Waste Biomass Valor (2010) 1:395–405 DOI 10.1007/s12649-010-9045-3
ORIGINAL PAPER
Production of Solid Biofuel from Agricultural Wastes of the Palm Oil Industry by Hydrothermal Treatment Ahmad T. Yuliansyah • Tsuyoshi Hirajima Satoshi Kumagai • Keiko Sasaki
•
Received: 24 April 2010 / Accepted: 25 September 2010 / Published online: 21 October 2010 Ó Springer Science+Business Media B.V. 2010
Abstract In this study, upgrading of agricultural waste, in the form of oil palm fronds and trunks, into solid biofuel was investigated using hydrothermal treatment. A slurry of 300 mL of water and 30 g of material was treated in a 500-mL batch autoclave equipped with stirrer, thermometer, and pressure sensor. Experiments were conducted in the temperature range 200–350°C at an initial pressure of 2.0 MPa. The slurry was gradually heated to the target temperature and held for a further 30 min. Approximately 35–65% of the original material was recovered as a solid product with favorable solid fuel characteristics. The gross calorific value ranged from 19.9 to 29.7 MJ/kg and the equilibrium moisture content was 7.6–4.5 wt%. The carbon content varied from 51.4 to 78.5 wt% and the oxygen content was 42.1–16.1 wt% after upgrading. Changes in the solid composition and carbon functional groups following upgrading were identified by FTIR and 13C NMR. In addition, analyses on the liquid product (by GC–MS) and the gas product (by GC) were carried out to clarify the decomposition behavior of material. Keywords Upgrading Agricultural waste Biomass Hydrothermal treatment Solid biofuel
A. T. Yuliansyah—On leave from Department of Chemical Engineering, Gadjah Mada University, Indonesia. A. T. Yuliansyah T. Hirajima (&) K. Sasaki Department of Earth Resources Engineering, Faculty of Engineering, Kyushu University, 744 Motooka, Nishi-ku, Fukuoka 819-0395, Japan e-mail:
[email protected] S. Kumagai Research and Education Center of Carbon Resources, Kyushu University, 744 Motooka, Nishiku, Fukuoka 819-0395, Japan
Introduction The rapid increase in energy consumption within the last few years, combined with the steady depletion of fossil fuel reserves, has prompted a global search for alternative energy resources. Biomass is a promising alternative owing to its sustainability and environmental friendliness. Various types of biomass products are available in large quantities and have potential for further utilization. One potential biomass source that is abundant in most tropical countries is oil palm waste. This waste can be categorized into two types: waste from harvesting and replanting activity in plantation fields and waste from the milling process to obtain palm oil. Numerous studies on utilization of milling waste (oil palm fiber, shell, and empty fruit bunches) have been conducted for energy applications [1–3], pulp and papermaking [4–6], bioadsorbents [7–10], construction materials [11, 12], and biocomposites [13–15]. However, most of the milling waste is fully utilized by the palm oil industry either as an additional energy source (oil palm fiber and shell) or as fertilizer (empty fruit bunches). By contrast, only a few studies have considered the utilization of harvesting waste. Currently, most of this waste is used conventionally as an organic fertilizer in plantation fields. However, unpleasant smells coupled with a slow release of CO2 and CH4 gas on decomposition, which can last for up to 1 year, are common problems. Nevertheless, oil palm fronds and trunks have great potential. In 2005, approximately 43.05 million ton of frond and 13.94 million ton of trunk wastes were generated in Indonesia, the largest crude palm oil producer in the world. This amount will continue increase with the rapid growth in the Indonesian palm oil industry. Thus, a better method to manage such wastes is highly desired.
123
396
Waste Biomass Valor (2010) 1:395–405
Various treatment methods for several agricultural wastes have been reviewed in several papers [16–19]. For energy application purposes, the methods can be simply divided into thermo-chemical and biological processes [20, 21]. Hydrothermal treatment is one of thermo-chemical processes have attracted much attention recently. In this process, biomass is treated in hot compressed water yielding in gases, aqueous organics and upgraded solid [22, 23]. After filtered and dried, the obtained solid can be used for solid fuel due to its high calorific value. Compared to other thermo-chemical conversion methods such as pyrolysis and gasification, the temperature for hydrothermal treatment is much lower (200–350°C for hydrothermal, compared with 450–550°C for pyrolysis and 900–1200°C for gasification) [16, 24]. In addition, biomass conversion takes place in a wet environment so high moisture content of feed biomass is not an issue. The role of water in the treatment is not only as a medium, but also a chemical reactant on decomposition. Therefore, such method is suitable for treating biomass with high moisture content, such as agricultural wastes which contain more than 50 wt. % of moisture in fresh condition. A contrast situation is found on pyrolysis and gasification which have a limitation on moisture content of the feed [24, 25]. Many studies using hydrothermal treatment have been conducted, but most of these used the method as a biomass Table 1 Composition of the raw materials Component
Frond
Trunk
Cellulose (wt. %, d.b)
31.0
39.9
Hemicellulose (wt. %, d.b)
17.1
21.2
Klason lignin (wt. %, d.b)
22.9
22.6
Wax (wt. %, d.b)
2.0
3.1
Ash (wt. %, d.b)
2.8
1.9
24.2
11.3
Others
pretreatment step in bio-ethanol production [26–29]. Few studies have considered benefits of the resulting solid. The focus of the present study was upgrading of solid material into solid biofuel by hydrothermal treatment. Therefore, we evaluated the feasibility of upgrading oil palm fronds and trunks and investigated their decomposition behavior during hydrothermal treatment.
Experimental Materials Oil palm waste in the form of fronds and trunks was collected from an oil palm plantation in southern Sumatra, Indonesia. Both raw fronds and trunks were chipped into pieces of approximately 2 cm in width of slabs. Prior to use, the chips were ground using a cutting mill to form powder with a maximum particle size of 1 mm. The composition of the waste material is listed in Table 1. Apparatus and Experimental Procedure Experiments were carried out in a 500-mL batch-type autoclave (Taiatsu Techno MA 22) equipped with a stirrer and an automatic temperature controller (Fig. 1). The autoclave had a maximum temperature of 400°C and a maximum pressure of 30 MPa. A slurry of 300 mL of water and 30 g of waste material was loaded into the autoclave. A stream of N2 gas was used to purge air from the autoclave and to maintain an initial internal pressure of 2.0 MPa. With stirring at 200 rpm, the autoclave was heated to the target temperature at an average rate of 6.6°C/min. The target temperature, ranging from 200 to 350°C, was automatically adjusted. Once the target temperature was reached, the sample was held for a further 30 min before the autoclave was cooled to ambient
Fig. 1 Schematic diagram of the experimental apparatus
Cooling water in
Cooling water out
PS TS
N2
Control Board
Vessel
PS = Pressure Sensor TS = Temperature Sensor
123
Waste Biomass Valor (2010) 1:395–405
397
300) with the following conditions: 10,000 scans; contact time, 2 ms; spinning speed,[12 kHz; pulse repetition time, 7 s. The spectrum was calibrated using hexamethyl benzene. Curve fitting analysis of the spectrum was performed using Grams/AI 32 ver. 8.0 software. Composition of the liquid products were analyzed by gas chromatography— mass spectrophotometer (GC–MS) Agilent 6890 N equipped with Jeol JMS-Q1000GC(A) system. Additionally, total organic carbon content of liquid was determined on a Shimadzu TOC-5000A instrument. Solid yield, energy densification ratio, and energy yield are three important parameters in this study which are defined as [31]: Fig. 2 Pressure and temperature profile for experiments (1 200°C; 2 240°C; 3 270°C; 4 300°C; 5 330°C; 6 350°C)
conditions. Pressure and temperature profile for the experiments was described in Fig. 2. After cooling, the gas products were fed into a gasometer (Shinagawa DC-1) to measure the volume. The gas was sampled using a microsyringe (ITO MS-GANX00) and its composition was determined by gas chromatography with thermal conductivity detection (Shimadzu GC-4C). The remaining slurry was filtered using an ADVANTEC 5C filter and a water aspirator. The solid part was dried in an oven at 105°C to yield the final solid product. Analysis The solid products were characterized using several techniques. The elemental composition was measured using Yanaco CHN Corder MT-5 and MT-6 elemental analyzer. The cellulose, hemicellulose, and lignin contents were measured using a procedure recommended by the US National Renewable Energy Laboratory [30] that is substantially similar to ASTM E1758-01. The mineral composition in ash was determined using X-ray fluorescence (XRF) on Rigaku ZSX Primus II equipment. In addition, proximate, total sulfur and gross calorific value (GCV) analyses were carried out according to JIS M 8812, JIS M 8819, and JIS M 8814, respectively. The equilibrium moisture content (EMC) of raw frond and trunk and the corresponding solid products was determined according to JIS M 8811. An aliquot of the sample was placed in a desiccator containing saturated salt solution at a constant relative humidity (75% RH). After equilibrium was reached, the moisture content of the solid was quickly measured using a Sartorius MA 150 analyzer. Identification of the chemical structure and functional groups was performed on a Fourier transform infrared (FTIR) spectrometer (JASCO 670 Plus) using the KBr disk technique. Cross polarization/ magic angle spinning (CP/MAS) 13C NMR spectra were measured on a solid-state spectrophotometer (JEOL CMX-
Solid yield ¼ mass of dried solid product= mass of dried feed material 100% Energy densification ratio ¼ GCV of product= GCV of feed material Energy yield ¼ solid yield energy densification ratio .
Results and Discussion Product Distribution Hydrothermal treatment led to thermal degradation of the feed material. Physical and chemical bonds in the material were broken, so that large long-chain compounds such as cellulose, hemicellulose and lignin were broken down into smaller and simpler molecules. Furthermore, some of the molecules were dissolved into liquid part and some others were degraded to gases. The remainder of feed material was recovered as a solid residue. Filtration and drying yielded a solid product from this residue. Figure 3 shows the distribution of organic compounds, represented by percentage carbon, in the gas, liquid, and solid phases. However, a small amount of material was not recovered during filtration and drying. The results indicate that most of the carbon remained in the solid phase and that the relative amount gradually decreased as the temperature increased. Conversely, the proportion of carbon in the gas phase steadily increased with temperature. For the liquid phase, a unique characteristic was observed. The relative amount of carbon increased with temperature to a maximum at *270°C and then decreased owing to polymerization of soluble compounds to produce solid precipitates. Solid products The properties of solid products for different temperature treatments are described in Table 2. Reaction temperature
123
398
Fig. 3 Distribution of carbon in the hydrothermal product a frond; b trunk
had a significant effect to the solid yield, energy densification ratio and energy yield of the solid products. Increasing temperature led to a decrease in solid yield and energy yield, while energy densification ratio increased. For the same reaction temperature, the solid product of trunk had a higher solid yield and energy yield than frond. However, the differences were smaller at higher temperature. At 200°C, the solid yields were 58.3 and 67.8 wt%, while the energy yields were 63.5 and 71.7%, for frond and trunk, respectively. The solid yield reduced to 35.1 and 35.3 wt%, while the energy yield decreased to 55.4 and 55.7% at 350°C treatment. Like other biomass materials, both fronds and trunks have very high volatile content of 82.5 and 83.9 wt%, in contrast to the low fixed carbon of 17.5 and 16.1 wt%, respectively. Progressive decomposition reactions occurred at higher temperature, leading to an increase in fixed carbon content and a decrease in volatile content. Treatment at 350°C increased the fixed carbon content to 54.8 and 55.0 wt% and decreased the volatile content to 45.2 and 45.0 wt% for fronds and trunks, respectively. This led to an increase in gross calorific value of the solid product.
123
Waste Biomass Valor (2010) 1:395–405
The calorific value is correlated with the elemental composition of a solid. The data in Table 2 show that an increase in calorific value is correlated with an increase in carbon content and a decrease in oxygen content. Compared to the raw material, the solid produced at 350°C has *59% higher carbon content and *58% lower oxygen content. These results suggest that components degraded and removed from the material were mainly oxygen-rich compounds. Dramatic changes in the calorific value and elemental composition took place in the temperature range 200–270°C. Due to thermal degradation, the content of cellulose, hemicellulose, and lignin in the solid decreased. Table 3 shows the percentage of these components in the solid products after treatment at 200–300°C. The data suggest that hemicellulose and cellulose were relatively easier to degrade than lignin. The treatment significantly degraded both hemicellulose and cellulose to produce a more ligneous solid. Slightly different behavior was observed for hemicellulose decomposition between frond and trunk material. The frond solid produced at 200°C still had a small amount of hemicellulose, which completely vanished on treatment at 240°C. On the other hand, no hemicellulose was found for the trunk products, even for treatment 200°C. This suggests that hemicellulose decomposition started at temperatures \200°C. By contrast, cellulose was gradually degraded at higher temperature and \0.5 wt% (on a solid product basis) remained after treatment at 270°C. This behavior is in agreement with previous reports [22, 32]. The obtained solids contained a small amount of ash ranging from 0.7 to 2.2 wt% (Table 2). The use of these low-ash products as fuel will be beneficial because it will reduce potential for solid deposition on burner equipment that commonly found on combustion of high-ash fuel. Table 4 shows ash analysis of frond and trunk solid products. As can be seen from this table, CaO and SiO2 were two major oxides accounted for 28.80–51.30 and 16.40–39.10 wt% for frond, and 15.20–27.30 and 22.80–44.30 wt% for trunk, respectively. The data in table 4 also indicates that some minerals such as CaO, MgO and K2O were selectively leached from material during treatment. Coal Band Under hydrothermal treatment, frond and trunk materials undergo a coalification-like process, as demonstrated in the Van Krevelen diagram in Fig. 4. The raw materials have high atomic H/C and O/C ratios, which both gradually decreased during treatment. The slope of the trajectories suggests that the O content decreased in proportion to the H content, probably due to dehydration. It is clear that the
Waste Biomass Valor (2010) 1:395–405
399
Table 2 Properties of the raw materials and the solid products Properties
Raw
Treated temperature (°C) 200
240
270
300
330
350
(a) Frond Proximate analysis (wt.%) Fixed carbon (d.a.f)
17.5
20.5
29.7
45.9
48.1
52.3
54.8
Volatile matter (d.a.f)
82.5
79.5
70.3
54.1
51.9
47.7
45.2
1.8
1.3
1.3
1.2
1.0
0.7
0.8
14.7
7.6
6.4
5.5
5.2
5.0
5.2
47.2
53.6
58.6
69.4
71.1
73.9
75.1
H
5.9
5.7
5.4
4.9
4.9
4.9
4.8
N
0.2
0.2
0.3
0.4
0.4
0.4
0.4
46.6
40.4
35.7
25.3
23.5
20.7
19.5
Ash (d.b) Equilibrium moisture Ultimate analysis (wt.%) (d.a.f) C
O (diff) S
0.1
Yield of solid product (wt.%) (d.b) Gross calorific value (MJ/kg) (d.a.f)
18.8
Energy densification ratio
0.1
0.1
0.1
0.1
0.1
0.1
58.3
52.0
42.5
38.4
36.7
35.1
20.5
23.0
26.7
27.3
29.0
29.7
1.09
Energy yield (%)
1.22
1.42
1.45
1.54
1.58
63.5
63.6
60.3
55.9
56.5
55.4
(b) Trunk Proximate analysis (wt.%) Fixed carbon (d.a.f)
16.1
16.2
26.6
45.1
48.8
52.8
55.0
Volatile matter (d.a.f)
83.9
83.8
73.4
54.9
51.2
47.2
45.0
2.2
1.8
1.8
2.2
2.1
1.9
2.1
13.6
7.5
6.5
5.1
4.8
4.6
4.5
C
47.5
51.4
57.5
69.3
71.4
73.4
75.3
H
5.9
5.9
5.6
5.1
5.0
4.9
4.9
N
0.5
0.4
0.6
0.8
0.8
1.0
1.0
45.9
42.1
36.2
24.6
22.6
20.6
18.6
Ash (d.b) Equilibrium moisture Ultimate analysis (wt.%) (d.a.f)
O (diff) S
0.1
Yield of solid product (wt.%) (d.b) Gross calorific value (MJ/kg) (d.a.f)
18.8
Energy densification ratio
0.1
0.1
0.1
0.1
0.1
0.2
67.8
56.9
41.7
38.7
36.9
35.3
19.9
22.6
27.0
28.0
29.4
29.7
1.06
Energy yield (%)
71.7
1.20 68.3
1.44 60.0
1.49 57.8
1.57 57.9
1.58 55.7
d.b dry basis, d.a.f dry ash free basis, diff. differences Table 3 Percent component in the solid products Product
Cellulose
Hemicellulose
Lignin & other
200°C
55.1
3.2
41.7
240°C
42.3
0.0
57.7
270°C
0.2
0.0
99.8
300°C
0.0
0.0
100.0
200°C
62.9
0.0
37.1
240°C
44.8
0.0
55.1
270°C
0.4
0.0
99.6
300°C
0.0
0.0
100.0
Frond
decrease in O and H content occurred mainly in the range 200–270°C. Less significant changes were observed at higher temperature. The solids resulting from higher temperature treatment had comparable compositions with typical solid fuels such as sub-bituminous coal. Although fronds and trunks have different raw compositions, the products after treatment at C300°C had almost identical compositions.
Trunk
Fourier Transform Infrared (FTIR) Analysis To understand changes in functional groups during hydrothermal treatment, FTIR analysis of the products was
123
400
Waste Biomass Valor (2010) 1:395–405
Table 4 Mineral composition of ash in the raw materials and the solid products Solid products
Oxide (wt.%) CaO
SiO2
MgO
Raw
52.70
12.20
16.00
200°C
51.30
28.70
6.04
240°C
39.50
32.90
270°C
28.80
39.10
300°C
30.80
330°C 350°C
K2O
Fe2O3
Al2O3
ZnO
SO3
P2O5
Others
8.29
0.39
0.21
0.05
7.21
2.43
0.52
3.66
1.41
0.79
0.06
5.78
1.10
1.16
7.52
4.36
1.97
1.49
0.07
8.31
2.54
1.34
7.60
4.64
1.22
1.61
0.05
12.40
2.95
1.63
29.60
6.95
5.50
2.19
2.17
0.05
15.20
5.36
2.18
37.10 36.50
16.60 16.40
8.49 8.51
6.03 5.71
2.54 2.96
2.62 2.67
0.07 0.08
16.30 16.00
8.14 7.83
2.11 3.34
Raw
23.50
14.90
15.60
24.50
0.71
0.25
0.12
10.70
8.71
1.00
200°C
18.80
39.80
8.37
13.20
1.14
0.67
0.11
9.54
7.27
1.10
240°C
15.20
44.30
5.83
9.21
1.43
0.76
0.10
9.23
13.10
0.84
270°C
16.40
44.00
4.64
7.38
1.21
0.83
0.06
9.95
14.80
0.73
300°C
18.20
41.40
4.64
5.38
1.69
1.49
0.06
7.97
18.20
0.97
330°C
27.30
22.80
5.05
6.76
2.57
1.22
0.15
9.64
23.00
1.51
350°C
22.10
22.80
11.80
5.16
3.13
0.76
0.12
9.33
18.40
6.41
(a) Frond
(b) Trunk
represents carbonyl (C=O) stretching vibrations. The peak at *1515 cm-1 reveals to aromatic skeletal vibration derived from lignin. The peak at *1050 cm-1 attributed to glycosidic bonds, indicating the presence of cellulose, steadily weakened and completely disappeared for temperatures [270°C, indicating that cellulose was totally degraded at this temperature. The decrease in intensity for C–O–C aryl–alkyl ether linkages at *1230 cm-1 suggests lignin decomposition. Conversely, solids derived from polymerization of intermediate compounds in the liquid phase increased the aromatic content, particularly at temperatures [300°C, as indicated by the increase in intensity for the peak at 1600 cm-1 attributed to aromatic skeletal vibrations and CO stretching. Fig. 4 Van Krevelen diagram for the solid products obtained at different temperatures in comparison with other solid fuel (1 raw material; 2 200°C; 3 240°C; 4 270°C; 5 300°C; 6 330°C; 7 350°C; a frond; b trunk)
13
13
performed. Peaks were assigned based on literature data [3, 33]. Figure 5 shows spectra of the raw materials and the corresponding solid products. The spectra for frond and trunk materials were similar. The intensity of the peak *3500 cm-1 attributed to –OH groups decreased at elevated temperature, indicating that water molecules within the solids were gradually expelled. In other words, dehydration of the feed material occurred. The peak at *2900 cm-1 attributed to aliphatic CHn groups also weakened, indicating that several long aliphatic chains were broken down. More distinctive peaks were observed in the region below 2000 cm-1. The peak at *1700–1740 cm-1
123
C Nuclear Magnetic Resonance (NMR) Results
C NMR measurements were conducted to complement FTIR in characterizing the molecular structure of the solid products. NMR is useful for making comparisons without the need for peak ratios. Each resonance peak can be measured relative to the total resonance intensity to give the relative amount of individual molecular groups. Typical 13C NMR spectra for raw biomass with peak assignment can be found in the literature [34–36]. In brief, resonance peaks in spectra for raw frond and trunk material were assigned to CH3 in acetyl groups (21 ppm), methoxyl groups in lignin (56 ppm), C-6 carbon atoms in cellulose (62–65 ppm), C-2/ C-3/C-5 atoms in cellulose (72–75 ppm), C-4 atoms in cellulose (84–89 ppm), C-1 atoms in hemicellulose (102 ppm),
Waste Biomass Valor (2010) 1:395–405
401
Fig. 6 13C NMR spectra with curve fitting for raw frond and products obtained at different temperatures
Fig. 5 FTIR spectra for the raw materials and the corresponding products a frond; b trunk
C-1 atoms in cellulose (105 ppm), unsubstituted olefinic or aromatic carbon atoms (110–127 ppm), quaternary olefinic or aromatic carbon atoms (127–143 ppm), olefinic or aromatic carbon atoms with OH or OR substituents (143–167 ppm), esters and carboxylic acids (169–195 ppm) including acetyl groups in hemicellulose (173 ppm), and carbonyl groups in lignin (195–225 ppm). Despite the various resonance peaks observed, for semi-quantitative analysis the spectra can be simply classified into aliphatic (0–59 ppm), carbohydrate (59–110 ppm), aromatic (110–160 ppm), carboxyl (160–188 ppm), and carbonyl regions (188–225 ppm) [35, 36]. Data for raw and treated fronds reveal that the peak resonance for hemicellulose and cellulose progressively decreased (Fig. 6). A similar pattern was observed for trunk spectra (Fig. 7). Products resulting from treatment at 200 to 240°C exhibited identical spectra to that of the raw material. However, treatment at 270°C led to extreme spectral changes to a more aromatic nature. The relative
aromatic content, which correlates with the lignin content, consistently increased in the range 200–270°C, whereas the carbohydrate content (hemicellulose and cellulose) decreased. This is in good agreement with the component analysis, which suggested that lignin was the predominant component for treatment at C270°C (Table 3). Equilibrium Moisture Content (EMC) Analysis Hydrothermal treatment greatly reduced the EMC of materials. Treatment at 200°C reduced the EMC from 14.7 to 7.6 wt% for fronds and from 13.6 to 7.5 wt% for trunks. Further treatment at 350°C led to EMC as low as 5.2 and 4.5 wt% for fronds and trunks, respectively. However, the decrease in EMC mainly occurred in the range 200–270°C, with only small changes observed at higher temperatures. These results are in agreement with the changes in solid components shown in Fig. 3. Based on the component characteristic on water adsorption, hemicellulose exhibits the strongest water adsorption, followed by cellulose and lignin [37]. Since hemicellulose was removed first from the solid at low temperature, it is reasonable that the EMC of the material dramatically decreased in this range. By contrast, solids with high lignin content adsorb only a small amount of moisture. The EMC results were confirmed by NMR
123
402
Waste Biomass Valor (2010) 1:395–405
Fig. 8 Relationship between percentage aromatic carbon and equilibrium moisture content
Fig. 7 13C NMR spectra with curve fitting for raw trunk and products obtained at different temperatures
results demonstrating an increase in aromatic content in the solid material. The presence of aromatic compounds, which are hydrophobic, results in resistance to humidity and water adsorption from air. Therefore, a higher aromatic content is correlated with lower EMC. The relationship between the relative amount of aromatic carbon and the EMC is presented in Fig. 8. EMC and calorific value are two important properties of solid fuels. When material is burned, some of the energy released by combustion is consumed to vaporize the water contained in the material. Material with a higher EMC will require more energy for moisture evaporation. Thus, a good solid fuel should have a high calorific value and a low EMC. Our experiments demonstrated that both properties were improved by hydrothermal treatment. Although combustion of the solid fuel products totally produces less energy than that of raw materials, the fuels offer other benefits. They are more resistant to undesired biological decomposition when they are kept in storage. In addition, as solid fuel which may be transported from one location to other, their higher energy density will affect on reduction of storage cost, as well as transportation cost. Liquid Product Under hydrothermal treatment, materials were degraded into numerous low-molecular weight compounds that subsequently leached into liquid. Figure 9 shows GC–MS
123
Fig. 9 GC-MS spectrogram for the liquid products obtained at various temperatures a frond; b trunk
chromatogram of the liquid product of frond and trunk, respectively. As indicated by those chromatograms, hydrothermal liquid product contained various organic compounds. Table 5 lists peaks for several major
Waste Biomass Valor (2010) 1:395–405 Table 5 Peak for identified compounds in GC-MS chromatograms
403
No. Peak
Retention time (min)
Compounds
Chemical formula
Molecular weight
1
00:04:50
Methanol
CH4O
32
2
00:05:45
Ethanol
C2H6O
46
3
00:06:16
Aceton
C3H6O
58
4
00:09:24
Acetic acid
C2H4O2
60
5
00:10:21
1-hydroxy 2-propanon
C3H6O2
74
6
00:10:59
Propanoic acid
C3H6O2
74
7
00:12:11
Cyclopentanone
C5H8O
84
8
00:12:57
Furfural
C5H4O2
96 82
9
00:13:00
2-cyclopenten-1-one
C5H6O
10
00:13:55
2-cyclopenten-1-one, 2-methyl
C6H8O
11 12
00:14:25 00:15:25
4-oxopentanethioic acid Phenol
C5H8O2S C6H6O
132 94
13
00:16:16
2-methoxy phenol
C7H8O2
124
96
14
00:17:15
3-pyridinol
C5H5NO
95
15
00:18:20
1,2 benzenediol
C6H6O2
110
16
00:18:21
5-HMF
C6H6O3
126
17
00:19:04
2,6-dimethoxy phenol
C8H10O3
154
18
00:19:43
2-methyl 1,3-benzenediol
C7H8O2
124
19
00:20:03
4-methoxy-3-(methoxymethyl) phenol
C9H12O3
168
compounds identified on GC/MS chromatogram. The chromatograms show that acetic acid and furfural were two predominant organics obtained at 200°C, followed by 5HMF, phenol, and 1-hydroxy 2-propanone. The intensity of acetic acid and phenol tended to increase at elevated temperature. However, peaks revealed furfural and 5-HMF, secondary decomposition products of hemicellulose and cellulose, were present only at 200–270°C. These data confirmed the results of chemical analysis and 13C NMR of solid that suggest a progressive decomposition of hemicellulose and cellulose took place within 200–270°C. On the other hand, phenol and other phenolic compounds such as 2,6-dimethoxy phenol and 1,2 benzenediol, indicating decomposition of lignin, were observed along the temperature range of 200–350°C. Due to their composition which mainly contained organic acids and phenolic compounds, the liquid product could be considered for disinfectant and organic preservatives. However, such applications need a further investigation. Gas Product Hydrothermal reactions involve the formation of gases derived from degradation of water-soluble compounds [38, 39]. A small amount of gas was produced at low temperature and steadily increased as the temperature was increased. However, CO2 was the predominant gas observed (C80 vol.%), followed by CO and H2. As shown
Fig. 10 Gas produced form hydrothermal treatment at various temperatures
in Fig. 10, the gas proportion strongly depended on the temperature. At 200°C, only CO2 was produced. The percentage of CO2 gradually decreased, accompanied by increases in CO and H2 at elevated temperature. A slight amount of CH4, indicating a methanation, was found at 350°C. The gas formed at this temperature comprised 5.3–5.5 vol.% H2, 3.5–5.3 vol.% CO, 83.5–85.2 vol.% CO2 and 5.6–6.0 vol.% CH4. Application of this gas product seems less essential. Although the gas product contained H2 and CH4, clean and high calorific gases, presence of very high of CO2 makes its application for fuel gas
123
404
inappropriate. On the other hand, its utilization for high purity CO2 source will be costly since an additional separation process is needed. Meanwhile, conversion into the gas products is less than 10%, as shown in the carbon balance (Fig. 3).
Waste Biomass Valor (2010) 1:395–405 Acknowledgments The authors are grateful for support of this research by a Grant-in-Aid for Scientific Research No. 21246135 from the Japan Society for the Promotion of Science (JSPS) and the Global COE program (Novel Carbon Resources Sciences, Kyushu University).
References Conclusions Upgrading of oil palm frond and trunk was investigated by hydrothermal treatment at 200–350°C for 30 min. Approximately 35–65 wt% of the original material was recovered after the process as solid fuel. The very high oxygen and volatile matter content of the original material were significantly reduced. By contrast, the fixed carbon content increased sharply due to carbonization. The van Krevelen diagram revealed that solids resulting from treatment at C330°C have a composition comparable to that of sub-bituminous coal. FTIR analysis confirmed that oxygen elimination due to dehydration in conjunction with decomposition of hemicellulose and cellulose occurred at 200–270°C. At temperatures [270°C, the structure of the solid dramatically changed and was dominated by lignin. This was indicated by an increase in aromatic compounds, as determined by 13C NMR spectroscopy. Hydrothermal treatment progressively changes the calorific value and EMC of materials. Treatment at 350°C produced solid with a gross calorific value as high as 29.7 MJ/kg (for both materials) and EMC of *5.2 and *4.5 wt% for frond and trunk material, respectively. However, significant changes in the calorific value and the EMC was observed in 200–270°C range, which can be attributed to progressive removal of hemicellulose and cellulose. This was in agreement with the GC–MS analysis results, showing that decomposition products of hemicellulose and cellulose (furfural and 5-HMF) were observed only at 200–270°C. The GC–MS result also indicated degradation of lignin took place along the temperature range of 200–350°C. After treatment, liquid containing various organic compounds, mainly organic acids and phenolic compounds, was produced. A small amount of gas was formed at low temperature and steadily increased as the temperature was increased. However, CO2 was the predominant gas observed (C80 vol.%). Although total energy produced from combustion of this solid product was less than that of raw material, the use of this fuel offers other benefits. The fuel had higher energy density and lower EMC which may improve its handling and storage properties. Thus, it is proposed that hydrothermal treatment could become an advantageous technology for producing solid fuel from biomass wastes.
123
1. Abdullah, N., Gerhauser, H.: Bio-oil derived from empty fruit bunches. Fuel 87, 2606–2613 (2008) 2. Nasrin, A.B., Ma, A.N., Choo, Y.M., Mohammad, S., Rohaya, M.H., Azali, A., Zainal, Z.: Oil palm biomass as potential substitution raw materials for commercial biomass briquettes production. Am. J. Appl. Sci. 5, 179–183 (2008) 3. Yang, H., Yan, R., Chen, H., Lee, D.H., Liang, D.T., Zheng, C.: Pyrolysis of palm oil wastes for enhanced production of hydrogen rich gases. Fuel Process. Technology 87, 935–942 (2006) 4. Tanaka, R., Kosugi, A., Mori, Y.: Utilization of lignocellulosic biomass discharged from palm oil industries. J. Jpn. Inst. Energy 86, 368–370 (2007) 5. Rodriguez, A., Serrano, L., Moral, A., Perez, A., Jimenez, L.: Use of high-boiling point organic solvents for pulping oil palm empty fruit bunches. Bioresour. Technol. 99, 1743–1749 (2008) 6. Gonzales, M., Canton, L., Rodriguez, A., Labidi, J.: Effect of organosolv and soda pulping processes on the metals content of non-woody pulps. Bioresour. Technol. 99, 6621–6625 (2008) 7. Hussein, M.Z., Tarmizi, R.S.H., Zainal, Z., Ibrahim, R., Badri, M.: Preparation and characterization of active carbons from oil palm shells. Carbon 34, 1447–1454 (1996) 8. Lua, A.C., Guo, J.: Preparation and characterization of activated carbons from oil-palm stones for gas-phase adsorption. Colloids Surf. A 179, 151–162 (2001) 9. Alam, M.Z., Muyibi, S.A., Mansor, M.F., Wahid, R.: Activated carbons derived from oil palm empty-fruit bunches: application to environmental problems. J. Environ. Sci. 19, 103–108 (2007) 10. Hameed, B.H., Tan, I.A.W., Ahmad, A.L.: Preparation of oil palm empty fruit bunch-based activated carbon for removal of 2, 4, 6-trichlorophenol: optimization using response surface methodology. J. Hazard. Mater. 164, 1316–1324 (2009) 11. Zaidon, A., Nizam, A.M.N., Nor, M.Y.M., Abood, F., Paridah, M.T., Yuziah, M.Y.N., Jalaluddin, H.: Properties of particleboard made from pretreated particles of rubberwood, EFB, and rubberwood–EFB blend. J. Appl. Sci. 7, 1145–1151 (2007) 12. Deraman, M., Zakaria, S., Husin, M., Aziz, A.A., Ramli, R., Mokhtar, A., Yusof, M.N.M., Sahri, M.H.: X-Ray diffraction studies on fiber of oil palm empty fruit bunch and rubberwood for medium-density fiberboard. J. Mater. Sci. Lett. 18, 249–253 (1999) 13. Rozman, H.D., Tay, G.S., Abubakar, A., Kumar, R.N.: Tensile properties of oil palm empty fruit bunch–polyurethane composites. Eur. Polym. J. 37, 1759–1765 (2001) 14. Rozman, H.D., Saad, M.J., Ishak, Z.A.M.: Flexural and impact properties of oil palm empty fruit bunch (EFB)–polypropylene composites–the effect of maleic anhydride chemical modification of EFB. Polym. Test 22, 335–341 (2003) 15. Khalid, M., Ratnam, C.T., Chuah, T.G., Ali, S., Choong, T.S.Y.: Comparative study of polypropylene composites reinforced with oil palm empty fruit bunch fiber and oil palm derived cellulose. Mater Des. 29, 173–178 (2008) 16. Arvanitoyannis, I.S., Kassaveti, A., Stefanatos, S.: Current and potential uses of thermally treated olive oil waste. Int. J. Food Sci. Technol. 42, 852–867 (2007) 17. Arvanitoyannis, I.S., Tserkezou, P.: Wheat, barley and oat waste: a comparative and critical presentation of methods and potential
Waste Biomass Valor (2010) 1:395–405
18.
19.
20. 21.
22.
23.
24.
25. 26.
27.
28.
29.
uses of treated waste. Int. J. Food Sci. Technol. 43, 694–725 (2008) Arvanitoyannis, I.S., Tserkezou, P.: Corn and rice waste: a comparative and critical presentation of methods and current and potential uses of treated waste. Int. J. Food Sci. Technol. 43, 958–988 (2008) Arvanitoyannis, I.S., Varzakas, T.H.: Vegetable waste treatment: Comparison and critical presentation of methodologies. Crit. Rev. Food Sci Nutr. 48, 205–247 (2008) McKendry, P.: Energy production from biomass (part 2): conversion technologies. Bioresour. Technol. 83, 47–54 (2002) Ni, M., Leung, D.Y.C., Leung, M.K.H., Sumathy, K.: An overview of hydrogen production from biomass. Fuel Process Technol. 87, 461–472 (2006) Ando, H., Sakaki, T., Kokusho, T., Shibata, M., Uemura, Y., Hatate, Y.: Decomposition behavior of plant biomass in hotcompressed water. Ind. Eng. Chem. Res. 39, 3688–3693 (2000) Sasaki, M., Adschiri, T., Arai, K.: Fractionation of sugarcane bagasse by hydrothermal treatment. Bioresour. Technol. 86, 301–304 (2003) Zhang, L.H., Xu, C.B., Champagne, P.: Overview of recent advances in thermo-chemical conversion of biomass. Energy Convers. Manage. 51, 969–982 (2010) Kruse, A.: Hydrothermal biomass gasification. J. Supercrit. Fluids 47, 391–399 (2009) Allen, S.G., Schulman, D., Lichwa, J., Antal, M.J., Laser, M., Lynd, L.R.: A comparison between hot liquid water and steam fractionation of corn fiber. Ind. Eng. Chem. Res. 40, 2934–2941 (2001) Ehara, K., Saka, S.: A comparative study on chemical conversion of cellulose between the batch-type and flow-type systems in supercritical water. Cellulose 9, 301–311 (2002) Hamelinck, C.N., Van Hooijdonk, G., Faaij, A.P.C.: Ethanol from lignocellulosic biomass: techno-economic performance in short-, middle- and long-term. Biomass Bioenerg. 28, 384–410 (2005) Laser, M., Schulman, D., Allen, S.G., Lichwa, J., Antal, M.J., Lynd, L.R.: A comparison of liquid hot water and steam
405
30.
31.
32.
33.
34.
35.
36.
37. 38.
39.
pretreatments of sugar cane bagasse for bioconversion to ethanol. Bioresour. Technol. 81, 33–44 (2002) Sluiter, A., Hames, B., Ruiz, R., Scarlata, C., Sluiter, J., Templeton, D., Crocker, D.: Determination of structural carbohydrates and lignin in biomass. The US national renewable energy laboratory technical report, (2005) Yan, W., Acharjee, T.C., Coronella, C.J., Vasquez, V.R.: Thermal pretreatment of lignocellulosic biomass. Environ. Prog Sustain. Energy 28, 435–440 (2009) Ehara, K., Saka, S.: Decomposition behavior of cellulose in supercritical water, subcritical water, and their combined treatments. J. Wood Sci. 51, 148–153 (2005) Kobayashi, N., Okada, N., Hirakawa, A., Sato, T., Kobayashi, J., Hatano, S., Itaya, Y., Mori, S.: Characteristics of solid residues obtained from hot-compressed-water treatment of woody biomass. Ind. Eng. Chem. Res. 48, 373–379 (2009) Capanema, E.A., Balakshin, M.Y., Kadla, J.F.: Quantitative characterization of a hardwood milled wood lignin by nuclear magnetic resonance spectroscopy. J. Agric. Food Chem. 53, 9639–9649 (2005) Wikberg, H., Maunu, S.L.: Characterisation of thermally modified hard- and softwoods by C-13 CPMAS NMR. Carbohydr. Polym. 58, 461–466 (2004) Wooten, J.B., Kalengamaliro, N.E., Axelson, D.E.: Characterization of bright tobaccos by multivariate analysis of C-13 CPMAS NMR spectra. Phytochemistry 70, 940–951 (2009) Morohoshi, N.: Woody biomass utilization. Buneido Shuppan, Tokyo, p. 84 (1991) (in Japanese) Kruse, A., Dinjus, E.: Hot compressed water as reaction medium and reactant: 2. degradation reactions. J. Supercrit. Fluids 41, 368–369 (2007) Srokol, Z., Bouche, A.G., van Estrik, A., Strik, R.C.J., Maschmeyer, T., Peters, J.A.: Hydrothermal upgrading of biomass to biofuel: studies on some monosaccharide model compounds. Carbohydr. Res. 339, 1717–1726 (2004)
123
