Direct Imaging of Carbon-dioxide Penetration in Cokes Pre- and Post-reaction

The suitability of carbon coke for use in a blast furnace is determined by a range of factors other than composition such as strength and reactivity, both of which are critically dependent on the coke microstructure. To improve the ability to predict coke behaviour through a better understanding of the impact the microstructure has on these properties and how the microstructure is in turn affected by the reaction of coke with carbon dioxide is a long term goal.

Project C27056 used Xenon gas as a CO2 analogue for imaging gas uptake and gas transport in these materials to determine the 3D distribution of the gas within the coke structure. Results obtained in regions where the gas was absent indicated that the sample is impenetrable to gas; conversely, in regions where there was an elevated density of gas compared to the free gas it can be concluded it must be sorbed onto surfaces due to the region's accessible high surface area. Thus, the gas density distribution in the image provides a measure of the distribution of accessible surface area in the sample.

Project C27056 found:

• In unreacted coke there was little gas uptake overall, indicating that there was little surface area to sorb gas. There were rare regions where gas was strongly sorbed, indicating these regions had high surface area.
• In reacted coke, there were large regions with strong sorption and some with none. The highly-sorbing regions were always associated with inertinite maceral derived component (IMDC). But not all IMDCs were created equal or react equally. Reaction also introduces new and extended surfaces for gas reaction.

Given the high gas concentrations of xenon sorbed by the coke after reaction in some regions it was predicted that you should be able to directly image the distribution of CO2 uptake in coke and reacted coke using x-ray imaging.

This led to the current project, which was designed to answer the following questions:

• To what extent could the uptake of CO2 gas into coke samples could be observed using synchrotron micro-tomography (micro-CT), and could micro-CT also detect changes in CO2 sorption onto coke?, and
• If CO2 sorption onto coke samples could be detected using imaging, then to what extent is sorption of CO2 onto coke similar to xenon sorption? Can xenon imaging provide a reasonable analogue of CO2 sorption?

Synchrotron micro-CT scans of coke samples subjected to pressurised with CO2 gas before reaction and after reaction to 20-50% mass loss with CO₂ were carried out. Unlike the earlier experiments with xenon which benefitted from the high x-ray contrast of xenon gas and could make use of a specialised technique known as K-edge subtraction, imaging CO2 itself is much more challenging. It has much lower contrast and we can't use K-edge subtraction, but use instead the comparison of images acquired with and without CO2 to determine the CO2 distribution. However, since CO2 is the gas of interest in a blast furnace, imaging CO2 directly rather than using an analogue is highly desirable.

It was found that:

• CO2 sorption onto coke, and especially reacted coke, was readily detectable using X-ray imaging (though not as well as with imaging using xenon, as expected);
• We could distinguish regions of high CO2 sorption from regions of low CO2 sorption;
• In agreement with the results for xenon imaging, the greatest sorption was seen in partially reacted IMDC;
• A direct comparison between CO2 sorption and xenon sorption found good agreement between the regions where the two were sorbed. However, there were minor exceptions;
• Xenon is a reasonable but not perfect analogue for exploring where CO2 is sorbed onto coke and identifying the regions in coke with greatest surface area and the greatest likelihood for fast reaction;

The behaviour of xenon and CO2 in coke are sufficiently similar that, given the relative ease of imaging and quantification with xenon compared to the much weaker/noisier signal with CO2, xenon is generally preferable for future experiments to determine regions of accessible surface area likely to react rapidly.