Reduction kinetics of magnetite concentrate particles by carbon monoxide

Abstract

In this study, the reaction kinetics of carbon monoxide reduction of magnetite ore concentrate in the temperature range of 973–1173 K (700–900°C) has been investigated using the thermal-gravimetric analysis. Various methods were used to obtain the governing mechanism and reaction rate constant of this reaction. It was observed that the sintering phenomenon affects the concentrate particles’ reduction at temperatures higher than 1173 K (900°C). CO reduction of magnetite concentrate to pure iron or iron carbide occurred as a single-step reaction. The first-order reaction model and diffusion models are two important mechanisms for this reaction in the operating temperature range. The following equation is the result of our research work relevant to carbon monoxide reduction of the magnetite concentrate particles in order to predict the progress of the reduction reaction (with R² about 0.98):

Keywords

Magnetite concentrate carbon monoxide reduction kinetics TGA

1 Introduction

Nowadays, the problem of increasing earth temperature and environmental pollutions exhort scientists to find alternative ironmaking technologies in order to replace the blast furnace. Owing to coke-making and coke-burning and also needing pelletization and sintering of ore concentrate, the blast furnace consumes high energy and emits high carbon dioxide. Direction of the new ironmaking processes, such as flash reduction of fine ore particles at high temperatures, is based on the eliminating sintering, coke-making and pelletization (Chen et al. 2015a, 2015b). Reformed natural gas or coal gas comprising carbon monoxide and hydrogen is a good candidate as a reducing gas in these processes.

Some researchers studied reduction of iron oxides by carbon monoxide through different steps of the reduction (Kawasaki et al. 1962; Trushenski et al. 1974; Hughes et al. 1982; Geva et al. 1990; Kang et al. 1998; Qiyuan et al. 1998; Mondal et al. 2004; Jozwiak et al. 2007; Noguchi et al. 2013). Nevertheless, a limited work considered the reduction reaction of iron ore particles by CO as a single step (Hughes et al. 1982; Chen et al. 2015b). Since the aim of this research work is providing a rate equation to use it in the design of a flash reduction reactor, conversion of the magnetic ore concentrate directly to iron metal was considered. Trushenski et al. (1974) investigated different steps of pure iron oxide reduction by carbon monoxide. They reduced disc-shaped pellets of the iron oxides by carbon monoxide at 1073–1173 K (800–900°C). They obtained activation energy of the triple steps of reduction of haematite to iron using the thermal-gravimetric analysis (TGA). Hughes et al. (1982) studied reduction of the iron ore particles by hydrogen, carbon monoxide and their mixtures at 1073–1223 K (800–950°C). They obtained activation energy of haematite ore particles to iron as a single-step reaction. Chen et al. (2015b) investigated reduction of haematite concentrate particles falling in carbon monoxide gas at 1423–1623 K (1150–1350°C) relevant to flash ironmaking process. The obtained value of activation energy for this reaction in the mentioned temperature range was higher than that of other similar studies.

In this study, the reaction kinetics of carbon monoxide reduction of iron ore concentrate in the temperature range of 973–1173 K (700–900°C) has been investigated using the TGA. First, a set of experiments were designed to insure the repeatability and accuracy of the experiments and also to obtain suitable conditions for the determination of the intrinsic reaction rate parameters. Fitting method and nucleation and growth model have been used to obtain the governing mechanism and reaction rate constant of the reduction reaction.

2 Experiments

2.1 Materials

The iron ore concentrate was provided from Gol-e-Gohar mine at south west of Iran, Sirjan, which contains about 66% total iron and 27% FeO. Impurities of the magnetite concentrate were sulphur, phosphor and oxides of silicon, aluminium, magnesium and calcium. The average size of the particles was 47 µm and about half of them have a size in the range of 44–53 µm. Helium and carbon monoxide have been supplied from the Air-Product Company with 99.99% and 99.999% purities, respectively.

2.2 Apparatus and procedure

A fabricated thermogravimetric unit was utilized to record the weight change of the solid sample by the reaction over time. The reduction reactor was an alumina tube with 34 mm inner diameter put vertically in an electrical furnace. A small ceramic crucible was suspended by a nickel-chrome wire to a digital scale (A&D-GF300 with 1 mgr accuracy) to weigh sample at each time. Reducing gas enters the reactor from the top and after reaction leaves from the bottom. The temperature of the reaction was measured using two different thermocouples (one of them placed at a point near the crucible inside the reactor and the other located at out of the reactor). The furnace started heating up the reactor to the desired temperature under an inert atmosphere. Helium was used as an inert gas; it protected the balance and also purged air content of the reactor. A stream of the reducing gas was injected to the reactor after purging the air. Three gas flowmeters measured volumetric flow rate of the gases. The weight change of solid content of the crucible was recorded by a connected computer to the digital balance, every 1 sec. At the end of each run, the reducing gas flow was turned off and the solid sample was discharged under an inert atmosphere. The reduced sample was packaged in a small under vacuum container in order to carry out further examination.

3 Results and discussions

3.1 Repeatability of the experiments

Figure 1 compares two different runs of the repeated experiments. As is clear in this figure, the reduction rate of the Tests and their repetitions are close and their differences are negligible (max differences are 8.37% and 8.27% for Tests 1 and 2, respectively). The vertical axis of the Figure shows progress of the reaction ( ), where w, w_i and w_f are instantaneous, initial and final weights of the solid sample, respectively (El-Rahaiby & Rao 1979).

Figure 1

Comparisons were carried out to verify repeatability of the experiments: Test 1, 1 gr iron concentrate in the presence of 4.571 Nlit.min⁻¹ flow rate of carbon monoxide at 1163 K (890°C) and test 2, 1 gr iron concentrate in the presence of 2.286 Nlit.min⁻¹ flow rate of carbon monoxide at 1163 K (890°C).

3.2 Factors affecting the overall rate

Four successive phenomena compete with each other for reducing solid particles in a fixed bed of iron concentrate inside the crucible:

Transfer of CO (reducing gas) and CO₂ (produced gas) into a formed gaseous film layer around the powders bed

Diffusion of CO and CO₂ within the powders bed

Diffusion of CO and CO₂ within the solid fine particles

The chemical reaction of solid iron oxide with CO

Owing to successive order of these steps in the reduction reaction, the slowest step will control the overall rate of the process. Since the main aim of this study was to determine the intrinsic kinetics of the reduction reaction, a set of experiments were designed to provide the necessary condition to increase mass transfer rates, consequently the reaction was so slower than the other steps and therefore controlled process of the reduction reaction. In the following paragraphs, it is described how the mass transfer effects (external gas film, intra-bed and inter-particle mass transfer resistances) were reduced.

If the thickness of gaseous film around the solid bed be large, CO molecules should spend considerable time to penetrate this layer and may have an important effect on the reaction rate. Greater speed from top of the suspended crucible in the reactor decreases the thickness of the formed gaseous film layer. By gradually increasing the CO flow rate and comparing the resulting X versus time, we are able to determine the minimum gas flow rate that minimizes the film layer effect. Figure 2 presents the effect of CO flow rate on the reduction rate of the iron concentrate. As seen in this Figure, the reaction rate was increased with increasing gas flow rate until about 4.6 Nlit.min⁻¹, which stops this trend. Therefore, other experiments were carried out with this flow rate (or higher). Note that the flow rate in normal conditions (an ambient temperature of about 298 K (25°C) and an ambient pressure of 0.813 atm) is much lower than the actual gas flow rate in the hot zone of the reactor.

Figure 2

Effect of gas flowrate on the reduction rate of 1 gr iron concentrate powder by carbon monoxide at 1163 K (890°C).

A thick layer of the powder bed makes it difficult to penetrate CO and also for the escape of CO₂ from it. A set of experiments, at same conditions but with different weights of the concentrate, were considered in order to investigate the effect of the solid bed thickness on the reduction rate (Figure 3). As seen in Figure 3, the reduction rate of concentrate was increased with solid mass decrease (solid bed thickness decrement) until reaching 1 gr. Therefore, 1 gr of iron concentrate is a lower limit below which there is insignificant effect on the progress of the reaction.

Figure 3

Effect of thickness of concentrate powder bed on the rate of reduction by 4.571 Nlit.min⁻¹ carbon monoxide at 1163 K (890°C).

It was necessary to design a set of experiments to examine the effect of particles size on the reduction rate. Figure 4 displays conversion of iron ore concentrate versus time at different particle sizes. As seen in the Figure, by decreasing the particle size, the reduction rate was increased. It is clear that penetration of reducing gas into the larger particles is more difficult. In the case of very small particles (<25 µm), a less solid bed porosity and therefore a more difficult gas penetrating medium will be provided within the crucible and cause to decay the reduction rate. As shown in Figure 4, iron concentrate, with its wide size distribution (about 1–300 µm), was reduced with a same rate as 44–53 µm particles, interestingly. However, because about half of the concentrate has about 47 µm particle size, this similarity is not surprising. Therefore, it can be concluded that iron concentrate (with its wide particle size distribution) has at least intra-particle mass transfer effect. All the remaining experiments (effect of CO concentration and effect of temperature on the reaction rate) have been carried out with 4.571 Nlit.min⁻¹ gas flow rate and 1 gr concentrate.

Figure 4

Effect of particle size on the reduction rate of iron concentrate by 4.571 Nlit.min⁻¹ carbon monoxide at 1163 K (890°C).

Figure 5 shows the effect of CO concentration on the reduction rate of iron concentrate. Various gaseous mixtures of helium and carbon monoxide with a constant overall flow rate (4.571 Nlit.min⁻¹) were injected into the reactor for this study. As is evident, decreasing of CO concentration decreases the reaction rate. The reason for this decrease is related to role of reducing gas concentration on the reaction progressing.

Figure 5

Effect of carbon monoxide concentration on the reduction rate, 1 gr iron concentrate at 1163 K (890°C).

Figure 6 shows the effect of temperature on the iron concentrate reduction by carbon monoxide. As is evident, a rise in temperature from 1000 to 1163 K (727–890°C) increases the reaction rate sharply. As temperature goes above this range, the process will begin to be slower. It is a result of sintering and more difficult penetrating of CO through subsequently formed pasty layer around concentrate particles. Slowing of the reaction rate by sintering confirms this fact that the reduction of iron concentrate particles was achieved by carbon monoxide gas and not by fine carbon particles, which were produced by the Boudouard reaction ( ). If carbon particles contribute in the reduction of iron ore concentrate, since the mobility of the solid carbon phase in the pasty layer of the sintered particles is increased, sintering phenomenon not only do not reduce rate of the reaction but also it increases rate of the reduction (Nakagawa et al. 2005; Takeuchi et al. 2007). We can conclude that the suitable temperature for determining the intrinsic reaction kinetics of the reduction is below 1173 K (900°C).

Figure 6

Effect of temperature on the reduction rate of 1 gr iron concentrate by 4.571 Nlit.min⁻¹ carbon monoxide.

3.3 Equilibrium conditions

A detailed study of this reduction requires the knowledge of the possible reactions at equilibrium condition. Equilibrium conditions of all possible reactions were studied using the HSC software (Antti Roine 2006). Flowing of CO might be preventing the equilibrium conditions by removing produced CO₂ from the crucible. However, equilibrium conditions show the possibility of each reaction occurrence.

Figure 7 represents the equilibrium condition of the reaction at temperature range from 298 to 1773 K (25–1500°C) when 1 gr iron concentrate has been maintained in a gaseous environment containing a certain amount of carbon monoxide (equivalent to the total moles of passed CO through the reactor during the experiment). Curves of ΔG versus temperature of the possible reactions between magnetite and carbon monoxide at the above-mentioned temperatures have been presented in Figure 8. Spontaneous reactions have non-positive ΔG at a constant temperature and pressure. Therefore, reactions with a smaller positive ΔG have more chances to change into a thermodynamically favourable reaction, when operating conditions are kept out from the equilibrium conditions (Silbey et al. 2004). Figures 7 and 8 reveal the following results:

At equilibrium conditions and low temperatures, the magnetite concentrate can be converted to the other types of iron oxide in a reducing CO environment. But, keeping away of the reaction from equilibrium conditions by flowing the reducing gas and also very low rate of producing FeO makes the possibility of achieving such compounds in the final product to be low.

At low temperatures which the Boudouard reaction has a significant chance to occur, CO has a very lower mass than CO₂ and carbon. But, increasing temperature and increasing ΔG of this reaction, subsequently, increase the concentration of CO and decrease the concentration of CO₂ and carbon sharply.

Increasing temperature decreases the oxygen content of iron concentrate and converts it to wustite (FeO), pure iron (Fe) and a small amount of iron carbide (Fe₃C), at 973 K (700°C). Therefore, at equilibrium conditions about 60%, metallization will be a maximum value at this temperature. A more temperature increase decays FeO production and increases the amount of Fe. It leads to more metallization at higher temperatures by CO reduction (see Table 3). Producing of Fe₃C is limited at above 1173 K (900°C) due to a decrease in the produced carbon in the reactor.

Oxygen separation from magnetite by CO can occur under a one-step (Fe₃O₄ to Fe or Fe₃C) or two-step (Fe₃O₄ to FeO and then FeO to Fe or Fe₃C) reaction. Only CO reduction of magnetite to wustite has a negative ΔG and thus is a spontaneous reaction, above 973 K (700°C). But the applied high flow rate of CO from top of the bed disturbs this equilibrium condition and makes other gas–solid reactions possible.

3.4 Reaction mechanism

In this section, the mechanism of iron concentrate reduction by carbon monoxide has been studied, based on fitting of the existing models. The rate of reduction progress of such reactions is usually based on the following equation (Brown et al. 1980):

(1) where t is the time, x is the solid conversion, f(x) is the reaction model as a function of the solid conversion and k is the apparent reaction rate constant. The reaction model and its integral form ( ) can take several forms. Some of its applicable forms have been presented in Table 1. A common analyzing method is fitting the experimental data (isothermal conversion versus time) with the models and then evaluating this accordance using correlation coefficient (R²) (Pourghahramani & Forssberg 2007). Figure 9 a–e shows fitting of the presented models with the experimental data at different temperatures.
Figure 7
Equilibrium curves of ore concentrate and carbon monoxide gas system at temperature range from 298 to 1773 K (25–1500°C): including 1 gr of iron concentrate in contact with 11.217 kmol carbon monoxide (amount of required CO during one hour reaction, a period relevant to experiment duration) (Jozwiak et al. 2007).

Figure 8
Comparing ΔG of the possible reactions at conditions mentioned relevant to Figure 7 (Jozwiak et al. 2007).

Figure 9
Fitting of the presented models in Table 1 with the experimental data of carbon monoxide reduction of iron concentrate at different temperatures.

Table 1
The common gas-solid reactions models that used to determine mechanism of iron concentrate reduction with different gases.

Number Model f(x) g(x)

Nucleation models

1 Power law

2 Power law

3 Power law

4 Power law

5 Avrami-Erofe'ev

6 Avrami-Erofe'ev

7 Avrami-Erofe'ev

8 Avrami-Erofe'ev

Geometrical contraction models

9 Contracting area

10 Contracting volume

Diffusion models

11 1D diffusion

12 2D diffusion

13 3D diffusion

14 Ginstling-Brounshtein

Reaction-order models

15 Zero-order

16 First-order

17 Second-order

18 Third-order

Figure 10 shows logarithm of the calculated k versus inverse temperature for different models. Models with low R² values (models with less importance) have been represented with lighter markers and lines in this Figure. The activation energy (E), according to the Arrhenius equation, was calculated for each model using slope of the lines.
Figure 10
Obtaining the activation energy of the carbon monoxide reduction of iron concentrate by the represented models in Table 1.

Figure 11 shows the calculated activation energy of the models and compares them through vertical bars. The width of the bars in Figure 11 has been calculated using the represented values of R² in Figure 9 a–e. i.e. degree of fitness of the models with the experimental data can be observed through width magnitude of the bars. This Figure reveals that some models, such as first-order reaction model or diffusion models, are able to describe the reaction. Therefore, a combination of the acceptable mechanisms controls this reaction.
Figure 11
Calculated activation energy of the carbon monoxide reduction of iron concentrate by the represented models in Table 1.
3.5 Reaction kinetics parameters

Number	Model	f(x)	g(x)
Nucleation models
1	Power law
2	Power law
3	Power law
4	Power law
5	Avrami-Erofe'ev
6	Avrami-Erofe'ev
7	Avrami-Erofe'ev
8	Avrami-Erofe'ev
Geometrical contraction models
9	Contracting area
10	Contracting volume
Diffusion models
11	1D diffusion
12	2D diffusion
13	3D diffusion
14	Ginstling-Brounshtein
Reaction-order models
15	Zero-order
16	First-order
17	Second-order
18	Third-order

A common model for the gaseous reduction of solid particles is nucleation and growth model ( ) (Nakagawa et al. 2005; Chen et al. 2015a, 2015b). Suitable fitting of the model equation with the experimental data gives the Avrami parameter (n) and simultaneously the apparent reaction rate constant. This equation has a suitable fitting with the obtained experimental data by the TGA. The apparent reaction rate constant is an exponential function of temperature (the Arrhenius equation) and a power function of particles size (d_P) and CO concentration (C_CO). Therefore, isothermal integral form of equation (1) can be represented as the following equation:

(2) Figure 12 shows the fitting of logarithmic form of this equation with the experimental data in order to obtain the apparent reaction rate constants at different particle size of the ore concentrate. All the experiments in Figure 12 have been carried out at the same temperature and CO concentration. The slope and intercept of the lines give n and k values for this reaction, respectively. Figure 13 shows the dependency of these k values to d_p. Based on this fitting, the value of α is equal to −1.871.
Figure 12
Determination of Avrami parameter and also apparent reaction rate constant of reduction of different particles sizes of iron ore concentrate by 8.518 mol.m⁻³ carbon monoxide at 1163 K (890°C).

Figure 13
Dependency of the apparent rate constant on particle size of the iron concentrate.

Figure 14 shows the accordance of logarithm form of equation (2) with the experimental data for obtaining k at different concentrations of CO. The experiments in Figure 14 have been carried out with a same particle size and temperature. Figure 15 shows dependency of the apparent reaction rate constant to CO concentration. It is obtained that β is equal to 2.738 for CO reduction of iron concentrate.
Figure 14
Determination of the apparent reaction rate constant of iron concentrate with carbon monoxide in different concentration at 1163 K (890°C).

Figure 15
Dependency of apparent reaction rate constant on the concentration of the reducing gases.

Now, it is possible to obtain intrinsic reaction rate constant. This quantity is just a function of temperature. Figure 16 shows the accordance of equation (2) with the experimental data for obtaining the apparent reaction rate constant at different temperatures. All the experiments in Figure 16 have been carried out at a same particle size and CO concentration. It is seen in this Figure that the values of Avrami parameter is about one for all of the experiments. This approves the suitable accordance of one-order reaction model with the experimental results of this reaction. Figure 17 shows dependency of the intrinsic reaction rate constant on temperature. This Figure represents the activation energy and also pre-exponential factor of this reaction based on the Arrhenius equation.
Figure 16
Determination of n and k by fitting the logarithm form of equation (2) with the experimental data.

Figure 17
Determination of E and k₀ of the iron concentrate reduction by carbon monoxide.
3.6 Evaluation of the results

The calculated activation energy of the reaction using nucleation and growth model (Figure 17) and some fitting models (Figure 11) have reasonable agreements with each other. Activation Energy of the similar studies for reduction of iron oxides by CO has been represented in Table 2. As can be observed from the Table, reported values are very different. The obtained value in this research is very close to the value relevant to in-flight reduction of the haematite ore concentrate.

Table 2
The previous studies relevant to iron oxides reduction by carbon monoxide.

T K (°C)	E (kJ.mol⁻¹)	Method	References
1073–1173 K (800–900°C)	113.8	TGA (Disk)	Trushenski et al. (1974)
623–673 K (350–400°C)	70	TGA (Particles)	Jozwiak et al. (2007)
1073–1173 K (800–900°C)	9.97	TGA (Particles)	Mondal et al. (2004)
1073–1223 K (800–950°C)	14.6	TGA (Particles)	Hughes et al. (1982)
1423–1623 K (1150–1350°C)	231	In-flight (Particles)	Chen et al. (2015b)
1032–1073 K (750–800°C)	114.7	TGA (Pellet)	Qiyuan et al. (1998)
1073–1173 K (800–900°C)	73.6	TGA (Disk)	Trushenski et al. (1974)
673–873 K (400–600°C)	55	TGA (Particles)	Jozwiak et al. (2007)
1073–1173 K (800–900°C)	69.4	TGA (Disk)	Trushenski et al. (1974)
1173–1373 K (900–1100°C)	47.6	TGA (Sintered particles)	Noguchi et al. (2013)
1073–1173 K (800–900°C)	14.13	TGA (Particles)	Mondal et al. (2004)
673–773 K (400–500°C)	104	TGA (Particles)	Jozwiak et al. (2007)

^aReactions were not balanced, stoichiometrically.

Some of the reduced solid samples in the TG experiments were analyzed by titration for comparing their progress of reaction (X) and the amount of oxygen taken from them (Y) with the percent of their total iron (Fe_total) and metal iron (Metallization). X and Y have been calculated using their weight changes in TGA and Fe_total and Metallization have been obtained using titration. Table 3 represents result of this comparison and also effect of cooling gas conditions, which surrounded the solid sample after each TGA experiment, where w_Oxygen is the weight of oxygen content of the solid sample. The first row of this Table presents the total iron and sulphur in the studied concentrate. The low sulphur content is due to the burning by air in the cooling time (see the cases that were cooled in the air). The low amount of sulphur in the reduced sample shows the high oxidation of it during cooling time in the air. This is a reason of existing differences between the calculated percent of removed oxygen from the sample (by TGA) and produced metal iron (by titration).

Table 3

Comparing the results of TGA and wet chemical analysis of the reduced solid samples.

Temperature K (°C)	Reducing gas	Cooling gas	Reduction time (sec)			Fe_total (%)	Metalization (%)	Sulphur (%)
–	–	–	–	–	–	66	–	1.27
1162 K (889°C)	CO	Air	1932	94	59	77.2	50.1	0.746
1163 K (890°C)	CO	CO	6989	99.1	90.6	82.69	90.98	0.983
1164 K (891°C)	CO	Air	2623	92.1	79.8	78.17	42.21	0.182
1348 K (1075°C)	CO	Air	2099	94.8	93	82.3	54.06	0.028

Figure 18 compares the predicted weight loss of the iron concentrate by nucleation and growth model and the recorded weight loss by TGA (R² = 0.976). This Figure confirms accuracy of the applied model.

Figure 18

Comparing model predictions with the experimental data for reduction of iron ore concentrate by 4.571 Nlit.min⁻¹ carbon monoxide at 1163 K (890°C).

4 Conclusion

In this study, the reaction kinetics of carbon monoxide reduction of iron ore concentrate particles in the temperature range of 973–1173 K (700–900°C) was investigated using the TGA. The fitting method and nucleation and growth model were used to obtain the governing mechanism and reaction rate constant of this reaction. It was deduced that the sintering phenomenon affects the reduction of the concentrate at temperatures more than 1150 K (900°C). The dependency of the reduction rate on the reducible solid particles was obtained first order. The dependency on the CO concentration and also particle size were obtained as power functions with 2.738 and −1.871 powers, respectively. The activation energy was obtained about 243.8 kJ.mol⁻¹ and finally, the following equation was obtained to predict progress of the reduction reaction: (3)

Footnotes

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes on contributors

Bahador Abolpour was born in Shiraz, Iran, in 1985. He received three BS degrees in Chemical Engineering, Aerospace Engineering and Computer Engineering from 2004 to 2016. He also received his MS and PhD degrees in Chemical Engineering from the Shahid Bahonar University of Kerman in 2011 and 2016, respectively. Gas-Solid reaction is the main field of his studies.

M. Mehdi Afsahi is an assistant professor of Chemical Engineering at Kerman University, Iran. He received his BS and MS degrees in Chemical Engineering from Tehran and Modares Universities in 1988 and 1992. He received a PhD degree in the field of non-catalytic gas–solid reactions from Amirkabir technical University in 2008. Since graduated in 2008 he has been concerned primarily with gas–solid processes in Chemical and Metallurgical industries.

Mehdi Azizkarimi was born in Rafsanjan, Iran, in 1981. He received the MS degree in extractive metallurgy from the Amirkabir University of Technology, Iran, in 2009. In 2010, he joined the Golgohar Iron ore and Steel Research Institute (GISRI), as a senior researcher in the field of pelletizing, ironmaking and steelmaking. His current research interests include kinetic of gas-solid reaction, new technology of ironmaking, oxidation–reduction process and pyrometallurgy.

References

Antti Roine. 2006. HSC Chemistry 6.0, Chemical reaction and equilibrium software with extensive termochemical database and flowsheet simulation, Outokumpu Research Oy Information Center.

Brown

, Dollimore

, Galwey

AK.

1980. Comprehensive chemical kinetics. Amsterdam: Elsevier.

Chen

, Mohassab

, Jiang

, Sohn

HY.

2015a. Hydrogen reduction kinetics of hematite concentrate particles relevant to a novel flash ironmaking process. Metall Mater Trans B. 46:1133–1145. doi: 10.1007/s11663-015-0332-z

Chen

, Mohassab

, Zhang

, Sohn

HY.

2015b. Kinetics of the reduction of hematite concentrate particles by carbon monoxide relevant to a novel flash ironmaking process. Metall Mater Trans B. 46(4):1716–1728. doi: 10.1007/s11663-015-0345-7

El-Rahaiby

, Rao

YK.

1979. The kinetics of reduction of iron oxides at moderate temperatures. Metall Trans B. 10:258–264. doi: 10.1007/BF02652470

Geva

, Farren

, John

, Hayes

PC.

1990. The effects of impurity elements on the reduction of wustite and magnetite to iron in CO/CO₂ and H₂/H₂O gas mixtures. Metall Trans B. 21:743–751. doi: 10.1007/BF02654253

Hughes

, Kam

EKT

, Mogadamzadeh

1982. The Reduction of iron ores by hydrogen and carbon monoxide and their mixtures. Thermochim Acta. 59:361–377. doi: 10.1016/0040-6031(82)87158-X

Jozwiak

, Kaczmarek

, Maniecki

, Ignaczak

, Maniukiewicz

2007. Reduction behavior of iron oxides in hydrogen and carbon monoxide atmospheres. Appl Catal A. 326:17–27. doi: 10.1016/j.apcata.2007.03.021

Kang

, Chung

, Murayama

, Ono

1998. Effect of iron ore shape on gaseous reduction rate. ISIJ Int. 38(11):1194–1200. doi: 10.2355/isijinternational.38.1194

10.

Kawasaki

, Sanscrainte

, Walsh

TJ.

1962. Kinetics of reduction of iron oxide with carbon monoxide and hydrogen. AICHE J. 8(1):48–52. doi: 10.1002/aic.690080114

11.

Mondal

, Lorethova

, Hippo

, Wiltowski

, Lalvani

SB.

2004. Reduction of iron oxide in carbon monoxide atmosphere – reaction controlled kinetics. Fuel Proc Tech. 86:33–47. doi: 10.1016/j.fuproc.2003.12.009

12.

Nakagawa

, Maeda

, Nishioka

, Shimizu

2005. Ironmaking-rapid reduction of fine iron ore transported with CH₄ gas. Tetsu-to-hagane. 91(6):521–527. doi: 10.2355/tetsutohagane1955.91.6_521

13.

Noguchi

, Ohno

, Maeda

, Nishioka

, Shimizu

2013. Effect of CO gas concentration on reduction rate of major mineral phase in sintered iron ore. ISIJ Int. 53(4):570–575. doi: 10.2355/isijinternational.53.570

14.

Pourghahramani

, Forssberg

2007. Reduction kinetics of mechanically activated hematite concentrate with hydrogen gas using nonisothermal methods. Thermochim Acta. 454(2):69–77. doi: 10.1016/j.tca.2006.12.023

15.

Qiyuan

, Hong

, Pingmin

, Guanzhou

, Jing

1998. Isothermal kinetics of the first stage magnetite cold bond pellets with carbon monoxide. J Cent Univ Technol. 5(1):1–3. doi: 10.1007/s11771-998-0020-0

16.

Silbey

, Alberty

, Bawendi

MG.

2004. Physical chemistry. New York: John Wiley and Sons.

17.

Takeuchi

, Nomura

, Ohono

, Maeda

, Nishioka

, Shimizu

2007. Kinetic analysis of spherical wustite reduction transported with CH₄ gas. ISIJ Int. 47(3):386–391. doi: 10.2355/isijinternational.47.386

18.

Trushenski

, Li

, Philbrook

WO.

1974. Non-topochemical reduction of iron oxides. Metall Trans. 5(5):1149–1158. doi: 10.1007/BF02644326

Reduction kinetics of magnetite concentrate particles by carbon monoxide

Abstract

Keywords

1 Introduction

2 Experiments

2.1 Materials

2.2 Apparatus and procedure

3 Results and discussions

3.1 Repeatability of the experiments

3.4 Reaction mechanism

Table 2 The previous studies relevant to iron oxides reduction by carbon monoxide.

Footnotes

References

Table 2
The previous studies relevant to iron oxides reduction by carbon monoxide.