Solar Energy 76 (2004) 345349 www.elsevier.

com/locate/solener

Biomass gasication integrated with pyrolysis in a circulating uidised bed

a,b,* ,

J. Andries b, H. Splietho b, M. Fang c, P.J. van de Enden

Bioenergy and Wastes Treatment Group, School of Environmental Science and Engineering, Tianjin University, Weijin Road 92, Tianjin 300072, Naikai District, China Section Thermal Power Engineering, Faculty of Design, Engineering and Production, Delft University of Technology, Mekelweg 2, 2628 CD Delft, The Netherlands c Institute for Thermal Power Engineering, Zhejiang University, Yugu Road 20, Hangzhou 310027, China Received 11 March 2003; received in revised form 18 August 2003; accepted 19 August 2003

Abstract The use of biomass for energy generation is getting increasing attention. At present, gasication of biomass is taken as a popular technical route to produce fuel gas for application in boilers, engine, gas turbine or fuel cell. Up to now, most of researchers have focused their attentions only on xed-bed gasication and uidised bed gasication under airblown conditions. In that case, the producer gas is contaminated by high tar contents and particles which could lead to the corrosion and wear of blades of turbine. Furthermore, both the technologies, particularly xed bed gasication, are not exible for using multiple biomass-fuel types and also not feasible economically and environmentally for large scale application up to 1050 MWth . An innovative circulating uidised bed concept has been considered in our laboratory for biomass gasication thereby overcoming these challenges. The concept combines and integrates partial oxidation, fast pyrolysis (with an instantaneous drying), gasication, and tar cracking, as well as a shift reaction, with the purpose of producing a high quality of gas, in terms of low tar level and particulates carried out in the producer gas, and overall emissions reduction associated with the combustion of producer gas. This paper describes our innovative concept and presents some experimental results. The results indicate that the gas yield can be above 1.83 N m3 /kg and the uctuation of the gas yield during the period of operation is 3.3% at temperature of 750 C. Generally speaking, the results achieved support our concept as a promising alternative to gasify biomass for the generation of electricity. 2003 Elsevier Ltd. All rights reserved.

1. Introduction Biomass is a renewable resource with almost zero net CO2 emission since carbon and energy are xed during the biomass growth (Johansson et al., 2000; Gielen et al., 2000). Compared with other renewable energy resources, biomass is huge in annual production with a geographically widespread distribution in the world (Larson, 1993). Furthermore, biomass is an easy-to-use energy source, based on present technical level and

Corresponding author. Tel./fax: +86-22-274-04833. E-mail address: g.chen@wbmt.tudelft.nl (G. Chen).

economics. Therefore biomass has been the focus of most of countries for sustainable energy production and also for the reduction of greenhouse gases (GHGs) emissions. European Union (2000) is taking the increase of biomass resource as a contribution to sustainable solutions to many of problems facing us now. As set out in the White paper, EUs goal is to signicantly increase the use of biomass, adding further 90 million tonnes oil equivalent (Mtoe) by year 2010 from current level of 45 Mtoe. If these targets were achieved, biomass would provide around 50% of the energy derived from renewables. In the longer term, it is anticipated that biomass could contribute 20% of the current EU primary energy demand, with more than 20 million hectares used for fuel crops (European Commission, 2000).

0038-092X/$ - see front matter 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.solener.2003.08.021

346

G. Chen et al. / Solar Energy 76 (2004) 345349

So far worldwide, much of biomass as an energy source is used in fairly simple, and sometimes polluting combustion system that is not very energy ecient (we call it as traditional application). Modern application of biomass including direct combustion for electricity and heat generation, thermalchemical processes to produce gas, an oil-like liquid and char, and biochemical processes to produce gaseous and liquid fuels, on the contrary, could overcome the disadvantage mentioned above. Gasication, a type of thermalchemical process, has been received attention throughout the world due to the following advantages: (1) not strict limitations to the size and type of raw biomass used (for example, municipal solid wastes, agricultural wastes and forest residues); (2) convenient use of the resultant gases for dierent applications (for instance, heat/power generation, production of syn-gas, methane, and hydrogen); (3) less pollution problems associated with the downstream applications compared with pyrolysis or combustion. Moreover, a very promising study supported by IEA (1998) and McGowin et al. (1998) shows that the gasication technology for heat and power generation has the similar cost when compared with direct combustion or even is competitive with the cost of all other candidate electricity supply options, including natural gas combined cycle. It is believed that in the near future, the use of gasication to convert solid wastes into ready-touse fuel could literally solve several environmental problems at the same time (Krigmont, 1999). However, the currently used gasication technologies are still far from satisfactory. The main challenges facing us are non-stable gas production process caused by local hot spots existing in the gasier, non-exibility to diverse biomass types, diculty of scale-up, and low quality of product gas. Although circulating uidised bed (CFB) technology has been used in coal combustion and petroleum industry for more than two decades with great success, unfortunately its practical application in biomass gasication is still quite less (Van de Drift et al., 2001). In consideration of the features of CFB technology, CFB gasication of biomass could be expected to be characterized by (i) acceptance of a wide variety of biomass feeds, (ii) exibility of scale-up from small to large capacity (Bentzen and Hummelshoj, 2000), (iii) ability of utilization in existing power plants and thus low specic investment cost. Based on these, CFB gasication of biomass is a quite promising way. However, uncertainties are still too high and challenges are very obvious for practical operation of a new gasier designed. The paper presents an innovative design concept for circulating uidised bed gasication system (CFBGs) aiming at producing high quality gas, in terms of low tar and particulate levels in the producer gas, and overall emissions reduction with respect to gasication process itself and the downstream combustion of the product

gas, as well as high gas yield while not compromising the heating value of producer gas, and also at making the overall system compact (Chen, 2001). In addition, gaining rst-hand experiences of operating CFBGs, in particular of keeping the operation in a stable and reliable way with the minimum uctuations in the producer gas volume and composition are also one of our objectives. In this paper some results obtained experimentally in the laboratory have been presented. Because of the physical constraints in the currently used test-rig, there are obvious shortcomings in results; but the technology still appears promising.

2. The technical concept The basic concept is to separate the pyrolysis, gasication and combustion processes in the same reactor and assures that only char from gasication is burned in the combustion zone. Furthermore, this gasication concept is to gasify biomass in a compactly constructed CFB reactor, where the combustion, gasication, pyrolysis and tar cracking processes take place only in the riser. In summary, the concept combines and integrates the following well-known processes: (i) partial oxidation; (ii) fast pyrolysis (with an instantaneous drying); (iii) gasication; and (iv) tar cracking, as well as (v) a shift reaction. Generally, at certain levels of high temperature, thermal tar cracking is always simultaneously occurs with gasication, but catalytically tar cracking happens only when the catalysts are introduced. Therefore, in the case of no catalyst, the processes (iii) and (iv) could be regarded as equivalent to a gasication-tarcracking process.

3. Experimental apparatus and approach The CFBGs used in the experiments is shown in Fig. 1, which consists of a riser, cyclone, downcomer, returnleg, fuel feeders and an upstream preheater as well as downstream facilities including a gas-burner, gas sampling (ltering unit and analyzers) and chimney. A computer-controlled data acquisition apparatus is also an important component of the whole CFBGs. The riser is 6 m in total height and 83 cm in diameter with a thermal capacity of 100 kWth using miscanthus pellets as feed. The pellet is cylindrical in shape with averaged diameter of 8 mm and length of 24 mm. Table 1 shows the composition of miscanthus pellet. Coarse particles in the hot gas are captured by a high-temperature cyclone immediately downstream of the riser and the captured particles are then recycled back to the bottom of riser via a return-leg. The preheater is connected to the bottom of riser to preheat the incoming air/nitrogen in order to keep the bed temperature at the desired level. The nat-

G. Chen et al. / Solar Energy 76 (2004) 345349 Table 2 Design data for CFBGs
Cyclone Biomass Hopper

347

Optional N2

Items Internal diameter of riser (mm) Total height of riser (m) Eective height of the riser (mm) Cyclone eciency Maximum thermal capacity (kW) Fluidising medium Biomass fuel staging Preheating capacity

Design data 83 6 5.2 95% above 100 Air, nitrogen, and steam 23 Feeding points 12 kW

Auxiliary N2 or Steam

Optional CO H2 Reaction + Tar Cracking Tar Cracking + Gasification Fast Pyrolysis

Optional Steam or Air

Air + Main Steam

Char Partial Oxidation Secondary Biomass Feeder

Auxiliary Air/N2

Table 3 Operating parameters recommended Operating parameters Values 36 1 1035 0.10.5 0.20.5 820 At certain percentage Velocity (m/s) Pressure (bar) Feed-rate (kg/h) Equivalence ratio Staging ratio (auxiliary gas input/ primary uidising medium, %) Temperature (averagely, C) Catalyst fraction (compared to feed-rate)

Fig. 1. The schematic description of CFB gasier.

Table 1 The composition of fuel used Items Proximate analysis (wt%) Fixed carbon Volatile Moisture Ash Ultimate analysis (wt%) C O H N S Cl Ash High heating value (MJ/kg) Miscanthus pellet 15.0 3.50 73.8 3.50 8.7 1.47 2.5 0.44 43.9 0.95 46.8 0.81 5.9 0.40 0.57 0.12 0.14 0.049 0.17 0.043 2.5 0.44 17.6

ural gas burner is located near the outlet of riser and used to investigate the producer gas combustion characteristics in terms of release of pollutants. The gas sampling is placed at the outlet of the gas-ltering unit, which is connected to the outlet of riser, for gas composition analysis. To heat the riser to ignition temperature of the biomass feedstock (400450 C) a 12 kW electrical heater is installed at the riser wall. All parts of CFBGs, except the downstream facilities, are well insulated by refractory material to prevent heat loss. The biomass particles can be fed into the riser at the dierent feeding points via a screw feeder coupled to a pneumatic transportation section. Table 2 shows the design data for CFBGs and Table 3 lists the recommended operating parameters.

The gas phase can be analyzed by a FTIR spectrometer; a multi-component gas chromatograph and a on-line single component analyzer (consisting of three non-dispersive infrared analyzer for CO, CO2 and SO2 , a non-dispersive ultra violet NO analyzer and a paramagnetic O2 , analyzer and a micro-gas chromatograph for H2 ). The analyzable gas species in our laboratory are CO, CO2 , H2 , H2 O, C1 C3 hydrocarbon (CH4 , C2 H4 , C2 H6 , C3 H6 ), O2 , N2 , NH3 , HCN, and nitrogenous oxides (NO, N2 O, and NO2 ), as well as SO2 , COS, CS2 (usually very low in concentration), HCl. In addition, C3 Hx C10 Hx is also conveniently measurable. The tar level of producer gas can be determined by a laboratoryowned measurement facility at relatively high precision. As for the ash components, samples are usually delivered to outside institutions for the measurement of ultimate elements (C, H, N, Cl, S, and O (by dierence), major mineral elements (Al, K, Na, Ca, Mg, P, Fe, Si) and trace elements (Co, Mo, Mn, Cr, Ni, Zn, Cu, V, Pt, Cd). At the beginning of the experiment, only air is fed into the riser as the uidising agent until the bed temperature reaches 400450 C due to the electrical heating. Thereafter, the electric heater is turned o and biomass feedstock with nitrogen/fuel gas is fed into the fast pyrolysis zone, where the heat from partial oxidation reaction pyrolyzes the feedstock. For the case of using steam, only

348

G. Chen et al. / Solar Energy 76 (2004) 345349

when the desired reaction temperature is reached, steam is introduced into the riser. The volatiles from the fast pyrolysis zone are led to the gasication zone, where oxidation air and gasication steam (the ratio is controlled based on the requirements) are added together. Hereby part of the char from the pyrolysis process is oxidized and part of the remains is gasied at the chosen temperature. The catalysts of dolomite/lime, nickel and even nickel-compounds can be introduced into the riser via three ways, i.e. before-hand direct mixing of bed material and catalyst, mixing of catalyst and biomass particles in the feeding pipe, and feeding of catalyst simultaneously in parallel with feeding of biomass at the separated ports of riser. The tar content is dramatically reduced due to cracking thermally and catalytically. Finally the gas passes by the shift reaction zone, where the gas composition is adjusted. When the CFBGs operation reaches steady state, the producer gas passing by the ltering unit is sampled and subsequently analyzed, and at the same time the emissions from gas-burner is monitored. All the data acquired by the data-acquisition goes to PC to make the control of operation at convenience. For this series of experiments, the start-up of CFBGs is from bubbling uidised combustion, to circulating uidised combustion and then to circulating uidised gasication. The advantages of this mode are as follows: In the circulating mode as the temperature rise is very slow and long time is needed to get sucient riser temperature, the gas sampling and analysis pipe is seriously fouled by ne char particles and afterwards cannot work properly. In addition high velocity in circulating mode could result in high rate of solid particle carryover from the riser to the downstream combustor and part of the sand may deposit on the surface of gas burner. This may lead to the failure of burner operation. In the bubbling mode, the above problems are greatly reduced. However, in the bubbling condition, the feeding should be done intermittently because the minimum feed rate of 5.9 kg/h is too high; the uidising air ow does not contain enough oxygen to properly combust the biomass at this rate (Van de Enden, 2001).

Fig. 2. The uctuation of the gas yield against the operating time.

Table 4 Operating condition for a gasication experiment Items Primary air Biomass injection air N2 recirculation N2 purge ows Average riser temperature Riser velocity (no producer gas) Estimated biomass ow Minimum stoichiometric airow Used air ow per kg fuel Lambda (used/stoichiometric airow) Additional gas production Gas yield Data 8.5 N m3 /h 13.3 N m3 /h 0.5 N m3 /h 0.9 N m3 /h 753 C 4.5 m/s 16.8 kg/h 4.24 N m3 air/kg fuel 1.29 N m3 /kg fuel 0.30 7.64 N m3 /h 1.83 N m3 /kg fuel

Table 5 Producer gas composition, heating value and tar concentration Items CO CO2 CH4 C2 H 4 H2 H2 O(g) N2 (calc.) LHV (wet, tar free) HHV (dry, tar free) Tar concentration Measured (mole fraction) Wet (mol%) 10.99 14.00 2.14 0.65 5.24 12.59 54.38 3.46 MJ/N m3 3.69 MJ/N m3 75 mg/N m3 Dry (mol%) 12.57 16.02 2.45 0.74 6.00 0.00 62.22

4. Experimental results Fig. 2 shows the uctuation of the gas yield varying with the operating time. As can be seen, the maximum uctuation of the yield of product gas is varying from 1.79 to 1.85 N m3 /kg. This could almost bring no negative inuence on the stability of the operation of downstream facility particularly like gas turbine/microturbine. Tables 4 and 5 show the experimental results of gasication. At a rst glance, the results can be seen to be not so good but still quite similar to those reported in

the literature. For example, the heating value of producer gas is around 4 MJ/N m3 , and the gas yield is close to 2 N m3 /kg fuel at supplied air to stoichiometric air ratio of 0.3. However, if the physical constraints to the currently used CFBGs are taken into account, these results are quite satisfactory suggesting thereby that the

G. Chen et al. / Solar Energy 76 (2004) 345349

349

proposed gasication concept is quite promising. Basically the CFBGs has the following constraints: low preheating capacity leading to insucient temperature level in the riser, low cyclone eciency, non-exible biomass feeding point, and non-ideal biomass particle size.

References
Bentzen, J.D., Hummelshoj, R., 2000. Low tar and high ecient gasication concept. In: Hirs, G.G. (Ed.), Proceedings of ECOS 2000 (additional papers), Eurotherm Seminar, Enschede, The Netherlands. Chen, G., 2001. Circulating uidised bed gasication of biomass: originality in system design and experimental approach. In: Liu, W. (Ed.), Preprints of the International Conference on Energy Conversion and Application, Wuhan, China, pp. 12631267. European Commission, Directorate-General for Research, 2000. Biomass: An Energy Resource for the European Union. Programme Energy, Environment and Sustainable Development, Brussels, Belgium. Gielen, D.J., De Feber, M.A.P.C., Bos, A.J.M., 2000. Biomass for energy or materials? A Western European systems engineering perspective. Energy Policy 29, 291302. IEA CADDET Centre for Renewable Energy, 1998. Advanced thermal conversion technologies for energy from solid waste. A joint report of IEA bioenergy and the IEA CADDET, Vienna, Austria. Johansson, T.B., Kelly, H., Reddy, A.N., 2000. Renewable Energy Sources for Fuel and Electricity. Petroleum Industry Press, Beijing. pp. 170184. Krigmont, H.V., 1999. Integrated biomass gasication combined cycle (IBGCC) power generation concept: the gateway to a clearer future. A white paper, QEP. Larson, E.D., 1993. Technology for electricity and fuels from biomass. Annul. Res. Energy Environ. 8, 567630. McGowin, C., Hughes, E., Holt, N., 1998. Economic & risk evaluation of the Brazil biomass-gasication/gas-turbine demonstration project. World Bank report. Van de Drift, A., Van Doorn, J., Vermeulen, J.W., 2001. Ten residual biomass fuels for circulating uidised-bed gasication. Biomass Bioenergy 20, 4556. Van de Enden, P.J., 2001. Commissioning of a Circulating Fluidised Bed Biomass Gasier. Delft University of Technology, Delft, The Netherlands.

5. Conclusions and future work The results from our CFBGs show that the yield of product gas is 1.83 N m3 /kg, and the heating value of product gas is around 3.6 MJ/N m3 . The maximum uctuation of gas yield duration 1 h operation is 3.3%. The above results indicate the innovative concept for biomass gasication proposed here looks quite promising despite fact that experimental results (particularly the heating value of product gas) are not very satisfactory. Further work is required to remove the physical constraints of the present CFBGs. Furthermore, the theoretical model of the CFBGs and its validation of the model are recommended.

Acknowledgements This work would not be possible if not for the nancial support from our university and our section. The authors would like to take this opportunity to thank all the colleagues for their great help, particularly for Mr. Maarten de Groot in experimental arrangement and Mr. Wieberen de Jong and Mr. Umal Unal in scientic discussions.