Reduction of Magnetite Concentrate Particles by H₂+CO at 1673 K

Abstract

The purpose of this research was to determine the effect of CO addition to H₂ on the reduction kinetics of magnetite concentrate particles at high temperatures. Experiments were carried out at 1673 K. The replacement of N₂ by CO enhanced the reduction rate considerably. The kinetics of CO reduction is slower than that by hydrogen but significant, and the contributions by the two gases are additive.

1. Introduction

A novel alternate ironmaking technology based on the direct gaseous reduction of fine magnetite concentrate particles is under development at the University of Utah.¹⁾ This technology produces iron directly from magnetite concentrate particles by gas-solid suspension reduction. A mixture of CO and H₂ produced by partial combustion of natural gas is to be used as the reducing reagents in this process.²⁾ The reduction of iron oxide with CO, H₂, and mixture of H₂ and CO has been widely investigated. However, most of the work has been done with pellets, briquettes, and iron oxide powder beds and thus at temperatures below ~1400 K.^3,4,5,6,7,8) Much less work has been performed on the reduction rate of fine magnetite concentrate particles with mixtures of H₂ and CO at temperatures above 1473 K to be used in the process under development.

It has been determined that magnetite concentrate particles of 20–53 μm sizes can be completely reduced to iron by hydrogen within several seconds above 1473 K.^9,10) The overall reaction rate was proved to be controlled by the chemical reduction reaction.⁹⁾

This communication describes the effect of CO addition to H₂ on the reduction kinetics of magnetite concentrate particles at 1673 K.

2. Experimental Work

Details of experimental work have been described in an earlier related work⁹⁾ and thus only brief salient information will be given here.

Magnetite concentrate from a taconite ore of the Mesabi Range of the U.S. was used in this study. A high temperature drop-tube reactor with a dilute particles-gas flow system, described in greater detail elsewhere,^9,11) was used to measure the chemical reaction rate of fine particles entrained in a reducing gas. As shown in Fig. 1, this apparatus consisted of a vertical tube furnace, a pneumatic powder feeder, gas delivery lines, a powder cooling and collecting system, and an off-gas outlet. The furnace system used a vertical alumina tube (5.6 cm ID, 193 cm long) in which a reaction zone was maintained at a constant temperature of 1173 to 1773 K by bar-type SiC elements. Carefully measured reaction zone was 91 cm long within ± 20 K lengthwise for typical downward gas flow conditions. A cylindrical alumina honeycomb was inserted in the tube and hung right above the beginning of the reaction zone as a flow straightener and a heat exchanger for the reducing gas.

Fig. 1.

Schematic diagram of drop-tube reactor system.

2.1. Definition of Parameters

2.1.1. % Excess H₂

The term % excess H₂ used in this work is defined as follows, taking into consideration the fact that the final stage of the iron oxide reduction, i.e. the reaction of FeO with H₂, is significantly limited by equilibrium. For example, at 1673 K,   

$$FeO\left(s\right)+H_2\left(g\right)=Fe\left(s\right)+H_{2}O\left(g\right); \mathrm{\Delta }{G}^{°}\cong +90 cal \left(1\mathrm{\hspace{0.17em} }673\mathrm{\hspace{0.17em} }K\right)$$(1a) ¹²⁾

  

$${\left[p_{{\text{H}}_2}/\left(p_{{\text{H}}_2}+p_{{\text{H}}_2\text{O}}\right)\right]}_{\text{eq}}=0.5$$(1b)

This reaction has a slightly positive standard Gibbs free energy, and the equilibrium gas product has an H₂/H₂O molar ratio of ~ 1 at this temperature, i.e. 50% H₂ and 50% H₂O when pure hydrogen is used as the reductant. Taking the equilibrium composition into consideration, % excess H₂ is defined in this work as follows:   

$${\text{Fe}}_3{\text{O}}_4+{\text{ H}}_2=3\text{FeO}+{\text{H}}_2\text{O}$$(2a)

  

$$\text{FeO}+{\text{H}}_2=\text{Fe}+{\text{H}}_2\text{O}$$(2b)

  

$$\text{Overall:}\mathrm{\hspace{0.17em} }\mathrm{\hspace{0.17em} }{\text{Fe}}_3{\text{O}}_4+4{\text{H}}_2=3\text{Fe}+4{\text{H}}_2\text{O}$$(2c)

  

$$K_{\left(2b\right)}={\left(p_{H_{2}O}/p_{H_2}\right)}_{eq}={\left(n_{H_{2}O}/n_{H_2}\right)}_{eq}$$(3)

  

$$n_{H_2,min}=n_o^i+\frac{n_o^i}{K_{\left(2b\right)}}=\left(1+\frac{1}{K_{\left(2b\right)}}\right)n_o^i$$(4)

  

$$\%\mathrm{\hspace{0.17em} }excess\mathrm{\hspace{0.17em} }{H}_2=\frac{n_{H_2,supplied}-n_{H_2,min}}{n_{H_2,min}}\times 100$$(5)

Reaction (2a), the hydrogen reduction of magnetite to wustite, has a large equilibrium constant, i.e. essentially irreversible, whereas Reaction (2b) is considerably limited by chemical equilibrium. The minimum amount of hydrogen, n_H2,min, is the amount of hydrogen used to remove the oxygen from the iron oxide, $n_o^i$ , plus the hydrogen required by equilibrium to be present with the water vapor produced by the reduction reaction, $n_o^i$ /K_(2b), with K_(2b) = 0.973 at 1673 K. The % excess H₂ is then calculated from the total amount of hydrogen fed into the reactor, n_H2,supplied, compared with the minimum amount of hydrogen, n_H2,min, as indicated in Eq. (5).

This definition was also applied in the presence of CO reduction. Amount of excess CO was calculated according to the same feeding rate of magnetite concentrate with H₂ reduction.

2.1.2. Reduction Degree

The total iron content in the particles after reduction was determined by titration methods and the reduction degree is obtained by tracing the difference of the mass of oxygen combined with iron in the particles before and after reduction. The percent reduction is calculated as follows.   

$$\text{\%Reduction }=\frac{{\text{m}}_{\text{o}\left(0\right)}-{\text{m}}_{\text{o}\left(\text{t}\right)}}{{\text{m}}_{\text{o}\left(0\right)}}\times 100\text{\%}$$(6)

  

$${\text{m}}_{\text{o}\left(0\right)}=\frac{{\text{m}}_{\text{TFe}\left(\text{t}\right)}}{{\left(\text{\%TFe}\right)}_0}\times {\left(\text{\%O}\right)}_0$$(7)

  

$${\text{m}}_{\text{o}\left(\text{t}\right)}=\text{m}-{\text{m}}_{\text{TFe}\left(\text{t}\right)}-\frac{{\text{m}}_{\text{TFe}\left(\text{t}\right)}}{{\left(\text{\%TFe}\right)}_{\text{t}}}\times \left[100-{\left(\text{\%TFe}\right)}_0-{\left(\text{\%O}\right)}_0\right]$$(8)

Here, m is the mass of reduced sample used in the titration, m_o(t) and m_o(0) are, respectively, the mass of the removable oxygen in the reduced sample collected after reaction for time t, and the corresponding mass of the removable oxygen in the unreduced dry concentrate. m_TFe(t) is the mass of total iron in the reduced sample (mass is m). (%TFe)₀ and (%O)₀ are the mass percentage of total Fe and removable oxygen of the corresponding unreduced concentrate, which are determined respectively by titration and complete reduction conducted in a separate horizontal tubular furnace. The weight change related to possible volatile species such as phosphorus and sulphur was neglected based on their small contents and by the negligible weight loss confirmed by heating under an inert atmosphere.

3. Results

3.1. Morphologies of Samples Reduced by H₂ at 1673 K

Micrographs of particles reduced at 1673 K and the cross-sections are shown in Fig. 2. The partially reduced particles were round due to the fact that the melting point of wustite is 1649 K. The product iron coalesced in the center and was surrounded by unreduced wustite, as shown in Fig. 2(b). A similar phenomenon was also reported in other investigations.¹³⁾

Fig. 2.

SEM micrographs of 69% reduced sample (32–38 μm) obtained at 1673 K: (a) Unpolished; (b) Polished section.

3.2. Comparison of Reduction Rates with H₂ and CO

From the rate equation of hydrogen reduction of magnetite concentrate, reported in an earlier paper,⁹⁾ the results of 200% pure H₂ reduction were calculated and shown in Fig. 3. The results with 200% excess pure CO obtained in this work are also shown in Fig. 3. It is seen that hydrogen reduction is much faster than CO reduction. Even when hydrogen partial pressure was decreased to 0.1 atm, the reduction rate by hydrogen was still faster than that by pure CO.

Fig. 3.

Comparison of reduction kinetics of 32–38 μm magnetite concentrate particles at 1673 K with H₂ and CO.

3.3. Reduction with Mixtures of H₂ and CO

To determine the effect of adding CO to H₂, the gas mixture of hydrogen and nitrogen was first tested to establish the baseline. The reduction rate is affected by the amount of produced water vapor, if there is insufficient amount of excess hydrogen. The examined amount of excess hydrogen varied from 0% to 800% with 0.1 atm partial pressure at 1673 K.

The results with H₂ are compared with those for gas mixtures in which N₂ was completely replaced by CO and the results are shown in Fig. 4. Based on previous analysis,¹⁴⁾ these data represent the particle reaction rates unaffected by external mass transport, i.e. the observed reduction degrees are much lower than the values expected if mass transfer were the rate-controlling step.

Fig. 4.

The effect of gaseous mixture of H₂ and CO on reduction kinetics of 32–38 μm magnetite concentrate particles at 1673 K with different amount of excess H₂. □ △ ○: p_H2/p_CO/p_N2 = 0.1 atm/0 atm/0.75 atm: [□: 800% excess H₂, △: 200% excess H₂, ○: 0% excess H₂]. ■ ▲ ●: p_H2/p_CO/p_N2 = 0.1 atm/0.75 atm/0 atm: [■: 800% excess H₂/2950% excess CO, ▲: 200% excess H₂/920% excess CO, ●: 0% excess H₂/240% excess CO]. [The total pressure was constant at 0.85 atm (86.1 kPa), the atmospheric pressure in Salt Lake City, Utah.]

The effect of replacing N₂ by CO was small under large excess H₂ but became significant with lower excess H₂. When there is large excess H₂, the reduction is largely done by H₂ because of its greater reducing power as seen in Fig. 3; thus the replacement of N₂ by CO has a moderate effect. Under low excess H₂, the driving force for reduction by hydrogen is decreased by the produced water vapor and the reduction by CO makes measurable contribution to the overall reduction of iron oxide. The kinetics of CO reduction is slower than that by hydrogen but significant and the contributions by the two gases are additive.

Both the CO+FeO reaction and the water-gas shift reaction that would produce H₂ are unfavorable at this high temperature,¹²⁾ as indicated below:   

$$\begin{array}{l}\text{FeO}\left(\text{s}\right)+\text{CO}\left(\text{g}\right)=\text{Fe}\left(\text{s}\right)+{\text{CO}}_2\left(\text{g}\right);\\ \mathrm{\Delta }Gº\text{at }1\mathrm{\hspace{0.17em} }673\mathrm{\hspace{0.17em} }\text{K}=+4\mathrm{\hspace{0.17em} }142\text{ cal }\left(K_{eq}=0.29\right)\end{array}$$(9)

  

$$\begin{array}{l}{\text{H}}_2\text{O}\left(\text{g}\right)+\text{CO}\left(\text{g}\right)={\text{H}}_2\left(\text{g}\right)+{\text{CO}}_2\left(\text{g}\right);\\ \mathrm{\Delta }Gº\text{at }1\mathrm{\hspace{0.17em} }673\mathrm{\hspace{0.17em} }\text{K}=+\mathrm{\hspace{0.17em} }4\hspace{0.17em}052\text{ cal }\left(K_{eq}=0.30\right)\end{array}$$(10)

Thus, when a mixture of H₂ + CO is used in a small excess amount, as would be done in an industrial process, the contribution by CO is expected to be limited because of the equilibrium limitation, which becomes more unfavorable at higher temperatures, in addition to the slower kinetics. Further, no carbon was observed on the reduced particles under the test conditions.

Work is planned in this laboratory to collect more comprehensive data on the reduction by H₂ + CO mixtures over wider ranges of conditions, including the effect of the presence of H₂O and CO₂.

4. Concluding Remarks

It was determined in this work that the replacement of N₂ by CO increased the reduction rate of magnetite concentrate particles significantly at 1673 K, and the contributions of H₂ and CO are approximately proportional to their individual reduction rates. At 1673 K, the reduction rate by pure CO is considerably slower than reduction by hydrogen.

This work had the specific aim of (1) determining the effect of CO addition to H2; (2) determining the behavior of CO as a reducing gas at a significantly higher temperature (1673 K) than previously studied levels (at most 1373 K typically much lower) because the study is related to a new process that uses this higher temperature; and (3) establishing the reduction rate of concentrate particles by CO+H2 mixtures unlike most previous work done with pellets and briquettes.

Acknowledgments

The authors acknowledge the financial support from American Iron and Steel Institute (AISI) through a Research Service Agreement with the University of Utah under AISI’s CO₂ Breakthrough Program. This material also contains results of work supported by the U.S. Department of Energy under Award Number DE-EE0005751.

References

• 1)   H. Y.  Sohn: Steel Times Int., 31 (2007), No. 4, 68.
• 2)   H. K.  Pinegar,  M. S.  Moats and  H. Y.  Sohn: Ironmaking Steelmaking, 39 (2012), 398.
• 3)   E.  Turkdogan and  J.  Vinters: Metall. Mater. Trans. B, 3 (1972), 1561.
• 4)   A. A.  El-Geassy,  K. A.  Shehata and  S. Y.  Ezz: Trans. Iron Steel Inst. Jpn., 17 (1976), 629.
• 5)   J.  Szekely and  Y.  El-Tawil: Metall. Mater. Trans. B, 7 (1976), 490.
• 6)   N.  Towhidi and  J.  Szekely: Ironmaking Steelmaking, 8 (1981), 237.
• 7)   O.  Nakazato,  H. T.  Yonezawa and  T.  Usui: ISIJ Int., 43 (2003), 1502.
• 8)   H.  Wang and  H. Y.  Sohn: Steel Res. Int, 83 (2012), 903.
• 9)   H.  Wang and  H. Y.  Sohn: Metall. Mater. Trans. B, 44 (2013), 13.
• 10)   H. Y.  Sohn,  M. E.  Choi,  Y.  Zhang and  J. E.  Ramos: Iron Steel Technol. (AISI Trans.), 6 (2009), No. 8, 158.
• 11)   M. E.  Choi and  H. Y.  Sohn: Ironmaking Steelmaking, 37 (2010), 81.
• 12)  Outokumpu HSC Chemistry 5.11, Outokumpu Research Oy, Pori, Finland, (2002).
• 13)   S.  Hayashi and  Y.  Iguchi: ISIJ Int., 34 (1994), 555.
• 14)   M. E.  Choi: PhD Dissertation, University of Utah, Salt Lake City, Utah, (2010).