CSFB applied to fluidized-bed gasification of special fuels

Fuel 88 (2009) 826–833

Contents lists available at ScienceDirect

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

CSFB applied to fluidized-bed gasification of special fuels Marcio L. de Souza-Santos * UNICAMP – University of Campinas, Faculty of Mechanical Engineering, Department of Energy, CP 6122, Campinas, 13083-970 SP, Brazil

a r t i c l e

i n f o

Article history: Received 16 June 2008 Received in revised form 26 September 2008 Accepted 18 October 2008 Available online 17 November 2008 Keywords: CSFMB Fluidized-bed Gasification Simulation University of California

a b s t r a c t The new version of a comprehensive simulation program of moving and fluidized-beds (CSFB or CSFMB) has been tested against data generated from a gasification unit at the University of California at Davis (UCD), Department of Biological and Agricultural Engineering. The pilot operated at atmospheric pressure under bubbling fluidized-bed and consumed various biomasses as well residues. Air was used as gasifying agent and electrical resistances around the reactor helped heating the reactor during start-up and were kept under low energy discharge during experiments. CSFMB was adapted to allow simulation of such cases as well to several other possibilities of additional heating systems to reactors. The present paper presents the results from cases of almond shells and walnut pruning gasification. Good to reasonable agreement between simulation and operational data have been obtained. Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction

2. Adaptation to simulate equipment with additional heating

Since its first version [1,2] the mathematical model and respective simulation program of a comprehensive simulator for moving and fluidized-bed equipment (CSFB or CSFMB) has been improved and applied to various classes of equipment consuming a wide range of fuels [3–24]. Among the simulated units, there were: boilers, gasifiers, shale retorting reactors, dryers and pyrolysers consuming various coal ranks, biomasses, and residues. Recently [25], a new version CSFMB was developed to simulate cases of units with special heating devices, such as:

The fundamental aspects of the mathematical model and simulation strategy are summarized at the Appendix. Detailed description of the model basic equations and correlations in which the last version is based upon can be found in recent publications [24,25]. That version already includes the possibilities of using jackets around the bed and freeboard regions. Hot or cold gases or liquids can be injected into the jacket or jackets to exchange heat with various regions of the equipment. Since the model takes into account terms related to the heat transfer between the equipment interior and jackets, any other energy source terms can also be added to the energy balances (represented by Eqs. A.5 and A.11 at Ref. [25]). Consequently, the adaptation to simulate the effect of electrical heating is a straightforward because the rate of energy delivered by resistances could be imposed as evenly distributed throughout the section coated by them.

(a) Electrical resistances around the reactor or in its interior. (b) Steam or hot gas passing through jackets around the reactor. (c) Steam or hot gas injected into tube banks inserted into the reactor. All those methods can be applied to any case of bubbling, circulating or even updraft and downdraft moving bed equipment. The simulation program has been successfully tested against many operations of boilers, gasifiers, and oil shale retorting [1– 6,13–15,18–25]. However, none of those included electrically heated units. Of course, application of electrical power might be difficult to justify for most of industrial-scale processes. Despite that, the tests were a good source for testing CSFMB in such situation and this paper is devoted to the two first cases reported by UCD team [26]. Other comparisons should follow soon. * Tel.: +55 19 35213278; fax: +55 19 32893722. E-mail address: [email protected] 0016-2361/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.fuel.2008.10.035

3. Experimental results Several tests were carried at a pilot gasification unit at the UCD Department of Biological and Agricultural Engineering consuming almond shells, walnut pruning, rice straw, whole tree wood chips, sludge and non-recyclable waste paper [26]. The basic characteristics of the reactor are described in Table 1. As mentioned above, the present paper concentrates in the cases of almond shells and walnut pruning with properties shown in Table 2. Table 3 lists the main operational conditions for each test. Alumina-Silicate (43.5% Al2O3 and 53.5% SiO2) was employed as inert material for the bed. Its apparent particle density was around

827

M.L. de Souza-Santos / Fuel 88 (2009) 826–833 Table 1 Main design data of UCD gasifier. Basic characteristic

Value

Bed internal diameter Freeboard internal diameter Average bed dynamic height Total internal heighta Average operational pressure Average temperature of injected air Position of fuel feedinga Thickness of reactor insulation Insulation average thermal conductivity Thickness of distributor porous plate Distributor average thermal conductivity

0.073 m 0.127 m 1.0 m 1.4 m 101 kPa 293 K 0.025 m 0.1 m 0.6 W m 0.1 m 0.3 W m

a

Table 4 Composition of gas produced during real operations of UCD gasifier and respective simulation results. Composition of produced gas (mol%, dry and tar free)

1

K

1

1

K

1

From the distributor internal surface.

Table 2 Characteristics of the solid fuels fed into the gasifier. Properities

Biomass

Proximate analysis (w.b. %) Moisture Volatiles Fixed carbon

Almond shell

Walnut pruning

9.00 53.51 13.47

14.3 69.25 13.54

36.27 3.94 0.79 32.38 0.05 26.57 15.1 1070 1500

48.20 4.41 0.59 44.34 0.03 2.43 19.0 1010 1490

Particle size distribution Sieve opening (mm) 2.000 1.410 0.851 0.420 0.149 <0.149

% Mass retained 0.6 3.3 6.5 27.5 32.2 29.7

0.0 0.1 3.0 65.2b 22.5 9.2

a

According to private communication with UCD team, few values are a bit different from the published in the original report [26]. b Corrected value from original report [26] to allow consistency.

Table 3 Main operational conditions during two tests of UCD gasification experiments.

Fuel feeding (kg/s) Injected air flow (kg/s) Oxygen ratio (%) Injected air temperature (K) Mass of inert initially in the bed (kg)

AS1

WP1

Real

Simulation

Real

Simulation

13 n.d. 0.503 n.d. n.d. 39 n.d. 1 n.d. 20 20 0 7 n.d. n.d. n.d. n.d. n.d. n.d.

13.6500 0.0022 0.9028 0.0000 0.0000 40.0834 0.0000 0.0000 0.0073 18.7722 17.7035 0.0014 7.7099 0.4307 0.1046 0.0000 0.0000 0.1575 0.4745

10 n.d. 0.199 n.d. n.d. 42 n.d. 3 n.d. 22 16 0 7 n.d. n.d. n.d. n.d. n.d. n.d.

10.5315 0.0005 0.0000 0.0000 0.0000 39.5875 0.0000 0.0000 0.0025 22.3692 17.6226 0.0000 9.3476 0.0446 0.0000 0.0000 0.0000 0.0244 0.4697

n.d.: not determined or reported.

Ultimate analysis (% d.b.)a C H N O S Ash HHV (d.b.) (MJ/kg) Particle app. density (kg/m3) Particle true density (kg/m3)

Operational condition

H2 H2S NH3 NO NO2 N2 N2O O2 SO2 CO CO2 HCN CH4 C2H4 C2H6 C3H6 C3H8 C6H6 Ar

Test

Test Almond shell first test (AS1)

Walnut prunings first test (WP1)

2.120  10 3 4.80  10 4 5.97 293 0.866

1.482  10 3 6.05  10 4 9.20 576 0.433

2700 kg/m3 and almost all particles passed through the 0.5 mm and were retained at 0.21 mm sieve apertures. Tables 4 and 5 illustrate the comparisons between real operations and simulation results. 4. Discussion Tables 4 and 5 demonstrate that CSFMB is able to reproduce relatively well the experimental tests carried by UCD. Not only the concentration of species in the produced gases, but also tempera-

tures and even rates of elutriated particles were simulated within acceptable deviations. Utilizing the features of CSFMB, few graphs showing the profiles of main process variables are presented at Figs. 1–7. Fig. 1 reveals that, for most of the bed, all temperatures are very close to the average, while Fig. 2 the same however with steady decreases of temperatures in the freeboard. In contrast, many bubbling fluidization processes lead to surges of temperature in the bed and not so linear profiles in the freeboard [1–25]. For instance, in many cases temperature of bubbles depart significantly from the average at intermediary regions of the bed. This is so due the following sequence of events: (1) At the distributor, part of the injected oxidant gas (air or mixtures with oxygen) forms bubbles and the remaining flows to the emulsion phase. (2) As the emulsion retains almost all solid fuel particles, little or no oxygen remains in that phase at positions not too far from the distributor. That reducing condition allows the increases in concentrations of fuel gases produced by gasification reactions. (3) On the other hand, bubbles are relatively free of fuel particles and remain relatively cold, even at regions well above the distributor. (4) In their progress toward the top, bubbles receive fuel gases migrating from the emulsion phase. While the bubbles remain relatively cold, the oxygen does not significantly react with those gases. However, once their temperature rise due to heat exchange with the emulsion, ignition of the combustible gases occur, leading to the temperature surge. (5) After that, the temperature of bubbles tends to follow the average in the bed. If the rate of oxygen (or air) injection into the bed is relatively low, the surge of temperature is not too pronounced, as illustrated by Fig. 1. As described above as well at the Appendix, the different concentrations of gases found in the emulsion and bubbles drive the intense mass transfers between these phases. Such process greatly influences combustion and gasification in bubbling fluidized-beds [1–25]. CSFMB provides the concentration profiles of 18 gaseous

828

M.L. de Souza-Santos / Fuel 88 (2009) 826–833

Table 5 Several additional operational conditions during real operations of UCD gasifier and respective simulation results. Condition or information

TEST AS1

Average temperature in the bed (K) Mass flow of produced gases (tar-free) (kg/s) Flow of particles at freeboard top (kg/s) Flow of tar leaving with gases (kg/s) Flow of particles discharged from the system (kg/s) Carbon conversion (%) TDH (m) Average residence time of particles in the reactor (min)a Power input by entering gas and solid streams (kW) Power provided to the process due to partial solid fuel combustion (kW)b Power input by electrical resistances (kW)

WP1

Real

Simulation

Real

1009 ± 20 n.c. 1.482  10 n.d. n.c. n.c. n.d. n.d. n.c. n.d. n.d.

1011.8 1.053  10 1.232  10 7.961  10 6.391  10 82.31 1.316 12.95 29.7 1.91 1.40

1017 ± 14 n.c. 2.35  10 n.d. n.c. n.c. n.d. n.d. n.c. n.d. n.d.

4

3 4 4 4

Simulation

5

1043.5 1.323  10 3.810  10 6.575  10 7.510  10 94.96 3.094 12.73 24.4 2.18 1.40

3 5 4 5

n.c.: not clear determination. n.d.: not determined or reported. a Based on feeding rate. b Approximation computed by multiplying the fuel energy input (mass flow x fuel combustion enthalpy) by the oxygen ratio (Table 3).

Fig. 1. Temperature profiles in the bed (AS1 test).

Fig. 2. Temperature profiles in the freeboard (AS1 test).

species at each phase throughout the bed and freeboard. Nonetheless, to simplify the discussion here, Fig. 3 just shows the average concentration of very important components in the bed and free-

board (this last region starts around one meter above the distributor). The rates of selected critical reactions in the bed (the notations at the legend are just reminders of the reactions and

M.L. de Souza-Santos / Fuel 88 (2009) 826–833

Fig. 3. Concentration profiles of CO, CO2, and O2 throughout the reactor (AS1 test).

Fig. 4. Concentration profiles of H2, H2O, and CH4 throughout the reactor (AS1 test).

Fig. 5. Concentration profiles of H2S, SO2, and Tar throughout the reactor (AS1 test).

829

830

M.L. de Souza-Santos / Fuel 88 (2009) 826–833

Fig. 6. Reaction rates of main heterogeneous reactions in the bed emulsion (AS1 test).

Fig. 7. Reaction rates of main homogeneous reactions throughout the freeboard (AS1 test).

their complete description can be found elsewhere [24,25]) are displayed by Fig. 6. The combination of Figs. 3 and 6, uncover that all oxygen is consumed by the combustion reaction (C + O2) within less that 10 cm above the distributor. Practically no O2 is left above that position, much less to be released with a gas mixture rich in hydrogen and other highly combustible gases. The decrease of CO and increase of H2 (Figs. 3 and 4) in the freeboard illustrates the important role played by the shift-reaction (CO + H2O = CO2 + H2). Fig. 7 displays its relatively high rate throughout that region. Therefore, complete chemical equilibrium is never achieved and that is one reason behind the usefulness of dimensional models to obtain a more reliable picture of complex processes taking place in fluidized-bed equipment [24].

5. Influence of electrical heating The injection rate of oxygen is among the most essential operational parameters in gasification processes. Its relative quantity

compared with the necessary to stoichiometric combustion of the fuel, is known as oxygen ratio. Of course, if no other power input is present, a sizable fraction of the injected oxidant stream is required in order to burn part of the fuel and therefore provide enough energy to properly sustain the main endothermic gasification reactions. When small oxygen ratios are used, temperatures in the bed are too low and the gasification reactions do not reach reasonable rates. In such instances, low concentration of fuel gases such as hydrogen and carbon monoxide is observed in exiting gas streams. Reversibly, if too much oxygen is provided, part of the fuel gases produced by the gasification reactions would oxidize, leading to poor gas quality as well. Usually, industrial gasification processes operate with air or oxygen ratios between 20% and 40% [7,23–25,27–29]. However, the experimental tests by UCD [26] reports extremely low ratios (Table 3), which corroborates the above discussion related to temperature profiles. Despite that, the achieved gas qualities (Table 4) and temperatures (Table 5) are found within the range of processes using much higher oxygen ratios [23–25,27–29]. Because electrical

M.L. de Souza-Santos / Fuel 88 (2009) 826–833

heating was used during the start-up, the unexpected trends could only be explained if at least a fraction of total energy input were applied by resistances throughout the experiments. Unfortunately, the precise electrical power inputs are not found the UCD report [26]. Since the new version is capable of reproducing cases were external heating to reactor is provided, few simulation trials and hints from the UCD team led to the value of 1.4 kW as the approximate power applied during the experiments. Table 5 shows the significance of such supplementary source of energy by comparing it with the amount obtained by partial combustion of the fuel. During test AS1 the electrical energy input was around 73% of the energy provided by partial combustion, while at test WP1 the value was near 64%. Of course, excluding very specific operations aimed to extract valuable chemical components from biomasses – mainly released during pyrolysis – application of electric power is hardly justifiable in industrial-scale processes. 6. Presence of oxygen in produced gases Another visible occurrence during the experiments [26] was the presence of oxygen in the exiting gas streams. In some cases, the concentrations achieved 3% molar (Table 4). Considering that the streams left the reactor at relatively high temperatures and contained considerable concentrations of very reactive gases – such as H2 – the possibility of any detectable oxygen was very unlikely, if not impossible. The theoretical limit for the oxygen concentration is the one obtained after chemical equilibrium. Nonetheless, if mixtures of oxygen and excess of hydrogen were exposed to temperatures similar

831

to those of gaseous streams leaving the UCD pilot, the concentration of O2 would remain beyond the sensitivity of most on-line analyzers. No other published experimental work on gasification reported such relatively high concentrations of oxygen in the exiting gases as those found during the test at UCD. The simulations have also shown that such oxygen concentrations in the exiting gas could not occur. Therefore, it is very likely that some air infiltrated into the line to the analyzers. During this process, oxygen would not significantly react with hydrogen (and other fuel gases) because of the relatively low temperature of the gas mixture. 7. Conclusions The new version of CSFMB (or CSFB) has been able to replicate the experimental work developed at the University of California at Davis. Electrical resistances coating the gasifier were used during the starting-up and maintained on operation, at least partially, during the tests. Earlier attempts to simulate the experiments were unsuccessful mainly because the rates of power input applied during the tests were not reported by UCD publication [26]. However, simulations showed that such power inputs were around the same order of magnitude of the obtained by partial combustion of the feeding solid fuel. Using that information, good reproductions of the experimental tests were achieved, which affirms the usefulness of CSFB in predicting and optimizing gasification processes and equipment designs. Acknowledgments The author is grateful to Drs. Bryan Jenkins and Brad Meister for additional details related to the experimental work.

Fig. 8. Simplied scheme of CSFMB mathematical model.

832

M.L. de Souza-Santos / Fuel 88 (2009) 826–833

Appendix A Fig. 8 presents a simplified scheme of CSFMB model. The basic assumptions and computational strategy can be summarized as follows: (1) The unit operates in steady-state regime. (2) The equipment is separated in two main regions: dense region (or bed in cases of bubbling condition) and lean region (or freeboard in bubbling processes). (3) The dense or bed region is divided in two main phases: bubble and emulsion. (4) There are three possible solid-phases: fuel, inert, and sulfur absorbent such as limestone, dolomite or mixture of those. Ash, eventually detached from the spent fuel, would constitute part of the inert solid phase. (5) The emulsion is composed by solid particles and percolating gas. (6) Bubbles are assumed free of particles. (7) Emulsion gas is considered inviscid, therefore rises through the bed in a plug-flow regime. (8) The same as above is assumed for the bubble gas. However, dimensions, raising velocity, fraction of bed volume occupied by bubbles, as well other characteristics of bubbles are considered in all calculations regarding that phase. (9) Bubbles and emulsion exchange mass and heat. (10) Mass transfers also occur between particles and emulsion gas. (11) Heat transfers also occur between all phases, including particles. (12) Gases are assumed transparent regarding radiative heat transfers. (13) Emulsion gas exchanges heat with the vessel or reactor walls. Therefore, all heat transfers between the walls and other phases (bubbles and particles) take place indirectly through the emulsion gas. (14) All phases exchange heat with surrounding or eventually immersed surfaces (such as tube banks or jackets) in or around the bed and freeboard. (15) Heat transfers to tube banks or jackets are computed pointby-point between those and bed as well freeboard. Eventual phase changes inside the tubes or jackets are also computed. (16) The average composition for each solid particle is computed in the bed or dense region through convergence procedures involving the solutions of differential mass and energy balances described elsewhere [24,25]. However, their composition may change in the freeboard. In addition, particles may exhibit large gradients of temperature and composition in the bed and freeboard. (17) Compositions and temperatures of all gas and solid phases vary in the freeboard and are computed using complete differential and energy balances [24,25]. (18) Particle size distributions modify due to chemical reactions, attritions between particles themselves, as well due to the entrainment and recirculation processes. Those are also taken into account to compute the size distributions of each solid phase in the bed and freeboard. (19) Heat and mass transfers in the axial or vertical direction within each phase are considered negligible when compared with the respective transfers in the radial or horizontal direction between a phase and neighboring ones. (20) At each axial position (z), mass transfers between phases result from differences of species average concentrations at each phase. As soon chemical species are consumed or formed by reactions, they are subtracted from or added to

(21)

(22)

(23)

(24)

(25)

(26)

(27)

(28)

(29)

the respective phase. Therefore, these effects appear as sink or source terms in the mass continuity equations for each phase. At each axial position (z), heat transfers between phases result from differences of temperature at each phase. These terms would appear as sinks or sources in the energy conservation equations [1,2,24,25]. At the basis of the dense region (z = 0), the two-phase model [1,2,24,25,30] is applied to determine the splitting of injected gas stream between emulsion and bubble phases. For points above that (z > 0), the mass flow in each phase is determined by fundamental equations of transport phenomena. Those include mass transfers between the various phase as well homogeneous and heterogeneous reactions. Boundary conditions for the gas phases concerning temperature, pressure and composition at (z = 0) are given by the values of injected gas stream. At each iteration, boundary conditions at z = 0 for the three possible solid-phases (carbonaceous, sulfur absorbent and inert) are obtained after differential energy balances involving conduction, convection, and radiative heat transfers between the distributor surface and the various phases. The solution of differential equations describing the energy and mass transfers proceed from the distributor (z = 0) to the top of freeboard or lean region (z = zF). The values at the top of the bed or dense region (z = zD) are used as boundary conditions for the bottom of lean one. For the first iteration, a carbon conversion is assumed. After solving the system of coupled non-linear differential equations throughout the equipment, the new carbon conversion is computed. Conversions of all other solid-phases components are computed as well. The cyclone system is simulated and all characteristics of the collected particles are obtained. If those are recycled to the bed, CSFMB includes such a stream into the mass and energy balances during iterations. Steps 25 to 28 are repeated until convergence regarding a weighted overall deviation is achieved. That weighing considers deviations between assumed and computed conversions of chemical species as well between assumed and computed heat transfers among phases and immersed surfaces in the bed and freeboard. This and the tight coupling of all chemical and physical phenomena involved in the equipment, ensures consistency regarding all mass and energy balances.

Once the simulation is concluded all internal and overall details of the equipment operation such as temperature, concentration, and all other variable profiles throughout the entire equipment, are printed. A graphical interface facilitates the input of data for simulations as well consultation and study of outputs. References [1] de Souza-Santos ML. Modelling and simulation of fluidized-bed boilers and gasifiers for carbonaceous solids. PhD thesis presented at the University of Sheffield, Department of Chemical Engineer and Fuel Technology, England, United Kingdom, July 1987. [2] de Souza-Santos ML. Comprehensive modelling and simulation of fluidizedbed boilers and gasifiers. Fuel 1989;68:1507–21. [3] de Souza-Santos ML, Cincoto MA, Pikman B, Ushima AH, de Souza MA. Fluidized-bed combustion of rice husks; experimental tests. In: IV Brazilian congress on energy, Rio de Janeiro, RJ, Brazil, 05–09 November 1990. [4] Ushima AH, Guardani R, de Souza-Santos ML. Determination of operational conditions for the reduction of CaSO4 to CaS in a fluidized bed. In: IX Brazilian congress on chemical engineering, Salvador, Bahia, Brazil, September 1992. [5] de Souza-Santos ML. Comprehensive Modeling and simulation of fluidized-bed reactors used in energy generation processes, meeting on energy modeling:

M.L. de Souza-Santos / Fuel 88 (2009) 826–833

[6] [7]

[8]

[9]

[10]

[11]

[12]

[13]

[14]

[15]

[16]

[17]

optimizing information and resources. Chicago: Institute of Gas Technology, 7– 8 June 1993. de Souza-Santos ML. Application of comprehensive simulation to pressurized fluidized bed hydroretorting of shale. Fuel 1994;73(9):1459–65. de Souza-Santos ML. Application of comprehensive simulation of fluidized-bed reactors to the pressurized gasification of biomass. J Braz Soc Mech Sci 1994;16(4):376–83. de Souza-Santos ML. A study on pressurized fluidized-bed gasification of biomass through the use of comprehensive simulation. In: Proceedings of the fourth international conference on technologies and combustion for a clean environment, vol. 2. Lisbon, Portugal, 7–10 July 1997. p. 7–13 [paper 25.2]. de Souza-Santos ML. Search for favorable conditions of atmospheric fluidizedbed gasification of sugar-cane bagasse through comprehensive simulation. In: 14th Brazilian congress of mechanical engineering, Bauru, São Paulo, Brazil, 8– 12 December 1997. de Souza-Santos ML. Search for favorable conditions of pressurized fluidizedbed gasification of sugar-cane bagasse through comprehensive simulation. In: 14th Brazilian congress of mechanical engineering, Bauru, São Paulo, Brazil, 8– 12 December 1997. de Souza-Santos ML. A study on pressurized fluidized-bed gasification of biomass through the use of comprehensive simulation. In: Chapter of the vol. 4 of the book on Combustion Technologies for a Clean Environment, Amsterdam, Holland: Gordon and Breach Publishers; 1998. de Souza-Santos ML. A feasibility study on an alternative power generation system based on biomass gasification/gas turbine concept. Fuel 1999;78:529–38. Costa AMS, de Souza-Santos ML. Studies on the mathematical modeling of circulation rate of particles in bubbling fluidised beds. Powder Technol 1999;103:110–6. Leveson PD, de Souza-Santos ML. Improvements on the modelling of sulphur dioxide emissions from fluidised bed combustors and gasifiers, ENCIT 2000. In: 8th brazilian congress of thermal engineering and sciences, Porto Alegre, RS, 36 October 2000. Rabi JA, de Souza-Santos ML. A two-parameter preliminary optimization study for a fluidized-bed boiler through a comprehensive mathematical simulator, ENCIT 2000. In: 8th Brazilian congress of thermal engineering and sciences, Porto Alegre, RS, Brazil, 3–6 October 2000. de Souza-Santos ML. Two trial models for the rate of combustion and gasification of liquid fuels sprayed on inert fluidised particles. In: Sixth international conference on technologies and combustion for a clean environment, Porto, Portugal, 9–12 July 2001. de Souza-Santos ML. Two trial models for the rate of combustion and gasification of slurries or liquid biomass sprayed on inert fluidised particles.

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25] [26]

[27]

[28]

[29] [30]

833

In: Fifth biomass conference of the Americas, USA, Orlando, Florida, 17–21 September 2001. Rabi JA, de Souza-Santos ML. Preliminary improvements in the radiative heat transfer modeling for fluidized bed biomass gasification: bed section. In: Fifth biomass conference of the Americas, USA, Orlando, Florida, September 17–21; 2001. Rabi JA, de Souza-Santos ML. Incorporation of a 2-flux model for radiative heat transfer in a comprehensive fluidized bed simulator. Part I: Preliminary model equations, ENCIT-2002. In: Nineth Brazilian congress on engineering and thermal sciences, Caxambu, Minas Gerais, Brazil, 15–18 October, 2002. Rabi JA, de Souza-Santos ML. Incorporation of a 2-flux model for radiative heat transfer in a comprehensive fluidized bed simulator. Part II: numerical results and assessment, ENCIT-2002. In: 9th Brazilian congress on engineering and thermal sciences, Caxambu, Minas Gerais, Brazil, 15–18 October, 2002. Rabi JA, de Souza-Santos ML. Incorporation of a two-flux model for radiative heat transfer in a comprehensive fluidized bed simulator. Part I: preliminary theoretical investigations. Therm Eng 2003;3:64–70. Rabi JA, de Souza-Santos ML. Incorporation of a two-flux model for radiative heat transfer in a comprehensive fluidized bed simulator. Part II: numerical results and assessment. Therm Eng 2004;4:49–54. van den Enden PJ, Lora ES. Design approach for a biomass fed fluidized bed gasifier using the simulation software CSFB. Biomass Bioenergy 2004;26:281–7. de Souza-Santos ML. Solid fuels combustion and gasification: modeling, simulation, and equipment operation. New York: Marcel Dekker Inc.; 2004 [ISBN: 0-8247-0971-3]. de Souza-Santos ML. A new version of CSFB, comprehensive simulator for fluidized bed equipment. Fuel 2007;86:1684–709. Meister BC, Williams RB, Jenkins BM. Utilization of waste renewable fuels in boilers with minimization of pollutant emissions; laboratory scale gasification screening experiments. Report CEC-500-2005-134 prepared for california energy commission, public interest energy research (PIER) under contract no. 500-99-013, December 2005. Ocampo A, Arenas E, Chejne F, Espinel J, Londono C, Aguirre J, Perez JD. An experimental study on gasification of colombian coal in fluidized bed. Fuel 2003;82:161–4. Evans RJ, Knight RA, Onischak M, Babu SP. Process and environmental assessment of the RENUAS process. In: Symposium on energy from biomass and wastes, Washington, DC: Sponsored by Institute of Gas Technology, 6–10 April 1986. Li XT, Grace JR, Lim CJ, Watkinson AP, Chen HP, Kim JR. Biomass gasification in a circulating bed. Biomass Bioenergy 2004;26:171–93. Toomey RD, Johnstone HF. Gaseous fluidization of solid particles. Chem Eng Progress 1952;48(5):220–6.