Resolution in dense ores is still an issue for X-ray CT from my limited experience. This is limiting if you intend to detect cracks generated by different grinding modes. But, I fully agree that fundamental research must go on on that topic.

Concerning 2D vs 3D liberation it is only realistic to achieve results on (very) simple ores. My personal experience is linked to 2D vs 3D size and shape analysis. This is up to now the most feasible application for a better understanding of analytical biases (in conventional particle analysis)

Sorry, but are we talking about conventional, laboratory X-ray sources or Synchrotron?

Sorry indeed. I was speaking about desktop X-Ray imaging and not having access to the very latest technologies -not to mention any brand 😉

I know of a lab that has largely ignored X-ray CT scanning and 3-d liberation data because we have a huge comparative database of 2-d liberation data from SEM-based sources. For mineral concentration circuits, one of the most frequent questions is why has the recovery changed? This can be due to ore source, or operational drift. To help decide, one of the things we most commonly look at is comparative liberation data between recent lower recovery days and historical data with higher recovery. I prefer to compare my apples to apples - so I stick with what works and what I have which is 2-d data. So I suppose there is a certain amount of data inertia for these systems to overcome. Note that it was easy for us to make the change to SEM-based systems from optical based, because we still had the same 2-d polished blocks. And there was the added bonus of not having to do all that dreaded point counting.

There will have to be a step-changing advantage for these systems to take hold - either drastically reduced price or increased speed or resolution that you just can't do without. Simply saying that your sample is 89% liberated in 3-d vs. 96% liberated in 2-d will not be enough.