Turkish Journal of Agriculture and Forestry Turk J Agric For

(2015) 39: 39-47

http://journals.tubitak.gov.tr/agriculture/
TBTAK
Research Article doi:10.3906/tar-1402-115

Vinyl acetate-modified microcrystalline cellulose-reinforced HDPE composites

prepared by twin-screw extrusion
1, 1 2
Nihat Sami ETN *, Nilgl ZMEN ETN , David Paul HARPER
1
Department of Forest Industry Engineering, Faculty of Forestry, zmir Katip elebi University, zmir, Turkey
2
Center for Renewable Carbon, University of Tennessee, Knoxville, TN, USA

Received: 24.02.2014 Accepted: 24.07.2014 Published Online: 02.01.2015 Printed: 30.01.2015

Abstract: In order to compare the effects of microcrystalline cellulose (MCC) as a filler, MCC and high-density polyethylene (HDPE)
composites were produced with 4 different wt.% MCC loadings (1 wt.%, 5 wt.%, 10 wt.%, and 20 wt.%). To improve the compatibility
between MCC and the matrix, MCC was acetylated with vinyl acetate (VA), and 2 different MCC-HDPE composites were produced
with 5 wt.% and 10 wt.% VA-modified MCC loadings. Untreated MCC and VA-modified MCC-reinforced HDPE pellets were produced
by using a twin-screw extruder. The MCC-HDPE composites were then compression-molded and the mechanical properties of the
composites were determined. The thermal and morphological properties of the MCC-HDPE composites were characterized by using
thermogravimetric and electron microscopy techniques, respectively. The VA-modified MCC-filled HDPE had a significantly higher
tensile strength than the unmodified MCC-filled or the neat HDPE composites. The thermal stability of MCC was also improved with
VA modification.

Key words: Acetylation, cellulose-based composites, high-density polyethylene, microcrystalline cellulose, vinyl acetate

1. Introduction due to their ability to be recycled and their biodegradability,

Cellulose is the most abundant, renewable, and degradable
biopolymer on earth. It can be obtained from various
sources, such as trees, plant fibers, various sea animals
(tunicin, chitin), and bacteria. Cellulose is a polydispersed
linear polymer of -(1,4)-D-glucose and contains both
amorphous and crystalline regions (Lima and Borsali,
2004; Hill, 2006). Microcrystalline cellulose (MCC) is
much stronger and stiffer than amorphous cellulose and
is expected to be a more effective reinforcing filler than
conventional cellulose fibers due to its high specific
surface area (Mathew et al., 2005). MCC has a high
crystalline content and can be easily produced through
the low-cost process of acid hydrolysis (Bai and Li, 2009).
MCC is a commercial product and a partially purified,
depolymerized -cellulose. It is an odorless, white, and
micrometer-sized type of crystalline cellulose powder.
MCC has been used for many years in different industries
like cosmetics, plastics, food, and pharmaceuticals among
others, due to its excellent properties.
In the past few years there has been some research on MCC
as a green reinforcing filler in ecofriendly polymer composites
(Reinsch and Kelley, 1997). Cellulose-based composites have
found an increasing number of applications in recent years
* Correspondence: nihatcetin@yahoo.com
abundance, low cost, and density (Eichhorn et al., 2010;
Ashori and Nourbakhsh, 2010). The main drawback of MCC
in the thermoplastic matrix is its highly polar and hydrophilic
character, which makes it poorly compatible with commonly
used nonpolar matrices (Spoljaric et al., 2009). In order to
achieve wide dispersion and good adhesion between MCC
and the polymer matrix, chemical modification methods are
often used to improve the hydrophobicity of the MCC surface.
Modified cellulose shows better adhesion as compared with
unmodified cellulose and improves the mechanical properties
of the composites.
The objective of this study was to evaluate the influence
of various wt.% MCC loadings (1 wt.%, 5 wt.%, 10 wt.%,
and 20 wt.%) and surface acetylation of MCC on the
mechanical properties of MCC-reinforced high-density
polyethylene (HDPE) composites. MCC was initially
acetylated with vinyl acetate (VA) of 0.7 degrees of
substitution, and then 2 different wt.% loadings (5 wt.%
and 10 wt.%) of VA-modified MCC were compounded
with HDPE. The morphologies and the thermal properties
of unmodified and VA modified MCC-HDPE composites
were determined by scanning electron microscopy (SEM)
and thermogravimetric analysis (TGA), respectively.
39
 ETN et al. / Turk J Agric For
2. Materials and methods
Microcrystalline cellulose (MCC) Avicel (PH 101) was
obtained from Fluka (nterlab, stanbul, Turkey) and
HDPE was purchased from Petkim (Petilen YYS0464,
stanbul, Turkey). Reagent-grade VA, potassium
carbonate, and dimethyl formamide were obtained from
Aldrich Chemical Company, Inc. (Milwaukee, WI, USA).
2.1. Vinyl acetate modification of microcrystalline
cellulose
The VA modification reaction was performed in a 2-L
round-bottomed flask equipped with a condenser.
Reactions were carried out at 100 C for 3 h with potassium
carbonate (K2CO3) as a catalyst and DMF as a solvent.
First, 30 g of MCC was introduced in the reagent solution
consisting of 1 L of DMF, 100 mL of VA (10.8 mmol), and
10 g of K2CO3 (0.7 mmol). After the modification, the
reaction was stopped with deionized water and filtered on
a 4-m Isopore polycarbonate filter, and then subsequently
rinsed with 50 mL of deionized water for 3 h. Finally, the
VA-modified MCC was further extracted with a mixture
of toluene acetone and ethanol (4/1/1, v/v) with a Soxhlet
apparatus. VA-modified MCC was transferred in deionized
water, sonicated overnight at 10 C, and then freeze-dried.
2.2. Calculation of degree of substitution
Degree of substitution (DS) measurements were performed
as described in an earlier publication (etin et al., 2009).
2.3. Infrared spectroscopy
Infrared absorption spectra of the unmodified and VA-
modified samples were obtained with the potassium
bromide (KBr) technique, using a Shimadzu 8400s FT-
IR spectrometer (Kyoto, Japan) at a resolution of 4 cm1
(40 scans). In each case, 1% w/w of oven-dried MCC was
dispersed in a matrix of KBr and pressed to form pellets.
2.4. Thermogravimetric analysis
The TGA of the produced composites was performed by
thermal gravimetric analyzer (Shimadzu TA60, Kyoto,
Japan). The sample weight was 20 mg. The measurement
Table. The compositions of produced MCC-HDPE composites.
ID HDPE (wt.%) MCC (wt.%) VA-modified MCC (wt.%)
HDPE 100 - -
MCC1 99 1 -
MCC5 95 5 -
MCC10 90 10 - -
MCC20 80 20 - -
VA5 95 - 5
VA10 90 - 10
40
was carried out at a heating rate of 10 C/min in an inert
nitrogen atmosphere with a flow rate of 20 mL/min.
Weight loss versus temperature was recorded.
2.5. Production of MCC-HDPE composites
The composites were compounded using a 10-mm co-
rotating twin-screw extruder with a length-to-diameter
ratio (L/D) of 20 (Rondol Microlab Twin Screw Extruder,
Staffordshire, UK) with gravimetric feeding systems and
equipped with a twin-screw side-feeder for addition of
the MCC. The temperature setting from the hopper to die
was 100 C / 150 C / 170 C / 170 C / 170 C and the
screw speed was 50 rpm. Following this, the extrudate was
passed through a water bath and automatically pelletized
at 2 mm in size by a Rondol pellet machine. Seven different
compositions were used for the production of composites
(Table).
The pellets were dried at 70 2 C for 24 h and then
composites were produced with the compression molding
technique using a Carver press (Model 12-12 H, Wabash,
IN, USA) according to the American Society for Testing
and Materials standards (2010). For composite production,
a Carver brand tile mold [25 cm (l), 25 cm (w), and 5 mm
(h)] was used. The pellets were initially heated for 10 min,
then pressed at a force of 9 t for 5 min at 177 C. Following
that, the composites were cooled down according to
Cooling Method B of the American Society for Testing
and Materials standards (2010). For each composition, 3
composites were produced (Figure 1a). The composites
were then cut into 1 dumbbell shape and 2 bar shapes by
using special molds to produce the samples for tensile,
flexural, and impact strength tests (Figure 1b).
Tensile and flexural tests were performed with a
universal testing machine (Zwick Roell Z010, Ulm,
Germany) following the American Society for Testing and
Materials standards (2001, 2005a) in order to determine
tensile strength and flexural strength for the samples after
being conditioned at 23 2 C with relative humidity of
65 5% for 7 days. The test rate was 5 mm/min. Impact
Degree of substitution (DS)
-
-
-
0.7
0.7
 ETN et al. / Turk J Agric For
a)
Figure 1. a) Photographs of neat HDPE and MCC-HDPE composites; b) samples of tensile strength testing, flexural
testing, and impact testing specimens.
strength was measured with a Zwick HIT5.5P impact
testing machine following the American Society for
Testing and Materials standards (2005b). The notches
of impact specimens were cut using a Polytest notching
cutter by Ray-Ran (Warwickshire, UK).
2.6. SEM analysis
MCC-reinforced HDPE composite samples were dipped
into liquid nitrogen and then snapped into 2 pieces to
get a fractured surface for SEM analysis. The surfaces
were analyzed directly with a JEOL scanning electron
microscope (Model NeoScope JCM-5000, Tokyo, Japan).
3. Results and discussion
3.1. Effect of various wt.% MCC loadings on mechanical
properties of MCC-HDPE composites
As illustrated in Figure 2, increasing MCC addition
changed the composite color from white to caramel at
20 wt.% loading. The effect of various MCC loadings on
the tensile strength values of HDPE composites were
determined (Figure 3). ANOVA ( = 0.05) results showed
Figure 2. MCC wt.% loading levels effects on the color of HDPE
composites.
b)
24
a
a a
22
b
Tensile strength (MPa)
20
c
18
16
0 1 5 10 20
MCC wt.% loading
Figure 3. MCC wt.% loading level effects on the tensile strength
values of the HDPE composites (bars with the same letter are not
significantly different).
that mean tensile strength values decreased when the MCC
loading reached 10 wt.% or above. According to the Tukey
test results, there was no significant difference between
neat HDPE composites and low-level MCC loading (up to
5 wt.% loading).
A significant decrease of tensile strength properties of
MCC-HDPE composites was achieved using 10 wt.% or
higher MCC loading as a filler. Mathew et al. (2005) studied
various MCC wt.% loadings in a PLA matrix. At the addition
of 10 wt.% and 20 wt.% MCC loading, the tensile strength
values showed 33% decrease compared with neat PLA.
In this study, at low levels of MCC loading (up to 5 wt.%)
in HDPE matrix, there was no change in tensile strength.
When MCC loading levels reached 10 wt.% and 20 wt.%, the
tensile strength values dropped 5% and 19% as compared to
neat HDPE due to dispersion problems.
The flexural modulus of the MCC-HDPE composites
increased with MCC loading (Figure 4). MCC-HDPE
41
 ETN et al. / Turk J Agric For

c
2.0

bc
1.8 b 180
a
160
a
Bending modulus (GPa)

1.6 a 140

Impact strength (J/m)

1.4 120
100 b
1.2
80
c c
1.0 60 d
40
0.8
0 1 5 10 20 20
MCC wt.% loading 0
Figure 4. MCC wt.% loading effects on the flexural modulus 0 1 5 10 20
values of HDPE composites (bars with the same letter are not MCC wt.% loading
significantly different). Figure 5. MCC wt.% loading effects on the impact strength
values of the HDPE composites (bars with the same letter are not
significantly different).
composite showed maximum flexural modulus at 20
wt.% loading; the flexural modulus was 36% better than Various wt.% MCC-filled HDPE composites were also
neat HDPE composites. Figure 5 shows the variations of characterized with FT-IR analysis. As seen from Figure 6,
impact strength of the untreated MCC-HDPE composites by increasing the amount of MCC in the HDPE matrix, the
as a function of MCC loading. Impact strength decreases intensities of absorption bands at the regions of 36003200
with MCC loading, reaching a minimum at 20 wt.% MCC cm1 (A) and 1240840 cm1 (B) attributed to the cellulose
loading. components also increased.

Figure 6. FT-IR spectra of HDPE composites with varying MCC wt.% loading.

42
 ETN et al. / Turk J Agric For
3.2. VA modification and characterization of MCC
Since hydrophilic MCC is difficult to disperse in
hydrophobic nonpolar polymer matrices, it is necessary to
improve the interfacial bonding by chemical modification
of the cellulose surface. For this reason, MCC was modified
with VA in the presence of DMF and potassium carbonate
as a solvent and a catalyst, respectively. The reaction
mechanism is shown in Figure 7.
In the current study, our goal was only surface
acetylation of MCC to prevent any decrease of crystallinity
of MCC. The extent of the modification was calculated
by measuring the DS with titration as 0.7 after a 3-h
reaction time. It was reported that when cellulose is used
as a reinforcing element in composites, any chemical
modification of the hydroxyl groups in cellulose must
be restricted to the surface. Otherwise, the high level of
modification causes morphological changes to the fiber
and results in decreased crystallinity, particle size, and
thickness (etin et al., 2009).
VA-modified MCC was first characterized by FT-IR
spectroscopy as shown in Figure 8. Modified samples
showed the following vibrations peaks: the carbonyl
stretching vibration at 1740 cm1 (C = O), the methyl in-
plane bending at 1375 cm1 (C-H), and the C-O stretching
at 1235 cm1 (C-O).
O
Cellulose OH +
O
VA
Figure 7. Reaction between vinyl acetate and MCC.
Figure 8. FT-IR spectra of a) untreated MCC and b) VA-modified MCC.
Thermal degradation patterns of unmodified and
VA-modified MCC are shown in Figure 9. TG analysis
indicated that the MCC did not show significant initial
degradation under 300 C, and the main degradation
took place at 356 C (Figure 9a). VA modification raises
the main degradation temperature from 356 C to 380
C (Figure 9b). In previous studies, VA modification
improved the main degradation temperature of hemp fiber
and Scots pine wood flour at around 16 C compared with
unmodified ones (zmen, 2012; zmen et al., 2013).
3.3. Effect of VA-modified MCC-reinforced HDPE
composites on mechanical and thermal properties
The tensile strength of the VA-modified MCC-filled
HDPE composites was higher than that of the untreated
MCC-filled HDPE composites (Figures 10 and 11): 5 wt.%
and 10 wt.% VA-modified MCC (DS 0.7)-filled composites
showed 5.5% and 5.3% higher tensile strength values
than untreated composites at the same wt.% loading,
respectively. The increase in tensile strength with VA
modification is attributed to better interaction between
MCC and HDPE. At 5 wt.% loading, VA modification
had no effect on the flexural modulus, whereas at 10
wt.% loading, VA modification caused a slight decrease.
However, all MCC-filled composites showed better
flexural modulus than the neat HDPE composite.
O
Cellulose O + HO
Acetylated cellulose
O
43
 356C 2.5
mcc
0
0 100 200 300 400 500 600
Temperature (C)

ETN et al. / Turk J Agric For

0.5
0.5
100 100
0.0
0.0
80 80 0.5
0.5

Weight loss (%)

Weight loss (%)

1.0
60 1.0 60

dta

dta
1.5
40 1.5 40
a b 2.0
2.0
20 20
380C 2.5
356C 2.5 va 3h
mcc 3.0
0 0
0 100 200 300 400 500 600 0 100 200 300 400 500 600
Temperature (C) Temperature (C)
Figure 9. TGA curves of a) MCC and b) VA-modified MCC (for 3 h, DS = 0.7).

24 2.0
b 0.5
100
a 0.0 b
a 1.8 b
a b b
Bending moulus (GPa)

2280 0.5
(MPa)

c
loss (%)

1.6 a
1.0
strength

60
dta

1.5
Weight

1.4
2040
Tensile

b 2.0
20 1.2
380C 2.5
va 3h
3.0
18 0 0 100 200 300 400 500 600 1.0
HDPE MCC5 VA5 MCC10 VA10 HDPE MCC5 VA5 MCC10 VA10
Temperature (C)
Figure 10. VA modification effects on tensile strength (TS) values Figure 11. VA modification effects on tensile modulus values of
of MCC-HDPE composites at 5 wt.% loading (bars with the same MCC-HDPE composites at 10 wt.% loading (bars with the same
letter are not significantly different). letter are not significantly different).

Izod impact strength of the MCC-filled composites
decreased in comparison with the neat HDPE composite
(Figure 12). At 5 wt.% loading, the composites modified
with VA showed a significant increase of 38% in impact
strength compared with the untreated MCC-filled ones.
This higher impact strength may be explained by the
improved compatibility between the MCC surface and the
HDPE matrix. At 10% loading, VA modification had no
statistically significant effect on the impact strength values,
being just 9.5% better than the untreated ones.
Figures 13a and 13b and Figures 14a and 14b show the
SEM micrographs of the freeze-fractured surface of the
unmodified and VA-modified MCC-HDPE composites
(5 wt.% and 10 wt.% MCC loading, respectively). The
VA-modified composites (VA5 or VA10) indicated an
improvement of physical interaction between acetylated
MCC and the HDPE matrix compared with unmodified
ones (MCC5 or MCC10). VA-modified composites showed
better MCC dispersion and improved adhesion between
MCC and the matrix (Figures 13b and 14b), which
resulted in significantly better tensile strength of VA5 and
VA10 composites.
The composites, filled with untreated or VA modified
MCC (at 5 wt.% loading), were evaluated by TG analysis
and the results are shown in Figure 15. MCC-HDPE
composite degradation was completed in 3 steps. The
initial weight lost due to the loss of moisture occurred
between 25 and 270 C (step 1). The decomposition of
MCC occurred between 270 and 400 C in step 2. Finally,
step 3 occurred between 400 and 500 C (revealing the
HDPE matrix decomposition). Complete decomposition
of the composite occurred at 500 C.
It was observed that the main degradation temperature
of MCC in HDPE was shifted from 347 to 371 C with
VA modification. This was an indication that VA-
modified MCC was more stable than untreated MCC
and the chemical modification played an important
role in the thermal decomposition process. MCC5 and

44
 ETN et al. / Turk J Agric For

170
a
160

Impact strength (J/m)

150

c
80
b b b
60

40
HDPE MCC5 VA5 MCC10 VA10
Figure 12. VA modification effects on impact strength values of
MCC-HDPE composites at 5 wt.% and 10 wt.% loading (bars
with the same letter are not significantly different).

a b

Figure 13. Scanning electron micrographs of freeze-fractured surface of 5 wt.% MCC-reinforced HDPE
composites for a) unmodified MCC-reinforced (illustrating voids and gaps) and b) VA-modified MCC-reinforced
(illustrating more effective cellulose wetting by the matrix) at 5 wt.% MCC loading.

a b

Figure 14. Scanning electron micrographs of freeze-fractured surface of 10 wt.% MCC-reinforced HDPE
composites for a) unmodified MCC-reinforced (illustrating voids and gaps) and b) VA-modified MCC-reinforced
(illustrating more effective cellulose wetting by the matrix) composites.

45
 20
3
a
0 487C
3
ETN et al. / Turk J Agric0For100 200 300 400 500 600 700 800
Temperature (C)
1 0.5
100 100
0 0.0
80 347 C 1 80 371 C 0.5
Weight loss (%)

Weight loss (%)

60 1 1.0
60

dtga

dtga
2 1.5
40 40
2 2.0
20 20
3 2.5
a b
0 487C
3 0 487 C 3.0
0 100 200 300 400 500 600 700 800 0 100 200 300 400 500 600 700 800
Temperature (C) Temperature (C)
Figure 15. TGA curves of a) unmodified MCC-reinforced
0.5 HDPE and b) VA-modified MCC-reinforced HDPE composites at 5 wt.%
cellulose
100 loading.
0.0
80 0.5 ash
371 Caround 0.63% and 0.81% final
VA5 composites gave main decomposition temperature of VA-modified MCC was
Weight loss (%)

contents,
60 respectively. 1.0 24 C higher than that of unmodified MCC. VA-modified
In conclusion, at up to 5 wt.% MCC loading, there was MCC in the HDPE matrix also showed a higher main
dtga

1.5
no effect
40 on the tensile strength, whereas at 10 wt.% or 20 decomposition temperature compared with untreated MCC.
wt.% MCC loading, tensile strength exhibited significant
2.0 SEM studies of the freeze-fractured surface of the
reduction.
20 Increasing the addition of unmodified MCC composites gave evidence for the enhanced interfacial
2.5
wt.% loading
b resulted in increased flexural modulus but adhesion and good dispersion between MCC and the
0
significantly decreased 487
the C
impact strength. 3.0 matrix upon VA modification of MCC. Unmodified
VA0 modification of MCC resulted
100 200 300 400 500 600 700 800 in improved MCC-HDPE composites (MCC5 and MCC10) fractured
compatibility between HDPE (C)
Temperature and MCC and led to surfaces showed that MCC and the matrix had 2 separate
significant increases in the values of tensile strength and phases and voids and gaps covered the whole surface,
impact strength compared with the unmodified MCC at while composites containing VA-modified MCC showed
the same wt.% loading. VA modification of MCC had no strong adhesion between MCC and the matrix.
effect on the flexural modulus: treated and untreated MCC-
HDPE composites gave similar flexural modulus values. VA Acknowledgment
modification showed a significant effect upon the thermal This study was carried out with financial support from
properties of MCC, making MCC thermally more stable. The TBTAK (Project 110O138).

References
American Society for Testing and Materials (2001). ASTM D638,
Standard Test Method for Tensile Properties of Plastics. West
Conshohocken, PA, USA: ASTM.
American Society for Testing and Materials (2005a). ASTM D6109,
Standard Test Methods for Flexural Properties of Unreinforced
and Reinforced Plastic Lumber and Related Products. West
Conshohocken, PA, USA: ASTM.
American Society for Testing and Materials. ASTM D256 (2005b).
Standard Test Methods for Impact Resistance of Plastics and
Electrical Insulating Materials. West Conshohocken, PA, USA:
ASTM.
American Society for Testing and Materials (2010). Standard
Practice for Compression Molding Thermoplastic Materials
into test Specimens, Plaques, or Sheets. ASTM D4703-10. West
Conshohocken, PA, USA: ASTM.
Ashori A, Nourbakhsh A (2010). Performance properties of
microcrystalline cellulose as a reinforcing agent in wood
plastic composites. Compos Part B-Eng 41: 578581.
Bai W, Li K (2009). Partial replacement of silica with microcrystalline
cellulose in rubber composites. Compos Part A-Appl S 40:
15971605.
etin NS, Tingaut P, zmen N, Henry N, Dadmun M, Sbe G (2009).
The acetylation of cellulose nanowhiskers with vinyl acetate
under moderate conditions. Macromol Biosci 9: 9971003.
Eichhorn SJ, Dufrense A, Aranguren M, Marcovich NE, Capadona
JR, Rowan SJ (2010) Review: Current international research
into cellulose nanofibres and nanocomposites. J Mater Sci 35:
201207.
Hill CAS (2006). Wood Modification: Chemical, Thermal and Other
Processes. London, UK: John Wiley and Sons.

46
 ETN et al. / Turk J Agric For
Lima MMD, Borsali R (2004). Rod-like cellulose microcrystals:
structure, properties, and applications. Macromol Rapid
Comm 25: 771787.
Mathew AP, Oksman K, Sain M (2005). Mechanical properties of
biodegradable composites from poly lactic acid (PLA) and
microcrystalline cellulose (MCC). J Appl Polymer Sci 97:
20142025.
zmen N (2012). A study of the effect of acetylation on hemp fibres
with vinyl acetate. BioResources 7: 38003809.
zmen N, etin NS, Mengelolu F, Birinci E, Karaku K (2013).
Effect of vinyl acetate modification on the properties of wood-
plastic composites. BioResources 8: 753767.
Reinsch VE, Kelley SS (1997). Crystallization of poly(hydroxybutyrate-
cohydroxyvalerate) in wood fiber-reinforced composites. J
Appl Polym Sci 64: 17851796.
Spoljaric S, Genovese A, Shanks RA. Polypropylene-microcrystalline
cellulose composites with enhanced compatibility and
properties. Compos Part A-Appl S 40: 791799.
47