Kinetic Model of Steam Gasification of Biomass in ... - ACS Publications

Sep 25, 2017 - Bijan Hejazi,*,†,§. John R. Grace,. †. Xiaotao Bi,. † and Andrés Mahecha-Botero. †,‡. †. Department of Chemical and Biolo...
0 downloads 0 Views 1MB Size
Subscriber access provided by University of Sussex Library

Article

Kinetic model of steam gasification of biomass in a dual fluidized bed reactor: comparison with pilot plant experimental results Bijan Hejazi, John R. Grace, Xiaotao Tony Bi, and Andres Mahecha-Botero Energy Fuels, Just Accepted Manuscript • DOI: 10.1021/acs.energyfuels.7b01833 • Publication Date (Web): 25 Sep 2017 Downloaded from http://pubs.acs.org on September 25, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Energy & Fuels is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Kinetic model of steam gasification of biomass in a dual fluidized bed reactor: comparison with pilot plant experimental results Bijan Hejazi †,* , John R. Grace †, Xiaotao Bi † and Andrés Mahecha-Botero †,‡ †

Department of Chemical and Biological Engineering, University of British Columbia, 2360 East Mall, Vancouver, BC, Canada, V6T 1Z3



Present Address: NORAM Engineering, 200 Granville Street, Suite 1800 Vancouver, BC, Canada, V6C 1S4

ABSTRACT By incorporating reaction kinetics and reactor hydrodynamics, a steady-state two-phase onedimensional reactor model for biomass steam gasification in the bubbling fluidized bed gasifier of a dual fluidized bed reactor is developed. The generic two-step kinetic model adopted for biomass pyrolysis allows for prediction of tar generation and cracking, as well as predicting pyrolysis products yield and composition based on CHO elemental balances. This model is capable of predicting species concentrations, solids hold-up, temperature, pressure and superficial gas velocity profiles along the gasifier. By performing mass and energy balances over the two interconnected fluidized beds, key operating parameters such as solids circulation rate

ACS Paragon Plus Environment

1

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 62

and additional fuel required for stable operation of the process are approximated. This predictive reactor model, which will provide a useful tool for designing, evaluating and improving a dual fluidized bed gasifier, is compared with experimental data from the pilot dual fluidized bed gasification unit at the University of British Columbia. KEYWORDS: Biomass gasification, Dual fluidized bed reactor, Kinetic modeling, Bubbling fluidized bed. INTRODUCTION Renewable energy is among the possible options to lower greenhouse gas emissions while satisfying the global demand for energy services. The share of renewable energy in the energy mix has increased substantially in recent years1. Bioenergy, a renewable energy, can be produced from a variety of biomass feedstocks, including forest, agricultural and livestock residues, shortrotation forest plantations, energy crops, municipal solid waste, and other organic waste streams. Through a variety of processes, these feedstocks can directly produce electricity or heat, or be used to produce gaseous, liquid, or solid fuels. The range of bioenergy technologies is broad, and the technical maturity varies substantially. Among the options for reducing greenhouse gas emissions is gasification of biomass. Gasification is defined as a high-temperature partial oxidation process in which a solid carbonaceous feedstock such as biomass is converted into gaseous products by gasifying agents such as air, oxygen, steam, carbon dioxide or their mixtures2. Biomass gasification in fluidized bed reactors has a history of proven operability in the field of solid fuel conversion to gaseous products. However, improvements are needed to meet the demands of commercial processes. A recent development of fluidized bed gasifiers is the twinbed configuration, known as dual fluidized bed (DFB) gasifiers. Configurations and operating

ACS Paragon Plus Environment

2

Page 3 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

conditions for dual bed gasifiers of different designs have been reviewed by Corella et al. 3, in particular a successful dual bed system developed at the University of Vienna, and put into commercial use by Hofbauer et al. 4. Due to its complexity, kinetic reactor modeling of biomass gasifiers is at an early phase. Important information is missing from the fluidized bed gasifier reactor model of Corella et al.5, in part due to its proprietary nature. The kinetic reactor model of Radmanesh et al. 6, applied to a single bubbling fluidized bed gasifier, contains useful reaction kinetic information. Lü et al. 7 developed a steady-state kinetic reactor model of biomass air–steam gasification in a fluidized bed reactor. Separate kinetic reactor models developed by Kaushal et al.8 for a bubbling fluidized bed gasifier and a circulating fluidized bed combustor9 are most relevant. In developing kinetic models of biomass gasifiers, various simplifying assumptions are included. For instance, while some assume instantaneous drying and pyrolysis, others consider these phenomena to be crucial components of the model8,10. Moreover, although the gasification product gas is often assumed to be free from tar11, some models include a sub-model for tar generation and cracking10. Two comprehensive reviews on modeling of fluidized bed biomass gasifiers are helpful to identify the knowledge gaps12,13. The unique features of the model developed in this study are revealed by comparing it with a number of kinetic models of fluidized bed biomass gasifiers in the literature2. Our comprehensive one-dimensional two-phase phase kinetic model evaluates the overall performance of a dual fluidized bed biomass gasifier by closing mass and energy balances over the entire system to predict the solids circulation rate and additional fuel requirements for stable operation of the process. Instead of assuming instantaneous biomass pyrolysis, as is common in the literature, the two-step biomass pyrolysis kinetic mechanism allows for tar generation and

ACS Paragon Plus Environment

3

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 62

cracking to be predicted, as well as addressing the knowledge gap in predicting pyrolysis products yield and composition. Furthermore, by using the ultimate analysis to define tar as a mixture of carbon, hydrogen and oxygen, uncertainties in tar measurement and analysis are addressed. The objective of this paper is to apply a steady-state kinetic reactor model to predict the performance of steam gasification of biomass in the dual fluidized bed reactor at the University of British Columbia (UBC). As shown schematically in Figure 1, this reactor consists of two interconnected fluidized beds, referred to as the gasifier and riser/combustor. Superheated steam fluidizes the gasifier in the bubbling flow regime to produce syngas by partial oxidation of solid biomass fed to the gasifier. Unreacted char (a by-product of biomass gasification), together with bed material (inert silica sand, SiO2), are circulated from the bottom of the gasifier to the bottom of the combustor, which operates at a higher temperature. The combustor is a riser fluidized by air at a high superficial gas velocity in the fast fluidization flow regime, as in a circulating fluidized bed riser. The heat requirements of the highly endothermic biomass gasification reactions are provided by circulation of hot bed material (mostly sand) from the top of the combustor through a cyclone to the gasifier. The combustion of unreacted char is insufficient to maintain the combustor temperature at the desired level due to the large heat losses experienced by a pilot scale unit caused by its high surface-to-volume ratio. Therefore, additional fuel (here natural gas) is burnt with air in an air pre-heater/natural gas burner prior to the combustor.

Figure 1. Schematic of UBC dual fluidized bed gasification pilot system14.

ACS Paragon Plus Environment

4

Page 5 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

MODEL DEVELOPMENT Upon entering a fluidized bed gasifier, biomass particles are exposed to very high heating rates, causing rapid increase of particle temperature, evaporation of moisture and devolatilization into volatile matter that may constitute more than 80% of the original particle mass. The rest remains in the form of solid char which is subsequently consumed in heterogeneous gasification reactions in the presence of an oxidizing agent such as air or steam. During biomass drying and pyrolysis, the gasifying agent may not reach the particle surface due to high evaporation and devolatilization fluxes, whereas during gasification, the gasifying agent from the bulk gas stream is transported to the char surface, where it reacts heterogeneously. Therefore, drying, pyrolysis and gasification can be modeled as consecutive processes.

Simplifying Assumptions To develop a realistic but workable model for steam gasification of biomass in the dual-bed reactor, the following simplifying assumptions are adopted: 1. The system operates under steady-state conditions. 2. The freeboard6,8,15 and the bubbles8 ,10,15,16 are modeled as being free of solids. 3. Uniform temperature and total pressure are assumed throughout the BFB17. 4. The solids in the BFB reactor are perfectly mixed8,1821, and there is a uniform distribution of drying and pyrolysis products throughout the dense bed height 8. 5. Gases and volatiles are in plug flow inside the reactor8,1821. 6. The freeboard region of the BFB is modeled as a plug flow reactor without heat loss 6,8,15,20. 7. The particles are thermally thin, with instantaneous heat-up to the reactor temperature.

ACS Paragon Plus Environment

5

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 62

8. Pyrolysis involves a two-step process, with primary and secondary pyrolysis described by three parallel heterogeneous reactions and a homogeneous tar cracking reaction, respectively20 25. 9. First-order Arrhenius type chemical reactions are assumed for pyrolysis20 25. 10. The sand particles are inert and do not catalyze reactions26. 11. Catalysis by ash is also neglected. 12. Chemical reaction is the rate-controlling mechanism for char gasification inside the bubbling fluidized bed. 13. The ideal gas law applies for all volatiles (gas and bio-oil) released from biomass pyrolysis. 14. For elemental balances, each species, including unreacted biomass, char, non-condensable gas and tar, is treated as a homogeneous mixture of carbon, hydrogen and oxygen. Other elements, including nitrogen and sulfur, are neglected due to their low contents. 15. Complete char (and natural gas) combustion occur inside the CFB riser.

Reaction Kinetics As illustrated in Figure 2, the biomass pyrolysis products distribution is approximated from the generic two-step kinetic mechanism 2224, with primary pyrolysis modeled by three parallel firstorder reactions producing non-condensable gas, tar and char. Secondary pyrolysis is modeled by a first-order reaction producing non-condensable gas from thermal cracking of tar. The selected kinetic parameters for primary pyrolysis are adopted from Chan et al.24, who verified their pyrolysis model predictions with experimental results from lodgepole pine (wood) devolatilization. The reaction kinetics equation of Boroson et al.27 is used to model thermal

ACS Paragon Plus Environment

6

Page 7 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

cracking of tar (bio-oil) to non-condensable gas in the temperature range of 500–800oC, with stoichiometry: Tar   CO CO  CO2 CO2   H2 H2   CH4 CH4   Tarinert Tarinert

(1)

and mass-based stoichiometric coefficients:

 CO  0.56333,  CO2  0.11093,  H2  0.01733,  CH4  0.08841,  Tarinert  0.22

(2)

The above stoichiometry suggests that 78% of the primary tar is cracked, with the rest remaining unchanged. However, studies show that operating parameters, such as fluidizing agent, temperature, residence time and ash, affect tar cracking8, 28. To account for these effects, the yield of inert tar is assigned a value of zero (  Tar

inert

 0 ), while the remaining yields are

loaded on the non-condensable gas species (i = H2, CO, CO2 and CH4): wi ,tar cracking   i 0.78

(3)

All reaction rate constants are expressed in first-order and Arrhenius form as: k j  k0 j exp  E j RT . 

(4)

The kinetic parameters and heats of reaction are summarized in Table 1. In addition, the kinetic rate expressions of the five major gasification reactions included in the model are listed in Table 2. Note that the catalytic effects of metal components present in the biomass ash, such as Ca, Na and K, are not taken into account, despite indications that they can have significant impact on the biomass gasification reactions26.

Figure 2. Two-step kinetic mechanism adopted for biomass pyrolysis2224.

ACS Paragon Plus Environment

7

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 62

Table 1. Kinetic parameters for biomass pyrolysis. Table 2. Kinetic expressions for major gasification reactions.

For small biomass particles that meet the criterion of being thermally thin (i.e. Biot numbers significantly smaller than 1), internal heat conduction is much faster than external heat transfer, and uniform temperature can be assumed throughout the particle volume35. For these particles, evaporation of the moisture content, biomass devolatilization and particle heat-up to the reactor temperature all occur almost instantaneously. Therefore, the yields of biomass pyrolysis products (per unit mass of dry biomass) are approximated at the reactor temperature by:

YG, pyro  k1  k1  k2  k3 

(5)

YT , pyro  k2  k1  k2  k3 

(6)

YC , pyro  k3  k1  k2  k3 

(7)

Assuming that tar, char and non-condensable gas are homogeneous mixtures of carbon, hydrogen and oxygen, the product gas composition is approximated from CHO elemental balances for an average particle experiencing pyrolysis inside the reactor. For a given reactor temperature, average elemental compositions are assigned to each lumped species, and the particle mass balance is broken down into CHO elemental balances. Details of the elemental balances are discussed elsewhere36. BFB Gasifier Two-Phase Kinetic Reactor Model In our steady-state gas-solid bubbling fluidized bed reactor model, while perfect mixing is assumed for solid particles, the gas phase is modeled by a generalized version of the one-

ACS Paragon Plus Environment

8

Page 9 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

dimensional two-phase model37, 38. As shown schematically in Figure 3, the water vapour, gas and tar generated from the drying, pyrolysis and gasification of biomass/char particles in the solid phase are assumed to be transferred uniformly to the high-density (emulsion) phase along the reactor height. Furthermore, there is a net flow of excess gas from the emulsion phase to the low-density (bubble) phase. The mixture of fluidizing agent (i.e. N2 and steam) and gaseous products released from gas-solid reactions are modeled as being in plug flow in both the bubble and emulsion phases.

Figure 3. Schematic of two-phase biomass steam gasification reactor model for the BFB.

Due to the very good heat transfer of bubbling fluidized beds, uniform temperature is assumed throughout the dense region of the BFB as a first approximation. In the absence of differential equations for energy balances and neglecting dispersion (second-order) terms, a system of differential equations for the gas mole balances in the low- and high-density phases, as well as a differential equation for the reactor pressure balance along the dense bed height, are solved together. For the bubble or low-density phase and i=1,..., Nc: 

1 d ''' ''' .  L .CiL   L .aI LH .kci ,LH .  CiH  CiL   L . Bulk .CiL  H . Bulk .CiH  L .RateiL  0 L H H L A dz

(8)

For the emulsion or high-density phase and i=1,..., Nc: 1 d ''' '''  .  H .CiH   H .aI H L .kci ,H L .  CiL  CiH   H . Bulk .CiH  L . Bulk .CiL  H .RateiH  0 H L LH A dz

(9)

Pressure balance:

ACS Paragon Plus Environment

9

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



dP  1    . s .g dz

Page 10 of 62

(10)

The variation in volumetric flow rate of each phase along the reactor is calculated by assuming that the ideal gas law applies: R.T NC   . Fi P i 1

(11)

with the molar flow rate of each species in phase φ:

Fi    .Ci

(12)

The bed superficial gas velocity at every height in the bed is:

U   L   H  A

(13)

The bed volume fractions must add up to unity:

 L  H  1

(14)

Table 3 summarizes the hydrodynamic equations used for kinetic modeling of BFB gasifier.

Table 3. Hydrodynamic equations used for BFB gasifier kinetic model.

According to the Mahecha-Botero38 calculation algorithm, to maintain fluidizing conditions at any integration point: a) If  H  m.U mf . A : Gas flows to the H-phase to avoid de-fluidization. ''' '''  Bulk  0 &  Bulk   m.U mf . A   H   H . A.z  H L L H

(15)

ACS Paragon Plus Environment

10

Page 11 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

. mf . A : There is no cross-flow. b) If  H  mU ''' '''  Bulk  0 &  Bulk 0 H L L H

(16)

. mf . A : Excess gas in the H-phase transfers to the L-phase. c) If  H  mU ''' '''  Bulk   H  m.U mf . A   H . A.z  &  Bulk 0 H L L H

(17)

Here m can be taken as 1.0 corresponding to the standard two-phase theory. Case c is often encountered for biomass gasification leading to the release of a significant amount of excess gas flow in the H-phase. Most of this excess gas then migrates to the L-phase to augment bubbles. To account for heterogeneous reactions, we need to know the distribution of solids between the two phases. From a solids material balance:

1     1   H  . H  1   L . L

(18)

and the mass fraction of solids going to phase φ is:

 1    1  

  .  

(19)

Given the perfect mixing assumption for particles in the BFB, we assume that the ratio of char hold-up (mChar) in adjacent cells of the fluidized bed is the same as the ratio of total solids holdup of those cells. With ∆z being the equal cell height, we have: mchar  z  z  1    z  z   . A.z  mchar ( z ) 1    z   . A.z

(20)

If MChar is the total char bed inventory and Lbed the expanded dense bed height, the local char hold-up in every cell is calculated as:

mchar  z  1    z   . A.z  M char 1   ave  A.Lbed

(21)

ACS Paragon Plus Environment

11

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 62

where the average bed voidage is given by

 ave

 Lbed      .dz  Lbed    0 

(22)

Therefore, the local char hold-up in phase φ is:  1   mchar  z  .   1 

 M char   .      Lbed

  1     .  .  .z   1   ave 

(23)

Assuming the abovementioned distribution of char particles between the cells and phases of the BFB also applies to biomass particles that go through drying and pyrolysis, we formulate the net reaction rates for different gaseous species. For solids-free bubbles, heterogeneous biomass pyrolysis and char gasification reactions only take place in the emulsion phase. The net reaction rate for tar is obtained from tar generation and cracking reaction rates:

 1    YT , pyro .m B,in MWTar RateTar ,     . k4 . 1   Tarinert .CTar ,  A.Lbed  1   ave 

 





(24)

Non-condensable gas species (i.e. i = H2, CO, CO2 and CH4) are generated /consumed from three sources including: I. Biomass pyrolysis occurring uniformly along the dense bed height:

 1     YG, pyro .m B,in .  wi , pyro MWi  Ratei, , pyrolysis   . A.Lbed  1   ave 

(25)

II. Homogeneous thermal tar cracking:

 





Ratei, ,tar cracking    . k4 . 1   Tarinert .CTar , .MWTar .  wi ,tar cracking MWi  III.

(26)

Homogeneous and heterogeneous gasification reactions:

 1      M Char MWChar RateCO, , gasification    .  rSMR  rWGS     . A.Lbed  1   ave  

  .  2.rC1  rC 2  

(27)

ACS Paragon Plus Environment

12

Page 13 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

 1      M Char MWChar RateCO2 , , gasification    .  rWGS     A.Lbed  1   ave  

  .   rC1  

(28)

 1     M Char MWChar RateCH4 , , gasification    .  rSMR     A.Lbed  1   ave 

  .  rC 3  

(29)

 1      M Char MWChar  RateH 2 , , gasification    .  3.rSMR  rWGS      .  rC 2  2.rC 3  A.Lbed  1   ave   

(30)

Combining Equations (25) to (30), the net reaction rate for non-condensable gas species i is:

Ratei,  Ratei, , pyrolysis  Ratei, ,tar cracking  Ratei, , gasification

(31)

Finally, the net reaction rate for water vapor/steam is obtained from drying, homogeneous and heterogeneous gasification reaction rates:

 1     m M ,in MWH 2O  1     M Char MWChar RateH 2O ,      .   rSMR  rWGS      A.Lbed A.Lbed  1   ave   1   ave 

  .   rC 2  

(32)

Given an initial guess for the char hold-up of the dense bubbling bed and using the MATLAB ODE solver (ODE45), we solve the above coupled ODEs numerically (13 ODEs). The code contains a set of algebraic subroutines to calculate the various model parameters used by the solver and the code complexity is significant (having around 1500 lines of code) to obtain the axial concentration profiles of steam, tar and non-condensable gas species, as well as the reactor pressure. The numerical scheme divides the BFB reactor into two zones (bottom dense bed and top freeboard) and meshes each zone with cells of equal size. A mesh independence study is carried out to choose the largest mesh that gives a mesh-independent solution. After running on an initial mesh and ensuring convergence of residual error to 10-4 with steady monitor points and imbalances below 1%, the mesh is refined globally so that we have finer cells throughout the domain and the process is repeated until a solution is reached that is independent of the mesh.

ACS Paragon Plus Environment

13

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 62

Given the thermally thin particle in a steady-state 1-D reactor model, the code is not computationally expensive. The mean solids residence time is defined as the solids hold-up inside the bubbling bed divided by mass flow rate of solid particles leaving the dense bed. In a dual fluidized bed reactor configuration, the transportation of particles out of the BFB takes place by circulation of solids to the combustor reactor. Hence, the mean solids residence time inside the BFB is calculated from experimental measurement of the sand inventory of dense bubbling bed ( M sand ) and the sand

 sand ): circulation mass flow rate ( m  s  M sand m sand

(33)

Therefore, the mass flow rate of char leaving the BFB is:

m Char ,out  M Char  s

(34)

A char balance over the dense bubbling bed at steady-state yields

m Char ,out  m Char , gen  m Char ,cons

(35)

where:

m Char , gen  YC, pyro .m B,in

(36)

Fluidized particles are often small enough that internal resistances to transfer are small. External transfer resistances are also likely to be smaller than those related to chemical reaction. Furthermore, the char particles are subject to severe attrition inside the bed. Therefore, it is reasonable to neglect the internal and external mass transfer resistances and assume that heterogeneous reactions are the rate-controlling mechanism for char gasification. Given the axial partial pressures of gasifying agents in the emulsion phase which contains all the char, the rate of char consumption due to heterogeneous gasification reactions (Table 2) is:

ACS Paragon Plus Environment

14

Page 15 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

m Char ,cons  

Lbed

0

 dm

Char , gas

dz  .dz  

Lbed

0

r  z  r C1, H

C 2, H

 z   rC 3, H  z   .  M char

Lbed  .dz

(37)

Substituting the above values in the char balance equation for a given mean solids residence time, the char hold-up of the bed is updated:

M char  YC, pyro .m B,in

1    s

Lbed

0

r  z  r C 1, H

C 2, H

 z   rC 3, H  z   .dz

Lbed



(38)

For a given pressure drop across the bubbling fluidized bed, the expanded bed height is obtained, iteratively:

Lbed 

  P  z  Lbed   P  z  0  

(39)

 sand . 1   ave  .g

Neglecting the char weight contribution, the total bed inventory (Wbed) is:

Wbed   sand . A. Lbed . 1   ave 

(40)

To reduce solids entrainment from the dense bubbling bed, the reactor diameter is expanded in the upper section. For the current model, the entrainment of char and sand particles are neglected, and the freeboard region is modeled as a plug flow reactor in which only homogeneous reactions (i.e. tar cracking, SMR & WGS) take place. Six ODE’s for the mole balances of gaseous species together with one ODE for temperature variation along the freeboard are solved simultaneously. As a first approximation and assuming a well-insulated freeboard, we ignore heat loss from the freeboard:  dCi , fb U dz  Ratei , fb  0 Freeboard :  Nr C  U dT   H .Rate   0 rxn ,k rxn ,k  Pg g dz  k 1

(41)

ACS Paragon Plus Environment

15

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 62

Boundary Conditions The schematic of the dual fluidized bed biomass steam gasification system in Figure 4 indicates the location of the boundary conditions. Except for the superheated steam used as the fluidizing gas, the concentrations of the other species at the bottom of the BFB gasifier are set equal to zero (assuming no air/gas carryover or leakage). The inlet gas flow is divided into the two phases, making use of a split factor:

FH 2O ,H ,in  Split H .FH 2O ,in

(42)

FH 2O , L ,in  FH 2O ,in  FH 2O , H ,in

(43)

with the split factor calculated as:



SplitH  m.U mf . A.CH 2O ,in



FH 2O ,in

(44)

The un-gasified char leaving the gasifier is burnt with excess air in the riser. As the energy released from char combustion is not able to provide enough heat for the process, natural gas is introduced as an additional fuel. To simplify the calculations, all the air required by the system, including that for char combustion, natural gas combustion and aeration air for transfer of solids from the gasifier to the combustor, is treated as a single stream introduced to the NG burner (air pre-heater) upstream of the combustor. Figure 4. Schematic of dual vessel system biomass steam gasification. Dashed and solid arrows denote energy and mass flows, respectively.

At a given stoichiometric O2 ratio (), the total oxygen flow rate required for complete combustion of natural gas ( CH4  2O2  CO2  2H2O ) and char ( C  s   O2  CO2 ) is:

ACS Paragon Plus Environment

16

Page 17 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels



FO2 ,total   m Char ,in 12  2  FCH 4 , Burner



(45)

where FCH , Burner is the molar flow rate of natural gas to the burner. 4 Therefore, the CFB riser inputs are:

FO2 ,in  FO2 ,total  2  FCH 4 , Burner

(46)

FN2 ,in   79 21  FO2 ,total

(47)

FH 2O ,in  2  FCH 4 ,Burner

(48)

FCO2 ,in  FCH 4 ,Burner

(49)

FCH 4 ,in  FH 2 ,in  FCO ,in  0

(50)

Overall Energy Balance Regardless of the reactions included in each reactor, energy balance calculations are only based on total thermodynamic enthalpies of all inlet and outlet streams crossing the system boundaries. In this manner, the thermodynamic states of streams and the heats of reactions are inherently taken into account. We have: Q   m j .H *j  Pj , T j    m i .H i*  Pi , Ti  j

(51)

i

To describe the thermodynamic states of the streams, substances are divided into four classes: ideal gases, inorganic solids, organic substances and pure water/steam48. NASA-polynomials are used to calculate isobaric heat capacities of ideal gases and inorganic solid species49. The empirical correlations of Boie48 and Merrick50 are applied to calculate the lower heating value of dry and ash-free biomass (LHVfuel) and the enthalpy and heat capacity of char as functions of

ACS Paragon Plus Environment

17

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 62

temperature. IAPWS-IF9751, 52 is used to estimate the enthalpy of sub-cooled liquid water (i.e. biomass moisture content) and superheated steam. Since the volumetric heat capacity of the solids is much higher than that of the gas, the outlet temperature and composition of the solid and gaseous products are the same as within the respective reactors. The constituents and temperatures of each stream are identified in Figure 4. The energy balance for the gasifier is solved for the circulation rate of sand between the two beds: * m fuel .H *fuel T fuel   m moisture .H H* 2O  Pfuel , T fuel   m steam .H H* 2O  Psteam , Tsteam   m sand .H SiO Tcomb  2 * * *  m char ,comb .H char Tcomb   m sand .H SiO Tgas   m char ,gas .H char Tgas   Qloss,gas   m i,gas .H i* Tgas   0 2 Nc

i 1

(52) Neglecting the circulation rate of char compared to sand, the net solids circulation flux for the given riser diameter is estimated to be:

Gs 

m sand 2 4  .Driser

(53)

As a fair approximation, we assume complete combustion of char in the riser and complete combustion of natural gas in the NG burner. The energy balance for the combustor, together with NG burner, is then solved to determine the flow rate of natural gas to the burner: * * * * m CH4 ,Burner .HCH Tair   m air .Hair* Tair   m sand .H SiO Tcomb  Tgas   m char,gas .Hchar Tgas   m sand .H SiO 4 2 2 Nc

* m char ,comb .Hchar Tcomb   Qloss,comb   m i,comb .Hi* Tcomb   0 i 1

(54) Thus, the riser superficial gas velocity is calculated from the CFB riser boundary conditions discussed above and the ideal gas law:

ACS Paragon Plus Environment

18

Page 19 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Nc R.Tcomb U  Fi,in 2 Pcomb .  .Driser 4  i 1

(55)

Note that the above riser superficial gas velocity and solids circulation flux are initial approximations from an energy balance point of view. In addition, a global pressure balance over the two interconnected fluidized bed reactors should be performed to find the solids circulation rate required to meet the hydrodynamic constraints of the DFB system under steady-state operation.

RESULTS AND DISCUSSION The predictions of the BFB gasifier model are compared with experimental results of Li et al.14 who studied the steam gasification of two types of pellets, hardwood (HW) and softwood (SW), in the BFB steam gasifier of the UBC DFB reactor. The properties of the feedstock and the experimental operating conditions are summarized in Tables 4 and 5, respectively. Temperatures were measured and averaged over six or seven axial thermocouple positions, and also over the gasification period. The particle properties and reactor dimensions are reported in Table 6. For particles of dp=790 μm (in this paper) and under typical operating conditions (superficial gas velocity~0.2 m/s, temperature=750-800oC, S/B~1), the Biot number is 0.2, not far above what is usually taken as the threshold for the lumped capacitance model (Bi=0.1). Therefore, over the range investigated, the assumption of thermally thin particles does not make a large difference in the final product gas composition.

Table 4. Feedstock analysis in experimental runs. Table 5. Operating conditions of UBC DFB gasifier.

ACS Paragon Plus Environment

19

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 62

Table 6. Particle and reactor conditions.

Model predictions of dry product gas composition are shown in Figure 5 as a function of reactor temperature, together with experimental data for steam gasification of wood pellets14. As seen, the model is successful at predicting the general trends and the CO2 concentration. However, there are significant quantitative differences in the CO, H2 and CH4 concentrations. In order to quantitatively evaluate the performance of the model, the absolute percent error for component i at each run is calculated by

% Eik  100  yie  yip yie where

yie

and

(56)

yip are the experimental and predicted volume percent of species i in dry

product gas, respectively. According to Table 7, the average percent error for overall model evaluation of H2, CO, CH4 and CO2 volume percentages for all runs are 66.6%, 40.3%, 72.4% and 16.9%, respectively. It is noted that the difference between the experimental and modeling results is likely to be influenced by heat loss effects in the freeboard and also experimental sampling methods that might have been affected by reverse reaction due to gas cooling38.

Table 7. Absolute percent error of dry gas composition (vol %).

At higher temperatures, H2 and CO production is promoted by the endothermic Boudouard, water-gas and steam methane reforming reactions. Elevating the temperature reverses the exothermic WGS reaction towards more H2 and CO production and greater CO2 consumption. Figure 5 shows that the model under-predicts the CO and CH4 contents of the product gas while over-predicting the H2 content. From the modeling point of view, under-prediction of

ACS Paragon Plus Environment

20

Page 21 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

methane in the syngas may be due to the very fast SMR kinetics, consuming most of the methane generated from biomass pyrolysis. On the other hand, the very high CH4 content measured experimentally in the UBC DFB gasifier suggests inadequate gas residence time in the reactor for effective reforming and cracking of hydrocarbons. In reality, the experimental steam conversion is less than predicted, suggesting that the steam fed to the gasifier reactor is not fully available for gasification reactions. Hence, low steam availability for the WGS and SMR reactions likely leads to the low H2 and high CO contents in the product gas observed experimentally. One of the features of the novel "sore-thumb" standpipe entrance used in the experimental tests14 is that the flow of steam leaving with the particles entering the U-bend from the bottom of the BFB gasifier is unknown. This portion of the total steam will not have been available for reaction in the gasifier. Therefore, several simulations were performed with different possible effective S/B ratios. As depicted by the dashed curves of Figure 5, for an effective steam-tobiomass ratio of 0.5, the hydrogen continues to be significantly over-predicted and the methane to be under-predicted, although their predictions are somewhat improved. The reasons for the lack of agreement with the H2 and CH4 predictions of the model are likely to be the stability of methane in the absence of steam reforming catalyst, as well as methanation (reverse reforming) reaction in the cooler freeboard of the gasifier. Note that our model does not account for heat losses from the freeboard and temperature drop in downstream components such as the cyclone, exit gas line and heat exchanger, given the difficulty of estimating the contribution of the different heating elements that are installed externally and insulated. Future work is needed to make better allowance for heat losses, temperature gradients and decreasing concentrations of

ACS Paragon Plus Environment

21

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 62

particles with height in the freeboard region of the gasifier, as well as catalytic effects of the components of the char and sand in the reactor system.

Figure 5. Effect of gasifier dense bed temperature on dry product gas composition for steam gasification of hard wood and soft wood. The unbroken and dashed curves are predictions from the model developed in this paper for S/B=1 and S/B=0.5, respectively, whereas the experimental data points are from Li et al.14.

In the following paragraphs the sensitivities of various operating parameters are explored by further simulations. Tables 6 and 8 summarize the input parameters used for the simulations. The operating conditions are typical of the UBC DFB gasifier. Given the difficulty of estimating heat loss, for simulation purposes we first assume a heat loss of 15%  m fuel .LHV fuel  for each reactor.

Table 8. Operating conditions for UBC DFB gasifier simulation.

The predicted changes in molar fractions of gaseous species in the bubble and emulsion phases along the height are illustrated in Figure 6. Dashed curves represent the emulsion phase, whereas the unbroken curves denote the bubble phase. Heterogeneous biomass gasification reactions are dominant in the emulsion phase where almost all of the solid char resides. Furthermore, the smaller gas velocity in the emulsion phase leads to increased gas residence time there. Hence, H2O is consumed more rapidly along the height in the emulsion phase. On the other hand, the H 2 concentration is higher in the emulsion phase than in the bubble phase, indicating that more

ACS Paragon Plus Environment

22

Page 23 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

hydrogen is produced in this phase. Due to the Boudouard reaction, the CO concentration is also larger in the emulsion phase than in the bubble phase. Figure 7 illustrates KWGS = equilibrium constant of the WGS reaction, together with the ratio κ = ([CO2].[H2])/([H2O].[CO]), as functions of height in the gasifier. Due to the drop in temperature along the freeboard, both KWGS and κ vary with height. As seen, at the top of the dense bubbling bed, the composition of gas mixture approaches equilibrium, and due to the increased gas residence time in the conical section of the reactor the composition there is also close to that at equilibrium. However, above the conical section, the gas mixture composition deviates substantially from that at equilibrium and almost flattens out due to the low reaction rates at the reduced temperatures.

Figure 6. Mole fractions of gas species along gasifier height, B: bubble, E: emulsion. Gasifier dense bed temperature=750oC, Steam/Biomass mass flow ratio=1.22.

Figure 7. Deviation from WGS reaction equilibrium constant (K WGS) as a function of height in the gasifier. Gasifier dense bed temperature=750oC, Steam/Biomass mass flow ratio=1.22.

The predicted dry product gas composition as a function of gasifier temperature is illustrated in Figure 8, where the unbroken curves are for dry product composition leaving the reactor and the dashed curves are for the gas composition just above the dense bubbling bed. Since the WGS reaction is exothermic, elevating the temperature reverses this reaction towards more CO production and CO2 consumption. CO and H2 production are also promoted by the endothermic SMR reaction. These trends are consistent with modeling and experimental data from the

ACS Paragon Plus Environment

23

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 62

literature53,54. The freeboard allows for the homogeneous reactions which enhance the consumption of CH4 and CO and the production of H2.

Figure 8. Predicted dry product gas composition as a function of gasifier dense bed temperature. Unbroken and dashed curves are with and without freeboard. Steam/biomass ratio=1.22. For other operating conditions, see Tables 6 and 8.

The predicted variations of different performance variables with gasifier temperature are plotted in Figure 9. As gasification is primarily an endothermic process, increasing temperature increases the product gas volumetric flow rate (QPG), at the cost of decreasing the lower heating value (LHVPG). Increasing the gasifier temperature at a fixed combustor temperature decreases the temperature difference between the two fluidized beds. In order to provide greater sensible heat transfer, required to maintain the gasifier at a higher temperature, the sand circulation rate must increase. Therefore, increasing gasifier temperature increases the solid circulation flux (Gs) and decreases the solids mean residence time (  s ) in the gasifier. By increasing the gasifier temperature, the mass flow rate of char transferred to the combustor ( m char ) decreases, at the cost of a slight increase in the required volumetric flow rate of natural gas (QNG) to the combustor.

Figure 9. Variation of key performance variables as a function of gasifier temperature. Steam/biomass ratio =1.22, combustor temperature=900oC.

We define the chemical efficiency of the process as

ACS Paragon Plus Environment

24

Page 25 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

chem 

QPG .LHVPG m fuel .LHV fuel

(57)

Figure 10 shows that both the carbon conversion and the chemical efficiency increase significantly with increasing gasifier temperature, resulting from enhanced gasification reactions which are highly endothermic. Figure 11 plots the predicted dry product gas composition as a function of the S/B ratio, where the unbroken and dashed curves are for the product composition with and without the freeboard, respectively. With increased steam, more H2 and CO2 are produced at the cost of increased consumption of CO and CH4 by means of the WGS and SMR reactions. Furthermore, because of the reactions occurring in the freeboard, the consumptions of CH4 and CO are enhanced, and the H2 production is increased.

Figure 10. Predicted chemical efficiency and carbon conversion as functions of gasifier temperature. Steam/biomass ratio =1.22.

Figure 11. Predicted dry product gas composition as a function of S/B ratio. Unbroken and dashed curves are for the composition with and without the freeboard. Gasifier temperature = 750oC. For other operating conditions, see Tables 6 and 8.

The predicted variations of different performance variables with steam/biomass ratio are illustrated in Figure 12. Increasing the steam/biomass ratio promotes gasification reactions towards increased product gas volumetric flow rate (QPG). Increasing the steam/biomass ratio decreases the product gas lower heating value (LHVPG), mainly due to increased CO2 content of the product gas.

ACS Paragon Plus Environment

25

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 62

As shown in Figure 13, the chemical efficiency does not change appreciably with variation of the steam/biomass ratio, whereas the carbon conversion is predicted to increase slightly with increasing steam/biomass ratio.

Figure 12. Performance variables as functions of S/B ratio. Gasifier dense bed temperature = 750oC, combustor temperature = 900oC. For other operating conditions, see Tables 6 and 8.

Figure 13. Predicted chemical efficiency and carbon conversion as functions of S/B ratio. Gasifier dense bed temperature = 750oC.

Unlike large plants where process heat integration and comparatively low heat losses lead to higher process efficiencies, for smaller reactors with higher ratio of heat transfer surface area to reactor volume, heat loss plays an important role in energy balance calculations. Note that the above definition of chemical efficiency does not truly represent the overall process efficiency because it does not take into account the external heating requirements of the process. In the literature 55, 56, the overall process efficiency is defined in two ways:



Overall , I   QPG .LHVPG  m fuel .LHV fuel  QCH 4 .LHVCH 4



Overall , II  QPG .LHVPG  QCH 4 .LHVCH 4

  m

fuel



.LHV fuel 

(58)

(59)

The first definition gives the energy obtained in the product gas divided by the total energy input to the system, where the input includes the biomass and any additional fuel such as CH 4. This definition can be generalized by including other energy inputs, such as the heat provided by electric heaters to preheat the air and to heat the superheated steam to the desired gasifier

ACS Paragon Plus Environment

26

Page 27 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

temperature. The second definition, which gives lower values of process efficiency, is particularly meaningful when one considers the special case of recycling a fraction of product gas to be combusted in the riser. Figure 14 demonstrates the effect of heat loss on the chemical and overall process efficiencies. The heat loss is taken into account as the fraction of biomass energy input ( m fuel . LHV fuel ) to the gasifier. As shown in Figure 14, if we assume that the entire heat loss of the dual fluidized bed system occurs from the gasifier, a lower limit for chemical efficiency is obtained. On the other hand, if the entire heat loss is from the combustor, one obtains an upper limit for chemical efficiency which does not change with variation of the heat loss from the combustor. When it comes to overall process efficiency, the distribution of heat losses in the system does not matter, and the overall process efficiency decreases sharply with increasing total heat loss. When heat loss is negligible, the maximum overall efficiency is around 75%, close to the chemical efficiency.

Figure 14. Chemical and overall process efficiencies as functions of overall heat loss. Gasifier dense bed temperature = 800oC, Combustor temperature = 900oC, steam/biomass ratio=1.22, excess air=20%.

CONCLUSIONS The performance of a dual fluidized bed gasifier was simulated with a steady-state predictive kinetic reactor model that uses a generic version of the two-phase fluidized bed reactor model. This model predicts species concentrations, temperature, pressure, superficial gas velocity and other hydrodynamic parameters such as voidage profiles along the reactor. By integrating the

ACS Paragon Plus Environment

27

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 28 of 62

reactor models of a bubbling fluidized bed (BFB) gasifier and a circulating fluidized bed (CFB) riser/combustor through mass and energy balances, key operating parameters such as solids circulation rate and natural gas flow rate required for stable operation of the process are predicted. Predictions of the BFB gasifier reactor model are compared with experimental results from the UBC DFB gasifier. Sensitivity analyses were performed with respect to the steam-tobiomass (S/B) ratio, gasifier temperature and system heat loss. Sensitivity analysis shows that for smaller S/B ratios, the dry product gas composition predicted by the model is in better agreement with experimental data. This suggests that low steam availability for WGS and SMR reactions, due to it being drawn down the U-bend, might have led to the low H2 and high CH4 contents in the product gas observed in the experimental measurements. High heat losses in the pilot scale dual bed may also have been a significant factor. Future work will focus on heat losses and catalytic effects of particles in the gasifier freeboard. The developed kinetic model gives the following insights: (1) Pyrolysis is a key step greatly affecting the model predictions. To develop a reliable model, in addition to the yields of pyrolysis products that are often modeled as lumped species, an attempt was made to address an existing gap of knowledge in predicting the composition of major compounds in pyrolysis gas, based on the generic two-step biomass pyrolysis kinetic mechanism and CHO elemental balances. (2) By defining tar as a mixture of carbon, hydrogen and oxygen, uncertainties in tar measurement and analysis are addressed. However, major effort is still needed to improve the predictability of char and tar conversion processes. (3) Uncertainties about the model structure can be reduces by obtaining and comparing further experimental dual bed results with model predictions.

ACS Paragon Plus Environment

28

Page 29 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

AUTHOR INFORMATION Corresponding Authors *2360 East Mall, Vancouver, Canada, V6T 1Z3, Phone: +98 915-580-4299. E-mail: [email protected] & [email protected] Present Addresses ‡ NORAM

Engineering, 200 Granville Street, Suite 1800 Vancouver, BC, Canada, V6C 1S4

ACKNOWLEDGMENT The authors gratefully acknowledge financial aid from Carbon Management Canada, BioFuelNet Canada, and the NSERC CREATE program. NOMENCLATURE A

Bubbling bed cross-sectional area, m2

a I L H

Interphase transfer area between phases per unit volume, m -1

Ar

Archimedes number, -

Bi

Biot number, -

Ci

Concentration of species i, mol/m3

Ci

Concentration of species i in phase φ, mol/m 3

Cp

Specific heat capacity, J/kg.K

db

Bubble diameter, m

Dij

Binary diffusion coefficient between species i and j, m 2/s

dp

Particle diameter, m

Driser

Riser internal diameter, m

ACS Paragon Plus Environment

29

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Ej

Activation energy of jth reaction, kJ/mol

Fi

Molar flow rate of species i in phase φ, mol/s

g

Acceleration of gravity, 9.81 m2/s

GS

Solid circulation flux, kg/m2.s

H i*

Total enthalpy of stream i, J/mol or J/kg

K

Equilibrium constant, -

kci ,LH

Interphase mass transfer coefficient of species i, m/s

k0 j

Pre-exponential factor of jth reaction, s-1

kj

Kinetic rate constant of jth reaction, s-1

Lbed

Dense bubbling bed height, m

m

Mass, kg

m

Parameter for modified two-phase theory, -

m

Mass flow rate, kg/s

MChar

Char hold-up inside reactor, kg

MW

Molecular weight, g/mol

NC

Number of gas components, -

Nor

Number of holes in distributor plate, -

Nr

Number of reactions, -

P

Reactor pressure, Pa

Pi

Partial pressure of ith species, Pa or bar

P0

Standard pressure, 1 bar

Q

Volumetric flow rate, m3/s or Nm3/h

Q

Net heat source/sink, kW

Qloss

Heat loss, kW

r

Reaction rate, s-1 or mol/m3.s

Page 30 of 62

ACS Paragon Plus Environment

30

Page 31 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

R

Universal ideal gas constant, 8.314 J/mol.K

Ratei,

Net reaction rate of species i in phase φ, mol m 3φ .s

SplitH

H-phase flow split factor, -

t

Time, s

T

Temperature, K

U

Superficial gas velocity, m/s

Ub

Bubble rise velocity, m/s

Uc

Superficial gas velocity at transition from bubbling to turbulent fluidization regime, m/s

Umb

Minimum bubbling velocity, m/s

Umf

Minimum fluidization velocity, m/s

Use

Transition velocity from turbulent to fast fluidization regime/significant entrainment, m/s

U gas

Convective gas velocity of phase φ, m/s

u

Absolute gas velocities in phase φ, m/s

w

Mass fraction, -

Wbed

Solids bed inventory, kg

XC

Overall carbon conversion, -

yi

Mole fraction of ith species, -

Y

Average yield, kg/kg dry biomass

z

Axial coordinate along reactor height, m

Greek

i

Stoichiometric coefficient of ith species in tar cracking reaction, -

o H rxn

Heat of reaction at standard condition, kJ/kg or kJ/mol

H rxn , j

Heat of jth reaction, kJ/kg



Bed voidage at height z, -

 ,ave

Average bubbling bed voidage, -

ACS Paragon Plus Environment

31

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 62



Void fraction of phase φ, -

 chem

Chemical efficiency, -

κ

Ratio of ([CO2].[H2])/([H2O].[CO])



Stoichiometric oxygen ratio for combustion, -



Viscosity, kg/m.s



Volumetric flow rate of phase φ, m3/s

'''  Bulk L H

Volumetric flow rate convectively transferred from L-phase to H-phase per unit volume of phase, s-1



Density, kg/m3

s

Mean solids residence time, s

φ

Phase L or H

S

Particle sphericity, -



Volume fraction of phase φ, -



Temperature-dependent collision integral

Subscripts b

Bubble

B

Biomass

C

Char, Carbon

comb

Combustor/riser

eq

Equilibrium

fb

Freeboard

fuel

Dry-ash-free biomass feed

g

Gas phase

G

Non-condensable gas

G2

Secondary gas

H

Hydrogen/High density phase (emulsion or annulus)

ACS Paragon Plus Environment

32

Page 33 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

i

Species number

j

Reaction number, Element number

L

Low density phase (bubble or core)

M

Moisture

mf

Minimum fluidization

NG

Natural gas

O

Oxygen

P

Particle

PG

Product gas

S

Solid/sand

T

Tar

V

Water vapor

ABBREVIATIONS BFB

Bubbling Fluidized Bed

CFB

Circulating Fluidized Bed

DFB

Dual Fluidized Bed

FICFB

Fast Internally Circulating Fluidized Bed

HW

Hardwood

LHV

Lower Heating Value

Rxn

Reaction

S/B

Steam/biomass ratio

SMR

Steam-Methane Reforming

SW

Softwood

WGS

Water-Gas Shift

ACS Paragon Plus Environment

33

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 62

REFERENCES (1)

Edenhofer, O.; Pichs-Madruga, O.; Sokona, Y.; Seyboth, K.; Matschoss, P.; Kadner, S.;

Zwickel, T.; Eickemeier, P.; Hansen, G.; Schlömer, S.; von Stechow, C. IPCC Special Report on Renewable Energy Sources and Climate Change Mitigation; Cambridge University Press: Cambridge, New York, 2011. (2)

Ahmed, T. Y.; Ahmad, M. M.; Yusup, S.; Inayat, A.; Khan, Z. Mathematical and

computational approaches for design of biomass gasification for hydrogen production: A review. Renewable Sustainable Energy Rev. 2012, 16, 23042315. (3)

Corella, J.; Toledo, J. M.; Molina, G. A review on dual fluidized-bed biomass gasifiers.

Ind. Eng. Chem. Res. 2007, 46, 6831–6839. (4)

Hofbauer, H.; Rauch, R.; Loeffler, G.; Kaiser, S.; Fercher, E.; Tremmel, H. Six years

experience with the FICFB-gasification process. In Proceedings of the 12th European Conference and Technology Exhibition on Biomass for Energy, Industry and Climate Protection; ETA: Florence, Italy (and WIP, Munich, Germany), 2002, 952985. (5)

Corella, J.; Sanz, A. Modeling circulating fluidized bed biomass gasifiers: A pseudo-

rigorous model for stationary state. Fuel Process. Technol. 2005, 86, 10211053. (6)

Radmanesh, R.; Chaouki, J.; Guy, C. Biomass gasification in a bubbling fluidized bed

reactor: experiments and modeling. AlChE J. 2006, 52, 42584272. (7)

Lü, P.; Kong, X.; Wu, C.; Yuan, Z.; Ma, L.; Chang, J. Modeling and simulation of

biomass air-steam gasification in a fluidized bed. Front. Chem. Eng. China 2008, 2, 209–213.

ACS Paragon Plus Environment

34

Page 35 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

(8)

Kaushal, P.; Abedi, J.; Mahinpey, N. A comprehensive mathematical model for biomass

gasification in a bubbling fluidized bed reactor. Fuel 2010, 89, 36503661. (9)

Kaushal, P.; Proell, T.; Hofbauer, H. Model for biomass char combustion in the riser of a

dual fluidized bed gasification unit: Part 1 – model development and sensitivity analysis. Fuel Process. Technol. 2008, 89, 651–659. (10) Hamel, S.; Krumm, W. Mathematical modelling and simulation of bubbling fluidised bed gasifiers. Powder Technol. 2001, 120, 105–12.

(11) Fiaschi, D.; Michelini, M. A two-phase one-dimensional biomass gasification kinetics model. Biomass Bioenergy 2001, 21, 121–32. (12) Gómez-Barea, A.; Leckner, B. Modeling of biomass gasification in fluidized bed. Prog. Energy Combust. Sci. 2010, 36, 444–509. (13) Basu, P.; Kaushal, P. Modeling of pyrolysis and gasification of biomass in fluidized beds: a review. Chem. Prod. Process Model 2009, 4, 1–45. (14) Li, Y. H.; Chen, Z.; Watkinson, P.; Bi, X.; Grace, J. R.; Lim, C. J.; Ellis, N. A novel dualbed for steam gasification of biomass, submitted to Biomass Conversion and Biorefinery. (15) Ji, P.; Feng, W.; Chen, B. Comprehensive simulation of an intensified process for H 2 production from steam gasification of biomass. Ind. Eng. Chem. Res. 2009, 48, 3909–20. (16) Sadaka, S,; Ghaly, A. E.; Sabbah, M. A. Two phase biomass air–steam gasification model for fluidized bed reactors: Part I – model development. Biomass Bioenergy 2002, 22, 439–62.

ACS Paragon Plus Environment

35

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 36 of 62

(17) Ammendola, P.; Chirone, R.; Miccio, F.; Ruoppolo, G.; Scala, F. Devolatilization and attrition behavior of fuel pellets during fluidized-bed gasification. Energy Fuels 2011, 25, 1260–66. (18) Kunii, D.; Levenspiel, O. Fluidization Engineering (2nd Ed.); Boston: Butterworth, 1991. (19) Yang, W. C. Handbook of Fluidization and Fluid-Particle Systems, Marcel-Dekker Inc., New York: Taylor & Francis Group LLC, 2003. (20) Kaushal, P.; Abedi, J. A simplified model for biomass pyrolysis in a fluidized bed reactor. JIEC 2010, 16, 748–755. (21) Liden, A.G.; Berruti, F.; Scott, D.S. A kinetic model for the production of liquids from the flash pyrolysis of biomass. Chem. Eng. Comm. 1988, 65, 207–221. (22) Grønli, M. G.; Melaaen, M. C. Mathematical model for wood pyrolysis: comparison of experimental measurements with model predictions. Energy Fuels 2000, 14, 791–800. (23) Shafizadeh, F.; Chin, P. P. S. Wood Technology: Chemical Aspects. In Chemical Society Symposium Series; Goldstein, I. S.,

American

Ed.; American Chemical Society:

Washington, DC, 1977, Vol. 43, 57. (24) Chan, W. C. R.; Kelbon, M.; Krieger, B. B. Modelling and experimental verification of physical and chemical processes during pyrolysis of a large biomass particle. Fuel 1985, 64, 1505–1513. (25) Di Blasi, C.; Branca, C. Kinetics of primary product formation from wood pyrolysis. Ind. Eng. Chem. Res. 2001, 40, 5547–5556.

ACS Paragon Plus Environment

36

Page 37 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

(26) Sutton, D.; Kelleher, B.; Ross, J. R. H. Review of literature on catalysts for biomass gasification. Fuel Process. Technol. 2001, 73, 155–173. (27) Boroson, M. L.; Howard, J. B,; Longwell, J. P.; Peters, W. A. Product yields and kinetics from the vapor phase cracking of wood pyrolysis tars. AIChE J. 1989, 35, 120–128. (28) Rapagnà, S.; Jand, N.; Kiennemann, A.; Foscolo, P. U. Steam-gasification of biomass in a fluidized-bed of olivine particles. Biomass Bioenergy 2000, 19, 187–197. (29) Barrio, M.; Hustad, J. E. CO2 gasification of birch and the effect of CO inhibition on the calculation of chemical kinetics. In Progress in Thermochemical Biomass Conversion; Bridgwater, A. V., Ed.; Blackwell Science Ltd.: Oxford, U.K., 2001. (30) Barrio, M.; Gobel, B.; Risnes, H.; Henriksen, U.; Hustad, J.; Sorensen, L. Steam gasification of wood char and the effect of hydrogen inhibition on the chemical kinetics. In Progress in Thermochemical Biomass Conversion; Bridgwater, A. V., Ed.; Blackwell Science Ltd.: Oxford, U.K., 2001. (31) Babu, B.; Sheth, P. Modeling and simulation of reduction zone of downdraft biomass gasifier: effect of char reactivity factor. Energy Convers. Manage. 2006, 47, 2602–2611. (32) Gómez-Barea, A.; Leckner, B. Estimation of gas composition and char conversion in a fluidized bed biomass gasifier. Fuel 2013, 107, 419–443. (33) Biba, V.; Macak, J.; Klose, E.; Malecha, J. Mathematical modeling for the gasification of coal under pressure. Ind. Eng. Chem. Process Des. Dev. 1978, 17, 92–98. (34) Parent, J. D.; Katz, S. Equilibrium compositions and enthalpy changes for the reaction of carbon, oxygen, and steam. Research Bulletin 2; IGT-Inst. Gas Tech., 1948.

ACS Paragon Plus Environment

37

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 38 of 62

(35) Murzin, D. Y. Chemical Engineering for Renewables Conversion; Academic Press, 2012. (36) Hejazi, B.; Grace, J. R.; Bi, X.; Mahecha-Botero, A. Kinetic model of steam gasification of biomass in a bubbling fluidized bed reactor. Energy Fuels 2017, 31, 1702−1711. (37) Abba, I. A.; Grace, J. R.; Bi, H. T.; Thompson, M. L. Spanning the flow regimes: Generic fluidized bed reactor model. AIChE J. 2003, 49, 18381848. (38) Mahecha-Botero, A. Comprehensive modeling and its application to simulation of fluidized-bed reactors for efficient production of hydrogen and other hydrocarbon processes. Ph. D. Thesis, University of British Columbia, 2009. (39) Grace, J. R. Fluidized-bed Hydrodynamics, in Handbook of Multiphase Systems; Hetsroni, G., Ed.; Hemisphere: Washington, DC, 1982, pp 5–64. (40) Wen, C. Y.; Yu, Y. H. Mechanics of fluidization. Chemical Engineering Program Symposium Series 1996, 62, 100–111. (41) Clift, R.; Grace, J. R. Continuous bubbling and slugging, Fluidization; Academic Press: London, 1985; pp 73–132. (42) Darton, R. C.; Lanauze, R. D.; Davidson, J. F.; Harrison, D. Bubble growth due to coalescence in fluidized-beds. Trans. Inst. Chem. Eng. 1977, 55, 274–280. (43) Davidson, J. F.; Harrison, D. Fluidized Particles; Cambridge University Press: Cambridge, New York, 1962. (44) Cussler, E. L. Diffusion: Mass Transfer in Fluid Systems (2nd ed.); Cambridge University Press: Cambridge, New York, 1997.

ACS Paragon Plus Environment

38

Page 39 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

(45) Hirschfelder, J., Curtiss, C. F., Bird, R. B. Molecular Theory of Gases and Liquids; Wiley, New York, 1954. (46) Wilke, C. R. Diffusional properties of muticomponent gases. Chem. Eng. Prog. 1950, 46, 95–104. (47) Sit, S. P., Grace, J. R. Effect of bubble interaction on interphase mass transfer in gas fluidized beds. Chem. Eng. Sci. 1981, 36, 327–335. (48) Pröll, T., Hofbauer, H. Development and application of a simulation tool for biomass gasification based processes. Int. J. Chem. Reactor Eng. 2008, 6, A89. (49) McBride, B. J.; Gordon, S.; Reno, M. A. Coefficients for calculating thermodynamic and transport properties of individual species. NASA Technical Memorandum 4513, NASA, 1993. (50) Merrick, D. Mathematical models of the thermal decomposition of coal Parts 16. Fuel 1988, 62, 534570. (51) Dooley, B. Release on the IAPWS industrial formulation 1997 for the thermodynamic properties of water and steam. IAPWS Secretariat, EPRI, Palo Alto CA, 1997. (52) Burcat, A.; McBride, B. Ideal gas thermodynamic data for combustion and air pollution use, Report No. TAE 804, Technion Israel Institute of Technology, Aerospace Engineering, 1997. (53) Schuster, G.; Loffler, G.; Weigl, K.; Hofbauer, H. Biomass steam gasification – an extensive parametric modeling study. Bioresour. Technol. 2001, 77, 7179.

ACS Paragon Plus Environment

39

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 40 of 62

(54) Hofbauer, H.; Rauch, R. Stoichiometric water consumption of steam gasification by the FICFB-gasification process. In Progress in Thermochemical Biomass Conversion, Innsbruck, Austria, 2000. (55) Pfeifer, C.; Rauch, R.; Hofbauer, H. In-bed catalytic tar reduction in a dual fluidized bed biomass steam gasifier. Ind. Eng. Chem. Res. 2004, 43, 1634–1640. (56) Higman, C.; van der Burgt, M. Gasification; Gulf Professional Publishing: USA, 2003.

ACS Paragon Plus Environment

40

Page 41 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 1. Schematic of UBC dual fluidized bed gasification pilot system 14.

ACS Paragon Plus Environment

41

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 42 of 62

Figure 2. Two-step kinetic mechanism adopted for biomass pyrolysis 2224.

ACS Paragon Plus Environment

42

Page 43 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 3. Schematic of two-phase biomass steam gasification reactor model for the BFB.

ACS Paragon Plus Environment

43

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 44 of 62

Figure 4. Schematic of dual vessel system biomass steam gasification. Dashed and solid arrows denote energy and mass flows, respectively.

ACS Paragon Plus Environment

44

Page 45 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 5. Effect of gasifier dense bed temperature on dry product gas composition for steam gasification of hardwood and softwood. The unbroken and dashed curves are predictions from the model developed in this paper for S/B=1 and S/B=0.5, respectively, whereas the experimental data points are from Li et al.14.

ACS Paragon Plus Environment

45

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 46 of 62

Figure 6. Mole fractions of gas species along gasifier height, B: bubble, E: emulsion. Gasifier dense bed temperature=750oC.

ACS Paragon Plus Environment

46

Page 47 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 7. Deviation from WGS reaction equilibrium constant (K WGS) as a function of height in the gasifier. Gasifier dense bed temperature=750oC, Steam/Biomass mass flow ratio=1.22.

ACS Paragon Plus Environment

47

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 48 of 62

Figure 8. Predicted dry product gas composition as a function of gasifier dense bed temperature. Unbroken and dashed curves are with and without freeboard. Steam/biomass ratio=1.22. For other operating conditions, see Tables 6 and 8.

ACS Paragon Plus Environment

48

Page 49 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 9. Variation of key performance variables as a function of gasifier temperature. Steam/biomass ratio =1.22, combustor temperature=900oC.

ACS Paragon Plus Environment

49

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 50 of 62

Figure 10. Predicted chemical efficiency and carbon conversion as functions of gasifier temperature. Steam/biomass ratio =1.22.

ACS Paragon Plus Environment

50

Page 51 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 11. Predicted dry product gas composition as a function of S/B ratio. Unbroken and dashed curves are for the composition with and without the freeboard. Gasifier temperature=750oC. For other operating conditions, see Tables 6 and 8.

ACS Paragon Plus Environment

51

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 52 of 62

Figure 12. Performance variables as functions of S/B ratio. Gasifier dense bed temperature = 750oC, combustor temperature = 900oC. For other operating conditions, see Tables 6 and 8.

ACS Paragon Plus Environment

52

Page 53 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Figure 13. Predicted chemical efficiency and carbon conversion as functions of S/B ratio. Gasifier dense bed temperature = 750oC.

ACS Paragon Plus Environment

53

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 54 of 62

Figure 14. Chemical and overall process efficiencies as functions of overall heat loss. Gasifier dense bed temperature = 800oC, Combustor temperature = 900oC, steam/biomass ratio=1.22, excess air=20%.

ACS Paragon Plus Environment

54

Page 55 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Table 1. Kinetic parameters for biomass pyrolysis a.

a

j

k0j (1/s)

Ej (kJ/mole)

∆Hrxn,j (kJ/kg)

1b

1.30  108

140

64

2b

2.00  108

133

64

3b

1.08  107

121

64

4c

1  105

93.3

-42

For mechanism, see Figure 2. Reference 27.

b

Reference 24.

c

ACS Paragon Plus Environment

55

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 56 of 62

Table 2. Kinetic expression for major gasification reactions. Gasification reaction Boudouard29 a

Reaction kinetic rate expression

C  s   CO2  2CO H

Water-Gas30 a

0 rxn

mol

 131 kJ

mol

C  s   2 H 2  CH 4 H

a

 172 kJ

C  s   H 2O  CO  H 2 H

Methanation31 a

0 rxn

0 rxn

 75 kJ

mol

Steam-Methane Reforming (SMR) 32

CH 4  H 2 O  CO  3 H 2

Water-Gas Shift (WGS) 33, 34

CO  H 2O  CO2  H 2

0  H rxn   206 kJ

H

0 rxn

 41 kJ

mol

mol

 215000  0.38 rC1  s 1   3.1 106 exp    PCO2  bar   R.T   237000  0.57 rC 2  s 1   2.62 108 exp    PH 2O bar   R.T 

 94800  0.93 rC 3  s 1   16.4 exp    PH 2  MPa   R.T 

 125000  rSMR  mol.m 3 .s 1   3  105 exp    CCH 4 .CH 2O  R.T   3968  KWGS  0.0265exp    T  CCO2 .CH2   1510   rWGS  mol.m3 .s 1   2.78exp      CCO .CH2O  KWGS   T  

First order reactions with respect to solid Carbon. R=8.314 J.mol -1.K-1, T [K] and Ci [mol.m-3].

ACS Paragon Plus Environment

56

Page 57 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Table 3. Hydrodynamic equations used for BFB gasifier kinetic model. Parameter

Equation

Volumetric flow rate of phase φ

     .  .U gas . A

Superficial gas velocity

U   L   H  A

Archimedes number

Ar   g .  s   g  .g.d 3p  2

Minimum fluidization velocity39

Remf   g .U mf .d p   27.22  0.0408. Ar  27.2

Voidage at minimum fluidization40

 mf  14.S 

Bed voidage41

  1  1   mf

Volume fraction of bubble phase

 L      mf

Bubble diameter42

db  0.56 g 0.2 U  U mf 

Bubble rise velocity43

U b  U  U mf  . 1   0.711 U  g.db

Diffusion coefficient in mixture46

transfer coefficient47

1  U  U   0.711



Dij 

g .d b

mf

 

L0

  mf 0.4





z  4



Binary diffusion coefficient44, 45

Inter-phase volumetric mass

1 3



A N or





1.858 103 T 3 2  MWi  MW j   MWi .MW j 

Di  1  yi 



1

2

P. ij2 . NC

 y j 1 j i

i

Dij 



aI LH .kci ,LH   6 db  U mf 3  2. Di . mf .U b  .db 



ACS Paragon Plus Environment

57

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 58 of 62

Table 4. Feedstock analysis in experimental runs. Type of feedstock

Hardwood (HW)

Softwood (SW)

Pellet Particle size

6.35  12.7 mm

6.35  12.7 mm

3.4

5.4

Moisture (%, wet basis)

Proximate analysis (wt% basis) Volatiles

82.3

81.7

Fixed carbon

17.1

17.3

Ash

0.6

1.0

Ultimate analysis (wt%, dry-ash-free basis) C

48.1

50.8

H

5.61

6.26

O

45.82

42.62

N

0.37

0.22

S

0.10

0.10

ACS Paragon Plus Environment

58

Page 59 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Table 5. Operating conditions of UBC DFB gasifier. Run

1

2

3

4

5

6

7

8

Biomass type

HW

HW

HW

SW

SW

SW

SW

SW

Biomass fuel flow rate (kg/h)

10.5

10.8

10.9

10.7

14.6

10.9

10.0

10.0

Steam flow rate (kg/h)

10.3

10.3

10.3

10.3

10.3

10.3

10.2

10.3

Gasifier dense bed temperature (°C )

690

750

790

830

800

824

831

750

Combustor top temperature (°C )

740

820

900

930

884

920

942

870

ACS Paragon Plus Environment

59

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 60 of 62

Table 6. Particle and reactor conditions. Riser diameter

0.1 m

Riser height

6.5 m

Gasifier diameter

0.3 m

Freeboard diameter

0.66 m

Gasifier height (w/o freeboard)

2.43-1.37 m

No. of holes in distributor plate

72

Gasifier bed inventory

120 kg

Sand particle average dia.

143 µm

Sand density

2650 kg/m3

Biomass particle average dia.

790 µm

Biomass particle density

570 kg/m3

ACS Paragon Plus Environment

60

Page 61 of 62

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Energy & Fuels

Table 7. Absolute percent error of dry gas composition (vol %). Run

1

2

3

4

5

6

7

8

H2

117.2

47.8

48.9

73.4

62.7

67.9

58.2

56.8

CO

58.1

38.1

31.1

40.7

35.9

34.3

34.3

49.4

CH4

53.9

65.2

74.0

83.1

71.5

81.2

83.6

66.5

CO2

7.0

9.5

8.0

27.2

25.3

10.3

20.0

28.0

ACS Paragon Plus Environment

61

Energy & Fuels

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 62 of 62

Table 8. Operating conditions for UBC DFB gasifier simulation. Biomass fuel flow rate

10 kg/h (as received)

Biomass moisture content

10 wt% (as received)

Steam flow rate

10 kg/h

Tfuel & Pfuel (at inlet)

25°C & 1.1 atm

Gasifier temperature

650-800°C

Tsteam

Same as gasifier

Psteam (at inlet)

1.2 atm

Combustor temperature

900°C

Excess air

20%

Tair & Pair (at inlet)

400°C & 1.1 atm

ACS Paragon Plus Environment

62