Effective Utilization Method for Surface Tension Based Coal Blending Technique -Effect of Coal Fluidity-

Abstract

In our previous paper, a new measurement method for the coal adhesion property called “surface tension of semi-coke” was devised. The surface tension of a semi-coke sample obtained by heat treatment of a coal sample at 500°C was measured as a unique adhesion property. Conventionally, it has been thought that adhesion is dominant under a low MF (Gieseler maximum fluidity) condition. Moreover, it is important for effective coal utilization to develop a technique that enables production of high strength coke under low MF conditions, which has been thought to deteriorate coke strength. However, in the previous paper, the effect of surface tension on coke strength was investigated only under a single MF condition without changing the level of MF.

In this paper, the effects of surface tension on coke strength under adhesion dominant conditions (low MF and high TI (total inert content)) were investigated. As a result, it was found that the effect of surface tension on coke strength was significant when MF was low or TI was high. Therefore, it is considered that high strength coke can be produced even under low-grade conditions (low MF or high TI) by controlling surface tension. Finally, based on the results, the concept of the conventional MOF diagram was extended. This technique enables effective selection and utilization of coal resources.

1. Introduction

In recent years, there has been increasing global demand to reduce CO₂ emissions associated with business activities as a countermeasure against global warming. Reducing CO₂ emissions is a particularly critical issue for the steel industry, which emits large amounts of CO₂. In the blast furnace process, one of the most important ways to reduce CO₂ emissions is to reduce the reducing agent ratio derived from carbon materials such as coke and pulverized coal. However, a decrease in the coke ratio deteriorates the gas permeability in the blast furnace and causes operational trouble. In order to achieve stable production of pig iron even at a low coke ratio, it is necessary to use coke with higher strength and maintain gas permeability in the blast furnace.¹⁾ Therefore, there is expected to be even stronger demand in the future for the development of advanced technologies that enable the production and stable supply of high strength coke.

As a method for producing high strength coke, it is important to develop a rational technology that takes into account the effective use of coal resources. In other words, rather than relying on high-grade coal, it is desirable to establish a technology that enables use of low-grade coal, which has conventionally been thought to deteriorate coke strength.

Coke strength is considered to be determined by the bond strength between coal particles. In the conventional view, it has been thought that there are two types of bonding between coal particles: fusion and adhesion.^2,3,4,5) Fusion is a form of bonding in which coal particles melt and coalesce with other surrounding particles. In this type of bonding, the boundaries of the original particles disappear, and the particles eventually form a continuous coke matrix. On the other hand, adhesion is a form of bonding in which coal particles soften slightly and only stick each other without formings a single integrated body. In this type of bonding, the boundaries of the original particles are clear and the adhering particles do not form a continuous coke matrix.^6,7) Previous studies have mainly focused on coal properties that related to the fusion phenomenon: dilatation,⁸⁾ fluidity,⁹⁾ viscosity,¹⁰⁾ rheometry^11,12,13) and permeation distance.¹⁴⁾ Regarding the adhesion phenomenon, Arima et al.^15,16,17) proposed the concept of “specific dilatation volume” and argued that voids remain between the coal particles when the amount of coal expansion is insufficient relative to initial void ratio of the coal packed bed. Although the specific dilatation volume can be used to evaluate the ratio of grain boundaries that do not come into contact after carbonization, it is considered difficult to evaluate adhesiveness itself using this index.

To evaluate coal adhesion, in our previous studies,^18,19,20) we developed a new evaluation method called “semi-coke surface tension measurement.” This method was devised based on the principle of the adhesion phenomenon. In general, the adhesion phenomenon is known to proceed in the following order:^21,22) First, the adhesive material become to have fluidity, which moves to the surface to be bonded. Then, the adhesive material wets the surface and obtains a contact state. Finally, the adhesive material hardens while remaining in contact to form a bond. It is known that the adhesive material can wet the surface more effectively and the bond strength increases as the surface tension difference at the interface decreases (i.e., interfacial tension becomes smaller). In the previous study, we concluded that the difference in surface tension during the thermoplastic state affects the adhesion between the coal particles, and investigated the relationship between coke strength and the surface tension evaluated by the developed method.

Although the previous study¹⁸⁾ showed that surface tension affects coke strength, it is not clear how the effect of surface tension on coke strength changes when the quality of the coal blend is degraded (lower Gieseler maximum fluidity (MF) or higher total inert content (TI)), which makes fusion difficult. When the blending quality is high (high MF or low TI), it is known that fusion is the dominant bonding form,^2,4) and in this case, the coal particles can fully coalesce with each other and no interface remains after fusion. Therefore, the influence of adhesion is considered to be small. On the other hand, when the blending quality is low (low MF or high TI), it is also known that adhesion is the dominant bonding form.^2,4) In this case, the coal particles do not have sufficient fluidity to coalesce with each other, and the adhesion strength is considered to be determined by the wettability of the coal particle interface. Therefore, it is considered that the effect of the surface tension difference (or interfacial tension) on coke strength becomes more significant when the blending quality is low (low MF or high TI). In this study, we investigated the relationship between surface tension and coke strength under both high blending quality conditions (high MF or low TI) and low blending quality conditions (low MF or high TI) in order to clarify how the effect of surface tension on coke strength changes under conditions where fusion is difficult.

2. Experimental

2.1. Investigation of Adhesive Strength Under Different MF Conditions (Carbonization Tests of Coal Tablets)

The effect of surface tension differences between coal brands on adhesion strength under different MF conditions was investigated by using binary coal blends.

The properties of the coal brands are shown in Table 1. The mean maximum reflectance of vitrinites (Ro), total inert content (TI),²³⁾ Gieseler maximum fluidity (MF) and proximate analysis values of the coal brands were measured according to JIS M 8816,²⁴⁾ M 8801²⁵⁾ and M 8812,²⁶⁾ respectively. Surface tension γ was measured as in our previous paper¹⁸⁾ and in brief, the measurement procedure comprises the following steps: Raw coal samples were crushed under 200 μm and charged in a crucible with a bulk density of 800 kg/m³. The coal samples were heated to 500°C at 3°C/min and quenched in liquid nitrogen, and the resultant semi-cokes were then crushed under 150 μm. The surface tension of the semi-coke samples was evaluated by the film floatation method.²⁷⁾ The semi-coke particles were powdered on the surface of ethanol aqueous solutions having ethanol concentrations adjusted to 0, 4, 8, 10, 13, 16, 20, 25, 30, 37, 50 or 75 mass%, respectively. The weight ratio distribution of the floating semi-coke particles for each of the 12 tests was obtained, and the weighted average of this distribution was calculated and is called γ.

Table 1. Properties of coal brands used in adhesive strength tests.

	Surface tension (mN/m)	Ro (%)	log MF (log/ddpm)	TI (%)	Ash (wt% d.b.)	VM (wt% d.b.)
Coal A	40.2	0.71	1.32	39.6	8.5	36.3
Coal B	37.8	1.62	1.28	22.2	9.7	17.3
Coal C	41.6	0.92	3.43	24.8	9.0	28.0
Coal D	38.9	1.07	2.09	39.9	10.4	23.3
Coal E	40.9	0.72	2.11	18.1	8.6	37.3
Coal F	40.1	1.03	2.15	39.2	11.3	25.7
Coal G	40.9	0.75	2.28	22.0	9.5	36.1
Coal H	39.6	1.00	2.43	39.2	8.0	28.1

The binary coal blends were selected from the coal brands in Table 1. The relationship between the surface tension difference between two coal brands and the adhesive strength of the resultant coke was investigated.

Adhesive strength was measured by the method described in the previous paper.¹⁸⁾ The raw coal samples were crushed under 75 μm. The aim of this procedure is to diminish the effect of pore growth and to increase the contact area of the coal particles during carbonization in order to evaluate adhesive strength. Next, two coal samples were mixed with a blending ratio of 50:50 in weight. The coal mixture was placed into a mold and a load of 14 MPa was applied to the mold for 10 s, forming a coal tablet with a diameter of 6.7 mm and thickness of 2.3 mm.

The coal tablets were placed in a carbonization vessel (dimension: W200 mm × L200 mm × H70 mm) packed with coke breeze whose size was under 1 mm. The vessel was heated to 1000°C at 3°C/min, and the resultant coke samples were cooled under a nitrogen atmosphere.

The tablet coke samples were compressed in the thickness direction with a pressing speed of 1 mm/min by using a universal testing machine (Shimadzu, AG-I 50 kN), and the breakage load was measured. The compressive strength was calculated by dividing the breakage load by the area that bears the load. The averaged compressive strength of 10 samples was evaluated as the adhesion strength.

2.2. Investigation of Adhesion Dominant Conditions (Carbonization Tests of Coal Blends to Produce Lump Coke)

The conditions under which the effect of surface tension on coke strength becomes significant were investigated with multi-component coal blends. Two carbonization tests were carried out to investigate the effect of surface tension on coke strength with different coal blend qualities. First, the effect of surface tension on coke strength under different MF conditions was evaluated, which is called the MF-test in the following. Second, the effect of surface tension on coke strength under different TI conditions was evaluated, which is called the TI-test below.

2.2.1. Effect of Surface Tension Under Different MF Conditions (MF-test)

The properties of the coal brands are shown in Table 2. The qualities and blending conditions of the coal blends are shown in Table 3. The interfacial tension of a heat-treated coal blend γ_inter was evaluated as described in the previous paper.¹⁸⁾ In brief, the procedure comprises the following steps: First, the surface tension γ_i which is the property of a single coal brand, was obtained. Next, interfacial tension γ_ij, which is the property of the interface between two coal brands i and j was calculated by Eq. (1).^28,29)   

$${\gamma }_{\mathrm{i}\mathrm{j}}={\gamma }_{\mathrm{i}}+{\gamma }_{\mathrm{j}}-2exp\left[-\beta {\left({\gamma }_{\mathrm{i}}-{\gamma }_{\mathrm{j}}\right)}^2\right]\sqrt{{\gamma }_{\mathrm{i}}{\gamma }_{\mathrm{j}}}$$(1)

β was a constant which is experimentally determined, and was calculated by D. Li and A. W. Neumann as 0.0001247 (m²/mJ)².²⁹⁾ Finally, the interfacial tension of the coal blend γ_inter, which is a property of the coal blend, was calculated as the weighted average of interfacial tension γ_ij by Eq. (3):   

$${\gamma }_{\text{inter}}=\sum _{i=1}^n\sum _{j=1}^n{w}_i{w}_j{\gamma }_{ij}$$(3)

where, w_i: mass fraction of coal brand i (i=1,2,…,i,…n).

Table 2. Properties of coal brands used in carbonization tests (MF-test).

Coal brand	Ro (%)	logMF (log/ddpm)	TI (%)	Ash (wt% d.b.)	VM (wt% d.b.)	Surface tension (mN/m)
M_A	1.24	2.48	38.4	8.4	23.3	39.3
M_B	1.23	1.04	45.6	8.0	21.6	40.2
M_C	0.98	2.58	33.8	8.3	27.1	41.1
M_D	0.99	0.48	46.6	9.8	25.7	41.3
M_E	0.97	1.79	32.4	8.7	29.0	40.2
M_F	0.81	2.92	8.6	8.4	37.0	41.0
M_G	1.54	0.85	21.4	10.7	18.6	38.7
M_H	0.64	3.47	21.8	6.8	42.1	41.6
M_I	1.18	2.85	30.8	10.4	23.3	39.8
M_J	1.17	2.65	29.2	10.2	22.8	39.8

Table 3. Coal blending conditions and coke strength in carbonization tests (MF-test).

Coal brand	Blend M_1	Blend M_2	Blend M_3	Blend M_4	Blend M_5
M_A	30.0	30.0	30.0
M_B				30.0	30.0
M_C				7.0	25.0
M_D	18.0	17.6	17.3	8.0
M_E	25.0	16.0	7.9	10.5
M_F
M_G	1.1	0.7		1.0
M_H	14.9	18.5	21.4	21.6	21.4
M_I	0.0	10.4	20.4	18.4	23.6
M_J	11.0	6.8	3.0	3.5
Ro (%)	1.03	1.03	1.03	1.03	1.03
logMF (log/ddpm)	2.10	2.24	2.37	2.09	2.37
TI (%)	34.7	34.3	34.0	34.8	34.1
γinter (mN/m)	0.026	0.028	0.029	0.017	0.015
${\text{DI}}_{\text{15}}^{\text{150}}$ (-)	79.4	79.7	80.0	80.3	80.6

Each coal sample was air-dried and crushed under 3 mm. As shown in Table 3, five types of coal blends were prepared by blending coal brands at specified blending ratios. The carbonization conditions are shown in Table 4. The moisture content of the coal blend was controlled to 8 wt%. The coal blend was charged in a carbonization vessel (W270 mm × L263 mm × H300 mm) with a bulk density of 775 kg/m³. The vessel was heated in an electric furnace while controlling the heating wall temperature to 1050°C for 360 min, and then cooled under a nitrogen atmosphere. The coke strength (drum strength index ${\text{DI}}_{\text{15}}^{\text{150}}$) of the resulting lump coke sample was measured according to JIS K 2151.³⁰⁾

Table 4. Carbonization test conditions (MF-test, TI-test).

	MF-test	TI-test
Moisture content (wt%)	8
Bulk density (kg-dry/m3)	775	750
Dimensions (mm)	W270×L263×H300
Wall temperature (°C)	1050
Coking time (min)	360

2.2.2. Effect of Surface Tension Under Different TI Conditions (TI-test)

The properties of the coal brands are shown in Table 5. The qualities and blending conditions of the coal blends are shown in Table 6.

Table 5. Properties of coal brands used in carbonization tests (TI-test).

Coal brand	Ro (%)	logMF (log/ddpm)	TI (%)	Ash (wt% d.b.)	VM (wt% d.b.)	Surface tension (mN/m)
T_A	1.46	1.72	26.8	10.0	19.1	39.9
T_B	1.56	0.00	37.5	8.3	17.2	37.7
T_C	0.91	3.64	21.6	7.9	33.4	42.2
T_D	0.95	3.00	31.4	7.9	28.8	40.8
T_E	0.73	2.46	23.2	8.8	36.2	42.8
T_F	1.37	1.04	44.3	7.0	19.3	41.0
T_G	0.75	2.73	5.9	5.2	39.9	40.7
T_H	0.98	1.54	37.6	9.6	25.8	41.1
T_I	0.73	2.59	19.9	9.1	34.2	41.4
T_J	0.98	2.78	35.7	8.6	28.2	40.1
T_K	1.03	3.10	37.1	9.6	28.2	40.7
T_L	1.03	3.09	35.3	9.1	27.8	40.7
T_M	1.62	0.70	17.9	9.5	18.8	37.7

Table 6. Coal blending conditions and coke strength in carbonization tests (TI-test).

Coal brand	Blend T_1	Blend T_2	Blend T_3	Blend T_4
T_A	8
T_B	9	5	7	3
T_C	10		15
T_D	10
T_E	12
T_F	15	29	18	26
T_G	17	35		26
T_H	19	11	20	22
T_I			25
T_J			10
T_K		20.0
T_L				23.0
T_M			5.0
Ro (%)	1.05	1.05	1.05	1.05
logMF (log/ddpm)	2.01	2.05	2.00	2.03
TI (%)	28.4	28.3	30.8	30.6
γinter (mN/m)	0.053	0.016	0.052	0.010
${\text{DI}}_{\text{15}}^{\text{150}}$ (-)	80.8	81.1	79.6	80.9

Each coal sample was air-dried and crushed under 3 mm. As shown in Table 3, five types of coal blends were prepared by blending the coal brands at specified blending ratios. The carbonization conditions are shown in Table 4. The moisture content of the coal blend was controlled to 8 wt%. The coal blend was charged in a carbonization vessel (W270 mm × L263 mm × H300 mm) with a bulk density of 750 kg/m³. The vessel was heated in an electric furnace while controlling the heating wall temperature to 1050°C for 360 min, and then cooled under a nitrogen atmosphere. The coke strength (drum strength index ${\text{DI}}_{\text{15}}^{\text{150}}$) of the resulting lump coke sample was measured according to JIS K 2151.³⁰⁾

3. Results and Discussion

3.1. Effect of Surface Tension on Adhesion Strength Under Different MF Conditions

The effect of surface tension on the adhesion strength between two coal brands was investigated. Table 7 shows the blended coal brands, the properties of coal blends, the surface tension difference between the two coal brands and the adhesive strength of the resultant coke. Figure 1 shows the relationship between MF of the coal blend and adhesive strength. The adhesive strength tends to increase with an increasing MF of the coal blend. Miyazu et al.²⁾ and Aramaki et al.⁴⁾ reported that high MF coal blends displayed fusion bonding and low MF coal blends showed adhesion bonding. Those reports suggest that coal blends with higher MF contain more fusion bonds. Therefore, from the results in Fig. 1, it was suggested that fusion is a stronger bonding form than adhesion. Figure 2 shows the relationship between adhesive strength and the absolute value of the surface tension difference between two coal brands (Δγ). Adhesive strength decreased with increasing Δγ under a low MF condition. This is same trend as reported in our previous paper.¹⁸⁾ On the other hand, adhesive strength did not decrease with increasing Δγ under a high MF condition. Miyazu et al.²⁾ and Aramaki et al.⁴⁾ reported that a clear particle interface remained when MF was low. In addition, Kanai et al.³¹⁾ considered that non-adhesion grain boundaries caused a decrease in strength due to large stress concentration. Therefore, it is thought that coke strength is determined by adhesion at the interface under low MF conditions, and this is considered to depend on the surface tension difference. On the other hand, when MF was high, Miyazu et al.²⁾ and Aramaki et al.⁴⁾ reported that the coke matrix became continuous and the interface disappeared, and Kanai et al.³¹⁾ considered that when the amount of low-grade coal brands in a coal blend is small, fracture of the coke matrix itself occurs, rather than fracture originating at the non-adhesion grain boundaries. Therefore, it is considered that when MF is low, there is more adhesion bonding and the surface tension difference affects the wettability of the interface and is controlling for coke strength.

Table 7. Blending conditions and measurement results of adhesive tests.

Sample	1	2	3	4	5	6	7	8	9	10
Coal brand	Coal A	Coal D	Coal E	Coal A	Coal H	Coal H	Coal E	Coal H	Coal C	Coal G
	Coal B	Coal B	Coal B	Coal D	Coal B	Coal D	Coal H	Coal F	Coal B	Coal H
Ro (%)	1.17	1.35	1.17	0.89	1.31	1.04	0.86	1.02	1.27	0.88
log MF (log/ddpm)	1.30	1.69	1.70	1.71	1.86	2.26	2.27	2.29	2.36	2.36
TI (%)	30.9	31.1	20.2	39.8	30.7	39.6	28.7	39.2	23.5	30.6
Difference of surface tension (mN/m)	2.4	1.1	3.1	1.3	1.8	0.7	1.3	0.5	3.8	1.3
Adhesive strength (MPa)	36.8	67.9	34.6	64.2	57.4	88.4	134.9	144.0	137.7	125.3

Fig. 1.

Relationship between MF and adhesive strength.

Fig. 2.

Relationship between surface tension difference and adhesive strength under different MF conditions.

The tablet test has the advantage of enabling evaluation of a large number of samples because it requires only a small sample size, and showed that the effect of the surface tension difference on coke strength is reduced under high MF conditions. However, because the tablet samples are small, large defects tend to occur under high MF conditions, and as a result, the test may be less accurate. Therefore, an evaluation with lump coke samples is necessary. In addition, the coke strength of a binary coal blend may be affected by other blending qualities (for example, Ro). More practically, it is necessary to produce coke with the same blending qualities. Therefore, in the following sections, carbonization tests were conducted to produce lump coke with the same blending qualities.

3.2. Effect of Surface Tension on Coke Strength Under Different MF Conditions

As a result of the MF-test, the relationship between the MF of the coal blend and coke strength is shown in Fig. 3. The coke obtained from the high γ_inter coal blends (Blend M_1, Blend M_2 and Blend M_3) showed lower strength than the coke obtained from the low γ_inter coal blends (Blend M_4 and Blend M_5). In addition, when γ_inter was high, the coke strength decreased greatly with a decrease in the MF of the coal blend. Figure 4 shows the relationship between coke strength and γ_inter. When the MF of the coal blend was low, coke strength decreased greatly with an increase in γ_inter. Our previous report¹⁸⁾ also confirmed that coke strength tends to decrease with increasing γ_inter, and the same trend was observed in the present study. In addition to this finding, the present study also suggests that the effect of γ_inter on coke strength becomes significant when the MF of a coal blend is low.

Fig. 3.

Relationship between MF and coke strength under different interfacial tension conditions.

Fig. 4.

Relationship between interfacial tension and coke strength under different MF conditions.

The mechanism by which the effect of surface tension on coke strength differs under different MFs of coal blends, that is, the MF dependency of the effect of surface tension on coke strength, was estimated. Sasaki et al.⁶⁾ and Mochida et al.⁷⁾ reported that low MF coal particles were in contact but carbonized independently. Based on those reports, when the MF of a coal blend is low, the coal blend is considered to be in the “adhesion” state shown in Fig. 5(a), in which the coal particles soften slightly and only stick each other, but do not coalesce into a single body. Specifically, the interfaces of the original particles remain clear, and the particles are not fully integrated. In addition, Kanai et al.³¹⁾ considered that the non-adhesion grain boundary was one of the areas prone to rupture because of the high stress concentration. Under a state like that in Fig. 5(a), coke strength is considered to be dominated by the adhesion strength at the interface between the coal particles. Since this adhesion strength is greatly affected by the difference in surface tension between the coal particles,¹⁸⁾ it is estimated that the effect of γ_inter on coke strength is significant when the MF of a coal blend is low.

Fig. 5.

Estimated mechanisms of MF dependency of effect of interfacial tension on coke strength.

On the other hand, Miyazu et al.²⁾ and Aramaki et al.⁴⁾ reported that when MF was high, the coke matrix became continuous and the interface between the coal particles disappeared. Based on those reports, when the MF of a coal blend is high, the coal blend is considered to be in the “fusion” state shown in Fig. 5(b). The coal particles melt and coalesce with each other, eventually forming a single continuous coke matrix with indistinct interfaces between the original particles. Kanai et al.³¹⁾ suggested that when the amount of a low-grade coal brand in a coal blend is small, fracture of the coke matrix itself occurs, rather than fracture originating at the non-adhesion grain boundaries. Under a state like that in Fig. 5(b), coke strength is considered to be dominated by the pore structure rather than adhesion. Therefore, the effect of γ_inter on coke strength is reduced.

3.3. Effect of Surface Tension on Coke Strength Under Different TI Conditions

As a result of the TI-test, the relationship between the TI of a coal blend and coke strength is shown in Fig. 6. The coke obtained from high γ_inter coal blends (Blend T_1, Blend T_2) showed lower strength than the coke obtained from low γ_inter coal blends (Blend T_3 and Blend T_4). In addition, when γ_inter was high, the coke strength decreased greatly with increasing TI of the coal blend. Figure 7 shows the relationship between coke strength and γ_inter. Here, when the TI of a coal blend was high, coke strength decreased greatly with increasing γ_inter. Therefore, it is suggested that the effect of γ_inter on coke strength becomes significant when the TI of the coal blend is high.

Fig. 6.

Relationship between TI and coke strength under different interfacial tension conditions.

Fig. 7.

Relationship between interfacial tension and coke strength under different TI conditions.

The mechanism of the effect of surface tension on coke strength under different TI of coal blends was estimated. Barriocanal et al.^32,33) reported that increasing the mixing ratio of inert macerals reduced the ratio of inert macerals enveloped by melting reactive macerals. Based on those reports, when the TI of a coal blend is high, the coal blend is considered to be in the “adhesion” state shown in Fig. 8(a). The melting reactive macerals cannot fully envelop the inert macerals, and the reactive macerals cannot coalesce completely with each other, resulting in a discontinuous reactive maceral coke matrix divided by inert macerals. Since the inert macerals do not melt, the bonding form at the interface between an inert maceral and a reactive maceral is adhesion. In addition, Roest et al.^34,35,36) argued that when there were many pores around the inert derived coke matrix, that area would become the starting point of rupture. Under a state like that in Fig. 8(a), the adhesion interface is large and the effect of adhesion on coke strengh is considered to be significant. Therefore, the effect of γ_inter on coke strength is considered to be significant when the TI of a coal blend is high.

Fig. 8.

Estimated mechanisms of TI dependency of effect of interfacial tension on coke strength.

When the TI of a coal blend is low, the coal blend is considered to be in the state shown in Fig. 8(b) based on the reports by Barriocanal et al.^32,33) In this case, the melting reactive macerals can envelop inert macerals completely and the reactive macerals can coalesce with each other, and a single continuous reactive maceral coke matrix is considered to be formed. Roest et al.^34,35,36) reported that when the adhesion between the inert macerals and reactive macerals was good, fracture of the coke matrix itself occured instead of fracture originating at the interface between the inert and reactive macerals. Under a state like that in Fig. 8(b), coke strength is considered to be dominated by the pore structure rather than adhesion. Therefore, the effect of γ_inter on coke strength is considered to be reduced.

If the estimated mechanism is correct, it is expected that the effect of γ_inter on coke strength would be significant under conditions where adhesion is dominant at the coal particle interface, not only when MF is low or TI is high.

3.4. Extension of Conventional Blending Theory

Finally, the findings obtained in this study were correlated with the conventional coal blending theory. Miyazu et al.^2,37) proposed the concept of the MOF diagram (thick line in Fig. 9). In this conceptual diagram, from a practical point of view, there was concern to clarify the threshold value for MF transitioning from the fluidity-controlled region to the coal rank-controlled region. However, the threshold value of MF is still unclear. In this study, it was found that the influence of adhesive strength was greater in the low MF region. Previous studies have not evaluated the effect of adhesion, and it may have been difficult to determine the threshold value of MF. This paper proposes an extended concept of the MOF diagram in which the effect of adhesive strength (interfacial tension) is added to the conventional concept of the MOF diagram in Fig. 9. This extended concept means that when γ_inter is low, the threshold value of the MF transition to the fluidity-controlled region is lower.

Fig. 9.

Extended MOF diagram including effects of surface tension.

4. Conclusions

In order to develop a more effective utilization technique for coal surface tension, the effect of surface tension on coke strength was investigated under different MF and TI conditions of coal blends. As a result, the following findings were obtained.

(1) It was found that the effect of surface tension on coke strength increased when the MF of the coal blend was low.

(2) It was found that the effect of surface tension on coke strength increased when the TI of the coal blend was high.

In conclusion, this study clarified the fact that control of interfacial tension γ_inter is important in the condition where adhesion is dominant. According to the findings obtained in this study, high strength coke can be produced by controlling the γ_inter of the coal blend, even under conditions where the MF of the coal blend is low, which is conventionally considered to reduce coke strength. This approach can contribute to broadening the range of coal brands that can be used in the coke production process and allows a more flexible selection of coals. For example, it is expected that high strength coke can be produced even under low MF or high TI conditions by adjusting the γ_inter of the coal blend to be small. Therefore, the findings in this paper are expected to contribute to reduction of the reducing agent ratio in blast furnaces and reduction of CO₂ emissions in the ironmaking process through the production of high strength coke.

References

• 1)   T.  Ariyama,  M.  Sato,  T.  Sato,  S.  Watakabe and  R.  Murai: Tetsu-to-Hagané, 92 (2006), 114 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.92.3_114
• 2)   T.  Miyazu,  Y.  Okuyama,  N.  Suzuki,  T.  Fukuyama and  T.  Mori: NKK Tech. Rep., 67 (1975), 125.
• 3)   T.  Aramaki,  K.  Kojima,  Y.  Sakurai and  M.  Sakawa: Coke Circular, 33 (1984), 35 (in Japanese).
• 4)   T.  Aramaki,  Y.  Miura,  M.  Sakawa and  Y.  Sakurai: J. Fuel Soc. Jpn., 69 (1990), 355 (in Japanese). https://doi.org/10.3775/jie.69.355
• 5)   K.  Nishioka and  S.  Yoshida: Tetsu-to-Hagané, 70 (1984), 351. https://doi.org/10.2355/tetsutohagane1955.70.3_351
• 6)   M.  Sasaki,  I.  Mochida and  R.  Takahashi: J. Jpn. Inst. Energy, 71 (1992), 256 (in Japanese). https://doi.org/10.3775/jie.71.256
• 7)   I.  Mochida,  K.  Maeda,  K.  Takeshita and  H.  Marsh: J. Jpn. Pet. Inst., 23 (1980), 127 (in Japanese).
• 8)   K.  Nishioka and  S.  Yoshida: Trans. Iron Steel Inst. Jpn., 23 (1983), 381. https://doi.org/10.2355/isijinternational1966.23.381
• 9)   T.  Fukuyama,  T.  Miyazu and  S.  Kimoto: J. Fuel Soc. Jpn., 51 (1972), 628 (in Japanese). https://doi.org/10.3775/jie.51.8_628
• 10)   K.  Fukada,  S.  Itagaki and  I.  Shimoyama: Tetsu-to-Hagané, 90 (2004), 641 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.90.9_641
• 11)   S.  Nomura,  K.  Kato,  I.  Komaki,  Y.  Fujioka,  K.  Saito and  I.  Yamaoka: J. Jpn. Inst. Energy, 78 (1999), 33 (in Japanese). https://doi.org/10.3775/jie.78.33
• 12)   T.  Takanohashi,  T.  Yoshida,  M.  Iino,  H.  Kumagai and  K.  Kato: Tetsu-to-Hagané, 85 (1999), 382 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.85.5_382
• 13)   K. M.  Steel,  R. E.  Dawson,  D. R.  Jenkins,  R.  Pearce and  M. R.  Mahoney: Fuel Process. Technol., 155 (2017), 106. https://doi.org/10.1016/j.fuproc.2016.04.027
• 14)   Y.  Dohi,  K.  Fukada,  T.  Yamamoto,  T.  Matsui,  H.  Sumi and  I.  Shimoyama: Tetsu-to-Hagané, 101 (2015), 407 (in Japanese). https://doi.org/10.2355/tetsutohagane.TETSU-2015-013
• 15)   T.  Arima,  T.  Suzuki,  K.  Kato and  I.  Komaki: CAMP-ISIJ, 10 (1997), 858 (in Japanese).
• 16)   T.  Arima: Tetsu-to-Hagané, 87 (2001), 274 (in Japanese). https://doi.org/10.2355/tetsutohagane1955.87.5_274
• 17)   S.  Nomura,  T.  Arima and  K.  Kato: Fuel, 83 (2004), 1771. https://doi.org/10.1016/j.fuel.2004.03.006
• 18)   M.  Nagayama,  K.  Fukada,  T.  Matsui,  H.  Fujimoto,  Y.  Dohi,  H.  Sumi and  I.  Shimoyama: ISIJ Int., 57 (2017), 989. https://doi.org/10.2355/isijinternational.ISIJINT-2016-660
• 19)   D.  Igawa,  K.  Fukada,  T.  Matsui,  H.  Fujimoto,  Y.  Dohi,  H.  Sumi,  M.  Nagayama and  I.  Shimoyama: 8th ICSTI 2018 Proc., The Austrian Society for Metallurgy and Materials, Franz-josef-straße, Leoben, (2018), 63.
• 20)   I.  Akishika,  K.  Fukada,  T.  Matsui,  H.  Fujimoto,  Y.  Dohi,  T.  Yamamoto,  H.  Sumi and  I.  Shimoyama: AISTech 2019 Proc. Iron & Steel Technology Conf., Curran Associates, Inc., Red Hook, NY, (2019), 213.
• 21)   T.  Hata: J. Soc. Chem. Ind., Jpn., 73 (1970), 10 (in Japanese). https://doi.org/10.1246/nikkashi1898.73.10
• 22)   T.  Hata and  Y.  Kitazaki: J. Adhes. Soc. Jpn., 7 (1971), 8 (in Japanese).
• 23)   N.  Schapiro,  R. J.  Gray and  G. R.  Ensner: Proc. Blast Furnace, Coke Oven, and Raw Materials Conf., Metallurgical Society of AIME, New York, (1964), 89.
• 24)  JIS M 8816: 1992, Solid mineral fuels−Methods of microscopical measurement for the macerals and reflectance (in Japanese).
• 25)  JIS M 8801: 2008, Coal-Testing methods (in Japanese).
• 26)  JIS M 8812: 2006, Coal and coke−Methods for proximate analysis (in Japanese).
• 27)   M. C.  Williams and  D. W.  Fuerstenau: Int. J. Miner. Process., 20 (1987), 153.
• 28)   J.  Girifalco and  R. J.  Good: J. Phys. Chem., 61 (1957), 904. https://doi.org/10.1021/j150553a013
• 29)   D.  Li and  A. W.  Neumann: J. Colloid Interface Sci., 148 (1992), 190. https://doi.org/10.1016/0021-9797(92)90127-8
• 30)  JIS K 2151: 2004, Coke-Testing methods (in Japanese).
• 31)   T.  Kanai,  Y.  Yamazaki,  X.  Zhang,  A.  Uchida,  Y.  Saito,  M.  Shoji,  H.  Aoki,  S.  Nomura,  Y.  Kubota,  H.  Hayashizaki and  S.  Miyashita: J. Therm. Sci. Technol., 7 (2012), 351. https://doi.org/10.1299/jtst.7.351
• 32)   C.  Barriocanal,  S.  Hanson,  J. W.  Patrick and  A.  Walker: Fuel, 73 (1994), 1842. https://doi.org/10.1016/0016-2361(94)90209-7
• 33)   C.  Barriocanal,  S.  Hanson,  J. W.  Patrick and  A.  Walker: Fuel, 75 (1996), 243. https://doi.org/10.1016/0016-2361(95)00233-2
• 34)   R.  Roest,  H.  Lomas,  K.  Hockings and  M. R.  Mahoney: Fuel, 180 (2016), 785. https://doi.org/10.1016/j.fuel.2016.01.058
• 35)   R.  Roest,  H.  Lomas,  S.  Gupta,  R.  Kanniala and  M. R.  Mahoney: Fuel, 180 (2016), 803. https://doi.org/10.1016/j.fuel.2016.04.019
• 36)   H.  Lomas,  D. R.  Jenkins,  M. R.  Mahoney,  R.  Pearce,  R.  Roest,  K.  Steel and  S.  Mayo: Fuel Process. Technol., 155 (2017), 183. https://doi.org/10.1016/j.fuproc.2016.05.039
• 37)   Y.  Okuyama,  T.  Miyazu,  H.  Sugimura and  M.  Kumagai: J. Fuel Soc. Jpn., 49 (1970), 736 (in Japanese). https://doi.org/10.3775/jie.49.10_736