Composites: Part A 38 (2007) 1455–1461

www.elsevier.com/locate/compositesa

Mechanical behavior and microstructural analysis of sugarcane

bagasse fibers reinforced polypropylene composites
S.M. Luz a, A.R. Gonçalves a,*
, A.P. Del’Arco Jr. b

a
Escola de Engenharia de Lorena, Universidade de São Paulo, P.O. Box 116, CEP 12602-810 Lorena/SP, Brazil
b
Divisão de Materiais, Instituto de Aeronáutica e Espaço, Comando Técnico Aeroespacial, Pç Mal do Ar Eduardo Gomes,
50, Vila das Acácias, 12228-904, São José dos Campos, SP, Brazil

Received 28 December 2004; received in revised form 23 November 2006; accepted 14 January 2007

Abstract

The compression and injection molding processes were performed in order to evaluate the better mixer method for fiber (sugarcane
bagasse, bagasse cellulose and benzylated bagasse) and matrix (polypropylene). The samples (composites and polypropylene plates) were
cut and submitted to mechanical tests in order to measure flexural and tensile properties. The morphological and microstructural anal-
yses of fracture surface and specimens from composites can be easily evaluated by microscopic techniques. The fracture surface was eval-
uated by SEM and selected specimens from composites were analyzed by reflected light in OM. The better tested method for composites
obtainment was the injection molding under vacuum process, by which composites were obtained with homogeneous distribution of
fibers and without blisters. The mechanical properties show that the composites did not have good adhesion between fiber and matrix;
on the other hand, the fiber insertion improved the flexural modulus and the material rigidity.
 2007 Elsevier Ltd. All rights reserved.

Keywords: A. Polymer–matrix composite (PMCs); B. Microstructure; D. Mechanical testing; E. Injection molding

1. Introduction renewable and less expensive sources [5]. Many natural

fibers have been used in automobiles, trucks, and railway
Polymeric composites may be understood as the combi- cars [4,6]. The sugarcane bagasse is a residue widely gener-
nation of two or more materials, for example, reinforce- ated in high proportions in the agro-industry [7]. A Fibrous
ment elements or filler involved by a polymeric matrix residue of cane stalks is left over after the crushing and
[1]. The introduction of fibers (treated or in natura) into a extraction of juice from the sugarcane. It is a lignocellulosic
polymer is known to cause substantial changes in the by-product of the sugar and alcohol industries and is
resulting composites [2,3], which may result, differently almost completely used by the sugarcane factories as fuel
from the original materials, in good mechanical properties for the boilers [8]. Dimensions of bagasse are of 1.2 mm
and rigidity [4]. length and 15 lm width with 80 L/D [9]. The bagasse is a
In recent years, studies about the utilization of lignocel- vegetable fiber mainly constituted by cellulose, that is a glu-
lulosic materials as reinforcement in polymeric composites cose–polymer with relatively high modulus, often found as
are increasing due to the improvements that natural fibers fibrillar component of many naturally occurring compos-
can provide to the product, such as low density and biode- ites (wood, sugarcane straw and bagasse) in association
gradability, besides the fact that these materials are from with lignin [10]. Many naturally occurring fibers are used
in composites, but mostly in applications involving not
*
Corresponding author. Tel.: +55 012 31595033; fax: +55 012
very high stresses [4].
31533165. Fibers and thermoplastic matrix can be placed together
E-mail address: [email protected] (A.R. Gonçalves). in contact in the forms of polymer solution, polymeric

1359-835X/$ - see front matter  2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compositesa.2007.01.014
Produced with a Trial Version of PDF Annotator - www.PDFAnnotator.com

1456 S.M. Luz et al. / Composites: Part A 38 (2007) 1455–1461

films, polymer in powder or granules, utilizing the mecha- composites fracture surfaces, fiber distribution and
nisms of compression molding by heating and injection microstructure.
molding [1,11,12]. In these two mechanisms, the fibers are
placed in intimate contact with the matrix. 2. Experimental
Thermoplastic polymers are materials thoroughly used
in emergent technologies having low processing cost and 2.1. Materials description
density, among other properties, such as transparency
and possibility of recycling [13]. Polypropylene (PP) was Depithed sugarcane bagasse was kindly furnished by
the thermoplastic polymer used in this work, being a Usina São José ZL (Brazil); the bagasse was washed, dried
semi-crystalline polymer and an important engineering and milled at particles <1 mm. Cellulose from sugarcane
thermoplastic with various industrial applications [14], bagasse was obtained by pulping with soda/anthraquinone
but it is also a commodity polymer by the low cost, low as described by Luz and Gonçalves [20]. Benzylated
level of mechanical resistance, processing facility and larger bagasse was obtained from sugarcane bagasse reacted with
production [15]. Therefore, the combination of lignocellu- benzyl chloride during 4 h, as described by Luz and
losic material with thermoplastic matrix can present a con- Gonçalves [21]. Sawdust/PP composites, material utilized
siderable problem: incompatibility between the polar and as intern coating in cars, with 20 wt% sawdust were fur-
hygroscopic fiber and the non-polar and hydrophobic nished by Simoldes Plásticos Ltda. Pinus elliottii is the
matrix. The possible solution for this problem has been wood species of sawdust inserted inside PP. Polypropylene
studied through the chemical modification of fibers due (PP), material presented in granules (H 503), was kindly
to the presence of hydroxyl groups, very reactive and sus- furnished by OPP (Brazil).
ceptible to chemical reactions. Non-polar groups can be
inserted in the fibers, resulting in hydrophobic characteris- 2.2. Compression molding
tics compatibles with thermoplastic matrices [1,16,17].
Other alternative to improve the compatibility between In an uniaxial press, containing heating elements (infe-
fiber and matrix is to modify the matrix, for example, with rior and superior), a molding tool was placed, containing
maleic anhydride [18]. the materials for compression [bagasse (5 wt%)/polypro-
Recent studies in respect to mechanical behavior of rein- pylene, and polypropylene]. The temperature, heating rate
forcement fibers in composites show that these materials and platform time were determinated utilizing the polypro-
can present structural and non-structural applications. pylene. The adjusted parameters were: 205 C, 2 C min1
The synergetic characteristics of composite materials heating rate and 3 h platform time. Both composites and
depend on the initial characteristics of the reinforcement polypropylene plates were obtained with these parameters.
or the matrix [2]. Damage in the composites is associated The bagasse/polypropylene composite and polypropyl-
to matrix crack, failed interfacial bond between fiber and ene plate were cut using a dual-tape saw (Mecânica Europa
matrix, fiber break and delamination. Composites fracture equipment with 1/400 sheet) in the necessary dimensions for
surfaces have been studied by scanning electron micros- the mechanical tests (flexural and tensile) and the material
copy (SEM), due to the great depth of focus exhibited by polishing was made with sandpaper A80 (Acerbi
this technique. Specimens are fractured by normal physical equipment).
testing procedures and SEM is used to evaluate size, distri-
bution and adhesion of the filler fibers or particles. The nat- 2.3. Injection molding
ure of adhesion between the matrix and the reinforcement
and information relating the structure of mechanical prop- Samples of 8.7–10 wt% bagasse, 6.7 wt% cellulose from
erties can be obtained by SEM assessment of the composite bagasse and 10 wt% benzylated bagasse were initially
fracture surface [19]. mixed with polypropylene. The mixture was placed in an
Reflected light microscopy is a typical optical micros- injector camera at 200 C and 2 C min1 heating rate.
copy (OM), which provides an overview of composite Starting from this temperature, the platform time was
microstructure. Specimens are polished and examined in 3 h. The melted material was injected with or without vac-
reflected light, generally at magnifications about 30–500· uum in a pre-warm mold (200 C) with 3.2 · 220 · 230 mm
[19]. The strength of fiber-reinforced composite depends dimensions. Cutting and polishing were performed as
on the fiber length and orientation, on the proportion of described in the compression molding.
fibers in the matrix and on interfacial bond between the
fiber and matrix [19]. 2.4. Tensile tests
This work aimed to evaluate the mechanical characteris-
tics of PP-composites obtained by compression and Composites were analyzed in an ‘‘Instron’’ universal-
injection molding processes utilizing the bagasse, benzy- testing machine (model 4301), equipped with pneumatic
lated bagasse and bagasse cellulose as reinforcement. Some claws at a cross-head speed of 2 mm min1. Five specimens
composites were analyzed by SEM and OM techniques in of composites were analyzed, with dimensions in agreement
order to evaluate the morphological characteristics of with the ASTM D 638 standard [22]: 19 mm width,
 S.M. Luz et al. / Composites: Part A 38 (2007) 1455–1461 1457

165 mm length and 3.2 mm thickness. The standard molding showed a homogeneous distribution of the fibers
demands specimens with ‘‘dogbone’’ or ‘‘tie’’ format, how- in the matrix. However, the distribution in the plate was
ever, the specimens were cut in rectangular form; hence, the not uniform. Larger and shorter fibers were found in the
results of the tests will be analyzed only comparatively. superior and inferior parts of the plate, respectively. The
press action is from up to down and the shorter fibers easily
2.5. Flexural tests penetrated into the melted polymer. Other problem found
in the composite was the presence of many blisters also
For these tests, the ‘‘Instron’’ universal-testing machine present in the polypropylene plate.
(model 4301) equipment was also used, where a load was The amount of added reinforcement did not contribute to
applied on the specimen with crosshead motion rate of the variation of the tensile and flexural moduli (Table 1).
1.3 mm.min1. The 5 specimens were analyzed with dimen- The fibers insertion can contribute to the modulus increase,
sions in agreement with the ASTM D 790 standard [23]: because the Young’s modulus of the fibers is higher than the
25 mm width, 76 mm length and 3.2 mm thickness. The thermoplastic modulus. However, to obtain a significant
adopted flexural analysis was four-point at 1/4 points increase, a good interfacial bond between fiber and matrix
method. is necessary [24]. The flexural and tensile strengths exhibited
a poor interaction between fiber and matrix, with decreases
2.6. Scanning electron microscopy of 24.1% and 31.8%, respectively, in relation to the poly-
propylene.
The samples submitted to tensile tests were cut and the The elongation decrease at break value (tensile test) for
composite intact fracture surface was analyzed in a LEO the composite in comparison to the polypropylene, is
1450 V scanning electron microscopy with tungsten fila- probably due to the defects generated in the material after
ment operating at 20 kV, utilizing low vacuum technique the fibers insertion, which were not well aggregated to the
and work distance of 12 mm. Magnifications of 60· and polypropylene. The rupture of the specimen was probably
4000· were made and a secondary electron detector was originated by the presence of blisters. The blisters are
utilized. empty spaces, caused probably by air entrance in the
mold. This showed that the compression molding is not
2.7. Optical microscopy the adequate method for thermoplastic composites
preparation.
The selected region of the sample was cut and placed in
molds. Polyester resin and hardeners were mixed, poured 3.2. Injection molding
into the mold and cured to provide support. The samples
grinding was made with sandpaper ranging from 120 to In injection molding, after the optimization of tempera-
1000 grit. The specimens were polished using graded alu- ture, platform time and heating rate conditions, various
mina suspensions in water. After the polishment, the spec- composites from variable reinforcements were obtained:
imens were observed by reflected light in a LEICA optical bagasse, cellulose from bagasse and benzylated bagasse.
microscopy. Sawdust/PP composite, material industrially obtained
was used for comparison.
3. Results and discussion The injected polypropylene did not present blisters
(Fig. 2a and b). In bagasse (8.7 wt%)/PP composite, the
The materials obtained by different methods (compres- fibers distribution was homogeneous and random.
sion and injection molding) were analyzed separately. The composites presented few blisters in the interior of
the plate.
3.1. Compression molding The composite with 10 wt% bagasse, utilizing polypro-
pylene as matrix, did not present visible blisters, because
The uniaxial press has two heating elements (inferior the injection was made under vacuum. The fibers distribu-
and superior). The composites obtained by compression tion was also homogeneous and random.

Table 1
Properties of the materials obtained by compression molding
Samples Elongation at break (tensile) (mm) Tensile strength (MPa) Tensile modulus (MPa)
Polypropylene (PP) 7.2 ± 1.2 27.4 ± 1.6 991.6 ± 57.8
Bagasse (5.0%)/PP composite 3.0 ± 0.3 18.7 ± 2.2 1001.9 ± 95.7
Elongation at break (flexural) (mm) Flexural strength (MPa) Flexural modulus (MPa)
Polypropylene (PP) 7.9 ± 0.4 25.3 ± 1.0 241.9 ± 18.4
Bagasse (5.0%)/PP composite 6.9 ± 0.5 19.2 ± 3.1 223.8 ± 21.4
Reinforcement in wt%.
Produced with a Trial Version of PDF Annotator - www.PDFAnnotator.com

1458 S.M. Luz et al. / Composites: Part A 38 (2007) 1455–1461

The cellulose from sugarcane bagasse and benzylated Sawdust (20%)/PP Tensile
bagasse was not adequate as reinforcement, since they left Cellulose bagasse (6.7%)/PP
Flexural

several fiber agglomerations in the matrix. The agglomera-

Benzylated bagasse (10%)/PP
tion is due to inefficient fibers dispersion inside matrix.
Bagasse (10%)/PP
Hydrophilic fibers are strongly attracted by themselves,
resulting in clusters. Bagasse (8.7%)/PP

The materials obtained by injection molding were jointly Polypropylene (PP)

analyzed in respect to the mechanical characteristics sup- 0 1 2 3 4 5 6 7 8 9 10 11 12 13

Elongation at break (mm)
plied by the tensile and flexural properties.
Fig. 1 shows that the samples elongated before the rup-
Sawdust (20%)/PP Tensile
ture. In respect to the elongation (tensile tests), the Flexural
Cellulose bagasse (6.7%)/PP
obtained composites presented shorter elongation com-
pared to the polypropylene. Two hypothesis can be consid- Benzylated bagasse (10%)/PP

ered: (A) fibers are failed points that can generate a crack, Bagasse (10%)/PP

and (B) there is a sliding of the fibers in relation to the Bagasse (8.7%)/PP
matrix. The elongation at break of the composites in the Polypropylene (PP)
flexural tests is smaller than that in the tensile tests, because 0 4 8 12 16 20 24 28 32
in this method the force application is different. Strength (MPa)

According to Sawer [19], the elongated materials present

ductile fracture and the failed points in composites exhibit Sawdust (20%)/PP

the brittle fracture (Figs. 3a–d and 5a–d). The Fig. 2a and b Cellulose bagasse (6.7%)/PP

show the SEM of polypropylene fracture surface and pres- Benzylated bagasse (10%)/PP
ent a brittle fracture in lateral region and ductile fracture in Bagasse (10%)/PP
the border of the sample. Bagasse (8.7%)/PP
Tensile
The strength properties in Fig. 1 show that the tensile
Polypropylene (PP) Flexural
strength of the composites decreased in relation to the 0 200 400 600 800 1000 1200 1400
polypropylene. The fibers distribution of the bagasse Modulus (MPa)
(10 wt%)/polypropylene composite was homogeneous
Fig. 1. Mechanical properties of composites materials obtained by
and without presence of blisters and the tensile strength
injection molding. The amount of reinforcement is expressed in weight
decreased only 33%, differently of the composite with percentage.
8.7 wt% of bagasse, which value decreased 38% in rela-
tion to the polypropylene. This is probably due to the the polymer and the adhesion between fiber and matrix
interference of other defects, generated by the presence failed.
of blisters. Fig. 3a shows a bagasse/polypropylene com- The optical micrograph (Fig. 4a) exhibits the shape of
posite region that presents fibers clumps and the fibers bagasse within matrix and a random distribution can be
preservation can be observed by the morphological char- observed. The higher magnified micrograph of bagasse
acteristic of lignocellulosics. The polypropylene (matrix) and polypropylene interface (Fig. 4b) shows a wetting or
presented a brittle fracture around the fibers. Fig. 3b major penetration of polypropylene into the fiber, resulting
shows the bagasse/matrix interface and that the failed in better adhesion between fiber and matrix.
propagation goes into fiber direction. The Fig. 3c and d Due to the agglomeration of reinforcement in some
show the higher magnified micrograph of fiber and points, the tensile strength, respectively for the composites
matrix interface and that the fibers were pulled out of of cellulose from bagasse and benzylated bagasse,

Fig. 2. SEM of fracture surfaces of injected polypropylene: (a) lateral detail exhibits classical brittle fracture, and (b) detail of extremity of specimens
fracture exhibits a ductile fracture.
 S.M. Luz et al. / Composites: Part A 38 (2007) 1455–1461 1459

Fig. 3. SEM of fracture surfaces of bagasse/polypropylene composite injected under vacuum. (a) Morphology of composite fracture surfaces towards the
bagasse the presence exhibits of brittle fracture. (b) Morphology of composite fracture surfaces exhibits the limit region between fiber and matrix (brittle
fracture). (c and d) Higher magnification of limit region between fiber and matrix.

Fig. 4. Optical micrograph with reflected light of composite polished surface (bagasse/polypropylene). (a) 50· Magnified micrograph exhibits the bagasse
forms and distribution within the matrix. (b) 100· Magnified micrograph shows the detail of limit region between fiber and matrix.

decreased 50% and 42%, when compared to the polypro- fiber clumps were observed within matrix, resulting in areas
pylene. In the composite with sawdust/polypropylene concentrated with fibers and areas with excessive thermo-
industrially produced, the tensile strength decreased 62.3%. plastic, which ultimately decreases mechanical properties.
The modified fiber (benzylated bagasse)/matrix interface The bagasse cellulose/matrix interface is shown in
is shown in Fig. 5a. The fibers presented circular cross-sec- Fig. 5d. Fibers exhibited poor fiber wetting, as shown by
tion shapes. In a recent work [20], in the chemical modifi- the clumps of fibers with no polypropylene. The fibers pre-
cation process, the lignin (natural fiber component) was sented circular cross-section shapes.
partially removed by NaOH addition and the morphology The formation of clumps in the material is disadvanta-
of benzylated bagasse was different in respect to in natura geous, due to the decrease of tensile strength (reinforcement:
bagasse as shown in Fig. 5b. In Fig. 5a, the fibers are trans- cellulose from bagasse and benzylated bagasse). However,
versally aligned to the fracture. The higher magnified in flexural strength, these materials act differently, resulting
micrograph of fiber and matrix interface (Fig. 5c) exhibits in a smaller decrease of these values. For the composite with
the pulled-out fibers of the matrix by tensile stress. The sawdust, as the particles are very small (about 20 mesh), the
Produced with a Trial Version of PDF Annotator - www.PDFAnnotator.com

1460 S.M. Luz et al. / Composites: Part A 38 (2007) 1455–1461

Fig. 5. SEM composite fracture surface injected under vacuum: (a) benzylated bagasse/polypropylene interface showing the clumping and morphological
fibers; (b) the micrograph shows the shapes of benzylated bagasse; (c) higher magnified micrograph shows the detail of region interface between fiber and
matrix; and (d) bagasse cellulose and matrix interface showing the morphology of their composites.

flexural strength decreased about 54% in respect to the poly- fibers and free of blisters. The flexural and tensile tests show
propylene, due to larger amount of failed points. that the obtained composites did not present good flexural
The tensile and flexural moduli are related to the mate- and tensile strengths. The blisters and/or the non-homoge-
rial rigidity, and as previously mentioned, the Young’s neity of the fibers are defects in the material that directly
modulus was higher for the fibers than for thermoplastics. interfere in the mechanical properties, harming the obtain-
If the fibers are distributed and aggregated to the matrix, a ment of a material with high resistance. The composites did
higher flexural modulus for the material can be resulted. not present, in general, a good interfacial contact between
Comparing the values of tensile modulus between the fiber and matrix. However, the fibers insertion caused an
composites (Fig. 1), bagasse (10 wt%)/PP and sawdust increase in the flexural modulus, turning the materials more
(20 wt%)/PP, the composites presented values very similar rigid. The mechanical properties of the obtained composites
to the neat polypropylene. The tensile modulus values for are comparable to the commercial material with 20 wt%
the bagasse (8.7 wt%)/PP, benzylated bagasse (10 wt%)/ sawdust and PP. The composites utilizing bagasse (treated
PP and cellulose from bagasse composites are smaller due or in natura) can be investigated, seeking a similar applica-
to the disadvantageous presence of defects. tion, for example, in coatings of pieces of automobiles. The
For the composites, the flexural modulus increased in SEM and OM techniques permit the study of the morphol-
relation to that of the neat polypropylene (Fig. 1; except ogy and microstructure of composites.
for the composite with benzylated bagasse). As the flexural
tests were made by the four-points method, the composite Acknowledgements
with random distribution of the fibers had a better perfor-
mance. Better performance of composites with random dis- The authors acknowledge FAPESP, Capes and CNPq
tribution into the matrix is attained due to homogeneous for fellowships and financial support.
distribution of fibers. The stress transference between fiber
and matrix is more effective, affecting positively the References
performance.
[1] Nogueira CL, Marlet JMF, Rezende MC. Processo de Obtenção de
Pré-Impregnados Poliméricos Termoplásticos Via Moldagem por
4. Conclusions
Compressão a Quente. Pol Ciênc e Tecnol 1999(3):18–27.
[2] Khalil HPSA, Ismail H, Rozman HD, Ahmad MN. The effect of
The best molding process is the injection under vacuum, acetylation on interfacial shear strength between plant fibres and
which leads to materials with homogeneous distribution of various matrices. Europ Polym J 2001;37:1037–45.
 S.M. Luz et al. / Composites: Part A 38 (2007) 1455–1461 1461

[3] Sanadi AR. Natural fiber–thermoplastic composites: a review. In: [15] Canevarolo SV. Ciência dos Polı́meros. São Paulo: Artliber; 2002.
Leão AL, Carvalho FX, Frollini E, editors. Natural Polymers and [16] Rowell RM, Sanadi AR, Caulfield DF, Jacobson RE. Utilization of
Composites IV. São Carlos; 2002. p. 409–15. natural composites: problems and opportunities. In: Leão AL,
[4] Chawla KK. Fibrous Materials. New York: Cambridge University Carvalho FX, Frollini E, editors. Lignocellulosic – Plastics Compos-
Press; 1998. ites. São Paulo; 1997. p. 23–52.
[5] Mattoso LHC. Conferência Internacional de Compósitos Reforçados [17] Hon DN-S. Functional Natural Polymers: A New Dimensional
com Fibras Vegetais. Pol Ciênc e Tecnol 1999(2):16. Creativity in Lignocellulosic Chemistry. In: Hon DN-S, editor.
[6] Rijswijk KV, Brouwer WD. Benefits of composites made of locally Chemical Modification of Lignocellulosic Materials. New
grown natural fibers. In: Leão AL, Carvalho FX, Frollini E, editors. York: Dekker; 1996. p. 1–10.
Natural Polymers and Composites IV. São Carlos; 2002. p. 422– [18] Joseph PV, Joseph K, Thomas S, Pillai CKS, Prasad VS, Groeninckx
28. G, et al. The thermal and crystallisation studies of short sisal fibre
[7] Paiva JMF, Frollini E. Matriz Termofixa Fenólica em Compósitos reinforced polypropylene composites. Compos Part A 2003;34:
Reforçados com Fibras de Bagaço de Cana-de-açúcar. Pol Ciênc e 253–66.
Tecnol 1999(2):78–87. [19] Sawyer LC, Grubb DT. Polymer Microscopy. London: Chapman &
[8] Rasul MG, Rudolph V, Carsky M. Physical properties of bagasse. Hall; 1996.
Fuel 1999;78:905–10. [20] Luz SM, Gonçalves AR. Evaluation of soda/Aq pulping for
[9] Young RA. Utilization of natural fibers: characterization, modifica- sugarcane bagasse and straw. In: Proceedings of 7th Brazilian
tion and applications. In: Leão AL, Carvalho FX, Frollini E, editors. Symposium on the Chemistry of Lignins and other Wood Compo-
Lignocellulosic – Plastics Composites. São Paulo; 1997. p. 1–21. nents. Belo Horizonte; September 2001. p. 131–6.
[10] Klemm D, Philipp B, Heinze T, Heinze U, Wagenknecht W. [21] Luz SM, Gonçalves AR, Del’Arco Jr AP. Evaluation of mechanical
Comprehensive Cellulose Chemistry, vol. 1. Weinheim: Wiley- properties of sugarcane bagasse/polypropylene composites. In: Leão
VCH; 2001. AL, Carvalho FX, Frollini E, editors. Natural Polymers and
[11] Chiellini E, Cinelli P, Corti A, Kenawy ER. Composite films based on Composites IV, São Carlos; 2002. p. 523–27.
waste gelatin: thermal–mechanical properties and biodegradation [22] ASTM D 638. Association standards testing materials. Standard test
testing. Polym Degrad Stabil 2001;73:549–55. method for tensile properties of plastics. Philadelphia; 1988.
[12] Curvelo AAS, Carvalho AJF, Agnelli JAM. Thermoplastic starch– [23] ASTM D 790. Association standards testing materials. Standard test
cellulosic fibers composites: preliminary results. Carbohyd Polym method for flexural properties of unreinforced and reinforced plastics
2001;45:183–8. and electrical insulating materials. Philadelphia; 1988.
[13] Brandrup J, Immergut EH, Grulke EA. Polymer Handbook. New [24] Luz SM, Gonçalves AR. Evaluation of chemically modified sugar-
York: Wiley Interscience; 1999. cane bagasse by weight variation and FT-IR analysis. In: Leão AL,
[14] Mano EB. Introdução a Polı́meros. São Paulo: Edgard Blücher; Carvalho FX, Frollini E, editors. Natural Polymers and Composites
1988. IV, São Carlos; 2002. p. 1494–497.