Search

The plastic layer formation is a critical step during coke formation in coke ovens, during which the physical structures of the coal change dramatically. To better characterize the pore structures ...

A reflux classifier system was employed to concentrate large sized inertinite-rich particles (1.6–2.0 mm) from two different ranked coking coals A and B. The impact of these concentrates on swelling, permeability and volatile evolution rates of another two coking coals C and D was investigated using the Computer Aided Thermal Analysis (CATA) and Dynamic Elemental Thermal Analysis (DETA) techniques. The results indicated that the concentrated inertinite-rich particles contained inertinite up to 75% for both coals A and B. When 25% inertinite particles were blended with 75% coal C or D by mass, the low rank inertinite concentrates A caused higher extents of effect on swelling and permeability than the high rank inertinite concentrates B, leading to higher softening temperatures with smaller plastic ranges. During swelling, inertinite concentrates B can compress volatile evolution rates of both coals C and D, however, inertinite concentrates A promoted volatile evolution rate of coal C. Concentrates from both coals A and B can compress the utilization of volatiles for driving swelling, the blend containing concentrate A used nearly half of volatiles that coals C and D themselves used for driving swelling. This study suggested that the additive internite concentrates not only provided a pathway for gas diffusion, but also interacted with the blended coals, which affected thermoplasticity of the whole blends.

An in-situ study on the formation mechanism of plsatic layer was carried out by using a 4 kg laboratory-scale coke oven under controlled heating conditions. The coke oven rig enabled in-situ measurements of internal gas pressure (IGP) and temperature profiles in five locations in the coal charge and allowed sampling of semi-coke which consisted of the coke/semi-cokes, the plastic layers and loose coal. Six Australian coking coals with varying properties were used in the coking experiments. The semi-coke samples were scanned by micro-CT at the Imaging and Medical Beamline at Australian Synchrotron to study the physical structure transitions inside the thermoplastic regions and investigate the fissure patterns at the coke/semi-coke sides. The thicknesses of the plastic layers in different locations inside the coal charge were estimated from the measured temperature profiles and the initial softening temperatures and the solidification temperatures measured from the Gieseler fluidity tests. The results showed that the thicknesses of the plastic layers increased from the heating wall to the center of the coal charge. The IGP curves measured in different locations were used to obtain the maximum IGP values and the IGP termination temperatures across the coal charge. The difference in the extents of fissuring appeared to influence the maximum IGPs measured at the center of the oven. The higher extent of fissurin appeared to facilitate the release of volatile matterfrom the thermoplsatic region through the semi-coke sides, thus leading to lower maximum IGPs at the center. It was found that the extents of the fissuring and the maximum IGPs at the center were sensitive to the coal properties such as the petrographic properties, Gieseler maximum fluidity and volatile matter content.

The aim of this study was to characterize the pyrolysis behavior of the plastic layer formed during the coking process in a 4 kg laboratory-scale coke oven facility. The 4 kg coke oven rig was used to produce semi-coke samples that included the plastic layer. The semi-coke samples were analyzed by using the Synchrotron Micro-CT to characterize their physical structures and identify the plastic layer features. Sectioned plastic layer samples corresponding to softening, maximum fluidity, and resolidification of coal were obtained for five coking coals of varying thermoplastic properties. Pyrolysis behavior and structural changes of the plastic layer samples were analyzed by a thermogravimetric analyzer coupled with a Fourier transform infrared spectroscopy (TG-FTIR) and an attenuated total reflection Fourier transform infrared spectroscopy (ATR-FTIR). The TG-FTIR analysis allowed the characterization of the changes in pyrolysis behavior including volatile matter yields, the changes in aliphatic and aromatic C–H structures in the sectioned plastic layer samples. These structure changes agreed well with the ATR-FTIR analysis which also showed chemical structure changes across the plastic layer. The results suggest that the aliphatic C–H bonds of the five coals underwent the greatest reduction with the progression of the coking process in the coal charge. It was observed that the aliphatic C–H across the thermoplastic layer region varied with the thermoplastic properties of the parent coal. This implies that the aliphatic structures in coal may have played a significant role in the development of thermoplastic properties during the formation of the plastic layer.

A coke produced using a custom-built sole-heated oven and a coke prepared in a pilot-scale oven from a matched coal, were compared using a range of analytical techniques. The aim of this comparison was to assess to what extent the small-scale sole-heated oven can successfully replicate the production of pilot-scale oven cokes, and thus be used to rapidly prepare and screen a wide range of cokes for particular characteristics, e.g. abrasion resistance. The techniques applied included conventional methods and novel methods developed by our research team. These included microstructural and microtextural analyses of samples of each coke, and tribological, scratch test and fractographic analyses, each of which elucidates different strength attributes. These include microstructural weaknesses, abrasion resistance, and the strength of microtextural interfaces. The level of replication achieved indicates that the sole-heated oven, used in combination with an annealing step in a muffle furnace, can be beneficially used to model the pilot-scale oven.

The chemistry of the plastic layer in a coke oven plays a critical role in coke-making and has a great impact on coke quality. In this study, progressive changes of the plastic layer were investigated using a combination of analytical methods. Three Australian coking coals with different vitrinite contents were used to produce plastic layer samples using a 4 kg lab-scale coke oven. The plastic layer samples were analysed by means of Synchrotron micro-CT scanning and attenuated total reflectance-Fourier transform infrared (ATR-FTIR) analysis to delineate the characteristic regions of each layer. The peak separation and quantitative calculation of FTIR spectra allowed the analysis of changes in structural parameters and functional groups of each plastic layer. The results showed drastic changes in physical and chemical structures of the plastic layer samples for all coals tested and were also strongly influenced by coal type. For the high-vitrinite coals, it seems that relatively larger amounts of the plastic mass were developed due to the timely rupture of large number of the bridge bonds in the intermediate plastic regions and efficient hydrogen transfer to stabilize the radicals generated. This led to a more drastic growth of pores and larger maximized porosities in the intermediate plastic regions. Conversely, the low-vitrinite coal was assumed to form lower amount of the plastic mass in the intermediate plastic region due to less rupture of aliphatic bridge bonds and inefficient hydrogen transfer. This was attributed to a lower degree of the pore development and the lower porosity observed in the region.

Tribological testing and analysis of metallurgical-grade cokes were conducted to elucidate the nature of the surface of each coke and the influence of the surface on coke abrasion resistance. The c...

The accurate prediction of coke quality is important for the selection and valuation of metallurgical coals. Whilst many prediction models exist, they tend to perform poorly for coals beyond which the model was developed. Further, these models general fail to directly account for physical interactions occurring between the blend components, through the assumption that the aggregate properties of the blend are suitably representative of the overall behavior of the blend. To study this assumption, a parameter termed the vitrinite distribution category was introduced to directly account for the distribution of one of these commonly aggregated parameters, the vitrinite reflectance. The introduction of this parameter in a regression model for coke quality prediction improved the model fit. The vitrinite distribution category was demonstrated to provide new information about coal blending decisions, and was found to be capable of providing insight into the behavior of different blending structures. Residual analysis was applied to explore the behavior of the coke quality prediction model, with the vitrinite distribution category found to explain more than just the presence or absence of coals within a blend. This work provides the foundation of future studies in examining coal blending decisions, with the proposed parameter having the potential to be applied as part of a coke quality prediction model to optimize coal blending decisions.

Coke samples, produced from coals having a range of rank and vitrinite content, were subjected to uniaxial loading in a universal tester. Cokes were also imaged at high resolution using micro-CT. The aim was to understand the relationship between the internal microstructure of the coke and coke strength under load. The loading was done in two separate ways, being either compression to failure or progressive loading. The results showed that the coke samples underwent a form of stiffening at low loads, potentially due to closing of fine-scale pores and/or re-alignment of graphitic layers in the RMDC. Measurements of plastic strain indicated that these changes were permanent. At higher loads, small load-bearing components of the microstructure were found to break, leading to a softening of the coke samples. Evidence of the breaking was observed using micro-CT images before and after loading of samples. The work has relevance to the understanding of the fundamentals of coke strength, as well as to issues relating to handling and preparation of coke for standard testing.

Acoustic emission profiles generated during scratch testing of a range of metallurgical coke samples were recorded and linked to the concurrent energy release, dispersal or absorption on coke fracture or damage. Three different signatures were identified, which were based on the simultaneous measurement of acoustic and total energy release profiles, and these signatures could be correlated with both the microstructure and microtexture of the coke being traversed at the time. The acoustic emission signature for fracture or damage to the coke reactive maceral derived constituents (RMDC) was correlated to the rank of the parent coal or coal blend, with the signature number generally increasing with increasing rank. Conversely, the signature numbers did not vary with parent coal rank for fracture or damage to the inertinite maceral derived constituents (IMDC), with the majority of IMDC fractures associated with a release of mechanical energy. The incidence of the signature associated with a release of mechanical energy (type 1) became increasingly dominant from RMDC to RMDC-IMDC interfaces to IMDC. Conversely, signature types associated with a dispersal (type 2) or absorption (type 3) of mechanical energy become increasingly dominant from IMDC to RMDC-IMDC interfaces to RMDC. The findings suggest acoustic emissions recorded during scratch testing and their subsequent characterisation can be used to indicate the fracture toughness of a given coke. This study contributes towards a broader program of research to improve understanding of the factors which influence the strength of coke and its microtextural constituents and interfaces, and how this relates to the properties of the parent coals.

The prediction of coke quality from global coal basins is critical to coke producers and steel makers for both the selection and effective utilisation of coals. This review analysed the methods described within published models for the prediction of coke quality. Of particular focus were methods that sought to predict coke strength after reaction (CSR) and the related coke reactivity index (CRI). Using the cross industry standard process for data mining (CRISP-DM) as an analysis framework, the models were compared in terms of their data treatment and use of analytical techniques. On reviewing these papers, our results indicate that it is difficult to apply models beyond the conditions under which they were derived, and that many models do not report enough detail to allow complete replication.

Within the coke making industry, the ability to accurately predict the quality of the coke produced from a variety of global coal basins is critical in both coal selection and blast furnace control. However, due to the complexity of the coke making process, the prediction of the resulting coke properties is a difficult task. This review analysed published models for the prediction of various measures of coke quality, with a particular emphasis on coke strength after reaction (CSR) and the related coke reactivity index (CRI). Focus was placed on the coal parameters selected as model inputs, and their reported behaviour with respect to the predicted coke quality. This review draws similar conclusions to previous analysis, namely there is a limited range of model applicability beyond the specific range of coals for which each model was derived. This conclusion is extended to suggest that the inconsistent utilisation of key attributes contributes to these limitations.

This paper has examined whether there are preferred pathways in the pore microstructure of coke. By modelling permeability and tortuosity in the digitised 3D structure obtained from 3D X-ray micro CT analysis of coke samples, preferred pathways were found to be perpendicular to the direction of applied force, i.e. parallel to the oven walls. The tortuosity was approximately doubled when moving from a direction of parallel to perpendicular with the oven wall and the permeability halved. It is hypothesised that applied force could squeeze the pore structure cutting off the openings in the direction toward the wall. Furthermore, by modelling mechanical properties of the digitised 3D structure it was also found that the Young’s modulus decreased by almost 25% going from the parallel to perpendicular direction (with respect to the oven wall). This result indicates that anisotropy in coke’s mechanical properties can occur. This anisotropy could have implications for coke strength. The coke might be more prone to breaking/shearing along particular planes when in the blast furnace. Furthermore, anisotropic permeability suggests that reactivity toward CO2 may also be anisotropic, and this could also have implications for strength deterioration behaviour in the blast furnace. The findings here are at a preliminary stage and it is recommended that the approaches developed here be applied to more coke samples to examine whether the findings here occur universally, and also the extent that anisotropic permeability and stiffness have implications for strength anisotropy during utilisation.

To better understand the connection between parent coal blend, coke structure, and coke strength novel structural identification and characterisation techniques for processing high resolution optical photomicrographs of coke were developed. Automated segmentation of inert maceral derived components (IMDC) and reactive maceral derived components (RMDC) allowed measurement of a broad range of parameters characterising coke structure. The characterisation included separate characterisation of IMDC and RMDC, IMDC boundaries including calculation of parameters characterising connections between IMDC and RMDC, coke porosity and, separately, IMDC porosity. Several other novel approaches using image analysis software were investigated, such as: • Separation of large porosity objects into individual pores (Separated Pores), • Removal of weak walls and hence potential identification and measurement of Weak Areas, • Identification of nodes (relatively thick agglomerations in the coke structure) and walls (part of the coke matrix connecting nodes), • Measurement of Modified Wall Thickness (when only walls are taken into account), Wall Neck Thickness (weakest points in walls) and Specific Neck Thickness (sum of all necks per unit area). Application of these novel characterisation techniques to a set of coke samples produced a substantial amount of data, which revealed many strong correlations between novel parameters characterising coke structure, and also between these parameters, coke strength indices and parent coal composition. This demonstrated the usefulness of the newly developed characterisation methodologies and gave a significantly improved understanding of coke structure and its connection with coke strength and parent coal blend. It should also be noted that these novel approaches to structural/textural characterisation can be applied to carbonaceous materials other than coke or other structural materials such as sinter, ore, and ceramics.

Laboratory washability testing on coal is used to determine the anticipated yield of coal versus density and coal quality data. The organic liquids which are commonly used are simple to use, but th...

Petrographic analysis of coke samples from the tuyere level of an operating blast furnace was carried out using a combination of advanced microscopy techniques including high resolution bireflectance and maximum reflectance measurements. A novel tribological approach was employed to investigate the wear mechanisms of the same cokes. The study showed that the wear mechanisms of coke are influenced by its microstructure and microtexture, as well as the strength and nature of the interfaces between the different coke textural constituents. The wear mechanisms of the cokes were found to vary with location from the tuyere entrance of the blast furnace and also differed to the feed coke. This is the first example of the application of petrographic analysis of cokes from an operating blast furnace to identify the difference in hardness and interaction strength of the distinct coke textural constituents. This research contributes towards improving understanding of the complex mechanisms of coke degradation in operating blast furnaces.

The effect of combinations of iron-bearing minerals and quartz on the reactivity of coke analogues with CO2 was examined in a TGA. Troilite–quartz, pyrite–quartz and magnetite–troilite binary combinations and the troilite–magnetite–quartz ternary combination were examined. For the troilite–quartz and pyrite–quartz binary combinations the reactivity in general decreased non linearly with increasing Si:Fe ratio. The magnetite–troilite combination had a non-linear effect on the reactivity in that mixes of these two minerals show a higher reactivity than would be expected from a simple proportional mixing approach. Some evidence of sulphur transfer from the troilite to the magnetite was observed. The reactivity of the ternary magnetite–troilite–quartz combination can largely be understood from the trends in the axes. These were largely followed across all compositions, allowing better understanding and prediction of the effect of these minerals on the reactivity of industrial cokes.

Metallurgical coke is a brittle composite material comprising predominantly carbon derived from both reactive and inertinite coal macerals. During cokemaking, these macerals form the porous coke matrix. The interfaces between the distinct maceral derived components, known as textures in the final coke structure, are well-known to be a major source of weakness in some metallurgical cokes. In this paper, we present a novel approach which applies techniques used in tribology to investigate the strength of the interfaces as well as the wear properties of the different textural constituents in metallurgical coke, at room temperature. The wear mechanisms which occur in coke were investigated using three pilot oven cokes of different single coal origin. A range of high resolution analytical techniques, including coke petrography, optical microscopy, stereo microscopy, and scanning electron microscopy, were used to determine the wear modes. The technique has also proven to be an effective method to determine the ...

Pore structure is known to play a role in the strength of coke. However, across cokes with similar porosities, strength can be variable, indicating that certain features of the pore structure are playing a role. An understanding of what those features are and how they form can pave the way for better coke strength prediction models and reveal new ways to manipulate the coking process to prevent or enable certain features forming. In this paper, rheometry has been used to characterise the viscoelastic properties and follow the pore structure development through normal force measurements as coal is heated at 3 °C/min. Coal samples were quenched in the rheometer at various temperatures corresponding to key events of bubble growth, bubble coalescence, minimum viscosity and resolidification and these quenched samples underwent micro-CT imaging, using the Imaging and Medical Beamline of the Australian Synchrotron to obtain 3D representations of the structure with approximately 10 μm resolution. The number of isolated pores, the pore size distributions of the isolated and connected pores, and the overall porosity were determined. Pore connectivity was found to be established very early on in the softening phase and coincides with the bubble coalescence phenomenon. For the high fluidity coals, virtually all of the pore space in the semi-coke is connected. It is proposed that bubble coalescence and applied forces play key roles in the contraction process and pore structure properties, in particular the size of the pore openings. In addition, during resolidification, the porosity reached a limit, however, the size of the pore openings continued to change, with a reduction in smaller openings and corresponding increase in larger openings. This trend suggests that the pore space becomes more channel-like, which could be due to the gas flow behaviour or could be due to surface tension effects. Semi-fusinite appears to reduce expansion. The precise mechanism by which it reduces expansion is not known. It may act as a conduit for the release of volatiles. This research is part of a broad set of activities which include understanding behaviour in the sole heated oven (SHO) and studying the mechanisms of coke breakage.

Micro-CT imaging of samples of metallurgical coke was achieved at high resolution (~ 10 μm), using the Imaging and Medical Beamline of the Australian Synchrotron. The coke samples were then stressed to breaking point and imaged again. The result is that we can examine, in fine detail in 3D, the locations of the fracture surfaces in the coke, and their relationship to the different aspects of the microstructure. In addition, we have performed 3D finite element linear elasticity analysis, using the microstructure of the cokes as input to the calculations. The result is a 3D map of the von Mises stress in the structure, which provides information about the likely locations of fracture. Finally, we have performed a fractographic analysis of the fracture surfaces, in order to characterise the nature of the fracture mechanisms. The imaging and analysis has been performed for both compressive and mixed-mode loading of the coke samples. The work provides the opportunity to identify the key mechanisms for fracturing of coke due to the aspects of their microstructure, and has the potential to aid in the production of stronger metallurgical coke, by means of blending coals to produce coke with improved strength properties.

Eight Australian coking coal samples (RvMax from 0.8 to 1.6) from six different basins were employed to investigate the effect of coal maceral compositions and rank on thermo-swelling behavior. Coal pyrolysis experiments were conducted at a heating rate of 5 °C/min from room temperature (25 °C) to 1000 °C with a top coal particle size of 212 μm. Thermoplasticity of coal was evaluated on the basis of the change of permeability of coal pellets, plastic range, swelling factor (defined by maximum swelling %/vitrinite %), and heat of exothermic reactions during the primary devolatilization. The endothermic and exothermic processes were identified by the estimate of the apparent specific heat using the computer-aided thermal analysis (CATA) technique, while swelling and permeability of gas flowing through coal pellets were simultaneously investigated with extended volumetric measurement downstream and pressure sensor upstream, respectively. Volatile evolution profiles of heating coal samples with temperature we...

Chemical changes of eight Australian coking coal samples from six different basins with various ranks (RvMax from 0.87 to 1.52) and maceral constituents (vitrinite from 48.2 to 76.5%) were dynamically investigated to explore the mechanism of coal coking linking to the synchronized physical and thermal changes. Volatile release during coal pyrolysis was monitored using a novel technique of dynamic elemental thermal analysis (DETA) that is able to differentiate tar and gas evolution in terms of carbon and hydrogen compositions. Condensed coal tars were characterized both chemically and thermally using the DETA and laser desorption ionization–time of flight–mass spectrometry (LDI–TOF–MS) techniques. Coal pyrolysis experiments were conducted at a heating rate of 5 °C/min from room temperature (25 °C) to 1000 °C with a top coal particle size of 212 μm, which kept the same experimental conditions as the physical and thermal measurements. The results indicated that, overall, the volatile evolution rates decrease...

Metallurgical coke is a complex brittle heterogeneous material consisting of carbon derived from fusible, semi-fusible and inert coal particles that forms a porous composite matrix. This paper presents a novel approach to assess and quantify the breakage behaviour and microstructural weaknesses in a pilot oven metallurgical coke. The approach uses fractography, a method commonly applied to determine the fracture behaviour and origin(s) in homogeneous materials, such as metals and ceramics. Determination of the fracture origin(s), paths of crack propagation and microstructural weaknesses in such a complex heterogeneous material as metallurgical coke represents a significant advance in both the application of fractography and the assessment of coke strength and breakage behaviour. Identification of the key features that contribute to the coke’s failure will facilitate better prediction of coke strength from coal properties and ultimately optimisation of the coal blending process. Key features and markings have been clearly identified on fracture surfaces that can either be traced back to the fracture origin or give an indication of the type of fracture or stresses to which the coke has been subjected, including the directionality and strength of those stresses. These markings include hackle, hackle twist and wallner lines, as well as markings generated by conchoidal and overload fractures. A three-step approach was applied to determine the breakage behaviour in stabilized lumps of the pilot oven coke, in which fractured coke surfaces were analysed at the macro, micro and submicron levels. The observed mechanisms of failure were quantified and summarised using a radar diagram.

Metallurgical coke used in blast furnaces is the result of blending coals to achieve the optimum coke strength properties. Prediction of coke strength from coal properties relies primarily on statistical methods. Improvement of these models is desirable. We believe a better understanding of the fundamental factors controlling coke strength is required to achieve this. In this second of a series of three papers, we apply our recently developed fractographic approach to examine and quantify the breakage behaviour and microstructural weaknesses in pilot oven metallurgical cokes from two series of single coals and binary blends of the coals. We have successfully applied fractography to perform an effective failure analysis of the complex, heterogeneous coke material. Summaries of the factors contributing to the failure of each coke studied have been presented as radar graphs. These graphs demonstrate that the failure profile of the cokes from blends is not simply a weighted average of the fracture behaviour of the cokes from the constituent single coals. Instead, non-additive effects are observed, and this may have implications for the mathematical models widely used to predict coke behaviour and strength in the blast furnace. Trends between the various fractographic parameters have been identified, and preliminary relation of these to both the maximum compressive strength of each coke and basic properties of the original coals, has been completed. Enhanced understanding of how the fracture behaviour of each coke relates to the properties of the coal blend used will help facilitate better prediction of coke strength from coal properties and ultimately optimisation of the coal blending process.

Metallurgical coke samples across the tuyere level of an operating blast furnace were analysed using a fractographic approach to identify and quantify various coke failure mechanisms. Tuyere level cokes displayed different failure mechanisms compared to the feed coke, implying a strong influence of the blast furnace reaction environment on the modification of coke failure mechanisms. The conchoidal and brittle overload type failures were noted to have an effect on the propagation of transgranular and intergranular crack growth in the coke microstructure. Feed coke breakage behaviour was strongly influenced by existing fissures and conchoidal fracture, while the breakage of the deadman coke lumps was dominated by pore wall collapse through intergranular crack growth in combination with erosive wear. The porosity at the IMDC–RMDC interface in coke has been identified as a critical parameter contributing to coke failure at tuyere regions. This fractographic study demonstrates that the fracture faces show evidence of the varying failure mechanisms presented in the preceding manuscripts (Part 1 and Part 2), in addition to the erosive wear resulting from the blast furnace exposure. However, further studies are required to establish the association between the identified coke failure mechanisms and coke texture, which has implications for improving coke performance in a blast furnace.

The changes in chemical structures over the plastic layer region during the coking of coals have a significant impact on coke formation and coke quality. This paper employed the Solid-state 13Carbon Nuclear Magnetic Resonance (13C NMR), and the Synchrotron attenuated total reflection Fourier transform infrared (ATR-FTIR) microspectroscopy (Synchrotron IR) to study the transformation of the chemical structures in plastic layer samples. The light gases (mainly methane and hydrogen) released from coking process were analyzed using micro gas chromatography (micro-GC) connected to a small coking reactor heated in an electric furnace that simulated the formation of the plastic layers. The results show clearly that the total aromaticity increased consistently in the plastic layers for all coals tested, while the amounts of side-chains decreased significantly during the plastic layer. There was a clear trend showing that the total number of bridge bonds and the looped structures, indicating that the degree of cross-linking would increase through the plastic layer. The plastic layer samples from low fluidity exhibited cross-linking structures with a high degree of branching and aromaticity, while those from high fluidity coals formed cross-linking structures with a relatively low degree of aromaticity and branching but with a large number of bridge bonds and looped structures. The transferable methyl, methylene and hydrogen were strongly correlated to the cross-linking reaction and side-chain elimination in the thermoplastic region, which is reflected by the release profiles of methane and hydrogen gas during the plastic layer stage.

This paper provides a review of the state-of-the-art research in the open literature on the carbon structure evolution in the semi-coke region following the last stage of the plastic layer transformation. Coking coals exhibit thermoplastic fluid-like behavior due to the change of chemical structures in plastic layers when heated in the coke ovens. Once the temperature of the coal charge exceeds the thermoplastic range, condensation, cross-linking, and repolymerization reactions take place. This results in the formation of a more ordered structure, referred to as semi-coke, with the final release of light gases. A further temperature increase leads to the release of hydrogen from the aromatic hydrocarbon structures with the formation of C–C bonds and, consequently, the carbon structure, which corresponds to the gradual transformation from the semi-coke to high-temperature coke. A variety of advanced analytical techniques have been employed, including infrared spectroscopy (IR), solid-state carbon-13 nuclear magnetic resonance (13C NMR) and X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), high-resolution transmission electron microscopy (HRTEM), and Raman spectroscopy. The carbon structure in coke generally is generally in the form of non-graphitic turbostratic structure, which exhibits isotropic property. There is a lack of information in terms of the 3D carbon structure model of coke/semi-coke and how these structures evolve from hydrocarbon sheets to more stable structures above 500 °C in the coke oven. This review also concludes future research scopes and the limitations of current knowledge.

The dispersion of Ca and the effect of Ca and Fe on the coke reactivity were studied using Sole Heated Oven (SHO) cokes prepared from washed Australian coal samples and added minerals, lime and magnetite. It was observed that the Ca from added lime were dispersed throughout the coke, with a smaller particle size, similar to the dispersion of Ca in the coke analogue, low ranked coals and coke prepared from low ranked coals. Ca had also reacted with minerals from the parent coals to form complex Ca-Al/Si oxy sulfides. The effects of Ca and Fe content on the coke reactivity with CO2 were studied using a TGA at 1373 K by adding varying amounts of lime and magnetite. Both lime and magnetite additions increased coke reactivity. It was found that the reactivity of lime added SHO cokes increased with increasing Ca content in the cokes. The reactivity of Fe containing SHO cokes increased with increasing Fe content in the cokes. However, the Ca addition had a much larger effect on coke reactivity, nearly doubling the coke reactivity compared to the Fe addition. The effect of the Ca addition on reactivity is likely to be accentuated by the dispersion of small Ca particles through the coke, which increased the mineral-carbon contact area.

When they are heated in coke ovens, coking coals become softened, coalesce into theremoplastic matter and then resolidify into a solid porous coke after the thermoplastic range. These stages take place layer by layer in the coke oven due to the temperature gradients. A coal charge in an industrial coke oven undergoes a thermoplastic transformation which corresponds to the formation of a plastic layer. The plastic layer is a foam-like thin layer consisting of the molten vitrinite and liptinite, inert coal components (i.e. inertinite macerals) and the volatile matter that is entrapped inside the layer. This is considered an important stage for the evolution of pore structures of a coke, which are closely related to the coke quality. This paper provides an overview of the current research in the operature literature into the thermoplasticity of coal, with a particular focus on the physical and chemical structure transitions inside the plastic layer using a variety of experimental techniques and the relevant phenomena such as IGP. The paper also analysed various factors affecting the formation and nature of the plastic layer such as coal rank, maceral compositions and rheological properties and coking conditions. The review included the most recent studies on plastic layers by the authors on pore structures analysed by using Synchrotron Micro-CT and chemical structure changes by employing FTIR techniques. The knowledge gaps and future research needs are identified and suggested.

A suite of coal maceral concentrates were prepared from a single coal using a water-based method for two discrete particle size fractions. Coal macerals were produced with the vitrinite content varying from 91.2% to 26.1% for the 106–212 μm particle size fraction and 96.0% to 38.2% for the 212–500 μm particle size fraction. Thermo-swelling of coal maceral concentrates were evaluated by the Computer Aided Thermal Analysis (CATA) technique with extended volumetric measurements. This novel CATA technique allows the apparent specific heat, thermal conductivity, and transient volumetric swelling of the coal sample to be measured simultaneously, as a function of temperature. The experimental results indicated that the maximum swelling (∼510 °C) and high-temperature contraction (600–1000 °C), as well as exothermic heat during primary devolatilization and thermal conductivity at maximum swelling all increased with vitrinite content, with the data inferring a linear relationship, and was independent of particle si...

Models of gas evolution and gas transport within softened coal particles, combined with bubble growth within the particles and gas flow between the particles, are used to develop a phenomenological model of the process of contraction and dilatation of coal samples in a dilatometer test. The model is fitted to dilatometer curves for a range of different coals, in order to determine unknown parameters of the model. The model is shown to provide excellent fits to the measured data for these coals, validating the assumptions underlying it. Finally, the model is used to predict the dilatation of blends. The comparison of the predictions with measured dilatation of blends is good for some cases, but there are discrepancies when non-expanding coals comprise a significant proportion of the blend. A hypothesis for the effect of the non-expanding coals upon the dilatation is proposed.

Dynamic measurements of physical, chemical and thermal changes in the transformation of coal maceral concentrates were made during heating at a rate of 10 °C/min to 1000 °C. The endothermic and exothermic processes were identified by measurements of apparent specific heat while the fluidity was indicated by the estimated thermal conductivity. Measurements of swelling and bed permeability were made, with continuous quantitative elemental analysis of gases and tars as they evolved. Data for two coal concentrates of high and moderate vitrinite indicate that the same reactions and events are occurring for the two samples, but to a greater extent for the high vitrinite sample. The research has noted the significance of evolved tars in the early events, being the lowest temperature event identified, with rapid tar evolution prior to the onset of swelling associated with permeability change. Further tar release and gas evolution is associated with a rapid swelling event, this event being substantially greater for the high vitrinite sample. The data has also quantified contraction at higher temperatures following swelling which is associated with the release of hydrogen containing gases. Evolved tars from the high vitrinite sample showed elevated H/C ratio indicating that vitrinite tars appear to be more aliphatic than those evolved from inertinite. A comparison of measured swelling with estimated volumetric flow rate of the volatiles has indicated that thermo-expansion of coal utilised up to 21% of the volatiles to drive bubble growth. This utilisation rate varied significantly across the plastic temperature range.

The fundamental mechanism(s) for high oven-wall pressure are still not completely understood. The hypothesis put forward in this series of papers is that bubble-growth combined with a lack of bubble-coalescence in the plastic-layer is the primary reason for high oven-wall pressure, and that a lack of bubble-coalescence occurs when the minimum viscosity and elasticity are above a certain threshold. Because it is known that coal oxidation decreases its fluidity, an examination of the changes to both viscoelastic properties of the plastic-layer and oven-wall pressure during the coking of oxidised coals was considered to be a promising way to test the hypothesis, as most of the other properties of the coal, such as volatile matter, remain unaltered. Viscoelastic properties were measured using high-temperature oscillatory shear rheometry. For the fresh coals studied, the viscoelastic properties were such that bubble-growth and bubble-coalescence occurred and the oven-wall pressure was low. After subjecting the coal to enhanced oxidation, the minimum viscosity and elasticity increased to a level whereby bubble-growth occurred but bubble-coalescence did not. A large increase in OWP was found to coincide with this change. It is envisaged that bubble-coalescence enables channels to form in the plastic-layer and that the continuous release of volatiles keeps the channels open. Therefore, bubble-coalescence is considered to strongly influence gas permeability. When gas permeability through the semi-coke is severely restricted, volatiles would be forced to move to the centre of the charge, condense, and then revolatilise as the temperature rises, leading to high late OWP peaks. If oxidation was allowed to progress further, it is anticipated that viscosity and elasticity would increase to a level whereby bubble-growth would be restricted and OWP would not be high, but the coke would be highly non-fused and very weak. These results may elucidate why variable OWP results are found for different coals before and after oxidation; it depends on the initial and final viscoelastic properties. For another coal, OWP remained low at

a b s t r a c t A new technique that allows dynamic measurement of thermal properties, expansion and the elemental chemistry of the volatile matter being evolved as coal is pyrolysed is described. The thermal and other properties are measured dynamically as a function of temperature of the coal without the need for equilibration at temperature. In particular, the technique allows for continuous elemental characterisation of tars as they are evolved during pyrolysis and afterwards as a function of boiling point. The technique is demonstrated by measuring the properties of maceral concentrates from a coal. The variation in heats of reaction, thermal conductivity and expansion as a function of maceral composition is described. Combined with the elemental analysis, the results aid in the interpretation of the chemical processes contributing to the physical and thermal behaviour of the coal during pyrolysis. Potential applications in cokemaking studies are discussed.

The properties of metallurgical coke are critical to the performance of the blast furnace for iron making. While extensive work has gone into understanding the relationship between the properties of coke and its behaviour in the blast furnace, the heterogeneous nature of coke makes understanding these relationships difficult. Herein we report the use of a laboratory produced coke analogue to examine the effects of minerals on the reactivity behaviour of coke in a pseudo-CRI test. The effects of a range of minerals on the reactivity of the coke analogue are demonstrated and the effects of binary combinations of silicon and iron bearing minerals are examined. The properties of the metallurgical coke used in the blast furnace for steel making have a direct effect on the quality and productivity of the steel making process. Coke provides three primary functions in the blast furnace. 1) It is fuel for the furnace, the carbon source for the CO to convert iron oxide to metallic iron and provides the structural support for the blast furnace burden. The structural support of the burden during operation of the blast furnace is critical as collapse of the burden reduces gas and liquid permeability 1) consequently reducing the efficiency of the blast furnace operation both in reducing through put and in increasing carbon dioxide emissions per unit of iron produced. A number of factors affect the performance of metallurgical coke in the blast furnace. These include carbon forms (microtextures), pore network and catalytic effects on the reactivity due to minerals present in the coke. 2) It is desirable to predict the reactivity of coke from its key characteristics. Unfortunately this is in part limited by unknown or non-quantified effects of minerals on coke reactivity. Elucidation of the effects of minerals on coke reactivity has proved difficult due to the complex heterogeneous nature of coke and potential non-additive effects of minerals on coke reactivity. Several different approaches have been taken to understanding the effects of minerals on the reactivity of metallurgical coke. Two principle methods have been utilised to study the effects of minerals on coke reactivity. Early attempts were made through doping of coke making coals with mineral additions 3,4) and studying the subsequent reactivity. Later techniques involved statistical analysis of a large range of cokes of known mineralogy and reactivity to attempt to identify the important mineral species with respect to coke reactivity. 5–8) Early studies examining doped coke reactivates 3) indicated that the mineral cations present coke could be ranked as follows in terms of their effect on reactivity

A novel thermal analysis technique – Dynamic Elemental Thermal Analysis (DETA) – is demonstrated for the characterisation of biomass. The technique provides a unique characterisation of the pyrolysis streams by conversion into combustion products using a custom built downstream O 2 lance. The result is a continuous characterisation of evolving total volatiles, light gases and tars and solid char, in terms of the elements CHNOS, with pyrolytic heating. The condensed tars were also characterised as CHNOS with boiling point. The generated suite of integrated data provided quantitative mass yields and elemental compositions that compares well with standard analytical methods. The results were used to estimate the thermal properties of the biochar and the heat of combustion of the volatile streams (i.e. total volatiles and gas only), allowing a dynamic evaluation of the energy balance of a biochar generation process. At an illustrative temperature of 500 °C for a Spotted Gum biomass, 17% of the biomass carbon is retained in the biochar. The results are used to estimate the potential revenue from each product of the heating process. The economic assessment showed that bio-oil (being the largest fraction) would provide the greatest income stream as either renewable electricity or when sold as crude liquid. The sequestration of carbon as biochar would be the smallest income stream for a carbon price of $23/tonne CO 2 (around 4% total revenue). Overall, the DETA technique was shown to provide data suitable for assessment and optimisation of a biochar production technology.

A novel technique is described which provides quantitative and continuous analysis of light gas and condensable tar components as they are evolved in terms of carbon, hydrogen and oxygen. The technique has also been used to directly characterise the total tar sample in terms of carbon distribution and boiling point. It has been found that changes to the dynamic tar-H/C ratio correspond well with particular temperatures measured by Geiseler Plastometer for softening, maximum fluidity and re-solidification. This technique can enhance the chemical understanding of mechanisms occurring during de-polymerisation and cross-linking of coal (i.e. metaplast development and the transfer of hydrogen) while also monitoring tar evolution. A tar collection and re-vaporisation method provides a means of identifying tar groups that contribute towards the metaplast phase and temperatures at which they evolve. Both methods are unique in studying chemical aspects of coal and tar behaviour with heating, in a field based on thermo-physical techniques (e.g. H1 NMR, high temperature rheology, Geiseler plastometry, dilatation). Overall, the Dynamic Elemental Thermal Analysis (DETA) technique can give new insight into the fundamental mechanisms prevalent in coal pyrolysis and provides quantitative chemical assessment of tar nature (i) during the heating of coal and (ii) as a final (total) condensed product.

A fundamental study on the behaviour of heating coal macerals has been undertaken using two novel thermal analysis techniques. The apparent specific heat was determined during heating using an inverse calorimetric method (computer aided thermal analysis, CATA) and combined with pressure and displacement measurements to correlate endothermic and exothermic behaviour with measurement of swelling. The second technique used a post-oxidation stage to combust the tars and gases into products which were analysed. This method was used to study the elemental character of volatiles release from coal maceral concentrates in terms of carbon and hydrogen. Extents of swelling and exothermicity during primary devolatilisation were found to be correlated with vitrinite content and were associated with tar evolution. For the highest vitrinite fraction (of 86.4% vitrinite) swelling was initiated at the same temperature range for exothermic reactions, and maximum swelling coincided with the peak release of light gases. Tar evolution was found to change in chemical character (as defined by H/C ratio) during progressive heating, initially rising in the early stages of tar formation (

This work follows on from two papers which presented a hypothesis for the mechanism behind oven wall pressure. It was proposed that oven wall pressure results when the period of pore growth is lengthy prior to coalescence and/or the period of coalescence is minor ahead of resolidification. When this occurs a thick swollen band of closed pores in the viscoelastic medium results which exerts pressure on the oven walls. The band is a resistance to volatile transfer and the largely closed pore network remains throughout resolidification further exacerbating pressure on oven walls. This hypothesis has resulted from a mapping of the viscoelastic properties obtained from a rheometer against the plate behaviour of the rheometer where pore nucleation and growth has been characterised by an increase in plate gap (Δ L ) when the plate is held under constant axial force (AF) or an increase AF when the plates are held at constant gap. Pore coalescence has been characterised by a peak in AF or a decrease in the rate of expansion of Δ L . Because the behaviour of either Δ L or AF has brought about our understanding of coking pressure this paper attempts to prove that pore coalescence can be monitored from plate behaviour. As such this paper attempts to track pore growth and coalescence using mercury porosimetry, permeability, scanning electron microscopy and optical microscopy, all of which have been carried out on quenched coke samples. Scanning electron microscopy images showed that the Δ L profile provides an accurate account of the onset of bubble nucleation, growth and coalescence. Porosity measurements indicate that the proportion of open pores in the 1–100 μm range increases significantly after the peak in AF or decrease in the rate of expansion of Δ L , indicating that these points coincide with a bubble coalescence phenomena. Optical microscopy showed a distinct difference in pore structure between the image corresponding to the peak in axial force and images after, whereby the number of isolated cells decreased and pore walls became thinner. Permeability measurements on quenched samples throughout the softening process show an initial drop in permeability due to the loss of interparticle voidage and the onset of swelling followed by a steady increase coinciding with the peak in axial force and decrease in the rate of Δ L expansion. This work has helped to reinforce the proposed mechanism for coking pressure. Furthermore, permeability measurements of quenched samples at 550 °C show that the high pressure coals tend to have the lower permeabilities at this temperature. The finding that pore growth and coalescence behaviour is what impacts on coking pressure has opened the door for finding novel ways to manipulate pore behaviour and therefore manipulate coking pressure. Understanding and manipulating pore growth behaviour has other implications, largely the understanding for coke strength development, which is currently being explored by the authors.

The reflux classifier is a water based device designed for classifying and separating particles using inclined plates. Particles of sufficiently high settling rate are refluxed back to the slurry, while slow settling particles (low density vitrinites) are removed in the overflow. In this study, the reflux classifier was used to separate coal macerals less than 212 μm in size. The thermo-swelling characteristics of separated macerals were evaluated on the basis of the Computer Aided Thermal Analysis (CATA) technique with extended volumetric measurements. A suite of maceral concentrates with various densities, ash contents, particle sizes and vitrinite percentages was obtained, which indicated that the reflux classifier can offer an alternative method of macerals separation/concentration without the use of dense media. The pyrolysis experiments of maceral concentrates indicated that the vitrinite-rich macerals showed a higher extent of volumetric swelling, exothermic reactions and thermal conductivity between 400 and 600 °C than inertinite-rich macerals. Addition of inertinite to vitrinite concentrates showed that inertinite can suppress volumetric swelling and shift the swelling to a higher temperature during the primary devolatilization, but does not have a significant influence on high temperature contraction. Further observations about the interaction among different maceral groups were also discussed in this study.

Large, uniform sized coke is desirable for blast furnace use. It has previously been shown that the coke oven flue temperature in the first few hours of coking is a key determinant of coke size. In this paper, the authors present a new programmed heating approach, which is called pulsed heating, aiming to increase coke mean size at a given average flue temperature. The approach takes into account the charging sequence in coke oven batteries and the authors demonstrate how existing operating practice can be modified in batteries with suitable heating systems to achieve the desired heating programme. A mathematical model of fissure formation provides a prediction of the increase in coke mean size using pulsed heating, compared with standard heating. Pilot scale experiments have also been performed to validate the modelling approach. The results of the modelling indicate that the mean coke size can be increased by several millimetres in some cases, although results from the pilot scale show that puls...

One of the most important aspects of the cokemaking process is to control and limit the coking pressure since excessive coking pressure can lead to operational problems and oven wall damage. Following on from a previous paper on plastic layer permeability we have studied the effect of contraction of semi-coke on coking pressure and the effect of organic additives on contraction. A link between contraction (or simulated contraction) outside the plastic layer and coking pressure was demonstrated. The interaction between this contraction, local bulk density around the plastic layer and the dependence of the permeability of the plastic layer on bulk density was discussed as possible mechanisms for the generation of coking pressure. The effect of blending either a high volatile matter coal or one of two semi-anthracites with low volatile matter, high coking pressure coals on the coking pressure of the binary blends has been explained using this mechanism.

A number of coal blends and pitch/coal blends were evaluated using rheometry, thermogravimetric analysis and microscopy to confirm and further elucidate the coking pressure mechanism previously proposed by Duffy et al. (2007) [1]. We confirm that blending a low rank, high fluidity, low coking pressure coal, with a high rank, low fluidity, high coking pressure coal can significantly reduce the coking pressure associated with the latter. Interestingly, blending does not necessarily result in a fluidity that is midway between that of the two coals; sometimes the fluidity of the blend is less than that of the low fluidity coal, especially when the coals are significantly different in rank. This occurs because the increase in complex viscosity (η*) through resolidification of the low rank, high fluidity coal counteracts the reduction in η* resulting from softening of the high rank, low fluidity coal. It has also been confirmed that the η* of the resultant blend can be estimated from the η* of each component coal using a logarithmic additivity rule commonly employed for polymer blends. Polarised light microscopy has indicated that the degree of mixing between coals of different rank is minimal, with fusion restricted to the particle surface. It is therefore inappropriate to think of such a coal blend in the same way as a single coal, since each component coal behaves relatively independently. This limited fusion is important for understanding the coking pressure mechanism for blends. It is proposed here that the lower rank coal, which softens at lower temperature, is able to expand into the interparticle voids between the high rank coal that is yet to soften, and these voids can create channels for volatiles to traverse. Then, and importantly, when the high rank coal begins to expand, the pore structure developed in the resolidified structures of the low rank coal can facilitate removal of volatiles, while the resolidified material may also act as a suitable sorbent for volatile matter. This is considered to be the primary mechanism by which coal blending is able to alleviate coking pressure, and applies to addition of inert material also. Addition of a coal tar pitch was found to increase fluidity but also to extend the thermoplastic range to lower temperatures. This caused an increase in the swelling range, which was accompanied by a long plateau in η*, a feature which has previously been observed for certain high fluidity, high pressure coals. Elasticity and η* at the onset of expansion were also higher for both the pitch impregnated coals and the high pressure blends, which supports previous findings for singly charged high pressure coals, and confirms the potential use of such criteria for identifying potentially dangerous coals/blends. © 2009 Elsevier Ltd. All rights reserved.

Pilot scale coke oven experiments, in which carbonisation has been arrested after 2 and 4 h, have been used to identify key aspects of the initial formation and propagation of fissuring in coke. The experiments show that the primary fissure network is formed within the first few hours of coking and the fissures propagate towards the centre of the oven as coking progresses. They also show that some of the initially formed fissures stop propagating, thus increasing the effective fissure spacing as coking continues. A model for the propagation of regular crack arrays has been used to identify the cause of the initial formation of the fissure network, evaluate the fissure spacing and explain the fissure coarsening effect. The coarsening is shown to be due to an instability, in the form of every second fissure stopping. The formation of lateral fissures is shown to be due to tensile stress formation near the tips of stopped fissures and also to increased shrinkage due to a maximum in the coke contraction coefficient at around 700 °C.

We investigate the effects of the heating rate, coke shrinkage and coke breakage strength upon the fissure pattern developed in a coke oven charge during carbonisation. This is done principally using a mechanistic model of the formation of fissures, which considers them to be an array of equally spaced fissures, whose depth follows a “period doubling” pattern based upon the time history of the fissures. The model results are compared with pilot scale coke oven experiments. The results show that the effect of heating rate on the fissure pattern is different to the effect of coke shrinkage, while the effect of coke breakage strength on the pattern is less pronounced. The results can be seen in both the shape and size of resulting coke lumps after stabilisation. The approach gives the opportunity to consider means of controlling the carbonisation process in order to tune the size of the coke lumps produced.

In this study a number of high coking pressure coals with different fluidities were evaluated alongside a number of low pressure coals also with differing fluidities. This was to confirm findings from an earlier study using a limited selection of coals, and to establish rheological parameters within which a coal may be considered potentially dangerous with regards to coking pressure.

A model of coke stabilization, based on a fundamental model of fissuring during carbonisation is used to demonstrate the applicability of the fissuring model to actual coke size distributions. The results indicate that the degree of stabilization is important in determining the size distribution. A modified form of the Weibull distribution is shown to provide a better representation of the whole coke size distribution compared to the Rosin–Rammler distribution, which is generally only fitted to the lump coke. A statistical analysis of a large number of experiments in a pilot scale coke oven shows reasonably good prediction of the coke mean size, based on parameters related to blend rank, amount of low rank coal, fluidity and ash. However, the prediction of measures of the spread of the size distribution is more problematic. The fissuring model, the size distribution representation and the statistical analysis together provide a comprehensive capability for understanding and predicting the mean size and distribution of coke lumps produced during carbonisation.