Monday, June 28, 2010

potential calamity averted by a careful change in the erosional surface orientation

My most recent experiment is one I’ve long been looking forward to analyzing—this is the highest-pressure run I’ve done yet. We gave it nearly 500 hours in the piston cylinder, while I was busy with lots of travel and a minor bout of food poisoning. Once we finally downloaded it and I rescued the capsules from their nest I discovered that they were stuck together. We always run two capsules (2 mm diameter gold tubes, welded shut around the experimental charge) in every experiment, with different compositions in each.

Usually it is enough to set them in the palm of my hand and rub them back and forth a bit to cause them to break apart before I carefully break off the last of the MgO, salt and graphite that clings to the capsules. This time they stayed stuck together no matter what I tried, so I consulted with my boss, who suggested that I try carefully grasping each with jeweler’s pliers and seeing if I can break them apart that way. Alas, they did not separate, instead one of the two capsules started to tear open.

Therefore I stopped trying and instead we put both capsules into a single epoxy mount. Knowing that it would require a fair bit of luck for the capsules to be oriented within the mount such that it would be possible to expose the insides of both capsules when I went to polish them I put off the task and spent the weekend visiting with friends. This morning I finally decided to look at the mount and try my hand a polishing them. When I looked I determined that the important ends of each capsule were not on the same plane with respect to the surface of the mount, but it looked like if I were careful to put more pressure on the one side than the other it might be possible to change the dip of the top of the mount such that both capsules were intersected. Much to my delight I managed to accomplish exactly this task. I now have one mount with two capsules, both exposed, all polished and turned in to be carbon-coated in preparation for tomorrow’s microprobe session.

This makes me very happy. I did not like the alternative, which would have been polish enough to expose one of the two, analyze and photograph everything in it and then polish it away to expose the other one for analysis. This way we will be able to go back and re-analyze anything we want at any time, rather than having one of them cease to exist in order to reach the other.

No comments: