by
lmh » Thu Oct 24, 2013 2:45 pm
... I'd add that a triple quadrupole is going to be a lot more expensive than a straightforward GC-MS (which is a single-quadrupole instrument relying on source-fragmentation to make its beautiful spectra), and that is already more expensive that a straightforward FID system. For many applications, there is absolutely no need to use the most illustrious instrument available; in fact there may be no benefit in doing so.
There's a real danger that somewhere, someone will use a stupidly over-specified piece of equipment to do a job because they happen to have it. They want to publish the results, so they think up some marginal (or downright spurious) reason why their equipment should confer benefits over the cheaper systems people have used for years. Once it's in print, everyone else is obliged to follow, or look cheap-and-sub-standard; not being "cutting edge" can be bad news for your scientific reputation, even if the cutting edge doesn't contribute anything new...
If you need a triple quad, buy one. If a simple GC-MS will do (and you don't envisage novel applications that need future-proofing*), get a simple GC-MS. And if you don't need MS at all, and an FID is doing fine, consider sticking with it.
(a related issue is people developing LC methods for analytes that work better in GC, and vice versa, because they happen to have one or other instrument/expertise; that's fine, but I hate it when they claim advantages that don't exist! How I wish we could write "We analysed this by LC because the GC was broken at the critical time in the project"!)
(* note: beware of future-proofing that actually gets you an instrument that does a worse job on the tasks that you know you will have to do, in order to make it able to do hypothetical tasks it may never have to do...)
[/mode: general complaint]