ACKNOWLEDGEMENT

A project is always a coordinated and scheduled team effort, but it can never reach
completion without proper guidence and encouragement.
Words are short for expressing our deepest and sincere thanks towards our project
guide S.D. P.V. MANE Bharati Vidya peeth Deemed University College of
Pune.
We are thankfull to the principal Dr. A.R. Bhalerao of Bharati Vidya peeth
Deemed University College of Engineering, Pune for graniting us a wonderful
opportunity.
We are also thankful to our principal Shri. S.V. Andhare & Head of the
Department Prof. (Mrs.) V.S. Sohoni ( Head of Civil Department) for their
support that has being given to us in the form of Infrastructure and Facilities.
Last but not. We must thank all the other teaching and non-teaching staff of civil
department for their assistance.

SR NO.

NAME

PAWAR OMEAR ZULFIKAR

TANWAR MOHD FAWAD MOHD SAKIL

EFFECT OF CLIMATE ON CONCRETE

Name:- PAWAR OMEAR ZULFIKAR
Membership no:- 84545
Course:- T.Engg. Civil Part II
AICTE Institution:
Bharati Vidyapeeth Deemed University College
Engineering, Pune 411043 (Maharashtra)

SR NO.

NAME

PAWAR
ZULFIKAR

MEMBERSHIP
NO.
OMEAR 84545

COURSE
T.ENGG. CIVIL
PART II

Academic Year :- 2015-2016

DEPARTMENT OF CIVIL ENGINEERING
Bharati Vidyapeeth Deemed University College
Engineering, Pune 411043 (Maharashtra)

EFFECT OF CLIMATE ON CONCRETE

A Project submitted in partial fulfillment of
T.Engg. Civil Part II in AMICE
Diploma in Civil Engineering
PAWAR OMEAR ZULFIKAR
By
Membership no. 84545
&
TANWAR MOHD FAWAD MOHD SAKIL
Membership no. 84208
Under the Guidance of
S.D. P.V. MANE
BVDUCOEP PUNE-43

The Institution of Civil Engineering (India)

2015-2016

Academic Year :- 2015-2016

DEPARTMENT OF CIVIL ENGINEERING

Bharati Vidyapeeth Deemed University College

Engineering, Pune 411043 (Maharashtra)
CERTIFICATE
This is to certify that the Project entitled Comparative Study Of Construction
Work Between Local Contractor And Qualified Contractor .
The modified work done under by guidance in partial fulfillment of the
students for the award of T.Engg. Civil Part II in Civil Engineering for
Academic Year 2015-2016.
By
PAWAR OMEAR ZULFIKAR
Project guide

Prof. S.D.

84545
Prof. (Mrs) V.S.Sohoni

( Head of Civil Department)

EFFECT OF CLIMATE ON CONCRET

Reinforced concrete more liable to damage under
climate change
Concrete structures, such as buildings, bridges and
harbours, reinforced with internal metal bars for added
strength are an integral part of social and economic
activities in modern societies, with some infrastructures
built to last hundreds of years. However, it is thought that
higher atmospheric CO2 and rising temperatures
projected under a changing climate could increase the
rate of corrosion of the reinforcing metal resulting in
serious cracking of reinforced concrete structures.
Weakened infrastructure will cost more to repair and
disrupt the use of facilities in the future. This study
investigated climate change impacts on the risk of
corrosive damage to concrete structures over 100 years,
from 2000 to 2100, by modelling changes in CO2
concentration, temperature and humidity in two
Australian cities: Sydney (representative of a temperate
climate) and Darwin (representative of a tropical climate).
Three scenarios were compared with a baseline of
keeping CO2 concentrations at 2000 levels: 1) high CO2
emissions, 2) medium CO2 emissions and 3) reduced CO2
emissions brought about by policy measures to mitigate
climate change, but still higher than baseline levels.
Concrete structures were located in different types of
sites exposed to a range of water impacts, for example,
submerged, in a tidal zone and dry inland. Corrosion

impacts were modelled from two sources: exposure to

CO2 in the atmosphere (carbonation) and exposure to
chloride, from salty water and air. It was found that
reinforced concrete structures were more susceptible to
corrosion by carbonation, affected by increased CO2
levels in the atmosphere, than by chloride-induced
corrosion under all three future scenarios. By 2100, it is
likely that 20-40 per cent of all concrete infrastructure, in
these two cities representing two climate types, will be
damaged, requiring maintenance or repairs. For the worst
case scenario, with the greatest CO2 emissions,
carbonation damage is up to 460 per cent higher than for
the base case (where emissions remain the same as year
2000 levels) for dry inland regions or temperate climates.
Structures here would need extra attention to adapt to
the more damaging environment. Although higher
temperatures under climate change will increase
chlorine-induced corrosion under all emission scenarios,
the risk of corrosive damage will increase by a maximum
of 15 per cent under the worst case scenario, compared
with the base case scenario. Nevertheless the risk of
corrosion is already high for marine structures located in
tidal splash areas, so these structures especially require
further protective measures to adapt to the extra risk of
damage caused by climate change. In planning future
infrastructure using reinforced concrete, the costs
associated with adaptations (such as extra concrete cover
or special coatings) to mitigate anticipated increases in
corrosion damage should be factored into the design of
the structures.

IMPACT OF CLIMATE CHANGE ON CORROSION AND

DAMAGE RISKS TO CONCRETE INFRASTRUCTURE
Keywords: Corrosion, climate change, risk assessment,
deterioration, carbonation
Abstract.
Increases in atmospheric CO2 concentrations, and
changes in temperature and humidity due to a changing
climate will, especially in the longer term, cause an
acceleration of deterioration processes and consequently
acceleration in the decline of the safety, serviceability
and durability of concrete infrastructure. An investigation
of concrete carbonation-induced deterioration in typical
Australian and Chinese cities under a changing climate is
described in this paper. It is based on Monte-Carlo
simulation analysis that involves three emission
scenarios, i.e. A1B, A1FI and 550 ppm stabilisation. The
probabilistic analysis included the uncertainty of climate
predictions, deterioration processes, material properties,
dimensions, and predictive models. Deterioration of

concrete structures is represented by the probability of

reinforcement corrosion initiation and corrosion induced
damage at a given calendar year between 2010 and
2100, and all of them are affected by the changing
climate depending on locations. It was found that
carbonation depths may increase by more than 45% for
inland locations in Australia. It was also found that
carbonation-induced damage risks can increase threefold
by 2100 to 2% for Canberra. The findings provide a basis
for the development of climate adaptation strategies
through the improved design of concrete structures.

INTRODUCTION
Concrete is the predominant construction type used in
critical infrastructure in many countries. The deterioration
rate of such structures depends not only on the
construction processes employed and the composition of
the materials used but also on the environment.
Increases in atmospheric CO2 concentrations, and
changes in temperature and humidity due to a changing
climate will, especially in the longer term, cause an
acceleration of deterioration processes and consequently
acceleration in the decline of the safety, serviceability
and durability of concrete infrastructure. The
Intergovernmental Panel on Climate Change fourth
assessment report (IPCC 2007) indicated a significant
increase of CO2 concentration in the atmosphere from
280 ppm in 1750 to 380 ppm in 2005 with an increasing

trend. In comparison with pre-industrial temperatures, the

best estimation of the temperature increase from 1990
caused by increasing atmospheric CO2 concentration can
be Professor and Director, Centre for Infrastructure
Performance and Reliability, The University of Newcastle,
NSW, 2308, Australia PhD Candidate, Centre for
Infrastructure Performance and Reliability, The University
of Newcastle, NSW, 2308, Australia CSIRO Climate
Adaptation Flagship and CSIRO Ecosystem Sciences,
Highett, Vic, Australia.

Mark G. Stewart, Lizhengli Peng and Xiaoming Wang

2.1C for 550 ppm CO2, 3.0C for 700 ppm, and 4.4C for
1000 ppm CO2 by 2100 (IPCC 2007). Increased
temperature will increase rates of carbonation and
chloride penetration, as well as corrosion rate. An
increase in temperature will increase the rate of
infiltration of deleterious substances (increased material
diffusivity) and increase the corrosion rate of steel. For
example, corrosion rates will increase by up to 15% if
temperature increases by 2o C (Stewart et al. 2011,
2012). Stewart et al. (2011, 2012) have developed
advanced probabilistic and reliability-based methods to

predict concrete deterioration under a changing climate

in Australia, in terms of changes in probability of
reinforcement corrosion initiation and corrosion induced
damage due to (i) increase in the concentration of CO2 in
the atmosphere, and changes to temperature. These time
and spatial variables will affect the penetration of
aggressive agents CO2 and chlorides into concrete, and
the corrosion rate once corrosion initiation occurs
(Stewart et al. 2011, 2012). It has been shown by Wang
et al. (2010c) that additional carbonation-induced
damage risks for the A1FI emission scenario is up to 16%
higher if there are no changes to how concrete structures
are designed or constructed. In practical terms, this is
equivalent to expecting that an additional 16% of all
concrete surfaces by the year 2100 will be damaged and
in need of costly maintenance or repair. While there is
much research on deterioration of concrete studies, there
is relatively little research on how deterioration is
affected by a changing climate. Stewart and Peng (2010)
used a simplified deterioration model and global IPCC
(2007) CO2 concentration and temperature change data
to conduct a life-cycle cost assessment to assess the
cost-effectiveness of increasing design cover as an
adaptation measure to mitigate the effects of carbonation
of concrete. This preliminary analysis found that
increasing design cover may not be cost-effective, but
results were sensitive to repair costs and user disruption.
Stewart et al. (2011) developed a more accurate
formulation of carbonation and chloride-induced corrosion
that predicted the probability of corrosion initiation and

damage (severe cracking) for concrete infrastructure for

the Australian cities of Sydney and Canberra. It was found
that carbonation-induced damage risks increases to 20%
to 40% for A1FI, A1B and 550 ppm emission scenarios.
The additional damage risks for chloride-induced
corrosion is only 3% over the same time period due to
temperature increase, but without consideration of ocean
acidity change in marine exposure. However, these
models still relied on relatively straightforward timedependent deterioration models, and ignored the effect of
changes in humidity on the deterioration process. The
effect of climate change on chloride-induced corrosion
has also been the subject of relatively little research,
however, Bastidas-Arteaga et al. (2010) have calculated
5-15% increases in probability of corrosion initiation due
to climate change. Talukdar et al. (2012) have predicted
carbonation depths in Canada for climate change
scenarios, but did so using a deterministic model and
assuming constant RH. An investigation of concrete
carbonation-induced deterioration in typical Australian
and Chinese cities under a changing climate is described
in this paper. Improved deterioration models are used
that accurately predict carbonation depth when CO2
levels, temperature and humidity are timedependent
variables. It is based on Monte-Carlo simulation analysis
that involves three emission scenarios, i.e. A1B, A1FI and
550 ppm stabilisation, representing medium, high and
policy-intervened GHG emission scenarios. The
probabilistic analysis included the time-dependent
changes in atmospheric CO2 concentration, temperature

and humidity, and the uncertainty of climate predictions,

deterioration processes, material properties, dimensions,
and predictive models. Deterioration of concrete
structures is represented by the probability of
reinforcement corrosion initiation and corrosion induced
damage at a given calendar year between 2010 and
2100, and all of them are more or less affected by the
changing climate depending on locations. This paper
seeks to provide insights into the likely impacts of climate
change on the durability and damage risks of concrete
structures in Australia and China, under a range of
climate change scenarios, which will have similar
implications for other countries. The findings from the
investigation provide a basis for the development of
climate adaptation through the improved design of
concrete structures. Since the main environmental driver
to increased concrete deterioration is CO2 concentration,
temperature and humidity, then this will affect all
concrete infrastructure globally, not just Australia or
China. While the present study focuses on concrete
infrastructure, changes in temperature and relative
humidity will also affect the corrosion of steel structures,
but these effects are beyond the scope of the present
paper.
2 PROBABILISTIC MODELLING OF CARBONATIONINDUCED DAMAGE RISKS
2.1 Anthropogenic Aspects of Climate Change Future
climate was projected by defining carbon emission
scenarios in relation to changes in population, economy,

technology, energy, land use and agriculture, represented

by a total of four scenario families, i.e., A1, A2, B1 and B2
(IPCC 2007). Sub-categories of the A1 scenario included
differing energy sources (fossil intensive, non-fossil
energy and a balance across all sources, for example,
A1FI, A1T and A1B, respectively). In addition, scenarios of
CO2 stabilisation at 450 and 550 ppm by 2150 were also
introduced to consider the effect of policy intervention
(Wigley 1996). Our entire investigation will cover A1FI,
A1B and 550 ppm stabilisation scenarios representing
high, medium emission scenarios and policy intervention
scenarios, respectively. An emission scenario based on
Year 2000 CO2 levels is also considered to provide a
reference for other emission scenarios. Figure 1 describes
the projection of CO2 concentrations from 2000 based on
the Model for Assessment of Greenhouse-gas Induced
Climate Change, known as MAGICC (Wigley et al. 1996),
specifically related to A1FI, A1B and 550 ppm CO2
stabilisation scenarios. Their low and upper bounds are
also described to consider CO2 projection modelling
errors. If low and high values shown in Figure 1 are taken
as 10th and 90th percentiles, respectively, of a normal
distribution then the statistical parameters for CO2
concentrations are: mean CO2(t) is equal to mid value,
and standard deviation CO2(t) is (high-low)/2.56. The
COV increases with time to a maximum value of
approximately 0.06 for all emission scenarios. For the
reference (best) emission scenario based on constant
year 2000 CO2 concentration then CO2(t) = 369.2 ppm
and CO2(t) = 0. In all cases the probability distributions

are censored at year 2000 CO2 concentration. To project

spatially dependent temperature increases in the future
under different emission scenarios, various climate
models or Atmosphere-Ocean General Circulation Models
(AOGCMs) have been developed based on physical
principles at the continental scale. Selecting an AOGCM to
be used in an impact assessment is not a trivial task,
given the variety of models. The IPCC suggested that due
to the varying sets of strengths and weaknesses of
various AOGCMs, no single model can be considered the
best. Therefore, it is necessary to use multiple models to
take into account the uncertainties of models in any
impact assessment. In the current study, climate
projections from nine climate models are used - for more
details see Wang et al. (2010) and IPCC (2007). From the
global CO2 concentration and temperature rise
projections obtained by MAGICC for a given emission
scenario, temperature and relative humidity changes in
Australia with different GCMs can be projected by OZClim.
OZClim is a climate change projection software developed
by the Commonwealth Scientific and Industrial Research
Organisation (CSIRO), which is based on the WCRP CMIP3
multimodel dataset developed by the Program for Climate
Model Diagnosis and Intercomparison (PCMDI), and the
Working Group on Coupled Modelling (WGCM) of World
Climate Research Programme (WCRP). A site-specific
correction factor (kmid) is applied to the mid global
temperature increase data for the IPCC emission
scenarios (Tmid) such that T(t) =Tref + kmidTmid(t)
where Tref is the reference temperature at year 2000. For

example, Figure 2 shows the projected median

temperatures for the lowest and highest of the nine GCM
projections, for A1FI, A1B, 550 ppm and Year 2000
emission scenarios for Sydney, Canberra and Xiamen.
The mid global temperature Tmid(t) is modelled as a
normally distributed random variable where the standard
deviation T(t) is (high-low)/2.56. The Coefficient of
Variation (COV) increases with time from 0.3 for t=2000
to a maximum value of approximately 0.4 to 0.5 for all
emission scenarios. In all cases the probability
distributions of Tmid(t) are censored at zero temperature
rise.

2.2 Time to Corrosion Initiation Carbonation depth

depends on many parameters: concrete quality, concrete
cover, relative humidity, ambient carbon dioxide
concentration and others. The impact of carbonation has
been studied by many researchers and various
mathematical models have been developed with the
purpose of predicting carbonation depths (for review see
e.g., Duracrete 1998, Stewart et al. 2002). It is observed
that corrosion may occur when the distance between the
carbonation front and the reinforcement bar surface is
less than 1-5 mm (e.g., Yoon et al. 2007). However,
probabilistic analyses for assessing durability design
specifications tend to ignore this effect (Duracrete 2000b,
fib 2006). Hence, time to corrosion initiation (Ti ) occurs
when carbonation front equals concrete cover. The
carbonation depth model recommended by Duracrete

(1998), Yoon et al. (2007) and others considers a wide

range of influencing parameters and so the carbonation
depth (xc in cm) is predicted as a diffusion process, as (1)
(2) where t defined in years starting from 2000, CCO2 is
the mass concentration of ambient CO2 (10- 3 kg/m3 )
with CO2(t) and COV equal to CO2(t)/CO2(t) obtained
from Figure 1 (using the conversion factor 1 ppm =
0.001910-3 kg/m3 ); DCO2 (t) is CO2 diffusion
coefficient in concrete; D1 is CO2 diffusion coefficient
after one year; nd is the age factor for the CO2 diffusion
coefficient; t0 is one year; Ce is cement content (kg/m3 );
CaO is CaO content in cement (0.65); H is a degree of
hydration; MCaO is molar mass of CaO and equal to 56
g/mol and MCO2 is molar mass of CO2 equal to 44 g/mol.
The age factor for microclimatic conditions (nm)
associated with the frequency of wetting and drying
cycles is nm=0 for sheltered outdoor and nm=0.12 for
unsheltered outdoor. The mean values for D1 and nd are
given in Table 1, and Yoon et al (2007) provided estimates
of maximum (95th percentile) values for D1 and nd. The
standard deviation for D1 is approximately 0.15, and COV
for nd is approximately 0.12 for all w/c ratios. These
statistics represent model error (or accuracy). The
diffusion coefficient D1 is less than 510-4 cm2 s-1 and
which is appropriate for good quality concrete. These
parameters are based on T=20 oC and RH=65%. w/c
D110-4 cm2 s-1 nd 0.45 0.65 0.218 0.50 1.24 0.235
0.55 2.22 0.240 Table 1: Mean Parameter Values (Yoon et
al. 2007) The degree of hydration for Ordinary Portland
Cement (OPC) after more than 400 days is estimated as

(de Larrard 1999): (3) Four new variables are now

included: 1. CO2(t) = time-dependent increase in
atmospheric CO2 concentration, 2. ksite = factor to
account for increased CO2 levels in non-remote
environments, 3. fT(t) = time-dependent change in
diffusion coefficient due to changes in temperature, and
4. fRH(t) = time-dependent change in diffusion coefficient
due to changes in relative humidity. A number of studies
have shown elevated CO2 levels in urban environments
due to higher pollution, exhaust fumes, etc. Stewart et al.
(2002) recorded CO2 concentrations of up to 575 ppm in
Brno (Czech Republic), with CO2 concentrations higher
near street level. George et al. (2007) found that CO2
concentrations in an urban site (Baltimore) were on
average 16% higher than a rural site, and increases of
21-31% were reported in the literature. Day et al. (2002)
observed an average enhancement over the course of the
day in CO2 concentration near an urban centre (Phoenix)
of 19 ppm. Moreover, the CO2 concentration predicted by
the IPCC is the global mean value which is collected at
remote marine sea level locations, such as Mauna Loa in
Hawaii and Wisconsin in the Great Lakes region. In that
case the CO2 level in urban, suburban and rural locations
is usually higher
than the global mean value because of the anthropogenic
and natural sources and carbon sinks. Therefore, a new
factor ksite is introduced as (4) Recorded CO2
measurements were obtained for rural, suburban and
urban locations in U.S., U.K., China, Italy, Poland, Japan,

and Kuwait and values of ksite calculated for these three

categories. As most infrastructure are close to the
ground, height impact has been taken into consideration
when observation height is larger than 30 m and so a
multiplication factor of 1.007 is applied to observation
values made at height larger than 30 m. Since the size of
every city is different, the definitions of rural, suburban
and urban areas depend on the size of the city. In order to
make ksite comparable, the definition of rural, suburban
and urban are given as follows. Table 2 defines an urban
area based on population size and distance from the
central business district (CBD) or downtown area. If the
population of the city is less than 0.5 million then the city
is considered as suburban. A rural area is defined as the
area outside of cities and towns, and typically much of
the land is devoted to agriculture. An area that is neither
rural nor urban is defined as suburban. Table 3 shows the
statistical parameters for ksite obtained from the data. As
expected, ksite increases for urban areas, most likely due
to higher pollution levels and urban domes. Population
(million) 0.5 - 1 1-2 2-5 5-10to 2.5. The RH factor fRH
exceeds one when RH is less than 65%, and reduces to
zero when RH is equal to 100%. As presented by
Richardson (1988) reports that insufficient water is
available for carbonation to commence for a relative
humidity of 25% and less; hence, a lower limit of
RH=25% is set. Carbonation tends to be highest for
relative humidities RH(t) of 50% to 70% (Russell et al.
2001) which is reflected in Eqn. (6). While Eqn. (1) was
used by Yoon et al. (2007) to predict carbonation depths

for increases in CO2 concentrations it needs to be

recognised that Eqn. (1) is a point-in-time predictive
model - i.e., the carbonation depth at time t assumes that
CO2 is constant, and assumes that D is constant for all
times up to time t. Stewart et al. (2011) considered this
phenomenon and calculated carbonation depths due to
enhanced atmospheric CO2 concentration conditions
using the average CO2 concentration over the time
period, and not the peak value at time t. If we consider
the carbonation process as a steady state modelled by
Ficks First Law, where CO2 concentration and
temperature and RH correction factors are timedependent, then (7) 2.3 Corrosion Propagation The
carbonation-induced corrosion rate is variable and highly
dependent on exposure conditions and atmospheric
situations. Some empirical models exist, but each isolates
only several variables. It is interesting to note that when
water or oxygen is limited, the corrosion process may
eventually stop. In fact, very high humidity in concrete
may reduce oxygen diffusion to the corrosion area and
slow the corrosion process of reinforcement, while the
shortage of water in dry concrete also reduces corrosion
activities. The corrosion rate for carbonation or chlorides
becomes negligible when relative humidity RH(t) is less
than 50%, and in this study a negligible corrosion rate is
defined as a corrosion current density (icorr) of 0.1
A/cm2 where a corrosion rate (icorr) of 1 A/cm2 =
0.0116 mm/year. The optimum relative humidity for
corrosion is 70-80% (Neville 1995). In the present study,
corrosion rate is assumed lognormally distributed with

statistical parameters for a temperature of 20 oC given

by Duracrete (1998), see Table 4. These values take into
account the concrete grades suggested for the
corresponding exposure classes. An increase in
temperature will increase corrosion rate, and the model
described by Duracrete (2000) is used: (8) where icorr-20
is the corrosion rate at 20 oC given in Table 4, and
K=0.025 if T(t)20 oC. Duracrete (2000) notes that Eqn.
(6) is a close correlation to Arrhenius equation, at least
for temperature below 20 o C, but may be conservative
for T(t)>20 oC. A 2o C temperature increase will increase
the corrosion rate by 15%. There is little data on timedependent effects on corrosion rate for carbonated RC
structures. Hence, the present analysis assumes a timeinvariant corrosion rate for carbonation. This is likely to
be a conservative assumption as corrosion rate will
generally decrease with time due to the build up of rust
products thus impeding the corrosion process (e.g., Vu
and Stewart 2000). Exposure Class Mean Standard
Deviation Distribution C1 - Dry 0.0a 0.0 Lognormal C2 Wet- rarely dry (unsheltered) 0.345 A/cm2 0.259
A/cm2 Lognormal C3 - Moderate humidity (sheltered)
0.172 A/cm2 0.086 A/cm2 Lognormal C4 -Cyclic wetdry (unsheltered) 0.431 A/cm2 0.259 A/cm2 Lognormal
a assume negligible = 0.1 A/cm2 Table 4: Carbonation
Corrosion Rates (icorr-20) for Various Exposures
(Duracrete 1998). 2.4 Time to Crack Initiation (T1st) As
there is a porous zone around the steel reinforcing bar
the corrosion products must firstly fill this porous zone
before the products start to induce internal pressure on

the surrounding concrete. Therefore, not all corrosion

products contribute to the expansive pressure on the
concrete. This approach to crack initiation has been used
by El Maaddawy and Souki (2007) and their model is used
herein. The thickness of the porous zone (0) is typically
in the range of 10 - 20 mm and can be described using a
normal distribution with mean equal to 15 m and COV of
0.1. It should be noted, that the accuracy of the time to
severe cracking is dominated by the accuracy of time to
corrosion initiation (ti ) and the time since crack initiation
to reach a limit crack width (tsev), and so service life
predictions are relatively insensitive to the crack initiation
model (Stewart and Mullard 2007). 2.5 Time to Severe
Cracking (Tsev) The time to severe cracking referred to
herein is the time when concrete cover cracking reaches
a limit crack width of 1 mm. Mullard and Stewart (2011)
have modelled rate of crack propagation which enables
the time for a crack to develop from crack initiation to a
limit crack width (Tsev). The time (after crack initiation)
for cand where icorr-20 is the corrosion current density
(A/cm2 ) at T=20 oC and constant with time, cp is the
cover cracking parameter, rcrack is the rate of crack
propagation in mm/hr, MErcrack is crack propagation
model error, w is the crack width (mm), concrete cover is
in mm, D is reinforcing bar diameter in mm, ft is the
concrete tensile strength in MPa, kR is a rate of loading
correction factor where icorr(exp)=100 A/cm2 is the
accelerated corrosion rate used to derive rcrack, and kc is
the confinement factor that represents an increase in
crack propagation due to the lack of concrete

confinement around external reinforcing bars. If the

reinforcing bar is in an internal location then kc=1, but
for rebars located at edges and corners of RC structures
then kc is in the range of 1.2 to 1.4. Although the data is
limited, there appears to be a trend where kc increases
as cp increases. In this study kc is taken as 1.0. A
statistical analysis of model accuracy to account for
variabilities between model prediction and experimental
data is essential for stochastic or reliability analyses
where statistics for model error are required. Hence, the
statistics for model error for rcrack (MErcrack) are:
mean(MErcrack) = 1.04 and COV(MErcrack) = 0.09
(Mullard and Stewart 2011). For more details of this
improved cover cracking model see Mullard and Stewart
(2011). The cover cracking model developed by Mullard
and Stewart (2011) was based on chloride-induced
corrosion. Concrete strength is time-variant, and the timedependent increase in concrete compressive strength
after one year using the ACI method is fc=1.162fc(28)
where fc(28) is the 28 day compressive strength. Timedependent gains in strength beyond one year are not
considered in the present analysis. 2.6 Time to Corrosion
Damage (Tsev) Since corrosion rate is a time-dependent
function of temperature then times to corrosion damage
need to be corrected since Eqns. (9-12) assume a timeinvariant (constant) corrosion rate. If we assume that the
amount of corrosion products needed to cause cracking
(mcorr) for a constant corrosion rate is directly
proportional to icorr(T1st+Tsev) then
mcorr=icorr(T1st+Tsev). The time to corrosion damage

for a variable corrosion rate (Tsp) is such that the

corrosion amounts (mcorr) for constant and variable
corrosion rates are equal. It follows that Tsp is obtained
from solving the unknown Tsp from the following
equation: (13racking of the concrete surface to reach a
crack width of w mm is:
2.6 Probability of Time to Corrosion Initiation
Corrosion will take place when the carbonation depth
reaches the surface of the reinforcing bar, and so the
cumulative probability of corrosion initiation at time t is
(14) where xc(t) is the carbonation depth obtained from
Eqn. (7).

Probability of Time to Corrosion Damage

Corrosion damage is defined as the time when concrete
cover severely cracks (crack width w=1.0 mm).
Therefore, the cumulative probability of corrosion damage
at time t is (15) where Tsp is the time to severe cracking
and spalling equal to T1st+Tsev. If the material,
dimensional and corrosion parameters are assumed
homogeneous for the structure (i.e. spatial variability is
ignored), then the mean proportion of corrosion damage
is ps(t)100% (Sudret et al. 2007).

Computational Method
Monte-Carlo simulation is used as a computational
method for the time-dependent reliability analysis. Note
that the CO2 concentration is fully correlated with time. 3
RESULTS 3.1 Durability Design Specifications in Australia
and China Environmental exposure in Australia is
classified by the Australian Concrete Structures Code
AS3600-2009 as three climatic zones (arid, temperate
and tropical), see Figure 4. The selected sites of Sydney
and Canberra represent two very different durability
design requirements with design cover for many locations
in Sydney being 40-50 mm due to its coastal location,
and 30 mm cover for Canberra due to its inland location .

Climatic Zones Defined by the Australian Concrete

CodeAS3600.
Concrete inside buildings with low air humidity, or that is
permanently submerged in water, generally has a low
exposure to carbonation, while concrete surfaces subject
to long-term periodic water contact, concrete inside
buildings with moderate-to-high air humidity, and

external concrete sheltered from rain have high

exposures to carbonation. For this reason the reliability
analyses to follow will focus on corrosion predictions for
sheltered structures for carbonation. The durability
design requirements specified in AS3600 relate to
minimum concrete cover and concrete compressive
strength, and assume standard formwork and
compaction. Table 5 shows the durability design
specifications related to the AS3600 exposure
classifications A1 to C for carbonation. Table 6 presents
the Chinese durability design requirements for bridges
(JTG.D62- 2004). Clearly, the environmental categories
are defined differently, however, the durability
requirements for bridges in Chinese standard are
generally the same as Australian standards, but the
requirements for normal buildings (GB50010-2010 2010)
are lower than that for bridges. Table 5 also shows the
parameter values for the deterioration models. The
exposure classifications of most relevance for concrete
infrastructure in Sydney and Canberra are near-coastal
(B1) and coastal excluding tidal and splash zones (B2) for
Sydney, and A2 for Canberra.
Table 6: Chinese Durability Design Specifications
(JTG.D62-2004) and Deterioration Models, for Carbonation
in Sheltered Conditions, and for Bridge Structures (Piers,
Earth Retaining Structures, Culverts, Beams, Plates,
Arches, Spandrel Structures). Table 7 shows the statistical
parameters for corrosion parameters, material properties
and dimensions - these are representative of concrete

structures in Australia. Clearly, the uncertainty and

variability of deterioration parameters is considerable.
However, improved deterioration modelling may reduce
this variability, as could Bayesian updating based on
conditions or other site specific data for existing or new
structures. Unless noted otherwise, all results in the
following sections refer to the average of nine GCM
temperature simulations. Reinforcement bar diameter is
20 mm, and ksite is based on statistical parameters for
an urban environment for Sydney and Xiamen, and
suburban environment for Canberra . Note that the same
value of ksite applies for all emission scenarios including
the year 2000 scenario.
Statistical Parameters for Corrosion Parameters, Material
Properties and Dimensions The impact assessment is
focused primarily on the relative change in corrosion
initiation and damage risks due to enhanced CO2 levels,
temperature and humidity when compared to year 2000
levels, and not on the absolute estimates of risk. The
deterioration models are mostly derived from the 19961999 European Duracrete project which has formed the
basis for the probabilistic durability design of many
important structures and the fib model code for service
life design (fib 2006). However, many other deterioration
models have been developed for concrete durability,
which if deemed more appropriate, can readily be
incorporated into the stochastic and reliability framework
developed in the present study. While different
deterioration models will produce different estimates of

absolute risk, deterioration model selection will have

significantly less influence on comparative risks. 3.2
Corrosion Damage Risks Figure 5 shows the mean
carbonation depth for four emission scenarios and
exposure classifications, for Sydney, Canberra and
Xiamen. The Australian Concrete Structures Code
AS3600- 2009 (AS3600 2009) specifies improved
concrete compressive strength and other enhanced
durability design specifications, which will result in a
reduced rate of carbonation. This is evident in Figure 5(a)
where exposure classification C with w/c=0.40 concrete
has a carbonation depth significantly less than exposure
classification A2 with w/c=0.56. The A1FI, A1B and 550
ppm emission scenarios have a significant effect on
carbonation depths, but the carbonation depths for these
emission scenarios vary by no more than 6 mm by 2100
in all the three cities. For example, in Figure 5(b), the A1FI
emission scenario increases carbonation depth by
approximately 45% when compared to reference year
2000 CO2 emissions for exposure classification A2 for
Canberra. The effect of the nine GCM temperature
prediction models on probabilities of corrosion initiation
and corrosion damage is shown in Stewart et al.s work
(Stewart, Wang et al. 2011), for A1FI emission scenario
and A1 exposure classification in Sydney. There are 6.8%
and 17.1% differences between maximum and minimum
predicted probabilities of corrosion initiation and damage,
respectively. This represents quite low variability; hence,
it is reasonable to only consider the average of nine GCM
temperature simulations when presenting results. The

likelihood of corrosion initiation is less than 0.03%, and

less than 0.003% for corrosion damage, for exposure
classifications B2 and C in Sydney, see Figure 6. These
probabilities are negligible irrespective of the emission
scenario. Corrosion initiation and corrosion damage risks
are highest for exposure classifications A2 and B1, as
these are exposures most susceptible to carbonation for
structures located well away from the coast. There is
unlikely to be any significant corrosion damage for the
first 40-50 years service life, but the likelihood of
corrosion damage then increases to 1% to 2.5% for A1FI,
A1B and 550 ppm emission scenarios for Sydney. In
practical terms, this is equivalent to expecting that 1% to
2.5% of every concrete surface by the year 2100 will be
damaged and in need of maintenance or repair. Of most
interest in this paper is the exposure classifications B1
and B2 for Sydney (up to 50km from coast). The
probability of corrosion damage for the worst case
scenario (A1FI) is up to 10 times higher than that
observed for the reference (best) mitigation scenario for
B1, and for B2 exposure classifications. For Canberra,
damage risks increase to 2% by 2100 for emission
scenarion A1FI. For Xiamens bridges there is up to 5%
probabilities of Mark G. Stewart, Lizhengli Peng and
Xiaoming Wang corrosion damage for A1FI, A1B and 550
ppm emission scenarios by 2100. This indicates that the
higher CO2 concentration could lead to a significant
likelihood and extent of corrosion damage that may need
costly and disruptive repairs during the service life of
many concrete structures in Australia, China, and

elsewhere. Because the definition of environmental

exposure in the two countries standards are different,
the results cannot be compared directly. However, in
order to make the comparison between structures in
Sydney and Xiamen more reasonable, the results of
similar structures could make sense. For example, we
could explore how does the damage risk for a bridge 5
km from the coast in Sydney differs from one in Xiamen
that is also 5 km from the coast for the same exposure
and same use. Therefore, the exposure classification for
structures in Sydney and bridges in Xiamen should be B1
and II, respectively. In both cases, ksite is based on
statistical parameters for an urban environment (see
Table 3). The results of all the possible exposure
categories for Sydney and Xiamen are presented in Table
8. It can be easily find out from Table 8 that the corrosion
of Xiamens bridges are more severe than in Sydney. For
beams of a bridge 5 km from the coast in Sydney and
Xiamen, the exposures for them are B1and II,
respectively. The probabilities of corrosion damage for
Xiamens bridges are double than that of Sydney. The
warmer weather and greater temperature increase in the
90 years prediction in Xiamen might be one cause of
higher damage risks, but lower Fc and cover might be
more critical.
CONCLUSIONS
This paper describes a probabilistic and reliability-based
approach that predicts the probability of corrosion
initiation and damage (severe cracking) for concrete

infrastructure subjected to carbonationinduced corrosion

resulting from elevated CO2 levels, RH and temperatures
due to a changing climate. The probabilistic analysis
included the uncertainty of CO2 concentration,
deterioration processes, material properties, dimensions,
and predictive models. It was found that
carbonationinduced damage risks can increase threefold
by 2100 to 2% for Canberra. The results were most
sensitive to increases in atmospheric CO2. These
structures may merit appropriate and cost-effective
climate adaptation measures to ameliorate the adverse
effects of a changing climate.
REFERENCES
AS3600. 2009. Concrete Structures. Sydney: Standards
Australia. Bastidas-Arteaga, E., Chateauneuf, A., SanchezSilva, M., Bressolette, Ph., and Schoefs, F. 2010. Influence
of weather and global warming in chloride ingress into
concrete: a stochastic approach, Structural Safety, 32:
238-249. Day, T.A., Gober, P., Xiaong, F.S. and Wentz, E.
2002. Temporal Patterns in Near Surface CO2
Concentrations over Contrasting Vegetation Types in the
Pheonix Metropolitan Area. Agriculture and Forest
Meteorology. 110: 229-245. de Larrard, F. 1999. Concrete
Mixtures Proportioning: a Scientific Approach. London:
E&FN Spon. DuraCrete. 1998. Modelling of Degradation,
DuraCrete - Probabilistic Performance based Durability
Design of Concrete Structures. EU Brite EuRam III.
Contract BRPR-CT95-0132. Project BE95- 1347/R4-5, 174
p. DuraCrete. 2000a. Statistical Quantification of the

Variables in the Limit State Functions. DuraCrete Probabilistic Performance based Durability Design of
Concrete Structures, EU - Brite EuRam III. Contract BRPRCT95-0132. Project BE95-1347/R9, 130 p. DuraCrete.
2000b. Probabilistic Calculations, DuraCrete - Probabilistic
Performance based Durability Design of Concrete
Structures. EU - Brite EuRam III. Contract BRPR-CT950132. Project BE95- 1347/R12-13, 41 p. El Maaddawy, T.
and Soudki, K.A. 2007. A Model for Prediction of Time froe
Corrosion Initiation to Corrosion Cracking. Cement &
Concrete Composite. 29(3): 168-175. Fib. 2006. Model
Code for Service Life Design. Lausanne: Fib Bulletin 34.
George, K., Ziska, L.H., Bunce, J.A. and Quebedeaux, B.
2007. Elevated Atmospheric CO2 Concentration and
Temperature Across an Urban-Rural Transect. Atmospheric
Environment. 41: 7654-7665. IPCC. 2007. Fourth
Assessment Report of the Intergovernmental Panel in
Climate Change. UK: Cambridge University Press. IPCC.
2000. Emission scenarios. Special report of the
intergovernmental Panel on Climate Change. UK:
Cambridge University Press. Kada-Benameur, H., Wirquin,
E. and Duthoit, B. 2000. Determination of Apparent
Activation Energy of Concrete by Isothermal Calorimetry.
Cement and Concrete Research. 30: 301-305. McGee, R.
1999. Modelling of Durability Performance of Tasmanian
Bridges. ICASP8 Applications of Statistics and Probability
in Civil Engineering. R.E. Melchers and M.G. Stewart
(eds.). 1: 297-306. Mirza, S.A., Hatzinikolas, M., and
MacGregor, J.G. 1979. Statistical Descriptions of Strength
of Concrete. Journal of the Structural Division. 105(ST6):

1021-1037. Mullard, J.A. and Stewart, M.G. 2011.

Corrosion-induced cover cracking: new test data and
predictive models. Aci Structural Journal 108(1):71-79
Pham, L. 1985. Reliability Analysis of Reinforced Concrete
and Composite Column Sections Under Concentric Loads.
Civil Engineering Transactions. IEAust. CE27(1): 68-72.
Richardson,M.G. 1988. Carbonation of reinforced
concrete: Its Causes and Management. Russel, D.,
Basheer, P.A.M., Rankin, G.I.B. and Long, A.E.. 2001. Effect
of relative humidity and air permeability on prediction of
the rate of carbonation of concrete. Proceeding of the
Institution of Civil Engineers-Structures and Buildings.
146(3):319-326. Stewart, M.G., Teply, B. and Kralova, H.
2002. The Effect of Temporal and Spatial Variability of
Ambient Carbon Dioxide Concentrations on Carbonation
of RC Structures. 9th International Conference on
Durability of Building Materials and Components. CSIRO.
Paper 246 (CD-ROM). Stewart, M.G. and Peng, J. 2010.
Life Cycle Cost Assessment of Climate Change Adaptation
Measures to Minimise Carbonation-Induced Corrosion
Risks, International Journal of Engineering Under
Uncertainty: Hazards, Assessment and Mitigation, 2(1-2):
35-46. Stewart, M.G., Wang, X.M. and Nguyen, M.N. 2011.
Climate change impact and risks of concrete
infrastructure deterioration. Engineering Structures 33(4):
1326-1337. Stewart, M.G., Wang, X. and Nguyen, M.
2012. Climate Change Adaptation for Corrosion Control of
Concrete Infrastructure, Structural Safety, 35(March): 2939. Talukdar, S., Banthia, N., Grace, J.R. and Cohen, S.
2012. Carbonation in Concrete Infrastructure in the

Context of Global Climate Change: Parts 1 and 2, Cement

& Concrete Composites (in press). Vu, K.A.T. and Stewart,
M.G. 2000. Structural Reliability of Concrete Bridges
Including Improved Chloride-induced Corrosion Models.
Structural Safety. 22(4): 313-333. Wang, X., Chen, D. and
Ren, Z. 2010. Assessment of climate change impact on
residential building heating and cooling energy
requirement in Australia, Building and Environment,
45(7): 1663-1682. Wigley T.M.L., Richels R, and Edmonds
J.A. 1996. Economic and environmental choices in the
stabilization of atmospheric CO2 concentrations. Nature:
379:240-3 Yoon, I.S., Copuroglu, O., and Park, K.B. 2007.
Effect of global climatic change on carbonation progress
of concrete. Atmospheric Environment. 41: 7274-7285.
Initial set, as defined by ACI 116R, is a degree of stiffening of a mixture
of cement less than final set, generally stated as an empirical value
indicating the time in hours and minutes required for the cement paste to
stiffen sufficiently to resist to an established degree, the penetration of a
weighted test needle. The concrete setting time at various temperatures
is given at table below:
Table 1 Setting Time of Concrete at Various Temperature
Temperature Approximate
Setting Time
(hours)
100oF (38oC) 1-2/3
90oF (32oC)

2-2/3

80oF (27oC)

70oF (21oC)

60oF (16oC)

50oF (10oC)

11

40oF (4oC)

14

30oF (-1oC)

19

20oF (-7oC)

Set will not occur

The retardation of initial setting time by the use of admixture is affected

by three factors, that is, the ambient temperature, the dosage used, and
the time of adding to the batch.

Temperature Effect on Retardation of Initial Setting Time

Temperature can have a detrimental effect to concrete strength
development. However, proper cold weather concrete curing will
enhance concrete strength development. Hot weather is defined as any
combination is high ambient temperature, high concrete temperature,
low relative humidity, and wind velocity. Cold weather period, as
defined by ACI Committee 306, is when one of the following conditions
occur for three consecutive days:
Average daily air temperature is less than 40oF
The air temperature is not greater than 50oF for more than one-half
of any 24 hour period.

The effect of concrete temperature and retardation of setting time is

given by PCA in the chart below. It is concluded from the chart that the
retardation effect is more pronounced when a higher concrete

temperature is used.

Effect of Concrete Temperature and Retarder on Setting Time

Retardation of setting time is influenced by the type of admixtures used.

The chart below illustrates the effect of various lignosufonates (1 and 2)
and carboxylic (3 and 4) admixtures on setting time.

The time of adding the admixture to the batch here is significant and
may affect final results. More retardation may take place if the
admixture is added as the last ingredient and the cement is wet.
Dosage Effect on Retardation of Initial Setting Time
Higher dosage may be used up to a certain level only prior to when any
rapid stiffening and slump loss occurs. This admixture is sensitive to
ambient temperature when introduced to the batch. The lower the
ambient temperatures, the longer time of setting for the concrete will be.
The following figure is used to estimate initial setting time according to
the dosage of retarder and concrete ambient temperature.

Increasing Initial Setting Time with Retarder Content

Remember When Experts Predicted Climate Change

Was "Global"? The U.S. Warming Pause

(click on to enlarge)

Per NOAA, the U.S. warming pause (aka the 'Hiatus') has
now achieved a 19-year stall (see adjacent chart). In
fact, a slight cooling has been the trend over this period.
Remember the predicted global warming by experts? The
same "experts" who predicted that hurricanes would
become stronger and more frequent as a result of the
global warming - which also did not happen.
As the empirical climate datasets reveal, the predicted
global warming has amounted to about nil for close to
two decades. And because of this, the global warming
scientists recently resorted to exceptional fabrications of
temperature datasets to produce "warming" that
disappears the 'Pause'.
Ginning up climate change fears in anticipation of
the Paris 2015 COP21 climate travesty show seems to be
the driving force behind the most recent wholesale fakewarming production.
Back to the included chart. As depicted, the 19-year
pause includes not only the continental U.S. (at -0.4F
per century cooling) but also the states of Virginia and
Maryland, both at -0.5F per century cooling.
Why depict those two state's temperature trends?
Because those states surround the metropolitan District
of Columbia where federal bureaucrats, U.S. elected
representatives and administration officials pontificate
about the rapid and dangerous "global warming". These
elites live and work in the D.C. micro-climate warming
bubble that is a direct result of federal taxpayer asphalt,
steel, concrete and airports with very hot jet exhausts,
which in combination have produced a rapidly
warmingurban heat island (UHI).

The NOAA scientific empirical evidence is rather clear and

undeniable. For most Americans, global warming is not
an issue and is definitely not impacting their daily lives.
But for a minority of governing elites, who obviously
created a hostile warming micro-climate for their work
environment, it has made them incapable of
distinguishing the climate forest from the micro-climate
trees, so-to-speak. Or, put another way, they can't
discern the difference between climate reality and climate
fantasy.
Hmmm....maybe the best solution for saving the elites
from their own, self-created hostile and climate change
triggering environment is to disperse the federal
government offices and personnel across rural locations
throughout the U.S.
EFFECTS OF EXTREME COLD ON MATERIALS

An understanding of the effect of extreme cold on the elasticity,

durability, strength, and other physical characteristics of materials, and
the treatment that these materials should receive when exposed to such
temperatures is important. Where applicable and when required,
information
on this subject can be obtained from manufacturers furnishing material
or

equipment, and from qualified research laboratories.

Water
Fresh Water . Under usual conditions, fresh water freezes at a
temperature of 32F., forming solid ice and expanding about 9% in
volume. It takes 80 calories to raise a cubic centimeter of ice across the
freezing point and more to bring it to a temperature at which it is
potable. Water weighs 62.5 lb. per cubic foot and ice at 32F. weighs
57.5 lb. per cubic foot. The strength
of ice is dependent on its structure (see Strength and Uses of Fresh-and
Salt-Water Ice.) Trautwin d e states that the expansive force of ice is
probably
not less than 30,000 per square inch. This force exceeds the yield
strength
of cast iron (25,000 + p.s.i.). Fresh-water ice is 2 to 3 times stronger
than
sea (salt) water ice. Pressure applied to ice causes momentary melting at
the point of pressure, producing a film of moisture. This is especially
true
at temperatures near 32F. At progressively lower temperatures, melting

due to pressure decreases to the vanishing point.

Sea (salt) water freezes at approximately 28.6F., depending on
salinity.
Water with a higher salt content freezes at lower temperatures. Newly
formed
sea-water ice is mushy because of high salinity; at lower temperatures
later,
when some of its the salt has become eliminated, it is flexible as
compared to
fresh-water ice. Old sea-water ice is usually stronger than new sea ice, is
darker in color, and lighter in weight. So long as it contains appreciable
salt it has a rough surface.

Antifreeze Solutions
Approximately two-thirds of the energy in the gasoline used in
operating
an automobile engine is converted into heat. It is, therefore, necessary to
provide special cooling facilities to prevent the metal parts from
reaching
excessive temperatures. The method generally used is an indirect one in

volving the transfer of heat from the engine to [ ?] a liquid, usually

water, and
Fig. 1 then cooling the liquid by air through the use of a radiator (see
Fig. 1).
Water was naturally selected as a cooling medium because of its
availability
and relatively high heat transfer properties. However, water has certain
short
comings, the most important of which are its high freezing point and its
corrosive action on metal parts of the cooling system, which may result
in
rust clogging and metal perforation. These two major disadvantages are
largely
overcome by adding materials to the water to prevent freezing in winter,
and
special chemical ingredients to inhibit corrosion. Oils, sug e a rs, and
inorganic
salt solutions are generally regarded as unsatisfactory antifreeze
materials.
In the United States and Canada, approximately one-third of the cars

requiring
antifreeze are protected with ethylene glycol (glycol) base products and
most of the remaining two-thirds employ methyl

alcohol (methanol) or ethyl alcohol (ethanol) type

solutions. In the Arctic, ethylene glycol base
products are used almost entirely.
The antifreeze effectiveness of methyl and
ethyl alcohols and ethyleneglycol types is shown in
Figure 2. These curves bring out several facts.
First, the methyl alcohol type give the greatest
freezing protection per unit volume, followed by
ethylene glycol, and then the ethyl alcohol.
Second, all three liquids are capable of depressing
the freezing point of water to thelowest
atmospheric temperatures likely to be
encountered. The first reason is based only on
freezing protection per gallon, and does not take
into con sideration the extra quantities of the lowboiling-point alcohol antifreeze solutions required
after the initial filling because of boil-away losses,
or the superiority of the comparatively high boiling
point of ethylene glycol solution in preventing such
losses. For antifreeze solutions protecting down to
20F., the boiling point of the ethylene glycol
solution is 22 8 3 F.while the boiling point of the

alcohol-base solution is 180F.Unlike water,

antifreeze solutions do not solidify when exposed
to
temperatures slightly below their freezing points
but instead tend to form slush. The minimum
temperatures to which solutions of the three types
of antifreeze having a freezing point of 0F. may be
exposed without giving rise to overheating or other
difficulties immediately after the engine is started
are: methyl alcohol, 2.5 to 5.5F.; ethyl alcohol,
5.5 to 8.0F.; ethylene glycol, 8.0 to
11.5F.
The lower the freezing point of the antifreeze
solutions used, the further below this freezing
temperature is it possible to expose the solution
without fear of overheating, resulting from
circulation restricted by ice crystals
Relation Between Concentrations and Freezing
Protections of Various Antifreeze Solutions or slush ice,
after the engine started. From Figure 2 it is noted that
antifreeze protection can be determined in volume per
cent concentration in water and easily reduced to pints
per gallon of solution (see Table I).
Table I. Pints of Antifreeze per Gallon Soulutions for
Protection

down to Various Temperatures.

Protection

Methyl

Ethyl

Ethylene

to, F.

alcohol

alcohol

glycol

+ 10

2 1/4

2 3/4

10

3 3/4

3 1/4

20

3 1/2

30

4 1/2

40

5 1/4

4 1/4

50

5 3/4

4 1/2

20

In the case of ethylene glycol, the greatest freezing protection that

can be obtained is 62F. which is given by a solution containing 60%
anti freeze and 40% water. Solutions containing more than 60%
ethyleglycol give less protection.

Glycerim (glycerol) is one of the acceptable nonvolatile antifreeze

materials, but because of its relatively high cost compared to ethylene
glycol, and its many other important commercial uses, it is not used to
any great extent.Kerosene, freezing point 60F., had been used in
standard automotive cooling systems in localities with extreme cold
climates. Its heat capacity is approximately one-half that of water, but
automobiles operating with kerosene as a coolant are subject to
overheating in warm weather. Additional disadvantages are its
unpleasant odor, flammability, and severe action on rubber hose.Care
should be taken to select an antifreeze containing heavy-duty inhibitors.
Two general types are in general use: soluble oils and salts.
The oil types are considered generally to be the most satisfactory.
Vehicle radiators filled with antifreeze should be tagged showing type of
antifreeze.
Fuels and Lubricants
During World War II, special fuels and lubricants were developed
to overcome the difficulties in star gasoline and diesel engines
previously encountered in the Arctic. (See Petroleum Products for
Arctic Winter Use in Automotive Equipment and Tractor-Type
Transportation Units for Arctic
Operations for details on the improvements made on the various
properties of fuels and lubricants for low-temperature use.)
In shipping fuels in drums it is important that only extra heavy
exporttype drums be used. This is necessary as this type drum can be
handled easier
in the cold and facilitates roping for dropping by parachute from planes.

The smooth drum is slippery when wet or covered with ice or snow and
it is difficult to rope and attach to a parachute.
The recommendations of the manufacturer of any equipment should
be consulted regarding lubrication under cold conditions. Many
excellent lubricants
have been developed and used successfully in northern operations.
However, it must be realized that at extreme temperatures oils and
greases become stiff.
If an engine has been shut down for any period of time the lubricant may
have become so stiff that a fully charged battery will not turn the engine
over.
This situation may be further aggravated because at such temperature
batteries lose much of their energy.
Rubber-like Material
The general effect of reduced temperatures is the same for all
rubber-like materials. As the temperature is decreased the rubber passes
from a soft (easily deformed) and elastic state to a more rigid state and
finally to a brittle glasslike condition. The various commercial rubbers
differ appreciably as to the temperature ranges in which they pass
through these various states.
None of the available commercial rubbers are truly elastic at extremely
low temperatures (below 40F.). New rubber products stand up better
under cold conditions than old rubber. The effect of temperature on

rubber materials is predominantly physical and any chemical changes

which may take place can, on a practical basis, be ignored.
Some new natural rubber materials are usable at low temperatures
approaching50F. but in the course of their use it is imperative that they
be not subjected to any force at an excessive velocity.
That is, rapid bending orflexing at or near such low temperatures will
result in breaking or even shattering of the rubber part. For example,
rubber tires will develop flat spots at low temperatures. The tread of old
rubber tires will chip due to cold embittlement when subjected to force
or flexing. New tires show less tendency to crack than do tires of old
rubber.
Lower-temperature rubber-like materials are made by specifically
compounding the integral parts for low-temperature service. Two
general classes of these have been developed: normal natural rubber
material to operate (with care) down to 40F., and special rubber-like
material (natural rubber and butyl rubber) for extreme low temperatures
to 70F. Many of the large rubber and chemical companies that
specialize in rubber and synthetic rubberproducts are working on the
problem of providing rubber-like materials for use under extreme cold
conditions.
Plastics
Most plastics contain a base material, the properties of which have
been modified by the incorporation of plasticizers or fillers. Each base
material is the foundation for a group of compositions related in general
behavior but differing from one another in individual physical
properties. Such basic
groups of plastics are: acrylics, celluloses, nylons, ethylene , polymers,

vinyl ester polymers, polyvinyl acetals, phenolics, urea resins, caseins

alkyds, neoprene, etc. groups which contain several different
compositions are subdivided into types. Each type represents one or
more compositions, each of which is designed to give superior value of
some specific property even at the expense of some other property.
There is, for example, Type neral; Type II, temperature resistant; Type
III, impact resistant; Type IV, moisture resistant; etc. Where further
subdivision is required, the typesare subdivided into grades. Each grade
represents, at broadest, a very restricted number of common commercial
materials which are quite similar both chemically and physically. These
groups, types, and grades usually correspond to those given in the
specifications of the American Society for Testing Materials.
The service success of an article of any plastic often depends as
much upon the design and fabrication processes as on the material itself.
The
importance of selecting items of good workmanship in both design and
fabrication for cold-weather operations cannot be overemphasized. The
plastics
industry has developed a background of practical experience in design,
fabrication, and testing of plastics, and should be consulted regarding
specific cold-weather problems.
The importance of selecting the proper material and consulting with
plastic manufacturers concerning cold-weather problems cannot be
overstressed. It is important not only to select the proper material but to
use it properly in the field. Too frequently, good plastics improperly
handled in the field failed, when the same material properly utilized
would have been entirely satisfactory.

As an aid in understanding this field of material, a list of the more

important plastics by resin group and subgroup, trade names, available
forms, and commercial uses is given in Table I. (The code for the
available forms is: F, filaments; M, molded; R, rods; S, sheeting; T,
tubing.) where applicable, comments on the effects of extreme
temperatures and care in use in the field are given in the text.
The acrylics are perfectly clear and transparent. They have the best
resistance of all transparent plastics to sunlight and outdoor weathering,
and will tolerate years of exposure without significant loss of properties.
They possess a good combination of flexibility with shatter resistance
and rigidity.
Their impact strength is lower than the celluloses, but the effect of
extreme low temperatures upon this property is much less pronounced;
hence, articles designed for use at ordinary temperatures will not show
excessive embrittlement
at 50F.
Cellulose nitrate is the toughest of all thermoplastics. It has low
water absorption and is resistant to mild acids. At 50F., its impact
strength
is about 35% of its impact strength at normal temperatures (77F.).
Cellulose nitrate is very flammable; it is not suitable for prolonged
service in outdoor sunlight for it turns yellow and becomes brittle.
Cellulose acetate is comparatively tough. Its low temperature
impact strength and embrittlement characteristics are inferior to those of
cellulose

Table I. Some Important Commercial Plastics.

Resin
group and

Trade
names

Forms

Uses

availa
ble

subgroup
Acrylics:

Lucite

Methyl
methacryla
te

Plexigla
s

M, R,
S, T

Windshields,
goggles,
dentures,

M, R, S
artificial eyes,
drafting in

resin
struments,
automotive
parts,
aircraft
enclosures
Celluloses:

Celluloid

Cellulose
nitrate

Nitron
Nixon
C/N

R, S, T,
F

Fountain pens
and pencils,

R, S, T

drawing
instruments,
spectacle

R, S, T

frames, bottle

Pyralin

R, S, T

caps, toilet
seats,
tool handles,
shoelace tips,
film

Cellulose
acetate

Fibestos

R, S, T

Lumarit
h

R, S, T,
M

Containers,
luggage, food
cases,
truck curtains

Nixon
C/A

R, S, T,
M

Plastoce
le

R, S, T

Chemac
o

M
M

Knobs, goggle
frames,
combs,

Hercules
M
Koppers
M

brushes, tool
handles,
safety

Tenite I
goggles, eye
shields,
automotive

parts and
housings
Cellulose
acetate

Tenite II

Telephones,
steering
wheels,

butyrate
film spools,
radio
housings,
knobs
and pulls, light
supports, coil
spools, brush
backs
Ethyl
Cellulose

Celcon

Chemac
o

Radio
housings,
toothbrushes,
pen

M, S
Ethocel
M
Hercules

and pencil
barrels, tool
handles,

M
Koppers

knobs and

M, S

pulls,
flashlight
cases

Textile fiber,
ropes, lines,
hose,

Nixon
E/C
Nylon:
Textile
filament

tents,
stockings,
clothing,

types

bristles,
surgical
sutures
Injection,
extrusion
and
alcoholsoluble

M, S

Injection and
compressed
molding,
covering for
wire and
sheets,

types
solution
castings, small
bearings,
specialty

containers,
electrical
coil forms and
insulators,
small
gears, cams,
coatings

Table I. Some Important Commercial Plastics (contd).

Resin group
and

Trade
names

Forms

subgroup

availa
ble

Ethylene
plymers:

F, M, S,
T

Polyethylene

F, M, S,
T

Polythene

Uses

Films, liners,
closures,
wrappings
for frozen
food,
primary
cable,
insulating
material,

coating for
weatherpro
of wire
Polytetrafluor
o

Teflon

M, R, S,
T

Films,
tubes, tapes
and special

ethylene
applications
made by
rolling,
drawing, or
machining
Polyvinyl
acetals:

Formvar

Polyvinyl
formal

Polyvinyl
butyral

Insulating
enamel,
base for
electric
wires,
phonograph
records

Butacite

M, S

Saflex

Plastic
interlayer,
laminated

Vinylite

M, S

for safety
glass,
sheeting,
and
coatings for
dustproof
and
waterproofi
ng fabrics

Vinyl ester
polymers:
Polyvinyl
chloride

Geon

M, S

Marvinol

Pliovic

M, S

Ultron

M, S

Vinylite

M, S

Jacketing
material on
electric
wires and
cables,
waterrepellent
garments,
shower
curtains,
garment
bags,
upholstery,
belts, floor

coverings,
overlays for
maps,
phonograph
records
Polyvinyliden
e resins:

Saran

F, M, T

Finylidene
chloride

Hoses,
flexible
tubing, rigid
pipe,
lined steel
pipe,
moistureresistant
films and
fabrics for
upholstery
and
transportati
on seating

Polystyrene

Bakelite

Cerex

Standoff
insulators,
antenna in

Chemac
o

M
M

sulators,
radio coil
forms,

Koppers
M
Loalex
M

telephone
equipment,
fluorescent

Loalin
M
Lustrex
M
Styron

light
fixtures,
wall til s e ,
combs,
knobs and
pulls,
shaver
housings,
camera
cases,
refrigerator
parts,
bottle caps

Polystrene
expanded

Styrofoa
m

Insulating
material in
refrigera

tion
construction
, buoyancy
agent
for life rafts
and small
metal
boats

Table I. Some Important Commercial Plastics (contd).

Resin group
and

Trade
names

Forms

Uses

availa
ble

subgroup
Phenolics:

Bakelite

Phenolformaldehyde

Durez

Durite

resin
Resinox

Camera
cases,
photographic
film
spools,
handles,
instruments,

boxes, radio
cabinets,
ignition
parts,
instrument
panels,
pulleys,
housings,
terminal
blocks,
telephone
parts, goggle
frames,
wheels
Melamine
resin:
Melamineformalde

Melmac

Plaskon

Resime
ne

fittings,
sockets, food
containers

Beetle

Buttons,
tableware,

hyde
Urea resins:

Compression
moldings,
electrical

Ureaformaldehyde

Plaskon

boxes,
electrical
parts and
lighting
reflectors

Synthetic
rubber:
Chlorobutadie
ne

Neopre
ne

M, S, T

Hose,
molded
parts,
weather
strip
ping, wire
and cable
jacketing
adhesive,
coated
fabric,
electrical
cable
construction,
inflatable
gear, sealing
strips

nitrate. Cellulose acetate is superior to cellulose nitrate in

resistance to
outdoor exposure and to burning. Sunlight has little effect
on tmaterial. Since there are many commercial
compositions of this material, it is advisable for a given
application to indicate the application and desired
properties, for example, for general use, resistance to
heat, cold, impact, or moisture.
Cellulose acetate butyrate material is tough and has dimensional
stability.Fluctuation in dimension must be considered when articles are
made of a combination of this material and glass or metal.
Ethyl cellulose material possesses toughness, high impact strength
at low temperatures, and excellent dimensional stability. When the
article is in combination with glass or steel, assurance must be made that
the wall thickness of the plastic is sufficient to withstand the strain
caused bytemperature changes. Type II of this plastic is specifically
designed for low-temperature resistance. At 50F., its impact strength is
about 40% of its impact strength at normal temperatures.
Nylon is a generic term for any long-chain synthetic polymeric
amide which has recurring amide groups as an intergral part of the main
polymer chain, and which is capable of being formed into a filament
whose structural elements are
oriented in the direction of the axis. Nylon textile filament materials are
noted for their toughness. The effect of extreme cold on the mechanical
properties of cords and ropes is small: tensile strength increases and
elongation

decreases. Woven fabrics will not be stiffened or embrittled by extreme

cold and remains soft and pliable at 40F. The effect of prolonged
exposure to sunlight and outdoor weather is not enough to impair
practical utility.
Several different types of nylon g plastic s are involved here atheir
properties are not identical. Impact strength is measurably decreased by
extreme cold but toughness and impact strength at low temperatures are
still so good that nylon plastics have been successfully used at low
temperatures. At40F., the impact strength of nylon is about 55% of its
impact strength at normal temperatures. The electrical properties of
nylon plastics are better at low temperatures than at normal
temperatures. Prolonged exposure of nylon plastics to sunshine and
weathering is not recommended.
Polyethylene and polythene materials are tough and durable. Their
toughness is not seriously effected by low temperatures. These materials
remain fairly flexible at moderately low temperatures, stiffen slightly at
temperatures of 30F. and lower, and become brittle at 94F. They
have excellent electrical properties, extremely low moisture vapor
transfer qualities, resist solvents and strong acids, and have other
desirable qualities such as nontoxicity.
Polytetrafluoroethylene has potential utility owing to its excellent
thermal stability, resistance to corrosive reagents, and low dielectric
loss.It is not embrittled by extremely low temperatures.
Films can be flexed at temperatures as low as 148F. without breaking.
Its resistance to outdoor weathering is excellent.

Polyvinyl acetal material provides a tough impact-resistant

adhesive layer for safety glass over a wide range of temperatures down
to about 40F., is stable to light and heat, relatively insensitive to
moisture, and has good
adhesive qualities. It is an excellent thermoplastic adhesive for leather,
rubber, paper, wood, canvas, laminated cellophane, and glass.
It is also excellent for coating fabrics for raincoats, water-repellent
garments, tentage, food and clothing bags, etc.
Polyvinyl chloride compositions are noteworthy for their heat
resistance,exceptional toughness, and ability to withstand continued
exposure to maximum temperature differences. Some of these
compounds have a low-temperature brittleness approaching 40 and
50F. when subjected to bending. Howeve such material if subjected to
sudden shock would fail at higher temperatures,possibly approaching
30F.
Vinylidene chloride material is tough, resistant to chemicals and
prolonged
immersion in water, nonflammable, and useful over a wide range of
temperatures.