5890 set up for both liquid and gas samples?

[NorthernLab]

I have a 5890 II with manual pneumatics. 7673b autosampler. FID and ECD. Currently have Rtx-5 columns on both front and rear split/splitless injectors. Chemstation software, not sure of the version.

I can accomplish what I'm doing now with just the front injector, autosampler, and the FID. I've found that I don't really need the ECD at this time. What I'd like to do is swap out the ECD with a TCD to start analyzing the lighter components (up to c7)of raw high and low pressure natural gas.

I have started to gather a small list of hardware required to add this capability.

6-port sampling valve, sample loop, metering valve, reference standard, ss tubing and fittings.

Each column would operate at very similar operating temps, so there's no problems there.

My initial research indicates that what I'd like to do is possible, but I could use some confirmation since I am unfamiliar with how a gas sampling system is configured. My one and only question is:

Can I have a gas sample loop, valve, etc plumbed to the rear injector while still maintaining use of the front injector port and autosampler for my liquids analysis?

I would be grateful for any other insight any of you could provide.

Thanks in advance.

Dan

[AICMM]

NorthernLab,

Yes, you should be able to plumb a gas sampling valve to the back SSL injection port and do gas analysis off of that (assuming you are using PLOT or capillary columns.) You need to think about how to deal with lights and heavies on one channel and possibly consider a C5+ or C6+ backflush configuration. You also want to consider that liquified gas may be too much sample with a gas sampling valve (typically done with a liquid injection valve.) If you are just interested in hydrocarbons, you might want to stick with another FID and if you are wanting to do other constituents (CO, CO2, etc....) remember you won't be able to do hydrogen very well with the HP TCD at the same time.

Best regards,

AICMM

[NorthernLab]

AICMM,

Thank you so much for your response.

I am planning on a PLOT column. I do need to determine non hydrocarbon content as well as hydrocarbons. The gas I will be testing has virtually undetectable hydrocarbon content past C6 and will always be in gas phase. I hypothesized that a backflush loop could possibly be eliminated by periodically 'baking out' any heavier hydrocarbons that may have built up in the column. Any thoughts about that? Would built up heavier hydrocarbons in the column affect analysis of the lighter ones?

I should say that the clients I will be serving will not be using my analysis to negotiate contracts and are mainly interested in CO2, N2, and BTU values of the gas. My analyses will only be used to verify the buyer's analysis and help exploration companies determine where their 'hottest' deposits of gas are hiding within an established field.

Thanks

Dan

[AICMM]

Dan,

You should be okay with baking out, but it would be better to backflush. You would be surprised at how much C6+ is in samples that don't appear to have C6+.... By the way, these are the high value components. The thing to watch for is peaks in the middle of your run that are much wider than the neighbors. It may be your bad luck that these sometimes land underneath peaks you are interested in. Since you are looking to buy a valve anyway, why not buy a 10 port and implement loop sampling with backflush?

Regarding TCD, since H2 is not on your list, you should be fine.

Best regards,

AICMM

[chemviren]

h