Recovery of solid fuel from municipal solid waste by hydrothermal treatment using subcritical water

Waste Management 32 (2012) 410–416

Contents lists available at SciVerse ScienceDirect

Waste Management journal homepage: www.elsevier.com/locate/wasman

Recovery of solid fuel from municipal solid waste by hydrothermal treatment using subcritical water In-Hee Hwang ⇑, Hiroya Aoyama, Toshihiko Matsuto, Tatsuhiro Nakagishi, Takayuki Matsuo Laboratory of Solid Waste Disposal Engineering, Faculty of Engineering, Hokkaido University, Kita 13 Nishi 8, Kita-ku, Sapporo 060 8628, Japan

a r t i c l e

i n f o

Article history: Received 24 March 2011 Accepted 4 October 2011 Available online 21 November 2011 Keywords: Hydrothermal treatment using subcritical water (HTSW) Municipal solid waste (MSW) Char Heating value Cl removal

a b s t r a c t Hydrothermal treatments using subcritical water (HTSW) such as that at 234 °C and 3 MPa (LT condition) and 295 °C and 8 MPa (HT condition) were investigated to recover solid fuel from municipal solid waste (MSW). Printing paper, dog food (DF), wooden chopsticks, and mixed plastic film and sheets of polyethylene, polypropylene, and polystyrene were prepared as model MSW components, in which polyvinylchloride (PVC) powder and sodium chloride were used to simulate Cl sources. While more than 75% of carbon in paper, DF, and wood was recovered as char under both LT and HT conditions, plastics did not degrade under either LT or HT conditions. The heating value (HV) of obtained char was 13,886–27,544 kJ/kg and was comparable to that of brown coal and lignite. Higher formation of fixed carbon and greater oxygen dissociation during HTSW were thought to improve the HV of char. Cl atoms added as PVC powder and sodium chloride to raw material remained in char after HTSW. However, most Cl originating from PVC was found to converse into soluble Cl compounds during HTSW under the HT condition and could be removed by washing. From these results, the merit of HTSW as a method of recovering solid fuel from MSW is considered to produce char with minimal carbon loss without a drying process prior to HTSW. In addition, Cl originating from PVC decomposes into soluble Cl compound under the HT condition. The combination of HTSW under the HT condition and char washing might improve the quality of char as alternative fuel. Ó 2011 Elsevier Ltd. All rights reserved.

1. Introduction Hydrothermal treatment using subcritical water (HTSW) has been widely employed for the solubilization, extraction, and liquefaction of target materials. Subcritical water has hydrolytic and pyrolytic reaction characteristics, which result from a decrease in the dielectric constant and increase in the ion product of water at temperatures and pressures below and near 374 °C and 22.1 MPa (Kang et al., 2001; Brunner, 2009). There have been several research works on HTSW for municipal and industrial solid-waste treatment. Most researchers have focused on solubilization and extraction to recover valuable organic compounds such as glucose and organic acids (Kang et al., 2001; Goto et al., 2004; Yoshida and Tavakoli, 2004; Ren et al., 2006; Watchararuji et al., 2008; Lamoolphak et al., 2008). Some researchers have dealt with hydrothermal treatment as pretreatment prior to fermentation, gasification, composting, and other processes (Eley et al., 1996; Sawayama et al., 1997; Kato and Matsumura, 2003; Papadimitriou et al., 2008). Only few researchers have focused on recovering solid fuel from municipal solid waste ⇑ Corresponding author. Tel./fax: +81 11 706 6828. E-mail address: [email protected] (In-Hee Hwang). 0956-053X/$ - see front matter Ó 2011 Elsevier Ltd. All rights reserved. doi:10.1016/j.wasman.2011.10.006

(MSW). Nouguchi and Inoue (2007) performed HTSW to recover char from waste at 150–350 °C but it was limited to model food waste. Recovery of solid fuel from MSW using HTSW has several advantages: the effect of carbonization (or pyrolysis) can be achieved under the relatively low temperature below 300 °C, compared with the temperature of 400–600 °C for MSW carbonization performed at little or no oxygen condition; moisture removal is not necessary for wet MSW unlike MSW carbonization; and conversely, the moisture included in MSW can be used as a heating medium, stream, to decompose MSW. A 37 t/day-scale HTSW plant has been operating in Shiraoi-Cho, Hokkaido, since April 2009 to treat combustible wastes including food waste collected from residential and commercial areas for solid-fuel conversion. Batch-type HTSW is performed using three autoclaves and saturated steam at 234 °C and 3 MPa. The total operating time per batch is 4–6 h including the time for waste input and product discharge (1–1.5 h). Obtained solid product, char, is pelletized with shredded wooden and plastic wastes and used as alternative fuel for the boiler of a paper-manufacturing plant. However, for better understanding the process, further information is needed on the effects of the temperature and pressure of HTSW the holding time in reactor, the physical composition of MSW, and

411

I.-H. Hwang et al. / Waste Management 32 (2012) 410–416

350

A

B

C

10

D

300 8 250 6

200 150

4 THT

100

TLT PHT

50

PLT

0

2. Material and methods

0

30

60

90

120

Fig. 1 shows a laboratory-scale experimental apparatus for HTSW. A 50 g of raw material and 150 ml of distilled water were put in a batch type of autoclave reactor and the reactor was then sealed and purged with nitrogen gas through valve A. After nitrogen gas purging, valves A and B were closed. The speed of stirrer was set up 150 rpm. Temperature and pressure were set at 234 °C and 3 MPa or 295 °C and 8 MPa. The former is the same operating condition as for the hydrothermal treatment plant in Shiraoi-Cho, Hokkaido and is referred to as the LT condition. The latter is the upper limits of temperature and pressure for the reactor and is referred to as the HT condition. The temperature and pressure within the reactor were monitored by sensors and signalized by a data logger connected with a computer system. As presented in Fig. 2, nitrogen gas in the reactor was discharged by opening valve A at 75 °C and the reactor was filled with steam in the temperature range of 75–115 °C (A). The reactor was heated to a set LT or HT condition (B) and maintained for 5 min after reaching set condition (C). Afterward, the heater was turned off and the reactor was decompressed by opening valve B. Some mixture of vapor and gas product in the reactor was discharged to condensing bottles at the same time and then the reactor was cooled down (D). Solid, liquid, and gas products were collected when the reactor had cooled to room temperature. Nitrogen gas was injected through valve A to discharge remaining gas in the reactor into a gas sampling bag completely. The volume of non-condensable gas was measured with a dry-gas meter and the entire volume was collected in a 10 L Tedlar bag to analyze the gas composition. Solid product, char, was collected first. Remaining solid and liquid mixture was collected by rinsing with distilled water and then filtered with a 1 lm pore size filter paper. Solid trapped by the filter paper was collected and weighted as char. Char was dried at 105 °C for 12 h and then weighed and pulverized to a grain size of 500 lm.

Fig. 2. Variations in temperature and pressure within the reactor under LT and HT conditions.

Filtrate was collected as liquid product. Liquid in the gas cooling and scrubbing bottles was collected and then filtered with a 1 lm pore size filter paper too. All filtrate was collected as liquid product and was provided for measuring the total carbon concentration. 2.2. Analytical method Raw material and char were provided for determination of ash and volatile matter (VM) based on ASTM D3172-89. Fixed carbon (FC) was estimated according to the mass balance (FC = 100  VM  ash). C, H, and N contents were measured with an elemental analyzer (CHN Corder MT-6, Yanaco Co.). S and Cl contents were measured employing combustion and gas-absorption methods. The sulfate concentration of the absorption solution was measured by ion chromatography (DX 500 series, Dionex Co.). The chloride ion concentration was measured employing the mercuric thiocyanate method and an absorption spectrophotometer (U-1101, Hitachi Co., Tokyo, Japan) at 460 nm. Total carbon and chloride ion concentration of liquid product were measured employing a total organic carbon analyzer (TOCV CPH, Shimadzu Co., Kyoto, Japan) and the mercuric thiocyanate method and an absorption spectrophotometer at 460 nm respectively. CO2, CO, and CH4 concentrations of non-condensable gas were measured by a GC using thermal conductivity detector (Type 164, Hitachi Co., column type: WG-100, flow rate of He: 33 ml/ min; detector temperature: 50 °C). HV of char was measured using a bomb calorimeter (CA-4PJ, Shimazu Ltd.). Washing test of char was carried out to investigate Cl removal. A 10 g of char was shaken with 100 ml of distilled water at 150 rpm for 20 min and the mixture was filtered using a 1 lm fil-

Pressure sensor

Temperature sensor

Valve B

Valve A

Stirrer N2 gas Gas bag

Reactor controller

0 150

Holding time (min)

2.1. Experimental apparatus and procedure

Computer system

2

Pressure (MPa)

o

Temperature ( C)

the presence of Cl compounds in MSW on the yield and quality of char. In this work, we performed HTSW to recover solid fuel from MSW. Paper, food, wood, and plastics were prepared as the main components of MSW and polyvinyl chloride (PVC) and NaCl were used as sources of Cl in MSW. Decomposition characteristics of each MSW component and yield and composition of obtained char were investigated under two conditions of HTSW. Moreover, the heating value (HV) and Cl removal of solid product were evaluated in terms of fuel utilization.

Autoclave reactor

Vapor and gas condensing bottles

Gas Gas Gas meter scrubbing dehydrating bottles bottle

Fig. 1. Laboratory-scale apparatus for HTSW.

412

I.-H. Hwang et al. / Waste Management 32 (2012) 410–416

ter paper. This procedure was repeated up to four times to recover the entire soluble fraction. The concentration of released chloride ion in distilled water was measured employing the mercuric thiocyanate method and an absorption spectrophotometer at 460 nm. 2.3. Raw material and Cl source Shredded printing paper (20  20 mm), dog food (DF) as food waste, shredded wooden chopsticks (<5 mm), and mixed plastic film and sheets (<20 mm; PE:PP:PS = 5:3:2) were prepared as raw materials. According to the ingredients table offered by the manufacturer, DF was composed of 50% carbohydrates, 25% protein, 11% fat, 6% fiber, and 9% ash on a dry basis. Table 1 presents the composition of the raw material. To investigate the behavior of Cl compounds during HTSW, NaCl and PVC powders were added to paper and wood respectively such that Cl was 2% of the total dry weight of the raw material. 3. Results and discussion 3.1. Carbon balance of product The reliability of the experiment was checked using the carbon balance between raw materials and products obtained by HTSW. Table 2 shows the carbon recovery ratio of product (CR) when the total carbon of raw material is normalized to one and the carbon distribution ratio of char (Cc), liquid (Cl), and gas (Cg) when the total carbon recovered as product is normalized to one respectively. More than 88% of carbon in raw material was recovered as products, which demonstrates the reliability of the experiment. For paper, DF, and wood, as shown by Cc, more than 75(=0.88 X 0.85 X 100)% of carbon in raw material was recovered in char under both LT and HT conditions, which was considered a merit in solidfuel recovery. Such a high value of Cc was due to the remaining VM, which was usually much more degraded during carbonization at 400–600 °C. The Cc for the LT condition was about 5–8% higher than that for the HT condition for the same input material. The sum of Cl and Cg was less than 19% of CR. Generated gas is composed of carbon dioxide and carbon monoxide (CO2 >> CO). Zhang et al. (2011) noted that carbon dioxide and carbon monoxide were primarily formed as a result of decarboxylation of organic fraction during HTSW. Liquid product was not provided for the qualitative analysis in this work but it was expected to contain organic compounds such as alcohols, ethers, aldehydes, phenols, carboxylic acids, etc. according to previous research (Qian et al., 2010; Zhang et al., 2011). On the other hand, Cc for plastics is 100% because plastics do not decompose under either LT or HT.

yield was determined from the weight ratio of char to input raw material on a dry basis. For the same raw material excepting plastics, a higher char yield was obtained under the LT condition. The largest difference in the char yield between LT and HT conditions was observed for char derived from paper. The main component of paper is fiber such as hemicellulose and cellulose, which have been reported to decompose rapidly in the temperature range of 200–400 °C under the pyrolytic condition (Sørum et al., 2001; Völker and Rieckmann, 2002; Myung et al., 2004; Shen and Gu, 2009). The greater degradation of hemicellulose and cellulose seems to drastically decrease the char yield. Compared with the char derived from paper, the difference in the char yield between the LT and HT conditions was not significant for DF and wood. Wood, another fibrous biomass in this work, has a relatively high yield compared with paper under the HT condition. Wooden chopsticks were made of broadleaf trees such as silver birch, which usually contains from 18% to 24% lignin (Murata, 2004). Considering that lignin starts to decompose at temperatures over 330 °C under a pyrolytic condition (Myung et al., 2004), the presence of lignin may explain why the char yield of wood is higher than that of paper under the HT condition. The yields of char derived from paper noticeably deviated under the LT condition (Table 3). This was probably due to difference in heating rates by fluctuation in the rising and falling conditions of temperature during HTSW. For this reason, the difference in the holding time of raw material in the reactor was checked. As shown in Fig. 2, the holding time indicated the entire elapsed time during HTSW but the extent of decomposition of raw material might be related to the holding time at higher temperature from stage B to stage D. Fig. 3 shows the relationship between the yield of VM + FC of char and the holding time of raw material at temperature over 220 °C. The yield of VM + FC of char means the weight of VM + FC of recovered char when the weight of raw material is normalized to one. As the holding time increased under the LT condition, the yield of VM + FC of char derived from paper decreased. This indicates that the variation in the holding time leads to the deviation of the char yield of paper. The relationship between the yield and holding time at temperatures exceeding 200 and 210 °C was also observed, but their correlation was not strong compared with that for temperatures exceeding 220 °C. On the other hand, no effect of holding time was observed for DF and wood char because the deviation of the holding time was small under both LT and HT conditions. Finally, the mixed plastics were expected to degrade under the high pressure of the HT condition but no weight loss of plastics was observed as shown by the yield in Table 3. The plastics appeared to melt partially through hydrothermal treatment and to harden again during cooling of the reactor. 3.3. Evaluation of the char composition and heating value

3.2. Variations in char yield and composition depending on raw material and HTSW condition Table 3 shows the yield and composition of char derived from each raw material depending on the LT and HT conditions. Char

Table 1 Composition of raw material.

Paper DF Wood Plastics

Ash

FC

VM

C

H

N

S

Cl

0.05 0.08 0 0

0.06 0.14 0.09 0.01

0.89 0.79 0.91 0.99

0.38 0.45 0.47 0.86

0.06 0.07 0.06 0.13

ND 0.04 ND 0.06

ND 0.004 ND ND

0.002 0.022 0.001 ND

Dry basis. Unit: weight ratio ().

Several chars, which were selected according to the type of raw material and HTSW condition, were provided for measurements of HVs. The results are presented in Table 4. HVs of char (HVchar) derived from paper, DF, and wood were 13,886–26,000, 24,627– 27,145, and 22,134–27,544 kJ/kg respectively, which are comparable to HVs of lignite and brown coal. Char yield decreased under the HT condition (Table 3) but HVchar was higher under the HT condition (Table 4). To clarify the relationship between char composition and HVchar the compositions and HVs recovered from 1 kg of raw material were calculated by multiplying the values char compositions and of HVchar by char yield respectively. These results are compared in Fig. 4. Although a small amount of char was obtained from 1 kg of raw material under the HT condition (Fig. 4b), the recovered energy per 1 kg of raw material did not differ between LT and HT conditions

413

I.-H. Hwang et al. / Waste Management 32 (2012) 410–416 Table 2 Carbon balance of product depending on raw material and HTSW condition. Carbon recovery ratio

Paper

LT (n = 8) HT (n = 8) LT (n = 4) HT (n = 2) LT (n = 8) HT (n = 6) LT (n = 2) HT (n = 2)

DF Wood Plastics

a

Carbon distribution ratio

b

CR

Cc

Cl

Cg

0.99 ± 0.03 1.00 ± 0.06 0.93 ± 0.03 0.88 ± 0.03 0.89 ± 0.02 0.92 ± 0.01 1.01 ± 0.01 1.01 ± 0.01

0.89 ± 0.02 0.81 ± 0.03 0.92 ± 0.03 0.85 ± 0.02 0.89 ± 0.01 0.84 ± 0.01 1.00 ± 0.00 1.00 ± 0.00

0.05 ± 0.01 0.08 ± 0.02 0.05 ± 0.02 0.07 ± 0.01 0.09 ± 0.01 0.08 ± 0.01 0.00 ± 0.00 0.00 ± 0.00

0.06 ± 0.02 0.12 ± 0.02 0.03 ± 0.01 0.09 ± 0.01 0.03 ± 0.01 0.07 ± 0.00 0.00 ± 0.00 0.00 ± 0.00

Unit: weight ratio (–). a when the total carbon of raw material is normalized to 1. b when the total carbon of recovered as product is normalized to 1.

Table 3 Yield and composition of char depending on raw material and HTSW condition. Compositiona

Yield

Paper

LT (n = 8) HT (n = 8) LT (n = 4) HT (n = 2) LT (n = 8) HT (n = 6) LT (n = 2) HT (n = 2)

DF Wood Plastics

0.66 ± 0.07 0.47 ± 0.02 0.66 ± 0.03 0.55 ± 0.00 0.59 ± 0.03 0.51 ± 0.02 0.98 ± 0.01 1.00 ± 0.01

Ash

FC

VM

C

H

N

S

Ob

0.08 ± 0.01 0.09 ± 0.02 0.13 ± 0.03 0.13 ± 0.01 0.01 ± 0.02 0.02 ± 0.00 0.00 ± 0.01 0.00 ± 0.01

0.25 ± 0.06 0.42 ± 0.01 0.25 ± 0.01 0.33 ± 0.02 0.36 ± 0.04 0.53 ± 0.01 0.01 ± 0.00 0.01 ± 0.00

0.67 ± 0.06 0.49 ± 0.02 0.62 ± 0.02 0.53 ± 0.03 0.63 ± 0.02 0.44 ± 0.00 0.99 ± 0.00 0.99 ± 0.00

0.49 ± 0.03 0.63 ± 0.02 0.56 ± 0.01 0.58 ± 0.01 0.61 ± 0.03 0.69 ± 0.01 0.85 ± 0.00 0.84 ± 0.01

0.05 ± 0.00 0.05 ± 0.00 0.06 ± 0.00 0.05 ± 0.00 0.05 ± 0.00 0.05 ± 0.00 0.12 ± 0.00 0.11 ± 0.01

0.00 ± 0.00 0.00 ± 0.00 0.05 ± 0.00 0.05 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00 0.00 ± 0.00

0.01 ± 0.00 0.02 ± 0.00 0.03 ± 0.00 0.03 ± 0.00 0.01 ± 0.01 0.01 ± 0.01 0.01 ± 0.01 0.01 ± 0.00

0.37 ± 0.03 0.22 ± 0.02 0.18 ± 0.02 0.15 ± 0.00 0.32 ± 0.01 0.24 ± 0.01 0.01 ± 0.01 0.03 ± 0.02

Unit: weight ratio (). LT: HTSW at 234 °C and 3 MPa. HT: HTSW at 294 °C and 8 MPa. a based on dry weight. b calculated by mass balance.

1.0

Yield of VM+FC of char when the weight of raw material is 1 (-)

0.9

0.8 P-L8

0.7

the different raw materials under LT and HT conditions were heated for 7 min at 900 °C to remove VM and were then taken to measure the HV of FC (HVFC). As listed in Table 3, HVFC ranged from 29,203 to 34,084 kJ/kg, with there being no large difference between LT and HT conditions. The average value of HVFC was 32,120 ± 1680 kJ/kg, which is near the heat of combustion of carbon (32,800 kJ/C-kg). Assuming that HVFC was equal to 32,800 kJ/ kg, HVVM was estimated using an Eq. 1:

Paper-LT Paper-HT DF-LT DF-HT Wood-LT Wood-HT Plastics-LT Plastics-HT P-L5

W-L2 W-L4 W-L1

0.6

P-L1 P-L2

P-L7 P-L3

W-L3 W-H1

D-L2

0.5

HV VM ðkJ=kg  VMÞ ¼

P-L6

W-H2

D-L1

D-H2 P-H5 P-H1

D-H1

P-H2 P-H3

P-H4

0.3 0

20

40

ð1Þ

P-L4

W-H3

0.4

HV char  FC  HV FC VM

60

80

100

Holding time over 220oC (min) Fig. 3. Variation of VM + FC of char obtained from HTSW of different raw material with holding time over 220 °C (P: paper, D: DF, W: wood, L: LT condition, H: HT condition).

(Fig. 4a). The increase in FC content was likely to improve the HV of char obtained under the HT condition. At the same time, the quality of VM might vary and affect the HV of char too. To prove the variation of VM quality depending on LT and HT conditions, the HVs of VM were estimated. Six chars obtained from

Unlike HVFC, HVVM varied from 7375 to 26,936 kJ/kg depending on raw materials and HTSW conditions (Table 4). To clarify the reason for the change in HVVM, the elementary composition of VM was estimated as described in Fig. 5 and the results are presented in Table 4. Variation of carbon and oxygen content was obvious between LT and HT condition. Fig. 6 shows the molar ratios of C/H and O/H of VM. At the HT condition, the drop in oxygen content was obvious for char derived from paper whereas the drop in carbon content was obvious for char derived from wood. In regard to char derived from DF, the variation of the molar ratios of C/H and O/H of VM was not significant. Comparing these results with HVVM in Table 4, oxygen dissociation seems to contribute to the rise in HVVM at HT condition. High oxygen dissociation might be related to the formation of carbon monoxide and carbon dioxide by decarboxlyation reaction during HTSW. As shown in Table 2, the Cg was highest for HTSW of paper under the HT condition might be one of good evidence for the highest oxygen drop in char derived from paper at HT condition.

414

I.-H. Hwang et al. / Waste Management 32 (2012) 410–416

Table 4 HVs of char, FC, and VM and composition of VM. IDa

Paper

LT

HT

DF

LT

HT

Wood

LT

HT

a b c

HVChar

P-L1 P-L2 P-L3 Avg ± st P-H1 P-H2 P-H3 Avg ± st D-L1 D-L2 Avg ± st D-H1 D-H2 Avg ± st W-L1 W-L2 Avg ± st W-H1 W-H2 Avg ± st

dev

dev

dev

dev

dev

dev

HVFC

HVVM

b

Composition of VM

c

()

(kJ/kg-char)

(kJ/kg-FC)

(kJ/kg-VM)

C

H

N

S

O

18,290 15,791 13,886 15,989 ± 2209 26,000 23,929 21,267 23,732 ± 2373 24,627 27,145 25,886 ± 1780 25,114 26,985 26,050 ± 1323 24,853 22,134 23,494 ± 1923 26,036 27,544 26,790 ± 1066

32,110

16,464 12,188 7375 12,009 ± 4547 24,097 22,755 16,467 21,106 ± 4073 24,911 31,325 28,118 ± 4535 26,936 29,698 28,317 ± 1953 20,288 18,895 19,592 ± 985 19,066 22,714 20,890 ± 2580

0.36 0.36 0.35 0.36 ± 0.01 0.42 0.41 0.43 0.42 ± 0.01 0.47 0.50 0.49 ± 0.02 0.45 0.48 0.47 ± 0.02 0.40 0.39 0.40 ± 0.01 0.32 0.33 0.33 ± 0.01

0.07 0.08 0.08 0.08 ± 0.01 0.10 0.10 0.10 0.10 ± 0.00 0.09 0.09 0.09 ± 0.00 0.10 0.10 0.10 ± 0.00 0.08 0.08 0.08 ± 0.00 0.10 0.10 0.10 ± 0.00

0.00 0.00 0.00 0.00 ± 0.00 0.00 0.00 0.00 0.00 ± 0.00 0.08 0.08 0.08 ± 0.00 0.10 0.10 0.10 ± 0.00 0.00 0.00 0.00 ± 0.00 0.00 0.00 0.00 ± 0.00

0.02 0.01 0.01 0.01 ± 0.01 0.02 0.04 0.04 0.03 ± 0.01 0.04 0.04 0.04 ± 0.00 0.06 0.06 0.06 ± 0.00 0.00 0.03 0.02 ± 0.02 0.03 0.00 0.02 ± 0.02

0.54 0.56 0.56 0.55 ± 0.01 0.46 0.45 0.42 0.44 ± 0.02 0.33 0.29 0.31 ± 0.03 0.29 0.27 0.28 ± 0.01 0.52 0.51 0.52 ± 0.01 0.56 0.57 0.57 ± 0.01

32,210

29,203

31,738

33,378

34,084

Refer Fig. 3. estimated using equation 1 assuming that HVFC is 32,800 kJ/kg. when the VM weight is normalized to 1.

Paper

P-L1 P-L2 P-L3 P-H1 P-H2 P-H3

DF

D-L1 D-L2 D-H1 D-H2

Wood

W-L1 W-L2

FC VM Ash

W-H1 W-H2

20000

16000

12000

8000

4000

0

Recovered energy as char (kJ/kg-raw material)

0.0

0.2

0.4

0.6

0.8

1.0

Recovered FC, VM, and ash as char (kg/kg-raw material)

(b)

(a)

Fig. 4. Recovered energy and composition as char from 1 kg raw material (P: paper, D: DF, W: wood, L: LT condition, H: HT condition).

CHAR FC VM

C

H NS

O

H NS

O

Ash

FC C

Fig. 5. Estimation of elementary composition of VM composition.

3.4. Cl content of char and Cl removal by washing Table 5 shows the Cl recovery ratio (ClR) when the total Cl added to raw material is normalized to one and the Cl distribution ratio of char (Clc) and liquid + gas (Cll+g) when total Cl recovered as products is normalized to one respectively. After HTSW under LT and HT conditions, more than 83% of input Cl was recovered in products and more than 85% of recovered Cl was found in char.

When PVC is pyrolyzed in the range of 200–300 °C, dehydrochlorination emits HCl gas, which dissolves in water (Mikata et al., 1996; Takeshita et al., 2004). As HTSW was performed over 234 °C in this work, a considerable quantity of Cl atoms in PVC was expected to shift to liquid products. However, more than 85% of Cl originating from PVC remained in char regardless of the LT or HT condition (Table 5). Cl added in the form of NaCl also remained in the char owing to its high melting point (Table 5). To examine the feature of Cl compounds remaining in char, char washing was performed. Fig. 7 shows the soluble and insoluble Cl contents in char. In the case of PVC, a greater amount of soluble Cl was observed in the char obtained under the HT condition (Fig. 7a). This indicates that dehydrochlorination advanced more and Cl atoms remained in char in the form of a soluble compound under the HT condition. It is possible that PVC once decomposes into HCl, which is dissolved in moisture at the surface and in the pores of char particles. Another possibility is that Cl atoms chemically

415

I.-H. Hwang et al. / Waste Management 32 (2012) 410–416

0.7

Paper-LT P-L3 P-L4

0.6

P-L5 P-L6

C/H molar ratio (-)

0.5 D-L2

Paper-HT

D-L1 W-L1

D-H2

0.4

W-L2

P-H3 P-H1

D-H1

P-L2

P-H5

P-L3

Wood-LT

Paper

P-H2

0.3

P-H4

P-L1

W-L3

Paper-LT

W-H2

Paper-HT

W-H1

W-L4

DF

0.2

Wood-HT

DF-LT

W-H2

DF-HT

W-H3

Wood

0.1

Wood-LT

0.0

Wood-HT

0.2

0.0 0.0

0.1

0.2

0.3

0.4

0.5

0.6

0.4

0.6

0.8

1.0

1.2

1.0

1.2

Cl content (mg-Cl/g-char) (a) Addition of PVC

0.7

O/H molar ratio (-) Paper-LT Fig. 6. Molar ratios of C/H and O/H of VM (P: paper, D: DF, W: wood, L: LT condition, H: HT condition).

P-L7

combine with alkali metals included in char during HTSW. PVC was added as 2% Cl (0.56 Cl-eq/kg) for raw material. However, the contents of alkali metals such as K, Na, Mg, and Ca were 1.47 and 0.05 Cl-eq/kg for paper and wood respectively. Thus, the latter seems slim for char derived from wood. On the other hand, Cl added in the form of NaCl was largely removable by washing char produced under both LT and HT conditions as shown in Fig. 7b. Soluble Cl of char derived from paper + PVC was obviously higher than that of char derived from wood + PVC under the LT condition (Fig. 7a). Considering that the yield of VM + FC decreased with increasing holding time over 220 °C (Fig. 3), the same correlation might be observed between the formation of soluble Cl and the holding time over 220 °C. Accordingly, a longer holding time for PVC over 220 °C led to the progression of dehydrochlorination, enhancing the generation of soluble Cl compounds in char that can be removed by washing.

Paper-HT

P-L8 P-H2 P-H3 0.0

0.2

0.4

0.6

0.8

Cl content (mg-Cl/g-char) (b) Addition of NaCl Fig. 7. Soluble and insoluble Cl contents of char (

soluble Cl,

insoluble Cl).

4. Conclusions HTSW at 234 °C and 3 MPa (LT) and 295 °C and 8 MPa (HT) was investigated as a method of recovering solid fuel from MSW. More than 75% of carbon in raw material was recovered as char by HTSW under LT and HT conditions, and the char had an HV comparable to

Table 5 Cl balance of product depending on the Cl source and HTSW condition. IDa

Paper + PVC

LT

HT

Wood + PVC

LT

HT

Paper + NaCl

LT

HT

a b c

Refer Fig. 3 when the amount of input Cl is 1. when the amount of recovered Cl is 1.

Cl recovery ratio

P-L3 P-L4 P-L5 P-L6 Avg ± stdev P-H4 P-H5 Avg ± stdev W-L3 W-L4 Avg ± stdev W-H2 W-H3 Avg ± stdev P-L7 P-L8 Avg ± stdev P-H2 P-H3 Avg ± stdev

b

Cl distribution ratio

c

ClR

Clc

Cll+g

1.01 0.89 1.00 0.89 0.95 ± 0.07 0.98 1.07 1.02 ± 0.06 0.93 0.98 0.95 ± 0.04 0.83 1.01 0.92 ± 0.13 1.02 1.04 1.03 ± 0.01 0.97 0.97 0.97 ± 0.00

0.98 0.98 1.00 0.98 0.98 ± 0.01 0.87 0.98 0.92 ± 0.08 0.97 0.99 0.98 ± 0.01 0.94 0.85 0.89 ± 0.07 0.97 0.98 0.97 ± 0.01 0.93 0.95 0.94 ± 0.01

0.02 0.02 0.00 0.02 0.02 ± 0.01 0.13 0.02 0.08 ± 0.08 0.03 0.01 0.02 ± 0.01 0.06 0.15 0.11 ± 0.07 0.03 0.02 0.03 ± 0.01 0.07 0.05 0.06 ± 0.01

416

I.-H. Hwang et al. / Waste Management 32 (2012) 410–416

that of brown coal and lignite. A higher concentration of carbon as FC and greater oxygen dissociation of VM during HTSW are thought to improve the HV of char. These reaction characteristics were obvious for char derived from paper and wood. Considering that paper occupies 30–50% of household waste, such reaction characteristics are important to the recovery of qualified fuel from MSW during hydrothermal treatment. Plastics did not degrade under LT and HT conditions in this work. If a large amount of plastic is mixed in the input waste, the quality of char may be uneven owing to the partial melting of plastics during HTSW under conditions considered in this work. Most Cl originating from PVC and NaCl remained in char after HTSW. However, PVC substantially degraded into soluble Cl compounds that could be removed by washing under the HT condition. Considering the presence of salt of food waste and PVC in MSW the combination of HTSW under the HT condition and char washing might be a proper method in case a strict standard of Cl content of char is required to use it as alternative fuel. From these results, the merits of HTSW as a method of recovering solid fuel from MSW are considered to require no drying process prior to HTSW, to produce char with minimal carbon loss, and to decompose PVC into soluble Cl compounds removable by washing. Acknowledgements The authors would like to acknowledge financial support from Kubota Co. Ltd. and the Support Office for Female Researchers at Hokkaido University (FResHU). References Brunner, G., 2009. Near critical and supercritical water. Part I. Hydrolytic and hydrothermal processes. The Journal of Supercritical Fluids 47, 373–381. Eley, M.H., Guinn, G.R., Bagchi, J., 1996. Cellulosic materials recovered from steam classified municipal solid wastes as feedstocks for conversion to fuels and chemicals. Applied Biochemistry and Biotechnology 51–52 (1), 387–397. Goto, M., Obuchi, R., Hirose, T., Sakaki, T., Shibata, M., 2004. Hydrothermal conversion of municipal organic waste into resources. Bioresource Technology 93, 279–284. Kang, K., Quitain, A.T., Daimon, H., Noda, R., Goto, N., Hu, H.-Y., Fujie, K., 2001. Optimization of Amino Acids Production from Waste Fish Entrails by Hydrolysis

in Sub- and Supercritical Water. The Canadian Journal of Chemical Engineering 79, 65–70. Kato, A., Matsumura, Y., 2003. Hydrothermal Pulping of Wet Biomass as Pretreatment for Supercritical Water Gasification Studied Using Cabbage as a Model Compound. Journal of the Japan Institute of Energy 82, 97–102 (Japanese). Lamoolphak, W., De-Eknamkul, W., Shotipruk, A., 2008. Hydrothermal production and characterization of protein and amino acids from silk waste. Bioresource Technology 99, 7678–7685. Mikata, N., Hashimoto, S., Takeuchi, T., Nishiyama, H., 1996. Research in Thermal Decomposition Characterisitics of Plastic Waste. Nippon Steel Technical Report 360, pp.38–45 (Japanese). Murata, S., 2004. Handbook of wood engineering, fourth ed. Maruzen, Tokyo (Japanese). Myung, S., Eom, Y., Dong, J.-I., Park, Y.-K., Kang, B.S., Jeon, J.-K., 2004. Characteristics of Thermal Decomposition of Major Components of Biomass Isolated from Wood. J. Korean Ind. Eng. Chem 15 (8), 896–900 (Korean). Nouguchi, M., Inoue, S., 2007. Carbonization of Model Garbage under Hydrothermal Condition. Journal of the Japan Institute of Energy 86, 475–479 (Japanese). Papadimitriou, E.K., Barton, J.R., Stentiford, E.I., 2008. Sources and levels of potentially toxic elements in the biodegradable fraction of autoclaved nonsegregated household waste and its compost/digestate. Waste Management & Research 26, 419–430. Qian, Q., Mochidzuki, K., Sakoda, A., 2010. Carbon and nitrogen balances for pyrolysis of methane fermentation sludge (MFS) using super-heated steam. Environmental Science 23 (1), 31–41. Ren, L.-H., Nie, Y.-F., Liu, J.-G., Jin, Y.-Y., Sun, L., 2006. Impact of hydrothermal process on the nutrient ingredients of restaurant garbage. Journal of Environmental Sciences. 18 (5), 1012–1019. Sawayama, S., Inoue, S., Minowa, T., Tsukahara, K., Ogi, T., 1997. Thermochemical Liquidization and Anaerobic Treatment of Kichen Garbage. Journal of Fermentation and Bioengineering 83 (5), 451–455. Shen, D.K., Gu, S., 2009. The mechanism for thermal decomposition of cellulose and its main products. Bioresource Technology 100, 6469–6504. Sørum, L., Grønli, M.G., Hustad, J.E., 2001. Pyrolysis characteristics and kinetics of municipal solid wastes. Fuel 80, 1217–1227. Takeshita, Y., Kato, K., Takahashi, K., Sato, Y., Nishi, S., 2004. Basic study on treatment of waste polyvinyl chloride plastics by hydrothermal decomposition in subcritical and supercritical regions. The Journal of Supercritical Fluids 31, 185–193. Völker, S., Rieckmann, Th., 2002. Thermokinetic investigation of cellulose pyrolysis– impact of initial and final mass on kinetic results. Journal of Analytical and Applied Pyrolysis 62, 165–177. Watchararuji, K., Goto, M., Sasaki, M., Shotipruk, A., 2008. Value-added subcritical water hydrolysate from rice bran and soybean meal. Bioresource Technology 99, 6207–6213. Yoshida, H., Tavakoli, O., 2004. Sub-critical Water Hydrolysis Treatment for Waste Squid Entrails and Production of Amino Acids, Organic Acids, and Fatty Acids. Journal of Chemical Engineering of Japan 37 (2), 253–260. Zhang, L., Champagne, P., Xu, C., 2011. Bio-crude production from secondary pulp/ paper-mill sludge and waste newspaper via co-liquefaction in hot-compressed water. Energy 36, 2142–2150.