Since iron smelting furnace process of ore with charcoal started in 1881 at Kamaishi, Iwate, Japan, coke manufacturing from coal was attempted by using various ovens, that is, Beehive-, Copee-, Semet-Solvay-, Koppers-, and Kuroda-oven. Manufacturing of coke with verified quality for blast furnace is dependent upon the nature of coal as a starting material. This review concerns with mainly carbonization characteristics of coal with respect to the coal science up to about 2000. Particular attentions are paid to the following topics. - Thermoplasticity including coking and caking properties of coals. - Mesophase, which is governing the performance of coke in blast furnace, appearing between coal and coke during carbonization. - Coalification as a natural carbonization and carbonization on heating. - Reflectance and some physical properties of coal as quality control parameters. Persons at the dawn of coke industry in Japan are introduced at the attached section.
Research and development of new cokemaking process (SCOPE21) was conducted in Japan from 1994 to 2003 by the Japan Iron and Steel Federation (JISF). Pilot plant scale test of SCOPE21 process was conducted successfully and targets of the project were confirmed. SCOPE21-type new coke oven battery was constructed at Nippon Steel Oita works and the operation of new coke plant started in 2008. The coke production capacity is 1 million ton per year. At present, high quality coke has been produced in this process using high blending ratio (over 50%) of non- or slightly-caking coal. The new cokemaking process is in operation very smoothly.
Coke in a blast furnace has many significant functions such as a heat source, a spacer for a stable gas and liquid flow, a carburization agent for modifying the dripping behavior of metal and a reducing agent of iron ore. As the blast furnace operation moved into a more difficult stage where a high productivity and a high level of pulverized coal injection is demanded in a large volume blast furnace, coke strength has been considered important. On the other hand, the blending ratio of slightly caking coal has been increased in cokemaking process to meet cost reduction. With the development of cokemaking technology, reactivity of coke is attracting attention in terms of improving blast furnace reaction efficiency and reducing CO2 emission. In this paper, reviewed are how production technology of highly reactive coke and improvement in blast furnace reaction efficiency through the use of highly reactive coke have developed.
Recent research progress on coke reactivity and its control are discussed. Based on a review of coke behavior in a blast furnace, the importance of increase in coke reactivity in the temperature range of <1127K is emphasized. The rate of the reaction between coke and CO2 is affected by carbon structure, heating conditions and catalyst, and its effect on coke degradation is also discussed. Research interests and application of highly reactive coke and its contribution to the realization of highly effective blast furnace operation are also mentioned.
Coal extract fractions were blended with coals, and their effects on coke strength were evaluated. Since the pyridine-soluble (PS) fraction of coal has a high thermoplasticity, its addition to a slightly-caking Kouryusho (KRS) coal greatly increased the strength of cokes. While, by the addition of the pyridine-insoluble (PI) fraction, the strength of cokes was decreased. The reason for decrease may be decline of thermoplasticity due to the PI addition, since its thermoplasticity is significantly low.
Eight poly-aromatic hydrocarbons as model compounds of coal extract were added to coal blends, in order to clarify the mechanism on the effect of coal extract (HyperCoal) on coke strength. The addition of large aromatic-ring compounds (coronene, perylene and naphtho[2,3-a]pyrene) greatly enhanced the coke strength, while, the three ring aromatics did not have a significant effect on the coke strength. Since large poly-aromatic compounds have better affinity to coal molecules, they co-fused with coal particles. As a result, formation of large pores during coking was suppressed, leading to the increased coke strength.
In this study, relation between the amount of transferable hydrogen of coal tar pitch (CP) and fluidity of the slightly caking coal adding CP was discussed in order to enhance fluidity of the slightly caking coal. In particular, additive effects of two constituents (toluene soluble but hexane insoluble; TSHI and pyridine soluble but toluene insoluble; PSTI) separated from solvent extraction of CP were estimated. As a result, an amount of transferable hydrogen of these constituents increased slightly by high temperature (360–420°C) hydrogenation. Fluidity of Enshu coal (slightly caking coal) increased by addition of these hydrogenated constituents (TSHI and PSTI). Because retrogressive reaction among the molecules in Enshu coal was suppressed by hydrogen donation from the increased transferable hydrogen in TSHI or PSTI to the molecules in Enshu coal. Further, additive effect of fluoranthene (FL) was estimated. For all tested coals, maximum fluidity of coal adding hydrogenated FL (HFL) was larger than that adding FL. The difference was large for coals containing much of oxygen. Accordingly, retrogressive reaction among the oxygen containing groups in the coal molecules was suppressed by hydrogen donation from the transferable hydrogen in HFL. Estimating from the polarizing micrographs of cokes from coal adding HFL, it was clarified that thermal plasticity of coal was enhanced by addition of HFL, not by thermal plasticity of HFL itself.
Japanese steel industries, which are importing all coal resources required, are facing the necessity of increasing usable coal resources. They are to increase the kinds and amounts of usable coal and to develop the technologies for upgrading low rank coals such as subbituminous coal and brown coal as substitutes. In this work we have developed methods for upgrading brown coal. The methods involve the fractionation of brown coal using sequential thermal solvent extraction, dewatering and upgrading of brown coal using hydrothermal treatment, and co-pyrolysis of upgraded brown coal and plastic derived wax. The fractionation method successfully fractionated an Australian brown coal into six fractions, some of which are expected as binders for cokemaking because of their thermal plasticity. The hydrothermal upgrading/extraction method can not only removes water from brown coal but also separates the coal into extract of low molecular mass compounds and upgraded coal having lower hydrophilicity and spontaneous combustibility and higher calorific value than the raw coal. Addition of the upgraded coal to the coking coals was found to enhance the gasification reactivity of the resulting coke. Co-pyrolysis of the upgraded coal and waxes formed from waste plastics was also investigated as one of the methods for further upgrading the hydrothermally upgraded coal. It was found that the waxes effectively removed oxygen atoms from the upgraded coal and tended to be retained as a solid carbon in the coke through co-pyrolysis. It is shown that these methods enable to produce binders for cokemaking and substitute of slightly/noncaking coal.
In the present study, we have conducted a fundamental study to understand the reasons for the deterioration by the pre-addition of catalysts such as Fe2O3 and CaCO3 to a coking coal, Goonyella. The addition of Fe2O3 resulted in the reductions in the maximum fluidity (log MF) and the drum index (DI), when the Fe2O3 of 0.31 μm were introduced to the coal. Of course the increase in the reactivity toward the solution-loss reaction was observed as previously reported. While in the case of the addition of CaCO3, no distinct changes in log MF, DI and CRI were observed irrespective of the size and the contents of CaCO3. The conversion of TI-PS (toluene insoluble and pyridine soluble) fraction to TS fraction was observed when the Fe2O3-added coal was heat-treated up to 440°C, while no changes were observed for the CaCO3-added coal. No signs of catalytic graphitization took place judged by X-ray diffraction study. The addition of Fe2O3 produced hollow balloon-like structure with 10 nm-thickness-amorphous carbon walls, while such formation was not observed for the case of CaCO3 addition. Finally, we concluded that the deterioration in the coke strength by the pre-addition of Fe2O3 is caused by the insufficient adhesion of coal particles due to the reduction of the TI-PS fraction as well as by the formation of the fragile balloon-like structures.
Coarse defect structure in coke was quantified and its control factor was investigated, which exerts a strong influence on coke strength, as part of the development of high strength cokemaking technology. First of all, X-ray CT images of several cokes were evaluated by image analysis and it was clarified that coke defect structure around millimeter scale obeyed Weibull distribution. Secondary, viscosity (η) and swelling pressure (ΔP) of coal during plastic phase, those affect bubble growth phenomena, were measured, based on the assumption that there is correlation between bubble growth behavior and coke defect generation. As a result, coarse defect structure in coke was strongly dependent upon the bubble growth factor ΔP/η which can be derivable from a governing equation for single bubble growth. Eventually, it was cleared that the bubble growth factor was one of the important parameter to control coke strength.
To control coking pressure is one of the most important aspects of the cokemaking process since excessive coking pressure increases the force needed for coke cake pushing and in some cases leads to operational problems such as hard pushes or stickers, causing wall damages. We investigated a selective fine crushing of high coking pressure coal as a way to reduce coking pressure. It was shown in a laboratory scale that fine crushing of high coking pressure coals increases permeability of the plastic coal layer, which decreases coking pressure (internal gas pressure). Based on the basic investigations, we tried fine crushing of high coking pressure coals at commercial cokemaking plants. It was confirmed in a long-term commercial scale experiment that fine crushing of high coking pressure coal decreases coking pressure and decreases maximum power current of coke pushing. A selective fine crushing of high coking pressure coal is a promising way to reduce coking pressure and to prolong coke oven life.
In order to obtain highly reactive cokes, a Ca-ion exchanged brown coal (Loy–Yang coal) was added to a caking coal (Goonyella coal) with different blending ratios and cokes were then prepared from these blended coal samples. With the addition of Ca-loaded coals, the CO2-gasification reactivity of the resulting cokes was significantly increased. Their initial reactivity was particularly enhanced, at a maximum, being thirty times as high as the reactivity of a coke prepared from a single Goonyella coal. This is because the reactive part derived from the brown coal in each coke was initially gasified with the other part from the caking coal almost intact. Since, as coke texture, the former part (from the brown coal) is included in a matrix of the latter part (from the caking coal), the mechanical strength was not lowered so much, even though the cokes were gasified up to a conversion of 10%. In conclusion, this study demonstrates that some addition of a Ca-ion exchanged brown coal to a caking coal is quite promising for the production of highly reactive and high strength cokes.
Reactivity of cokes produced by pre-addition of Fe2O3 and CaCO3 to coals was investigated. Cokes with 0, 2, 5, and 10 wt% Fe2O3 and CaCO3 were produced. Fe2O3 and CaCO3 loaded cokes were deashed by hydrofluoric acid treatment and their gasification reactivity was measured and compared with Fe2O3 and CaCO3 loaded original cokes under CO2 and CO+CO2 (Fe2O3 only) mixed environment. Carbon crystallite size of the cokes and Fe/Ca loaded cokes were determined by XRD analysis and correlated with the reactivity. Addition of both Fe and Ca increased the reactivity of cokes. The observed enhancement of coke reactivity was discussed by considering the contribution from the change in cokes properties brought upon by Fe/Ca during coke making stage on reactivity and contribution from their catalytic effect in enhancing the reactivity. At lower temperatures (900°C), increase in reactivity due to catalytic effect was small for Fe2O3 but significant for CaCO3.
The performance of coke in blast furnace (BF) used to prefer a high strength and low reactivity, generally, while the high reactivity coke commonly has a low strength. To overcome the contradiction, we are going to use the catalytic effect on the coke gasification. It is important to clarify the mechanism of coke gasification concerning to the coke microstructure and pore. In this study, the coke gasification was analyzed using μ-X-ray CT (Computed Tomography). The optimum conditions for the image processing of the data from the X-ray CT were obtained through the comparison with the cross section of the coke embedded in a resin. Nondestructive observation became possible. To improve resolution of the nondestructive observation, the sample shape was changed from spherical to cylindrical one and the observation conditions were adjusted to an optimum one. It was found that the resolution of the nondestructive observation of cylindrical sample was better than that of spherical sample. Using high precision μ-X-ray CT, the position of the catalyst added and the reaction behavior became clear. Furthermore, using new developed image processing of the data from the μ-X-ray CT, macroscopic reaction behavior of the catalyst became clear.
The performance of coke in blast furnace (BF) used to prefer a high strength and low reactivity, generally, while the high reactivity coke commonly has a low strength. To overcome the contradiction, it is important to develop a catalyst to increase the reactivity of carbon, so that the reaction mode can be changed from chemical reaction control to diffusion control. In this study, the reaction behavior of a ferro-coke was investigated, in which an iron oxide were added into a raw coal. Using high temperature laser microscope, in situ observation was carried out with the ferro-coke sample ground into tetragonal shape with 3.4×3.4×2.2 mm. Behavior of catalyst in coke was clarified through the in situ observations. It was found that the temperature at peak of reaction without catalyst was 1270°C, while the coke with catalyst showed two peaks at 900°C and 1270°C. It was considered that redox reactions would exist in the system. The mechanism of redox reaction is shown as Eqs. (1), (2) and (3), mainly.
On the other hand, the coke gasification was analyzed using high precision μ-X-ray CT. The optimum conditions for the image processing of data from the μ-X-ray CT were obtained through the comparison with the cross section image of the coke embedded in the resin. Because of the smaller sample than the previous study, higher precision analysis was able to carry out.
The intensive studies on the strength degradation of blast furnace cokes after gasification reaction are essential for developing mechanically strengthened coke with highly reactive nature to gasification. In this study, the degradation behavior of the mechanical properties of several types of cokes after their gasification reaction were investigated in instrumented spherical indentation and compression tests. The mechanical degradations were examined in the instrumented spherical indentation test for two types of cokes that were treated under about 20% gasification reaction. The indentation test results confirmed that the discrepancy in the degradation behaviors of mechanical properties (elastic modulus, yielding stress, and work-of-indentation) of these two cokes are insignificant, whereas there exists a significant discrepancy in the values of their drum indices (DI). In the compression test of coke grains, several types of cokes (differences in coal species, grain sizes, reaction temperatures, manufacturing conditions (formed coke, catalyst-added coke)) were tested. The Weibull statistic was applied to the results of compressive failure tests, where the concept of the work-of-compression was introduced, and successfully utilized in quantitatively examining the strength degradation of reacted cokes, and then it was clearly demonstrated that the grain size, reaction temperature, and the manufacturing condition are all essential in providing mechanically strengthened coke with highly reactive nature to gasification reaction.
To investigate the effect of mechanical properties and non-adhesion region boundaries in coke on behavior of coke, fracture analyses using RBSM (Rigid Bodies-Spring Model) were carried out for coke. First, the fracture analyses with RBSM assuming 4-point bending tests were carried out for two cases by applying the approximate equations of the shear strength. The approximate equation of the shear strength derived from internal friction angle and the one for brittle materials like glasses and rocks were applied to analytical objects. When the approximate equation of the shear strength for brittle materials like glasses and rocks was applied, analytical results reproduce the fracture behavior better than the other one. It is known that non-adhesion region boundaries in coke influences coke strength, so the fracture analyses with RBSM assuming 4-point bending tests were also carried out to investigate the effect of non-adhesion region boundaries in coke. Analytical results showed that the fracture started at non-adhesion region boundaries when non-adhesion region boundaries were located under region of high stress. It was also indicated that fracture loads varied due to the difference in fracture strength at non-adhesion region boundaries even if non-adhesion region boundaries existed at the same point. As a result, it is supposed that non-adhesion region boundaries in coke influence crack propagation and fracture load and are the important intensity factor.
Recent years various trials to improve operation efficiency of blast furnace by combining the decrease in operation temperature and the increase in reaction rates. An increase in coke reactivity is one of the approaches to decrease thermal reserve zone temperature in blast furnace. Although the mechanism to improve the efficiency by increasing the reactivity is explained by the static models like the Rist diagram, these models are incapable of estimating the temperature decrease by the reactivity increase. In this study numerical simulations of blast furnace operation using a mathematical model based on the kinetic theories were performed. In the simulations, descending motions of burden materials were tracked and the reaction behaviors of coke particles along the trajectories were discussed in details through the method of chemical reaction engineering. The results showed that the increase in the coke reactivity lowers the temperature of upper part of the furnace. In the studied range of the reactivity, the reaction scheme of the solution loss reaction in thermal reserve zone varied from uniform reaction to surface reaction with increase in reactivity. Although the coke consumption by the solution loss reaction increases, the solution loss reaction in the central part of the coke particle is suppressed by the increase in coke chemical reactivity.
Coke strength is mainly determined by pores and cracks that cause fracture of coke. In this study, connected pores that were considered to cause fracture of coke were investigated. In order to evaluate connected pores quantitatively, low roundness pores (whose roundness were below 0.2) were measured by image analysis technique of microscopic photographs of cokes. The relationship between the amount of low roundness pores and coke strength (DI1506) showed a good correlation. It is thought that the amount of low roundness pores is one of the factors determining coke strength.
While high strength coke is required for the high productivity of blast furnace, price of coking coal has been increasing. High accuracy of coke strength (drum index) estimation based on the degradation mechanism and on the physical propertyof coke is necessary to achieve cost reduction of raw coal by avoiding the excess coke strength. In this paper, the behavior of fine coke generation during the drum test was investigated considering coke breakage mechanism. Change in fine generation rate, its Weibull analysis and size distribution of fines was examined. Coke fines after drum test was divided into following three classes; (1) Under 0.5 mm fines generated by compressive breakage of coke matrix. (2) 6 to 15 mm fines generated by volume breakage induced by cracks and defects in coke lump. (3) 0.5 to 6 mm fines influenced by both compressive breakage and volume breakage. Also, amount of these fines were related to the coke and coal properties. Amountof 0.5 mm under fine and Brinell hardness, which represents strength of coke matrix, had good correlation. On the other hand, volume breakage was influenced by the volatile matter content and fluidity of raw coal.
Due to enlargement, stabilized and efficient operation of blast furnaces (BF), cokes with high strength have been required in BF in recent years. Recently, on the coke reactivity with carbon dioxide, it is reported that high reactive cokes can improve the reduction of iron ore in BF and reduce carbon dioxide emissions from BF. These coke strength (DI) and coke reactivity (CRI) have been primarily controlled by the blending operation of coals in coke production. However, it is difficult to get the cokes which possess both high DI and high CRI, because the blending operation for increasing DI usually results in the decrease of CRI. To get high DI, we investigated the effect of coating on coke with polyvinyl alcohol (PVA) on coke strength. Coke particles were dipped into the PVA aqueous solution and dried out. DI and CRI/CSR of the PVA coated coke prepared by the above method were measured. Those results showed that 0.011 g/g-coke PVA coating on the surface of the coke resulted in the increase of the DI by 4.2 points and no change in the CRI/CSR values. It may be possible to control the coke strength (DI) and the coke reactivity (CRI) independently without any change of coal blending by using our new method.