Before pride, comes a fall. At the end of term I had a whole slew of results, they were inexplicable at this stage, I'd got them reproduced, the highest melting point that had ever been obtained for iodine monochloride, therefore it was very pure. I said, okay, Norm, take a break, go skiing.

So I went skiing and, to Sestriere, with a girlfriend whom I'd met through my cousin. She was at Newnham College, and we went to Sestriere with the Oxford and Cambridge Ski Club, which was an interesting story I might come to later. But we had this skiing holiday, I came back, and the first thing in those days that any budding chemist did when he'd been away for an hour... for a month, on vacation, is to go to Chemical Abstracts and see what had happened in the month. And there amongst the first papers I saw, with Fialkov in the Neorganicheskoi Khimii, the Russian Journal of Inorganic Chemistry, Fialkov and Shor: The Conductivity of Iodine Monochloride. So I quickly got a copy of this paper and I didn't need Russian to see a graph with a maximum... and the wretched guy had got there just before me.

So that wasn't good news, and so I naturally, of course, told Emmy about this straightaway, and we decided that the best thing to do was to continue, because from the results in Fialkov's paper, it was quite clear, and I have to say it, ours were much better results: the melting point was better, the conductivity was lower which meant it was purer material, and so forth. So I was very confident that those results were good, but the saviour – bit of good luck – iodine, as you know, has another chloride, iodine trichloride, ICl₃. That's nowhere near as easy to work with. First point, it doesn't melt until 100 degrees, so that's a higher temperature. Second, it doesn't melt at all unless it's under pressure. If you just heat it, it loses chlorine, goes back to iodine monochloride, and evolves chlorine. So, you have to heat it under a sealed condition, and you would normally use something like a Carius tube for that, which is reinforced glass, a special Pyrex, so that would withstand the pressure. Well, I calculated the pressure that it would be under, and I worked out it would be about 15 atmospheres, that's 15 times atmospheric pressure, so I had to make a conductivity cell which would withstand that, and remember that we've got glass-to-metal seals in this.

Well, I did the first experiment with platinum electrodes which are the standard electrodes, and that was the first bit of chemistry I learnt: iodine monochloride dissolves platinum when it's liquid. So Emmy wasn't particularly pleased that a big sheet of platinum had dissolved in Norm's iodine trichloride. However, I found out, we tried molybdenum and that didn't work, tried tungsten. That was okay. But the trouble with tungsten is twofold. Firstly, it doesn't seal readily into glass when it's thick so you have to get a thin wire of it, and if you have a thin wire, you can't weld it on to a flat sheet of tungsten, it's too brittle. So the way around that, that I devised, was to get a rod of tungsten about 5mm thick, and then to turn it down on a lathe so it was only 1mm diameter at one end, so it was just like a thin rod, and a thick rod on top of it, and then to seal that into glass. But of course you had to clean that first, and the way to clean tungsten, Brian, if you don't know, is to treat it with molten sodium nitrite. That takes the oxygen layer off, and then you have to swill it with freshly-distilled water, de-gas it, and then of course dry it under vacuum, and so forth, and then you can seal it into the cell. So I got some results on iodine trichloride at high pressure, which certainly no one had ever done before, tied that in with the iodine monochloride paper, and we published that, and that was the first paper that I got from the Emeléus lab.