Sie sind auf Seite 1von 10

Journal of Materials Research and Technology

w w w. j m r t . c o m . b r


Thermogravimetric Stability of
Polymer Composites Reinforced with Less Common
Lignocellulosic Fibers an Overview
Sergio Neves Monteiro1,*, Vernica Calado2,
Rubn J. S. Rodriguez3, Frederico M. Margem3
Instituto Militar de Engenharia (IME), Materials Science Department, Rio de Janeiro, Brazil.
Escola de Qumica, Universidade Federal do Rio de Janeiro, Rio de Janeiro, Brazil.
Universidade Estadual do Norte Fluminense, Campos dos Goytacazes, Brazil.

Manuscript received April 14, 2012; in revised form June 19, 2012

Sergio Neves Monteiro (June 24, 1943) graduated as a Metallurgical Engineer (1966) at the Universidade
Federal do Rio de Janeiro (UFRJ). Received his MSc (1967) and PhD (1972) from the University of Florida,
followed by a course (1975) in Energy at the Escola Superior de Guerra (Brazilian War College) and
Post-doctorate (1976) at the University of Stuttgart. Joined the Metallurgy Department (1968) and was
appointed (1977) Full Professor of the Post-Graduation Program in Engineering (COPPE) of the UFRJ. He
was elected Head of the Department (1978), Coordinator of COPPE (1982) and Under-Rector for Research
(1983). Invited as Under-Secretary of Science for the State of Rio de Janeiro (1985) and Under-Secretary
of College Education for the Federal Government (1989). Retired in 1993 and joined the Universidade Es-
tadual do Norte Fluminense (UENF). Published over 900 articles in journals and conference proceedings.
Has been honored with several awards including the ASM Fellowship. Is presently top researcher (1A) of
the Conselho Nacional de Desenvolvimento Cientco e Tecnolgico (Brazilian Council for Scientic and
Technological Development CNPq), vice-president of the Fundao de Amparo Pesquisa do Estado do Rio de Janeiro (Superior
Council of the State of Rio de Janeiro Research Foundation FAPERJ), consultant for the main Brazilian R&D agencies and member
of the Editorial board of three international journals.

Environmental, economic, and technical reasons justify research efforts aiming to provide natural materials with possibility of
replacing synthetic ber composites. Commonly known lignocellulosic bers, such as jute, sisal, ax, hemp, coir, cotton, wood, and
bamboo have not only been investigated as reinforcement of polymeric matrices but already applied in automobile components. Less
common bers, such as curaua, henequen, que, buriti, olive husk, and kapok are recently being studied as potential reinforcement
owing to their reasonable mechanical properties. The relatively low thermal stability of these bers could be a limitation to their
composites. The works that have been dedicated to analyze the thermogravimetric stability of polymer composites reinforced with
less common lignocellulosic bers were overviewed.

*Corresponding author.
E-mail address: (S. N. Monteiro)

2012 Brazilian Metallurgical, Materials and Mining Association. Published by Elsevier Editora Ltda.
Este um artigo Open Access sob a licena de CC BY-NC-ND

J. Mater. Res. Tecnol. 2012; 1(2):117-126

118 Monteiro et al.

KEY WORDS: Natural bers; Polymer-matrix composites (PMCs); Thermogravimetric analysis.

2012 Brazilian Metallurgical, Materials and Mining Association. Published by Elsevier Editora Ltda.
Este um artigo Open Access sob a licena de CC BY-NC-ND

1. Introduction It was emphasized[21,22] that, as compared to berglass,

the lignocellulosic ber composites are lighter, cheaper,
The rst engineered composites materials were probably and less abrasive in contact with processing equipments.
reinforced with natural bers and produced in the primor- Furthermore, berglass represents a problem to the envi-
dium of mankind. Evidence of adobe construction blocks of ronment with restriction to nal destination by incineration
sun-dried clayey mud incorporated with straw was found in in thermo-electric plants. Glass ber particulates also rep-
the ancient Egyptian civilization. According to our current resent potential health hazardous both in the initial com-
interpretation, these rst composites would be considered posite processing and latter at the end-of-life degradation.
environmentally friendly owing to their all natural compo- By contrast, the thermal stability of a lignocellulosic ber
nents. Even today, the rudimentary technique used by our composite is inferior to similar matrix composite reinforced
ancestors to fabricate such composites persists in many low with glass ber. This could be a critical restriction for condi-
income regions of the world. As mentioned by Bledzki and tions associated with relatively high temperatures attained
Gassan[1], in a seminal review on natural ber composites, during the curing of the composite or its in-service use.
only at the end of the 19th century the industrial manu- In fact, temperature usually causes an initial degradation
facture of natural bers with small content of polymeric of the ber organic structure and thus limits the polymer
binder was reported for airplane seats and fuel tanks. Large composite application.
amounts of sheets, tubes, and pipes for electronic pur-
pose were continuously fabricated as earlier as 1908, using 2. Thermal Decomposition of Lignocellulosic
natural bers, paper, or cotton incorporated into phenolic Material
matrix[1,2]. With the rising performance of petroleum-based
polymers and synthetic bers as well as the facility to com- Several review articles[2326] have, since more than ve de-
bine them into more uniform and stronger composites, the cades ago, been dedicated to the thermal decomposition
industrial production of traditional natural ber composites of lignocellulosic materials. Beall and Eickner[23] reviewed
declined. works on thermal analysis results of wood and its cellulose,
In another classic review on biobres and biocomposites, hemicelluloses, and lignin constituents. Kilzer[24] reviewed
Mohanty et al.[2] indicated that the synthetic ber compos- works on the thermal decomposition of cellulose. Nguyen
ites reached commodity status in the 40s with glass ber et al.[25,26] conducted subsequent reviews on the application
reinforcing (berglass) unsaturated polyesters. These syn- of differential thermal analysis (DTA), differential scanning
thetic ber composites experienced an exponential growth calorimetry (DSC), and thermogravimetry (TG) to the study
after the World War II to become the most successful class of lignocellulosic materials as well as modied forms of li-
of engineering materials[35] with application in practically gnocelluloses. These review articles[2326] indicated the fol-
all elds of human interest, from appliances and sports to lowing results:
surgical prosthesis and aerospace components. At the end
of the last century, increasing environmental concerning 2.1 Cellulose
over generalized pollution caused by non-degradable ma-
terials, especially long lasting plastics, and climate changes Cellulose decomposition takes place by major reaction in-
resulting from CO2 emission, promoted a growing tendency volving depolymerization, thermoxidation, dehydration,
towards the substitution of synthetic bers composites. and formation of glycosans, depending on the presence of
Natural materials, particularly cellulose-rich bers, also oxygen or an inert atmosphere. In a non-oxidative atmo-
known as lignocellulosic bers, were renewed as reinforce- sphere, dehydration occurs in a range of 210C260C and
ment of polymer composites. A considerable number of re- depolymerization with volatilization of levoglucosans, at
searches have, in the past few decades, been dedicated about 310C. Under oxygen, thermoxidative reactions oc-
to lignocellulosic bers as engineering materials and their cur in the temperature range of 160C250C. The degra-
reinforced polymer composites for applications in substi- dation of cellulose by pyrolysis has been assumed to follow
tution of synthetic ber composites. Several reviews and a rst-order kinetic. In the specic case of wood, cellu-
general articles covered this trend[1,2,615]. In these publica- lose decomposition in air begins at 320C with a maximum
tions, advantages are emphasized and drawbacks discussed weight loss at 350C. In helium it also begins at 320C, but
aiming at reduce the limitation, which exists in practical the maximum rate is shifted to 375C. Cotton cellulose dis-
use. Despite the drawbacks, a growing industrial applica- plays two endothermic DTA peaks at 100C and 367C.
tion of lignocellulosic ber composites is nowadays occur-
ring in sectors such as building construction, packaging, 2.2 Hemicellulose
sport devices, electrical parts, and vehicle components[14].
Automotive industries, initially the Europeans followed by Hemicellulose constituents decompose at temperatures as
Americans and Japanese, are adopting this type of compos- low as 159C175C, as in the case of acetyl galactogluco-
ites in several interior and exterior parts[1620]. mannan. DTA of xylan in oxygen exhibits a rst exothermic

J. Mater. Res. Tecnol. 2012; 1(2):117-126

Thermogravimetry of Composites with Less Common Natural Fibers 119

peak at 215C. Both arabinogalactan and deacetylated ga- density polyethylene[34], polypropylene[35], and polylactic
lactoglucomannan begin degradation at 195C, reaching a acid[36] matrices;
maximum at 280C. b) Hemp Fiber Composites, in polyester[37], cashew nut shell
resin[38], epoxy[39], polypropylene[40,41], and starch-base ther-
2.3 Lignin moplastic[42] matrices;
c) Sisal Fiber Composites, in polypropylene[43,44], blend of
Obtained by hydrolysis of wood, it was found to decompose polypropylene with high density polyethylene[43], polysty-
in three stages. At temperatures below 220C250C, con- rene[45], polypropylene and maleic acid anhydride grafted
densation and splitting of the side chains takes place. Be- styrene-ethylene-co-butylene-styrene copolymer[46,47], epoxy[48],
tween 300C400C, active pyrolysis leads to the formation phenolic and lignophenolic[49], and soy protein blended with
of free radicals. Above 400C, decomposition is associated gelatin[50] matrices;
with a series of degradation and condensation reactions d) Flax Fiber Composites, in polypropylene[51], epoxy[39], and
with accumulation of aromatic products. polylactic acid[52] matrices;
e) Coir Fiber Composites, in copolymer of starch with ethyl-
2.4 Wood ene vinyl alcohol[53], and polyester[54] matrices;
f) Cotton Fiber Composites, in phenolic thermoset[55]
In its natural and unmodied form, showed, under helium, matrix;
a rst endothermic peak at 107C, corresponding to wa- g) Kenaf Fiber Composites, in epoxy[39] chitosan[56,57], and
ter loss. Weak endothermic peaks were found at 207C and starch-based thermoplastic[40] matrices;
330C, corresponding to dehydration and depolymerization h) Wood Fiber Composites, in polypropylene[5860], low and
of various constituents, probably including hemicellulose. high density polyethylene[59,61], and polyhydroxy(butyrate-
A very strong endothermic peak at 367C was attributed to co-valerate)[60] matrices;
cellulose decomposition. An exothermic peak rises above i) Pineapple Fiber Composites, in polyethylene[62],
400C, presumably related to recombination of cellulose polyhydroxy(butyrate-co-valerate)[63], and polycarbonate[64]
and lignin fragments. Under oxygen, it was reported an en- matrices;
dothermic peak at 87C, owing to evaporation of water, and j) Bamboo Fiber Composites, in polylactic acid[65], polybu-
two exothermic peaks at 343C and 470C, probably due to tylene succinate[65], epoxy[66], and polyhydroxy(butyrate-co-
cellulose and lignin. valerate)[67] matrices;
k) Ramie Fiber Composites, in polylactic acid[36] matrix;
3. Thermal Analysis of Common l) Banana Fiber Composites, in polyvinyl chloride[68] matrix;
Lignocellulosic Fiber Composites m) Bagasse Fiber Composites, in polyurethane (PU)[69], and
recycled high density polyethylene[61] matrices.
A short review on the thermal stability of polymer com- As a general comment, one may conclude from all these
posites reinforced with few common lignocellulosic bers works on the thermal stability of polymer matrix composites
was, for the rst time, presented as one of the sections of reinforced with commonly known and used lignocellulosic
the Nabi Saheb and Jog[6] review on natural ber polymer bers that their processing and applications are restricted
composites. They indicated that the thermal degradation to a safe temperature of 250C, or to a maximum of 367C,
of natural bers is a crucial aspect in the development of in case of more stable specic polypropylene and maleic
their composites and thus has a bearing on the curing tem- anhydride grafted styrene-ethylene-co-butylene-styrene
perature in the case of thermosets and extrusion tempera- copolymer matrix reinforced with sisal ber[46]. The reader
ture in thermoplastic matrix composites. Nabi Sahed and should also bear in mind that all these composites, owing to
Jog[6] also stated that thermal stability improvement have the contribution of the lignocellulosic ber, display a DTG
been attempted by coating and/or grafting the bers with water loss peak which is found at temperatures as low as
monomers, quoting the works of Mohanty et al.[27] and Sa- 37C[32] and inferred at about 140C144C[48,53].
baa[28]. The effect of the composite fabrication ambient was
discussed by the authors[6] as a possibility of lignocellulosic 4. Thermogravimetric Stability of Less
ber degradation, quoting the work of Sridhar et al.[29], and Common Lignocellulosic Fiber Composites
indicating that the actual practice is carried out under air
and that thermal degradation can lead to inferior mechani- In principle, a complete assessment of the thermal be-
cal properties. As a nal remark, Nabi Saheb and Jog[6] con- havior of a material would require not only temperature
cluded that the thermal degradation of the lignocellulosic difference, DTA, thermogravimetric, TG/DTG, differential
ber inside the polymeric composite matrix also results in calorimetric, DSC, and dynamic-mechanical (DMA) thermo
production of volatiles at processing temperatures above analyses, but also the evaluation of properties such as ther-
200C. This could result in porous composites with lower mal conductivity, specic heat, and thermal diffusivity. The
densities and inferior mechanical properties. scope of this overview is limited to thermal stability re-
Since this rst short review[6], numerous works have sults associated with weight loss variation with tempera-
investigated the thermal stability of polymer composites ture obtained by thermogravimetric analysis. These results,
reinforced with common lignocellulosic bers. The reader displayed as TG thermograms as well as its derivative DTG,
may nd specic data and conclusions on the following ar- will be covered for relevant works on polymer composites
ticles listed by employed reinforced ber: reinforced with less known bers published in internation-
a) Jute Fiber Composites, in phenol formaldehyde[30], poly- ally recognized sources. Numerous less common lignocel-
ester[31], vinyl ester[32], polyester with acrylic acid[33], high lulosic bers such as curaua, henequen, que, buriti, olive

J. Mater. Res. Tecnol. 2012; 1(2):117-126

120 Monteiro et al.

husk, kapok, abaca, isora, piassava, artichoke, milkweed, shoulder peak around 370C, which is more pronounced
sabei, okra, caroa, sansevieria, palmyrah, maize, phrag- for higher volume fractions of curaua ber, as well as a
mite, esparto, malva, paina, sponge gourd, communis, pita major peak almost coincident with that of the neat poly-
oja, roselle, canary, rye, mesta, barley, raphia, oat, and ester. Fig. 1 illustrates the TG/DTG curves of neat poly-
rape are commercially available and currently being inves- ester and composites reinforced with up to 30 vol.% of
tigated for their potential as engineering materials[14,15,70]. continuous and aligned curaua bers, adapted from the
Only a few of these less common bers, however, have been Ferreira et al.[74] work. In this gure, an enlarged insert
studied for thermogravimetric stability in association with reveals shoulder peaks associated with curaua ber ther-
polymer composites. These few bers curaua, henequen, mal degradation in the composite. By considering these
que, buriti, olive husk, and kapok serve as sub-titles in overviewed works, and based on the results for the iso-
reviewing the following related thermogravimetric stability lated curaua bers[71,72,7577] with shoulder at 268C290C
works on their polymer composites. and major peak at 310C365C, one should conclude that
PU[71], HDPE[72], and polyester matrices[74] composites are
4.1 Curaua Fiber Composites thermally more stable than the pure curaua ber.

Moth and Arajo[71] showed TG/DTG and DTA curves for 4.2 Henequen Fiber Composites
PU matrix composites reinforced with up to 20 wt.% of
curaua bers. For the neat PU, the DTG curve displays a Sgriccia and Hawley[39] presented thermogravimetric re-
small initial peak at 260C that was assigned to the de- sults on epoxy matrix composites reinforced with 15 wt.%
composition of additives in the PU. A broad shoulder peak of hemp (section 3, item b), ax (section 3, item d), kenaf
around 360C followed by a major peak at 422C were (section 3, item g), and henequen bers. These compos-
attributed to decomposition of the rigid and soft urethane ites were both oven and microwave cured. For compari-
bonding, respectively. The DTA curve shows three endo- son, similar glass ber composites were also investigat-
thermic events at 250C, 330C, and 420C that corrobo- ed. Although no TG curves were shown, the authors[39]
rate those found in the DTG peaks. The investigated com-
posites[71] also display faint DTG peaks around 60C, due (a)
to the release of water. Two other decomposition peaks
with temperatures (wt.% of curaua bers) can be seen at
358C and 418C (5 wt.%); 356C and 420C (10 wt.%);
and 356C and 418C (20 wt.%). Endothermic DTA peaks
are also seen at the same temperatures and conditions.
These results indicate that the thermal stability of the
composites is practically the same of the PU. Moreover,
the amount of curaua ber in the composite causes no
apparent change in the thermal stability.
Arajo et al.[72] presented results from TG/DTG curves
of isolated curaua ber as well as high density polyethyl-
ene (HDPE) and HDPE matrix composites reinforced with
20 wt.% of either unmodied (natural) or poly(ethylene-
co-vinyl-acetate) (EVA) and maleic anhydride grafted
polyethylene (PE-g-MA) compatibilized curaua bers.
The DTG curve of the neat HDPE shows a peak at 478C.
The DTG curves of the composites display two distinct
peaks. The rst, coinciding for all composites at 349C,
occurred closer to the main decomposition peak of the
curaua ber, 363C. Additionally, the authors calculated a
weighted mean expected DTG curve if there was no inter-
action among the degradative process for the composites,
quoting the work of Waldman and de Paoli[73]. The second
peak was found at about 468C471C for the composites.
A comparison of TG curves with the calculated curve al-
lowed the authors[72] to indicate that the composite com-
patibilized with PE-g-MA is less stable, while those with
EVA as well as with no compatibilization are more stable.
Ferreira et al.[74] showed TG/DTG curve, obtained at a
heating rate of 10C/min in nitrogen, for polyester com-
posites reinforced with up to 30 vol.% of untreated curaua
bers. A small initial peak observed for all composites
around 70C was attributed to moisture release. The neat
polyester begins to decompose around 250C and displays Fig. 1 TG/DTG curves of polyester composites reinforced with different
a major DTG peak at about 410C related to the degra- volume fraction of curaua bers. (a) Plain curves; (b) enlarged DTG curves.
dation of its molecular chains. The composites display a Adapted from Ferreira et al.[74].

J. Mater. Res. Tecnol. 2012; 1(2):117-126

Thermogravimetry of Composites with Less Common Natural Fibers 121

presented the degradation temperature associated with 4.3 Fique Fiber Composites
5 wt.% to 75 wt.% of weight loss. Tables 1 and 2 repro-
duces these results and the reader can observe that the Gan and Mondragon[78] performed thermogravimetric anal-
isolated henequen ber shows degradation temperature ysis on both polypropylene (PP) and polyoxymethylene (POM)
(weight loss) of 61C (5 wt.%) up to 392C (75 wt.%). On matrices composites reinforced with 20 wt.% que bers,
the other hand, the microwave cured composites show a untreated as well as modied with maleic anhydride (MA),
signicant increase: 355C (5 wt.%) up to 483C (75 wt.%). propionic acid (PA), glycidyl-methacrylate (G) and formalde-
These values are closer (Tables 1 and 2) to those corre- hyde (F) or compatibilized with a copolymer of polypropylene
sponding to the neat epoxy microwaved of 403C (5 wt.%) (Fluka) and maleic anhydride (MAPP). The discussed compos-
and 484C (75 wt.%). Fig. 2, adapted from Sgriccia and ite results were limited to some of the ber treatments. For
Hawley[39] with necessary corrections, illustrated ESEM instance, a major decomposition peak was reported for the
images of microwave cured composites. The authors[39] neat PP at 473C, while the untreated as well as PA and
indicated that the natural ber composites investigated, MAP treated ber composites display three peaks. Fig. 3,
including the henequen/epoxy, both oven and microwave adapted from the work of Gan and Mondragon[78], shows
cured, composites would not be suitable for applications TG/DTG curves for neat PP and PP composites reinforced with
at temperatures as high as those of similar glass ber 20 wt.% of que bers as well as neat POM and POM compos-
composites. ites reinforced with 20 wt.% of que bers. The authors[78]
indicated that the rst (310C) and second (387C) peaks for
the untreated ber composites, as well as the corresponding
Table 1 Degradation temperatures for glass and henequen bers rst (344C) and second (393C) peaks for the MAPP treated
associated with different levels of TG weight loss ber composites, are related to those of the isolated que
ber, 301C and 356C, respectively, quoting their previous
T(C) 5 T(C) 25 T(C) 50 T(C) 75 work[79]. The third peak at 451C for untreated and at 476C
Weight loss (%) wt.% wt.% wt.% wt.%
for MAPP treated ber composites correspond to the decom-
Henequen 61.1 301.4 337.4 391.6 position of the PP matrix. As a general comment, Gaan and
Mondragon[78] stated that the lower thermal stability of POM
Glass 586.4
with respect to PP does not allow for separating the con-
Adapted from Sgriccia and Hawley [39]
. tributions for thermal degradation of que bers and POM
matrix in related composites. Higher temperatures than that
for the neat POM are necessary for complete degradation of
Table 2 Degradation temperatures for glass and henequen ber its composites, possibly due to ber degradation delaying
reinforced epoxy matrix composites associated with different and matrix crystallinity variations.
levels of TG weight loss
4.4 Buriti Fiber Composites
T(C) 5 T(C) 25 T(C) 50 T(C) 75
Weight loss (%) wt.% wt.% wt.% wt.% Santos et al.[80] conducted thermogravimetric experiments
Glass oven 400.3 N/A 401.8 538.6 on cardanol-formaldehyde (CFR-thermoset resin) matrix
composite incorporated with 5 wt.%, 10 wt.%, and 15 wt.%
Glass microwave 400.4 N/A 401.7 481.7
of buriti bers obtained from leaf straw. Buriti bers were
Henequen oven 305.8 374.6 391.0 455.9 both untreated and subjected to NaOH alkali (merceriza-
Henequen 355.0 389.1 392.0 483.0 tion) or silanization treatments. TG curves show that the
microwave (a) (c)
Neat epoxy oven 396.6 N/A 397.9 489.8
Neat epoxy 403.3 N/A N/A 484.3
Adapted from Sgriccia and Hawley[39].

(b) (d)

Fig. 2 ESEM micrographs of microwave cured epoxy composites reinforced Fig. 3 TG/DTG curves: (a) neat PP; (b) PP composites with 20wt.% of que
with: (a) henequen and (b) glass bers. bers; (c) neat POM; (d) POM composites with 20 wt.% of que bers.
Adapted from Sgriccia and Hawley[39]. Adapted from Gan and Mondragon[78].

J. Mater. Res. Tecnol. 2012; 1(2):117-126

122 Monteiro et al.

CFR has, apparently, an onset temperature for thermal maleic-anhydride-polypropylene (PPMA). The pure untreated
degradation, around 320C, which is higher than those for OHF begins to degrade at 210C, while the VTAS treated OHF
the 10 wt.% buriti ber composites, around 240C. In fact, at 201C. This reduction was attributed by the authors[81] to
the authors[80] indicated that both composites present an the elimination of hydrogen bonds that requires signicant
intermediate thermal stability in relation to the isolated energy. DTG peaks were also related to the OHF degradation.
buriti ber and the matrix. Furthermore, the mercerized The rst, around 100C, was assigned to water evaporation;
ber composite curve shows an inection at about 305C the second, at 260C, for untreated and 250C for treated
because of the thermal ber degradation and another at OHF were ascribed to both hemicellulose and glycidic bonds
400C owing to the thermal degradation of the thermo- of cellulose decomposition. The third peak, at 325C, was
set resin. Fig. 4, reproduced from the work of Santos et attributed to cellulose decomposition, while the fourth
al.[80], shows SEM micrographs of untreated, mercerized peak, at 350C, to the lignin decomposition. The authors[81]
and silanized buriti bers as well as 10 wt.% buriti ber indicated that these results are in agreement with those of
reinforced CFR matrix composites. As indicated by the au- Mder et al.[35] and Pracella et al.[40].
thors[80], the mercerized ber (Fig. 4a) displays a rough As for the neat matrix, a single DTG peak was observed
aspect, probably due to the removal of low molar mass at 397C in association with 97% of weight loss. The au-
compounds, which leaves cavities at the surface. The si- thors[81] also concluded that the grafting reaction of MA
lanized ber (Fig. 4c) presents a more regular surface. onto the PP, for the OHF treatment, generates a 16C re-
This suggests that the silanization produces a lm coating duction in thermal stability. TG/DTG curves for the com-
on the entire ber surface. The composite (Fig. 4d) dis- posites showed an intermediate behavior between the pure
plays a lamellar-like morphology with evidence of excel- OHF and the neat PP. In fact, according to Amar et al.[81],
lent adhesion between the matrix and mercerized ber. the thermal degradation of the olive husk ber composites
This good coverage of the ber by the resin may contrib- occurred in a three step degradation process. However,
ute to improve its thermal stability. the 20 wt.% OHF composites apparently display four DTG
peaks. A small rst peak at about 100C, not mentioned by
4.5 Olive Husk Fiber Composites authors[81], is probably due to water release from the OHF.
A second, in the temperature range of 232C308C, and
Amar et al.[81] performed thermogravimetric analysis at a a third at 308C350C, shoulder peaks were attributed to
heating rate of 10C/min in nitrogen on polypropylene (PP) the decomposition of hemicellulose and cellulose, respec-
matrix composites added with 10 wt.% and 20 wt.% of olive tively. The main DTG peaks around 400C, related by the
husk our (OHF), the solid portion remaining after pressing authors to the third decomposition stage between 350C
olives, which contains signicant amounts of lignocellulosic 454C, were assigned to the PP matrix and OHF lignin joint
ber. The OHF were both untreated and subjected to vinyl- degradation. It was also indicated by the authors that the
triacetoxysilane (VTAS) chemical treatment or grafted with composite modied by PPMA revealed a better thermal
stability than those with untreated or VTAS treated OHF.

4.6 Kapok/ Cotton Fibers Hybrid Composites

Mwaikambo et al.[82] employed hybrid kapok/cotton -

bers wove as a fabric, which was both untreated and
chemically treated by either one hour soaking in acetic
anhydride at 70C (acetylation) or dipped in 2% NaOH
solution for 48 hours (alkali/mercerization), reinforcing
both conventional isotactic polypropylene (iPP) and an-
hydride grafted polypropylene resins (MAiPP) matrices
composites. In addition, the authors[82] also investigated
the effect of accelerated weathering of the composites by
immersion in boiling water for two hours before conduct-
ing thermogravimetric analysis at a heating rate of 10C/
min in nitrogen. Table 3, reproduced from their work[82],
summarizes the main parameters obtained from TG curves
for all distinct composites.
Although not mentioned in the abstract, the authors[82]
indicated in the experimental procedure that the plain
weave kapok/cotton fabric was used together with non-
woven glass mat. Along the article, no comment or discus-
sion was given regarding the role of the glass mat on the
thermogravimetric analysis. As shown in Table 3, all com-
posites display a dehydration temperature in the range of
59C78C, which is probably related to the water release
Fig. 4 SEM micrographs: (a) untreated; (b) mercerized; and (c) silanized
buriti bers as well as (d) composite with 10 wt.% mercerized bers,
from the kapok and cotton bers, as usually reported in
indicated by white arrows. Inserts with high magnication details. other lignocellulosic ber composites[3032,50,53,54,56,71,74]. As
Reproduced from Santos et al.[80]. major thermal stability results, Mwaikambo et al.[82], em-

J. Mater. Res. Tecnol. 2012; 1(2):117-126

Thermogravimetry of Composites with Less Common Natural Fibers 123

Table 3 Thermogravimetric results of polypropylene reinforced with kapok/cotton for different alkali- or acetylated-treated fabrics
conventional or maleic anhydride grafted matrices and unweathered and weathered composites.

Alkali-treated kapok/ Acetylated kapok/ Unweathered kapok/

Properties cotton-iPP composite cotton-iPP composite cotton-iPP composite

Dehydration temperature (oC) 74.1 59.3 76.3

Onset temperature (oC) 332.3 339.5 316.4
Degratation temperature 1 (oC) 372.8 372.63 370.4
Degratation temperature 2 (oC) 431.7 428.2
Decomposition temperature ( C)
656.7 655.3 656.4

Unweathered kapok/ Weathered kapok/ Weathered kapok/

Properties cotton-MAiPP composite cotton-iPP composite cotton-MAiPP composite

Dehydration temperature (oC) 77.5 71.8 71.8

Onset temperature ( C)
339.5 318.5 339.2
Degratation temperature 1 (oC) 381.9 380.6 384.2
Degratation temperature 2 (oC) 459.1 423.7 462.5
Degratation temperature 2 (oC) 657.9 658.7 655
Reproduced from Amar B et al.[81].

phasized that the weathered kapok/cotton-iPP composite tensile strength above 800 MPa and elastic modulus above
shows the highest degradation temperature followed by 4 GPa for okra bers. These less common lignocellulosic
kapok/cotton-MAiPP composite. As for the fabric treat- bers certainly possess an engineering potential, of both
ment, the acetylated kapok/cotton fabric-iPP composites strength and stiffness, for polymer composite reinforce-
display a higher onset temperature of degradation than ment. In addition to future specic works on the mechani-
both the weathered composite and mercerized kapok/ cal properties of these less common natural ber com-
cotton fabric-iPP composites, but lower than MAiPP com- posites, the thermal stability of such composites will also
posites. The authors[82] stated that the maleic anhydride need to be investigated for practical use.
is likely to be the possible source for the improved ther- A second point of the present overview is the alert
mal properties of PP as also evidenced by the rise in the on the growing demand for studies covering the several
degradation temperature of the composites. By contrast, less common lignocellulosic bers with potential applica-
mercerization of bers, which is known to result in re- tion as engineering materials. Consequently, the general
duced crystalline cellulose, has been found to decrease properties and, in particular, the TG/DTG analysis of nov-
the thermal stability of the kapok/cotton fabric-iPP com- el composites based on these bers have to be assessed.
posites. As a general conclusion, it has also been found, Despite the limited and fragmented information existing
with exception of mercerization kapok/cotton fabric-iPP in the literature, this overview conveys relevant conclu-
(Table 3), that all composites exhibit two degradation sion regarding the thermogravimetric stability of polymer
temperatures. Mwaikambo et al.[82] suggested that the composites reinforced with less common lignocellulosic
rst degradation temperature is a result of depolymer- bers. Apparently, an initial weight loss associated with a
ization of the cellulose materials. The second degrada- water release DTG peak below 200C is a common feature
tion temperature was caused by the polymeric matrices of these composites[39,71,74,81,82], although not emphasized
breakdown into monomers and/or decomposition of the in some works[73,78,80]. This initial peak is most probably
levoglucosan. a result of the evaporation of water from the ber sur-
face, since the polymeric matrix contribution, if existing,
5. Concluding Remarks should be relatively small. For practical use, however, the
temperature related to the onset of thermal degradation
A few points are worth being discussed concerning the re- can be considered the composite thermal stability limit.
sults presented in this overview. First, the reader should In the overviewed works, this limit was found to be in
noticed the relatively small number of papers[39,71,72,74,78,8082] the range of 240C[80]355C[39], and attributed to the li-
covering the thermogravimetric stability of polymer com- gnocellulosic ber decomposition. According to Santos et
posites reinforced with less common lignocellulosic bers. al.[80], these onset degradation temperatures are interme-
Indeed, it was rather surprising that only such limited diate between the isolated ber and the polymer matrix.
works have so far been dedicated to these composites in Other higher temperature DTG peaks, 422C [71]463C[81],
spite of the potential presented by less common lignocel- are related to the polymer matrix macromolecular deg-
lulosic bers. For instance, Fiore et al.[83] reported ten- radation or depolymerization but not as important as the
sile strength above 300 MPa and elastic modulus above onset degradation temperature to dene the composite
1.5 GPa for artichoke bers, while De Rosa et al.[84] found thermogravimetric stability.

J. Mater. Res. Tecnol. 2012; 1(2):117-126

124 Monteiro et al.

Acknowledgements 23. Beall FC, Eickner HW. Thermal degradation of wood compo-
nents: a review of literature. USDA Forest Service Research
The authors thank the support to this investigation by the Paper 1970; FPL 130.
Brazilian agencies: CNPq, CAPES and FAPERJ. 24. Kilzer FJ. Cellulose and Cellulose Derivatives. New York: Wiley-
Interscience, 1971, p. 101722.
References 25. Nguyen T, Zavarin E, Barral EM. Thermal analysis of lignocel-
lulosic materials. Part I - Unmodied materials. J Macromol
1. Bledzki AK, Gassan J. Composites reinforced with cellulose- Sci Rev Macromol Chem 1981; C 20:165.
based bers. Prog Polym Sci 1999; 4:22174. 26. Nguyen T, Zavarin E, Barral EM. Thermal analysis of lignocel-
2. Mohanty AK, Misra M, Hinrichsen G. Biober, biodegradable lulosic materials. Part II - Modied materials. J Macromol Sci
polymers and biocomposites: an overview. Macromol Mat Eng Rev Macromol Chem 1981; C 21:160.
2000; 276/277:124. 27. Mohanty AK, Patnaik S, Singh BC, Misra M. Graft copolymeriza-
3. Agarwal BD, Broutman LJ. Analysis and Performance of Fiber tion of acrylonitrile onto acetylated jute bers. J Appl Polym
Composites. 2.ed. New York: Wiley, 1990. Sci 1989; 37(5):117181.
4. Ashbee KH. Fundamental Principles of Fiber Reinforced Com- 28. Sabaa MW. Thermal degradation behavior of sisal bers graft-
posites. 2.ed. Lancaster, PA: Technomic, 1993. ed with various vinyl monomers. Polym Degrad Stab 1991;
5. Chawla KK. Composite Materials Science and Engineering. 32(2):20917.
2. ed. New York: Springer-Verlag, 1998. 29. Sridhar MK, Basavarajjappa G, Kasturi SS, Balsubramanian N.
6. Nabi Sahed D, Jog JP. Natural ber polymer composites: a re- Thermal stability of jute bers. Indian J Tex Res 1982; 7:8791.
view. Adv Polym Technol 1999; 18(4):35163. 30. Das S, Saha AK, Chourdhury PH, Basak RK, Mitra BC, Todd T
7. Eichhorn SJ, Baillie CA, Zafeiropoulos N, Mwaikambo LY, An- et al. Effect of steam pre-treatment of jute ber on dimensional
sell MP, Dufresne A et al. Review Current international re- stability of jute composite. J Appl Polym Sci 2000; 76:165261.
search into cellulosic bres and composites. J Mat Sci 2001; 31. Dash BN, Rana AK, Mishra HK, Nayak SK, Tripathy SS. Novel low-
36:210731. cost jute-polyester composites. III. Weathering and thermal
8. Mohanty AK, Misra M, Drzal LT. Sustainable bio-composites from behavior. J Appl Polym Sci 2000; 78:16719.
renewable resources: opportunities and challenges in the green 32. Ray D, Sarkar BK, Basak RK, Rana AK. Thermal behavior of vinyl
materials world. J Polym Environ 2002; 10:1926. ester resin matrix composites reinforced with alkali-treated
9. Netravali AN, Chabba S. Composites get greener. Mater Today jute bers. Appl Polym Sci 2004; 94:1239.
2003; 6:229. 33. Manfredi LB, Rodriguez ES, Wladyka-Przybylak M, Vazquez A.
10. Mohanty AK, Mishra M, Drzal LT. Natural Fibers, Biopolymers Thermal degradation and re resistance of unsaturated poly-
and Biocomposites. Boca Raton: CRC press, 2005. ester, modied acrylic resins and their composites with natural
11. Satyanarayana KG, Guimares JL, Wypych F. Studies on lignocel- bres. Polym Degrad Stab 2006; 91:25561.
lulosic bers of Brazil. Part I: Source, production, morphology, 34. Mohanty S, Verma SK, Nayak SK. Dyamic mechanical and ther-
properties and applications. Compos Part A 2007; 38:1694709. mal properties of MAPE treated jute/HDPE composites. Compos
12. Crocker J. Natural materials innovative natural composites. Sci Technol 2006; 66:53847.
Mater Technol 2008; 23:1748. 35. Doan T-T-L, Brodowsky H, Mder E. Jute bre/polypropylene
13. Monteiro SN, Lopes FPD, Ferreira AS, Nascimento DCO. Natural composites. II. Thermal, hydrothermal and dynamic mechani-
ber polymer matrix composites: cheaper, tougher and envi- cal behavior. Compos Sci Technol 2007; 67:270714.
ronmentally friendly. JOM 2009; 61:1722. 36. Yu T, Li Y, Ren J. Preparation and properties of short natural
14. Kalia S, Kaith BS, Kaurs I (eds.). Cellulose Fibers: Bio- and ber reinforced poly(lactic acid) composites. Trans Nonferrous
Nano- Polymer Composites. New York: Springer, 2011. Met Soc China 2009; 19:s6515.
15. Monteiro SN, Lopes FPD, Barbosa AP, Bevitori AB, Silva ILA, 37. Aziz SH, Ansell MP. The effect of alkalinization and bre align-
Costa LL. Natural lignocellulosic bers as engineering materi- ment on the mechanical and thermal properties of kenaf and
als. Metal Mater Trans A 2011; 42:296374. hemp bast bre composites: Part 1 Polyester resin matrix.
16. Larbig H, Scherzer H, Dahlke B, Poltrock R. Natural ber re- Compos Sci Technol 2004; 64:121930.
inforced foams based on renewable resources for automotive 38. Aziz SH, Ansell MP. The effect of alkalinization and bre align-
interior applications. J Cell Plast 1998; 34:36179. ment on the mechanical and thermal properties of kenaf and
17. Marsh G. Next step for automotive materials. Mater Today hemp best bre composites: Part 2 Cashew nut shell liquid
2003; 6:3643. matrix. Compos Sci Technol 2004; 64:12318.
18. Holbery J, Houston D. Natural-ber-reinforced polymer com- 39. Sgriccia N, Hawley MC. Thermal, morphological, and electrical
posites in automotive applications. JOM 2006; 58:806. characterization of microwave processed natural ber compos-
19. Zah R, Hischier R, Leo AL, Brown I. Curaua bers in automo- ites. Compos Sci Technol 2007; 67:198691.
bile industry A sustainability assessment. J Clean Prod 2007; 40. Pracella M, Chionna D, Anguillesi I, Kulinski Z, Piorkowska E.
15:103240. Functionalization, compatibilization and properties of polypro-
20. Alves C, Ferro PMC, Silva AJ, Reis LG, Freitas M, Rodrigues LB. pylene composites with hemp bre. Compos Sci Technol 2006;
Ecodesign of automotive components making use of natural 66:221830.
jute ber composites. J Clean Prod 2010; 18(4):313-27. 41. Beckermann GW, Pickering KL. Engineering and evaluation of
21. Wambua P, Ivens I, Verpoest I. Natural bers: can they replace hemp bre reinforced polypropylene composites: bre treatment
glass and bre reinforced plastics? Compos Sci Technol 2003; and matrix modication. Composites Part A 2008; 39:97988.
63:125964. 42. Moriana R, Vilaplana F, Karlsson S, Ribes-Greus A. Improved
22. Joshi SV, Drzal LT, Mohanty AK, Arora S. Are natural ber com- thermo-mechanical properties by the addition of natural bres
posites environmentally superior to glass ber reinforced in starch-based sustainable biocomposites. Composites Part A
composites? Compos Part A 2004; 35:3716. 2011; 42:3040.

J. Mater. Res. Tecnol. 2012; 1(2):117-126

Thermogravimetry of Composites with Less Common Natural Fibers 125

43. Albano C, Gonzalez J, Ichazo M, Kaiser D. Thermal stability of 61. Lei Y, Wu Q, Yao F, Xu Y. Preparation and properties of recy-
blends of polyolens and sisal ber. Polym Degrad Stab 1999; cled HDPE/natural ber composites. Composites Part A 2007;
66:17990. 38(7):166474.
44. Joseph PV, Joseph K, Thomas S, Pillai CKS, Prasad VS, Groen- 62. George J, Bhagawan SS, Thomas S. Thermogravimetric and
inckx G et al. The thermal and crystallization studies of short dynamic mechanical thermal analysis of pineapple bre re-
sisal bre reinforced polypropylene composites. Composites inforced polyethylene composites. J Thermal Anal Cal 1996;
Part A 2003; 34:25366. 47(4):112140.
45. Nair KCM, Thomas S, Groeninckx G. Thermal and dynamic me- 63. Luo S, Netravali AN. Mechanical and thermal properties of envi-
chanical analysis of polystyrene composites reinforced with ronment-friendly green composites made from pineapple leaf
short sisal bres. Compos Sci Technol 2001; 61:251929. bers and poly (hydroxybutyrate-co-valerate) resin. Polym Com-
46. Xie XL, Fung KL, Li RKY, Tjong SC, May Y-W. Structural and pos 1999; 20(3):36778.
mechanical behavior of polypropylene maleated styrene- 64. Threepopnatkul P, Kaerkitcha N, Athipongarporn N. Effect of sur-
(ethylene-co-butylene-styrene/sisal ber composites prepared face treatment on performance of pineapple leaf berpolycar-
by injection molding. J Polym Sci Part B: Polym Phys 2002; bonate composites. Composites Part B 2009; 40(7):62832.
40:121422. 65. Lee S-H, Wang S. Biodegradable polymers/bamboo ber biocom-
47. Xie XL, Li RKY, Tjong SC, May Y-W. Structural properties and me- posite with bio-based coupling agent. Composites Part A 2006;
chanical behavior of injection molded composites of polypropyl- 37(1):8091.
ene and sisal ber. Polym Compos 2002; 23(3):31928. 66. Shih Y-F. Mechanical and thermal properties of waste water
48. Gan P, Garbizu S, Llano-Ponte R, Mondragon I. Surface modi- bamboo husk ber reinforced epoxy composites. Mater Sci
cation of sisal bers: effects on the mechanical and thermal Eng A 2007; 4456:28995.
properties of their epoxy composites. Polym Compos 2005; 67. Singh S, Mohanty AK, Sugie T, Takai Y, Hamada H. Renewable
26:1214. resource based biocomposites from natural ber and polyhy-
49. Paiva JMF, Frollini E, Macromol J. Unmodied and modied sur- droxybutyrate-co-valerate (PHBV) bioplastic. Composites Part
A 2008; 39(5):87586.
face sisal bers as reinforcement of phenolic and lignophenolic
68. Zainudin ES, Sapuan SM, Abdan K, Mohamad MTM. Thermal
matrices composites: Thermal analyses of bers and composites.
degradation of banana pseudo-stem lled unplasticized poly-
Mater Eng 2006; 291:40517.
vinyl chloride (UPVC) composites. Materials and Design 2009;
50. Kim JT, Netravali AN. Mechanical and thermal properties of sisal
ber-reinforced green composites with soy protein/gelatin res-
69. Moth CG, de Arajo CR, de Oliveira MA, Yoshida MI. Thermal
ins. Biobased Mater Bioenergy 2010; 4(4):33845.
decomposition kinetics of polyurethane-composites with ba-
51. Arbelaiz A, Fernandez B, Ramos JA, Mondragon I. Thermal and
gasse of sugar cane. J Therm Anal Cal 2002; 67(2):30512.
crystallization studies of short ax bre reinforced polypropyl-
70. Jawaid M, Abdul Khalil HPS. Cellulosic/synthetic bre rein-
ene matrix composites: effect of treatments. Thermochim Acta
forced polymer hybrid composites: A review. Carbohydrate
2006; 440(2):11121.
Polym 2011; 86(1):118.
52. Aydin M, Tozlu H, Kemaloglu S, Aytac A, Ozkoc G. Effects of Alkali
71. Moth CG, de Araujo CR. Thermal and mechanical character-
Treatment on the Properties of Short Flax FiberPoly(Lactic Acid)
ization of polyuretane composites with curaua bers (in Portu-
Eco-Composites. J Polym Environm 2011; 19(1):117.
guese). Polmeros: Cincia e Tecnologia 2004; 14(4):27488.
53. Rosa MF, Chiou B-S, Medeiros ES, Woods DF, William TG, Mat-
72. Arajo JR, Waldman WR, de Paoli M-A. Thermal properties of
toso LHC et al. Effect of ber treatments on tensile and thermal
high density polyethylene composites with natural bres: Cou-
properties of starch/ethylene vinyl alcohol copolymers/coir bio- pling agent effect. Polym Degrad Stab 2008; 93(10):17705.
composites. Bioresource Technol 2009; 100(21):5196202. 73. Waldman WR, de Paoli M-A. Thermo-mechanical degradation
54. Santaf Jr. HPG, Rodriguez RJS, Monteiro SN, Castillo TE. Char- of polypropylene, low-density polyethylene and their 1:1
acterization of thermogravimetric behavior of polyester compos- blend. Polym Degrad Stab 1998; 60(23):3018.
ites reinforced with coir ber. In: Characterization of Minerals, 74. Ferreira AS, Rodriguez RJS, Lopes FPD, Monteiro SN. Thermal
Metals and Materials Symposium TMS 2011 Annual Conference, analysis of curaua ber reinforced polyester matrix compos-
San Diego, CA, USA, March 2011, p. 16. ites. In: Characterization of Minerals, Metals and Materials
55. Silva CG, Benaducci DB, Frollini E. Lyocell and cotton bers as Symposium TMS Conference (TMS 2010), Seattle, WA, USA,
reinforcements for a thermoset polymer. BioResources 2012; March 2010, p. 17.
7(1):7898. 75. Santos PA, Spinac MAS, Fermoselli KKG, de Paoli M-A. Poly-
56. Julkapli NM, Akil HM. Thermal Properties of Kenaf-Filled Chito- amide-6/vegetal ber composite prepared by extrusion and
san Biocomposites. Polym-Plast Technol Eng 2010; 49(2):14753. injection molding. Composites Part A 2007; 38(12):240411.
57. Akil HM, Omar MF, Mazuki AAM, Saee S, Ishak ZAM, Bakar AA. 76. Tomczak F, Satyanarayana KG, Sydenstricker THD. Studies
Kenaf ber reinforced composites: A review. Mater Design 2011; on lignocellulosic bers of Brazil: Part III Morphology and
32(8-9):410721. properties of Brazilian curaua bers. Composites Part A 2007;
58. Coutinho FMB, Costa THS, Carvalho DL, Gorelova MM, de San- 38(10):222736.
ta Maria LC. Thermal behaviour of modied wood bers. Polym 77. Spinac MAS, Lambert CS, Fermoselli KKG, de Paoli M-A. Char-
Test 1998; 17(5):299310. acterization of lignocellulosic curaua bres. Carbohydrate
59. Doh GH, Lee S-Y, Kang I-A, Kong Y-T. Thermal behavior of liq- Polym 2009; 77(1):4753.
ueed wood polymer composites (LWPC). Compos Struct 2005; 78. Gan P, Mondragon I. Thermal and degradation behavior of
68(1):1038. que ber reinforced thermoplastic matrix composites. J
60. Bhardwaj R, Mohanty AK, Drzal LT, Pourboghrat F, Misra M. Re- Therm Anal Cal 2003; 73(3):78395.
newable resource-based green composites from recycled cel- 79. Gan P, Mondragon I. Surface modication of que bers.
lulose ber and poly(3-hydroxybutyrate-co-3-hydroxyvalerate) Effect on their physico-mechanical properties. Polym Compos
bioplastic. Biomacromolecules 2006; 7(6):204451. 2002; 23(3):38394.

J. Mater. Res. Tecnol. 2012; 1(2):117-126

126 Monteiro et al.

80. Santos RS, Souza AA, de Paoli M-A, Souza CML. Cardanolform- 82. Mwaikambo LY, Martuscelli E, Avella M. Kapok/cotton fabric
aldehyde thermoset composites reinforced with buriti bers: polypropylene composites. Polym Test 2000; 19(8):90518.
Preparation and characterization. Composites Part A 2010; 83. Fiore V, Valenza A, di Bella G. Artichoke (Cynara carduncu-
lus L.) bres as potential reinforcement of composite struc-
tures. Compos Sci Technol 2011; 71(8):113844.
81. Amar B, Salem K, Hocine D, Chadia I, Juan MJ. Study and
84. De Rosa IM, Kenny JM, Puglia D, Santulli C, Sarasimi F. Mor-
characterization of composite materials based on polypro- phological, thermal and mechanical characterization of okra
pylene load with olive husk our. J Appl Polym Sci 2011; 122: (Abelmoschus esculentus) bres as potential reinforcement in
138294. polymer composites. Compos Sci Technol 2010; 70(1):11622.

J. Mater. Res. Tecnol. 2012; 1(2):117-126