Chromatography Forum

Hi,

Anyone here experience in H2S analysis in gas sample?
I am told that prior to injection of H2S calibration gas to GC, the gas cylinder need to be heated up to 50C, this is to prevent the H2S from sticking to the cylinder wall.

How true is it?

Thanks
Punt

Hi,

Any expert here can help me with H2S analysis in gas sample?
when i try inject H2S calibration gas into my GC, the area count goes down everyday. Can warming the gas cylinder helps to prevent H2S from sticking on the cylinder wall?

Thank you

punt,

I have no experience warming cylinders, sorry. I will say that many metal surfaces need deactivating so one thing you might try is multiple injections with the same cylinder to deactivate the transfer lines, valve, etc.... Also, absolutely no brass anywhere.

Best regards.

Hydrogen sulphide sticks very tightly to metals. I doubt that warming to 50C would make a useful difference. You need deactivated surfaces, Restek have a specific sulphur treatment, although I've never needed to try it.

Peter

Heating the cylinder will probably not yield reproducible results. The absorption in the cylinder is probably not reversible for quantitative results. The Restek silco treatments have always worked exceptionaly well for sulfur analyses. Peter has the answer to your problem. You should also make sure any sample "wetted" surfaces are silco treated. That is especially important for ultra trace determinations.

We get our H2S standards in aluminum cylinders. The vendor says the cylinders were passivated before the standard is made. I believe our 1 mol% standard has had a fairly constant area count by TCD for over a year.

I have had samples in carbon steel cylinders where the sulfur compounds nearly completely disappeared within a few days. The sulfur compounds were in the range of 10 to 50 ppm and there may have also been some moisture in the samples.

Dropping area counts could also be due to other things such as the detector or leaks.

Hi,

Thanks for the feedback, currently I am using Antek SCD for my GC detector. I noticed there is a trend in the area count, from injection to injection within the same hour, the area is much more consistent.
But for injection on different days, the peak area drop in a trend
Day 1 - area count-10,000,
Day 2 area count-9000,
on day 3 it drop to 8000 and so on
Does anyone has experience in this SCD detector or have experience such a problem before.

Thank you

From 10,000 to 9,000 (10 %) in just one day? that is too much, H2S
usually loss concentration because it get combined with cylinders walls in long terms periods, but no so fast.

Also the connection from the gas cylinder to your inlet, your gas sampling valve and connections from the GSV to the column, and the column itself should all be treated for sulfur analysis (sulfinert / silcosteel). Regarding the detector you could always contact the people from Antek or your local distributor for Antek see www.paclp.com

I used almost 10 GC over 5 years with capillarity columns a and micro TCD detectors not treated for H2S and never have any problem like this, by the way I remember a old gow-mac with packed column an long volume that works ok for many years as well, the concentration of H2S is 5ppm aprox, run is made every 2 minute, 10% loss signal per day should be something else that h2s poison.