Using Coal Resources with Optimal Bursting Pressure for the Production of High-Quality Metallurgical Coke

Abstract

When applying advanced technologies and technological methods for the preparation of coal raw materials (technology for coking stamped batch, technology for coking dry or thermally prepared batch), the problem of developing high bursting pressure arises. The aim of this research is to assess the possibility of predicting the bursting pressure of coal blends taking into account their technological properties and petrographic characteristics, as well as to study the effect of bursting pressure on the metallurgical properties of coke. Standardized methods were used to study the technological properties of coal and coal blends (determination of technical and petrographic analyses). The qualitative characteristics of coke were studied using physical, mechanical, and thermochemical methods for the study of standardized indicators: crushability (M₂₅), abrasion (M₁₀), reactivity (CRI), and post-reaction strength (CSR). The regression equations for predicting the bursting pressure of coal blends, taking into account the volatile matter in the blend, vitrinite content, and grinding, which are characterized by high correlation coefficients (0.89 and 0.9), were proposed. Their use will make it possible to optimize the composition of coal batches, control the bursting pressure during regrinding, and reduce the number of experimental measurements in a particular coke production. It was also found that an increase in the bursting pressure by 1 kPa can be expected to increase the mechanical strength of coke in terms of crushability M₂₅ by about 2.6% and reduce the abrasion of coke M₁₀ by 1%.

1. Introduction

Modern requirements for blast furnace coke quality remain quite high: mechanical strength in terms of crushability M₂₅ ≥ 88.0–90.0% and abrasion M₁₀ ≤ 6.0–6.5%; post-reaction strength of coke CSR—60.0–75.0%; coke reactivity CRI—25–30.0%, content of +80 mm fraction—no more than 5%; content of −25 mm fraction—no more than 5%; moisture fluctuations on both sides—no more than 0.5% [1].

The need to produce high-quality metallurgical coke and the shortage of high-sintering coal resources have led to the formation of a raw material base for Ukrainian coke plants from coal from different basins. Differences in the technological properties and material composition of imported and Ukrainian coal concentrates require clarification and improvement of the main technological methods of preparation when they are used in coal batches. Achieving high-quality coke for blast furnace smelting is possible provided an integrated, scientifically based approach to improving the technology of coal preparation for coking is taken, which involves the development of methods and technological measures [1,2].

The quality of coke depends on both the component composition of the coal batch and the technology of preparing coal concentrates for the coking process in terms of such indicators as particle size distribution, bulk density, moisture content, and initial temperature of the coal mass. Recently, technologists have been focusing on the introduction of advanced technologies such as modification of batchеs with additives, including inorganic (carbonates, carbides), desulphurizing (anthracite, coke fines, and dust), and organic (liquid and solid products of petrochemical and coke production [3,4,5]; coking of stamped batchеs [6]; and the use of special methods of preparing coal feedstock for coking (technology for coking stamped batchеs, technology for coking dry batchеs and technology for coking thermally prepared batchеs) [7,8,9] and briquetting/partial briquetting of batchеs [10,11,12].

In the process of compaction of the coal batch, its bulk density is increased due to the mutual sliding of coal grains and their reorientation and convergence [6,13]. In a stamped coal batch, difficult conditions are created for the evacuation of vapor and gaseous products from the plastic layer. This leads to an increase in gas pressure. The higher the gas pressure, the more actively the vapor and gaseous products of pyrolysis participate in the following reactions with the organic mass of decomposing coal. At the same time, hydrogen saturation reactions are intensified and free bonds of macromolecular fragments (free radicals) are formed. This leads to an increase in the amount of relatively low molecular weight compounds that can be in a plastic state at certain temperatures and take an active part in the sintering process. This makes it possible to use the sintering potential of the batch to a greater extent and produce coke of high mechanical strength from batches of reduced sintering [6].

The technology of coking pre-dried charge developed by Nippon Steel Corporation was implemented as a CMC (charge moisture control) process in technological and hardware design, then improved and implemented as DAPS (charge drying and pelletizing) [8].

According to the CMC technology, coal is dried in a pipe-dryer, while its moisture content is reduced from 10% to 5–6% before being loaded into the coke oven [8].

The essence of the DAPS process is to reduce the moisture content of the batch to 2% by drying in a fluidized bed. The resulting fine coal dust is sent to pelletizing. Dry coarse coal and pelletized coal are mixed and fed into a coke oven [14].

The thermal preparation technology involves the rapid heating of the batch to 150–250 °C before coking under conditions that prevent its oxidation. It makes it possible to produce coke from blends with a high content of highly volatile low-blowing coal that is superior in mechanical strength to low-reactive coke from the best coal blends [7].

Thus, the reduction in coal moisture content achieved in these processes leads to an increase in the bulk density of coal, an increase in coke oven productivity, energy saving, and an increase in coke strength. However, it has also been established that a high bulk density of coal loading increases the bursting pressure and leads to operational problems.

Thus, on the one hand, the use of advanced technologies and technological methods makes it possible to achieve high-quality coke while expanding the raw material base for coking, increasing the share of low-burning coal in coal batchеs, thereby reducing their cost. On the other hand, there is the problem of increased bursting pressure of thermally treated, stamped batchеs. It is in this case that the issue of optimizing the component composition and predicting the bursting pressure, taking into account the properties of coal batchеs, becomes acute.

Bursting pressure is also one of the most important aspects of industrial coking, which determines the operating conditions, condition, and service life of the coke oven stock. And given that in order to achieve the required coke quality and high productivity of coke ovens the current practice in the coke industry is to use coal batchеs that include up to 12 coal concentrates of different geographical origin (USA, Australia, China, etc.) [15], the development of methods for predicting the bursting pressure remains an urgent issue.

The bursting pressure should be understood as the pressure exerted by a coal mass that has passed into a plastic state, provided that it is deprived of the ability to expand freely. The main reason for the development of the burst pressure is the pressure in the plastic layer of the vapor-gas products of coal thermal decomposition, which is transferred through the semi-coke–coke to the refractory wall of the heating wall of the coking chamber, causing its deflection. According to the Rules for Technical Operation of Coke and Chemical Enterprises, the bursting pressure of the plant batch should not exceed 7 kPa, which ensures the normal operating conditions of coke oven batteries and a long service life [13,16,17].

The value of the bursting pressure will depend on the nature of the coal feedstock (petrographic characteristics), the properties of the plastic mass (fluidity and gas permeability), and the amount of volatile products formed during pyrolysis, as well as the coking rate, the width of the coke oven chamber, and the method of its loading [13].

The difficulty in blending and planning coal batch compositions with regard to bursting pressure and flowability is that the values are not additive. Therefore, coke plants experimentally determine the bursting pressure of coal concentrates and coal batchеs for coking. The proposed equations and methods [3,18,19] for predicting the bursting pressure are cumbersome, and their use is complicated by the need to determine the component composition of steam and gas products for the components of the batch, which is not always possible in the laboratories of coke plants. When new coal with unknown bursting pressure is introduced, it becomes necessary to study experimental blends and determine their bursting pressure.

Thus, in order to preserve the furnace stock and produce blast furnace coke with the required quality indicators with the available coal resources, various methods and techniques are used to improve certain quality parameters of the main product and ensure optimal operating conditions for the equipment. It should also be borne in mind that the introduction of the stamped batch coking technology allows for the involvement of cheaper and less scarce G, GZh, KP, and KS ranks of coal, which are used in limited quantities in layer coking, in the raw material base of the enterprise. In turn, an increase in the content of coking coal, which contributes to the development of a high bursting pressure, requires study with further modeling of batch compositions, both for layer coking and stamped batch coking.

Therefore, given the instability of the coking raw material base, limited coal resources, the introduction of the latest technologies for preparing coal for coking, and the need for proper operation of coke ovens and preservation of the furnace stock, the issue of monitoring and regulating the bursting pressure of coal batchеs that they develop during coking is relevant. The aim of this research is to evaluate the possibility of predicting the bursting pressure of coal batchеs, taking into account their technological properties and petrographic characteristics, and to study the effect of bursting pressure on the metallurgical properties of coke.

2. Materials and Methods

In this study, 17 coal concentrates were used as the object of research. Of the seventeen coal concentrates, six are represented by grade G, five by grade Zh, three by grade K, and one each by ranks DG, GZh, and PS. Depending on the vitrinite reflectance (R₀), volatile matter (V^daf), calorific value, and sintering power, as measured by the plastic layer thickness (Y) and Rog Index (RI), Ukrainian coal is divided into grades in accordance with the National Standard of Ukraine 3472-96 “Lignite, Stone and Anthracite Coal Classification” [20]. Highly volatile coals of the gas group include the G, DG, and GZh grades. The sintering base of the charge is made up of the Zh and K grades.

Sampling of coal concentrates and coal batches was carried out manually from the surface of the stopped conveyor using a device (frame). A frame was installed on the conveyor belt, which was two parallel walls vertically located at a distance that was at least twice the size of the maximum piece. The frame was immersed in the coal raw material to the transport surface perpendicular to the flow direction. The selected sample was delivered to the coal testing room for further preparation for the test. The preparation included successive operations of grinding, reduction, and separation of the sample. Equipment and tools for sample preparation met the requirements according to the state standard of Ukraine 4096-2002 “Brown coal, hard coal, anthracite, combustible shale and coal briquettes. Methods of sample selection and preparation for laboratory tests” [21].

The coal batch sample was poured onto the sampling table and sifted through a sieve with cell sizes of 3 mm. Grains that did not pass through the sieve were crushed and sieved again until the entire sample passed through the sieve. Next, the sample was thoroughly mixed with the shoveling method, which consists of placing the sample on the table, then scooping it evenly around the perimeter with scoops, pouring it into one point, and taking it to the center to form a cone. The operation was repeated three times with a change in the location of the cone. Then, the sample was reduced by the quartering method in the following order:

-

The cone obtained after mixing was leveled so that the upper part was in the shape of a circle with a uniform thickness of the layer. The center of the circle had to coincide with the center of the cone;

-

The circle was divided into four parts using a cross;

-

Two diametrically opposite parts were discarded, and the remaining parts were combined; the reduction operation was repeated until the weight was at least 2 kg;

-

In order to prevent an error, two opposite parts were rejected at each subsequent operation. Next, the sample was divided into four parts with the help of a cross to determine the quality parameters of the coal raw material.

The characteristics of the coal concentrates were determined according to the following standard methods:

-

ІSО 1171-97 “Solid mineral fuels. Methods for determination of ash” [22];

-

ІSО 589-81 “Hard coal–Determination of total moisture” [23];

-

ІSО 7404-3-84 “Methods for the petrographic analysis of bituminous coal and anthracite—Part 3: Method of determining maceral group composition” [24];

-

ІSО 7404-5-85 “Methods for the petrographic analysis of coals–Part 5: Method of determining microscopically the reflectance of vitrinite” [25];

-

State standard of Ukraine 7722:2015 “Hard coal. Method of determining plastometric characteristics” [26].

The burst pressure was determined in accordance with DSTU 8724:2017 [17] using an apparatus that reproduces on a small scale the same process of burst pressure development that occurs in an industrial furnace. The design of the installation is shown schematically in Figure 1 and Figure 2 [16].

The installation consists of an electric furnace, a 20 A transformer, a temperature control unit, a bursting-pressure display unit, a 100 kPa piezoelectric pressure sensor, a bracket for the pressure sensor, and coking retorts made of steel grade St. 3. The retort provides heating up to 880 ± 5 °C. It has the following external dimensions, mm: length—161 ± 1, width—105 ± 1, height—191 ± 1. In its upper part, there is a thick flange for connection to the lid, equipped with a gas outlet (7) with a plate (8) for collecting water and coal tar partially condensed in the gas outlet pipe. The retort is installed in the heating chamber of the furnace with insignificant gaps. The bursting pressure of the coal load is absorbed by the thrust plate, and the pressure from it is transmitted to the pressure sensor through the quartz rod. Since the pressure sensor is stationary (it rests against the adjusting screw), the system “stop plate—quartz rod” is also stationary, although they have freedom of movement. Therefore, the bursting pressure of the coal load develops at a constant volume, i.e., under conditions similar to coking in an industrial furnace. When measuring the bursting pressure, the furnace is first connected to a 220 V network. Line the side walls of the retort coking chamber with a 4 mm thick asbestos sheet. The retort chamber is loaded with coal (batch) in three equal portions slightly pressed so that the loading height is 94 mm.

Table 1. Preparation parameters of the coal sample loaded into the coking chamber.

Loading	Loading Weight, g	Level of Crushing, Class ≤ 3 mm	Moisture, %	Bulk Density, g/cm3
Bulk	600	80 ± 3	10 ± 2	0.80 ± 0.01

shows the required parameters for preparing the coal batch.

Insert a retort with a coal batch into the preheated furnace. Close the upper part of the furnace with the retort with a heat-insulating lid, then bring the pressure sensor close to the end of the quartz rod with the help of a set screw. The pressure value in kilopascals is displayed on the electronic display of the microprocessor-based pressure indicator. The technological properties of coal concentrates are shown in

Table 2. Technological properties of coal concentrates.

Coal Components, Sample Number	Proximate Analysis, %				Plastometric Parameter, mm	Bursting Pressure, kPa
	Wtr, %	Аd, %	Sd, %	Vdaf, %	Y	Рb
DG1	11.6	6.4	1.48	42.2	9	1.5
G1	9.4	7.2	0.29	42.6	13	2.2
G2	8.3	8.1	0.45	39.1	11	3.2
G3	6.9	5.3	1.66	39.6	15	2.8
G4	8.0	6.0	1.24	38.2	14	2.8
G5	6.7	5.9	0.27	33.6	10	3
G6	9.9	7.6	0.48	38	12.5	2
GZh1	8.4	8.3	1.18	32.7	22	7
Zh1	6.4	7.6	0.87	34.2	18	5.2
Zh2	7.2	8.2	0.68	30.6	22	5.2
Zh3	8.2	8.1	0.93	31.9	22.5	5.6
Zh4	7.8	8.6	1.16	31.8	20	7.4
Zh5	7.9	8.3	0.49	32.3	29	5.6
К1	9.1	8	0.69	27.1	14	15.4
К2	6.1	8.5	0.82	27.4	22	7.9
К3	8.8	7.6	0.69	26.2	14	16.3
PS1	8.5	7.3	0.64	18.0	12	23.3

, and the petrographic analysis indicators are shown in

Table 3. Petrographic characteristics of the studied coal concentrates.

Coal Components, Sample Number	Petrographic Composition (Without Mineral Impurities), %					Mean Vitrinite Reflection Coefficient, %
	Vt	L	Sv	I	ΣFC	R0
DG1	65	11	0	24	24	0.59
G1	68	7	0	25	25	0.59
G2	70	3	0	27	27	0.65
G3	67	10	0	23	23	0.73
G4	68	7	0	25	25	0.75
G5	59	2	1	38	39	0.73
G6	79	5	0	16	16	0.91
GZh1	79	5	0	16	16	0.91
Zh1	81	2	0	17	17	1.04
Zh2	84	2	0	14	14	0.97
Zh3	88	2	0	10	10	0.98
Zh4	84	2	0	10	10	0.95
Zh5	85	2	0	11	11	0.9
К1	88	2	0	10	10	1.15
К2	78	1	0	21	21	1.12
К3	86	2	0	12	12	1.15
PS1	78	0	0	22	22	1.53

.

The ash content (A^d) of coal concentrates varied across a wide range: 5.9–14.4%. The content of total sulfur (S^d_t) in coal concentrates was in the range of 0.27–1.66%. The volatile matter (V^daf) on a dry ash-free basis ranged from 32.7 to 42.6% for the gas group, 32.3 to 34.2% for Zh-grade coal, 26.2 to 27.4% for K-grade coal, and 18% for PS-grade coal.

It can be noted (Table 2) that the bursting pressure definitely depends on the properties of the coal, determined by its degree of metamorphism. Given that the nature, degree of metamorphism, transformation of the organic mass of coal, and its behavior during coking change, the highest bursting pressure develops during the coking of coking and lean refractory coal; the minimum values are typical for gas-group coal; and fatty coal, which forms the largest amount of liquid products, is characterized by the maximum thickness of the plastic layer and does not develop a dangerous bursting pressure.

For a deeper assessment of the influence of raw material factors on the bursting pressure, the characteristics of the plastic–viscous properties of coal concentrates were determined by the Gieseler method according to ISO-FDIS 13029 Coal—Determination of plastic properties—Constant-torque Gieseler plastometer method. Technical Committee: ISO/TC 27/SC 5 ICS: 73.040, 2017 [27].

During the experiment, the following temperatures were recorded, °C:

-

The beginning of softening, t₁;

-

Maximum fluidity, t_max;

-

Hardening, t_h;

-

Plasticity range, Δt = t₁ − t_h.

The most important indicator is the maximum fluidity (F_max, ddpm), which characterizes the viscosity of the plastic mass [27]. The properties of plastic masses and the nature of the sintering process have their own characteristics depending on the degree of metamorphism and maceral composition. Low-metamorphosed coal during pyrolysis forms liquid-phase products characterized by low thermal stability and low fluidity. The plastic mass of coal at the middle stage of metamorphism is more homogeneous in composition, contains fewer low-molecular-weight components, and the liquid-phase components have a pronounced plasticizing effect. In turn, the homogeneity of the plastic zone is a factor that determines its gas permeability.

Characteristics of the plastic–viscous properties are shown in

Table 4. Characterization of plastic–viscous properties of coal raw materials.

Coal Components, Sample Number	Plastic Properties by the Gieseler Method
	Fmax, ddpm	t1, °С	tmax, °С	th, °С	Δt, °С
DG1	11	403	428	448	45
G1	189	400	435	453	53
G2	58	396	429	449	54
G3	3601	398	434	470	72
G4	4102	398	435	477	79
G5	4	424	442	451	27
G6	151	397	434	463	66
GZh1	2734	394	442	483	89
Zh1	14,054	395	440	479	84
Zh2	7665	397	451	484	87
Zh3	12,650	392	445	487	95
Zh4	1993	396	445	484	88
Zh5	48,932	381	438	480	99
К1	42	413	452	482	69
К2	8070	388	454	496	108
К3	86	413	449	479	66
PS1	9	447	477	498	51

.

The characteristics of the obtained cokes (indicators of technical analysis, physical and mechanical properties, thermomechanical properties) were determined according to the following standard methods:

-

State standard of Ukraine ISO 579-2002 “Coke-determination of total moisture” [28];

-

ISO 556-80 “Coke (greater than 20 mm in size)-determination of mechanical strength” [29];

-

ISO 18894:2006 “Coke-determination of coke reactivity index (CRI) and coke strength after reaction (CSR)” [30].

The statistical analysis of the results and the development of mathematical dependencies were performed using the licensed computer program Microsoft Excel. The following statistical estimates were used to evaluate the obtained equations: r—multiple correlation coefficient; D—coefficient of determination, %; σ—standard deviation, units; and cv—coefficient of variation, %.

3. Results

3.1. Formation of Bursting Pressure Depending on Properties of Coal Concentrates

After evaluating the data obtained, it can be stated that the maximum plastic-viscosity characteristics were recorded for coal of grade G, and the minimum—for coking and low-sintering coals. Thus, analyzing the plastic-viscosity properties of coal raw materials, it should be noted that for gas group coals (grades DG, G and GZh), the maximum fluidity F_max was in the range from 4 to 4102 scale divisions with a plastic layer thickness of 13 mm on average, and a small plasticity interval Δt_mean = 60 °C was also recorded; for fatty coal, the thickness of the plastic layer was on average 22 mm, and F_max varied in the range from 1993 to 14054 scale divisions with a fairly wide plasticity interval, which averaged Δt_mean = 90 °C, for coking coal (grade K) with a sufficiently high sintering (the thickness of the plastic layer was 14 mm and 22 mm), low yield values were obtained, respectively, 86 to 8070 scale divisions, the lean sintered coal has the lowest yield value of 9 and Δt = 51 °C.

Thus, in the case of coal of the gas group, a small amount of plastic mass formed during coking is heterogeneous, quickly begins to solidify, so steam and gas products can freely pass through the heterogeneous layer, as a result of which the burst pressure has minimal values (P^b_mean = 2.5 kPa). For fatty coal, a large amount of plastic mass is formed, its homogeneity causes complications in the conditions of evacuation of volatile products, but due to its high fluidity it does not create much resistance and does not lead to the development of a critical burst pressure (P^b_mean = 5.9 kPa). Coking coal is also characterized by the formation of a sufficient amount of liquid products, but at the same time, low fluidity, high viscosity and, as a result, low gas permeability of the plastic mass are observed, which leads to an increase in pressure inside the plastic zone and causes critical values of the bursting pressure (P^b_mean from 7.9 to 25.3 kPa).

Using regression analysis tools, the relationships between the bursting pressure of coal concentrates and the temperatures of the onset of softening, t₁, and maximum yield, t_max, were established (correlation coefficients are shown in

Table 5. Correlation coefficients between the bursting pressure and plastic–viscous properties of coal raw materials.

	Рb, kPа	Fmax, ddpm	t1, °С	tmax, °С	th, °С	Δt, °С
Рb, kPa	1
Fmax, ddpm	−0.12	1
t1, °С	0.66	−0.5	1
tmax, °С	0.87	−0.08	0.62	1
th, °С	0.63	0.26	0.04	0.76	1
Δt, °С	−0.01	0.546	−0.68	0.12	0.72	1

).

These indicators of the plastic–viscous properties of coal depend on the degree of its metamorphism. The high correlation coefficients between P^b and t₁, t_h, and t_max (r = 0.66, r = 0.63, and r = 0.87, respectively) confirm the close connection and influence of the nature and degree of metamorphism on the development of the burst pressure. In addition, a characteristic feature of these indicators is their additivity, which makes their use promising for optimizing the composition of batches and predicting the bursting pressure. Graphical dependences are shown in Figure 3 and Figure 4.

We have also established relationships between petrographic and technological indicators of coal raw material quality and bursting pressure (Table 2, Table 3 and Table 4), which are characterized by the corresponding correlation coefficients (presented in

Table 6. Correlation coefficients between bursting pressure and genetic and technological characteristics of coal raw materials.

	Wtr	Аd	Sd	Vdaf	Y	I	Vt	Рb	R0	ΣFC
Wtr	1
Аd	−0.06	1
Sd	0.05	−0.32	1
Vdaf	0.28	−0.42	0.21	1
Y	−0.44	0.6	0.06	−0.26	1
I	−0.16	−0.57	0.05	0.16	−0.55	1
Vt	0.06	0.71	−0.28	−0.35	0.58	−0.95	1
Рb	0.002	0.27	−0.15	−0.9	−0.02	−0.24	0.37	1
R0	−0.22	0.41	−0.14	−0.95	0.23	−0.22	0.6	0.89	1
ΣFC	−0.21	−0.61	0.03	0.14	−0.5	0.96	−0.93	−0.2	−0.23	1

).

Thus, the influence of genetic and technological properties of coals on the formation of their bursting pressure during coking was established. The close relationship is confirmed by the high correlation coefficients between P^b and Vt, V^daf, and R₀ (r = 0.6, r = 0.9, and r = 0.89, respectively). The corresponding graphical dependences of the bursting pressure on the volatile matter (V^daf) and the vitrinite reflectance (R₀) are linear in nature, as shown in Figure 5, Figure 6 and Figure 7.

The obtained mathematical dependences of the influence of the yield of the above raw material characteristics on the value of the bursting pressure and their statistical estimates are shown in

Table 7. Mathematical equations.

№	Mathematical Equations	Statistical Assessment
		r	D, %	σ, un.
(1)	${\mathrm{Р}}^{\mathrm{b}}=0.2594·{\mathrm{t}}_1-97.413$	0.66	43.8	4.65
(2)	${\mathrm{Р}}^{\mathrm{b}}=0.4455·{\mathrm{t}}_{\mathrm{m}\mathrm{a}\mathrm{x}}-190.48$	0.87	76.02	3.04
(3)	${\mathrm{Р}}^{\mathrm{b}}=0.0334·{\mathrm{V}}^{\mathrm{d}\mathrm{a}\mathrm{f}2}-2.9585·{\mathrm{V}}^{\mathrm{d}\mathrm{a}\mathrm{f}}+66.97$	0.95	89.66	4.19
(4)	${\mathrm{Р}}^{\mathrm{b}}=19.409·{\mathrm{R}}_0^2-{16.749·\mathrm{R}}_0+4.7534$	0.92	85.49	2.86
(5)	${\mathrm{Р}}^{\mathrm{b}}$$=0.45054 ·$ Vt − 23.868	0.6	33.03	5.77

.

After evaluating the presented equations, it can be argued that the most accurate of the studied coal quality indicators, the bursting pressure, is predicted by the maximum fluidity temperature, the volatile matter, the vitrinite reflection index, and (Equations (2)–(4)). The mathematical models are characterized by high correlation coefficients r (0.87, 0.95, and 0.92, respectively).

It was found that an increase in the temperature at the beginning of coal softening and the temperature of maximum fluidity by 1 °C leads to an increase in pressure by 0.2594 kPa and 0.4455 kPa, respectively. When increasing the content of volatile matter by 1%, a decrease in the bursting pressure by about 0.8 kPa was observed. When increasing the degree of carbonization by R₀ by 0.1%, an increase in the bursting pressure by about 2 kPa was recorded.

Upon analyzing the calculated values of the coefficients of variation, we can note the homogeneity of the sample in terms of V^daf, Vt, R₀, t₁, and t_max.

3.2. Prediction of the Bursting Pressure of Coal Mixtures and Its Influence on Physical and Mechanical Properties of Coke

Coal blends of various component compositions were prepared from the studied coal concentrates (data are given in

Table 8. Component composition of coal batches.

Batch Samples	Coal Components
	G	Zh	К	PS
1	45	2	52	1
2	46	3	50	1
3	45	2	52	1
4	46	2	51	1
5	40	6	53	1
6	38	4	57	1
7	43	4	52	1
8	44	4	52	-
9	32	13	54	1
10	31	11	57	1
11	35	10	54	1
12	37	9	53	1

). The main characteristics and technological properties of the blends are given in

Table 9. Qualitative characteristics of coal batch samples.

Batch Samples	Wtr, %	Аd, %	Sd, %	Vdaf, %	Y, mm	Ro, %	Vt, %	Рb, kPa	Class Content ˂3 mm,%
1	10.3	7.7	1.04	32.6	14	0.95	79	7.0	80
2	9.2	7.7	1.10	32.8	13	0.94	80	6.6	80
3	9.5	7.8	1.08	32.6	14	0.95	79	6.8	81
4	9.0	7.0	1.14	33.6	14	0.91	75	6.7	80
5	9.0	7.9	1.11	32.6	15	0.95	79	7.1	81
6	8.8	7.9	1.04	32.1	14	0.97	80	8.1	81
7	10.3	7.9	1.06	32.6	13	0.95	81	7.1	81
8	10.3	7.8	1.05	32.8	14	0.94	79	6.9	81
9	8.5	7.8	0.88	31.8	14	0.98	82	7.9	80
10	8.7	8.2	1.44	31.9	16	0.98	81	8.1	79
11	9.0	7.4	1.23	31.8	15	0.99	80	7.6	79
12	9.3	8.0	1.44	32.5	16	0.96	79	7.3	79

.

The main characteristics and technological properties of the blends are presented in Table 9.

We also investigated the relationships between petrographic and technological indicators of the quality of the experimental coal batches and the fracture pressure, which are characterized by the corresponding correlation coefficients (

Table 10. The correlation coefficients between the bursting pressure and technological and petrographic characteristics of coal batchеs.

	G	Zh	К	Wtr	Аd.	Sd	Vdaf	Class Content ˂3 mm	Y	Ro	Vt	Рb
G	1
Zh	−0.94	1
К	−0.81	0.57	1.00
Wtr	0.78	−0.70	−0.64	1.00
Аd	−0.41	0.31	0.48	−0.28	1.00
Sd	−0.34	0.31	0.28	−0.17	0.26	1.00
Vdaf	0.85	−0.77	−0.75	0.68	−0.53	−0.10	1.00
class content ˂3 mm	0.57	−0.65	−0.26	0.33	0.00	−0.74	0.26	1.00
Y	−0.64	0.62	0.50	−0.50	0.32	0.78	−0.36	−0.67	1.00
Ro	−0.86	0.78	0.74	−0.70	0.48	0.21	−0.99	−0.35	0.43	1.00
Vt	−0.66	0.66	0.49	−0.51	0.69	−0.05	−0.84	−0.09	0.09	0.79	1.00
Рb	−0.90	0.73	0.95	−0.70	0.43	0.18	−0.84	−0.35	0.43	0.83	0.62	1.00

).

The ash content (A^d) of the coal batches was in the range of 7.0–8.2%. The content of total sulfur (S^d_t) in the batch was in the range of 0.88–1.44%. The volatile matter (V^daf) on a dry ash-free basis ranged from 31.8 to 33.6%. The grindability of coal batchеs in the ˂3 mm size class was in the range of 79–81%. According to the petrographic characteristics, the coal blends had the following indicators: vitrinite content 79–82% and vitrinite reflection index 0.94–0.99%.

The characteristics of the obtained cokes are presented in

Table 11. Characteristics of coke produced.

Batch Samples	Wtr, %	Аd, %	Sd, %	М25, %	М10, %	CRI, %	CSR, %
1	3.5	10.4	0.66	87.0	7.2	40.2	43.3
2	3.5	10.4	0.72	86.4	7.6	41.1	42.7
3	3.4	10.7	0.67	86.4	7.6	41.0	41.7
4	4.6	10.4	0.7	85.9	8.0	43.5	38.3
5	4.7	10.6	0.7	88.0	6.9	41.1	41.8
6	4.7	10.4	0.66	90.0	6.2	40.2	43.0
7	4.8	10.4	0.69	87.7	7.4	41.5	41.7
8	4.7	10.4	0.68	86.0	7.8	40.1	43.4
9	4.8	10.6	0.67	89.0	6.8	37.1	47.6
10	4.8	10.6	0.74	89.0	6.8	39.7	44.6
11	4.8	10.9	0.74	88.0	6.9	38.3	45.7
12	4.8	10.7	0.71	87.5	7.0	40.1	43.4

.

Thus, it was found that mixtures are also characterized by the influence of genetic and technological properties on the formation of their bursting pressure during coking. The close relationship is confirmed by the high correlation coefficients between P^b and V^daf and Vt (r = 0.84 and r = 0.62, respectively). The corresponding graphical dependences of the bursting pressure on the volatile matter (V^daf), the vitrinite content (Vt), and mean vitrinite reflection coefficient (R₀) are linear in nature, as shown in Figure 8, Figure 9 and Figure 10.

The equations that describe the resulting graphical dependencies are shown in

Table 12. Mathematical equations.

№	Equations	Statistical Assessment
		r	D, %	σ, un.
(6)	${\mathrm{Р}}^{\mathrm{b}}=0.4419·{\mathrm{V}}^{\mathrm{d}\mathrm{a}\mathrm{f}2}-29.669·{\mathrm{V}}^{\mathrm{d}\mathrm{a}\mathrm{f}}+504.56$	0.89	79.58	0.3
(7)	${\mathrm{Р}}^{\mathrm{b}}$$=0.1957·$ Vt − 8.3183	0.62	38.68	0.44
(8)	${\mathrm{Р}}^{\mathrm{b}}=163.19·{\mathrm{R}}_0^2-290.95·{\mathrm{R}}_0+136.19$	0.85	72.63	0.31

.

Upon evaluating Equations (6) and (8), it can be argued that they are characterized by rather high correlation coefficients r (0.89 and 0.85, respectively).

To test the possibility of predicting the rupture pressure of coal mixtures, we used regression Equations (9)–(11) that we developed from our study of coal concentrates (volatiles, vitrinite reflectivity, and vitrinite consistency).

These equations are shown in

Table 13. Regression equations.

№	Equations	Statistical Assessment
		r	D, %	σ, un.
(9)	${\mathrm{Р}}^{\mathrm{b}}=-0.829·{\mathrm{V}}^{\mathrm{d}\mathrm{a}\mathrm{f}}-0.071·{\mathsf{\gamma }}_{\left(3-0\mathrm{m}\mathrm{m}\right)}+39.878$	0.86	73.5	0.31
(10)	${\mathrm{Р}}^{\mathrm{b}}=-0.91385·{\mathrm{V}}^{\mathrm{d}\mathrm{a}\mathrm{f}}-1.16944·{\mathrm{R}}_0+38.04789$	0.84	71	0.32
(11)	${\mathrm{Р}}^{\mathrm{b}}=-1.243·{\mathrm{V}}^{\mathrm{d}\mathrm{a}\mathrm{f}}-0.094·\mathrm{V}\mathrm{t}+51.28$	0.85	73	0.3

.

The results based on the actual measurement data of P^b, the values calculated by the additivity rule, and Formulas (9)–(11), are presented in

Table 14. The values of the bursting pressure according to actual measurements, calculated by the additivity rule and calculated by Formulas (9)–(11).

Batch Samples	Actual Values of Pb According to the Measurement Results, kPa	Pb Values Calculated by the Additivity Rule, kPa	The Values of Pb Calculated by Equation (9), kPa	The Values of Pb Calculated by Equation (10), kPa	The Values of Pb Calculated by Equation (11), kPa
1	7.00	11.07	7.2	7.1	3.3
2	6.60	11.80	7	7.0	3.0
3	6.80	12.11	7.1	7.1	3.3
4	6.70	12.07	6.3	6.3	2.5
5	7.10	10.66	7.1	7.1	3.3
6	8.10	11.38	7.5	7.6	3.9
7	7.10	12.01	7.1	7.1	3.1
8	6.9	11.93	6.9	7.0	3.1
9	7.9	11.1	7.8	7.8	4.0
10	8.1	11.4	7.8	7.8	4.0
11	7.6	13.1	7.9	7.8	4.1
12	7.3	10.4	7.3	7.2	3.5

.

A comparison of the results of measurements and calculations of P^b is shown in Figure 11.

Comparing the results of calculations with the actual measurement results, it should be noted that there is no additivity of the P^b indicator in multicomponent coal mixtures. Moreover, the value of the bursting pressure P^b calculated by the additivity rule is higher in all blends than the experimental values. Equation (11), taking into account the vitrinite content, gives inaccurate results; the obtained P^b values are also significantly lower than the actual ones. The best prediction result was recorded when using mathematical models that take into account the volatile yield as the main parameter and additional factors such as vitrinite content and batch grinding. Thus, Equations (9) and (10) can be recommended for predicting the bursting pressure.

3.3. Influence of Tensile Pressure on Coke Quality

The effect of the bursting pressure on the metallurgical properties of coke in terms of mechanical strength (M₂₅, M₁₀) was evaluated using mathematical statistics. Thus, the following correlation coefficients were obtained (

Table 15. Correlation coefficients between bursting pressure and quality characteristics of cokes from batches.

	Рb, kPa	М25, %	М10, %
Рb, kPa	1
М25,%	0.934805	1
М10,%	−0.87333	−0.94906	1

).

Graphical dependencies are shown in Figure 12 and Figure 13.

The equations describing the obtained graphical dependences are given in

Table 16. Mathematical dependencies.

№	Equations	Statistical Assessment
		r	D, %	σ, un.
(12)	${\mathrm{М}}_{25}=2.2559·{\mathrm{Р}}^{\mathrm{b}}+71.185$	0.93	87.39	0.2
(13)	${\mathrm{М}}_{10}=-0.8379·{\mathrm{Р}}^{\mathrm{b}}+13.272$	0.87	76.27	0.27

.

It should be noted that a close relationship between the bursting pressure and the physical and mechanical properties of coke for blast furnace melting has been confirmed. The mathematical models are linear in nature and have high statistical estimates (correlation coefficients, r, are 0.93 and 0.87).

An increase in the burst pressure indicates a complicated gas release, which increases the time spent by thermal degradation products in the plastic state and deepens the interaction between them. This can lead to the formation of additional liquid products, the softening of coal grains, and more complete contact between them, which leads to an increase in the mechanical strength of coke. Thus, with a 1 kPa increase in the bursting pressure, we can expect an increase in the mechanical strength of coke in terms of crushability M₂₅ by about 2.6% and a decrease in the abrasion of coke M₁₀ by 1%.

4. Discussion

The results obtained in this work are in line with the conclusions and statements given in [19,31,32].

In [32], the authors studied the possibility of reducing the bursting pressure by fine grinding coal, which is capable of developing a high bursting pressure, and the effect of fine grinding inert coal on coke strength. According to the results of the research, it was found that the fine grinding of coal, which is characterized by a high bursting pressure, causes an increase in the total surface of coal particles, which increases the gas permeability of the plastic layer, which leads to a decrease in bursting pressure and an improvement in coke strength in terms of ${\mathit{DI}}_{50}^{150}$. It is noted that the improvement in coke strength is associated with the fine grinding of inertinite grains (≤1 mm).

In [33], it was found that an increase in the degree of grinding of the batch with a significant content of lean (fusinized) components (ΣОК = 19–21%, R_o = 0.94–0.95%) from 82.7 to 90.3% leads to a slight increase in the fluidity of the plastic mass (from 100 to 125 ddpm) and the bursting pressure (from 3.4 to 3.7 kPa). This is due to a more uniform distribution of petrographically heterogeneous coal grains in the coal mass and results in a 0.6% increase in coke strength by M₂₅. In the case of increased grinding for coal blends with high sintering capacity (fatty coal content ≥ 70%, ΣОК = 11%. R_o = 1%), a significant decrease in the fluidity of the plastic mass (from 335 to 135 ddpm), an increase in its viscosity and, as a result, an increase in the bursting pressure from 4.2 to 7.4 kPa were observed. At the same time, the crushability index increased by 1.8%, and the abrasion resistance decreased by 0.8%. Thus, the increase in the viscosity of the plastic mass makes it difficult to evacuate the vapor-gas destruction products and increases the time they stay in the plastic zone, which causes an increase in the bursting pressure.

It should be noted that despite the established positive effect of the bursting pressure on the physical and mechanical characteristics of coke, to ensure the normal operation of coke ovens, it is necessary to control and prevent the pressure from rising above 7 kPa.

To preserve the furnace stock and produce blast furnace coke with the required quality parameters with the available coal base, various methods and techniques are used to improve certain quality parameters of the main product and ensure optimal operating conditions for the equipment. The main mechanism is the optimization of the composition of coal blends, taking into account the bursting pressure and the degree of oxidation of each component of the mixture.

Additionally, targeted control can be used by introducing non-sintering additives into coal batches: coke dust and fines of anthracite, semi-coke, as well as inorganic additives (fine oxides of titanium, silicon, iron, and aluminium) [4].

With the introduction of desulphurizing additives, the plastic mass becomes heterogeneous, so the gases formed during coking are easier to evacuate, which reduces their pressure on the chamber walls.

Thus, work [3] presents the results of studying the effect of coke fines on the bursting pressure. It is shown that this additive is a rather effective means of reducing the bursting pressure of both bulk and stamped batchеs. The addition of coke fines (<3.0 mm) to the bulk batch in the amount of up to 3.0% can reduce the bursting pressure by 12–15%. The bursting pressure of stamped batchеs is also reduced by the same amount. Adding finely crushed (<0.25 mm) coke fines to the stamped batch in an amount of up to 7.0% reduces the bursting pressure by 45–46%. Similar results are achieved when coke dust with a similar particle size distribution (100% < 0.25 mm) is added to the batch. The advantage of this solution compared to increasing the content of gas-coal in the batch is a much smaller amount of additive, which makes it possible to practically preserve the sintering properties of the batch, as well as to slightly increase the yield of the target product (coke).

Paper [34] describes the effect of additions of coal with a high volatile content (35.8%) and semi-anthracite to three different grades of sintering coal, which are characterized by low or medium volatile yields (17.8%, 22.6%, and 17.5%). The organic additives proved to be very effective in reducing the coking pressure of low refractoriness coals (coal A, for which the plastic layer thickness was Y = 8 mm). But at the same time, a decrease in coke strength was recorded. Thus, the ${\mathit{DI}}_{50}^{150}$ index decreased from 82.1 to 76.6%. In the case of the addition of coal with a high volatile content (HV) to coal (B) with a better sintering ability (Y = 16.5 mm), a smaller decrease in coking pressure was observed, but an increase in coke strength from 85.65 to 86.3% was noted.

In [35], it is proposed to determine the bursting pressure using the Redlich–Kwong equation of the state of a real gas, taking into account the component composition of steam and gas products, which depends on the nature of the coal (degree of metamorphism) and heating conditions. Thus, it is noted that according to the calculations, the maximum bursting pressure is typical for coals of technological grades coking K and lean sintering PS. However, the equation is cumbersome, and its use is complicated by the need to determine the component composition of steam and gas products for the components of the batch, which is not always possible in the laboratories of coke plants.

It is also necessary to take into account that the bursting pressure depends significantly on the oxidation of coals and coal blends [36], which may be the subject of a special study.

The authors of [3] proposed a scheme for calculating the bursting pressure of a multicomponent mixture based on the developed types of bursting-pressure dependencies in binary mixtures. According to the methodology, it is first necessary to determine the weighted average bursting pressure within each coal grade included in the batch. In this case, the bursting pressure of coal belonging to the same grade, regardless of the coal deposit in binary mixtures, follows the additivity rule. The relative percentage of the required coal grades in the batch is then calculated. When a new coal with an unknown bursting pressure is introduced, it becomes necessary to prepare experimental blends and determine their bursting pressure.

The use of the proposed Equations (9) and (10) makes it possible to predict the bursting pressure using the data on the properties of coal concentrates, which are necessarily determined at a coke plant to control the quality of raw materials.

5. Conclusions

The aim of this study was to investigate the factors influencing the formation of the bursting-pressure value, which will make it possible to plan and adjust the component composition of the blends of production batches of coal accordingly.

The interrelation of the bursting pressure with the component composition of the coal mixture, the volatile matter of the batch, and petrographic characteristics (vitrinite content) was confirmed.

The possibility of predicting the bursting pressure of coal blends with regard to the volatile matter and the vitrinite reflection index is shown. The use of the proposed Equations (9) and (10), which are characterized by high correlation coefficients (0.86 and 0.84), makes it possible to predict the bursting pressure and optimize the composition of coal batches using the data on the properties of coal concentrates, which must be determined at a coke plant to control the quality of raw materials. The use of the equation decreases the number of experimental measurements of bursting pressure in the conditions of a particular coke production.

The mechanism of influence of the bursting pressure on the strength characteristics of coke was confirmed. Thus, an increase in the bursting pressure causes an increase in the amount of liquid products, deepening their interaction with softened coal grains, which improves the sintering and coking properties of the batch and, as a result, increases the mechanical strength of coke. It was also found that an increase in the bursting pressure by 1 kPa can be expected to increase the mechanical strength of coke in terms of crushability M₂₅ by about 2.6% and reduce the abrasion of coke M₁₀ by 1%.

The next step in the research will be to check the impact of raw materials and technological factors on the distribution pressure for ramming the charge to study the possibility of predicting its value and to develop additional criteria for compounding coal mixtures using ramming technology.