The effect of steam explosion on the production of sugarcane
bagasse/polyester composites
Ricardo José Brugnago
a,1, Kestur Gundappa Satyanarayana
b,c,⇑,2, Fernando Wypych
a,1,
Luiz Pereira Ramos
a,1a
Research Center in Applied Chemistry (CEPESQ), Department of Chemistry, Federal University of Paraná, P.O. Box 19081, Curitiba, PR 81531-990, Brazil
bAcharya R&D Center, Acharya Institute of Technology, Soladevanahalli, Hesaraghatta Road, Chikkabanasandra Post, Bangalore 560 090, Karnataka, India c
BMS College of Engineering, Bangalore and Poornaprajna Institute of Scientific Research, Bangalore, India
a r t i c l e
i n f o
Article history:Received 23 April 2010
Received in revised form 20 November 2010 Accepted 11 December 2010
Available online 16 December 2010
Keywords:
A. Polymer matrix composites (PMCs) E. Surface treatment
B. Thermal properties D. Chemical analysis
a b s t r a c t
This paper describes a pretreatment option to sugarcane bagasse fibers for their use in composite prep-aration with unsaturated polyester. Sugarcane bagasse fibers modified by (i) steam explosion and (ii) alkali washing after steam explosion, along with (iii) as-received bagasse fibers were characterized. Steam explosion significantly reduced the amount of hemicelluloses and acid-soluble lignin of bagasse fibers, while acid-insoluble lignin increased proportionally. Alkaline washing of steam-exploded fibers removed nearly 60% of their acid-insoluble lignin. Polyester matrix composites containing 10 wt.% of these fibers were prepared by compression molding. Density, thermal stability, water absorption and thermomechanical analysis of the composites containing steam explosion treated bagasse fibers showed improvement in these properties over those of the untreated fiber containing composite. These are explained in terms of the chemical modifications that occurred due to the steam explosion treatments.
Ó 2010 Elsevier Ltd.
1. Introduction
Sugarcane is an energy crop that is largely grown on flat land in countries such as Brazil, China, India, Thailand and Australia. Brazil is the largest world producer of this crop[1], where it is cultivated as a tropical plantation [2] with an annual production that in-creased from about 387 Mt in 2004/2005 to about 570 million in 2008/2009[3]. About 60% of the sugarcane grown in Brazil is used to produce ethanol[4], underscoring the importance of this crop for the local economy. On the other hand, sugarcane bagasse, the main agro-industrial residue of sugarcane processing, is reported to be 101 Mt from 340 Mt of sugarcane[5].
Sugarcane bagasse is a rather heterogeneous material, consist-ing of three main components: fibers (73%), pith (5%) and rind (22%). This lignocellulosic material is reported to have about 40 different uses including both short-term and medium-term appli-cations for power generation[6,7]. One such application is the pro-duction of composites, particularly with polymer matrices.
Towards this end, some of the available literature report about the chemical composition of the fibers, as well as the physical (density, crystallinity, microfibrillar angle, moisture absorption, etc.) and mechanical (tensile, flexural and impact for composites) properties of both fibers and some of their polymeric composites with and without surface-treated fibers [8–29]. Morphological studies of fibers (fiber dimension, shape and size of cells) and their polymeric composites are also reported, the latter being used to mainly explain the observed strength properties. The surface treat-ment is usually made by extraction procedures such as alkali washing to improve the fiber adhesion to the polymer matrices.
Of all of the proposed and tested fiber pretreatment methods to date for lignocellulosic materials[30–32], the best ones are those in which both physical and chemical treatments are combined. High-pressure steaming (steam explosion, SE) belongs to this and is one of the most promising techniques for fractionating lignocel-lulosic materials including wood into its three major components while enhancing the susceptibility of cellulose to chemical and/or biological conversion. SE allows the fractionation and recovery of the three main components (lignin, cellulose and hemicelluloses) of the cell wall in high yields. Further, removal of hemicelluloses and lignin results in fibers with large pore volumes and increased surface area, while the cellulose component will be of high degree of crystallinity due to hydrolysis of its amorphous regions and reorganization of its supramolecular structure during sample prep-aration and handling[30]. Therefore, upon optimization, steam
1359-835X Ó 2010 Elsevier Ltd. doi:10.1016/j.compositesa.2010.12.009
⇑Corresponding author. Present address: Acharya R&D Center, Acharya Institute of Technology, Soladevanahalli, Hesaraghatta Road, Chikkabanasandra Post, Bangalore 560 090, Karnataka, India. Tel.: +91 (80) 2666 0140; fax: +91 (80) 2661 4357.
E-mail address:kgs_satya@yahoo.co.in(K.G. Satyanarayana).
1
Tel.: +55 (41)3361 3470; fax: +55 (41)3361 3186.
2 Formerly at PIPE/Department of Chemistry, Federal University of Paraná, Curitiba,
PR Brazil.
Composites: Part A 42 (2011) 364–370
Contents lists available atScienceDirect
Composites: Part A
j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c o m p o s i t e s a
Open access under the Elsevier OA license.
Open access under the Elsevier OA license.
CORE Metadata, citation and similar papers at core.ac.uk
explosion may result in materials with lower fiber density, im-proved strength properties, higher thermal stability and lower moisture absorption, depending on the chemistry and distribution of the lignin component in the steam-treated fibers. More details including the advantages and disadvantages of this process can be found in Ref.[32], which also reports that this technology can lead to a zero emissions [ZERI] model for industries dealing with lignocellulosic materials.
The present work is part of the systematic studies carried out by the authors on the characterization of Brazilian lignocellulosic fi-bers for their use in the preparation of composites. Some of the re-sults of the hitherto published work by the authors on Brazilian fibers include: details on the sources, extraction, availability, struc-ture-properties and uses of these fibers[28], tensile properties as functions of fiber dimensions, thermal properties and morphology of coir and curauá fibers[33,34], chemical, thermal and morpho-logical characterization of banana, sugarcane bagasse and sponge gourd (Luffa cylindrica) fibers[35]and use of sugarcane bagasse fi-bers as reinforcement in PHB and corn starch matrices[29,36]. The present study deals with (i) characterization of sugarcane bagasse fibers with and without pretreatment by steam explosion; (ii) preparation of their composites with unsaturated polyester, and fi-nally, (iii) characterization of these composites to assess the effect of steam explosion on surface properties of the fiber for composite making.
2. Materials and methods
Sugarcane bagasse was obtained from a local supplier (Usina Melhoramentos S/A, Maringá, Brazil). They were conditioned prior to the steam explosion (pretreatment) by hot water extraction. This consists of water reflux for 1 h, which helps in removing any residual sucrose, followed by drying at 50 °C until constant weight. Sugarcane bagasse with a typical moisture content of 10% (m/ m) was subjected to steam explosion treatment (hereafter also re-ferred as ‘steam-treated’) at the Pilot Plant Unit of Federal Univer-sity of Paraná using a 5 L steam gun. The bagasse was impregnated with dilute phosphoric acid (H3PO4) and pretreated at a saturated
steam temperature of 203 °C for 3 min. These processing parame-ters were based on preliminary studies reported elsewhere [29,36]. Acid impregnation was carried out by spraying the air-dried sugarcane bagasse with 0.3 mol L 1H
3PO4at room
tempera-ture (RT). Acid-impregnated fibers had a final moistempera-ture content of 50 wt.%. Steam explosion was performed by loading the acid-impregnated bagasse samples directly into the steam reactor. After treatment under the conditions mentioned above, the ‘‘steam-trea-ted bagasse’’ (STB) was released from the steam reactor by rapid depressurization of the pressure vessel, causing the material to ex-pand (explode) into a stainless steel cyclone from which it was recovered by scrubbing and washing the internal walls with run-ning water. The resulting STB fibers were filtered in a Büchner fun-nel, re-suspended in water and washed once again at a 5 wt.% solids for 1 h at room temperature with constant stirring. After draining the washed water under low pressure, the water-insolu-ble fraction was recovered and stored at 4 °C for further use. This fraction was called ‘STB-WI’ (water-insoluble steam-treated ba-gasse), while the water-soluble fraction was called ‘STB-WS’ (water-soluble steam-treated bagasse).
Water-washed steam-exploded fibers were then subjected to an alkaline washing with 1 mol L 1of NaOH at 373 K for 1 h under
reflux at a consistency of 4 wt.%. After this, the fibrous material was filtered and exhaustively washed with water (until neutral pH was observed) for subsequent use in the preparation of composites.
The chemical analyses of the as-received (hereafter referred as ‘‘AR’’), STB-WI and steam-treated and alkali-washed fibers
(hereafter referred as ‘‘STB-AW’’) were carried out always in triplicate, following the modified Klason lignin method[37]. The acid-insoluble lignin was measured gravimetrically and the total carbohydrates (cellulose and polyoses) were characterized by high-performance liquid chromatography (HPLC). The Klason lignin hydrolysates were centrifuged and analyzed using a Shimadzu LC10AD system, provided with a SIL10A auto sampler, a RID10A model refraction index detector and a SPD-M10A ultraviolet detec-tor. The analysis was carried out in an Aminex HPX-87H column (Bio-Rad) at 65 °C, preceded by a pre-column Cation-H with 8 mmol L 1H
2SO4as the mobile phase at a flow rate of 0.6 mL min 1. The
re-sults were quantified by external calibration using standard solu-tions of celobiose, glucose, xylose, arabinose and acetic acid.
The diameter (width) of each fiber was measured using projec-tion microscopy as well as with a micrometer screw gauge (for comparison) to the nearest 0.01 mm, and the length was measured to the nearest 0.1 mm with digital calipers.
Laminates of plain polyester were first prepared as per Brazilian NBR 7143 standard for the compression molding of thermoplastics. In short, about 100 g of polyester was placed in a square steel mold (250 250 mm) and subjected to a pressure of 0.94 MPa for 1 h at 150 °C, after increasing the temperature and the load/pressure en-ough to make the material flow without any thermal decomposi-tion. The temperature, load/pressure and time were chosen after some preliminary experiments. Then, the material was cooled at a cooling rate of 20 °C min 1 down to a temperature at which the laminates could be easily removed from the mold. The same procedure was followed for the composites using both treated and untreated bagasse fibers of 1 mm in length and 0.25 mm in diameter (width) with the substitution of 10 g of polyester by 10 g of dry fibers in the three conditions mentioned above (AR, STB-WI and STB-AW). These samples are marked as follows: (i) composites with as-received bagasse fibers as ‘‘AR-PC’’, (ii) com-posites containing STB-WI fibers as ‘‘STB-WIPC’’ and (iii) compos-ites containing STB-AW fibers as ‘‘STB-AWPC’’. PPC stands for the plain polyester matrix. Hereafter the above nomenclatures will be used in the paper.
A homogeneous fiber–matrix combination was obtained by compounding them in a mechanical mixer. Triplicates for each of the conditions described above were obtained. In all cases, the fi-bers were previously dried and Willey-milled to eliminate possible fiber aggregates, which can hinder the homogeneity of the final ba-gasse/polyester composite.
The density of AR, STB-WI and STB-AW fibers was determined as per ISO 1183-1 standard using n-pentane. On the other hand, the density of the composite (10 mm wide and 50 mm long) was determined according to ABNT 119931/119937. In both cases, a pycnometer was used for the determination. A minimum of five replicates were carried out and the average values are reported.
Water absorption of the composites was determined according to ASTM D 570-95. Briefly, samples measuring 60 20 0.5 mm were conditioned in a hot air oven for 24 h. at 105 °C. The weights of the samples were measured after they were put in distilled water, and the measurement continued for 7 days. The experiment was repeated until the sample showed constant weight, indicating no significant absorption of water had taken place.
The dynamic-mechanical analysis (DMA) of the composites was carried out using a Netzsch instrument, model DMA 242, with three-point bending at constant load and heating rate of 3 °C min 1
in the temperature range of 150 °C to 180 °C in both N2
(non-oxi-dizing) and O2 (oxidizing) atmospheres. The gas flowrate was
maintained at 50 mL min 1at a frequency of 1 Hz.
The thermal stability of the composites was evaluated through thermal analysis using a Netzsch DSC 200 and a TG 209 analysers under N2and O2atmospheres with a heating rate of 20 °C min 1up
3. Results and discussion 3.1. Density of sugarcane bagasse
Density values of three types of bagasse fibers (AR, STB-WI and STB-AW) were found to be 932, 695 and 437 kg m 3respectively.
Lower density values in treated fibers can be understood as due to the partial degradation of cellulose and complete degradation of both hemicelluloses and lignin in the treated fibers. While the degradation of hemicelluloses and cellulose has been reported [38] as due to hydrolysis of glycosidic bonds, lignin is modified through hydrolysis and radical through radical reactions involving its units. Accordingly, bigger pores would result from removal of both hemicelluloses and lignin by organic solvents or aqueous media[38].
3.2. Chemical composition
The chemical composition of sugarcane bagasse in the as-re-ceived condition along with that reported in the literature is shown inTable 1. The analyses were performed and the reported values are expressed in relation to the dry weight of raw bagasse. It can be seen that the cellulose content (38.8%) obtained in the present study is lower than that reported by others for these type of fibers from Brazil and China[2,8,19,20], but comparable with those from elsewhere[15]. The other constituents are somewhat comparable giving the total mass balance of 94.3%. Balance of about 6% could be due to ash and other extracts, which were not determined in this study. The difference in chemical composition of the fiber by various researchers is understandable since the values reported are usually not normalized taking into account all the constituents. Also, these discrepancies could be due to the locality from where the fibers were obtained, their age, etc., in addition to the method used for analysis, their accuracy and experimental conditions [30,35].
Table 2lists the amount of insoluble (STB-WI) and water-soluble (STB-WS) fractions that were recovered from the phospho-ric acid-catalyzed steam-exploded sugarcane bagasse. The overall mass balance of the pretreated fractions revealed a recovery of 80.2% in relation to the dry weight of the starting material. This corresponds to yields of 65.5% in STB-WI, 14.7% in STB-WS and 19.8% in mass loss as volatiles and/or other unidentified compounds.
The chemical analysis of the STB-WI fraction is shown inTable 3. On average, the STB-WI fraction showed 4.2% of ash, 27.7% of lig-nin, 58.5% of cellulose and 9.1% of hemicelluloses, for a total mass balance of 99.5%. Furthermore, the hemicellulose component was composed primarily of heteroxylans containing 8.0% xylose (ex-pressed as xylan), 0.6% arabinose (ex(ex-pressed as arabinan) and 0.5% acetyl groups in relation to the original bagasse dry weight. These values compare well with earlier values reported elsewhere [9–12,31,32]. Also, the higher ash contents obtained in this study are still within the range reported earlier[20].
Alkali washing after steam explosion resulted in removal of approximately 60% of the lignin component found in the STB-WI
fraction. This fraction (STB-AW) contained 11% of total lignin (1.1% acid soluble + 9.9% acid insoluble), 86% cellulose and 2% hemicelluloses (mostly xylan containing trace amounts of arabi-nose and 0.6% acetyl groups), giving the total mass balance of 99%. The amount of soluble lignin decreased with the SE pretreat-ment of the fiber but increased with the subsequent alkaline wash-ing. On the other hand, the amount of acid insoluble lignin showed the reverse trend in both the cases. The percentage of cellulose in-creased along the process, while that of hemicelluloses dein-creased. These properties were considered interesting for the fundamental objective of this work.
3.3. Density of composites
Using the measured density of the three types of sugarcane ba-gasse fibers mentioned in Section3.1and the reported the density of polyester, theoretical density of all composites samples were calculated using the rule of mixtures. Both theoretical and experi-mental density values are listed inTable 4.
It is interesting to note that theoretical density values of com-posites follow a definite trend (decreasing with treatments) while those of experimental do not follow any definite trend. For exam-ple, the experimental density value is the lowest for the AR-PC fol-lowed by that of STB-AWPC and PPC, while the STB-WIPC sample
Table 1
Chemical composition of sugarcane bagasse (%).
Cellulose Lignin Hemicellulose Ash Silica/extracts Moisture content Refs.
54.3 24.4 29.7 7 – 45–52 [2] 40–50 15–35 25–35 – – – [8] 40 20 30 – 10 [15] 32–48 19–24 27–32 1.5–5 0.7–3.5 – [15] 55.2 25.3 16.8 0.5–1.0 9.5 [19,20] 38.8 24.1 27.6 3.80 – – Present study Table 2
Recovery yields of water-soluble (STB-WS) and water-insoluble (STB-WI) fractions of sugarcane bagasse after pretreatment by steam explosion.
Experiment STB-WI (%) STB-WS (%) Total (%)
1 64.8 11.5 76.3
2 68.2 13.8 82.0
3 63.4 18.9 82.3
Table 3
Chemical composition (%) of sugarcane bagasse after SE pretreatment + water washing and alkali washed after SE + water washing.
Constituent Steam exploded and water washed
Steam exploded and water washed and alkali washed
Cellulose 58.5 86 Hemicellulose 9.1 11 Lignin 27.7 2 Ash 4.2 n.d. n.d.: Not determined. Table 4
Theoretical and experimental density values of plain polyester matrix (PPC) and its composites containing treated and untreated bagasse fibers.
Composite Theoretical density (kg m3) Experimental density (kg m 3)
PPC – 1180
AR-PC 1111.3 1120
STB-WIPC 1045.7 1260 STB-AWPC 973.9 1150
showed the highest experimental density values. On the other hand, the theoretical density values were highest for the AR-PC (al-most equal to that of its experimental value), followed by STB-WIPC and STB-AWPC. However, all these theoretical values were lower than that of PPC.
Experimental density value of STB-WIPC higher than those of AR-PC (both experimental and theoretical) and even that of PPC could be attributed to the presence of lignin (27.7%), which has a complex poly-phenolic structure. Upon heating, lignin can diffuse into the pores of the fibers created by the steam explosion and also cement the fibers together, leading to a greater degree of molecular interactions, which may translate into higher density values. As mentioned earlier, the steam explosion of bagasse fibers results in partial degradation of cellulose and structure modification of both hemicelluloses and lignin. Accordingly, bigger pores would be resulted due to the removal of both hemicelluloses and lignin by organic solvents or aqueous media, which will result in decreas-ing the available surface area of SE fibers[38]. However, absence of small air bubbles in these composite laminates [STB-WIPC and STB-AWPC] compared to those of ARPC might also be another fac-tor. Further, polyester resin itself might have seeped into the fibers, which could have increased the density of the treated fibers and hence that of their composites. Both the mechanisms may contrib-ute to the observed higher density of the composite. On the other hand, lower theoretical density values for WIPC and STB-AWPC are understandable since both lignin and hemicelluloses contents decreased with the high-pressure steaming (SeeTable 3). Low density values (experimental) of AR-PC are due to the reported low density values of the natural fiber itself, which is also the case for other natural fiber containing polymer composites[38]. In addi-tion, unlike treated fibers, absence of pores for the polyester to seep in and presence of small air bubbles in these laminates even after the application of load during sample preparation could ex-plain the lower density value of ARPC laminates.
3.4. Thermal analysis
The thermal analyses of the polyester matrix and all fiber-con-taining composites are shown in Figs. 1a–1c. The thermo-gravi-metric analysis (DTA and TGA curves on Figs. 1a and 1b, respectively) demonstrated that, at temperatures below 150 °C, the composites do not show any significant decomposition, even in the presence of oxygen. However, above 150 °C, PPC presents two decomposition stages. The first, at 390 °C, consumes about 90% of the mass, whereas the second, up to 550 °C, degrades the remainder of the sample. However, the incorporation of both AR
and STB-WI fibers into the polyester matrix provides slightly lower thermal stability (AR-PC and STB-WIPC inFigs. 1a and 1b), respec-tively, while the reverse is observed for STB-AWPC. This is due to the fact that lignin favors a better interaction between bagasse fi-bers and the polyester matrix, whose polarity is not compatible with that of cellulose.
Likewise, the STB-WIPC sample had better thermal stability than the AR-PC. Steam explosion increases the relative amount of lignin in the bagasse fibers, and this lignin is in a state of smaller molecular aggregation with the cellulose because most of the hemicellulose interface was removed by acid hydrolysis. This al-lows the interpenetration of lignin in the polyester matrix and bet-ter fiber reinforcement, as well as a betbet-ter homogenization of the sample, as evidenced by the visual evaluation of the fiber compos-ites. These observations are fundamentally related to the events occurring at temperatures below 400 °C because, above this tem-perature, the behavior of the matrix was not the same. The reason for such behavior is not known at present and needs further investigation.
The DSC curves (Fig. 1c) shows that the glass transition temper-ature of PPC, which was identified as 64 °C at the maximum peak height, shifted to 67 °C after the incorporation of the STB-WIPC
100 200 300 400 500 600 700 800 900 STB-AWPC
PPC STB-WIPC
AR-PC
Weight first derivative (a.u.)
Temperature (ºC)
Fig. 1a. TGA curves of polymer matrix and the composite with as-received and surface-treated bagasse fibers.
100 200 300 400 500 600 700 800 900 0 20 40 60 80 100 STB-AWPC AR-PC PPC STB-WIPC Temperature (°C) Weight (%)
Fig. 1b. TGA curves of polymer matrix and the composite with as-received and surface treated bagasse fibers.
40 50 60 70 80 90 100 0.26 0.28 0.30 0.32 0.34 0.36 0.38 0.40 Endo 67°C 64°C STB-AWPC AR-PC STB-WIPC PPC
Energy flow (a.u.)
Temperature (°C)
Fig. 1c. DSC curves of polymer matrix and the composite with as-received and surface treated bagasse fibers.
fibers, and the same trend was observed in the DMA studies (see discussion below).
In general, the dynamic-mechanical analysis (DMA) illustrates the deformation that takes place in a solid or viscous environment under a variation of frequency or temperature due to the applica-tion of a small oscillatory mechanical tension. The DMA curves of the PPC and the fiber composites are shown inFig. 2. All the sam-ples showed that there is an abrupt fall in the modulus [shown on the left side in the figure], in temperatures above 0 °C. This charac-terizes a phase change (glass transition), which indicates that the polymeric matrix displays a visco-elastic behavior[39]. The maxi-mum value of tan d may be due to the high conversion of mechan-ical energy into heat through the Brownian motion of the molecular chains. These motions are related to the cooperative dif-fusion movements of the segments of the main chain. The glass transition temperatures for the untreated and treated fibers con-taining composites are thus found to be 60 °C and 72 °C, respec-tively. In other words, the Tgvalue shifted upward by about 10 °C
for the steam-treated fibers. This effect is larger for the STB-WIPC and STB-AWPC. Further, the analyses by DMA showed that the incorporation of fibers into the polyester matrix improves the stor-age modulus, indicating the formation of a more rigid structure. Besides this, the Tgdisplacement [shown on the right side in the
figure] can be observed at higher temperatures for the composites with pretreated fibers compared to the pure matrix or to the as-re-ceived bagasse fiber-containing composites. This behavior can be attributed to the better interfacial interaction of the matrix with the treated fibers, which contributed to a decrease in the mobility of the polymeric chains.
It should be noted that the surface modification of lignocellu-losic fibers promote chemical or physical changes at the interface [40]and the interfacial properties can be improved by appropriate modifications of the components, which give rise to changes in physical and chemical interactions at the interface [41]. Several classes of compounds are known to promote adhesion apparently by chemically coupling the adhesive to the adherents. The promo-tion of physical and chemical changes at the interface enhances the final bulk properties of the composites. In the present case, steam
explosion (SE) is used to modify the surface of sugarcane bagasse fibers. This leads to cellulose with increased degree of crystallinity and crystallites of higher perfection and size as part of the reorga-nization of amorphous and/or paracrystalline cellulose regions, releasing strains that arise due to the interactions with hemicellu-loses and lignin[42,43]. Also, there will be higher availability of hydroxyl groups particularly on the surface of cellulose crystallites or in the amorphous regions. It is also mentioned that some of the effects of SE of fibers include providing greater activity towards chemical and biochemical reagents, better surface roughness, and higher crystallinity of cellulose.
In agreement with the published literature[44], the magnitude of the tan d peak can be related to the toughness of the material. Therefore, the composites containing pretreated bagasse fibers are expected to show superior toughness compared to that of AR-PC sample suggesting these composites could be good candi-dates for structural applications, such as automotive uses, requir-ing good damprequir-ing.
3.5. Water absorption
Another important property of the composites reinforced by lig-nocellulosic fibers is water absorption as this affects their thermal stability and limits their use in many industrial applications.Fig. 3 shows the water absorption curves for all composites prepared in this study. These are comparable with those observed earlier for PP composites containing as-received, steam exploded and acety-lated wood fibers[30].
All samples showed an increase in their weight with increasing time of immersion. A water uptake was observed up to 14 days of immersion, followed by a gradual increase for about 35 days. Fur-thermore, STB-AWPC showed the highest level of water uptake, while the STB-WIPC and AR-PC did not show any significant mu-tual difference. High water absorption by the composites is under-standable due to the hydrophilic nature of AR as well as STB-WI and STB-AW fibers [30]. Besides, there exists the possibility of water absorption by capillarity at the fiber edges of the cut samples of composite laminates as the polyester matrix is hydrophobic. In
AR-PC
STB-WIPC
PPC
STB-AWPC
PPC
STB-WIPC
Fig. 2. DMA curves of polyester matrix and its composites (a) polyester matrix (PPC) and (b) polyester-as-received bagasse (AR-PC); (c) polyester-surface treated bagasse (STB-WI and WIPC).
addition, this could be indicative of some processing problems, such as the incomplete curing of the thermoset matrix, which may contain fissures that could have appeared during the prepara-tion of the laminates. Reduced water absorpprepara-tion in both AR-PC and STB-WIPC samples is due to their lignin content, which minimizes the absorption of water. Similar results are reported for PP com-posites containing untreated as well as steam-exploded and alka-li-treated wood fibers[30]. This is probably due to the observed higher density of these composites and to the presence of poly-phenolic structure of lignin, which is hydrophobic by nature. The presence of lignin leads to a highly packed structure, leaving no space for swelling. Besides, its absence results in the opposite ef-fect. On the other hand, STB-AWPC, containing partially delignified STB-WI fibers, displayed the highest water absorption values. This is because these fibers are rich in a chemically modified cellulose component whose water accessibility has been enhanced by steam explosion and alkali extraction.
4. Conclusions
Steam explosion is quite efficient in the separation of the main components of sugarcane bagasse, lowering the amounts of hemi-celluloses and increasing the relative amount of lignin. The incor-poration of steam-exploded fibers in unsaturated polyester had a positive reinforcing effect in terms of DMA analysis and higher thermal stability (TGA/DTG/DSC studies) compared to composites containing as-received bagasse fibers. This is due to the superior compatibility of these fibers with the polyester matrix compared to that of as-received and might be related to their increased crys-tallinity and high content of a modified lignin that is loosely con-nected to the matrix.
In general, the present study demonstrates that acid-catalyzed steam-exploded fibers derived from sugarcane bagasse are useful materials for the partial substitution of polymeric matrices such as polyesters. This can be a new and interesting alternative for the use of these renewable resources instead of discarding them as wastes, therefore providing an environmental effect. However, further work is required to assess the beneficial effects of such ptreatment for these fibers in comparison to other processes re-ported in the literature.
Acknowledgements
The authors thank Drs. Gabriel Pinto de Souza and Helena Maria Wilhelm for their help at several stages of this work. The group also thanks to the Research Council for Scientific and Technological
Development of Brazil (CNPq) and to the Institute of Technology for the Development (LACTEC) for providing financial support and easy access to several facilities mentioned in this work, respec-tively. Dr. KGS also acknowledges the interest and encouragement of the three organizations with whom he is presently associated with in India.
References
[1] Ho K. The potential of bagasse-based cogeneration in the US, MECE 4211.001, Columbia University; 12 January 2006. <http://www.columbia.edu/kgh2103/ US-Bagasse> [accessed 06.06.08].
[2] Filho PA, Badr O. Biomass resources for energy in north-eastern brazil. Appl Energy 2004;77(1):51–67.
[3] UNICA – Produção de cana-de-açúcar no Brasil. <http://www.unica.com.br/ dadosCotacao/estatistica/> [accessed 04.1110].
[4] Zanin G, Santana CC, Bon EPS, Moraes FF, Macedo I, Coelho Neto C, et al. Brazilian bioethanol program. Appl Biochem Biotechnol 2000;84–86:1147–61. [5] Sartori MMP, Florentino HO, Basta C, Leão AL. Determination of optimum quantity of crop residues for energy in sugarcane crop using linear programming in variety selection and planning strategy. Energy 2001;26:1031–40. [6] Paiva JMF, Frollini E. Thermoset matrix-sugarcane bagasse reinforced composite
(Matriz termofixa fenólica em compósitos reforçados com fibras de bagaço de cana-de-açúcar). Polímeros- Ciência e Tecnologia 1999;9(2):78–87.
[7] Pandey A, Soccol CR, Nigam P, Soccol VT. Biotechnological potential of agro-industrial residues I: sugarcane bagasse. Biores Technol 2000;74:69–80. [8] Sun JX, Sun XF, Zhao H, Sun RC. Isolation and characterization of cellulose from
sugarcane bagasse. Polym Degrad Stab 2004;84:331–9.
[9] Rasul MG, Rudolph V, Carsky M. Physical properties of bagasse. Fuel 1999;78:905–10.
[10] Chiparus OI, Chen Y. An image method to evaluate bagasse fiber dimensions. Biores Technol 2003;90:305–9.
[11] Reis JML. Fracture and flexural characterization of natural fiber-reinforced polymer composite. Construct Build Mater 2006;20:673–8.
[12] Fukumoto I, Mekaru S, Cao Y. Development of fiber reinforced plastic using bagasse by injection moulding. In: The 10th JSME/ASME material and processing conference, Hawaii, USA; 2002. p. 507–10.
[13] Mattoso LHC, Leão AL, Frollini E, editors. In: Proceedings of the second international symposium on natural polymers and composites (ISNaPol 1998), Atibaia, Brasil; 10–13 May 1998. ISBN: 85.86463-04-3.
[14] Mattoso LHC, Leão AL, Frollini E, editors. Natural polymers and composites. In: Proceedings of the third international symposium on natural polymers and composites (ISNaPol-2000), Brasil; 14–17 May 2000. ISBN: 85-86463-07-8. [15] Frollini E, Leão AL, Mattoso LHC, editors. Natural polymers and agro fibers
based composites, USP/UNESP and Embrapa, São Carlos, Brasil; 2000. ISBN: 85-86463-06-X.
[16] Hassan ML, Rowell RM, Fadl NA, Yacoub SF, Christiansen AW. Thermo-plasticization of bagasse I – preparation and characterization of esterified bagasse fibers. J Appl Polym Sci 2000;76:561–74;
Hassan ML, Rowell RM, Fadl NA, Yacoub SF, Christiansen AW. Thermo-plasticization of bagasse. II-Dimensional stability and mechanical properties of esterified bagasse composites. J Appl Polym Sci 2000;76:575–86.
[17] Stael GC, Tavares MIB, d’Almeida JRM. Impact behavior of sugar cane bagasse waste – EVA composites. Polym Testing 2001;20:869–72.
[18] Proceedings of international symposium on natural polymers and composites – ISNaPol, São Pedro, Brasil; 2002. ISBN: 85-86463-108.
[19] Trindade WG, Hoareau W, Razera IAT, Ruggiero R, Frollini E, Castelan A. Phenolic thermoset matrix reinforced with sugar cane bagasse fibers: attempt to develop a new fiber surface chemical modification involving formation of quinines followed by reaction with furfuryl alcohol. Macromol Mater Eng 2004;289:728–36.
[20] Mei LHI, Mattoso LHC, Curvelo, AAS, editors. Proceedings of fifth international symposium on natural polymers and composites (ISNaPol-2004), São Pedro, SP, Brasil; 12–15 September 2004. ISBN: 85-86463-13-2.
[21] Souza MV, Monteiro SN, d’Almeida JRM. Evaluation of pretreatment, size, molding pressure on flexural mechanical behavior of chopped bagasse– polyester composites. Polym Testing 2004;23:253–8.
[22] Shibata S, Cao Y, Fukumoto I. Press forming of short natural fiber-reinforced biodegradable resin: effects of fiber volume and length on flexural properties. Polym Testing 2005;24:1005–11.
[23] Shibata S, Cao Y, Fujimoto I. Study of flexural modulus of natural fiber/ polypropylene composites by injection molding. J Appl Polym Sci 2005;100:911–7.
[24] Shibata S, Cao Y, Fujimoto I. Effect of bagasse fiber on the flexural properties of biodegradable composites. Polym Compos 2005;100:689–94.
[25] Cao Y, Shibata S, Fukumoto I. Mechanical properties of biodegradable composites reinforced with bagasse fiber before and after alkali treatments. Composites A 2006;37:423–9.
[26] Hodzic A, Coakley R, Curro R, Brendt CC, Shanks RA. Design and optimization of biopolyester bagasse fiber composites. J Biobased Mater Bioenergy 2007;1:46–55;
Hodzic A, Coakley R, Curro R, Brendt CC, Shanks RA. Bagassse fiber for sustainable manufacturing. J Biobased Mater Bioenergy 2007;1(3):289–300. Fig. 3. Moisture absorption of polyester–bagasse composites containing (a)
as-received bagasse (AR-PC); (b) steam exploded and washed bagasse (STB-WIPC); (c) steam exploded + alkali washed bagasse (STB-WIPC).
[27] Luz SM, Gonçalves AR, Del’Arco Jr AP. Mechanical behavior and microstructural analysis of sugarcane bagasse fibers reinforced polypropylene composites. Compos Part A 2007;38:1455–61.
[28] Satyanarayana KG, Guimarães JL, Wypych F. Studies on lignocellulosic fibers of Brazil: part I – source, production, morphology, properties and applications. Compos Part A 2007;38(8):1694–709.
[29] Pinto CES, Arizaga GGC, Satyanarayana KG, Wypych F, Ramos LP. Studies on the effect of pressure and incorporation of sugarcane bagasse fibers on the structure and properties of poly (hydroxybutyrate). Compos Part A 2009;40:573–82.
[30] Fakirov S, Bhattacharyya D, editors. Handbook of engineering biopolymers. Munich: Hanser Publishers; 2007. p. 49–77. ISBN: 978-1-56990-405-3 [chapter 2].
[31] Ramos LP, Silva TA, Martins LF, Satyanarayana KG. Conversion of lignocellulosics to fuels, chemicals and environmentally-friendly materials. Met Mater Proc 2005;17(3–4):299–318.
[32] Satyanarayana KG. Steam explosion – a boon for value addition to renewable resources. Metal News 2004;22:35–40.
[33] Tomczak F, Sydenstricker THD, Satyanarayana KG. Studies on lignocellulosic fibers of Brazil: part II – structure and properties of Brazilian coconut fibers. Compos Part A 2007;38(8):1710–21.
[34] Tomczak F, Satyanarayana KG, Sydenstricker THD. Studies on lignocellulosic fibers of Brazil: part III – structure and properties of Brazilian Curauá fibers. Compos Part A 2007;38(10):2227–36.
[35] Guimarães JL, Frollini E, Da Silva CG, Wypych F, Satyanarayana KG. Characterization of banana, sugarcane bagasse and sponge gourd fibers of Brazil. J Ind Crops Prod 2009;30(3):407–15.
[36] Guimarães JL, Satyanarayana KG, Cursino ACT, Wypych F, Saul SK, Ramos LP. Studies on the processing and characterization of corn starch and its composites with banana and sugarcane fibers of Brazil. Carbohydr Polym 2010;80(1):130–8.
[37] Gomide JL, Demuner BJ. Determinação do teor de lignina em material lenhoso: método Klason modificado. O Papel 1986;47:36–8 [in Portuguese]. [38] Excoffier G, Tousasaint B, Vignon MR. Saccharification of steam-exploded
popular wood. Biotechnol Bioenergy 1991;38:1308–17.
[39] Kevin PM. Dynamic mechanical analysis – a practical introduction. CRC Press LLC; 1999.
[40] Mat R, Taib ZA, Mohd. Ishak HD, Rozman WG. Glasser, steam exploded wood fibers as composite reinforcement In: Fakirov S, Bhattacharyya D, editors. Handbook of engineering biopolymers. Munich: Hanser Publishers; 2007. p. 611–653. ISBN: 978-1-56990-405-3 [chapter 20].
[41] Medeiros ES, Mattoso LHC, Sreekumar PA, Joseph K. Role of interface in lignocellulosic fiber reinforced polymer composites. In: Thomas S, Pothan LA, editors. Natural fiber reinforced polymer composites. USA: Publisher Old City Publishing; 2009. p. 270–314. ISBN: 978-1-933153-09-4 [chapter 11]. [42] Focher B, Marzetti A, Beltrame PC, Avella M. SE biomass for the preparation of
conventional and advanced biopolymer-based materials. Biomass Bioenergy 1998;14:187–94.
[43] Avella M, Cassale L, Dell’erba R, Focher B, Martuscelli E, Marzetti E. Broom fibers as reinforcing materials for PP-based composites. J Appl Polym Compos 1998;68:1077–89.
[44] Satyanarayana KG, Arizaga GGC, Wypych F. Biodegradable composites based on lignocellulosic fibers – an overview. Prog Polym Sci 2009;34:982–1021.
