Physical and Chemical Interactions Occurring During Cokemaking and Their Influence on Coke Strength

This project builds onto a previous project , C24055 in which macerals were separated and then reblended in order to identify whether any interactions were occurring between vitrinite and inertinite. There were strong signs that inertinite can either trap or aid volatile release which affects the driving force for liquid evaporation and leads to either an increase or decrease in the amount of liquid, respectively. An increase or decrease in the amount of liquid decreases or increases the viscosity of the melt which controls bubble growth and coalescence behaviour. Therefore, it was proposed that inertinite may indirectly influence adhesion and pore structure for a given coal in either a positive or negative way. This project focussed on endeavouring to understand these mechanisms in more detail.

In C24055 the reflux classifier was used, however, it was found that it wasn't able to obtain high purity concentrates of inertinite without also having high levels of mineral matter. There was also strong signs that the inertinite concentrate was oxidised. Oxidation is a major problem when studying interactions as it can lead to inaccurate assessments of the mechanisms occurring. In this project a different method was tried which involved hand-picking within an inert atmosphere. It was successful in obtaining high-purity concentrates. It is important to note that this method can be time consuming, particularly when the vitrinite and inertinite are well mixed. Therefore, only small samples can be obtained.

The vitrinite and inertinite were remixed and studied viscoelasticity and volatile release behaviour using rheometry, dynamic elemental thermal analysis (DETA), and thermogravimetric analysis (TGA). Using the behaviour of the high purity concentrates predictions could be made on how the blends should behave and then compared to what is measured to identify whether interactions are occurring.

The rheometry tests for all three coals studied showed that the inertinite causes a delay in softening and expansion by providing more time for the vitrinite derived liquid to flow into the void space and allowing more time for volatiles to diffuse out before the liquid fills the void space and bubble nucleation and growth (expansion) occurs. Regarding the minimum viscosity values, for a high rank coal, the measured values were very close to the predicted. For lower rank coals, the measured values were slightly lower than the predicted, which indicates that inertinite may aid in trapping liquid matter. For the DETA tests slightly lower volatile release rates were obtained when the level of inertinite concentrate was low (25-30%), and then higher when the level was high. For TGA, the volatile release rate was as predicted for the high rank coal, but it was reduced during the region of maximum volatile release rate for the lower rank coals.

These results indicate that inertinite may act to trap and keep volatile matter and therefore liquid matter in the melt. The transfer of a molecule from liquid to gas depends on temperature and pressure and its ability to diffuse to a gas-liquid interface and evaporate into the gas phase. The inertinite may be altering the diffusion path length such that the volatile matter cannot easily diffuse out of the liquid without forming a bubble. Another possibility is that inertinite reduces the gas-liquid interfacial area available for evaporation.

To gain a greater appreciation of the role of inertinite a range of solid additives was studied. The inertinites studied in this project behave similar to graphite. This work showed that the pressure in the vitrinite melt can build up to a greater extent when graphite is added, which provides evidence for the hypothesis that inertinite may trap and keep volatile matter in. Regarding viscosity, inertinite and graphite appear to increase the viscosity in a predictable manner according to the following equation where η∗ is the viscosity of the blend, η_v∗ is the viscosity of the vitrinite concentrate and φ_I is the wt fraction of solid added.

It was found that mesoporous solids do not follow this equation; the exponent is much higher such that small additions result in dramatic losses of liquid and increases in viscosity. Correcting for density differences does not explain the loss of liquid. It was initially speculated that the liquid may simply be flowing into the pores. To study whether this might be occurring the charcoal and graphite was added to a silicone oil which has a similar viscosity to the melt of 100 Pa.s. The increase in viscosity when either graphite or charcoal is added to silicone oil is identical, so it seems unlikely that the loss of liquid is due to it flowing into the pores. The only plausible explanation at this stage for the loss of liquid for mesoporous solids is that volatiles are being adsorbed onto the surface area of the solid's porous structure, lowering the partial pressure of those molecules in the vapour space. A lowered partial pressure would establish a driving force for enhanced evaporation of those molecules, thereby causing more liquid to evaporate and an increase in the viscosity. Why is this result for mesoporous solids important? Whilst this effect was not noticeable with the inertinites studied here it does show that a highly porous inertinite could have a substantial impact on the behaviour of liquid and volatile components, affecting expansion, adhesion, and pore structure properties.

'Cross-fertilisation' studies were undertaken by mixing the inertinite from one coal with the vitrinite from another coal. No significant interactions were noticed, even given the proportions of fusinite and semifusinite were different. The possibility of a chemical interaction was noticed when the high rank inertinite was mixed with the lower rank vitrinites. There was a substantial loss of liquid which is considered to be due to an early onset of resolidification. This may be due to the vitrinite donating (losing) hydrogen-rich radicals to the inertinite.