ACARP Project Number: C20010      Published: June 12

Despite the simple chemical composition of metallurgical coke, the structure is complex, consisting of a network of variously sized pores and walls of varying thickness. Adding to this complexity is the varying forms of the carbon in the pore walls which have different microproperties and vary in the optical anisotropy dependent upon the parent coal properties. It is well known that the microscopic structure and texture of coke is of intrinsic importance to both the hot and cold strength of coke. Many papers have been written about coke characterisation, and the relationship that the parent coal's maceral composition has with the produced coke's microtexture and microstructure. As a result a large number of classification systems have been developed to differentiate the various microscopic features of cokes. Despite the large variation in classification systems around the world coke microtextures have been and are increasingly used to assist in coke quality assessments manually and automatically by microscopic imaging techniques. While it is recognised that microscopic coke composition is directly related to coke quality parameters, differences in classification are a limiting factor in the use of microtextures in the international community.

For the purposes of this report, coke microtexture describes the nature of the carbon in coke, the crystalline development and the degree of anisotropy present. Whereas coke microstructure describes the spatial relationships present in the coke material for example porosity and pore wall thickness.

The objectives of this project were to review the coke classification systems in existence to assist in the development of a comprehensive and uniform classification that is applicable to Australian and international coals. A reference document developed as part of the project that outlines this classification, including photomicrographs, can be used to assist petrographers when applying the classification to the analysis of metallurgical cokes. The application of the classification system selected and the usefulness of the reference document produced will be tested by a round robin by Australian petrographers using two Australian cokes.

As a result of the review of the various classification systems in existence a reference document was produced detailing a method of microtextural classification selected to be tested by a round robin of various petrographers within Australia. The classification method selected is known to be used within Australia and is able to be applied to cokes that have been made from coals with a range of rank and coal types. Additionally the method uses the results generated to give an overall measurement of anisotropy known as the coke anisotropy quotient (CAQ) which has been related to coke quality parameters such as coke strength after reaction (CSR).

Two cokes were dispatched to a number of petrographers for analysis, the cokes selected were manufactured from two coals being tested at ALS Coal's Coking Research Centre, the two coals had similar rank but different maceral concentrations. Only two petrographers from one laboratory returned results for the round robin. It was found that between petrographers using the same classification system the fused carbon was able to be determined with good reproducibility. Despite differences in the classifications of the domains within the fused carbon the overall anisotropy measured by the CAQ was very similar between the two petrographers. It is suggested that it is most important for the CAQ to be close between analysts as this is the parameter that is used to relate the coke microtexture to other coal quality parameters.

The round robin samples were also sent to two commercial laboratories that conduct coke microtextural analysis. One of which conducts manual coke microtextural analysis and the other automated imaging analysis. The results obtained from these laboratories were compared to the mean results from the round robin. It was found that for one coke the total proportion of fused carbon is similar between the three methods but for the other coke the total proportion of fused carbon is quite different. It is possible that this is due to the classification of partly fused carbon. Additionally differences between the analytical methods and classification methodologies used to differentiate the texture of the fused material do not allow for an easy comparison of the results between laboratories.

It is recommended that further dialogue between coal producers and petrographers occurred at Australian and International levels to determine the most suitable method for characterising coke at a microscopic level with the aim to develop an AS or ISO standard method.