H2S in a natural gas sample

[NorthernGC]

Any help would be greatly appreciated...
I have a sample of natural gas with a known quantity of 10ppm H2S and 7% propane. We have been analyzing using a PFPD and are only measuring <1ppm H2S. I thought it may be propane quenching so have spent most of the week tuning: split flow ratios, temperature ramping, and changing the combuster from 3mm to 2mm. Nothing seems to help and I have complete confidence that 10ppm H2S is present.
Does anyone have a technique I could try or advice? Appreciate your help.

[chromatographer1]

Have you tried to improve your separation of H2S and propane?

Rod

[NorthernGC]

Thanks, Rod. I was advised that temp ramping would create the needed separation and I have observed some progress. I'm still not seeing the peak. I'm using He as a carrier gas. Is Hydrogen better?

[GasMan]

Somet times detectors tell you the truth. Are you 100% sure that your sample now contains 10 ppm H2S? We have just gone through a similar experience with a natural gas sample containing low levels of sulfur compunds, to find that some compounds are missing entirely and new ones have appeared.

I assume that you are using an inert sample path in your GC. If you are able to inject liquid samples into your GC, I suggest that you check out your system with a known good sample such as the check out sample that your detector manufacturer supplies.

Gasman

[chromatographer1]

Hydrogen is not superior to helium in this case.

Can you tell us ANYTHING about your sample system, your injection, or your column, your source of the gas standard, its storage container, your GC parameters, other than the minimum you have already told us.

Either the H2S is there or it is not. If not then one direction to proceed is indicated. If it is, then another direction is mandated.

It is difficult to be helpful when the question leaves so much to the imagination. Facts are much superior to supposition and cuts to the chase a lot faster, if that is important.

Rod

[NorthernGC]

Thanks Gentlemen. I hope th following details help:
We have a Varian 3800 with FID, TCD and PFPD. The column is a new Agilent GS-GasPro, 30m, 0.32 mm ID. We are using a split injector and have tried ratios from 5:1 to 50:1 with 15:1 showing the best response. Important to note that a portion of this sample is also going to the TCD which is why the lower split ratio seems to be giving the best response. We are using an EFC with Helium carrier gas at 2 mls/min. Standard is 25 ppm in air contained in an aluminum bottle.
Gate is set at 6; Air 1 at 17, H2 at 10, Air 2 at 13. The 3 mm ID combustor is installed.
Source gas is from a Sulfinert treated 500cc cylinder sampled from a producing gas field. Sample is days old. Gas detector tube readings of sample are consistently 10ppm.
Looking forward to your response. Thanks.

[chromatographer1]

Wow, NEAT ! Information at last.

Thanks.

Now, back to business.

You see 10 ppm in gas sample when using a tube detection. Perhaps the tube is seeing something other than H2S, a thiol or a mercaptain? Just a thought. Anything else showing up on the 'gram ?

The other thing to try is to use another column. like a GS-Q which profoundly separates propane from H2S, especially at temperatures lower than 100°C.

There are many packed columns which will do the separation. If the Gaspro does not have enough separation at the higher concentrations of propane then there is little you can do to improve it other than temperature or adding another column to the Gaspro or substituting for it.

Good luck,

Rod

[NorthernGC]

Thanks again, Rod.
We did not see any other sulfur species in the gram, however, we diid see good peaks of same in a calibration gas. So I am not convinced there is thio or mercap contmination of gas detector tube.
Thanks for the suggestion on the GS-Q. Sample gram shows very early H2S peaks which may be the trick. I need to think about feeling the pain of another $800 column after just buying the GS GasPro!
It is great to be able to get advice from someone who knows their stuff.

[chromatographer1]

H2S elutes betwen ethane and propane on porous polymer capillary or packed columns. A micropacked column would work in a capillary GC as you could get flows at 3-5 cc/min to work. I know Hayesep and Porapak both have packings that will work after some preparation work.

Supelco has a special porous polymer packing that is used in the paper industry for H2S. Of course you need to use fused silica lined tubing and thin deactivated screens for terminations.

But capillary columns will also work but packed is SO much cheaper.

Imagine what a trip to the grocery store would cost you if you had to charter a 767 to get there, and where would you find a place to park?

Good luck with your analysis.

Rod

[alchemist]

I would suggest that if you have conditioned your GC at certain conditions, you have first of all to check the resolution of the seperation. And to do that is by using reference standard (External standard), then you'd be able to check whether your GC is detecting H2S or not and what is the RT for the H2S. Then you have to run your sample across the standard readings.

I hope it could trigger your approach at least

Regards,

[Dank]

My company does H2S analysis all the time. We use packed columns ~ 18'' of 30% DC 200/500. I actually make them and might have some floating around for sale. PM if you would like to know more.

[Dank]

My company does H2S analysis all the time. We use packed columns ~ 18'' of 30% DC 200/500. I actually make them and might have some floating around for sale. PM if you would like to know more.

Gave you precolumn specs the main column is 27 1/2' of DC 30% 200/500

[chromatographer1]

Dank,

I am sure NorthernGC appreciates your offer but I doubt your columns are up to the task of a ppm amount of H2S in a gas sample. Your packings are based on diatomaceous earth and while acceptable for higher levels of H2S, they are not reproducible for low levels, which is why other packings were developed.

NorthernGC would be better served with a thick film PDMS phase on a .32mm ID. capillary. But even better for the separation of propane which inhibits the detection of H2S would be a Carbopack column or a porous polymer specially treated for sulfur analysis.

NorthernGC call the vendors and get suggestions from them for high propane containing samples analysis for H2S. It is a problem addressed long ago by the process control and pipeline custody transfer industries in which I worked as an analytical chemist in the 1980s and in the late 1990s.

best wishes,

Rod

[Dank]

Chromatographer,


Man you are telling me, I am trying to get a Clarus 500 PE equipped with a TCD and a 30% DC 200/500 column to detect 75 ppm. Is this even possible?


Sorry for all the questions which may seem rather obvious just graduated from college a year ago so I am new to the world of chromatography.

[chromatographer1]

1% probably.

2000 ppm maybe, different lots of support differ in performance.

75 ppm no, not reproducibly.

I mean, do you even have bare SS tubing for the columns? ( shudder ! ) I do wish you luck, Dank.

Talk to vendors who know the business. I left Supelco over two years ago and I don't know if there is anyone out there except in the process control analyzer business or the NG pipeline industry who can help you. Nobody wants to pay for my knowledge, they will gladly pry all they can for free but nobody wants to spend a nickel for expertise but are willing to spend thousands on wasted time.

But such is our world.

Rod