Numerical Investigation of Applying High-carbon Metallic Briquette in Blast Furnace Ironmaking

Abstract

Application of high-carbon metallic briquette (HCMB) in the blast furnace (BF) ironmaking for coke saving was previously proposed. This paper is focused on clarifying the in-furnace phenomena and demonstrating the advantages of applying the HCMB in BF ironmaking. A mathematical model has been formulated based on the gas-solid counterflow and its validity was confirmed by the comparison of the simulation results with the averaged industrial data from a BF with a productivity of 6250 tHM/day. Afterward, BF operations under two HCMB mixing ratios (5% and 10%) were simulated. Simulation results indicate that charging HCMB in BF can suppress the coke gasification and improve the ore reduction above the CZ. Coke of approximately 12 kg could be saved for producing one-ton hot metal from the ore under an HCMB mixing ratio of 5%, and approximately 17 kg under an HCMB mixing ratio of 10%. Simulation results still indicate that the reasonable mixing ratio of HCMB is less than 5%, under which, significant changes of the BF operation conditions for HCMB charging are not required.

1. Introduction

The blast furnace (BF) ironmaking currently is the unchallenged method of producing hot metal with many features as high productivity and high thermal efficiency. For the last 35 years, about 65% of the world’s steel has been produced from iron ore and the blast furnace has contributed the major volume of this iron.^1,2) Coke is an important raw material in the BF ironmaking and causes a majority of production costs of the hot metal. Because of the scarcity of the caking coal resources, the environmental pollution in the coke production, and the desire to reduce hot metal costs, tremendous efforts have been made on the low-coke operation technology in the developments of BF ironmaking sectors.^3,4)

Many of the low-coke BF operation technologies concentrate on substituting the coke by non-caking coals or other reducing agents (e.g. oil, gas, and biomass). In the previous study, the authors of the current study proposed to apply high-carbon metallic briquette (HCMB) for further reduction of the coke consumption in addition to the PCI (pulverized coal injection) operation in BF ironmaking.⁵⁾ Lab-scale tests showed that the HCMB are with satisfying strength and high CO₂ reactivity under the simulated BF in-furnace environment. However, for its application in industrial scale, the behavior of the HCMB in BF and its influence on the BF performance need thorough investigations. Nowadays, numerical simulation is becoming an effective method to evaluate various technologies proposed for the BF ironmaking. Such simulations offer a powerful tool that can be used to (1) gain fundamental insights, (2) investigate the impact of key parameters, and (3) develop strategies for process optimization.⁶⁾ To understand the important aspects of the HCMB charging and to clarify the in-furnace phenomena in the BF, a detailed mathematical model based on the fundamental phenomena is required.

In this paper, a mathematical model has been established to simulate the BF operation with HCMB charging. After that, the influence of charging HCMB on the coke-based BF process was investigated and the potential of coke saving was evaluated.

2. Description of HCMB Preparation and Its Properties

Details of the HCMB preparation method and its properties have been given in ref. [5]. The following is the outline. The HCMB is prepared by roasting the cold-boned iron-oxide-coal briquette. The raw materials of the cold-boned briquette are pure hematite fines (size: approximately 2 μm) and pulverized non-caking coal fines (size: approximately 60 μm), and the mass ratio of hematite fines to coal fines in the briquette is 2.0. During the slow roasting under the inert atmosphere, the hematite fines are reduced to metallic iron forming an iron network in the briquette and bonding the carbon particles together. The HCMB has a carbon content of 25 wt% and a metallic iron content of approximately 75 wt% while the contents of iron oxides (e.g. Fe₃O₄ and FeO) are negligible. Size of the HCMB is similar to that of the sinter ore particles. The HCMB has a cold strength (CS) of 1300 N/briquette, and a crushing strength after reaction (CSR) of 2500 N/briquette, showing satisfying strength characteristics for BF application. Under the BF in-furnace environment, threshold temperature and activation energy of the HCMB carbon gasification by CO₂ are approximately 973 K, and 166 kJ/mol respectively, indicating that the HCMB has a far higher CO₂ reactivity than the conventional metallurgical coke.

3. Model Development

3.1. Computational Mesh

For ensuring a high convergence of the developed model, the multiphase flow in BF is simplified to be a gas-solid counterflow. Other phases as PC and liquid phase are treated in a way different from other researchers and the treatment is detailed in the following.

The geometry of the model is based on a blast furnace of an iron and steel company in China. The blast furnace is with an inner volume of 2500 m³, producing 6250 ton/day of hot metal. 30 tuyeres are installed on its lower sidewall. The BF is illustrated in Fig. 1(a). The mathematical model is two-dimensional and axis-symmetric. The computational domain covers the stack part and the bosh part, from burden surface to hearth liquid surface. A two-dimensional structure grid is developed and is shown in Fig. 1(b). The grid includes 810 cells. Calculation of the cell volumes is based on 12 degrees in the circumferential direction. The deadman zone and the raceway zone are predetermined in the grid according to the research of Austin^7,8,9) and Gupta.¹⁰⁾

Fig. 1.

Size of the modeled blast furnace (a) and structure grid (b).

3.2. Chemical Reactions

Reactions included in the model are listed in Table 1. For each reaction, the dependence of reaction heat with temperature is regressed using the thermochemical data given by Ye and Hu.¹¹⁾

Table 1. Chemical reactions involved in the model.

No.	Reaction	Reaction heat (106 J/kmol)	Note
1	3Fe2O3(s)+CO(g)=2Fe3O4(s)+CO2(g)	0.00008T2 – 0.07T – 36.62 (273 ≤ T ≤ 853) 0.05T2 + 11.23 (853 < T ≤ 983) 0.000005T2 + 0.004T – 38.77 (983 < T ≤1473)	Indirect reduction
2	Fe3O4(s)+CO(g)= 3FeO(s)+CO2(g)	−0.00007T2 + 0.05T + 23.90 (273 ≤ T ≤ 853) 0.00003T2 + 0.06T + 44.89 (853 < T ≤ 1473)
3	FeO(s)+CO(g)= Fe(s)+CO2(g)	0.00005T2 – 0.03T – 8.57 (273 ≤ T ≤ 1193) −0.004T – 10.65 (1193 < T ≤ 1673)
4	C(s)+1/2O2(g)=CO(g)	−0.0045T – 106.68	Combustion
5	C(s)+CO2(g)=2CO(g)	−0.0013T + 185.15	Solution loss reaction
6	Fe(s)=Fe(l)	14.22	Melting
7	FeO(s)=FeO(l)	18.29
8	Gangue(s)=Slag(l)	15.24 (MGangue=60)
9	FeO(l)+C(s)= Fe(l)+CO	147.0	Direct reduction
10	C(s,HCMB)+CO2(g)=2CO(g)	−0.0013T + 185.15	Reactions of HCMB
11	Fe(s,HCMB)=Fe(l)	14.22

Rates of reactions (1–3) are described using Eq. (1), which is the three-interface shrinking core model. The values of the parameters in Eq. (1) are given by Natsui.¹²⁾   

$$R_{\text{i}}=1.0\times {10}^{-3}{\alpha }_{\text{ore}}{A}_{\text{ore}}{P}_{\text{g}}\text{/(}8.314T_{\text{g}})\sum _{\text{k}=1}^3{a}_{\text{k}}$$(1)

Rates of reactions (4,5) are described using Eq. (2).¹³⁾   

$$\begin{array}{l}{R}_{\text{i}}=1.0\\ \hspace{1em}\times {10}^{-3}{\alpha }_{\text{coke}}{A}_{\text{coke}}{P}_{\text{re}}\text{/(}8.314T_{\text{g}}\text{)}/(\text{1/}{k}_{\text{f}}+\text{6/(}{d}_{\text{coke}}{k}_{\text{P}}{E}_{\text{f}}{\rho }_{\text{coke}}\text{)})\end{array}$$(2)

where, E_f = 3(φcoth(φ)−1)/φ², φ = (d_coke/2)(ρ_cokek_P/D)^1/2, D = 6.7×10⁻¹⁰(T_s)^1.78, and $k_{\text{f}}=2\times {Re}_{\text{gs}}^{-0.336}\left|\stackrel{\rightharpoonup }{U_g}\right|/{\epsilon }_{\text{s}}$. For reaction (4), P_re=P_O2, k_P = 6.52×10⁵ exp(−22000/T_s)×(T_s)^0.5. For reaction (5), P_re=P_CO2, k_P = 4.0×10¹⁰ exp(−40400/T_s).

Reactions (6–8, and 11) occur in the cohesive zone (CZ). The CZ is defined as the region where the solid temperature is from 1473 to 1673 K. Melting rates of these reactions are described using enthalpy model.^14,15) For each cell in the CZ, their reaction rates are expressed as Eqs. (3), (4).   

$$\begin{array}{l}{R}_{\text{i}}=\\ \sum _{i\text{nflow}}^{\text{k}=\mathrm{1..4}}((1/(1\text{-}{\eta }_{\text{k}})Max(0,\eta -{\eta }_k){\rho }_{\text{s,k}}{y}_{\text{j,k}}\left|{\stackrel{\rightharpoonup }{V}}_{\text{s}}\cdot \stackrel{\rightharpoonup }{n}\right|A_{\text{k}}/(M_{\text{j}}{V}_{\text{Cell}}))\end{array}$$(3)

  

$$\eta =(T_{\text{S}}-1\mathrm{\hspace{0.17em} }473)/200$$(4)

where, η is the fraction of liquid phase, k is the index of the cells surrounding the calculated cell; j (j=i) represents Fe, FeO and Gangue, and HCMB Fe for reactions 6, 7, 8, and 11, respectively; A_k is the face area between cell k and the calculated cell, $\stackrel{\rightharpoonup }{n}$ is the normal unit vector on the face of the calculated cell.

3.3. Treatment of PC and Liquid Phases

The PC particles are gasified in the raceway zone reaching a burnout rate of more than 90% within 20 ms.¹⁶⁾ Therefore, the combustion products of the blast and the PC through reaction (4) form the inlet condition for the gas phase in the model.

The liquid phase includes the molten iron and the molten slag. Droplets of the molten iron and the molten slag are generated in the CZ with an initial temperature equivalent to the local solid temperature. After generation, they flow down through the dripping zone (DZ), getting heated by the gas phase and the coke bed, and reaching the final tapping temperature in the hearth. On their flowing path, these droplets undergo coalescing, splitting, or flying a short distance with the strong bosh gas,¹⁷⁾ so it is difficult to give precise mathematical descriptions upon the gas-liquid and solid-liquid heat exchanges. In the BF bosh, heat is mainly generated by the combustion of oxygen with coke and PC in the raceway and the gas phase has the highest temperature. As the gas flows upward, the heat is transferred from the gas to the coke bed, and to the liquid droplets; simultaneously, the heat is also transferred from the coke bed to the liquid droplets. This analysis shows that the required heat for the liquid phase could be simplified as an energy source of the gas phase. The liquid temperature in the hearth is considered to be 1753 K, therefore, the overall heat loss rate (Q_l) from the gas-solid system to the liquid phase is Eq. (5), Assuming that the heat loss rate is uniformly distributed in the DZ, an enthalpy source (Eq. (6)) is added to the energy equation of the gas phase in the DZ.   

$$\begin{array}{l}{Q}_{\text{l}}=\\ \sum _{\text{i}}^{C\mathrm{Z}}\text{(}{M}_{\text{Fe}}(R_{\text{6}}+R_{\text{11}}\text{)}+M_{\text{FeO}}{R}_{\text{7}}+M_{\text{Gangue}}{R}_8)C{p}_{\text{l}}{V}_{\text{cell}}(1\mathrm{\hspace{0.17em} }753-T_{\text{S}})\end{array}$$(5)

  

$$E_{\text{gl}}=Q_{\text{l}}/\sum _i^{\text{DZ}}{V}_{\text{cell}}$$(6)

Molten FeO in the slag droplets is reduced fast in the DZ through reaction (9). In the view of the mass balance of molten FeO in CZ and DZ, the rate of Reaction (9) is described using Eq. (7), in which, reaction (9) is assumed to uniformly proceed in the DZ.   

$$R_9=\sum _{\text{i}}^{C\mathrm{Z}}\text{(}{R}_7{V}_{\text{cell}}\text{)/}\sum _{\text{i}}^{\text{D}\mathrm{Z}}{V}_{\text{cell}}$$(7)

The above method of treating the behavior of the PC and the liquid phase was demonstrated to be helpful for the model to reach a high convergence.

3.4. Treatment of HCMB

In the model, the chemical composition of the HCMB is assumed to be composed of 75 wt% metallic iron and 25 wt% carbon, and the HCMB is mixed well with the ore burden (sinter ore). The addition level in the ore burden is represented by the HCMB mixing ratio (β, the mass ratio of the HCMB to the ore burden). Properties such as size, bulk density, and porosity of the HCMB are considered to be the same as those of the sinter ore, however, the gasification kinetics of the HCMB carbon differs from that of the coke. The gasification rate expression of the HCMB carbon is obtained from the experimental results of the previous researches.

3.5. Gas-solid Heat Transfer

Gas-solid heat transfer is described using Eqs. (8), (9).^7,18,19)   

$$E_{\text{gs}}=h_{\text{gs}}({\alpha }_{\text{ore}}{A}_{\text{ore}}+{\alpha }_{\text{coke}}{A}_{\text{coke}})(T_{\text{g}}-T_{\text{s}})$$(8)

  

$$h_{\text{gs}}=\gamma {\lambda }_g/d_{\text{s}}(2.0+0.6{Pr}_g^{0.33}{\text{(9}\text{.0}{Re}_{\text{gs}}\text{)}}^{\text{0}\text{.5}})$$(9)

where, γ is the scaling factor, and γ=0.2.⁷⁾

The permeability of the solid phase becomes poor in the CZ owing to the occurring of the liquid phase, so heat exchange between gas and solid here changes to between gas and slab, and the heat transfer coefficient is Eq. (10).^19,20,21)   

$$h_{\text{gs}}={\lambda }_{\text{g}}/d_{\text{s}}(0.2\text{03}{Re}_{\text{gs}}^{0.33}{Pr}_{\text{g}}{}^{0.33}+0.220{\text{Re}}_{\text{gs}}^{0.8}{\text{Pr}}_{\text{g}}{}^{0.4})$$(10)

3.6. Governing Equations

In the BF, the gas moves upward through the slow descending packed bed with typical gas and solids residence time of 2–3 seconds and 4–5 hours, respectively. Therefore, the gas flow is considered as the gas flow through the porous bed. The general governing equation for the gas phase is Eq. (11), in which, the superficial gas velocity is adopted. Terms to represent ϕ , Γ_ϕ and S_ϕ in Eq. (11) are listed in Table 2.   

$$\text{div(}{\rho }_{\text{g}}{\stackrel{\rightharpoonup }{U}}_{\text{g}}\varphi )=\text{div}({\mathrm{\Gamma }}_{\varphi }\text{grad}\varphi )+S_{\varphi }$$(11)

Table 2. Terms in Eq. (11).

	ϕ	Γϕ	Sϕ
Mass	1	0	$M_O\sum _{i=1}^3{R}_i+M_C{R}_4+M_C{R}_5+M_{CO}{R}_9+M_C{R}_{10}$
Momentum	${\stackrel{\rightharpoonup }{U}}_{\text{g}}$	μg	$-\nabla P_g+\stackrel{\rightharpoonup }{F_{gs}}$
Enthalpy	Hg	μg/Prg	$\begin{array}{l}0.5\sum _{i=1}^3{R}_i(-\mathrm{\Delta }{H}_i)+0.5(R_4(-\mathrm{\Delta }{H}_4)+R_5(-\mathrm{\Delta }{H}_5)\\ +R_9(-\mathrm{\Delta }{H}_9)+R_{10}(-\mathrm{\Delta }{H}_{10}))-E_{gs}-E_{gl}\end{array}$
Species	yN2	μg/Scg	0
	yCO	μg/Scg	MCO(−R1−R2−R3+R4+2R5+R9+2R10)
	yCO2	μg/Scg	MCO2(R1+R2+R3)
	yO2	μg/Scg	MO2(-0.5R4)

The BF gas is considered to be an ideal gas, so its properties follow the mixing law of the ideal gas.¹²⁾ The flow resistance from the solid phase (the porous bed) is described using the Ergun’s equation (Eq. (12)).   

$$\stackrel{\rightharpoonup }{F_{\text{gs}}}=\left(C_{\text{1}}+C_{\text{2}}\left|\stackrel{\rightharpoonup }{U_{\text{g}}}\right|\right){\stackrel{\rightharpoonup }{U}}_g$$(12)

where, C₁=−150μ_g(1−ε_s)²/$({\epsilon }_{\text{s}}^3{d}_{\text{s}}^2)$, and C₂=−1.75ρ_g(1−ε_s)/$\text{(}{\epsilon }_{\text{s}}^3{d}_{\text{s}})$.

Porosity and particle size of the solid burden vary in the BF. The method given by Austin^7,8) is used in defining the local particle size and the local porosity of the porous bed.

A non-slip wall condition for the gas velocity and an impermeable condition for the gas species are defined on the BF wall. The heat loss rate of the gas energy on the BF wall is calculated by h_wall(T_g−T_Wall), where, h_wall=5.0 W/(m²·K),²¹⁾ and T_wall=353 K. Gas inlet conditions are determined according to the BF operation conditions. At the gas outlet, a fully-developed gas flow is assumed.

The slow-moving structure of the solid burden is mainly determined by the rate at which the molten iron and slag are produced in the BF and subsequently tapped from the hearth. In a steady BF operation, this is matched by an equivalent rate of charging burden at the BF top. For these reasons, the solid flow in BF could be treated as a viscous flow.^22,23) The viscous model given by Chen²⁴⁾ and the treatment method of the deadman zone given by Austin⁸⁾ are applied for modeling the solid flow. The general governing equation of the solid phase is Eq. (13), in which, the solid bulk density and the solid physical velocity are adopted. Terms to represent ϕ, Γ_ϕ, and S_ϕ in Eq. (13) are listed in Table 3. In Table 3, the mass fraction of the HCMB iron in the solid phase is given as a dependent variable is to facilitate the calculation of the ore reduction fraction, and the mass fraction of the HCMB carbon in the solid phase as a dependent variable is owing to the different gasification kinetics between the HCMB carbon and the coke.   

$$\text{div(}{\rho }_{\text{s}}{\stackrel{\rightharpoonup }{V}}_{\text{S}}\varphi )=\text{div}({\mathrm{\Gamma }}_{\varphi }\text{grad}\varphi )+S_{\varphi }$$(13)

Table 3. Terms in Eq. (13).

/	ϕ	Γϕ	Sϕ
Continuity	1	0	−MO(R1+R2+R3)−MFe(R6+R11)−MFeOR7−MGangueR8−MCR10
Momentum	$\stackrel{\rightharpoonup }{V_s}$	μs,eff	−∇Ps
Enthalpy	HS	λs,eff/Cps	$\begin{array}{l}{E}_{gs}+0.5\sum _{i=1}^3{R}_i(-\mathrm{\Delta }{H}_i)+\sum _{i=6}^8{R}_i(-\mathrm{\Delta }{H}_i)\\ +R_{11}(-\mathrm{\Delta }{H}_{11})+0.5(R_4(-\mathrm{\Delta }{H}_4)\\ +R_5(-\mathrm{\Delta }{H}_5)+R_9(-\mathrm{\Delta }{H}_9)+R_{10}(-\mathrm{\Delta }{H}_{10}))\end{array}$
Species	yC,HCMB	0	MC(−R10)
	yC,COKE	0	0
	yFe2O3	0	MFe2O3(−3R1)
	yFe3O4	0	MFe3O4(2R1−R2)
	yFeO	0	MFeO(3R2−R3−R7)
	yFe	0	MFe(R3−R6)
	yFe,HCMB	0	MFe(−R11)
	ygangue	0	Mgangue(−R8)

The solid bulk density is calculated by Eqs. (14), (15). For ensuring a stable solid flow, the reaction rate of the coke is not included in Eq. (13) (Table 3).

The heat capacity of the solid phase is calculated using Eq. (16). Other effective thermophysical properties of the solid phase could be found in refs. (7,8).   

$$\begin{array}{l}{\rho }_{\text{s}}=((1.0/(1.0+\beta ){\alpha }_{\text{ore}}{\rho }_{\text{ore}}+\beta /(1.0+\beta ){\alpha }_{\text{ore}}{\rho }_{\text{Fe,HCMB}}\\ \hspace{1em}\hspace{1em}+{\alpha }_{\text{coke}}{\rho }_{\text{ash, coke}})(1-\eta )+\beta /(1.0+\beta ){\alpha }_{\text{ore}}{\rho }_{\text{C,HCMB}}\\ \hspace{1em}\hspace{1em}+{\alpha }_{\text{coke}}{\rho }_{\text{C, coke}}\end{array}$$(14)

  

$${\rho }_{\text{ore}}={\rho }_{\text{ore,0}}-{\rho }_{\text{O,0}}{f}_{\text{ore}}$$(15)

where, ρ_ore,0 is the initial bulk density of the ore, ρ_O,0 is the initial bulk density of the removable O in the ore burden.   

$$\begin{array}{c}C{p}_{\text{s}}=1\mathrm{\hspace{0.17em} }\text{2}00.0y_{\text{C,coke}}+460.0y_{\text{Fe}}+\text{720}\text{.0}{y}_{\text{FeO}}\\ +\text{920}\text{.0(1}\text{.0}-y_{\text{Fe}}-y_{\text{FeO}}-y_{\text{C,coke}})\end{array}$$(16)

Local volume fractions of the ore and of the coke are determined using Eqs. (17), (18). It is assumed that, along the flowing path of the solid particles, the coke/ore ratio remains unchanged.   

$$\text{div(}{\stackrel{\rightharpoonup }{V}}_{\text{s}}{\alpha }_{\text{coke}})=0$$(17)

  

$${\alpha }_{\text{coke}}+{\alpha }_{\text{ore}}=1.0$$(18)

A fluid-slip boundary is applied for the solid velocity on the BF wall.²⁵⁾ Heat loss of the solid phase on the BF wall is not considered. The inlet conditions of the solid phase are established according to the BF operation conditions. At the solid outlet, the solid phase reaches a fully-developed flow.

3.7. Solution Strategy

The above two sets of equations (Eqs. (11) and (13)) are solved numerically using the commercial software package PHOENICS²⁶⁾ and an in-house developed code. All equations are solved simultaneously. The calculation procedure is similar to the flowchart given in ref. (27). In addition to the examination of the convergence of gas and solid flow fields, the mass balance of the removable element O and of the element Fe are examined and the convergence criteria are Eqs. (19), (20).   

$$\left|m_{\text{O}}-\sum M_{\text{O}}(R_1+R_2+R_3+R_9)V_{\text{cell}}\right|\text{/}{m}_{\text{O}}<\text{0}\text{.01}$$(19)

  

$$\left|m_{\text{Fe}}-\sum M_{\text{Fe}}(R_6+R_9+R_{11})V_{\text{Cell}}\right|/m_{\text{Fe}}<\text{0}\text{.01}$$(20)

where, m_O is the mass supply rate of the element O in the solid phase at the solid inlet, and m_Fe is the mass supply rate of element Fe in the solid phase at the solid inlet.

4. Results and Discussion

4.1. Base Results and Model Validation

The simulation was performed first with the conditions being determined by the normal BF data (Table 4) and the charge pattern of coke and ore is given in Fig. 2.

Table 4. BF operation data.

Variables	Values
Productivity (tHM/day)	6250
Blast temperature (K)	1523
Blast rate (Nm3/min)	4800
Oxygen enrichment (mol%)	4.0
Top absolute pressure (Pa)	2.8×105
PC injection rate (kg/tHM)	180
Ore rate (kg/tHM)	1680
Coke rate (kg/tHM)	335
Batch weight of ore (ton)	76
Batch weight of coke (ton)	15
Solid inlet temperature (K)	300
Ore particle property	Composition: TFe: 55.8 wt%, FeO: 6.8 wt%, CaO: 4.60, SiO2: 9.97 wt%, Al2O3: 2.19 wt%, TiO2: 2.0 wt% and V2O5 1.0 wt%; Porosity: 0.35; Bulk density: 1750 kg/m3; Average particle size: 20 mm.
Coke particle property	Composition: Fixed Carbon: 90 wt%, and Ash: 10 wt%; Porosity: 0.50; Bulk density: 500 kg/m3; Average particle size: 40 mm.
PC property	Composition: C: 80.0 wt%, H: 4.0 wt%, O: 3.5 wt%, N: 2.0 wt%, and S: 0.32 wt%; H2O: 4.0 wt%, and Ash: 7.0 wt%.
Liquid phase (molten iron and slag) property	[%C]: 4.0 wt%, Temperature: 1753 K, Average heat capacity: 1000 J/kg, and Slag rate: approximately 400 kg/tHM

*tHM: ton hot metal

Fig. 2.

Radial variation of coke/ore volume ratio.

Table 5 is the comparison of model predictions on some typical BF operation indexes and averaged industrial data. The agreement of the comparison is acceptable considering the simplification of the BF gas composition and the BF reactions in the model.

Table 5. Comparison of model predictions with BF industrial data.

Item	Measured Value	Calculated Value
Top gas temperature near the center (K)	573	618
Top gas temperature near the wall (K)	423	349
Top gas utilization efficiency (%)	50.0	49.8
Pressure drop in BF (MPa)	0.140	0.120

Figure 3 shows some in-furnace characteristics of the gas and solid phases. In Fig. 3, the black zone represents the CZ, which presents a typical inverted V shape. Figure 3(a) displays the gas pressure drop profile, showing that the pressure drop from the BF top to the CZ is 0.04 Mpa, which is nearly 30 percent of the total in-furnace pressure drop. Figure 3(b) shows the gas temperature profile. Around the raceway zone, the gas temperature reaches a temperature of 2473 K. The gas temperature decreases with its rising due to the heat exchange with the solid phase. In the CZ, exchanged heat from gas to solid is consumed as melting latent heat, so the cooling rate of the gas phase gets quicker. Figure 3(c) shows the solid flow pattern. Three flowing zones could be clearly identified in the motion of the solid particles: (1) plug flow in the upper part; (2) main funnel flow region through which the majority of the particles flow smoothly into the raceway; and (3) quasi-stagnant region, bounded by the stagnant zone and the main flow region, where particles move slowly compared to the funnel flow region. Figure 3(d) shows the solid temperature profile. The solid is heated up with its descent by the heat exchange from the gas phase flowing upward. The high temperature is observed in the shaft central part and forms a steep temperature gradient. These simulation results agree with the actual BF in-furnace state, suggesting that the developed model is reliable.

Fig. 3.

Simulation results under the actual BF operation conditions: (a) gas pressure drop profile, (b) gas temperature profile, (c) solid flow pattern, and (d) solid temperature profile.

4.2. Gasification Rate Expression of HCMB Carbon in Model

Research of Tang⁵⁾ showed that the conversion of the carbon in a single HCMB briquette could be described by Eq. (21) under the simulated in-furnace environment of the BF.   

$$df/dt=1\mathrm{\hspace{0.17em} }200exp(-20\mathrm{\hspace{0.17em} }000/T){(1-f)}^{2/3}(P_{\text{CO2}}/1.0\times {10}^5)$$(21)

where, f is the gasification fraction of the briquette, T is the briquette temperature, P_CO2 is the pressure of CO₂, and t is time.

Figure 3(c) shows that the particles of the solid phase present a pattern of plug flow, indicating their mixing in the radial and axial directions is negligible. Therefore, for a specific cell on the flowing path of the solid particles, the gasification rate of the HCMB carbon is expressed with Eqs. (22), (23).   

$$\begin{array}{l}{R}_{10}=\\ {\rho }_{\text{s,0}}{y}_{\text{BC,0}}1\mathrm{\hspace{0.17em} }200exp(-2\mathrm{\hspace{0.17em} }0000/T_{\text{s}}){(1-f_{\text{C,HCMB}})}^{2/3}(P_{{\text{CO}}_{\text{2}}}/1.0\times {10}^5)\\ /M_{\text{C}}\end{array}$$(22)

  

$$f_{\text{C,HCMB}}=1.0-({\rho }_{\text{s}}{y}_{\text{C,HCMB}})/({\rho }_{\text{s,0}}{y}_{\text{BC,0}})$$(23)

where, ρ_s,0 and y_BC,0 are the bulk density of the solid phase and the HCMB carbon mass fraction in the initial cell of the streamline going through the investigated cell.

4.3. Influence of Charging HCMB on Gas and Solid Behavior

Three cases were investigated and compared. Simulation conditions of the three cases are shown in Table 6. Case A (the base case) represents the BF normal operation, of which the simulation conditions are determined using the data in Table 4, and the results are given in Table 5 and in Fig. 3. The results of case A are kept as reference values for other cases. For cases B and C, most of the simulation conditions are the same as those of case A. The mixing ratio of HCMB is 5% in case B, and 10% in case C. The coke supply rates in cases B and C are determined using a trial and error method. The convergence criteria for the mass balance of element C in the coke is Eq. (24).   

$$\begin{array}{l}\left|m_{\text{C,coke}}-\sum M_{\text{C}}(R_4+R_5+R_9)V_{\text{cell}}-m_{\text{Fe,ore}}[\%\text{C]}-m_{\text{C,other}}\right|\text{/}{m}_{\text{C,coke}}\\ <\text{0}\text{.01}\end{array}$$(24)

where, m_C,coke and m_Fe,ore are the mass supply rates of element C in the coke, and element Fe in the ore at the solid inlet, respectively; [%C] is the carbon content of the hot metal in the hearth; and m_C,other is the mass rate of carbon consumed by other reactions as silica and titanium oxide reductions, which is determined by the base case.

Table 6. Simulation conditions of different cases.

	Variables	Case A (base case)	Case B	Case C
Solid inlet conditions	Ore supply rate (kg/s)	4.03
	HCMB supply rate (kg/s)	0	0.20	0.40
	Coke supply rate (kg/s)	0.804	–	–
	Solid temperature (K)	300
Gas inlet conditions	Gas supply rate (kg/s)	3.88
	Gas composition (mass fraction)	CO: 0.20, O2: 0.13, and N2: 0.67
	Gas temperature (K)	2350

*Productivities of case A, case B, and case C are 6250, 6644, and 7038 tHM/day, respectively.

Figure 4 shows the simulation results on the HCMB gasification in different cases. The region where the HCMB gasification occurs enlarges by increasing the HCMB mixing ratio. The overall gasification rate of the HCMB above the CZ of each case was calculated by $\sum M_{\text{C}}{R}_{\text{10}}{V}_{\text{cell}}$ and the results are shown in Fig. 5(a). The overall HCMB gasification rate is more than 2.58×10⁻² kg/s after the mixing ratio reaches 5%. The overall gasification fraction of the HCMB above the CZ was evaluated by $\left(\sum M_{\text{C}}{R}_{\text{10}}{V}_{\text{cell}}\right)\text{/}{m}_{\text{HCMB,C}}$ and the results are shown in Fig. 5(b). The overall gasification fraction of the HCMB reaches 52% under the mixing ratio of 5%; however, it decreases to 41% under the mixing ratio of 10%. This suggests that increasing the HCMB addition level may also lead to an increase of the ungasified HCMB carbon in the ore burden. Since the behavior of the ungasified carbon particles from the HCMB has not been studied, its influence on the simulation results is not considered in the present research.

Fig. 4.

Profile of HCMB gasification rate.

Fig. 5.

HCMB gasification behavior: (a) variation of overall gasification rate with HCMB mixing ration, and (b) variation of overall gasification fraction with HCMB mixing ratio.

Simulation results on the coke gasification above the CZ in different cases are shown in Fig. 6. The region where the coke gasification occurs shrinks with the increase of the HCMB mixing ratio. The HCMB has a lower initial temperature and a faster gasification rate than the coke under the BF in-furnace environment. Therefore, by mixing the HCMB into the ore burden, the coke gasification above the CZ is suppressed. The overall coke gasification rate above the CZ was calculated using $\sum M_{\text{C}}{R}_5{V}_{\text{cell}}$ and the results are shown in Fig. 7. It could be seen that the overall coke gasification rate decreases fast before the mixing ratio reaches 5%, and its decrease becomes slow after.

Fig. 6.

Profile of coke gasification rate above CZ.

Fig. 7.

Variation of overall coke gasification rate above CZ with HCMB mixing ratio.

Combing the HCMB gasification and the coke gasification, the total carbon transfer rate from the solid phase to the gas phase above the CZ can be obtained and the results are shown in Fig. 8. It could be seen that more carbon in the solid phase would enter the gas phase as the HCMB mixing ratio increases.

Fig. 8.

Variation of overall carbon transfer rate above CZ with HCMB mixing ratio.

As mixing HCMB increases the carbon transfer rate from the solid phase to the gas phase above the CZ, the BF gas composition hence could be changed. Simulation results on CO and CO₂ volume fractions in BF gas are shown in Figs. 9 and 10, respectively. In Fig. 9, the line of 0.4 near the center moves upward as the HCMB mixing ratio increases, indicating an increase of the CO volume fraction in BF gas. In Fig. 10, profiles of CO₂ volume fraction in case B and C are similar, however, the area with CO₂ volume fraction larger than 0.35 in case C is much smaller than in the other two cases, indicating mixing 10% HCMB in the ore causes an observable drop of CO₂ volume fraction in the BF gas. Corresponding to CO profiles in Fig. 9 and CO₂ profiles in Fig. 10, the BF top gas utilization efficiency exhibits an obvious drop after the HCMB mixing ratio reaches 10% (Fig. 11).

Fig. 9.

Profile of CO volume fraction.

Fig. 10.

Profile of CO₂ volume fraction.

Fig. 11.

Variation of BF top gas utilization efficiency with HCMB mixing ratio.

Results on the ore reduction in different cases are shown in Fig. 12. Near the center and in the mid-radial region, the distance between the line of 0.95 and the CZ increases by increasing the HCMB mixing ratio, reflecting an increase of ore reduction above the CZ by charging HCMB.

Fig. 12.

Profile of ore reduction fraction.

Results on the solid temperature in the BF is shown in Fig. 13. By charging HCMB, the increase of the solid temperature is obviously delayed near the wall and in the mid-radial region (e.g. the lines of 1073 and 1273 K move downward as the HCMB mixing ratio increases). Results on the BF gas temperature are shown in Fig. 14. in all individual cases, the gas temperature profile is similar to that of the solid temperature, which is due to the efficient gas-solid heat exchange in the BF. Variation of BF top gas temperature with the HCMB mixing ratio is displayed in Fig. 15, showing charging HCMB has a cooling effect on the BF gas.

Fig. 13.

Profile of solid temperature.

Fig. 14.

Profile of gas temperature.

Fig. 15.

Variation BF top gas temperature with HCMB mixing ratio.

Gasification of the HCMB carbon could increase the CO volume fraction in the BF gas (Fig. 9), therefore, Charging HCMB into BF has the positive effects of promoting the ore reduction, and suppressing the coke gasification above the CZ. As the HCMB carbon gasification proceeds being accompanied by the reduction of coke gasification, its thermal effect on the BF performance is not important. However, HCMB is a composite of iron and carbon, so charging HCMB introduces extra iron into the BF to increase the BF productivity resulting in a delayed temperature increase of the solid phase and a temperature drop of the gas phase. In the BF, rates of reactions (1-3) are sensitive to temperature and gas reducing potential (P_CO/P_CO2), and higher temperature or higher gas reducing potential is favored. In case B, as the HCMB mixing ratio is low (β=5%), the thermal effects of HCMB iron on the BF gas and the ore burden is not significant, therefore, the positive effect of HCMB carbon gasification on the ore reduction are evident and the BF performance could be improved. In case C, as the mixing ratio is large (β=10%), the thermal effect of HCMB iron becomes important and improving the BF performance by charging HCMB is difficult.

4.4. Coke Saving Analysis

Simulation results of some BF indexes involving reducing agent consumption are listed in Table 7. In Table 7, it could be seen that the HCMB carbon rate increases with the increase of the HCMB mixing ratio. In case C, the HCMB carbon rate reaches a high level of 37.3 kg/tHM. However, its gasification fraction is low as 41%. Based on the knowledge on the behavior of the unburnt PC particles in the lower part of the BF, the ungasified HCMB carbon have three pathways of (1) being consumed by the carburization of the HCMB iron during the melting, which occupies only a small portion; (2) forming part of the dust in the BF gas, and (3) being entrapped by the molten slag and iron deteriorating the permeability of the coke bed and leading to a unstable operation of the BF.²⁸⁾ Therefore, high HCMB mixing level with a low gasification fraction of HCMB carbon above the CZ is not favored.

Table 7. Reducing agent consumption rate in different cases.

Item	Case A	Case B	Case C
HCMB carbon rate (kg/tHM)	0.0	19.7	37.3
PC rate (kg/tHM)	180	169	160
Coke rate (kg/tHM)	335	304	283

Table 7 also shows that the coke consumption rate significantly decreases with the increase of the HCMB mixing ratio. This is mainly attributed to the introduction of metallic iron in the HCMB. To remove the effect on the coke rate reduction induced by the HCMB iron, coke consumption rate distribution is calculated on the basis of producing 1.063 ton hot metal in case B (1.0 ton hot metal from the ore and 0.063 ton hot metal from the HCMB iron), and 1.126 ton hot metal (1.0 ton hot metal from the ore and 0.126 ton hot metal from the HCMB iron). Results of the coke consumption rate distribution of the three cases are shown in Table 8. It is assumed that the ungasified HCMB carbon is only for the carburization of the iron in HCMB in Table 8. Results in Table 8 indicate that the coke consumption for producing one-ton hot metal from the ore could be lowered to approximately 323 kg under the HCMB mixing ratio of 5%, and to approximately 318 kg under the HCMB mixing ratio of 10%.

Table 8. Distribution of coke consumption rate in different cases.

Item	Case A (kg/tHM)	Case B (kg/(1.063 tHM))	Case C (kg/(1.126 tHM))
Combustion	175.1	175.1	175.1
Gasification above CZ	44.4	33.7	28.9
Reduction of molten FeO	63.2	62.3	61.9
Carburization of molten iron	45.0	45.0	45.0
Other reactions	7.3	7.3	7.3
Total	335.0	323.4	318.2

4.5. Applicability of HCMB

Presently the HCMB is prepared using pure hematite fines and non-caking coal fines. All the results on the HCMB are obtained through lab-scale tests and theoretical analysis. However, this research provides an alternative way for a partial substitution of coke by non-caking coal in addition to the PCI. In industrial practices, the main problems arise from the cost of HCMB preparation and the adjustment of BF operation conditions.

Regarding the cost of HCMB preparation, it is considered that the cost could be controlled by utilizing the resources in the iron and steel making plants. The raw materials for the HCMB include coal and ultra-fine hematite powders. the coal powders could be directly obtained from the PCI sector, and the ultra-fine hematite powders are available by processing the iron-oxide rich metallurgical dust as sinter dust, BF dust, and steelmaking dust. As these dust fines are collected from sinter flue gas, BF gas, or steelmaking off-gas, they are with an average size of approximately 40 μm.²⁹⁾ Energy consumption on grinding these iron-oxide rich fines to the size satisfied for the HCMB preparation is acceptable.

Regarding the adjustment of the BF operation conditions, it is considered that no significant adjustment of the operation conditions is required in case that the HCMB mixing ratio is within 5%. Under an addition level of 5% HCMB in the ore burden, the temperature drop of the BF top gas is approximately 10 K, and the HCMB reaches an overall gasification fraction of 52%, indicating changes of the BF thermal state and of the permeability in the BF lower part are negligible; nevertheless, a distinguished coke saving effect for producing one-ton hot metal from the ore could be realized.

5. Conclusions

A numerical investigation has been carried out by applying HCMB in the BF ironmaking. The following conclusions could be drawn.

(1) A BF model based on gas-solid counter flow has been developed. Reliability of the model is confirmed by the comparison with the simulation results and the industrial data of an actual BF.

(2) The simulation results show that charging HCMB in BF can suppress the coke gasification and improve the ore reduction above the CZ. Coke of approximately 12 kg could be saved for producing one-ton hot metal from the ore under an HCMB mixing ratio of 5%, and approximately 17 kg under a HCMB mixing ratio of 10%.

(3) The simulation results show that the reasonable mixing ratio of HCMB in the ore burden is within 5%, by which, change of the BF thermal state and permeability of the BF lower part are negligible. Therefore, mixing HCMB does not require significant changes of the BF operation conditions.

Acknowledgments

The authors would like to thank the National Natural Science Foundation of China (No. 51144010), and the State Key Laboratory of Advanced Metallurgy of University of Science and Technology Beijing (USTB) (No. 41618012) for supporting this work.

Nomenclature

A_face: cell face area, m²

A: specific area, m²/m³

a_k: rate coefficient of Interface k of the three-interface unreacted core model, m/s

Cp: heat capacity, J/(kg·K)

d: diameter, m

E: enthalpy source, J/(m³·s)

f: reduction fraction, gasification fraction, -

H: total enthalpy, J/kg

ΔH: reaction heat, J/kmol

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

k: reaction rate constant/mass transfer coefficient, 1/s, m/s

M: molar weight (kg/kmol)

m: mass supply/consumption rate of the given element, kg/s

P: pressure, Pa

Pr: Prandtal number, -

R_i: reaction rate of reaction i, kmol/(m³·s)

r: r directional, m

Re: Reynolds number, -

S: source term, units vary

Sc: Schmidt number, -

T: temperature, K

V_Cell: volume of cell, m³

y: mass fraction, -

z: z directional, m

Vector

${\stackrel{\rightharpoonup }{F}}_{\text{gs}}$: gas flow resistance, N/(m³·s)

$\stackrel{\rightharpoonup }{n}$: normal unit vector on the cell face

${\stackrel{\rightharpoonup }{U}}_{\text{g}}$: gas superficial velocity vector, m/s

${\stackrel{\rightharpoonup }{V}}_{\text{S}}$: solid physical velocity vector, m/s

Greek Symbols

ϕ: general dependent variable

Γ: general diffusion coefficient

α: volume fraction, -

β: HCMB mixing ratio, -

η: fraction of the liquid phase, -

λ: thermal conductivity, W/(m·K)

ε: porosity, -

μ: fluid viscosity, kg/(m·s)

ρ: density, bulk density of solid phase and its species or element, kg/m³

Subscripts

s: solid variable

g: gas variable

l: liquid variable

coke: coke variable

ore: ore variable

HCMB: HCMB variable

species or element name: variable of assigned species or element

References

• 1)   J.  Zhang and  G.  Wang: Energy, 33 (2008), 525.
• 2)   J.  Kopfle and  R.  Hunter: Ironmaking Steelmaking, 35 (2008), 254.
• 3)   I. G.  Tovarovskiy: Energy Sci. Technol., 6 (2013), 1.
• 4)   M. M.  Mollah,  W. R.  Jackson,  M.  Marshall and  A. L.  Chaffee: Fuel, 148 (2015), 104.
• 5)   H.  Tang,  S.  Liu and  R.  Rong: ISIJ Int., 59 (2019), 22.
• 6)   C.  Zhou,  G.  Tang,  J.  Wang,  D.  Fu,  T.  Okosun,  A.  Silaen and  B.  Wu: JOM, 68 (2016), 1353.
• 7)   P. R.  Austin,  H.  Nogami and  J. I.  Yagi: ISIJ Int., 37 (1997), 458.
• 8)   P. R.  Austin,  H.  Nogami and  J. I.  Yagi: ISIJ Int., 37 (1997), 748.
• 9)   P. R.  Austin,  H.  Nogami and  J. I.  Yagi: ISIJ Int., 38 (1998), 246.
• 10)   G. S.  Gupta and  V.  Rudolph: ISIJ Int., 46 (2006), 195.
• 11)   D.  Ye and  J.  Hu: Thermodynamic Data Manual of Inorganics, Metallurgical Industry Press, Beijing, (2002), 48.
• 12)   S.  Natsui,  T.  Kikuchi and  R. O.  Suzuki: Metall. Mater. Trans. B, 45 (2014), 2395.
• 13)   N. N.  Viswanathan,  N.  Srinivasan and  A. K.  Lahiri: ISIJ Int., 38 (1998), 1062.
• 14)   Y.  Dutil,  D. R.  Rousse,  N. B.  Salah,  S.  Lassue and  L.  Zalewski: Renew. Sustain. Energy Rev., 15 (2001), 112.
• 15)   H.  Hu and  S. A.  Argyropoulos: Model. Simul. Mater. Sci. Eng., 4 (1996), 371.
• 16)   J. G.  Mathieson,  J. S.  Truelove and  H.  Rogers: Fuel, 84 (2005), 1229.
• 17)   T.  Kon,  S.  Natsui,  S.  Ueda,  R.  Inoue and  T.  Ariyama: ISIJ Int., 52 (2012), 1565.
• 18)   M.  Hatano and  K.  Kurita: Trans. Iron Steel Inst. Jpn., 22 (1982), 448.
• 19)   X. F.  Dong,  A. B.  Yu,  S. J.  Chew and  P.  Zulli: Metall. Mater. Trans. B, 41 (2010), 330.
• 20)   X.  Yu and  Y.  Shen: Metall. Mater. Trans. B, 49 (2018), 2370.
• 21)   S. B.  Kuanga,  Z. Y.  Li,  D. L.  Yan,  Y. H.  Qi and  A. B.  Yu: Miner. Eng., 63 (2014), 45.
• 22)   Z. Y.  Zhou,  H. P.  Zhu,  A. B.  Yu,  B.  Wright and  P.  Zulli: Comput. Chem. Eng., 32 (2008), 1760.
• 23)   P.  Zhou,  H. L.  Li,  P. Y.  Shi and  C. Q.  Zhou: Appl. Therm. Eng., 95 (2016), 296.
• 24)   J.  Chen,  T.  Akiyama,  H.  Nogami,  J. I.  Yagi and  H.  Takahashi: ISIJ Int., 33 (1993), 664.
• 25)   S. J.  Zhang,  A. B.  Yu,  P.  Zulli,  B.  Wright and  P.  Austin: Appl. Math. Model., 26 (2012), 141.
• 26)  CHAM: PHOENICS User Document, CHAM, London, (2000).
• 27)   Z.  Zhang,  J.  Meng,  L.  Guo and  Z.  Guo: JOM, 67 (2015), 1936.
• 28)   K.  Wu,  R.  Ding,  Q.  Han,  S.  Yang,  S.  Wei and  B.  Ni: ISIJ Int., 50 (2010), 390.
• 29)   G.  Zhan and  Z.  Guo: J. Environ. Sci. (China), 25 (2013), 1226.