Microstructure Variability in Coke and its Effect on Coke Properties

The objectives of this project were to:

· Obtain micro-CT images of the microstructure of various cokes;

· Use the microstructure information to estimate strength and reactivity properties of the cokes using a finite element approach;

· Compare the estimated properties with measured indices of coke strength and reactivity; and

· Seek correlation with the coal properties of the various cokes.

Firstly, a set of cokes, from coals covering a reasonable range of rank, were obtained from Australian coking coal companies. These coals were then sampled and imaged using the CSIRO micro-CT scanner, as well as the Australian synchrotron. The data produced was then analysed using (a) 3D image analysis techniques and (b) 3D finite element analysis. This analysis gives us detailed quantitative information about the microstructure of the coke(s) and how they relate to the macroscopic properties which affect coke quality, such as strength and reactivity. Finally, the measures of strength and reactivity that were determined from the microstructural analysis were compared with industry measures of (abrasive) strength and reactivity (CRI).

Some key results from this work are that:

· We can identify differences between cokes using this technique. The differences can be identified visually (by simply observing the 3D structure) but more importantly quantitatively, by developing measures of the structure and its effect upon coke quality;

· We have developed quantitative measures of the coke quality based on the microstructure of the coke and shown how these relate to key coke quality properties.

We envisage a range of future activities for micro-CT that will enhance the development of our understanding of the factors that control coke quality and, ultimately, aid in identifying means for obtaining optimal quality for given coal inputs. As a result, we see this work as a precursor to the development of a suite of capability that will be useful in both cokemaking research and in helping to value the coal assets of Australian coking coal miners.

In particular, we are proposing to image and analyse more coke samples using this technique, to provide a larger dataset to strengthen the results developed here. We also propose using the technique to analyse the development of coke microstructure, which will guide the development of optimal coke properties.