Why not IC before ICP?

delta_G · Dec 20, 2016

I've been checking out this article about the Triton methods.

http://packedhead.net/2015/triton-lab-icp-oes-testing-of-a-certified-artificial-saltwater-standard/

I've seen a few more of these companies starting to try to pop up. As cheap as Ion Chromatography is, why don't they put an IC in front of their ICP instruments? Is it a time / throughput thing? Is it cost for consumables?

Randy Holmes-Farley · Dec 21, 2016

In front? You mean a coupled machine? What would be the benefit?

The issue is that if you use IC (or anything else) to separate some of the ion before the ICP, then you need to run many ICP samples to get the values rather than one sample.

delta_G · Dec 21, 2016

The benefit would be that you get some separation between analytes instead of trying to read them all at once with all the potential interference issues that brings. You don't need total separation but it seems like some would help.

You wouldn't need to run many many samples, just one for anions and one for cations with the eluent stream from the ICgoing straight to the ICP or perhaps some suppression of the mobile phase and then direct to the ICP. If you get the right instrument with two pumps and plan the chromatography out right you could even run the anions and cations simultaneously on two columns and recombine into the ICP.

It just seems like a little bit of separation would alleviate some of the interference issues. Maybe I'm just overthinking it.

Randy Holmes-Farley · Dec 21, 2016

But you need to run many ICP scans of the ongoing effluent from the IC. That's what I meant.

You'd also have some issues integrating the values over time in the effluent if you did not have almost continuous ICP scans of the effluent.

delta_G · Dec 22, 2016

OK. I really don't know enough about the ICPs. That was how I thought they worked. I haven't worked with any. I was under the impression that they constantly scanned. So my thought was that once you know the elution times you could focus on only the appropriate wavelengths at the appropriate times. So a run is now much longer for sure, but it seems to me like it would also be much more accurate.

Randy Holmes-Farley · Dec 22, 2016

delta_G said:
OK. I really don't know enough about the ICPs. That was how I thought they worked. I haven't worked with any. I was under the impression that they constantly scanned. So my thought was that once you know the elution times you could focus on only the appropriate wavelengths at the appropriate times. So a run is now much longer for sure, but it seems to me like it would also be much more accurate.

In a normal ICP run, you send in a burst of fluid all the same concentration, and scan it while the fluid is in the plasma state. Maybe a few repeat scans to get better signal to noise. It is not continuous, but discrete scans.

But after an IC, the concentration of all the ions is continuously changing, so you'd need to be scanning each little bit of fluid that leaves the IC to tell what is in it. You would send in fluid from the IC, scan it, and then send in more fluid, etc. Usually you flush between runs, but in such a situation a flush would be needed each time. But lots of scans woulds.

So it would take many more scans, and hence more time. Maybe 100 times as long.