A Tale of Two Supervisors

Something scary happened to me the other day. It occurred to me that about now was the fiftieth anniversary of my submitting my PhD thesis, as well as that of my friend, who I shall call John, in part because that is his name. The two of us had started our PhDs at about the same time, and we finished about the same time. But the effects were not the same. PhD projects tend to involve somewhat unexciting work, largely because when you set out you have to nominate the title, and the final thesis work has to reflect that title. Accordingly, supervisors tend to come up with projects that cannot fail to end up somewhere, but somewhere is usually very unexciting. John elected to work on chemical transformations of steroids, which, at the time, were considered very important. A steroid is a hydrocarbon molecule with four rings fused in a specific configuration (See Wikipedia for a diagram) and steroids were becoming important because they involve the sex hormones, and the contraceptive pill was just making its appearance. The different steroids differ in substitution, so a lot of chemistry could be done to see what sort of new steroids could be made. (Much later, this seems to have continued with effort into anabolic steroids for athletes to cheat, but at the time this was more or less unheard of.)

So John worked in a “hot” topic, and he got lucky. Something quite unusual happened – the so-called “Backbone rearrangement” was discovered. This led to a good number of papers being published, and John’s supervisor worked this to the hilt, with good use of the dreaded “letter”, a short paper that effectively lets the same work be reported first in parts in sketches, and later in detail. John’s career was well on the way to a sound scientific career, with lots of publications in a “hot” area.

At the time of thesis submission, I was highly excited because while I was not going to get many publications, in many ways against the odds, I thought I was making a real contribution to science. My project had a rocky start. I had selected my supervisor to be, he had given me a project, and I started a review of what had previously been done. Three and a half weeks later, and a few days before Christmas, I gave him the answer to the project: it had just turned up in the latest edition of Journal of the American Chemical Society. It was a good project, but we had been scooped. (Actually, this was good. It would have been awful if it had turned up, say, two years into the project, particularly as it then turned out my supervisor had no fall-back position.) So, after a couple of days, he gave me the choice of two projects, and we all went off for Christmas, which is in the summer in New Zealand.

I came back reasonably promptly, and the literature review of these projects was anything but encouraging. One involved a nightmare synthesis that required heating stuff with about 2/3 of a kg of a substance that was highly explosive, and the explosion could be triggered by scratches on glass. Then, very small amounts of the desired material could be separated, in principle, from a little under 4 kg of tar. And that only gave the starting material for some rather long and horrible syntheses to follow, which might or might not have worked. The second project was simpler: to measure the rates of chemical reaction of a class of compounds that were at least tolerably easy to make. That looked good until I found a reference that the rate was zero. The reaction simply did not go. My supervisor was still on holiday, so when the Head of Department saw me in a somewhat depressed state, he suggested I find a project. That was probably to get rid of me, but I took up the challenge. There was a debate underway in the chemical literature, and I elected to join in. So, that started my PhD. The project was to decide whether the electrons in a cyclopropane ring could delocalize, and I intended to use to be the first (I wasn’t, as it turned out) to use the Hammett equation to actually solve a problem.

I soon found out the compounds were a lot harder to make than I expected, and I also found out my supervisor was not going to be much help with advice. I had another problem. One of the syntheses was outlined in one of those dreaded “letters”. My problem was, the method did not work on my compounds. Since the original publication was on quite different types of molecules, I thought that was just life. Several years later, the full paper came out and the reason was obvious: in the letter a very key condition was omitted. I did get another quick method to work, but to be useful I needed to order a number of compounds from the US. Eventually, they turned up – three weeks after my thesis was submitted, and apparently after having gone via Hong Kong, Singapore, and spent time in warehouses in each place! So I was back to my rather tortuous fall-back synthesis route.

Shortly after the end of year 1, my supervisor took a year sabbatical in the US, so I had to do all my thinking myself. Meanwhile, the debate was settling down, and most had come to the conclusion that there was delocalization, and more particularly, some big names were coming down on that side. Somewhere in this period two publications came out relating to using this Hammett equation on the carboxylic acids I was making (Scooped!). One showed there was no delocalization; the other claimed to show there was. In my view, neither was likely to be correct. In the first there was a huge scatter in the data fit and I knew the acids were barely soluble in the solvent used. In my opinion, the second contained a logic error (see explaining post if interested). Since I intended to work on amines, which were more sensitive to this equation, I was still in business.

That was until I tried to measure my equilibrium constants. Everything went like a charm until I had to use the trans 2-(p-nitrophenyl)cyclopropylamine. I introduced this as the hydrochloride (because hydrochlorides were easier to purify) but when I introduced the buffer to generate the equilibrium, the solution gradually darkened. I got a fast result before the darkening became significant, technically the result was consistent with no delocalization, but I did not trust the result. Unfortunately, this was the key compound. I was out of business.

It was now that my supervisor made a brief appearance before heading back to North America to try to get a better-paid position. He made the suggestion of reacting the acids with a diazo compound. His stated reason for the suggestion: I needed to pad out the thesis. Nevertheless, this turned out to be a great suggestion because the required low dielectric solvent had the effect of amplifying changes between compounds. And there it was: I had an unambiguous result lying well outside any experimental uncertainty. The answer was no, the cyclopropane ring did not delocalize. Pity that by this time everyone else was starting to believe it did because the big names said it did. I saw my supervisor once more, briefly, then he was off for good. No advice on writing the thesis from him.

Needless to say, I was excited because I had settled a dispute (or so I thought), and more to the point, I had worked out what was really going on. The writing of that in the thesis was a bit obscure, first because I wanted the work to stand on the results, and secondly, I had decided that if my supervisor wanted to get any benefit from my theoretical work, he would have to make some progress himself, or at least talk to me. One of the jobs of a supervisor is to write up scientific papers based on the thesis results, after all the supervisor puts his name on those papers so he should do something. (A lot of supervisors, like John’s, do a lot more; this one did not.) After some considerable delay he wrote up one paper, relating to the work on the amines. This was not really conclusive, but I had made some compounds for the first time. But what really annoyed me was that he never wrote up the one part to which he had contributed, and which had unambiguously proven the absence of such delocalization. Why not? As far as I could tell, either he did not understand it, or he was too afraid to stick his head above the parapet and challenge the big names.

Needless to say, there is more to this story, but that can wait for another day. Meanwhile, in the post below, I shall try to explain the chemistry for those who are interested.


2 thoughts on “A Tale of Two Supervisors

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s