by
lmh » Wed Oct 23, 2019 9:20 am
A simple split won't work anyway. ESI gives a signal related to concentration, not to total amount of stuff reaching the source (that's why nano-flow is so good for proteomic work), so if you merely split 100:1 you'll still have the same issue that the signal at the MS will be enormous if you couple it with a much less sensitive detector.
If you want to make this work, you need to split the flow 100:1 and then add a make-up flow on the MS side of the split, to dilute the tiny flow of concentrated peak and make it a rapid flow of diluted peak. This overcomes the problem of diffusion/peak dispersion in the tubing, and also balances the signals.
This is the typical approach used in mass-directed fractionation, which has a similar problem (desire to inject a tremendously large amount of stuff, preferably to recover most of it in the fraction, but not overload the MS detector).
But this is a world of pain. You need an extra pump. You now have two t-pieces, one splitting off the PDA, and the other adding in the make-up flow. If you get the balance of resistances wrong, the pressure at the second tee will slightly exceed that at the first, and the make-up flow will go the wrong way, not only supplying the entire flow to the MS (which will detect nothing) but putting a little excess out the other way to the PDA. Your only saving grace is that you can at least measure the flow to the PDA, but working out whether it is 594uL/min (good) or 610uL/min (disastrous) is not going to be easy. People who make this work, tend to set it up once, and never touch the plumbing! you can buy, off the shelf, splitters that do the job, with an input for the make-up pump, but remember that they are, in the end, just resistances, and everything you add to the outputs of the splitter (tubing to detectors, the detectors themselves) will affect the split, so the world of pain doesn't disappear. Also, the moment anything gets clogged up, you have a horrible diagnosis problem to work out why it's not working any more.
100:1 is quite an extreme split. If you can get away with a less drastic split-ratio, then the world will be decidedly less painful. It may still need a make-up flow, but it will be easier to get it right. Can you sacrifice quality in one of the two detectors in order to get enough quality in the other? For example, if you're doing MRM's or single ion monitoring, can you work with an isotope peak rather than the main peak in order to reduce MS sensitivity? If you must use PDA with a very good MS, it helps to get the very best PDA you can.