ISIJ International
Online ISSN : 1347-5460
Print ISSN : 0915-1559
ISSN-L : 0915-1559
Ironmaking
Numerical Investigation of Hematite Flash Reduction-biomass Steam Gasification Coupling Process
Xingnan WangGuiqin FuWei Li Miaoyong Zhu
Author information
JOURNAL OPEN ACCESS FULL-TEXT HTML

2021 Volume 61 Issue 5 Pages 1450-1458

Details
Abstract

The flash ironmaking process is a novel ironmaking technology; the direct use of biomass as the reductant and fuel in this process can take full advantage of the heat and syngas produced during the biomass gasification. This study establishes a three-dimensional computational fluid dynamics model that incorporates turbulent flow, mass transfer, and heat transfer to describe the complex gas-particle reaction behavior of the hematite flash reduction-biomass steam gasification (FR-BSG) coupling process in an entrained flow reactor to explore its feasibility. The temperature and species distributions in the FR-BSG coupling process are analyzed, and the effects of steam/carbon molar ratio (S/C) and ore/biomass mass ratio (O/B) are investigated. The results show that the reduction degree of hematite particles reaches 76.67% in the residence time of 1.65 s under the conditions of S/C=0.1, O/B=1.0 and T=1673 K. The increase of S/C can enhance the production of H2 but reduce the molar fractions of H2 and CO in biomass syngas, which leads to the decrease of hematite reduction degree. A higher reduction degree of hematite and lower carbon conversion of biomass can be obtained at lower O/B values. These results provide a theoretical basis for the use of biomass as energy in flash ironmaking technology.

1. Introduction

Blast furnace technology is the foremost ironmaking technology, and has made substantial contributions to the development and progress of the iron and steel industry. While this technology has been matured after a long period of development, it is characterized by a long process with high energy consumption, high requirements for resource adaptability, and a large investment in construction, as well as increasingly prominent environmental pollution. In order to overcome these shortcomings, non-blast furnace ironmaking technology has become a research focus. Recently, flash ironmaking has been developed as a novel technology that aims at the direct reduction of fine iron ore (<100 μm) using high-temperature reductive gas in an in-flight process that lasts only a few seconds.1)

At present, the H2–CO syngas used in the ironmaking process is mainly derived from fossil fuels, which are gradually becoming in short supply as their consumption increases. As a sustainable energy source, biomass energy is expected to replace fossil fuels and play an important role in the steel industry, which can satisfy the technical requirements of the sustainable development of the steel industry.2,3,4,5,6,7) Biomass gasification, in which biomass is transformed into reductive gas through a rapid pyrolysis and gasification process, is an important research direction for the development of biomass energy, and is expected to be applied to the ironmaking process in the future.

This study proposes the flash ironmaking technology with biomass as energy, i.e., the iron ore flash reduction-biomass gasification coupling process. The key concept of this process is to obtain a high temperature reductive gas via the biomass gasification process while simultaneously flash reducing iron ore powder to metallic iron in the same reactor. During the biomass gasification process, reductive gases are produced, and a large amount of heat is released. In the flash reduction process, the reduction of iron ore requires heat and reductive gases. The flash reduction-biomass gasification coupling process integrates the exothermic gasification process and the endothermic reduction process without repeated conversions of the material and energy, which has theoretical advantages in material utilization and energy efficiency. At present, biomass gasification agents mainly include air, oxygen-rich and steam. The presence of large amounts of nitrogen in the air tends to take away the heat from biomass syngas. While pure oxygen can promote the gasification process, its production cost is high. High-quality gasification syngas without N2 can be produced by steam gasification.8,9,10,11) As compared with air and oxygen-rich gasification, biomass steam gasification (BSG) can produce more hydrogen, which is both highly efficient and clean.12) Hence, the BSG process can be combined with flash ironmaking technology to form the flash reduction-biomass steam gasification (FR-BSG) coupling process. This is a sustainable ironmaking process, as it attains a high hydrogen utilization ratio during the reduction process and fundamentally reduces the emission of CO2.

The FR-BSG coupling process is complex multiphase flow reduction process; it includes both mass and heat transfer, and both homogeneous and heterogeneous reactions. Compared with complex and time-consuming experiments, numerical simulation can explore the the reaction flow in the process provide effective guidance for reactor design and experimental optimization. Mathematical models employed in various studies have been used to investigate the flash reduction behavior of iron oxides in H213,14) and CO,15) as well as in synthetic gas.16) Yang et al.17,18) analyzed the gasification and reduction characteristics of a combined coal gasification and flash ironmaking process via a three-dimensional (3D) computational fluid dynamics (CFD) model. However, no specific work has been reported on the flash reduction behavior of iron ore particles when coupled with biomass gasification. For exploring the feasibility of the FR-BSG coupling process and studying the effects of operating conditions on the reduction behavior of iron ore in FR-BSG coupling process, it is therefore necessary to analyze this reaction process.

In this study, a 3D CFD model is established to study the complex gas-particle reaction behavior of the FR-BSG coupling process in an entrained flow reactor. The effects of molar ratio of steam and carbon (S/C) in biomass and ore/biomass mass ratio (O/B) are analyzed in details. The coupling process is characterized based on gas species, temperature, biomass carbon conversion, and the hematite reduction degree. This study explores the feasibility of FR-BSG coupling process and provides a theoretical basis for future research and applications.

2. Mathematical Model

2.1. Governing Equations

For the CFD modeling of the FR-BSG coupling process, the steady state governing equations were solved using the Euler-Lagrange method. Gases were regarded as continuous phase with uncompressible ideal gases. The governing equations for the mass, momentum, energy, and species of the gas phase can be represented by the following equations.   

x i ( ρ g u i ) = S p,m (1)
  
x i ( ρ g u i u j ) =- p g x j + x i ( τ ij - ρ g u i u j ' ¯ ) + ρ g g+ S p,mom (2)
  
x i ( ρ g u i h ) = x i ( λ eff T g x i ) + S h + S p,h + S rad (3)
  
x i ( ρ g u i Y j ) = x i ( ρ g D eff Y j x i ) + S Y i + S p, Y i (4)
where, Sp,m, Sp,mom, Sp,h, and SY,p are the source terms that describe the interphase exchange terms for the mass, momentum, energy, and species, respectively. Additionally, Sh and SY are the source terms for homogeneous reactions, and Srad is the radiation source term. In Fluent software, only the P-1 and discrete ordinates (DO) radiation models allow for particle radiation. The DO radiation model can be adopted for a wide range of optical thicknesses, while the P-1 radiation model is applied when the optical thickness is larger than 1. Thus, the DO model was applied in this model to calculate the radiative heat-transfer involving both gas and particles. Moreover, the realizable k-ε model was used to describe the turbulent flow in the reactor. The transport equations for k and ε can be calculated by:   
x i ( ρ g k u i ) = x i [ ( μ+ μ t σ k ) k x i ]+ G k + G b - ρ g ε (5)
  
x i ( ρ g ε u i ) = x i [ ( μ+ μ t σ ε ) ε x i ]+ ρ g C 1 Sε       - ρ g C 2 ε 2 k+ vε + C 1ε ε k C 3ε G b (6)

The biomass and hematite particles were spherical, and were treated as the discrete phase because of the low volume fraction of particles in the reactor. The Lagrange method was applied to the tracking of particles. Moreover, no fragmentation, attrition, or agglomeration of solids took place. The effect of turbulence on the particles was realized via the stochastic tracking model. The governing equations of mass, momentum, and energy for the biomass and hematite particles are respectively given by Eqs. (7), (8), (9) and (10), (11), (12).   

d m p,biomass dt = ( d m p,biomass dt ) drying + ( d m p,biomass dt ) pyrolysis      + ( d m p,biomass dt ) reactions (7)
  
d u p,biomass dt = F D ( u g - u p,biomass ) + ( ρ p,biomass - ρ g ) g ρ p,biomass (8)
  
m p,biomass c p,biomass d T p,biomass dt =h A p,biomass ( T g - T p,biomass ) + A p,biomass ε P σ( θ R 4 - T p,biomass 4 ) + h fg ( d m p,biomass dt ) drying + f h ( d m p,biomass dt ) pyrolysis + f h ( d m p,biomass dt ) reaction H reaction (9)
  
d m p,ore dt = ( d m p,ore dt ) H 2 + ( d m p,ore dt ) CO (10)
  
d u p,ore dt = F D ( u g - u p,ore ) + ( ρ p,ore - ρ g ) g ρ p,ore (11)
  
m p,ore c p,ore d T p,ore dt =h A p,ore ( T g - T p,ore ) + A p,ore ε P σ( θ R 4 - T p,ore 4 ) + f h ( d m p,ore dt ) H 2 H H 2 + f h ( d m p,ore dt ) CO H CO (12)
The drag force FD in Eq. (8) can be expressed as:   
F D = 18 μ g ρ p d p 2 C D R e p 24 (13)
where, the particle Reynolds number Rep and the drag coefficient CD for the spherical particles are calculated as:   
R e p = ρ g d p | u g - u p | μ g (14)
  
C D = a 1 + a 2 R e p + a 3 R e p 2 (15)
where a1, a2, and a3 are constants given by Morsi and Alexander.19) The particle temperature was calculated considering heat transfer, including by convection, radiation, and chemical reactions, given by Eqs. (9) and (12). The radiation temperature θR can be calculated by the following equation:   
θ R = ( G 4σ ) 1/4 (16)
The equations of the reaction rate in the particle governing equations are detailed in the following section.

2.2. Chemical Reaction Models

The chemical reaction models included the drying, pyrolysis, and char reactions of biomass particles, the reduction reactions of hematite particles, and gas-phase reactions. The reactions relevant to the sulfur and nitrogen elements in the biomass particles were ignored.

2.2.1. Drying of Biomass

When the biomass temperature reached the water vaporization temperature, the moisture evaporated from the biomass particle surface at the rate of Eq. (17), and when the biomass temperature reached the boiling point, the moisture evaporated at the rate of Eq. (18).   

( d m p,biomass dt ) drying =π d p,biomass 2 M w k g ( C s - C ) (17)
  
( d m p,biomass dt ) drying = π d p,biomass k c c p,p ( 2+0.46R e p 0.5 ) ln( 1+ c p ( T g - T p,biomass ) h fg ) (18)

2.2.2. Pyrolysis of Biomass

The biomass pyrolysis process can be expressed by a one-stage global single reaction as follows:   

BiomassChar+Volatiles Volatiles x 1 CO+ x 2 C O 2 + x 3 H 2 + x 4 C H 4 +tar (19)
Here, x1, x2, x3, and x4 represent the respective proportionality factors of CO, CO2, H2, and CH4 in the volatiles, which varied with the biomass composition. In this study, these values were determined by fitting both experimental and simulation data under different conditions. The pyrolysis rate was solved by the single kinetic rate model, which can be determined by:   
( d m p,biomass dt ) pyrolysis =- k r,pyrolysis [ m p -( 1- f v,0 ) ( 1- f w,0 ) m p,0 ] (20)
kr,pyrolysis is the pyrolysis reaction rate constant, which is calculated by the Arrhenius equation.   
k r,pyrolysis =Aexp( - E R T p ) (21)
where A = 5.0×106 s−1, and E = 1.2×108 J/kmol.20)

2.2.3. Char-gas Reactions of Biomass

After the pyrolysis process, only char and ash remained in the biomass particles. The ash was carried along with particles without taking part in any reactions. In the BSG process, char reacted with CO2 and H2O, which were respectively converted to CO and H2.

The char consumption process is affected by the diffusion process and reaction kinetics. The char consumption rate, which includes the effects of both the chemical kinetic effect and diffusion effect, can be expressed as Eqs. (22), (23), (24).21)   

d m C-i dt =- A p p i,g r diff,i r kin,i r diff,i + r kin,i (22)
  
r diff,i = C i [ ( T p + T g ) /2 ] 0.75 d p (23)
  
r kin,i = A i exp( - E i R T p ) (24)
where mC−i is the mass of the char remaining in the biomass particle when char reacts with gasifying species i (i = CO2 or H2O), pi,g is the partial pressure of the gasifying species. rdiff,i and rkin,i are the diffusion rate and the kinetic rate, respectively. Ci is the mass diffusion rate constant, 5.0×10−12 (s/K0.75). Ai and Ei are the parameters of the Arrhenius kinetic rates, as reported in Table 1.

Table 1. Reaction rates of char-gas reactions.22,23)
ReactionsAi·(s/m)Ei (J/kmol)
C+CO2→2CO3.0×10−12.0×108
C+H2O→CO+H22.0×10−31.96×108

2.2.4. Reduction Reactions of Hematite

In this study, hematite particles were assumed to be pure Fe2O3 without moisture or gangue in this study. The reduction process of hematite goes through a stepwise procedure (Fe2O3→Fe3O4→FeO→Fe). But measuring the reduction kinetics of each step reaction in a few seconds is extremely difficult. Thus, only the overall reduction process (Fe2O3→Fe) was considered in this study. The overall reaction kinetic constants of Fe2O3→Fe in H2 and CO as reported by Fan et al.14) were used in this model.   

Fe 2 O 3 + 3H 2 2Fe+3 H 2 O (25)
  
( d m p,ore dt ) H 2 =- w O,i m p,i 8.47×1 0 7 × e ( - 218   000 RT ) ( p H 2 - p H 2 O K H 2 ) ( 1-X ) (26)
  
Fe 2 O 3 +3CO2Fe+ 3CO 2 (27)
  
( d m p,ore dt ) CO =- w O,i m p,i 5.18×1 0 7 × e ( - 241   000 RT ) ( p CO - p C O 2 K CO ) ( 1-X ) (28)
  
X= m p,i - m p m p,i w O,i ×100% (29)

2.2.5. Gas Phase Reactions

The finite-rate/eddy-dissipation model was applied to solve the rates of the homogeneous reactions. The rates of the homogeneous reactions were controlled by a lower Arrhenius reaction rate and eddy dissipation reaction rate, as follows:   

R i,gas =min( R i,chem , R i,tur ) (30)
  
R i,tur = min[ ν i,r M w,i Aρ ε k min R ( Y R v R,r M w,R ) , ν i,r M w,i ABρ ε k ( Σ p Y R j N v j,r M w,j ) ] (31)
  
R i,chem = M w,i r=1 N R Γ( ν i,r - ν i,r ) ( R f,r j=1 N r [ C j,r ] η j,r - R b,r j=1 N r [ C j,r ] η j,r ) (32)
  
R b,r = R f,r K r (33)
  
K r =exp( - Δ G r 0 RT ) ( p atm RT ) i=1 N ( ν i,r - ν i,r ) (34)
The Arrhenius reaction rates of the homogeneous reactions considered in this model are presented in Table 2.

Table 2. Reaction rates of gas phase reactions.24,25,26)
ReactionsReaction rate (kmol m3 s−1)
CO+H2O↔CO2+H2 R f =7.4× 10 8 exp( - 2.883× 10 5 8.314T ) [ CO ] 0.5 [ H 2 O ]
R b =1.09× 10 7 exp( - 2.22× 10 5 8.314T ) [ CO 2 ] [ H 2 ] 0.5
K=2.89× 10 -2 exp( 3.21× 10 4 8.314T )
CH4+H2O↔CO+3H2 R f =7.0× 10 6 exp( - 1.26× 10 5 8.314T ) [ CH 4 ][ H 2 O ]
R b =( R f /K ) [ CO ] [ H 2 ] 3 [ C H 4 ][ H 2 O ]
K=8.99× 10 12 exp( - 2.2× 10 5 8.314T )

2.3. Solving Method

The simulation was solved by ANSYS FLUENT 17.0 software together with user defined functions to simulate the heterogeneous reactions. The SIMPLE algorithm was utilized for pressure-velocity coupling. The second-order scheme was applied for pressure discretization. The second-order upwind scheme was chosen to evaluate the momentum, energy, and species, and the first-order upwind scheme was adopted for other terms.

3. Model Setup

The simulations were performed on a laboratory-scale entrained-flow reactor with a length of 200 cm and a diameter of 8 cm. The diameter of particle inlet was 2.5 cm. More details about this reactor can be found in a previous study.27) This study explored the reaction heat transfer behavior of particles-fluids in the reactor. Figure 1 depicts the computational geometry and mesh of this reactor. To obtain accurate solutions, the grid in the center zone of reactor was locally refined. The grid of the reactor consisted of 148680 cells, and was therefore dense enough to obtain grid-independent solutions, as reported in Section 4.1. The mass flow inlet boundary was defined at the gas inlet and particle inlet, and the reactor outlet was regarded as the pressure outlet. A non-slip condition was applied to the reactor wall for the gas flow inside the reactor. The standard wall function was adopted for near-wall treatment.

Fig. 1.

(a) Three-dimensional computational grid and (b) the inlet top view of the entrained-flow reactor.

Biomass and hematite particles at room temperature were carried by a 10 NL/min N2 stream to enter the reactor from the center particle inlet. The steam preheated at 1673 K was blown into the reactor from the outer ring of the gas inlet. The steam mass flow was set as 1.0 g/min. The mass rates of biomass and hematite were varied by the S/C and O/B values. The initial velocity of particles was assumed to be equal to the velocity of N2 stream. The operation pressure was fixed at 1.0 atm. The wall temperature was set as 1673 K.

The proximate and elemental analysis of the wood used in this study is summarized in Table 3. The hematite ore was assumed to be pure Fe2O3. The diameters of hematite and biomass particles were set as 74 μm and 100 μm, respectively.

Table 3. The proximate and elemental analysis of the wood.27)
Proximate analysis (wt%, ar)Elemental analysis (wt%, daf)
MoistureVolatileFixed carbonAshCHOOther
9.0476.713.650.6149.96.443.60.1

4. Results and Discussion

4.1. Model Verification

In consideration of the grid independence analysis, numerical procedures were respectively performed for four different mesh systems of 104304 cells, 148680 cells, 209940 cells, and 280200 cells. An N2 stream at 293 K with a uniform inlet velocity was used in the test. Figure 2 presents the gas temperature profile in the center of the reactor along the reactor height and the gas velocity profile at the bottom reactor exit along the reactor diameter. It can be seen that the results of the four meshes are almost the same. Considering the requirement of mesh independence and a shorter computational time, the mesh system of 148680 cells was used in this model.

Fig. 2.

A comparison of simulation results based on different meshes: (a) gas temperature profiles along the reactor center line, (b) gas velocity profiles along the reactor outlet diameter. (Online version in color.)

The accuracy of the CFD model was verified by a hematite flash reduction experiment and biomass gasification experiment, respectively. In order to validate the accuracy of the model of the flash reduction process, the predicted reduction degree of hematite at the reactor exit was compared with the experimental results reported by Chen et al.28,29) In our previous work,30) the maximum relative error between the experimental and simulated results was 9.30%, which demonstrates that this CFD model could reasonably predict the flash reduction process.

This model about the biomass gasification process was validated by the experimental data reported by Qin et al.27) in terms of the gas production at the reactor exit. Biomass particles were gasified in a steam atmosphere at 1673 K. The molar ratio of steam and carbon in biomass (S/C) varied from 0 to 1.0, and the detailed conditions are displayed in Table 4. The experimental and predicted gas productions at the reactor exit are displayed in Fig. 3. Although the model was established according to experimental conditions, some extremely complicated conditions in the experiment were simplified, such as assuming that the biomass was spherical. Therefore, there was a partial gap between the simulated and experimental data. However, in general, the experimental data were consistent with the simulated data, so it is believed that the model can reasonably measure the biomass gasification behavior. Thus, the proposed CFD model can be used as an effective tool to study the FR-BSG coupling process.

Table 4. Detailed conditions in biomass steam gasification for validation.
S/C (–)T (K)dp (μm)Biomass feeding rate (g/min)Feeder N2 flow rate (NL/min)Steam flow rate (g/min)
0167331012.810.00
0.5167331012.810.04.3
1.0167331012.810.08.6
Fig. 3.

Comparison of species production between simulation and experiment for steam gasification. (Online version in color.)

4.2. Flash Reduction-biomass Steam Gasification Coupling Process

Figure 4 shows the FR-BSG coupling process in the reactor in the steady state at 1673 K under the conditions of S/C=0.1 and O/B=1.0, and includes the gas temperature and species distribution. When moving in the reactor, biomass particles are heated by high-temperature gas, and then undergo the pyrolysis process together with the release of a large amount of volatiles. This leads to increases in the molar fractions of H2, CO, CO2, and CH4 molar fraction at the upper zone of reactor. After biomass pyrolysis, the char in the biomass reacts with steam to produce H2 and CO. Accordingly, the molar fractions of H2 and CO increase, and the molar fraction of H2O in the reactor decreases. Due to the reactions of char with CO2 (C+CO2=2CO) and the methane steam reforming reaction (CH4+H2O→CO+3H2), the molar fractions of CO2, CH4 and H2O gradually decrease. Figure 5 compares the profiles of species molar fraction along the reactor centerline in the BSG process and FR-BSG coupling process. It is evident that the reduction process of hematite decreases the molar fractions of H2 and CO, and the molar fractions of H2O and CO2 accordingly increase.

Fig. 4.

FR-BSG coupling process in the vertical center plane of reactor at 1673 K with S/C=0.1 and O/B=1.0: (a) temperature, (b-f) species molar fraction. (Online version in color.)

Fig. 5.

Comparison the profiles of main species molar fraction along the reactor centerline in BSG process and FR-BSG coupling process. (Online version in color.)

4.3. Effect of Steam/carbon Molar Ratio

As reported by Qin et al.,27) the increase of S/C can increase the production of H2 at the reactor exit during the BSG process. Syngas with a high content of hydrogen causes the reduction process of hematite to be more complete. In this section, the effect of S/C on the FR-BSG coupling process is investigated within the range of 0.1-1.0 at the fixed steam mass flow rate of 1.0 g/min and O/B value of 1.0.

Because S/C can influence the FR-BSG process by controlling the biomass gasification behavior, the effect of S/C on the BSG process was first explored, as shown in Fig. 6. Figure 6(a) presents the gas produced at the reactor exit as a function of S/C, without H2O content. With the increase in S/C, the productions of H2 and CO2 gradually increased, the CH4 production reduced, while the CO production underwent through a process of first increasing and then reducing. This is because the char in biomass particles reacted with the introduced steam to form H2 and CO, and the augmented steam contributed to the forward water-gas shift reaction (CO+H2O→CO2+H2) and the methane steam reforming reaction (CH4+H2O→CO+3H2). Figure 6(b) depicts the effect of S/C on gas molar fraction at the reactor exit. In terms of the species molar fractions at the reactor exit, the increase of S/C was found to enhance the H2O molar fraction but decreases the molar fractions of H2 and CO in the biomass syngas. This implies that the reduction capacity of biomass syngas may be impaired with an increase in S/C, despite the increase of H2 production.

Fig. 6.

Effect of S/C on (a) species production and (b) species molar fraction at the reactor exit in the BSG process. (Online version in color.)

The carbon conversion efficiency (CC) is also an important parameter by which to characterize biomass gasification behavior, and it can be obtained by the following equation.   

CC= m out,CO ( 12 28 ) + m out,C O 2 ( 12 44 ) + m out,C H 4 ( 12 16 ) m in,biomass Y C ×100% (35)

Note that the particle velocity can be obtained by the governing equations of momentum for the biomass and hematite particles. The residence time of particle in Figs. 7, 8, 9 can be obtained by the particle velocity and reactor length. Note that each particle stream has a unique trajectory, so that the residence time are obtained by averaging all the particle streams.

Fig. 7.

Effect of S/C on carbon conversion and residence time of biomass in the BSG process.

Fig. 8.

Effect of S/C on hematite flash reduction behavior in the FR-BSG process at O/B=1.0.

Fig. 9.

Effect of O/B on hematite flash reduction behavior in the FR-BSG process at S/C=0.5.

Figure 7 shows the predicted biomass average residence time in the reactor and biomass CC at the reactor exit during the BSG process with different values of S/C. The results show that the biomass CC reached above 85.06% in the residence time of 2.57 s under the condition of S/C=0.1. With the increase of S/C, the biomass residence time increased significantly, whereas the variation of the CC of biomass was negligible. In summation, the increasing S/C can enhance the production of H2 but reduces the molar fraction of H2 in biomass syngas, whereas it has a very small impact on the CC.

The effect of S/C on the reduction of hematite in the FR-BSG coupling process was explored by the residence time in the reactor and the reduction degree at the reactor exit. Figure 8 reveals that the reduction degree of hematite particles reached 76.67% with the residence time of 1.65 s at S/C=0.1. Increasing S/C led to a decrease of the hematite reduction degree, despite of an increase in the particle residence time. This can be explained by the fact that increasing S/C decreases the molar fractions of H2 and CO in biomass syngas, thereby weakening its reduction capacity. Thus, S/C should be controlled within limits to prevent the excessive steam from affecting the hematite reduction in the FR-BSG coupling process.

Table 5 compares the biomass gasification behavior in the BSG and FR-BSG processes. The CC in the FR-BSG process was found to be higher than that in the BSG process, which means that the molar fractions of H2, CO and CH4 at the reduction exit may be higher in the FR-BSG process. However, higher H2, CO and CH4 molar fractions were found in the BSG process. This is because the reduction process of hematite consumes H2 and CO, which accordingly decreases the molar fractions of H2O and CO2. The increases of the contents of H2O and CO2 further promote the biomass gasification process, which leads to an increase of CC in the FR-BSG process.

Table 5. Comparison of biomass gasification behavior in the BSG and FR-BSG process at S/C=0.1 and O/B=1.0.
ProcessBiomassSpecies molar fraction (%)
t (s)CC (%)H2H2OCOCO2CH4
BSG2.5785.0631.043.9131.500.502.88
FR-BSG2.4090.0923.0613.3331.832.091.09

4.4. Effect of Ore/biomass Mass Ratio

The effect of O/B on the hematite reduction behavior in the FR-BSG coupling process is presented in Fig. 9. The O/B value was varied from 0.1 to 2.0 by changing the ore mass rate. Under a certain composition of biomass syngas obtained at S/C=0.5, increasing the ore mass was found to reduce the reduction degree. As was expected, a higher reduction degree of hematite can be obtained at a lower O/B value. With the enhancement of O/B from 0.1 to 1.0, the reduction degree decreased from 83.13% to 54.12%. Table 6 compares the biomass gasification behavior in the BSG and FR-BSG processes at different O/B values. The increase of O/B was found to decrease the reduction degree of hematite, but enhanced the CC of biomass. This can be explained by the fact that a higher ore mass flow rate consumes more H2 and CO, and accordingly produces more H2O and CO2, which are further used in the gasification process.

Table 6. Comparison of biomass gasification behavior in the BSG and FR-BSG process at different O/B with S/C=0.5.
ProcessO/BCC (%)Species molar fraction (%)
H2H2OCOCO2CH4
BSG88.1316.776.3914.010.700.20
FR-BSG0.590.0814.678.4913.961.070.15
1.091.2413.669.4913.731.480.13
1.591.9413.209.9413.481.840.11

5. Conclusions

In this study, a three-dimensional computational fluid dynamics model that incorporates mass and heat transfer, as well as chemical reactions, was established to explore the feasibility of hematite flash reduction-biomass steam gasification (FR-BSG) coupling process. After validation, the complex gas-particle reaction behavior in the millisecond process of hematite flash reduction and biomass gasification in an entrained flow reactor were described by demonstrating the gas temperature and species distributions. The effects of steam/carbon molar ratio (S/C) and ore/biomass mass ratio (O/B) on the FR-BSG coupling process were then analyzed. The results demonstrate that the molar fractions of H2 and CO in biomass syngas decreased with the increase of S/C, thereby weakening the reduction capacity of biomass syngas. Hence, at a fixed steam mass flow rate, the increase of S/C decreases the reduction degree. A higher O/B value leads to a lower reduction degree of hematite but a higher carbon conversion of biomass. These results provide a theoretical basis for the application of biomass as energy in the flash ironmaking process.

Acknowledgment

This work is financially supported by National Natural Science Foundation of China (51904066), Liaoning Revitalization Talents Program (XLYC1802032), Fundamental Research Funds for the Central Universities (N182503032), Postdoctoral Foundation of Northeastern University (20190201) and Postdoctoral International Exchange Program (Dispatch Project, 20190075).

Nomenclature

A = Pre-exponential factor

Ap = Particle surface area, (m2)

CC = Carbon conversion, (%)

CD = Drag coefficient, (kg/(m3·s))

cp,g, cp,p = Heat capacity of gas and particle, (J/(kg·K))

Deff = Effective diffusivity coefficient, (m2/s)

dp = Particle diameter, (m)

E = Activation energy, (kJ/mol)

FD = Drag force per unit particle mass

fh = Particle reaction heat absorption ratio

g = Gravitational acceleration, (m2/s)

G = Incident radiation

Gk = Generation term for turbulence kinetic energy

h = Convective heat transfer coefficient, (W/(m2·K))

HH2, HCO = Reaction heat of Fe2O3 reduced by H2 and CO, (J/kg)

k = Turbulence kinetic energy, (m2/s2)

kr,pyrolysis = Pyrolysis reaction rate constant, (1/s)

KH2, KCO = Reaction equilibrium constant of FeO reduced by H2 and CO

KWGSR = Equilibrium constant of WGSR

mp = Instantaneous particle mass, (kg)

mp,i = Initial particle mass, (kg)

mout,i = Species production mass at the reactor exit

min,biomass = Biomass mass at the reactor inlet

Nu = Nusselt number

p = Gas pressure, (Pa)

PH2, PH2O, PCO, PCO2 = Partial pressure of H2, H2O, CO and CO2, (atm)

Pr = Prandtl number

rdiff,i = Diffusion rate

rkin,i = Kinetic rate

R = Universal gas constant, (J/(kmol·K))

Rf, Rb = Reaction rates of forward and backward

Rep = Particle Reynolds number

Sp,m, Sp.mon, Sp,h, SpYi = Interphase exchange terms for mass, momentum, enthalpy and species

Sh, SYi = Source terms due to the homogeneous gas-phase reactions

Srad = Radiation source term

Sct = Turbulent Schmidt number

t = Particle residence time, (s)

Tg, Tp = Temperature of gas and particle, (K)

θR = Radiation temperature, (K)

ug , up = Velocity of gas and particle, (m/s)

wo,i = Initial oxygen mass fraction in particle, (%)

X = Reduction degree, (%)

Yi = Mass fraction of chemical species

Yc = Mass fraction of carbon in biomass

Z = Distance from the particle inlet, (m)

τij = Viscous stress tensor

ε = Dissipation rate of turbulence kinetic energy, (m2/s3)

εp = Particle emissivity

λeff = Effective thermal conductivity, (W/(m·K))

μg, μt = Dynamic viscosity and turbulent viscosity, (kg/(m·s))

ρg, ρp = Density of gas and particle, (kg/m3)

σ = Stefan-Boltzmann constant, (W/(m2·K4))

References
 
© 2021 The Iron and Steel Institute of Japan.

This is an open access article under the terms of the Creative Commons Attribution-NonCommercial-NoDerivs license.
https://creativecommons.org/licenses/by-nc-nd/4.0/
feedback
Top