Accepted Manuscript

Enhancement of Cellulose Acetate Degradation under Accelerated Weathering by

Plasticization with Eco-Friendly Plasticizers
Robert Quintana, Olivier Persenaire, Yahia Lemmouchi, John Sampson, Stuart
Martin, Lela Bonnaud, Philippe Dubois
PII: S0141-3910(13)00203-6
DOI: 10.1016/j.polymdegradstab.2013.06.032
Reference: PDST 7031
To appear in: Polymer Degradation and Stability
Received Date: 22 September 2012
Revised Date: 24 June 2013
Accepted Date: 27 June 2013
Please cite this article as: Quintana R, Persenaire O, Lemmouchi Y, Sampson J, Martin S, Bonnaud
L, Dubois P, Enhancement of Cellulose Acetate Degradation under Accelerated Weathering by
Plasticization with Eco-Friendly Plasticizers, Polymer Degradation and Stability (2013), doi: 10.1016/
j.polymdegradstab.2013.06.032.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
Enhancement of Cellulose Acetate Degradation under
Accelerated Weathering by Plasticization with Eco-Friendly
Plasticizers

Robert Quintana
1,2
, Olivier Persenaire
1
, Yahia Lemmouchi
3
, John Sampson
3
, Stuart Martin
3
,
Lela Bonnaud
1
, Philippe Dubois
1,*


1
Materia Nova asbl, Parc Initialis, Avenue N. Copernic 1, and University of Mons UMONS,
Place du Parc 20, 7000 Mons, Belgium.
2
Institute of Materials Research and Engineering (IMRE), A*STAR,
3 Research Link, Singapore 117602
3
Group R&D Centre, British American Tobacco, Regents Park Road, Millbrook,
Southampton SO15 8TL, UK.

Abstract
Cellulose acetate (CA) with a degree of substitution (DS) of 2.5 has been plasticized using eco-
friendly plasticizers such as triacetin, tripropionin, triethyl citrate, tributyl citrate, tributyl 2-acetyl
citrate and poly(ethylene glycol) of low molecular weight. Thermo-mechanical properties and
hydrophilicity of the modified CA have been measured and correlated with the content and
nature of the plasticizer used and compared with unplasticized CA. The increase in toughening
and the change in the hydrophilicity by the plasticization were evaluated in terms of aging and
weathering stability under accelerated conditions. Samples were exposed to UV-degradation
with water spray periods. The treated samples were removed periodically and characterized by
several analytical techniques. The results are discussed with particular emphasis toward the
effects of plasticization on enhancement of the degradation rate of CA. The plasticization of CA
triggered an increase of the weight loss between 50 and 90%, where low molecular weight
plasticizers were shown to be more effective. A right balance between hydrophilicity and
plasticization efficiency (reduction of T
g
) is needed to increase the degradation rate of CA.


Keywords: cellulose acetate, degradation, weathering, plasticization, blend.
Acknowledgements: LPCM is very much indebted to the Rgion Wallonne and
European Community for financial support (FSE-FEDER) in the frame of Objectif 1,
Phasing-out: Materia Nova and Plan de convergence. LPCM thanks the Belgian
Federal Government Office Policy of Science for general support in the frame of the
PAI 6/27. British American Tobacco financial support is acknowledged.

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
* Corresponding author. Address: University of Mons UMONS, Place du Parc 20, 7000 Mons,
Belgium. Fax: +32 65 373484; E-mail address: philippe.dubois@umons.ac.be (Ph. Dubois)
1. Introduction
One of the most important organic esters is cellulose acetate (CA), which is
widely used for many industrial applications in the form of film and fibre [1]. CA is
derived from cellulose using the acetylation process and its main properties are its
hardness, good resistance to impact, high shine, transparency, pleasing texture, lack of
static electricity and resistance to hydrocarbons [1, 2]. Furthermore, CA has recently
been reported to be potentially biodegradable [3, 4] and that the biodegradation rate
depends on its degree of acetyl-group substitution (DS). As DS of CA decreases,
biodegradation rate increases. However, for some applications, a faster degradation
rate is desirable in minimising the impact of CA products in the environment.
Plasticizers have long been known for their effectiveness in producing flexible
plastics for applications ranging from the automotive industry to medical and consumer
products. Plasticizers are used for biodegradable plastics as well. There are more
stringent requirements on these plasticizers, though, since by definition they will be
released into the environment during product use. Health and safety issues dominate
the research in this field, with benign, often natural substances having nearly as great
importance as the ability to lower the T
g
of the polymer. Because of the prevalence and
commercial applications of biodegradable polymers, recent research has focused on
developing compatible plasticizers that also biodegrade [5,6]. Citrate-based plasticizers
can be derived from naturally occurring citric acid. They are non-toxic and are used as
plasticizers with some biodegradable polymers [7-11]. Ghiya et al. melt blended
cellulose acetate (CA) with two different citric acid esters: triethyl citrate and acetyl
triethyl citrate. The authors observed that both plasticizers are miscible with CA and the
addition of plasticizer reduces the tensile modulus and increases the elongation of CA.
The biodegradation rates are increased dramatically with an increase in plasticizer
content [12]. Simple triglycerides as triacetin can be used as a food additive and is
already applied to cigarette filter as a plasticizer [13]. The high boiling point of triacetin
compared to citrate-based plasticizers (Table 1) reduces the amount of plasticizer loss
occurring during melt processing, especially with the lower MW citrates [10].
Table 1

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT

Polyols are another class of compounds which have been studied as
plasticizers for biodegradable polymers. Glycerol, which is often used with
biodegradable polymers, was found to reduce thermal degradation of thermoplastic
starch (TPS) reinforced with cellulosic fibers [14]. However, the lower miscibility of
glycerol in acetone reduces its application by solvent blending with CA. Poly(ethylene
glycol) (PEG) is a flexible and water-soluble polymer and also soluble in acetone. Rosa
et al. reported that blending CA with PEG(400) or PEG(1500) decrease the T
g
of CA
and enhances the tensile strength of CA [15]. The authors pointed out the effect of the
interactions between the free hydroxyl groups of PEG and the chains of CA as an
explanation of the enhanced the resistance of CA [15].
The present study focused on the use of eco-friendly plasticizers as modifiers
for CA2.5 (DS=2.5) and their effects on thermo-mechanical properties and weathering
stability of the related polymer materials. To our knowledge, the relationship between
CA plasticization and weathering stability has never been reported. The aim of the
study was to increase the degradation rate of CA by blending CA with water soluble
plasticizers.

2. Experimental
2.1. Materials
Cellulose Acetate (flakes, M
n
=65.400 and polydispersity index of 3.4) (CA) with
a degree of substitution of 2.5 was kindly supplied by British American Tobacco.
Triacetin (TA) was purchased from Acros. Tripropionin (TP), triethyl citrate (TeC),
tributyl citrate (TbC) and tributyl 2-acetyl citrate (Tb2C) were purchased from Merck
and Poly(ethylene glycol) 200 (PEG) from Fluka. These chemicals and solvents used in
this study were all guaranteed reagent grade and used without further purification.

2.2. Procedures

2.2.1. Film Preparation by solvent casting and melt processing
Films of blends of CA and plasticizers were prepared by both solvent casting
and melt processing methods. The films were cast by being drawn on Petri dishes 10%
solutions (w/w) of the materials in acetone. The thickness of the films was
approximately 0.5 mm after the removal of the solvent. Melt blending of CA (dried
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
overnight at 105C in a ventilated oven) with plast icizers was performed in a Brabender
internal mixer (model 50 EHT, 80 cm free volume) equipped with cam blades at 235
C and 60 rpm during a total time of 10 minutes. 500-m films were prepared,
immediately after blend preparation, by hot-pressing moulding at 235 C using an Agila
PE20 hydraulic press (low pressure for 120 s without degassing cycle, followed by a
high-pressure cycle at 150 bars for 180 s and cooling by tap water at 50 bars for 120
s).

2.2.2. Aging
The plasticized CA films were boxed and aged at room conditions for 3 months.
The films were then characterized with MDSC in order to investigate the stability of the
blends.

2.2.3. Accelerated weathering
Accelerated weathering testing was performed in a Q-Sun Xenon arc test chamber with
a wavelength range of 295 800 nm and the Daylight-Q filter. The Daylight-Q filter
produces a spectral power distribution equivalent to noon summer sunlight. The
irradiance setting used was 0.65 W/m
2
calibrated at 340 nm. The irradiance is relatively
homogeneous in the chamber (only around a 10% drop from the maximum exposure in
the centre of the shelf to the periphery). Square (3x3 cm) specimens were exposed to
0.65 W/m
2
at a chamber temperature of 50 C and a relative h umidity of 50% up to
475 h. The samples were water sprayed during 30 minutes of each 5 hour cycle.
Samples were placed perpendicular to the irradiation into sample cups of 92 mm of
diameter with three films placed in each cup. After each time period, a film was
removed for further characterization.
2.3. Measurements
The number average molecular weights and molecular weight distributions of
the polymers were determined in THF at 23C using a Agilent size exclusion
chromatograph equipped with a Knauer 2320 refractometer index detector and two
PLGel columns (MIXED-D and 103A). Samples were dissolved in THF (5 mg/1 ml), 20
L of the solutions were injected into the columns using a flow rate of 1 mL/min.
Monodisperse polystyrene standards (Polymer Laboratories Ltd.) were used for the
primary calibration.

Differential scanning calorimetry (DSC) measurements were performed by
using a DSC 2920 from TA Instruments calibrated with indium under nitrogen flow (50
mL/min). The following procedure was used: first heating at 20 C/min from room
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
temperature to 150 C, keeping at this temperature for 60 s to eliminate the
dehydration, which occurs around 120 C [16, 17]. Then the temperature was reduced
to 30 C at 10 C/min and finally a heating scan at 10 C/min to 270 C.
Modulated DSC spectra were obtained using a MDSC 2920CE from TA
Instruments calibrated with indium under nitrogen flow (50 mL/min). The samples was
heated at 20 C/min from room temperature to 150 C and held there for 60 s. Then the
temperature was reduced to 30 C at 10 C/min and finally the sample was heated to
270 C at 5 C/min while applying a temperature oscillation of 1 C/min.

Thermogravimetric analyses (TGA) were performed by using a TGA Q500 (TA
instruments) with a heating rate of 20 C/min in air, from room temperature to 800 C
(platinum pans, 60 cm
3
/min air flow rate). High resolution TGA analyses were
performed on a Hi-Res TGA 2950 from TA Instruments, using nitrogen as purge gas
and a resolution parameter of 5, which means that a continuously variable heating rate
is applied in response to changes in the sample decomposition rate. This resolution
parameter can be tuned within an eight-step scale to maximize weight loss resolution.

Tensile testing measurements were performed by using a Lloyd LR 10K tensile
bench in accordance with ASTM D 882 "Standard Test Method for Tensile Properties
of Thin Plastic Sheeting", standard at room temperature using a crosshead speed of
20 mm/min and a distance of 25.4 mm between grips. Rectangular test specimens (64
x 10 mm) were cut from 500-m films and were previously conditioned for at least 48
hours at 201C under relative humidity of 45 5 % and values were averaged over
five measurements.

The static contact angle of water drop deposited onto film surface was
measured using a drop shape analysis system (DSA 10 MK2, Krss) at 25 C. A drop
of deionized water (20 L) was placed onto the sample surface and the images of the
water menisci on the sample surface were recorded through a digital camera. These
images were analyzed by DSA software to yield the contact angle values. A total of 4
measurements on different areas of the surface were averaged.

3. Results and discussion
3.1. Plasticization of CA
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
Table 2 presents the molecular structure of the eco-friendly plasticizers used to prepare
plasticized cellulose acetate. Two families of low molecular weight plasticizers have
been considered: citrate esters and simple triglycerides, together with poly(ethylene
glycol) of low molecular weight. Good clear and transparent films of neat CA and CA
blends with 10 and 20 wt% of the selected plasticizers were prepared by either solvent
casting in acetone or by melt processing methods. Only when tributyl 2-acetyl citrate
was used white and opaque films were obtained by casting. Further DSC analysis
demonstrated the presence of a crystalline melting peak. The macroscopic bending
presented on these films was then attributed to contraction at the film surface produced
by crystalline phase formation during solvent evaporation. However, those problems
were avoided when films were prepared by melt processing. Indeed, clear transparent
films were easily prepared.

Table 2
3.2. Tensile properties
Plasticization is assumed to increase the flexibility of chains and lead to a
decrease in the stiffness. The strength usually also decreases, whereas deformability
becomes stronger at the same time. The effect of PEG concentration on the
mechanical properties of CA/PEG films has been investigated by tensile tests. Figure 1
presents the Youngs modulus, stress at break and elongation at break results for the
complete series of CA/PEG blends (0, 10, 15 and 20 wt%).
Figure 1

The effect of plasticizing is clearly observed with an increase of the ductility of
the films containing PEG. However, no significant differences on tensile properties
were observed between the three compositions studied. This result is discussed later
on in terms of miscibility threshold. For this reason, we have focused the study only on
10 and 20% compositions for the other plasticizers. Then, the tensile properties of
CA2.5/plasticizer films with 10 and 20 wt% of plasticizer content and prepared by either
solvent casting or melt processing method are compared in Table 3. As observed, the
main differences between results obtained by film casting and melt processing concern
the tensile strength. However, observed tendencies in function of the nature of the
plasticizer remained unaffected. Comparison with neat CA2.5 results reveals that
blends are slightly more ductile and have similar ultimate properties.
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
Table 3
Taking into account the effect of plasticizer concentration, PEG and TP are
more effective at low concentration and blends prepared with TA and TeC are more
sensitive to plasticizer concentration. At the 20 wt% content the final properties of the
films are almost independent of the plasticizer.
3.3. Thermal properties and hydrophilicity.
Thermal properties (via DSC and TGA) and hydrophilicity (via contact angle
measurements) of the plasticized cellulose acetates have also been studied. Table 4
summarizes the transition temperature and contact angle values for the blends
examined in this study, together with the corresponding data for the neat CA as a
reference sample.
Table 4
The results of the DSC measurements confirm the miscibility of the plasticizers
with CA. T
g
shifted as

the plasticizer content increased, with a single T
g
, which indicated
miscibility with CA [18]. The low molecular weight plasticizers are able to decrease T
g

of CA in a range of 60 and 80 C. However, at 10 wt% all plasticizer showed the same
effect on T
g
, a decrease of 30 and 40 C. Furthermore, for these plasticizers a
proportional relationship of approx. of 4 C/wt% between concentration and T
g
drop can
be observed. It is well known that PEG with a MW=400 is miscible with CA at this
concentration range [18-20]. However, our results indicate that a reduction is reached,
about 35 C, but shows no dependence on the PEG content. For this reason, we
interpret our data in terms of miscibility threshold: at a concentration lower than 20 wt%
the miscibility threshold was reached, implying that these films could be over-
plasticized. This interpretation agrees with data reported by Yuan et al. [20] regarding
the preparation of CA/PEG blends using PEGs of different molecular weights (400,
1000 and 3350). Their results indicate that increasing the molecular weight of PEG
permits to increase the content on PEG in the blend without incurring on film over-
plasticization.
The thermal stability of plasticized CAs is not strongly depending on the nature
of the plasticizer as revealed by oxidative TGA results. Contact angle measurements
included in Table 4 were conducted only few days (1 to 3 days) after film preparation.
In this study, the contact angles are used as a measure to follow the changes in
surface hydrophilicity due to the type and concentration of the plasticizer. Although all
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
plasticizers studied are water soluble, only PEG is considered as hydrophilic
plasticizer. This characteristic is reflected on CA blends prepared with PEG which
shown a significant increase of surface hydrophilicity. In the other hand, the low
molecular weight plasticizers studied are hydrophobic, with water solubility ranging
from 30 to 65 g/L at room conditions following the trend TP<TATeC. The lowest
solubility of TP regarding TA is related to the longest chain from the corresponding fatty
acid of the former. In all cases, these hydrophobic plasticizers increased surface
hydrophobicity of the blend. These data together with the decrease of Tg discussed
above evidence the positive performance of hydrophobic low weight compounds as
effective plasticizers of hydrophilic CA2.5. Stability of the CA blend films at room
temperature has been examined by MDSC analysis at 1 and 3 months after the
preparation. Table 5 summarizes the MDSC data obtained from the first heating scan
from 30 to 270 C.
Table 5
As Table 5 shown, only very small differences could be measured for CA/PEG blend,
as T
g
increase up to 2-6 C. This loss of plasticization could be explained as an effect
due to migration of the plasticizer to the surface of the film which has been reported for
PEG of higher molecular weight (i.e. 600) [18]. Results could be also correlated with
the over-plasticized interpretation of CA/PEG films suggested before. In contraposition,
the stability of films containing the low molecular weight plasticizers was not
significantly altered by time.

3.4. Accelerated weathering
Film degradability was studied by using an in-house weatherometer method,
which uses UV light and water to simulate an accelerated degradation. Selected film
samples were subjected to weathering conditions up to 475 hours. Figure 2 shows the
visual appearance of series of degraded films after 225 and 425 hours, together with
the original ones for comparison. Change in visual appearance of the films could be
clearly correlated with the content and the nature of the plasticizer. Table 6 reports the
remaining weight of the films as a function of the weathering time, for up to 425 hours;
after this time, it was not possible to correctly determine the weight of the samples due
to the high level of degradation reached.
Figure 2
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
Table 6
It appears that the samples showed a gradual weight loss over time (Table 6).
As the accelerated weathering method used in this study has water spray periods and
the plasticizers used are water-soluble, a reasonable explanation for the weight
changes observed is the fact that a certain amount of plasticizer was dissolved from
the films by water. High Resolution TGA (Hi-Res TGA) measurements have been
conducted to quantify the amount of the plasticizers remained in the samples after 0
and 425 hours of the accelerated weathering. Plasticizer content is determined as the
weight loss observed on the Hi-Res TGA curves (as it is shown on Figure 3) due to
volatilization of the plasticizer. Table 7 summarizes these data in relation to the
remaining weight percentage.
Figure 3
Table 7
Table 7 shows that the weight loss by the samples subjected to the accelerated
weathering could be correlated with the plasticizer loss, at least to some extents. For
example, the highest loss of weight (20% at 425 hours) of the CA/PEG blend could be
correlated with the higher hydrophilicity of this blend.

Molecular weights of the samples subjected to accelerated weathering were
estimated by GPC. As illustration, in Figure 4 are shown the GPC curves of CA blend
with 10 wt% PEG samples before and after 200 and 425 h of accelerated weathering.
No significant differences in solubility were observed in the sample preparation for GPC
analysis in THF, even for those samples subjected to 425 hours of accelerated
weathering. In Figure 5 are represented the evolution of the remaining number- and
weight-average molecular weight fractions for neat and plasticized CA with 20 wt%
plasticizer. The blended CA showed a clear reduction of the molecular weight,
significantly higher than that of the neat CA. In general, plasticization of CA triggered
an increase of the molecular weight loss between 50 and 90%, for which the low
molecular weight plasticizers are more effective than PEG.

Figure 4
Figure 5
Table 8
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
In order to correlate the visual appearance with the molecular weight of the
representative samples presented in Figure 2, in Table 8 are summarized their number
average molecular weights after 0, 200 or 225 and 425 hours of accelerated
weathering. Film breakdown is evident when M
n
is under the value of 10.000 Da
(aprox. 40 anhydroglucose units) and indicates that the critical molecular weight for
entanglements has been reached. It is remarkable that the breakdown rate is tripled
when the plasticizer concentration is doubled.
Differences observed in Table 8 on molecular weight of the initial samples were
attributed to thermo-mechanical degradation of CA occurring during melt processing.
As plasticizer content increases the degradation effect on molecular weight of CA
decreases. It is well known that plasticization reduces the melt viscosity and thus the
shearing stress during processing, which promotes chain scission.


4. Conclusions
Plasticized cellulose acetates have been prepared by blending with eco-friendly
plasticizers such as triacetin, tripropionin, triethyl citrate, tributyl citrate, tributyl 2-acetyl
citrate and poly(ethylene glycol). Transparent films of CA and CA blends with 10 and
20 wt% of the selected plasticizers could be prepared by either solvent casting in
acetone or by melt processing methods. Thermo-mechanical properties of these films
were assessed by means of DSC, TGA and tensile testing. The low molecular weight
plasticizers were able to decrease T
g
of CA (190 C) by 60 to 80 C and by about 35 C
in case of PEG, while thermal stability of the CA2.5/plasticizer blends proved not to be
highly dependent upon the plasticizer nature. In comparison with neat CA, blends were
slightly more ductile and their ultimate properties were also almost independent of the
plasticizer nature. Hydrophilicity of the films was estimated via contact angle
measurements. Only the CA/PEG blend showed higher hydrophilicity than CA. In
general, CA/plasticizer blends present good stability. Uniquely, CA/PEG blends
displayed some plasticizer migration with time.
Accelerated weathering tests results showed that plasticization of CA triggered
an increase of the molecular weight loss, up to 90%. A good balance between
hydrophilicity and plasticization efficiency (reduction of T
g
) seems needed to increase
the degradation rate of CA.

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
References
[1] Ach A. Biodegradable Plastics Based on Cellulose Acetate. Journal of
Macromolecular Science, Part A: Pure and Applied Chemistry. 1993;30(9):733 - 740.
[2] Brydson J. Plastics Materials (7th Edition). Elsevier; 1999.
[3] Buchanan CMG, R. M.; Komarek, R. J. Journal of Applied Polymer Science.
1993;47:1709-1719.
[4] Komarek RJG, R. M.; Buchanan, C. M.; Gedon, S. Journal of Applied Polymer
Science. 1993;50.
[5] Rahman M, Brazel CS. The plasticizer market: an assessment of traditional
plasticizers and research trends to meet new challenges. Progress in Polymer Science.
2004;29:1223-1248.
[6] Quintana R, Persenaire O, Bonnaud L, Dubois Ph. Polymer Chemistry. 2012 ;
3(3) :591-595.
[7] Nadia Ljungberg BW. The effects of plasticizers on the dynamic mechanical and
thermal properties of poly(lactic acid). Journal of Applied Polymer Science.
2002;86(5):1227-1234.
[8] Massimo Baiardo GF, Mariastella Scandola, Michel Rimelen, David Lips, Kurt
Ruffieux, Erich Wintermantel,. Thermal and mechanical properties of plasticized poly(L-
lactic acid). Journal of Applied Polymer Science. 2003;90(7):1731-1738.
[9] Matko Erceg TKaIK. Thermal degradation of poly(3-hydroxybutyrate) plasticized
with acetyl tributyl citrate. Polymer Degradation and Stability. 2005;90(2):313-318.
[10] Labrecque LV, Kumar RA, Dav V, Gross RA, McCarthy SP. Citrate esters as
plasticizers for poly(lactic acid). Journal of Applied Polymer Science. 1997;66(8):1507-
1513.
[11] Mohanty AK, Wibowo A, Misra M, Drzal LT. Development of renewable resource
based cellulose acetate bioplastic: Effect of process engineering on the performance of
cellulosic plastics. Polymer Engineering & Science. 2003;43(5):1151-1161.
[12] Ghiya VP, Dave V, Gross RA, Mccarthy SP. Biodegradability of Cellulose Acetate
Plasticized with Citrate Esters. Journal of Macromolecular Science, Part A.
1996;33(5):627 - 638.
[13] Teufel E, Sexauer W, Willmund R. Filter cigarette. In: AG RA, editor.2000.
[14] Carvalho AJF, Zambon MD, Curvelo AAS, Gandini A. Size exclusion
chromatography characterization of thermoplastic starch composites 1. Influence of
plasticizer and fibre content. Polymer Degradation and Stability. 2003;79(1):133-138.
[15] Rosa DS, Guedes CGF, Casarin F, Bragana FC. The effect of the Mw of PEG in
PCL/CA blends. Polymer Testing. 2005;24:542-548.
[16] Bragana FC, Rosa DS. Thermal, mechanical and morphological analysis of
poly(epsilon-caprolactone), cellulose acetate and their blends. Polym Adv Technol.
2003;14(10):669-675.
[17] Buchanan CM, Pearcy BG, White AW, Wood MD. The relationship between blend
miscibility and biodegradation of cellulose acetate and poly(ethylene succinate) blends.
Journal of Environmental Polymer Degradation. 1997;5(4):209-223.
[18] Yamashita Y, Endo T. Deacetylation behavior of binary blend films of cellulose
acetate and various polymers. Journal of Applied Polymer Science. 2006;100(3):1816-
1823.
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT
[19] Guo JH. Effects of plasticizers on water permeation and mechanical-properties of
cellulose-acetate - antiplasticization in slightly plasticized polymer film. Drug Dev Ind
Pharm. 1993;19(13):1541-1555.
[20] Jinghua Yuan PPS, Stephen H. Wu. Effects of Polyethylene Glycol on Morphology,
Thermomechanical Properties, and Water Vapor Permeability of Cellulose Acetate-free
Films. Pharmaceutical Technology. 2001;25:62-73.



Figure Captions
Fig. 1. Evolution of Young modulus, stress and strain at break of CA/PEG blends with
PEG content.
Fig. 2. Visual appearance of series of film samples of a) Pure CA, b) PEG 10 wt %, c)
PEG 20 wt %, d) TP 10 wt%and e) TP 20 wt% subjected to 0, 225 and 425 h of
accelerated weathering.
Fig. 3. Hi-Res TGA curves of CA/PEG (20 wt %) blend before and after 425h of
accelerated weathering.
Fig. 4. GPC curves of CA blend with 10 wt% PEG samples before and after 200 and
425 h of accelerated weathering.
Fig. 5. Evolution of remaining number (left) and weight (right) average molecular
weight fractions, for CA and plasticizer blends subjected to accelerated weathering
tests.


M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT

Table 1. Boiling points and densities of triacetin and citrate-based plasticizers.
Plasticizer Boiling point (C) (g/cm
3
)
Triacetin 259 1.15
Tributyl citrate 169 1.10
Triethyl citrate 126 1.14
Acetyl tributyl citrate 173 1.05
Acetyl triethyl citrate 131 1.14
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT

Table 2. Chemical structures of plasticizers used in this study.
O O
O
O
O
O


O O
O
O
O
O

Glycerol triacetate, Triacetin
(TA)

Glyceryl tripropionate, Tripropionin
(TP)
O O
O
OH
O
O
O


O O
O
OH
O
O
O

Triethyl citrate (TeC) Tributyl citrate (TbC)
O O
O
O
O
O
O
O


O
n


Tributyl 2-acetyl citrate (Tb2C) Poly(ethylene glycol) (PEG)
Average Mn = 200

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT




Table 3. Comparison of the tensile properties of CA2.5/plasticizer films prepared by
solvent casting and melt processing with 10 and 20 wt%of plasticizer content.
Young modulus
(MPa)
Stress at break
(MPa)
Strain at break
(%)

Plasticizer
wt
%
Cast. Melt Proc. Cast. Melt. Proc. Cast.
Melt
Proc.
None 0 2073 81 2158 77 75.4 3.5 76.8 21.1 10 2 6 3
10 1711 50 1644 81 49.0 1.8 50.6 1.3 11 4 31 15
PEG(200)
20 1754 25 1600 77 49.0 1.7 38.3 3.3 19 13 35 15
10 1817 69 2153 163 43.0 2.6 59.3 1.3 4 1 11 5
Triacetin
20 1550 105 1712 72 51.3 4.0 49.9 2.9 9 2 29 17
10 1995 66 1776 21 50.0 6.9 52.7 8.8 5 2 27 13
Tripropionin
20 1540 44 1959 193 45.0 3.4 52.5 4.8 11 3 28 16
10 2027 101 1861 196 57.0 1.4 50.8 64 6 1 9 6
Triethyl
citrate
20 1534 86 1435 59 52.7 3.2 21.8 11.2 11 3 3 0.4
Note: Mean values along with its standard deviation ().
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT


Table 4. Effect of plasticizer nature and content on thermal properties and contact
angle of CA2.5 films prepared by hot-pressing moulding.
Plasticizer DSC
a


TGA
b
Type
content
(wt %)


Tg
(C)


o
T
d

(C)

T
d

(C)

Contact angle
c

()
None 0 190 337 372 58.5
10 154 327 370 55.4
PEG(200)
20 156 296 372 38.0
10 151 313 369 82.7
Triacetin
20 118 279 371 76.2
10 157 327 371 76.6
Tripropionin
20 122 288 372 66.9
10 153 315 367 83.5
Triethyl citrate
20 109 300 372 81.6
a
Glass transition temperature calculated from the first heating scan from 30 to 270 C.
b
Decomposition temperatures measured by TGA for a 10% of weight loss (
o
Td ) and for the maximum
decomposition rate (Td).
c
Average value of 4 different measurements on different areas of film surface just after its preparation.

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT


Table 5. Glass transition temperature of CA blends boxed at room conditions
after 1 and 3 months.
Plasticizer T
g
(C)
a
Nature
Content (wt
%)

Initial After 1 month After 3 months
10 154 156 151
PEG(200)
20 156 162 160
10 151 152 146
Triacetin
20 118 117 116
10 157 158 151
Tripropionin
20 128 127 128
10 153 152 151
Triethyl citrate
20 110 109 110
a
Glass transition temperature calculated from the first heating scan from 30 to 270 C.

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT




Table 6. Remaining weight of samples subjected to accelerated weathering testing.
Accelerated weathering time (hours)
0 125 175 325 425
Sample % remaining weight
a
CA2.5 100 100 98 98 100
20 wt%PEG 100 92 89 81 80
20 wt%TA 100 100 100 95 94
20 wt%TP 100 100 100 94 87
20 wt%TbC 100 98 95 96 94
a
Measurement error 4 %

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT










Table 7. Plasticizer content, determined by Hi-Res TGA, and Remaining Weight
of samples subjected to accelerated weathering testing.
Hi-Res TGA
Plasticizer content (wt %)
Remaining weight (RW)
after 425 h
Sample Initial

After 425 h

Difference 100 - RW (%)
CA2.5 0 0 0 0
20 wt%PEG 20.3 2.7 17.6 20
20 wt%TA 17.4 12.3 5.1 6
20 wt%TP 18.5 11.0 7.5 13
20 wt%TbC 20.9 18.6 2.3 6
M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT

Table 8. Evolution of Mn of plasticized CA with accelerated weathering time.
CA2.5 PEG TP
Time 10 wt% 20 wt% 10 wt% 20 wt%
0 35400 51200 53800 56500 62700
200/225
a
22500 41900 26900 36300 12300
425 19000 24200 8200 11100 4300
a
200 or 225 hours for CA plasticized with 10 or 20 wt%, respectively


M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT

M
A
N
U
S
C
R
I
P
T

A
C
C
E
P
T
E
D
ACCEPTED MANUSCRIPT