Chromatography Forum

There is no reason why you can't run the GC at 10:1 or even 5:1 split ratio. To isolate where the problem is, you can connect the carrier gas line directly to the injector port by bypassing HT3. Try run the GC @10:1 or 5:1 split ratio, if it still shuts down, it's probably a leak in the injector port. If it runs fine, it's a HT3 issue.

Hi Schmitty

I'll go with Rod, you sample is sticking in the 4-5 inch bit of unheated metal tube. Getting this up to the temperature of the transfer line is going to be tricky. just insulating it will probably not be enough becuase its only sources of heat are via conduction from the inlet and transfer line. Can you wrap a bit of heater tape around it, or just to confirm that it is the source of the problem, blow on it with ahot air gun just for a couple of test runs.

Peter

The split does work fine with the HT3 bypassed. I will look for a leak when I get some time, and will ask around for some heat tape or some better insulation.

Thanks all!

Can you tell us exactly when the GC starts to beep? My guess is it happens during the desorb of HT3. If that's the case, the leak is in the carrier gas flow path during the desorb.

Schmitty,

I would guess your problem is water from the headspace. You are probably not scanning that low (a wise thing) and you are probably getting a large amount of water coming over. The split scheme (have used, works great) is a good starting point but the early eluters are probably sitting in a sea of water at the head of the column. Evidence of this is the peak shape when you did a solvent injection. Try a long dry purge time or perhaps a J trap. Best of luck.

I am awaiting a small rheostat and heat tape combination from Fisher right now. I can try a longer dry purge, thanks for the tip.

A type J trap seems to be significantly different than the K. I am looking at the Tekmar conditions now, and I see for their Vocarb 3000 trap (same as a "K"), a time of 4-8 minutes on the dry purge is recommended. Currently I am at 2 minutes.

AICMM has a good point about the water distorting alcohol peaks by a solvent effect at the head of the column, but this alone would probably not put a step onto the ether peak.

Peter

I tried a high level sample overnight, with better results using a dry purge time of 6 minutes.

Next test will be to use the method modification software to adjust the dry purge time in 1 minute increments to see if a real change is taking place in a single run.

Thanks all.

Well, after months of accepting mediocre peakshape, I have made another commitment to improving the chromatography. Below are two chromats. One bad, and one much better (but still not good). I have been messing with some of the HT3 purge settings, like desorb time, sweep time, and dry purge time, all with little success. The recent peakshape degradation seems to be associated with going to an 80°C sample temp from a 60°C sample temperature.

Also of note: All of the other compounds (hexanes, benzene, xylenes, butyl ether, methylene chloride and 10 others) all have great peak shapes and reproducibility (for the most part).

Any further suggestions or help would be greatly appreciated.

Hi Schmitty

Increasing the sample temp will be putting more water vapour onto the column, so AICMM's explanation is looking strong.

Just for troubleshooting, can you run a dry sample - just put the analytes, with no water, into a HS vial and purge onto the trap. Getting small enough quantities into the vial might be tricky, but it really starts to look as if water is the problem.

Peter

Hi Schmitty

Increasing the sample temp will be putting more water vapour onto the column, so AICMM's explanation is looking strong.

Just for troubleshooting, can you run a dry sample - just put the analytes, with no water, into a HS vial and purge onto the trap. Getting small enough quantities into the vial might be tricky, but it really starts to look as if water is the problem.

Peter

Hi Schmitty

Increasing the sample temp will be putting more water vapour onto the column, so AICMM's explanation is looking strong.

Just for troubleshooting, can you run a dry sample - just put the analytes, with no water, into a HS vial and purge onto the trap. Getting small enough quantities into the vial might be tricky, but it really starts to look as if water is the problem.

Peter

Peter,

I will try that, as my stock is in DMA.

Thanks,

Schmitty

Schmitty,

Several things occured to me in reviewing your post. First, I still think water is a big part of your problem. Raising from 60 to 80 would put a lot more water on the trap so I am not sure why you would want to do that. Increased split should help with that if you have the sensitivity. Second, the fact that you are dynamically purging the headspace means that you are also moving the analytes deeper and deeper into the trap. The non-polars don't go as deep, probably stuck in the first phase of the trap, but the polars are going to go very deep, especially since they are low carbon compounds. A long equilibration time with a short purge and hot desorb should indicate whether this is the case. This would also explain the stepped shape as you are getting two loadings on the head of the column, one from the first trap material and one from the second trap material.

One other thing to think about trying. Throw a chunk of dry ice in the oven and go for the low initial oven temp (18 or 20) and see what happens to the peak shape. May not be a solution but could provide insight.

Best regards.

Schmitty,

Sorry, one more comment. How long do you sweep (dynamic headspace?) Earlier post shows great peak shape for static headspace, sweeping may be the problem (see my earlier post) , solvent injection looked good you said, so why not try long hot equilibrium and static sample or very short sweep time?

Best regards.

The other thing you can do if you do not chose to change parameters in your purge and trap is to use a short piece of thick film Supelcowax capillary column in front of your OV-1301 type column.

This will help focus your alcohol and water plug so you might get better chromatography.

best wishes,

Rod