Preserving the old ways

During a recent meeting at my workplace, someone mentioned a problem that could be solved by taking an infrared spectrum. ’Do we even have an IR?’ I wondered. ’Sure we do,’ came the reply, ’it’s down there on the first floor . somewhere. I think.’ Someone mentioned the need to blow the dust off of the machine first. 

And so it goes, I should think, at many other industrial research sites. The front-line techniques for analysing samples are LCMS and NMR, and that’s usually where things stop, because that’s usually all that you need. Within twenty minutes, those instruments will tell you more about the structure and purity of your unknown compound than a chemist in 1960 could have worked out in a month. ’Worked’ is exactly the right verb, too, given the techniques available then. Everyone knew where the IR machine was in those days, that’s for sure. 

I’ll resist the temptation to go off into a rant about how easy chemists have it today, with their ready access to new-fangled high-field NMR machines. I’m not quite old enough for that, although I do remember those water-cooled 80 MHz beasts, and having access to a 300 MHz machine only once a week. But I am old enough to have a continuing sense of wonder at the ease and availability of, for example, walk-up LCMS systems. They’re not cheap, but they’re so spectacularly useful that they’ve spread throughout industrial labs. A reliable, practical liquid chromatography front end to a mass spectrometer didn’t quite exist when I was a grad student, so the modern machines will always impress me.  

Goodbye TLC 

Yet one thing they’re doing is slowly killing off a technique that I never would have believed was vulnerable: thin-layer chromatography. Chemists have spent decades running TLC plates to check the progress of their reactions, but if your reaction is amenable to LCMS, that’s clearly a better way to go. Far more information is gathered, and it’s stored in digital form for later use. If you decide a year later to go back and look for another molecular mass, that’s no problem. Just try finding (and trusting) that old TLC plate you ran last July. 

In all these cases, the advanced applications of these techniques are lost as the method slowly contracts to its basic utility. IR is brought out mainly to look for nitriles and azides, with their unmistakable sharp peaks, and TLC gets used to figure out the right ethyl acetate/hexane combination for the automated flash columns. Meanwhile, hardly anyone realises that silyl protecting groups have unique IR signatures, for example, or remembers exactly the order in which carbonyl stretching frequencies change with substitution. And when was the last time you saw anyone modify a TLC plate with silver nitrate to separate alkene isomers, or run a two-dimensional plate with different eluents? Your answer may vary depending on the funding of your research environment, true, but these trends are working their way through as the equipment becomes cheaper. 

Old technique, new tricks 

There’s still real use in all these technologies, so the question is how to fit them into a modern lab. That has to be done by playing to their strengths. IR, for example, can be miniaturised much more easily than NMR or LCMS could ever be. A new use for it has been the advent of small spectroscopic probes that can be immersed in ongoing reactions. A trace of selected peak changes with time can be very informative. I can imagine such probes being used routinely if they become inexpensive enough, collecting data in real time and alerting chemists when a reaction has run to completion. And as for TLC, combining it in some way with mass spec might be a way to increase its utility. The number of techniques available for MS sample collection from solid surfaces might point the way to an automated TLC-MS system, which as far as I know doesn’t yet exist. If you bounced an IR beam off it at the same time, you might really have something. 

Derek Lowe is a medicinal chemist working on preclinical drug discovery in the US