- Email: [email protected]
Pyrolytic characteristics of microalgae as renewable energy source determined by thermogravimetric analysis
Bioresource Technology 80 (2001) 1±7
Pyrolytic characteristics of microalgae as renewable energy source determined by thermogravimetric analysis Weimin Peng, Qingyu Wu *, Pingguan Tu, Nanming Zhao Department of Biological Sciences and Biotechnology, Tsinghua University, Beijing 100084, People's Republic of China Received 18 December 2000; received in revised form 9 April 2001; accepted 10 April 2001
Abstract Two kinds of autotrophic microalgae, Spirulina platensis (SP) and Chlorella protothecoides (CP) were pyrolyzed at the heating rates of 15, 40, 60 and 80°C/min up to 800°C in a thermogravimetric analyzer to investigate their pyrolytic characteristics. Three stages (dehydration, devolatilization and solid decomposition) appeared in the pyrolysis process. SP and CP mainly devolatilized at 190±560°C and 150±540°C, respectively. A total volatile yield of about 71% was achieved from each microalga. As the heating rate increased, a lateral shift to higher temperatures was observed in their thermograms, and the instantaneous maximum and average reaction rates in the devolatilization stage were increased while the activation energy was decreased. The value of activation energy for CP pyrolysis was 4:22±5:25 104 , lower than that of SP 7:62±9:70 104 , and the char in ®nal residue of CP was 14.00±15.14%, less than that of SP by 2±3%. This indicated that CP is preferable for pyrolysis over SP. The experimental results may provide useful data for the design of pyrolytic processing systems using planktonic microalgae as feedstock. Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: Spirulina platensis; Chlorella protothecoides; Pyrolysis; Bio-fuel; Eutrophication
1. Introduction Pyrolysis is the degradation of macromolecular materials with heat alone in the absence of oxygen (Meier and Faix, 1999). As a good renewable energy source, biomass has been considered as a potential feedstock for pyrolysis to produce fuels including oil and gas, which can be used as substitutes for petroleum or natural gas for internal-combustion engines, power stations and heating supplies (Raveendran and Ganesh, 1996a). These bio-fuels are cleaner than fossil fuels such as coal and petroleum because of their low nitrogen and sulfur contents (Churin and Delmon, 1989; Ahuja et al., 1996). Many researches on biomass pyrolysis have been carried out to meet the energy demand during the past two decades (Maschio et al., 1992; Meier and Faix, 1999). Thermogravimetric analysis (TGA) is extensively used to understand the pyrolytic characteristics and determine the kinetic parameters (Williams and Besler, 1993; Mansaray and Ghaly, 1998). Many TGA studies *
Corresponding author. Tel.: +86-10-6278-1825; fax: +86-10-62781825. E-mail address: [email protected] (Q. Wu).
on the pyrolytic kinetics of wastes from domestic, industrial and agricultural activities have been reported (Williams and Besler, 1993; Ahuja et al., 1996; Raveendran et al., 1996b; Encinar et al., 1997; Wu et al., 1997; Mansaray and Ghaly, 1998, 1999; Sharma and Rao, 1999). However, there are only a few reports concerning the pyrolytic behaviors of microalgae. Cyanobacterium Spirulina platensis (SP) and green alga Chlorella protothecoides (CP) are two kinds of microalgae arti®cially cultured worldwide for commercial goals of producing food, feed, bait and ®ne chemicals (Oh-Hama and Miyachi, 1988; Richmond, 1988). However, their potential for fuel production is neglected. To determine the eect of temperature and heating rate on their pyrolytic characteristics, the samples were pyrolyzed under non-isothermal conditions, which are more practical and reasonable in simulating large-scale pyrolysis processes (Dogan and Uysal, 1996), in a thermogravimetric analyzer. The kinetic parameters including activation energy, frequency factor and reaction order were measured. These data are important to the ecient design, operation, and modeling of pyrolysis and related thermochemical conversion systems for microalgae.
0960-8524/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 0 - 8 5 2 4 ( 0 1 ) 0 0 0 7 2 - 4
2
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
2. Methods 2.1. Algae culture and sample preparation The strain of SP was furnished by Prof. Shang, China Agricultural University (Beijing, Poeple's Republic of China). The strain of CP was obtained from the Culture Collection of Algae at the University of Texas (Austin, USA). SP was grown in Zarrouk's medium (Zarrouk, 1966) at 28°C with continuous illumination of about 60 lE= m2 s. The culture was agitated by aeration. The preparation of medium for CP was described as before (Wu et al., 1992, 1996). CP was axenically grown at 23°C with continuous illumination of 40 lE= m2 s. The culture was also agitated by aeration. The algal cells from a late log phase were harvested by centrifugation and washed one time with distilled water and then dried under infrared light at 30°C. Samples were prepared by pulverization in a mortar to be small enough to eliminate heat transfer eects during the pyrolysis process and then stored in a desiccator.
monly known as the Freeman±Carroll method. Generally the reaction rate da=dt (%/min) for a decomposition reaction is described as da=dt A exp E=RT 1
an ;
1
where a is the conversion of reactant, t (min) is the time, A min 1 is the pre-exponential factor (frequency factor), E (J/mol) is the activation energy, R (J/mol K) is the universal gas constant (8.314), T (K) is the temperature, n is the reaction order. As the heating rate b (°C/ min) is dT =dt, the general non-isothermal kinetic equation is given as da=dT A=b exp E=RT 1
an ;
2
where da=dT is the instantaneous reaction rate (weightloss rate, %/°C), which can be directly obtained from DTG. The logarithmic form for Eq. (2) is ln da=dT ln A=b
E=RT n ln 1
a:
3
For two dierent T and the corresponding a within the same pyrolysis zone,
2.2. Quantitation of the main chemical components in cells
D ln da=dT
The contents of crude protein, crude fat and carbohydrate were determined by Kjeldehl method, Soxhletextract method and phenol-sulfuric acid method, respectively (Kochert, 1978a,b). Protein content was calculated using a factor of 6.25. The solution for lipid extraction was CHCl3 . Glucose was used for calibration during the determination of carbohydrate. The moisture content was determined from the weight loss on heating at 105°C for 24 h, and the ash content was determined as the residue after ignition at 600°C for 3 h. The measurements were replicated three times.
where D means the dierence of two values. The following equation is obtained:
2.3. Pyrolysis in a thermogravimetric analyzer The algal cells were subjected to TGA in a nitrogen atmosphere at heating rates of 15, 40, 60 and 80°C/min from ambient temperature to 800°C in a Dupont 2100 Thermal Analyzer (Dupont TA Instruments, 109 Lukens Drive, New Castle, DE, USA). All the TGA experiments at dierent heating rates were replicated three times. A high-purity nitrogen gas (99.99%) was fed at a constant ¯ow rate of 60 ml/min as an inert purge gas to displace air in the pyrolysis site, thus avoiding unwanted oxidation of the sample. The continuous on-line records of weight loss and temperature were obtained to plot the TGA curve and the derivative thermogravimetric analysis (DTG) curves. 2.4. Kinetic parameters The kinetic parameters for the main pyrolysis process could be calculated by the following equations com-
E=RD 1=T nD ln 1
D ln da=dT =D ln 1
a
E=RD 1=T =D ln 1
a n;
a;
4
5
E and n can be obtained from a plot of D ln da=dT =D ln 1 a versus D 1=T =D ln 1 a that gives a straight line with a slope of E=R and an intercept of n. The value of A may be calculated by substituting the values of E and n back into the Eq. (5) in conjunction with the data of T, a and da=dT . 2.5. Data analyses The results on variance and coecient of determination R2 were obtained from the experimental data analyzed by the Excel 2000 software included in Oce 2000 package produced by Microsoft Corporation.
3. Results 3.1. Main chemical components in cells SP (Table 1) contained 10.30% crude lipid, which was less than that in CP (14.57%), whereas crude protein in SP (61.44%) was higher than that in CP (52.64%). The carbohydrate content in each sample was approximately 10.60%. The total amount of the three components in SP reached 82.31%, which was greater than that in CP (77.83%).
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
3
Table 1 Contents of the main chemical components in two microalgal cellsa
a b
Component (%)
S. platensis
C. protothecoides
Protein Lipid Carbohydrate Ash Moisture Othersb
61.44 0.31 10.30 0.10 10.57 0.13 4.34 0.03 8.04 0.06 5.31 0.63
52.64 0.26 14.57 0.16 10.62 0.14 7.38 0.05 8.74 0.07 6.05 0.68
Values are the average of three measurements. Calculated by dierence.
3.2. Thermal degradation process The TGA and DTG curves of two microalgae at dierent heating rates 15, 40, 60, 80°C/min in nitrogen (Figs. 1 and 2) revealed that three stages existed in the pyrolysis process, as diagramed in Fig. 3. The ®rst stage (Stage I) was from the starting temperature to the temperature of initial devolatilization Ti ; the second stage (Stage II) was from Ti to the end of the main devolatilization Te ; the third stage (Stage III) was from Te to the ®nal temperature (800°C). Te was taken as the point of equal value of instantaneous thermal degradation rate to Ti . The temperature of maximum reaction rate was referred as Tm . Fig. 2. The TGA and DTG curves of C. protothecoides at the heating rates of 15, 40, 60 and 80°C/min.
As shown in Figs. 1±3, a slight weight loss appeared in Stage I. It could be due to the elimination of water (dehydration). Stage II was characterized by a major weight loss, which corresponded to the main pyrolysis process (devolatilization). Most of the volatiles were released in this stage. It proceeded with a high rate and led to the formation of the pyrolysis products. During Stage III, a slow continued loss of weight revealed that the carbonaceous matters in the solid residue continuously decomposed at a very slow rate and the solid residue reached an asymptotic value. The instantaneous maximum and average reaction rates in Stage II for CP and SP were increased as the heating rate increased (Table 2). Both rates for SP were higher than those for CP at each heating rate. In addition, the average rate in the devolatilization stage was about twice as much as that of the whole process. 3.3. Temperature characteristics
Fig. 1. The TGA and DTG curves of S. platensis at the heating rates of 15, 40, 60 and 80°C/min.
The values of Ti ; Tm and Te at dierent thermograms are given in Table 3. There was a lateral shift to higher temperatures for almost all the temperature points in the thermograms of each microalga as the heating rate was increased (Figs. 1, 2 and Table 3). The decomposition of
4
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
Table 2 Reaction rates in the pyrolysis process of two microalgaea ;b Microalga
Heating rate (°C/min)
Devolatilization in stage II (%)
Instantaneous maximum reaction rate (%/min)
Average reaction rate (%/min) Stage II
Overallc
S. platensis
15 40 60 80
68:14 0:03 69:15 0:03 68:27 0:03 68:89 0:04
8.34 0.01 22.06 0.02 30.86 0.02 43.75 0.04
2.84 0.01 7.90 0.02 12.05 0.03 16.21 0.04
1:47 0:00 3:92 0:00 5:87 0:00 7:85 0:01
C. protothecoides
15 40 60 80
68:38 0:03 68:51 0:03 68:23 0:03 69:18 0:04
6.62 0.01 18.15 0.01 27.28 0.02 36.16 0.03
2.63 0.01 7.61 0.02 11.70 0.03 15.81 0.04
1:45 0:00 3:92 0:00 5:89 0:00 7:86 0:01
a
Values are the average of three measurements. The values of reaction rates (%/min) listed here are the products of reaction rates (%/°C) from DTG and the corresponding heating rates. c For the entire pyrolysis process. b
CP started earlier and also ended earlier than that of SP. The initial temperatures for decomposition of CP and SP at dierent heating rates were 150±190°C and 190± 220°C respectively, while the ®nal temperatures for CP and SP were approximately 540°C and 560°C, respectively (Table 3). This implied that the main pyrolysis reactions including depolymerization, decarboxylation and cracking took place over the temperature range 150±560°C. 3.4. Products of pyrolysis by thermogravimetric analysis The yields of initial moisture, volatile and ®nal residue obtained from TGA are presented in Table 4. The initial moisture yield was calculated as the weight loss in Stage I and the total volatile yield was calculated as the weight loss in both Stage II and Stage III. The initial moisture yields of SP and CP were 6.68±7.57% and 6.85± 7.64%, respectively. They were slightly lower than the moisture content shown in Table 1. The dierence may have been caused by the dierent heating times and temperatures. Each microalga gave a total volatile yield of about 71%. The thermal degradation in Stage III contributed little to the total volatile yield (Table 4). Correspondingly, the amounts of ®nal residue at 800°C Table 3 Temperature characteristics for the pyrolysis of two microalgaea Microalga
Fig. 3. The diagrams for TGA and DTG curves of S. platensis and C. protothecoides.
a
Heating rate (°C/min)
Temperature (°C) Ti
Tm
Te
S. platensis
15 40 60 80
190 210 220 220
330 360 360 360
550 560 560 560
C. protothecoides
15 40 60 80
150 180 190 190
330 340 340 340
540 540 540 540
Ti ; Tm and Te are de®ned as in Fig. 3.
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
5
Table 4 The yields of initial moisture, volatile and ®nal residuea Microalga
a
Heating rate (°C/min)
Initial moisture (%)
Total volatile (%)
Volatile evolved between Te and 800°C (%)
Final residue at 800°C (%) Total
Ash
Char
S. platensis
15 40 60 80
7:57 0:01 6:68 0:01 7:48 0:01 6:97 0:01
70:66 0:02 71:72 0:03 70:72 0:03 71:51 0:03
2:52 0:00 2:57 0:00 2:45 0:00 2:62 0:01
21:77 0:01 21:60 0:01 21:80 0:01 21:52 0:01
4:34 0:03 4:34 0:03 4:34 0:03 4:34 0:03
17:43 0:01 17:26 0:01 17:46 0:01 17:18 0:01
C. protothecoides
15 40 60 80
6:85 0:01 6:97 0:01 7:64 0:01 7:07 0:01
70:63 0:02 71:41 0:02 70:87 0:03 71:55 0:04
2:25 0:00 2:90 0:00 2:93 0:00 2:68 0:01
22:52 0:01 21:62 0:01 21:49 0:01 21:38 0:01
7:38 0:05 7:38 0:05 7:38 0:05 7:38 0:05
15:14 0:01 14:24 0:01 14:11 0:01 14:00 0:01
Values are the average of three measurements.
Table 5 Kinetic parameters of pyrolysis under non-isothermal conditions in nitrogena Microalga
a
Heating rate (°C/min)
A min 1
n
R2
4
7
E (J/mol)
S. platensis
15 40 60 80
9:70 10 8:25 104 7:68 104 7:62 104
9:16 10 6:41 106 2:17 106 2:44 106
1.98 1.75 1.55 1.55
0.98 0.97 0.96 0.96
C. protothecoides
15 40 60 80
5:25 104 4:52 104 4:49 104 4:22 104
1:11 104 4:43 103 4:63 103 3:52 103
1.88 1.79 1.25 1.29
0.99 0.93 0.94 0.96
E: activation energy; A: pre-exponential factor; n: reaction order; R2 : coecient of determination.
of SP and CP were 21.52±21.80% and 21.38±22.52%, respectively. The quantity of char in the ®nal residue of SP was 17.18±17.46%, greater than that of CP by about 2±3%. It seemed that the weight of ®nal residue including char was not in direct proportion to the ash content as the ash content in SP was lower than that in CP by 3%. Little dierence was observed in the ®nal residual weights for both microalgae at 800°C at different heating rates. 3.5. Kinetic parameters The results on the kinetic parameters from TGA calculated by the equations mentioned in Section 2 had a con®dence (R2 , coecient of determination) in the range 0.93±0.99 (Table 5). The activation energies in the main pyrolysis stages exhibited a decrease with increasing heating rate up to 80°C/min. The activation energy for the pyrolysis of SP was higher than that of CP, which suggests that CP is preferable for pyrolysis as compared with SP. The reaction orders for their pyrolysis at dierent heating rates of 15±80°C/min showed the same trend as the activation energies. The reaction orders for SP and CP were 1.55±1.98 and 1.25±1.88, separately. The pre-exponential factor for SP varied between 2:17 106 ±9:16 107 ; while for CP, it varied between 3:52 103 ±1:11 104 .
4. Discussion Microalga is considered to be an optimal candidate for pyrolytic fuel production because of the advantages of larger biomass, faster growth, and higher content components preferable for pyrolysis (Dote et al., 1994; Ginzburg, 1993; Milne et al., 1990; Minowa et al., 1995), compared to lignocellulosic materials. Some previous studies in our laboratory on the pyrolysis of Oscillatoria tenuis (cyanobacterium), Chlorella protothecoides (green alga), Emiliania huxleyi and Gephyrocapsa oceanica (coccolithophore) have also revealed the competitive potential of hydrocarbon production from microalgal biomass (Wu et al., 1989, 1996, 1999a,b). Since the late 1970s in China, the input of waste and sewage from municipal, agricultural and industrially sources has resulted in more and more serious eutrophication of lakes (Jin, 1995). Every year large-area microalgal blooms consisting of Microcystis, Chlorella and other genera (mainly blue-green algae and green algae) burst forth in the eutrophic lakes such as Chaohu Lake, Taihu Lake and Dianchi Lake. Owing to the toxins in cells and the heavy metals in aquatic environment, microalgae from eutrophic lakes are not suitable for food and feed. However, they may be used as pyrolytic feedstock for fuel production since most of them have similar chemical compositions to Chlorella and
6
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
Spirulina. Utilization of planktonic microalgae for pyrolysis may decrease the eutrophication of lakes; meanwhile it could supply considerable fuel potential to oset the cost of disposal. As shown in Table 3 and Table 4, SP and CP mainly devolatilized at 150±560°C with a total volatile yield of about 71% and the maximum devolatilization rate reached at 330±370°C; while for many lignocellulosic materials especially hard woods rich in lignin, higher temperature is needed to gain an equal volatile yield and to reach the maximum devolatilization rate (Raveendran et al., 1996b; Mansaray and Ghaly, 1998). This dierence was probably caused by the dierent compositions between two kinds of feedstocks. SP and CP have a high total content of protein, lipid and water-soluble carbohydrate (Table 1). These compounds are preferable to be pyrolyzed rather than cellulose, lignin and hemicellulose, the main chemical components in lignocellulosic biomass (Raveendran et al., 1996b; Meier and Faix, 1999). In addition to the composition of pyrolysis materials, the heating rate is a major factor that aects the thermal behavior of biomass together with temperature (Maschio et al., 1992; Raveendran et al., 1996b; Williams and Besler, 1996). During the devolatilization stage, the instantaneous and average reaction rates increased as the heating rate was increased (Table 3), similar to the results on rice straw (Mansaray and Ghaly, 1998, 1999). Conversely, the activation energies in the devolatilization zones decreased with the increasing heating rate (Table 5). This was in agreement with results on rice husks (Williams and Besler, 1993). Moreover, the increasing heating rates caused a lateral shift to higher temperature values for almost all the characteristic temperature points mentioned above in the thermograms of each microalga (Figs. 1, 2 and Table 2). This trend was also shown in the studies of other feedstocks and has been suggested as mainly due to changes in the kinetics of the decomposition (Lipska-Quinn et al., 1985; Williams and Besler, 1993). It seemed that the ®nal residual weight at 800°C was little changed by the heating rate (Table 4), similar to that observed in a TGA study of rice husks (Williams and Besler, 1993). However, dierent results had been reported in an additional TGA study of rice husks which showed the ®nal residue weight increased as the heating rate was increased (Mansaray and Ghaly, 1998). The pyrolysis of biomass is complicated since the decomposition represents a large number of reactions in parallel and series (Sharma and Rao, 1999). TGA measures the overall weight loss due to these reactions. Therefore TGA provides general information on the overall reaction kinetics rather than individual reactions. The data including the apparent kinetic parameters obtained from TGA of CP and SP are useful for preliminary assessment of planktonic microalgae as feedstocks for thermochemical convertion systems.
Acknowledgements This study belongs to Project 3987006 supported by the National Natural Science Foundation of China and is also partly supported by the State Key Basic Research and Development Programs of China (G1998010100 and G19990433). References Ahuja, P., Singh, P.C., Upadhyay, S.N., Kumar, S., 1996. Kinetics of biomass and sewage sludge pyrolysis: thermogravimetric and sealed reactor studies. Indian J. Chem. Technol. 3, 306±312. Churin, E., Delmon, B., 1989. What can we do with pyrolysis oils? In: Ferrero, G.L., Maniatis, K., Buekens, A., Bridgwater, A.V. (Eds.), Pyrolysis and Gasi®cation. Elsevier, London, pp. 326±333. Dogan, O.M., Uysal, B.Z., 1996. Non-isothermal pyrolysis kinetics of three Turkish oil shales. Fuel 75, 1424±1428. Dote, Y., Sawayama, S., Inoue, S., Minowa, T., Yokoyama, S., 1994. Recovery of liquid fuel from hydrocarbon-rich microalgae by thermochemical liquefaction. Fuel 73, 1855±1857. Encinar, J.M., Beltran, F.J., Gonzalez, J.F., Moreno, M.J., 1997. Pyrolysis of maize, sun¯ower, grape and tobacco residues. J. Chem. Tech. Biotechnol. 70, 400±410. Ginzburg, B., 1993. Liquid fuel (oil) from halophilic algae: a renewable source of non-polluting energy. Renewable Energy 3, 249±252. Jin, X., 1995. The eutrophication of Chinese lakes. In: Jin, X., Liu, S., Zhang, Z., Tu, Q., Xu, N. (Eds.), The Environment of Chinese Lakes, vol. 1. Ocean Press, Beijing, China, pp. 234±302 (in Chinese). Kochert, G., 1978a. Quantitation of the macromolecular components of microalgae. In: Hellebust, J.A., Craigie, J.S. (Eds.), Handbook of Phycological Methods. Physiological and Biochemical Methods. Cambridge University Press, Cambridge, pp. 189±195. Kochert, G., 1978b. Carbohydrate determination by the phenolsulfuric acid method. In: Hellebust, J.A., Craigie, J.S. (Eds.), Handbook of Phycological Methods. Physiological and Biochemical Methods. Cambridge University Press, Cambridge, pp. 95±97. Lipska-Quinn, A.E., Zeronian, S.N., McGee, K.M., 1985. Thermal degradation of rice straw and its components. In: Overend, R.P., Milne, T.A., Mudge, K.L. (Eds.), Fundamentals of Thermochemical Biomass Conversion. Elsevier, Amsterdam, pp. 453±471. Mansaray, K.G., Ghaly, A.E., 1998. Thermal degradation of rice husks in nitrogen atmosphere. Biores. Technol. 65, 13±20. Mansaray, K.G., Ghaly, A.E., 1999. Kinetics of the thermal degradation of rice husks in nitrogen atmosphere. Energy Sources 21, 773±784. Maschio, G., Koufopanos, C., Lucchesi, A., 1992. Pyrolysis, a promising route for biomass utilization. Biores. Technol. 42, 219±231. Meier, D., Faix, O., 1999. State of the art of applied fast pyrolysis of lignocellulosic materials ± a review. Biores. Technol. 68, 71±77. Milne, T.A., Evans, R.J., Nagle, N., 1990. Catalytic conversion of microalgae and vegetable oils to premium gasoline, with shapeselective zeolites. Biomass 21, 219±232. Minowa, T., Yokoyama, S., Kishimoto, M., Okakura, T., 1995. Oil production from algal cells of Dunaliella tertiolecta by direct thermochemical liquefaction. Fuel 74, 1735±1738. Oh-Hama, T., Miyachi, S., 1988. Chlorella. In: Borowitzka, M.A. Borowitzka, L.J. (Eds.), Micro-Algal Biotechnology. Cambridge University Press, Cambridge, pp. 3±26. Raveendran, K., Ganesh, A., 1996a. Heating value of biomass and biomass pyrolysis products. Fuel 75, 1715±1720. Raveendran, K., Ganesh, A., Khilar, K.C., 1996b. Pyrolysis characteristics of biomass and biomass components. Fuel 75, 987±998.
W. Peng et al. / Bioresource Technology 80 (2001) 1±7 Richmond, A., 1988. Spirulina. In: Borowitzka, M.A., Borowitzka, L.J. (Eds.), Micro-Algal Biotechnology. Cambridge University Press, Cambridge, pp. 85±121. Sharma, A., Rao, T.R., 1999. Kinetics of pyrolysis of rice husk. Biores. Technol. 67, 53±59. Williams, P.T., Besler, S., 1993. The pyrolysis of rice husks in a thermogravimetric analyzer and static batch reactor. Fuel 72, 151± 159. Williams, P.T., Besler, S., 1996. The in¯uence of temperature and heating rate on the slow pyrolysis of biomass. Renewable Energy 7, 233±250. Wu, C.H., Chang, C.Y., Lin, J.P., 1997. Pyrolysis kinetics of paper mixtures in municipal solid waste. J. Chem. Tech. Biotechnol. 68, 65±74. Wu, Q., Sheng, G., Fu, J., 1989. Components of natural gas resulting from thermal degradation of the blue-green alga (cyanobacterium) Oscillatoria tenuis. J. Appl. Phycol. 1, 285±287.
7
Wu, Q., Kuang, M., Grant, N.G., 1992. Growth, AEC and pigments of Chlorella protothecoides in two strains under the autotrophic and heterotrophic conditions. Acta Plant Physiol. Sinica. 18, 293±299. Wu, Q., Zhang, B., Grant, N.G., 1996. High yield of hydrocarbon gases resulting from pyrolysis of yellow heterotrophic and bacterially degraded Chlorella protothecoides. J. Appl. Phycol. 8, 181±184. Wu, Q., Dai, J., Shiraiwa, Y., Sheng, G., Fu, J., 1999a. A renewable energy source-hydrocarbon gases resulting from pyrolysis of the marine nanoplanktonic alga Emiliania huxleyi. J. Appl. Phycol. 11, 137±142. Wu, Q., Shiraiwa, Y., Takeda, H., Sheng, G., Fu, J., 1999b. Liquidsaturated hydrocarbons resulting from pyrolysis of the marine coccolithophores Emiliania huxley and Gephyrocapsa oceanica. Mar. Biotechnol. 1, 346±352. Zarrouk, C., 1966. Contribution a I'etude d'une facteure physiques et la photosynthese de Spirulina platensis. (Setch et Gardner). Geitter, Ph.D. Thesis, University of Paris.
Pyrolytic characteristics of microalgae as renewable energy source determined by thermogravimetric analysis Weimin Peng, Qingyu Wu *, Pingguan Tu, Nanming Zhao Department of Biological Sciences and Biotechnology, Tsinghua University, Beijing 100084, People's Republic of China Received 18 December 2000; received in revised form 9 April 2001; accepted 10 April 2001
Abstract Two kinds of autotrophic microalgae, Spirulina platensis (SP) and Chlorella protothecoides (CP) were pyrolyzed at the heating rates of 15, 40, 60 and 80°C/min up to 800°C in a thermogravimetric analyzer to investigate their pyrolytic characteristics. Three stages (dehydration, devolatilization and solid decomposition) appeared in the pyrolysis process. SP and CP mainly devolatilized at 190±560°C and 150±540°C, respectively. A total volatile yield of about 71% was achieved from each microalga. As the heating rate increased, a lateral shift to higher temperatures was observed in their thermograms, and the instantaneous maximum and average reaction rates in the devolatilization stage were increased while the activation energy was decreased. The value of activation energy for CP pyrolysis was 4:22±5:25 104 , lower than that of SP 7:62±9:70 104 , and the char in ®nal residue of CP was 14.00±15.14%, less than that of SP by 2±3%. This indicated that CP is preferable for pyrolysis over SP. The experimental results may provide useful data for the design of pyrolytic processing systems using planktonic microalgae as feedstock. Ó 2001 Elsevier Science Ltd. All rights reserved. Keywords: Spirulina platensis; Chlorella protothecoides; Pyrolysis; Bio-fuel; Eutrophication
1. Introduction Pyrolysis is the degradation of macromolecular materials with heat alone in the absence of oxygen (Meier and Faix, 1999). As a good renewable energy source, biomass has been considered as a potential feedstock for pyrolysis to produce fuels including oil and gas, which can be used as substitutes for petroleum or natural gas for internal-combustion engines, power stations and heating supplies (Raveendran and Ganesh, 1996a). These bio-fuels are cleaner than fossil fuels such as coal and petroleum because of their low nitrogen and sulfur contents (Churin and Delmon, 1989; Ahuja et al., 1996). Many researches on biomass pyrolysis have been carried out to meet the energy demand during the past two decades (Maschio et al., 1992; Meier and Faix, 1999). Thermogravimetric analysis (TGA) is extensively used to understand the pyrolytic characteristics and determine the kinetic parameters (Williams and Besler, 1993; Mansaray and Ghaly, 1998). Many TGA studies *
Corresponding author. Tel.: +86-10-6278-1825; fax: +86-10-62781825. E-mail address: [email protected] (Q. Wu).
on the pyrolytic kinetics of wastes from domestic, industrial and agricultural activities have been reported (Williams and Besler, 1993; Ahuja et al., 1996; Raveendran et al., 1996b; Encinar et al., 1997; Wu et al., 1997; Mansaray and Ghaly, 1998, 1999; Sharma and Rao, 1999). However, there are only a few reports concerning the pyrolytic behaviors of microalgae. Cyanobacterium Spirulina platensis (SP) and green alga Chlorella protothecoides (CP) are two kinds of microalgae arti®cially cultured worldwide for commercial goals of producing food, feed, bait and ®ne chemicals (Oh-Hama and Miyachi, 1988; Richmond, 1988). However, their potential for fuel production is neglected. To determine the eect of temperature and heating rate on their pyrolytic characteristics, the samples were pyrolyzed under non-isothermal conditions, which are more practical and reasonable in simulating large-scale pyrolysis processes (Dogan and Uysal, 1996), in a thermogravimetric analyzer. The kinetic parameters including activation energy, frequency factor and reaction order were measured. These data are important to the ecient design, operation, and modeling of pyrolysis and related thermochemical conversion systems for microalgae.
0960-8524/01/$ - see front matter Ó 2001 Elsevier Science Ltd. All rights reserved. PII: S 0 9 6 0 - 8 5 2 4 ( 0 1 ) 0 0 0 7 2 - 4
2
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
2. Methods 2.1. Algae culture and sample preparation The strain of SP was furnished by Prof. Shang, China Agricultural University (Beijing, Poeple's Republic of China). The strain of CP was obtained from the Culture Collection of Algae at the University of Texas (Austin, USA). SP was grown in Zarrouk's medium (Zarrouk, 1966) at 28°C with continuous illumination of about 60 lE= m2 s. The culture was agitated by aeration. The preparation of medium for CP was described as before (Wu et al., 1992, 1996). CP was axenically grown at 23°C with continuous illumination of 40 lE= m2 s. The culture was also agitated by aeration. The algal cells from a late log phase were harvested by centrifugation and washed one time with distilled water and then dried under infrared light at 30°C. Samples were prepared by pulverization in a mortar to be small enough to eliminate heat transfer eects during the pyrolysis process and then stored in a desiccator.
monly known as the Freeman±Carroll method. Generally the reaction rate da=dt (%/min) for a decomposition reaction is described as da=dt A exp E=RT 1
an ;
1
where a is the conversion of reactant, t (min) is the time, A min 1 is the pre-exponential factor (frequency factor), E (J/mol) is the activation energy, R (J/mol K) is the universal gas constant (8.314), T (K) is the temperature, n is the reaction order. As the heating rate b (°C/ min) is dT =dt, the general non-isothermal kinetic equation is given as da=dT A=b exp E=RT 1
an ;
2
where da=dT is the instantaneous reaction rate (weightloss rate, %/°C), which can be directly obtained from DTG. The logarithmic form for Eq. (2) is ln da=dT ln A=b
E=RT n ln 1
a:
3
For two dierent T and the corresponding a within the same pyrolysis zone,
2.2. Quantitation of the main chemical components in cells
D ln da=dT
The contents of crude protein, crude fat and carbohydrate were determined by Kjeldehl method, Soxhletextract method and phenol-sulfuric acid method, respectively (Kochert, 1978a,b). Protein content was calculated using a factor of 6.25. The solution for lipid extraction was CHCl3 . Glucose was used for calibration during the determination of carbohydrate. The moisture content was determined from the weight loss on heating at 105°C for 24 h, and the ash content was determined as the residue after ignition at 600°C for 3 h. The measurements were replicated three times.
where D means the dierence of two values. The following equation is obtained:
2.3. Pyrolysis in a thermogravimetric analyzer The algal cells were subjected to TGA in a nitrogen atmosphere at heating rates of 15, 40, 60 and 80°C/min from ambient temperature to 800°C in a Dupont 2100 Thermal Analyzer (Dupont TA Instruments, 109 Lukens Drive, New Castle, DE, USA). All the TGA experiments at dierent heating rates were replicated three times. A high-purity nitrogen gas (99.99%) was fed at a constant ¯ow rate of 60 ml/min as an inert purge gas to displace air in the pyrolysis site, thus avoiding unwanted oxidation of the sample. The continuous on-line records of weight loss and temperature were obtained to plot the TGA curve and the derivative thermogravimetric analysis (DTG) curves. 2.4. Kinetic parameters The kinetic parameters for the main pyrolysis process could be calculated by the following equations com-
E=RD 1=T nD ln 1
D ln da=dT =D ln 1
a
E=RD 1=T =D ln 1
a n;
a;
4
5
E and n can be obtained from a plot of D ln da=dT =D ln 1 a versus D 1=T =D ln 1 a that gives a straight line with a slope of E=R and an intercept of n. The value of A may be calculated by substituting the values of E and n back into the Eq. (5) in conjunction with the data of T, a and da=dT . 2.5. Data analyses The results on variance and coecient of determination R2 were obtained from the experimental data analyzed by the Excel 2000 software included in Oce 2000 package produced by Microsoft Corporation.
3. Results 3.1. Main chemical components in cells SP (Table 1) contained 10.30% crude lipid, which was less than that in CP (14.57%), whereas crude protein in SP (61.44%) was higher than that in CP (52.64%). The carbohydrate content in each sample was approximately 10.60%. The total amount of the three components in SP reached 82.31%, which was greater than that in CP (77.83%).
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
3
Table 1 Contents of the main chemical components in two microalgal cellsa
a b
Component (%)
S. platensis
C. protothecoides
Protein Lipid Carbohydrate Ash Moisture Othersb
61.44 0.31 10.30 0.10 10.57 0.13 4.34 0.03 8.04 0.06 5.31 0.63
52.64 0.26 14.57 0.16 10.62 0.14 7.38 0.05 8.74 0.07 6.05 0.68
Values are the average of three measurements. Calculated by dierence.
3.2. Thermal degradation process The TGA and DTG curves of two microalgae at dierent heating rates 15, 40, 60, 80°C/min in nitrogen (Figs. 1 and 2) revealed that three stages existed in the pyrolysis process, as diagramed in Fig. 3. The ®rst stage (Stage I) was from the starting temperature to the temperature of initial devolatilization Ti ; the second stage (Stage II) was from Ti to the end of the main devolatilization Te ; the third stage (Stage III) was from Te to the ®nal temperature (800°C). Te was taken as the point of equal value of instantaneous thermal degradation rate to Ti . The temperature of maximum reaction rate was referred as Tm . Fig. 2. The TGA and DTG curves of C. protothecoides at the heating rates of 15, 40, 60 and 80°C/min.
As shown in Figs. 1±3, a slight weight loss appeared in Stage I. It could be due to the elimination of water (dehydration). Stage II was characterized by a major weight loss, which corresponded to the main pyrolysis process (devolatilization). Most of the volatiles were released in this stage. It proceeded with a high rate and led to the formation of the pyrolysis products. During Stage III, a slow continued loss of weight revealed that the carbonaceous matters in the solid residue continuously decomposed at a very slow rate and the solid residue reached an asymptotic value. The instantaneous maximum and average reaction rates in Stage II for CP and SP were increased as the heating rate increased (Table 2). Both rates for SP were higher than those for CP at each heating rate. In addition, the average rate in the devolatilization stage was about twice as much as that of the whole process. 3.3. Temperature characteristics
Fig. 1. The TGA and DTG curves of S. platensis at the heating rates of 15, 40, 60 and 80°C/min.
The values of Ti ; Tm and Te at dierent thermograms are given in Table 3. There was a lateral shift to higher temperatures for almost all the temperature points in the thermograms of each microalga as the heating rate was increased (Figs. 1, 2 and Table 3). The decomposition of
4
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
Table 2 Reaction rates in the pyrolysis process of two microalgaea ;b Microalga
Heating rate (°C/min)
Devolatilization in stage II (%)
Instantaneous maximum reaction rate (%/min)
Average reaction rate (%/min) Stage II
Overallc
S. platensis
15 40 60 80
68:14 0:03 69:15 0:03 68:27 0:03 68:89 0:04
8.34 0.01 22.06 0.02 30.86 0.02 43.75 0.04
2.84 0.01 7.90 0.02 12.05 0.03 16.21 0.04
1:47 0:00 3:92 0:00 5:87 0:00 7:85 0:01
C. protothecoides
15 40 60 80
68:38 0:03 68:51 0:03 68:23 0:03 69:18 0:04
6.62 0.01 18.15 0.01 27.28 0.02 36.16 0.03
2.63 0.01 7.61 0.02 11.70 0.03 15.81 0.04
1:45 0:00 3:92 0:00 5:89 0:00 7:86 0:01
a
Values are the average of three measurements. The values of reaction rates (%/min) listed here are the products of reaction rates (%/°C) from DTG and the corresponding heating rates. c For the entire pyrolysis process. b
CP started earlier and also ended earlier than that of SP. The initial temperatures for decomposition of CP and SP at dierent heating rates were 150±190°C and 190± 220°C respectively, while the ®nal temperatures for CP and SP were approximately 540°C and 560°C, respectively (Table 3). This implied that the main pyrolysis reactions including depolymerization, decarboxylation and cracking took place over the temperature range 150±560°C. 3.4. Products of pyrolysis by thermogravimetric analysis The yields of initial moisture, volatile and ®nal residue obtained from TGA are presented in Table 4. The initial moisture yield was calculated as the weight loss in Stage I and the total volatile yield was calculated as the weight loss in both Stage II and Stage III. The initial moisture yields of SP and CP were 6.68±7.57% and 6.85± 7.64%, respectively. They were slightly lower than the moisture content shown in Table 1. The dierence may have been caused by the dierent heating times and temperatures. Each microalga gave a total volatile yield of about 71%. The thermal degradation in Stage III contributed little to the total volatile yield (Table 4). Correspondingly, the amounts of ®nal residue at 800°C Table 3 Temperature characteristics for the pyrolysis of two microalgaea Microalga
Fig. 3. The diagrams for TGA and DTG curves of S. platensis and C. protothecoides.
a
Heating rate (°C/min)
Temperature (°C) Ti
Tm
Te
S. platensis
15 40 60 80
190 210 220 220
330 360 360 360
550 560 560 560
C. protothecoides
15 40 60 80
150 180 190 190
330 340 340 340
540 540 540 540
Ti ; Tm and Te are de®ned as in Fig. 3.
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
5
Table 4 The yields of initial moisture, volatile and ®nal residuea Microalga
a
Heating rate (°C/min)
Initial moisture (%)
Total volatile (%)
Volatile evolved between Te and 800°C (%)
Final residue at 800°C (%) Total
Ash
Char
S. platensis
15 40 60 80
7:57 0:01 6:68 0:01 7:48 0:01 6:97 0:01
70:66 0:02 71:72 0:03 70:72 0:03 71:51 0:03
2:52 0:00 2:57 0:00 2:45 0:00 2:62 0:01
21:77 0:01 21:60 0:01 21:80 0:01 21:52 0:01
4:34 0:03 4:34 0:03 4:34 0:03 4:34 0:03
17:43 0:01 17:26 0:01 17:46 0:01 17:18 0:01
C. protothecoides
15 40 60 80
6:85 0:01 6:97 0:01 7:64 0:01 7:07 0:01
70:63 0:02 71:41 0:02 70:87 0:03 71:55 0:04
2:25 0:00 2:90 0:00 2:93 0:00 2:68 0:01
22:52 0:01 21:62 0:01 21:49 0:01 21:38 0:01
7:38 0:05 7:38 0:05 7:38 0:05 7:38 0:05
15:14 0:01 14:24 0:01 14:11 0:01 14:00 0:01
Values are the average of three measurements.
Table 5 Kinetic parameters of pyrolysis under non-isothermal conditions in nitrogena Microalga
a
Heating rate (°C/min)
A min 1
n
R2
4
7
E (J/mol)
S. platensis
15 40 60 80
9:70 10 8:25 104 7:68 104 7:62 104
9:16 10 6:41 106 2:17 106 2:44 106
1.98 1.75 1.55 1.55
0.98 0.97 0.96 0.96
C. protothecoides
15 40 60 80
5:25 104 4:52 104 4:49 104 4:22 104
1:11 104 4:43 103 4:63 103 3:52 103
1.88 1.79 1.25 1.29
0.99 0.93 0.94 0.96
E: activation energy; A: pre-exponential factor; n: reaction order; R2 : coecient of determination.
of SP and CP were 21.52±21.80% and 21.38±22.52%, respectively. The quantity of char in the ®nal residue of SP was 17.18±17.46%, greater than that of CP by about 2±3%. It seemed that the weight of ®nal residue including char was not in direct proportion to the ash content as the ash content in SP was lower than that in CP by 3%. Little dierence was observed in the ®nal residual weights for both microalgae at 800°C at different heating rates. 3.5. Kinetic parameters The results on the kinetic parameters from TGA calculated by the equations mentioned in Section 2 had a con®dence (R2 , coecient of determination) in the range 0.93±0.99 (Table 5). The activation energies in the main pyrolysis stages exhibited a decrease with increasing heating rate up to 80°C/min. The activation energy for the pyrolysis of SP was higher than that of CP, which suggests that CP is preferable for pyrolysis as compared with SP. The reaction orders for their pyrolysis at dierent heating rates of 15±80°C/min showed the same trend as the activation energies. The reaction orders for SP and CP were 1.55±1.98 and 1.25±1.88, separately. The pre-exponential factor for SP varied between 2:17 106 ±9:16 107 ; while for CP, it varied between 3:52 103 ±1:11 104 .
4. Discussion Microalga is considered to be an optimal candidate for pyrolytic fuel production because of the advantages of larger biomass, faster growth, and higher content components preferable for pyrolysis (Dote et al., 1994; Ginzburg, 1993; Milne et al., 1990; Minowa et al., 1995), compared to lignocellulosic materials. Some previous studies in our laboratory on the pyrolysis of Oscillatoria tenuis (cyanobacterium), Chlorella protothecoides (green alga), Emiliania huxleyi and Gephyrocapsa oceanica (coccolithophore) have also revealed the competitive potential of hydrocarbon production from microalgal biomass (Wu et al., 1989, 1996, 1999a,b). Since the late 1970s in China, the input of waste and sewage from municipal, agricultural and industrially sources has resulted in more and more serious eutrophication of lakes (Jin, 1995). Every year large-area microalgal blooms consisting of Microcystis, Chlorella and other genera (mainly blue-green algae and green algae) burst forth in the eutrophic lakes such as Chaohu Lake, Taihu Lake and Dianchi Lake. Owing to the toxins in cells and the heavy metals in aquatic environment, microalgae from eutrophic lakes are not suitable for food and feed. However, they may be used as pyrolytic feedstock for fuel production since most of them have similar chemical compositions to Chlorella and
6
W. Peng et al. / Bioresource Technology 80 (2001) 1±7
Spirulina. Utilization of planktonic microalgae for pyrolysis may decrease the eutrophication of lakes; meanwhile it could supply considerable fuel potential to oset the cost of disposal. As shown in Table 3 and Table 4, SP and CP mainly devolatilized at 150±560°C with a total volatile yield of about 71% and the maximum devolatilization rate reached at 330±370°C; while for many lignocellulosic materials especially hard woods rich in lignin, higher temperature is needed to gain an equal volatile yield and to reach the maximum devolatilization rate (Raveendran et al., 1996b; Mansaray and Ghaly, 1998). This dierence was probably caused by the dierent compositions between two kinds of feedstocks. SP and CP have a high total content of protein, lipid and water-soluble carbohydrate (Table 1). These compounds are preferable to be pyrolyzed rather than cellulose, lignin and hemicellulose, the main chemical components in lignocellulosic biomass (Raveendran et al., 1996b; Meier and Faix, 1999). In addition to the composition of pyrolysis materials, the heating rate is a major factor that aects the thermal behavior of biomass together with temperature (Maschio et al., 1992; Raveendran et al., 1996b; Williams and Besler, 1996). During the devolatilization stage, the instantaneous and average reaction rates increased as the heating rate was increased (Table 3), similar to the results on rice straw (Mansaray and Ghaly, 1998, 1999). Conversely, the activation energies in the devolatilization zones decreased with the increasing heating rate (Table 5). This was in agreement with results on rice husks (Williams and Besler, 1993). Moreover, the increasing heating rates caused a lateral shift to higher temperature values for almost all the characteristic temperature points mentioned above in the thermograms of each microalga (Figs. 1, 2 and Table 2). This trend was also shown in the studies of other feedstocks and has been suggested as mainly due to changes in the kinetics of the decomposition (Lipska-Quinn et al., 1985; Williams and Besler, 1993). It seemed that the ®nal residual weight at 800°C was little changed by the heating rate (Table 4), similar to that observed in a TGA study of rice husks (Williams and Besler, 1993). However, dierent results had been reported in an additional TGA study of rice husks which showed the ®nal residue weight increased as the heating rate was increased (Mansaray and Ghaly, 1998). The pyrolysis of biomass is complicated since the decomposition represents a large number of reactions in parallel and series (Sharma and Rao, 1999). TGA measures the overall weight loss due to these reactions. Therefore TGA provides general information on the overall reaction kinetics rather than individual reactions. The data including the apparent kinetic parameters obtained from TGA of CP and SP are useful for preliminary assessment of planktonic microalgae as feedstocks for thermochemical convertion systems.
Acknowledgements This study belongs to Project 3987006 supported by the National Natural Science Foundation of China and is also partly supported by the State Key Basic Research and Development Programs of China (G1998010100 and G19990433). References Ahuja, P., Singh, P.C., Upadhyay, S.N., Kumar, S., 1996. Kinetics of biomass and sewage sludge pyrolysis: thermogravimetric and sealed reactor studies. Indian J. Chem. Technol. 3, 306±312. Churin, E., Delmon, B., 1989. What can we do with pyrolysis oils? In: Ferrero, G.L., Maniatis, K., Buekens, A., Bridgwater, A.V. (Eds.), Pyrolysis and Gasi®cation. Elsevier, London, pp. 326±333. Dogan, O.M., Uysal, B.Z., 1996. Non-isothermal pyrolysis kinetics of three Turkish oil shales. Fuel 75, 1424±1428. Dote, Y., Sawayama, S., Inoue, S., Minowa, T., Yokoyama, S., 1994. Recovery of liquid fuel from hydrocarbon-rich microalgae by thermochemical liquefaction. Fuel 73, 1855±1857. Encinar, J.M., Beltran, F.J., Gonzalez, J.F., Moreno, M.J., 1997. Pyrolysis of maize, sun¯ower, grape and tobacco residues. J. Chem. Tech. Biotechnol. 70, 400±410. Ginzburg, B., 1993. Liquid fuel (oil) from halophilic algae: a renewable source of non-polluting energy. Renewable Energy 3, 249±252. Jin, X., 1995. The eutrophication of Chinese lakes. In: Jin, X., Liu, S., Zhang, Z., Tu, Q., Xu, N. (Eds.), The Environment of Chinese Lakes, vol. 1. Ocean Press, Beijing, China, pp. 234±302 (in Chinese). Kochert, G., 1978a. Quantitation of the macromolecular components of microalgae. In: Hellebust, J.A., Craigie, J.S. (Eds.), Handbook of Phycological Methods. Physiological and Biochemical Methods. Cambridge University Press, Cambridge, pp. 189±195. Kochert, G., 1978b. Carbohydrate determination by the phenolsulfuric acid method. In: Hellebust, J.A., Craigie, J.S. (Eds.), Handbook of Phycological Methods. Physiological and Biochemical Methods. Cambridge University Press, Cambridge, pp. 95±97. Lipska-Quinn, A.E., Zeronian, S.N., McGee, K.M., 1985. Thermal degradation of rice straw and its components. In: Overend, R.P., Milne, T.A., Mudge, K.L. (Eds.), Fundamentals of Thermochemical Biomass Conversion. Elsevier, Amsterdam, pp. 453±471. Mansaray, K.G., Ghaly, A.E., 1998. Thermal degradation of rice husks in nitrogen atmosphere. Biores. Technol. 65, 13±20. Mansaray, K.G., Ghaly, A.E., 1999. Kinetics of the thermal degradation of rice husks in nitrogen atmosphere. Energy Sources 21, 773±784. Maschio, G., Koufopanos, C., Lucchesi, A., 1992. Pyrolysis, a promising route for biomass utilization. Biores. Technol. 42, 219±231. Meier, D., Faix, O., 1999. State of the art of applied fast pyrolysis of lignocellulosic materials ± a review. Biores. Technol. 68, 71±77. Milne, T.A., Evans, R.J., Nagle, N., 1990. Catalytic conversion of microalgae and vegetable oils to premium gasoline, with shapeselective zeolites. Biomass 21, 219±232. Minowa, T., Yokoyama, S., Kishimoto, M., Okakura, T., 1995. Oil production from algal cells of Dunaliella tertiolecta by direct thermochemical liquefaction. Fuel 74, 1735±1738. Oh-Hama, T., Miyachi, S., 1988. Chlorella. In: Borowitzka, M.A. Borowitzka, L.J. (Eds.), Micro-Algal Biotechnology. Cambridge University Press, Cambridge, pp. 3±26. Raveendran, K., Ganesh, A., 1996a. Heating value of biomass and biomass pyrolysis products. Fuel 75, 1715±1720. Raveendran, K., Ganesh, A., Khilar, K.C., 1996b. Pyrolysis characteristics of biomass and biomass components. Fuel 75, 987±998.
W. Peng et al. / Bioresource Technology 80 (2001) 1±7 Richmond, A., 1988. Spirulina. In: Borowitzka, M.A., Borowitzka, L.J. (Eds.), Micro-Algal Biotechnology. Cambridge University Press, Cambridge, pp. 85±121. Sharma, A., Rao, T.R., 1999. Kinetics of pyrolysis of rice husk. Biores. Technol. 67, 53±59. Williams, P.T., Besler, S., 1993. The pyrolysis of rice husks in a thermogravimetric analyzer and static batch reactor. Fuel 72, 151± 159. Williams, P.T., Besler, S., 1996. The in¯uence of temperature and heating rate on the slow pyrolysis of biomass. Renewable Energy 7, 233±250. Wu, C.H., Chang, C.Y., Lin, J.P., 1997. Pyrolysis kinetics of paper mixtures in municipal solid waste. J. Chem. Tech. Biotechnol. 68, 65±74. Wu, Q., Sheng, G., Fu, J., 1989. Components of natural gas resulting from thermal degradation of the blue-green alga (cyanobacterium) Oscillatoria tenuis. J. Appl. Phycol. 1, 285±287.
7
Wu, Q., Kuang, M., Grant, N.G., 1992. Growth, AEC and pigments of Chlorella protothecoides in two strains under the autotrophic and heterotrophic conditions. Acta Plant Physiol. Sinica. 18, 293±299. Wu, Q., Zhang, B., Grant, N.G., 1996. High yield of hydrocarbon gases resulting from pyrolysis of yellow heterotrophic and bacterially degraded Chlorella protothecoides. J. Appl. Phycol. 8, 181±184. Wu, Q., Dai, J., Shiraiwa, Y., Sheng, G., Fu, J., 1999a. A renewable energy source-hydrocarbon gases resulting from pyrolysis of the marine nanoplanktonic alga Emiliania huxleyi. J. Appl. Phycol. 11, 137±142. Wu, Q., Shiraiwa, Y., Takeda, H., Sheng, G., Fu, J., 1999b. Liquidsaturated hydrocarbons resulting from pyrolysis of the marine coccolithophores Emiliania huxley and Gephyrocapsa oceanica. Mar. Biotechnol. 1, 346±352. Zarrouk, C., 1966. Contribution a I'etude d'une facteure physiques et la photosynthese de Spirulina platensis. (Setch et Gardner). Geitter, Ph.D. Thesis, University of Paris.