Simultaneous production of bio-solid fuel and bio-crude from vegetal biomass using liquefied dimethyl ether

Simultaneous production of bio-solid fuel and bio-crude from vegetal biomass using liquefied dimethyl ether

Fuel 116 (2014) 370–376 Contents lists available at ScienceDirect Fuel journal homepage: www.elsevier.com/locate/fuel Simultaneous production of bi...

1MB Sizes 0 Downloads 22 Views

Fuel 116 (2014) 370–376

Contents lists available at ScienceDirect

Fuel journal homepage: www.elsevier.com/locate/fuel

Simultaneous production of bio-solid fuel and bio-crude from vegetal biomass using liquefied dimethyl ether Peng Li ⇑, Hideki Kanda 1, Hisao Makino Energy Engineering Research Laboratory, Central Research Institute of Electric Power Industry (CRIEPI), Kanagawa 240-0196, Japan

h i g h l i g h t s  DME method has robust dewatering ability for the high-moisture vegetal biomass.  This method also efficiently removes the organic bio-crude from the vegetal biomass.  The properties of dewatered biomass should be suitable for used as bio-solid fuel.  DME method can efficiently extract lipids from lipid-rich vegetal biomass as same as hexane Soxhlet.

a r t i c l e

i n f o

Article history: Received 7 May 2013 Received in revised form 7 August 2013 Accepted 7 August 2013 Available online 23 August 2013 Keywords: Vegetal biomass Solvent extraction Dewatering Bio-solid fuel Dimethyl ether

a b s t r a c t Liquefied dimethyl ether (DME) was used for moisture and bio-crude extraction in 4 common vegetal biomasses. The process removed approximately 81.3–88.7% of the water from these biomasses and yielded 5.8–16.8% bio-crude of dry weight of sample. The properties of the original samples, dewatered bio-solids, extracted bio-crude and removed water were studied by elemental analysis, gel permeation chromatography (GPC), and gas chromatography-mass spectrometry (GC–MS), respectively. In addition, the oxygen activity of the bio-solids was investigated. The results indicate that both bio-solids and biocrude could be further exploited for energy production and other purposes. Ó 2013 Elsevier Ltd. All rights reserved.

1. Introduction Fossil fuel depletion and global warming problem have led to the development of alternative energy source, particularly by utilising renewable resources such as municipal waste and agricultural vegetal biomass [1,2]. However, the vegetal biomass generally contains organic components (the so-called bio-crude) and excessive water, which cause environmental problems and requiring capital-intensive methods of biofuel production. Currently, the main methods for biofuel production from vegetal biomass are gasification and/or liquefaction [3], and/or making them bio-solid as refuse-derived fuels (RDF) [4]. The biomass with or without drying is usually treated by a physical or chemical method to recover their combustible chemical components such as lipids and hydrocarbons. An example of these methods is pyrolysis, which can convert biomass into useful liquid and gaseous fuel and solid char. However, the high energy consumption is still a ⇑ Corresponding author. Tel.: +81 46 856 2121; fax: +81 46 856 3346. E-mail address: [email protected] (P. Li). Present address: Department of Chemical Engineering, Nagoya University, Nagoya 464-8603, Japan. 1

0016-2361/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.fuel.2013.08.020

drawback of pyrolysis due to the large amount of heat required for this process [5]. RDF, on the other hand, has various merits such as easy transportation and combustible stability. However, direct combustion of RDF results in waste of value-added chemicals in the biomass and the production of environmental pollutants such as dioxins and polychlorinated biphenyls (PCBs) [6]. In previous studies, we have developed a new method, which uses liquefied dimethyl ether (DME) as extractant to remove the water from wet materials. In addition, simultaneously with dewatering, DME can also extract the bio-crude from the materials because DME is an organic solvent and can dissolve a wide range of organic components. Thus far, this method has been studied on both laboratory and bench scale and reported on (i) drying of low-rank coal at normal and frozen temperature [7,8]; (ii) removal of PCBs from polluted soil [9]; (iii) extraction of lipids and hydrocarbons from highmoisture microalgae [10–12] and (iv) extraction of bioactive components from the green tea leave [13]. Our institute is also attempting to carry out a pilot-scale study on this technology. The advantages of the proposed method are that it is high energy-efficient and environmentally friendly, with its principle being based on the following DME characteristics: (i) High affinity

P. Li et al. / Fuel 116 (2014) 370–376

to organic compositions and partial miscibility with water. The phase equilibrium relationship for the liquefied DME/water system is presented in detail in a previous paper [14]. At normal temperature, water is soluble in DME in the range of 7–8 wt.% [14]. Therefore, the weight of DME required for the extraction of water is 1/(0.07 0.08) times the weight of water. (ii) The standard boiling point of DME is 24.8 °C, and it exists in a gaseous state under normal conditions. Its saturated vapour pressure is mentioned and explained in detail in Ref. [15]; for example, it is 0.51–0.59 MPa in the normal temperature range (20–25 °C). (iii) The European Food Safety Authority concluded that there are no safety concerns with regards to the use of DME as an extractant in food processing [16]. The principle of energy-saving of this process is shown in Fig. 1. First, the process involves mixing the water/organic componentbearing materials with liquefied DME and subsequent extraction of water and bio-crude from the materials using DME. In the next step, DME in the mixture was evaporated at around room temperature and then the bio-crude are separated from water in the aftertreatment process. For large-scale application, a heating source of unharnessed waste heat of about 40 °C is desirable for DME evaporation. DME gas is then liquefied again at a slightly lower temperature for recirculation. In this stage, a cooling source of about 10 °C such as geo-heat, which is the temperature within the first 50 m of the Earth’s surface, is desirable [17,18]. Taking all these factors into consideration, we performed experiments to investigate the applicability of this method for dewatering and extraction of bio-crude for high-moisture vegetal biomass. In this study, 4 representative common vegetal biomasses including spent coffee ground, tea leaf waste, orange peel and gramineous weed were selected for testing. The former three are main industrial food wastes with very huge production worldwide, thus, the effective utilisation of them is important. For example, the world annually produces spent coffee grounds of around 6 million tons from the beverage factory [19]. According to a credible report, the spent coffee grounds contain around 10–15% oily substance depending on the coffee species, which can be easily converted to biodiesel [20]. As an unexplored biomass resource, the gramineous weed is an herbaceous plant which can be seen and available anywhere, therefore, it is an ideal sample to investigate the performance of DME method on such vegetal biomass. In this paper, the properties of dewatered bio-solids, extracted bio-crudes and removed water were analysed and evaluated. In addition, the oxygen reactivity of the dewatered bio-solids was examined. 2. Experimental 2.1. Materials The spent coffee grounds (pulverized coffee bean) were supplied by a Japanese brewing company. The species of tea leaves

Fig. 1. Schematic illustration of the DME extraction method.

371

is Camellia sinensis obtained from a producing plant in Shizuoka, Japan. The initial water content and profile of the tested biomasses before and after DME extraction and the extracted bio-crudes are shown in Fig. 2. Because the size of the coffee grounds was larger than that of the extraction column, the coffee grounds were again pulverised to particle size to be able to load them into the extraction column. Gramineous weed and orange peel were cut into cylinders and rectangles. The tea leaves were loaded into the column in their received form. The following sizes of samples were average values obtained from 10 gauging stations, with maximum and minimum values. The diameter of pulverised coffee grounds was 2.5 (3.6–1.5) mm. The as-received tea-leaf sizes were between 1.0 (0.5–1.7) mm and was 13.6 (7.0–19.0) mm, and the thickness was 0.5 (0.1–1.1) mm. Orange peel was cut into sizes between 6.3 (4.8–7.5) mm and 19.2 (13.0–28.0) mm, and the thickness was 2.3 (1.3–3.1) mm. Gramineous weed was cylindrically cut into a diameter of 4.3 (between 1.5 and 5.2) mm, and the length was 28.1 (22.4–33.7) mm, and the thickness was 2.3 (1.3–3.1) mm. 2.2. Outline of extraction method The experimental apparatus has been described in detail in our previous paper [10]. Fig. 3 is a schematic diagram of the Lab-scale DME extraction apparatus. It consists of 2 main parts: an extraction column (diameter, 11.6 mm; length, 190 mm; HPG-10-5, Taiatsu Techno Corp., Saitama, Japan) and a storage vessel for the mixture of DME, water, and bio-crude (HPG-96-3, Taiatsu Techno Corp.). The extraction column was loaded with the test sample. On average of 3 tests, the amounts of spent coffee ground, tea leaf waste, orange peel and gramineous weed were 6.85 (±0.1) g, 4.84 (±0.2) g, 5.11 (±0.2) g and 5.21 (±0.2) g, respectively. At the column outlet, extracted bio-crude passed through a filter (pore diameter < 0.65 lm). The DME flow rate was 10 (±1) cm3 min 1, and the extraction temperature and pressure were 20 °C and 0.51 MPa, respectively. 2.3. Property analysis The proximate analysis, ultimate analysis, and high heating value (HHV) measurements were carried out for the original dry biomass samples and the bio-solids obtained from DME extraction. The analytical methods were according to the Japanese Industrial Standard (JIS) and were the same as those used in our previous study [10]. The biomass samples and dewatered bio-solids were pre-treated for this analysis by continual drying at 105 °C up to constant weight in vacuum. The concentrations of minor elements, Na and K, in the biocrudes and removed water were determined by flame atomic absorption spectrometry (Z-2000; Hitachi High-Technologies Corp., Tokyo, Japan) according to the Japanese industrial standard methods JIS K 0102 48.2 (Na) and 49.2:2011 (K). The Ca and Mg concentrations were determined by inductively coupled plasma atomic emission spectroscopy (Optima3300XL; Perkin Elmer Inc., USA) according to JIS K 0102 50.3 (Ca) and 51.3:2011(Mg). The P concentrations in the bio-crudes were determined by sulphuricperchloric acid digestion molybdenum blue absorption spectroscopy (U-2001; Hitachi High-Technologies Corp., Tokyo, Japan) according to the analytical method JIS K 0102 46.3.2:2011. The P concentrations in the removed water were determined by potassium peroxodisulphate digestion molybdenum blue absorption spectroscopy (U-2001) according to the analytical method JIS K 0102 46.3.1:2011. The N concentrations in the removed water were determined by the copperised cadmium column method according to the analytical method JIS K 0102 45.4:2011. The molecular weight distributions (MWDs) of the bio-crudes were determined by gel permeation chromatography (GPC) per-

372

P. Li et al. / Fuel 116 (2014) 370–376

Fig. 2. Pictures of original samples, dewatered bio-solids and extracted bio-crudes. The value in parentheses is the initial water content of biomass.

Fig. 3. Schematic diagram of the apparatus for the DME extraction.

formed at 40 °C by diluting the bio-crudes in either chloroform or tetrahydrofuran (THF), which are solvents typically used for GPC measurements. A bio-crude concentration of 0.1% (wt./vol.) was used for each measurement, and the injection volume was 100 lL. When chloroform was used, a SC-8010 controller and a refractive index detector (Toso Co., Tokyo, Japan) with Shodex K800D chromatographic columns and 2K-805L (Showa Denko K.K., Tokyo, Japan) were employed. For THF, a HLC-8220GPC controller and a refractive index detector (Toso Co., Tokyo, Japan) with a Shodex KF-G chromatographic column and KF-805L and KF-800D (Showa enko K.K., Tokyo, Japan) were used. Gas chromatography-mass spectrometry (GC–MS) was performed on a GC: 6890N, MS: 5975B system (Agilent Technologies

Inc., CA, USA. An HP-1 column was used (15 m  0.25 mm i.d.; Agilent Technologies Inc.). The temperature program of GC analysis is as follows: an oven temperature of 50 °C for 2 min, then increased from 50 °C to 120 °C at a rate of 15 °C min–1.; thereafter, the oven temperature was increased from 120 °C to 270 °C at a rate of 25 °C min 1, and then maintained at 270 °C for 3 min. The extracted bio-crudes (0.17 g) were diluted with 10 mL of n-hexane, and then ultrasonicated for 30 min for analysis. The mass scanning ranged between m/z 50 and m/z 500. Further investigation of the compositions of bio-crude derived from coffee grounds was carried out. In brief, the bio-crude was directly chemical converted to corresponding methyl esters and then analysed by GC–MS. Qualitative analysis of the detected fatty acids methyl esters (FAMEs) were conducted by comparison of their retention time and mass spectra with standard chemicals (Supelco 37 Comp. FAME Mix. Sigma–Aldrich St. Louis USA). To study the safety in the transportation and conservation of the dewatered bio-solids, we examined the oxygen reactivity of the samples before and after DME extraction with a spontaneous ignition tester (SIT-2; Shimazu Co., Ltd., Kyoto, Japan). The sample weights were 0.8 g. At the pre-processing stage, dry nitrogen gas at 50 °C was passed through the samples for 4 h to ensure that the samples are dried in a steady state. After the pre-processing stage, we supplied dry air instead of nitrogen and measured the temperature increase by oxidation of the samples. During the measurement, the temperature indicator was used under an adiabatic condition by synchronously increasing circumference temperature. Here, the time duration which it took to reach 250 °C was used as the time index for oxygen reactivity.

P. Li et al. / Fuel 116 (2014) 370–376

373

3. Results and discussion 3.1. Dewatering and bio-crude extraction using the DME method in various vegetal biomasses As shown in Fig. 2, the biomasses were atrophic and their surface became bright after DME dewatering due to the water and pigment compositions contained in the samples were decreased substantially. In addition, this process also yielded the bio-crude as a paste form as shown in Fig. 2. Because of the differences in the biological properties of the tested biomasses, the efficiency of DME extraction on these biomasses could not be directly compared. However, for a comprehensive analysis, the dewatering and bio-crude extraction efficiencies of liquefied DME on these biomasses were integrated together as shown in Figs. 4 and 5. Here, the DME consumptions (abscissa) for dewatering and extraction are expressed as the ratio of DME consumption to the total biocrude yield and initial water amount of the tested samples. As shown in Fig. 4, the bio-crude extraction rates of spent coffee ground, tea leaf waste, orange peel and gramineous weed at maximum were 16.8 ± 1.0%, 16.2 ± 1.5%, 6.2 ± 0.5%, and 5.8 ± 0.5%, respectively. Apparently, the difference in the bio-crude extraction rate is due to the differences in extractable bio-crude contents and chemical compositions of these biomasses. However, as shown in Fig. 5, the differences in the final dewatering rate of these biomasses are very small, between 81.3% and 88.7%, although their initial water contents are quite different, between 42.0% and 79.0%. As a comparison, the bio-crude extraction yields from these biomasses were tested by using the conventional solvent hexane with a Soxhlet extractor. As shown in Table 2, the extraction yields of bio-crudes using the DME method relative to the hexane Soxhlet method for spent coffee ground, tea leaf waste, orange peel and gramineous weed were 97.6%, 852.6%, 688.8% and 214.8% respectively. The differences in bio-crude extraction yield between the two methods are due to the different chemical compositions of the tested biomasses and the different chemical properties of DME and hexane as extraction solvents. The main composition of spent coffee ground is lipids [21], which can be extracted by both liquefied DME and hexane facilely. On the other hand, the tea leaf waste, orange peel and gramineous weed contain more polar molecular components such as polyphenols [22], which are apt to be extracted by DME than by hexane. Another possible cause

Fig. 4. Extraction of bio-crude from vegetal biomasses using liquefied DME.

Fig. 5. Removing of water from vegetal biomasses using liquefied DME. The value in parentheses is the initial water content of biomass.

of the low extraction yield of tea leaf waste, orange peel and gramineous weed by the hexane Soxhlet method is that the volatile substances were lost during the high temperature and long-time condition used in the method [23]. On the other hand, the DME method has the advantage of requiring only room temperature for such biological materials; therefore, this process can avoid volatilisation and degradation of target chemicals. The results of dewatering and bio-crude extraction revealed that these highmoisture vegetal biomasses can be efficiently dewatered by the proposed method. In addition, the DME method has similar lipid extraction yield to that of hexane Soxhlet for coffee grounds, and can also extract more products from other three vegetal biomasses compared with hexane Soxhlet.

3.2. Properties of bio-solids obtained by DME extraction The results of the proximate analysis and the main elemental compositions of the vegetal biomasses before and after DME extraction are shown in Table 1. Among these biomasses, the raw coffee grounds had the highest volume of volatile matter, carbon, hydrogen and HHV and lower fixed carbon and oxygen because of the high lipid content. All these values were reduced obviously after DME extraction implying that no oily substance remained after DME treatment. The HHVs of original moist biomasses were estimated at between 3.93 and 14.0 MJ kg 1 based on their water content. The HHVs of the bio-solids were between 18.1 and 21.1 MJ kg 1, indicating the possibility of utilising these dewatered biomasses as carbon neutral fuel. The bio-solid from tea leaf waste had the highest remaining N and S contents of 4.05% and 0.28%; in contrast, the remaining N and S contents in other dewatered biosolids ranged between 0.96% and 2.43% and between 0.07% and 0.22%, respectively. In our previous study, the N and S remaining in the dewatered coal were 1.15% and 0.18% [7]. As comparison, the concentration of N in the bio-solids is higher than that of the coal about twofold, thus, the NOx emission of these dewatered biomasses should be further investigated and considered, although they have sufficient caloric density used as bio-solid fuel blended combustion with coal. The oxygen reactivity of the bio-solids by DME extraction was examined, and their adiabatic temperature profile was compared to that of hot dried raw samples, as shown in Fig. 6. Curves coloured black, red, blue and orange represent spent coffee ground,

374

P. Li et al. / Fuel 116 (2014) 370–376

Table 1 Proximate analysis and main elemental compositions of the vegetal biomass before and after DME extraction. Analysis (wt.% dry basis)

Proximate analysis Ash yield (±0.1) Volatile matter (±0.4) Fixed carbon (±0.4) Ultimate analysis C (±0.3) H (±0.1) N (±0.4) O (±0.6) S HHV (MJ kg 1)

Coffee

Tea

Orange

Weed

B

A

B

A

B

A

B

A

0.7 84.4 14.9

1.9 81.1 17.0

4.7 71.6 23.7

5.3 78.0 16.7

2.4 77.0 20.6

2.3 81.5 16.2

7.0 76.7 16.3

6.9 77.7 15.4

55.7 8.12 2.03 33.4 0.13 24.3 14.0a

51.5 6.97 2.43 37.1 0.15 21.1

52.7 6.34 3.91 32.3 0.24 21.2 5.10a

48.8 6.43 4.05 35.4 0.28 20.0

48.6 6.55 0.82 41.6 0.07 18.7 3.93a

46.3 6.59 0.96 43.9 0.07 18.1

48.7 6.78 2.05 35.4 0.21 20.1 8.44a

46.9 6.55 2.10 37.4 0.22 18.8

B: vegetal biomass before DME extraction. A: bio-solid obtained from DME extraction. Data were means of triplicate results with standard error in parentheses. a High heat value of moist sample by calculation.

Table 2 Extraction yields of bio-crudes compared to the hexane Soxhlet method, ultimate analysis of extracted bio-crudes and concentration of minor elements in the extracted biocrudes and removed water. Analysis (wt.% dry basis)

Coffee

Tea

Orange

Weed

BC

RW

BC

RW

BC

RW

BC

RW

Extraction ratio based on hexane (%)a Ash yield (±0.1)

97.6 LOQ

– –

852.6 0.4

– –

688.8 NS

– –

214.8 0.1

– –

Ultimate analysis C (±0.3) H (±0.1) N (±0.4) O (±0.6) S

77.0 11.40 1.96 9.6 LOQ

– – – – –

68.4 7.59 5.68 17.7 LOQ

– – – – –

NS NS NS NS NS

– – – – –

80.7 12.1 3.60 3.4 LOQ

– – – – –

Minor elements (ppm) Na K Mg Ca P N (in water) HHV (MJ kg 1)

210 LOQ LOQ LOQ 110 – 38.9

100.4 40.5 7.0 10.6 9.0 LOQ –

110 210 95 LOQ 3300 – 31.8

200.4 440.2 10.6 10.3 280.2 50.4 –

49 LOQ LOQ LOQ 990 – 40.8

80.7 30.2 5.0 10.1 10.2 50.6 –

89 LOQ 1900 LOQ 160 – 42.1

250.2 130.9 820.0 20.5 10.3 30.9 –

BC: bio-crude; RW: removed water; NS: not enough samples for the analysis; LOQ: data below the limit of quantification. HHV: high heat value. a Extraction ratio of bio-crude using DME based on hexane.

gramineous weed, tea leaf waste and orange peel, respectively. The temperature increase caused by oxidation of the residues was faster than that of hot dried samples, except for the gramineous weed for which the adiabatic temperature did not change up to 10,000 min. Considering the above results, pelletising of the biosoild will be necessary to prevent hypergolicity during their transportation or long-time storage. 3.3. Properties of extracted bio-crude and removed water Two different solvents, chloroform or THF, were used for the GPC measurements of the MWDs of the bio-crude, as shown in Fig. 7. In the measurement using chloroform, the weight average molecular weight (Mw) and the number average molecular weight (Mn) of spent coffee ground, gramineous weed, tea leaf waste and orange peel were 1400/1200, 1100/800, 700/400 and 1400/ 1400 g mol 1, respectively. In the measurement using THF, the Mw and Mn were 1300/900, 1200/700, 1100/1000 and 1100/ 600 g mol 1, respectively. The molecular weight distribution curves differed for chloroform and THF due to the differences in

solvency and chemical properties of these solvents such as polarity. The results of the GC–MS analysis showed that the detectable chemical compositions of the bio-crudes derived from these biomasses were quite complex and different. The main compositions of bio-crude extracted by DME method from green tea leaf have been reported in our previous study [13]. In this study, only the compositions of bio-crude derived from coffee grounds were determined. As shown in Fig. 8, the gas chromatogram of the bio-crude obtained via DME extraction resembled those of by hexane Soxhlet extraction, the predominant components were lipids with carbon number from C16 to C18 which consists of both saturated and unsaturated fatty acids. This result is almost identical to a previous report on the crude oil derived from used coffee grounds [20]. The above result indicates that the bio-crude from lipid-rich feedstock by the DME method could be used for further production of refined biofuel such as biodiesel [24]. The main elemental compositions of extracted bio-crudes are shown in Table 2. They consisted mainly of C, H and O. The large HHV of the bio-crude derived from gramineous weed was due to

P. Li et al. / Fuel 116 (2014) 370–376

375

Fig. 8. Gas chromatogram of bio-crude derived from coffee grounds via DME and hexane Soxhlet extraction.

Fig. 6. Adiabatic temperature increase by oxidation of biomass bed by air at 50 °C. Broken and solid curves represent adiabatic temperature of dewatered residues and hot dried samples respectively.

these vegetal biomasses ranged between 49 and 210 ppm and between 110 and 3300 ppm, respectively. The extremely high concentration of P in the bio-crude from tea waste was due to the presence of phosphorus-containing compounds in the tea leaf [25]. The analytical results presented above indicate that further purification is required if the extracted bio-crude is to be used for the production of refined fuel. The concentrations of Na, K, Mg, Ca and P in the removed water from these vegetal biomasses were in the range of 80.7–250.2, 30.2–440.2, 5.0–820.0, 10.1–20.5 and 9.0–280.2 ppm, respectively. The concentrations of these minor elements in removed water are not suitable for direct domestic use; the concentration limits of World Health Organization [26] for domestic use of Na, K, Mg and Ca are 50–200, 100–200, 30–150 and 75–200 ppm, respectively. However, considering that the water is removed from vegetal feedstock and there is no remaining of DME in water at normal temperature, the removed water may be suitable for use in agricultural irrigation [27].

4. Conclusions

Fig. 7. Molecular weight distribution curve of the extracted bio-crudes. The solvent used for A was chloroform, and that used for B was THF.

its higher C and H and lower O contents. The S content in the biocrudes derived from these biomasses was below the limit of quantification (LOQ), and the N content ranged between 1.96% and 5.68%. The concentrations of minor elements Na, K, Ca, Mg and P in both extracted bio-crudes and removed water were determined. The LOQ for K and Ca was 500 ppm and that for Mg was 50 ppm. The concentrations of K, Ca and Mg in bio-crudes derived from all these vegetal biomasses were below the LOQ, except for the Mg concentration in the bio-crude from tea waste, which was 95 ppm, and that from gramineous weed, which was 1900 ppm. The concentrations of Na and P in the bio-crudes derived from

The DME method exhibits robust dewatering ability and efficiency in removing the bio-crude for the tested high-moisture vegetal biomasses. Taken together, the results of the ultimate analysis and HHV analyses reveal that the dewatered biomass should be suitable for use as bio-solid fuel. However, the spontaneous combustion of the bio-solids should be considered, as only dewatered gramineous weed is stable in the oxygen reactivity test. In addition to its use in the production of bio-solid fuel, the DME method can efficiently extract lipids from coffee grounds, implying its utilisation to produce bio-liquid fuel from lipid-rich biomass. Evaluation of the concentrations of minor elements in the removed water exhibits the possibility of using the removed water for agricultural purposes. In future, the transesterification of extracted oil to biodiesel, and application of DME method in the extraction of other value-added by-products from vegetal biomass will be investigated, and a pilot-scale study will also be reported from the viewpoint of energy balance.

Acknowledgements This research was supported by the Environment Research and Technology Development Fund (No. K-112033, H. Kanda and P. Li) of the Ministry of the Environment, Japan. The authors would like to express their gratitude to Dr. Otaka, Mr. Shoji, and Mr. Sakuragi who in the same research group of CRIEPI for providing the coffee grounds sample and assistance in the GC–MS analysis.

376

P. Li et al. / Fuel 116 (2014) 370–376

References [1] Li F, Zeng L, Fan L-S. Biomass direct chemical looping process: process simulation. Fuel 2010;89:3773–84. [2] Demirbas A. Political, economic and environmental impacts of biofuels: a review. Appl Energy 2009;86:S108–17. [3] Swain PK, Das LM, Naik SN. Biomass to liquid: a prospective challenge to research and development in 21st century. Renew Sustain Energy Rev 2011;15:4917–33. [4] Sakka M, Kimura T, Sakka K. Comparison of microbial consortia in refusederived fuel (RDF) preparations between Japan and Germany. Biosci Biotechnol Biochem 2006;70:2868–73. [5] Lam K-L, Lee C-W, Hui C-W. Multi-stage waste tyre pyrolysis: an optimization approach. Chem Eng Trans 2010;21:853–8. [6] Lee J-M, Kim D-W, Kim J-S, Na J-G, Lee S-H. Co-combustion of refuse derived fuel with Korean anthracite in a commercial circulating fluidized bed boiler. Energy 2010;35:2814–8. [7] Kanda H, Makino H. Energy-efficient coal dewatering using liquefied dimethyl ether. Fuel 2010;89:2104–9. [8] Kanda H, Li P, Makino H. Direct water removing from frozen brown coal at 20 °C. Fuel 2013;111:880–882. [9] Oshita K, Kitade M, Takeda N, Kanda H, Makino H, et al. Extraction of PCBs and water from river sediment using liquefied dimethyl ether as an extractant. Chemosphere 2010;78:1148–54. [10] Kanda H, Li P. Simple extraction method of green crude from natural bluegreen microalgae by dimethyl ether. Fuel 2011;90:1264–6. [11] Kanda H, Li P, Ikehara T, Yasumoto-Hirose M. Lipids extracted from several species of natural blue-green microalgae by dimethyl ether: extraction yield and properties. Fuel 2012;95:88–92. [12] Kanda H, Li P, Yoshimura T, Okada S. Wet extraction of hydrocarbons from Botryococcus braunii by dimethyl ether as compared with dry extraction by hexane. Fuel 2013;105:535–9. [13] Kanda H, Li P, Makino H. Production of decaffeinated green tea leaves using liquefied dimethyl ether. Food Bioprod Process 2013. [in press]. [14] Holldorff H, Knapp H. Binary vapour-liquid-liquid equilibrium of dimethyl ether-water and mutual solubilities of methyl chloride and water: experimental results and data reduction. Fluid Phase Equilib 1988;44: 195–209.

[15] Wu J, Zhou Y, Lemmon EW. An equation of state for the thermodynamic properties of dimethyl ether. J Chem Eng Data 2011;40:023104. [16] Scientific Opinion of the Panel on Food Contact Materials, Enzymes, Flavourings and Processing aids (CEF) on dimethyl ether as an extraction solvent. EFSA J 2009; 984:1–13. doi:10.2903/j.efsa.2009.984. [17] Kanda H, Makino H, Morita M, Takegami K, Toshikoshi A, Takahashi M. Hydrous matter treating system. Patent number: JP 4 748 154 B2 WO2008/ 111483, Japan; 2008. [18] Delrue F, Setier P-A, Sahut C, Cournac L, Roubaud A, Peltier G, et al. An economic, sustainability, and energetic model of biodiesel production from micro algae. Bioresour Technol 2012;111:191–200. [19] Mussatto SI, Carneiro LM, Silva JPA, Roberto IC, Teixeira JA. A study on chemical constituents and sugars extraction from spent coffee grounds. Carbohydr Polym 2011;83:368–74. [20] Kondamude N, Mohapatra SK, Misra M. Spent coffee grounds as a versatile source of green energy. J Agric Food Chem 2008;56:11757–60. [21] Couto RM, Fernandes J, da Silva MDRG, Simoes PC. Supercritical fluid extraction of lipids from spent coffee grounds. J. Supercrit. Fluids 2009;51:159–66. [22] Wang L, Gong L-H, Chen C-J, Han H-B, Li H-H. Column-chromatographic extraction and separation of polyphenols, caffeine and theanine from green tea. Food Chem 2012;131:1539–45. [23] Bimakr M, Rahman RA, Taip FS, Ganjloo A, Salleh LM, Selamat J, et al. Comparison of different extraction methods for the extraction of major bioactive flavonoid compounds from spearmint (Mentha spicata L.) leaves. Trans Ind Chem Eng C 2011; 89: 67–72. [24] Niza NM, Tan KT, Lee KT, Ahmad Z. Biodiesel production by non-catalytic supercritical methyl acetate: thermal stability study. Appl Energy 2013;101:198–202. [25] Lin Z-H, Qi Y-P, Chen R-B, Zhang F-Z, Chen L-S. Effects of phosphorus supply on the quality of green tea. Food Chem 2012;130:908–14. [26] World Health Organisation, Geneva, Guidelines for drinking-water quality. Recommendations 1997; 1: 1–4. [27] Singh AK, Mahato MK, Neogi B, Tewary BK, Sinha A. Environmental geochemistry and quality assessment of mine water of Jharia coalfield, India. Environ Earth Sci 2012;65:49–65.