- Email: [email protected]
Effect of subcritical water pretreatment on cellulose recovery of water hyacinth (Eichhornia crassipe)
ARTICLE IN PRESS
JID: JTICE
[m5G;January 10, 2017;18:51]
Journal of the Taiwan Institute of Chemical Engineers 0 0 0 (2017) 1–7
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Effect of subcritical water pretreatment on cellulose recovery of water hyacinth (Eichhornia crassipe) Bich Thuyen Nguyen Thi, Lu Ki Ong, Dieu Thuy Nguyen Thi, Yi-Hsu Ju∗ National Taiwan University of Science and Technology, Department of Chemical Engineering, 43 Sec. 4, Keelung Rd., Taipei 10607, Taiwan, R.O.C
a r t i c l e
i n f o
Article history: Received 3 September 2016 Revised 14 December 2016 Accepted 18 December 2016 Available online xxx Key words: Water hyacinth Subcritical water treatment Cellulose recovery
a b s t r a c t The effect of subcritical water (SCW) pretreatment on cellulose recovery of water hyacinth was investigated in this study. Before SCW treatment, cellulose and hemicellulose contents of dried water hyacinth sample were 25.0% and 11.0%, respectively. After being treated at 165 °C and 50 bar for 30 min with a water to dried sample ratio of 10:1 (ml/g), cellulose content in the treated sample was 68.2% which is 131.5% of the untreated sample. Pretreatment using H2 SO4 was also carried out in order to understand the effect of the chemical on lignocellulose degradation and compare it with that of SCW pretreatment. The results revealed that SCW treatment was an environmentally friendly method to recovery cellulose and remove lignin from water hyacinth which is promising for bioethanol production. The effects of SCW treatment on the composition and structure of water hyacinth were studied by TGA, FTIR and SEM. © 2016 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
1. Introduction
2. Materials and methods
Water hyacinth is a widespread aquatic weed in tropical and subtropical areas. The growth rate of water hyacinth is 220 kg/ha/day. Its growth causes deteriorated native ecosystems, clogs up lakes and rivers [1–3]. However, biomass of water hyacinth has high cellulose and hemicellulose contents (Table 1.) which can provide sugars for bioconversion to fuel ethanol [2–11]. Among conversion steps to ethanol, pretreatment is a major challenge because of structural linkages in lignocellulose, which are difficult to break at initial conditions [3]. Among pretreatments, SCW also known as technology leads to a liquid phase rich in hemicellulose sugars and a solid residue rich in cellulose [12,13]. SCW treatment can reduce lignin content by breaking ether and ester bonds of lignin and hemicellulose [12] leading to improved sugar yield. In our understanding, optimal conditions of SCW treatment have not been studied for water hyacinth. Therefore, this study aimed at optimizing SCW pretreatment conditions to obtain maximum cellulose yield of water hyacinth and determining its suitability for ethanol production. Effects of temperature, time, water to solid ratio and pressure on cellulose recovery were studied. Thermogravimetric analysis (TGA), Fourier transform infrared (FTIR) analysis and scanning electron microscopy (SEM) analysis of SCW treated water hyacinth were carried out to investigate the effect of the treatment on sample’s structure.
2.1. Chemicals and materials
∗
Corresponding author. E-mail address: [email protected] (Y.-H. Ju).
Glucose (anhydrous, ≥99.5%) and 3,5-dinitro-2-hydroxybenzoic acid were purchased from Sigma Aldrich (St. Louis, USA). Fresh water hyacinth used in this study was harvested from Hau River, Viet Nam. The plants (without roots) were cleaned, chopped to about 5 cm in length and sun dried for two days, then oven dried for two days at 70 °C. The dried sample was ground and screened through 0.7 mm wire-mesh sieve. The sample was stored in airtight plastic bags at 4 °C for further use. 2.2. SCW pretreatment The equipment includes a reactor with heater, a thermocouple and a pressure gauge. The reactor is made of stainless steel with an inner volume of 200 ml that can tolerate pressure up to 100 MPa. For SCW pretreatment, nitrogen gas (99.9% purity) was used to maintain pressure in the reactor. One gram dried sample was mixed with 10 ml deionized water (DIW) and put in the reactor. The experiment was studied at different temperature (100 to 200 °C), time (15–90 min), DIW to solid ratio (7:1 to 20:1 ml/g) and pressure (0–150 bar). After pretreatment, the reactor was cooled to room temperature, the solid was removed by filtration, washed by DIW, then dried and its lignocellulosic content analyzed by TGA. Reducing sugars (RDS) content in the liquid hydrolysate was determined by DNS (3,5-dinitrosalicylic acid) method [14]. All data were expressed as average of duplicate experiments.
http://dx.doi.org/10.1016/j.jtice.2016.12.028 1876-1070/© 2016 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth (Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
ARTICLE IN PRESS
JID: JTICE 2
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7 Table 1 Lignocellulose contents of water hyacinth (% dry biomass). Cellulose
Hemicellulose
Lignin
References
Treatment methods
23.3 18.2 18.3 24.3 36 49 25.0 68.2
22.1 29.3 23.3 22.5
not shown 2.8 17.7 8.6 6.5 not shown 2.5 0.9
[2] [11] [10] [10] [3] [3] this study this study
Untreated Untreated Untreated DMSO, 120 °C, 24 h Untreated Ultrasound, 20 min Untreated Subcritical water, 165 °C, 30 min, 50 bar
11.0 0
The yield of RDS and cellulose Eqs. (1) and (2) as shown below:
RDS yield (% ) =
were
calculated
using
RDS (mg/g ) obtained from DNS method theoretical RDS (mg/g ) (1)
Where theoretical RDS was obtained by the method of Xia et al. [2]
Cellulose yield (% ) Cellulose (% ) after treatment which based on remained solid = Cellulose (% ) of untreated sample (2) Where cellulose (%) is obtained from TGA analyis.
Fig. 1. Effect of temperature on lignocellulose and RDS content of water hyacinth. (a) cellulose & hemicellulose; (b) cellulose. Reaction conditions: 30 min, 50 bar, DIW to solid ratio = 10: 1 (ml/g).
2.3. Subcritical H2 SO4 (SC H2 SO4 ) pretreatment Dried water hyacinth was treated by H2 SO4 (0.25−5% w/v) under the following conditions: 165 °C, 30 min, 50 bar and a H2 SO4 to dried biomass ratio of 10:1 ml/g.
2.4. Statistical analysis T-test: two samples assuming equal variance was used for considering significant difference and insignificant difference of investigated points in this study.
2.5. Thermal analysis Dried water hyacinth sample (5−7 mg) was pyrolyzed by TGA machine (Model: Perkin-Elmer) in the temperature range of 30−800 °C at an increasing rate of 10 °C/min. Thermogravimetry (TG) trials were carried out at a nitrogen flow rate of 40 ml/min [3].
2.6. SEM analysis Surface morphology of samples were observed by SEM (JSM−6390LV, JEOL, Japan) at an accelerated voltage of 15−20 kV after gold or platinum coating by a JEOL JFC-1100 E sputtering device for 85 s prior to SEM observation.
2.7. FTIR analysis A FTIR Bio-Rad FTS-3500 spectroscopy was employed in this study. The spectrum was obtained in the transmission mode in 40 0 0−40 0 cm−1 at a resolution of 4 cm−1 with 40 scans per sample.
3. Results and discussion 3.1. Effect of temperature on RDS and lignocellulose content SCW is hot water between 100 and 374 °C using pressure to maintain water in liquid state. Fig. 1. shows that SCW temperature has strong effect on both liquid and solid compositions. RDS content increased from 48.2 mg/g to 111.3 mg/g as temperature was increased from 100 to 170 °C, then decreased to 100.4 mg/g and 68.6 mg/g as temperature was increased to 180 °C and 200 °C, respectively. These results are consistent with that of Xia et al. [2], Pronyk & Mazza [13] and Wei et al. [15] who reported that pretreatment at high temperature decreased sugars concentration due to degradation of sugars into smaller molecules such as furfural, acetic acid, 5-hydroxymethyl-2-furaldehyde (HMF) and formic acid. In lignocellulosic materials, cellulose, hemicellulose and lignin are linked together in tight structure which is difficult to degrade at normal conditions. To release cellulose and hemicellulose, pretreatment is required to break links between cellulose, hemicellulose and lignin. Cellulose and hemicellulose content in the SCW treated water hyacinth in this study are shown in Fig. 1. Cellulose and hemicellulose content increased from 60.3% to 64.5% as temperature was raised from 100 to 155 °C. These results can be attributed to the role of SCW in the digestion of lignocellulosic component of water hyacinth. At 25 °C, the dielectric constant ɛ is 78.5. This value decreases to 55.43 at 100 °C/50 bar and 43.95 at 150 °C/50 bar because hydrogen bonding is weakened by increasing temperature. Hence polarity of water decreases with increasing temperature and water becomes more non-polar and a good solvent for organic compounds (ɛdimethyl sulfoxide = 46.6 at 20 °C, ɛmethanol = 32.6 at 25 °C) [12]. In addition, as temperature increases water density decreases which leads to increase in diffusivity and hence to increase in degradation of cellulose, hemicellulose and lignin in the tight biomass matrix. It was observed
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
ARTICLE IN PRESS
JID: JTICE
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7
3
Table 2 Mass balance of subcritical water pretreatment. No.
Solid used (g)
1 2 3 4 5
0.9995 1.0 0 05 1.0040 1.0 0 0 1.0030
± ± ± ± ±
0.0035 0.0 0 07 0.0014 0.0 0 03 0.0057
Solid remained (g) 0.5103 0.4823 0.4823 0.4823 0.3241
± ± ± ± ±
0.0018 0.0109 0.0 0 09 0.0 0 01 0.0089
Conditions 155 °C, 165 °C, 165 °C, 165 °C, 165 °C,
in this study that as treatment temperature was raised from 100 to 155 °C, cellulose and hemicellulose content was increased from 60.3% to 64.5%, corresponding to a decrease of lignin content from 2.0% to 1.3%. This finding is similar to that of Ciftci & Saldaña [12]. As seen in Table 2., increasing temperature from 155 °C to 165 °C decreased residue which is due to digestion of hemicellulose and lignin leading to the releasing of cellulose (Fig. 1.) which was expelled from the tight lignin and hemicellulose structure. Although the loss and increase of residue were observed, increase in the produced RDS was not fit with calculated number of loss. This was due to generation of components such as furfural and HMF [2,20]. In this study, almost all hemicellulose was degraded as treatment temperatue was raised from 165 °C to 200 °C, therefore, the main compound in residue was cellulose (shown in thermal analysis & Fig. 1.). Moreover, between 165 °C and 200 °C, lignin content in the solid (0.9%) was lower than that treated at temperatures lower than 165 °C which means that lignin disruption was higher. As a result, cellulose content was the highest (∼68.2%) between treatment temperature of 165 °C to 200 °C since cellulose was not tied by linkage with hemicellulose and lignin and expelled from the closed matrix. Wei et al. [15] showed that as treatment temperature was raised from 160 to 200 °C, cellulose content in eucalyptus was enhanced from 44.83% to 53.96%. Xia et al. [2] reported that lignin in water hyacinth slightly decreased from 92% at 120 °C to 78% at temperatures above 160 °C. Moreover, Ando et al. [16] reported that cellulose of bamboo, chinquapin, and Japan cedar started to degrade at temperatures above 230 °C. From the above discussion, in view of cellulose recovery, temperatures higher than 200 °C were not investigated in this study. By using statistical analysis, Fig. 1. shows that there are no significant difference of cellulose content among treatment temperature of 165 °C, 170 °C, 180 °C and 200 °C (P > 0.05), and no significant difference of sugars yield between treatment temperature of 165 °C and 170 °C (P>0.05). Thus, 165 °C is preferable in terms of yield of cellulose and sugars. As expected, at 165 °C the RDS content obtained was only 111.3 mg/g. Whereas, cellulose recovery after treatment was 68.2% which is 131.5% of the untreated sample. Therefore, 165 °C was chosen as the temperature for recovering cellulose in this study. 3.2. Effect of time on RDS and cellulose content As shown in Fig. 2., RDS content increased with treatment time, reaching a maximum of about 111.3 mg/g at 30 to 35 min, then decreased to 82.2 mg/g as time was prolonged to 90 min. The reason is that at 165 °C RDS content mostly were produced from hemicellulose hydrolysis and very little from hydrolysis of cellulose for a treatment time of 15–35 min [2,12,13,15] and [17]. However, sugars were decomposed into smaller molecules at prolonging time which resulted in a decline of sugars yield. This finding is similar to the study on water hyacinth by Xia et al. [2] who reported that total RDS remained constant (19.0 g/100 g) when microwave pretreatment time was between 5 and 25 min, but decreased to 16.0 g/100 g as time was extended to 45 min owning to forma-
30 min, 50 bar, DIW to solid ratio = 10: 1 (ml/g) 30 min, 50 bar, DIW to solid ratio = 10: 1 (ml/g) 30 min, 50 bar, DIW to solid ratio = 20: 1 (ml/g) 30 min, 150 bar, DIW to solid ratio = 10: 1 (ml/g) 50 bar, 30 min, 1% H2 SO4 to dried biomass ratio = 10:1 ml/g)
Fig. 2. Effect of time on cellulose and RDS content of water hyacinth (conditions: water to biomass ratio = 10: 1 ml/g, 165 °C, 50 bar). (a) cellulose & hemicellulose; (b) cellulose.
tion of furfural and HMF by hemicellulose and cellulose decomposition. Similar trend was shown in a study using liquid hot water pretreatment on eucalyptus [15], the highest xylose yield was 13.09 g/100 g raw biomass which was generated after treated at 180 °C for 20 min, but only limited amount of glucose (2.5 g/100 g raw material) was released. The results implied that decomposition of hemicellulose started at lower temperature (180 °C) than that of cellulose in liquid hot water. As pretreatment time was increased from 20 to 30, then to 40 min; the corresponding xylose yield declined from 13.09 to 11.37, then to 9.57 g due to the formation of inhibitors [15]. As shown in Fig. 2., cellulose content reached 68.2% at 30 min and kept almost constant until 90 min. This may be attributed to lignin removal which was significant at 30 min and prolonged time more than 30 min did not remove more lignin. In addition, hemicellulose decomposition started at 15 min and was completed at 30 min, resulting in the highest liberation of cellulose at 30 min. Increasing of time from 30 to 90 min did not improve cellulose content. These results are consistent with that of Xia et al. [2] and Perez et al. [18]. Product distribution can be controlled thermodynamically or kinetically, as indicated by the change of reaction product ratio as temperature changes. Since a temperature change caused a change in the RDS to cellulose ratio (Fig. 9.), it is possible that kinetic controlled biomass degradation. Further proof was given in Fig. 10. which shows that RDS/cellulose ratio increased with increasing temperature (from 170 to 200 °C). Increasing RDS/cellulose ratio reflected different equilibrium constant, which is inconsistent with that of thermodynamic control, which theoretically possesses the same Go over a modest temperature range. Higher RDS in the product at higher temperature was probably due to higher kinetic rate of glucose formation from cellulose. 3.3. Effect of pressure on RDS and cellulose content As shown in Fig. 3., both RDS and cellulose content increased with increasing pressure from 0 to 50 bar, then kept at 68.2% for cellulose yield and 111.3 mg/g for RDS as pressure was increased
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
JID: JTICE 4
ARTICLE IN PRESS
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7
Fig. 3. Effect of pressure on cellulose and RDS content of water hyacinth (reaction conditions: water to biomass ratio = 10: 1 ml/g, 165 °C, 30 min). (a) cellulose & hemicellulose; (b) cellulose.
Fig. 4. Effect of DIW to dried biomass ratio (mL/g) on RDS and cellulose content of water hyacinth (reaction conditions: 30 min, 165 °C, 50 bar). (a) cellulose & hemicellulose; (b) cellulose.
from 80 to 150 bar (P>0.05) since most hemicellulose was removed at 165 °C and 50 bar resulting in the huge increase of cellulose. Pressure higher than 50 bar was not effective for obtaining higher yield of sugars and cellulose since the main role of pressure is to hold water in liquid condition [12,13]. This tendency agrees with that of Hanim et al., [17], who showed that at 190 °C hemicellulose yield in palm increased significantly from 21% to 66% as pressure was raised from 500 psi to 600 psi. Ciftci & Saldaña [12] examined the effect of pressure (50–200 bar) on lignocellulosic degradation by analyzing the phenolic content of lupin hull. In lignocellulosic biomass, hemicelluloses link to lignin by phenolic compounds. SCW treatment breaks such linkages, increasing total phenolic content in the extracts. Their results showed that at 200 bar, phenolic content increased from 0.07 mg gallic acid/g biomass at 160 °C to 0.72 mg gallic acid/g biomass at 220 °C. In contrast, pressure had insignificant effect on total phenolic content and cellulose yield. Cantero et al. [19] reported that pressure did not affect cellulose content at these conditions: 18−27 MPa/350 °C and 10−27 MPa/300 °C. A pressure of 50 bar was chosen in this study. 3.4. Effect of water to dried biomass ratio on RDS and cellulose content The effects of water to solid ratio on cellulose composition and RDS content are shown in Fig. 4. It can be seen in Fig. 4. that both RDS and cellulose content increased as water to biomass ratio was increased from 7:1 to 20:1 (ml/g). However, the increase
Fig. 5. Effect of H2 SO4 concentration on RDS and cellulose content (reaction conditions: 30 min; 165 °C; 50 bar, H2 SO4 to dried biomass ratio = 10:1 ml/g.) (a) cellulose & hemicellulose; (b) cellulose.
of both RDS and cellulose yield as water to biomass ratio was increased from 10:1 to 20:1 is insignificant (P>0.05). This is consistent with mass balance shown in Table 2. which shows that there was no change in the residue weight as water to biomass ratio was increased from 10:1 to 20:1 (ml/g). At a fixed amount of solid (1 g), 7 ml of water was not enough for blending of water and solid resulting in incomplete disruption of biomass. Increasing water amount from 7 ml to 10 ml facilitated water penetration into biomass to break down lignocellulosic structure. As discussed previously, lignin was mostly degraded and hemicellulose was removed completely at 165 °C, 50 bar, 30 min and a water to biomass ratio of 10: 1 ml/g. Under these conditions, the highest recovery of cellulose was obtained. These results are in agreement with that of Xia et al. [2]. Hence the optimal conditions for cellulose recovery in this study were found to be temperature = 165 °C, pressure = 50 bar, time = 30 min and water to dried biomass ratio = 10:1 ml/g. Furthermore, the use of excessive water to penetrate dried water hyacinth structure may reflect the result obtained for the SCW process using fresh water hyacinth. Excessive water may plasticize the lignin matrix structure in the original dried wooden biomass resulting in the same reaction characteristic with fresh biomass, which has been saturated with water [24]. 3.5. Comparison of effectiveness for cellulose recovery between SCW pretreatment and SC H2 SO4 pretreatment As shown in Fig. 5., there was an increase in both cellulose and RDS content as H2 SO4 concentration was raised from 0.25% to 1%. RDS content reached maximum (264.4 mg/g) at 2% H2 SO4 , then dropped to 114.6 mg/g as H2 SO4 concentration was increased to 5% due to the decomposition of sugars into smaller molecules [2,20]. H2 SO4 is able to dissolve hemicellulose at low concentration [2,20], therefore, raising concentration of H2 SO4 from 0.25−1% released cellulose from the confinement of hemicellulose and lignin. However, as H2 SO4 content was raised form 1% to 5%, cellulose content dramatically decreased from 69.4% to 23.9% which may be caused by the disruption of crystalline cellulose at high acid concentration. These results are in agreement with that of Xia et al. [2]. In this study the highest RDS content obtained was only 264.4 mg/g, while cellulose concentration was 69.4% but the cellulose yield was only 89.9%. The low yield is due to the low residue left after treatment, as can be seen in Table 2. In addition, using acid will generate toxic components such as furfural and HMF which are toxic to micro-organisms and requires neutralization of acid [2,20]. Meanwhile cellulose yield obtained by SCW was 131.5%.
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
JID: JTICE
ARTICLE IN PRESS
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7
5
Fig. 6. TGA curves of water hyacinth. (a) untreated sample, (b) treated at 130 °C, 50 bar, 30 min, water/solid (10:1 ml/g), (c) treated at 165 °C, 50 bar, 30 min, treated at 165 °C, 50 bar, 30 min, H2 SO4 1%/solid (10:1 ml/g), (e) treated at 165 °C, 50 bar, 30 min, H2 SO4 5%/solid (10:1 ml/g). water/solid (10:1 ml/g), (d)
3.6. TGA Moisture in water hyacinth evaporated at 30–200 °C. Hemicellulose and cellulose decomposed at 200−350 °C and lignin degradation occurred at 350−470 °C [3], 23]. Fig. 6a indicates that hemicellulose was digested at 265 °C while the main peak at 327 °C is attributed to cellulose degradation. Untreated sample curve (Fig. 6a) shows that hemicellulose peak is the only shoulder in the derivative thermogravimetric curve which means that hemicellulose is the minor component in water hyacinth. The peak gradually disappeared with increasing treating temperature from 100 to 155 °C and disappeared at 165 °C. This observation was similarly found for the H2 SO4 (0.25−3%) treated sample. These results imply that hemicellulose was completely degraded at 165 °C by SCW and H2 SO4 (0.25−3%). However, cellulose peak significantly decreased at high H2 SO4 concentration (4−5%) indicating that cellulose was decomposed at high acid concentration. The small peak appeared at 468 °C was attributed to lignin. This observation agrees with that of Harun [3] and Sundari [23]. 3.7. SEM Fig. 7a shows surface of the untreated water hyacinth. The surface is smooth and randomly tight. The structure was partly broken after SCW treatment at 130 °C and 50 bar for 30 min (Fig. 7b). Nevertheless, the overall structure of lignocellulose was still intact. This finding indicates that SCW at low temperature cannot break the intact structure of lignin and hemicellulose which encloses cellulose. Increasing temperature to 165 °C dramatically degraded water hyacinth. The surface became more crumbled, bunches of fibers were broken revealing that the compact lignocellulose structure was significantly degraded leading to release of cellulose from the tight enclosure of hemicellulose and lignin (Fig. 7c). Therefore, cellulose yield of SCW pretreatment at 165 °C was higher than that at other temperatures. This was similarly detected for H2 SO4 (0.25%) treated sample (Fig. 7d). It seems that there is an agglomeration on the surface of H2 SO4 (5%) treated sample. Fig. 7e shows that the surface was smoothened because of serious collapse of structure caused by using higher concentration H2 SO4 . As a result, agglomeration of lignocellulose structure was formed. 3.8. FTIR Fig. 8 shows FTIR spectra of the untreated and the treated water hyacinth samples. In the untreated sample, the peaks observed at
Fig. 7. SEM images of water hyacinth. (a) untreated sample, (b) treated at 130 °C, 50 bar, 30 min, water/solid (10:1 ml/g), (c) treated at 165 °C, 50 bar, 30 min, water/solid (10:1 ml/g), (d) treated at 165 °C, 50 bar, 30 min, H2 SO4 1%/solid (10:1 ml/g), (e) treated at 165 °C, 50 bar, 30 min, H2 SO4 5%/solid (10:1 ml/g).
1249 and 1379 cm−1 are those of C–O stretching of syringyl lignin and C–O–C of hemicellulose oscillation in anomeric region, respectively [4,21]. The wave at 1735 cm−1 is originated from acetyl and uronic ester groups of hemicelluloses [22,23]. The peaks at 1249 cm−1 and 1735 cm−1 disappeared for the 165 °C SCW treated sample indicating that pretreatment removed hemicellulose and partial lignin. For acid treated samples, bands at 1249 and 1735 cm−1 are not seen in the curves which means that pretreatment by acid under the studied subcritical conditions
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
JID: JTICE 6
ARTICLE IN PRESS
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7
tions are in agreement with Abral [21,22,23] and in line with TGA and SEM results. 4. Conclusions A cellulose content of 69.4% (cellulose yield 89.9%) was obtained from pretreatment using 1% H2 SO4 under conditions of 165 °C, 50 bar, a water to dried biomass ratio of 10 ml/g and a treatment time of 30 min. Besides, using acid has several disadvantages as mentioned previously. Meanwhile, SCW pretreatment is not only an environmentally friendly method but also generated high cellulose (68.2%) which is 131.5% of the untreated sample. Functional groups and morphology of all samples were inspected by FTIR, SEM and TGA. Results indicate degradation of lignocellulosic materials which explains the enhancement of cellulose content. These results are promising for ethanol production, therefore, ethanol production will be investigated in the subsequent study. References Fig. 8. FTIR spectra of untreated and treated water hyacinth.
Fig. 9. RDS to cellulose ratio vs. time. Reaction conditions: water to biomass ratio = 10: 1 ml/g, 165 °C, 50 bar
Fig. 10. RDS to cellulose ratio vs. temperature. Reaction conditions: water to biomass ratio = 10: 1 ml/g, 50 bar, 90 min.
probably dissolved hemicellulose and partial lignin. However, high concentration of sulfuric acid also disrupted cellulose, therefore, cellulose content in the H2 SO4 treated sample sharply dropped from 69.4% to 23.9% with increasing acid concentration from 1% to 5%. The result was supported by the disappearance of peak at 808 cm−1 in the 5% H2 SO4 treated sample which corresponds to the C–H rocking vibration of cellulose. The peaks at 3408 cm−1 and 2985 cm−1 are –OH and –CH stretching vibrations. These observa-
[1] Bayrakci AG, Kocar G. Second-generation bioethanol production from water hyacinth and duckweed in Izmir: a case study. Renew Sustain Energy Rev 2014;30:306–16. [2] Xia A, Cheng J, Song W, Yu C, Zhou J, Cen K. Enhancing enzymatic saccharification of water hyacinth through microwave heating with dilute acid pretreatment for biomass energy utilization. Energy 2013;61:58–166. [3] Harun MY, Radiah ABD, Abidin ZZ, Yunus R. Effect of physical pretreatment on dilute acid hydrolysis of water hyacinth (Eichhornia crassipes). Bioresour Technol 2011;102:5193–9. [4] Ahmed IN. Bioethanol production from Melaleuca leucadendron shedding bark: assessing pretreatments. Solid and enzyme loading; 2013. PhD thesis p. 8–9. [5] Cheng J, Lin R, Song W, Xia A, Zhou J, Cen K. Enhancement of fermentative hydrogen production from hydrolyzed water hyacinth with activated carbon detoxification and bacteria domestication. Int J Hydrogen Energy 2015;40:2545–51. [6] Yan J, Wei Z, Wang Q, He M, Li S, Irbis C. Bioethanol production from sodium hydroxide/hydrogen peroxide pretreated water hyacinth via simultaneous saccharification and fermentation with a newly isolated thermotolerant Kluyveromyces marxianu strain. Bioresour Technol 2015;193:103–9. [7] Guragain YN, Coninck JD, Husson F, Durand A, Rakshit SK. Comparison of some new pretreatment methods for second generation bioethanol production from wheat straw and water hyacinth. Bioresour Technol 2011;102:4416–24. [8] Singh A, Bishnoi NR. Ethanol production from pretreated wheat straw hydrolyzate by Saccharomyces cerevisiae via sequential statistical optimization. Ind Crops Prod 2013;41:221–6. [9] Aswathy US, Sukumaran RK, Devi GL, Rajasree KP, Singhania RR, Pandey A. Bio-ethanol from water hyacinth biomass: an evaluation of enzymatic saccharification strategy. Bioresour Technol 2010;101:925–30. [10] Gao J, Chen L, Yan Z, Wang L. Effect of ionic liquid pretreatment on the composition, structure and biogas production of water hyacinth (Eichhornia crassipes). Bioresour Technol 2013;132:361–4. [11] Ma F, Yang N, Xu C, Yu H, Wu J, Zhang X. Combination of biological pretreatment with mild acid pretreatment for enzymatic hydrolysis and ethanol production from water hyacinth. Bioresour Technol 2010;101:9600–4. [12] Ciftci D, Saldaña MDA. Hydrolysis of sweet blue lupin hull using subcritical water technology. Bioresour Technol 2015;194:75–82. [13] Pronyk C, Mazza G. Kinetic modeling of hemicellulose hydrolysis from triticale straw in a pressurized low polarity water flow-through reactor. Ind Eng Chem Res 2010;49:6367–75. [14] Miller GL. Use of dinitrosalicylic acid reagent for the determination of reducing sugars. Anal Chem 1959;31:426–8. [15] Wei W, Shubin W, Liguo L. Combination of liquid hot water pretreatment and wet disk milling to improve the efficiency of the enzymatic hydrolysis of eucalyptus. Bioresour Technol 2013;128:725–30. [16] Ando H, Sakaki T, Kokusho T, Shibata M, Uemura Y, Hatate Y. Decomposition behavior of plant biomass in hot-compressed water. Ind Eng Chem Res 20 0 0;39:3688–93. [17] Hanim SS, Norsyabilah R, Suhaila MHN, Noraishah A, Kartina AKS. Effects of temperature, time and pressure on the hemicelluloses yield extracted using subcritical water extraction. Proc Eng 2012;42:562–5. [18] Pérez JA, Ballesteros I, Ballesteros M, Negro MJFS, Manzanares P. Optimizing liquid hot water pretreatment conditions to enhance sugar recovery from wheat straw for fuel-ethanol production. Fuel 2008;87:3640–7. [19] Cantero DA, Tapia AS, Bermejo MD, Cocero MJ. Pressure and temperature effect on cellulose hydrolysis in pressurized water. Chem Eng J 2015;276:145–54.
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
JID: JTICE
ARTICLE IN PRESS
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7 [20] Ganguly A, Chatterjee PK, Deyb. Studies on ethanol production from water hyacinth—A review. Renew Sustain Energy Rev 2012;16:966–72. [21] Abral H, Kadriadi D, Rodianus A, Mastariyanto P, Ilhamdi, Arief S, Sapuan SM, et al. Mechanical properties of water hyacinth fibers – polyester composites before and after immersion in water. Mater Des 2014;58:125–9. [22] Fang JM, Sun RC, Tomkinson J. Isolation and characterization of hemicelluloses and cellulose from rye straw by alkaline peroxide extraction. Cellulose 20 0 0;7:87–107.
7
[23] Sundari MT, Ramesh A. Isolation and characterization of cellulose nanofibers from the aquatic weed water hyacinth—Eichhornia crassipes. Carbohydr Polym 20 0 0;87:1701–5. [24] Ibbett R, Gaddipati S, Davies S, Hill S, Tucker G. The mechanisms of hydrothermal deconstruction of lignocellulose: new insights from thermal–analytical and complementary studies. Bioresour Technol 2011;102:9272–8.
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
JID: JTICE
[m5G;January 10, 2017;18:51]
Journal of the Taiwan Institute of Chemical Engineers 0 0 0 (2017) 1–7
Contents lists available at ScienceDirect
Journal of the Taiwan Institute of Chemical Engineers journal homepage: www.elsevier.com/locate/jtice
Effect of subcritical water pretreatment on cellulose recovery of water hyacinth (Eichhornia crassipe) Bich Thuyen Nguyen Thi, Lu Ki Ong, Dieu Thuy Nguyen Thi, Yi-Hsu Ju∗ National Taiwan University of Science and Technology, Department of Chemical Engineering, 43 Sec. 4, Keelung Rd., Taipei 10607, Taiwan, R.O.C
a r t i c l e
i n f o
Article history: Received 3 September 2016 Revised 14 December 2016 Accepted 18 December 2016 Available online xxx Key words: Water hyacinth Subcritical water treatment Cellulose recovery
a b s t r a c t The effect of subcritical water (SCW) pretreatment on cellulose recovery of water hyacinth was investigated in this study. Before SCW treatment, cellulose and hemicellulose contents of dried water hyacinth sample were 25.0% and 11.0%, respectively. After being treated at 165 °C and 50 bar for 30 min with a water to dried sample ratio of 10:1 (ml/g), cellulose content in the treated sample was 68.2% which is 131.5% of the untreated sample. Pretreatment using H2 SO4 was also carried out in order to understand the effect of the chemical on lignocellulose degradation and compare it with that of SCW pretreatment. The results revealed that SCW treatment was an environmentally friendly method to recovery cellulose and remove lignin from water hyacinth which is promising for bioethanol production. The effects of SCW treatment on the composition and structure of water hyacinth were studied by TGA, FTIR and SEM. © 2016 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
1. Introduction
2. Materials and methods
Water hyacinth is a widespread aquatic weed in tropical and subtropical areas. The growth rate of water hyacinth is 220 kg/ha/day. Its growth causes deteriorated native ecosystems, clogs up lakes and rivers [1–3]. However, biomass of water hyacinth has high cellulose and hemicellulose contents (Table 1.) which can provide sugars for bioconversion to fuel ethanol [2–11]. Among conversion steps to ethanol, pretreatment is a major challenge because of structural linkages in lignocellulose, which are difficult to break at initial conditions [3]. Among pretreatments, SCW also known as technology leads to a liquid phase rich in hemicellulose sugars and a solid residue rich in cellulose [12,13]. SCW treatment can reduce lignin content by breaking ether and ester bonds of lignin and hemicellulose [12] leading to improved sugar yield. In our understanding, optimal conditions of SCW treatment have not been studied for water hyacinth. Therefore, this study aimed at optimizing SCW pretreatment conditions to obtain maximum cellulose yield of water hyacinth and determining its suitability for ethanol production. Effects of temperature, time, water to solid ratio and pressure on cellulose recovery were studied. Thermogravimetric analysis (TGA), Fourier transform infrared (FTIR) analysis and scanning electron microscopy (SEM) analysis of SCW treated water hyacinth were carried out to investigate the effect of the treatment on sample’s structure.
2.1. Chemicals and materials
∗
Corresponding author. E-mail address: [email protected] (Y.-H. Ju).
Glucose (anhydrous, ≥99.5%) and 3,5-dinitro-2-hydroxybenzoic acid were purchased from Sigma Aldrich (St. Louis, USA). Fresh water hyacinth used in this study was harvested from Hau River, Viet Nam. The plants (without roots) were cleaned, chopped to about 5 cm in length and sun dried for two days, then oven dried for two days at 70 °C. The dried sample was ground and screened through 0.7 mm wire-mesh sieve. The sample was stored in airtight plastic bags at 4 °C for further use. 2.2. SCW pretreatment The equipment includes a reactor with heater, a thermocouple and a pressure gauge. The reactor is made of stainless steel with an inner volume of 200 ml that can tolerate pressure up to 100 MPa. For SCW pretreatment, nitrogen gas (99.9% purity) was used to maintain pressure in the reactor. One gram dried sample was mixed with 10 ml deionized water (DIW) and put in the reactor. The experiment was studied at different temperature (100 to 200 °C), time (15–90 min), DIW to solid ratio (7:1 to 20:1 ml/g) and pressure (0–150 bar). After pretreatment, the reactor was cooled to room temperature, the solid was removed by filtration, washed by DIW, then dried and its lignocellulosic content analyzed by TGA. Reducing sugars (RDS) content in the liquid hydrolysate was determined by DNS (3,5-dinitrosalicylic acid) method [14]. All data were expressed as average of duplicate experiments.
http://dx.doi.org/10.1016/j.jtice.2016.12.028 1876-1070/© 2016 Taiwan Institute of Chemical Engineers. Published by Elsevier B.V. All rights reserved.
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth (Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
ARTICLE IN PRESS
JID: JTICE 2
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7 Table 1 Lignocellulose contents of water hyacinth (% dry biomass). Cellulose
Hemicellulose
Lignin
References
Treatment methods
23.3 18.2 18.3 24.3 36 49 25.0 68.2
22.1 29.3 23.3 22.5
not shown 2.8 17.7 8.6 6.5 not shown 2.5 0.9
[2] [11] [10] [10] [3] [3] this study this study
Untreated Untreated Untreated DMSO, 120 °C, 24 h Untreated Ultrasound, 20 min Untreated Subcritical water, 165 °C, 30 min, 50 bar
11.0 0
The yield of RDS and cellulose Eqs. (1) and (2) as shown below:
RDS yield (% ) =
were
calculated
using
RDS (mg/g ) obtained from DNS method theoretical RDS (mg/g ) (1)
Where theoretical RDS was obtained by the method of Xia et al. [2]
Cellulose yield (% ) Cellulose (% ) after treatment which based on remained solid = Cellulose (% ) of untreated sample (2) Where cellulose (%) is obtained from TGA analyis.
Fig. 1. Effect of temperature on lignocellulose and RDS content of water hyacinth. (a) cellulose & hemicellulose; (b) cellulose. Reaction conditions: 30 min, 50 bar, DIW to solid ratio = 10: 1 (ml/g).
2.3. Subcritical H2 SO4 (SC H2 SO4 ) pretreatment Dried water hyacinth was treated by H2 SO4 (0.25−5% w/v) under the following conditions: 165 °C, 30 min, 50 bar and a H2 SO4 to dried biomass ratio of 10:1 ml/g.
2.4. Statistical analysis T-test: two samples assuming equal variance was used for considering significant difference and insignificant difference of investigated points in this study.
2.5. Thermal analysis Dried water hyacinth sample (5−7 mg) was pyrolyzed by TGA machine (Model: Perkin-Elmer) in the temperature range of 30−800 °C at an increasing rate of 10 °C/min. Thermogravimetry (TG) trials were carried out at a nitrogen flow rate of 40 ml/min [3].
2.6. SEM analysis Surface morphology of samples were observed by SEM (JSM−6390LV, JEOL, Japan) at an accelerated voltage of 15−20 kV after gold or platinum coating by a JEOL JFC-1100 E sputtering device for 85 s prior to SEM observation.
2.7. FTIR analysis A FTIR Bio-Rad FTS-3500 spectroscopy was employed in this study. The spectrum was obtained in the transmission mode in 40 0 0−40 0 cm−1 at a resolution of 4 cm−1 with 40 scans per sample.
3. Results and discussion 3.1. Effect of temperature on RDS and lignocellulose content SCW is hot water between 100 and 374 °C using pressure to maintain water in liquid state. Fig. 1. shows that SCW temperature has strong effect on both liquid and solid compositions. RDS content increased from 48.2 mg/g to 111.3 mg/g as temperature was increased from 100 to 170 °C, then decreased to 100.4 mg/g and 68.6 mg/g as temperature was increased to 180 °C and 200 °C, respectively. These results are consistent with that of Xia et al. [2], Pronyk & Mazza [13] and Wei et al. [15] who reported that pretreatment at high temperature decreased sugars concentration due to degradation of sugars into smaller molecules such as furfural, acetic acid, 5-hydroxymethyl-2-furaldehyde (HMF) and formic acid. In lignocellulosic materials, cellulose, hemicellulose and lignin are linked together in tight structure which is difficult to degrade at normal conditions. To release cellulose and hemicellulose, pretreatment is required to break links between cellulose, hemicellulose and lignin. Cellulose and hemicellulose content in the SCW treated water hyacinth in this study are shown in Fig. 1. Cellulose and hemicellulose content increased from 60.3% to 64.5% as temperature was raised from 100 to 155 °C. These results can be attributed to the role of SCW in the digestion of lignocellulosic component of water hyacinth. At 25 °C, the dielectric constant ɛ is 78.5. This value decreases to 55.43 at 100 °C/50 bar and 43.95 at 150 °C/50 bar because hydrogen bonding is weakened by increasing temperature. Hence polarity of water decreases with increasing temperature and water becomes more non-polar and a good solvent for organic compounds (ɛdimethyl sulfoxide = 46.6 at 20 °C, ɛmethanol = 32.6 at 25 °C) [12]. In addition, as temperature increases water density decreases which leads to increase in diffusivity and hence to increase in degradation of cellulose, hemicellulose and lignin in the tight biomass matrix. It was observed
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
ARTICLE IN PRESS
JID: JTICE
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7
3
Table 2 Mass balance of subcritical water pretreatment. No.
Solid used (g)
1 2 3 4 5
0.9995 1.0 0 05 1.0040 1.0 0 0 1.0030
± ± ± ± ±
0.0035 0.0 0 07 0.0014 0.0 0 03 0.0057
Solid remained (g) 0.5103 0.4823 0.4823 0.4823 0.3241
± ± ± ± ±
0.0018 0.0109 0.0 0 09 0.0 0 01 0.0089
Conditions 155 °C, 165 °C, 165 °C, 165 °C, 165 °C,
in this study that as treatment temperature was raised from 100 to 155 °C, cellulose and hemicellulose content was increased from 60.3% to 64.5%, corresponding to a decrease of lignin content from 2.0% to 1.3%. This finding is similar to that of Ciftci & Saldaña [12]. As seen in Table 2., increasing temperature from 155 °C to 165 °C decreased residue which is due to digestion of hemicellulose and lignin leading to the releasing of cellulose (Fig. 1.) which was expelled from the tight lignin and hemicellulose structure. Although the loss and increase of residue were observed, increase in the produced RDS was not fit with calculated number of loss. This was due to generation of components such as furfural and HMF [2,20]. In this study, almost all hemicellulose was degraded as treatment temperatue was raised from 165 °C to 200 °C, therefore, the main compound in residue was cellulose (shown in thermal analysis & Fig. 1.). Moreover, between 165 °C and 200 °C, lignin content in the solid (0.9%) was lower than that treated at temperatures lower than 165 °C which means that lignin disruption was higher. As a result, cellulose content was the highest (∼68.2%) between treatment temperature of 165 °C to 200 °C since cellulose was not tied by linkage with hemicellulose and lignin and expelled from the closed matrix. Wei et al. [15] showed that as treatment temperature was raised from 160 to 200 °C, cellulose content in eucalyptus was enhanced from 44.83% to 53.96%. Xia et al. [2] reported that lignin in water hyacinth slightly decreased from 92% at 120 °C to 78% at temperatures above 160 °C. Moreover, Ando et al. [16] reported that cellulose of bamboo, chinquapin, and Japan cedar started to degrade at temperatures above 230 °C. From the above discussion, in view of cellulose recovery, temperatures higher than 200 °C were not investigated in this study. By using statistical analysis, Fig. 1. shows that there are no significant difference of cellulose content among treatment temperature of 165 °C, 170 °C, 180 °C and 200 °C (P > 0.05), and no significant difference of sugars yield between treatment temperature of 165 °C and 170 °C (P>0.05). Thus, 165 °C is preferable in terms of yield of cellulose and sugars. As expected, at 165 °C the RDS content obtained was only 111.3 mg/g. Whereas, cellulose recovery after treatment was 68.2% which is 131.5% of the untreated sample. Therefore, 165 °C was chosen as the temperature for recovering cellulose in this study. 3.2. Effect of time on RDS and cellulose content As shown in Fig. 2., RDS content increased with treatment time, reaching a maximum of about 111.3 mg/g at 30 to 35 min, then decreased to 82.2 mg/g as time was prolonged to 90 min. The reason is that at 165 °C RDS content mostly were produced from hemicellulose hydrolysis and very little from hydrolysis of cellulose for a treatment time of 15–35 min [2,12,13,15] and [17]. However, sugars were decomposed into smaller molecules at prolonging time which resulted in a decline of sugars yield. This finding is similar to the study on water hyacinth by Xia et al. [2] who reported that total RDS remained constant (19.0 g/100 g) when microwave pretreatment time was between 5 and 25 min, but decreased to 16.0 g/100 g as time was extended to 45 min owning to forma-
30 min, 50 bar, DIW to solid ratio = 10: 1 (ml/g) 30 min, 50 bar, DIW to solid ratio = 10: 1 (ml/g) 30 min, 50 bar, DIW to solid ratio = 20: 1 (ml/g) 30 min, 150 bar, DIW to solid ratio = 10: 1 (ml/g) 50 bar, 30 min, 1% H2 SO4 to dried biomass ratio = 10:1 ml/g)
Fig. 2. Effect of time on cellulose and RDS content of water hyacinth (conditions: water to biomass ratio = 10: 1 ml/g, 165 °C, 50 bar). (a) cellulose & hemicellulose; (b) cellulose.
tion of furfural and HMF by hemicellulose and cellulose decomposition. Similar trend was shown in a study using liquid hot water pretreatment on eucalyptus [15], the highest xylose yield was 13.09 g/100 g raw biomass which was generated after treated at 180 °C for 20 min, but only limited amount of glucose (2.5 g/100 g raw material) was released. The results implied that decomposition of hemicellulose started at lower temperature (180 °C) than that of cellulose in liquid hot water. As pretreatment time was increased from 20 to 30, then to 40 min; the corresponding xylose yield declined from 13.09 to 11.37, then to 9.57 g due to the formation of inhibitors [15]. As shown in Fig. 2., cellulose content reached 68.2% at 30 min and kept almost constant until 90 min. This may be attributed to lignin removal which was significant at 30 min and prolonged time more than 30 min did not remove more lignin. In addition, hemicellulose decomposition started at 15 min and was completed at 30 min, resulting in the highest liberation of cellulose at 30 min. Increasing of time from 30 to 90 min did not improve cellulose content. These results are consistent with that of Xia et al. [2] and Perez et al. [18]. Product distribution can be controlled thermodynamically or kinetically, as indicated by the change of reaction product ratio as temperature changes. Since a temperature change caused a change in the RDS to cellulose ratio (Fig. 9.), it is possible that kinetic controlled biomass degradation. Further proof was given in Fig. 10. which shows that RDS/cellulose ratio increased with increasing temperature (from 170 to 200 °C). Increasing RDS/cellulose ratio reflected different equilibrium constant, which is inconsistent with that of thermodynamic control, which theoretically possesses the same Go over a modest temperature range. Higher RDS in the product at higher temperature was probably due to higher kinetic rate of glucose formation from cellulose. 3.3. Effect of pressure on RDS and cellulose content As shown in Fig. 3., both RDS and cellulose content increased with increasing pressure from 0 to 50 bar, then kept at 68.2% for cellulose yield and 111.3 mg/g for RDS as pressure was increased
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
JID: JTICE 4
ARTICLE IN PRESS
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7
Fig. 3. Effect of pressure on cellulose and RDS content of water hyacinth (reaction conditions: water to biomass ratio = 10: 1 ml/g, 165 °C, 30 min). (a) cellulose & hemicellulose; (b) cellulose.
Fig. 4. Effect of DIW to dried biomass ratio (mL/g) on RDS and cellulose content of water hyacinth (reaction conditions: 30 min, 165 °C, 50 bar). (a) cellulose & hemicellulose; (b) cellulose.
from 80 to 150 bar (P>0.05) since most hemicellulose was removed at 165 °C and 50 bar resulting in the huge increase of cellulose. Pressure higher than 50 bar was not effective for obtaining higher yield of sugars and cellulose since the main role of pressure is to hold water in liquid condition [12,13]. This tendency agrees with that of Hanim et al., [17], who showed that at 190 °C hemicellulose yield in palm increased significantly from 21% to 66% as pressure was raised from 500 psi to 600 psi. Ciftci & Saldaña [12] examined the effect of pressure (50–200 bar) on lignocellulosic degradation by analyzing the phenolic content of lupin hull. In lignocellulosic biomass, hemicelluloses link to lignin by phenolic compounds. SCW treatment breaks such linkages, increasing total phenolic content in the extracts. Their results showed that at 200 bar, phenolic content increased from 0.07 mg gallic acid/g biomass at 160 °C to 0.72 mg gallic acid/g biomass at 220 °C. In contrast, pressure had insignificant effect on total phenolic content and cellulose yield. Cantero et al. [19] reported that pressure did not affect cellulose content at these conditions: 18−27 MPa/350 °C and 10−27 MPa/300 °C. A pressure of 50 bar was chosen in this study. 3.4. Effect of water to dried biomass ratio on RDS and cellulose content The effects of water to solid ratio on cellulose composition and RDS content are shown in Fig. 4. It can be seen in Fig. 4. that both RDS and cellulose content increased as water to biomass ratio was increased from 7:1 to 20:1 (ml/g). However, the increase
Fig. 5. Effect of H2 SO4 concentration on RDS and cellulose content (reaction conditions: 30 min; 165 °C; 50 bar, H2 SO4 to dried biomass ratio = 10:1 ml/g.) (a) cellulose & hemicellulose; (b) cellulose.
of both RDS and cellulose yield as water to biomass ratio was increased from 10:1 to 20:1 is insignificant (P>0.05). This is consistent with mass balance shown in Table 2. which shows that there was no change in the residue weight as water to biomass ratio was increased from 10:1 to 20:1 (ml/g). At a fixed amount of solid (1 g), 7 ml of water was not enough for blending of water and solid resulting in incomplete disruption of biomass. Increasing water amount from 7 ml to 10 ml facilitated water penetration into biomass to break down lignocellulosic structure. As discussed previously, lignin was mostly degraded and hemicellulose was removed completely at 165 °C, 50 bar, 30 min and a water to biomass ratio of 10: 1 ml/g. Under these conditions, the highest recovery of cellulose was obtained. These results are in agreement with that of Xia et al. [2]. Hence the optimal conditions for cellulose recovery in this study were found to be temperature = 165 °C, pressure = 50 bar, time = 30 min and water to dried biomass ratio = 10:1 ml/g. Furthermore, the use of excessive water to penetrate dried water hyacinth structure may reflect the result obtained for the SCW process using fresh water hyacinth. Excessive water may plasticize the lignin matrix structure in the original dried wooden biomass resulting in the same reaction characteristic with fresh biomass, which has been saturated with water [24]. 3.5. Comparison of effectiveness for cellulose recovery between SCW pretreatment and SC H2 SO4 pretreatment As shown in Fig. 5., there was an increase in both cellulose and RDS content as H2 SO4 concentration was raised from 0.25% to 1%. RDS content reached maximum (264.4 mg/g) at 2% H2 SO4 , then dropped to 114.6 mg/g as H2 SO4 concentration was increased to 5% due to the decomposition of sugars into smaller molecules [2,20]. H2 SO4 is able to dissolve hemicellulose at low concentration [2,20], therefore, raising concentration of H2 SO4 from 0.25−1% released cellulose from the confinement of hemicellulose and lignin. However, as H2 SO4 content was raised form 1% to 5%, cellulose content dramatically decreased from 69.4% to 23.9% which may be caused by the disruption of crystalline cellulose at high acid concentration. These results are in agreement with that of Xia et al. [2]. In this study the highest RDS content obtained was only 264.4 mg/g, while cellulose concentration was 69.4% but the cellulose yield was only 89.9%. The low yield is due to the low residue left after treatment, as can be seen in Table 2. In addition, using acid will generate toxic components such as furfural and HMF which are toxic to micro-organisms and requires neutralization of acid [2,20]. Meanwhile cellulose yield obtained by SCW was 131.5%.
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
JID: JTICE
ARTICLE IN PRESS
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7
5
Fig. 6. TGA curves of water hyacinth. (a) untreated sample, (b) treated at 130 °C, 50 bar, 30 min, water/solid (10:1 ml/g), (c) treated at 165 °C, 50 bar, 30 min, treated at 165 °C, 50 bar, 30 min, H2 SO4 1%/solid (10:1 ml/g), (e) treated at 165 °C, 50 bar, 30 min, H2 SO4 5%/solid (10:1 ml/g). water/solid (10:1 ml/g), (d)
3.6. TGA Moisture in water hyacinth evaporated at 30–200 °C. Hemicellulose and cellulose decomposed at 200−350 °C and lignin degradation occurred at 350−470 °C [3], 23]. Fig. 6a indicates that hemicellulose was digested at 265 °C while the main peak at 327 °C is attributed to cellulose degradation. Untreated sample curve (Fig. 6a) shows that hemicellulose peak is the only shoulder in the derivative thermogravimetric curve which means that hemicellulose is the minor component in water hyacinth. The peak gradually disappeared with increasing treating temperature from 100 to 155 °C and disappeared at 165 °C. This observation was similarly found for the H2 SO4 (0.25−3%) treated sample. These results imply that hemicellulose was completely degraded at 165 °C by SCW and H2 SO4 (0.25−3%). However, cellulose peak significantly decreased at high H2 SO4 concentration (4−5%) indicating that cellulose was decomposed at high acid concentration. The small peak appeared at 468 °C was attributed to lignin. This observation agrees with that of Harun [3] and Sundari [23]. 3.7. SEM Fig. 7a shows surface of the untreated water hyacinth. The surface is smooth and randomly tight. The structure was partly broken after SCW treatment at 130 °C and 50 bar for 30 min (Fig. 7b). Nevertheless, the overall structure of lignocellulose was still intact. This finding indicates that SCW at low temperature cannot break the intact structure of lignin and hemicellulose which encloses cellulose. Increasing temperature to 165 °C dramatically degraded water hyacinth. The surface became more crumbled, bunches of fibers were broken revealing that the compact lignocellulose structure was significantly degraded leading to release of cellulose from the tight enclosure of hemicellulose and lignin (Fig. 7c). Therefore, cellulose yield of SCW pretreatment at 165 °C was higher than that at other temperatures. This was similarly detected for H2 SO4 (0.25%) treated sample (Fig. 7d). It seems that there is an agglomeration on the surface of H2 SO4 (5%) treated sample. Fig. 7e shows that the surface was smoothened because of serious collapse of structure caused by using higher concentration H2 SO4 . As a result, agglomeration of lignocellulose structure was formed. 3.8. FTIR Fig. 8 shows FTIR spectra of the untreated and the treated water hyacinth samples. In the untreated sample, the peaks observed at
Fig. 7. SEM images of water hyacinth. (a) untreated sample, (b) treated at 130 °C, 50 bar, 30 min, water/solid (10:1 ml/g), (c) treated at 165 °C, 50 bar, 30 min, water/solid (10:1 ml/g), (d) treated at 165 °C, 50 bar, 30 min, H2 SO4 1%/solid (10:1 ml/g), (e) treated at 165 °C, 50 bar, 30 min, H2 SO4 5%/solid (10:1 ml/g).
1249 and 1379 cm−1 are those of C–O stretching of syringyl lignin and C–O–C of hemicellulose oscillation in anomeric region, respectively [4,21]. The wave at 1735 cm−1 is originated from acetyl and uronic ester groups of hemicelluloses [22,23]. The peaks at 1249 cm−1 and 1735 cm−1 disappeared for the 165 °C SCW treated sample indicating that pretreatment removed hemicellulose and partial lignin. For acid treated samples, bands at 1249 and 1735 cm−1 are not seen in the curves which means that pretreatment by acid under the studied subcritical conditions
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
JID: JTICE 6
ARTICLE IN PRESS
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7
tions are in agreement with Abral [21,22,23] and in line with TGA and SEM results. 4. Conclusions A cellulose content of 69.4% (cellulose yield 89.9%) was obtained from pretreatment using 1% H2 SO4 under conditions of 165 °C, 50 bar, a water to dried biomass ratio of 10 ml/g and a treatment time of 30 min. Besides, using acid has several disadvantages as mentioned previously. Meanwhile, SCW pretreatment is not only an environmentally friendly method but also generated high cellulose (68.2%) which is 131.5% of the untreated sample. Functional groups and morphology of all samples were inspected by FTIR, SEM and TGA. Results indicate degradation of lignocellulosic materials which explains the enhancement of cellulose content. These results are promising for ethanol production, therefore, ethanol production will be investigated in the subsequent study. References Fig. 8. FTIR spectra of untreated and treated water hyacinth.
Fig. 9. RDS to cellulose ratio vs. time. Reaction conditions: water to biomass ratio = 10: 1 ml/g, 165 °C, 50 bar
Fig. 10. RDS to cellulose ratio vs. temperature. Reaction conditions: water to biomass ratio = 10: 1 ml/g, 50 bar, 90 min.
probably dissolved hemicellulose and partial lignin. However, high concentration of sulfuric acid also disrupted cellulose, therefore, cellulose content in the H2 SO4 treated sample sharply dropped from 69.4% to 23.9% with increasing acid concentration from 1% to 5%. The result was supported by the disappearance of peak at 808 cm−1 in the 5% H2 SO4 treated sample which corresponds to the C–H rocking vibration of cellulose. The peaks at 3408 cm−1 and 2985 cm−1 are –OH and –CH stretching vibrations. These observa-
[1] Bayrakci AG, Kocar G. Second-generation bioethanol production from water hyacinth and duckweed in Izmir: a case study. Renew Sustain Energy Rev 2014;30:306–16. [2] Xia A, Cheng J, Song W, Yu C, Zhou J, Cen K. Enhancing enzymatic saccharification of water hyacinth through microwave heating with dilute acid pretreatment for biomass energy utilization. Energy 2013;61:58–166. [3] Harun MY, Radiah ABD, Abidin ZZ, Yunus R. Effect of physical pretreatment on dilute acid hydrolysis of water hyacinth (Eichhornia crassipes). Bioresour Technol 2011;102:5193–9. [4] Ahmed IN. Bioethanol production from Melaleuca leucadendron shedding bark: assessing pretreatments. Solid and enzyme loading; 2013. PhD thesis p. 8–9. [5] Cheng J, Lin R, Song W, Xia A, Zhou J, Cen K. Enhancement of fermentative hydrogen production from hydrolyzed water hyacinth with activated carbon detoxification and bacteria domestication. Int J Hydrogen Energy 2015;40:2545–51. [6] Yan J, Wei Z, Wang Q, He M, Li S, Irbis C. Bioethanol production from sodium hydroxide/hydrogen peroxide pretreated water hyacinth via simultaneous saccharification and fermentation with a newly isolated thermotolerant Kluyveromyces marxianu strain. Bioresour Technol 2015;193:103–9. [7] Guragain YN, Coninck JD, Husson F, Durand A, Rakshit SK. Comparison of some new pretreatment methods for second generation bioethanol production from wheat straw and water hyacinth. Bioresour Technol 2011;102:4416–24. [8] Singh A, Bishnoi NR. Ethanol production from pretreated wheat straw hydrolyzate by Saccharomyces cerevisiae via sequential statistical optimization. Ind Crops Prod 2013;41:221–6. [9] Aswathy US, Sukumaran RK, Devi GL, Rajasree KP, Singhania RR, Pandey A. Bio-ethanol from water hyacinth biomass: an evaluation of enzymatic saccharification strategy. Bioresour Technol 2010;101:925–30. [10] Gao J, Chen L, Yan Z, Wang L. Effect of ionic liquid pretreatment on the composition, structure and biogas production of water hyacinth (Eichhornia crassipes). Bioresour Technol 2013;132:361–4. [11] Ma F, Yang N, Xu C, Yu H, Wu J, Zhang X. Combination of biological pretreatment with mild acid pretreatment for enzymatic hydrolysis and ethanol production from water hyacinth. Bioresour Technol 2010;101:9600–4. [12] Ciftci D, Saldaña MDA. Hydrolysis of sweet blue lupin hull using subcritical water technology. Bioresour Technol 2015;194:75–82. [13] Pronyk C, Mazza G. Kinetic modeling of hemicellulose hydrolysis from triticale straw in a pressurized low polarity water flow-through reactor. Ind Eng Chem Res 2010;49:6367–75. [14] Miller GL. Use of dinitrosalicylic acid reagent for the determination of reducing sugars. Anal Chem 1959;31:426–8. [15] Wei W, Shubin W, Liguo L. Combination of liquid hot water pretreatment and wet disk milling to improve the efficiency of the enzymatic hydrolysis of eucalyptus. Bioresour Technol 2013;128:725–30. [16] Ando H, Sakaki T, Kokusho T, Shibata M, Uemura Y, Hatate Y. Decomposition behavior of plant biomass in hot-compressed water. Ind Eng Chem Res 20 0 0;39:3688–93. [17] Hanim SS, Norsyabilah R, Suhaila MHN, Noraishah A, Kartina AKS. Effects of temperature, time and pressure on the hemicelluloses yield extracted using subcritical water extraction. Proc Eng 2012;42:562–5. [18] Pérez JA, Ballesteros I, Ballesteros M, Negro MJFS, Manzanares P. Optimizing liquid hot water pretreatment conditions to enhance sugar recovery from wheat straw for fuel-ethanol production. Fuel 2008;87:3640–7. [19] Cantero DA, Tapia AS, Bermejo MD, Cocero MJ. Pressure and temperature effect on cellulose hydrolysis in pressurized water. Chem Eng J 2015;276:145–54.
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028
JID: JTICE
ARTICLE IN PRESS
[m5G;January 10, 2017;18:51]
B.T. Nguyen Thi et al. / Journal of the Taiwan Institute of Chemical Engineers 000 (2017) 1–7 [20] Ganguly A, Chatterjee PK, Deyb. Studies on ethanol production from water hyacinth—A review. Renew Sustain Energy Rev 2012;16:966–72. [21] Abral H, Kadriadi D, Rodianus A, Mastariyanto P, Ilhamdi, Arief S, Sapuan SM, et al. Mechanical properties of water hyacinth fibers – polyester composites before and after immersion in water. Mater Des 2014;58:125–9. [22] Fang JM, Sun RC, Tomkinson J. Isolation and characterization of hemicelluloses and cellulose from rye straw by alkaline peroxide extraction. Cellulose 20 0 0;7:87–107.
7
[23] Sundari MT, Ramesh A. Isolation and characterization of cellulose nanofibers from the aquatic weed water hyacinth—Eichhornia crassipes. Carbohydr Polym 20 0 0;87:1701–5. [24] Ibbett R, Gaddipati S, Davies S, Hill S, Tucker G. The mechanisms of hydrothermal deconstruction of lignocellulose: new insights from thermal–analytical and complementary studies. Bioresour Technol 2011;102:9272–8.
Please cite this article as: B.T. Nguyen Thi et al., Effect of subcritical water pretreatment on cellulose recovery of water hyacinth
(Eichhornia crassipe ), Journal of the Taiwan Institute of Chemical Engineers (2017), http://dx.doi.org/10.1016/j.jtice.2016.12.028