Effect of Temperature on Reaction and Degradation Behaviors during CO₂ and H₂O Gasification Reactions of Coke in Same Conversion Ratio

Abstract

The efficiency of blast furnaces is adversely affected by coke degradation via gasification. Considering the utilization of hydrogen-enriched blast furnaces, it is essential to investigate the reaction and degradation behaviors of coke at different temperatures. In this study, coke gasification experiments were conducted under CO₂ and H₂O atmospheres at different temperatures to prepare cokes with a conversion ratio of 0.2. The reaction rate of the H₂O gasification reaction was higher than that of the CO₂ gasification reaction at the same temperature. The activation energies for CO₂ and H₂O gasification were 150.2 and 126.0 kJ/mol, respectively. After gasification, the shrinkage ratio was low by H₂O gasification at 1273 K and increased with increasing temperature, indicating that the surface reaction became the control factor that consumed the coke matrix with increasing temperature. On the other hand, the shrinkage ratio by CO₂ gasification tended to be stable from 1273 to 1673 K. Furthermore, the increase in the porosity of coke by H₂O gasification was lower than that by CO₂ gasification at higher temperatures. In addition, the strength of the coke via H₂O gasification was higher than that of the coke via CO₂ gasification.

1. Introduction

CO₂ emissions corresponding to iron and steel manufacturing in Japan were approximately 145 million tons in 2021, accounting for 12% of the country’s total emissions in that year.¹⁾ Therefore, iron and steel manufacturing industries have received significant attention owing to their large-scale and substantial emissions, as global warming has recently become one of the most critical issues. As the blast furnace is the dominant process for producing iron and steel, it is imperative for the iron and steel industry to seek ways to achieve highly efficient operation and reduce CO₂ emissions from blast furnaces.^2,3,4,5) Coke is one of the most important materials for ironmaking process and serves multiple essential functions. It supplies the critical carbon source necessary for the reduction reactions within the blast furnace during the ironmaking process. It also acts as a source of heat that facilitates the necessary temperatures for the reduction of iron ore to iron and enhances ventilation permeability within the furnace.⁶⁾ Furthermore, coke acts as the structural skeleton supporting the charge in the furnace, which is a key reason why it has not been replaced in blast furnace operation.⁷⁾ Coke quality is intrinsically related to its structure, as the strength mainly correlates with its pore characteristics, which tend to diminish as porosity increases.^8,9,10,11) Additionally, as the reaction progresses and coke matrices are consumed, the solution loss reaction significantly affects strength.^12,13) Adequate ventilation ensures that gases can move freely through the packed bed of materials within the furnace, thereby allowing for uniform heating and efficient reactions. Poor ventilation permeability can lead to operational inefficiencies and instabilities in the furnaces. One of the primary causes of poor ventilation permeability is resistance loss in the softened cohesive zone, which accounts for approximately 85% of the total resistance loss within the furnace.¹⁴⁾ The cohesive zone represents the region where the iron ore softens and begins to melt, resulting in an area of high resistance that restricts gas flow, typically within the temperature range of 1373–1623 K.¹⁵⁾ In traditional blast furnaces, coke undergoes a series of complex reactions, such as the solution-loss reaction, leading to its degradation. The gasification reaction has an important effect when the reaction rates are low, thereby allowing the gasification agents (CO₂/H₂O/CO/H₂) to penetrate deeper into the internal pore structure of the lump coke.⁶⁾ In traditional blast furnaces, coke reacts with CO₂ via a CO₂ gasification reaction, leading to its degradation. Kim et al.¹⁶⁾ concluded that CO₂ gasification reaction rate is influenced by different temperature.

In response to a low-carbon society, a series of plans have been proposed for the use of hydrogen in the iron and steel industry. To achieve this goal, research has been undertaken to develop hydrogen-enriched blast furnace technologies aimed at reducing carbon emissions and energy consumption.^17,18,19,20) These blast furnaces inject hydrogen gas through tuyeres, partially replacing coke to reduce CO₂ emissions and energy consumption. In a hydrogen-enriched blast furnace, the injection of hydrogen generates large quantities of water vapor. This water vapor reacts with coke through the H₂O gasification reaction, which also affects the structure and degradation behavior of coke in different temperature ranges in hydrogen-enriched blast furnaces. Therefore, elucidating the reaction mechanisms and kinetics of coke at different temperatures in hydrogen-enriched blast furnaces is essential. To elucidate the reaction mechanism and kinetics of coke, previous studies^{21,22,23,24,25)} have focused on changes in coke after CO₂ gasification at 1173–1873 K and concluded that the change in strength after the CO₂ gasification reaction is caused by changes in the composition and pores of the texture, which reduce the mechanical strength of coke. Takatani et al.²⁶⁾ investigated the kinetic parameters and the gasification reaction characteristics of lump coke using the kinetic parameters obtained. Ono et al.²⁷⁾ investigated the strength of coke after gasification at 1373 K and found that coke was damaged over a wide range of temperatures after CO₂ gasification and was locally damaged after H₂O gasification. Shin et al.²⁸⁾ investigated the porosity and macrostrength from 1373 to 1773 K and observed that H₂O showed a tendency to react with coke at the surface. Chang et al.²⁹⁾ and Wang et al.³⁰⁾ analyzed the effects of CO₂ and H₂O on the gasification dissolution reaction from 1223 K to 1423 K, indicating that water vapor is more likely to diffuse into the coke interior at lower temperatures. Higher temperatures can reduce the extent of the solution loss within the coke interior, particularly in reactions with H₂O. However, research on the degradation behavior of coke has primarily focused on the effects of CO₂ gasification, with few studies examining the influence of H₂O gasification. Most of these studies have compared the reaction mechanism and kinetics of coke for the same reaction time and different conversion ratios for CO₂ and H₂O gasification; however, the initial reaction stage during CO₂ and H₂O gasification at the same conversion ratio remains ambiguous.

In this study, we considered the reaction and degradation behaviors of coke lumps at the initial reaction stage during CO₂ and H₂O gasification at different temperatures and a conversion ratio of 0.2. Gasification experiments using a single coke lump were conducted in a CO₂ or H₂O gas atmosphere at different temperatures in a batch-type vertical tube furnace at a conversion ratio of 0.2. Reaction rate analysis was performed at different temperatures for each gasification reaction. In contrast to previous studies that mostly selected partial images of the cross section for comparison when analyzing porosity, this study analyzed the overall cross section by a photomontage technique (combining multiple backscattered electron (BSE) images by gluing, rearranging, and overlapping them into a complete image of the overall cross section in coke) using scanning electron microscopy (SEM). The effects of atmospheric gas and temperature on the reaction and pore structure of coke lumps with the same conversion ratio were investigated in detail.

2. Gasification Experimental Methods

2.1. Experimental Sample

The coke samples used in this study had a diameter of 10–30 mm (average diameter of the coke was approximately 15 mm). Table 1 lists the quality indices of the samples, and Table 2 lists their properties.

Table 1. Quality index of coke sample.

	Drum Index [%]	Coke Reactivity Index [%]	Coke Strength after Reaction [%]
Coke	85.4	20.8	68.0

Table 2. Properties of coke sample.

Proximate analysis [wt%, dry basis]
	Ash	Volatile matter	Fixed carbon
Coke	11.81	0.86	87.33

2.2. Experimental Methods

The gasification experiments were conducted using a batch-type vertical tube furnace, as shown in Fig. 1. The sample was placed in a Pt basket and hung on a Pt wire connected to an electronic balance. After reaching the predetermined temperature, the furnace was raised and the sample was held in the center of the alumina tube (ID: 60 mm). The heating rate of coke was approximately 5 K/s. To remove volatile matter, the sample was held at a predetermined temperature under an N₂ gas atmosphere for 10 min, with N₂ gas introduced into the reaction tube. The atmosphere was then changed to an N₂–CO₂ or N₂–H₂O gas mixture. The weight losses of the samples were continuously measured using an electronic balance. When weight loss reached the target conversion ratio, the furnace was refilled with N₂ gas. The samples were then placed in water-cooling jacket. Table 3 lists the experimental conditions. Each experiment was performed more than twice to reduce the effect of sample differences.

Fig. 1. Schematic of batch-type vertical tube furnace. (Online version in color.)

Table 3. Experimental conditions.

Conditions	CO2 gasification	H2O gasification
Atmospheres [vol%]	N2:CO2 = 85:15	N2:H2O = 85:15
Gas flow rate [L/min]	4
Temperature [K]	1273, 1473, 1673
Conversion ratio (X) [-]	0.2

To evaluate the degradation behavior of the samples, an I-type tumbler was used to measure the coke strength before and after the gasification experiment.

To track the progress of the CO₂ or H₂O gasification reactions in the coke samples, conversion ratio (X) was used, as described in Eq. (1).

  

$$X=\frac{W_0^{\prime }-W_{exp}}{W_0\times FC}$$(1)

where X [-] is the conversion ratio, W′₀ [g] is the weight of the coke sample after the VM removal in the gasification experiment, W_exp [g] is the weight of the coke sample after the gasification experiment, and W₀ [g] is the initial weight of the coke sample, FC [-] is the fixed carbon ratio of the coke sample.

The shrinkage ratio [%] of the coke samples was calculated using Eq. (2).

  

$$Shrinkage\mathrm{\hspace{0.17em} } ratio=\frac{d_0-d}{d_0}\times 100$$(2)

where d₀ [mm] is the initial average diameter of the sample and d [mm] is the average diameter of the coke sample after the gasification experiment. The coke diameter was measured at three locations using a Vernier caliper to calculate the average diameter.

The kinetics of the reaction between gas and solid can be described by the following equation:³¹⁾

  

$$\frac{dX}{dt}=k(P_g,T)f(X)$$(3)

where k represents the reaction rate constant [1/s], which is determined by the partial pressure of the gasifying agents and the temperature T [K]. f(X) is a kinetic function describing the evolution of the coke sample properties as the reaction proceeds. In this study, the partial pressure (P_g) of the gasifying agent was maintained constant throughout the gasification process. Therefore, k(P_g, T) can be simplified as k(T), which can be described by the Arrhenius equation.

  

$$k(T)=Aexp\left(-\frac{E}{RT}\right)$$(4)

where A is the pre-exponential factor [s⁻¹], E is the activation energy [kJ/mol], R is the universal gas constant [8.3145 kJ/(mol·K)].

To observe the initial stage of the reaction, X = 0.2 of coke conversion was used to determine the kinetic parameters. Taking the logarithm of both sides of Eq. (4) yielded a linear equation, allowing the activation energy E to be determined without specifying the reaction model f(X).

  

$$ln\left(\frac{dX}{dt}\right)=ln\left(Af(0.2)\right)-\frac{E}{R}\frac{1}{T}$$(5)

2.3. Image Analysis Method

To investigate the change in the porosity of coke by gasification, overall cross-sectional images of the coke samples were analyzed. The coke samples were embedded in resin and vacuum-sealed to remove air. Once the resin hardened sufficiently over a day, it was removed from the mold, and a cross-sectional sample was prepared. Because non-conductive samples, such as resin, often experience charge-up and abnormal contrast when observed under SEM, the observation surface was coated with carbon using an auto carbon coater (JEOL: JEC-560) to prevent this problem. The photomontages of the overall cross section are shown in Fig. 2. Around 150–250 pieces BSE images at 100× magnification were combined to obtain the overall cross-sectional images (the number of images varied with the size of the coke sample). In the BSE image of the overall cross section, the white, light-gray, and deep-gray areas represent the ash particles, coke matrix, and pores in the coke, respectively. To distinguish the pores in the coke sample, we binarized the photomontages using particle analysis software (ImageJ), as shown in Fig. 2.

Fig. 2. Photomontages and binarized images of pore at 1673 K. (Online version in color.)

The porosity [%] of the overall cross section was calculated using the following equation:

  

$$Porosity=\frac{A_p}{A_p+A_{a,c}}\times 100$$(6)

where A_p is the area of the pores in the overall cross section of the coke sample [mm²], and A_a,c is the area of the ash and coke matrix in the overall cross section of the coke sample [mm²].

2.4. Coke Strength Measurement

To investigate the degradation behavior of coke during gasification, strength tests were conducted on the coke samples before and after gasification. Because the coke samples used in this study weighed only 1–2 g, it was difficult to perform general industrial strength tests. Therefore, the coke strength of single particles was measured using an I-type tumbler tester,³²⁾ shown in Fig. 3, to evaluate the coke degradation. In this experiment, the rotational speed was maintained at 30 rpm, samples were taken every 5 min, and sieved through a 1-mm mesh sieve. The weight of each sample (> 1 mm) was measured on a sieve. The experiments were conducted for 25 min. The tumbler strength index (TI) of the coke samples were calculated using Eq. (7).

  

$$TI=\frac{W_{+1mm}}{W_0}\times 100$$(7)

where TI [%] is the tumbler strength index, W₀ [g] is the initial weight of the sample, and W_+1mm [g] is the weight of the lump larger than 1 mm on the sieve after 25 min.

Fig. 3. Schematic of the I-type tumbler tester. (Online version in color.)

3. Results and Discussion

3.1. Reaction Behaviors during CO₂ and H₂O Gasification at Each Temperature

The curves of the reaction time to 1−X = 0.8 for the gasification experiments in the temperature range of 1273–1673 K are depicted in Fig. 4. H₂O gasification reaction was faster than the CO₂ gasification reaction at the same temperature. In CO₂ gasification, the reaction time to reach the target conversion ratio (X = 0.2) at 1273 K was 6.4 times longer than that at 1473 K and 34.0 times longer than that at 1673 K. In H₂O gasification, on the other hand, the reaction time to reach the target conversion ratio (X = 0.2) at 1273 K was still 6.4 times longer than that at 1473 K, but it was only 20.7 times longer than that at 1673 K. This result shows that the influence of temperature on the reaction time for CO₂ gasification is greater than that for H₂O gasification.²²⁾

Fig. 4. Curves of non- conversion ratio (1−X) for CO₂ (a) and H₂O (b) gasification experiments in the different temperature range. (Online version in color.)

The relationship between shrinkage ratio and temperature during CO₂ and H₂O gasification is shown in Fig. 5. During CO₂ gasification, the shrinkage ratio of the coke changed slightly from 1273 K to 1673 K. In H₂O gasification, on the other hand, the shrinkage ratio of coke increased by the higher temperature, indicating that the tendency of surface reaction of coke became stronger with the increase in temperature. Therefore, it is assumed that the volumetric reaction becomes the control factor that consumes the coke matrix during CO₂ gasification from 1273 to 1673 K.⁷⁾ In H₂O gasification, the volumetric reaction is the control factor that consumes the coke matrix at low temperatures; however, the surface reaction becomes the control factor at higher temperatures.^20,25)

Fig. 5. Relationship between shrinkage ratio and temperatures in CO₂ and H₂O gasification in temperature range of 1273–1673 K. (Online version in color.)

3.2. Reaction Rate of the CO₂ and H₂O Gasification

Figure 6 shows the relationship between the gasification rate (dX/dt) and the non-conversion ratio (1−X) for CO₂ and H₂O gasification in the temperature range of 1273–1673 K. The CO₂ gasification rates at 1473 and 1673 K were 6.4 and 33.0 times than those at 1273 K, respectively. The H₂O gasification rates at 1473 and 1673 K were 6.4 and 20.7 times that those at 1273 K, respectively. The gasification reaction rate increases with increasing temperature.

Fig. 6. Relationship between the gasification rate (dX/dt) and non-conversion ratio (1−X) for CO₂ (a) and H₂O (b) gasification in temperature range of 1273–1673 K. (Online version in color.)

Figure 7 shows the Arrhenius plot in the temperature range of 1273–1673 K using the gasification rates at 1−X = 0.8 by applying Eq. (3). Table 4 lists the activation energies E calculated using Eq. (5) and the coefficient of determination, R². The activation energy E for CO₂ and H₂O gasification in the range of 1173 to 1573 K were 158.8–270.1 and 121.8–205.9 kJ/mol, respectively.^19,28) The corresponding E for CO₂ and H₂O gasification were 150.2 and 126.0 kJ/mol. These values are consistent with those reported in literature, thereby validating the reliability of our experimental results. These results indicated that the temperature dependence of the reaction rate for H₂O gasification was smaller than that for CO₂ gasification.

Fig. 7. Arrhenius plot in temperature range of 1273–1673 K. (Online version in color.)

Table 4. The activation energy E and coefficient of determination R².

Atmosphere	E [kJ/mol]	R2
CO2	150.2	0.9966
H2O	126.0	0.9934

3.3. Change in Porosity in Coke Sample after CO₂ and H₂O Gasifications

Figure 8 shows the porosity of the raw coke sample and coke samples (X = 0.2) after CO₂ and H₂O gasification in the temperature range of 1273–1673 K. In existing evaluation methods for calculating porosity,^9,24,33) the porosity of raw coke was about 50%–60%. In this study, the porosity of the area ratio inside the overall cross section of coke was calculated; therefore, the porosity values were different from the porosity of the volume ratio results from methods such as mercury porosimetry. The raw coke sample used in this study had a porosity of 36.0%, which is consistent with the porosity values reported by Shin et al.²⁸⁾ Therefore, the results of this experiment are still within a reasonable range. Compared with the raw coke, the overall cross-sectional porosity of coke after gasification increased in the temperature range of 1273–1673 K. This was because the carbonaceous matter inside the coke was consumed by gasification. At 1273 K, the porosities of the CO₂ and H₂O gasification were similar. This result indicates that CO₂ and H₂O gasification proceeded as volumetric reaction models at 1273 K. The porosity after the CO₂ gasification reaction decreased slightly with increasing temperature. On the other hand, the porosity after the H₂O gasification reaction decreased significantly with increasing temperature. It is assumed that this significant decrease in porosity at high temperatures during H₂O gasification was due to the consumption of carbonaceous matrix on the coke surface during gasification. Therefore, H₂O gasification proceeded as a surface reaction model at high temperatures. This is consistent with the results of shrinkage ratio.

Fig. 8. The porosity of the raw coke and cokes (X = 0.2) after CO₂ and H₂O gasification in temperature range of 1273–1673 K. (Online version in color.)

3.4. Change in Strength in Coke Sample by CO₂ and H₂O Gasification

Figure 9 shows the tumbler strength index (TI) of the raw coke sample and coke samples (X = 0.2) after the gasification experiments in the temperature range of 1273–1673 K. Compared with the TI of raw coke (92.8%), the TI of the coke samples decreased after gasification. This result demonstrates that the degradation behavior is related to the porosity of the coke.⁹⁾ The porosity of the coke increases after gasification, leading to a reduction in coke strength. Because the CO₂ gasification reaction proceeded as a volumetric reaction model in the temperature range of 1273–1673 K, the porosity of the coke was similar after CO₂ gasification. Therefore, the tendency of TI reduction during 5–25 min was almost the same from 1273 to 1673 K. For H₂O gasification, the tendency of TI reduction during 5–25 min was almost the same as that of CO₂ gasification because of the volumetric reaction model at 1273 K. However, the decrease rate of TI during 10–25 min at 1473–1673 K was slower than that at 1273 K. It is supposed that the surface reaction model of H₂O gasification of coke at higher temperatures leads to this result, and the porosity of the inner part of the coke was close to that of raw coke because the carbon matrix was almost not consumed in the inner part. Therefore, the decrease rate of TI decreased at 1473–1673 K. Particularly at 1673 K, the decrease rate of TI in 10–25 min was close to that of raw coke. Therefore, the overall TI of coke after H₂O gasification was higher than that after CO₂ gasification in the high-temperature range because of the surface reaction model, as shown in Fig. 10.²⁰⁾ From these results, it can be inferred that the degradation behavior of coke in a hydrogen-enriched blast furnace is improved compared to that in a conventional blast furnace in the higher temperature range.

Fig. 9. TI of coke samples (X = 0.2) under different gasification atmospheres in temperature range of 1273–1673 K. (Online version in color.)

Fig. 10. Relationship between the decrease in TI and temperature. (Online version in color.)

4. Conclusions

In this study, the reaction and degradation behaviors of coke during the initial reaction stages of CO₂ and H₂O gasification at different temperatures were investigated. Coke lumps with a conversion ratio (X) of 0.2 were produced in CO₂ and H₂O atmospheres. The porosity and tumbler strength index (TI) of the coke after the gasification experiments were measured using SEM and an I-type tumbler tester. The following results were obtained:

(1) The rate of the H₂O gasification reaction was higher than that of the CO₂ gasification reaction at the same temperature. The influence of temperature on the reaction time for CO₂ gasification was greater than that for H₂O gasification at 1273–1673 K.

(2) The shrinkage ratio of coke changed slightly from 1273 to 1673 K during CO₂ gasification. On the other hand, for H₂O gasification, the shrinkage ratio of coke increased with increasing temperature, indicating that the surface reaction of coke became stronger with increasing temperature.

(3) The activation energy E for CO₂ and H₂O gasification were 150.2 and 126.0 kJ/mol, respectively. This result indicated that the temperature dependence of the reaction rate for H₂O gasification was smaller than that for CO₂ gasification.

(4) The porosity of coke after CO₂ and H₂O gasification was similar at 1273 K. With increasing temperature, the porosity after the CO₂ gasification reaction decreased slightly; however, the porosity after H₂O gasification reaction decreased significantly.

(5) The degradation behavior was related to the porosity of the coke. Therefore, the TI values of the coke samples decreased after gasification. Moreover, the TI of coke after H₂O gasification was higher than that after CO₂ gasification around the higher temperature range because the carbonaceous matrix on the surface of the coke was mainly consumed during H₂O gasification.

Statement for Conflict of Interest

The authors have no conflicts of interest directly relevant to the content of this article.

Acknowledgements

This work was partially performed by the Research Group for Lumpy Zone Control for Next Generation Hydrogen-Enriched Blast Furnace in The Iron and Steel Institute of Japan (the chief examiner is Prof. K. Ohno from Kyushu University). The authors would like to acknowledge the contributions of all research group members. Part of this study was financially supported by JST SPRING (grant number JPMJSP2125). The authors would like to take this opportunity to thank the “THERS Make New Standards Program for the Next-Generation Researchers”.

References

• 1)  National Institute for Environmental Studies: Japan’s Greenhouse Gas Emissions Data, (Final figures for 2021), (April 21, 2023), https://www.nies.go.jp/gio/archive/ghgdata/index.html (in Japanese), (accessed 2024-10-03).
• 2)   S.  Watakabe,  K.  Miyagawa,  S.  Matsuzaki,  T.  Inada,  Y.  Tomita,  K.  Saito,  M.  Osame,  P.  Sikström,  L. S.  Ökvist and  J.-O.  Wikstrom: ISIJ Int., 53 (2013), 206. https://doi.org/10.2355/isijinternational.53.2065
• 3)   S. K.  Sibagatullin,  A. S.  Kharchenko,  G. Yu.  Beginyuk and  V. A.  Beginyuk: CIS Iron and Steel Review, 16 (2018), 11. https://doi.org/10.17580/cisisr.2018.02.02
• 4)   K.  Nishioka,  Y.  Ujisawa and  T.  Inada: Tetsu-to-Hagané, 100 (2014), 127077 (in Japanese). https://doi.org/10.2355/tetsutohagane.100.1347
• 5)   R.  Higashi,  D.  Maruoka and  T.  Murakami: ISIJ Int., 63 (2023), 1972. https://doi.org/10.2355/isijinternational.ISIJINT-2023-329
• 6)   Y.  Ono,  Y.  Fukuda,  Y.  Sumitani,  Y.  Saito,  Y.  Matsushita,  H.  Aoki,  T.  Shishido and  N.  Okuyama: The 29th Annual Conference of the Japan Institute of Energy, Toyama, (2020), 6 (in Japanese). https://doi.org/10.20550/jietaikaiyoushi.29.0_6
• 7)   M.  Sun,  J.  Zhang,  K.  Li,  K.  Guo,  Z.  Wang and  C.  Jiang: Int J Miner Metall Mater., 26 (2019), 1247. https://doi.org/10.1007/s12613-019-1846-1
• 8)   Y.  Numazawa,  Y.  Matsushita,  H.  Aoki and  A.  Komiya: ISIJ Int., 60 (2020), 2686. https://doi.org/10.2355/isijinternational.ISIJINT-2020-198
• 9)   Y.  Sumitani,  Y.  Ono,  Y.  Saito,  Y.  Matsushita,  H.  Aoki,  T.  Shishido and  N.  Okuyama: ISIJ Int., 61 (2021), 119. https://doi.org/10.2355/isijinternational.ISIJINT-2020-292
• 10)   S.  Pusz,  M.  Krzesińska,  L.  Smędowski,  J.  Majewska,  B.  Pilawa and  B.  Kwiecińska: Int. J. Coal Geol., 81 (2010), 287. https://doi.org/10.1016/j.coal.2009.07.013
• 11)   Z.  Zheng,  Y.  Ueki,  R.  Yoshiie and  I.  Naruse: ISIJ Int., 63 (2023), 1810. https://doi.org/10.2355/isijinternational.ISIJINT-2023-207
• 12)   Y.  Ueki,  K.  Teshima,  R.  Yoshiie and  I.  Naruse: ISIJ Int., 60 (2020), 1427. https://doi.org/10.2355/isijinternational.ISIJINT-2019-692
• 13)   Y.  Numazawa,  D.  Igawa,  S.  Matsuo,  Y.  Saito,  Y.  Matsushita,  H.  Aoki,  T.  Shishido and  N.  Okuyama: ISIJ Int., 58 (2018), 1420. https://doi.org/10.2355/isijinternational.ISIJINT-2018-144
• 14)   C.  Lan,  S.  Zhang,  X.  Liu,  Q.  Lyu and  M.  Jiang: Int. J. Hydrog. Energy., 45 (2020), 14255. https://doi.org/10.1016/j.ijhydene.2020.03.143
• 15)   X.  Xing,  H.  Rogers,  G.  Zhang,  K.  Hockings,  P.  Zulli and  O.  Ostrovski: ISIJ Int., 56 (2016), 786. https://doi.org/10.2355/isijinternational.ISIJINT-2015-704
• 16)   R. G.  Kim,  C. W.  Hwang and  C. H.  Jeon: Appl. Energy., 129 (2014), 299. https://doi.org/10.1016/j.apenergy.2014.05.011
• 17)   Y.  Kashiwaya and  K.  Ishii: Tetsu-to-Hagané, 79 (1993), 1305 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.79.12_1305
• 18)   W. H.  Chen,  M. R.  Lin,  A. B.  Yu,  S. W.  Du and  T. S.  Leu: Int. J. Hydrog. Energy, 37 (2012), 11748. https://doi.org/10.1016/j.ijhydene.2012.05.021
• 19)   C.  Lan,  S.  Zhang,  X.  Liu,  R.  Liu and  Q.  Lyu: ISIJ Int., 61 (2021), 167. https://doi.org/10.2355/isijinternational.ISIJINT-2020-401
• 20)   P.  Wang,  Y.  Zhang,  H.  Long,  R.  Wei,  J.  Li,  Q.  Meng and  S.  Yu: ISIJ Int., 57 (2017), 643. https://doi.org/10.2355/isijinternational.ISIJINT-2016-488
• 21)   S.  Watakabe and  K.  Takeda: Tetsu-to-Hagané, 87 (2001), 467 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.87.7_467
• 22)   P.  Wang,  J.  Zhang and  B.  Gao: High Temp. Mater. Process., 36 (2017), 101. https://doi.org/10.1515/htmp-2015-0112
• 23)   X.  Xing,  H.  Rogers,  G.  Zhang,  K.  Hockings,  P.  Zulli and  O.  Ostrovski: Energy Fuels., 30 (2016), 161. https://doi.org/10.1021/acs.energyfuels.5b02152
• 24)   H.  Yamaoka and  S.  Suyama: ISIJ Int., 43 (2003), 338. https://doi.org/10.2355/isijinternational.43.338
• 25)   Y.  Numazawa,  Y.  Saito,  Y.  Matsushita and  H.  Aoki: Fuel., 266 (2020), 117080. https://doi.org/10.1016/j.fuel.2020.117080
• 26)   K.  Takatani and  Y.  Iwanaga: Tetsu-to-Hagané, 75 (1989), 594 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.75.4_594
• 27)   Y.  Ono,  Y.  Fukuda,  Y.  Sumitani,  Y.  Matsukawa,  Y.  Saito,  Y.  Matsushita,  H.  Aoki,  T.  Shishido and  N.  Okuyama: Fuel., 309 (2022), 122061. https://doi.org/10.1016/j.fuel.2021.122061
• 28)   S. M.  Shin and  S. M.  Jung: Energy Fuels., 29 (2015), 6849. https://doi.org/10.1021/acs.energyfuels.5b01235
• 29)   Z.  Chang,  P.  Wang,  J.  Zhang,  K.  Jiao,  Y.  Zhang and  Z.  Liu: Int J Miner Metall Mater., 25 (2018), 1402. https://doi.org/10.1007/s12613-018-1694-4
• 30)   P.  Wang,  S.  Yu,  H.  Long and  R. Y.  Zhang: Ironmaking & Steelmaking., 44 (2017), 595. https://doi.org/10.1080/03019233.2016.1223792
• 31)   Y.  Wang,  R.  Yoshiie,  Y.  Ueki and  I.  Naruse: Fuel., 108 (2023), 101238. https://doi.org/10.1016/j.joei.2023.101238
• 32)   Y.  Kashiwaya,  M.  Takahata,  K.  Ishi,  K.  Yamaguchi,  M.  Naito and  H.  Hasegawa: Tetsu-to-Hagané, 87 (2001), 259 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.87.5_259
• 33)   J.  An,  J.  Seo,  J.  Choi,  J.  Lee and  H.  Kim: ISIJ Int., 56 (2021), 226. https://doi.org/10.2355/isijinternational.ISIJINT-2015-482