Formation of Improved Metallurgical Properties and Carbon Structure of Coke by Optimizing the Composition of Petrographically Heterogeneous Interbasin Coal Batches

Abstract

Given the multi-basin raw material base for coking that has been formed at most industry enterprises, there is an urgent need to optimize the component composition and improve the basic technological methods of coal raw material preparation, taking into account the petrographic characteristics of coal batches. A comprehensive study of the components included in a coke chemical enterprise’s coking raw material base was carried out. The work used standardized methods for studying coal and coal batches’ technological and plastic–viscous properties. The qualitative characteristics of coke were determined using physical–mechanical and thermochemical methods of studying standardized indicators: crushability (M₂₅), abrasion (M₁₀), reactivity (CRI), post-reaction strength (CSR), and specific electrical resistance (ρ). The results were analyzed using the licensed Microsoft Excel computer program. Based on the results of proximate, plastometric, and petrographic analyses of the studied coal samples and data from experimental industrial coking, proposals were made to optimize the component composition, properties of the coal batch, and technology for its preparation for coking. The established inverse dependence of Gibbs free energy (ΔG_f,total) on the reaction capacity of coke CRI and its direct reliance on its post-reaction strength CSR confirmed the feasibility of using ΔG_f,total as a thermodynamic predictive parameter for optimizing and compiling coal batches that produce less reactive, stronger coke. This made it possible to improve the quality indicators of metallurgical coke. Thus, according to the M₂₅ crushability index, the mechanical strength increased by 0.6%, and the M₁₀ abrasion decreased by 0.4%. Significant improvements in thermochemical properties and an increase in the orderliness of the carbon structure were recorded: the CRI reactivity decreased by 3.1%, the CSR post-reaction strength increased by 8.3%, and the specific resistance decreased by 8.4%.

1. Introduction

At present, blast furnace production sets higher requirements for coke quality, the standards of which are fixed both in the current technical specifications (A^d, S^d_t, M₁₀, and M₂₅) and in ISO 18894-2006 (CRI and CSR). Thus, the current requirements for coke quality are M₂₅ ≥ 88.0–90.0%, M₁₀ ≤ 6.0–6.5%, CSR—60.0–75.0%, and CRI—25–30.0%.

Analysis of coal resources shows that the domestic raw material base cannot produce blast furnace coke of improved quality on its own. Consequently, the question of attraction and rational use of imported components (coals from far abroad) in coal batches for coking arises [1,2,3,4,5,6].

For coking of imported raw coal, which differs in technological properties and petrographic characteristics and comes from different countries, basins, and continents, the composition of the coal batch according to the ranks of coal does not reflect the actual composition of the mixture and its characteristics and does not fully take into account the peculiarities of the composition or properties of specific components of the organic mass of coal. Classification of distant foreign coal and its aggregation into classes and groups [1] differs from the accepted classification of ranks in accordance with DSTU 3472-2015 [7]. Therefore, the use of imported coal is complicated by different approaches to classification and inconsistencies between domestic and foreign regulatory documents.

An essential factor that necessitates the consideration of petrographic characteristics when compiling blends is that most enrichment plants in Ukraine do not have a permanent coal raw material base. As a result, they are forced to enrich coal of two or more ranks, which leads to mutual contamination and a decrease in the technological value of the obtained concentrate [8]. In this regard, the declared rank of concentrates often does not correspond to reality, which complicates the production of coke of planned quality.

Therefore, it is obligatory to formulate the coal mixture taking into account the maceral composition, the indicator of vitrinite reflection, the distribution of vitrinite components by stages of metamorphism, and the optimal ratio between clinkering (ƩCC) and lean (fusinized) components (ƩFC). When predicting the output and properties of coke, we should consider that the macerals of the groups of vitrinite, liptinite, and one-third of semivitrinite constitute the sum of reactive and clinkering components. In contrast, the whole of inertinite and two-thirds of semivitrinite have an oppressive effect on the system [9].

Various models have been developed to optimize charge composition and predict coke quality. The first generation of models focuses on predicting cold mechanical strength indicators (i.e., ASTM stability and indices obtained when processing coke in a MICUM drum). The second-generation models use CRI and CSR as coke quality parameters. No universal models for predicting coke quality have been developed due to the specific characteristics of coal from different coal basins, countries, and continents, which can vary significantly in composition, structure, and properties.

Analyses of mathematical models for predicting coke quality considering the properties of coal have been conducted by steel companies and research institutes and are presented in [10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26]. The main variables considered in some mathematical models are presented in

Table 1. CSR prediction models and the variables they consider.

Company /Research Institute/ Researchers	Variable Characteristics of Coal
	Petrographic Characteristics	Rheological Properties of Coal in the Plastic State	Characteristics of the Inorganic Part	Others
British Steel	Vitrinite reflectance		Coke ash
Nippon Steel	Inertinite content, %, vitrinite reflectance	Maximum fluidity	Ash basicity index
BCRA	Inertinite content, %	Maximum fluidity	Content of basic oxides in coal	Carbon and oxygen content in coal, coke, and porosity
Kobe Steel	Vitrinite reflectance	Maximum fluidity	Ash basicity
BHP	Content of inert components	Maximum fluidity	Coal ash basicity	Yield of volatile matter
Inland Steel		The temperature of transition of coal into a plastic state	Ash basicity index	Sulfur content in coal
ISCOR	Vitrinite reflectance, content of fusite components	Maximum fluidity	Content of basic oxides in coal ash
Tata Steel	Vitrinite reflectance, the ratio between cohesive and inert macerals, reflectogram of the vitrinite component	Maximum fluidity, free swelling index	Ash, SiO2/Al2O3 ratio, alkali index	Composite coking potential (CCP), type of coke residue according to the Gray–King method

.

Analyzing the data in the table, most mathematical models consider the vitrinite reflectance and the content of inert (fused) components, the maximum fluidity, and the basicity of coal ash.

Given the multi-basin raw material base for coking, which determines differences in the technological properties and material composition of coal concentrates, it is necessary to refine and improve the basic technological methods of preparation when using them in coal batches. The study aims to develop recommendations for obtaining blast furnace coke of stable quality while optimizing the composition, properties, and degree of crushing of the batch, taking into account its petrographic characteristics.

When blending coal to obtain coke of a particular strength, including the maximum possible strength, the criterion for the technological properties of the blend that meets these requirements should not be the rank (quality) but rather the optimality, based on the calculation of specific values of the properties of the charge according to the properties of its components. One such unique indicator could be the value of the Gibbs free energy of formation (ΔG_f,total). The uniqueness and practical value of introducing the thermodynamic index (ΔG_f,total) lie in the fact that it considers the content of vitrinite and inertinite and the reflection coefficients for all maceral types. This approach allows the overall thermodynamic stability of the coal mixture to be assessed using a single integral value.

2. Materials and Methods

Samples were collected and prepared per 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” [27].

The research methods used to study the properties of coal raw materials and coke are presented in

Table 2. Methods used to characterize materials.

Materials	Indicator	Methods
Characteristics of coal raw materials (coals, batches)	Ad	ISO 1171-97 “Solid mineral fuels. Methods for the determination of ash”	[28]
	Wa	ISO 589-81 “Hard coal—Determination of total moisture”	[29]
	Vd, Vdaf	ISO 562:2024 “Hard coal and coke—Determination of volatile matter “	[30]
	Sdt	ISO 334:2020; Coal and Coke. Determination of Total Sulfur. Eschka Method	[31]
	Ro, Vt, L, Sv, I	ISO 7404-3-84 “Methods for the petrographic analysis of bituminous coal and anthracite—Part 3: Method of determining maceral group composition”; ISO 7404-5-85 “Methods for the petrographic analysis of coals—Part 5: Method of determining microscopically the reflectance of vitrinite”	[32,33]
	X, Y	State standard of Ukraine 7722:2015 “Hard coal. Method of Determining Plastometric Characteristics”	[34]
	RI	State standard of Ukraine 7723:2015 Coal. Determination of the sintering index by the Rog method	[35]
	Fmax, t1, t2, Δt	ISO-FDIS 10329 Coal—Determination of plastic properties—Constant-torque Gieseler plastometer method	[36]
	tI, tII, tIII, a, b	ISO 349:2020 Hard coal—Audibert–Arnu dilatometer test	[37]
	FSI	ISO 501:2012 Hard coal. Method for determining the free swelling index	[38]
	G	ISO 502:2025 Hard coal—Determination of caking power—Gray–King coke test	[39]
Characteristics of coke	M25 M10	ISO 556-80 “Coke (greater than 20 mm in size)—Determination of mechanical strength”	[40]
	CRI CSR	SO 18894:2006 “Coke—Determination of coke reactivity index (CRI) and coke strength after reaction (CSR)”	[41]
	ρ	State standard of Ukraine 8831:2019 Coke. Method for the determination of the resistivity of coal coke powder	[42]

.

The degree of crushing of the coal batch according to the above method [43], depending on its petrographic characteristics and sinterability, can be determined by the following equation:

$$k={\displaystyle \frac{75\cdot Vt\cdot \frac{\sum \left(0.9-1.39\right)}{100}+90\cdot Vt\cdot \frac{\sum \left(0.5-0.89\right)+\sum \left(1.4-2.6\right)}{100}+90\cdot \sum FC}{100}},\%$$

(1)

where Σ (0.5–2.6) is the content of the vitrinite components with reflectance in the range 0.5–2.6%, corresponding to different coal ranks; 75 is the recommended degree of crushing of Zh and K coal, %; 90 is the recommended degree of crushing of G, PS, and T coal, %; and ΣFC is the sum of fusinized components (I + 2/3 Sv), %.

The variants of experimental batches were developed to test them by box coking. Box coking was carried out using the following methodology. Experimental batches weighing 5.5 kg were loaded into metal crates. With the crate volume of 7.3 dm³, the density of the experimental batch in them was approximately 0.760 t/m³, which corresponds to the average practical data on the density of the batch in the chamber.

Photos of the boxes are presented in Figure 1.

Coking of experimental batches was carried out on coke batteries (adequate volume of the chamber: 21.6 m³) in specially selected ovens with good condition of masonry of heating partitions and a stable thermal regime.

Loading of boxes was carried out through the middle hatch after the release of the batch from the coke and machine hoppers of the coal loading car so that the boxes with the batch were located at a level of approximately 2.5 m from the bottom of the coking chamber, that is, in the zone of optimal temperature conditions. Four boxes were loaded into each chamber according to the number of experimental batches. Experimental coking was carried out under the thermal regime, the parameters of which are presented in

Table 3. Coking conditions for coal batches.

Coking Period, Hours	Temperatures in the Control Verticals, °C		Coke Cake Axis Temperature, °C
	Machine Side	Coke Side
15.5	1320	1350	1100

.

Once the coking process was complete, the cake was delivered using standard technology (from the chamber to the quenching car, then quenching and unloading onto the coke slide). If necessary, the boxes were quenched with water and pre-cooled in air.

The obtained experimental coke was extracted from the crates, weighed, and dried to constant weight to determine the yield of gross coke from the batch and for subsequent analyses.

3. Results

3.1. Detailed Study of American Coal

To determine the optimal conditions for preparation and coking of a batch with a large content of petrographically heterogeneous coal in the laboratory and experimental industrial conditions, studies were conducted to determine their rational rank and component composition, preparation schemes, and coking mode, as well as to assess the impact of these measures on the quality of blast furnace coke (in terms of mechanical strength and specific electrical resistance).

Carrying out the above measures is especially relevant in coke chemical enterprises in Ukraine, as the raw material base contains coal from different basins.

Samples of coal concentrates, which constitute the raw material base of the coke plant, and production batches were selected and studied. These samples’ technological properties, petrographic characteristics, and plastic–viscous properties were determined. The obtained data are presented in

Table 4. Technological properties of coal samples and batch.

Sample	Rank of Coal DSTU 3472-2015	Proximate Analysis, %				Plastometric Parameters, mm
		Ad	Sdt	Vd	Vdaf	X	Y
A	G	9.4 ± 0.2	0.32 ± 0.05	37.3 ± 0.3	41.2 ± 0.3	35 ± 3	14 ± 1
B (coal USA)	Zh	6.9 ± 0.2	1.03 ± 0.05	30.8 ± 0.3	33.1 ± 0.3	17 ± 3	20 ± 1
C	Zh	9.1 ± 0.2	0.54 ± 0.05	29.8 ± 0.3	32.7 ± 0.3	31 ± 3	19 ± 1
D	K/KO/KZh	12.2 ± 0.2	0.67 ± 0.05	21.5 ± 0.3	24.5 ± 0.3	28 ± 3	15 ± 1
E	K	8.8 ± 0.2	0.72 ± 0.05	26.6 ± 0.3	29.2 ± 0.3	17 ± 3	12 ± 1
F	K	12.8 ± 0.2	1.75 ± 0.05	24.0 ± 0.3	27.5 ± 0.3	36 ± 3	14 ± 1
G	PS	9.9 ± 0.2	0.40 ± 0.05	19.0 ± 0.3	21.1 ± 0.3	16 ± 3	11 ± 1
H	K	8.6 ± 0.2	0.41 ± 0.05	18.2 ± 0.3	20.0 ± 0.3	25 ± 3	11 ± 1
Base batch	-	10.4 ± 0.2	0.83 ± 0.05	26.5 ± 0.3	29.6 ± 0.3	34 ± 3	13 ± 1

,

Table 5. Petrographic characteristics of the samples studied.

Sample	Petrographic Composition (Without Mineral Impurities), %					Mean Vitrinite Reflection Coefficient, %
	Vt	Sv	I	L	Σ FC	Ro
A	85 ± 4	-	12 ± 2	3 ± 2	12	0.69 ± 0.02
B (coal USA)	73 ± 5	-	22 ± 5	5 ± 2	22	1.02 ± 0.02
C	79 ± 5	-	20 ± 5	1 ± 2	20	0.94 ± 0.02
D	91 ± 4	-	8 ± 2	1 ± 2	8	1.06 ± 0.02
E	89 ± 4	1 ± 2	8 ± 2	2 ± 2	9	1.31 ± 0.02
F	76 ± 5	3 ± 2	21 ± 5	-	23	1.14 ± 0.02
G	89 ± 4	1 ± 2	10 ± 2	-	11	1.53 ± 0.02
H	38 ± 6	2 ± 2	60 ± 6	-	62	1.28 ± 0.02
Base batch	74 ± 5	1 ± 2	22 ± 5	3 ± 2	23	1.07 ± 0.02

and

Table 6. Characteristics of the plastic–viscous properties of the samples studied.

Sample	Rank of Coal DSTU 3472-2015	Plastometric Parameter, mm, Y мм/(Rog’s Index, Units)	Free Swelling Index	Plastic–Viscous Properties According to Gieseler					Dilation According to Odbert-Arnou, °C			Sintering According to Gray–King, GK
		Y/(RI)	FSI	t1, °C	tmax, °C	th, °C	∆t, °C	Fmax, ddpm	t1	tII	tIII	a, %	b, %
Coal USA	Zh	20/80	6 1/2	370	435	480	110	>2800	360	415	475	29	191	G8
Batch	-	13/58	5	370	430	475	105	692	380	430	460	23	18	G4

.

Based on the results of the analysis, the conclusion is that American coal is characterized by low ash and sulfur content. In terms of its overall genetic characteristics and plastic–viscous properties, it is typical fatty coal. The obtained indicators for characterizing thermoplastic properties (dilatometric indicators according to the Odier–Arnou method: contraction (a), dilatation (b), and maximum fluidity according to Gieseler (F_max)) give an idea of the behavior of American coal under real coking conditions. Thus, the plastic mass of coal is homogeneous in composition, contains fewer low-molecular components, and the liquid phase components exhibit a well-defined plasticizing effect. During thermal destruction, a significant amount of non-volatile, slow-moving products are formed, characterized by high fluidity and the ability to bake the residual material of destructive coal grains and a large quantity of non-clinkering additives. This is evidenced by data obtained using the Rog and Gray–King methods. Like all bituminous coal in the medium stage of metamorphism, the plastic mass of this coal has a pronounced tendency to swell, which is confirmed by high values of the free swelling and dilatation index according to the Odibier–Arnou method.

Equimetamorphosed American bituminous coals are very similar to Ukrainian bituminous coals. The American coals studied in terms of the nature of the thermochemical transformations occurring in the temperature range of plasticity are very similar to the processes occurring with Ukrainian coals, which allows us to judge their good compatibility during coking in mixtures. This, in turn, gives reason not to make any significant adjustments to the thermal regime of their coking. However, given that the maximum rate of mass loss of Ukrainian coals occurs in the range of maximum fluidity according to Gieseler (t_max = 430–440 °C), it can be assumed that it is this feature of their thermal transformations that determines the production of coke with better mechanical strength (in particular, abrasion resistance) due to improved plastic contact between grains under the action of increased pressure from the gaseous products of thermal destruction in a closed volume.

Thus, based on the entire complex of studied properties, American coal is analogous to Ukrainian bituminous coal of a medium stage of metamorphism, differing from the latter in terms of a lower ash and sulfur content as well as a slightly higher content of lean (fusinized) macerals of the inertinite group.

3.2. Using American Coal in Coking Batches

The production batch of the plant has increased ash (A^d = 10.4%) compared to the batch of other plants of Ukraine, where the ash of the batch, as a rule, varies within 8.5–9.0%. Sulfur content in the batch is low (S_t^d = 0.83%). It should be noted that the composition of the batch vitrinite includes coals from almost all stages of metamorphism, from low metamorphosed to highly metamorphosed. The principal amount of vitrinite constituents has reflection coefficient R_o values in the range of 0.65–0.84 (20%) and 0.90–1.19 (59%).

The total content of vitrinite constituents with a mean vitrinite reflection coefficient of 0.9–1.39% is 66%. However, taking into account that the absolute content of vitrinite in the batch is only 74%, the actual amount of vitrinite components having a mean vitrinite reflection coefficient of 0.9–1.39% in the organic mass amounts to only 46.9%, which is far from the recommended minimum value of 53%, which ensures the production of high-strength metallurgical coke. Reduced amounts of well-clinkering components of vitrinite and increased content of non-clinkering macerals of the inertinite group, as well as high ash, cause low sinterability and coking ability of the batch as a whole, which is reflected in the results of the study of plastic and viscous properties of the batch.

The production batch was sorted into size classes, and the technical analysis indicators, sintering ability according to the Rog method, and petrographic characteristics were determined. In addition, the content and quality of fractions lighter and heavier than 1400 kg/m³ were determined to be in charge size classes greater than 1 mm. These data are presented in

Table 7. Properties of production batch size classes.

Batch	Granulometric Composition, %		Proximate Analysis, %				Sintering According to Rog, Units
	Class, mm	Yield, %	Wa	Ad	Vd	Vdaf	RI
	>10	3.2	1.4	17.5	26.4	32.0	45
	10–6	5.5	1.6	11.0	26.3	29.5	53
	6–3	8.6	2.0	9.4	28.7	31.7	48
	3–2	11.6	1.8	9.3	27.5	30.4	53
	2–1	11.5	1.7	8.4	27.3	29.8	66
	1–0.5	14.0	2.2	8.5	26.7	29.2	65
	0.5–0.25	13.0	2.0	8.1	27.5	29.9	65
	<0.25	32.6	2.1	12.8	25.0	28.6	56
Total batch (actual)		100.0	1.7	10.4	26.5	29.6	58
Total batch (calculated)		100.0	2.0	10.4	26.5	29.6	58
Fractional composition of the batch (class > 1 mm)
Fraction < 1400 kg/m3		78.7	2.1	6.4	29.3	31.3	60
Fraction > 1400 kg/m3		21.3	2.2	28.4	19.6	27.4	21

and

Table 8. Petrographic characteristics of production batch size classes.

Class, mm	Yield of Grain Size, %	Petrographic Composition (Without Mineral Impurities), %					Mean Vitrinite Reflection Coefficient, %	Stages of Vitrinite Metamorphism, %
								0.50–	0.66–	0.90–	1.20–	1.40–	1.70–	2.60
								0.65	0.89	1.19	1.39	1.69	2.59	And More
								Coal Ranks, Which Roughly Correspond to The Stages of Vitrinite Metamorphism
		Vt	Sv	I	L	Σ FC	Ro	DG	G	Zh	K	PS	T	A
1	2	3	4	5	6	7	8	9	10	11	12	13	14	15
>10; 10–6; 6–3	17.3	61	1	34	4	35	1.11	4	25	55	3	4	4	5
3–2; 2–1	23.1	70	1	28	1	29	1.08	4	23	58	6	5	1	3
1–0.5; 0.5–0.25; <0.25	59.6	72	1	25	2	26	1.10	4	14	61	11	6	3	1
Total batch (actual)	100.0	74	1	23	3	23	1.07	4	20	59	7	5	2	2
Total batch (calculated)	100.0	73	1	24	2	24	1.10	4	18	59	8	5	3	3
Fractional composition of the batch (class > 1 mm)
Fraction < 1400 kg/m3	78.7	79	-	18	3	18	0.98	8	31	48	6	4	3	-
Fraction > 1400 kg/m3	21.3	54	2	39	5	41	1.21	2	16	57	8	7	4	6

.

The study’s results indicate a suboptimal distribution of particle sizes and quality. Thus, the charge contains 45.6% dust-like fractions with a size of less than 0.5 mm, which significantly complicates the process of smokeless loading, contributing to the entry of the smallest coal particles into the tar, which increases its ash and density.

At the same time, the batch contains many high-ash, poorly clinkering large grains larger than 6 mm, which are centers of crack formation in coke, reducing its resistance to crushing forces.

Grains of batches larger than 1 mm contain 21.3% industrial product fractions with an ash content of 28.4% and reduced clinkering ability, which is characterized by the Rog method and is only 21 units. At the same time, the rest of the load has a value of 60 units.

Thus, the calculation of the optimal degree of crushing [43], which ensures maximum strength indicators of metallurgical coke, taking into account the data in Table 7, showed that it should be according to Equation (1):

$$k={\displaystyle \frac{75\cdot 74\cdot \frac{66}{100}+90\cdot 74\cdot \frac{24+9}{100}+90\cdot 23}{100}}=79\%$$

(2)

In the studied batch sample, the content of the 0–3 mm class is 82.7%, which is definitely higher than the optimal value. At the actual level of crushing, the content of dust-like fractions in the batch is already very high, which leads to a decrease in the bulk density of the batch in the coking chamber and several technological difficulties when loading the batch into the hopper of the loading car, smoke-free loading, etc.

In this regard, it is necessary to install devices to screen out minor ranks of charge before its final crushing to avoid significant over-crushing.

Three variants of experimental batches were developed and compiled. The component composition of the base (production) batch and experimental variants of the batches is shown in

Table 9. Component composition of the batches.

	Component Composition of the Batches, %
	Base Batch	Batch 1	Batch 2	Batch 3
A	5	10	-	10
B (coal USA)	30	30	30	30
C	8	10	10	10
D	35	40	40	40
E	-	10	10	-
F	12	-	-	-
H	10	-	-	-
G	-	-	10	10
	100.00	100.00	100.0	100.0

. The data in

Table 10. Technological properties of the experimental batches.

Sample	Proximate Analysis, %				Plastometric Parameters, mm
	Ad	Sdt	Vd	Vdaf	X	Y
Base batch	10.4	0.83	26.5	29.6	34	13
Batch 1	9.7	0.65	28.0	30.9	27	16
Batch 2	9.9	0.65	26.1	28.8	25	16
Batch 3	9.8	0.61	27.2	30.0	27	16

characterize the technological properties of the batches. The petrographic characteristics of the pilot batches are given in

Table 11. Petrographic characteristics of the experimental batches.

Sample	Petrographic Composition (Without Mineral Impurities), %					Mean Vitrinite Reflection Coefficient, %
	Vt	Sv	I	L	Σ FC	Ro
Base batch	74	1	22	3	23	1.07
Batch 1	79	-	20	1	20	1.00
Batch 2	78	1	19	2	20	1.10
Batch 3	80	1	18	3	18	1.05

.

During the development of variants of experimental batches, we were guided by the fact that, first of all, it was necessary to reduce the content of non-binding macerals of the inertinite group in the experimental batches in comparison with the basic one at the expense of coal H. Also, it was necessary to exclude coal F from the batch, which did not contain vitrinite with a reflection index of 1.2–1.39 and had an increased number of components in its composition, with R_o = 0.65–0.84%. R_o = 0.90–1.19% and R_o = 1.7–2.59%. The introduction of petrographically homogeneous G coal into the batch and the increase in the content of G and E coal, which had the minimum content of inertinite group macerals, also contributed to solving the above problems.

The data analysis in

Table 12. Qualitative indicators of box coking, coke, and ashing and desulphurization coefficients.

Sample	Proximate Analysis, %			Ashing Coefficient	Desulphurization Coefficient	CRI, %	CSR, %	ρ, Om·cm
	Ad	Std	Vdaf
Base batch	12.9	0.61	0.9	1.303	0.806	38.3 ± 3.5	40.5 ± 3.5	0.0597
Batch 1	12.9	0.53	0.6	1.330	0.815	36.5 ± 3.5	44.0 ± 3.5	0.0568
Batch 2	12.6	0.52	0.5	1.299	0.800	36.1 ± 3.5	44.9 ± 3.5	0.0562
Batch 3	12.8	0.53	1.1	1.312	0.869	35.2 ± 3.5	48.8 ± 3.5	0.0547

shows that the objectives were achieved.

The content of lean (fusinized) macerals (the sum of lean (fusinized) components ƩFC) in the experimental batch compared to the base batch decreased from 23% to 19–20%. The amount of vitrinite components with a mean vitrinite reflection coefficient of 0.9–1.39% in organic mass increased from 45.1% in the base batch to 51.4, 49.9, and 48.8%, respectively, in the experimental variants.

The quality results of the experimental cokes are shown in Table 12 and

Table 13. Qualitative indicators of coke.

Sample	Yield and Strength of Coke, %
	Yield	M25	M10
Base batch	76.31	92.1 ± 3	7.3 ± 1
Batch 1	75.26	92.3 ± 3	7.5 ± 1
Batch 2	76.60	92.4 ± 3	7.0 ± 1
Batch 3	75.82	92.7 ± 3	6.9 ± 1

.

As can be seen from the data presented, the ash content of the base and experimental cokes is high and ranges from 12.6 to 12.9%. At the same time, the sulfur content is low and varies between 0.52 and 0.61%. The ash content coefficients depend on the yield of volatile substances from the charge, decreasing with lower yield and increasing with higher yield. The residual sulfur coefficient in coke ranges from 0.800 to 0.869.

The influence of coal raw material properties on the quality characteristics of coke is illustrated by the graphical dependencies represented in Figure 2, Figure 3, Figure 4, Figure 5 and Figure 6.

The equations describing the obtained graphical dependences are given in

Table 14. Mathematical equations.

№	Mathematical Equations	Statistical Assessment
		r	D, %
(3)	CRI = 0.6255·ΣFC + 23.859	0.99	98
(4)	CSR = −1.6196·ΣFC + 77.347	0.98	95.86
(5)	CRI = −0.4711·Vt + 73.152	0.95	90.51
(6)	CSR = 1.1639·Vt − 45.94	0.9	80.56
(7)	M25 = −0.1157·ΣFC + 94.718	0.95	91.01

.

Specific resistivity is a measure of the temperature effect during the process of thermal destruction of coal and coke conversion. Under the influence of temperature, structural changes in the organic matrix of coal transform its physical properties and therefore its electrical resistivity [44]. Depending on the final temperature of the process, the specific electrical resistivity decreases as the temperature increases. Since determining electrical resistivity is simpler, does not require reagents, and takes significantly less time than measuring reactivity, it can be considered an operational method for determining coke’s structural characteristics and readiness [45,46]. Therefore, the specific electrical resistance index can characterize the degree of coke readiness.

To ensure a rational temperature regime for coking petrologically heterogeneous interbasin coal batches, it is necessary to systematically control the coking temperature regime, increase the readiness of coke in the head zones of the coking chamber, and systematically control the completeness of loading. Heights and temperatures of the subbed space were measured. The study of coke structure is based on the analysis of coke carbon, the distribution of carbon atoms in hybrid states, and the degree of ordering, i.e., the ratio of ordered and disordered carbon. Coal exhibits dielectric properties [47], which is explained by the large number of σ-bonds formed by the carbon atom in the sp³ hybridization state. Under the influence of high temperatures in coking and deep cracking, macromolecules are practically deprived of side chains, leading to condensed polyaromatic structures containing carbon in sp hybridization. This improves the orderliness of the coke structure, bringing it closer to the structure of graphite, whose single crystals consist of a large number of parallel layers formed by hexagons of carbon atoms. In the interlayer space, π electrons form a unified electron cloud with sufficient mobility. This explains the high electrical conductivity of carbon materials. Thus, the resistivity value can control the degree of orderliness of the structure of carbonaceous materials.

It should also be noted that the organic carbon matrix consists of carbon atoms and other heteroatoms, such as hydrogen, oxygen, nitrogen, halogen, sulfur, phosphorus, etc. These atoms are bonded to the carbon edges and determine the chemical properties of the coal surface. The presence of oxygen functional groups (phenols, quinones, and carboxyl groups) on the surface significantly affects these materials’ reactivity, dispersibility, and electrical properties [48]. Surface chemistry affects carbon’s chemical behaviors in specific chemical reactions [49].

Thus, the readiness of the obtained coke is good, as indicated by the data on the yield of volatile substances, equal to 0.5–1.1%, and the specific electrical resistivity, which is in the range of 0.0547–0.0597 Om·cm. Increased readiness and orderliness of the coke structure lead to decreased electrical resistivity. Thus, an increase in coke strength after reaction (CSR) leads to a decrease in resistivity. A decrease in coke reactivity leads to a rise in electrical resistivity. The dependence of the specific resistivity of coke on its reactivity is illustrated by the graph in Figure 7, with a correlation coefficient of 0.83.

Indices of reactivity (CRI) and post-reaction strength (CSR) determined by the “Nippon Steel” method are the worst for coke from the base batch, and the best values are obtained for coke from batch variant 3. Coke from the batch of variants 1 and 2 is also better than coke from the basic batch, primarily because F coal, which has a high ash content of 12.8%, was removed from the composition of the experimental batch.

The analysis shows that the best values of M₂₅ and M₁₀ are coke from the batch of variants 2 and 3 containing G coal.

Transitioning to batch coking without H and F coals improves mechanical strength and CRI and CSR indicators. Quality indicators of metallurgical coke are enhanced due to the introduction of 10% G coal into the batch.

The effect of petrographic nonuniformity within coal batches on a specific coke property was studied by calculating each batch’s weighted Gibbs free energy of formation (ΔG_f,total). This provides a single representative thermodynamic descriptor of the batch’s tendency to undergo structural transformation during coking:

$${\mathsf{\Delta }\mathrm{G}}_{\mathrm{f},\mathrm{total}}={\sum }_i^n{\displaystyle \frac{{\mathrm{\%}}_i}{100}}\times {\mathsf{\Delta }\mathrm{G}}_{\mathrm{f},\mathrm{i}}$$

(3)

where %_i is the percentage of maceral group i, and ΔG_f,i is the Gibbs free energy of formation for maceral type i.

The thermodynamic parameters for the coal macerals were taken from [50]; these values are also visualized in Figure 8, where ΔG_f,_i is presented as a function of the mean reflection coefficient (R_o). Results of the calculations are presented in

Table 15. Thermodynamic assessment of petrographic nonuniformity of coal batches.

Parameter		Base Batch	Batch 1	Batch 2	Batch 3
Reflection coefficient Ro, %	Vt	1.07	1.00	1.10	1.05
	Sv	1.28	-	1.32	1.26
	I	1.98	1.85	2.04	1.94
	L	0.58	0.54	0.59	0.57
ΔGf,i, kJ/mol	Vt	−64.9	71.8	−112.5	−10.7
	Sv	−1227.2	-	−1203.8	−1239.0
	I	−1039.0	−1069.6	−1026.6	−1047.7
	L	−1665.4	−1810.0	−1614.8	−1712.4
ΔGf,total, kJ/mol		−338.8	−175.3	−327.2	−260.9

.

The table presents a comparative analysis of four coal batches using the calculated total Gibbs free energy of formation (ΔG_f,total) based on the weighted contributions of their maceral groups (vitrinite, semivitrinite, inertinite, and liptinite).

Batches with more negative ΔG_f,total (e.g., the base batch and batch 2) exhibit a higher thermodynamic driving force for carbon structure ordering during coking. Batch 1, which has the least negative ΔG_f,total, is thermodynamically the least favorable for this transformation.

Analysis indicates that the coke reactivity index (CRI) decreases as ΔG_f,total becomes more negative. This suggests that cokes produced from thermodynamically more favorable batches (more negative ΔG_f,total) are less reactive, likely due to higher structural ordering and lower microporosity.

Conversely, the coke strength after reaction (CSR) increases with more negative ΔG_f, total, reinforcing the link between thermodynamic favorability and the mechanical integrity of the resulting coke. These correlations support using ΔG_f,total, as a predictive thermodynamic parameter for optimizing coal batch selection and anticipating coke performance.

The values of ΔG_f,i were empirically derived from experimental assessments of maceral-specific thermodynamic behavior based on elemental composition and calorimetric data. Accordingly, ΔG_f,total functions as an integrated descriptor of the maceral composition, directly influencing plasticity development, mesophase formation, and the eventual structural ordering of carbon. The proposed Gibbs energy index for the coal charge accounts not only for the relative content of maceral groups but also for their degree of metamorphism, as indicated by the vitrinite reflectance.

This index is reproducible within the known uncertainty limits of petrographic analysis. Since ΔG_f,total is calculated from quantitatively determined maceral proportions (%i) and fixed ΔG_f,i values are assigned to each maceral type, and it can be reliably recalculated for any coal blend with available petrographic data.

Thus, ΔG_f,total provides a thermodynamically grounded basis for comparing coal blends. Its value lies in bridging petrographic heterogeneity with the inherent thermodynamic propensity for structural ordering, offering a complementary indicator to traditional assessments of coking performance.

The uniqueness and practical value of introducing the thermodynamic index (ΔG_f,total) lie in considering the contents of vitrinite and inertinite and incorporating reflection coefficients for all macerals. This approach allows the estimation of the overall thermodynamic stability of a coal blend using a single integral value.

4. Conclusions

The coke quality for blast furnace smelting largely depends on the properties of coal raw materials and the efficiency of their preparation. Coking coal has been and remains a critically crucial raw material resource of strategic importance for industry functioning and economic development in many countries. Ensuring the economic efficiency and effectiveness of coke production in the conditions of a multi-basin raw material base for coking requires optimizing the component composition of coal batches and improving the primary technological preparation methods considering its petrographic characteristics.

This work used standardized methods for studying coal and coal batches’ technological and plastic–viscous properties. The qualitative characteristics of coke were determined using physical–mechanical and thermochemical methods of studying standardized indicators: crushability (M₂₅), abrasion (M₁₀), reactivity (CRI), post-reaction strength (CSR), and specific electrical resistance (ρ). Statistical analysis of the results and analysis of the influence of raw material factors on the mechanical and thermochemical properties of coke were performed using the licensed Microsoft Excel computer program.

Based on the results of experimental industrial coking of scientifically based batch compositions developed with the presence of imported coal, two variants of coal batches where the coke had the best quality indicators were proposed. The quality indicators of metallurgical coke were improved by adding 10% G coal to the batch. Thus, according to the M25 crushability indicator, the mechanical strength increased by 0.6%, and the M₁₀ abrasion decreased by 0.4%. Significant improvements in thermochemical properties and an increase in the orderliness of the carbon structure were recorded: the CRI reactivity decreased by 3.1%, the CSR post-reaction strength increased by 8.3%, and the specific resistance decreased by 8.4%.

The inverse relationship of ΔG_f,total with CRI and its direct relationship with CSR validate its use as a thermodynamic predictor for selecting coal batches that yield less reactive, stronger coke.

The next stage of research involves accumulating a large amount of experimental data and developing general recommendations for coke plants in conditions of deteriorating coking raw material base and irregular supply of coal concentrates, focusing on optimizing the component composition and determining the optimal degree of crushing while taking into account the petrographic composition and sinterability of the batches.