard to the use of recycled aggregates in
concrete.
In order to bridge the gap between the potential offer and demand, in 2013 the project
NIB-Stortklaar Beton voor de Toekomst (Eng: Ready-mix concrete for the future) was
initiated, with the specific goal to offer technical and practical guidance to the whole
construction team (recycler, concrete producer, contractor, client, designer, 3rd party
verification agent, ) in order to realise actual building projects as demonstrators and pilots
for green concrete.
The last two years 10 construction projects were executed. 5 more projects are in
preparation for execution in 2015-2016. The guidance delivered and follow-up on these
projects allows to learn several lessons, on the one hand in terms of process to follow
towards realisation, on the other hand on actual results on the performance of the concrete.
2.
OVERVIEW OF COMPLETED PROJECTS
An overview of the completed pilot projects is presented in Table 1, at the end of the
article. The projects are varying in terms of application domain (roadworks: 3, industrial
concrete floors: 5, building structures: 2) and thus in concrete composition and requirements.
Most pilots opted to replace virgin aggregate by recycled concrete aggregate (8) or
manufactured aggregates like slags (2); in only 1 project, alternative binders were used as way
to reduce the environmental impact. The replacement ratio varies between 20% to 100% (of
the coarse aggregate fraction).
The overview shows clearly that green or recycled concrete can be applied in several
application domains, with different levels of ambition. However, there appears to be a
preference to use well-known and high quality coarse recycled aggregates as a first step in
greening the concrete, mainly in horizontal applications like floor slabs. This is due to the
lack of knowledge and confidence in other aggregate types like mixed C&DW aggregate
(including bricks etc.) or sands, and the fact that more demanding concrete applications like
beams and columns in constructions require more investigation (e.g. on creep behaviour).
3.
PROCESS
A guidance procedure or step-by-step approach was developed within the NIB-project by

the partners (BBRI, FPRG, FedBeton & NAV), and implemented in each of the pilot projects.
This was done to assure a well-thought and qualitative application of recycled concrete in
each case for the client and the designer. The step-by-step procedure puts forward the most
relevant and frequently asked questions (FAQs) in the development of a recycled concrete
project.
3.1
Motivation
The starting point of each project consists in a motivated choice to use recycled aggregates
in concrete. As stated, in Belgium, this is not a common practice, and thus a client must make
a definite choice to start the process. In most projects, a mix of environmental, economical
and social arguments led to the choice for greener concrete. Two main clusters of projects
can be identified:
- Industry driven (= offer side): producers of recycled aggregates, or concrete companies
that already had some experience in the use of concrete aggregates in concrete, want to
show the possibilities and to investigate or explore the limits of applications. The main
drivers are economical value (e.g. to create added value for the industry, to work at
lower cost) and sustainability (e.g. limited use of resources, avoidance of waste).

Demand driven: several clients explicitly choose to use concrete with recycled
aggregates, because of environmental concerns and societal needs: closing the material
loop, responsible attitude towards resource consumption, One client did it because
of the requirements of a sustainability certification scheme (BREEAM).
3.2
Design choices
Since 2012, Belgium has a concrete standard that allows for use of recycled concrete
aggregates and specific types of slags in structural concrete: NBN B15-001:2012 [2].
Replacement up to 20% (in volume) of the coarse fraction is allowed in concrete to be used in
interior applications (classes X0 and XC1 according to EN 206-1). Other application domains
are allowed, on condition that specific suitability is proven. However, it remains difficult to
obtain a quality label for concrete with recycled aggregates since the regulations for obtaining
the label are not adapted to use of recycled aggregates [3]. This usually forms a difficulty in
convincing the client or the engineer responsible for the design to implement recycled
concrete. Also for road applications, specifications exist since 2013 for concrete for linear
elements and the base layer of two-layer-concrete pavements in the Flemish Standard
Specification for roadworks, SB250.
From scientific knowledge and previous experience [4], it is known that higher ambition
levels can be justified in terms of application domain and replacement ratio, but this implies
that the client or designer surpasses the limits stated by the current Belgian normative
framework. Thus, at the design stage, choices have to be made: either applications are
identified in the project that fall within the current prescriptive context and can be considered
as generally suitable, either other applications and/or replacement levels are chosen. In this
last case, this requires extra testing/calculation/preparation/ to ensure the specific
suitability.
There are also specific combinations of replacement levels and applications where little
knowledge is available (e.g. higher replacement ratios in combination with sea water
environment), and that should not be part of real scale applications at this moment.
3.3
Preparation phase
Although not implemented for all of the projects due to timing constraints, it was advised
that a preparation phase took place, in order to assure the quality of the aggregates and the
concrete on beforehand. The most important aspects during this preliminary phase are:
- Classifying & characterising the recycled aggregates - especially if the technical
characteristics are not well known from standard quality assurance procedures. The
most relevant parameters are the grading (including fines), the composition, the particle
density and the water absorption as indicators for the quality of the recycled aggregate.
- Defining the concrete composition performance: in first instance general aspects like
workability, effective W/C-ratio, compressive strength, should be evaluated, but in
function of the chosen application(s) other specific aspects had to be investigated, more
specifically in terms of durability. In conditions where carbonation and frost attack
(with/without de-icing salts) can take place, specific testing on the concrete had to be
done in order to demonstrate the suitability for use.
- Gaining practical experience on aggregate preparation methods (pre-humidification),
mixing methods, workability behaviour, finishing, etc.
Some results of the preparation phase and tests on executed concrete are discussed in
paragraph 4 below.

3.4
Execution
The goal in all projects was to make concrete with recycled aggregates that performs
equally to concrete with primary aggregates, and thus, no difference should be noticeable on
the construction site during the execution of the works. However, differences were noticed
between projects where more experienced manufacturers (who had made recycled aggregate
concrete before) delivered the recycled concrete and sites where producers who used the pilot
projects as first experience. In the latter case, a drop in workability during the works and a
different moment of hardening and finishing of the concrete were noticed more often. This is
probably due to a lack of experience with the pre-treatment of the recycled aggregates. In the
projects where the aggregates were pre-wetted before application in the concrete mix, no
problems were noticed. However, a larger variation in workability (slump) also was noticed in
these projects.
These findings have led to further research within BBRI to assess the water household of
concrete with recycled aggregates & the effects on workability and finishing. In any case,
contractors should be informed on beforehand on the possible differences in workability and
finishing protocol between conventional and recycled aggregate concrete.
4.
RESULTS & CONCRETE PERFORMANCE
This section presents some of the concrete research results obtained from the pilot projects.
Interesting to notice is that most of the results come from actual concrete used on site, and
thus from industrial scale installations and not from labo-scale mixers or laboratory
environment. This allows to compare the real concrete results to the vast amount of laboratory
research available.
Fresh & hardened concrete
4.1
The concrete with recycled aggregates had similar properties in fresh condition (slump,
workability) as conventional concrete besides the attention points noticed in 3.4: variance
in slump & a faster drop in workability. Additionally, tests on fresh concrete showed that
concrete with recycled aggregates can have a higher air content (a surplus of 1-2%). A
possible explanation is the use of a specific type of superplasticizer that interacts with the
concrete mix. This has a negative effect on the compressive strength, but can be beneficial for
the frost-resistance of the concrete.
In all projects, the required compressive strength was obtained. There was however a slight
decrease (-5%) in compressive strength noticeable when a limited amount (<30%) of recycled
aggregate was used, which became more clear (-10%) when replacing higher volumes (>
30%). This is in line with the vast amount of research available, e.g [4].
4.2
Durability
An important issue hindering the application of recycled concrete in Belgium is the lack of
trust in the durability of the concrete. Therefore, the 10 executed pilot projects were an ideal
opportunity to collect as much information as possible by testing different concrete
compositions. Since the application domain was reinforced concrete exposed to frost and
rain, focus was laid on resistance to carbonation and resistance to frost-thaw attack. For road
applications, also the water absorption of the concrete was measured.
The resistance to carbonation was measured according to NBN EN 13295, on prisms
(10x10x40cm) made on site, thus with the concrete applied in practice. In none of the projects
a significant difference was noticed between concrete with recycled aggregates and
conventional concrete.

The figure below shows the results of one of the case studies (project n 8 in Table 1). The
average measured carbonation depth (average of 3 samples) after 56 days in an environment
containing 1%CO2 for different types of concrete is shown. Within project n8 two types of
concrete were tested with a different W/C-ratio (0.50 and 0.45) containing 0%, 30% and
100% of recycled coarse concrete aggregate.
It should be noted that the recycled concrete aggregates used are of a very good quality
see Table 2 for some characteristics. This is in line with most literature results, stating that
coarse aggregate replacement, when having a very good quality aggregate, has a limited
impact on the carbonation behaviour of the concrete.[5]
Figure 1 Average carbonation depth in mm after 56 days of exposure, of concrete with

(30R, 100R) and without (0R) recycled aggregate, with 2 different W/C-ratios.
Table 2: Recycled concrete aggregate properties used in project n8
Property
Test method
Result
Requirement (NBN
B15-001:2012)
Fines content
EN 933-1
0.4%
1.5%
Flakiness Index
EN 933-3
20
Particle density (rd)
EN 1097-6
2410 kg/m
2200 kg/m
Water absorption (24h)
EN 1097-6
4.2 %
10%
Resistance to freezing
& thawing
EN 1367-1
2.63%
2% (deicing agents) or
4% (no deicing agents)
Altering rate of setting

& hardening
EN 1744-6
3 min.
10 min. or 40 min.
Acid-soluble chlorides
EN 1744-5
0.03%
Water soluble sulfates
EN 1744-1
0.08%
For the relevant projects, frost-thaw-resistance was measured, according to CEN/TS

12390-9 - slab test method. In general, the resulting mass losses after 28 or 56 cycles clearly
show:

The influence of air entraining agent. The compositions with air entrained perform
much better. The influence of the use of air entraining agent is much bigger than the
use of recycled aggregates.
The influence of the use of recycled aggregates is only visible when higher
replacement ratios were applied: below 50% replacement of the coarse fraction, no
clear differences were noticed between conventional concrete and RCA-concrete.
Replacement up to 100% resulted in higher mass losses and thus a less frost resistant
concrete.
The concrete compositions for roads (lower W/C, air entrained) all perform well, even
with 100% recycled aggregate. Although, the water absorption rate of the concrete
(which is used as a proxy to frost resistance) is higher than allowed by Flemish
specifications (6.5%). This probably has to do with the lower volume mass and the
higher air content entrapped by the recycled aggregates.
The figure below shows these effects on samples taken from project n 8 (produced with
aggregates from Table 2). The air-entrained concrete performs the best, even with use of
100% recycled aggregate. 30% replacement in concrete with a W/C-ratio of 0.45 has a small
influence, whereas 100% replacement causes higher mass losses.
Figure 2 Average mass loss (kg/m) in slab-test with de-icing salts after 7-14-28-44-56 cycli
for different concrete compositions: W/C = 0.45 or 0.50, with/without air entraining agent and
applying 0-30-100% of recycled coarse concrete aggregate
All-in-all, the results of the different projects are in line with the vast amount of research
available: it is feasible to make concrete on an industrial scale, using recycled aggregates up
to a certain replacement ratio, that is equivalent to conventional concrete in terms of fresh
concrete properties, compressive strength and durability. An important element to notice is
that in all projects efforts were made to work with high quality concrete aggregate. This
aggregate was produced from selected sources (e.g. demolished road pavements containing
pure and qualitative concrete) and/or produced in a specific crusher process (e.g. crushed
twice to retain only the best quality of concrete aggregates).
5.
OUTCOME
An important result of the project was the elaboration of a technical specification

document that can be used in projects. It allows the different parties (client, designer,

contractor, concrete producer, ) to agree terms on the use of recycled aggregates in

concrete. It gives examples on how to prescribe concrete with recycled aggregates, and how
to cope with quality assurance, where the regular Belgian scheme cannot be used (due to lack
of clauses on recycled aggregates). In this way, concrete with recycled aggregates can be
applied in various projects with a high level of confidence. The 5 remaining projects will use
this document as test cases, in order to gain more knowledge on agreements between parties,
responsibilities, contracts, .
6.
CONCLUSIONS & OUTLOOK
The executed pilot projects demonstrate the feasibility of using concrete with recycled
aggregates in Belgian practice. Several application domains and ambition levels can be
covered. There are some technical attention points in the process, mainly in the workability of
the concrete. The step-by-step procedure developed and the technical specifications should
assist project managers with the implementation in practice. In terms of concrete
performance, it becomes clear that real scale concrete performs in the same line as concrete
made on laboratory scale: smaller replacement ratios have less influence. Once 50-100%
replacement is targeted, durability and strength loss have to be investigated on beforehand.
The executed projects are all well documented, and there will be a follow-up stage in the
coming years to assess the real life performance of the executed projects.
ACKNOWLEDGEMENTS
The authors wish to thank the Vlaamse Overheid, Agentschap Ondernemen who funded
the NIB-project Stortklaar beton voor de toekomst in the framework of Vlaanderen in Actie,
as well as all the construction parties (manufacturers, contractors, designers, ) involved in
the pilot cases.
REFERENCES
[1] Vrijders, J. and Desmyter. J., Recycling in Belgium: an overview of the present situation in
Belgium from the recycling industry and research about CDW, in Vazquez E. (ed.), Progress of
Recycling in the Built Environment: Final Report of the RILEM Technical Committee 217-PRE,
RILEM, Netherlands, 2013, 37-41
[2] NBN, NBN B15-001 - Beton - Specificatie, eigenschappen, vervaardiging en conformiteit Nationale aanvulling bij NBN EN 206-1 : 2001., 2012
[3] Vyncke, J. and Vrijders, J., Recycling of C&D Waste in Belgium: State-of-the-Art and
Opportunities for Technology Transfer in Proceedings of the Second International Conference
of Sustainable Construction Materials and Technologies, Ancona, June, 2010, 2010.
[4] De Brito, J. and Saikia, N., Recycled Aggregate in Concrete Springer London, 2013.
[5] Silva, R. et al., Carbonation behaviour of recycled aggregate concrete, Cement & Concrete
Composites 62, p.22-32, 2015.
Road (bicycle
farmer path)
concrete
Fresh
execution
& Compression
strength
Frost resistance is very good

Tests in progress
Good
Tests in progress
(linear Recycled concrete aggregate Normal

(50%)
Floor Recycled concrete aggregate Normal

(30-50%)
Road
elements)
Industrial
(outside)
10
0.45-0R: 60.51 MPa,

0.45-30R : 55.97 MPa,
0.45-100R : 59.15 MPa)
Good but smaller No influence on carbonation

resistance. High replacement
margins to limits
(0.50-0R: 56.9 MPa, ratio has influence on frost
0.50-100R: 52.74 MPa, resistance, but still good
Floor Recycled concrete aggregate Normal

(30-100%)
Industrial
(outside)
Not considered
Building
(foundations
floor slab)
Recycled concrete aggregate Slump normal, but loss of Good

workability in time
& (30-100%)
MPa, 100R : 58.1 MPa)
Floor Recycled concrete aggregate Normal but higher variation Good but losses due 50% replacement performs
Industrial
replacement
100%
well,
to recycling
between concrete types
(50-100%)
(outside & inside)
frost test.
on
worse
performs
(0R-air: 59.8 MPa, 50Rair: 54.3 MPa, 0R : 66.7 Carbonation: no difference
Good frost resistance is

attention point
Good
Building (internal Slag (20%)

walls & floors)
Normal
(0R: 51.5 MPa, 20R:

50.4 MPa, 30R: 47
MPa)
Floor Recycled concrete aggregate Normal. Shorter time to Good small losses Recycled concrete performs
Industrial
due to recycling better than reference
finish concrete.
& lead slag (20-30-40%)
(outside & inside)
Initial mass loss in slab test

(due to inferior finishing)
Frost resistance is very good
Too cold for a decent Good

finishing process
better Good (small loss)
Road
hall)
storage Lead slag (50%)
even
& Recycled concrete aggregate Normal,
workability
(100%)
Carbonation is attention point.

Frost resistance is good.
Durability
(to
Industrial
(outside)
Green aspects
Floor Alternative binder, Recycled Too dry, slump too low Good
(due to absorption by RCA)
concrete aggregate (100%)
Application
domain
Project
n

ANNEX TABLE 1 OVERVIEW OF PROJECTS

CHEMICAL-MINERALOGICAL SORTING OF MINERAL

CONSTRUCTION AND DEMOLITION WASTES FOR A BETTER
RECOVERY AGAINST THE STATE-OF-THE-ART
M. Landmann (1), I. Dring (1), L. Raumann (2), E. Lin (3), A. Mller (1), U. Palzer (1)
(1) IAB - Institut fr Angewandte Bauforschung Weimar gemeinntzige GmbH
(2) IZF - Institut fr Ziegelforschung Essen e.V.
(3) Bauhaus-Universitt Weimar
Abstract
One means of bringing recycled construction materials back into the materials cycle is to reuse
them as feedstock in production. To establish a closed-loop cycle in this sector it is necessary to
ensure a sufficient purity of crushed and sorted recycled material. The increasing material
diversity in mineral construction and demolition wastes (C&DW) as well as the arrival of more
complex composite materials and structures requires sophisticated approaches to mineral
processing in the construction sector. One approach for achieving better recovery of usable
materials from mixed C&DW is presented in this paper. Higher product quality of recycled
materials could be achieved through the implementation of modern, automatic sorting techniques
in C&DW recycling that can differentiate between materials based on their chemicalmineralogical characteristics, especially for the middle-sized fractions 4/45 mm and the recycled
sands 0/4 mm.
Keywords: Construction and demolition waste, sensor-based sorting, composites, processing
1.
INTRODUCTION
At present, construction and demolition waste (C&DW) recycling is mainly realized by

simple but proven techniques in order to process large quantities of construction rubble in a
short time. However, these techniques will not be able to handle the increasing complexity of
composite materials and structures in the mineral building materials industry for several
reasons:
- In the mid- and long-term, the increasing number of building materials in the
construction industry will lead to an increasing heterogeneity of recycled materials. It is
anticipated that the material characteristics of recycled C&DW are subject to certain
fluctuations which make value-added recycling more difficult. Without a response to
these changes, the quality of recycled products will decrease. Modern sorting techniques
are required to turn those mixtures into usable and valuable material fractions, e.g. for
closed material cycles.
- The increase of composite materials and structures in the construction industry requires
adequate liberation techniques to separate materials. Adhesive bonds that normally have

to withstand external mechanical influences for as long as possible must be broken

down during processing.
Figure 1: Heterogeneity in C&DW; mixed recycled C&DW 0/45 mm (left) and composite
structures (right) with integrated insulation material that must firstly be crushed under a nonhand sortable grain size smaller than 45 mm
The challenges of processing composite material waste streams from modern structures
have been already discussed in [1]. The main process is characterized by manual sorting
followed by a crushing step. This sequence will presumably no longer be suitable for the
production of homogeneous recycled materials from increasingly complex composite
materials. In order to physically unlock usable material, it is necessary to liberate particles
through comminution to a grain size smaller than 45 mm, which is no longer sortable by hand.
Subsequently, an automated sorting step is required to sort the fractions 0/45 mm.
In order to separate complex composite materials, the sorting and crushing steps must be
reversed. To increase the recovery ratio in the short term, conventional crushers can be used
to liberate valuable aggregates by liberation comminution. Through such a modification, the
previously non-sortable fraction (mostly 0/45 mm), as well as diverse composites, could be
processed to a high level of quality. Through this change, the usable material can be liberated
from structures to the extent of the production of more fine fractions 4 mm.
The following sorting technologies must be still developed in order to finally turn the
recycled mineral mixtures into homogeneous and thus usable materials.
2.
STATE-OF-THE-ART IN SORTING
Sorting techniques are needed to separate bulk materials with characteristic material
properties and to remove impurities from the usable and valuable material fractions. Sorting
of minerals can be achieved according to either their physical or chemical-mineralogical
characteristics.
Sorting by particle density or shape, e.g. through wind sifting, is already state-of-the-art. It
is used to separate light from heavy mineral fractions. Another example is the setting
technique as wet process.

Density separation methods are suitable to separate lightweight materials in the coarser
particle size range. However, it remains difficult to sort fine and medium sized particles, in
particular when wet materials are present. The particle size and shape are determining factors.
Due to the large material diversity with a wide range of densities and the emergence of
new building materials in mixed C&DW - e.g. due to energy saving regulations for
buildings - sorting by physical properties, such as the bulk density, may be difficult in the
future. This is due to the fact that densities are very variable.
In that regard, conventional techniques may be limited in their effectiveness, and cannot
assure the required product quality for recycled materials. Therefore, sorting processes
focusing on chemical-mineralogical characteristics will increase in importance [2].
3.
DENSITY-INDEPENDENT SORTING TECHNIQUES
3.1
Importance of density-independent sorting techniques
Sorting is the prerequisite for turning waste materials into usable materials with a defined
product quality. Figure 2 illustrates the importance of (new) sorting techniques. In particular,
improved quality seems to play a major role in future recycling.
Recycling techniques
should respond to the
increasing complexity in
the composition of
building materials and
structures.
Complexity
Quality
Reuseability
The quality of recycled

materials determines their
suitability for further use.
Improvements can be
anticipated through the
application of appropriate
sorting techniques. Example:
Elimination of impurities,
installation of control systems
New fields of recycling need to be
opened up. Consideration of the
chemical-mineralogical composition
of CDW offers new possibilities.
Example: Closed material cycles
as option for a better recovery and increasing recycling ratio

against the state of the art of separation just by particle size, density and color
Figure 2: Importance of (new) sorting techniques

Sorting can occur according to the composition of the primary material:
Materials recycling: The material is transformed in its particle size to recycle it in their
original or similar purpose (e.g. closed-material cycles for each material)
- Raw material or feedstock recycling: The sorted material will be further processed in a
chemical or thermal process in which the material characteristics are converted in a new
product. One example could be the production of lightweight aggregates, where the
mixed recycled material must have an average chemical composition: [3].
-

Figure 3 illustrates how C&DW could be differentiated according to the primary materials
composition. In this case the ternary system SiO2 CaO Other oxides and loss of ignition
(1025 C) is shown.
Non-conventional sorting techniques can be used. One technology is presented in the
following chapters. Each one has advantages and disadvantages that have to be evaluated.
Table 1: Chemical composition of selected main masonry building materials and roof tiles
from milled samples
Main oxides from X-ray fluorescence spectroscopy [M.-%]
Masonry building materials
SiO2
CaO Al2O3 Fe2O3 Na2O K2O
SO3
Min
66.6
0.8
14.4
1.2
0.2
1.6
0.0
Facing brick
Max 74.9
0.9
22.5
6.8
1.6
3.5
0.1
Min
48.6
2.6
15.6
4.4
0.7
2.4
0.2
Vertical cored clay
bricks
Max 56.9
19.2 22.7
7.9
2.3
4.0
1.1
Min
58.2
0.4
16.9
4.2
1.4
1.8
0.0
Roof tiles
Max 73.6
1.9
22.9
8.2
2.3
5.0
0.3
Min
57.5
7.9
1.2
0.2
0.4
0.2
0.1
Sand-lime brick
Max 83.0
23.3 9.5
1.2
1.0
4.8
0.3
44.1
27.8 1.2
0.6
0.3
0.3
0.5
Autoclaved aerated Min
concrete
Max 58.7
36.3 3.9
1.2
0.3
0.8
2.5
Min
39.5
14.1 14.1
2.4
2.6
2.3
0.9
Lightweight
concrete
Max 58.7
33.7 16.7
3.7
3.5
4.1
2.5
Min
1.2
36.6 0.3
0.1
0.0
0.1
46.5
Gypsum plaster
Max 11.6
45.8 0.4
0.3
0.0
0.4
56.8
3.
EXAMPLE - SENSOR-BASED SORTING BY NEAR-INFRARED
One of the most promising techniques is sensor-based sorting technology. Through the
installation of such sensor-based single particle sorting devices in building materials recycling
heterogeneous bulk materials could be converted into homogeneous material fractions. The
performance limitations of the prevalent sensor techniques, as well as economic aspects, must
be considered. At the present level of technology, only coarser aggregates can be technically
and economically sorted by this technique.
Figure 4 shows the functional principle of a linescan (pushbroom) configuration of a nearinfrared sorting system. The particles to be sorted are excited on a conveyor belt to emit
electro-magnetic spectra that are measured by a sensor system above. This radiation in the
near-infrared range contains information about the chemical-mineralogical composition of the
material. Through the use of chemo-metrical methods the materials can be distinguished even
if their spectra only differ slightly. There are overlaps of each individual constituent and an
overall spectrum is emitted.

Vertical cored clay bricks
SiO2
Facing clay bricks

Roof tiles
Sand-lime brick
Autoclaved aerated concrete
Lightweight concrete
Gypsum plaster
Some concrete blocks
CaO 100
Other oxydes and

loss of ignition 1025 C
CaO [mass-%]
Figure 3: Main masonry building materials by the average composition (X-ray fluorescence
spectroscopy on milled samples) in a ternary system
Focal plane
detector
(2-dimensional)
Spectrograph
Hyperspectral
image
(3-dimensional)
Control system
(Software)
Dispersed light
Chemometrics
PLS
Emitter of
electromagnetic
radiation
Slit
Feeder
Conveyor
movement
(turning 2D into 3D)
Diffuse reflectance of
electromagnetic radiation
Conveyor
Good
material
Bad
material
Inspired by:
James E. Burger: Hyperspectral NIR Image Analysis, Doctoral Thesis 2006, Swedish University of Agricultural Sciences
Figure 4: Near-infrared sorting of a linescan (pushbroom) configuration [2]

Intensity
Investigations in a German project showed that mineral C&DW can be principally differed
in the near-infrared region (Figure 5). Sand-lime brick, autoclaved aerated concrete, gypsum
plasters, and most concretes, among other examples, emit clear spectra. The only exceptions
are clay bricks where only special types show measurable spectra. The circumstances have
not yet been definitively clarified. The recognition of gypsum plasters is very good. Similar
investigations have shown that the removal of gypsum-based materials from the valuable
fraction is possible by near-infrared, also organic material: [4].
1255
1347
1439
1532
1624
1716
1809
Wavelength [nm]
1901
1993
2085
2178
Figure 5: Spectra from main building materials

Cameras with a high spatial resolution are needed for sorting mineral particles and
recognizing composite particles. The surface is scanned and the spectrum for each spot is
measured (Figure 6). Near-infrared is a surface near measurement method. For this reason, the
liberation ratio (appearance of composite particles) and the texture influence directly the
recognition and sorting quality. Also the dust adhesion from processing has a significant
influence on the sorting quality.
The near-infrared region is also sensitive to the moisture content of the samples, especially
in the near-surface area. For this reason, the chemical-mineralogical composition on the
surface is of importance. From a wavelength of 1.45 m the water absorption increases
significantly (IR-B region 1.4 3.0 m) and also affects the spectra reflected of the
intrinsically material.

How this technology can improve the quality of recycled materials in industrial scale is
still subject to research. There are also other sensor technologies that can help to sort mineral
C&DW, such as laser-induced breakdown spectroscopy (LIBS).
Figure 6: Measurement of spectra from particle surface on a conveyor belt [2]
4.
CONCLUSIONS
Building materials are becoming more and more complex. Therefore, future C&DW
will change in their composition and will be subject to higher variability. To what extent
this will influence recyclability has not been sufficiently answered yet and requires
more research. Challenges in future will be in particular: High gypsum plaster inputs in
building structures, a decrease in particle density and compressive strength as well as an
increase of the water adsorption, as side effect to stricter insulation regulations.
Increasing demand for closed-material cycles in the mineral construction industry
requires homogeneous recycled materials with consistent product qualities. There is a
need for innovative techniques for comminution liberation and sorting for separation of
impurities from the usable material fractions.
The innovation and integration of new liberation and sorting techniques in industrial
scale should go hand in hand, especially for technological sensibility. Processing of
mineral C&DW is complex and requires sophisticated concepts that also integrate
sustainable and economic aspects.
One option for an improved quality and to remove impurities is to apply near-infraredbased sorting systems. Most mineral main building materials emit spectra in the nearinfrared region. There are other technologies that can be used in the building materials
industry. There is a need of appropriate sorting techniques for recycled sands.

ACKNOWLEDGEMENTS
The results were derived from several German research projects, supported through the
AiF within the frameworks of Industrial Collective Research (IGF) and Central Innovation
Programme Mittelstand (ZIM), run by Germanys Federal Ministry of Economic Affairs and
Technology based on a decision of the German parliament (Bundestag): 18101 BG (IGF)
between IZF - Institut fr Ziegelforschung Essen e.V. and IAB - Institut fr Angewandte
Bauforschung Weimar gemeinntzige GmbH, 16617 BG (IGF) between Bauhaus-Universitt
Weimar and IAB - Institut fr Angewandte Bauforschung Weimar gemeinntzige GmbH,
KF3033025 (ZIM) between Bauhaus-Universitt Weimar, LLA Instruments GmbH, T.B.R.
Teltower Baustoffrecycling GmbH, Sesotech and IAB - Institut fr Angewandte
Bauforschung Weimar gemeinntzige GmbH.
REFERENCES
[1] Landmann, M., Mller, A., et al., Efficiency of processing methods for the recovery of unmixed
material fractions from masonry Part I and II, in AT MINERAL PROCESSING EUROPE 03
and 04 (2014) (Volume 55).
[2] Landmann, M., Mller, A., Karrasch, A., Lin, E., Ulsen, C., Chemical-mineralogical sorting of
mineral construction and demolition waste, in Minerals in Circular Economy Book of Abstracts
(VTT Technology 192) of the 1st Int. Conf. on Minerals in the Circular Economy, Espoo,
Finland, 26-27 November 2014.
[3] Mller, A., The manufacture of Lightweight Aggregates from Recycled Masonry Rubble, in
Mll und Abfall 11 (2014) 625-633.
[4] Vegas, I., et al., Upgrading the quality of mixed recycled aggregates from construction and
demolition waste by using near-infrared sorting technology, in Construction and Building
Materials 75 (2015) 121-128.

USE OF FINE RECYCLED AGGREGATE AS SUPPLEMENTARY

CEMENTITIOUS MATERIAL
M.E. Sosa (1), C.J. Zega (1,2), G.B Coelho Dos Santos (1), A.A. Di Maio (1,2)
(1) Laboratorio de Entrenamiento Multidisciplinario para la Investigacin Tecnolgica
(LEMIT-CIC) Calle 52 e/121 y 122 (1900) La Plata. Buenos Aires. Argentina.
(2) Consejo Nacional de Investigaciones Cientficas y Tcnicas (CONICET). Argentina.
Abstract
Concrete wastes occupy up to 50% of overall construction waste. The use of those as
coarse aggregate in new concrete production is a topic widely studied since some decades.
The use of fine recycled aggregate (FRA) began to be important in last decades, due to a
shortage of natural sand suitable for the concrete industry. In several studies, the use of FRA
in concrete has been evaluated from different aspects and contradictory results can be found
about its influence in fresh and hardened state properties. However, most of the researches
agree with the fact taht the finest fraction of this waste (sizes smaller than 2 mm) must be
discarded, considering that it is the most damaging for the properties of concrete.
Nonetheless, recent studies concluded that the use of water-reducing admixtures could rectify
this behaviour. Other authors found that it can be used as filler in self-compacting concrete,
improving fresh state without prejudice to the mechanical performance. In this work a
measurement of cementing properties by means of strength activity index of the smaller
fraction of recycled aggregate obtained from crushing concrete as well as its action as filler is
evaluated. From the results obtained it appears that the use of this fraction as a supplementary
cementitious material is feasible.
Keywords: fine recycled aggregate; cementitious material; filler action.
1. INTRODUCTION
Fine recycled aggregates (FRA) present attached mortar in different proportions in its
composition, which may be considered as the main difference compared to fine natural
aggregate. Thus, a particle of FRA may be composed of natural aggregate, mortar, and/or a
combination thereof. The mortar attached to the FRA, owing to its high porosity, is
responsible for the increased absorption respect to natural aggregate [1-3]. Due to this, it is
expected that different properties of both fresh and hardened state of concrete can be altered
when such aggregates are used as replace for the fine natural aggregates (FNA). In this sense,
contradictory results can be found regarding to the influence of FRA in different properties of
concrete, such as consistency, mechanical behaviour and durability [4-7]. However, most of

them agree about the properties such as static and dynamic modulus of elasticity, and
ultrasonic pulse velocity, which are diminished by using FRA even in small percentages.
Additionally, Hansen [8] concludes that the use of the FRA is technically and
economically feasible, if the fraction smaller than 2 mm is discarded. According to him, the
use of this fraction causes increase of drying shrinkage, decrease in compressive strength and
impairs substantially the workability of the mixtures. About this, Selmo and Miranda [9]
found that the amount of particles smaller than 75 m have a great influence on all the
properties of mortars. In dry blends the authors concludes that the amount of this material
(smaller than 75 microns) is better to indicate shrinkage behaviour compared to the
water/cement ratio (w/c).
On the other hand, different authors [10-12] have suggested that in the smaller fraction
than 2 mm of FRA are unhydrated cement particles, which can supply some additional
product of C-S-H. Regarding to this, Arm [13] found that unbound layer for pavement
structure made with aggregate from crushed concrete (without any binder material) showed a
greater stiffness than identical layers made from natural crushed aggregates, under identical
conditions, and that the stiffness increased significantly with the age at which was evaluated.
Poon et al [14] found that fractions smaller than 0.15 mm of FRA had C2S in their
composition, which may be the cause of the cementitious properties. Although this
component was not found in the fraction between 0.30 mm and 0.60 mm, they also showed
cementing capacity.
The aim of this study is to evaluate the cementing capacity and filler effect of the fractions
of FRA powder. For that, the smaller milled materials than 600 m and 75 m were analysed.
Both fractions were evaluated by means of strength activity index (SAI), the mortars used in
SAI were evaluated in fresh and hardened state, by means of properties such as consistency,
unit weight, density and water absorption. Additionally, petrographic analysis by means of
microscopy on thin section was done.
2.
2.1
EXPERIENCES
Materials and Methodology
Recycled fine aggregate (FRA) were obtained by crushing concrete with a resistant level of
35 MPa and with crushed granite stone as coarse aggregate. Fraction smaller than 600 m
(P600) was milled by means of a laboratory ball mill, while the fraction smaller than 75 m
(P75) was used without any processing. This materials, also are designated as powder of FRA
Table 1 shows the density and the Blaine fineness of powder of FRA (P600 and P75) and
ordinary Portland cement (OPC) used as reference, with dv values 10, 50 and 90 and their
respective standard deviations (dv represents the maximum particle size below which is 10%,
50% and 90% of the sample found, respectively). The average size distribution of materials
P600, P75 and OPC, evaluated by laser diffractometer CILA 990 is shown in Figure 1.
Figure 2 shows the XRD analysis of the studied materials, P600 and P75. It can be
identified different components, some of which correspond to the original natural aggregate,
quartz (Q) and feldspar (F), and other connected with the attached mortar, calcite (C) and
portlandite (CH). Any evidence about the presence of unhydrated cement particles (C2S)
could be observed.

Table 1. Physical properties of fines

Blaine
Density
dv10
Materials
fineness
3
(g/cm )
(m)
(m2/kg)
1.91
OPC
3.00
323
(0.08)
0.45
P600
2.70
11430
(0.02)
0.57
P75
2.60
269
(0.02)
dv50
(m)
dv90
(m)
16.83
(0.09)
9.68
(0.65)
32.18
(0.73)
49.54
(0.53)
53.94
(4.11)
72.04
(1.93)
Cumulative Values (%)
100
90
80
70
60
50
40
30
20
10
0
0,1
P600
10
Diameter (m)
P75
100
1000
OPC
Figure 1. Size distribution

P600
P75
Figure 2. XRD analysis

Figure 3 shows the microscopic analysis of the powder of FRA and the OPC. It can be
observed that the P600 and P75 particles are more angular and flaky than those of OPC,
which presents rounded shape.
Figure 3. Microscopic image

Table 2 shows the mix proportions of mortars with OPC, P600 and P75 (MOPC, MP600
and MP75 respectively) made in order to determine the strength activity index (SAI),
according to IRAM 1654 [15]. This method involves the comparison of the compressive
strength between a reference mortar, made with ordinary portland cement, and a mortar made
with partial replacement of cement (30% by volume) with the material on study. The SAI is
considered positive when, at the age of 28 days, compressive strength of studied mortars is
greater than or equal to 75% that of reference mortars.
Additionally, mortars were evaluated by means of flow table (according to ASTM C1437)
and unit weight in fresh state, while density, water absorption and petrographic analysis by
optical microscope were determined in hardened state.
Table 2. Mix proportions (g)
Materials
MOPC
MP600
MP75
Water
Cement
water/binder ratio
P600
P75
Sand
225
500
225
325
225
325
----1500
152
--1500
--146
1500
3.
The properties determined at fresh and hardened state of mortars are summarized in
Table 3. It can be seen that the flow is similar for MOPC and MP600 and 50% lower for
MP75. This fact can be attributed to a more flaky, angular and coarser particles of P75 (see
figure 3). Similar flow values between MP600 and MOPC mortars, can be due to a filler
effect as result of particle size of PM60 (see figure 2), which may prevent flocculation of the
cement particles.
Mortars with FRA powders and OPC presents similar density, while absorption is
approximately 17% higher for mortas madewith FRA powders, which could be as
consequence of increased porosity of the matrix as result of the lower hydration products in
MP600 and MP75.
Table3. Fresh and hardened state properties
Flow Unit weight
Density
Absorption
Mortar
(%)
(Kg/dm3)
(g/cm3)
(%)
MOPC
69
2430
2.26
8.25
2.23
9.59
MP600
63
2455
2.23
9.75
MP75
35
2370
Table 4 shows compressive and flexural strengths and the SAI at 28 and 90 days. Despite
that at both ages the SAI is negative, a further increase in the compressive strength of mortars
with the FRA powder compared to the MOPC can be noted. This increment is in compressive
strenght around of 12% for MP600 and MP75 and 4% for MOPC. Similar behaviour can be

observed in flexural strength, where the increment is around of 33%, 16% and 10% for
MP600, MP 75 and MOPC, respectively. These higher increments in MP600 and MP75
respect to MOPC, although the lesser amount of cement, might indicate a slow cementing
properties of the powder of FRA used. Assessments at older ages are necessary to confirm
this fact.
Table 4. Mechanical properties and SAI
Compressive
Flexural strength
strength (MPa)
(MPa)
Mortar
28 days 90 days 28 days 90 days
MOPC
49.8
51.6
6.3
6.9
MP600
31.8
35.8
4.3
5.8
MP75
27.6
30.9
4.0
4.6
SAI (%)
28 days
100
65
55
90 days
100
69
60
In order to establish qualitative differences in the composition of cement mortars and in the
textural characteristics of interface areas of each mortar, a petrographic analysis was
performed on thin sections. The samples are shown in Figure 5.
Figure 5. Petrographic analysis. Cement paste (cp); aggregate (A); remnant grain cement
clinker; cement paste with calcium hydroxide (cp-CH); grains with carbonation (G-cc).
From Figure 5, the following characteristics could be established:
MOPC: cement paste in good condition was observed, consistent with small amount of
pores (<5%) with circular morphology and sizes smaller than 1 mm. A moderate proportion
of unhydrated cement particles (<20%) (sizes from 20 m to 100 m) uniformly distributed
throughout the paste was found. Crystals of calcium hydroxide with sizes smaller than 10
microns, in moderate proportion were identified. The interfacial transition zones (ITZ) are
shown in good condition, cohesive and with a great area of portlandite (Ca2OH).

MP75: Finely carbonated angular particles were observed with sizes between 40 m and
60 m (possibly mortar attached to FRA). Carbonation observed on these particles extends
into the paste, forming a halo of carbonation. Unhydrated Cement particles (sizes smaller than
100 m) and calcium hydroxide crystals (smaller than 10 m) were also observed in
approximately the same proportion as that for reference mortar (MOPC). A higher number of
pores regarding to the MOPC was identified, particularly in areas of ITZ, however not
exceeding a 10%; these latter present irregular morphology and sizes smaller than 1 mm.
MP600: A higher amount of calcium hydroxide crystals (sizes smaller than 10 m)
regarding to MP75 and MOPC were found, particularly in the ITZ. The porosity of the paste
is similar to that of MP75, with diameters of 1 mm. Unhydrated cement particles were
identified (sizes smaller than 200 m) in smaller amount than for MOPC and MP75. Particles
of irregular morphology finely carbonated and with fuzzy edges were also found.
4.
CONCLUSIONS
From the evaluation of the cementing properties and the filler effect of two fractions of
fine recycled aggregate (FRA) (material smaller than 600 m and 75 m), by means of the
strength activity index (SAI), regarding to a reference mortar (MOPC) made with ordinary
portland cement, the following can be stated:
Neither fractions of the powders of FRA (P600 and P75) evidence unhydrated cement
particles in their composition, from the XRD evaluation.
The flow of the mortar MP600 was similar to that of MOPC due to a filler effect of the
material P600, and both were 50% higher than that of MP75, which was attributed to a
different shape of the particles of P75.
Density of all mortars was similar, while absorption of MP600 and MP75 was increased
by 17% with respect to MOPC. This fact suggests a higher porosity in recycled mortars,
which was verified by the petrographic analysis.
The strength activity index (SAI) determined at 28 and 90 days was negative, but the
increase in compressive and flexural strength at 90 days was substantially higher in
mortars made with P600 and P75.
Although the SAI was negative, the increment in the mechanical strengths could suggest
a slow cementing reaction of FRA.
From the results presented in this paper, additional analysis at older ages are needed in
order to prove the cementing capacity of the smaller sizes of FRA.
ACKNOWLEDGEMENTS
The authors wish to thank the financial support by the National Technological University,
Argentina, and the collaboration of LEMaC (Road Researches Centre, FRLP-UTN) and the
National University of the Buenos Aires Province for the tests carry out.
REFERENCES
[1] Zega, C.J.; Sosa, M.E.; Di Maio, A.A. Propiedades de los agregados reciclados procedentes de
hormigones con diferentes tipos de agregados gruesos naturales. 18 Reunin Tcnica de la
Asociacin Argentina de Tecnologa del Hormigon. (2010) 33-38

[2] Sosa, M.E.; Zega, C.J.; Di Maio, A.A. y Peralta, J.P., Evaluacin de morteros con agregado fino
reciclado en reemplazo de arena natural de trituracin.19Reunion Tcnica de la Asociacin
Argentina de Tecnologa del Hormign. (2012) 415-422.
[3] Hincapi Henao, A.M y Aguja Lpez, E.A. (2003). Agregado reciclado para morteros. Revista
Universidad Eafit. 132 (39) 76-89.
[4] Dhir, R. K.; Limbachiya, M. C. y Leelawat, T. "Suitability of recycled concrete aggregate for use
in BS 5328 Designated mixes." Proceedings of the ICE - Structures and Buildings, 134 (3) (1999).
257-274.
[5] Zega, C.J., Di Maio, A.A., Use of recycled fine aggregate in concretes with durable
requirements, Waste Management, 31, (2011) 2336-2340.
[6] Masood, A., Ahmad, T., Arif, M., Mahdi, F., Waste management strategies for concrete,
Environmental Engineering and Policy, 3, (2002) 15-18.
[7] de Brito, J.; Pereira, A. S. y Correia, J. R.. "Mechanical behaviour of non-structural concrete
made with recycled ceramic aggregates." Cement and Concrete Composites, 27 (4) (2005) 429433.
[8] Hansen, T. C. "Recycled aggregate and recycled aggregate concrete. Second state-of-the-art.
Report developments." Materials and Structures, 19 (3) (1986) 1845-1985.
[9] Miranda, L.F.R; Selmo, S.M.S. CDW Recycled aggregate renderings: Part I- Analysis of the
effect of materials finer than 75 m on mortar properties Construction and Building Materials,
20 (2006) 615-624.
[10] Khoshkenari, A.G.; Shafigh, P.; Moghimi, M.; Mahmud, H.B., "The role of 0-2 mm fine recycled
concrete aggregate on the compressive and splitting tensile strengths of recycled concrete
aggregate concrete." Materials and Design. 64 (2014) 345-354.
[11] Katz, Ammon. Properties of concrete made with recycled aggregate from partially hydrated old
concrete Cement and concrete research, 33 (2003) 703-711.
[12] Khatib, J.M. Properties of concrete incorporating fine recycled aggregate. Cement and Concrete
Research. 35 (4) (2005), 763-769.
[13] Arm Mara. Self-Cementing properties of crushed demolished concrete in unbound layers:
results from triaxial test and field test. Waste Management 21 (2001) 235-239.
[14] Poon, Chi Sun; Dixon Chan, Qiao. The Cause and influence of self-cementing properties of fine
recycled concrete aggregates on the properties of unbound sub-base Waste Management 26
(2006) 1166-1172.
[15] IRAM 1664. Puzolanas. Mtodos de ensayos generales. Instituto Argentino de Normalizacin Y
Certificacin. Argentina

PROPERTIES OF LIGHTWEIGHT CONCRETE MANUFACTURED

WITH RECYCLED MIXED AGGREGATES
Agrela F. (1), Rosales J. (1), Beltrn M.G. (1), Cabrera M. (1), Ayuso J. (1)
(1) Area of Construction Engineering, University of Cordoba, Cordoba, Spain
Abstract
The use of recycled aggregates in the manufacture of concrete is increasing significantly mainly
due to the need for an environmental solution to this type of material. Therefore, numerous
investigations have been performed to study the mechanical properties and durability in concrete made
with recycled aggregates from different origins (concrete or mixed C&DW). However, the use of
these materials has not been very much studied in their application in lightweight concrete. Due to
the current increasing trend in the use of lightweight concrete for producing pre-manufactured
elements in the building sector, it would be interesting to study the mechanical and durability
properties of lightweight concrete manufactured with recycled mixed aggregates. Therefore, in this
work, a study of this type of concrete has been made with recycled mixed and concrete aggregates to
evaluate the mechanical properties.
Keywords: Lightweight concrete; Recycled aggregates, Durability; Mechanical Behavior

1.
INTRODUCTION
The study and use of recycled aggregates (RA) from construction waste and demolition
(C&DW) or other sources is on the rise due to a raising awareness on environmental and new
environmental policies. Internationally, the construction and building industries generate a
substantial waste stream including materials from brick and concrete demolition. The use of
these RA is promoted, as well as causing a profit to the environment, reducing the amount of
natural aggregates from quarries and river beds, which are used in the production of concrete
and other applications [1].
In recent years, several studies have been performed on the concrete application, substituting
natural aggregates by recycled concrete aggregate (RCA), to verify their mechanical
properties and durability [2]. There are many research focused on the influence of RCA and
mixed (RMA) as replacement coarse fraction for the manufacture of concrete [3, 4].
Alternatively, International union of laboratories and experts in construction materials,
systems and structures (RILEM), in its report containing recommendations for coarse
recycled aggregates, included RMA in a mix of crushed masonry and concretes for the
manufacture of new concretes [5].
However, the use of RMA from crushed bricks and masonry is increasingly being
investigated for their use in concrete and other construction elements. Other studies analyzed
the behaviour of non-structural concrete made with different amounts of RMA, and the
physical and mechanical properties of friction concretes made with RMA [4, 6].

Fewer studies exist for the application of RA in the manufacture of lightweight concretes
(LWC). The low density and high thermo-insulating capacity of LWC are its most distinct
characteristics in comparison to normal-weight concrete [7].
The properties of the LWC mainly depend on the amount and the properties of the lightweight
aggregates used. Thus, number of studies has been conducted to examine how the aggregate
properties affect the mechanical properties of lightweight aggregate concrete [8].
As discussed above, RMA present lowest density values [9] so the possibility of making
lightweight concrete by applying a substitution of natural aggregate coarse fraction by this
type of RA, may lead to a decrease of the density in LWC.
In this study, RMA was studied for its application in several substitutions natural aggregates
in the manufacture of non-structural lightweight recycled concretes. The characterisation
properties of RMA were determined and then applied in the manufacture of lightweight
recycled concrete. Additionally, physical, mechanical (compressive and flexural strength) and
durability properties (chloride and water penetration under pressure, shrinkage) of lightweight
recycled concretes with different replacement percentages of natural aggregates were studied.
2.
2.1.
Materials
2.1.1 Cement
The cement used in the study was Portland cement with limestone with high initial strength,
CEM II / AL type medium strength 42.5 MPa at 28 days. All the chemical properties are
summarised in Table 1.
Table 1: Properties of cement
SiO2
Al2O3
Fe2O3
%
%
%
18.22
4.27
3.06
2.1.2 Recycled mixed aggregates
CaO
MgO
SO3
K2O
Na2O
%
62.21
%
0.9
%
3.17
%
0.70
%
0.26
Recycled mixed aggregate (RMA) from the crushing of concrete blocks and masonry blocks
from the treatment of CDW in Gecorsa Plant (Crdoba) were used in this research whose
properties are summarised in Table 2 and particle-size distribution is shown in Fig. 1.
2.1.3 Natural aggregates
Two types of natural aggregates were used in the manufacture of concrete: natural coarse
gravel (NG), and natural sand (NS). The properties of the aggregates are shown in Table
2 and their particle-size distribution is shown in Fig. 1.
Table 2: Properties of recycled and natural aggregates
Apparent density (Kg/dm3)

Back to dry density on stove (Kg/dm3)
Saturated density. Dry surface area (Kg/dm3)
Absorption (%)
Plasticity index
NG
2.68
2.61
2.67
1.53
NP
RMA
2.51
2.24
2.39
6.94
NP
NS
2.61
2.5
2.59
0.81
12

Figure 1: Particle size distribution curves

2.1.4 Expanded clay
Expanded clay (EC) is a brown lightweight aggregate with a nominal particle size of 3mm to
8mm. The bulk density of expanded clay is 0.6 kg/dm3.
2.1.5 Water-reducing admixture
A super-plasticiser additive (SP), type BASF Rheobuild 1222, was used. The objective was to
reduce the mixing water, improving the mechanical strength in the short term by increasing
the durability and allowing the concrete to be manufactured with a low effective water/cement
ratio (w/c) because of its high water reduction.
2.2. Concrete mix proportions
For the study of the properties of LWC, six batches were made by applying different
percentage of recycled aggregates substitutions (Table 3).
The first batch, called "CONV." was made with natural aggregates in order to be the reference
concrete, not these being classified as lightweight aggregate.
The control lightweight concrete (CONTROL) has been manufactured applying NG, NS and
EC, to lighten up in its fraction 3-8mm.
Three batches were made applying identical amount of EC and 25%, 50%, 75% and 100%, as
well as applying substitutions of GN by RMA, according to the data of table 3.
Table 3: Concrete mix proportions (kg/m3)
NG
NS
RMA
EC
CEM
WATER
TOTAL
CONV
857.6
1039.1
0
0
300
180
2376.7
CONTROL
494.2
710.3
0
108.7
300
180
1793.2
25RMA
370.65
710.3
108.8
108.7
300
180
1778.5
50RMA
247.1
710.3
217.8
108.7
300
180
1763.9
75RMA
123.55
710.3
326.6
108.7
300
180
1749.1
100RMA
0
710.3
435.5
108.7
300
180
1734.5
The effective w/c ratio was 0.6 for all cases. Similarly, the amount of cement (300 kg/m3) and
slump (89 cm) were the same for all manufactured concretes. The Bolomey method was
used to calculate the mixture proportions shown in Table 3.
The total water shown in this table corresponds to the mix water calculated in 300 kg/m3 of
cement plus the saturation water of the recycled mixed aggregates (RMA), and natural

aggregates (NG and NS). This amount decreased depending on the moisture present in each
of the aggregates.
2.3.
Experimental methods
After exposure to physical characterization, method of manufacture and experimental

methodology of lightweight concrete cured are shown here.
2.3.1 Compressive strength
UNE EN 12390-3 specifies a test method for determining the compression strength of test
specimens of hardened concrete. The compressive strength of the hardened concrete was
determined at the ages of 7 and 28 days on 100 mm cubes. These cubes were removed from
the moulds after 1 day and were cured in moist chamber before testing.
2.3.2 Flexural strength
Flexural strength was determined using prismatic specimens with the dimensions
100 100 400 mm3 cured during 28 days. This property was determined according to
UNE-EN 12390-5.
2.3.3 Penetration of water under pressure
To determine the penetration of water in concrete, water was applied to the cylindrical
specimens at 5 atmospheres of pressure according to UNE-EN 12390-8 at 28 days.
2.3.4 Drying shrinkage test
To study the durability of the concrete, drying shrinkage measurements were obtained on the
concrete prisms measuring 100 100 500 mm, according to ASTM C157. The specimens
were exposed to conditions of 50% relative humidity and 20 C and the measurements were
taken for 1, 4, 7, 14, 28, 56, 90, 180 and 360 days.
3.
3.1 Physical and mechanical results.

3.1.1 Compressive strength
For all ages at break, CONVENTIONAL concrete presented the highest values of
compressive strength. LWC with addition of expanded clay (series EC) showed less
resistance, being a lower resistance than in mixtures where substitutions of RMA were
applied. Resistances obtained in recycled LWC (R-LWC) were significantly lower compared
to conventional concrete, in comparison with mixtures where 25%, 50%, 75% and 100%
substitutions of NG by RMA were applied. It was observed a decrease of resistance in
concrete with natural aggregates and incorporation of RMA respect to CONTROL (Fig.2).
Strength loss was higher with the incorporation of RMA (1.9%, 4.1%, 16.8% and 25%
respectively for replacement ratios 25%, 50%, 75% and 100% GN by RMA).

Figure 2: Compressive strength at 28 days and evolution of lightweight concrete

An increase is observed according to the curing age (Fig. 2), as expected by Kou et al. [10].
Similarly, approximately 75% of the compressive strength at 28 days was obtained at only 7
days for the entire series, as obtained by Pereira et al. [11].
3.1.2. Flexural strength
Referring to the flexural strength, the CONVENTIONAL concrete presented the highest value
(Fig. 3). For LWC down resistance, the higher the amount of added ARM. In the results of
flexural strength (Fig. 3), it can be possible to observe that the presence of RMA in LWC
produces an inferior loss of resistance than in compressive strength behaviour.
Figure 3: Flexural strength and correlation between compressive and flexural strength
LWC with 25% of RMA (serie 25RMA) presented similar resistances than in those where CE
was applied ( with 0% of RMA - CONTROL), indicating that the replacement of 25, 50% and
75% of RMA in lightweight concrete produced a similar behaviour to LWC with CE and NA
(0.5%, 2% and 5.7% in flexural respectively). LWC with 100% of RMA substitution, the
flexural strength loss was more significant, over 9%.
Fig. 3 shows the excellent correlation between the compressive strength and flexural strength
for all concretes (R2 = 0.98). Compressive strength can reliably determine the flexural
strength for each concrete and vice versa.
3.1.3. Penetration of water under pressure
Water penetration increased significantly with the addition of CE and substitutions of natural
aggregates by recycled aggregates (Table 4). This progressive increase of concrete with
recycled aggregates is observed, this growth was greater with greater the amount of ARM,
mainly due to the high porosity of recycled aggregates.

Table 4: Durability properties of hardened concrete.

AGE
(Days)
CONV CONTROL 25RMA 50RMA 75RMA 100RMA
Water Penetration
28
44.6
65.7
70.6
72.9
77.2
81.7
(mm)
Chloride
28
19.3
24.9
30.4
35.1
44.7
48.4
Penetration (mm)
Shrinkage
90
380
434
450
479
485
501
(mm/m)
As it happened with the absorption, water penetration increased significantly by introducing
EC and RMA, due to the high absorption capacity of RMA, greater with larger amount of
RMA added (7.5%, 11%, 17.5% and 24.4% for series 25RMA, 50RMA, respectively 75RMA
and 100RMA.
Mas et al. [4] concluded that the penetration of water under pressure increased upon the
addition of mixed recycled aggregates with a high linear proportionality. This is inconsistent
with the work of Martinez-Lage et al. [12], where water penetration was less with the
incorporation of RMA.
3.6. Drying shrinkage test
To study the durability of the concrete, drying shrinkage measurements were obtained on the
concrete prisms measuring 100 100 500 mm, according to ASTM C157. The specimens
were exposed to conditions of 50% relative humidity and 20 C and the measurements were
taken for 1, 4, 7, 14, 28, 56, 90, 180 and 360 days.
Shrinkage to different age was reflected in Fig. 6, where higher values for concrete shrinkage
are observed with RMA. It is observed that from 7 days old concrete with ARM obtained
significant differences from CONTROL, most notable difference still 28 days old. On the
other hand, 56 days old stabilization observed in increased shrinkage, increasing from that
date less than in previous ages.
Figure 6: Evolution of shrinkage with respect to reference concrete

As shown in Figure 6, 90 days shrinkage has progressively increased by introducing ARM.
For 25RMA, 50RMA, 75RMA and 100RMA series, increased retraction from the CONTROL
was 3.7%, 10.4%, 11.8% and 15.4% respectively.

4.
CONCLUSIONS
The following conclusions were obtained:

- The addition of EC and RMA allows to obtain lightweight concrete (LWC) but the
mechanical properties of lightweight concrete decreases over conventional concrete up to
25% to 100% additions.
- The density decreases significantly with the addition of recycled aggregates, RMA, and the
lightening aggregates (CE), due to the lower density of particles with respect to natural
aggregates. This decrease in density is required to keep the LWC category.
- Therefore, the ARM incorporation is beneficial for obtaining LWC. Moreover, a lower
density for the same amount results in decreased weight as pores. Therefore, the introduction
of aggregates with a high porosity improves significantly the properties of this type of
concrete.
- Water penetration, penetration of chlorides and shrinkage get worst with the incorporation of
RMA, in comparison to those where Natural Aggregates are applied.
In summary, the replacement of NG by RMA adversely affects the mechanical properties and
durability in the manufacture of LWC, and a higher density is obtained. Therefore, it is
advisable to use recycled LWC as both RMA at smaller replacement rates in non-structural
lightweight concrete which do not require high technical exigencies.
REFERENCES
[1] Chini AR, Kuo SS, Armaghani JM, Duxbury JP. Test of recycled concrete aggregate in accelerated
test track. J Transp Eng 2001;127(6):48692.
[2] Wai Hoe Kwan, Mahyuddin Ramli , Kenn Jhun Kam , Mohd Zailan Sulieman, 2012. Influence of
the amount of recycled coarse aggregate in concrete design and durability properties. Construction
and Building Material, 26, 575-573
[3] Etxeberria, M.; Vazquez, E.; Mar, A. y Barra, M., 2006. Influence of amount of recycled coarse
aggregates and production process on properties of recycled aggregate concrete. Cement and
Concrete Research 37 (5), 735-742.
[4] Mas, B.; Cladera, A.; Del Olmo, T. y Pitarch, F., 2012. Influence of the amount of mixed recycled
aggregates on the properties of concrete for non-structural use. Construction and Building Materials
27, 612-622.
[5] RILEM 121-DRG-specifications for concrete with recycled aggregates Materials and Structures,
27 (1994), pp. 557559
[6] I. Martnez-Lage, F. Martnez-Abella, C. Vzquez-Herrero, JL Prez-Ordez Propiedades de
hormign en masa hechas con mezcla de rido grueso reciclado Constr Mater Generar, 37 (2012), pp.
171-176
[7] Chandra S, Berntsson L.2003. Lightweight aggregate concrete science, technology and
applications. Noyes Publications; 2003.
[8] Erhan Gneyisi, Mehmet Gesolu, Emad Booya, Kasm Mermerda.2014. Strength and
permeability properties of self-compacting concrete with cold bonded fly ash lightweight aggregate.
Construction and Building Materials 74, 17-24.
[9] F. Agrela, M. Snchez de Juan, J. Ayuso, VL Geraldes, JR Jimnez.2011. Limiting properties in
the characterisation of mixed recycled aggregates for use in the manufacture of concrete.
Construction and Building Materials 25, 3950-3955.
[10] S.C. Kou, C.S. Poon, H.W. Wan. 2012. Properties of concrete prepared with low-grade recycled
aggregates. Construction and Building Materials, 36, 881889.
[11] P. Pereira, L. Evangelista, J. De Brito. 2012. The effect of superplasticizers on the workability
and compressive strength of concrete made with fine recycled concrete aggregates. Construction and
Building Materials, 28, 722729.

[12] I. Martinez-Lage, F. Martnez-Abella, C. Vzquez-Herrero, J.L. Prez-Ordez. 2012.

Properties of plain concrete made with mixed recycled coarse aggregate. Construction and Building
Materials, 37, 171176
[13] S. Otsuki, N. Nagasaki, K. Nakashita. 1992.Evaluation of AgNO3 solution spray method for
measurement of chloride penetration into hardened cementitious matrix materials. ACI Mater J, 89.
587592.

IMPROVE THE QUALITY OF RECYCLED AGGREGATE CONCRETE

BY SORTING IN AIR JIG
Carlos H. Sampaio (1), Weslei M. Ambrs (1), Leonardo R. Miranda (2), Gerson L.
Miltzarek (1), Mrcio A. Kronbauer (1)
(1) Department of Metallurgy, Federal University of Rio Grande do Sul, Brazil
(2) Department of Civil Construction, Federal University of Paran, Brazil
Abstract
The goal of this work is investigate the use of air jigging to separate out concrete from
brick and impurities (gypsum, wood and paper). The mixture of aggregates and impurities
were comminuted individually in the size range 12 to 20 mm. The tests were performed with a
base mixture of concrete and brick, in which was added increasing amounts of impurities for
each trial (1%, 2.5% and 5% mass). During the tests, high contents of concrete in concentrates
were reached. The results give good perspectives in relation to industrial application of air
jigging in a recycling plant.
Keywords: recycled aggregates, air jigging, gravity concentration
1.
INTRODUCTION
In some countries, the increase of urbanization level and city rebuilding has results in
greater amounts of construction and demolition wastes (CDW) generated. CDW in the US is
estimated more than 325 million tons generated annually [1], while to Europe around 970
million tons was generated in 2006 [2]. In Brazil, CDW generation was about 77 million tons
in 2013 which represents an increase in 4.1% in relation to 2012 [3].
CDW normally needs to be processed in waste management plants which aim the removal
of contaminants, separation of mixed waste fractions and to obtaining marketable products
[4]. Currently, the most part of recycled aggregates from waste plants are reused as road base
material [5].
In order to be used for concrete production, recycled aggregates must be fully free of
contaminants. Constituents such as plastic, wood, paper, metal and roofing materials are
relatively easy to separate out, but this is not the case for concrete, ceramics and gypsum. The
presence of contaminants produces cement with inferior strength, with cause direct impact in
the cement consumption, in the way that environmental and economic balance of replace
coarse aggregates with recycled concrete aggregates may be negative [6].

Several studies have been done about available techniques to treat CDW [7,8,9]. Among
them, automatic sorting and jigging has given good results in sorting of coarse aggregates.
The use of near-infrared sorting technology can be useful to remove inorganic and organic
contaminants [10]. The wet jigging is an efficient technique to separate concrete from
masonry rubbles, especially for coarse aggregates [11]. However, in the face of economical
and environmental restrictions, the use of dry processes can be more appropriate.
Air jigs can operate at high feed rates with a reasonable performance [12] and can
efficiently beneficiate particles with sizes over than 4 mm [13]. The operation with only
single stage can deliver concrete with more than 90% purity in mass from a mixture of
concrete, brick and gypsum [6].
In order to improve the quality of the current recycled aggregates, this paper investigates
the use of air jig to separate mixtures of conditioned samples of brick, gypsum, concrete,
wood and paper. The process performance is discussed in order to verify the industrial
application in a CDW recycling plant.
2.
EXPERIMENTAL METHOD
The materials used in the tests were prepared by size reduction of separated samples of
concrete, clay bricks, gypsum, wood and paper without the presence of contaminants. The
concrete, brick and gypsum were obtained from concrete laboratory compressive cylinders
(type 30 MPa at 28 days), from solid clay bricks and from solid gypsum blocks, respectively.
The wood was collected from transport pallets while the paper was obtained directly in a
building site and consisting in wastes of drywall panels (Figure 1).
Figure 1: Samples used in jigging tests.

All tests were conducted with particles having a size of 12-20 mm. This particle size is
within the range considered typical for coarse aggregates, that is, 4 to 20 mm. For stony
fraction the skeletal, enveloped and bulk densities were measured (Table 1). To measure the
skeletal density was used a helium pycnometer (Quantachrome) while enveloped density
was measured by water displacement after surface impermeabilization. The bulk density was
determined by weighing the aggregate on a box of known volume and a standard deviation
was estimated for three repetitions.
Table 1: Skeletal, enveloped and bulk density of stony materials.
Material
Concrete
Brick
Gypsum
Skeletal density
(g/cm)
2.67
2.59
2.30
Enveloped
density (g/cm)
2.39
2.26
1.86
Bulk density standard

deviation (g/cm)
1.47 0.0103
1.07 0.0468
0.63 0.0075

The device used to perform the sorting tests was an air jig model Allair S 500 from
AllMineral (Figure 2). The jig is composed by a chamber assembled with two different kinds
of boxes one over the other, in which the particle bed is placed. The volume of each box used
in tests was 50 x 50 x 2.5 cm. The particle bed is subjected to the action of two inputs of air
flow which enters simultaneously by the bottom chamber. The first air flow is continuous and
causes the expansion of particle bed; the second air flow is discontinuous and is responsible
by vibration of particle bed. This continuous cycle of expansion and contraction characterize
the jigging phenomenon and results in a stratification of particle bed with heavier particles
concentrated in inferior layer and the lighter particles concentrated in superior layers.
Figure 2: AllMineral jig, model AllAir S 500, used in the tests.

The operational conditions which are possible to adjust in the air jig are the expansion ratio
and the frequency with which the air flows into the chamber. Beyond that, the time of the
sorting operating can be varied on several time intervals.
The tests were performed aimed to obtain a concrete with high purity from a mixture with
brick and containing other components, which were considered as impurities. A basic mixture
of brick and concrete was contaminated with gypsum, wood and paper. Three impurities
levels were used with concentrations of 1%, 2.5% and 5% in mass in relation to the basic
mixture (Table 2). The proportion in mass used in the contaminant fraction was ever 40%
wood, 40% paper and 20% gypsum. This mass proportion was adopted because consisted in a
similar volume of each contaminant.
Table 2: Mass of each material used in the tests.
Materials (g)
Concrete
Brick
Gypsum
Wood
Paper
Contaminants level
1% impurity 2.5% impurity 5% impurity
18302.1
18312.9
18346.6
13355.3
13456
13623
127.7
318.2
639.7
127
318.4
634.1
63.5
159.2
319.7
31975.6
32564.7
33759.9
After samples preparation, the mixed particles corresponding to each test were put inside
the jig chamber. The two inferior boxes were defined as "bottom layer", the two intermediate
boxes as "middle layer" and the superior box as "top layer". For the segregation analysis was
assigned a number for each box, beginning in the bottom (layer 1) until top (layer 5).

Figure 3: Particle bed scheme with the designation of each layer.

All tests were performed with the same jigging parameters, defined as: expansion ratio of
80%; frequency of 160 rpm; and test duration of 120 s. These parameters values were adapted
from a previous work with the same equipment [6]. After each test, the boxes containing the
stratified material were removed separately and the materials were separated by hand. Each
fraction was weighed to get the percentage of each component in each box.
3.
Figure 4 shows the jigging evolution for the test with 5% mass of impurities. By visual
observation it was possible to see that the stratification began rapidly, with the paper, wood
and gypsum moving towards the top of chamber. The wood and particularly the paper move
to the top very quickly while the gypsum ascends more slowly. As time increases, a
qualitative segregation could be seen after the test end at t = 120 s, with almost all impurities
concentrated in the layer 5. This behavior of bed stratification was similar in the other tests.
Figure 4 - Sequences of observation on the jig wall during the sorting test. (a) Mixture before
the test; (b) particle bed expansion during test; (c) after test.
The stratification results for test with 1% mass of impurities are shown in Figure 5, in
terms of mass fraction in each layer. It is possible to observe that the most part of concrete
tended to concentrate in the bottom layers (layers 1 and 2), a behavior which is expected due
its higher density in relation to other materials. Brick was more concentrated in the middle
layers (layers 3 and 4) while virtually all wood and paper were concentrated in the top layer
(layers 5).

Mass fraction
100%
75%
50%
25%
0%
1
Paper
0.00%
Wood
0.00%
Gypsum 0.08%
Brick
7.66%
Concrete 92.25%
2
0.00%
0.00%
0.35%
25.17%
74.47%
3
0.01%
0.00%
0.53%
63.10%
36.36%
4
0.12%
0.74%
0.76%
91.31%
7.09%
5
27.31%
39.07%
0.00%
33.62%
0.00%
Figure 5 Results for test with 1% mass of impurities.

If we consider only layer 1 as the product, a high purity concrete can be obtained (more
than 92% mass) with a concrete recovery rate about 50% of the input concrete. Even if we
consider also the layer 2 as the product, we can have an output material with more than 84%
mass of concrete, without any wood or paper and with less than 0.4% mass of gypsum. In this
case, the concrete recovery rate would be of 84%, that is, we would have a more quantity of
concrete, but its purity level would be reduced.
The removal of paper and wood was easy to be realized once that they both are very less
dense than concrete. In relation to gypsum, there is a significant difference between its density
and the enveloped and bulk densities of concrete (Table 1), in the way that there is a
significant driving force for their separation by jigging. Indeed, more than 95% of gypsum
was removed from the concrete in the layer 1 and the occurrence of gypsum particles was
very rare in this layer. The brick particles became more spread out along the layers, although
had a tendency to concentrate in the layers 3 and 4. Once that the enveloped density of brick
is close to the concrete, its separation is more complicated to be done than gypsum. However,
the difference in particle shape of concrete and brick results in a significant difference in the
bulk densities, which can act as the driving force for segregation. Thus, the product in layer 1
has only 7.66% of brick, which represents a removal level of 66% of initial brick content.
Table 3 show the stratification results for the other two tests. The increasing impurity level
had a variable impact according the contaminant and layer. The gypsum concentration was
kept relatively unchanged in layer 1 while intermediate and superior layers were significant
affected by gypsum content in impurities. If only the concrete in layer 1 was considered as the
product, we have gypsum content less than 0.2% in mass. The same may be observed
regarding to wood and paper, which were again almost totally removed from the concrete. As
well as gypsum, it content in inferior layers increase with the initial impurity level. This
behavior occurs because, once that we have more impurity material along the mixture, the
effect of enclosures that occurs in certain points of chamber is more marked. As result, a
small amount of wood and paper remains confined in the bottom and near wall regions, in
which the particle bed expansion was not enough to allow paper and wood pieces move up.
However, if we use a high expansion ratio, the wood and paper enclosed tends to be totally
carried by the ascending air flow.

For all tests, a significant separation between concrete and brick could be achieved. The
brick content in the bottom product decrease from about 41% in mass in the initial mixture to
an average concentration of 10.7% for the three tests.
Table 3 Stratification results for tests with 2.5% and 5% mass of impurities.
Impurity
level
2.5%
5%
Layer
Concrete
Brick
Gypsum
Wood
Paper
1
2
3
4
5
1
2
3
4
5
85.66%
67.41%
41.63%
16.51%
1.69%
89.32%
65.35%
35.53%
10.35%
0.35%
14.14%
31.69%
57.27%
80.37%
28.40%
10.48%
34.00%
61.98%
82.92%
28.98%
0.08%
0.82%
1.07%
2.32%
1.66%
0.15%
0.59%
2.45%
5.91%
4.26%
0.00%
0.05%
0.02%
0.70%
46.13%
0.02%
0.03%
0.04%
0.75%
43.99%
0.11%
0.03%
0.01%
0.09%
22.12%
0.03%
0.03%
0.00%
0.08%
22.41%
For three kinds of tests the concrete concentration in the layer 1 was between 85% and
92% in mass. On the assumption that this layer corresponds to desired product, we could
induce that if an optimization is made on air jig parameters, a higher purity recycled concrete
could be obtained with more than 90% in mass, and possibly exceeding 95%. This
composition is in agreement with the standard guidelines required in most countries for
delivery a high quality recycled concrete [14]. The gypsum removal level in the tests
produced an output product with less than 0.2% in mass of gypsum for all cases.
4.
CONCLUSIONS
The enormous amount of CDW generated all over the world can be handled as a simple ore
treatment problem, with good separation possibilities. Among the separation techniques, the
air jigging is one of the most promising to sort coarse aggregates in the modern recycling
plants. Experimental results showed that the difference of enveloped and bulk densities drive
the separation of gypsum and concrete while the separation of brick and concrete is mainly
drive by difference in bulk densities [6]. Since the air jig can separate gypsum and brick from
concrete with a good performance, the separation of lighter impurities as wood and paper is
an easy task.
The use of air jigs to remove lighter impurities as paper and wood is not usual in recycling
plants, in which air sifters normally perform this function at modern facilities [7]. However,
the results show that air jigging can be useful both for removal of lighter and stony
contaminants from concrete, which gives good perspective to industrial application. In a
recycling plant, this could represent one single operation to remove several contaminants,
which would decrease the economical and environmental costs related to energy and
maintenance.
Further studies need to be performed to identify the optimal jigging parameters and
specially the influence of initial composition of input CDW into the jig. A careful preparation
of material feed is essential to achieve a reasonable separation level with the air jigging.
Similarly, the use of a widely particle size range could affected the bed stratification in

different ways [13]. Furthermore, there are other phenomena that occur in parallel with
jigging, as the formation of granular convection cells in regions near the walls, which is
typical in vibrated granular beds [15] but is still not completely understood for jigs.
REFERENCES
[1] 1. CDRA Construction & Demolition Recycling Association. http://www.cdrecycling.org/
(accessed 04.2015).
[2] Monier, V., Hestin, M., Trarieux, M., Mimid, S., Domrose, L., Van Acoleyen, M., Hjerp, P.,
Mudgal, S., 2011. Study on the Management of Construction and Demolition Waste in the EU.
Contract 07.0307/2009/540863/SER/G2, Final report for the European Commission (DG
Environment).http://www.eusmr.eu/cdw/docs/ (accessed 04.2015).
[3] Abrelpe - associao brasileira de limpeza pblica e resduos especiais. Panorama dos resduos
slidos no Brasil 2013. Abrelpe, 2013. http://www.abrelpe.org.br/ (accessed 04.2015).
[4] Hiete, M., 2013. Waste management plants and technology for recycling construction and
demolition (C&D) waste: state- of-the- art and future challenges. Handbook of Recycled Concrete
and Demolition Waste, pages 76-107.
[5] Mulder, E., Tako, P.R. de Jong, Feenstra, L., 2007. Close Cycle Construction: An Integrated
Process for the Separation and Reuse of C&D Waste. Waste management, Volume 27, Issue 10,
pages 1408-1415.
[6] Cazacliu, B., Sampaio, C.H., Miltzarek, G., Petter, C., Le Guen, L., Paranhos, R., Huchet, F.,
Kirchheim, A.P., 2014. The potential to using air jigging to sort recycled aggregates. Journal of
Cleaner Production, Volume 66, pages 46-53.
[7] Coelho, A., De Brito, J., 2013. Preparation of concrete aggregates from construction and
demolition waste (CDW). Handbook of Recycled Concrete and Demolition Waste, pages 210245.
[8] Tam, V., Tam, C.M., 2006. A review on the viable technology for construction waste recycling.
Resources Conservation and Recycling, Volume 47, Issue 3, pages 209-221.
[9] Xing, W., Pietersen, H., Hendriks, Ch., Rem, P., 2002. Improve the quality of construction and
demolition waste by separation techniques. Advances in Building Technologies, Volume 2, pages
1439-1446.
[10] Vegas, I., Broos, K., Nielsen, P., Lambertz, O., Lisbona, A., 2015. Upgrading the quality of mixed
recycled aggregates from construction and demolition waste by using near-infrared sorting
technology. Construction and Building Materials, Volume 75, pages 121-128.
[11] Hendriks, C., Xing, W., 2004. Suitable separation treatment of stony components in construction
and demolition waste (CDW). International RILEM Conference on the Use of Recycled Materials
in Building and Structures, pages 166-172.
[12] Sampaio, C.H., Aliaga, W., Pacheco, E.T., Petter, E., Wortruba, H., 2008. Coal beneficiation of
Candiota mine by dry jigging. Fuel Processing Technology, Volume 89, Issue 2, pages 198-202.
[13] Sampaio C.H., Tavares L.M.M. (2005). Beneficiamento Gravimtrico: Uma introduo aos
processos de concentrao mineral e reciclagem de materiais por densidade, ed. Editora UFRGS,
1, p.603, Porto Alegre, Brazil.
[14] Martin-Moralez, M., Zamorano, M., Valverde-Palacios, I., Cuenca-Moyano, G., Snchez-Roldn,
Z., 2013. Quality control of recycled aggregates (RAs) from construction and demolition waste
(CDW). Handbook of Recycled Concrete and Demolition Waste, pages 270-303.
[15] Xue, K., Zheng, Y., Fan, B., Li, F., Bai, C., 2013. The origin of granular convection in vertically
vibrated particle beds: the differencial shear flow field. The European Physical Journal, Volume
36.

THE INFLUENCE OF A SUPER PLASTICIZER ON RECYCLED

GYPSUM PLASTER PROPERTIES
Sayonara Maria de Moraes Pinheiro (1), Gldis Camarini (2)
(1) UFES - Federal University of Esprito Santo, Department of Civil Engineering, Brazil
(2) UNICAMP - University of Campinas, School of Civil Engineering, Architecture and
Urban Design
Abstract
The gypsum plaster waste is generated during the components production and building
construction processes. This material should be recycled to avoid the contamination of soil
and groundwater. The reactions reversibility enables the recycling of the material by a simple
recycling process which produced a recycled gypsum plaster. However, the recycled gypsum
has shown a reduction in workability. This waste material needs to ensure the workability to
be applied to components production. In this way, this study analyses the influence of using a
superplasticizer on the properties of recycled gypsum plaster. The properties were evaluated
in fresh (consistency and setting times) and in hardened state (compressive strength, air
permeability and microstructure). Results show better workability with the superplasticizer
content, but the strength diminished significantly. The setting times were increased with
increasing the admixture content and the microstructure showed the formation of large
dihydrate crystal during recycled gypsum plaster hydration.
Keywords: Recycled gypsum plaster; workability; super plasticizer, fresh properties; hardened
properties
1.
INTRODUCTION
Construction gypsum plaster is a biding material with large use in building construction.
the chemical composition of this material is calcium sulphate hemihydrate (CaSO40.5H2O),
Anhydrite III (CaSO4H2O, with 0.06 < < 0.11), Anhydrite II (CaSO4), calcium sulphate
dihydrate (CaSO42H2O) and smaller amounts of impurities [1], [2], [3].
In Brazil, this binder is obtained by the calcination of gypsum mineral rock (CaSO42H2O)
at temperatures ranging from
C to 160 C, as foll
[1].
140
ows (1)
(1)
CaSO42H2O + heat (140 C - 160 C)
CaSO40.5H2O + 1.5H2O
The calcination conditions, such as the dry or humid calcination process, the temperature
in the kiln, the particle size distribution of the raw material, and the residence time, lead to the
final product having different calcium sulphate phases. This results in different behaviour by
changing the kinetic of hydration reaction and their mechanical properties [3].
In Brazil, the chemical composition of the commercial gypsum plaster is hemihydrate
(54.8% - 96.2%), Anhydrite III (0.0% - 19.0%) and dihydrate (0.0% - 10.0%). It is used in

building constructions as pastes for wall renderings, for components (plasterboards,

masonry blocks and ceilings) and for decorative elements [1], [2].
In contact with water, gypsum hemihydrate particles react (hydration reaction) and
become a hard mass by an exothermic reaction, showed in (2) [4].
CaSO40.5H2O + 1.5H2O
CaSO42H2O + heat
(2)
In Brazil, there is a significant amount of gypsum plaster waste: 4% to 15% of

construction and demolition waste [5], and values up to 47% of material loss when used as
wall rendering [6].
This waste material is mainly composed of calcium sulphate dihydrate (CaSO42H2O)
which is not inert and contaminates soil and groundwater. Brazilian Environmental
Legislation establishes that the gypsum waste can be recycled or placed in special landfills.
Due to the chemical characteristics of gypsum plaster waste; this material can be recycled
as a binding material by a simple recycling process, similar to the industrial process [5].
Recent studies showed similar characteristics between recycled gypsum plaster (RGP) and
commercial gypsum plaster (CGP) assessed by thermal analyses (TG/DTG) [7] and crystal
morphology (SEM) [8], as well as it can be recycled successively [9].
In spite of that, there is a lack of information about the behaviour of this recycled material.
Pinheiro [5] has found a reduction on the workability of the recycled gypsum plaster (RGP),
and to be used in buildings this property is very important.
The RGP in the production of components for buildings is necessary to obtain an adequate
workability for component moulding. It can be reached by using an admixture that modifies
the material properties in fresh and hardened states. In this way, this work aims to study the
influence of a superplasticizer (polycarboxylate) on RGP properties.
2.
HYDRATION, WORKABILITY AND PROPERTIES
The mechanism of gypsum hydration has been explained by the theory of crystallization
described by Lavoisier in 1798, and by Le Chatelier in 1877, whereas the gypsum
crystallization occurred in the saturated solution of the hemihydrates. The chemical
phenomenon had three steps: the chemical phenomenon of hydration; the physical
phenomenon of crystallization; and the mechanical phenomenon of hardening [4].
When mixing with water, hemihydrate (CaSO40.5H2O) dissolves in water, forming a
saturated solution of Ca2+ and SO42- ions, followed by precipitation of acicular dihydrate
crystals (CaSO42H2O). So, the crystallization nuclei are formed and around it the crystals
grow, forming an interlocking net of long crystals providing the strength of the material [10],
[11].
During the hydration process the microstructure of the hardened paste is defined. It
depends on plaster constituents, the water/gypsum ratio and the admixtures used.
Initially, the water/gypsum ratio is responsible to lead the hydration reactions and to keep
the workability/fluidity of gypsum slurry. In general, when the water content increases the
void fractions also increase, and the bond between the gypsum crystals decrease, leading to
lowering the strength of the material [12], [13].
To achieve the workability/fluidity required for gypsum plaster it is also employed certain
types of admixtures. The admixtures often used to improve the workability of plaster are
retardants that increase the time available for its handling, and superplasticizers, which
provide the fluidity/plasticity necessary for moulding [14], [15].
Superplasticizers, when added to gypsum paste, modify the mechanism of the RGP
hydration. The admixture is chemically adsorbed on the surface of calcium sulphate

hemihydrate grains, slowing down the dissolution process, the formation of nucleation sites
and the crystal growth, resulting in changing the material microstructures decreasing their
physic and mechanical properties [16].
3.
3.1
Materials
The process to obtain the RGP is followed [5]: The gypsum plaster waste (GPW) was
ground in a ball mill reaching a fineness modulus less than 1.10. After grinding the powder
was calcined in a stationary kiln at 150 C for one hour, obtaining the RGP. Tables 1 and 2
show the physical properties and chemical characteristics of RGP, respectively.
Table 1: Physical properties of recycled gypsum plaster.
Material
Fineness
modulus
Bulk unit weight

(kg/m3)
Specific gravity
(kg/m3)
Specific surface area

(m2/kg)
RGP
0.38
429
2550
6345
Table 2: Physical properties of recycled gypsum plaster.

Mass (%)
Levels (%)
Material
Free
water
Water
crystallization
CaO
SO3
RGP
0.85
2.96
39.0
49.4
Hemihydrate
Dihydrate
(CaSO40.5H2O) (CaSO42H2O)
89.49
10.51
The superplasticizer used in this experimental work was a polycarboxylate based

admixture. The solids are 21.88%, determined by drying in infrared light balance.
Methods
3.2
Pastes were produced with a water/plaster ratio of 0.7. The content of superplasticizer used
were 0%, 1%, 1.5%, 1.75%, 2.0% and 2.2%.
It was observed: the consistency by the mini-slump test [5], setting times by Vicat
needle [17], compressive strength [18], air permeability [5] and SEM observations.
4.
4.1
Superplasticizer content
Table 3 shows the values of mini-slump test (spreading) and compressive strength from the
pastes with 0%, 1%, 1.5%, 1.75%, 2.0% and 2.2% of superplasticizer content.
The RGP increased the workability/fluidity by increasing the spreading of the paste. On the
other hand, the compressive strength decreased with increasing the admixture content, as
reported by Milln [15] and Sing and Middendorf [16].

Table 3: Spreading and compressive strength results from RPG pastes with different
superplasticizer content.
Pastes
Properties
RGP-0 RGP-1 RGP-1.5 RGP-1.75 RGP-2.0 RGP-2.2
Spreading (mm)
0.0
50
71
71
77
79
6.2
5.0
3.8
3.2
2.5
2.7
4.2
Microstructure evaluation
The microstructure was evaluated on the RGP pastes whose results of spreading (higher
than 70 mm) and compressive strength were suitable for the production of components (good
workability and minimum value of compressive strength according to the Standards). In this
case it was chosen the mixture RPG-1.5. This mixture had spreading of 71 mm and
compressive strength of 3.8 MPa (Table 3). Results of setting times are important to these
pastes to know how long they maintain the workability to make the components (Table 4).
Table 4: Results of spreading and setting times from RGP pastes with 0% and 1.5% of
superplasticizer
Paste
Spreading
(mm)
RGP-0
RGP-1.5
71
Setting times (min)
Air permeability
Compressive
strength (MPa)
start (t1)
end (t2)
t (t2 - t1)
(mm)
12
25
13
2.18 x 10-14
28
36
18.47 x 10
-14
6.2
3.8
These results show that the setting times increased with the admixture, showing more time
to work with the RPG paste and diminishing the possibility to this mixture be lost and
produce again gypsum plaster waste.
The compressive strength to make components is in accordance with the
Brazilian Standards for soil-cement blocks (2 MPa) [19] and ceramic block (1MPa) [20].
The air permeability increased significantly (8.5 times) showing the high permeability and
the possibility of fungal growth; so, this component must be protected. But this high
permeability can improve the thermal comfort inside the ambient when using blocks made of
RGP.
Figure 1 shows the microstructure of the RGP pastes without and with admixture. The
crystal formation is quite different. In the RGP without superplasticizer the crystals are needle
like with thickness dimensions varying from 1 m to 2.5 m, while the paste RGP-1.5 with
1.5% of superplasticizer the thickness are higher than 3 m, the pores are large and prismatic.
There is not the interlocking of crystals to bring the necessary bond to improve the RGP
properties.
When the RGP has superplasticizer addition, results can be summarised as follows: (i) an
increase in the workability/fluidity of the material; (ii) retarding the setting time; (iii) an
increase in the air permeability; and (iv) decrease in compressive strength with respect to the
RGP0.

The superplasticizer addition modifies the hydration mechanism of the material. The
admixture is chemically adsorbed on the hemihydrate surface becoming slower the process of
dissolution, the formation of nucleation sites, crystal growth and hardening [11], [15], [16].
So, the same occurred with RGP-1.5 paste: an increase in the workability/fluidity and the
retarding of setting time.
Figure 1. Microstructure morphology obtained by SEM from (a) RGP-0 and (b) RGP-1.5.
(a) RGP-0: long needle-like dihydrate crystals
(b) RGP-1.5: large and prismatic dihydrate

crystals
During the hydration mechanism the microstructure of hardened gypsum plaster is formed.
It is constituted by interlocking dihydrate crystals in the form of plates and needles with high
content of voids. This microstructure arrangement leads the physic-mechanical properties of
hardened plaster [12].
When superplasticizer is added to the RGP, it modifies the morphology and habit of
growing crystals resulting in microstructures formed by larger and less interlocked crystals,
with high voids content. This new microstructure arrangement results in changes, increasing
the porosity and reducing the mechanical properties of the resulting material [16].
The RGP1.5 presented an increase in the air permeability and decrease in compressive
strength in relation to RGP0. The change of microstructure (Figure 1) is similar as related by
Sing and Middendorf [16].
5.
CONCLUSIONS
This work studied the influence of superplaticizer admixture on the properties of recycled
gypsum plaster (RGP). The superplasticizer addition on RGP pastes had the following.
Increased the workability/fluidity and decreased the compressive strength of RGP;
The setting times were increased; and increase the air permeability;
The microstructure of RGP-1.5 had large dihydrate crystals, with low interlocking, and
high content of voids.
ACKNOWLEDGEMENTS
The authors are grateful to CAPES and CNPq for the scholarships, the LNLS for the SEM
observations, LARES-FEC-UNICAMP where the tests took place, and DEC-CT-UFES by the
participation at the conference.

REFERENCES
[1] M.A. Cincotto, V. Agopyan, M. C. Florindo. Gypsum plaster as a construction material, in:
Technology of Buildings, Ed (IPT, So Paulo, Brazil, 1988), 53-60. (in Portuguese).
[2] V. M. John, M. A. Cincotto, Gypsum for Civil Construction, in: ISAIA, G. C.Civil
Construction Materials, Ed (IBRACON, So Paulo, Brazil, 2007), 727-760. (in Portuguese).
[3] J. Moisset, Review of the main process used for plaster products, in Gypsum Fair,
Proceeding of Gypsum Fair, Olinda, Brazil, 1997. (FINEP, Brazil, 1997), 33p.
[4] W. C. Hansen, Hydration of calcined gypsum, Industrial and Engineering Chemistry.
22, (1930), 611-613.
[5] S. M. M. Pinheiro, Recycled gypsum plaster: properties evaluation for using in
components, PhD Thesis, Graduate Program in Civil Engineering, School of Civil

Engineering, Architecture and Urban Design FEC, University of Campinas
UNICAMP, Campinas, Brazil, 2011. (in Portuguese).
[6] G. Camarini, L. L. Pimentel, N. H. R. Sa, Assessment of the material loss in wall
renderings with B-hemihydrate paste, Applied mechanics and Materials, 71-78,
(2011),1242-1245.
[7] G. Camarini, S. M. M. Pinheiro, and K. Tannous, Thermal analysis of recycled gypsum
from construction and demolition waste, Applied Mechanics and Materials, 260-261,
(2012), 977-980.
[8] G. Camarini, S. M. M. Pinheiro, Microstructure of recycled gypsum plaster by SEM,
Advanced Materials Research, 912-914, (2014), 243-246.
[9] S. M. M. Pinheiro, G. Camarini, Characteristics of gypsum recycling in diferente cycles,
International Journal of Engineering and Technology, 7, (2015), 215-218.
[10] J. Karni, E. Karni, Gypsum in construction: origin and properties, Materials and
Structures, 28, (1995), 92-100.
[11] A. J. Lewry, J. Williamson, The setting of gypsum plaster: part I the hydration of
calcium sulphate hemihydrate, Journal of Materials Science, 29, (1994). 5524-5528.
[12] J. Lewry, J. Williamson, The setting of gypsum plaster: part II The development of
microstructure and strength, Journal of Materials Science, 29, (1994), 524-5528.
[13] Q. L. Yu and H. J. H. Brouwers, Microstructure and mechanical properties of bhemihydrate produced gypsum: An insight from its hydration process, Construction and
Building Materials, 25, (2011), 3149-3157.
[14] L. V. Domnguez, A. G. Santos, Gua del Yeso, Edn (Tcnica y la Associacin de
Empresas de Yeso, Espaa, 2001).
[15] M. L. Milln, Influencia de los aditivos en las caractersticas especficas de las
formulaciones base yeso in Gypsum Fair, Proceeding of Gypsum Fair, Olinda, Brazil,
1997. (FINEP, Brazil, 1997), 17p.
[16] N. B. Sing, B. Middendorf, Calcium sulphate hemihydrate hydration leading to gypsum
crystallization, Progress in Crystal Growth and Characterization of Materials, 53,
(2007), 57-77.
[17] Gypsum for Construction: determination of phisical properties from paste, ABNT
Standard NBR 12.128, Rio de Janeiro, 1991.
[18] Gypsum for Construction: determination of mechanical properties from paste, ABNT
Standard NBR 12.129, Rio de Janeiro, 1991.
[19] Soil-cement block Dimensional analysis, compressive strength determination and
water absorption Test method, ABNT Standard NBR 10.836, Rio de Janeiro, 2013.
[20] Ceramic components Part 1: Hollow ceramic blocks for non load-bearing masonry Terminology and requirements, ABNT Standard NBR 15.270-1, Rio de Janeiro, 2005.

TREATMENT OF RESIDUAL DIATOMITE EARTH FOR USE AS

POZZOLAN IN CEMENTITIOUS MIXTURES
Jlia G. Vieira(1), Csar A. Ponciano (1), Welles J. Oliveira (1), Leila A. C. Motta (1)
(1) College of Civil Engeneering, Federal University of Uberlndia, 38400-902, UberlndiaMG, Brazil
Abstract
The problems of environmental pollution caused by industrial waste have attracted great
interest in recent years, which has motivated the search for viable alternatives for the disposal
of these, especially solid residues, as is the case of the diatomite earth used by industries in
filtration processes. Construction is a sector of society that lies in accelerated growth, being
responsible for much of the urban pollution, however this is also one of the sectors where part
of waste generated can be reused. Pozzolanic materials enable the production of special
cimentitious mixtures with lower energy consumption and cost of manufacture. The
pozzolanic additions contribute to the greater durability of composites by reducing the
calcium hydroxide content free. In this sense, treatments were investigated to adapt diatomite
earth to the use as pozzolan in mixtures with Portland cement, since this present contaminants
that affect the cement hydration reactions. Different combinations of times and temperatures
of calcination were investigated to remove impurities present in it. Treated samples were
characterized through the techniques of Thermogravimetric Analysis (TGA), X-Ray
Diffraction (XRD) and pozzolanic activity index (Modified Chapelles method), in which the
best result obtained was for the treatment of 2 hours to 650 C.
Keywords: residue, diatomite earth, thermal treatment, pozzolan.
1.
INTRODUCTION
The problems of environmental pollution caused by waste from industries have attracted
great interest in Brazil in recent years. The environmental control laws have become more
stringent and the environmental inspection agencies more efficient. On the other hand,
disposal costs of waste in an environmentally correct way are high. This has motivated the
search for viable technological alternatives for the disposal of industrial waste, especially for
the solid residues, which can introduce reactivity and occupy large space, as is the case of the
diatomite earth used by industries in filtration processes [1].

Construction is a sector that currently lies in accelerated growth, being responsible for
much of the urban pollution that affects the entire environment. However this is also one of
the sectors where part of waste generated can be reused or incorporated into the construction
process [2]. In this context, pozzolanic materials enable the production of special
cimentitious mixtures with lower energy consumption and therefore lower cost of
manufacture [3,4]. The pozzolanic additions contribute to the greater durability of composites
by reducing the calcium hydroxide content free [5,6].
The pozzolans are natural or artificial substances of siliceous composition or silicoaluminous, that not being itself cementitious, react with calcium hydroxide at room
temperature resulting in compounds with cementitious properties. The pozzolans can be
divided into two groups: natural and artificial. The natural pozzolans are all the rocks with
natural pozzolanic activity and also the products of calcination of these rocks. As for the
artificial are industrial by-products [4].
Among the natural pozzolans, are the diatomite earths. Diatomaceous are photosynthetic
organisms that are of a carapace or frustule of non-crystalline silica; they live in a wide
variety of aquatic environments, from the fresh water or brackish water until the regions
frankly navies. The diatomite or spongolite is a rock powdery appearance, light, composed of
carapaces of microscopic algae or frustules of spongiary, whose structure is amorphous
hydrated silica. They also include small amounts of inorganic substances, such as alumina,
iron and alkali metals, varying amounts of organic matter and common components, such as
sand, silt and clay [4].
The diatomite earth can be applied for many different purposes, such as: industrial
filtration, thermal and acoustic insulation, catalysts, absorbent, loads, abrasives,
pharmaceutical industry, chemical industry, building materials and pozzolans. Among the
aforementioned applications, diatomaceous earth can be used as natural pozzolans, due to
their physical and chemical characteristics. Depending on the degree of purity, they can be
mixed directly to cement, replacing a portion of Portland clinker or added after thermal
treatment [3].
The diatomite earth used in this study is a residue of the food industry, where it is
employed in filtration processes, being generated a large amount of waste, about 400 t/month.
In this sense, treatments were investigated to adapt it to the use as pozzolan in mixtures with
Portland cement, whereas the residual diatomite earth present contaminants that affect the
cement hydration reactions. In this case, the contaminants are organic compounds from
filtration process, which are adsorbed on the structure of the diatomite earth. Different
combinations of times and temperatures of calcination were investigated to remove impurities
present in it. Treated samples were characterized through the techniques of
Thermogravimetric Analysis (TGA), X-Ray Diffraction (XRD) and pozzolanic activity index
(Modified Chapelles method).
2.
EXPERIMENTAL
The diatomite earth used in this study comes from a food industry of the city of Uberlndia
- Minas Gerais, where it is used in filtration processes. The diatomite earth was received in its
saturated form, so moist, and initially was dried at 60 C for 24 h. This procedure was
performed to remove the moisture and so that it could be stored until further use.

2.1. Thermal treatment

Diatomite earth samples were subjected to different temperatures and times of calcination
based on literature and also based on TGA thermogram of residual diatomite earth, to improve
its properties as a pozzolanic material. Once defined the best calcining temperature
investigated the possibility of reducting the time for optimizing the treatment. Thermal
treatment of samples was carried out in a muffle furnace model FORNITEC N1110, under the
following conditions: 2 h at 650 C and 2 h at 650 C with thermal shock, 2 h at 750 C, 2 h at
850 C, 2 h at 950 C, 1.5 h at 650 C, 1 h at 650 C and 0.5 h at 650 C.
2.2. Determination of pozzolanic activity with Portland cement

The test of determination of pozzolanic activity with Portland cement was performed
according to ABNT NBR 5752/2012 [7].
2.3. Thermogravimetric Analysis (TGA)/DTA
Thermogravimetric analysis was performed in order to determine the temperature of
degradation of organic compounds present in residual diatomaceous earth. TGA curves were
obtained in a DTG-60H of Shimadzu. About 10 mg of sample were heated in alumina
crucibles, with a heating rate of 10 C min-1, under continuous nitrogen flow of 30 mL min-1.
The weight loss was recorded from 25 to 1000 C. DTA measurements were performed under
the same conditions.
2.4. X-Ray Diffraction (XRD)
The Diffraction X-ray technique was used to evaluate changes in mineralogical structure of
diatomaceous earth, before and after thermal treatment. X-ray diffraction patterns were
obtained with a Shimadzu Diffractometer XRD-6000 with Ni filtered, empregando-se a
radiao K do Cu (1,5406 ) e filtro de Ni, com 2 variando 5 a 80e velocidade de
varredura de 4 min-1, operando a uma potncia de 40kV comcorrentede 30mA.
2.5. Pozzolanic activity index (Modified Chapelles method)
The determination of the pozzolanic activity index is a very important characterization,
since the pozzolanic additions contribute to the greater durability of composites by reducing
the calcium hydroxide content free. This procedure was performed according to ABNT NBR
15895/2010 [8].
The pozzolanic activity index, which corresponds to the calcium hydroxide content fixed
in mg of Ca(OH)2 by g of pozzolanic material was calculated by the formula (1):
(1)
Where:

ICa(OH)2: is the pozzolanic activity index Chapelle, obtained in the test, that corresponds to
the fixed content of calcium hydroxide (mg Ca(OH)2/g pozzolanic material);
m2: is the mass of pozzolanic material (g);
V2: is the volume of HCl 0.1 M consumed in the test with the sample (mL);
V3: is the volume of HCl 0.1 M consumed in the blank test (mL);
Fc: correction factor for the solution of HCl 0.1 M;
1,32 is the molecular ratio Ca(OH)2/CaO.
3.

3.1. Determination of pozzolanic activity with Portland cement
Through the determination of pozzolanic activity with Portland cement test, it was found
that only the specimens of the mortars prepared with new diatomite earth and treated by 2 h at
650 C dried, while those prepared with residual diatomite untreated not dried within a period
of 24 h, showing the interference of contaminants present in diatomite earth on cement
hydration reactions. Therefore, it was only possible to carry out the cure process of these.
This essay also highlighted the need for thermal treatment before any use or application of
the residual diatomite earth.
3.2. Thermogravimetric Analysis (TGA)/DTA
Figure 1 shows the TGA and DTA curves of diatomite earth saturated with organic matter,
residue generated in the filtration process in a food industry.
100
50
95
40
30
85
80
20
75
10
70
0
65
100
200
300
400
500
600
700
800
900
Temperature difference
Weight Loss (%)
90
1000
Temperature ( C)
Figure 1: TGA and DTA thermograms for residual diatomite earth.

By the result of TGA were defined temperatures above 650 C for the thermal treatment of
diatomite earth samples. Assuming that this temperature is above the second stage of weight
loss, this can ensure the elimination of organic matter adhered to diatomite earth used in
filtration process.
Figure 2 shows the thermograms for samples of new diatomite earth (ND), residual
diatomite earth treated for 2 h at 650 C (RD 650) and treated by 2 h at 650 C with thermal
shock (RD 650TS), in comparison with the residual untreated diatomite earth (RD).
ND
RD
RD 650
RD 650TS
100,0
Weight Loss (%)
97,5
95,0
92,5
90,0
87,5
85,0
82,5
80,0
100
200
300
400
500
600
700
800
900
Temperature ( C)
Figure 2: TGA curves for the samples of new diatomite earth (ND), residual (RD), residual
diatomite earth treated for 2 h at 650 C (RD 650) and residual treated by 2h at 650 C with
thermal shock (RD 650TS).
From the thermograms is possible to verify that the weight loss to the residual
diatomite earth was approximately 18%, evidencing the loss of organic matter adhered to the
material, while for new diatomite earth was only 2.5%. The sample of diatomite earth treated
by 2 h at 650 C presented weight loss of about 3% and the sample treated with thermal shock
around 4.5%.
3.2. X-Ray Diffraction (XRD)

In Figure 3 are presented the X-ray diffraction patterns concerning samples: new diatomite
earth (ND) and new diatomite earth treated by 2 h at 650 C (ND 650) and 2 h at 950 C (ND
950); residual diatomite earth untreated (RD) and residual diatomite earth treated by 2 h at
650 C (RD 650), 750 C (RD 750) and 850 C (RD 850); residual diatomite earth treated by 2

h at 650 C with thermal shock (RD 650TS) and residual diatomite earth treated by 1.5 h (RD
1.5), 1 h (RD 1.0) and 0.5 h (RD 0.5).
ND
ND 650
ND 950
RD
RD 650
RD 750
RD 850
RD 650TS
RD 1.5
RD 1.0
RD 0.5
4000
3500
Intensity (CPS)
3000
2500
2000
1500
1000
500
0
10
20
30
40
50
60
70
80
2degrees)
Figure 3: X-ray diffraction patterns for new diatomite earth, residual diatomite and residual
treated under various conditions of time and temperature.
It is observed that there were no changes in the position and relative intensity of the
diffraction peaks, suggesting that the thermal treatment did not alter the structure of the
mineral. In the study of Souza et al. [9] there has been changes in crystal structure in a type
of diatomaceous earth found in the region of Campos dos Goytacazes - RJ, only at
temperature close to 950 C, in this case there was not even changes to the material treated for
2 h at 950 C.
3.3. Pozzolanic activity index (Modified Chapelles method)
Studies have shown that it is possible to minimize the deleterious effects on vegetable
fibres and ensure the durability of composites by reducing the calcium hydroxide content free
of the matrix, with the use of pozzolanic additions, such as silica fume, rice husk ash,
metakaolin, pozzolans as diatomaceous earth and blends with two or more pozzolans,
including ceramic residue and metakaolin [6]. It is also known advantages of replacing the
part of clinker by pozzolan in cement manufacturing and production of concretes, improving
performance, durability and reducing cost, energy consumption and emission of pollutants in
the atmosphere [5]. In this regard, it is important to determine the pozzolanic activity index of
the studied samples of diatomaceous earth. The results of this test are shown in Table 1.

Table 1: Results of pozzolanic activity index

Samples
New diatomite earth

New diatomite earth 2 h 650 C
New diatomite earth 2 h 950 C
Residual diatomite untreated
Residual diatomite earth 2 h 650 C
(thermal shock)
Residual diatomite earth 1.5 h 650 C
Residual diatomite earth 0.5 h 650 C
Pozzolanic activity index

(Modified Chapelles
method)
752.04
424.76
122.83
163.71
655.37
290.22
230.59
465.28
540.04
441.25
398.04
Analyzing the results of Table 1, it was found that the higher pozzolanic activity index
was shown by the new diatomite earth, followed by residual diatomite earth treated for 2 h at
650 C. On the other hand, the residual diatomite earth treated for 2h at 650 C with thermal
shock, showed less than optimal results. It was also observed that the residual diatomite earth
without thermal treatment shows a very low pozzolanic activity index, justifying the need for
thermal treatment for removing of impurities before its use.
Whereas the calcination temperature with best result for the pozzolanicity index was to
650 C, investigated the possibility of calcination time reduction, for the purpose of reducing
process costs. However, the reduction of calcination time prejudiced the pozzolanic activity of
residue.
Figure 4 shows the images of new diatomite earth and residual diatomite earth after
thermal treatment for 2 h at 650 C.
(a)
(b)
Figure 4: New diatomite earth (a) and residual diatomite earth thermal treated for 2 h at
650 C (b).

4.
CONCLUSION
From the TGA thermogram was possible to set the temperature for the thermal treatment of
residual diatomaceous earth samples, ensuring the elimination of organic matter adhered to
diatomaceous earth during the filtration process. For examining X-ray diffraction patterns it
was found that there were no changes in the position and relative intensity diffraction peaks,
suggesting that thermal treatment did not alter the structure of the mineral. The results of the
pozzolanic activity index indicate the residual diatomaceous earth treated by 2 hours at 650 C
as appropriate for use as pozzolan in cementitious mixtures.
ACKNOWLEDGEMENTS
The authors acknowledge CAPES for the financial support and for providing the
periodicals portal (www.periodicos.capes.gov.br). Vieira thanks CAPES for Post-Doctoral
scholarship and Ponciano thanks CNPq for Scientific Initiation scholarship. The authors thank
Cargill company - Unidade de Negcio de Amidos e Adoantes for providing the diatomite
earth used in this study.
REFERENCES
[1] Goulart, M.R., Silveira, C.B., Campos, M.L., Almeida, J.A., 'Metodologias para reutilizao
do resduo de terra diatomcea, proveniente da filtrao e clarificao da cerveja', Quim. Nova. 34
(4) (2011) 625-629.
[2] H. Savastano Jr, 'Sistemas de cobertura para construes de baixo custo: uso de fibras vegetais
e de outros resduos agroindustriais', in: J. C. Rocha, V. M. John. (Eds.), Utilizao de resduos na
construo habitacional, Finep, Rio de Janeiro, 2003, 94-123.
[3] Montanheiro, T.J., Yamamoto, J.K., Santagostino, L.M., Kihara, Y., Saito, M.M., 'Terras
diatomceas: uma pozolana natural na bacia do Paran, estado de So Paulo', Revista do Instituto
Geolgico, 23 (2) (2002b) 01-17.
[4] Montanheiro, T.J., Yamamoto, J.K., Kihara, Y., 'Caractersticas e propriedades pozolnicas de
arenitos opalinos da Serra de Itaqueri, SP', Revista do Instituto Geolgico, 23 (1) (2002a), 13-24.
[5] Lima, P.R.L., Toledo Filho, R.D., 'Uso de metacaulinita para incremento da durabilidade de
compsitos base de cimento reforados com fibras de sisal', Ambiente Construdo (Online), 8 (4)
(2008), 07-19.
[6] Farias Filho, J., Toledo Filho, R.D., Lima P.R.L., 'Efeito da argila calcinada sobre a
durabilidade de argamassas reforadas com fibras curtas de sisal', Revista Brasileira de
Engenharia Agrcola e Ambiental, 14 (10) (2010) 1109-1114.
[7] ABNT NBR 15895/2010. Pozzolans - Determination of pozzolanic activity with Portland
cement - Test method, 2012.
[8] ABNT NBR 15895/2010. Pozzolanic materials - Determination of calcium hydroxide fixed Modified Chapellesmethod, 2010.
[9] Souza, G.P., Filgueira, M., Rosenthal, R., Holanda, J.N.F., 'Caracterizao de material
compsito diatomceo natural', Cerm. 49 (2003), 40-43.

RELEASE OF DANGEROUS SUBSTANCES FROM CONSTRUCTION

AND DEMOLITION RECYCLED MATERIALS USED IN ROAD
PAVEMENTS - LABORATORY AND FIELD LEACHING TESTS
Martins I.M (1), Roque A.J. (1), Freire A.C. (1), Neves J. M. (2), Antunes M.L. (1)
(1)
Laboratrio Nacional de Engenharia Civil LNEC, Portugal
(2)
Instituto Superior Tcnico IST, Universidade de Lisboa, Portugal
Abstract
The use of Construction and Demolition Recycled Materials in road pavements is
envisaged as a recycling solution with environmental and economic benefits, allowing the
incorporation of large amounts of those materials from different sources. The environmental
concerns regarding the potential contamination of surface and ground waters and soil forces
the evaluation of the leaching behavior of the recycled materials. In Portugal the application
of Construction and Demolition Waste (CDW) in civil engineering works demands their
compliance with the requirements of leachability for waste admissible in landfills for inert
waste. Nevertheless some issues concerning this criterion and the leaching procedure should
be better studied whereby leaching tests, laboratorial batch and column leaching tests and in
situ lysimeter leaching tests, were performed, or are still ongoing, on recycled aggregates
crushed mixed concrete, crushed reclaimed asphalt and milled reclaimed asphalt and on a
natural aggregate to provide a comparison with the results obtained with recycled materials.
Crushed and milled reclaimed asphalt aggregates do not fulfill the criteria to be disposed at
landfill for inert waste owing to the content of the hydrocarbons C10-C40. Nevertheless,
based only on the thresholds for leachability defined in the Portuguese technical specifications
no restrictions apply for their use from an environmental perspective.
Keywords: Construction and demolition recycled materials; Road pavements; Leaching
1.
INTRODUCTION
In the European Union the overall annual material consumption per inhabitant is around
16 tonnes with almost one third being wasted and about half of this being landfilled, thereby
hindering a closed loop material cycle. One of the aims of the Europe 2020 strategy and also
an opportunity according to the Vision 2050 of the World Business Council for Sustainable
Development is the efficient use of resources. This concern is further reinforced in the
Construction Products Regulation (CPR) No. 305/2011 [1] which points out the need for
recyclability of materials from construction and demolition works and the use of raw
materials and secondary materials with low environmental impact.
The aggregates sector, with a world estimated annual demand of 5.2% is the non-energy
extractive industry using more natural resources and can be seen as a priority sector to
improve resource efficiency by increasing the use of recycled aggregates, i.e. materials from

CDW that, currently in Europe, represents only 5% of the aggregates production.

Nevertheless, to boost the use of recycled aggregates the materials from CDW stream should
be considered secondary raw materials whereby they must obtain the end of waste (EoW)
status under the Waste Framework Directive [2].
According to EoW criteria one important condition is that the materials that cease to be
waste must not lead to overall adverse environmental or human health impacts, an issue also
recognised in the basic requirement of CPR. To provide evidence on this topic the leachability
of recycled aggregates must be addressed and compared to threshold values for the relevant
pollutants. Development of a methodology to establish limit values for pollutants in the EoW
framework at European level is ongoing but in its lack each Member State defines their own
criteria [3]. Similarly to other countries, Portugal implemented in their technical specifications
for the use of recycled CDW aggregates a leachability requirement. According to this
requisite the eluates of recycled aggregates, obtained using the batch leaching test of
EN 12457-4 [4] should comply with the limits for waste to be disposed in landfill for inert
waste.
Recycled aggregates are increasingly common in road infrastructures although the use of
crushed or milled reclaimed asphalt (RAP) as unbound materials for base and subbase layers
is unusual. The National Laboratory for Civil Engineering (LNEC) and the University of
Lisbon (IST) carried out the research project, SUPREMA, regarding the use of recycled
aggregates in road infrastructures including those materials [5, 6]. To assess their potential
release of pollutants, the results from leaching tests were evaluated against the leaching
behaviour of an often used recycled aggregate - crushed mixed concrete and of a natural
aggregate limestone for the abovementioned application.
The European standard regarding the use of unbound and hydraulically bound materials in
road construction foresees a laboratory batch leaching test, according to EN 1744-3 [7]. In
theory, leaching according to EN 12457-4 [4] and EN 1744-3 [7] evaluate the short term
release of contaminants based on the principle that equilibrium or near equilibrium conditions
between solid and liquid phases were achieved, but some debate around this subject can be
found in the literature [8]. The repeatability of the test is uncertain owing to the extended
particle size allowed and there is a mix of release mechanisms.
Moreover, the release of pollutants was also evaluated using lysimeters in which a layer of
each recycled aggregate, placed under a drainage layer, was compacted with the optimum
water content of the Modified Proctor test. The tests conditions of the lysimeters simulate the
field conditions of the road test sections constructed during the project but no wearing course
was placed above the recycled aggregates, corresponding to the extreme situation of a
damaged pavement due to traffic and climatic solicitations, whereby it would result in higher
leachability.
The main objective of this paper is to present and analyse the leachability observed in
different leaching tests, assuming that in all situations the recycled aggregates were tested
under similar conditions.
2.
The leaching tests performed and the constituents of the crushed mixed concrete, CMC,
crushed reclaimed asphalt, CRAP, and milled reclaimed asphalt, MRAP are shown in Table 1
and in Table 2, respectively.
Often the leaching of heavy metals and salts are considered the main categories of concern
regarding the use of recycled aggregates. Nevertheless, this study was mainly focused in RAP
aggregates whereby an analysis of organic analytes was also carried out.

Table 1: Test methods for assessment of leaching

EN 1744-3
EN 12457-4
In situ
Preparation of eluates by leaching of aggregates

Characterization of waste. Leaching. Compliance test for leaching of
granular waste materials and sludges Part 4: One stage batch test at a
liquid to solid ratio of 10l/kg for materials with particle sizes below
10mm (without or with size reduction)
Recovery and analysis of leachates from lysimeters percolated by
rainwater, with cumulative L/S ratio from 0.1 to 10.0.
Table 2: Proportion of the constituents in the recycled aggregates [9]

Constituents
Concrete, concrete products,
mortars
Unbound aggregate, natural
Ru [%]
stone,
hydraulically
bound
aggregates
Ra [%]
Bituminous materials
Clay masonry units (i.e. bricks
and tiles), calcium silicate
Rb [%]
masonry units, aerated nonfloating concrete
Rs [%]
Soils
Rg [%]
Glass
Other: cohesive materials (i.e.
clay and soil), plastics, rubbers,
X [%]
metals (ferrous and nonferrous),
non-floating wood, gypsum
plaster
3
FL [cm /g] Floating particles
Rc [%]
3.
CMC
CRAP
MRAP
60
6.1
0.0
24
29
0.0
12
64
99
3.9
0.9
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.1
0.0
1.0
0.0
0.0
0.0
RESULTS
3.1
Batch leaching tests
Table 3 shows the results of leaching tests performed according to EN 12457-4 [4] as well
as the allowed limits of leachability for recycled aggregates to be used in civil engineering
works, as foreseen in the Portuguese specifications for these materials. As can be seen the
released contents of substances of concern are too far from those limits.
In what concerns to eluates from RAP aggregates, the dissolved organic carbon, especially
for MRAP, was higher when compared to CMC owing to the high content of the bituminous
binder (see Table 2) tough it is lower that the limit for landfills for inert waste. Nevertheless,
an increased mobility of some determinands due to the higher content of dissolved organic

carbon in RAP aggregates, as seen in eluates of other residues [10], was impossible to identify
in this leaching test.
If all the requirements for inert waste, stated in the Landfill Waste Framework, were taken
into account to allow the use of recycled aggregates it is also necessary to analyse the total
content of organic contaminants. RAP aggregates show a high total organic carbon content,
particularly MRAP, that is not a problem since the content of dissolved organic carbon in the
eluate is below the threshold for acceptance of waste at landfills for inert waste. Other organic
contaminants PAH, BTEX and PCB show lower contents on RAP aggregates but for
hydrocarbons C10-C40 the threshold of 500 mg/kg is exceeded: the contents were 6200
mg/kg for MRAP and 3020 mg/kg for CRAP. Hence, RAP aggregates will not be classified as
inert and could not be used as unbound aggregates.
As regards the inorganic metal analytes, chromium shows the higher content in CMC
eluates tough it is of the threshold. Cadmium, copper and zinc, usually assigned as metal
pollutants from RAP do not pose any environmental concern in these aggregates [11].
Regarding soluble species sulphate is the main critical anion, particularly for CMC.
Chromium and sulphate were also considered critical contaminants for concrete and mixed
concrete recycled aggregates to be used in structural road layers by Galvn et al. [12]. Based
only on the leachability criteria the low release of contaminants on the short term for the
different recycled aggregates point to the feasibility of their use as alternative materials to
natural aggregates in road construction. For RAP aggregates this conclusion opposes the
given conclusion when total content of organic contaminants is considered.
Table 3: Contents of chemical species in eluates prepared according to EN 12457-4 [4]
Parameter
[mg/kg, dry matter]
Cadmium, Cd
Chromium, Cr
Lead, Pb
Zinc, Zn
Copper, Cu
Nickel, Ni
Chloride, ClSulphate, SO42Dissolved organic carbon, DOC
pH
CMC
CRAP
MRAP
Limestone
<0.005
0.126
<0.024
0.010
0.045
<0.026
44
267
46
11.29
<0.005
0.046
<0.024
<0.006
0.030
<0.026
35
124
71
10.95
<0.005
<0.010
<0.024
<0.006
<0.014
<0.026
35
13
94
9.67
<0.005
<0.010
<0.024
0.035
<0.014
<0.026
44
145
71
7.90
Leaching
limits
0.04
0.5
0.5
4
2
0.4
800
1000
500
The content of potential contaminants in the eluates prepared according to EN 1744-3[7} is

shown in Table 4. The environmental risk of using EN 1744-3 [7] in a comparative context to
EN 12457-4 [4] put in evidence the higher zinc, sulphate and chloride release in the former
test. These differences were assigned to the influence of the pH and oxidation-reduction
potential on controlling the solubility for different particle size distribution of the materials
under test. RAP aggregates show lower release against crushed mixed concrete and a very
similar release to the natural aggregate. This behaviour, underestimation of release, is in
agreement with Van der Sloot and Mulder analysis regarding EN 1744-3 [8].

3.2
Lysimeter leaching tests
Table 5 lists the cumulative content of the analytes in the leachates from lysimeters. Lower
and upper limits of release are reported for specific components whenever their
concentrations in the leachate fractions were below the limit of detection.
Table 4: Contents of chemical species in eluates prepared according to EN 1744-3 [7]
Parameter
[mg/kg, dry matter]
Cadmium, Cd
Chromium, Cr
Lead, Pb
Zinc, Zn
Copper, Cu
Nickel, Ni
Chloride, ClSulphate, SO42pH
CMC
CRAP
MRAP
Limestone
<0.005
0.103
<0.024
0.035
0.022
<0.026
53
343
9.96
<0.005
0.022
<0.024
0.023
0.016
<0.026
53
163
8.12
<0.005
<0.010
<0.024
0.053
<0.014
<0.026
35
18
8.14
<0.005
<0.010
<0.024
0.021
<0.014
<0.026
35
159
6.86
It should be mentioned that milled RAP was combined with limestone aggregate to be in
accordance with the composition used in the road pilot section constructed during the
SUPREMA project. The selected ratio, 30% MRAP/70% limestone, allowed fulfilling the
grading envelope of the Portuguese Road Administration.
Table 5: Cumulative contents of chemical species in leachates from lysimeters (L/S =5.0)
Parameter
[mg/kg dry matter]
Cadmium, Cd
Chromium, Cr
Lead, Pb
Zinc, Zn
Copper, Cu
Nickel, Ni
Chloride, ClSulphate, SO42Dissolved organic carbon, DOC
Hydrocarbons (C10-C40)
pH
CMC
CRAP
MRAP/Limestone
Limestone
0 - 0.003
0.083
0 - 0.005
0 - 0.011
0 - 0.007
0 - 0.016
58
627
21
7.20-7.63
0 - 0.003
0 - 0.025
0 - 0.005
0 - 0.010
0.009
0 - 0.015
47
295
37
0 - 0.15
7.35-7.66
0 - 0.003
0 - 0.025
0 - 0.005
0 - 0.011
0 - 0.005
0 - 0.015
32
57
14
0 - 0.15
7.33-7.74
0 - 0.003
0 - 0.025
0 - 0.005
0 - 0.018
0 - 0.005
0 - 0.016
37
272
7
7.27-7.76
The most relevant factors regarding leaching from lysimeters are the pH, the redox
conditions, the infiltration rate and volume, the exposed surface area and compaction. Table 5
shows the lower pH on the leachates from the lysimeters when compared to the eluates from
the batch leaching tests which affect the release. In what concerns heavy metals, it is
noteworthy that chromium is the metal showing higher content in the leachate from CMC,
though lower for the batch leaching tests, which is compatible with lower leachate pH. It was
also found a slight mobilization of copper in the lysimeter with CRAP, probably assigned to

the high content of dissolved organic carbon of this aggregate. For MRAP similar effect was
not seen probably due to the mixture of this aggregate with limestone
For the lysimeter with crushed mixed concrete sulphate is again critical showing higher
content when compared to batch leaching tests. The leaching in lysimeters in the absence of
the wearing layer corresponds to the worse situation whereby the increased release of soluble
anions should not be regarded as the real behaviour in the field. According to Reid et al. [13],
the wearing layer could reduce the water infiltration in 90%, therefore in undamaged
pavements it is expected a significant decrease of sulphate leaching, corresponding to a lower
environmental impact.
Towards clarifying the problems that could be related to the high total content of
hydrocarbons C10-C40, an evaluation of these organic contaminants in the leachates from
lysimeters containing RAP aggregates have shown that the release is always very low, below
the detection limit (50 g/l), corresponding to maximum leached content from 0 to
0.15 mg/kg. The released C10-C40 can represent up to 2% of the dissolved organic carbon
which indicates that most part of the leached organic compounds did not come from the
fraction C10-C40. In the literature similar low values of release of these hydrocarbons can be
found for end-of-life tyres used as granulates in several civil engineering works [14].
Moreover, it is important to stress that the assessment of organic pollutants based on content
is due to the lack of European leaching standards for these chemical species Therefore, it does
not seem acceptable to restrict the use of RAP aggregates based on the fail of the criterion for
the total hydrocarbon C10-C40 content when they pass all the remaining criteria for waste
admissible in landfills for inert waste, as well as the established criteria in the Portuguese
technical specifications for CDW to be used as recycled aggregates.
4.
CONCLUSIONS
Based on the results from the leaching test, stated in the Portuguese Technical
specifications, the release of contaminants from aggregates to be used in unbound pavement
layers was higher for CDW aggregates when compared to natural aggregates. Nevertheless,
the contaminants contents shows the feasibility of applying the different types of CDW since
the chemical species analysed in the eluates comply with the limits for waste to be disposed in
landfills for inert waste.
Regarding leachates from lysimeters, containing unbound recycled aggregates without a
bituminous layer as wearing course, i.e., representing a totally damaged pavement owing to
traffic and climate solicitations, the results for a ratio L/S = 5 show low release of
contaminants. An estimation of the release for L/S=10 leads to contaminants contents below
the admissibility criteria for waste into landfill for inert waste, in accordance to the
laboratorial leaching test EN 12457 [4]. For intact pavements the contact with water occurs
mainly by the shoulders of the road and a substantial decrease of leaching is foreseen, thereby
strengthening the abovementioned viability of using the different recycled aggregates.
The critical constituents in the batch leaching tests were similar to those found in the
leachates from lysimeters: chromium and sulphate for crushed mixed concrete aggregates and
sulphate and dissolved organic carbon for reclaimed asphalt aggregates
In what concerns RAP aggregates, the high content of hydrocarbons C10-C40 had a minor
release in leachates from lysimeters, which represents up to 2% of the dissolved organic
carbon, and it should not hinder aggregates use. It is essential to establish leaching limits for
organic pollutants to avoid restrictions based on the total contents since the released fraction
maybe a very minor portion of the total content.

ACKNOWLEDGEMENTS
The authors acknowledge the financial support of FCT from Portuguese Ministry of
Economy and Employment to the project SUPREMA Sustainable application of
Construction and Demolition Recycled Materials (C&DRM) in Road Infrastructures, and the
Ambigroup SGPS SA and Teodoro Gomes Alho SA for materials supply.
REFERENCES
[1] OJEU, Regulation (EU) No 305/2011 of the European Parliament and of the Council laying
down harmonised conditions for the marketing of construction products and repealing Council
Directive 89/106/EEC. Official Journal of European Union, L088, 0004 0043, 2011.
[2] OJEU, Directive 2008/98/EC of the European Parliament and the Council of 19 November 2008
on Waste and Repealing certain Directives. (Text with EEA relevance). Official Journal of
European Union L312/3, 22.11.2008, 2008.
[3] Saveyn, H., Eder, P., Garbarino, E., Muchova, L., Hjelmar, O., Van der Sloot, H., Comans, R.,
Van Zomeren, A., Hyks, J. and Oberender, A., Study on methodological aspects regarding limit
values for pollutants in aggregates in the context of the possible development of end-of-waste
criteria under the EU Waste Framework Directive. JRC-IPTS, 2014.
[4] EN 12457-4. Characterization of waste. Leaching. Compliance test for leaching of granular
waste materials and sludges- Part 4: One stage batch test at a liquid to solid ratio of 10l/kg for
materials with particle sizes below 10mm (without or with size reduction). Comit Europen de
Normalisation, 2002.
[5] A.C. Freire, J. Neves, A. Roque, I.M. Martins, M.L. Antunes, G. Faria, Sustainable application
of construction and demolition recycled materials (C&DRM) in road infrastructures, in
Proceedings of the 1st International Conference on WASTES: Solutions, Treatments and
Opportunities, pp. 23-29. Guimares, Portugal, 2011.
[6] J. Neves, A.C. Freire, A. Roque, I.M. Martins, M.L. Antunes, G. Faria, Use of construction and
demolition recycled materials (C&DRM) in road pavements validated on experimental test
sections, in Proceedings of the 9th International Conference on the Bearing Capacity of Roads,
Railways and Airfields. Trondheim, Norway, 2013.
[7] EN 1744-3. Tests for chemical properties of aggregates - Part 3: Preparation of eluates by
leaching of aggregates. Comit Europen de Normalisation, 2002.
[8] Van der Sloot HA, Mulder E., Test methods to assess environmental properties of aggregates in
different applications: The role of EN 1744-3. Energieonderzoek Centrum Nederland, ECN;
2002.
[9] EN 933-11. Tests for geometrical properties of aggregates - Part 11: Classification test for the
constituents of coarse recycled aggregate. Comit Europen de Normalisation, 2009.
[10] Olsson, S., Gustafsson, J.P., Berggren Kleja, D., Bendz, D. and Persson, I., Metal leaching from
MSWI bottom ash as affected by salt or dissolved organic matter. Waste Management, 29(2): p.
506-512. 2009.
[11] Mangani, G., Berloni, A., Bellucci, F., Tatno, F. and Maione M., Evaluation of the Pollutant
Content in Road Runoff First Flush Waters. Water, Air, and Soil Pollution, 160(1-4): p. 213-228.
2005.
[12] Galvn, A.P., Ayuso, J., Garca, I., Jimnez, J.R. and Gutirrez, F., The effect of compaction on
the leaching and pollutant emission time of recycled aggregates from construction and demolition
waste. Journal of Cleaner Production, 83: p. 294-304. 2014.
[13] Reid, J.M., Evans, R.D., Holnsteiner, R., Wimmer, B., Gaggl, W., Berg, F., Pihl, K.A., MilvangJensen, O., Hjelmar, O., Rathmeyer, H., Francois, D., Raimbault, G., Johansson, H.G.,
Hkansson, K., Nilsson and U., Hugener, M. ALT-MAT: Alternative materials in road
construction. Final Project Report, Transport Research Laboratory, Crowthorne, 2001.

[14] Moretto, R. Environmental and health evaluation of the use of elastomer granulates (virgin and
from used tyres) as filling in third-generation artificial turf. Aliapur EEDEMS Fieldturf Tarkett,
2007.

RECYCLED FINE AGGREGATE IN NEW CONCRETE

A. Katz (1), D. Kulisch (1)
(1) National Building Research Institute, Faculty of Civil and Environmental Engineering,
Technion Israel Institute of Technology
Abstract
The quantities of construction and demolition waste (C&DW) are getting larger and larger
nowadays as the construction market increases in the modern world. Many studies have
proved the possibility of incorporating the coarse fraction of the waste obtained after sorting,
crushing and treating. Replacing 10-20% of the virgin coarse aggregate by coarse recycled
aggregate is acceptable in many standards without damaging the properties of the new
concrete, or with minor effect that can easily be corrected with slight increase of the cement
content. As for the fine fraction of C&DW, there is a wide consensus that its effect on the
properties of new concrete is more significant and the risk of damaging the properties of the
new concrete is high. The study presented here examines the effect of replacing recycled fine
aggregate (RFA) on the properties of new concrete. Aggregate was obtained from two
regional recycling facilities thus they represent mixed and non-uniform waste. The results
showed that the recycled fine aggregates are more porous and weaker compared to natural
aggregate obtained by crushing natural stone. It also may contain some impurities that affect
the hydration of the new concrete. When all the fine aggregate was replaced, the compressive
strength has reduced by 34%-53% in concrete having low water to cement ratio (w/c=0.40)
and by 19%-53% for higher w/c of 0.60. When only 30% of the aggregates were replaced by
RFA the reduction was 3%-29% for w/c=0.40 and 9%-29% for w/c=0.6. Other properties also
showed inferior results compared with the reference concrete prepared with virgin aggregates.
Key words: construction and demolition waste; recycled fine aggregate; concrete.
1.
INTRODUCTION
Construction activities are responsible for several environmental issues, including resource
use, generation of construction and demolition waste (C&DW), soil contamination and
different forms of pollution. C&DW corresponds to a large portion of solid waste. In Spain,
70% of total waste comes from construction activities; in the United Kingdom, over 50%; in
Australia, 44%; in Hong Kong, 38%; in Japan, 36%; in USA, 29% [1].
Concrete cycle results in three basic environmental problems: use of high amounts of raw
materials, consumption of high amounts of energy, and creation of big volumes of waste [2].

One of the raw materials used in concrete production is sand. Natural sand can be extracted
from pits or terrace sources, river sources, beach deposits, dune deposits, dredges sources
(marine source) and hard rock quarries [3].
Shortage of good quality natural sand and the environmental pressure to minimize sand
extraction from rivers have increased the interest of the construction industry to find suitable
alternatives to replace fine aggregate [4]. Considering the use of fine aggregate from recycled
concrete as a replacement for natural sand, the advantages are economic aspects, reduction of
environmental impacts and saving resources. Recycling minimizes solid waste, which saves
landfill spaces (extend the useful life), reduces consumption of resources (preserve finite
natural resources), and reduces environmental damages caused by incorrect disposal [1].
This paper reports how recycled aggregate affect mortar and concrete when used as fine
aggregate.
2.
EXPERIMENTAL RESEARCH PROGRAM
2.1
Fine aggregate
The recycled fine aggregate used in the study was obtained from two different recycling
plants in Israel: one is located in Herzliya (central) and the other is located in Kyriat Ata
(north), according to Table 1. These aggregates represent mixed and non-uniform waste from
C&DW. In this study, only the fine portion of each aggregate was used (< 4.75 mm).
Table 1: Reference of each recycled material
Herzliya - Fine aggregate resultant from beneficiation process
Recycled aggregate 1
(crushing; double sieving, with and without air; magnetic separation;
(RA-1)
washing; hydrocyclone)
Herzliya - Coarse aggregate resultant from beneficiation process
Recycled aggregate 2
(crushing; double sieving, with and without air; magnetic separation;
(RA-2)
washing; secondary crushing)
Recycled aggregate 3 Herzliya Aggregate resultant from the first part of beneficiation
(RA-3)
process (crushing and double sieving, with and without air)
Recycled aggregate 4 Kiryat Ata - Aggregate resultant from the process (sieving; air knife;
(RA-4)
crushing and sieving)
2.2
Mortar mix
The materials used to produce mortar specimens were: cement CEM I 52.5 N, water and
sand (four types of recycled sand and crushed sand for reference). Two different water/cement
ratios are tested (0.4 and 0.6). Total replacement (100%) and partial replacement (30%) of
crushed sand by recycled sand are compared. The composition of each mix is described in
Table 2.
2.3
Testing
Laboratory tests of recycled aggregates were performed, including content of fines, organic
impurities, absorption capacity, specific gravity (oven-dry and saturated-surface-dry). In fresh
state, setting time was evaluated. In hardened state, properties evaluated were compressive
strength [5] and Torrent air permeability [6]. All the tests were performed according to the
relevant ASTM standard.

Table 2: Mortar mix composition (kg/m3)

Replacement
ratio
REF
0.4
0%
RA-1
0.4
30%
RA-2
0.4
30%
RA-3
0.4
30%
RA-4
0.4
30%
RA-1
0.4
100%
RA-2
0.4
100%
RA-3
0.4
100%
RA-4
0.4
100%
REF
0.6
0%
RA-1
0.6
30%
RA-2
0.6
30%
RA-3
0.6
30%
RA-4
0.6
30%
RA-1
0.6
100%
RA-2
0.6
100%
RA-3
0.6
100%
RA-4
0.6
100%
* Aggregates in SSD condition.
Aggregate
3.
w/c
Cement
Water
667
656
656
645
648
622
611
594
599
603
598
593
591
577
567
553
530
517
267
263
262
258
259
249
245
238
240
362
359
356
355
346
340
332
318
310
Recycled
aggregate*
394
393
387
389
1245
1223
1189
1198
359
356
355
346
1133
1106
1061
1035
Crushed
sand
1335
919
918
903
908
1207
837
830
828
808
-
Characterization of the recycled material

3.1
Properties of the recycled material are presented as follows (Table 3). All the tests were
performed according to the relevant ASTM standard.
Aggregates RA-1 and RA-2 present lower content of fines (7.6% and 0.85%, respectively),
while RA-3 and RA-4 did not pass a washing phase thus present a higher content of fines
(18.2% and 25.1%, respectively). It becomes clear that the washing part of the process
reduces significantly the content of fines.
RA-1, RA-2 and RA-3 present low amounts of organic impurities while RA-4 presents
significantly high amount of organic impurities. Considering that RA-1, RA-2 and RA-3 are
originated from the Herzliya recycling plant and RA-4 is originated from the Kyriat Ata
recycling plant, we can see that each recycling plant receives different material from different
places (central portion and north portion of the country). Also, the initial separation process,
before the sieving, is more effective in Herzliya than in Kyriat Ata, where more organic
matter is retained with the material (RA-4).
RA-1 presents a similar absorption value to the reference, while RA-2, RA-3 and RA-4
present higher values. The higher absorption capacity leads to a higher porosity, which may
be originated from the old cement paste in the recycled material as was reported by others [510]. The high absorption value of RA-3 is intriguing, because RA-1 and RA-2, which are
sourced from RA-3, present lower values. The high value of RA-3 could mean that the
washing phase of the process removes porous material, together with fines. Also this could
mean that a portion of the washed fines is composed of old cement paste, which is more

porous. It should be remembered that it is difficult to determine accurately absorption

capacity of fines, which may present some inaccuracy in the values of RA-3 and RA-4.
Table 3: Properties of the recycled materials and reference
RA-2
RA-3
RA-4
Crushed sand
(REF)
Content of material finer

7.6
than 75 m (%)
0.85
18.2
25.1
12.4
Organic impurities
Lighter
Lighter
Lighter
Darker
Absorption capacity (%)
2.5
6.6
8.7
7.8
2.4
Specific gravity OD
2.48
2.22
2.13
2.17
2.60
Specific gravity SSD
2.55
2.37
2.32
2.34
2.65
RA-1
All recycled materials presented lower specific gravity than the reference, both on ovendry (OD) state and saturated-surface-dry (SSD) state.
Normalized setting time
3.2
Fresh state properties
Figure 1 presents normalized setting time with respect to reference.
Setting time
1.80
1.60
1.40
1.20
1.00
0.80
0.60
0.40
0.20
0.00
REF
RA-1
RA-2
RA-3
RA-4
REF
w/c = 0.4
RA-1
RA-2
RA-3
RA-4
w/c = 0.6
Initial
Final
Figure 1: Initial and final setting times, normalized to the reference mix.
Recycled aggregate have an influence on the setting time, both initial and final. For both
w/c (0.4 and 0.6), all the times are higher, which means that recycled aggregate have a
retarding effect on the setting time. This increase may be related to the presence of organic
impurities.

For low w/c (0.4), RA-1 and RA-2 present an increase of 19%-23% in initial setting time
and 4%-6% in final setting time. RA-3 presents an increase of 40% and 32% for initial and
final setting time, respectively. RA-4 presents a higher increase (63% and 43% for initial and
final setting time, respectively). For high w/c (0.6), RA-1 presents an increase of 18%-23% in
initial and final setting time. RA-2 and RA-3 present an increase of 20% in initial setting time
and 32%-37% in final setting time. RA-4 presents the highest increase (43%-55%).
3.3
Hardened state properties
Compressive strength values are presented in Table 4. The test was performed according to
the SI 1 part 1 [5].
Table 4: Compressive strength
w/c = 0.4
w/c = 0.6
REF
RA-1 30%
RA-1 100%
RA-2 30%
RA-2 100%
RA-3 30%
RA-3 100%
RA-4 30%
RA-4 100%
REF
RA-1 30%
RA-1 100%
RA-2 30%
RA-2 100%
RA-3 30%
RA-3 100%
RA-4 30%
RA-4 100%
Compressive strength (MPa)

3 days
7 days
28 days
56.5
64.6
73.8
43.7
63.6
71.7
35.0
42.6
48.8
42.4
52.6
62.5
23.8
32.0
39.7
42.1
51.4
55.5
23.0
30.0
40.3
41.5
48.3
52.7
22.2
25.0
34.4
24.6
33.3
47.2
22.4
33.3
42.9
19.2
26.6
38.2
27.0
35.3
43.7
13.7
19.3
30.1
25.0
31.9
40.2
13.7
15.9
25.9
23.7
31.4
33.5
12.0
14.4
22.0
Change (%)
D 3 days D 7 days
-23%
-2%
-38%
-34%
-25%
-19%
-58%
-50%
-26%
-20%
-59%
-54%
-27%
-25%
-61%
-61%
-9%
0%
-22%
-20%
+10%
+6%
-44%
-42%
+2%
-4%
-44%
-52%
-4%
-6%
-51%
-57%
D 28 days
-3%
-34%
-15%
-46%
-25%
-45%
-29%
-53%
-9%
-19%
-7%
-36%
-15%
-45%
-29%
-53%
The results from compressive test are presented in Table 4 together with the rate of change
relative to the reference mix. Figure 2 presents the normalized compressive strength with
respect to reference. For low w/c (0.4), compressive strength is reduced for all types of
aggregates and for both partial and total replacement. For high w/c (0.6) the reduction was
less pronounced. It is noted that for low w/c (stronger new cement paste), the influence of the
recycled aggregate is stronger than for high w/c (weaker new cement paste). Considering 28
days, for low w/c, it can be noticed that partial replacement of RA-1 presents no significant
impact while the replacement of RA-2, RA-3 and RA-4 presents a reduction of 15%-29%.
The total replacement however presents a more significant decrease (34%-53%) for all

aggregates, as expected. For high w/c, RA-1, RA-2 and RA-3 present a smaller reduction
(7%-15%) for partial replacement, and RA-4 presents a higher decrease (29%). For total
replacement, RA-1 presents a reduction of 19% and higher reduction values are presented for
RA-2, RA-3 and RA-4 (36%-53%).
Normalized compressive strength
Compressive Strength
1.2
1
0.8
0.6
0.4
0.2
0
REF RA-1 RA-2 RA-3 RA-4 RA-1 RA-2 RA-3 RA-4 REF RA-1 RA-2 RA-3 RA-4 RA-1 RA-2 RA-3 RA-4
0
30
100
0.4
3 days
0
7 days
30
28 days
100
0.6
Figure 2: Normalized compressive strength (with respect to reference).

Figure 3 presents normalized air permeability with respect to reference. Mortar samples
made with recycled aggregates present higher air permeability values than with natural
aggregates (Figure 3). For low w/c (0.4) RA-1 presents an increase of 31% for partial
replacement and 75% for total replacement. The increase of RA-2 is 82% for partial
replacement and almost 600% for total replacement. RA-3 presents an increase of 37% for
partial replacement and 147% for total replacement. The increase of RA-4 is 15% for partial
replacement and 76% for total replacement. For high w/c (0.6), RA-1 presents an increase of
19%-23%. The increase of RA-2 is 30% for partial replacement and 193% for total
replacement. RA-3 presents an increase of 21% for partial replacement and 114% for total
replacement. The increase of RA-4 is 33% for partial replacement and almost 200% for total
replacement.
The significantly high increase of air permeability with RA-2 in the low w/c mix can be
explained by the coarse granulometry, low content of fines in this aggregate and much larger
amount of old cement paste, thus increasing the permeability. In low w/c mix, RA-2 is much
more porous than surrounding new cement paste, thus creating a continuous path for air
permeation. This phenomenon is less pronounced in the high w/c.

Air permeability
6
5
4
3
2
1
w/c = 0.4
RA-4 100%
RA-4 30%
RA-3 100%
RA-3 30%
RA-2 100%
RA-2 30%
RA-1 100%
RA-1 30%
REF
RA-4 100%
RA-4 30%
RA-3 100%
RA-3 30%
RA-2 100%
RA-2 30%
RA-1 100%
RA-1 30%
0
REF
Normalized air permeability
w/c = 0.6
Figure 3: Normalized Air permeability (with respect to reference).

4.
CONCLUSIONS
Recycled fine aggregates are more porous and weaker, presenting higher absorption and
lower specific gravity compared to natural aggregate obtained by crushing natural stone. It
may also contain some impurities that affect the hydration rate of the new concrete, as organic
matter.
When all the fine aggregate was replaced, the 28 days compressive strength has reduced by
34%-53% (w/c=0.40) and by 19%-53% (w/c=0.60). When only 30% of the aggregates were
replaced by RFA the reduction was 3%-29% for low w/c and 9%-29% for high w/c.
Other properties also showed inferior results compared with the reference concrete
prepared with virgin aggregates, as increased setting time and air permeability.
REFERENCES
[1] Tam, V.W.Y and Tam, C.M., Re-use of construction and demolition waste in housing
developments, (Nova Science Publishers, Inc, New York, 2008).
[2] Oikonomou, N.D., Recycled concrete aggregates, Cement and Concrete Composites, 27 (2005)
315-318.
[3] Alexander, M.G. and Mindess, S., Aggregates in concrete (Modern Concrete Technology, 13).
(Taylor & Francis Group, London and New York, 2005).
[4] Elavenil, S. and Vijaya, B., Manufactured Sand, a solution and an alternative to river sand and in
concrete manufacturing, Journal of Engineering, Computers & Applied Sciences (JEC&AS), 2 (2)
(2013) 20-24.
[5] SI 1 part 1. Cement: Common cements, 2010.
[6] Torrent, R. J., A two-chamber vacuum cell for measuring the coefficient of permeability to air of
the concrete cover on site, Materials and Structures, 25 (6) (1992) 358-365.
[7] Tam, V.W.Y., Gao, X.F., Tam, C.M., Microstructural analysis of recycled aggregate concrete
produced from two-stage mixing approach, Cement and Concrete Research, 35 (2005) 1195-1203.
[8] Tam, V.W.Y., Tam, C.M., Le, K.N., Removal of cement mortar remains from recycled aggregate
using pre-soaking approaches, Resources, Conservation and Recycling, 50 (1) (2007) 82101.

[9] Juan, M.S. de, Gutierrez, P.A., Study on the influence of attached mortar content on the properties
of recycled concrete aggregate, Construction and Building Materials, 23 (2009) 872877.
[10] Ismail, S., Ramli, M., Engineering properties of treated recycled concrete aggregate (RCA) for
structural applications, Construction and Building Materials, 44 (2013) 464476.
[11] Duan, Z.H. and Poon, C.S., Properties of recycled aggregate concrete made with recycled
aggregates with different amounts of old adhered mortars, Materials and Design, 58 (2014) 19-29.
[12] Akbarnezhad, A., Ong, K.C.G., Zhang, M.H., Tam, C.T., Foo, T.W.J., Microwave-assisted
beneficiation of recycled concrete aggregates, Construction and Building Materials, 25 (2011) 3469
3479.

STUDY ON THE RECYCLING METHOD OF PVC BUILDING WASTES

Kohei Yamaguchi (1) and Akio Koyama (2)
(1) Graduate School of Science and Technology, Meiji Univ., Japan
(2) Dept. of Architecture Faculty of Science and Technology, Meiji Univ., Japan Dr.Eng
Abstract
It is common that building finishing materials that have durability and design properties are
essential for buildings. In recent, maintenance and repair-work except the new construction of the
building increases, and it is expected that the amount used of the building finishing materials increase
in future. However, it remains to be elucidated the recycling method for the building finishing wastes
especially the PVC building wastes that have the majority on the building finishing materials. The
main reason of low recycling rate is that the PVC building wastes are usually discharged as mixed
waste with impurities. Therefore, we aim to develop the recycling method for PVC building wastes.
Here we found the new method that is able to separate recycling PVC powder from PVC building
wastes. Specifically, it targeted tile carpets, wallpapers and PVC waterproof sheet. Moreover, here we
made the recycling sheets by the PVC recycling powder, and evaluated the quality of the recycling
sheets. Finally, we evaluated that the effect of the new recycling method can reduce damage of
environment and energy use. These observations indicate that the new recycling method is able to
reduce PVC building wastes and intend for practical use. These results are not only reducing the
amount of the building wastes, but also contributing for a sustainable society.
Keywords: PVC resin, Recycle, PVC building wastes, Separation apparatus, Recycled PVC
sheet, LCA evaluation
1.
INTRODUCTION
Vinyl chloride resin (PVC) building materials that have printability, workability and durability
are used as multi-use for wallpaper, tile carpet, and so on. Therefore the materials that let the
building enhance design and long-life are typical of building finishing materials [1]. However,
on the other hand, the recycling rate of the PVC building materials in Japan are very low
compared with the other building materials (for instance concretes, woods, asphalt concretes)
[2]
. The fundamental cause is that the materials are discharged as mixing wastes with the other
kind of materials. For instance, if a wallpaper is discharged, it is discharged as mixing waste
with PVC material and pulp, thus it is so difficult to recycle these materials together. From now
on, the amount of the building finishing materials discharged will increase with expansion of
reform business by long-life building. Hence it is essential for the global environmental problem
to develop the recycling method of the PVC building wastes.
From the above mentioned background, we aim to develop the recycling method for PVC
building wastes. Therefore, here we developed the new method that was able to separate PVC

recycled powder from the mixing wastes. Moreover, here we made recycled sheet by the PVC
recycled powder, and evaluated the quality of the recycled sheet. In addition, we evaluated the
effect of the new recycled method by life cycle assessment to reduce environment impact and
energy use.
2.
THE SEPARATION METHOD
2.1
The outline of the separation method
First, hitting the recycling of the PVC building wastes, it must be the separation of the mixing
wastes. Therefore, here we develop the new separation apparatus. The apparatus has the unit
which is able to do pounding and untying the wastes by high-speed rotation [3], [4]. Fig.1
shows the separation apparatus. The separation method are as follows:
1 Rough crushing PVC building wastes to 30 or less
2 Putting the crushed PVC building wastes in the separation apparatus that high speed turned
a rotary tool at 130m per second
3 Letting the PVC building wastes finer to 300 or less by frictional force between the
wastes and the apparatus inner wall surface with centrifugal force (the finer PVC wastes is
separated PVC recycled powder and other materials by relative density)
4 Putting aspirator in the separation apparatus makes separating efficient and continuous
Fig.1 : Separation apparatus

2.2
The quality of the PVC recycled powders
Table.1 shows the grain size of the PVC recycled powder by the separation apparatus, Fig.2
shows the some kinds of separated PVC building wastes. As a result, the separation method
was able to separate just PVC powder from the PVC building wastes, the mixture rate of other
materials was less than 1%.
Table 1 : Grain size of the PVC recycled powder
Peripheral velocity
(m/s)
Grain size ()
300 Gs
200 Gs 300
100 Gs 200
Gs 100
110
130
150
170
90
5
4
1
5
30
50
15
3
20
60
17
1
10
70
19

Fig.2 : Separated PVC building wastes

3.
THE PVC RECYCLED SHEETS
3.1
Experimental procedures
3.1.1 The materials

The PVC recycled powders for this study were four kinds of the PVC building wastes origin,
Tile carpets (called as T), Wallpapers (called as W), Waterproof sheets (called as P) and pipes
(called as H). These PVC recycled powders were made finer to 300 or less by the separation
method.
The virgin PVC powder (called as V) for this study were compounded with PVC resin (degree
of polymerization is 1300), plasticizer (phthalic ester) and filler (calcium carbonate). The
compounding proportion (weight) of the virgin PVC powder is that, PVC resin is 56%,
plasticizer is 33% (phr : 60%) and filler is 11% (phr : 20%).
3.1.2 The Compounding of the PVC recycled powders
Table.2 shows the compounding of the PVC recycled powders for the PVC recycled sheets.
The recycled sheets which made of one kind of the PVC recycled powders with the virgin PVC
powder were as series 1, the recycled sheets which made of two and more kind of the PVC
recycled powders with the virgin PVC powder were as series 2.
Table 2 : The compounding of the PVC recycled powders
Ser.
sample name
V100
R25
ser.1 R50
R75
R100
R20-R10
R15-R15
R10-R20
R40-R20
R30-R30
R20-R40
ser.2
R67-R33
R50-R50
R33-R67
R10-R10-R10
R20-R20-R20
R33-R33-R33
recycling rate
combine rate (%)
virgin R
R
R
(%)
0
100
25
75
25
50
50
50
75
25
75
100
0
100
30
70
20
10
30
70
15
15
30
70
10
20
60
40
40
20
60
40
30
30
60
40
20
40
100
0
67
33
100
0
50
50
100
0
33
67
30
70
10
10
10
60
40
20
20
20
100
0
33
33
33
remarks
R = Tile carpet, Wallpaper,
Waterproof sheet or Hard PVC
(R , R) = (Tile carpet, Wallpaper)

or (Waterproof sheet, Hard PVC)
R = Tile carpet
R = Wallpaper
R = Waterproof sheet

3.1.3 The PVC recycled sheets productive method

Fig.3 shows PVC recycled sheets productive method. The recycled PVC sheets are as follows:
1 Compounding some PVC recycled powder to a predetermined ratio
2 Putting the compounded PVC powder into mixing roll machine set to 160
3 Kneading and mixing the PVC powder until it becomes a sheet and thickness 0.5
Fig.3 : Recycled PVC sheets productive method

3.1.4 Test procedures
Table.3 shows the test items and methods. The purpose of the PVC recycled sheets are
waterproof sheet or carpet, therefore, these tests evaluate the application to them.
Table 3 : The test items and methods
Test item
Test method
Tensile properties
JIS K 6251
Weathering
JIS A 1415
3.2
Specification
Tensile strength : 1000N/ and more
Elongation at break : 200% and more
Tensile strength : 80% and more
Elongation at break : 70% and more
Results and discussions
3.2.1 Tensile properties

1) Series 1
Fig.4 shows the effect of each compounding (series 1) on tensile properties. As the recycling
rate increases, the tensile strength of the recycled sheets decreased proportionally except the
recycled sheets made of pipes, the elongation at break of the recycled sheets decreased
proportionally except the recycled sheets made waterproof sheets. It is assumed that these
results were from the qualities of the PVC recycled powders. For instance, pipes are made of
hard PVC resin, waterproof sheets have much content of the plasticizer, therefore, the recycled
sheets made of pipes are hard and the recycled sheets made of waterproof sheets are highelongation. Basically, PVC building materials are composition with PVC resin, a plasticizer
and filler, however, each of the PVC recycled powder is composition with different
compounding rate, it is also the reason why each of the quality of the recycled sheets was
different. In addition, the recycled sheets were able to be increased tensile properties by mixing
with the virgin PVC powder. In addition, as the recycling rate of each compounding, 25% and
less in mixing the recycled powder of the tile carpets, 50% and less in the wallpapers, 100%
and less in the waterproof sheet, 75% and less in the pipes meet the JIS standard.

5000
1000
tensile strength
elongation at break
800
3000
600
2000
400
1000
200
elongation at break (%)
tensile strength (N/)
4000
Fig.4 : Effect of each compounding on tensile properties (series 1)

: The dotted line, JIS standard
2) Series 2
Fig.5 shows the relationships between the recycling rate and tensile properties in series 2. If the
recycled sheets made of two difference kinds of the PVC recycled powders, the tensile property
of the recycled sheets was between two of them. Similarly, the recycled sheet made of three
difference kinds of the PVC recycled powders, the tensile property of the recycled sheet was
between three of them too. Moreover, in the compounding of the multiple kind of the powders,
as the compounding rete of one of the powder increases, the quality of the recycled sheet is
biased to one. As a result, it was not declined for the tensile properties of the recycled sheet to
mix two and more difference kinds of the PVC recycled powder.
700
5000
Waterproof sheet (P)
P:H=2:1
P:H=1:1
P:H=1:2
PVC pipe (H)
3000
600
500
4000
2000
1000
0
0
20
40
60
recycling rate (%)
80
300
200
100
100
5000
20
40
60
recycling rate (%)
20
40
80
100
60
80
recycling rate (%)
100
700
Tile carpet (T)
Wallpaper (W)
Waterproof sheet (P)
T:W:P=1:1:1
3000
600
500
4000
400
2000
1000
0
20
40
60
80
recycling rate (%)
100
400
300
200
100
Fig.5 : Relationships between the recycling rate and tensile properties (series 2)

3.2.2 Weathering test

Fig.6 shows relationships between each compounding and the proportion of the tensile
properties of weathering test against the default. As the weathering test, the tensile strength and
the elongation at break of the recycled sheets decreased. As the recycling rate increases, the
proportion of the tensile properties increased. The reason of the relation is suspected that depend
to the virgin sheet that has high-tensile property and the recycled powders have already had
some stabilizer for durability [5]. In addition, as the recycling rate of each compounding, 75%
and more in mixing the recycled powder of the tile carpets, 25% and less in the wallpapers and
the waterproof sheet, 25% and more 75% and less in the pipes meet the JIS standard.
120
tensile strength
elongation at break
130
110
100
90
80
70
60
50
40
30
20
10
elongation at break proportion (%)
tensile strength proportion (%)
140
Fig.6 : relationships between each compounding and the proportion of the tensile properties
(weathering test) : The dotted line, JIS standard
4.
LCA EVALUATION
4.1
Evaluation procedures
4.1.1 Case study

Focus production stage of the raw materials that is a part of the life cycle, this study evaluated
environmental influence of production stage of the PVC recycled powder with this separation
method. Here we calculated energy consumption and CO emissions about the PVC recycled
powder (made of wallpaper) and the virgin PVC powder.
4.1.2 Areas and conditions of analysis
Fig.7 shows the areas and conditions of analysis about making the PVC recycled powder, Fig.8
shows the areas and conditions of analysis about making the virgin PVC powder. The areas and
conditions of analysis about each PVC powder are as follows:
1) The PVC recycled powder
Collecting wallpaper wastes 1000t
The yield of the PVC recycled powders is 60% (From the measurement value)
Making 600t of the PVC recycled powders made of the wallpapers wastes
The power consumption of the separation apparatus is 513kwh
2) The virgin PVC powder
Purposing of the virgin PVC sheet to PVC waterproof sheet
Compounding the virgin PVC powder as PVC waterproof sheet
Making the virgin PVC powder 600t

Recycle
Areas of analysis
Transport 1
100
Deconstruction
Wallpaper : 1000t
1st disposing
Rough crush
Transport 2
50
Recycled PVC powder

600t
2nd disposing
The separating system
Pulp 400t
Fig.7 : The areas and conditions of analysis

(making the PVC recycled powder)
Virgin
Table 4 : Energy consumption basic unit

and CO emission coefficient
Areas of analysis
PVC resin : 45%
Virgin PVC : 100%
600t
Phthalic ester : 25%
Material
Calcium carbonate : 30%
PVC resin
Phthalic ester
Calcium carbonate
Fig.8 : The areas and conditions of analysis

(making the virgin PVC powder)
Energy consumption
basic unit
(M/t)
5992
5457
22
CO emission
coefficient
(-CO/t)
1430
1375
0.005
4.1.3 Calculating methods

Table.4 shows energy consumption basic unit and CO emission coefficient [6], [7]. The basic
unit of electric power consumption and fuel-consumption of transport for the PVC recycled
powder used examples from the reference [8]. The calculating methods for LCA also used
examples from the reference [8].
3000
700
2500
600
Calcium
carbonate
2000
Phthalic
ester
1500
1000
500
transport
PVC resin
disposing
CO emissions -CO/t
Energy consumption 10Mcal)
4.2
Results and discussions
Fig.9 shows the energy consumption and CO emissions of making the PVC recycled powder
and virgin PVC powders to PVC waterproof sheet. The energy consumption and CO emissions
were decreased by making the PVC recycled powder, moreover, the effect of decreasing was
around 70% down them comparing with the virgin PVC powder. This evaluation yields that the
recycle method was valuable for resources effective utilization and environmental load
lowering.
Calcium
carbonate
500
Phthalic
ester
400
300
200
PVC resin
trasport
100
disposing
0
Recycle
Virgin
Recycle
Virgin
Fig.9 : Energy consumption and CO emissions of making the PVC powders

5.
CONCLUSION
The main conclusions obtained are as follows:

Being able to separate just the PVC recycled powder from the PVC building wastes by the
separation methods developed in this study
As the kind and mix proportion of the PVC recycled powder, the tensile properties of the
PVC recycled sheet are changed
The tensile properties and durability of the recycled sheets are able to be increased by
mixing with the virgin PVC powder.
It is not decreased for the tensile properties of the recycled sheet by mixing two and more
difference kinds of the PVC recycled powder.
The energy consumption and CO emissions are decreased around 70% by making the
recycled methods
ACKNOWLEDGEMENTS
This study was supported by JSPS KAKENHI Grant Number 22560573.
REFERENCES
[1] Brad, E., Global PVC markets: threats and opportunities, Plastics, Additives and Compounding,
(2008) 2830.
[2] Ministry of Land, Infrastructure, Transport and Tourism, Survey on construction byproduct,
http://www.mlit.go.jp/sogoseisaku/region/recycle/pdf/fukusanbutsu/jittaichousa/H24sensuskekka_
sankou.pdf, (2015.07 confirmation)
[3] Akio, K., Junya, M., Takao, N., Masafumi, K., Development of recycling technology of waste
PVC floorcovering (Materials and Construction), AIJ J. Technol. Des. No.22, 43-48, Dec., 2005
[4] Akio, K., Naoki, M., Takao, N., Masafumi, K., Development of recycling technology of waste
PVC sheets, AIJ J. Technol. Des. Vol.14, No.28, 399-404, Oct., 2008
[5] Mikiya, I. and Kazukiyo, N., Degradation behavior and application of recycled PVC sheet made
of floor sheet for railway vehicle, Polymer Degradation and Stability, 92 (2007) 1692-1699.
[6] Architectural Institute of Japan, The LCA guide to the building,
http://news-sv.aij.or.jp/tkankyo/s0/tkankyo/arc08_files/AIJ-LCAandLCW_Manual_Ver.4.02.pdf,
(2015.07 confirmation)
[7] Plastic Waste Management Institute, Survey Report for LCI Data of Petrochemical Products, 1999
[8] Katsuhito, N., Seiji, S., Keiichi, K., Itaru, Y., Life Cycle Analysis of Environmental Load from
Polyvinyl Chloride Products, Journal of the Japan Institute of Energy, Vol.86 (2007) No.5, P.339345

A STUDY ON THE ACTUAL STATE FROM GENERATION TO

TREATMENT OF CONSTRUCTION AND DEMOLITION WASTES
Ken-ichi Imagawa (1), Koyama Akio(2)
(1) Graduate School of Science and Technology, Meiji Univ., Japan
(2) Dept. of Architecture Faculty of Science and Technology, Meiji Univ., Japan Dr.Eng
Abstract
To clarify the actual state from generate of the construction and demolition waste to treatment and
recycling, firstly, the authors made the field survey for 60 demolition works of wooden house in Japan
to grasp the actual state of the generation of demolition wastes and clarified the relationship between a
kind and quantity of the construction waste and the scale of houses. Additionary, to clarify the criteria
for recycling in Japan, the authors made questionnaire survey of the criteria for intermediate treatment
facilities when they accepted of the construction waste. As a result, it was revealed that intermediate
treatment facilities look from a severe viewpoint for contamination on the occasion of acceptance of
construction and demolition wastes. Finally, we investigated Life Cycle Assessment in the
intermediate treatment to clarify the actual state. As for the breakdown of the Carbon dioxide emission
in the intermediate treatment of construction wastes, transportation 50%, electricity 30%, heavy
machine use 20%, the disposal less than 10%, and as for the amount of Carbon dioxide emissions was
clarified that a high precision prediction is enabled from waste acceptance quantity.
Keywords: Construction and Demolition Waste, CO2 emission rate, Generation of wastes by
demolition works, Life Cycle Assessment
1.
INTRODUCTION
In many countries, construction and demolition wastes (CDW) has a big ratio among the
industrial waste. Actions for the recycling of construction wastes are serious responsibilities
of construction industries so that the world will accomplish development continuously in the
future.
Not only development and the spread of recycling technologies but also the facilitation of
the material flow of the building materials are important to promote recycling of CDW[1][2][3].
For the facilitation of the material flow, it is necessary to grasp the actual state from
generation of construction wastes to treatment and recycling.
However, there are many parts which information lacks[4][5][6][7] about the actual situation
of generate of the construction waste which is a premise of the facilitation and processing and
the recycling.
Therefore, the authors decided to clarify the actual situation from an on-the-spot survey
and questionnaire survey for 2 large contents (shown below) that information in particular
was lack at present.

The actual state of generation waste by demolition works of wooden house (generate
stage)
Life Cycle Assessment in intermediate treatment of construction wastes (treatment
stage)
2.
THE QUANTITY OF CDW FROM DEMOLITION WORKS OF WOODEN
HOUSE
2.1
THE PURPOSE AND OUTLINE OF THE SURVEY
Demolition work based on an appropriate processing plan and recycling plan is

indispensable for appropriate treatment and recycling of CDW.
Therefore, the authors conducted on-the-spot survey of the 60 demolition works of wooden
houses to grasp the actual state of the generation of CDW to be necessary when mapping
appropriate treatment and recycling plans.
This survey was based on data of an industrial waste management slip, and conducted by
calculating generation rate of various wastes from every demolition works. In addition, shown
kinds of the wastes classified in this survey in table 1.
2.2
SURVEY RESULTS
1)
GENERATION RATE OF WASTES
table.1 : Kinds of the wastes

Including
Survey items
Shown relations between generation rate of No. asbestos
wreckage and the total floor area in fig.1.
1
wreckrages
2
waste glass,concrete,ceramics
Wreckage was generated by demolition works of
3
waste plastics
foundation concrete, and the rate shows a
4
waste metal were

tendency to increase so that total of floor area is
5
waste wood
enlarged, and, the correlation is high.
6
waste textiles
Shown relations between generation rate of
7
waste plaster board

8
constructionmixedwaste
glass, ceramics, and concrete wastes, and the
9
total floor area in fig.2.These wastes were
10
waste plastics
classified as generated from concrete blocks or
11
waste plaster board

tiles, the correlation with the generation rate and
12
constructionmixedwaste
the total of floor area is low.
Shown relations between generation rate of plastic wastes, and the total floor area in fig.3,
and, between generation rate of metal wastes, and the total floor area in fig.4. Both of
generation rates show a tendency to increase so that total of floor area are enlarged, but, the
correlations are relatively low. Incidentally, a similar tendency was confirmed in waste
textiles
Shown relations between generation rate of wood wastes, and the total floor area in fig.5.
Because of the subjects in this survey are wooden houses, there is more generation rate than
other wastes. Additionally, the rate shows a tendency to increase so that total of floor area is
enlarged, and the correlation is relatively high.
Shown relations between generation rate of plasterboard wastes, and the total floor area in
fig.6. The rate shows a tendency to increase so that total of floor area is enlarged. In this
survey, there were few cases which used a plasterboard to all interior groundwork materials,
but the number of them are supposed to increase in the future.
Shown relations between generation rate of Construction mixed wastes, and the total floor
area in fig.7. The rate shows a tendency to increase so that total of floor area is enlarged.

2)
GENERATION RATE OF WASTES INCLUDING ASBESTOS
There was the thing which contained asbestos in wastes generated as "glass, ceramics, and
concrete wastes", "waste plastics" and "Construction mixed wastes ". Most of them are
classified in "glass, ceramics, and concrete wastes". Shown relations between generation rate
of glass, ceramics, and concrete wastes including asbestos, and the total floor area in fig.8.
3)
SUMMARY OF WASTES GENERATION RATE
Each kind of wastes, generation rates show a tendency to increase so that total of floor area
are enlarged.
Therefore, generation rates could be estimated from total floor area.
Waste glass, concrete, ceramics
wreckage
16.0
quantity of waste generatedm3
70.0
60.0
50.0
40.0
30.0
20.0
y = 0.2133x - 6.7453
R = 0.4602
10.0
0.0
0
50
100
150
Total floor aream2
200
14.0
12.0
y = 0.0021x + 3.4436
R = 0.0008
10.0
8.0
6.0
4.0
2.0
0.0
250
Fig.1 :Relations between generation

rate of wreckage and the total floor area
50
35.0
250
Waste metal were
waste plastics
9.0
8.0
7.0
6.0
y = 0.0141x + 1.3293
R = 0.1048
5.0
4.0
3.0
2.0
1.0
30.0
25.0
y = 0.0388x + 3.578
R = 0.0508
20.0
15.0
10.0
5.0
0.0
0.0
0
50
100
150
Total floor aream2
200
250
Fig.3 :Relations between generation rate

of waste plastics and the total floor area
50
100
150
Total floor aream2
200
250
Fig.4 :Relations between generation rate of

waste metal were and the total floor area
waste plaster board
waste wood chip

14.0
90.0
200
Fig.2 :Relations between generation

rate of glass, concrete, ceramics and the
total floor area
10.0
80.0
y = 0.1839x + 15.674
R = 0.2596
70.0
60.0
50.0
40.0
30.0
20.0
10.0
12.0
10.0
y = 0.0425x - 0.7976
R = 0.422
8.0
6.0
4.0
2.0
0.0
0.0
0
50
100
150
Total floor aream2
200
250

of waste wood chip and the total floor area
50
100
150
Total floor aream2
200
250

of waste wood chip and the total floor area
Waste glass, concrete, ceramics including asbestos
Construction mixed waste

25.0
12.0
100
150
Total floor aream2
20.0
y = 0.0549x - 1.2528
R = 0.3758
15.0
10.0
5.0
10.0
y = 0.0263x - 1.1705
R = 0.193
8.0
6.0
4.0
2.0
0.0
0.0
0
50
100
150
Total floor aream2
200
250
50
100
150
Total floor aream2
200
250
Fig.7 :Relations between generation rate Fig.8 :Relations between generation rate of
of mixed waste and the total floor area waste including asbestos and the total floor area

3.
INVESTIGATION LIFE CYCLE ASSESSMENT (LCA) IN INTERMEDIATE
TREATMENT OF CDW
3.1
THE PURPOSE AND OUTLINE OF THE INVESTIGATION
1)
Basic surveys concerning intermediate treatment facilities

To clarify the criteria of acceptance of CDW enables to simplify considerartion fractional
processing in recycling. Thus, prior to investigate of LCA in Intermediate treatment, basic
questionnaire surveys concerning waste acceptance are conducted for 141 intermediate
treatment facilities in Japan[8].
2)
Environmental load by intermediate treatment

To grasp the actual state of environmental load in intermediate treatment of CDW
(treatment stage), conducted precise questionnaire survey for an intermediate treatment
facility that of a model case for calculating environmental load, and questionnaire survey for
other 10 intermediate treatment facilities. Shown items questionnaire in this survey in table 2.
Besides, the CO2 emission of each items were calculated by using methods shown below.
1. Electricity
Shown an equation of CO emission by electricity in the facilities in eq. (1), emission
coefficients[9] in the eq. in table 3. Model case (company A) has photovoltaic power
generation systems, and the electricity sale by them subtracted it from the total.
CO emissions [t-CO] =Electricity consumption[kWh]*Emission coefficient of the (1)
power company [t-CO/kWh]
2. Heavy machine use
Shown an equation of CO emission by heavy machine use in the facilities in eq. (2),
emission coefficients[10][11] in the eq. in table 3.
CO emissions [t-CO] =Fuel consumption[m3]*Emission coefficient of the fuel (2)
[t-CO/m3]
3. Transportation
An equation of CO emission by transportation is the same in eq. (2).
4. Final disposal
Shown an equation of CO emission by final disposal in the site in eq. (3), emission
coefficients[10][11] in the eq. in table 3.
CO emissions [t-CO] =Waste commitment [t]* Emission coefficients of final (3)
disposal [t-CO/t]
table.2 : Questionnaire items
No.
1
2
3
4
5
6
7
Questionnaire items
Annual total energy consumption
Annual energy consumption
by using electricity
by using heavy machines
by heavy machine use
by transportation
Kinds,quantities,and processing
method of wastes
Details from facility to commitment
table.3 : CO2 emission coefficient

Categories
(basic unit)
electricity
(t-co 2/kWh)
fuel
(t-co 2/m3)
Types
Emission coefficient
Tokyo Electric Power Co.
0.000406
gasoline
diesel fuel
VP fuel
Final disposal waste plastics
construction waste rubber
waste
waste metel were
(co2-/)
wreckages
incineration;
incineration;
incineration;
incineration;
incineration;
2.587
0.033
0.033
0.033
0.033
land
land
land
land
land
fill:
fill:
fill:
fill:
fill:
2.32
2.58
2.49
0.038
0.038
0.038
0.038
0.038

3.2
SURVEY RESULTS
1)
Basic surveys related to acceptance of CDW

Shown an actual state of setting criteria related to waste acceptance by intermediate
treatment facilities in fig.9. Some difference is exist in every kind of wastes, nevertheless
shapes and dimensions, material composition and presence of contamination are setting as
criteria. Especially, the setting ratio concerning contamination is higher. As a result,
intermediate treatment facilities look from a severe viewpoint for contamination.
Additionally, the setting ratio of acceptance criteria is higher than other items in waste
concrete. The difference caused by concrete waste that a use of the recycling is clear.
2)
LCA investigation
Shown annual CO2 emissions by electricity of model case (company A) in table 4, by using
of heavy machines in table 5, by transportation in table 6, by final disposal in table 7.
Additionally, annual total CO2 emissions of the facility was around 1700 tons, shown the
details in fig. 9.
Additionally, shown attribute of other intermediate facilities in table 8, those of CO2
emission rates in table 9, and Average of breakdown CO2 emission rates in fig.10. In this
survey results shows that as for the breakdown of the CO2 emission in the intermediate
treatment of CDW, transportation 50%, electricity 30%, and heavy machine use 20%, final
disposal less than 10%. Quantity of acceptance and CO2 emission are greatly different every
each facility.
Meanwhile, the emission rate shows a tendency to increase so that quantities of acceptance
are increasing. Additionally, the table shows that amount of energy consumption by treatment
varies according to kind of waste to manage.
Shown the relation between the quantity of acceptance and total emission rate by facility
and in fig.11. The relation is proportional, and the correlation is extremely high. Hence, a high
precision prediction of CO2 emission rate is enabled from waste acceptance quantity.
table.4 : Annual CO2 emissions (electricity)
CO2 emissions
t-CO2
Energy consumption
kWh/year
Items
Set of Criteria
Waste Gypsum board

Other conditions
Shapes and Demensions
300354
139.1
13722
6.4
1615
0.7
18
146.2
14
Kind (division by makers)
21
34
Limit of contamination
23
Items
Energy consumption
m3/year
24
gasoline
1.68
21
Content of PVC
15
11
Kind of esin
3.9
15
Waste Plastic
Other conditions
CO2 emissions
t-CO2
13
12
Limitation of the acceptance item
table.5 : Annual CO2 emissions (heavy machine)
14
Wastes Glassceramics
Other conditions
total
14
10
Kind (division by the discharge state)
electricity
(facilities)
electricity
(office)
solar generation
(sell)
Not set of Criteria
14
13
13
11
Waste Wood
diesel fuel
36.6
VP fuel
94.4
196.01
total
diesel fuel
CCA Wood or not
22
586.4
Presence of adhesion of the paint
t-CO2
emissions
0.00
315.64
814.30
814.3
27
24
32
17
Waste Concrete
Other conditions
11
89
Presence of contamination
0.00
28
25
66
Presence of rebar
34
61
39
76
0%
total
42
488.1
gasoline
7
21
table.6 :
Annual
CO2
Energy
consumption
CO emissions
(transportation)
Items
m3/year
Other conditions
Incorporation of ironware
20%
40%
24
60%
80%
Fig.9 :Actual state of setting criteria
100%

table.7 : Annual CO2 emissions (final disposal)

Processing form Quantityofannualcommitment CO2 emissions
in commitment
t/year
t-CO2
final disposal
waste plastics
927.1
35.1
(land fill)
final disposal
waste rubber
0.1
0.0
(land fill)
final disposal
waste metal were
427.4
16.2
(land fill)
waste
final disposal
glass,concrete,ce
501.3
19.0
(land fill)
ramics
final disposal
wreckages
2157.8
81.8
(land fill)
total
4013.7
152.1
152t
waste types
9%
146t
Electricity
9%
586t
34%
814t
48%
Fuel consumption by
using heavy machine
Fuel consumption by
transportation
Disposal(landfil,incin
erate)
Fig.9 :Details of CO2 emission in model case
table.8 : Attribute of intermediate facilities

Questionnaire results
Conpany Numbersof Acceptance Quantity of
Quantity of commission
Commission/
names acceptance of the mixed acceptance
million t
Acceptance
items
waste
million t Recycling Disposal
%
total
1
A
8
5.42
4.03
0.40
4.43
82
2
B
7
0.56
0.40
0.00
0.40
71
3
C
1
1.97
1.96
0.00
1.96
99
4
D
2
26.41
5
E
1
18.00
17.50
0.00
17.50
97
6
F
8
6.32
6.57
0.57
7.14
113
7
G
8
11.65
10.88
0.95
11.82
101
8
H
7
39.61
32.64
6.94
39.85
100
9
I
1
0.23
0.20
0.00
0.20
88
10
J
1
10.00
11
K
1
14.72
14.80
0.00
14.80
101
average
3
4.4
4
0.3
4.3
99
No.
121t
6%
Electricity
559t
27%
970t
48%
396t
19%
Fuel consumption by
using heavy machine
Fuel consumption by
transportation
Disposal(landfil)
Fig.10 :Details of CO2 emission in average

table.9 : Emission rates of intermediate facilities
Calculation of CO2 emissions
Conpany
No.
names
Emissionsperquantity ofacceptance(t-CO2 /million t)

Electricity
1
A
2
B
3
C
4
D
5
E
6
F
7
G
8
H
9
I
10
J
11
K
average
4.
26.7
96.8
52.4
5.6
60.3
122.3
189.1
156.6
324.9
6.5
12.3
128.2
Heavy
machine
Tranport
108.1
14.7
48.4
16.2
114.6
81.6
7.2
54.9
23.2
49.4
90.9
150.2
152.7
243.5
1.3
246.6
245.3
98.1
222.5
disposal
28.0
0.0
0.0
0.0
34.2
30.8
66.4
0.0
0.0
27.8
Total
313.0
264.2
344.3
23.1
60.3
517.7
546.7
230.1
477.9
29.7
61.7
469.3
Fig.11 : the relation between the quantity

of acceptance and total emission rate
CONCLUSIONS
The major conclusions derived from this survey are given as follows:
1. Each kind of wastes from demolition work of wooden house, generation rates show a
tendency to increase so that total of floor area are enlarged.
Therefore, generation rates could be estimated from total floor area.
2. Intermediate treatment facilities in Japan set criteria as follows: shapes and dimensions,
material composition and presence of contamination, and look from a severe viewpoint
for contamination on the occasion of acceptance of the CDW.
3. As for the breakdown of the CO2 emission in the intermediate treatment of CDW,
transportation 50%, electricity 30%, and heavy machine use 20%, final disposal less
than 10%

4. The relation between quantities of acceptance CDW and total emission rate by
intermediate treatment facility is proportional, and the correlation is extremely high. AS
a result, a high precision prediction of CO2 emission rates are enabled from waste
acceptance quantities.
REFERENCES
[1] Masafumi, K., Koji, S., Akio, K., Koichi, H., Murakami, Y., Norikazu, E., 'Investigation about
Development of the Unification System Models Contributing to Resources Circulation of
Construction Waste: Part 1. Investigation of the middle processing trader', AIJ Summaries of
Technical Papers of Annual Meeting, A-1, (2006), 611-612.
[2] Norikazu, E., Masaaki, O., Koichi, H., Junichiro, M., Akihiko, G., Makoto, K., Takeshi, I.,
'Uniform Handling System regarding Material Recycle of Plastics Generated from Construction
Work: Part.1 Amount of Waste Plastics from Construction Work and Possibility of Material
Recycle', AIJ Summaries of Technical Papers of Annual Meeting, A-1, (2007), 607-608.
[3] Toshiyuki, F., Takashi, S., Chikahiro, S., Masaaki, M, Sou, K., 'DEVELOPED INTEGRATED
SYSTEM WITH DISASSEMBLING AND RECYCLING TECHNOLOGIES PART1. CO2
EMISSIONS FROM INTERMEDIATE TREATMENT FACILITIES', RILEM EnvironmentConscious Materials and Systems for Sustainable Development (2004), 73-80.
[4] Kensuke, K., Takayuki, I., Kiyotaka, T., Takasi, I., 'ANALYSIS OF BUIDLING WASTE
TREATMENT IN VIEW OF ENVIRONMENTAL LOAD REDUCTION', AIJ Journal of
Environmental Engineering, 74(635), (2009), 97-104.
[5] Shuzo, O., Masayuki, T., Yoshihisa, N., Torao, K., Hioshi, K., 'Investigation of the Field Study of
Generation Rate and Fractionation of Construction System Wastes in the Large-scale new
Construction Field', AIJ Summaries of Technical Papers of Annual Meeting, A-1, (2002), 913914.
[6] Ryusuke, U., Hidetoshi, E., 'A fact-finding study of waste in new construction and dismantling
construction: Examination of a waste basic unit', AIJ Summaries of Technical Papers of Annual
Meeting, A-1, (2006), 607-608.
[7] Yoko, N., Takamasa, M., 'A INVESTIGATION ON CONSTRUCTION BYPRODUCTS AND
MIXED ONES IN BUILDING CONSTRUCTION PROCESS' , AIJ Journal of Technology and
Design (18), (2003), 325-328.
[8] Yasushi, M., Akio, K., Kanako, S., Masafumi, K., Yuichi, H., 'Development of estimation system
of cost and carbon-dioxide emissions on demolition work of wooden houses', AIJ Summaries of
Technical Papers of Annual Meeting, A-1, (2010), 229-230.
[9] Japanese Ministry of the Environment 'About the publication of direct discharge coefficient and
adjustment discharge coefficient of electric companies in 2012',
http://www.env.go.jp/press/17512.html
[10] Japanese Ministry of the Environment 'Example of the carbon footprint according to the fuel',
https://www.env.go.jp/council/16pol-ear/y164-04/mat04.pdf
[11] Japanese Ministry of the Environment ' List of calculation method and discharge coefficients in
calculation, reports, and the publication system', http://ghgsanteikohyo.env.go.jp/files/calc/itiran.pdf

RECYCLED CONCRETE AGGREGATES FROM DIFFERENT

SOURCES: CHARACTERIZATION AND USE IN STRUCTURAL
CONCRETE
Guilherme C. Cordeiro (1), Gabriela V. Maioli (1), Amanda P. Vieira (1), Mnica B. Leite
(2)
(1) Laboratory of Civil Engineering, Universidade Estadual do Norte Fluminense Darcy
Ribeiro, Brazil
(2) Department of Civil Engineering, Universidade Estadual de Feira de Santana, Brazil
Abstract
The use of recycled aggregates has been adopted as an environmentally responsible practice to
mitigate the impact caused by the waste generation in Civil Construction and the great amount of
natural resources used to produce aggregates. The present study proposes to compare the
characteristics of two recycled concrete coarse aggregates with a natural coarse aggregate aiming the
use in structural concrete. The first recycled aggregate was produced in laboratory using one-year-old
specimens of concrete. The second recycled aggregate was produced from the demolition waste of a
soccer stadium concrete structure, which was built in 1954 in the state of Rio de Janeiro, Brazil. After
the collection, both materials were crushed and classified for a maximum size of 12.5 mm. A known
quality natural aggregate originated from a granitic rock was used as the reference. All aggregates
were characterized by particle size distribution, Micro-Deval abrasion, water absorption, density and
packing density tests. After that, five different concrete mixtures were produced: one of reference with
100% of granite as coarse aggregate, two concrete mixtures with 50% of granite and 50% of each
recycled aggregate, and two concrete mixtures using 20% of rice husk ash, by volume, as cement
replacement material, and using the same RCA and granite contents as the two prior mixtures. The
concrete characterization was carried out by slump, compressive strength at 7 and 14 days tests. The
results showed that both RCAs presented adequate density, gradation and particle shape to
applications in concrete. Moreover, the RCAs had Micro-Deval higher than the usual limit stablished
for using in structural concretes. However, blending RCAs with the natural aggregate resulted in
abrasion loss values within the range of the structural concrete aggregates.
Keywords: Recycled concrete aggregate, Construction and demolition waste, Concrete waste,
Micro-Deval abrasion test, Rice husk ash.
1.
INTRODUCTION
Nowadays, construction and demolition wastes (CDW) have extensively been used as source
for aggregate production in different construction activities, such as buildings, roads, and
runways. In some countries, particularly in Europe, recycled aggregates from CDW are used as
partial replacements of natural crushed aggregates. According to the Eurostat [1], CDW is one
of the heaviest and most voluminous waste streams generated in the European Union. It is

estimated that about 25-30% of all waste generated in the European Union consist of CDW,
such as concrete, bricks and gypsum, many of which can be recycled. It is important to note
that approximately 850 million tonnes of CDW have been generated in the EU per year [2].
In Brazil, as in other countries, there is currently a public control over the generation of
CDW. In 2002, it was approved the governmental resolution CONAMA 307 [3], defining that
large public and private generators are required to develop and deploy a waste management
plan. Thus, generators shall indicate reuse, recycling or other environmentally friendly
disposition for CDW.
Recycled concrete aggregates (RCA) can be defined as old concrete that has been adequately
crushed to produce aggregates. RCA may be further defined as the most valuable type of CDW,
considering that the main problems associated with CDW aggregates, such as high water
absorption and heterogeneous composition, are considerably minimized in this particular type
of CDW. Thus, RCA has been used in high strength concretes [4],[5], self-consolidating
concretes [6][8] and fiber reinforced composites [9]. RCA from roadway materials are typically
generated and used at the same construction site, providing decrease in both money and time
[10]. However, since RCA is generally collected from different types of concrete, its major
characteristics such as mechanical strength, density, absorption, shape and texture have varied
significantly. RCA are composed of natural aggregates and adhered mortar and its physical
properties depend on both quality and amount of adhered mortar [11].
In this study, two different RCAs were compared with a natural coarse aggregate (granite)
aiming the use as a partial coarse aggregate substitute in structural concrete. The first RCA was
produced in laboratory using maximun one-year-old specimens of concrete. The second RCA
was produced from the demolition waste of a soccer stadium concrete structure, which was built
in 1954 in the state of Rio de Janeiro, Brazil. All aggregates with the same particle size
distribution and shape index were characterized by water absorption, density, Micro-Deval
abrasion, and packing density tests. After that, five different concrete mixtures were produced:
one of reference, with 100% of natural aggregates, two concrete mixtures with 50% of natural
aggregate and 50% of each recycled aggregate, and two concrete mixtures using 20% of rice
husk ash, by volume, as a cement replacement material, and using the same granite and RCA
contents as the two prior RCA-mixtures. The concrete characterization was carried out by slump
and compressive strength at 7 and 14 days tests.
2.
2.1
Production and characterization of aggregates
Three different coarse aggregates were used in this work, two of which were RCAs. A
commercial granite (igneous rock) was initially selected based on its high quality as a concrete
aggregate. Fig 1 shows the particle size distribution [12] of the granite, which had maximum
size of 12.5 mm, fineness modulus of 6.1 and density of 2668 kg.m-3. Two recycled materials,
one laboratory concrete, and one old structural concrete were used in this investigation. The
first RCA was produced using a Pulverisette 1 (Fritsch) jaw crusher from laboratory maximum
one-year-old specimens of concrete. This recycled aggregate was named as RCA-Lab and had
a density of 2639 kg.m-3. Additionally, the second RCA (RCA-Old) was produced from a
demolition project of a soccer stadium concrete structure, which was built in 1954 at the State
of Rio de Janeiro, Brazil. In this case, collected fragments of concrete (~ 20 kg) were firstly
crushed to small sizes using a hand hammer. After that, the small pieces were crushed using the
same procedures early described for the RCA-Lab. The density of RCA-Old was 2615 kg.m-3.

Dry sieving of both Lab and Old recycled aggregates was used to separate them into fractions
which were recombined later to produce the same particle size distribution of the granite sample
(Fig. 1).
100
Passing (%)
80
60
40
Fine aggregate
20
Coarse aggregates
(granite and RCAs)
0.1
10
100
Sieve size (mm)
Figure 1: Particle size distribution of fine and coarse aggregates.

The aggregate water absorption was obtained in this study based on two methodologies. The
first test was the total water absorption in accordance to ASTM C127-15 [13]. In this case, the
absorption (Aw) was determined after soaking the aggregate in water for a prescribed duration
using Equation 1. The second absorption test was conducted after the drying of 1.0 kg sample
at 60C for 7 days. Immediately after that, the sample was placed in an immersed sieve
connected to a scale to record the submerged mass to obtain the water absorption at different
time intervals. The mass of the sample was measured during 12 hours after immersion. Then,
the mass variation from water absorption (Mw) was determined using Equation 2. All mass
variation (absorption) average values were taken from three different samples of the same
aggregate.
Aw = 100.(MSSD MD).MD1
(1)
Considering:
Aw: water absorption in percentage;
MSSD: mass of the saturated-surface-dry test sample in air;
MD: mass of the oven-dry test sample in air.
Mw = 100.(Mt M0).M01
(2)
Considering:
Mw: mass variation (water absorption) in percentage;
Mt: apparent mass of the sample in water at t time;
Mi: initial apparent mass of the sample in water.
The aggregate shape index was determined by Brazilian standard NBR 7809-06 [14]. Shape
index in this case is defined as the average ratio between the length and the thickness of the

aggregate grains, weighted by the amount of each grain size distribution. The shape index test
was performed using samples with 200 oven-dry grains.
Micro-Deval abrasion tests were also performed according to ASTM D6928-10 [15] to
determine the aggregate abrasion resistance in the presence of water. In these cases, a 1.5 kg
aggregate sample (0.75 kg of 12.5-9.5 mm, 0.375 kg of 9.5-6.3 mm, and 0.375 kg of 6.34.75mm) was placed in a steel jar mill and rotated for 2 hours with 2 L of water and 5 kg of
grinding steel balls (diameter of 9.5 mm). Besides the individual aggregates, Micro-Deval tests
were carried out on granite-RCA-Lab and granite-RCA-Old blends with a 50%-50% proportion
in volume, which was the same proportion used in the concrete mixtures. The abrasion loss is
the mass of aggregate passing the 1.18 mm sieve expressed as a percentage of the original
sample mass for each aggregate (or blend). Three replicates were performed and the result was
expressed as an average value.
The particle packing density of aggregates was determined by the compaction and vibration
test developed by De Larrard [16]. Initially, a 7.5 kg dry sample was handily put in a steel
cylinder (diameter of 160 mm and height of 320 mm). After that, a steel piston was introduced
in the cylinder, applying a pressure of 10 kPa on the top of the aggregate sample. Then the
cylinder was fixed on a vibrating table (68 Hz), and submitted to vibration for 3 minutes. The
packing density was calculated from Equation 3. The test was made in duplicate samples.
= 4MD.(.D.H.)1
(3)
Considering:
: experimental packing density;
MD: mass of the sample;
D: diameter of the steel cylinder;
H: final sample height;
: density of the sample.
2.2
Complementary materials
Brazilian type V Portland cement (high early strength with 3100 kg.m-3 density), comercial
rice husk ash (D50 of 15.5 m and density of 2010 kg.m-3), fine aggregate of siliceous river sand
(particle size distribution showed in Fig. 1; fineness modulus of 2.4; and density of 2660 kg.m3
), and water-reducing high-range admixture polycarboxylate-based with 33% solids content
and density of 1210 kg.m-3 were used for the concrete mixtures.
2.3
Mix-design, production and characterization of concretes
Five concrete mixtures were prepared during the study and their compositions are provided
in Table 1. Initially, a 35 MPa-reference concrete mixture (C-REF) was designed within the
framework of the Compressible Packing Model [16] using the computer code Betonlab Pro2.
After that, two concretes were designed with RCA-Lab and RCA-Old using 50% volume
replacement of the granite used in the C-REF. An equal water and cement ratio of 0.59 was
used in such three concrete mixtures. Another concrete group was also designed. Considering
both RCA concretes, 20% of the cement was replaced by an equivalent volume of the rice husk
ash. The water dosage was the same of the C-REF, which resulted in a water-cementitious
materials ratio of 0.64. Previous laboratory tests indicated that the pozzolanic effect of the rice
husk ash compensated the increase in water-cementitious materials ratio, although the early
compressive strength were smaler than the C-REF strength. Slump values were kept constant
at 80-100 mm with specific dosages of superplasticezer for all concretes.

Table 1: Composition of concretes in kg.m-3.

Material
C-RCA-Lab- C-RCA-OldRHA
RHA
C-REF
C-RCA-Lab
C-RCA-Old
387.4
387.4
387.4
309.9
309.9
50.2
50.2
River sand
779.2
779.2
779.2
779.2
779.2
Granite
893.7
446.9
446.9
446.9
446.9
RCA-Lab
442.0
442.0
RCA-Old
438.0
438.0
Water
229.3
229.3
229.3
229.0
229.0
Superplasticizer
0.30
0.35
0.35
0.75
0.75
Cement
Rice husk ash
All concretes were made in a planetary mixer with the following mixing sequence: the fine
and coarse aggregates and 50% of water were mixed for 0.5 minute; the mixing was interrupted
for 5 min; the cementitious materials, the remaining water, and the superplasticizer were added
and mixed for 3 more minutes; the mixing was interrupted for 1 min to scrape the mixer; then,
the mixing was concluded after another 1 minute. It is important to note that if aggregates are
employed in dry conditions in a conventional mixing procedure, the workability of concrete is
greatly reduced. The stopped time of 5.0 minutes was performed due to the water absorption
results.
Cylindrical specimens with 100 mm diameter and 200 mm height were prepared for
compressive strength tests. All specimens were cast in two layers and compacted by using a
vibration table (1 minute for each layer). Also, the specimens were left to cure for 24 h under
damp cloth, and then demolded and cured in saturated limewater at laboratory conditions until
the age of the tests. After 7 and 14 days of curing, four replicates cylindrical specimens of each
concrete were submitted to compressive strength test using a servohydraulic machine
(Shimadzu UHI-500kNI) with displacement rate of 0.5 mm/min. Before the tests, cylinder ends
were prepared using a grinding machine.
3.
3.1
Characterization of aggregates
A summary of the aggregates properties is presented in Table 2, where the results confirmed
that the RCA samples had higher water absorptions and lower densities than the granite sample.
In this case, the density and water absorption were strongly affected by adhered mortar on the
aggregate particles. Both RCAs presented water absorption [13] values within the typical range
from 3% to 10% [5][11][17]. The granite presented only 0.9% of water absorption, indicating
the good performance of this material as aggregate.

Table 2: Properties of aggregates

Property
Granite
RCA-Lab
RCA-Old
2668
2639
2615
0.9
6.4
5.6
Shape index
1.4
1.5
1.5
0.5476
0.4419
0.4930
-3
Density (kg.m )
Packing density
The mass variation due the water absorption as a function of square root of time for both
RCAs is provided in Fig 2. It can be noted that the absorption values of both RCAs increased
in time and they were more pronounced in first 5 minutes of immersion. This information was
very relevant to definition of the mixing procedure previously described. After a fast increase
in water absorption due to the completion of larger pores in the aggregate, there was a marked
decrease in the absorption rate with formation of a typical plate. After 12 hours of test, RCALab had 4.8% of absorption, while the RCA-Old exhibited 3.8%. These results confirmed the
water absorption values from the ASTM C127-15 [13] (Table 2) and the need to maintain the
recycled aggregate for long periods until the mass equalization during the absorption test, as
also reported by Tegguer [18].
6
5
4
3
2
RCA-Lab
1
RCA-Old
0
0
0.8
1.6
2.4
Time-0.5 (h-0.5)
3.2
Figure 2: Mass variation over exposure time for different recycled aggregates.
Table 2 also shows the shape indices for all studied aggregate. Note that there were not
expressive differences between the shape index values, indicating the predominance for angular
cubical particles. The same mechanism of crushing and processing operations were used for
both RCAs, which reproduced in laboratory the similar conditions observed in the aggregate
producer plant. It should be noted that aggregate particles that are angular require more cement
to maintain the same water-cement ratio.
The average packing density results are provided in Table 2. The natural aggregate resulted
in an experimental packing density of 0.5476, while the RCA-Lab and RCA-Old samples
resulted in 0.4419 and 0.4930, respectivelly. Although the packing density protocol [16]
establishes that the packing density is governed by particle size distribution and shape particle,
the studied aggregates with similar granulometries and shape indices presented different
packing density values. This can be explained by differences in surface texture and possible
breakdown of recycled particle during the compaction and vibration test.

Fig 3 shows the comparison between the results of the Micro-Deval abrasion loss for all
aggregates and aggregate blends, which revealed expressive differences for the distinct
aggregates. The abrasion characterisation is very important due to recycled aggregates must be
tough and abrasion resistant to prevent crushing and degradation during the different phases of
concrete production. Results indicated that the granite, RCA-Lab and RCA-Old experienced
average losses of 13%, 27% and 20%, respectively. As expected, the used granite presented the
lowest abrasion loss followed by the RCAs. In fact, early research studies [19][20] indicated
good performances of the same granite used in this work in different concretes. Similar results
of Micro-Deval abrasion loss for distinct granites were highlighted by Cooley Jr. et al. [21].
Butler et al. [22] reported similar abrasion losses for RCAs.
Considering the maximum abrasion loss (17%) indicated by Rogers [23] to define an
aggregate as satisfactory for structural concrete, only the natural aggregate would be accepted.
However, the abrasion tests were conducted using 100% of each aggregate, which is not in
accordance with their use in concrete. To evaluate the effects of RCA content in blends on the
abrasion loss, new tests were carried out considering each aggregate content on the concrete
composition (Table 1). Fig.3 shows the abrasion losses of both granite-RCA-Lab and graniteRCA-Old blends. Thus, both blends presented lower values than each particular aggregate, with
19% granite-RCA-Lab and 17% granite-RCA-Old. Granite-RCA-Old blend had lower abrasion
loss than both the RCA-Old aggregate and the maximum loss indicated by Rogers [23] for
structural concrete aggregates. On the other hand, granite-RCA-Lab sample exhibited slightly
higher loss than the 17% limit. It is interesting to observe that pondered average values
considering each aggregate content on the concrete composition were also calculated. In this
case, the differences between theoretical and tested values were 5% lower.
By the observation of the coarse 1.18 mm particles of RCA samples after the Micro-Deval
tests, it was possible to see that the primary cause of the degradation was thought to be the
stripping of old-cement mortar from the particle. This degradation is an indicative of an increase
in fines that could affect each of the two RCAs differently.
30
27
25
Abrasion loss (%)
20
19
20
17
15
13
10
5
0
Granite
RCA-Lab
RCA-Old
Granite/
RCA-Lab
Granite/
RCA-Old
Figure 3: Micro-Deval abrasion loss values for all distinct aggregates and blends.

3.2
Application in concrete
At the fresh state, concrete workability was not influenced by the RCA samples since all
slump values met the required specified range (80-100 mm). This behavior was due to the
mixing procedure adopted to avoid the intense water absorption during first 5 min of wateraggregate contact. Moreover, all concretes produced with the both RCAs had higher
superplasticizer dosages than the content of the C-Ref, as can be seen in Table 1. These
increases in superplasticizer content were more pronounced in the mixtures containing rice husk
ash due to the high specific surface area of the pozzolan [24].
Average 7 and 14-day compressive strength test results for all concrete mixtures along with
the standard deviation are presented in Fig. 4. At 7 days, the partial replacement of the granite
by the RCA-Lab led to a slight reduction (approximately 12%) in the compressive strength of
the concrete. On the other hand, the blended RCA-Old and granite concrete experienced
compressive strength that was not significantly (Analysis of Variance with 0.05 significance
level) different of the C-Ref strength. Similar results were previously reported [5][11][17].
The same trend was observed after 14 days of curing. In this case, the compressive strength
of C-RCA-Lab was about 13% smaller compared to the C-Ref and C-RCA-Old. The distinct
behaviour of the concretes with the studied RCAs can be attributed to the different nature of
the sources that contributed to generate RCAs with different properties. Thus, it appears that
the better RCA (RCA-Old) experienced the better results in both individual and concrete
characterizations despite the same production procedures used for both RCAs.
Fig. 4 also makes clear that the better performance of the RCA-Old in relation to the RCALab considering the compressive strength results of the second concrete group (with rice husk
ash as pozzolan). As expected, in general there was a decrease in compressive strength due to
the reduction of cementitious materials content and, consequently, an increase in watercementitious ratio. Therefore, both C-RCA-Lab-RHA and C-RCA-Old-RHA concretes showed
decreases in compressive strength of approximately 20% and 15%, respectively, after both 7
and 14 days of curing in comparison with the C-Ref. It is expected, however, which for long
age of the concretes the compressive strength values are considerably influenced by the
pozzolanic reactions of the rice husk ash.
Visual observations of the fracture surfaces of the 14-day cylindrical specimens after failure
revealed that fractures occurred both around and through the coarse aggregates in spite of the
source. This behavior indicated that the mortar particle bond was not the only fracture
mechanism on the compressive rupture. There also was an influence of the coarse aggregate
strength.

40
7 days
35
14 days
30
25
20
15
10
5
0
C-Ref
C-RCA-Lab
C-RCA-Old
C-RCALab-RHA
C-RCAOld-RHA
Figure 4: Average 7 and 14-day compressive strength results for all concretes.
4.
CONCLUSIONS
Based on the research and results presented in this paper, the following conclusions are
reached:
Density, gradation and particle shape have been shown that the RCAs presented
adequate characteristics to applications in concrete.
Although the RCAs had Micro-Deval higher than the usual limit stablished for using
in structural concretes, blending RCAs with the natural aggregate resulted in
intermediate abrasion loss. Granite-RCA-Old blend (50%-50% in volume) had
abrasion loss value within the range of the structural concrete aggregates.
At the fresh state, concrete workability was not influenced by the RCA samples due
to adjustments on the mixing procedure (with pre-wetting of the aggregates at the
mixture starting) and increase in superplasticizer dosage.
Both RCAs concretes exhibited adequate values of compressive strength until 14
days of curing, specially the mixture with RCA-Old, which presented similar strength
comparing to the reference. These results confirmed the effectiveness of RCA to be
used in production of structural concrete.
ACKNOWLEDGEMENTS
The authors gratefully acknowledge the financial support received from Brazilian Agencies
FAPERJ and CNPq for conducting this research work. Donations of the Pedreira Itaoca
(Campos dos Goytacazes, Brazil) and Areal Anastcia (Campos dos Goytacazes, Brazil) are
also acknowledged.
REFERENCES
[1] Eurostat, 'Waste statistics'. http://ec.europa.eu/eurostat/statistics-explained/index.php/Waste_
statistics. Accessed May 2015.
[2] Fischer, C. and Werge, M. 'EU as a recycling society present recycling levels of municipal waste
and construction & demolition waste in the EU - ETC/SCP working paper 2/2009', 1st Edn
(ETC/SCP, Copenhagen, 2009).

[3] CONAMA. 'Resoluo 307, de 05 de julho de 2002. Dispe sobre a gesto dos resduos da
construo civil', (Dirio Oficial da Unio, Braslia, DF, 17 de jul. 2002. Seo 1, p. 95-96).
[4] Ajdukiewicz, A. 'Influence of recycled aggregate on mechanical properties of HS/HPC', Cem.
Concr. Compos. 24 (2) (2002) 269-279.
[5] Manzi, S., Mazzotti, C. and Bignozzi, M.C. 'Short and long-term behavior of structural concrete
with recycled'. Cem. Concr. Res. 37 (2013) 312-318.
[6] Fakitsas, C., Papakonstantinou, P., Kiousis, P. and Savva, A. 'Effects of recycled concrete
aggregates on the compressive and shear strength of high-strength self-consolidating concrete', J.
Mater. Civ. Eng. 24 (4) (2012) 356-361.
[7] Kou, S.C. and Poon, C.S. 'Properties of self-compacting concrete prepared with coarse and fine
recycled concrete aggregates', Cem. Concr. Compos. 31 (9) (2009) 622-627.
[8] Tuyan, M., Aghabaglou, A.M. and Ramyar, K. 'Freezethaw resistance, mechanical and transport
properties of self-consolidating concrete incorporating coarse recycled concrete aggregate', Mater.
Des. 53 (2014) 983-991.
[9] Carneiro, J.A., Lima, P.R.L., Leite, M.B. and Toledo Filho, R.D. 'Compressive stress-train behavior
of steel fiber reinforced-recycled aggregate concrete', Cem. Concr. Res. 46 (2014) 65-72.
[10] USGC, 'Recycled aggregates - Profitable resource conservation', (U.S. Geological Survey, Denver,
CO, 2000).
[11] Etxeberria, M., Vzquez, E., Mar, A. and Barra, M. 'Influence of amount of recycled coarse
aggregates and production process on properties of recycled aggregate concrete', Cem. Concr. Res.
37 (2007) 735-742.
[12] ASTM C136 / C136M-14, 'Standard test method for sieve analysis of fine and coarse aggregates',
ASTM International, West Conshohocken, PA - USA, 2014.
[13] ASTM C127-15, 'Standard test method for relative density (specific gravity) and absorption of
coarse aggregate', ASTM International, West Conshohocken, PA - USA, 2015.
[14] ABNT NBR 7809-06, 'Coarse aggregate - Determination of shape index by the caliper - Method of
test', Brazilian Association of Technical Standards, Rio de Janeiro, RJ - Brazil, 2008.
[15] ASTM D6928-10, 'Standard test method for resistance of coarse aggregate to degradation by
abrasion in the Micro-Deval apparatus', ASTM International, West Conshohocken, PA - USA,
2010.
[16] De Larrard, F. 'Concrete mixture proportioning: a scientific approach', 1st Edn (E & FN Spon,
London, 1999).
[17] Wardeh, G., Ghorbel, E. and Gomart, H. 'Mix design and properties of recycled aggregate
concretes: applicability of Eurocode 2', Int. J. Concr. Struct. Mater. 9 (2015) 1-20.
[18] Tegguer, A.D. 'Determining the water absorption of recycled aggregates utilizing hydrostatic
weighing approach', Constr. Build. Mater. 27 (2012) 112-116.
[19] Cordeiro, G.C. and Sales, C.P. 'Pozzolanic activity of elephant grass ash and its influence on the
mechanical properties of concrete', Cem. Concr. Res. 55 (2015) 331-336.
[20] Cordeiro, G.C., Barroso, T.R. and Toledo Filho, R.D. 'Enhancement of pozzolanic activity of boiler
sugar cane bagasse ash with high carbon content by re-calcination processes', Fuel, Submitted May
5, 2015.
[21] Cooley Jr., L.A., Huner, M.S. and James, R.H. 'Micro-Deval testing of aggregates in the Southeast
- NCAT Report No. 02-09', (NCAT, Auburn, 2002).
[22] Butler, L., West, J.S. and Tigher, S.L. 'Effect of recycled concrete coarse aggregate from multiple
sources on the hardened properties of concrete with equivalent compressive strength', Constr. Build.
Mater. 77 (2013) 1292-1301.
[23] Rogers, C., 'Canadian experience with the Micro-Deval test for aggregates', Engineering Geology
Special Publications, 13 (1998) 139-147.
[24] Cordeiro, G.C., Toledo Filho, R.D., Tavares, L.M. and Hempel, S. 'Influence of particle size and
specific surface area on the pozzolanic activity of residual rice husk ash', Cem. Concr. Res. 33
(2011) 529-534.

EVALUATION OF MORTAR BASED ON ELECTRIC

MEASUREMENTS: EFFECT OF PRE-WETTING CDW RECYCLED
AGGREGATE
T. STEFANI , M. CHERIAF, J. C. ROCHA
Universidade Federal de Santa Catarina Departamento de Engenharia Civil Lab ValoRes
Abstract
This paper reports on the effect of the moisture kinetics of CDW recycled aggregate as a
granular material and the pre-wetting condition on the mortar hydration. Mixtures were made
with CDW aggregate as a natural sand replacement (CDW/NS), with 0%, 50% and 100% of
replacement. Electrochemical impedance spectroscopy (EIS), capillary absorption tests were
carried out during hydration period (at 1, 3, 7 and 28 days of age). The impedance test (EIS)
realized during the hydration period shows that the saturation condition of the aggregate used
for the mix affects the stage of hydration. The addition of CDW, especially in wet condition
can influence on reduction of water capillary absorption.
Keywords: CDW recycled aggregate, moisture, electrochemical impedance EIS, capillary
absorption.
1.
INTRODUCTION
Mortars mix made with CDW aggregate require special consideration when assessing
mortar on pre-wetting and air-dried conditions. The porous structure of aggregate can retain
water by capillarity and could release this water during hydration period. This present
research has for purpose investigate the effect of CDW recycled aggregate, in air-dried or
saturated conditions, on mortar hydration using electric measurements with the analysis of the
impedance spectrum obtained. From the differences of the spectra, it is expected to visualize
water diffusion from the aggregate to the mortar due to gradient of a humidity that appears
during the curing process, once the aggregate has been mixed saturated.
Previous works has demonstrated the contributions of the impedance spectra on the
analyses of the solid phase and the electrolyte filling the pores [1-3]. Water diffusion can be
identified based on the confirmation of Warburg impedance or diffusive impedance [4].
IES has the possibility to measure the capacitance as well as the resistance of material
and can be used to obtain information about porous structure arrangement and the
microstructural and electrical properties relationships [5].
2.
EXPERIMENTAL
Mixtures were made with CDW aggregate as a natural sand replacement (CDW/NS),
with 0%, 50% and 100% of replacement. Mortar was made with a grain-size distribution
curves standardized (2.4/0.15) and cement aggregate ratio 1:7.5 (by weight). Five different
series were made according to the Table 1. The water-cement ratio has been maintained
constant in the way that only the aggregate and it is saturation condition (wet or dried) could
governing change in paste porosity. Electrochemical impedance spectroscopy (EIS), capillary
absorption tests were carried out during hydration period (at 1, 2, 5, 7, 14, 28 and 60 days).

After mortar casted the samples were sealed with a film, to avoid evaporation, and
maintained in climate chamber (temperature of 22 C 2) until the age of test. CDW
aggregate was a porous (red and grey mixed) from a processing plant located at Sao Paulo.
The chemical analysis shows the presence of CaO (40.8 %), SiO2 (20.4 %), Al2O3 (9.4%),
Fe2O3 (3.2%), K2O (1.84%), loss ignition of 23% and other elements (less than 1%).
Table 1: Composition of the mortar mixes.
Sample
NS
CDW
NS/CDW
CDW-S
NS/CDW-S
Mix
composition
(by weight)
1: 7.5: 1.37
1: 7.5: 1.37
Natural Sand
CDW
Condition
100%
0%
50%
0%
50%
0%
100%
50%
100%
50%
Dried
Dried
Dried
Saturated
Saturated
2.1. EIS during the hydration

Electric impedance measurements were performed on mortar samples to establish the
influence of the pre-wetting of CDW aggregate on the electrical properties of mortar, during
hydration process. For the measurements a Tektronix AFG3120C Oscilloscope was used, in
the frequency range of 1Hz to 100MHz (Fig. 1). The mortar samples were performed at 1, 2,
5, 7, 28 and 60 days. The impedance spectra were taken from the measurements and from it
the resistances R1 and R2 that compound the representative electric circuit [1] could be
determined. The resistance R2 reflects the behavior of the solid-liquid interface of the cement
and the total real impedance is given by the sum of R1 and R2 [1].
Figure 1: Device layout sample measurements.
Hydration process implies on the free-water consumed by the formation of the cement
hydrates compounds which contributes for both resistances, electric and mechanical strengths
[5]. The decreasing of water volume inside the voids and its filling with hydrates products
contributes for both electric and mechanical strengths [5] and it is graphically seen by an
rising arc diameter for later ages [2,3,5]. The spectrums of electrical impedance showed the
evolution of the arcs in the diameters and in the height (Fig. 2).

CDW
Reactance (K)
Reactance (K)
NS/CDW
3
2
1
0
0
2
3
4
5
Resistance (K)
0
0
Reactance (K)
Reactance (K)
1
1
2
3
4
5
Resistance (K)
3
4
5
Resistance (K)
CDW-S
6
5
4
3
2
1
0
1
NS/CDW-S
5
4
3
2
1
0
2
3
4
5
Resistance (K)
Reactance (K)
NS
6
5
4
3
2
1
0
0
2
3
4
Resistance (K)
Figure 2: Impedance spectrum development (Nyquist plots obtained during mortar hydration).
Mortars with CDW aggregates presented higher resistance than mortars with only
natural sand. Samples with pre-wetting CDW-aggregate showed a diffusive behavior mostly
expressive at 14 days age, reducing the resistive response of samples (Fig. 3). On samples
with dry condition, the resistive response increases with CDW replacement. Over 14 days, the
resistance of CDW-S was lower than resistance of NS/CDW-S. The influence of R2 to on the
total resistance of the sample is more important than mortar with CDW replacement (Fig 4).

Resistance (K)
Resistance (K)
7
6
5
4
3
2
1
0
0
50
100
CDW aggregate (%)
(a)
7
6
5
4
3
2
1
0
50
100
CDW aggregate (%)
(b)
Figure 3: Contribution of total resistance of mortar. (a) dried-air CDW aggregate (b) prewetting CDW aggregate
8.0
Rtotal
R2
R1
6.0
4.0
Resistace (K)
Resistance(K)
8.0
2.0
4.0
2.0
0.0
0.0
1
Rtotal
R2
R1
6.0
5
7 14
Time (days)
28
60
5
7 14
Time (days)
28
60
Figure 4: Contribution of R1 and R2 to the total resistance. (a): NS , (b) CDW.

2.2. Capillary test
Hydration process was stopped at age of 1, 7, 28 and 60 days and the water capillary
test was carried out on all mortar samples to evaluate the resistive response with the porous
configuration (Fig. 5). Sorptivity coefficient and wetting angle (28 days) were determined.
As it is shown on Table 2, sorptivity decreases with hydration. At the age of 1 day, a
secondary sorptivity was found on samples with dry-CDW aggregate. This behavior can be
related to the water filling of larger pores after the smaller has been filled. Samples with prewetting CDW have not shown this behavior, which indicate the presence of higher diameter
pore with less variability. CDW aggregate replacement causes a more expressive reduction on
sorptivity at first ages, demonstrated by the inclination angle of the curves at figure 5.

CDW
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
i (cm/cm)
i (cm/cm)
NS/CDW
0
1 day
10
20
30
Time (min)
7 days
28 days
40
0
60 days
1 day
CDW-S
NS/CDW-S
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
i (cm/cm)
i (cm/cm)
10
20
30
40
Time (min)
7 days 28 days 60 days
0
1 day
10
20
30
40
Time (min)
0
1 day
10
20
30
40
Time (min)
NS
i (cm/cm)
3.5
3.0
2.5
2.0
1.5
1.0
0.5
0.0
0
1 day
10
20
30
40
Time (min)
Figure 5: Capillary absorption as function square of time (min).
Mortars using recycled aggregate reduced capillary absorption (Table 2). This could be
explained by the internal cure of the samples, due the release of water contained inside the
grains.

Table 2: Results for samples stopped with 1 and 28 days.

Mortar
Capilarity
Sorptivity
Absorption
(a)
(28 days)
(1day)
(%)
(cm x min1/2)
NS
10.64
CDW
10.29
NS/CDW
3.66
CDW-S
6.81
NS/CDW-S
7.31
(a): primary (b): secondary
3.
22.10 x 10-2
13.90 x 10-2
8.59 x 10-2
5.75 x 10-2
8.28 x 10-2
Sorptivity
(b)
(1day)
(cm x min1/2)
3.74 x 10-2
4.87 x 10-2
5.10 x 10-2
NO
NO
Sorptivity
(28 days)
(cm x min1/2)
5.16 x 10-2
5.38 x 10-2
1.58 x 10-2
3.38 x 10-2
4.12 x 10-2
Wetting
Angle
()
75.2
67.5
86.7
77.7
69.1
RESULTS DISCUTION
During hydration process the water is consumed and conductive parcel of mortars
decrease (Fig. 1 e). Addition of CDW causes an increasing on total resistance of mortars at
the same age. This behavior may be related to the CDW higher water absorption capacity.
Water release from the aggregate to the paste as can be observed at 14 days on the samples
CD-S and NC-S (Fig. d, e). It indicates that the condition of the CDW aggregate as well the
percentage of CDW on the mix, change the cement paste. CDW addition an increase on
mortars resistive proprieties and lower capillary water absorption, that indicates less porous
connection and also more discontinuous paths.
The pre-wetting provides a water reserve inside the CDW grains. The cement
hydration causes a decrease on the volume of water filling the pores. And, therefore, a
humidity gradient appears, forcing the grains to release the water reserve. CDW replacement
percentage affects the water reservation capacity. A higher diffusion has seen at day 14 on
mortars with 50 and 100% of pre-wetting CDW replacement, but for mortars with 100% of
sand replacement the diffusion starts at 5 days age and continues up to 28 days age (fig.2).
It can be observed in the graphic the total resistances contribution (Fig. 4). At early age
there are no hydration products blocking the pore space and consequently the resistances are
lower. The resistance of mortars with natural sand is low when compared with total resistance
of mortar with CDW aggregate. However when used in dry condition CDW content affect this
measures changing cement paste and a large pores are formed. As hydration continues,
porosity decreases and necks form between the pores and the resistance increase. The
dependence of this measure can be associated to water filling the pores, once it is responsible
to the electric conductance. This effect confirms, at 14 days, an ionic motion inside the
electrolyte, named Warburg diffusion.Changes on porous structure can also be analyzed
comparing the results between void index and capillary absorption. At 60 days, the void
index gets closer to the day 1 index, besides the NC-S sample. At 1 day of hydration almost
all of pores are connected with smaller diameter, while at 60, pores with larger diameter and
fewers connections predominate (Fig 6).

Void Index (%)
30
25
20
15
1
7
28
60
Time (days)
NS CD NC CD-S NC-S
Figure 6: Void Index evolution during de hydration.
4. CONCLUSION
The main contributions of this paper are:
- CDW aggregate has influence on the internal process of cement hydration by the
diffusion of water;
- Percentage of CDW replacement has influence on water reservation and diffusion;
- CDW, mostly when its pre-wetted, causes changes on mortars porous structure;
- EIS is an important implement for cement hydration and microstructural development
study.
ACKNOWLWDGEMENTS
The authors gratefully acknowledge the National Council for Scientific and Technological
Development (CNPq), and the Program of Scholarships for Scientific Initiation for financial
support of this work.
REFERENCES
[1] Xu, G., Beaudoin, J. J., Jolicoeur, C. e Pag, M., The effect of a polynaphthalene
sulfonate superplasticizer on the contribution of the interfacial transition zone to the electrical
resistivity of mortars containing silica and limestone fine aggregate. From the Cement and
Concrete Research. Volume 30, Ed. 5, Pages 683-691(2000)
[2] Cabeza, M., Merino, P., Miranda, A., Nvoa, X.R., Sanchez, I., Impedance spectroscopy
study of hardenes Portland cement paste. Form the Cement and Concrete Research. Volume
32, Pages 881-891 (2002)
[3] CABEZA, M., KEDDAM, M., NVOA, X. R., SNCHEZ, I., TAKENOUTI, H.
Impedance spectroscopy to characterize the pore structure during the hardening process of
Portland cement paste. From the Electrochimica Acta. Volume 51, Pages 18311841, (2006)
[4] Vedalakshmi, R., Saraswathy, V., Ha-Won song, Palaniswamy, N., Determination of
diffusion coefficiente of chloride in concrete using Warburg diffusion coefficient. From the
Corrosion Science, Volume 52, Pages 1229-1307, (2009)
[5] Snchez, I., Antn, C., de Vera, G., Ortega, J.M., Climent, M.A., Moisture Distribution
in Partially Saturated Concrete Studied by Impedance Spectroscopy From the J. Nondestruct
Eval, Volume 32, Pages 362-371, (2013).

RECOVERY OF PACKAGING GLASS REFUSE BY MECHANICAL

BIOLOGICAL TREATMENT PLANT- CASE STUDY
Nilmara Dias ; M. Teresa Carvalho
CERENA, Instituto Superior Tcnico, Universidade de Lisboa
Av. Rovisco Pais, 1049-001 Lisboa
Abstract
It is well known that packaging glass can be considered an excellent example of circular
economy due to the characteristics which allow the recycling and reuse without loss of
product quality. However, in some countries, the selective collection systems together with
campaigns to mobilize the population are not completely efficient and there is still packaging
glass in municipal solid waste which is sent to Mechanical Biological Treatment plants
(MBT), incinerators or landfills. In a study developed in Portugal it has been found that there
is a significant amount of glass sent to MBT which is rejected and currently landfilled. Thus,
a diagram was developed to recover the glass refused by MBT for further recycling. The
diagram includes magnetic separation, screening, RecGlass (a device developed to separate
stones from glass particles) and optical sorting. Using samples from 3 different plants, with
glass content between 36 and 60% and different composition and characteristics, the diagram
was tested at laboratory and pilot scale. It was possible to produce a glass concentrate with
approximately 99% in glass content, with a glass recovery ranging from 24 to 64%.The paper
describes the work done and the results achieved.
Keywords: Municipal Solid Waste (MSW); MBT plants; Glass recovery;
1.
INTRODUCTION
The recovery of resources through a sustainable waste management has been seen as
crucial to create sustainable cities. To find out a solution that minimizes the environmental
impacts caused by the increase in waste generation, the scarcity of natural resources and
limited space for the construction of landfills is a global challenge (Kinnaman, 2006;
Halvorsen, 2012).
Today, the waste management models in Europe are based on integrated systems that
maximize materials recovery in order to create a sustainable resource management. The
Packaging and Packaging Waste (PPW) Directive, sets targets for recycling and recovery of
packaging waste. This Directive had a significant impact on the waste management
throughout the European Union (EU).
Selective deposition and collection are common practices in EU, but due to the inefficiency
of the system or to the inadequate population behaviour some recyclables are sent to the
Mixed Municipal Solid Waste (MMSW) (Halvorsen, 2012).
The Mechanical Biological Treatment Plants (MBT) are part of a strategy to comply with
targets to reduce the Biodegradable Municipal Waste sent to landfill, as established in Europe

by Directive 1999/31/EC. However, the high level of recyclable materials contained in the
MMSW and the need to maximize recovery of these materials, stimulated investments in the
MBT facilities to use technology to treat mixed material flows aiming at the recovery of
recyclables beyond the valorization of the biodegradable waste (Lornage, et al., 2007;
Montejo, et al., 2010; Tintner, et al., 2010; Pires, et al., 2011).
In the MBT facilities the MMSW is firstly mechanically sorted, producing, normally, three
main streams: 1) dry waste, composed mainly by paper/cardboard, plastic and wood, where
some recyclables are recovered and generally, part of those products that are not suitable to be
recyclable can be used to produce Residue Derivate Fuel (RDF); 2) organic stream, that is
biologically processed; 3) fine material (normally <20 mm), mainly composed by inert
material as stones, ceramic and glass (Di Maria, et al., 2013; DEFRA, 2013; Dias, et al.,
2014a). The glass contained in this fine fraction is not recovered today due to the lack of a
viable technical and economical solution.
The aim of the work presented here was to develop a process that results in a recyclable
glass product from the MBT heavy reject (MBTr). In this paper the steps to develop a diagram
to recover the glass contained in the MBTr are described. This paper is divided in 4 sections:
the first section shows the cases studied and the MBTr samples composition and particle size
distribution and the glass contained in the MBTr samples from the different plants is
quantified; the second section describes the work carried out at laboratory scale; the third
section shows the main results obtained in the tests carried out at pilot scale. Finally the
conclusions are presented.
2.
CASES STUDY
2.1 Samples
The six samples of MBTr used in the first part of the study were supplied by the six MBT
facilities fed with MMSW operating in Portugal in 2012: RESIESTRELA (RE), SULDOURO
(SD), VALNOR (VN), VALORLIS (VL), TRATOLIXO (TL) and AMARSUL (AS). Figure
1 shows the location of the MBT facilities.
Besides the differences in the composition of the plants feed, the MBT and the stage of the
process where the MBTr is refused is different from plant to plant, resulting in different
characteristics of the MBTr. The MBT equipment and settings are different resulting in MBTr
with different characteristics (more detailed information can be found in Dias, et al., 2014a).

Figure 1- Case studies- Localization and type of biologic treatment of the MBT plants treating MMSW
2.2. Characterisation of the MBTr

The MBTr samples were analyzed in terms of moisture, particle size and composition. The
moisture content of the samples varied from 3 to 22%. The high variability is mainly related
with differences in the MBT plant flowsheets, mainly addition of structuring material.
Figure 3 shows the particle size distribution of the six samples of MBT. As can be seen,
the samples can be divided in three groups: the finer samples, AS and VN, mostly finer than
5.6 mm; the coarser ones, VL and SD, mostly with particles larger than 11 mm); and RE and
TL, with an intermediate particle size distribution.
Figure 4 shows the composition of the MBTr samples, where it can be seen that with the
exception of the VN sample, the MBTr are mainly composed by glass. Due to the very low
content in all samples, ceramics, plastics, brick and metal were aggregated in one class,
labelled as CPBM. As can be seen at Figure 3, the main contaminants in the MBTr are
others, class composed by organic matter, and stones.

100
AS
80
% weight
RE
60
SD
VN
40
VL
20
TL
0
1
4
8
Particle size (mm)
16
Figure 2- Particle size distribution of the MBT samples (Dias, et al., 2014a)
100
% weight
80
31.50
60
26.19
7.28
6.99
28.86
11.96
49.05
11.09
13.88
CPBM
82.87
40
53.29
20
14.38
9.75
Stone
58.34
56.68
40.96
Other
14.76
Glass
33.18
0
AS
RE
SD
TL
MBT Plants
VN
VL
Figure 3 - Composition of each MBTr samples (Dias, et al., 2014a)
Based in the glass content obtained in the characterization carried out (Figure 3) and the
available values of the amount of MBTr produced in 2012 in the plants, Dias et al (2012)
estimated that the packaging glass contained in the MBTr in the six plants was approximately
17000 tonnes of glass, which were landfilled due to the lack of an economic and technical
viable solution for its recovery.
The separation of organic matter is not a technical problem today and ceramic particles are
efficiently removed by optical sorting. Nevertheless, this technology, commonly used in the
decontamination of glass cullet selectively collected, is not efficient in the removal of stones
due to the sub-spherical shape that leads to erratic trajectories (Dias, et al., 2014b). So, the
study of the separation of stones from glass particles based in the shape of particles was
undertaken at laboratory scale.

3. LABORATORY SCALE TESTS

The tests were carried out using the RecGlass, a device specially designed for this
application, which is inspired in the vanner, process used in the mineral concentration, and the
shaking table, a process widely used in the mineral field to concentrate ores by density,
although shape may play a significant role in the process.
2.1. RecGlass
RecGlass is a dry process, efficient in the particle size operating range between
approximately 6 and 11 mm. As represented in figures 4A and B, its main part is an inclined
belt moving upwards. The glass particles, mostly flat, move predominantly upwards,
constituting the glass concentrate, while stones mostly roll down and become part of the
process reject (Dias & Carvalho, 2012b)
Figure 4-RecGlass device diagram (A). Pilot unit (B) photo of the equipment. (Dias, et al., 2014b; Carvalho, et al., 2015a)
An exhaustive number of tests were carried out to optimize the RecGlass operation. Figure
5 shows the relative variation rate, the ratio between the difference in the contents of the
material in the concentrate and in the feed and the contents in the feed, in percentage,
obtained when 5 samples were processed in the optimized process conditions. This rate
measures how much the process feed was enriched or impoverished and allows the
comparison of results obtained when the process is operated with different feed.
RVR (%)

40
20
0
-20
-40
-60
-80
-100
Glass
Stones
AS
RE
SD
VN
MBT Plant
VL
Figure 5 Relative Variation Rate (RVR) obtained in the processing of the 5 MBTr>5.6 Tests carried out with the global
particle size range (Dias, et al., 2014b)
It is clear that RecGlass is able to enrich significantly MBTr in glass through the
elimination of a significant percentage of stones. Nevertheless, the glass content of the
concentrates obtained is not enough for the glass recycling industries (at least 98% in glass
content). Thus, the re-treatment of the separation products must be considered.
The re-treatment was tested, firstly by simulation and then experimentally. Table 1 shows
the results obtained with the samples from VN and SD. Through the re-treatment of the MBTr
by the RecGlass it was possible to remove more 10% of the stones, without significant loss of
glass (less than 7%) in both cases (Landim, et al., 2014).
Table 1- Recovery of the glass and stones in the treated and re-treated TMBr by RecGlass using samples from VN and SD
TMB
Feed
Materials % w/w
50
Glass
VN (fraction >4mm)
20
Stones
65
Glass
SD (Fraction >6 mm)
11
Stones
2.2.
Treatment
Concentrate
% w/w F(%)
65
76
7
20
76
89
4
27
Concentrate of the
re-treatment
% w/w
F(%)
65
62
4
10
77
82
2,7
18
Shaking Table
The shaking table is a versatile gravity concentration device commonly used in mineral
processing (Burt, 1975). A Wilfley type shaking table (Figure 6), from Outokumpu OY, was
used to process the finer fraction of MBTr.

Figure 6- Adapted Wilfley shaking table.
The original surface, riffled and painted with epoxy resin, was not adequate for the
separation due to the low friction coefficient between the particles and the surface. So, a plate
of rubber was placed on the table covering the overall surface of the table in order to be flat
with no rifles. The steady feed of the samples was done through a vibrating feeder.
It was concluded that the shaking table is adequate to reduce the content in stones
occurring in the MBTr. Nevertheless, RecGlass is more efficient, as can be seen in Figure 7.
Around 35% of the stones were recovered on the concentrated product using shaking table
while, using the RecGlass this recovery was around 20%.
Recovery (%)
100
80
60
RecGlass
40
Shaking tables
20
0
Glass
Stones
Recovery in the concentrate product (%)
Figure 7- Recovery in the concentrate product obtained through the processing of the MBTr by RecGlass and shaking table
using VN sample (Martins, et al., 2014)
3.
PILOT SCALE TESTS
An experimental program was carried out at pilot scale with the MBTr from VN, SD and
RE in three different campaigns of tests. The samples of these MBT facilities were chosen
among the six MBTr to guarantee the testing with samples with different characteristics in
terms of particle size distribution and composition. The glass content in the samples was,
respectively, 33%, 43% and 67%.

The flowsheet development was based in the characterization of the samples, the
laboratory work carried out in RecGlass and in the industrial experience in Maltha GlassRecycling Portugal. The pilot plant was built in Malthas facility.
The moisture content of the three MBTr samples was higher than 5%, reaching up to 22%
in the case of VN sample. The first step of the processing was to dry the samples by solar
exposure for some days. Each sample was spread on the floor on top of a plastic film for
natural drying. An overband magnetic separator over the conveyor belt feeding the screen
removed any existing ferrous metal.
The characterization study (Dias, et al., 2012a; Dias, et al., 2014a) showed that the finer
fraction of the MBTr is mainly composed by organic matter. So, the elimination of the
fraction < 6 mm removes most of this material without loss of a significant percentage of
glass. It was observed during the laboratory scale tests that the processing with close particle
size intervals is more efficient than the processing with the whole particle range. A
classification of the remaining > 6 mm fraction was made by screening with a 6 and a 16 mm
opening square mesh. The particle size fractions were separately processed.
The fraction 6-16 mm was pre-processed by RecGlass, in cascade in two stages, before
optical sorting. This process was performed in 3 stages, in cascade too.
The > 16 mm was directly sent to the optical sorting. This is due to the fact that the
particles are large enough to be efficiently processed by optical sorting and the efficiency of
RecGlass is lower in this particle size range due to the smaller difference in the shape of glass
particles and stones (see Dias et al.,2012). The optical sorting was carried out in cascade in
three stages. The flowsheet used is depicted in Figure 8.
Figure 8- Diagram developed to recover glass contained in MBTr (Dias, et al., 2015b)
The optical sorters used are Sesotec, model Spektrum (Figure 9). Recglass and the optical
sorters had vacuum systems above the feed entrance to remove light particles such as
remaining organic matter, plastics, paper, fine bones and wood fiber.

Figure 9- Optical sorter installed in cascade (Dias, et al., 2015b)

The detailed information regarding the whole experimentation procedure and results can be
seen in (Dias, et al., 2015b). Currently, there are no specifications established by the
Portuguese Green Dot (SPV) for the glass cullet recovered from MBT, only from single
stream bottle bank collection (glass from selective collection). Nevertheless, an exercise was
made comparing the existing specifications for the glass cullet recovered from selective
collection and the glass concentrate product characteristics obtained in the present study.
Table 2 shows the results obtained with the three samples (SD, VN and RE) when the two
particle size fractions (> 16 mm and 6-16 mm) were used. As referred, the >16 mm fraction
doesnt exist in VN sample.
It was possible, with the three samples, to obtain a glass concentrate with over than 98%
purity. However, the contaminant content, in organic matter, does not comply with the
specifications considering both fractions (6-16 mm and >16 mm) as can be seen in Table 2,
meaning that a more effective drying would be needed. The recovery of the glass in the glass
concentrate was about 25% to 64%.
Table 2- Composition of the glass concentrate obtained and the Portuguese Green Dot Company specifications for glass
cullet from selective collection (Dias, et al., 2015b)
Contaminants
% (w/w)
Materials
SPV
> 16 mm
Specifications SD
RE
98
99
99
Product Cullet
Infusible with size 40 mm
0.05
0.02 0.03
Infusible with size 40 mm
0.5
0.00 0.00
0.00 0.00
Ferrous metals
0.75
0.00 0.00
Nonferrous metals
0.2
Organic matter (including
0.5
0.91 1.12
other recyclable materials)
6-16 mm
SD
VN RE
98
99
98
0.04 0.15 0.03
0.00 0.00 0.00
0.00 0.00 0.00
0.19 0.00 0.08
1.13
0.76
0.79

5. Conclusions
In Portugal, the inert residual fraction refused by Mechanical Biological Treatment (MBTr)
plants treating mixed waste (MBTr) presents a high content in packaging glass. This product
is not accepted in Glass Recycling plants due to the high degree of contamination with other
materials and the small particle size, decreasing drastically the efficiency of the optical
sorting, the main process commonly used to purify the glass cullet. The MBTr samples
analysed were composed mainly by glass (ranging from 33-65%). It was estimated that
around 17000 ton of packaging glass contained in the MBTr was landfilled in 2012, in
Portugal by the case study plants.
The laboratory work carried out showed that it was possible to remove the main
contaminants of the MBTr (stones) exploiting the differences in the shape of the glass
particles and stones, using equipment RecGlass. However, the final glass product was still
significantly contaminated with other materials.
Based on the laboratory tests and on the characteristics of the contaminants, a flowsheet
was developed to recover the glass contained in the MBTr. The flowsheet included drying,
magnetic separation, RecGlass and optical sorting.
The pilot testing was carryed out using samples from 3 different MTBr plants. The results
show that, starting with samples of MBTr with 33%, 43% and 67% content in glass, it was
possible to increase the content in glass up to approximately 99%. The recovery of the glass in
the glass concentrate ranged from 25% to 64%, depending on the samples characteristics.
ACKNOWLEDGEMENTS
The authors acknowledge the financial support by the Portuguese Green Dot (SPV),
Suldouro, Valnor, Resiestrela, Amarsul, Tratolixo and Valorlis for providing the samples, and
Maltha for the construction of a pilot installation and participation in the tests carried out. We
are also thankful for the PhD grant of one of the authors (ND), provided by the Brazilian
program Science Without Border, process n200883/2012-7 and to FCT
(UID/ECI/04028/2013).
REFERENCES
1. Bonifazi, G. & Serranti, S., 2006. Imaging spectroscopy based strategies for ceramic glass
contaminants removal in glass recycling. Waste Management, 26(6), pp. 627-639.
2. 2. Burt, R. O., 1975. Development of the bartles crossbelt concentrator for the gravity
concentration of fines. International Journal of Mineral Processing, Volume 2, pp.
219-234.
3. Carvalho, M. T., Dias, N. & Brogueira, P., 2015a. Separation by particle shape the
RecGlass device. International Journal of Mineral Processing- In Press.
4. Cimpan, C. et al., 2015. Central sorting and recovery of MSW recyclable materials: A review
of technological state-of-the-art, cases, practice and implications for materials recycling.
Journal of Environmental Management, pp. 181-199.
5. Dias, N., Carvalho, M. & Pina, P., 2012a. Characterization of Mechanical Biological
Treatment reject aiming at packaging glass recovery for recycling. Minerals
Engineering, March, 29(5), p. 72-76.
6. Dias, N. & Carvalho, M. T., 2012b. Novel equipment to separate contaminants from

glass cullet. Bodrum, Turkey, s.n., pp. 931-937.

7. Dias, N., Carvalho, M. & Pina, P., 2012b. Characterization of Mechanical Biological
Treatment reject aiming at packaging glass recovery for recycling. Minerals
Engineering, March, 29(5), p. 72-76.
8. Dias, N., Mximo, A., Belo, N. & Carvalho, M., 2014a. Packaging glass contained in
the heavy residual fraction refused by Portuguese Mechanical and Biological
Treatment plants. Resources, Conservation and Recycling, Volume 85, pp. 98-105.
9. Dias, N., Belo, N., Mximo, A. & Carvalho, M. T., 2014b. Recovery of glass contained in the
heavy residual fraction of Portuguese mechanical Biological Treatment Plants.
doi:10.1016/j.jclepro.2014.05.028. Journal of Cleaner Production, Volume 79, pp. 271-275.
10. Dias, N. et al., 2015. Recovery of glass from the inert fraction refuse by MBT plants
in a pilot plant. Waste Management- Submited
11. Halvorsen, B., 2012. Effects of norms and policy incentives on household recycling:An
international comparison. Resources, Conservation and Recycling , Volume 67, pp. 18-26.
12. Lornage, R. et al., 2007. Performance of a low cost MBT prior to landfilling: study of
biological treatment of size reduced MSW without mechanical sorting. Waste Management,
Volume 27, pp. 1755-1764.
13. Martins, M. M., Dias, N. & Carvalho, M. T., 2014. Application of the shaking table to
separates stones from glass cullet in the fine heavy reject of the Mechanical and
Biological Treatment of MSW. IMPC-Santiago-Chile, s.n.
14. Montejo, C., Ramos, P., Costa, C. & Mrquez, M., 2010. Analysis of the presence of improper
materials in the composting process performed in ten MBT plants. Bioresource Technology,
Volume 101, pp. 8267-8272.
15. Navarotto, P. & Domingues, R., 2012. Materials Recovery from Municipal Solid Waste:
Ecoparc 4, Barcelona a case study. [Online] Available at:
http://www.mater.polimi.it/mater/images/Meetings/Documents/20120627_Milano/navarotto_
11.pdf
[Accessed 2014].
16. Ofori-Sarpong, G. & Amankwah, R. K., 2011. Comminution environment and gold
particle morphology: Effects on gravity concentration. Minerals Engineering ,
Volume 24, pp. 590-592.
17. Pires, A., Martinho, G. & Chang, N.-B., 2011. Solid waste management in european countries:
a review of systems analysis techniques. Journal of Environment Management, Volume 92,
pp. 1033-1050.
18. Smith, P. & Butts, G., 1980. Design of Mineral Processing Flowsheet Case History,chapter
42, AIME. In: 2. e. AL Mular and RB Bhappu (eds), ed. Design of Mineral Processing Plant
Design. s.l.:s.n., pp. 856-871.
19. SPV, 2012. Gesto de Resduos. [Online] Available at:

http://www.pontoverde.pt/[Accessed Maio 2013].
20. Tchobanoglous, G., Theissen, H. & Vigil, S., 1993. Integrated Solid Waste Management
Engineering. Principles and Management Issues. s.l.:McGraw-hill International Editions.
21. Tintner, J., Smidt, E., Bhm, K. & Binner, E., 2010. Investigations of biological processes in
Austrian MBT plants. Waste Management, Volume 30, pp. 1903-1907.

ALTERNATIVES FOR THE PRODUCTION OF STRUCTURAL

CONCRETE WITH FULL RECYCLED AGGREGATES
Taim Castao Cardoza (1), Sandra Lichtblau (2), Stefan Linsel(2), Jos Fernando
Martirena Hernndez (3).
(1) Civil Engineering Department, Constructions Faculty, Universidad de Oriente.
Santiago de Cuba
(2) University of Applied Science, Karlsruhe, Germany
(3) Center for the Research and Development of structures and materials (CIDem),
Universidad Central de las Villas (UCLV), Cuba
ABSTRACT
Aggregates from recycled materials are softer and more porous than natural aggregates, and
thus an increase in water demand is induced. This paper presents the production of a structural
concrete made entirely with mixed recycled aggregates. The influence of material under 63
m in mortar is discussed. The use of 6% of this material in mortars increases the water
demand and brings about a reduction of strength, but it decreases the effective porosity of the
matrix, increases the density of the ITZ and contributes to increase impermeability of the
matrix, as proven on alkalis expansion tests carried out in mortars. The high porosity of the
coarse fraction of recycled aggregate is reduced by coating of the 4-8 mm fraction with a
0.16-0.23 mm cement film, which compensates the high water demand of recycled aggregate,
as demonstrated in the porosity- and the frost-thaw-tests. For the same water to cement ratio,
the amount of superplasticizer, used in concrete made with encapsulated aggregate, was
reduced to a half in comparison with to the non-coated reference. Furthermore the concrete
reaches a compressive strength about more than 30 MPa. The encapsulation leads to a
significant improvement of the impermeability in the concrete structure, which is probably
caused by the densification of the pore structure.
Keywords: recycled, aggregates, strength, porosity, encapsulation, expansion

INTRODUCTION
The use of recycled aggregates from construction and demolition wastes (CDW) is not
widespread because of its variable nature. Impurities, such as mortar adhering on old
aggregate crushed bricks, ceramic materials and gypsum, lead to deleterious reactions
affecting the durability of the material [1]. Adhered mortar has a significant effect on recycled
aggregate properties because it is porous and has numerous cracks, which can cause adverse
effects on recycled aggregate properties compare to natural aggregate (e.g., high water
absorption capacity, lower density and lower strength [2] and can increase the presence of
sulphates, that induces deleterious effects on durability of concrete[3]. In coarse aggregate,
the adhered mortars induces the formation of various Interfacial Transition Zones (ITZ)
between natural aggregate and old cement paste (old ITZ), and the new ITZ between new
cement-pastes and the recycled aggregates [4].
The water absorption of the aggregate could be significantly reduced with a reduction of the
porosity of the coarse aggregate. An encapsulation of the coarse aggregate with cement paste
could provide a suitable solution, by creating a thin shell of hardened cement surrounding the
aggregate grain, which locks the pore system of the grain and provides a less porous surface
which could potentially improve the ITZ [5]. The encapsulation reduces the water absorption
of the coarse aggregate, and thus water demand in concrete, while still benefitting from a
denser ITZ between the treated coarse aggregate and the cement matrix. This method can also
improve the workability, increase the bulk density and reduce the water absorption of
hardened concrete [2, 6].
During the manufacturing process of mixed recycled aggregates a great quantity of very fine
material under 63 m is produced. This fine material, which is mainly composed of the
adhered mortar and cement paste of the recycled material grains, induces an increase of the
water demand, which is reflected in a moderate decrease of the mechanical strength [7]. The
presence of other components like wastes of aggregates with content of reagents materials
could also generate expansion problems and cracking of the matrix [8]. However, these small
particles could compensate discontinuities between the cement grains, and thus improve the
compactness of the cementitious matrix, which could indirectly reduce the water absorption
and the connected porosity in hardened state [9]. These fine grains could contribute to
increase the compactness density of the recycled aggregate-paste interface (ITZ), therefore, it
could reduce locally the relationship of water/cement in the ITZ and diminish the circulation
of liquid by percolation [1].
This paper presents an alternative to reduce porosity by coating of the 4-8 mm fraction of the
mixed recycled coarse aggregate with a thin layer of cement. It is expected that encapsulation
has a positive impact on rheology and mechanical properties of structural concrete entirely
made with recycled aggregate. The properties of coated recycled aggregate as well as concrete
made out of it, shall be assessed and the comparison to the original recycled material concrete
shall be done. This procedure may provide an alternative to produce a medium strength
concrete made entirely with recycled aggregate. Further, the influence on the properties of
mortar due to the very fine fraction - under 63 m - of recycled aggregate from mixed
construction debris, is assessed. The impact on the rheology and the pore structure in mortars
is assessed, as well as the durability of mortars, subjected to alkaline attack.

MATERIALS
Recycled aggregates were produced by crushing the CDW in a recycling plant in the south of
Germany. The recycled aggregate is referred as M2. The natural calcareous aggregate,
originated from a plant in Germany, was taken as reference for the study and is referred as
M5. Both aggregates contain a fine fraction of 0-4 mm and a coarse fraction of 4-8 mm. The
very fine particles under 63 m have not been removed. Both materials were completely
characterized. Table 1 presents the results of the characterization. The grain size distribution
of the used coarse and fine aggregates (recycled and reference) were assessed following to the
standard DIN EN 933-1:2012 and complies with specifications.
Table 1. Physical- and mechanical properties of the used natural and recycled aggregate
Properties
Current density fraction 0-4mm(g/cm3)
Apparent density fraction 4-8mm( g/cm3)
Water absorption (%) 0-4mm
Water absorption (%) 4-8mm
Unit Weight compacted fraction 4-8mm(kg/m3)
Unit Weight compacted fraction 0-4mm(kg/m3)
Organic impurities
Clay particles (%)
Abrasion resistance, Los Angeles (%)
Frost-Thaw resistance (%).
M2
2.344
2.535
4.75
5.34
1229.16
1248.50
plate 2
10.80
28.75
6.05
M5-D
2.568
2.769
1.02
1.08
1553.40
1452.00
plate 1
3.60
23.50
2.31
Aggregate M2 has a lower density than the reference aggregate, what is the characteristic of a
more porous material and what possibly leads to a higher water absorption. However, the
resistance to fragmentation of the recycled aggregate, measured following the standard EN
1097-2:2010 for Los Angeles test, does not result a high decrease compared to the reference.
The mineralogical composition of both aggregates was assessed aided by X-ray diffraction
tests. The most common phases, which can also be found in most recycled aggregates, are
quartz, mica, clay particles and unhydrated cement.
Reference aggregate has a calcareous origin with calcite and dolomite as main phases,
whereas mixed recycled aggregate shows several phases, including anhydrate cement, which
is typical for these aggregates [10].
EXPERIMENTAL METHODS
Mortar prisms with the dimension of 40 x 40 x 160 mm were produced and cured until 7 and
28 days in order to assess the influence of the very fine fraction on the properties of mortars.
The prisms were subjected to the compressive strength tests according to the standard UNE
83-821:92 and also to the water absorption test according to the standard DIN EN
480-5:1997-02.

The impact of the very fine fraction in the durability of mortars was evaluated according to
the procedures established by ASTM C33. Volume changes taking place in mortar prisms
through alkaline attack and were measured according to the ASTM C227 norm.
Concrete cubes 100 x 100 x 100 mm were cast to test the compressive strength and the
porosity as a means of assessing the influence of the encapsulation of the 4-8 mm fraction.
Five series of concrete were cast, depending on the properties of coarse aggregates: (1)
Normal concrete with the reference aggregate M5, (2) Concrete with the recycled aggregate
M2, (3) Concrete with recycled aggregate M2, where the coarse fraction 4-8 mm is coated
with cement, M2CO, (4) Concrete with recycled aggregate M2, where the coarse fraction 4-8
mm is coated with cement/fly ash (70/30), M2CFA70-30; and (5) Concrete with recycled
aggregate M2, where the coarse fraction 4-8 mm is coated with cement/fly ash (50/50),
M2CFA50-50. The fine aggregate was used as received, so it was not modified or improved.
In table 3 are shown the concrete mixture designs.
Fresh properties of concrete were measured. The spread at the flow table was fixed at 40 cm,
measured according to the standard DIN EN 12350-5:2009. The Compressive strength was
measured at 7 and 28 days and the water absorption was measured at the age of 28 days,
following the Cuban standard NC 345:2005.
DISCUSSION OF RESULTS
Influence of the fraction under 63 m on the properties of mortars
The studied rheological properties of the mortars, presented in table 2, are compromised by
the presence of the very fine particles. The mortar series produced with fines, M2f, demanded
the maximum dose of the manufacturers information for superplasticizer.
Table 2. Evaluation of the impact of very fine fraction in the rheology of mortars
Sample
M5
M2
M2f
Cement
(g)
450
450
450
Fine aggregate
(g)
1350
1350
1350
w/c
0.61
0.61
0.61
Additive
(%)
0
0.5
2
Consistency
(cm)
15.00
14.50
15.00
The presence of fines in the mixture M2f causes a decrease about 7-8% of the strength with
respect to the sample without fines M2, as presented in the following Figure 1. This can be
related to the presence of adhered mortar and the high absorption of the recycled fine
aggregate, which propitiates an increment in the water demand, a weaker mortar matrix and
therefore a decrease of the compressive strength.

60.0
0.80
7d
0.70
0.69
28d
0.60
0.55
40.0
0.50
30.0
Water absorption, %
compressive strength, MPa
50.0
20.0
10.0
0.46
0.40
0.30
0.20
0.10
0.00
0.0
M2
M2
M2f
M2f
M5-D
M5-D
Figure 1.Compressive strength of mortar

samples at 7 and 28 days
Figure 2.Water absorption for the mortar

samples M2, M2f and M5
Figure 2 shows the percentage value of water absorption for the samples M2, M2f and the
reference M5. The series M2f yielded better results than those of the series M2, without fines.
Sample M2f, exhibited slightly smaller values for mechanical strength than series M2. The
lower and therefore better value of the water absorption for M2f in comparison to M2 reflects
improvement in the pore structure, especially capillary pores, through the presence of very
fine particles which increase packing of the matrix.
These results are confirmed by tests which measure the porosity, aided by mercury intrusion,
presented in figure 3. The mortar series M2f increases the number of pores in the size around
0.01-0.1 m, which corresponds to a refinement of the pore structure consistent to the
reduction of water absorption. [11]
Figure 4 exhibits the results of the test for volume changes in mortars according to the
standard ASTM C 227. The standard ASTMC C33 refers, that an aggregate is potentially
reactive, if the expansion after three month is bigger than 0.05% and after six months beyond
0.1% from beginning of the test. The standard also specifies, that the expansion is not
considered as excessive, if it is higher than 0.05% after three month, but lower than 0.1% after
six month.
Sample M2c (without fines) reaches the highest value of expansion, above the established
maximum limit after both, three and six months. The series M2f, with fines, attains a decrease
of the expansion probably through the presence of very fine particles leading to a more
impermeable matrix.

Figure 3. Relative volume of pores for mortar

samples
Figure 4. Expansions of mortars after 304 days

according to ASTM C 227
Impact of encapsulation on the properties of recycled aggregate

The encapsulated aggregate studied was subjected to a frost-thaw test according to DIN EN
1367-1:2007. This test is used to describe the pore structure of the outer surface of the
material, and thus predicts indirectly water absorption, the porosity and pore structure of the
future concrete. [12] Figure 5 presents the values of factor FS for all aggregates. Figure 6
presents the results of the water absorption test carried out for the aggregates subjected to this
study. The aggregate M2 does not fulfill aggregate standards in terms of water absorption.
The encapsulation decreases water absorption and FS to a half. The best results were obtained
for the aggregate which is coated only with Portland cement.
7
Frost Test
5
Freeze-thaw soundness loss %
M5-D
M2
M2CO
M2CFA 70-30
M2 CFA 50-50
0
M5-D
M2
M2CO
M2CFA70-30
M2CFA50-50
Figure 5: Results of freeze-thaw tests carried out on Figure 6: Results of water absorption for the
the aggregates used for the study
aggregates used for the study
Impact of encapsulation of aggregate on fresh concrete properties

Table 3 presents the properties of the fresh concrete mixtures. By coating the aggregate with
cement, the demand of admixture can be reduced to a half, in respect to M2, to reach the
desired consistency with the same W/C ratio. That indicates the decrease in water absorption
for encapsulated coarse aggregate and could apparently compensate the extra water demand.

Table 3. Rheology and mechanical properties of concrete

Series
Cement
(g)
Sand
(g)
Coarse
aggregate
(g)
w/
c
SP
(%)
Consist.
(cm)
Strength 7d
(MPa)
Strength 28d
(MPa)
M5
M2
M2CO
M2CFA
70-30
M2CFA
50-50
504
504
504
570
570
570
897
897
897
0.5
0.5
0.5
1.1
1.6
0.9
42
42.5
40.5
35.56
34.15
36.78
46.28
45.32
43.86
504
570
897
0.5
1.2
40.0
36.27
38.96
504
570
897
0.5
0.98
39.5
36.54
Impact of encapsulation on hardened concrete properties

Results of 7 and 28 days compressive strength results of concrete series are presented at Table
3. All mixtures have been cast with a similar W/C ratio. The superplasticizer has been added
until the desired consistency was reached -spread 40 cm-, so one would expect similar
strength results in all series. Concrete series M2CO reached similar strength values as the
reference aggregate, for a similar water demand. Concrete series M2 also reached similar
strength values, but at the cost of significantly increasing the dosage of SP. Concrete made
entirely with recycled aggregate reaches strength above 30 MPa, which is at the same level as
normal concrete. The mixture, where the coarse aggregate has been coated with a blend of
cement and fly ash, have reached a slightly lower strength, especially at 28 days; this is
probably caused by a weaker ITZ due to the poorly reactive fly ash [13].
Impact of encapsulation on pore structure of concrete
Water absorption was measured on concrete cylinders at the age of 28 days to assess the
impact of the encapsulation on the pore structure and its connectivity. Figure 7 presents the
results. As expected, the series M2 has a higher water absorption as the reference series, most
likely because of the high porosity of the coarse aggregates. The series M2CO reaches similar
values to the reference series M5, but a significant reduction of porosity in comparison to the
non-encapsulated material. Coating with pure cement seems to improve water absorption,
probably due to the produced changes in the pore structure by the encapsulation of the coarse
aggregate, especially at the ITZ [14]. Figure 8 presents the results of the effective porosity,
measured in the samples M2, M2CO and M5, which also confirm the trend explained before.
Encapsulation with plain cement has reduced the effective porosity of concrete to a half, what
is proportional to the reduction of water absorption on the coarse aggregates, as presented
above.

8,000
Porosidad efectiva(%)
7,000
6,000
5,000
4,000
7,393
3,000
2,000
3,903
3,880
1,000
0,000
M5-D
M2
Muestras
M2CO
Figure 7: Results of water absorption for the Figure 8: Results of the effective porosity
concrete series after 28 days
for the concrete series at 28 days
CONCLUSIONS
The presence of a very fine fraction in mortars produced with mixed recycled aggregate
increases the water demand, which implies the necessity to use more superplasticizer, or more
effective additives. However, this fall of resistance is compensated with a refinement of pores
in the cementitious matrix, which helps to diminish the water absorption and impedes the
movement of harmful substances through the matrix. This has had a positive impact on the
evaluation of the alkaline reaction of the aggregate in mortars.
Encapsulation of the 4-8 mm fraction of coarse recycled aggregate brings about a significant
improvement on the quality of the aggregate, and thus enables the possibility to produce a
structural concrete with a strength around 30 MPa by using 100% recycled aggregates.
Encapsulation increases impermeability of the aggregate, which reflects on a significant
reduction of porosity of the aggregate (50% reduction), and thus a much lower water demand
in concrete.
THANKS GIVING
The authors want to thank the German Alexander von Humboldt Foundation for the support
provided for carrying out this research. Further, the authors appreciate the support of the
Laboratory for Building Materials at the Ecole Polytechnique Federal de Lausanne and the
University of Applied Sciences in Karlsruhe in Germany for the realization of the
experimental tests.
LITERATURE
[1] Morales M., Zamorano M.,Ruiz M.A., Valverde E.I., Characterization of recycled aggregates
construction and demolition waste for concrete production following the Spanish Structural Concrete
Code EHE-08, in Construction and Building Materials 25 (2011) 742748,
doi:10.1016/j.conbuildmat.2010.07.012.
[2] Sallehan I. and Mahyuddin R., Influence of surface/treated coarse aggregate on compressive
strength of concrete in International Journal of Civil, Architectural, Structural and Construction
Engineering Vol. 8 Nr. 8, 2014. International Science Index Vol:8, No:8, 2014, doi:
waset.org/Publication/9999071.

[3] Marinkovic S., Radonjanin V., Maleev M., Ignjatovic I. Comparative environmental assessment
of natural and recycled aggregate concrete, in Waste Management 30 (2010) 22552264,
doi:10.1016/j.wasman.2010.04.012.
[4] Xiao J.,Wengui L., Sun Z.,, Lange D.A, Shah S.P., Properties of interfacial transition zones of
recycled aggregate concrete tested by nanoindentation in Cement & Concrete Composites 37 (2013)
276292, doi.org/10.1016/j.cemconcomp.2013.01.006
[5] Jiusu L., Hanning X., Yong Z. Influence of coating recycled aggregate surface with pozzolanic
powder on properties of recycled aggregate concrete in Construction and Building Materials 23
(2009) 12871291, doi:10.1016/j.conbuildmat.2008.07.019.
[6] Mller H. S., Linsel S., Garrecht H., Wagner J.P., Thienel K.C., Hochfester Konstruktiver
Leichtbeton, Teil 1: Materialtechnologische Entwicklungen und Betoneigenschaften (High strength
lightweight construction concrete, Part 1:Technology development and concrete properties), in:
Beton- und Stahlbetonbau (Journal of Concrete and Reinforced concrete), 7(2000).
[7] Zega C., Di Maio A., Use of recycled fine aggregate in concretes with durable requirements. In
Waste Management 31 (2011) 23362340.
[8] Ge Z. et al., Mix design of concrete with recycled clay brick powder using the orthogonal design
method in Construction and Building Materials 31 (2012) 289293.
[9] Pereira P. et al.,The effect of superplasticizer, on the workability and compressive strength of
concrete made with fine recycled concrete aggregates in Construction and Building Materials 28
(2012) 722729.
[10] Limbachiya M.C., Marrocchino E., Koulouris A., Chemicalmineralogical characterisation of
coarse recycled concrete aggregate in Waste Management 27 (2007) 201208,
doi:10.1016/j.wasman.2006.01.005.
[11] Yang J. et al.: Concrete with recycled concrete aggregate and crushed clay bricks. Construction
and Building Materials 25 (2011) 19351945.
[12] Gokce A., Nagataki S., Saeki T., Hisada M., Identification of frost-susceptible recycled concrete
aggregates for durability of concrete in Construction and Building Materials 25 (2011) 24262431,l
doi:10.1016/j.conbuildmat.2010.11.054
[13] Casuccio M., Torrijos M.C., Giaccio G., Zerbino R., Failure mechanism of recycled aggregate
concrete
in
Construction
and
Building
Materials
22
(2008)
15001506,
doi:10.1016/j.conbuildmat.2007.03.03.
[30] Kubissaa W., Jaskulski R, Measuring and Time Variability of The Sorptivity of Concrete in
Procedia Engineering 57 (2013) 634 641, doi: 10.1016/j.proeng.2013.04.080.

STUDY OF RECYCLED AGGREGATES MANUFACTURED WITH

BIOMASS BOTTOM ASH FOR APPLICATION IN CEMENT TREATED
GRANULAR MATERIALS
Cabrera M (1), Agrela F (1), Rosales J (1), Dorado C (2), Galvn A.P. (1)
(1) Area of Construction Engineering, University of Cordoba, Cordoba, Spain
(2) English Department, IES N.S. de la Cabeza, Junta de Andaluca, Spain
Abstract
Recently, environmental protection has a great concern in world, where recycling of has become a
viable option to be incorporated into roads applications. A specific example of this may be the reuse of
residues as construction and demolition waste (C&DW) of buildings or bottom ash from biomass (BBA).
The topic of the present research is to evaluate the use of recycled mixed aggregates (RMA) and
recycled concrete aggregates (RCA) from C&DW, and BBA in the manufacture of Cement Treated
Granular Materials (CTGM). The materials were characterized under standard laboratory tests and the
effect of compaction was tested in different specimens. For that, their physical properties were
determined according to standard laboratory procedures and their mechanical behavior was analyzed
according to the following parameters: compressive strength, resilient modulus and dimensional changes.
However, it is essential not only to evaluate their physical and mechanical properties but also
the environmental impact derived from their second cycle-life must be assessed in detail.
The present research is focused also on the analysis of the environmental pollution potential of
RMA manufactured with BBA (with 3.5% cement). This manuscript studies the maximum potential
leachable of the granular materials. Then, based on the compliance test (UNE 12457-3) materials are
classified based on their hazardous potential according to the Landfill Directive 2003/33/EC.
Subsequently, the performance of the test tank (NEN 7343) allowed to analyze the behavior of
RMA mixed with BBA samples as monoliths. Finally, different groups of pollutants were established
according to the highest pollution levels and the different behavior leaching patterns at long-term were
identified.
Keywords: Recycled aggregate; Biomass; Bottom ash; Mechanical Behavior.

1.
INTRODUCTION
Aggregates from recycling construction and demolition waste (CDW) are increasingly used in
new construction due to various factors such as economic savings, but, above all, because of
the environmental benefit that this process entails. Possible applications of these aggregates
include the construction of earthworks, base or sub-base layers of pavements, or the
manufacture of new concrete
In recent decades, several studies have been undertaken to investigate the possibility of using
recycled aggregates from construction and demolition waste (C&D) in road bases or subbases. Inside this C&D, we can find recycled concrete aggregates (RCA) and recycled
aggregates mixed (RMA) [1-3].
Recent research has shown that the mechanical behavior of ARM and recycled aggregate
concrete (ARH) laboratory was adequate, showing resistance and compression 7 and 28 days
very similar to that of a natural aggregate used in the manufacture of Cement Treated
Granular Materials (CTGM) [4 -7].

In addition to using C&D for application in bases and sub-bases for roads, it is becoming
more common the incorporation of industrial by-product as a raw material in the construction
sector [8,9].
Industrial by-products of the combustion of biomass have been investigated for use in
construction materials. Biomass fly ash (BFA) has been studied as a filler in self-compacting
concrete [10].While fly ash utilisation has been extensively studied, similar studies on the
effective management and utilisation of bottom ash have been scarce. BBA are traditionally
disposed of in landfills.
However, some investigations were conducted to study the mechanical properties of BBA and
its possible use in the construction sector. Hinojosa et., al 2014 [9] studied the chemical
properties of BBA. Cabrera et., al 2014 [8] studied the physical properties of BBA and its
possible use on roads. Beltran et al (2014) [11] conducted an investigation related to the
application of BBA for non-structural concrete made with recycled aggregates.
Therefore, new opportunities and recommendations should be sought for the application of
other types of aggregates and waste. Thus, this work aims to promote the recycling of the
C&W and BBA, by studying the possibilities of applying the mixed and concretes recycled
aggregates, studying its mechanical behavior in the manufacture of CTGM with different
additions of BBA as well as its environmental performance.
2.
MATERIALS
2.1.
Cement
The cement used in this work according to the ASTM-C150 was CEM II/B-L 32.5 N (referred
to as CEM-II). The main properties of the cement are shown in Table 1.
Table 1: Properties of cement
SiO2
Al2O3
Fe2O3
%
%
%
16.29
3.74
2.16
2.2.
CaO
%
60.01
MgO
%
0.72
SO3
%
2.57
K2O
%
0.61
Loss of ignition
%
12.07
Recycled Aggregates
Recycled mixed aggregate (RMA) from the crushing of concrete blocks and masonry blocks
from the treatment of CDW in Gecorsa (Crdoba) were used in this research, also Recycled
concrete aggregate (RCA) were used, RCA were obtained from the recycling plant, this
process was performed in the firm Aristerra, located in Mlaga.
The both properties are summarised in table 2.

Table 2:Properties of the recycled aggregates

PROPERTIES
RCA
RMA
UNE - EN 1097 - 01
Density-SSD (kg/m3)
0-4 mm
4-31.5 mm
2.51
2.55
2.03
2.11
UNE - EN 1097 - 01
Waterabsorption (%)
0-4 mm
4-31.5 mm
Water-soluble sulphate content

Acid-soluble sulphatecontent
Composition
Bituminuos %
Ceramicparticles %
Concrete and mortars %
Natural aggregates
Gypsum %
test method
6.28
5.33
0.17
0.26
9.12
10.83
0.51
0.79
0.68
0.25
82.77
16.3
0
0.10
33.83
54
12.11
0
UNE - EN 1744-1
UNE - EN 1744-1
prEN 933-11
The grading curves of the RCA and RMA materials are shown in Fig. 1. Additionally, the
particle size distribution curve range of the material known as soilcement 40 mm is
represented. This material is named SC-40 by the Spanish General Technical Specifications
for Road Construction [12].
Figure 1: Particle size distribution curves compared with the granulometric limits.
In Fig. 1 we can see particle size distribution of both recycled aggregates is within the limits
established by regulation for SC-40. The particle size distribution of BBA was continuous,
allowing an appropriate particle size distribution.
2.3.
Biomass Bottom ash
Biomass Bottom Ashes (BBA) from the thermal plant located in Puente Genil, Crdoba
(Spain) was used as an addition to manufacture of CTGM. They are obtained from the
combustion of several agricultural wastes as olive pruning and other plant compounds, called
biomass. First, Biomass is introduced into a chamber and then, is combusted at a temperature
of 405 C to generate steam that flows through a closed-loop system. BBA is the noncombusted waste generated. Its properties are exposed in Table 3.

Table 3: Properties of biomass bottom ash

PROPERTIES
Size (mm)
Density (kg/dm3)
Friability ratio (%)
Si
Ca
K
Elemental content
Mg
(%)
Fe
Al
Na
Ti
Sulphur content (%)
Soluble sulphate (%SO3)
Organic matter content (%)
BBA
0-4
2.02
28.10
34.10
22.17
17.26
15.01
2.46
1.77
0.72
0.19
0.09
0.31
0.45
4.57
test method
UNE - EN 1097 - 01
UNE 83-115
UNE 83-115
UNE 80-215
UNE - EN 1744-1
As it is showed in Table 3, BBA presented a low density. Respect to water absorption

capacity, BBA obtained a high value (28.1%). However, the friability values were higher in
BBA respect to a natural sand. Regarding chemical properties, it is highlighted the high
organic content (4.57%) and chloride content (0.21%).
2.4.
Mixtures of recycled materials and dosage
Four mixtures of were produced in laboratory. Different types of aggregates were used. In
Table 4 are shown the different quantities of materials in Kg/m3applied in the mixtures used
for this performance.
Table 4: Dosages of the CTGM mixtures
DOSAGE (Kg/m3)
RCA100
RCA93-BBA7
RMA100
RMA93-BBA7
RCA
2100
1911
-
RMA BBA
138
1935
1707 129
Cement
74
69
68
64
3.
EXPERIMENTAL TESTS AND RESULTS
3.1
Modified Proctor test UNE-EN 103-501
Water
252
238
213
261
%Cement
3.50
3.26
3.50
3.31
The Proctor compaction test consists of compacting soil samples at given water content in a
standard mould with standard compaction energy. The standard modified Proctor test uses a
152 mm diameter mould with the compaction of five separate layers of soil using 60 blows by
a 4.5 kg hammer falling 457 mm.

RMA100
2.15
RMA93-BBA7
2.1
2.15
RCA 100
2
1.95
1.9
1.85
1.8
1.75
RCA
2.1
RCA93-BBA7
maximum density g/cm3
Dry Density (g/cm3)
2.05
RCA+BBA
RMA
2.05
RMA+BBA
2
1.95
1.9
1.85
1.8
1.7
1.75
1.65
7
11
13
15
Moisture Content (%)
17
1.7
19
10
15
Time sg
20
25
30
Figure 2: Moisture-density relationships and Vibrating hammer times

The relationship between moisture and dry density is shown in the curves plotted in Fig. 2.
BBA substitution batch values are lower than the rest because of the BBA density. The results
of the Modified Proctor test. All the values of maximum dry density range from 1.83 to 2.10
g/cm3.
3.2.
Vibrating hammer times NLT-310/90
The compaction time was calculated using a vibrating Kango hammer with a mass of 10 kg, a
ram steel of 3.5 kg, an application frequency of 1900 rpm and an electrical power of 750 W.
The test was performed using the CBR mould, a cylinder 152.5 mm in diameter and 179.8
mm in length. The specimens were made with the dosages shown in Table 4 on the dry weight
and the moisture content found in the Modified Proctor test.
Different compaction energies were used varying the time of application of the load exerted
by the vibrating hammer (10, 15, 20 and 25 seconds). Compaction was applied in three layers
with different compaction times in order to obtain a density greater than 98% of the Modified
Proctor Test. Fig 2 shows that the time required to obtain a vibration hammer of 98% Proctor
density was 7 s for RCA, 9 s for RCA+BBA, 18 s for RMA and 25 s in the case of
RMA+BBA.
3.3.
Compressive strength NLT-305
Twelve specimens, were manufactured for the determination of the compressive strength of
the CTGM, the specimens were unmolded and stored in a moist chamber following the
standard specifications. Spanish regulation requires a resistance higher than 2.5 MPa at 7 d in
specimens manufactured in a CBR mould.
Table 5 shows the results obtained in every CTGM produced using recycled aggregates (RA)
and mixtures of RA+BBA at 7 and 28 d. The highest compressive strength values were
obtained in the cement-treated mixtures manufactured with BBA.
Table 5: Compressive strength results
Age (Days)
RCA
RCA+BBA
RMA
7
2.94
3.28
3.19
28
3.87
4.05
4.31

3.67
4.92
RMA+BBA
Table 5 shows the results obtained in every CTGM produced using recycled aggregates (RA)
and mixtures of RA+BBA at 7 and 28 d. The highest compressive strength values were
obtained in the cement-treated mixtures manufactured with BBA.
3.4
Splitting test UNE EN 12390-6
A test for tensile strength in which a cylindrical specimen is loaded to failure in diametral
compression applied along the entire length. This test was performed according to UNE EN
12390-6. This test was carried out in a concrete compression testing machine. Three
specimens were manufactured per mixture.
Table 6: Splitting test and modulus of elasticity
Splitting test (MPa)
Modulus of Elasticity (Mpa)

28 days
RCA
RCA+BBA
RMA
RMA+BBA
0.412
0.794
0.588
0.897
3274.508
3762.617
3384.287
4717.911
Table 6 illustrates the individual results of splitting test and modulus of elasticity. Both tests
show that when adding 7% of BBA in mixtures the values increases when they are compared
to control samples (RCA and RMA). Observing the increasing of splitting test data in 72%
for RMA+BBA referred to the control sample (RMA)
3.5
Modulus of Elasticity UNE 83316
This test was performed according to UNE 83316. The development of elastic modulus of
CTGM materials was investigated using the stress-strain relationships of the mixtures
identified from the strength tests. Modulus of elasticity was determined as the initial secant
line from the origin to the point in the curve stress-deformation where the 20 percent of the
ultimate stress is signed.
The mean values obtained of modulus of elasticity after 28 days are reported in table 6, which
present an approximate increase in the RCA+BBA mixtures of 15% respect to RCA and the
RMA+BBA mixtures present an 40 % increase respect to the RMA.
3.6.
Compliance test UNE-EN 12457-3
The compliance test according to the standard UNE-EN 12457-3 was conducted to estimate
whether recycled construction materials and biomass bottom ash satisfy European regulations.
To classify those materials according to the Landfill Directive [13].
The procedure consists of a two-step batch leaching test reaching two liquid/solid ratios. This
method involves stirring the solution in both steps at the natural pH of the material. In the first
step, the solution is shaken at L/S of 2 L/kg, and the second step uses an L/S ratio of 10 L/kg.
Table 7 shows the green inert materials, inert value limits that are exceeded are given in bold
and yellow, while non-hazardous limits that are exceeded are underlined and in red.

Table 7: Leachate concentrations (mg/kg)

BBA
L/S 2
1.92
0.03
1.24
0.15
0.49
0.13
0.68
0.01
0.18
0.01
Cr
Ni
Cu
Zn
As
Se
Mo
Sb
Ba
Hg
RCA
L/S 10
2.19
0.03
1.04
0.09
0.25
0.05
0.59
0.01
0.35
0.35
L/S 2
0.18
0.01
0.03
0.03
<0.001
0.047
0.16
<0.001
0.11
<0.005
L/S 10
0.50
0.05
0.1
0.17
0.01
0.075
0.17
0.01
0.28
0.02
L/S 2
0.27
0.01
0.01
0.02
0
0.01
0.15
0.01
0.1
0
RMA
L/S 10
0.37
0.01
0.02
0.24
0.01
0
0.2
0.02
0.32
0
According to the results obtained in Table 7, As and Hg are identified as the most conflictive
elements. Other relevant elements are Cr, Ni, Cu, Se and Mo (exceeding the inert limit values
in most cases).
Based on the results of the compliance test, due to the high potential contaminants of BBA,
they are unable to be applied in civil engineering as isolated materials. That is why this study
proceeded to analyse mixtures of BBA with other materials and 3.5% cement to evaluate if
reducing the volume of BBA in the sample reduced the contaminant load.
3.7
Tank leach test NEN 7345
In the test a monolithic specimen is subjected to leaching in a closed tank to evaluate surface
area related release. The leachant demineralised water is renewed after 8 hours and 1, 2, 4, 9,
16, 36, 64 days using a leachant to product volume ratio (L/V) of 4. The results are expressed
in mg/m2.
Fig. 3 shows the diffusion curves of the most conflictive elements detected in monolithic
samples cemented with CEM II/BL 32.5.
BBA+ RMA + CEM
BBA+ RMA+ CEM
1
Hg
Release (mg/m2)
Release (mg/m2)
As
0.1
0.01
0.01
0.001
0.0001
0.001
0.1
0.1
1 Time (days)
10
100
0.1
Time (days)
10
100
Figure 3: Diffusion release curves of more conflictive elements for monolithic mixtures.
Figure 3 illustrates the diffusion release curves of the elements As and Hg Additionally, a
slope of 1:0.5 (the blue line) is represented graphically to facilitate the identification of the
mechanisms that govern the release. Previous researchers have proven that pure diffusioncontrolled release implies a 1:0.5 slope.

4.
CONCLUSIONS
This research present a technical assessment of recycled mixed aggregates (RMA), recycled
concretes aggregates (RCA) and biomass bottom ash (BBA) as material for construction in
road bases and sub-bases, proving that it is possible to apply these materials in CTGM.
The following particular conclusions can be identifying:
-The low density of BBA caused a decreasing of percentage of cement added when the
materials were mixtured with BBA
-Similar pattern of mechanical behaviour was observed for compressive strength, splitcylinder test and modulus of elasticity, increasing the mechanical properties in all the
samples.
-Data from the tank test performed on monolithic samples indicated that this demonstrates the
effectiveness of the S/S treatment for materials that present a high pollution potential.
The present work has proven for certain percentages in dosage the mechanical ability of BBA
used as material construction in civil infrastructures.
REFERENCES
[1] Poon CS and Chan D. Feasible use of recycled concrete aggregates and crushed clay brick as
unbound road sub-base. Construction and building Materials 20 (2006), 578-585.
[2] Molenaar AAA, van Niekerk AA. Effects of gradation, composition, and degree of
compaction on the mechanical characteristics of recycled unbound materials. Transportion Res.
Rec. 2002;1787:7382.
[3] Chini AR, Kuo SS, Armaghani JM, Duxbury JP. Test of recycled concrete aggregate in
accelerated test track. J Transp Eng 2001;127(6):48692.
[4] Xuan DX, Houben LJM, Molenaar AAA, Shui ZH.(2012). Mechanical properties of cementtreated aggregate material, a review. Material Design, 33:496502.
[5] Nataatmadja A. and Tan Y.L. (2001). Resilient response of recycled concrete road aggregates.
Journal of Transportation Engineering. 127(5): 450-453.
[6]Vegas, I.; Ibaez, J.A.; Lisbona, A. (2011). Pre-normative research on the use of mixed
recycled aggregates in unbound road sections. Construction and Building Materials, 25, 26742682.
[7]Poon CS, Chan D. (2006). Feasible use of recycled concrete aggregates and crushed clay brick
as unbound road sub-base. Construction and Building Materials, 20, 57885.
[8] Cabrera, M., Galvin, A. P., Agrela, F., Carvajal, M. D., & Ayuso, J. (2014). Characterisation
and technical feasibility of using biomass bottom ash for civil infrastructures. Construction and
Building Materials, 58, 234-244.
[9] Hinojosa, M. J. R., Galvn, A. P., Agrela, F., Perianes, M., & Barbudo, A. (2014). Potential
use of biomass bottom ash as alternative construction material: Conflictive chemical parameters
according to technical regulations. Fuel, 128, 248-259.
[10] Cuenca J, Rodriguez J, Martin-Morales M, Sanchez-Roldan Z, Zamorano M. (2013). Effects
of olive residue biomass fly ash as filler in self-compacting concrete. Construction and Building
Materials, 40,702709.
[11] Beltrn, M.G., Agrela, F., Barbudo, A., Ayuso, J., Ramrez, A., 2014. Mechanical and
durability properties of concrete manufactured with biomass bottom ash and recycled coarse
aggregates. Construction and Building Materials 72, 231-238
[12] PG-3. Spanish code for technical general specification for road and bridge construction.
[Pliego de Prescripciones Tcnicas Generales para Obras de Carreteras y Puentes. PG-3]. B.O.E.
7 Julio, Spain; 1976.
[13] Council Decision DC 2003/33/EC of 19 December 2002. Establishing criteria and

procedures for the acceptance of waste at landfills pursuant to Article 16 and Annex II to
Directive 1999/31/EC

SECONDARY BUILDING MATERIALS FROM CONSTRUCTION AND

DEMOLITION WASTE: ENVIRONMENTAL BENEFITS AND COSTS
Karin Weimann (1) and Christian Adam (1)
(1) BAM Federal Institute for Materials Research and Testing, Department 4: Materials and
Environment, Germany
Abstract
Due to the large quantities of construction and demolition waste (CDW) in Europe, its reuse or
recycling is of particular importance. Although several countries already recycle high amounts of
CDW, the use as secondary raw materials is often limited by inferior building material properties.
Specific characteristics, like high porosity and low density are caused by hardened cement paste in
crushed concrete and the content of mortar and plaster in brick debris. Impurities like wood or gypsum
and also harmful substances like organic pollutants may be a major problem for a reuse and should be
minimized. Therefore unwanted materials and impairing substances have to be separated from the
secondary building material stream. This can be done during the demolition process by using
techniques for selective dismantling or during the subsequent treatment of the resulting rubble. Since
almost all processing steps are associated with environmental impacts, the benefits of saving natural
resources by applying secondary building materials should be weighed carefully. An environmental
performance evaluation was undertaken to assess different techniques for reducing gypsum in recycled
concrete aggregates, aiming at a minimization of elutable sulfates. These results were compared to the
environmental impacts of the extraction of natural aggregates for concrete.
Keywords: demolition, selective dismantling, CDW processing, gypsum, environmental

evaluation
1.
INTRODUCTION
Regarding the growing importance of conserving natural resources worldwide, the

resource potential of CDW is of particular interest. The large quantities of construction and
demolition waste and also the high consumption of raw materials in the construction industry,
point out the importance of the reuse of building materials for an improvement of resource
efficiency. The utilization of used or recycled building materials as secondary building
materials can meet the requirements of positive environmental impacts in several aspects: the
extended time availability of primary raw materials and the related protection of natural
resources as well as the saving of landfill site. Accordingly a reduction of construction and
demolition waste (CDW) can be environmental advantageous.
Nevertheless, the use of secondary building materials is only possible if the specific
technical functionality of the certain material is secured. Many studies investigated the
building material properties of recycled concrete aggregates (RCA) and also recycled brick
material [1, 2]. In general recycled aggregates have inferior building material properties like
higher porosity and lower density that are caused by adherent old cement paste in crushed
concrete. Furthermore the content of mortar and plaster affects the quality of brick debris.
Therefore quality requirements for various possible applications have been elaborated and

were implemented in technical rules and standards. The influences of different techniques for
demolition or selective dismantling and of processing methods for the improvement of the
building material properties of the treated CDW were investigated as well. Several studies
have shown that the purity of the recycled fractions is an important prerequisite for efficient
and environmentally friendly reuse of secondary building materials [3]. Beside impurities like
wood or gypsum also harmful substances like heavy metals or organic pollutants can impair
the quality of the recycled building material.
In recent years the content of gypsum in CDW received growing attention. Two reasons
can be mentioned: firstly in many countries the utilization of gypsum containing building
materials increased strongly in the last decades (see Figure 1). Because of the growing
gypsum stock in buildings an increase of gypsum in future CDW has to be expected.
Secondly gypsum is usually the main source of sulfates in CDW and under certain conditions
sulfates might damage constructions by the formation of voluminous mineral phases,
particularly Ettringite but also Thaumasite, resulting in cracking, expansion or a loss of bond
between cement paste and aggregate [4]. Furthermore, sulfates can also impair the setting
behavior of fresh concrete. Accordingly technical standards and guidelines restrict sulfates in
secondary building materials.
Gypsum Consumption in the US
Gypsum Consumption in Germany

10
40
2012
1998
2005
1984
1991
1970
1977
1956
1963
1942
1949
1928
1935
1914
1921
1998
2005
1984
1991
1970
1977
1956
1963
1942
1949
1928
1935
1914
1921
10
1900
1900
20
1907
[Mio t/a]
1907
[Mio t/a]
30
Figure 1: Gypsum consumption in Germany (blue) and for the United States (red) in Mt,
1900 2010/2012 [5-7]
Gypsum in construction waste usually originates from interior fittings like gypsum walls,
floor screeds, plasterboards and also plaster. Several gypsum building materials like
plasterboards, dry walls or floating floor screeds can be deconstructed either by manual labour
or by using mechanical equipment e.g. to remove floor screeds by milling. Useful are
techniques for selective dismantling. Nevertheless, bonded gypsum screeds and plaster are
usually quite difficult to remove during the deconstruction process.
Depending on the specific deconstruction site it can be more effective and/or more
environmentally compatible to remove sulfates by treating the crushed concrete in a stationary
CDW processing plant. Applicable treatments for sulfate reduction include manual labour like
manual sorting of gypsum wall blocks as well as mechanical treatments like intensive
crushing. As a result, a quality improvement of recycled concrete aggregates is achievable.
On the other hand, the implementation of these dismantling and/or processing steps usually
causes additional environmental impacts e.g. by energy consumption. Therefore benefits of

reusing CDW as secondary building materials and the correlated environmental costs for
further dismantling or processing steps have to be calculated and evaluated carefully.
This paper presents results of an investigation of the environmental impacts of
conventional demolition or selective dismantling of different building types, each with a
subsequent CDW treatment, aiming a generation of RCA with a low content of sulfates [8].
The results were compared with data for the production of natural aggregates.
2.
METHODOLOGY
Environmental Evaluation
2.1
The basis for the environmental evaluation was an estimation of the process flows for the
model scenarios. The data for the process flows were taken from manufacturers data for
building machines or tools and also from the life cycle inventory (LCI) database ecoinvent
data 2.2 [9]. Using the impact assessment model CML 2002-method the categories climate
change, land consumption, ozone depletion potential, eutrophication potential and
acidification potential were evaluated. Impact categories and the related units are shown in
Table 1. The calculation of material flows and energy consumption was carried out with
Umberto, a software for material flows and lifecycle analysis [10].
Table 1: Impact categories, description and related units
Impact category
name of the impact

category in CML
unit
Climate change
Global warming potential,

GWP 100a
kg CO2 and
equivalents
Land use
Land use competition
guide parameter ma
Acidification potential
average Europe
kg SO2equivalents
Eutrophication potential
average Europe
kg NOxequivalents
Stratospheric ozone
depletion potential
Stratospheric ozone
depletion potential
kg CFC11equivalents
CO2-
All environmental impacts were calculated in relation to the production of a ton of RCA as
the functional unit. Data for the environmental comparison to the extraction of natural
aggregates were also taken from the database, ecoinvent data 2.2 [9].
2.2
System Boundaries
Since the investigation focused on the production of RCA, the end-of-life-phase of
buildings, C stage according to European Standards (CEN/TC 350), was investigated [11, 12].
This evaluation included the working steps
demolition/selective dismantling (C1)
related transports (C2)
CDW processing (mobile or stationary) (C3) and
final disposal/landfilling of residues (C4).
The comparison of the environmental impacts of the six scenarios was performed in these
system boundaries.

Since the production of RCA is not the objective of the demolition of a building, an
environmental comparison between natural and recycled aggregates must exclude the working
steps demolition resp. dismantling and related transports as well as related landfilling. The
resulting environmental impacts could be seen as business-as usual costs. These
environmental costs should only be taken into account if their sum exceeds the environmental
costs of comparable conventional demolition. Therefore the environmental comparison
between RCA and natural aggregates was carried out using the environmental impacts of
CDW processing and the related material transports. If the environmental costs for selective
dismantling were higher than the costs for demolition, the difference had to be included in the
evaluation. In the investigated scenarios this was not the case. The compared environmental
impacts of natural aggregates were calculated the same way.
2.3
Assumptions
Transport distances were estimated according to literature and data from the Federal
Statistical Office [6, 13, 14].
Investigated Building Types
2.4
Three different types of model buildings with practice orientated proportions of gypsum
containing interior fittings were designed for the environmental evaluation:
one-family dwelling
multiple dwelling unit and
industrial building.
For each type of building (already without expandable components like windows or
technical facilities and if necessary after removal of hazardous substances) two different
model scenarios were designed: conventional demolition and selective dismantling with
regard to gypsum containing building materials. Thus, six different model scenarios for the
demolition of the buildings and the subsequent CDW treatment were evaluated. The
composition of the main materials of the three building types (in tonnes) is shown in Table 2.
All building types contained bonded floor screeds and gypsum plasterboards. Selective
dismantling steps and the subsequent CDW treatment (stationary for CDW from one-family
dwelling and multiple dwelling unit, mobile for CDW from the industrial building) were
evaluated related to the production of RCA.
Table 2: Material composition [t] of the three building types
one-familiy dwelling
Floated Floor-screed
25.20
multiple dwelling
unit
77.04
industrial building
164.06
Concrete walls and ceilings
100.34
1,790.86
10,481.25
Concrete base plate
33.45
330.74
1,050.00
Gypsum plaster boards
0.45
4.50
5,000.00
198.87
0.00
0.00
Brick walls

3.
The results of the environmental assessment provided a unified picture: the environmental
impacts in the categories climate change and land consumption dominated all model
scenarios. In contrast the calculation of the stratospheric ozone depletion potential resulted in
very low values therefore these results are not listed in the following.
While the environmental impacts of the deconstruction of multiple dwelling unit and
industrial building had a similar level, the environmental costs of the demolition resp.
selective dismantling of the one-family dwelling were much higher. This is due to the fact,
that this building type included also brick walls. Since all environmental impacts were
referred to the functional unit one ton RCA, the content of bricks resulted in a lower
proportion of concrete and accordingly higher environmental costs per ton RCA.
Furthermore, compared to conventional demolition the approach of selective dismantling
turned out to be environmental advantageous for all investigated building types. Table 3
shows the results of the investigated impact categories in all six scenarios.
Table 3: Results of the environmental evaluation for the selective dismantling (SD)
respectively the conventional demolition (CD) of different building types referring to the
functional unit 1 t RCA
Global warming potential

[kg CO2-eq]
Land consumption
[ma]
Acidification potential
[kg SO2-eq]
Eutrophication potential
[kg NOx-eq]
One-family
dwelling
Multiple dwelling
Industrial building
unit
SD
46.17
18.43
17.31
CD
58.81
22.05
23.45
SD
1.55
0.55
0.33
CD
2.70
0.87
0.87
SD
0.20
0.08
0.11
CD
0.26
0.10
0.14
SD
0.34
0.14
0.20
CD
0.43
0.17
0.25
Regarding the different working steps, selective dismantling procedures lead to higher
environmental impacts in all categories. If the CDW processing is included, the expenses
result finally in lower environmental costs. These benefits can be attributed to good sorting
accuracy and, thereby, less land consumption for depositing residues and also less energy
consumption by reducing transports and energy required for the machines used to deposit
residues. As expected transportation was an important factor in all scenarios. Figure 2 shows
the results for the environmental impacts climate change and land consumption for itemized
working steps during the production of one ton RCA made of CDW from the deconstruction
of the multiple dwelling unit.

25
1,25
20
1,00
15
0,75
10
0,50
0,25
0,00
0
dismantling
without transports
total
climate change [kg CO2-eq/t RCA]

selective
conventional
dismantling
without transports
total
land consumption [ma/t RCA]

selective
conventional
Figure 2: Results of the environmental impact categories climate change and land use,
multiple dwelling unit
The environmental evaluation showed that the actual use of selective dismantling
techniques is usually connected with a higher consumption of energy and therefore higher
environmental impacts than the use of conventional methods for the demolition of a building.
Regarding the complete procedural processes (i.e. including CDW processing, transports and
landfilling) the picture is inverted: the use of selective dismantling techniques for the
separation of gypsum containing building elements caused less environmental impacts. In
contrast, the efforts for processing CDW and also for depositing residues of not selectively
gained CDW were, as a rule, higher than for material that origins from selective dismantling
procedures.
One of the most important aspects in the recycling of materials is the environmental
compatibility of their production. Decisive criterion: the utilization of a secondary product
should be environmental advantageous compared to the use of the primary product. In that
context a comparative environmental evaluation is a common procedure for the verification of
the sustainability of the aimed substitution.
In this investigation significant differences between recycled aggregates and natural
aggregate could only be determined in the impact category land consumption. Nevertheless,
the substitution of natural aggregates by RCA could help saving natural resources through
reducing land consumption. Figure 3 shows the results of the comparison of the production of
one ton natural gravel and one ton RCA from a one-family dwelling and from a multipledwelling-unit.

Comparison natural aggegates / recycled concrete aggregates
6
2
4
1
2
0
natural aggregates
RCA (OFD)
RCA (MDU)
climate change [kg CO2-eq./t ]

selective
conventional
natural aggregates
RCA (OFD)
RCA (MDU)
land consumption [ma/t ]

selective
conventional
Figure 3: Results of the environmental comparison in the impact categories climate change
and land use between natural aggregates and RCA produced from a one-family dwelling
(OFD) and from a multiple-dwelling unit (MDU)
However, although it could be verified that environmental costs can be reduced by
replacing natural aggregates by RCA, due to the multitude of influencing parameters such as
transport distances, composition of the building and accordingly the CDW, deconstruction
projects need to be individually planned and implemented to result in environmental benefits.
4.
CONCLUSIONS
The results obtained in this investigation lead to the following conclusions:
The environmental evaluation of the six model scenarios to produce RCA from different
building types showed that steps for the reduction of sulfates usually have an
environmental impact which must be considered. Climate change and also land
consumption dominated the environmental impacts of all scenarios.
Compared to conventional demolition techniques the selective dismantling was
environmental advantageous for all investigated building types. Although the selective
dismantling of a building usually needs higher efforts these environmental costs were
more than compensated by less disposal expenses.
Compared to the production of natural aggregates, the RCA showed a clear
environmental advantage in the impact category land consumption but similar results in
the other investigated categories. Therefore an environmental benefit could be reached
by substituting natural aggregates by RCA in concrete production within the
framework of existing regulations and standards.
ACKNOWLEDGEMENTS
We thank the Federal Ministry for the Environment, Nature Conservation and Nuclear
Safety and the Federal Environment Agency, Germany, for funding the research for reducing
gypsum in C&D waste in the project no. Ufoplan 3709 33 317

REFERENCES
1.
Katz, A., Properties of concrete made with recycled aggregate from partially hydrated old
concrete. Cement and Concrete Research, 2003. 33: p. 703-711.
2.
Mueller, A. Lightweight aggregates from masonry rubble. in Use of recycled materials in

building and structures. 2004. Barcelona: RILEM Publications S.a.r.l., Bagneux, France.
3.
Poon, C.S. and D. Chan, The use of recycled aggregate in concrete in Hong Kong. Resources
Conservation & Recycling, 2007. 50(3): p. 293-305.
4.
Rahman, M.M. and M.T. Bassuoni, Thaumasite sulfate attack on concrete: Mechanisms,
influential factors and mitigation. Construction and Building Materials, 2014. 73: p. 652-662.
5.
Bundesverband Baustoffe - Steine und Erden e.V., Aufgaben, Themen und Ziele 2010/2011 Jahresbericht der Baustoffindustrie, 2010, bbs: Berlin. p. 21.
6.
Statistisches Bundesamt (Federal Statistical Office), Statistisches Jahrbuch 2010, 2010:

Wiesbaden, Germany. p. 745.
7.
US Geological Survey (USGS). Gypsum Statistics. 2014

http://minerals.usgs.gov/ds/2005/140/ds140-gypsu.pdf.
8.
Weimann, K., et al., Optimization of demolition/dismantling of buildings for the recovery and
treatment of building materials considering the reduction of harmful substances (in particular
sulphates) in the recycled building material and aspects of life-cycle analyses, Federal
Environment Agency, Editor 2012, Bundesanstalt fr Materialforschung und -prfung and
Bauhaus-Universitt Weimar, Germany: Dessau, Germany. p. 225.
9.
Ecoinvent Centre, Database ecoinvent data v2.2 - life cycle inventory (LCI) data, 2011, St.
Gallen, Switzerland.
10.
Ifu, Umberto - Software for Life Cycle Assessment, IFU Hamburg, http://www.ifu.com/en/.
11.
CEN, Sustainability of construction works, in CEN/TC 350, Technical Committee 350 of the
European Committee for Standardization, Editor 2011.
12.
Silvestre, J.D., J. de Brito, and M.D. Pinheiro, Environmental impacts and benefits of the endof-life of building materials - calculation rules, results and contribution to a "cradle to
cradle" life cycle. Journal of Cleaner Production, 2014. 66: p. 37-45.
13.
Gallenkemper, B., et al., kologischer Nutzen des Recyclings und der Kreislaufwirtschaft im
Bauwesen. Mll und Abfall, 2004(6): p. 260-266.
14.
Jeske, U., et al., Beton mit rezyklierter Gesteinskrnung - eine kobilanz, 2004,
Forschungszentrum Karslruhe: Karlsruhe. p. 6.
[cited 2015]; Available from:

FATIGUE FAILURE OF RECYCLED AGGREGATE CONCRETE

UNDER COMPRESSIVE INCREASING AMPLITUDE TEST
C. Thomas (1); J.A. Polanco (1); J. Setin (1); C. Medina (2)
(1) LADICIM, University of Cantabria, ETSI Caminos, Canales y Puertos, Avda. Los Castros
S/N, 39005 Santander, Spain. PH (0034) 942 200 915; FAX (0034) 942 201 818; email:
thomasc@unican.es
(2) Escuela Politcnica de Cceres, Universidad de Extremadura, Spain.
Abstract
In this research, different mixtures under fatigue loadings using the Locati steps or the increasing
amplitude test method have been studied. Control concretes with natural limestone aggregate and recycled
aggregate concretes with substitutions of 20%, 50% and 100% of coarse natural aggregate by recycled
aggregates coming from concretes have been analyzed. Results show that the Locati methodology can be
used to obtain a good approximation of the limit interval where the fatigue limit. Furthermore, the use of
recycled aggregate implies a reduction in the number of cycles able to resist. These differences are
increased for low water/cement ratios, as in this case the main factor is the lower strength capacity of the
aggregate.
Keywords: fatigue, recycled concrete, LOCATI, failure, waste, recycled aggregate

1.
INTRODUCTION
In recent years, a great deal of research work has been carried out to find alternative ways
to reduce construction and demolition wastes, 31% of all waste produced in the European
Union [1], by the use of recycled aggregate (RA) in the production of concrete. Various
authors have studied the use of the recycled aggregates from concrete (RAC). Padmini et al.
[2] discusses the properties of recycled coarse aggregates derived from of different concretes,
also the fine aggregate has been studied [3] showing suitable characteristics for use of them in
concrete. However, the incorporation of RA supposes a significant loss of fluidity of the
mixture [4] caused by the attached mortar content of the RA [5]. Thomas et al. [6] suggest
that the defect of water can be compensated by adding a percentage of water to the mixture
which depends on the absorption capacity and the amount of the RCA. Also, sanitary ware
industry waste as coarse aggregate have been studied showing a good freeze-thaw behavior
[7] but a reduction on the gas permeability is observed [8].
It is generally agreed that there is a loss in the physical and mechanical properties but the
fatigue behavior has not been extensively analyzed. The review of the literature suggests that
the RAC requires more water for the same workability than conventional concrete; the
density, compressive strength and modulus of elasticity of RAC are relatively lower than
those of the control concretes; and for a given water/cement ratio (w/c), gas and water
permeability, rate of carbonation and risk of reinforcement corrosion are higher [2]. Also,
authors have confirmed the presence of three stages of behavior with a lifespan of 10%, 80%
and 10% of the fatigue life [9,10]. However, the lack of a well-established test procedure for
executing and evaluating fatigue tests makes it difficult to correlate, compare or extend

published test results [11]. In the present paper the Locati methodology for determining the
fatigue limit of the plain concrete is proposed and compared with other available
methodologies.
2.
MATERIALS
The control and recycled concretes, with replacements of 20%, 50% and 100% of coarse
NA by RCA, have been studied under fatigue loadings using the Locati method. The mix
designs are shown in Table 1.
Table 1 : Mix proportions.
Designation
Replacement
RCA
Eff. w/c
H-0.65AS
0%
20%
50%
100%
275
275
275
275
179
179
179
179
948
961
978
1010
490
362
199
0
513
378
209
0
0
185
408
640
0.67
0.68
0.67
0.70
C is the Cement (kg); W is the Water (kg); S is the Sand (kg); G is the Gravel with a size of 12/20 (kg); g is the
gravel with a size of 6/12 (kg); RCA is the recycled concrete aggregate (kg); Eff. w/c is the effective water/cement
ratio.
Cylindrical specimens, according to [12] were prepared and according to [13] cured for
365 days in a controlled atmosphere of 205 C and 97 2% humidity environment. The
reason for this age of curing corresponds to the long duration of fatigue tests. In the test time
interval, the properties of concrete can significantly vary in young concretes [6]. The
specimens were instrumented with two strain gauges glued in two opposite generatrix.
3.
LOCATI TEST
The Locati method, or stepwise increasing amplitude testing, is an accelerated method used
for the reliable determination of the material fatigue limit. The method proposes a test on a
single specimen as an alternative to the classical Whler diagrams. Whlers method requires
the use of several tests and specimens to obtain the fatigue limit of a component. Locatis
proposal consists in a fatigue test with constant amplitude for a relatively small number of
cycles after which the upper limit of loading, M, is increased by an amount, keeping the
other parameters unchanged: lower limit load, m, number of cycles, N, and the frequency, .
This process (step) is repeated until the failure of the specimen under testing. An alternative
configuration of the Locati method could be to keep constant the amplitude of the step and
increase the mean load by a constant amount.
The Locati fatigue tests used here were performed in load control. A sinusoidal wave of 10
Hz frequency was applied. The lower limit, common for all steps in this case, is set at m = 0.
The upper limit varies for the different steps j (j = 1,2,3 ...) as M = j. Thus, the upper stress
level is defined by the value of the increment and the value of the previous stress level:
j j 1
(j = 1,2,3 ...)
(1)
Thus, with the obtained average compressive strength, values of 0 = 5 MPa, 1 = 10 MPa
and = 5 MPa are adequate for the characterization of all concretes. If the specimen

undergoes the established number by step of 1105 cycles without failure, the second step
(with the same specimen) is performed maintaining the lower limit and increasing the upper
limit by the constant value = 5 MPa. The process described is repeated until the specimen
failure. A lower value of should provide a higher accuracy in the fatigue limit
determination, although more steps and time are required.
4.
Figure 1 shows the results of the Locati tests: maximum average cycles, strain and strength
of the tested specimens of a specimen of 20% substitution. The rest of the mixtures have
shown a similar behavior. Nevertheless, some differences between them can be observed: the
total number of supported cycles in the failure step, the maximum strength and the maximum
and minimum strain on each cycle. Each step has been represented by Si where i=1,2,3 f
and Sf is then the failure step. All Si<f exceed the 1105 cycles established for each step. In
these steps, two different types of behavior have been identified. During the steps Si<f the
strain of the material presents a similar behavior with different slopes for the different steps
and proportional to the stress level. However, the step Sf, coincident with the step in which the
fatigue failure take place, presents an additional exponential type of strain behavior.
Moreover, the ultimate step has a number of cycles lower than the others because this is the
step in which the accumulated damage due to successive cycles causes the final collapse of
the material.
The step Sf, in which the failure occurs due to the microcracks generated in the preceding
steps and especially due to the damage suffered in this step where the maximal rate of
increase of maximum and minimum strain and gap between them are found. Also, a new
exponential behavior is found, in which the rate of increase of the maximum and minimum
strain and the gap between them is greater than in all the previous rates.
S2
S1
S3
S4
S5
S6
Sf
2000
Strain [um/m]
1500
1000
500
0
1 10
2 10
3 10
4 10
5 10
6 10
7 10
Cycle [N]
Figure 1 : Maximum (top) and minimum (bottom) strain of a H-065-20% specimen during the
Locati test.
The damage on the material caused by the accumulation of cycles can be analyzed by the
study of the evolution of the gap between the maximum and minimum strain versus the
cycles. The maximum strain shows a higher variation with the number of cycles than the
minimum strain. Therefore the analysis of the fatigue behavior of concrete will be more

reliable using the maximum strain. Using the minimum strain, the highest value of strain of
each step is coincident with the smallest value of the strain of the next step because the
minimum stress is kept constant during the test. At first sight, the linear behavior of the
maximum and minimum strain in each step increasing with the number of cycles might be
considered as an indication of fatigue damage.
Therefore, from these results it can be deduced that, assuming that the damage caused by
the pre-failure Locati steps is not relevant, the concretes with 0%, 20%, 50% and 100% of
incorporation of recycled aggregate are not able to resist more than 1105 cycles with loading
intervals of over 75%, 70%, 65% and 60% of its compressive strength. The fatigue limit,
or stress under which the concrete could resist infinite cycles, must then be in a lower step
than the Locati failure step Sf.
The results show that the collapse of the specimen tested under fatigue takes place in steps Sf-1
and Sf. Thus, the fatigue limit must be situated in the Sf-2 Locati method step, the last step in
which the specimens are able to surpass the 2106 cycles.
In order to find in which step the fatigue limit is situated, results were compared with
monotonous fatigue tests have been carried out in other research [6,14,15]. These specimens
were tested until failure or up to a maximum of 2106 cycles maintaining the stress limits in
the ranges of the Sf-1 for the first test and the Sf-2 step for the second test. In this way, it has
been considered that the concrete has infinite life for a given stress interval when the fatigue
life exceeds 2106 cycles.
The results obtained also show that there is a critical strain rate value that could be used as
an indicator of the critical strength for fatigue damage.
5.
CONCLUSIONS
The use of this recycled concrete is a solution in order to avoid two situations: removing
waste and reducing the need for significant amounts of natural aggregate, preserving in both
ways the natural resources. It is well known that recycled concrete presents a good behavior
to static loading, but the study of its response to dynamic loading is quite novel.
It has been proven that the LOCATI method is suitable for obtaining a good approximation
to the fatigue limit. It is a much faster method than traditional methods and is valid for
establishing comparisons between different types of recycled concrete. One LOCATI test
supposes the application of different steps with increasing strength until the collapse where
the three types of behavior to the fatigue testing can be observed in a relatively short time.
The results obtained showed that the structural recycled concrete exhibits an adequate
behavior to fatigue loading. However, the observed losses in the fatigue life, due to the
recycled aggregate, are slightly higher than those which are observed in the compressive
strength. For the same water/cement ratio, the recycled aggregate concrete presents a higher
loss of stiffness at the same strength conditions. The loss of stiffness rate is inversely
proportional to the fatigue life.
ACKNOWLEDGMENT
The authors wish to thank to the Spanish Ministry of Environment for the funding of the
Project.

REFERENCES
[1] F Rodrigues, MT Carvalho, L Evangelista, J de Brito. Physicalchemical and mineralogical
characterization of fine aggregates from construction and demolition waste recycling plants,
J.Clean.Prod. 52 (2013) 438-445.
[2] AK Padmini, K Ramamurthy, MS Mathews. Influence of parent concrete on the properties of
recycled aggregate concrete, Constr.Build.Mater. 23 (2009) 829-836.
[3] CJ Zega, A Di Maio. Use of recycled fine aggregate in concretes with durable requirements,
Waste Manage. 31 (2011) 2336-2340.
[4] H Mefteh, O Kebali, H Oucief, L Berredjem, N Arabi. Influence of moisture conditioning of
recycled aggregates on the properties of fresh and hardened concrete, J.Clean.Prod. 54 (2013) 282288.
[5] MS de Juan, PA Gutirrez. Study on the influence of attached mortar content on the properties of
recycled concrete aggregate, Constr.Build.Mater. 23 (2009) 872-877.
[6] C Thomas, J Setin, JA Polanco, P Alaejos, M Snchez de Juan. Durability of recycled aggregate
concrete, Constr.Build.Mater. 40 (2013) 1054-1065.
[7] C Medina, MI Snchez De Rojas, M Fras. Freeze-thaw durability of recycled concrete containing
ceramic aggregate, J.Clean.Prod. 40 (2013) 151-160.
[8] C Medina, M Fras, MI Snchez de Rojas, C Thomas, JA Polanco. Gas permeability in concrete
containing recycled ceramic sanitary ware aggregate, Construction and Building Materials. 37
(2012) 597-605.
[9] C Thomas, A Cimentada, JA Polanco, J Setin, D Mndez, J Rico. Influence of recycled
aggregates containing sulphur on properties of recycled aggregate mortar and concrete, Composites
Part B: Engineering. 45 (2013) 474-485.
[10] XW Luo, HL Yao. Experimental Research on Influence of Loading Method on Dynamic
Behavior of Recycled Concrete, Advanced Materials Research. Advances in Mechanical Design
(2011) 1171-1174.
[11] MK Lee, BIG Barr. An overview of the fatigue behaviour of plain and fibre reinforced concrete,
Cement and Concrete Composites. 26 (2004) 299-305.
[12] EN 12390-1. Testing hardened concrete. Part 1: Shape, dimensions and other requirements for
specimens and moulds. (2001).
[13] EN 12390-2. Testing hardened concrete - Part 2: Making and curing specimens for strength tests,
(2009).
[14] C Thomas, J Setin, JA Polanco, I Lombillo, A Cimentada. Fatigue limit of recycled aggregate
concrete, Constr.Build.Mater. 52 (2014) 146-154.
[15] C Thomas, I Sosa, J Setin, JA Polanco, AI Cimentada. Evaluation of the fatigue behaviour of
recycled aggregate concrete, J.Clean.Prod. 65, 397-405 (2013).

EFFECT OF SUPERPLASTICIZER ON MORTARS WITH RECYCLED

FINE AGGREGATES
Gabriela A. Valencia (1), Winnie F. Santos (1), Srgio C. ngulo (1), Rafael G. Pileggi(1)
(1) Department of Civil Engineering, University of Sao Paulo, Brazil
Abstract
Many researchers have proposed alternatives to improve mortars with construction and
demolition waste (CDW) recycled aggregate as it demands more water due to high porosity,
which requires a further increase of cement. The proposal seeks to study the effect of
superplasticizer on fresh mortars replacing natural aggregate by CDW recycled aggregates
trying to reach same consistency and desirable eco-efficiency. Size and shape distributions of
aggregates were similar. For the same mortars designs, only replacing the aggregates, mixing
and stepped flow tests were carried out to investigate rheological behaviour assessed by a
planetary rheometer. An optimum superplasticizer content of 0.175% was determined.
Increasing 5 times superplasticizers optimum content, natural aggregate mortars showed 20%
higher mixing energy and similar rheological behaviour at stepped flow test. Otherwise,
recycled aggregate mortars showed 7% reduction of mixing energy and, at stepped flow tests,
reduction of the torque at low rotation speed. Adding 5 times more SP in recycled aggregates
mortars the water reduction was about 23%. Increasing only superplasticizer 250 times
without the addition of water did not allow achieving the set consistency.
Keywords: rheology, recycled aggregate, superplasticizer, mixing stage, stepped flow test.
1.
INTRODUCTION
Several studies with recycled aggregate mortars and concretes have been developed by
replacing natural aggregates [1][4] only taking into account the main differences such as
porosity and water absorption. For a recycled aggregate mixture with a comparable quality to
those traditional, it must concern physical- mechanical performance [5].
Construction and Demolition Wastes (CDW) recycled concrete aggregates grains have a
layer of porous cement paste on its surface, which increases substantially specific surface area
and porosity [6][9] influencing mortars rheological properties as workability. To control
rheological behaviour of recycled aggregate mortars there is a variety of alternatives: increase
the quantity of water according to the demand; pre- saturation the aggregate for 30 minutes or
24 hours before using; increase cement content [10]; mix the cement paste before adding the
aggregate [11]. Among these options, the most common practice is the addition of water,
which requires a further increase of cement content to keep the mechanical strength and,
therefore, is neither environmental nor economical desirable.
Another possibility is increasing superplasticizer when replacing aggregates [12] to reduce
water demand. It is know that superplasticizer adsorbs on cement surface due to electrostatic
and van der Waals forces, promoting dispersing effect[13], [14]. In recycled aggregate the

admixture probably will adsorb on the remaining cement paste surrounding aggregate
particles due to its higher porosity and specific surface area, demanding more superplasticizer
[15].
The scope of this study is to understand the effect of the superplasticizer on fresh mortars
with recycled aggregates, in comparison with natural aggregates mortars. The effect was
assessed investigating the rheological behaviour of mortars with different superplasticizer
contents.
2.
Materials used in this study were: a) Portland cement CPV-ARI (~90% of clinker), b)
standard natural sand (NA) sieved in four fractions (100/150m, 150/300m, 300/600m, and
600/1200m), c) recycled concrete aggregates (RA) from a commercial recycling plant
(URBEM - Sao Bernardo do Campo), and d) a sodium polycarboxylate (Melflux F2651)
superplasticizer (SP).
Particle size distribution and density of cement, NA and RA were determined. Morphology
and porosity (by mercury intrusion) were determined for NA and RA.
NA and RA particles <106m had been eliminated, so the cement was the only material
with particle size lower than 106 m. A mixture of various natural (quartz) aggregate
fractions (NAc) was made to achieve a grain size distribution similar to that of RA.
Mortar (~3 kg-dry mass) composition was 25% (kg/kg) of cement and 75% (kg/kg) of
aggregates. The tests were carried in mass since both aggregates have different envelope
density. Consistency, measured by the torque value, was set as mortar control parameter. The
target value for all reference mixtures was close to 1 Nm. The water content determined for
reference mixes was always used also for the corresponding RA mortars. Additional tests on
same RA mortars were carried out increasing the water content to reach the target torque
value; time was limited to 600s.
Rheological behaviour was assessed submitting the mortars to a stepped flow test.
The effect of the SP was assessed increasing its content (% g/g of cement) by 5 and 10
times. The optimum SP content (0.175% by weight of cement), determined in preliminary
continuous flow tests on cement paste, was set as starting value.
Each mortar was named in accordance with the following acronyms: Z-Xw-Ysp, where Z
is the type of aggregate (NAc=natural or RA=recycled), XW is the water content and Ysp is
the SP content.
2.1
Material characterization
Particle size and shape distributions: particle size distribution of cement was determined by
laser diffraction granulometer (Malvern Mastersizer LongBed - detection range from 0.05 to
555m). Grain size distribution and morphology of NA fractions, RA and NAc was carried
out by dynamic image analysis technique (Retsch- Camsizer - detection range from 0.001 to
30 mm), according to the methods described in ISO standard 13322-2/2006. The shape
parameters analysed were the aspect ratio (b/l) and sphericity (S) [17][19]. The aspect ratio
(Equation 1) is the ratio of the minimum (x_min) to the maximum (x_max) Feret diameter.It
gives an indication for the elongation of the particle (1 represents a sphere). The sphericity
(Equation 2) is the relation between the perimeter of a circle having the same projection area
of the particle (A) with the real perimeter of the particle (P). It ranges between 0 and 1 (values
close to 0 represent a rough angular particle, whereas close to 1 a smooth spherical surface).
(b/l)=(x_min/x_max)
(1)

S = (4A)/P2
(2)
True Density: density was determined for cement, NAc and RA. Particles volume was
determined by a helium pycnometer (Quantachrome MVP 5DC). Determined value includes
solid material volume, closed pores but not open pores.
Porosity and total pore area: measured for NAc and RA by a mercury intrusion porosimetry
(Micromeritics Autopore III 9420) according to [20], [21].
2.2
Rheological characterization
Rheological behaviour of mortars was assessed by a planetary rheometer developed at PoliUSP [22]. Aggregates and cement were manually homogenized and then introduced into the
rheometer container. The mixing was directly made into the rheometer at a constant engine
rotation speed (500 rpm for 180s). Water was added at t =10 s (flow rate of 44 cm3/s). After
the mixing, the mortar was submitted to the stepped flow test (from 50 to 1250 rpm and
back). The duration of each step was 5 seconds and the values of revolution per minute were
50, 100, 150, 250, 500, 750, 850, 1000, and 1250.
3.
3.1
Material characterization
Figure 1 shows the grain size distributions of NA fractions, RA and cement (a), whereas
the composed natural fraction (NAc) and the RA are presented in (b). The proportion of each
fraction was: 19.5% (fraction 100/150m), 27% (150/300m), 30.5% (300/600m), and 23%
(600/1200 m).
8
NA 600/1200
30
Discrete Distribution (%)
Discrete Distribution (%)
35
NA 300/600
25
NA 150/300
20
NA 100/150
RA
15
Cement
10
NA composed
RA
5
4
3
2
1
0
0.1
10
100
Diameter (mm)
(a)
1000
50
500
Diameter (mm)
5000
(b)
Figure 1 Grain size distributions of materials (a) and of aggregates mixture (b)
Figure 2 presents the aspect ratio (a) and the sphericity (b) of aggregates. NAc exhibits
aspect ratio of 0.7, while RA is 0.75. The determined sphericity was 0.8 for NAc and 0.85 for
RA. RA is slightly less angular than NA, contrary to found in other studies [1], [23], [24].
Roughly, both aggregates presented similar grain size and shape distributions.

30
NAc- Discrete
25
RA - Discrete
NAc- Accumulated
20
35
100
90
80
70
60
50
40
30
20
10
0
RA- Accumulated
15
10
5
30
100
90
80
70
60
50
40
30
20
10
0
NAc- Discrete
25
RA - Discrete
20
NAc- Accumulated
15
RA- Accumulated
10
5
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
0
0.05
0.1
0.15
0.2
0.25
0.3
0.35
0.4
0.45
0.5
0.55
0.6
0.65
0.7
0.75
0.8
0.85
0.9
0.95
Volume percentage (%)
Volume percentage (%)
35
Aspect Ratio
Sphericity
(a)
(b)
Figure 2 - Aspect ratio (a) and sphericity (b) of NAc and RA aggregates
Table 1 presents physical properties of cement, RA and NAc. Considerable differences
were observed between the aggregates. Due to the adhered old cement mortar/paste, RA
shows lower true density, porosity 30 times higher and total pore area 20 times higher.
Table 1 Physical properties of cement, RA and composed NAc
Physical Properties
True Density (g/cm3)
Porosity (% cm/cm3)
Total pore area (m2/g)
Cement
3.01
-
RA
2.58
23.40
12.10
NAc
2.65
0.76
0.60
3.2
Mortar mixing
Figure 3 shows the mixing rheograms of NAc and RA mortars without (a) and with (b)
superplasticizer. The mixing energy value is presented in Figure 3-c.
6
6
15% water and 0% SP
4
RA mortar
12.5% water
Torque (N.m)
Torque (N.m)
0.175% SP
4
1.75% SP
3
2
RA mortars
0.875% SP
1
NAc mortar
0
0
60
120
180
Time (s)
0.175% SP
60
120
Time (s)
Torque required: 1 N.m
(a)
NAc mortars
(b)
180
Torque required: 1 N.m

Mixing Energy (N.m.s)
15% of water
700
600
RA mortar
12% of water
680
500
400
NAc mortar
522
505
488
300
200
100
148
208
174
(c)
Figure 3 NAc and RA mortars mix curves (a - b) and mix energy (c). Nac with SP 1.75%
was not assessed
7
6
5
4
3
2
1
0
RA-31.1w-0sp
RA-26.1w-0.175sp
RA-24w-0.875sp
1 N.m
40
Torque (N.m)
Torque (N.m)
It is observed that the required energy for mixing mortars with RA are at least the double
for the same amount of water and additive used with NAc. This occurs due to the total pore
area of RA, 20 times higher, and its porosity 30 times higher than NAc. Thus, there was no
water wetting the surfaces and disperse the particles, resulting in a dry and not cohesive
mortar.
Increasing the SP content from 0 to 0.175% by weight of cement, NAc mortar showed
similar final torque values however the resulting mixing energy was around 18% higher. The
final torque was also similar for RA mixes but the mixing energy decreased of about 23%.
Increasing the SP content from 0.175 to 0.875%, NAc mortar showed 20% higher mixing
energy and higher final torque. It is possible that 0.175% is very close to the optimum content
of additive for NAc mortar to ensure the best consistency with the minimum content of water.
For RA, results showed 7% lower mixing energy and lower final torque. However, the RA
mortar with 1.75% of SP looked still dry and not cohesive, being impossible to apply.
The same RA mortars (SP content of 0, 0.175 and 0.875%) were mixed increasing the
water content until achieving the target torque (Figure 4-a).
RA-12.5w-1.75 to 43.5sp
Addition of SP
30
20
10
1 Nm
0
180
300
420
Time (s)
(a)
540
180
420
660
Time (s)
900
(b)
Figure 4 RA Mortars mix curves (a) with water addition to reach torque 1 N.m and (b) with
SP addition until 43.5%

The variation of water content was about 2.07, 2.09 and 1.92 for RA mortar with 0%,
0.175% and 0.875%, respectively, comparing to corresponding NA moratrs.
It was observed that the higher amount of additive demanded less water. So, the effect of
RAs in mortar was the absorption of water and the possible adsorption of SP due to the old
adhered mortar/cement paste [12].
Figure 4-b shows results obtained increasing the amount of SP (fixed amount of water) for
the mortar (RA-12.5w-1.75sp). The target value was not reached even with a SP content 250
times higher. The increase of torque was probably associated to the sticking of the mortar on
the rheometers stirrer due to excess of SP.
1,4
1.4
1,2
1.2
Torque (N.m)
Torque (N.m)
3.3
Mortar stepped flow test
Figure 5 presents the rheograms of the stepped flow test.
1,0
0,8
0,6
NAc-15w-0sp
NAc-12.5w-0.175sp
NAc-12.5w-0.875sp
0,4
0,2
1.0
0.8
0.6
RA-31.1w-0sp
RA-26.1w-0.175sp
RA-24w-0.875sp
0.4
0.2
0,0
0.0
0
250
500
750
1000
Rotation speed (rpm)
(a)
1250
250
500
750
1000
Rotation speed (rpm)
1250
(b)
Figure 5 Stepped flow tests of NAc (a) and RA mortars (b) with 0,0.175% and 0.875% by
weight of cement
Despite all mortars reached the target torque (1 Nm), NAc and RA mortars presented
different curves. The increase of SP content affected the torque values at high rotations for
NAc mortars, and at low rotations for RA. This different behaviour might be due to water
content, time during mixing stage and type of aggregate.
NAc mortars present remarkable positive hysteresis area (area between curves). During
acceleration, particles de-agglomerate faster than re-agglomerate in the deceleration stage.
They probably were not homogenized as RA mortars, which mixture time was at least twice
(NAc: 180s; RA without SP: 600s and RA with SP: 400s).
Without SP, NAc mortar presents lower torque values in high rates. For the same water
content, higher SP contents do not alter the rheological behaviour. It is possible that the
increase of SP content above the value of 0.175% does not favour further dispersion.
For RA mortars, the increase of SP content did not alter the torque values at high
revolutions per minute, whereas it decreased that parameter at low rpm, even with lower
water content.
4.
CONCLUSIONS
RA in this study, compared to NAc, showed 30 times higher porosity and total pore area 20
times larger, though particle size and shape distributions are similar.
About mixing stage test, adding same percentage of SP, 0.175% of cement mass determined
in cement paste, in NAc mortar, water content reduces 17%. Nevertheless, adding 5 times

more SP, there was no water reduction. Replacing NAc by RA influences the water
absorption and the probable SP adsorption on the old adhered cement paste of RA. This is still
valid when a higher amount of SP led to lower water content. Adding 0.175% of SP in RA
mortar, water content reduces about16%. Adding 5 times more SP, the water reduction was
about 23%.
Results showed that, for a water content of 12.5% (kg/kg by dry mass of solids), the increase
of SP (250 times higher than 0.175%) did not allow achieving the set consistency (target
torque value). Thus, the effect of water is predominant on the system.
Stepped flow tests showed that increasing SP in RA mortars contributed to reduce the torque
at low rpm but no changes were observed at high rpm. Therefore, multiple point tests are
necessary for a better understanding of the rheological behaviour.
ACKNOWLEDGEMENTS
Gabriela Araujo Valencias work is supported by CNPq Conselho Nacional de
Desenvolvimento Cientfico e Tecnolgico (grant 133866/2014-9). Winnie Franco Santoss
work is supported by FAPESP - Fundao de Amparo Pesquisa do Estado de So Paulo
(grant 2013/14307-0). The information and views set out in this study are those of the authors
and do not necessarily reflect the opinion of CNPq nor FAPESP. Authors would like to thank
the research financing support of Conselho Nacional de Desenvolvimento Cientfico e
Tecnolgico - CNPq - Brasil, Process 485340/2013-5
REFERENCES
T. C. Hansen, Recycled aggregates and recycled aggregate concrete second state-of-the-art
report developments 19451985, Mater. Struct., vol. 19, no. 3, pp. 201246, May 1986.
[2] M. B. Leite, Avaliao de propriedades mecnicas de concretos produzidos com agregados
reciclados de resduos de construo e demolio, 2001.
[3] S. C. ngulo, Caracterizao de agregados de resduos de construo e demolio reciclados e
a influncia de suas caractersticas no comportamento de concretos, Tese de Doutorado,
Universidade de So Paulo, 2005.
[4] F. Agrela, M. Snchez de Juan, J. Ayuso, V. L. Geraldes, and J. R. Jimnez, Limiting properties
in the characterisation of mixed recycled aggregates for use in the manufacture of concrete,
Constr. Build. Mater., vol. 25, no. 10, pp. 39503955, Oct. 2011.
[5] F. Faleschini, C. Jimnez, M. Barra, D. Aponte, E. Vzquez, and C. Pellegrino, Rheology of
fresh concretes with recycled aggregates, Constr. Build. Mater., vol. 73, pp. 407416, Dec.
2014.
[6] S. Nagataki, A. Gokce, T. Saeki, and M. Hisada, Assessment of recycling process induced
damage sensitivity of recycled concrete aggregates, Cem. Concr. Res., vol. 34, no. 6, pp. 965
971, Jun. 2004.
[7] F. Pacheco-Torgal, V. W. Y. Tam, J. A. Labrincha, and J. de Brito, Handbook of recycled
concrete and demolition waste, 1st ed. Cambridge, UK: Woodhead Publishing Limited,, 2013.
[8] C. Ulsen, H. Kahn, G. Hawlitschek, E. A. Masini, S. C. Angulo, and V. M. John, Production of
recycled sand from construction and demolition waste, Constr. Build. Mater., vol. 40, pp.
11681173, Mar. 2013.
[9] J. P. B. Vieira, J. R. Correia, and J. de Brito, Post-fire residual mechanical properties of
concrete made with recycled concrete coarse aggregates, Cem. Concr. Res., vol. 41, no. 5, pp.
533541, May 2011.
[10] C. S. Poon, Z. H. Shui, L. Lam, H. Fok, and S. C. Kou, Influence of moisture states of natural
and recycled aggregates on the slump and compressive strength of concrete, Cem. Concr. Res.,
vol. 34, no. 1, pp. 3136, 2004.
[1]

[11] V. W. Y. Tam, X. F. Gao, and C. M. Tam, Microstructural analysis of recycled aggregate

concrete produced from two-stage mixing approach, Cem. Concr. Res., vol. 35, no. 6, pp. 1195
1203, Jun. 2005.
[12] P. Pereira, L. Evangelista, and J. De Brito, The effect of superplasticisers on the workability
and compressive strength of concrete made with fine recycled concrete aggregates, Constr.
Build. Mater., vol. 28, no. 1, pp. 722729, 2012.
[13] H. Uchikawa, S. Hanehara, and D. Sawaki, The role of steric repulsive force in the dispersion
of cement particles in fresh paste prepared with organic admixture, Cem. Concr. Res., vol. 27,
no. 1, pp. 3750, 1997.
[14] A. Kauppi, K. M. Andersson, and L. Bergstrm, Probing the effect of superplasticizer
adsorption on the surface forces using the colloidal probe AFM technique, Cem. Concr. Res.,
vol. 35, no. 1, pp. 133140, Jan. 2005.
[15] G. Hawlitschek, Caracterizao das propriedades de agregados midos reciclados e a influncia
no comportamento reolgico de argamassas., text, Universidade de So Paulo, 2014.
[16] A. Griesser, Cement-Superplasticizer Interactions at Ambient Temperatures - Rheology, Phase
Composition, Pore Water and Heat of Hydration of Cementitious Systems, Doctoral and
Habilition Theses, SWISS FEDERAL INSTITUTE OF TECHNOLOGY, 2002.
[17] T. Allen, Powder Sampling and Particle Size Determination. Elsevier, 2003.
[18] S. J. Blott and K. Pye, Particle shape: a review and new methods of characterization and
classification, Sedimentology, vol. 55, no. 1, pp. 3163, Feb. 2008.
[19] ISO 9276-6, Representation of results of particle size analysis - Part 6: Descriptive and
quantitative representation of particle shape and morphology, 2008. .
[20] ISO 15901-1:2005 - Pore size distribution and porosity of solid materials by mercury
porosimetry and gas adsorption -- Part 1: Mercury porosimetry. .
[21] P. Webb, An Introduction To The Physical Characterization of Materials by Mercury Intrusion
Porosimetry with Emphasis On Reduction And Presentation of Experimental Data.
Micromeritics Instrument Corporation, 2001.
[22] F. A. Cardoso, Mtodo de formulao de argamassas de revestimento baseado em distribuio
granulomtrica e comportamento reolgico., Tese de doutorado, Escola Politcnica da USP,
So Paulo, 2009.
[23] J. de Brito and S. Nabajyoti, Recycled Aggregate in Concrete Use of Industrial, Construction
and Demolition Waste. London: Springer, 2013.
[24] C. Lima, A. Caggiano, C. Faella, E. Martinelli, M. Pepe, and R. Realfonzo, Physical properties
and mechanical behaviour of concrete made with recycled aggregates and fly ash, Constr.
Build. Mater., vol. 47, pp. 547559, Oct. 2013.

REHYDRATION OF CEMENT FINES: A TG /CALORIMETRY STUDY

Srgio C. Angulo (1,2); Mrio S. Guilge (2); Valdecir A. Quarcioni (2); Raphael
Baldusco (2); Maria A. Cincotto (1)
(1) Department of Construction Engineering, University of Sao Paulo (USP), Brazil
(2) Institute for Technological Research (IPT), Sao Paulo state, Brazil
Abstract
Research on new sources of cementitious materials is required to reduce global antropogenic CO2
emissions of cement industry. Cementitious powders from construction and demolition waste can be
dehydrated and rehydrated and present some residual strength according to some initial studies.
Systematic studies including characterization techniques (thermogravimetry and calorimetry) are
lacking for a better understanding of this material. This paper aims to investigate rehydration of
cement powders using these techniques, in order to establish relations between rehydrates water
content, heat of rehydration and achieved compressive strength. The compressive strengths of treated
cement fines were related to heat of rehydration and rehydrates water content.
Keywords: dehydration, rehydration, cementitious fines, thermogravimetry; calorimetry
1.
INTRODUCTION
Cement production is the industrial sector responsible for ~5% of antropogenic CO2
emissions in the world [1]. Since the cement production tends to double in the next 20years,
this sector would represent more than 30% of global antropogenic CO2 emissions in 2050.
Research on new sources of cementitious materials and additions are required for future.
Portland cement is constituted of 95-97% (in mass) of clinker and 3-5% (in mass) of
gypsum (2 - CaSO4.2H2O) [2]. Clinker contains mostly alite C3S (3CaO.SiO2) and
belite - C2S (2CaO.SiO2), but it also includes around 25% of calcium-aluminate phases, such
as C3A (3CaO.Al2O3) and C4AF (4CaO.Al2O3. Fe2O3). In contact with water, such crystalline
phases form the compounds described below.
23 + 62 3 2 3 + 3
(1)
22 + 42 3 2 3 +
(2)
33 + 32 + 262 6 3 32
(3)
33 + 6 3 32 + 42 34 12
(4)
In short, hydration reactions can be described as follows. Tricalcium-aluminate, in
presence of gypsum, is the first phase to be hydrated near the cementsurface grains forming
ettringite (C6 AS3 H32 ) and calcium monosulfoaluminate just after the contact cement-water.
After the induction period of the hydration, silicate hydrate (C-S-H) and portlandite (CH) are
formed followed by the complete formation of ettringite as result of the total consumption of
sulfate. Between 20 and 30 hours, most of ettringite is converted to calcium

monosulfoaluminate (34 12 ). In-situ X-ray difraction and thermogravimetry (TG) over

time are powerful techniques for understanding early cement hydration over time. It is now
well understood and validated by thermodynamic model [3].
Figure 1 Thermodynamic modelling of Portland cement hydration [3].
As reported by Lothenbach et al. [3], after 28 days of hydration, the paste volume is 45%
of C-S-H, 25% of CH, and 20% of calcium sulfolauminates phases, remaining near 5% of
anhydrous calcium silicate phases (C3S and C2S). Alite is more reactive than belite, that may
persist unhydrated over month or years. Even a long term hydrated cement may contain
certain amount of belite phase, still active in cementitious waste. When cementitious waste is
milled as powder (< 100m) it will present certain residual strength, as reported by [4][6].
Recently, some authors have been studied dehydration and rehydration of cementitious
powders in order to recover this residual strength [7][9]. The environmental advantage of
this type of recycled cement is the absence of decarbonation of the raw materials in the
process, avoiding CO2 emissions during its production.
Different techniques have been applied aiming to analyse the obtained product. Shui et al.
[8] studied the dehydration of cementitious powders and pure cement until 800C. They
identified by X-ray diffractometry (XRD) that ettringite disapears after thermal treatment at
200C, followed by an increase of intermediate silicate after 500C. The characteristics peaks
of this intermediate silicate are close to that of C2S. Compressive strengths from 4.7 to 8.3
MPa were achieved with the rehydratin of cementitious powders. Using nuclear magnetic
ressonance (NMR), Alonso and Fernandes [7] also observed that, after 200C, this
intermediate silicate starts to be formed and increases near 500C, being completed formed at
750C.
Guilge [6] also confirmed the rehydration of cementitious powders using complementary
characterisation techniques (XRD, thermogravimetry TG, and isothermal calorimetry).
Initial results of this dissertation are presented here. In this paper the investigation of recycled
cement powders rehydration was made by complementary techniques thermogravimetry and
calorimetry. A comprehensive study using thermogravimetry (TG) is necessary to check the
recovering ability of C-S-H and other hydrates (C-A-H, ettringite, etc) in different
dehydration temperatures. Other relevant aspect is the use of isothermal calorimetry to follow
rehydration reactions through heat released. Relationships among rehydrates content, heat of
rehydration and achieved compressive strength were investigated here.

2.
An Ordinary Portland Cement (95% of clinker) was hydrated, dehydrated at 300 C, 500
C and 650 C and rehydrated. Thermogravimetry (TG) and Isothermal Calorimetry (IC) were
used to evaluate dehydration and rehydration of these materials, as well as by compressive
strength essays at 28 days.
2.1
Materials
The cement (C) of the study contained 95% of clinker in mass (CPV-ARI Brazilian
standard type). The chemical composition of the cement type CPV-ARI is shown in Table 1.
Table 1 Chemical composition of the cement.
Oxides
Content (%)
SiO2
18.9
CaO
62.2
Al2O3
5.4
Fe2O3
2.8
SO3
4.2
MgO
1.9
K2O
1.0
Na2O
0.21
LOI
3.8
2.2
Methods
Hydrated cement (HC) specimens (cylinders of 5 x 10 cm) were prepared with
water/cement ratio of 0.48 (kg/kg) and cured during 28 days (at room temperature with RH
100%). After the curing period, the specimens were dried at 40C for 24 h, crushed and
milled as powders (< 150 m sieve aperture).
HC powders were then submitted to dehydration in a controlled heating condition up
to each selected temperature: 300, 500, 650 C. The furnace was heated at a rate of 10
C/min, starting from room temperature. When the desired temperature was reached, the
powders were kept in this condition during 2 h. The whole heating scheme [6], adapted from
Shuis procedure [8], is shown in Fig. 1. Dehydrated cement (DC) powders were then
submitted a rapid cooling with a fan to room temperature and covered with a plastic lm and
kept into a desiccator until characterization tests. They were named DC300, DC500 and DC650,
respectively.
Cooling abrupt to
room temperature
Figure 2 - Heating scheme of HC [6], [8].
HC and DC powders (DC300, DC500 and DC650) were then rehydrated with
water/cement ratio of 0.48, moulded in specimens (cylinders of 5 x 10 cm) and cured during
28 days. After 28 days, the rehydrated cement (RC) specimens were dried at 40C for 24 h,
crushed and milled again as powders (< 150 m sieve aperture). The powders were named
RC, RC300, RC500 and RC650, respectively.

Thermogravimetric (TG/DTG) analyses were conducted with HC, DC300, DC500,

DC650, RC, RC300, RC500 and RC650 samples, using a TA Instruments equipment, model SDT
2960. The essays were carried out under the following conditions: 50 mg sample were heated
in an alumina crucible 110l without cover, nitrogen gas flow of 50 ml/min ultrapure; heating
flow of 10C/min, and temperature interval from 10C to 1,000C. Table 2 shows the thermal
events of hydrated cement considered for the DTGs interpretation [2].
Table 1 Thermal events of hydrated cement for DTGs interpretation [2].
Thermal events
Dehydroxilation of C-S-H
Dehydroxilation of ettringite (AFt)
Dehydroxilation of calcium sulfoaluminates (AFm)
Dehydroxilation of hydrogarnet - C3AH6
Dehydroxilation of hydrotalcite- [M0,75A0,25(OH)2]C0,125H0,5
Dehydroxilation of portlandite Ca(OH)2
Decarbonation of calcite - CaCO3
Temperature range
(C)
0 400
70 140
185 200
250 550
330-430
430 550
550 - 900
To monitor rehydration process, isothermal calorimetric (IC) analyses were carried out
with the C, HC, DC300, DC500, DC650 samples, at 25C, using equipment Thermometric TAM
AIR with 8 channel, with 16s frequency. The mass used was 10 g each sample and data were
collected during 72 hours.
Compressive Strength (CS) tests of C, RC, RC300, RC500 and RC650 were made using
Brazilian standard procedure (NBR 7215: 1997): 1:3 proportion in mass (cement:quartz
reference sand); water/cement ratio of 0.48; four layers of material manually compacted; and
compressive strength tests with four specimens (5 x 10 cm) at 28 days.
3.
RESULTS
3.1 TG/DTG analyses

DTG results of HC, DC300, DC500, DC650 are presented in Fig. 3. The first thermal
event (temperature range from 0 to 330C) is related to the presence of ettringite (discrete
peak near 95C), C-S-H (peak near 120C) and calcium monosulfoaluminate (discrete peak
near 180C). All these events are clearly noticeable in HC. For DC300, DC500, DC650 samples,
the presence of ettringite and calcium monosulfoaluminate is not observed anymore, only a
progressively reduction of the amount of remained combined water of C-S-H.
The second thermal event (temperature range from 330 to 430C) is related to brucites
prsence (Mg(OH)2). This event is only observed for HC and DC300. Due to higher dehydration
temperature, this event is not more observed for DC500 and DC650 samples.
The third thermal event (temperature range from 430 to 550C) is related to
portlandites dehydration (Ca(OH)2). The event is clearly observed to HC and DC300, but
partially observed to DC500 and DC650 samples, since the temperature of thermal treatment
partially decomposed this compound.
The fourth event (temperature > 550C) is related to calcites presence (CaCO3). This
event is observed for HC, DC300, and DC500. A slightly reduction of this compound occurred
for DC650 sample.

Figure 3- Dehydration of HC samples.

Table 3 shows the mass losses of HC, DC300, DC500, DC650 samples for each of these
events. A gradual reduction of combined water of C-S-H is noticed when temperature of
dehydration is increased. A reduction of portlandite is also observed for DC500 and DC650
samples; probably forming higher amounts of dehydrated CaO compounds, easily rehydrated
in presence of humidity at room condition (Fig. 3). Dehydration removes 55%, 91% and 94%
(in mass) of the combined water of ettringite, C-S-H, C-A-H and calcium
monosulfoaluminates, respectivelly.
Table 3 Mass losses of HC and DC samples in the TG/DTG temperatures ranges.
Samples Thermal event
HC
DC300
DC500
DC650
Dehydroxilation of Aft, C-S-H, C-A-H, Afm

Dehydroxilation of hydrotalcite
Temperature
range (C)
25-330
330-430
430-550
550-1,000
25-330
330-430
430-550
550-1,000
25-330
330-430
430-550
550-1,000
25-330
330-430
430-550
550-1,000
Mass loss
(%)
15.30
1.95
4.31
4.71
6.87
2.69
6.57
5.38
1.35
0.00
3.88
6.25
0.95
0.00
2.54
4.46

Figure 4 compares DTGs of the rehydrated samples (RC, RC300, RC500 and RC650)
with the hydrated cement (HC). HC and RC are quite similar. The combined water of C-S-H
is partially recovered for all dehydration temperatures, as well as the water of brucite, and
portlandite. Difference may be attributed to non-recovery of ettringite. RC650 also recovered
less amount of water if compared with the other temperatures of dehydration (Table 4).
Figure 4- Rehydration of DC samples.

Table 4 Mass losses of RC samples in the TG/DTG temperatures ranges.
Samples Thermal event
RC
RC300
RC500
RC650

Temperature
range (C)
25-330
330-430
430-550
550-1,000
25-330
330-430
430-550
550-1,000
25-330
330-430
430-550
550-1,000
25-330
330-430
430-550
550-1,000
Mass loss
(%)
14.6
2.35
4.26
5.09
15.20
2.22
4.46
4.82
14.4
1.73
4.95
5.04
12.50
1.65
4.45
4.64
Rehydration recovers 54%, 85% and 75% (in mass) of the combined water of
ettringite, C-S-H, C-A-H and calcium monosulfoaluminates, respectively. Thermal treatment
at 500C achieved the best condition in terms of cement rehydration.

3.2 Isothermal Calorimetry (IC)

Figure 4 presents the isothermal calorimetry results. Hydration of conventional cement (C)
generated roughly 360 J/g. Rehydration of HC sample generated 49 J/g, probably due to
hydration of some residual anhydrous calcium silicates. Rehydration of DC samples (DC300,
DC500, DC650) generated less amount of heat: 120, 235 and 177 J/g, respectively. The most
reactive sample was obtained by the thermal treatment at 500C. High heat liberation in the
first minutes may be attributed to the presence of CaO in the samples. The rehydration
phenomena occurred in the first 30 hours, faster than the hydration phenomena. This finding
corroborates with that found by Shui et al. [8], [9].
400
300
250
Heat (J/g)
Heat flux (w/kg)
C
HC
DC 300
DC 500
DC 650
350
C
HC
DC 300
DC 500
DC 650
200
150
100
1
50
0
0
10
20
30
40
50
60
70
Time (hours)
0
0
10
20
30
40
50
Time (hours)
60
70
80
(b)
(a)
Figure 4- Heat flux and cumulative heat released during rehydration of DC samples.
3.3 Compressive strength (CS)
The average compressive strength of RC, RC300, RC500 , RC650 and HC were 1.50, 4.40,
7.60, 4.20 and 42.0 MPa, respectively (Table 5). Shui et al. achieved similar strength for a
dehydrated cement at 400-500C.
Table 5 Compressive strengths of RC samples.
Samples
RC
RC 300
RC 500
RC 650
Compressive Strength (MPa).

spec 1
1.4
4.1
7.4
4.0
spec 2
1.4
4.6
7.8
4.1
spec 3
1.6
4.3
7.2
4.6
spec 4
1.4
4.5
7.9
4.2
Average
1.5
4.4
7.6
4.2
Std. Dev.
0.1
0.2
0.3
0.3
3.4 Correlation between the parameters

Figure 5a shows a correlation between rehydrates content (% in mass) and heat of
rehydration (J/g). The higher rehydrates content is the higher the heat of (re)hydration is.
Similar conclusion can be obtained between heat of rehydration and compressive strength for
a fixed value of water/cement ratio (0.48) (Figure 5b).

50
400
rehydrated samples
Heat of rehydration (J/g)
cement hydrated
300
R = 0,95
200
100
rehydrated samples
40
cement hydrated
R = 0,95
30
20
10
(a)
(b)
10
15
20
rehydrates water content (%)
25
100
200
300
400
Heat of rehydration (J/g)
Figure 5-Rehydrates water content x Heat of rehydration (a) and

Heat of rehydration x compressive strength (b).
3.
CONCLUSIONS
Dehydration and rehydration of Portland cement fines was quantitatively determined by

thermogravimetry and calorimetry. The compressive strengths observed were related to heat
of rehydration and rehydrates water content. Restricted to this study and type of cement
applied, the most appropriate temperature for dehydration of cement fines in terms of
reactivity and strength was 500C.
ACKNOWLEDGEMENTS
This research was developed and funded by IPT. Authors also thank Intercement/BNDES
for technical partnership and financial support on the on-going research project (Development
of technology for cement production from CDW) motivated by this initial research.
REFERENCES
[1] WBCSD e IEA, Cement Technology Roadmap: Carbon emissions reductions up to 2050, World
Business Council for Sustainable Development; International Energy Agency, Report, 2009.
[2] H. F. Taylor, Cement Chemistry, 2o ed. London: Thomas Telford.
[3] B. Lothenbach, G. Le Saout, E. Gallucci, e K. Scrivener, Influence of limestone on the hydration
of Portland cements, Cem. Concr. Res., vol. 38, no 6, p. 848860, jun. 2008.
[4] M. Arm, Self-cementing properties of crushed demolished concrete in unbound layers: results
from triaxial tests and field tests, Waste Manag., vol. 21, no 3, p. 235239, jun. 2001.
[5] C.-S. Poon, X. C. Qiao, e D. Chan, The cause and influence of self-cementing properties of fine
recycled concrete aggregates on the properties of unbound sub-base, Waste Manag., vol. 26, no
10, p. 11661172, jan. 2006.
[6] M. S. Guilge, Desenvolvimento de ligante hidrulico a partir de resduos de cimento hidratado e
de tijolo cermico, Dissertao de Mestrado, Instituto de Pesquisa Tecnolgica, So Paulo, 2011.
[7] C. Alonso e L. Fernandez, Dehydration and rehydration processes of cement paste exposed to
high temperature environments, J. Mater. Sci., vol. 39, no 9, p. 30153024, maio 2004.
[8] Z. Shui, D. Xuan, H. Wan, e B. Cao, Rehydration reactivity of recycled mortar from concrete
waste experienced to thermal treatment, Constr. Build. Mater., vol. 22, no 8, p. 17231729, ago.
2008.
[9] Z. Shui, D. Xuan, W. Chen, R. Yu, e R. Zhang, Cementitious characteristics of hydrated cement
paste subjected to various dehydration temperatures, Constr. Build. Mater., vol. 23, no 1, p. 531
537, jan. 2009.

PRELIMINARY CHARACTERIZATION OF EXCAVATED SOILS

FROM A MINERAL CDW SO PAULO CITY LANDFILL
Karen Kataguiri (1), Sergio Cirelli Angulo (1), Maria Eugenia Boscov (1)
(1) Escola Politcnica, Universidade de So Paulo, Brazil
Abstract
A case study was conducted to characterize materials received in a mineral construction
and demolition waste (CDW) landfill located on Metropolitan Region of So Paulo. The study
consisted in: 35 sample collection of materials received in the landfill, 8 samples random
selection, sample mass reduction, preliminary visual classification (soil or other CDW types)
and some basic geotechnical characterizations. 38% of mass the samples can be classified as
excavated soil using particle size distribution and visual classification criteria. 62% of mass
samples were classified as mixed materials of excavated soils and other CDW types
(especially cement based materials). For mixed samples with almost 10% of granulometric
classified clay fraction, a separation between excavated soils and other CDW types was
possible if a sieving in 2 mm sieve aperture was adopted. In this case, a recovery 48% of
excavated soils is possible. However, most of the material received is still a mixture of
excavated soils and other CDW mineral types. Therefore, the need of a proper segregation
between these CDW types is necessary at construction building sites, prior to its final
disposal, to ensure optimal recovery.
Keywords: construction and demolition waste, soil reuse, soil characterization
1.
INTRODUCTION
Construction industry is responsible for generating more than 50% of the urban solid waste
mass in Brazil [1, 2]. Rocks, soils, ceramics, concrete, mortars, and other mineral origin
materials account for about 90% of the mass of construction and demolition waste (CDW) in
this country [3].
According to one study conducted by the Construction Industry Union of So Paulo State
(SINDUSCON-SP), excavated soil generation during construction of multi-storey buildings
has waste generation indexes ranging between 0.031 and 0.983 m/m of built floor area,
demonstrating that soil waste generation is about six times higher than other construction
waste generation [4]. Densely urbanizes regions in continuous growth, as the Metropolitan
Region of So Paulo (MRSP), require underground infrastructure construction for urban
mobility, which tend to generate large amounts of surplus soils. In European Countries, the
soil generation is about three to four times higher the other CDW generation [5].
Even though representing a very significant fraction, little research has been conducted on
the soil fraction present in the CDW. Excavated soils are generally disposed into inert mineral
waste landfills, along with other CDWs, and it is not segregated and valued even when CDWs
are sent to recycling plants. The grounding practice of the soil waste accelerates lifetime
depletion of the few inert mineral CDW landfills available in the MRSP. Most part of this

material could be reused on earthworks and trenchworks, vegetation replacement, paving,

among other geotechnical uses, or even as raw material for industrial products [6, 7, 8, 9]. To
recover the soil from mineral CDW landfills, a characterization study is necessary and it may
encourage its reuse as building material.
1.1
Objective
This work aims to characterize mineral CDW landfill samples collected at the Metropolitan
Region of So Paulo in terms of particle size distribution, visual composition (excavated soils
and other CDW types especially concrete, mortar) and fine fraction consistency
determination test. The on-going research aims to establish, based on variability of its
characteristics, the establishment of technical and environmental criteria for reuse of
excavated soils, mainly for geotechnical applications.
2.
METODOLOGY
A case study was conducted to characterize mineral CDW landfill samples obtained at the
Metropolitan Region of So Paulo (MRSP). The study consisted in two steps: a) sampling
consisted of collection of 35 samples during three months, b) characterization consisted of
random selection of 8 from 35 samples, particle size distribution analysis, visual composition
classification (soil almost pure material - or mixture soil-CDW) regarding different particle
size intervals, and fine fraction consistency determination test.
2.1
Sampling
C&D waste landfill was located in the south region of MRSP. It contains 360,000 m and
design to receive 2,000 tons per day of mineral C&D waste (concrete, mortar, ceramic,
soils).It is the biggest mineral C&D waste landfill in Latin America. The landfill receives
material from south region and part of west region of So Paulo, as well as other cities from
MRSP like Embu, Embu Guau, Itapecerica da Serra, Santo Andre, Sao Bernardo do Campo,
etc.
35 representative samples of the materials received in mineral C&D waste landfill were
collected; three collections per week (on Mondays, Wednesdays and Fridays) filling up a 40l
barrel per day; half in the morning and half in the afternoon during eight weeks. Samples were
collected at the moment of the material discharge from the truck, in scattered conical pile,
with tractor loader assistance, in order to perform the collection at different positions of the
scattered pile, preferably in the middle and on the sides of it so as to obtain a more
representative sample [10]. It was collected about 40kg sample material.
2.2
Characterization
Among the 35 collected samples, 8 samples were randomly selected for conducting the
tests. Samples collection and mass reduction procedures were judicious to ensure samples
representativeness for the tests, considering the heterogeneity of the materials received at the
landfill [10]. The initial material mass collected (about 40kg) was reduced to the necessary
masses for the tests using the Jones splitter type.
Particle size distribution and consistency limits determination (liquid limit, plastic limit
and plasticity index) were evaluated for the selected 8 samples according to Brazilian
standards ABNT NBR 6457: 1986, ABNT NBR 6459: 1984, ABNT NBR 6508: 1984, ABNT
NBR 7180: 1988 e ABNT NBR 7181: 1988 [11, 12, 13, 14, 15]. Visual observations of the

samples were conducted to analyze the presence of excavated soils and other CDW types
(cement based materials concrete and mortar and red ceramic).
To analyze if separation between excavated soils and other CDW types according to
particle size fractions, the mixed samples were sieved on the following particle size intervals:
2.0 to 1.2mm, 1.2 to 0.6mm, 0.6 to 0.4mm, 0.4 to 0.1 mm, and 0.1 to 0.075mm. A minimum
representative mass was calculated for each particle size interval [11] and the granulometric
fraction of each sample was reduced to the minimum representative mass using a rotary
sample splitter for fine particulates. A microscope equipped with a digital camera, and
objective zoom lenses up to 325x was used to obtain the images of the size fractions. In each
image, the grains were numbered and classified as cement-based material (an artificial
material present in high amount in C&D waste - Figure 1a) or soil (the typical clay and silt
material). The grains which did not present cement paste were considered as soil grains,
however, this procedure has limitations, because ceramic particles (Figure 1b) can be visually
similar of clay and silt grains (Figures 2a e 2b) mostly present in excavated soils.
b
a
Figure 1: (a) Cement and (b) ceramic based grains of typical C&D waste observed in digital
microscopy
Figure 2: (a) Clay and (b) silt grains observed in digital microscopy
3.
PRELIMINARY RESULTS AND DISCUSSION
Fig.3 presents the particle size distributions of mineral CDW landfill samples. B-7, B-15
and B-23 samples presented the lowest gravel fraction ranging between 7.2% and 23.0%, and
presented the highest fine fraction ranging between 31.4% and 48.6%. Samples B-4, B-5, B12, B-19 and B-22 were preliminary and visually classified as mixed CDW samples, and its
granulometry results presented the highest gravel fraction ranging between 34.2% and 59.9%,
and the lowest fine fraction (< 0.075 mm) ranging between 9.2% and 23.9%.

SAMPLES GRANULOMETRY
Sieves (mesh)
200 100
50 40 30
16
10
100
90
80
Cumulative passed material (% g/g)
70
60
50
B-4
B-5
40
B-7
B-12
30
B-15
B-19
20
B-22
B-23
10
0
0.000
0.001
0.010
0.100
1.000
10.000
100.000
Grains diameter (mm)

Clay
Silt
Fine Sand
Medium Coarse Sand
Gravel
Figure 3: Particle size distributions of mineral CDW landfills samples

Particle size distributions of B-7, B-15 and B-23 samples were finer than the others (B-4,
B-5, B-12, B-19, and B-22), being, in accordance with visual observation, mainly constituted
of excavated soils (Fig.4). The B-4, B-5, B-12, B-19 and B-22 samples were composed of a
mixture of soils and other CDW types (especially cement based materials, the most
predominant fraction in this kind of waste). According to these results, only 38% of mass
mineral CDW landfill samples can be considered as almost pure excavated soils; most of
them ae mixtures of excavated soils and other CDW types.
B-7
B-15
B-23
Figure 4: Preliminary visual classification of B-7, B-15, B-23 samples, mainly constituted of
excavated soils by visual observation

B-4
B-5
B-19
B-22
B-12
Figure 5: Preliminary visual classification of B-4, B-5, B-12, B-19, B-22 samples, constituted
of a mixture of excavated soils and other CDW (mainly cement based materials)
Table 1 presents consistency limits and plasticity indexes of the mineral CDW landfill
samples. 2 of 3 samples (B-15 and B-23; ~67% of mass) mainly composed of excavated soils
presented fine fraction (< 0.42 mm) with plasticity. 3 of 5 samples (B-4; B-12 and B-22;
~60% of mass) composed of mixture of excavated soils and other CDW types presented fine
fraction with plasticity. Two of these samples (B-4; B-12) also presented high amount of clay
fraction without a significant increase of silt fraction (fig.3). Samples were classified
according to the Unified Soil Classification System to describe its texture and grain size [16].
Table 1: Consistency limits and plasticity indexes of mineral CDW landfill fine fraction (<
0.42 mm) samples
Liquid
Plastic
Limit (%) Limit (%)
B-4
29
21
B-5
28
*
B-7
27
*
B-12
32
19
B-15
30
21
B-19
*
*
B-22
27
18
B-23
33
27
(*) fine fraction with no plasticity
Sample
Plasticity
Index (%)
8
*
*
13
9
*
9
6
Unified Soil Classification System

No applicable (mixed material)
Well-graded Sand (SW)
Clayey Sand (SC)
Silty Sand (SM)
Table 2 summarizes the cement and soil grains counting. The digital microscopy image
criteria to determine the presence of cement material could not be applied for grains smaller
than 0.4 mm due to the difficulty to differentiate the cement paste grains from some feldspar
minerals.
For B-4 and B-12 samples (mixed soil-other CDW types samples with fine fraction
plasticity), separation of excavated soils (silt, clay and sand) was possible for particle size

fraction lower than 2 mm (fig.5 and fig.6). Excavated soil separation according to particle
sizes was not possible for the other mixed samples, confirmed by Fig. 7, 8 and 9 (B-22
sample, with fine fraction plasticity).
Table 2: Composition classification (soil and cement based grains) of different particle size
fractions of the mixed samples by digital microscopy.
Sample
Preliminary Visual
Classification
B-4
Excavated soils with

other CDW types
B-5

other CDW types
B-12

other CDW types
B-19

other CDW types
B-22

other CDW types
Particle
Size (mm)
2.0 1.2
1.2 0.6
0.6 0.4
2.0 1.2
1.2 0.6
0.6 0.4
2.0 1.2
1.2 0.6
0.6 0.4
2.0 1.2
1.2 0.6
0.6 0.4
2.0 1.2
1.2 0.6
0.6 0.4
Cement Fraction
(%)
49
65
61
84
89
98
37
45
53
92
92
91
98
92
96
Soil Fraction
(%)
51
35
39
16
11
2
63
55
47
8
8
9
2
8
4
Figure 5: Digital microscopy image of B-4 sample for particle size ranging from (a) 2.0 to 1.2
mm (b) 1.2 to 0.6 mm, (c) 0.6 to 0.4mm.
Figure 6: Digital microscopy image of B-12 sample for particle size ranging from (a) 2.0 to
1.2 mm, (b) 1.2 to 0.6 mm, (c) 0.6 to 0.4mm.

Figure 7: Digital microscopy image of B-5 sample for particle size ranging from (a) 2.0 to 1.2
mm, (b) 1.2 to 0.6 mm, (c) 0.6 to 0.4mm.
1.2 mm, (b) 1.2 to 0.6 mm, (c) 0.6 to 0.4mm.
1.2 mm, (b) 1.2 to 0.6 mm, (c) 0.6 to 0.4mm.
4.
CONCLUSIONS
This study presents a characterization of the samples collected at a mineral CDW landfill
located on the Metropolitan Region of So Paulo (MRSP). 38% of mass samples can be
classified as excavated soils using particle size distribution and visual classification criteria.
62% of mass samples were classified as mixed materials of excavated soils and other CDW
types (especially cement based materials).
Most of the excavated soil samples (2/3) presented fine fraction with plasticity, but this
fine fraction plasticity was also observed for some samples constituted of a mixture of
excavated soils and other CDW types. Most of these mixed samples with fine fraction
plasticity presented near 20% of fines (0.075mm) and almost 10% of granulometric classified
clay fraction.
Digital microscopy was used to classify the composition of mixed samples in different
particle size intervals according to the presence or not of cement based materials. For the

samples with almost 10% of granulometric classified clay fraction, a separation between
excavated soils and other CDW types was possible if a sieving in 2 mm sieve aperture was
adopted.
Preliminary investigations in this study shows that adopting some particle size distributions
criteria, using visual criteria and doing fine fraction consistency determination tests in a
mineral CDW landfill, a recovery 48% of excavated soils is possible. However, most of the
material received is still a mixture of excavated soils and other CDW mineral types.
Therefore, the need of a proper segregation between these CDW types is necessary at
construction building sites, prior to its final disposal, to ensure optimal recovery.
ACKNOWLEDGEMENTS
We gratefully acknowledge Building Contractor Union of the State of So Paulo
(Sinduscon-SP) for the technical and financial support provided for research about
construction waste generation indicators. We also acknowledge IPT and Intercement for
providing mineral CDW landfill samples used in this characterization study.
REFERENCES
[1] Associao Brasileira de Limpeza Pblica e Resduos Especiais (Abrelpe), Panorama dos resduos slidos
no Brasil 2012, Abrelpe, 2012. Available at: <http://www.abrelpe.org.br/Panorama/panorama2012.pdf>.
Acess in: 18th may 2014.
[2] Secretaria do Meio Ambiente do Estado de So Paulo (SMA), Sindicato da Indstria da Construo do
Estado de So Paulo (Sinduscon-SP), Resduos da Construo Civil e o Estado de So Paulo, So Paulo:
SMA/Sinduscon-SP, 2012, 85 p.
[3] Angulo, S.C., Caracterizao de Agregados de Resduos de Construo e Demolio Reciclados e a
Influncia de suas Caractersticas no Comportamento Mecnico dos Concretos, Tese de Doutorado, Escola
Politcnica, Universidade de So Paulo, 2005, 149p.
[4] Sindicato da Indstria da Construo do Estado So Paulo (Sinduscon-SP), Gesto ambiental de resduos
da construo civil: a experincia do Sinduscon-SP, Pinto,T.P. et al. Induscon-SP, 2005.
[5] Monier, V., Hestin, M., Trarieux, M., Mimid, S., Domrse, L., Van Acoleyen, M., Hjerp, P. and Mudgal,
S., Study on the management of construction and demolition waste in the E.U, Final report for the
European Commission (DG Environment), 2011.
[6] Department for Environment Food and Rural Affairs (Defra), Construction code of practice for the
sustainable use of soils on construction sites, United Kingdom, 2009.
[7] Swiss Agency for the Environment, Forests and Landscape (Saefl), Guideline: reuse of excavated soils,
Swiss, 2002.
[8] Waste & Resources Action Programme (Wrap), Recycling of excavated spoil from utility works within
local authority areas, and promoting the use of recycled materials through good practice in procurement,
United Kingom, 2005.
[9] Waste & Resources Action Programme (Wrap), Recycled and stabilised materials in trench reinstatement,
United Kingom, 2007.
[10] Petersen, L., Minkkinen, P. and Esbensen, K.H., Representative sampling for reliable data analysis: theory
of sampling, Chemometrics and Intelligent Laboratory Systems. 77(1) (2005) 261-277.
[11] ABNT (Associao Brasileira de Normas Tcnicas), NBR 6457: Amostras de solo - Preparao para
ensaios de compactao e ensaios de caracterizao. Rio de Janeiro, 1986.
[12] _________, NBR 6459: Solo - Determinao do Limite de Liquidez Mtodo de Ensaio. Rio de Janeiro,
1984.
[13] _________, NBR 6508: Gros de solos que passam na peneira de 4,8 mm - Determinao da massa
especfica. Rio de Janeiro, 1984.
[14] _________, NBR 7180: Solo - Determinao do Limite de Plasticidade Mtodo de Ensaio. Rio de Janeiro,
1988.
[15] _________, NBR 7181: Solo - Anlise granulomtrica Mtodo de Ensaio. Rio de Janeiro, 1988.
[16] Standard, A.S.T.M., D2487: Standard practice for classification of soils for engineering purposes (Unified
Soil Classification System). West Conshohocken, 2011.

Anette Mueller
Recycled Aggregate Characterization Methods
Abstract
Recycled aggregates are mixtures of various materials that can occur either in bound
or liberated form. Thus, the first characterization step is to extract a representative
sample. Various properties must be determined depending on the intended recycling
paths.
If processed rubble is to be used as a course material in road construction or as
recycled concrete aggregate, characterization must be carried out in accordance with
certain German and European standards. Material composition, particle size
distribution and shape, resistance against abrasion and freeze-thaw cycles must be
determined, as well as the concentration of sulfates, chlorides, certain heavy metals
and PAHs present in the eluate. For recycled concrete aggregates, particle density
and water absorption must also be measured. This paper provides information about
the sample mass required to determine the composition of the material by manual
sorting. Results taken from test reports of recycling companies are given for the most
significant properties.
A more sophisticated characterization is necessary to get a better understanding of
the effects of recycled concrete aggregates on concrete production and performance.
Material composition, which varies between pure cement paste and pure aggregate,
reactivity and chemical and mineralogical compositions must be analyzed in detail.
Considerations of the composition are also the basis to develop new recycling paths.

1
Construction and Demolition
Quantities and Sectors of Reuse
Waste
in
Germany:
Classification,
German and European environmental policy objectives include the goal to convert
the linear economy that consumes resources and creates waste to a closed-loop
economy. For this purpose, legislation is required to define the paths and determine
the rules. At the same time, favorable starting conditions must be created for this new
sector of the economy, which will need to be developed in competition with the
conventional construction materials industry. It is equally important to develop the
scientific and technological bases to implement closed material cycles. It is crucial to
strengthen research and improve the knowledge base in this field.
Waste first needs to be classified to enable comparisons of the quantities of
generated waste or to devise options for the disposal of certain types of waste. In
Europe, the European Waste Catalogue (EWC) was introduced in 2000, which
makes it possible to systematically classify waste generated in the course of
production processes or at the end of product use. Construction waste encompasses
all types of waste generated in conjunction with constructing, converting, refurbishing
or demolishing buildings and structures. These are covered in Chapter 17 of the
catalogue. This chapter includes a total of 38 types of waste, which are grouped into
various categories according to their origin: excavated soil, rubble from road
demolition, rubble from demolished buildings and structures, waste from construction
sites, and gypsum-based construction waste.
Since 1996, the quantities of construction waste generated in Germany have been
published at regular intervals [1] [2]. As shown in Fig. 1, there was hardly any change
during the 16-year period under consideration. Earlier forecasts that predicted a
significant increase from 2005 have not materialized to date.
Fig. 1: Time series for the various types of construction waste according to [1], [2]
With the exception of soil and stones, the types of waste referred to above can be
reprocessed to create recycled building materials. Rubble from the demolition of
roads mainly consists of asphalt and is almost completely reused to produce asphalt
base and binder courses as well as, in some cases, surface courses (Fig. 2). Rubble
from the demolition of infrastructural facilities, industrial, residential and public
buildings is composed of concrete (Fig. 3) and masonry (Fig. 4) in a 50/50 ratio.

Concrete rubble is reused in road construction (Fig. 5). Recently, it has also been
increasingly used as a concrete aggregate for residential and administrative buildings
(Fig. 6). Masonry is largely excluded from being reused in road construction and
concrete production; it can only be used as a filler in civil engineering projects.
Fig. 2: Fragments of asphalt from road demolition
Fig. 3: Fragments of concrete from the demolition of a building

Fig. 4: Masonry rubble from the demolition of a building
Fig. 5: Reuse of recycled concrete rubble as a base course in road construction [3]

Fig. 6: Reuse of recycled concrete aggregate in waterproof concrete [4]

The use of recycled building materials for asphalt production, base courses in road
construction and concrete production is governed by technical rules and standards
see [5], [6], [7], [8], [9], [10], [11], [13] . Recycled building materials to be used in
these sectors must be subjected to quality control procedures at regular intervals.
The following sections of this paper describe the methods applied to determine
selected quality control parameters. Measured values taken from test records are
used to derive typical ranges and correlations between specific properties.
However, the parameters to be determined in the quality control process are not
sufficient to exhaustively describe recycled building materials. Additional parameters
such as their chemical and mineralogical composition, reactivity, behavior during heat
treatment etc. need to be determined to describe their properties and behavior more
thoroughly, which requires more sophisticated tests that are outlined in the
subsequent parts of this paper. The outcomes of these tests can also be used to
develop new recycling paths.
Quality Control: Methods and Results
Various parameters need to be determined as part of the quality control process of

recycled building materials for base courses or concrete aggregates. These
parameters are listed in Table 1, including the composition and environmentally
relevant characteristics of the material as well as its particle size distribution and
abrasion and freeze/thaw resistance values.

Table 1: Parameters of quality control of recycled aggregates for road construction

and concrete production
Composition
Granulometric parameters Physical parameters
Reuse as a base course material in road construction

Material composition
Finely
divided
organic
constituents
Leachable constituents
Particle size distribution

Content < 0.063 mm
Content of oversize
Particle shape
Portion
of
crushed
particles
Water content
Apparent density
Proctor density
Resistance to abrasion
Freeze/thaw resistance
Reuse as recycled concrete aggregate

Material composition
Particle size distribution
Lightweight
organic Content < 0.063 mm
constituents
Particle shape
Content of
water-soluble chloride
acid-soluble chloride
water-soluble sulfate
acid-soluble sulfate
Total sulfur content
Leachable constituents
Water absorption
Apparent density
Resistance to abrasion
Freeze/thaw resistance
Required sample mass

Manual sorting is used to determine the composition of the material. The sample
taken from the processed material is screened in accordance with the applicable
standard, i.e. at a passing size of 4 mm. Screening residues are visually inspected
and grouped into the individual types of building material. If and when required, the
material is wetted before to be able to identify color variations and structures more
accurately. The reliability of the results of this sorting analysis mainly depends on the
sample size and the process applied to taking the sample. This is particularly relevant
in the entire recycling industry because the materials to be processed are strongly
heterogeneous whereas their content of certain foreign substances is strictly limited
to prevent a deleterious effect on the properties of secondary products [14], [15], [16].
Examples include the ratio of ceramic materials, porcelain and stones, as well as offcolors in glass cullets or the ratio of impairing substances in substitute fuels (Fig. 7).
Only if a sufficient sample mass is obtained will the required quality control process
lead to reliable results. The required sample mass depends on the properties of the
material to be analyzed, including its maximum particle size and heterogeneity, as
well as on the content to be determined and the requirements for the reliability of the
outcome. Statistically speaking, the sample must at least have a size that makes it
possible to capture even relatively minor constituents with a certain probability.

Fig. 7: Sample mass for sorting analyses to determine the content of impairing
constituents in glass cullets (taken from [14]), substitute fuels [15] and recycled
aggregates
Various models were developed to calculate the required sample mass [17], [18],
[19]; these models include the influential parameters referred to above. Assuming a
normal distribution of the sample composition for the sake of simplicity, the number of
particles z that a sample must contain in order for them to be representative of the
total quantity of the material is determined using the following equation:
2
()
1
= (
)
()
factor for confidence level

relative deviation in [g/g]
content of the component in [g/g]
The number of particles can be used to derive the required sample size whilst
assuming spherical particles and a certain particle size distribution and a known bulk
density. Table 2 lists the specifications for the composition of the material required for
reuse in road construction or concrete production. This table also includes the
calculated number of particles to be sorted to prove that individual requirements are
met.

Table 2: Particle number and sample mass for sorting analyses to determine the
material composition of recycled materials (confidence level 95%, z(S) = 1.65,
relative deviation 10%, particle size distribution according to Gates, Gaudin,
Schuhmann % passing = (x/xmax)0.5 with xmax = 45 mm, density 2.3 g/cm)
Reuse as an unbound base course in road construction
Requirements for material composition acc. to [8]
Number
particles
4 mm
of
Sample
>
mass [kg]
Asphalt
30 m.-%
635
0.4
Clinker, clay brick, stoneware
30 m.-%
635
0.4
Calcium silica brick, mortar and similar

5 m.-%
materials
5173
3.2
Mineral lightweight and insulation

materials,
non-floating
aerated
1 m.-%
autoclaved concrete and pumice
concrete
26,953
16.7
Foreign materials such as wood,

0.2 m.-%
rubber, textiles, plastics
135,886
84.0
Additional requirements acc. to [20]

Glass
5 m.-%
5,173
3.2
Gypsum
0.5 m.-%
54,178
33.5
Ferrous and non-ferrous metals
2 m.-%
13,343
8.2
Volume of floating material
specify value
Reuse as recycled concrete aggregate

Requirements for material composition acc. to [12]
Number
particles
4 mm
Concrete, concrete
concrete masonry
rock
grains,
hydraulically bound
Ru
30
products, mortar,
units + unbound
natural
stone, 90 m.-%
rock grains Rc +
Bricks and roof tiles from clay, calcium

silica bricks, non-floating aerated 10 m.-%
autoclaved concrete Rb
of
Sample
>
mass [kg]
2,450
1.51
1 m.-%
26,953
16.7
Cohesive materials (i.e. clay, earth),

ferrous and non-ferrous metals, non 1 m.-%
floating wood, plastic, rubber, plaster +
glass X + Rg
26,953
16.7
Bituminous materials Ra
Volume of floating material
FL 2 cm/kg

A sample with a mass of 3.2 kg must be sorted to prove compliance with the limits for
calcium silica brick and mortar for base course materials. Overall, 5,173 particles
must be analyzed, of which 4,648 are in the 4-8 mm range. A particular challenge is
posed by the future requirement to determine the gypsum content of materials to be
reused in road construction. To prove the limit of 0.5 m.-% of gypsum, a sample with
a mass of 33.5 kg must be sorted, which is equivalent to 54,178 particles, of which
about 48,675 particles are in the 4-8 mm range. Thus, this approach is rather
impracticable. Furthermore, the sorting analysis only covers particle sizes in excess
of 4 mm. Particles smaller than 4 mm, which account for about one third of the total
sample, are not captured, which leads to the conclusion that the methods applied to
determine the material composition should be modified. The following approaches
appear to be feasible:
Sensors are used to detect and count or weigh the particles in order to replace
manual sorting.
Chemical parameters are introduced for certain groups of substances to
replace manually determined contents. For instance, the gypsum content
could be replaced with the ratio of water-soluble sulfate when assuming that
gypsum is the main sulfate source. Organic constituents could be captured as
TOC.
Material composition of recycled building materials

Recycled materials used for road construction primarily contain concrete and natural
mineral aggregates. Examining the sorting analyses of 25 test records showed strong
variations in the ratios of concrete and mineral aggregates (Fig. 8). The behavior of a
recycled material that contains over 85% of concrete rubble will be different from a
material with only 10% of concrete. Fig. 9 shows the mean ratios and variations in the
constituents, which are limited in the case of recycled materials for road construction.
The mean values lie within the limits listed in Table 2. The spreads reveal that
asphalt is the only material that comes very close to the threshold.

Fig. 8: Content of concrete and natural rock and gravel; data from the quality control
of 25 recycled aggregates
Fig. 9: Content of minor constituents, data from the quality control of 25 recycled
aggregates
Construction rubble may contain various sulfate carriers, including plaster boards,
gypsum renders and screeds as sources of water-soluble sulfate. Sulfate-containing
compounds such as ettringite and monosulfate, which are formed during cement
hydration and are present in concrete, are the sources of acid-soluble sulfate.
Sulfates may have an adverse effect in both structural and environmental terms,
which is why applicable test rules and procedures refer to the related requirements in
several instances:
In future, the gypsum content must be determined in order for the material to
be reused for road construction purposes. In addition, the concentration of
SO42- ions in the eluate must be measured. In Germany, the shaking method
is still used to produce the eluate; this process involves addition of water to the
sample at a 1:10 ratio and a 24-hour overhead turning of the sample.
The ratios of water- and acid-soluble sulfates and eluate parameters must be
determined if the recycled material is intended for use in concrete production.
Compared to sorting analyses, determining these ratios using chemical methods is
associated with the advantage of measuring analytical samples that contain particles
smaller than 100 m. Several consecutive disintegration and separation steps result
in a representative sample with a sufficient number of particles so that the sources of
error referred to above are irrelevant. Measurements of samples extracted from
laboratory concrete showed that the mean ratio of acid-soluble sulfate amounted to
0.54 m.-%. Recycled materials that consist of concrete rubble exhibit almost the
same mean sulfate content, but with greater variations. Any value greater than the
base value of 0.54 m.-% is indicative of the presence of gypsum. Values below this

threshold are possible if smaller quantities of concrete and greater quantities of

natural aggregates are present (Fig. 10).
Fig. 10: Comparison of the content of acid-soluble sulfate in pure concrete and
recycled concrete aggregates, data from [35][21], [23], [35], [74], [81], [87], [92], [93],
[95], [96], [99], [101], [102], [104]
Granulometric analyses of recycled building materials

The most important granulometric parameter of recycled building materials is their
particle size distribution. However, unlike material composition, this parameter is less
critical because it can be influenced by appropriate processing whilst relevant
requirements are not very demanding. The broad ranges listed in Table 3 apply to
reuse in frost protection or gravel base courses. Defined particle size ranges of
recycled materials are used as concrete aggregates.
Table 3: Requirements for the particle size distribution of recycled building materials
in road construction
Material for frost protection
Material for base courses
courses
Particle
[mm]
size
Percentage passing [m.-%]
Percentage passing [m.-%]
35
40
16
47
5.6
22
60
11.2
35
68
0.063
0.5
1
15
75
22.4
47
87
55
85
45
90
99
90
99

63
100
100
100
100
Screening is used to determine the particle size distribution. In addition, photo-optical

methods have been available for quite some time to determine sizes and shapes of
aggregate particles. However, these methods are currently primarily used in research
because they are not yet covered by applicable codes and standards [24].
Fig. 11 analyzes the particle size distributions taken from the test records. A total
27 grading curves was established for the recycled building materials; only two
these curves lie outside the required range. The specified limit of up to 5 m.-%
fines smaller than 0.063 mm, which is crucial to determine water permeability,
complied with.
of
of
of
is
Fig. 11: Grading curves of recycled building materials for road construction, data from
the quality control of 27 recycled aggregates
Most of these grading curves can be accurately described using the distribution
function according to Gates, Gaudin and Schuhmann:
= 100 (
particle size in [mm]

maximum particle size in [mm]
distribution exponent in [-]
The graphs highlighted in red apply to xmax = 56 mm and n = 0.55 and xmax = 32 mm
and n = 0.4, respectively.

Practices commonly applied in the recycling industry show that high brick ratios in the
feed material lead to higher ratios of recycled sand with particles smaller than 4 mm.
This tendency was not revealed by the data analyzed for the purpose of this paper.
However, the brick content varied only between 0 and 24%, which might have led to
this outcome.
Physical properties
Bulk density is the most important parameter to describe the physical properties of
recycled materials because it captures the porosity of the material. It is closely
related to water absorption. Both parameters are features that are helpful to classify
recycled building materials. According to the standard applicable in Germany,
recycled materials intended for concrete production must have a bulk density of at
least 2,000 kg/m and a water absorption rate of less than 15 m.-%.
Bulk density is derived from the mass of the solids contained in the sample and the
gross volume of the particles that includes the voids within the particles. Bulk density
can relate to the dry or water-saturated state of the sample. Dry bulk density (OD) is
lower than the materials bulk density in its water-saturated, surface-dry state (SSD).
When opening up the voids, which is achieved by disintegration to particle sizes of
less than 63 m, pure density becomes the characteristic feature. To a certain extent,
it reflects the mineralogical composition of the material. According to applicable rules
and standards, the liquid pycnometer or measuring cylinder method must be used to
measure bulk density. Either of these methods relies on liquid displacement to
determine the volume of the material. Particles must be saturated with the testing
liquid prior to the measurement because liquid absorption by the porous recycled
building materials would otherwise distort the outcome of the volume measurement.
This step may lead to subjective errors particularly for small particle sizes. The solid
pycnometer method is a good alternative solution [25]. This setup uses a free-flowing
dry powder that replicates the irregular contours of the particles. Volume and density
of the material particles are determined by displacement of this dry powder. This
method can be used for particle sizes starting at 4 mm. The testing powder may be
contaminated if samples with particles from 1 mm are analyzed, which is why it is
necessary to replace the powder more often.
When controlling the quality of recycled building materials intended for road
construction, their dry bulk densities must be determined on the basis of liquid
displacement, but there are no specific requirements related to this method. What is
striking, though, is that these values are relatively high. The mean derived from 25
measured values amounts to 2,544 kg/m. The analysis published by Silva et al [26]
includes 292 measured values and indicates a mean of 2,327 kg/m. A plausible
explanation of this phenomenon might be the natural stone and/or asphalt content of
the recycled materials. However, no correlations could be derived from available
data.
Unlike dense natural aggregates, recycled building materials absorb water that
mainly penetrates through capillary pores into the particles of the building material.
Water absorption is time-dependent: in the first few minutes after contact of the
recycled materials with water, the water absorption rate is very high. It then
decelerates up to the point of water saturation. After ten minutes, the material will

usually have absorbed 85 to 90% of the amount of water absorbed after 24 hours
[27], which is why determining water absorption leads to reliable results that are
characteristic of the material even after shorter measuring periods.
The dependency of water absorption on bulk density is a feature typical of recycled
building materials. This property has been described in many publications, including
Hansen [28] and Silva et al [26]. This dependency arises from the proportional
relationship between water absorption and porosity. If the entire porous portion of the
material consists of open capillary pores, the following equations apply to the volume
of absorbed water or to the water absorption rate relative to the sample mass:
2 =
= 100 (
2 ,
volume of water in pores, volume of particles

total porosity
water absorption [m.-%]
density of water [kg/m]
OD particle density in oven-dry stage [kg/m]
absolute (powder) density [kg/m]
If only a certain part of the pores is, or can be, filled with water, the saturation value
must be considered as an additional influential parameter. Bulk density becomes the
dominant parameter in types of materials with similar pure densities, such as in the
case of mineral building materials.
Fig. 12 compares water absorption values measured for various building materials
with calculated values. Values measured for concrete, clay brick and calcium silica
brick are slightly below the calculated values, which is indicative of incomplete water
saturation. Lightweight materials such as lightweight or aerated autoclaved concrete
reveal greater variations, which might be due to the fact that, in lightweight concrete,
not all of the pores are accessible to water or that the pores present in aerated
concrete are so large that no capillary suction takes place. The correlation calculated
by Silva et al confirms the water absorption/f(bulk density) curve derived from bulk
and pure densities. The differences are due to incomplete water saturation of the
particles as well as to a uniform mean pure density that was applied to all building
materials under study in order to simplify the calculation process.
Bulk density is the guiding parameter for a number of material characteristics that are
associated with mechanical damage to the particles caused by external actions. At
an identical intensity of such an action, a material with a lower bulk density will
usually be disintegrated to a larger extent than a material whose bulk density is
comparatively greater. This principle applies to both material processing as such and
actions in service, for example by traffic loads or freeze/thaw cycles.

Fig. 12: Water absorption of different building materials as a function of OD particle

density, data from [79]
The Los Angeles method is most commonly applied to determine particle strength.
For this purpose, the material is subjected to abrasion and impact. The sample is
placed in a rotating drum and impacted by steel balls. After completion of this
process, the mass percentage of the sample is determined that passes through an
analytical sieve with a certain mesh width. The abrasion measured for recycled
building materials in the Los Angeles test ranges from 20 to 40 m.-% (Fig. 13). It is
thus slightly greater than the reference values for natural mineral aggregates. Both
types of aggregate show an almost identical increase in the generated abrasion in
line with decreasing bulk density, as well as almost the same degree of variation.
The freeze/thaw resistance of recycled building materials is measured by subjecting
the aggregates to ten freeze/thaw cycles after 24 hours of immersion in water.
Thereafter, the loss of mass of the aggregates is determined by frost-induced
spalling. The freeze/thaw resistance of recycled concrete materials is smaller than
that of natural mineral aggregates due to their increased water absorption (Fig. 14).
The dependency on particle bulk density is distorted because of other influential
factors superimposed on this effect, which include the type of cement, cement
content, water/cement ratio and freeze/thaw resistance of the aggregates that were
used to produce the parent concrete. Air-entraining agents in the parent concrete
significantly improve the freeze/thaw resistance of the aggregates derived from it.

Fig. 13: LA abrasion value as a function of OD particle density, data from [29], [30],
[31], [32], [80]

Fig. 14: Freeze/thaw resistance as a function of OD particle density, data from [33],
[34], [35], [36]
The results of freeze/thaw tests are influenced not only by the greater porosity of
recycled materials but also by the test methods developed for dense mineral
materials. For instance, pre-drying of the samples at 105C causes increased
spalling [37].
Sophisticated Characterization
Concrete as composite material

Concrete is a composite material. Already crushing into aggregates for example 0/32
mm leads to differentiations (Fig. 15). Pure mortar particles occur as well as pure
gravel particles. Therefore, the particles of a sample of recycled concrete aggregates
have different contents of cement paste. This results in variations of properties. For
example, the particle density spread over the range from 1,900 kg/m to 2,700 kg/m.
The lower limit corresponds approximately to the density of pure cement paste; the
upper limit to the density of natural aggregates.
Fig. 15: Different composition of concrete grains taken from a stockpile of processed
concrete rubble
The hardened cement paste content is usually determined as the content of acidsoluble constituents. This method is not applicable when the concrete contains
calcite or dolomite aggregates. Florea [38] proposes an alternative method. The

transition of -SiO2 in -SiO2 quantified by differential scanning calorimetry is used as

a measure of the quartz content of recycled concrete aggregates.
From the physical parameters of the endpoints pure cement paste and pure natural
aggregate of the mixture series a state diagram can be simplistically developed. The
densities of the composite particles can be calculated from the values of pure
materials and the cement paste content:

=
+
+ = 1
,
,
portions of cement paste and natural aggregates

densities of cement paste and natural aggregates
In Fig. 16, the calculated changes of absolute density, OD particle density and total
porosity depending on the cement paste content are compared with the
corresponding measured values. Moreover, the mortar content of recycled
aggregates that is discussed more frequently in the literature is introduced in the
diagram. The measured values fit well with the curves derived from the calculations.
Fig. 16: State diagram of recycled concrete aggregates

Experimental data measured at aggregates from a recycling plant [22], from defined
concretes [98], from model concretes with graded content of cement paste [94],
fractions from processed concrete [38], mortar content of recycled aggregates from
[39]

Due to the variability of the composition of recycled aggregates the amount of old
cement paste, which is introduced into the concrete of the second generation can
vary considerably. If it is assumed that 50 vol.-% recycled aggregates and 50 vol.-%
natural aggregates are used and the total aggregate quantity is 1800 kg per cubic
meter of concrete, the following cases result in dependence on the cement paste
content of recycled materials:
Case 1: The cement paste content of the recycled aggregates, which have
undergone a specific treatment, is 5 m.-%. The introduced amount of old
cement paste is only 40 kg per cubic meter of concrete and thus remains at
approximately one order of magnitude below the new cement pate content.
Case 2: The cement paste content of the recycled aggregates is 25 m.-%. The
introduced amount of old cement paste is 210 kg per m of concrete, i.e. about
half of the new cement paste content.
If the example is pursued further, the additional porosity amounts approximately 0.5
vol-% in the Case 1 and 3 vol-% in the Case 2. An increase of the content of capillary
pores of 1 vol.-% results in a strength loss of about 3 to 4 N/mm.The decrease in the
compressive strength of recycled aggregate concrete with increasing amount of
recycled aggregates is thus explainable as well as the large variations that occur
(Fig. 17).
Fig. 17: Relative compressive strength of concrete in dependence on the content of

reycled aggregates, data from [40]-[58]
The input of old cement paste has effects on the compressive strength and the
modulus of elasticity as important design parameters of concrete:
The compressive strength is seriously dependent on the porosity. In recycled
aggregate concrete the porosity arises on the one hand from the excess water
of the "current" hydration. On the other hand the porosity is introduced by the
old cement paste of the recycled aggregates.

For the modulus of elasticity, the porosity and the amount of hydration
products, which are more easily deformable than aggregates, are the most
important factors. Again, the use of secondary aggregates causes compared
to concrete made of natural aggregates elevated levels.
Generally the decrease of elastic modulus is greater than the decrease of

compressive-strength [59] because in addition to the porosity, the deformability of the
hydration products has influence. Some relations proposed to calculate the elastic
modulus of recycled aggregate concrete take into account this dual influence. By the
parameter compressive strength the influence of the porosity is received. By the
relative concrete density the content of cement paste is taken into account.
2
0,33
= 9100
(
) (1
) acc. to [60]
2400
500
1,5
0,5
) (
) acc.
(
2000
2300
to [61]
= 190000
modulus of elasticity of concrete [N/mm]
compressive strength [N/mm]

density of fresh concrete [kg/m]
density of hardened concrete [kg/m]
portion of clay brick [mass-%]
The impact of the hardened cement paste content of crushed concrete can be
illustrated by the imaginary experiment "multiple recycling". The starting point is a
concrete of the 1st generation, which is made from natural aggregates. This concrete
is crushed at end of its life to a recycled aggregate. From these aggregates a
concrete of the second generation is prepared, in turn, is crushed and processed to
the 3rd generation concrete. With each cycle increases the content of cement paste
increases and the density decreases. The concrete of the n-th generation consists
only of cement paste. Its density has declined to the density of cement paste (
Fig. 18. As limits for the decline of strength and elastic modulus the values typical for
cement paste can be expected. The development of a series makes it possible to
calculate these parameters depending on the number of completed cycles.
Fig. 18: Changes in bulk density and cement paste content for multiple concrete
recycling stages, data from [62], [35], [97]

Multiple recycling stages of concrete will remain an exception due to the long service
life of the material. Furthermore, the loss of quality that is also observed in other
recycling processes will be limited if only a certain percentage of the mineral
aggregate is replaced. If 50% of the aggregates are replaced with recycled materials,
a cement paste content of 0.35 kg/kg and a bulk density of 2,230 kg/m will be
reached in the recycled material after only three cycles. Both parameters will remain
almost constant even when going through subsequent cycles.
Reaction potential of recycled aggregates

In contrast to natural aggregates, recycled aggregates are not inert. Reactions of the
components of the cement paste can run directly in the particles or in the concretes
produced from them. Two factors favor the progress of the reactions:
The porosity of the recycled aggregates. Through that the transport of water
as necessary partner for all reactions is possible.
The increase in the specific surface area through the treatment. Compared
with the demolished concrete as starting material, after crushing the recycled
aggregates have a many times increased surface. Through this surface the
reactants can penetrate into the grain. When a concrete fragment with
dimensions 1000x1000x140 mm is compared with a particle with a diameter
of 32 mm, the surfaces differs by a factor of 800.
The reaction potential of recycled materials is caused by the components of the old
cement paste - non-hydrated cement residues, calcium hydroxide, calcium silicate
hydrates, sulfate-free and sulfate-containing aluminum hydrates. Already during
storage, reactions can take place. During utilization of recycled aggregates in
unbound base courses in road constructions reactions can occur when water can
penetrate into the layers. When recycled aggregate concrete is produced, it depends
on the tightness, how the old cement paste behaves. If residues of unhydrated
cement are available, that become accessible by the processing, hydration reactions
can take place in the presence of water. Experimental studies confirm the hydraulic
potential however only if very young concrete is crushed and used as aggregate for
new concrete [62]. No hydraulic potential was found at recycled aggregates from
recycling plants. Even if the aggregates were ground at cement fineness and a
percentage of the cement was substituted by this concrete powder the strength
decreases. Simplified the strength loss corresponds to the "dilution" of the cement
[64].
The calcium hydroxide and the calcium silicate hydrates react in the presence of
water and CO2 according to the following simplified equations:

The products of these reactions are calcium carbonate and silica. Investigations at
defined mortars and crushed concretes confirm the course of both reactions and the
strong influence of particle size [65]. In the fractions 0/0.5 mm and 0.5/2 mm the
originally present calcium hydroxide was converted into calcium carbonate after 28
days dry-wet alternation. Between the 28th and the 91st day, there was a further
increase in the carbonate content, which must be attributed to the conversion of the
calcium silicate hydrates. The carbonation will begin already during storage of
recycled aggregates and will continue in unbound courses of roads or traffic areas.
However, it is mainly related to the sand fractions. The often-observed phenomena of
solidification occurring especially at stockpiles of fine aggregates can be caused
thereby. The coarse particles are not fully carbonated, because the products of
carbonation form a dense layer, which hinders the progress of the reaction.
For concrete, the phenomenon of structural damage caused by ettringite formation is
known for more than 100 years. If sulfate-containing water attacks the concrete,
ettringite 3CaOAl2O33CaSO432H2O is formed from the calcium aluminate hydrates
of the cement paste and the sulfate. An increase of volume up to 8 times of the
original volume occurs. In parallel to the formation of ettringite also thaumasite
CaSiO3CaCO3CaSO416H2O can be formed as the result of a sulfate attack. In this
case the reactants are silica, carbonate, sulfate and water. Ettringite and thaumasite
are related minerals with similar crystal structures. Thaumasite leads in contrast to
ettringite to a softening up to a "dissolution of the cement matrix, so that hardened
concrete is converted into a pulpy mass" [66].
It can be expected that concrete aggregates will behave - similar to concretes - not
inert, if they are exposed to sulfate-containing water. The sources of the sulfatecontaining water are on the one hand gypsum particles, which are present as minor
component and will be dissolved by contact with water. On the other hand, sulfate in
soils can be a source. If recycled concrete aggregates contain gypsum it can not be
excluded that the ettringite or thaumasite formation already starts at the stockpile at
the recycling plant. However, expansion phenomena are not apparent because the
pore space between the particles is sufficient to accommodate the formed ettringite.
This argument is no longer true in base courses. The contact with water can cause
the formation of ettringite and the associated increase of volume. The results are
uneven elevations and bulges.
At some roads - constructed with recycled aggregates in the recent years - such
phenomena has been observed. In all examined cases of damage ettringite and
partly thaumasite were detected. The damages occurred mostly after periods of frost.
From the concrete research is known that ettringite and thaumasite formation takes
place preferentially at low temperatures.

Construction and demolition waste as a raw material

Any recycled material can be considered to be reused as a raw material if such reuse
is associated with a process that converts or transforms the constituents the material
is composed of. Its chemical composition and reactivity are key parameters in this
regard. Other material characteristics that are important in traditional recycling paths
will become less significant. Powders or fine-grained input materials can be
processed. The reuse of waste as a raw material appears to be feasible in brick or
cement production if this consideration is limited to building material production
processes. Other construction materials whose production relies on transformative
processes include glass and rock wool, foamed glass and expanded clay.
The main constituents of construction waste include concrete, brick, calcium silica
brick, aerated autoclaved concrete and natural stone whereas mortar and render are
secondary constituents. Chemical analyses of the pure main constituents were
compiled to initially determine the range within which the chemical composition of the
material lies. Compositions of recycled building materials from concrete rubble and
masonry rubble were compared to each other (Table 4).
Table 4: Oxide composition of pure-grade main constituents of construction waste
and of recycled building materials, data from [21], [23], [67]-[73], [81]-[93], [95], [96],
[99]-[104]
[m.-%]
Pure-grade main constituents
Mean
Clay brick
Standard
Number: 40
deviation
Mean
Calcium silica brick
Standard
Number: 4
deviation
Aerated autoclaved Mean
concrete
Standard
Number: 13
deviation
Concrete
with Mean
siliceous
Standard
aggregates
deviation
Number: 30
Concrete
with Mean
calcitic
Standard
aggregates
deviation
Number: 16
Recycled building materials
Recycled
building Mean
materials
from
Standard
masonry rubble
deviation
Number: 65
Recycled
building Mean
materials
from
Standard
concrete rubble
deviation
Number: 62
LOI
SiO2
Al2O3 Fe2O3 CaO
MgO
K2O
Na2O SO3
1.02
66.50 14.52 5.89
4.42
2.20
2.92
0.74
0.38
1.00
6.96
2.78
1.92
3.99
1.45
0.72
0.38
0.60
4.40
80.70 3.54
1.20
7.85
0.29
1.09
0.27
0.14
0.75
4.34
1.20
0.70
0.67
0.30
0.57
0.22
0.10
14.36 50.26 2.60
1.05
25.94 0.77
0.92
0.30
2.53
5.20
3.96
0.80
0.26
3.82
0.38
0.44
0.08
1.07
5.56
74.03 3.78
1.86
11.61 0.89
0.94
0.40
0.52
1.62
9.45
2.66
1.88
3.37
1.36
1.04
0.58
0.27
28.57 23.34 2.99
1.86
34.92 6.05
0.77
0.30
0.66
8.34
9.48
2.58
1.52
5.86
2.4
1.12
0.52
0.41
5.57
66.75 9.67
3.72
8.21
1.60
2.12
0.77
0.84
2.64
6.68
1.92
1.24
3.68
1.04
0.57
0.28
0.72
9.73
65.09 6.03
2.45
11.93 1.20
1.58
0.80
0.58
5.46
10.18 1.51
0.68
5.30
0.67
0.32
0.18
1.02

All building materials (except concrete with calcitic aggregates) are dominated by
their
SiO2 content. Other constituents reveal the following differences:
Compared to the other building materials under study, brick is characterized
by significantly higher Al2O3 and Fe2O3 contents. The same applies to their
alkali content. Its loss on ignition is low because brick is produced in a thermal
process.
Compared to brick, mineral bound building materials (i.e. calcium silica brick,
aerated autoclaved concrete and concrete) contain a greater amount of CaO
and exhibit a higher loss on ignition. Both parameters are due to a
characteristic common to these materials, which is that their strength
development depends on calcium silicate hydrates.
Even recycled materials of various origins continue to reveal the differences between
pure-grade concretes and brick.
The oxide compositions of the pure-grade main constituents were integrated in the
ternary systems typical of building materials, which comprise a combination of SiO2CaO-(Al2O3+Fe2O3) or SiO2-fluxing agent (Fe2O3+CaO+MgO+K2O+Na2O)-Al2O3, to
identify construction materials for which recycled portions are suitable for use as a
raw material. The so-called SiO2-CaO-(Al2O3+Fe2O3) Rankin diagram can be used to
evaluate the materials that are suitable for cement clinker production (Fig. 19), as
well as to assess the potential of ground additives and admixtures. In the latter case,
reactivity is another key parameter besides their chemical composition. The SiO2(Fe2O3+CaO+MgO+K2O+Na2O)-Al2O3 ternary system is an appropriate means to
evaluate ceramic raw materials, including those used for producing expanded clay,
and glass and rock wool (Fig. 20). This analysis results in the following conclusions:
In both ternary system diagrams, mineral bound, pure-grade building materials
are located in a virtually identical position. Their Al2O3 content remains almost
constant at about 4 m.-%. With the exception of concrete with calcitic
aggregates, their SiO2 contents exceed 60 m.-%. Their composition
significantly differs from the range of raw materials or additives used for the
production of cement or of ceramic materials. Concrete with calcitic
aggregates is the only material that might be suitable as a raw material
component for cement production. All other building materials under study
could serve as SiO2 sources.
In both ternary system diagrams, the parameters for brick are concentrated in
a certain range. Their reuse for brick production is possible. However, there is
an upper limit for the possible ratio of recycled brick that results from the
plasticity of the raw mix that is required to create a moldable material. With
one exception, the composition of all types of brick under study lies in the
range typical of expanded clay, which makes them a promising option for
reuse in this area. Furthermore, brick could be used as an Al2O3-containing
raw material component in cement production.
When entering the compositions of the recycled building materials in the ternary
system diagrams, the variations between the main materials are less significant
compared to their original state Fig. 19, Fig. 22). Their chemical compositions were
largely aligned with each other in the demolition and reprocessing phases.

In terms of reuse as a raw material, the chemical compositions of the recycled

materials permit the conclusion that they could be used as corrective components in
cement production. However, recycled materials from concrete rubble have an Al2O3
content of only 10 m.-%, which is why they cannot compensate aluminum deficits
present in raw materials for cement. Their CaO content ranges from 10 to 30 m.-%
and makes it possible to use the recycled materials as a raw material component.
However, their contribution to improving the carbon footprint of the cement burning
process is low if the calcium oxide is not present in carbonate form. Recycled
materials from masonry rubble are suitable as an Al2O3 source for cement clinker
production; they are already being used for this purpose [77]. Due to their
composition, these materials also provide the potential for use as pozzolanic
admixtures if their reactivity can be increased by mechanical and/or thermal pretreatment.

Fig. 19: Position of the main constituents of construction and demolition waste in the
SiO2-CaO-(Al2O3+Fe2O3) cement ternary system
Fig. 20: Position of the main constituents of construction and demolition waste in the
SiO2-(Fe2O3+CaO+MgO+K2O+Na2O)-Al2O3 ceramic ternary system
Ranges for expanded clay according to Riley [75] and Wilson [76]

Fig. 21: Position of recycled building materials in the SiO2-CaO-(Al2O3+Fe2O3)

cement ternary system
Fig.
22:
Position
of
recycled
building
materials
(Fe2O3+CaO+MgO+K2O+Na2O)-Al2O3 ceramic ternary system
Ranges for expanded clay according to Riley [75] and Wilson [76]
in
the
SiO2-

Ceramic-based products provide more favorable conditions for reuse as raw

materials. It appears feasible to reuse these materials in brick production to at least
partially replace natural clay. Another option to reuse them as input materials is to
produce lightweight aggregates from masonry rubble. In this case, an expanding
agent is added to the masonry rubble, which is then ground, granulated and burned
in a rotary kiln. Production parameters are similar to those applied to the production
of expanded clay. Variations in the raw materials brick content between 25 and 75
m.-% have almost no influence on the product. Transfer of laboratory results to pilotscale production is currently being prepared [78].
Summary
The individual types of construction waste constitute the most significant waste
stream in Germany. 70% of construction waste are already being reused,
predominantly in road construction and civil engineering. However, the reuse of
processed concrete rubble for concrete production continues to be of secondary
significance. On the other hand, we are currently seeing increased activities to
develop and publicly promote practically feasible processes to produce concrete with
recycled aggregates that conform to applicable codes and standards.
Recycled materials for road construction and concrete production must meet specific
quality requirements in terms of their composition, particle size distribution and
particle shape, and abrasion and freeze/thaw resistance. The composition of the
material is determined by manual sorting. This method may result in errors because
sorted quantities are too small or particles are identified incorrectly. Recycled
materials comply with the resistance parameters specified in applicable standards,
but they are less resistant to actions than natural aggregates.
The detailed evaluation of recycled materials reveals quite a number of knowledge
gaps, for instance with respect to the unambiguous characterization of concrete
(which is a composite material) and of cement paste reactivity in the presence of
water, sulfate or carbon dioxide. Parameters that describe the chemical composition
of recycled materials need to be determined to reuse them as raw materials that
provide an alternative to current recycling practices. For this purpose, initial data will
be presented and conclusions be drawn.
References
[1]
Arbeitsgemeinschaft Kreislaufwirtschaftstraeger Bau (KWTB): 1.-5.

Monitoring-Bericht Bauabfaelle. Erhebungen 1996, 1998, 2000, 2002,
2004. Berlin/ Duesseldorf/ Duisburg.
[2]
Bundesverband Baustoffe Steine und Erden e.V.: Mineralische

Bauabfaelle. Monitoring 2006, 2008, 2010, 2012. Berlin.
[3]
Bundesanstalt fuer Strassenwesen: Jahresbericht 2009/2010. Bergisch

Gladbach, Mai 2011.
[4]
http://www.stadtentwicklung.berlin.de/umwelt/abfall/rc-beton/ from
09.05.2015.

[5]
Technische Lieferbedingungen fuer Asphaltmischgut fuer den Bau von

Verkehrsflaechenbefestigungen TL Asphalt-StB. Forschungsgesellschaft
fuer Straen- und Verkehrswesen, FGSV-Verlag, Koeln 2007.
[6]
Technische Lieferbedingungen fuer Asphaltgranulat TL AG-StB.

Forschungsgesellschaft fuer Straen- und Verkehrswesen, FGSV-Verlag,
Koeln 2009.
[7]
Merkblatt fuer die Verwertung von Asphaltgranulat M VAG.

Koeln 2009.
[8]
Technische Lieferbedingungen fuer Gesteinskoernungen im Straenbau TL

Gestein-StB. Forschungsgesellschaft fuer Straen- und Verkehrswesen,
FGSV-Verlag, Koeln 2004/Fassung 2007.
[9]
Technische Lieferbedingungen fuer Baustoffgemische und Boeden zur

Herstellung von Schichten ohne Bindemittel im Straenbau TL SoB-StB.
Koeln 2004/Fassung 2007.
[10]
Deutscher Ausschuss fuer Stahlbeton: DAfStb-Richtlinie Beton nach DIN

EN 206-1 und DIN 1045-2 mit rezyklierten Gesteinskoernungen nach DIN
EN 12620 - Teil 1: Anforderungen an den Beton fuer die Bemessung nach
DIN EN 1992-1-1, Beuth-Verlag, Berlin 2010.
[11]
Deutscher Ausschuss fuer Stahlbeton: DAfStb-Richtlinie Vorbeugende

Manahmen gegen schaedigende Alkalireaktionen im Beton (AlkaliRichtlinie), Beuth-Verlag, Berlin 2007.
[12]
DIN EN 12620: Gesteinskoernungen fuer Beton. DIN Deutsches Institut

fuer Normung, Beuth-Verlag, Berlin 2013.
[13]
DIN 4226-100: Gesteinskoernungen von Beton und Moertel, Teil 100:

Rezyklierte Gesteinskoernungen. DIN Deutsches Institut fuer Normung,
Beuth-Verlag, Berlin 2002.
[14]
Rasemann, W. et al.: Probenahme bei Recyclingglas. Freiberger

Forschungshefte A864, Vortraege zum 52. BHT Freiberg, (2001) S. 89-106.
[15]
Ketelhut, R.: Sortieranalysen zur Qualitaetssicherung von Abfaellen.

Urschrift zum 10. Dialog Abfallwirtschaft M-V am 12. Juni 2006.
stoffstromdesign ralf ketelhut.
[16]
Angulo, S.C.; Mueller, A.: Determination of construction and demolition

recycled aggregates composition, in considering their heterogeneity.
Materials and Structures (2009) 42, pp. 739748.
[17]
Gy, P.: The sampling of particulate materials a general theory.

International Journal of Mineral Processing Vol. 3 (1976), pp. 289-312.
[18]
Sommer, K.: Probenahme von Pulvern und koernigen Massenguetern,

Grundlagen, Verfahren, Geraete. Springer Verlag, Berlin 1979.
[19]
Petersen, L.; Minkkinen, P.; Esbensen, K. H.: Representative sampling for

reliable data analysis: Theory of Sampling. Chemometrics and Intelligent
Laboratory Systems. 74, (2005), Mai, pp. 261-277.

[20]
Technische Lieferbedingungen fuer Gesteinskoernungen im Straenbau.

TL Gestein-StB 13_E4. Draft 2013.
[21]
Heinz, D.; Schubert, J.: Nachhaltige Verwertung von Betonbrechsand als

Betonzusatzstoff. Schlussbericht F250. Technische Universitaet Muenchen
2006.
[22]
Hauer, B.; Klein, H.; Mueller, C.; Ruppert, J.; Schaefer, S.; Spanka, G.;
Wassing, W.; Zunzer, U.; Ramolla, S.; Ruebner, K.; Meng, B.: Potenziale
des Sekundaerstoffeinsatzes im Betonbau Teilprojekt B1. Deutscher
Ausschuss fuer Stahlbeton, Heft 572, Beuth-Verlag, Berlin 2007, S. 182198.
[23]
Mahmoud Solyman: Classification of Recycled Sands and their

Applications as Fine Aggregates for Concrete and Bituminous Mixtures.
Dissertation. Universitaet Kassel 2005.
[24]
Stark, U.; Mueller, A.: Korngroee und Kornform von Gesteinskoernungen

und ihr Potenzial fuer die Betonforschung. Beton (2006), H. 9, S. 390
397.
[25]
Webb, P.A.: Volume and Density Determinations for Particle Technologists.

Micromeritics Instrument Corp. 2/16/01.
[26]
Silva, R.V.; de Brito, J.; Dhir, R.K.: Properties and composition of recycled
aggregates from construction and demolition waste suitable for concrete
production. Construction and Building Materials 65 (2014), pp. 201217.
[27]
Gruebl, P.; Ruehl, M.: German Committee for Reinforced Concrete

(DAfStb)-Code: Concrete with Recycled Aggregates. International
Symposium Sustainable Construction: Use of Recycled Concrete
Aggregate Dundee 1998.
[28]
Hansen, T.C.: Recycled aggregates and recycled aggregate concrete.

Second state-of-the-art report. Developments 1945-1985. Rilem Technical
Committee-37-RC Demolition and Recycling of Concrete. 1985.
[29]
Marta Snchez de Juan; Pilar Alaejos Gutirrez: Study on the influence of

attached mortar content on the properties of recycled concrete aggregate.
Construction and Building Materials 23 (2009), pp. 872877.
[30]
Springenschmid, R.; Schmiedmayer, R.; Friedl, L.: Zwischenbericht zum

BMBF-Forschungsvorhaben "Baustoffkreislauf im Massivbau" Teilprojekt
D/01. Muenchen Maerz 1997.
[31]
Schiel, P.; Schmiedmayer, R.; Friedl, L.: Zwischenbericht 1/1998 zum

BMBF-Forschungsvorhaben Baustoffkreislauf im Massivbau Teilprojekt
D/01. Muenchen Juli 1998.
[32]
Barbudo, A.; Agrela, F.; Ayuso, J.; Jimnez, J.R.; Poon C.S.: Statistical
analysis of recycled aggregates derived from different sources for sub-base
applications. Construction and Building Materials 28 (2012) pp. 129138.
[33]
Dombrowski, K.: Einfluss von Gesteinskoernungen auf die Dauerhaftigkeit

von Beton. Dissertation. Bauhaus-Universitaet Weimar 2003.
[34]
Manns, W.; Wies, S.: Frostwiderstand von Sekundaerzuschlag aus

Bauschutt. Zwischenbericht zum BMBF-Forschungsvorhaben

"Baustoffkreislauf im Massivbau" Teilprojekt D/04. Stuttgart

September1998.
[35]
Mueller, C.: Beton als kreislaufgerechter Baustoff. Dissertation. Beuth

Verlag Berlin Wien Zuerich 2001.
[36]
Eden, W. et al.: Eignung von rezykliertem Kalksandstein-Mauerwerk fuer

Tragschichten ohne Bindemittel. Forschungsvereinigung Kalk-Sand eV
Forschungsbericht Nr.111, Hannover Maerz 2010.
[37]
Aurstad, J.: Evaluation of unbound crushed concrete as road building

material Mechanical properties vs field performance. The 26th
International Baltic Road Conference, 28.-30. August 2006.
[38]
Florea, M.V.A.; Brouwers, H.J.H.: Properties of various size fractions of

crushed concrete related to process conditions and re-use. Cement and
Concrete Research 52 (2013), pp. 1121.
[39]
Marta Sanchez de Juan; Pilar Alaejos Gutierrez: Study on the influence of

attached mortar content on the properties of recycled concrete aggregate.
Construction and Building Materials Vol. 23 (2009), pp. 872-877.
[40]
Pauw, C.: De Bton Recycl. Centre Scientifique et Technique de la

Construction. CSTCRevue Nr. 2, 15 (1980) 6, S. 2 15.
[41]
Wesche, K.; Schulz, R. R.: Beton aus aufbereitetem Altbeton, Technologie

und Eigenschaften. Beton 32(1982) 2, S. 64 68.
[42]
Ravindrarajah, S.; Tam, C.T.: Properties of concrete made with crushed

concrete as coarse aggregate. Magazine of Concrete Research 37 (1985)
3, No. 130, pp. 29 38.
[43]
Dillmann, R.: Beton mit rezyklierten Zuschlaegen. Beton (1999) 2, S. 86

91.
[44]
Lukas, W.: Konzept fuer die Herstellung von Recycling-Beton aus

Baurestmassen-Zuschlaegen. BFT Beton + Fertigteiltechnik 60 (1994) 10,
S. 68 75.
[45]
Stewart, M.; Greco. D.: An investigation into the use of crushed concrete
aggregates for the production of fresh concrete. University of Technology,
Sydney, Faculty of Engineering December 1997.
[46]
Huenninghaus, U.: Gebrochener Beton als Betonzuschlag. Baustoff

Recycling Deponietechnik 6 (1990) 6, S. 29 30.
[47]
Lukas, W.: Auswirkung auf technologische Kenngroeen von Beton bei

Verwendung von Recycling-Material. Zement und Beton 38 (1993) 3, S. 33
35.
[48]
Bairagi, N.K.; Kishore Ravande; Pareek, V.K.: Behaviour of concrete with

different proportions of natural and recycled aggregates. Resources,
Conservation and Recycling, 9 (1993), pp. 109-126.
[49]
Park, S. G.: Recycled Concrete Construction Rubble as Aggregate for New

Concrete. Study Report No. 86 (1999) Branz, S. 1 11.
[50]
Ajdukiewicz, A.; Kliszczewicz, A.: Influence of recycled aggregates on

mechanical properties of HS/HPC. Cement & Concrete Composites Vol. 24
(2002), pp. 269279.

[51]
Gmez-Sobern, J.: Porosity of recycled concrete with substitution of

recycled concrete aggregate - An experimental study. Cement and
Concrete Research 32 (2002) 8, pp. 1391 1311.
[52]
Jianzhuang Xiaoa; Jiabin Lia; Ch. Zhang: Mechanical properties of recycled

aggregate concrete under uniaxial loading. Cement and Concrete
Research Vol. 35 (2005), pp. 1187 1194.
[53]
Etxeberria, M.; Vzquez, E.; Mari, A.: Microstructure analysis of hardened

recycled aggregate concrete. Magazine of Concrete Research Vol. 58
(2006), No. 10, December, pp. 683690.
[54]
Etxeberria, M.; Vzquez, E.; Mar, A.; Barra, M.: Influence of amount of
recycled coarse aggregates and production process on properties of
recycled aggregate concrete. Cement and Concrete Research Vol. 37
(2007), pp. 735742.
[55]
Belen Gonzlez-Fonteboa; Fernando Martinez-Abella; Javier Eiras-Lopez;

Sindy Seara-Paz: Effect of recycled coarse aggregate on damage of
recycled concrete. Materials and Structures Vol. 44 (2011), pp. 17591771.
[56]
M. Chakradhara Rao; S.K. Bhattacharyya; S.V. Barai: Behaviour of

recycled aggregate concrete under drop weight impact load. Construction
and Building Materials Vol. 25 (2011), pp. 6980.
[57]
Boedefeld, J.; Reschke, T.: Verwendung von Beton mit rezyklierten

Gesteinskoernungen bei Verkehrswasserbauten. BAW-Mitteilungen Nr. 93
2011, S. 49 -60.
[58]
Duangthidar Kotrayothar: Recycled aggregate concrete for structural

applications. PhD Thesis University of Western Sydney 2012.
[59]
Mueller, A: Chancen und Grenzen des Betonrecyclings. BFT International

(2013) 04, S. 78 92.
[60]
Roos, F.: Ein Beitrag zur Bemessung von Beton mit Zuschlag aus
rezyklierter Gesteinskoernung nach DIN 1045-1. Dissertation. Technische
Universitaet Muenchen 2002.
[61]
Kakizaki, M.; Harada, M.; Soshiroda, T.; Kubota, S.; Ikeda, T.; Kasai, Y.:
Strength and elastic modulus of recycled aggregate concrete. Proceedings
of the 2nd International RILEM Symposium on Demolition and Reuse of
Concrete and Masonry, Tokyo, Japan, 711 November 1988, pp. 565574.
[62]
Sumaiya Binte Huda, M. Shahria Alam: Mechanical behavior of three

generations of 100% repeated recycled coarse aggregate. Construction
and Building Materials 65 (2014) pp. 574582.
[63]
Katz, A.: Properties of concrete made with recycled aggregate from partially
hydrated old concrete. Cement and Concrete Research 33 (2003) 5, pp.
703 - 711
[64]
Mueller, A.; Liebezeit, S.; Badstuebner, A.: Verwertung von

Ueberschusssanden als Zusatz im Beton. Steinbruch und Sandgrube
(2012) 5, S.46-49.
[65]
Toshifumi Kikuchi; Yasuhiro Kuroda: Carbon Dioxide Uptake in Demolished

and Crushed Concrete. Journal of Advanced Concrete Technology 9
(2011) No. 1, pp. 15-24.

[66]
Sulfatangriff auf Beton - Stellungnahme des DAfStb. Deutscher Ausschuss

fuer Stahlbeton. Berlin, 31.03.2003.
[67]
Mueller, C.; Wiens, U.; Dora, B.: Verwendungsmoeglichkeiten von

Materialien, die bei der Aufbereitung von Altbeton/Bauschutt anfallen und
nicht wiederverwertbar sind. Baustoffkreislauf im Massivbau BiM-ProjektNr. B/04. Statusseminar September 1998.
[68]
Copernicus Research Project: Recycling of waste clay brick and tile

material for the partial replacement of cement in concrete. First annual
report. February 1st 1995 - January 31 st 1996.
[69]
Van Dijk, K.: Closing the clay brick cycle. PhD Thesis. Delft 2004.
[70]
Vejmelkova, E.; Keppert, M.; Rovnanikova, P.; Ondracek, M.; Kerner, Z.;
Cerny, R.: Properties of high performance concrete containing fine-ground
ceramics as supplementary cementitious material. Cement & Concrete
Composites 34 (2012) pp. 5561.
[71]
Joris Schoon, J.; Buysser, K.; Van Driessche, I.; De Belie, N.: Feasibility
study on the use of cellular concrete as alternative raw material for Portland
clinker. Construction and Building Materials 48 (2013) 7, pp. 25733.
[72]
Portland Cement Association, Skokie, Illinois, USA, 2003.
[73]
Winkler; A.: Herstellung von Baustoffen aus Baurestmassen. Forschungsbericht, Bauhaus-Universitaet Weimar, Shaker-Verlag, Aachen 2001.
[74]
Weimann, K.: Untersuchungen zur Nassaufbereitung von Betonbrechsand

unter Verwendung der Setzmaschinentechnik. Dissertation. BauhausUniversitaet Weimar 2008.
[75]
Riley Ch.M.: Relation of Chemical Properties to the Bloating of Clays.

Journal of Am. Ceram. Soc. Vol. 34 (1951),pp 123128.
[76]
Wilson, H.: Lightweight Aggregates for the Construction Industry. Journal of

Can. Ceram. Soc. Vol 22 (1953), pp 4450.Wilson
[77]
Lampl, C.: Sekundrrohstoffe Anforderungen und Einsatzgebiete fr die

Baustoffe der Zukunft. Nachhaltige Nutzung von Baurestmassen.
sterreichischer Wasser- und Abfallwirtschaftsverband, Wien 2009.
[78]
Mueller; A.: Die Herstellung von Aufbaukrnungen aus Mauerwerkbruch.

The Manufacture of Lightweight Aggregates from Recycled Masonry
Rubble. Mll und Abfall (2014), H. 11, S. 625-633.
Data from unpublished Diploma Thesis and Research Reports

[79]
Escher, M.: Untersuchungen zur Sortierbarkeit von heterogenen

Abbruchgemischen mittels Nahinfrarot-Technik. Diplomarbeit. BauhausUniversitaet Weimar 2010.
[80]
Donner, M.: Untersuchungen an Gesteinskoernungen aus dem Thueringer

Raum zum Widerstand gegen Schlag-Abrieb-Beanspruchungen nach dem
Los Angeles-Verfahren. Diplomarbeit. Bauhaus-Universitaet Weimar 2001.

[81]
Engelhardt,
K.: Untersuchungen zur Verwendung von
Ziegelrestmassen zur Herstellung von Moertel und Beton. Diplomarbeit.
Hochschule fuer Architektur und Bauwesen Weimar 1993.
[82]
Winkler, A.: Moertelherstellung aus Ziegelrestmassen. 2. Studienarbeit.

Hochschule fuer Architektur und Bauwesen Weimar 1994.
[83]
Winkler, A.: Eigenschaften von Moerteln und Betonen aus

Ziegelrestmassen. Diplomarbeit. Hochschule fuer Architektur und
Bauwesen Weimar 1995.
[84]
Jelen,
T.: Verwertungsmoeglichkeiten von verschiedenen
Dachbedeckungsmaterialien. Diplomarbeit. Bauhaus-Universitaet Weimar
1996.
[85]
Hechler, N.: Eigenschaften und Verwendungsmoeglichkeiten fuer RCSande aus Mauerwerks- und Betonschutt. Diplomarbeit. BauhausUniversitaet Weimar 1997.
[86]
Latocha, M.:
Ermittlung der Produkteigenschaften von Baurestmassen
nach der Zerkleinerung in unterschiedlichen Brechertypen. 2. Studienarbeit.
Bauhaus-Universitaet Weimar 1998.
[87]
Schroedter, T.: Vergleichende Untersuchungen zur Mahlbarkeit

ausgewaehlter, selektierter Baureststoffe. Diplomarbeit. BauhausUniversitaet Weimar 1998.
[88]
Arnold, T.: Vergleichende Untersuchung zum Einflu der Aufbereitung auf

die Puzzolanitaet von Hochziegelmehlen, Klinkermauerziegelmehlen und
Steinkohlenflugasche. Diplomarbeit. Bauhaus-Universitaet Weimar 1998.
[89]
Latocha, M.: Verwertungsmoeglichkeiten der Feinfraktionen aufbereiteten

Abbruchmaterials als Betonzusatzstoff. Projektarbeit. Bauhaus-Universitaet
Weimar 1999.
[90]
Neugebauer, S.: Herstellung von Moertel mit Zuschlaegen aus

Recyclingsand. Diplomarbeit. Bauhaus-Universitaet Weimar 1999.
[91]
Tischer, M.: Vergleichende Messungen zum Schmelzverhalten. BauhausUniversitaet Weimar, Fakultaet Bauingenieurwesen, Diplomarbeit, Weimar
1999.
[92]
Kehr, K.: Untersuchungen zur Homogenitaet an Altbeton und

Abbruchziegel. Diplomarbeit. Bauhaus-Universitaet Weimar 2006.
[93]
Walther, C.: Die BBW Recycling Mittelelbe GmbH Reportage ueber ein
erfolgreiches Recyclingunternehmen. Diplomarbeit. Bauhaus-Universitaet
Weimar 2006.
[94]
Wolff, E.: Qualitaetskriterien fuer rezyklierte Zuschlaege fuer die

Betonherstellung. Diplomarbeit. Bauhaus-Universitaet Weimar 2007.
[95]
Laub, K: Systematisierung der Mahlbarkeit von reinen und binaeren

Modellgemischen Untersuchungen zu Anreicherungseffekten.
Diplomarbeit. Bauhaus-Universitaet Weimar 2008.
[96]
Lehmann, D.: Verwertung von RC-Mauerwerk. Diplomarbeit. BauhausUniversitaet Weimar 2009.

[97]
Thomas, H.: Doppelrecycling von Beton Festigkeits- und

Verformungseigenschaften eines zweimal recycelten Betons. Diplomarbeit,
Technische Universitt Berlin 2001.
[98]
Baustoffkreislauf im Massivbau. Teilprojekt C 05. Einfluss der

Brechwerkzeuge auf die Eigenschaften von Recycling-Granulaten im
Hinblick auf eine Eignung als Zuschlag fuer Beton nach DIN 1045.
Gesellschaft zur Aufbereitung von Baustoff mbH, Bremen 1998.
[99]
Abschlussbericht: Verwertungsmoeglichkeiten fuer die Sandfraktion von

Recyclingbaustoffen aus Beton. DFG-Forschungsvorhaben 2001.
[100] Abschlussbericht: Granulate aus Porenbeton und Ziegelbruch. Gefoerdert

vom Thueringer Ministerium fuer Wissenschaft, Forschung und Kunst 2001.
[101] Abschlussbericht: Entwicklung eines Trennverfahrens fuer
gipskontaminierten Betonbruch. Kooperationsprojekt. Gefoerdert vom
Bundesministerium fuer Wirtschaft und Technologie 2009.
[102] Abschlussbericht: Entwicklung von selbsterhaertenden RecyclingBaustoffen fuer Tragschichten im Straenbau. Kooperationsprojekt.
Gefoerdert vom Bundesministerium fuer Wirtschaft und Technologie 2009.
[103] Abschlussbericht: Entwicklung von kostenoptimierten und qualitativ
verbesserten Betonrezepturen fuer Betonprodukte auf der Basis von
hydraulisch aktiviertem Betonsand als Zementzusatzstoff.
Kooperationsprojekt. Gefoerdert vom Bundesministerium fuer Wirtschaft
und Technologie 2009.
[104] Daten aus dem BMBF-Verbundvorhaben: Steigerung der
Ressourceneffizienz im Bauwesen durch die Entwicklung innovativer
Technologien fuer die Herstellung hochwertiger Aufbaukoernungen aus
sekundaeren Rohstoffen auf der Basis von heterogenen Bau- und
Abbruchabfaellen. Gefoerdert vom Bundesministerium fuer Bildung und
Forschung 2009 bis 2012.

DESIGN OF A SEALING BIOMANSORY SINCE CRUDE LAND: THE

ADOBE BRICK CASE PROJECT
Daniel Augusto de Moura Pereira (1). Alusio Braz de Melo (2)
(1) CDSA, Universidade Federal de Campina Grande, Brazil
(2) CT, Universidade Federal da Paraba, Brazil
Abstract
Raw earth is the worlds most ancient building material and the use of this kind of technique is
present since man felt the need to build his own shelter. In Brazil, there are many regions, mainly
at the north and northeast, which have potential to use bio-construction through raw earth usage.
In this context, the goal of the present research is to analyse soils in a region at Parabas semiarid
in order to build masonry walls through raw earth to build social interest housing in two cities at
Parabas Cariri region, which shows a humid and dry climate. Results indicated that 50% of the
sample has presented a light-loam kind of soil, 25% was of a clay kind, 8,33% light sandy kind of
soil, 8,33% silty kind of soil and 8,34% was of a sandy-clay kind of soil, which points out that one
of those soils located at Sum is indicated to product adobe bricks to be used for building social
housing at that location.
Keywords: construction material, land use, project, masonry, adobe
1.
INTRODUCTION
The actual preoccupation with the bigger and bigger lack of production resources has been
leading the productive environment to the need to use non-conventional building technologies
which are compatile with the sustainability scenario already installed worldwide. Considering
the great habitational deficit Brazil has been through and the pursuit of non-polluting,
renewable and low-cost material, as opposed to conventional and industrialized technology,
which sometimes consumes too much energy, it is fundamentally important that ecologically
correct building techniques, such as bioarchitecture, green engineering and bio-construction
are properly disseminated, for it not only generates work and citizenship but also provides the
the rupture of the professional barrier in front of the community in self-construction projects
and public-private parnerships.
Construction in land is the most ancient and most used by humans constructive technique
throughout the world. Buildings in lands are present since men felt the need to build their
shelter, instead of using only those provided by nature [1]. In locations where it was difficult
to obtain rocks, walls were built by the usage of available materials, mainly land and wood.
Engineering and technological entities are searching for clean alternatives to reverse
habitational and environmental problems. Millenarian techniques rehabilitation, such as brick
and adobe, as well as the creation of new technologies and materials are ways to continue
developing in a sustainable way.

In Brazil, there are many regions, mostly at the northern and northeastern part of the
coutnry, which show potential to use raw land as a technology to build social housing.
However, not all soils are appropriate to produce adobe bricks. Ideal granulometric
composition for the soil mixture suitable for adobe has a certain percentage of sand, clay and
silt [2]. Therefore, it is essential that the soil is analyzed throughout these field and laboratory
tests in order to verify the need or not of granulometric correction with sand. This way, the
usage of raw earth and a main input at bioalvenary construction emerges as an important
option in low-cost buildings, for it saves logistics, when the adequate land is available at the
very same building location, due to the abundance of raw material at the already mentioned
regions, as well as because it does not require qualified labour force and only consumes solar
energy in its fabrication, among other benefits.
Water consumption to its production is sixty times less than required in cement production,
does not generate contaminant vapors and does not represent any danger in its manipulation.
manipulao [3]. On the other hand, it is an artisanal process that, depending on the mixture
composition and water contact, may suffer deterioration and erosion.
Adobe developed from raw land and sun-dried offers the following advantages: economic,
since its raw material is abundant and low-cost; energetic, for it can be produced at the same
place of the building, dispensing costs with transportation and the usage of an oven, since the
land dispenses the burning of any fuel, obtaining great energy economy; social and cultural,
since adobe technologies allow a reduction of costs in social housing, which permits the user
to build their own housing; and physical performance of the material, for the housing in raw
land may provide thermal acoustic comfort. [4].
In this sense, the main objective of the present paperwork is to project a sealing
bioalvenary from raw land in order to build social interest housing.
2.
THE ADOBE BRICK
Adobe brick is a very ancient building material. It consists of a clay brick formed in a
mechanical, manual or moulded process. Its dying occurs naturally. It is a technique in
accordance to sustanability practives, for it does not requires cement and does not spend fuel
to dry the bricks, since they are not burned. Adobe buildings may last many decades when
they are properly done.
Adobe is an arab word which was assimilated by spanish and broadcasted to the Americas,
where it was adopted also by the english language, meaning raw land bricks. Its most
common usage is throughout Vernacular Architecture.
Adobe is formed by clay, silt and sand particules. Soil used to build, for it can be found in
the nature, is not padronized, therefore many combinations might be used. Depending on the
used technique, the mixture has to be dryer or more humid. Clay ought to be the mixture of its
soil with water and sometimes other certain aditives might be used in order to improve the
final product.
The correct proportion of sand and clay normally stays between 1:1 up to 2:1 [5]. Smallest
proportions leave the mass too soft and the clay starts to lose its adherence. A combination
that provides an stable and dense clay is obtained by adding grains, fibers, dry and clean
leaves to the mixture. These additives stabilize the mass and internally tie soil and sand
molecules. Another form of creating stable mixtures is to add cement, lime or ashes that ends
up providing a more stable and durable alloy. A third option is to add vegetal oils, latex, sap
or asphaltic bitumen. This way the mixture becomes more waterproof and with less water,
which makes it more resistent to weather changes. The ideal land is the one with a yellow

color, but brown and red soils may also be used to produce bricks [6]. White or black colored
lands may be discarded.
The appropriate land to be used in the building is the one located below 50cm. Above that,
on the superficial layer, vegetal land is found and it contains organic matter and
microorganisms. This layer is not suitable for building, since it becomes fragile when dried,
due to material decomposition and besides it may compromise health inside environments.
2.1
Adobe brick manufacturing
The mixture is mashed and sieved in order to acquire a very homogeneous form. The
mixture may rest in the shade for approximately two days before it is again mixed with water
to create a plastic alloy. After that, clay is taken to the forms which can be made of wood,
metal or resistant plastic. Usually, the form is twice long than its own width.
The clay, already with certain consistence, may be discharged in the form and fill every
corner equally. Alloy plasticity may be concise and some minutes after filling in the forms,
the brick taken out of it shall not look like rubber.
Brick also shall not collapse, show torn corners or sinkings in its faces. Therefore, after
being taken out of the form, it might be put to dry. Before starting this part, it is ideal to
smooth brick faces with a spatula.
Adobe brick drying process depends on the places climatic conditions. When done in a
dry and hot weather, it completes the cycle faster and better. It is also needed to evaluate the
kind of mixture and the amount of water it contains. On the first phase of this step, brick
drying occurs with direct solar radiation, for it is in this moment that most part of brick
humidity is lost throughout a natural convection process. On the rest of the time, blocks shall
rest in the shade. It is worth mentioning that during the drying process bricks shall be flipped
over in regular spaces of time, in order to homogenize drying, therefore avoiding deformation
and disproportionate retractions. Finally, brick storage shall be done in a dry place. Lead time
of that productive process may take 28 to 35 days, according to local climate conditions.
Bio-construction
3.3
Bio-construction is the nomination used to refer to buildings where the ecological concern
is present since its conception up to occupation [6]. Since the project conception, throughout
execution and finally at the utilization of the thing, Bio-construction searches for the
maximum usage of productive resources available with a mininum impact. Treatment and
reutilization of waste, rain water collection, usage of renewable and non-polluting energies,
maximum reutilization of natural illumination instead of artificial ones are examples of those
preoccupations inside the conception of these projects
Bio-construction encompasses many architecture techniques, some of them with hundreds
of years of history and experience, characterized by the preference for local materials, such as
land, diminishing costs with manufacturing and transportation and building housing with
reduced costs and which offer great thermal comfort. [7]. Usually, bio-construction
techniques are relatively simple, therefore allowing them to be easily widespread.
Classic bio-construction techniques use Land (Rammed earth, Adobe, Wattle and doub,
etc), Renewable fibers (Parda, bamboo, etc), Vegetal coverages (Natural or planted
vegetation), Eco-sanitation (Banana trees circle and evapotranspiration basins) and Mosaics
(broken tiles, floor, pastilles, glass etc).
Building with land requires technical precautions in order to avoid future problems, so the
land used to manufacture might be slightly clayey to avoid material contraction and
consequent cracking. Adobe must be seated in a rock or concrete structure, and it shall not be

put directly on the soil, therefore avoiding the increase of humidity. It is also advisable that
buildings have a salient double pitched roof, in a way to better avoid the infiltration of water
from the rain through the walls (13).
3.
METHODOLOGY
12 soils between the cities of Sum and Serra Branca, located at Parabas Cariri region,
were analysed. These cities have a dry and hot weather. In Sum, a kind of soil at Centro de
Desenvolvimento Sustentvel do Semirido (CDSA), at Universidade Federal de Campina
Grande (UFCG) as well as a soil from Stio Riacho were analysed, while at Serra Branca, 9
soils were analysed at Comunidade Ligeiro de Baixo. Soil analysis in this research was
carried out by Laboratrio de Solos e gua, from Unidade Acadmica de Engenharia
Florestal, at Centro de Sade e Tecnologia Rural da UFCG and defined their textures, that
meaning it designed the proportion related to clay, sand and silt at the soil. These are different
between one another for the size of their particules (granulometry). Soils selected as
compatible to adobes production in this study are those which can take into account the
percentages of 54% to 75% of sand and 25% to 43% of binders, that being 10% to 25% of silt
or 15% to 18% of clay [8]
Aggregates separation was carried out by the humid way, throughout the usage of a
Yoder, oscilating in a vertical way. This machine contains a set of sieves with meshes of five
different diameters (2,00, 1,00, 0,500, 0,250 and 0,106 mm).
In order to determine soil composition, the used method was the Bouyoucos one, which
consists basically of determining the concentration of clay in a suspension in which the sands
were previously separated. Silt is obtained from the difference between the total amount
(100%) and the sum of clay and sand percentages. After the determination of the composition
of those 12 analysed soils, a comparison was made between the granulometry of each of them
according to Sociedade Brasileira de Cincia do Solo (SBCS)s pyramid, which contains 13
types of textural soil classes. Results were compared to the ideal soil composition to adobe
brick buildings proposed by the specialized literature.
In the adobe brick project which will be used to build social housing it was taken into
account technical-constructive aspects such as masonry geometry, measurements (height,
lenght and width), weight and number of holes. The ideal brick must have twice the width in
lenght [8]. Brick weight simulation was based on the product between adobe bricks specific
mass and the volume of the studied prism. The specific mass used for this calculation was
1700 kg/m3. It is worth pointing that the equation used in this study is valid only for a gravity
acceleration equal to 9.80665 m/s2 (sea level). To the bricks form project there were evaluated
materials which provided lightness, ergonomy, resistance to humidity, mechanical and cost
characteristics. Complementarily, in order to optimize the production process, it was also
evaluated the amount of bricks to be moulded each time. For the adobe brick project and its
form there were utilized the softwares Autocad and Sketchup.
4.
The area under study is the semiarid part of the state of Paraba, specifically the cities of
Serra Branca and Sum, located at the microrregion of Occidental Cariri in the referred state,
located at approximately 250 and 264 km from the capital, respectively. Semiarid climate
characteristics go from big range temperature during the day to big hot air masses conducting
powder airborne particles in their displacement at the dry period [9]. This climate is also
called hot-dry and shows two well distinctive seasons: a periof of drought and another rainy

one, direct and intense radiation and low relative air humidity. Annual precipitation stands
between 250m and 1.000mm, representing a great dficit during all year. Annual medium
relative humidity is 64%. Annual medium temperature stands around 27,5C, with minimum
temperature of 18C and maximum of 33C [10].
Since the cariri region of Paraba shows a hot and dry climate, houses shall be built with
high thermal inertia materials. This means that heat penetrates during the day and is slowly
liberated at night. The interior part of the house is cooler than the exterior one during the heat
of the day and warmer than the exterior at the cold of the night. Houses built from raw land
show the particular characteristic of breathing, meaning they allow diffusion through their
walls. Raw clay, for it has not been under high temperatures for its cooking and crystallization
keeps empty spaces between its molecules, what gives the material its characteristic of
porosity and permits the air flow through it. This characteristic makes the house an example
of a biologically healthy architecture, for the breathing of the walls provides constant air
renewal.
Adobe bricks buildings may be executed on common rocks foundations, normally schist,
around 60cm from the floor, in order to avoid contact with ascending humidity (infiltration),
which would degrade adobe. It is also important to build coverage with big overhang in order
to protect the walls from rain water. Walls shall be covered for greater durability. In this
context, the kind of brick selected to build this social interest housing in this study is the one
whose application is to houses covered by tile and whose wall will sustain the structures
weight. The massive brick is part of this classification. Adobe bricks will be dimensioned
15cm x 9,5cm x 30,5cm (WxHxL).
Adobe bricks projected in this work shall be settled in form of half time masonry, which
consists of putting the elements with the bigger area face turned down, the bigger transversal
brick dimension coincidental to the wall width, making the structure more resistant and stable.
Brick weight simulation was based on the product between specific adobe brick mass and the
volume of the studied prism. Specific adobe brick mass is 1700 kg/m3. The volume of the
projected prism is 4346,25 cm3. Converting the mentioned volume to m3, this volume equals
0,00434625 m3. Therefore, the estimated weight for the projected brick equals 7,388625 kg.
The broth used for the adobe brick confection will be made of wood, in order to be light,
easy to handle and resistant, to withstand nails without showing cracks or deformities, as well
as resist humidity without deforming itself. The broth will be dimensioned 15cm x 9,5cm x
68,5cm and will shape two bricks each batch. Broth internal walls shall be smooth so the
brick will loose itself easily. In its extremity the broth will have flaps, measured 20cm x 5cm
x 2,5cm, which will make it easier to handle the broth, also facilitating the logistics of the
process.
6.
CONCLUSIONS
Adobe brick is a traditional bio-construction technique, with high sustainability potential,

of easy application in hot and dry climates and which can be executed with local material and
little specialized labour. If it is correctly executed, it may provide termal comfort, due to the
high termal inertia that the brick possesses, as well as fireproof characteristics, since the used
soil is a non-combustible and non-toxic material.
The adobe brick project developed in this study has indicated that the kind of bioalvenary
indicated for building social interest housing at the region of Serra Branca and Sum, located
at the cariri region in the state of Paraba, with hot and dry climate characteristics is that one
of a massive type, whose application is for buildings covered with tiles and whose walls will

sustain the weight of the structure, 15cm x 9,5cm x 30,5. The broth for the adobe brick
confection will be made of wood, in order to be light, easy to handle and resistant, and will
mould two adobe bricks each batch.
To fabricate social interest housing in small towns, with great land availability and a
building site that can be organized, it is possible to produce many buildings in a semi-artisanal
way, guaranteeing adobes advantages and a greater standardisation (and consequente quality
control). A initiative of this kind could provide local housing for local population with a
higher quality than what is done nowadays by the civil construction industry.
REFERENCES
[1] Soares, Antnio F. [et. al] - O norte-lusitano Sub-bacia (Portugal) a partir do Trissico ao
Jurssico Mdio.: um modelo de um "rift abortado". "Rendus de l'Acadmie des Sciences". ISSN
0764-4450. 317: 12 (1993) 1659 1666.
[2] Perazzo, N.B. Curso de construo em terra crua, Construo de adobe. Curso apresentado a
Especializao em Tecnologia do Ambiente Construdo. CEFETCE, Fortaleza, 2006.
[3] Corra, A. A. R. et al.; Avaliao das propriedades fsicas e mecnicas do adobe (tijolo de terra
crua). Lavras MG, junho/2006.
[4] Ferreira, Saulo Rocha; Cabral da Silva, Adeildo. Construes Sustentveis: Avaliao das
Caractersticas Fsico-Mecnicas de Adobe com Adio de Fibras Vegetais do Coco Verde. In:
Anais do V Encontro Nacional e III Encontro Latino-Americano Sobre Edificaes e
Comunidades Sustentveis; 2009; Recife, BR.
[5] Minke, G. Manual de Construccin en tierra: la tierra como material de construccin y sus
aplicaciones en la arquitectura actual.Montevido: Nordan- comunidad, 2010.
[6] Cantarino,
Carol. Bioconstruo
combina
tcnicas
milenares
com
inovaes
tecnolgicas. Inovao Uniemp. 2006, vol.2, n.5, pp. 46-47. ISSN 1808-2394.
[7] Soares, A. L. J. Conceitos bsicos sobre permacultura - Braslia : MA/SDR/PNFC, 1998. 53 p.
[8] Mchenry, Paul Graham. Adobe and rammed earth buildings: design and construction. Tucson:
University of Arizona Press, 1989.
[9] Romero, M. A. B. Princpios Bioclimticos Para o Desenho Urbano. So Paulo: Projeto, 1988.
[10] Bueno, Mariano. O Grande Livro da Casa Saudvel. Traduo: Jos Luiz da Silva. So Paulo:
Roca, 1995.

Rilem Pro

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Rilem Pro

Uploaded by

Copyright:

Available Formats

PROCEEDINGS

03-05 August 2015 So Paulo, Brazil

III Progress of Recycling in the Built Environment

III Progress of Recycling in the Built Environment

INFLUENCE OF THE USE OF RECYCLED PRECAST DEBRIS IN THE

GRAVITY CONCENTRATION OF CONSTRUCTION AND DEMOLITION WASTE

GRAVITY CONCENTRATION OF CONSTRUCTION AND DEMOLITION WASTE

SELECTIVE COMMINUTION OF CONCRETE WASTE AND RECOVERY OF THE

RESEARCH AND MODIFICATION ON RECYCLED AGGREGATES

OUTLINE OF AIJ RECOMMENDATION FOR MIX DESIGN, PRODUCTION AND

CHANGES OF THE MATERIAL PROPERTIES BY SELECTIVE CARBONATION

THE ECONOMIC RISK ANALYSIS TECHNIQUE APPLIED TO DIFFERENT

RATIONAL MIX DESIGN APPROACH FOR RECYCLED AGGREGATE

III Progress of Recycling in the Built Environment

PUTTING RECYCLED CONCRETE INTO PRACTICE: LESSONS LEARNED

CHEMICAL-MINERALOGICAL SORTING OF MINERAL CONSTRUCTION AND

USE OF FINE RECYCLED AGGREGATE AS SUPPLEMENTARY CEMENTITIOUS

IMPROVING THE QUALITY OF RECYCLED AGGREGATES BY SORTING IN

THE INFLUENCE OF USING A SUPER PLASTICIZER IN RECYCLED GYPSUM

TREATMENT OF RESIDUAL DIATOMITE EARTH FOR USE AS POZZOLAN IN

RELEASE OF DANGEROUS SUBSTANCES FROM CONSTRUCTION AND

RECYCLED FINE AGGREGATE IN NEW CONCRETE

STUDY ON THE RECYCLING TECHNIQUE OF PVC BUILDING WASTES

A STUDY ON THE ACTUAL STATE FROM GENERATION OF CONSTRUCTION

RECYCLED CONCRETE AGGREGATES FROM DIFFERENT SOURCES:

III Progress of Recycling in the Built Environment

EVALUATION OF MORTAR BASED ON ELECTRIC MEASUREMENTS: EFFECT

RECOVERY OF PACKAGING GLASS REFUSED BY MECHANICAL

ALTERNATIVES FOR THE PRODUCTION OF STRUCTURAL CONCRETE WITH

TECHNICAL AND ENVIRONMENTAL STUDY OF CEMENT TREATED

SECONDARY BUILDING MATERIALS FROM CONSTRUCTION AND

FATIGUE FAILURE OF RECYCLED AGGREGATE CONCRETE UNDER

EFFECT OF SUPERPLASTICIZER ON MORTARS WITH RECYCLED FINE

PRELIMINARY CHARACTERIZATION OF EXCAVATED SOILS

III Progress of Recycling in the Built Environment

MECHANICAL PROPERTIES OF CONCRETE WITH RECYCLED

III Progress of Recycling in the Built Environment

III Progress of Recycling in the Built Environment

III Progress of Recycling in the Built Environment

RESULTS AND DISCUSSION

Recycled Aggregates (%)

Compressive Strength (MPa)

Compressive Strength (MPa)

Glass Size (mm)

% Recycled Aggregates 28 days

III Progress of Recycling in the Built Environment

Table 2: ANOVA results for compressive strength.

Glass Size (mm)

Flexural Strength (MPa)

% Glass Powder 28 days

% Recycled Aggregates 28 days

Glass Size 28 days

III Progress of Recycling in the Built Environment

Table 3: ANOVA results for flexural strength.

Sum of squares (S)

III Progress of Recycling in the Built Environment

[6] Gonzalez, A., andEtxeberria, M. Experimental analysis of properties of high performance

III Progress of Recycling in the Built Environment

III Progress of Recycling in the Built Environment

INFLUENCE OF THE USE OF RECYCLED PRECAST DEBRIS IN THE

III Progress of Recycling in the Built Environment

Natural and recycled aggregates

Natural and recycled cement

Pozzolanic cement, equivalent to ASTM type P cement is used. Different percentages of