Properties investigation of recycled polylactic acid reinforced by cellulose nanofibrils isolated from bagasse
In this research, an industrial-scale approach was developed for preparing bio-nanocomposites from recycled polylactic acid (rPLA) and cellulose nanofibrils (CNFs). In this regard, several steps were conducted consisting of extracting CNFs, preparing CNF/rPLA master batch, and melt compounding which was finally followed by compression molding. The influence of adding CNFs on rPLA properties was investigated by morphological, mechanical, thermo-mechanical, and degradability studies. Images from scanning electron microscopy (SEM) revealed an increase in the fracture surface roughness of rPLA after adding CNFs. In addition, compared to unreinforced rPLA, the modulus and strength of bio-nanocomposites containing 3 wt% CNFs were enhanced from 527.5 and 23.9 MPa to 716.5 and 32.6 MPa, respectively. Other mechanical properties including elongation at break and work of fracture were decreased by 35.5 and 33% at this CNF percentage. Furthermore, the storage modulus, obtained from dynamic mechanical analysis (DMA), was significantly improved from 1,024 to 8,214 MPa after adding 3 wt % CNF. Similarly, at this CNF percentage, glass transition temperature (Tg) was enhanced from 59.5 to 64°C. According to biodegradation study, the highest biodegradability resistance was also obtained for bio-nanocomposite containing 3 wt % CNF. POLYM. COMPOS., 2017.
Properties Investigation of Recycled Polylactic Acid
Reinforced by Cellulose Nanofibrils Isolated from
Pejman Heidarian ,
Department of Chemical Engineering, Isfahan University of Technology, Isfahan 8415681167, Iran
Department of Chemical Engineering, Center for Biocomposites and Biomaterials Processing, Faculty of
Forestry, University of Toronto, Toronto M5S 3B3, Canada
In this research, an industrial-scale approach was devel-
oped for preparing bio-nanocomposites from recycled
polylactic acid (rPLA) and cellulose nanofibrils (CNFs). In
this regard, several steps were conducted consisting of
extracting CNFs, preparing CNF/rPLA master batch, and
melt compounding which was finally followed by com-
pression molding. The influence of adding CNFs on rPLA
properties was investigated by morphological, mechani-
cal, thermo-mechanical, and degradability studies. Images
from scanning electron microscopy (SEM) revealed an
increase in the fracture surface roughness of rPLA after
adding CNFs. In addition, compared to unreinforced rPLA,
the modulus and strength of bio-nanocomposites contain-
ing 3 wt% CNFs were enhanced from 527.5 and 23.9 MPa
to 716.5 and 32.6 MPa, respectively. Other mechanical
properties including elongation at break and work of frac-
ture were decreased by 35.5 and 33% at this CNF percent-
age. Furthermore, the storage modulus, obtained from
dynamic mechanical analysis (DMA), was significantly
improved from 1,024 to 8,214 MPa after adding 3 wt %
CNF. Similarly, at this CNF percentage, glass transition
) was enhanced from 59.5 to 648C. Accord-
ingtobiodegradationstudy,the highest biodegradability
resistance was also obtained for bio-nanocomposite con-
taining 3 wt % CNF. POLYM. COMPOS., 00:000–000, 2017. V
2017 Society of Plastics Engineers
Nowadays, in the field of biomaterials, nanocelluloses,
owing to their unique properties, have gained great atten-
tions as novel bio-reinforcements for manufacturing high-
performance bio-nanocomposites . These nanomaterials
naturally exist in plant cell walls and are mainly isolated
from various wooden and non-wooden sources by chemo-
mechanical treatments . Depending on treatment proce-
dures, there exist different types of nanocelluloses with
different morphologies such as cellulose nanofibrils
(CNFs), microfibrillated celluloses (MFCs) [3–5], cellu-
lose nanowhiskers (CNWs) and cellulose nanocrystals
On the other hand, polylactic acid (PLA), an aliphatic
polyester, can be considered as a substitute for most
petroleum-based polymers because of its environmentally
friendly characteristics and good thermo-mechanical prop-
erties. As a whole, PLA obtains from completely renew-
able resources such as starch [9–11], and its monomer,
lactic acid (LA), because of having L-andD-optical enan-
tiomers controls PLA properties . However, although
there exists so many advantages for using PLA instead of
petroleum-based polymers, its mechanical properties have
already some weaknesses limiting use of PLA in some
applications [10, 12]. Therefore, using nanocellulose as a
bio-reinforcement in PLA matrix can be deemed a possible
solution for the mentioned issue. Recently, several studies
have been conducted to investigate the influence of adding
nanocellulose on PLA properties through melt compound-
ing [13, 14], solvent casting [15, 16] and kneading or
calendaring techniques [13, 17, 18]. Nevertheless, after
closer scrutinizing at these studies, a good dispersion of
nanocelluloses into PLA matrix by an industrial-scale
approach is still an ongoing struggle and major challenge.
The first attempt to develop CNF/PLA nanocomposites
was investigated by Chakraborty et al. . In this study,
a water slurry of CNF was directly pumped into PLA melt,
but their results led to a weak dispersion of CNFs in PLA
and undesirable thermo-mechanical properties. Afterward,
Oksman et al. performed several studies on developing
nanocellulose and PLA composites [20–22]. They firstly
used N,N-dimethylacetamide/lithium chloride (DMAc/LiCl)
solution and polyethylene glycol (PEG) in order to
improve the dispersion of CNWs in PLA . Their
results indicated that using DMAc/LiCl solution as separat-
ing agent caused nanocellulose whiskers to degrade at high
Correspondence to: P. Heidarian; e-mail: email@example.com
Published online in Wiley Online Library (wileyonlinelibrary.com).
C2017 Society of Plastics Engineers
temperatures. Moreover, the nonuniform dispersion of
nanocellulose was resulted by adding PEG in PLA . In
another study, untreated, tert-butanol treated, and surfactant
treated CNWs were dispersed through solution casting
technique in PLA to prepare bio-nanocomposites .
Based on the results obtained by this method, the surfac-
tant treated CNWs showed more flow birefringence behav-
ior in chloroform than untreated and tert-butanol treated
nanocelluloses and led to better dispersion of CNWs in
PLA . In another research, polyvinyl alcohol (PVA)
was chosen as a dispersing phase in order to enhance the
dispersion of nanocellulose in PLA. The results revealed
that the mechanical properties of PLA were slightly
increased due to the low dispersion of nanocellulose in
PLA chains . On the other hand, Yano et al. [13, 18]
prepared MFC/PLA nanocomposites using organic solvents
and investigated the influence of adding MFCs in PLA
using kneading process. Their results showed that MFCs
were uniformly dispersed in PLA, and consequently, its
mechanical properties were increased.
Although microscopic images in the aforementioned
studies shed light on a good dispersion of nanocellulose
in PLA, the major drawback was using various and high
contents of chemicals to manufacture PLA nanocompo-
sites. To meet this need, Jonoobi et al.  prepared melt
compounded PLA bio-nanocomposites with developing
MFC/PLA master batch through a twin screw extrusion.
In this way, a good dispersion of MFC in PLA matrix
was achieved which resulted in the desirable thermo-
mechanical properties of PLA bio-nanocomposites.
In this study, CNFs, as the bio-reinforcement, were
firstly isolated from depithed bagasse using an optimized
sulfur-free treatment in various chemo-mechanical stages.
Furthermore, an industrial-scale procedure was used in
order to prepare master batch. For this purpose, to reduce
chemicals used in kneading technique, the master batch
melt-compounding process, and chloroform was eliminated
from the solvent system. Scanning electron microscopic
(SEM) was employed to study the fracture surface of rPLA
and its bionanocomposites. In addition, mechanical and
dynamic mechanical (DMA) properties, as well as biode-
gradability behavior were studied in order to investigate the
effect of different CNF contents on bio-nanocomposites
properties. To the authors’ best knowledge, this is the first
study on producing bio-nanocomposites based on rPLA,
reinforced by bagasse CNFs.
Recycled Polylactic Acid (rPLA) Characterization
Recycled polylactic acid (rPLA) flakes were obtained
from the Center for Biocomposites and Biomaterials Proc-
essing (Toronto, Canada). Gel permission chromatography
(GPC) technique was used to measure the average molecu-
lar weights (M
) and polydispersity index (PDI) of rPLA
based on Marques and coworkers study . To determine
) and glass transition (T
) temperatures, differen-
tial scanning calorimetry (DSC) was employed. In addition,
melt flow index apparatus, pycnometer, and polarimeter
were used to calculate melt flow index (MFI), density and
L=Denantiomer ratio, respectively.
Cellulose Nanofibrils (CNFs) Isolation
Bagasse was obtained from a sugar producer in Khuze-
stan, Iran (Mirza Kuchak Khan). It was first depithed and
cut into small pieces (<20 mm). Next, its non-structural
materials and extractives were extracted by ethanol and
water using a Soxhlet apparatus based on a standard proce-
dure (NREL/TP-510-42619) . Having done this stage, in
order to remove hemicellulose, hot water pre-hydrolysis was
conducted in a stainless steel mini-batch reactor . The
prehydrolysed bagasse was then digested in the same reactor
by a soda-anthraquinone pulping process, obtained through
the current study, at optimized conditions . Finally,
bleaching steps were performed to remove insoluble lignin
according to an elemental chlorine free (ECF) sequence in
three stages using sodium chlorite, acetic acid, hydrogen
peroxide, and sodium hydroxide . Table 1 summarizes
the conditions of the chemical treatments.
To initiate the mechanical steps, the bleached pulp was
firstly refined using a laboratory PFI-mill with 25,000
rotations per minute. Afterward, to produce CNF, a water
slurry of refined fibers (1 wt%) was passed two times
through an ultra-fine friction grinder (MKCA6-3; Masuko
Sangyo, Japan) at 1,500 rpm.
CNF/rPLA Master Batch Preparation. To prepare
CNF/rPLA master batches, the required amount of rPLA
flakes were first dissolved in acetone with a solid to
TABLE 1. Chemical treatment conditions.
Auto hydrolysing Pulping Bleaching
Chemical concentrations (wt %) 0 NaOH (17.5);
Solid to liquor ratio 1:8 1:4 1:10 1:10 1:10
Maximum temperature (8C) 170 150 60 70 80
Time to maximum temperature (min) 90 60 – – –
Time at maximum temperature (min) 10 60 60 90 120
2 POLYMER COMPOSITES—2017 DOI 10.1002/pc
liquid ratio of 1:9 under vigorous stirring for 60 min at
558C. The water slurry of CNFs was then transferred into
acetone by a solvent exchange method through a series
centrifugation at 5,000 rpm for 15 min at 20C (ethyl
alcohol was used as a medium solvent). To avoid the
aggregation and precipitation of CNFs at nonpolar media,
the obtained suspension was re-dispersed by a homogenizer
for 10 min followed by a sonication through an ultrasonic
processor (Misonix, USA) at 100 W for another 5 min in
each stage. Next, an adequate amount of the prepared
CNFs slurry was immediately dispersed in the rPLA solu-
tion, stirred for another 30 min, and sonicated with the
same ultrasonic processor for 5 min. Finally, the obtained
master batch was cast in Petri dishes, moved into a hood
to evaporate excess solvent over the night, and crushed
into small flakes with 15 315 mm
CNF/rPLA Compound Preparation. The rPLA flakes
and CNF/rPLA master batches were dried by a vacuum
oven at 55C for 24 and 8 h, respectively. Afterward, all
compounds with various CNF contents (1, 3, and 5 wt %)
were prepared in two continuous stages including premix-
ing rPLA flakes and CNF/rPLA master batches followed
by melt-compounding them through an internal mixer
(Brabender Plastograph, Germany) at 170C and 60 rpm
for 5 min under a flow of dry nitrogen to minimize any
thermo-oxidative and hydrolysis degradations. Table 2
shows the compounding formulation during the process.
CNF/rPLA Composite Preparation. All CNF/rPLA
compounds were pelletized and vacuum oven dried at
55C for 6 h and compression molded at 160C for 3 min
in 30 bars followed by another 5 min in 50 bars. The
final thickness of rPLA bio-nanocomposite films was
obtained to be 0.3 mm.
Bagasse Fiber and rPLA Characterization
Chemical composition of raw and chemically treated
bagasse fibers was determined by a standard procedure
(NREL/TP-510-42618) . For this purpose, monomeric
sugars were measured based on the dissolution of carbo-
hydrates (cellulose and hemicellulose) under strong acid
hydrolysis condition (H
72% w/w) through an auto-
mated HPLC analyzer (HPLC-RL UV-VIS Detector,
Jasco). Afterward, cellulose and hemicellulose contents
were calculated by the following equations:
Cellulose 5Glucose=1:11 (1)
5Xylose 1Arabinose 1Galactose 1Mannose½=1:13
where glucose, xylose, arabinose, galactose, and mannose
are the monomeric sugars of lignocellulosic materials
. To calculate lignin content, acid-soluble lignin was
obtained by IR spectroscopy in 240 nm peak, and insolu-
ble lignin was determined through solid residue weight
Infrared spectra of rPLA were studied by FTIR spec-
troscopy in attenuated total reflectance (ATR) mode. For
this purpose, the powdered rPLA was characterized with
KBr pellet technique and each spectrum was obtained
within the range of 500–4,000 cm
with a wavelength
resolution of 4 cm
. On the other hand, differential scan-
ning calorimetry (DSC) was used to calculate rPLA ther-
mal properties, including glass transition (T
) and melting
) temperatures. Calorimetric scans were conducted on a
DSC6100 (SII Nanotechnology Inc., Japan) within the tem-
perature range of 25–200C on 8 mg of dry rPLA. The
rPLA thermal properties were analyzed at non-isothermal
conditions at the same heating and cooling rates of
in three scans: heating, cooling, and heating.
The optical purity of rPLA was calculated via an automatic
polarimeter (AA-10R, UK) at 25C and a wavelength of
589 nm, with a concentration of 1 g dL
The following equation was used to calculate the percent-
age of rPLA optical purity where 2156 is the specific
rotation of PLA with just L-enantiomer in PLA structure.
CNF Size Distribution Analysis
To determine CNF size distribution, dynamic light
scattering (DLS) were used in different suspensions by
assuming a spherical structure for CNF. For this purpose,
DLS measurements were performed on each sample with
the solid content of about 0.1 wt %. All samples were
sonicated by ultrasonication (Misonix- S 3000, USA) in a
detection angle of 1738at 258C using a 4 mW He-Ne
laser operating at a wavelength of 633 nm. The analysis
was carried out in triplicate for each sample and the aver-
age value was reported.
Fiber and Bionanocomposite Morphology
The morphology of fibers and the fracture surface of
bio-nanocomposites were observed by scanning electron
microscopy (SEM, 1450EP, Zeiss, Germany). For this
TABLE 2. Formulations in the processing and in the final
PLA (g) CNF (g)
PLA (g) Final composition (%)
rPLA 10 2150 40 rPLA (100)
rPLA/CNF1 9.5 0.5 150 40 rPLA/CNF (99/1)
rPLA/CNF3 8.5 1.5 150 40 rPLA/CNF (97/3)
rPLA/CNF5 7.5 2.5 150 40 rPLA/CNF (95/5)
DOI 10.1002/pc POLYMER COMPOSITES—2017 3
purpose, fibers and the cross-section of bio-nanocomposites
were coated by gold through vacuum coater and observed
under SEM operating at 20 and 10 kV acceleration vol-
tages. Afterward, the diluted suspension of CNF with the
solid content of about 0.05–0.1 wt% was well dispersed in
ethyl alcohol by an ultrasonic (Misonix- S 3000, USA),
cast on a perforated carbon coated grid Cu Mesh 300, and
observed via a transmission electron microscopy (TEM,
EM10C-80Kv, Zeiss, Germany). Finally, TEM images
were analyzed by an image tool analyzer (UTHSCSA) to
determine CNF diameters.
Bionanocomposite Mechanical Properties
Tensile analysis was conducted in order to investigate
the influence of CNF on the mechanical properties of
bio-nanocomposites, including ultimate strength, elonga-
tion at break, Young’s modulus, and work of fracture
using a universal testing machine (Zwick, 1446-60,
Germany) according to ASTM D638 Type 5 with a load
cell of 10 kN and a crosshead speed of 3 mm min
least, four samples were analyzed for each test and the
obtained average values were reported.
Bionanocomposite Dynamic Mechanical (DMA)
The dynamic mechanical properties of bio-nanocomposites
were studied by a mechanical analyzer (DMA-Triton, UK) in
tensile mode. To meet this need, the measurements were per-
formed at a constant frequency of 1 Hz and a strain amplitude
of 0.05% with a temperature range of 408Cto1208Cata
heating rate of 38Cmin
. The strips of samples were cut
from the films with 5 35cm
Bionanocomposite Biodegradation Analysis
The biodegradation analysis of bio-nanocomposites
was carried out according to Mathew et al.  in com-
posting condition. Specimens of bio-nanocomposites with
dimensions and 0.3 mm thickness were buried
in soil. Soil conditions were set at 60% humidity and
temperature about 58 62C. The weight loss of bio-
nanocomposites was measured after a period of time (14,
30, 60, and 75 days) and the digital images of specimens
RESULTS AND DISCUSSION
The rPLA characterization results are presented in
Table 3. As mentioned above, LA monomer, the PLA
building block, contains D- and L- enantiomers. It is found
that, the proportion of these enantiomers influences the
thermo-mechanical properties of PLA . For instance, L-
enantiomer controls the crystallinity of PLA, so high L/D
ratio means a highly crystallized PLA which is favorable
for melt-compounding process . In this study, the L/D
ratio was found to be 90/10 meaning that rPLA contains
90% crystalline enantiomer.
During melt-compounding process, having information
are of importance since temperature
must be set greater than these temperatures so as to pro-
vide a homogenous compound. In this study, T
were found to be around 60C and 156C (Fig. 1a). It is
worth mentioning that similar T
obtained for PLA with 90/10 enantiomer ratio [9, 31].
According to Fig. 1a, characterized by Day et al. , it
can be observed that the T
were decreased from
169 and 62C to 156 and 60C in comparison with non-
recycled PLA showing that recycling process declined
. These changes can be originated from
increasing oligomers which trigger the mobility of shorter
chains at low temperatures, as well as enhancing chain-
TABLE 3. Characteristics of rPLA.
Property Value Unit Measurement method
PDI 2.9 – GPC
33.5 – GPC
60 8C DSC
156 8C DSC
120 8C DSC
MFI 9 gr/10 min MFI
Density 1.25 gr cm
L=Dratio 90/10 – Polarimeter
FIG. 1. (a) DSC results for rPLA and PLA, (b) FTIR results for rPLA
4 POLYMER COMPOSITES—2017 DOI 10.1002/pc
ends per mass which boosts the mechanism of backbiting
reactions . Moreover, the rPLA M
was found to be
around 3.5 310
. A similar result was also reported for
the processed PLA with 90/10 enantiomer ratio after 30
min at 160C . In addition, PDI was found to be 2.9
proving the presence of oligomers in consequence of
depolymerization and/or hydrolytic processes in rPLA.
Figure 1b shows the FTIR spectra of neat PLA and rPLA.
As can be seen, in comparison with neat PLA, not only
no significant change can be observed in rPLA, but also
most peaks are in the same order; for instance, dominant
peaks around 3000 cm
belong to OAH bond in both
PLA and rPLA. Furthermore, absorbance peaks between
2200 and 2500 cm
attribute to symmetric and antisym-
metric CAH bonds, located on CH
observed at 1700, 1500, and 1200 cm
assign to C@O
bonds, the stretching vibration of CH
CAOAC bonds, respectively . Therefore, no consider-
able chain scission at CAO bond can be found in rPLA
chains; however, the intensity of OAH bond was
decreased which might be referred to the hydrolytic deg-
radation of rPLA. According to Al-Itry et al.  the pos-
sible mechanism of thermal degradation in PLA are
thermal hydrolysis and random main-chain scission.
TABLE 4. Chemical composition results.
Raw fibers 42.2 24.2 23.5 10.1
Bleached fibers 91.8 7.38 0.82 0
FIG. 2. (a) Microscopic images of microfibers, (b) CNF, (c) DLS analysis of hydrodynamical diameters of
water suspended CNF, (d) acetone suspended CNF. [Color figure can be viewed at wileyonlinelibrary.com]
DOI 10.1002/pc POLYMER COMPOSITES—2017 5
Fiber and CNF Chemical Composition and Morphology
Chemical composition results are summarized in Table
4. As can be seen, the chemical treatments not only suc-
cessfully isolated cellulose microfibers with an average
diameter of about 5.9 61lm (Fig. 2a), but also they
removed the major portion of noncellulosic materials,
including hemicellulose, lignin, and extractives from
bagasse fibers. Therefore, after having completed chemi-
cal step, cellulose percentage was increased up to 91.8%,
and hemicellulose, lignin, and extractives were decreased
to 7.38, 0.82, and 0%, respectively.
Figure 2b shows a TEM image from the isolated bagasse
CNF. As can be observed, this image reveals a web-like
morphology from CNF with an average diameter of 35 nm,
calculated by an image tool analyzer (UTHSCSA), so it can
be claimed that the proposed chemo-mechanical treatments
were completely efficacious to isolate CNF.
Dynamic Light Scattering (DLS)
The aggregation of nanofibers can be considered as the
main challenge during re-dispersing CNF in nonpolar sus-
pensions as it causes to increase CNF dimensions and con-
sequently limits its nano-scale reinforcing characteristic
[34, 35]. In this study, DLS analysis was therefore per-
formed to quantitatively evaluate the CNF dimensions in
various suspensions (water, acetone, and chloroform).
According to Fig. 2c, the size distribution curve of the
water suspended CNF disclosed a single peak at 860 nm
with the average hydrodynamic diameter of 820 nm. In
addition, two sharp peaks appeared for the acetone sus-
pended CNF at 791.6 and 4166 nm with the average
hydrodynamic diameter of 1558 nm, and no peak was
detected for the chloroform suspended CNF, revealing no
nano-scale dimensions for CNF. Based on the information
provided by this test, acetone was then used to prepare
Bio-nanocomposite fracture surfaces are shown by
SEM micrographs (Fig. 3). As can be observed, while
neat rPLA (Fig. 3a) depicted a smooth fracture surface,
its roughness was gradually increased by enhancing the
CNF percentage (Fig. 3b–d). In addition, by looking at
SEM images, it can be noticed that bionanocomposites
containing 1 and 3 wt% CNF have a homogenous mor-
phology unveiling the good dispersion of nanofibers in
FIG. 3. SEM images of fracture surface: (a) rPLA, (b) rPLA/CNF (1 wt%), (c) rPLA/CNF (3 wt%), (d)
rPLA/CNF (5 wt%).
6 POLYMER COMPOSITES—2017 DOI 10.1002/pc
FIG. 4. (a) Tensile stress–strain curves, (b) photographs of CNF/rPLA, (c) Young’s modulus, (d)
elongation at break, (e) tensile strength, (f) work of fracture. [Color figure can be viewed at wileyonline-
FIG. 5. DMA analysis of samples: (a) log of storage modulus, (b) tan delta.
DOI 10.1002/pc POLYMER COMPOSITES—2017 7
their matrices (Fig. 3b and c). This observation could be
explained by good interfacial interactions between cellu-
lose fibers and polymer matrix up to a specific percentage
. However, by increasing CNF percentage to 5 wt%
(Fig. 3d), nanofiber aggregations became dominant caus-
ing to increase numerous voids in rPLA matrix and stress
concentration points which would trigger growing micro
cracks in bio-nanocomposites during mechanical stresses.
Approximately, similar results were found by Jonoobi
et al. using neat PLA .
Bionanocomposite Mechanical Properties
Figure 4a shows the typical stress-strain curves for
rPLA ant its nanocomposites. As can be seen, using CNF
improved the ultimate strength and modulus of rPLA
even at low percentage (1 wt%). This observation could
directly be related to the superb crystallinity and high sur-
face area of CNF  which could be entangled with
polymer chains . The digital images of samples are
also shown in Fig. 4b. It can be seen that the opacity of
samples was increased by increasing the CNF percentage.
According to Fig. 4c and d, it can be observed that add-
ing 3 wt% CNF increased the rPLA modulus and ultimate
strength from 527.59 and 23.9 MPa to 716.5 and 32.6
TABLE 5. Storage modulus and tan delta peak temperatures of the
rPLA and its nanocomposites.
at 25 C8(MPa)
at 70 C8(MPa)
Tan delta peak
rPLA 821.4 650 3.875 616660
rPLA/CNF1 990.4 660 4.771 62 66.4 60
rPLA/CNF3 1002 650 8.803 64 67.6 61
rPLA/CNF5 974.5 640 7.95 63 66.1 61
FIG. 6. Different stages of biodegradation. [Color figure can be viewed at wileyonlinelibrary.com]
8 POLYMER COMPOSITES—2017 DOI 10.1002/pc
MPa, respectively, corresponding to 35.8 and 36.4%
improvement. These results also indicated that CNF was
uniformly dispersed into rPLA matrix up to 3 wt%.
Approximately, similar trends have been reported for
mechanical properties by adding 3 wt% CNF into the
neat PLA [5, 13, 18]. Furthermore, elongation at break
and work of fracture were reduced by 35.5 and 33% (Fig.
4e and f) at this CNF percentage.
Dynamic Mechanical Properties (DMA) Analysis
To investigate the thermo-mechanical properties of bio-
nanocomposites, dynamic mechanical analysis (DMA) was
employed. For this purpose, tan delta and storage modulus
against temperature are plotted in Fig. 5. Based on the
results, it was found that rPLA storage modulus and its
bio-nanocomposites were decreased in transition tempera-
ture around 608C. According to Fig. 5a, the rPLA T
its nanocomposites containing 1 and 3 wt% CNF were
found to be 59.5, 61.9, and 648C, respectively. One reason
for the observed positive shift might be referred to the seg-
mental immobility of polymer chains because of increasing
the rigid states between rPLA and CNF. Similar trends
have also been observed by Jonoobi et al. . In addition,
due to the cold crystallization of amorphous chains in
rPLA, the storage modulus of samples was increased again
at a temperature around 1008C [18, 29]. Moreover, the tan
delta peak was slightly shifted towards higher temperatures
by adding CNF (Fig. 5b). The observed increment might
also be related to entangling CNF and rPLA chains ,
restricting segmental motions in rPLA chains, and requir-
ing wider temperature span for chain transitions [29, 38].
Table 5 shows the thermo-mechanical results of biona-
nocomposites in more details. The results revealed that by
loading 3 wt% CNF into rPLA matrix, the storage modulus
of rPLA was increased from 821.4 and 3.875 MPa to 1002
and 8.803 MPa in glassy and rubbery regions, correspond-
ing to 24.66 and 127.17% improvements, respectively.
This could be related to homogenous dispersion of CNF in
rPLA matrix, immobility of rPLA chains, and improve-
ment of rPLA crystallinity.
Biodegradability behavior of bio-nanocomposites was
investigated based on Mathew et al. study . According
to their results, the high rate of PLA hydrolytic degradation
is obtained at 60% humidity and 58C62C, and the main
driving force for this hydrolytic degradation is the water
uptake ability of samples, which causes to cleave ester
groups, form oligomer fractions, solubilize oligomers, and
diffuse soluble oligomers through bacteria [10, 29, 39].
Figures 6 and 7 show the degradation of samples after dif-
ferent periods, and the residual weight percentage (R
bio-nanocomposites versus time (days).
As can be seen, after 30 days, the hydrolytic degradation
and brittleness of samples were started in consequence of
diffusing water from the soil . Moreover, the weight loss
percentage of rPLA and bio-nanocomposites after 60 days
was analysed. It was found that after 60 days, the residual
weight percentage of samples, especially for rPLA, was
decreased. Mathew et al. obtained similar results for their
composites . It can be explained that the less degradation
rate of bionanocomposites was due to the more diffusion
resistance ability of them compared to the unreinforced poly-
mer samples. The highest degradation rate of rPLA was
reached after 75 days. It can be suggested that the water
absorption process begins from the amorphous domains in
polymer, and the presence of nanocellulose between these
areas delays the degradation of these bionanocomposites
. In this work, the low biodegradability rate can be
related to the good dispersion of reinforcement in the biona-
nocomposite with 3 wt% CNF content. This result is in good
agreement with the obtained mechanical properties.
Recycled polylactic acid (rPLA) was successfully rein-
forced by bagasse cellulose nanofibrils (CNFs) with
92% cellulose percentage and an average diameter of
about 35 nm through an industrial-scale approach. The
fracture surface of bio-nanocomposites showed that sam-
ples containing 1 and 3 wt% CNF were well dispersed in
rPLA, and this feature led to improving the strength and
modulus of the rPLA, studied by tensile analysis. More-
over, by adding 3 wt% CNF, the storage modulus of
rPLA was enhanced up to 24.66 and 127.17% in glassy
and rubbery regions. In addition, by adding CNF the deg-
radation rate of rPLA was decreased compared to unrein-
forced rPLA due to the good interfacial interactions
between CNF and rPLA chains.
The authors thank Center for Biocomposites and Bioma-
terials Processing (Toronto, Canada) for supplying rPLA.
FIG. 7. Residual weight percentage (R
%) of the samples vs. time.
[Color figure can be viewed at wileyonlinelibrary.com]
DOI 10.1002/pc POLYMER COMPOSITES—2017 9
1. M.A. Hubbe, O.J. Rojas, L.A. Lucia, and M. Sain, Biore-
sources,3, 3 (2008).
2. A. Dufresne, Natural Polymers: Volume 2 Nanocomposites.
The Royal Society of Chemistry, London (2012).
3. A. Alemdar, and M. Sain, Bioresour. Technol.,99, 1664 (2008).
4. W. Chen, H. Yu, Y. Liu, Y. Hai, M. Zhang, and P. Chen,
Cellulose,18, 433 (2011).
5. M. Jonoobi, J. Harun, A. Shakeri, M. Misra, and K.
Oksmand, Bioresources,4, 626 (2009).
6. N. Johar, I. Ahmad, and A. Dufresne, Ind. Crop. Prod.,37,
7. A. Mandal, and D. Chakrabarty, Carbohydr. Polym.,86,
8. K. Oksman, J.A. Etang, A.P. Mathew, and M. Jonoobi, Bio-
mass. Bioenerg.,35, 146 (2011).
9. R.A. Auras, L.-T. Lim, S.E. Selke, and H. Tsuji, Poly (lac-
tic acid): Synthesis, Structures, Properties, Processing, and
Applications, Wiley (2011).
10. J.-M. Raquez, Y. Habibi, M. Murariu, and P. Dubois, Prog.
Polym. Sci.,38, 1504 (2013).
11. Y.-F. Shih, C.-C. Huang, and P.-W. Chen, Mater. Sci. Eng.,
A,527, 1516 (2010).
12. R. Liu, Y. Peng, and J. Cao, Polym. Compos.,37, 1971 (2015).
13. A. Iwatake, M. Nogi, and H. Yano, Compos. Sci. Technol.,
68, 2103 (2008).
14. M. Jonoobi, J. Harun, A.P. Mathew, and K. Oksman, Compos.
Sci. Technol.,70, 1742 (2010).
15. D. Liu, X. Yuan, D. Bhattacharyya, and A. Easteal, Express
Polym. Lett.,4, 26 (2010).
16. P. Qu, Y. Gao, G.-F. Wu, and L.-P. Zhang, BioResources,
5, 1811 (2010).
17. K. Okubo, T. Fujii, and E.T. Thostenson, Compos. Part A
Appl. S,40, 469 (2009).
18. L. Suryanegara, A.N. Nakagaito, and H. Yano, Compos. Sci.
Technol.,69, 1187 (2009).
19. A. Chakraborty, M. Sain, and M. Kortschot, Holzforschung,
59, 102 (2005).
20. K. Oksman, A. Mathew, D. Bondeson, and I. Kvien, Compos.
Sci. Technol.,66, 2776 (2006).
21. L. Petersson, I. Kvien, and K. Oksman, Compos. Sci. Tech-
nol.,67, 2535 (2007).
22. D. Bondeson, and K. Oksman, Compos. Interfaces,14, 617
23. D.A. Marques, S. Jarmelo, G. Baptista, M. Cristina, and M.
Gil, Macromol. Symp.,296, 63 (2010).
24. A. Sluiter, R. Ruiz, C. Scarlata, J. Sluiter, and D. Templeton,
Determination of extractives in biomass. National renewable
energy laboratory (NREL). Available at: http://www.nrel.gov/
biomass/analytical_procedures. html. Accessed 17 July 2005
25. Y. Hamzeh, A. Ashori, Z. Khorasani, A. Abdulkhani, and
A. Abyaz, Ind. Crop. Prod.,43, 365 (2013).
26. P. Heidarian, T. Behzad, and K. Karimi, Wood Sci. Tech-
nol.,50, 5 (2016).
27. A. Sluiter, B. Hames, R. Ruiz, C. Scarlata, J. Sluiter,
D. Templeton, and D. Crocker, Determination of structural
carbohydrates and lignin in biomass. National renewable
energy laboratory (NREL). Available at: http://www.nrel.
gov/biomass/analytical_procedures.html. Accessed 17 July
28. F. Monlau, A. Barakat, J. Steyer, and H. Carrere, Bioresour.
Technol.,120, 241 (2012).
29. A.P. Mathew, K. Oksman, and M. Sain, J. Appl. Polym.
Sci.,97, 2014 (2005).
30. L.-T. Lim, R. Auras, and M. Rubino, Prog. Polym. Sci.,33,
31. D. Garlotta, J. Polym. Environ.,9, 63 (2001).
32. M. Day, A. Nawaby, and X. Liao, J. Therm. Anal. Calorim.,
86, 623 (2006).
33. A.J.R. Lasprilla, G.A.R. Martinez, and B. Hoss, Chem.
Eng.,24, 985 (2011).
34. A. Ferrer, E. Quintana, I. Filpponen, I. Solala, T. Vidal, A.
ıguez, J. Laine, and O.J. Rojas, Cellulose,19, 2179
35. K. Dimic-Misic, A. Puisto, P. Gane, K. Nieminen, M.
Alava, J. Paltakari, and T. Maloney, Cellulose,20, 2847
36. S. Wang, Y. Lin, X. Zhang, and C. Lu, RSC Adv.,5,63
37. M. Jonoobi, A.P. Mathew, M.M. Abdi, M.D. Makinejad,
and K. Oksman, J. Polym. Environ.,20, 991 (2012).
38. K.M.Z. Hossain, I. Ahmed, A.J. Parsons, C.A. Scotchford,
G.S. Walker, W. Thielemans, and C.D. Rudd, J. Mater.
Sci.,47, 2675 (2012).
39. B.S. Ndazi, and S. Karlsson, Express Polym. Lett.,5, 119
10 POLYMER COMPOSITES—2017 DOI 10.1002/pc