- Email: [email protected]
Selective component degradation of oil palm empty fruit bunches (OPEFB) using high-pressure steam
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
Available online at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
Selective component degradation of oil palm empty fruit bunches (OPEFB) using high-pressure steam Azhari Samsu Baharuddin a,*, Alawi Sulaiman c, Dong Hee Kim d, Mohd Noriznan Mokhtar a, Mohd Ali Hassan b, Minato Wakisaka d, Yoshihito Shirai d, Haruo Nishida d a
Department of Process and Food Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia b Department of Bioprocess Technology, Faculty of Biotechnology and Biomolecular Sciences, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia c Faculty of Plantation and Agrotechnology, Universiti Teknologi MARA, 40410 Shah Alam, Selangor, Malaysia d Department of Biological Functions and Engineering, Graduate School of Life Science and System Engineering, Kyushu Institute of Technology, 2-4 Hibikino, Wakamatsu-ku, Kitakyushu, Fukuoka 808-0916, Japan
article info
abstract
Article history:
In order to accelerate the bioconversion process of press-shredded empty fruit bunches
Received 14 June 2011
(EFB), the effect of high-pressure steam pre-treatment (HPST) in degrading the lignocellu-
Accepted 11 February 2013
losic structure was investigated. HPST was carried out under various sets of temperature/
Available online 13 March 2013
pressure conditions such as 170/0.82, 190/1.32, 210/2.03, and 230 C/3.00 MPa. It was noted that after HPST, the surface texture, color, and mechanical properties of the treated EFB
Keywords:
had obviously altered. Scanning electron micrographs of the treated EFB exhibited effective
Empty fruit bunches
surface erosion that had occurred along the structure. Moreover, the Fourier transform
High-pressure steam
infrared and thermogravimetric analyses showed the removal of silica bodies and hemi-
Selective degradation
cellulose ingredients. X-ray diffraction profiles of the treated EFB indicated significant in-
Lignocellulosic structure
creases in crystallinity. These results reveal that HPST is an effective pre-treatment
Pre-treatment
method for altering the physicochemical properties of the EFB and enhancing its biodegradability characteristics for the bioconversion process. ª 2013 Elsevier Ltd. All rights reserved.
1.
Introduction
The empty fruit bunches (EFB) which generated at palm oil mills are extremely abundant, renewable, and readily available lignocellulosic materials. In recent years, palm oil mill industries have been keen to examine the potential of the cocomposting process using the EFB and palm oil mill effluents (POME) as an effective approach to utilizing the abundant biomass generated at palm oil mills. In utilizing the shredded or
press-shredded EFB, the co-composting treatment critically depends on the bio-conversion rate of EFB. This achieves maturity within 40e80 days using a conventional windrow in the field scale [1e3]. The longer treatment in the composting process corresponded to the phenolic compound in the cell walls of the EFB consisting of lignocellulosic materials, which could restrict the rate and extent of polysaccharide degradation [4]. Therefore, an appropriate pre-treatment method of EFB was essential to accelerate the degradation in compositing.
* Corresponding author. Tel.: þ60 3 89464424; fax: þ60 3 89464440. E-mail address: [email protected] (A.S. Baharuddin). 0961-9534/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biombioe.2013.02.013
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
Many physical, chemical, and microbial pre-treatment methods have been reported for enhancing EFB’s bioconversion by improving the hydrolysis rate of lingocellulose [5,6] through, for example, removal of lignin and hemicellulose ingredients, reduction in crystallinity, and an increase in porosity. Umikalsom et al. [7], proposed a method combining the physical, thermo-chemical, and biological procedures to pre-treat EFB for bio-sugar production. Additionally, chemicals like citric acid, hydrochloric acid, EDTA, and sodium hydroxide have been used for the pre-treatment of EFB. However, such chemical pre-treatments are often expensive and inappropriate for environmentally-benign bioconversion. This is especially true in the large scale commercial operations, involving chemical pre-treatments, which require additional precautionary steps such as, use of corrosion-resistant apparatus, safe disposal of used chemicals, etc. [8]. Steam explosion is one of the alternative approaches that can provide an efficient pre-treatment method for lignocellulosic materials [9], and it has been executed at the high-pressure and temperature ranges of 1.05e3.61 MPa and 180e240 C, respectively. Steam explosion breaks up the lignocellulose structure and decomposes the hemicellulose into uronic and acetic acids, which then function as catalysts in the auto-catalytic degradation of hemicellulose and lignin ingredients [8]. However, the pre-treatments of lignocellulosic materials through steam explosion require a high energy input and might increase the cost of operation at a commercial scale. Based on our observation of a palm oil mill operation, an appropriate method using high-pressure steam (HPST) was newly suggested for the pre-treatment of EFB. The HPST is expected to alter the lignocellulosic structure of EFB so that it becomes more susceptible to the cellulose-degrading enzymes. However, the optimal conditions in terms of pressure and holding time are required to apply the HPST method in practice. Interestingly, the requirement of such high-pressure inputs can be provided by the steam line infrastructure and excess steam generated at the palm oil mills. Moreover, in order to make HPST an effective pre-treatment method, knowledge of quantitative and comprehensive analysis of the physicochemical changes of EFB during HPST becomes important especially since it may suggest that EFB is a suitable substrate in many applications such as composting. In this article, in order to accelerate the bioconversion process of the press-shredded EFB, the effects of HPST on the lignocellulosic structures were investigated in detail. The physicochemical changes of EFB were determined using scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), thermogravimetric/differential thermal analysis (TG/DTA), X-ray diffraction analysis (XRD), and inductively coupled plasma-optical emission spectrometry (ICP-OES) under various sets of temperature/pressure conditions that steam is put through.
269
Malaysia). The EFB samples were stored under conditions of 4 C in plastic bags prior to use for HPST.
2.2.
High-pressure steam pre-treatment (HPST)
HPST of press-shredded EFB was conducted in a 500 mL highpressure autoclave (START 500, Nitto Kouatsu, Co. Ltd) equipped with a temperature and pressure control system that was capable of operating up to 250 C and 9.4 MPa, respectively. About 10 g of EFB was oven-dried at 60 C for 24 h, weighed, and put into the metal wire cage in the autoclave to be treated under various HPST conditions of 170/0.82, 190/1.32, 210/2.03, and 230 C/3.00 MPa for 2, 4, 8, and 10 min. At the end of each treatment, one exhaust valve of the autoclave was opened to release the steam, and then, the treated EFB sample was oven-dried at 105 C for 24 h and weighed. Hydrolysates condensed in the autoclave after HPST were also collected as liquid portions. Changes in surface texture, color, and odor of the treated EFB samples were examined, and pH values of hydrolysates were measured on a pH meter (F-51S, Horiba Ltd., Japan).
2.3.
SEM analysis
Surface textures of raw and pre-treated EFB samples were observed on a scanning electron microscopy (SEM) (S-3400N, Hitachi, Japan) at an accelerating voltage of 10 kV. The samples were air-dried and coated with goldepalladium in a sputter coater (E-1010, Hitachi, Japan). SEM images of all the samples were taken at the same magnification of 1500 and 3000.
2.4.
Fourier transform infrared (FTIR) analysis
The Fourier transform infrared (FTIR) spectra were recorded on a Perkin Elmer GX2000R infrared spectrophotometer in a range of 500e4000 cm1 at a resolution of 4 cm1. Reflection spectra of EFB samples were measured on a Golden Gate Diamond attenuated total reflectance (ATR) (10500) module with a germanium crystal, by the single-reflection ATR method.
2.5.
X-ray diffraction (XRD) analysis
Measurements of wide-angle X-ray diffraction of EFB samples were made on a Rigaku XRD-DSC-X II diffractometer system. Cu-Ka radiation (X-ray wave length: l ¼ 0.15418 nm) was used as the source, and X-ray diffraction patterns were recorded in a range of 2q ¼ 10e40 at a scan rate of 2 min1. The degree of crystallinity was calculated from the diffracted intensity data according to Vonk’s method [10]. The crystalline size was calculated according to the Scherrer equation (1): Kl bcosq
2.
Experimental
L¼
2.1.
Materials
where L is the crystalline size in nm, b is the half-value breadth of a diffraction peak on the 2q scale in radians, and K is a constant: 0.9 [11]. Crystalline size was calculated along the characteristic (002) and (040) plane directions.
Press-shredded EFB fibers (length: 15e20 cm) were obtained from Seri Ulu Langat Palm Oil Mill (Dengkil, Selangor,
(1)
270
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
2.6. Thermogravimetric and differential thermal analyses (TG/DTA)
Table 2 e Changes in pH values of hydrolysates during steam treatment of EFB samples.
Thermogravimetric and differential thermal analyses (TG/DTA) were conducted on a simultaneous thermal analyzer EXSTAR 6200 TG/DTA system (Seiko Instruments Inc., Japan) under a constant nitrogen flow (100 mL min1). About 5 mg of oven-dried EFB sample was scanned from 25 to 500 C at a heating rate of 9 C min1. A blank aluminum pan was used as a reference.
2.7.
Chemical composition analysis
Chemical compositions of raw and treated EFB samples were analyzed on a CNHS analyzer 2000 (Leeco, USA) to determine contents of carbon, nitrogen, and nutrients. Contents of heavy metal elements were measured on an inductively coupled plasma-optical emission spectrometer (ICP-OES) (Perkin Elmer, USA). Compositions of cellulose, hemicellulose, and lignin ingredients were determined according to the Goring’s method [12].
3.
Results and discussion
3.1. Macroscopic observation of treated EFB samples and their hydrolysates After the HPST was performed, the weight and color of EFB materials were noted as having changed. This is displayed in Table 1 and Fig. S1. The treated EFB had acquired a dark brown color because of an increase in steam temperature. Similar color changes were observed in liquid portions including hydrolysates from EFB (Fig. S2 in Supplementary data). The pH values of the liquid portions were detected in an acidic range and gradually decreased from 6.0 (170 C/0.82 MPa) to 3.7 (230 C/3.00 MPa) with increases in the treatment temperature and time (Table 2). These acidic values were attributed to the acetic acid relatives and phenolic hydrolysates from the hemicellulose and lignin ingredients, respectively, which corresponded to the considerable weight loss noted in the EFB materials. Moreover, the treated EFB and liquid portions released a caramel smell after the steam treatment. Sun et al. [13] reported that brownish products from wheat straws after steam treatment were the result of degradation of carbohydrates at high steam temperatures. Interestingly, the steamtreated EFB samples, especially, at 230 C/3.00 MPa were
Table 1 e Changes in weight, crystallinity, and crystalline size of EFB samples during steam treatment. Sample
Weight loss, WL (%)
Raw EFB e Steam-treated EFB for 10 min 15.2 Ttr ¼ 170 C 23.3 190 C 39.7 210 C 46.8 230 C
Crystallinity, Xc (%)
Crystalline size (nm) L002
L040
39.3
2.68
6.03
41.7 43.1 52.5 57.4
2.91 3.38 3.75 4.14
3.44 4.25 4.12 3.96
Time (min) 2 4 8 10
pH value
Ttr ¼ 170 C
190 C
210 C
230 C
6.0 5.4 5.3 5.3
4.9 4.7 4.6 4.6
4.5 4.2 4.1 4.1
3.9 3.9 3.7 3.7
found to be so brittle that they were easily broken up by a hammer mill as compared to raw EFB. The change in mechanical properties of the EFB after the steam-treatment was expected to be advantageous for the following bioconversion process.
3.2.
Microscopic changes in morphology of treated EFB
SEM observations were conducted in order to confirm the effects of HPST on EFB surface structures. Fig. 1 shows the SEM micrographs of raw and steam-treated EFB sample surfaces at various steam temperatures after 10 min. There are obvious differences between the raw and steam-treated EFB sample surfaces, indicating that HPST had some effects on the EFB modification as follows: First, the SEM micrographs for raw EFB sample surfaces (Fig. 1a) exhibited silica bodies (phytoliths) embedded in the EFB structure similar to a micrograph shown in the previous work [3]. After HPST, the first important change was that the silica bodies were removed. The remaining holes had homogenous dimensions of around 10 mm in diameter at the outer surface (Fig. 1bec). These holes were assumed to be effective in the swelling of the EFB structures, attracting the microbes and enzymatic reactions for the subsequent bioconversion. It was observed that most of the debris and silica bodies on EFB surfaces were removed when the steam temperature was increased to 190 C (Fig. 1c). This is an important change, because the silica bodies could retard the approach of microbes to the internal cellulose layers in the lignocellulosic structure [14]. Second, most of the outer layers of the raw EFB surface were found disrupted along the structure. When the steam temperature was increased to 210 C (Fig. 1d), the outer layers of EFB decomposed and cracked along the inner structures, and after the treatment at 230 C/ 3.00 MPa for 10 min, most of the outer layers were almost completely disrupted (Fig. 1e). The formation of the holes and cracks along the EFB inner structures indicates that HPST effectively contributed to the modification of lignocellulosic structure in raw EFB. It is also suggested that the structural modification performed at 230 C/3.00 MPa corresponded to a disruption level of the EFB structure achieved by a 20 day composting process [1,2]. According to Wang’s results [9], the steam treatment could alter cellulose fibers from Lespedeza stalks and increase their reactive areas, resulting in an enhancement of the accessibility of enzymes such as cellulase. Based on the above findings, it is seen that if the raw EFB is first pre-treated with high-pressure steam, the following composting processes of EFB and POME will be expedited and will take less than 40 days of treatment given the enhancement of microbial action.
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
271
Fig. 1 e SEM micrographs at 31500 magnification of (a) raw and steam-treated EFB samples at conditions of (b) 170/0.82, (c) 190/1.32, (d) 210/2.03, and (e) 230 C/3.00 MPa for 10 min. Bar: 30 mm.
3.3.
Chemical structure of treated EFB
Fig. 2 shows the FT-IR spectra of the raw and treated EFB samples under various steam temperatures after 10 min. With
increase in the steam temperature, some absorption bands at 2916, 2850, 1730, 1635, 1232, and 892 cm1 diminished or disappeared, and other bands at 1602, 1590, 1507, 1157, 1100, and 1053 cm1 notably increased.
Fig. 2 e FTIR spectra of (a) raw EFB and steam-treated EFB samples at conditions of (b) 170/0.82, (c) 190/1.32, (d) 210/2.03, and (e) 230 C/3.00 MPa for 10 min.
272
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
Absorptions at 2916 and 2850 cm1, which are attributable to the CeH stretching vibrations, gradually diminished with an increase in the steam temperature. The CeH bands corresponded to the aliphatic moieties mainly in polysaccharides of cellulose and hemicellulose [9]. The bands at 1730 and 1232 cm1 were attributable to nC]O of acetyl and uronic ester groups, and nCeO of the carboxyl group in the hemicellulose component, respectively [15,16]. The band at 1730 cm1 was also attributable to the ester linkages of carboxylic groups of the ferulic acid and p-coumaric acids of lignin [17]. These bands had almost disappeared when the steam temperature was increased up to 230 C. The bands at 1635 and 892 cm1 were reported as being characteristic bands of OeH bending of adsorbed water [16] and a b-glycosidic linkage between the sugar units [18], respectively. The diminution and disappearance of these absorptions indicated the preferential decomposition of the hemicellulose component. On the other hand, the increased absorptions at 1602, 1590, and 1507 cm1 were characteristic bands of aromatic skeletal vibrations [9,14,18], with the absorptions at 1157, 1100, and 1053 cm1 being attributable to the dCeH (aromatic in-plane deformation) of the lignin component, in which the absorptions at 1157 and 1053 cm1 included the other characteristic bands of CeOeC asymmetric vibration and CeO stretch, respectively, in cellulose [19]. These increases in absorption of the characteristic bands meant that after the steam treatment, the compositions of lignin and cellulose ingredients increased following the decomposition of the hemicellulose ingredient. Hemicellulose consists of various saccharides having many branches, which form amorphous regions. It has been reported that hemicellulose can be easily degraded and removed from the main stem as volatile products such as CO, CO2, and some hydrocarbon at 220e315 C [20]. Meanwhile, cellulose is more stable and degrades at 315e400 C, with lignin being the most stable despite degrading in a wide temperature range of 160e900 C [20] due to its aromatic ring network having various branch structures. Characteristic bands used to identify the cellulose component are at 1420 and 1430 cm1 attributable to amorphous cellulose II and crystallized cellulose I [9]. In the present study, the absorption of the band around 1420 cm1 increased in intensity slightly with increase in the steam temperature, but diminished at 230 C. Cellulose is relatively stable up to 210 C. However, by steam treatment at 230 C, the cellulose structure was broken down in a similar manner to previous results reported by Sun et al. [13]. Difference spectra of the steam-treated samples against the raw EFB spectrum are illustrated in Fig. S3 (Supplementary data). Besides, the changes in absorption bands which were clearly visible, other changes over some broad absorption regions were also detected. Typical changes were diminutions in absorption of bands at 3200e3000, 1100e1050, and 950e900 cm1 attributable to nOeH of silica ingredients, nSieOeSi, and dSiOeH of silica ingredients; and increases in the absorption of a band at 3500e3200 cm1 attributable to nOeH of hydrogen-bonded hydroxyl groups. These changes suggest a decrease in silica ingredient during the steam treatment [21].
Fig. 3 e TG and DTG profiles of raw EFB and steam-treated EFB samples at conditions of 170/0.82, 190/1.32, 210/2.03, and 230 C/3.00 MPa for 10 min.
3.4.
Thermal analysis of steam-treated EFB
Thermal responses of the raw and steam-treated EFB samples were investigated with a thermal gravimeter (TG). TG and differential TG (DTG) profiles are illustrated in Figs. 3 and 4. Fig. 3 shows the TG/DTG profiles measured at various steamtreatment temperatures (Ttr) for 10 min. In Fig. 3a, the onset point of weight loss shifted into high temperatures with increase in Ttr, suggesting removal of some heat-instable factors by the steam treatment. All the DTG profiles (Fig. 3b) of treated samples had their main peaks shifted into a temperature range of 320e390 C, which was higher than the range of 200e350 C for raw EFB. This main peak is attributed to the degradation of cellulose ingredient. Although the main peak shifted into a temperature range of 325e400 C up to Ttr ¼ 210 C, at Ttr ¼ 230 C the main peak returned to a lower temperature range of 320e380 C. In Fig. 4, changes in TG/DTG profiles at Ttr ¼ 170 and 210 C with time are shown, respectively. The TG/DTG profiles of treated samples at Ttr ¼ 170 C exhibited the same two-step degradation behavior together with a constant residual weight of around 18 wt% without considerable change during the treatment for 2e10 min (Fig. 4aeb). These profiles indicate that a discontinuous event occurred within 2 min and caused at least three changes: 1) separation of the degrading temperature ranges of hemicellulose and cellulose components, 2) stabilization of cellulose component, and 3) decrease in the residual weight from 28.7 wt% to 18.2 1.1 wt % (Table S1). The decrease in residual weight was attributed to the removal of silica bodies from the outer surface of the EFB materials.
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
273
Fig. 4 e TG and DTG profiles of raw EFB and steam-treated EFB samples at conditions of a, b) 170 C/0.82 MPa and c, d) 210 C/ 2.03 MPa for 2, 4, 8, and 10 min.
At Ttr ¼ 210 C, all the steam-treated samples showed smooth weight loss curves (Fig. 4c) and single model DTG peaked at 370 C (Fig. 4d), with these peaks being attributable to cellulose degradation. The weight loss of the DTG peaks corresponded with the hemicellulose degradation. The shift of the peaks could be seen even after the treatment time had been reduced to 2 min. Moreover, the degradation peaks for the remaining cellulose ingredients in the EFB structure was noticed at a high temperature range of 365e370 C. Similar temperature ranges for cellulose degradation peaks were determined after 10 min of steam treatment. Residual weight values at 450 C (16.9 1.6 wt% in Table S1) were slightly decreased in comparison with those measured at Ttr ¼ 170 C. At Ttr ¼ 230 C, the weight loss curves and DTG peaks attributable to the hemicellulose degradation of steamtreated samples disappeared within 2 min (Fig. S4 in Supplementary data) similarly with Ttr ¼ 210 C. The main DTG peak for cellulose degradation was observed around 370 C after the treatment for 2e4 min, and thereafter, the peak temperature gradually decreased with treatment time, reaching 355 C after 10 min. Moreover, the residual weight at 450 C decreased to around 13.9 1.2 wt% (Table S1) less than the 18.2 1.1 wt% after the treatment at Ttr ¼ 170 C. In accordance with Yang et al. [20], these results meant that the treatment at Ttr ¼ 230 C not only completed the hemicellulose degradation, but also partially caused the cellulose decomposition. In the above results, an interesting point is the stabilization of the cellulose ingredient, which occurred suddenly after the treatment at 170 C for 2 min. Taking into consideration the
changes in microscopic observation and chemical structure of treated EFB, this stabilization of the cellulose ingredient may be capable of being induced by the removal of silica bodies. In Fig. 5, DTA profiles of the steam-treated EFB samples have been plotted. The samples treated at 170/0.82, 190/1.32, and 210 C/2.03 MPa for 10 min showed similar endothermic peaks around 370 C, but when treated at 230 C/3.00 MPa the endothermic peak appeared around 350 C. The changes in the peak-top temperature were almost the same as the changes in the DTG peak maximum shown in Fig. 3b. According to Yang’s report [20], this endothermic peak indicated the fast devolatilization of decomposed fractions from cellulose ingredients. Interestingly, no endothermic peak was observed for the raw
Fig. 5 e DTA profiles of (a) raw EFB and steam-treated EFB samples at conditions of (b) 170/0.82, (c) 190/1.32, (d) 210/ 2.03 and (e) 230 C/3.00 MPa for 10 min.
274
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
EFB, suggesting an overlap with other exothermic reactions such as the hemicellulose degradation as shown in Fig. 3. The higher the treatment temperature, the wider the endothermic peak area became. This meant that the devolatilization of cellulose ingredient became more efficient after treatment at higher temperatures. Overall, it was found that the steam treatment at Ttr ¼ 210 C under 2.03 MPa produced the best conditions for selective degradation of EFB to separate the cellulose and lignin ingredients from the hemicellulose and silica ingredients.
3.5. Changes in crystallinity of cellulose ingredients in treated EFB Fig. 6 shows the wide-angle X-ray diffraction (WAXD) curves of raw and treated EFB samples under various steam conditions. Three diffraction peaks were detected in 2q ranges of 13e18, 19e25, and 33e36 C. The diffraction patterns showed the typical crystalline structure of cellulose I, which has its main diffraction peaks at 2q values of 14.9 , 16.3 , 22.5 , and 34.6 assigned to diffractions from planes (101), (101), (002), and (040), respectively [9]. The diffraction peaks from the (101) and (101) plains merged together. The diffraction peak from plane (002) became sharpened with an increase in the steamtreatment temperature. Crystallinity values (Xc%) of raw EFB and steam-treated EFB samples were calculated from the WAXD profiles with obtained Xc% values listed in Table 1 along with crystalline size values for plains (002) and (040). The Xc% value increased from the 39.3% of the raw EFB to 57.4% for the sample treated at 230 C for 10 min. The crystalline size L002 also increased with Ttr in a similar manner with Xc%, but, the size L040 was decreased after the treatment at 170 C, and thereafter, remained relatively unchanged after increases in temperature up to 230 C. This meant that the decomposition of the integrated lignocellulosic structure resulted in the hemicellulose degradation and the extension of the crystalline in the direction of the c-axis. According to previous reports [22,23], these results could be due to a relocation of hydrolyzed paracrystalline and amorphous cellulose to crystalline regions.
Fig. 6 e XRD analysis of (a) raw EFB and steam-treated EFB samples at conditions of (b) 170/0.82, (c) 190/1.32, (d) 210/ 2.03, and (e) 230 C/3.00 MPa for 10 min.
Table 3 e Mass fraction composition of the dry raw and steam-treated EFB samples at different temperatures for 10 min. Parameters
Raw EFB Ttr ¼ 170 190 210 230 C
C and N contents C (%) 44.3 44.2 N (%) 0.8 0.8 C/N ratio 55.4 55.3 Composition of main components Cellulose (%) 47.6 50.7 Hemicellulose (%) 28.1 25.9 Lignin (%) 13.1 13.2 Others (%) 11.2 10.2 Composition of nutrients and metal elements Phosphorus (%) 0.06 0.04 Potassium (%) 1.72 1.41 Calcium (%) 0.34 0.33 Sulphur (%) 0.11 0.10 Iron (%) 0.06 0.04 Magnesium (%) 0.10 0.08 41.0 40.8 Zinc (mg kg1) 23.8 23.0 Manganese (mg kg1) 21.3 18.7 Copper (mg kg1) 8.3 7.5 Nickel (mg kg1) n.d n.d Cadmium (mg kg1)
3.6.
44.3 0.8 55.4
44.2 0.7 62.2
44.0 0.7 62.8
53.7 18.3 14.7 13.3
70.6 2.8 16.6 10.0
65.9 1.2 13.7 19.2
0.04 1.43 0.31 0.09 0.04 0.06 39.0 22.8 19.0 4.0 n.d
0.03 1.03 0.26 0.08 0.03 0.05 35.4 19.7 13.9 3.6 n.d
0.02 0.98 0.24 0.08 0.03 0.02 33.1 19.1 13.8 3.6 n.d
Chemical compositions of treated EFB
Table 3 shows changes in the nitrogen and carbon contents, the compositions of main components, such as cellulose, hemicellulose, lignin, nutrients, and metal elements in EFB during the steam treatment. The composition analysis of cellulose, hemicellulose, and lignin was conducted using Goring’s method [12], resulting in 47.6, 28.1, and 13.1%, respectively. With increase in Ttr from 170 C to 230 C, the composition of hemicellulose dramatically decreased to 1.2%, while the compositions of cellulose and lignin steadily increased up to 70.6% and 16.6% at 210 C, respectively. These changes in composition support the above-mentioned FT-IR and TG/DTA results, which suggest the degradation and removal of hemicellulose ingredient during the steam treatment, and the consequent increases in relative contents of cellulose and lignin ingredients. These results are in agreement with a previous result reported by Sun et al. [13], in which almost 90% of hemicellulose was released from wheat straw after steam treatment at 210 C. According to Seong et al. [4], the cellulose ingredient in lignocellulosic materials of oil palm fruit fiber was composed of crystalline (85%) and amorphous (15%) regions. From the observed crystallinity values of 39.3e57.4% (Fig. 6 and Table 1) and cellulose contents of 47.6e65.9% (Table 3), the crystallinity of cellulose ingredient was calculated as being in a range of 82.6e87.1%, which agreed with the previously reported value. At Ttr ¼ 230 C, the compositions of cellulose and lignin ingredients decreased to 65.9% and 13.7%, respectively. These changes could be owed to increases in the amount of thermally-converted products from cellulose and lignin, which have been listed as “others%” in Table 3. This partial degradation of cellulose ingredient at Ttr ¼ 230 C was confirmed by the DTG analysis in Fig. 3. The C, N, nutrient, and metal element contents are important parameters that have to be evaluated for use in
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
bioconversion processes such as composting. The C and N contents hardly changed during the treatments, whereas the nutrient and metal contents decreased steadily with increase in the steam temperature. After the treatment at 230 C for 10 min, the phosphorus, iron, magnesium, and nickel contents were considerably reduced to less than half of their initial values for raw EFB. The nutrients and metal elements removed by the steaming were concentrated in the liquid portions (Fig. S2), which could be used as liquid fertilizer. Cadmium element was not detected in any of the samples. The low level of nutrient content in the treated EFB samples could be compensated for easily by the addition of other high nutrient sources [3], by which a rapid co-composting process could be made workable to produce the final matured compost containing a considerable amount of nutrients.
4.
Conclusions
The EFB structure was effectively changed by the HPST process and it also enhanced its degradability characteristics. It was noticed that the steam-treated EFB samples were easily disrupted. Morphological studies revealed that the alteration of treated EFB surfaces was attributed to the removal of silica bodies from the outer surface of the EFB materials. FTIR and TG/DTA analyses also showed the evidence of hemicellulose degradation after steam treatment. In addition, at Ttr ¼ 210 C/ 2.03 MPa for 10 min of treatment, the silica bodies and hemicellulose components were removed selectively. Meanwhile, at 230 C/3.00 MPa for 10 min, the cellulose crystalline was partially disrupted. Thus, the HPST seems to be a promising method for enhancing the EFB as suitable substrate for bioconversion processes such as composting.
Appendix A. Supplementary data Supplementary data related to this article can be found at doi: 10.1016/j.biombioe.2013.02.013.
references
[1] Baharuddin AS, Wakisaka M, Shirai Y, Abd-Aziz S, Rahman NA, Hassan MA. Co-composting of empty fruit bunches and partially treated palm oil mill effluents in pilot scale. Int J Agric Res 2009;4(2):69e78. [2] Baharuddin AS, Nakamura M, Tabatabaei M, Abd-Aziz S, Abdul Rahman NA, Wakisaka M, et al. Characteristics and microbial succession in co-composting of oil palm empty fruit bunch and partially treated palm oil mill effluent. Open Biotechnol J 2009;3:87e95. [3] Baharuddin AS, Hock LS, Yusof MZM, Rahman NA, Umikalsom MS, Hassan MA, et al. Effects of palm oil mill effluent (POME) anaerobic sludge from 500 m3 of closed anaerobic methane digested tank on pressed-shredded empty fruit bunch (EFB) composting process. Afr J Biotechnol 2010;9(16):2427e36. [4] Seong KT, Hassan MA, Ariff AB. Enzymatic saccharification of pretreated solid palm oil mill effluent and oil palm fruit fiber. Pertanika J Sci Technol 2008;16(2):157e69.
275
[5] Alam MZ, Mahamat ME, Muhammat N. Production of cellulase from oil palm biomass as substrate by solid state bioconversion. Am J Appl Sci 2005;2(2):569e72. [6] Baharuddin AS, Razak NA, Abdul Rahman NA, Budiatman S, Shirai Y, Hassan MA. Bioconversion of oil palm empty fruit bunch by Aspergillus niger EB4 under solid-state fermentation. Pertanika J Trop Agric Sci 2009;32(2):143e51. [7] Umikalsom MS, Ariff AB, Shamsuddin ZH, Tong CC, Hassan MA, Karim MIA. Production of cellulose by a wild strain of Chaetomium globusum using delignified oil palm empty fruit bunch fibre as substrate. Appl Microbiol Biotechnol 1997;47(5):590e5. [8] Nigam PSN, Gupta N, Anthwal A. Pre-treatment of agroindustrial residues. In: Nigam PSN, Pandey A, editors. Biotechnology for agro-industrial residues utilisation. Dordrecht: Springer Science and Business Media BV; 2009. p. 13e32. [9] Wang K, Jiang JX, Xu F, Sun RC. Influence of steaming pressure on steam explosion pretreatment of Lespedeza stalks (Lespedeza crytobotrya): part 1. Characteristics of degraded cellulose. Polym Degrad Stab 2009;94(9):1379e88. [10] Vonk CG. Computerization of Ruland’s X-ray method for determination of the crystallinity in polymers. J Appl Cryst 1973;6(P 2):148e52. [11] Xu F, Sun RC, Lu Q. Characteristics of cellulose isolated by a totally chlorine-free method from Caragana korshinskii. J Appl Polym Sci 2006;101(5):3251e63. [12] Goring HK. Forage fiber analysis (apparatus, reagents, procedures and some applications). In: Agricultural handbook no. 379. Washington DC: Agricultural Research Service-United States Department of Agriculture; 1970. p. 1e20. [13] Sun XF, Xu F, Sun RC, Fowler P, Baird MS. Characteristics of degraded cellulose obtained from steam-exploded wheat straw. Carbohyd Res 2005;340(1):97e106. [14] Dietrich D, Hinke S, Baumann W, Fehlhaber R, Baucker E, Ruhle G, et al. Silica accumulation in Triticum aestivum L. and Dactylis glomerata L. Anal Bioanal Chem 2003;376(3):399e404. [15] Kristensen JB, Thygesen GL, Felby C, Jorgensen H, Elder T. Cell wall structural changes in wheat straw pretreated for bioethanol production. Biotechnol Biofuels 2008;1(1):5. [16] Smidt E, Schwanninger M. Characterization of waste materials using FTIR spectroscopy: process monitoring and quality assessment. Spectrosc Lett 2005;38(3):247e70. [17] Zakaria S, Hamzah H, Murshidi JA, Deraman M. Chemical modification on lignocellulosic polymeric oil palm empty fruit bunch for advance material. Adv Polym Technol 2001; 20(4):289e95. [18] Chaa L, Joly N, Lequart V, Faugeron C, Mollet JC, Martin P, et al. Isolation, characterization and valorization of hemicelluloses from Aristida pungens leaves as biomaterial. Carbohyd Polym 2008;74(3):597e602. [19] Pandey KK. A study of chemical structure of soft and hardwood and wood polymers by FTIR spectroscopy. J Appl Polym Sci 1999;71(12):1969e75. [20] Yang H, Yan R, Chen H, Lee DH, Zeng C. Characteristics of hemicellulose, cellulose and lignin pyrolysis. Fuel 2007; 86(12e13):1781e8. [21] Leyden DE, Murthy RSS, Blitz JP, Atwater JB, Rachetti A. Reflectance FTIR investigations of the reactions of silanes on silica surface. Mikrochim Acta 1988;95(1e6):53e6. [22] Tanahashi M. Characterization and degradation mechanisms of wood components by steam explosion and utilization of exploded wood. Wood Res 1990;77:49e117. [23] Yamashiki T, Matsui T, Saitoh M, Okajima K, Kamide K. Characterization of cellulose treated by the steam explosion method. Part 1: influence of cellulose resources on changes in morphology, degree of polymerization, solubility and solid structure. Br Polym J 1990;22(1):73e83.
Available online at www.sciencedirect.com
http://www.elsevier.com/locate/biombioe
Selective component degradation of oil palm empty fruit bunches (OPEFB) using high-pressure steam Azhari Samsu Baharuddin a,*, Alawi Sulaiman c, Dong Hee Kim d, Mohd Noriznan Mokhtar a, Mohd Ali Hassan b, Minato Wakisaka d, Yoshihito Shirai d, Haruo Nishida d a
Department of Process and Food Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia b Department of Bioprocess Technology, Faculty of Biotechnology and Biomolecular Sciences, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia c Faculty of Plantation and Agrotechnology, Universiti Teknologi MARA, 40410 Shah Alam, Selangor, Malaysia d Department of Biological Functions and Engineering, Graduate School of Life Science and System Engineering, Kyushu Institute of Technology, 2-4 Hibikino, Wakamatsu-ku, Kitakyushu, Fukuoka 808-0916, Japan
article info
abstract
Article history:
In order to accelerate the bioconversion process of press-shredded empty fruit bunches
Received 14 June 2011
(EFB), the effect of high-pressure steam pre-treatment (HPST) in degrading the lignocellu-
Accepted 11 February 2013
losic structure was investigated. HPST was carried out under various sets of temperature/
Available online 13 March 2013
pressure conditions such as 170/0.82, 190/1.32, 210/2.03, and 230 C/3.00 MPa. It was noted that after HPST, the surface texture, color, and mechanical properties of the treated EFB
Keywords:
had obviously altered. Scanning electron micrographs of the treated EFB exhibited effective
Empty fruit bunches
surface erosion that had occurred along the structure. Moreover, the Fourier transform
High-pressure steam
infrared and thermogravimetric analyses showed the removal of silica bodies and hemi-
Selective degradation
cellulose ingredients. X-ray diffraction profiles of the treated EFB indicated significant in-
Lignocellulosic structure
creases in crystallinity. These results reveal that HPST is an effective pre-treatment
Pre-treatment
method for altering the physicochemical properties of the EFB and enhancing its biodegradability characteristics for the bioconversion process. ª 2013 Elsevier Ltd. All rights reserved.
1.
Introduction
The empty fruit bunches (EFB) which generated at palm oil mills are extremely abundant, renewable, and readily available lignocellulosic materials. In recent years, palm oil mill industries have been keen to examine the potential of the cocomposting process using the EFB and palm oil mill effluents (POME) as an effective approach to utilizing the abundant biomass generated at palm oil mills. In utilizing the shredded or
press-shredded EFB, the co-composting treatment critically depends on the bio-conversion rate of EFB. This achieves maturity within 40e80 days using a conventional windrow in the field scale [1e3]. The longer treatment in the composting process corresponded to the phenolic compound in the cell walls of the EFB consisting of lignocellulosic materials, which could restrict the rate and extent of polysaccharide degradation [4]. Therefore, an appropriate pre-treatment method of EFB was essential to accelerate the degradation in compositing.
* Corresponding author. Tel.: þ60 3 89464424; fax: þ60 3 89464440. E-mail address: [email protected] (A.S. Baharuddin). 0961-9534/$ e see front matter ª 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.biombioe.2013.02.013
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
Many physical, chemical, and microbial pre-treatment methods have been reported for enhancing EFB’s bioconversion by improving the hydrolysis rate of lingocellulose [5,6] through, for example, removal of lignin and hemicellulose ingredients, reduction in crystallinity, and an increase in porosity. Umikalsom et al. [7], proposed a method combining the physical, thermo-chemical, and biological procedures to pre-treat EFB for bio-sugar production. Additionally, chemicals like citric acid, hydrochloric acid, EDTA, and sodium hydroxide have been used for the pre-treatment of EFB. However, such chemical pre-treatments are often expensive and inappropriate for environmentally-benign bioconversion. This is especially true in the large scale commercial operations, involving chemical pre-treatments, which require additional precautionary steps such as, use of corrosion-resistant apparatus, safe disposal of used chemicals, etc. [8]. Steam explosion is one of the alternative approaches that can provide an efficient pre-treatment method for lignocellulosic materials [9], and it has been executed at the high-pressure and temperature ranges of 1.05e3.61 MPa and 180e240 C, respectively. Steam explosion breaks up the lignocellulose structure and decomposes the hemicellulose into uronic and acetic acids, which then function as catalysts in the auto-catalytic degradation of hemicellulose and lignin ingredients [8]. However, the pre-treatments of lignocellulosic materials through steam explosion require a high energy input and might increase the cost of operation at a commercial scale. Based on our observation of a palm oil mill operation, an appropriate method using high-pressure steam (HPST) was newly suggested for the pre-treatment of EFB. The HPST is expected to alter the lignocellulosic structure of EFB so that it becomes more susceptible to the cellulose-degrading enzymes. However, the optimal conditions in terms of pressure and holding time are required to apply the HPST method in practice. Interestingly, the requirement of such high-pressure inputs can be provided by the steam line infrastructure and excess steam generated at the palm oil mills. Moreover, in order to make HPST an effective pre-treatment method, knowledge of quantitative and comprehensive analysis of the physicochemical changes of EFB during HPST becomes important especially since it may suggest that EFB is a suitable substrate in many applications such as composting. In this article, in order to accelerate the bioconversion process of the press-shredded EFB, the effects of HPST on the lignocellulosic structures were investigated in detail. The physicochemical changes of EFB were determined using scanning electron microscopy (SEM), Fourier transform infrared spectroscopy (FTIR), thermogravimetric/differential thermal analysis (TG/DTA), X-ray diffraction analysis (XRD), and inductively coupled plasma-optical emission spectrometry (ICP-OES) under various sets of temperature/pressure conditions that steam is put through.
269
Malaysia). The EFB samples were stored under conditions of 4 C in plastic bags prior to use for HPST.
2.2.
High-pressure steam pre-treatment (HPST)
HPST of press-shredded EFB was conducted in a 500 mL highpressure autoclave (START 500, Nitto Kouatsu, Co. Ltd) equipped with a temperature and pressure control system that was capable of operating up to 250 C and 9.4 MPa, respectively. About 10 g of EFB was oven-dried at 60 C for 24 h, weighed, and put into the metal wire cage in the autoclave to be treated under various HPST conditions of 170/0.82, 190/1.32, 210/2.03, and 230 C/3.00 MPa for 2, 4, 8, and 10 min. At the end of each treatment, one exhaust valve of the autoclave was opened to release the steam, and then, the treated EFB sample was oven-dried at 105 C for 24 h and weighed. Hydrolysates condensed in the autoclave after HPST were also collected as liquid portions. Changes in surface texture, color, and odor of the treated EFB samples were examined, and pH values of hydrolysates were measured on a pH meter (F-51S, Horiba Ltd., Japan).
2.3.
SEM analysis
Surface textures of raw and pre-treated EFB samples were observed on a scanning electron microscopy (SEM) (S-3400N, Hitachi, Japan) at an accelerating voltage of 10 kV. The samples were air-dried and coated with goldepalladium in a sputter coater (E-1010, Hitachi, Japan). SEM images of all the samples were taken at the same magnification of 1500 and 3000.
2.4.
Fourier transform infrared (FTIR) analysis
The Fourier transform infrared (FTIR) spectra were recorded on a Perkin Elmer GX2000R infrared spectrophotometer in a range of 500e4000 cm1 at a resolution of 4 cm1. Reflection spectra of EFB samples were measured on a Golden Gate Diamond attenuated total reflectance (ATR) (10500) module with a germanium crystal, by the single-reflection ATR method.
2.5.
X-ray diffraction (XRD) analysis
Measurements of wide-angle X-ray diffraction of EFB samples were made on a Rigaku XRD-DSC-X II diffractometer system. Cu-Ka radiation (X-ray wave length: l ¼ 0.15418 nm) was used as the source, and X-ray diffraction patterns were recorded in a range of 2q ¼ 10e40 at a scan rate of 2 min1. The degree of crystallinity was calculated from the diffracted intensity data according to Vonk’s method [10]. The crystalline size was calculated according to the Scherrer equation (1): Kl bcosq
2.
Experimental
L¼
2.1.
Materials
where L is the crystalline size in nm, b is the half-value breadth of a diffraction peak on the 2q scale in radians, and K is a constant: 0.9 [11]. Crystalline size was calculated along the characteristic (002) and (040) plane directions.
Press-shredded EFB fibers (length: 15e20 cm) were obtained from Seri Ulu Langat Palm Oil Mill (Dengkil, Selangor,
(1)
270
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
2.6. Thermogravimetric and differential thermal analyses (TG/DTA)
Table 2 e Changes in pH values of hydrolysates during steam treatment of EFB samples.
Thermogravimetric and differential thermal analyses (TG/DTA) were conducted on a simultaneous thermal analyzer EXSTAR 6200 TG/DTA system (Seiko Instruments Inc., Japan) under a constant nitrogen flow (100 mL min1). About 5 mg of oven-dried EFB sample was scanned from 25 to 500 C at a heating rate of 9 C min1. A blank aluminum pan was used as a reference.
2.7.
Chemical composition analysis
Chemical compositions of raw and treated EFB samples were analyzed on a CNHS analyzer 2000 (Leeco, USA) to determine contents of carbon, nitrogen, and nutrients. Contents of heavy metal elements were measured on an inductively coupled plasma-optical emission spectrometer (ICP-OES) (Perkin Elmer, USA). Compositions of cellulose, hemicellulose, and lignin ingredients were determined according to the Goring’s method [12].
3.
Results and discussion
3.1. Macroscopic observation of treated EFB samples and their hydrolysates After the HPST was performed, the weight and color of EFB materials were noted as having changed. This is displayed in Table 1 and Fig. S1. The treated EFB had acquired a dark brown color because of an increase in steam temperature. Similar color changes were observed in liquid portions including hydrolysates from EFB (Fig. S2 in Supplementary data). The pH values of the liquid portions were detected in an acidic range and gradually decreased from 6.0 (170 C/0.82 MPa) to 3.7 (230 C/3.00 MPa) with increases in the treatment temperature and time (Table 2). These acidic values were attributed to the acetic acid relatives and phenolic hydrolysates from the hemicellulose and lignin ingredients, respectively, which corresponded to the considerable weight loss noted in the EFB materials. Moreover, the treated EFB and liquid portions released a caramel smell after the steam treatment. Sun et al. [13] reported that brownish products from wheat straws after steam treatment were the result of degradation of carbohydrates at high steam temperatures. Interestingly, the steamtreated EFB samples, especially, at 230 C/3.00 MPa were
Table 1 e Changes in weight, crystallinity, and crystalline size of EFB samples during steam treatment. Sample
Weight loss, WL (%)
Raw EFB e Steam-treated EFB for 10 min 15.2 Ttr ¼ 170 C 23.3 190 C 39.7 210 C 46.8 230 C
Crystallinity, Xc (%)
Crystalline size (nm) L002
L040
39.3
2.68
6.03
41.7 43.1 52.5 57.4
2.91 3.38 3.75 4.14
3.44 4.25 4.12 3.96
Time (min) 2 4 8 10
pH value
Ttr ¼ 170 C
190 C
210 C
230 C
6.0 5.4 5.3 5.3
4.9 4.7 4.6 4.6
4.5 4.2 4.1 4.1
3.9 3.9 3.7 3.7
found to be so brittle that they were easily broken up by a hammer mill as compared to raw EFB. The change in mechanical properties of the EFB after the steam-treatment was expected to be advantageous for the following bioconversion process.
3.2.
Microscopic changes in morphology of treated EFB
SEM observations were conducted in order to confirm the effects of HPST on EFB surface structures. Fig. 1 shows the SEM micrographs of raw and steam-treated EFB sample surfaces at various steam temperatures after 10 min. There are obvious differences between the raw and steam-treated EFB sample surfaces, indicating that HPST had some effects on the EFB modification as follows: First, the SEM micrographs for raw EFB sample surfaces (Fig. 1a) exhibited silica bodies (phytoliths) embedded in the EFB structure similar to a micrograph shown in the previous work [3]. After HPST, the first important change was that the silica bodies were removed. The remaining holes had homogenous dimensions of around 10 mm in diameter at the outer surface (Fig. 1bec). These holes were assumed to be effective in the swelling of the EFB structures, attracting the microbes and enzymatic reactions for the subsequent bioconversion. It was observed that most of the debris and silica bodies on EFB surfaces were removed when the steam temperature was increased to 190 C (Fig. 1c). This is an important change, because the silica bodies could retard the approach of microbes to the internal cellulose layers in the lignocellulosic structure [14]. Second, most of the outer layers of the raw EFB surface were found disrupted along the structure. When the steam temperature was increased to 210 C (Fig. 1d), the outer layers of EFB decomposed and cracked along the inner structures, and after the treatment at 230 C/ 3.00 MPa for 10 min, most of the outer layers were almost completely disrupted (Fig. 1e). The formation of the holes and cracks along the EFB inner structures indicates that HPST effectively contributed to the modification of lignocellulosic structure in raw EFB. It is also suggested that the structural modification performed at 230 C/3.00 MPa corresponded to a disruption level of the EFB structure achieved by a 20 day composting process [1,2]. According to Wang’s results [9], the steam treatment could alter cellulose fibers from Lespedeza stalks and increase their reactive areas, resulting in an enhancement of the accessibility of enzymes such as cellulase. Based on the above findings, it is seen that if the raw EFB is first pre-treated with high-pressure steam, the following composting processes of EFB and POME will be expedited and will take less than 40 days of treatment given the enhancement of microbial action.
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
271
Fig. 1 e SEM micrographs at 31500 magnification of (a) raw and steam-treated EFB samples at conditions of (b) 170/0.82, (c) 190/1.32, (d) 210/2.03, and (e) 230 C/3.00 MPa for 10 min. Bar: 30 mm.
3.3.
Chemical structure of treated EFB
Fig. 2 shows the FT-IR spectra of the raw and treated EFB samples under various steam temperatures after 10 min. With
increase in the steam temperature, some absorption bands at 2916, 2850, 1730, 1635, 1232, and 892 cm1 diminished or disappeared, and other bands at 1602, 1590, 1507, 1157, 1100, and 1053 cm1 notably increased.
Fig. 2 e FTIR spectra of (a) raw EFB and steam-treated EFB samples at conditions of (b) 170/0.82, (c) 190/1.32, (d) 210/2.03, and (e) 230 C/3.00 MPa for 10 min.
272
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
Absorptions at 2916 and 2850 cm1, which are attributable to the CeH stretching vibrations, gradually diminished with an increase in the steam temperature. The CeH bands corresponded to the aliphatic moieties mainly in polysaccharides of cellulose and hemicellulose [9]. The bands at 1730 and 1232 cm1 were attributable to nC]O of acetyl and uronic ester groups, and nCeO of the carboxyl group in the hemicellulose component, respectively [15,16]. The band at 1730 cm1 was also attributable to the ester linkages of carboxylic groups of the ferulic acid and p-coumaric acids of lignin [17]. These bands had almost disappeared when the steam temperature was increased up to 230 C. The bands at 1635 and 892 cm1 were reported as being characteristic bands of OeH bending of adsorbed water [16] and a b-glycosidic linkage between the sugar units [18], respectively. The diminution and disappearance of these absorptions indicated the preferential decomposition of the hemicellulose component. On the other hand, the increased absorptions at 1602, 1590, and 1507 cm1 were characteristic bands of aromatic skeletal vibrations [9,14,18], with the absorptions at 1157, 1100, and 1053 cm1 being attributable to the dCeH (aromatic in-plane deformation) of the lignin component, in which the absorptions at 1157 and 1053 cm1 included the other characteristic bands of CeOeC asymmetric vibration and CeO stretch, respectively, in cellulose [19]. These increases in absorption of the characteristic bands meant that after the steam treatment, the compositions of lignin and cellulose ingredients increased following the decomposition of the hemicellulose ingredient. Hemicellulose consists of various saccharides having many branches, which form amorphous regions. It has been reported that hemicellulose can be easily degraded and removed from the main stem as volatile products such as CO, CO2, and some hydrocarbon at 220e315 C [20]. Meanwhile, cellulose is more stable and degrades at 315e400 C, with lignin being the most stable despite degrading in a wide temperature range of 160e900 C [20] due to its aromatic ring network having various branch structures. Characteristic bands used to identify the cellulose component are at 1420 and 1430 cm1 attributable to amorphous cellulose II and crystallized cellulose I [9]. In the present study, the absorption of the band around 1420 cm1 increased in intensity slightly with increase in the steam temperature, but diminished at 230 C. Cellulose is relatively stable up to 210 C. However, by steam treatment at 230 C, the cellulose structure was broken down in a similar manner to previous results reported by Sun et al. [13]. Difference spectra of the steam-treated samples against the raw EFB spectrum are illustrated in Fig. S3 (Supplementary data). Besides, the changes in absorption bands which were clearly visible, other changes over some broad absorption regions were also detected. Typical changes were diminutions in absorption of bands at 3200e3000, 1100e1050, and 950e900 cm1 attributable to nOeH of silica ingredients, nSieOeSi, and dSiOeH of silica ingredients; and increases in the absorption of a band at 3500e3200 cm1 attributable to nOeH of hydrogen-bonded hydroxyl groups. These changes suggest a decrease in silica ingredient during the steam treatment [21].
Fig. 3 e TG and DTG profiles of raw EFB and steam-treated EFB samples at conditions of 170/0.82, 190/1.32, 210/2.03, and 230 C/3.00 MPa for 10 min.
3.4.
Thermal analysis of steam-treated EFB
Thermal responses of the raw and steam-treated EFB samples were investigated with a thermal gravimeter (TG). TG and differential TG (DTG) profiles are illustrated in Figs. 3 and 4. Fig. 3 shows the TG/DTG profiles measured at various steamtreatment temperatures (Ttr) for 10 min. In Fig. 3a, the onset point of weight loss shifted into high temperatures with increase in Ttr, suggesting removal of some heat-instable factors by the steam treatment. All the DTG profiles (Fig. 3b) of treated samples had their main peaks shifted into a temperature range of 320e390 C, which was higher than the range of 200e350 C for raw EFB. This main peak is attributed to the degradation of cellulose ingredient. Although the main peak shifted into a temperature range of 325e400 C up to Ttr ¼ 210 C, at Ttr ¼ 230 C the main peak returned to a lower temperature range of 320e380 C. In Fig. 4, changes in TG/DTG profiles at Ttr ¼ 170 and 210 C with time are shown, respectively. The TG/DTG profiles of treated samples at Ttr ¼ 170 C exhibited the same two-step degradation behavior together with a constant residual weight of around 18 wt% without considerable change during the treatment for 2e10 min (Fig. 4aeb). These profiles indicate that a discontinuous event occurred within 2 min and caused at least three changes: 1) separation of the degrading temperature ranges of hemicellulose and cellulose components, 2) stabilization of cellulose component, and 3) decrease in the residual weight from 28.7 wt% to 18.2 1.1 wt % (Table S1). The decrease in residual weight was attributed to the removal of silica bodies from the outer surface of the EFB materials.
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
273
Fig. 4 e TG and DTG profiles of raw EFB and steam-treated EFB samples at conditions of a, b) 170 C/0.82 MPa and c, d) 210 C/ 2.03 MPa for 2, 4, 8, and 10 min.
At Ttr ¼ 210 C, all the steam-treated samples showed smooth weight loss curves (Fig. 4c) and single model DTG peaked at 370 C (Fig. 4d), with these peaks being attributable to cellulose degradation. The weight loss of the DTG peaks corresponded with the hemicellulose degradation. The shift of the peaks could be seen even after the treatment time had been reduced to 2 min. Moreover, the degradation peaks for the remaining cellulose ingredients in the EFB structure was noticed at a high temperature range of 365e370 C. Similar temperature ranges for cellulose degradation peaks were determined after 10 min of steam treatment. Residual weight values at 450 C (16.9 1.6 wt% in Table S1) were slightly decreased in comparison with those measured at Ttr ¼ 170 C. At Ttr ¼ 230 C, the weight loss curves and DTG peaks attributable to the hemicellulose degradation of steamtreated samples disappeared within 2 min (Fig. S4 in Supplementary data) similarly with Ttr ¼ 210 C. The main DTG peak for cellulose degradation was observed around 370 C after the treatment for 2e4 min, and thereafter, the peak temperature gradually decreased with treatment time, reaching 355 C after 10 min. Moreover, the residual weight at 450 C decreased to around 13.9 1.2 wt% (Table S1) less than the 18.2 1.1 wt% after the treatment at Ttr ¼ 170 C. In accordance with Yang et al. [20], these results meant that the treatment at Ttr ¼ 230 C not only completed the hemicellulose degradation, but also partially caused the cellulose decomposition. In the above results, an interesting point is the stabilization of the cellulose ingredient, which occurred suddenly after the treatment at 170 C for 2 min. Taking into consideration the
changes in microscopic observation and chemical structure of treated EFB, this stabilization of the cellulose ingredient may be capable of being induced by the removal of silica bodies. In Fig. 5, DTA profiles of the steam-treated EFB samples have been plotted. The samples treated at 170/0.82, 190/1.32, and 210 C/2.03 MPa for 10 min showed similar endothermic peaks around 370 C, but when treated at 230 C/3.00 MPa the endothermic peak appeared around 350 C. The changes in the peak-top temperature were almost the same as the changes in the DTG peak maximum shown in Fig. 3b. According to Yang’s report [20], this endothermic peak indicated the fast devolatilization of decomposed fractions from cellulose ingredients. Interestingly, no endothermic peak was observed for the raw
Fig. 5 e DTA profiles of (a) raw EFB and steam-treated EFB samples at conditions of (b) 170/0.82, (c) 190/1.32, (d) 210/ 2.03 and (e) 230 C/3.00 MPa for 10 min.
274
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
EFB, suggesting an overlap with other exothermic reactions such as the hemicellulose degradation as shown in Fig. 3. The higher the treatment temperature, the wider the endothermic peak area became. This meant that the devolatilization of cellulose ingredient became more efficient after treatment at higher temperatures. Overall, it was found that the steam treatment at Ttr ¼ 210 C under 2.03 MPa produced the best conditions for selective degradation of EFB to separate the cellulose and lignin ingredients from the hemicellulose and silica ingredients.
3.5. Changes in crystallinity of cellulose ingredients in treated EFB Fig. 6 shows the wide-angle X-ray diffraction (WAXD) curves of raw and treated EFB samples under various steam conditions. Three diffraction peaks were detected in 2q ranges of 13e18, 19e25, and 33e36 C. The diffraction patterns showed the typical crystalline structure of cellulose I, which has its main diffraction peaks at 2q values of 14.9 , 16.3 , 22.5 , and 34.6 assigned to diffractions from planes (101), (101), (002), and (040), respectively [9]. The diffraction peaks from the (101) and (101) plains merged together. The diffraction peak from plane (002) became sharpened with an increase in the steamtreatment temperature. Crystallinity values (Xc%) of raw EFB and steam-treated EFB samples were calculated from the WAXD profiles with obtained Xc% values listed in Table 1 along with crystalline size values for plains (002) and (040). The Xc% value increased from the 39.3% of the raw EFB to 57.4% for the sample treated at 230 C for 10 min. The crystalline size L002 also increased with Ttr in a similar manner with Xc%, but, the size L040 was decreased after the treatment at 170 C, and thereafter, remained relatively unchanged after increases in temperature up to 230 C. This meant that the decomposition of the integrated lignocellulosic structure resulted in the hemicellulose degradation and the extension of the crystalline in the direction of the c-axis. According to previous reports [22,23], these results could be due to a relocation of hydrolyzed paracrystalline and amorphous cellulose to crystalline regions.
Fig. 6 e XRD analysis of (a) raw EFB and steam-treated EFB samples at conditions of (b) 170/0.82, (c) 190/1.32, (d) 210/ 2.03, and (e) 230 C/3.00 MPa for 10 min.
Table 3 e Mass fraction composition of the dry raw and steam-treated EFB samples at different temperatures for 10 min. Parameters
Raw EFB Ttr ¼ 170 190 210 230 C
C and N contents C (%) 44.3 44.2 N (%) 0.8 0.8 C/N ratio 55.4 55.3 Composition of main components Cellulose (%) 47.6 50.7 Hemicellulose (%) 28.1 25.9 Lignin (%) 13.1 13.2 Others (%) 11.2 10.2 Composition of nutrients and metal elements Phosphorus (%) 0.06 0.04 Potassium (%) 1.72 1.41 Calcium (%) 0.34 0.33 Sulphur (%) 0.11 0.10 Iron (%) 0.06 0.04 Magnesium (%) 0.10 0.08 41.0 40.8 Zinc (mg kg1) 23.8 23.0 Manganese (mg kg1) 21.3 18.7 Copper (mg kg1) 8.3 7.5 Nickel (mg kg1) n.d n.d Cadmium (mg kg1)
3.6.
44.3 0.8 55.4
44.2 0.7 62.2
44.0 0.7 62.8
53.7 18.3 14.7 13.3
70.6 2.8 16.6 10.0
65.9 1.2 13.7 19.2
0.04 1.43 0.31 0.09 0.04 0.06 39.0 22.8 19.0 4.0 n.d
0.03 1.03 0.26 0.08 0.03 0.05 35.4 19.7 13.9 3.6 n.d
0.02 0.98 0.24 0.08 0.03 0.02 33.1 19.1 13.8 3.6 n.d
Chemical compositions of treated EFB
Table 3 shows changes in the nitrogen and carbon contents, the compositions of main components, such as cellulose, hemicellulose, lignin, nutrients, and metal elements in EFB during the steam treatment. The composition analysis of cellulose, hemicellulose, and lignin was conducted using Goring’s method [12], resulting in 47.6, 28.1, and 13.1%, respectively. With increase in Ttr from 170 C to 230 C, the composition of hemicellulose dramatically decreased to 1.2%, while the compositions of cellulose and lignin steadily increased up to 70.6% and 16.6% at 210 C, respectively. These changes in composition support the above-mentioned FT-IR and TG/DTA results, which suggest the degradation and removal of hemicellulose ingredient during the steam treatment, and the consequent increases in relative contents of cellulose and lignin ingredients. These results are in agreement with a previous result reported by Sun et al. [13], in which almost 90% of hemicellulose was released from wheat straw after steam treatment at 210 C. According to Seong et al. [4], the cellulose ingredient in lignocellulosic materials of oil palm fruit fiber was composed of crystalline (85%) and amorphous (15%) regions. From the observed crystallinity values of 39.3e57.4% (Fig. 6 and Table 1) and cellulose contents of 47.6e65.9% (Table 3), the crystallinity of cellulose ingredient was calculated as being in a range of 82.6e87.1%, which agreed with the previously reported value. At Ttr ¼ 230 C, the compositions of cellulose and lignin ingredients decreased to 65.9% and 13.7%, respectively. These changes could be owed to increases in the amount of thermally-converted products from cellulose and lignin, which have been listed as “others%” in Table 3. This partial degradation of cellulose ingredient at Ttr ¼ 230 C was confirmed by the DTG analysis in Fig. 3. The C, N, nutrient, and metal element contents are important parameters that have to be evaluated for use in
b i o m a s s a n d b i o e n e r g y 5 5 ( 2 0 1 3 ) 2 6 8 e2 7 5
bioconversion processes such as composting. The C and N contents hardly changed during the treatments, whereas the nutrient and metal contents decreased steadily with increase in the steam temperature. After the treatment at 230 C for 10 min, the phosphorus, iron, magnesium, and nickel contents were considerably reduced to less than half of their initial values for raw EFB. The nutrients and metal elements removed by the steaming were concentrated in the liquid portions (Fig. S2), which could be used as liquid fertilizer. Cadmium element was not detected in any of the samples. The low level of nutrient content in the treated EFB samples could be compensated for easily by the addition of other high nutrient sources [3], by which a rapid co-composting process could be made workable to produce the final matured compost containing a considerable amount of nutrients.
4.
Conclusions
The EFB structure was effectively changed by the HPST process and it also enhanced its degradability characteristics. It was noticed that the steam-treated EFB samples were easily disrupted. Morphological studies revealed that the alteration of treated EFB surfaces was attributed to the removal of silica bodies from the outer surface of the EFB materials. FTIR and TG/DTA analyses also showed the evidence of hemicellulose degradation after steam treatment. In addition, at Ttr ¼ 210 C/ 2.03 MPa for 10 min of treatment, the silica bodies and hemicellulose components were removed selectively. Meanwhile, at 230 C/3.00 MPa for 10 min, the cellulose crystalline was partially disrupted. Thus, the HPST seems to be a promising method for enhancing the EFB as suitable substrate for bioconversion processes such as composting.
Appendix A. Supplementary data Supplementary data related to this article can be found at doi: 10.1016/j.biombioe.2013.02.013.
references
[1] Baharuddin AS, Wakisaka M, Shirai Y, Abd-Aziz S, Rahman NA, Hassan MA. Co-composting of empty fruit bunches and partially treated palm oil mill effluents in pilot scale. Int J Agric Res 2009;4(2):69e78. [2] Baharuddin AS, Nakamura M, Tabatabaei M, Abd-Aziz S, Abdul Rahman NA, Wakisaka M, et al. Characteristics and microbial succession in co-composting of oil palm empty fruit bunch and partially treated palm oil mill effluent. Open Biotechnol J 2009;3:87e95. [3] Baharuddin AS, Hock LS, Yusof MZM, Rahman NA, Umikalsom MS, Hassan MA, et al. Effects of palm oil mill effluent (POME) anaerobic sludge from 500 m3 of closed anaerobic methane digested tank on pressed-shredded empty fruit bunch (EFB) composting process. Afr J Biotechnol 2010;9(16):2427e36. [4] Seong KT, Hassan MA, Ariff AB. Enzymatic saccharification of pretreated solid palm oil mill effluent and oil palm fruit fiber. Pertanika J Sci Technol 2008;16(2):157e69.
275
[5] Alam MZ, Mahamat ME, Muhammat N. Production of cellulase from oil palm biomass as substrate by solid state bioconversion. Am J Appl Sci 2005;2(2):569e72. [6] Baharuddin AS, Razak NA, Abdul Rahman NA, Budiatman S, Shirai Y, Hassan MA. Bioconversion of oil palm empty fruit bunch by Aspergillus niger EB4 under solid-state fermentation. Pertanika J Trop Agric Sci 2009;32(2):143e51. [7] Umikalsom MS, Ariff AB, Shamsuddin ZH, Tong CC, Hassan MA, Karim MIA. Production of cellulose by a wild strain of Chaetomium globusum using delignified oil palm empty fruit bunch fibre as substrate. Appl Microbiol Biotechnol 1997;47(5):590e5. [8] Nigam PSN, Gupta N, Anthwal A. Pre-treatment of agroindustrial residues. In: Nigam PSN, Pandey A, editors. Biotechnology for agro-industrial residues utilisation. Dordrecht: Springer Science and Business Media BV; 2009. p. 13e32. [9] Wang K, Jiang JX, Xu F, Sun RC. Influence of steaming pressure on steam explosion pretreatment of Lespedeza stalks (Lespedeza crytobotrya): part 1. Characteristics of degraded cellulose. Polym Degrad Stab 2009;94(9):1379e88. [10] Vonk CG. Computerization of Ruland’s X-ray method for determination of the crystallinity in polymers. J Appl Cryst 1973;6(P 2):148e52. [11] Xu F, Sun RC, Lu Q. Characteristics of cellulose isolated by a totally chlorine-free method from Caragana korshinskii. J Appl Polym Sci 2006;101(5):3251e63. [12] Goring HK. Forage fiber analysis (apparatus, reagents, procedures and some applications). In: Agricultural handbook no. 379. Washington DC: Agricultural Research Service-United States Department of Agriculture; 1970. p. 1e20. [13] Sun XF, Xu F, Sun RC, Fowler P, Baird MS. Characteristics of degraded cellulose obtained from steam-exploded wheat straw. Carbohyd Res 2005;340(1):97e106. [14] Dietrich D, Hinke S, Baumann W, Fehlhaber R, Baucker E, Ruhle G, et al. Silica accumulation in Triticum aestivum L. and Dactylis glomerata L. Anal Bioanal Chem 2003;376(3):399e404. [15] Kristensen JB, Thygesen GL, Felby C, Jorgensen H, Elder T. Cell wall structural changes in wheat straw pretreated for bioethanol production. Biotechnol Biofuels 2008;1(1):5. [16] Smidt E, Schwanninger M. Characterization of waste materials using FTIR spectroscopy: process monitoring and quality assessment. Spectrosc Lett 2005;38(3):247e70. [17] Zakaria S, Hamzah H, Murshidi JA, Deraman M. Chemical modification on lignocellulosic polymeric oil palm empty fruit bunch for advance material. Adv Polym Technol 2001; 20(4):289e95. [18] Chaa L, Joly N, Lequart V, Faugeron C, Mollet JC, Martin P, et al. Isolation, characterization and valorization of hemicelluloses from Aristida pungens leaves as biomaterial. Carbohyd Polym 2008;74(3):597e602. [19] Pandey KK. A study of chemical structure of soft and hardwood and wood polymers by FTIR spectroscopy. J Appl Polym Sci 1999;71(12):1969e75. [20] Yang H, Yan R, Chen H, Lee DH, Zeng C. Characteristics of hemicellulose, cellulose and lignin pyrolysis. Fuel 2007; 86(12e13):1781e8. [21] Leyden DE, Murthy RSS, Blitz JP, Atwater JB, Rachetti A. Reflectance FTIR investigations of the reactions of silanes on silica surface. Mikrochim Acta 1988;95(1e6):53e6. [22] Tanahashi M. Characterization and degradation mechanisms of wood components by steam explosion and utilization of exploded wood. Wood Res 1990;77:49e117. [23] Yamashiki T, Matsui T, Saitoh M, Okajima K, Kamide K. Characterization of cellulose treated by the steam explosion method. Part 1: influence of cellulose resources on changes in morphology, degree of polymerization, solubility and solid structure. Br Polym J 1990;22(1):73e83.