Acetylene tail

Chromatography Forum: GC Archives: Acetylene tail
Top of pagePrevious messageNext messageBottom of pageLink to this message  By Andy King on Wednesday, May 26, 1999 - 01:14 pm:

We routinely measure 10 to 300 ppm ethylene in a 0.5 ml sample of 10% acetylene in air using an old varian vista 6000 gc. We recently had to purchase a new varian cp3380 gc and our method is not working as well. The problem is that the acetylene peak used to be back to baseline within 1.6 min after injection and now it takes 3.5 min.

The method is pretty much the same, we are using the same column type (a new 2m by 1/8 inch column containing 80/100 mesh poropak N). Column oven is at 80 C, injector and detector are at 120 C. Nitrogen mobile phase reads 36 psi on the column (don't know the actual flow rate). The only difference between old and new gc's is that column pressure was 52 psi on old machine and we are using a new column (which was broken in at 150 C overnight the same as the previous column).

The ethylene peak comes off at relatively the same time on both machines (0.66 min on vista 6000 and 0.63 min on new cp3380). The acetylene peak begins at appoximately the same time (0.83 min old machine and 0.75 min new machine) but tails very badly on the new gc.

We have tried several things without success:
1. Increased column temp to 90 and 100 C.
2. Increased injector temp to 150 C.
3. New acetylene source.
4. changed carrier gas flow rate (but peaks are coming off as quickly as before except ror extreme tail on acetylene). Note: if we try to crank the carrier gas up to get the 52 psi on the column we had before it blows the detector out at just over 40 psi.
5. Decreased injection volume to 0.25 ml.

Top of pagePrevious messageNext messageBottom of pageLink to this message  By MBSetzer on Thursday, May 27, 1999 - 11:38 am:

Hi Andy:
I continue to use a recently restored VA6000, I still have the original sales literature for when the 3400 series came out, it clearly indicates the 6000 as the top of the line.

A reason for higher head pressure on the 3700 & 6000 series is because there is usually many more feet of 1/16 inch stainless tubing (small coils with narrow ID) between the gauge and the head of the column, compared to other GC's.

With the RT's matched to the old GC it is logical to expect about the same flow, but there may be a big difference between old & new columns. If the old column is as old as the GC, it could also have an uncertain history which might have inadvertently prepared it for superior acetylene analysis. Oven temperature too, I would check with a regular thermometer. I would definitely measure the flow rate of each column at the actual operating temperature to see if there is any difference. If the flow rates were not the same, it would tend to indicate a difference between the columns. Could be that the new one has less retention, and you are slowing down the carrier enough to delay purging the highly adsorbed acetylene from the polymer pores.

Also, the other day I visited a lab where I brought them a column of Porapak Q (1976 vintage, previously unused) for some air testing. They were only familiar with capillary and had not used a TCD before. Brought them up to speed fairly easily. It was a VA3400 and the chemist said the 6-port gas injection valve was installed by their service operator (may not have been an actual Varian engineer). The valve was plumbed logically correct, except that the outlet from the valve ended up *injecting* the sample into the carrier inlet line of the ordinary syringe injector. That might be fine if you were going into a capillary with enough split flow to allow rapid purging of the injector body, but without split, the packed column result was asymmetrical tailing peaks.

Easily cured by eliminating the syringe injector from the system, just ran a 1/16 inch line from the valve exit through a hole in the insulation into the oven and connected directly to the column with a reducing union. The valve & line did not need to be heated since they were testing noncondensable gases. You might check this on your new GC.

Also it might be most helpful to exchange columns between the GC's, that old one may just have the magic . . .

Hope this helps,
Mike Setzer
Quality Analytical Services, Inc.

Top of pagePrevious messageNext messageBottom of pageLink to this message  By William A. Jenko on Saturday, September 11, 1999 - 05:13 pm:

Have you tried swapping the columns between the 2 GC's? This could be a problem with either the column packing or the tubing, rather than the GC. We have experienced some problems with trace acetylene measurements that seem to be associated with the tubing. We changed to Restek Silcosteel tubing for the column in one case and this helped. Also, if you are using a very old Porapak column in your older GC, you might be dissappointed with the performance of newer Porapak N. I suggest you try a Hayesep N as a substitute if the column seems to be the source of the tailing. It will do essentially the same separation.

Good Luck,

Bill Jenko
Sr. Chemist
Seimens Applied Automation, Inc.

Top of pagePrevious messageNext messageBottom of pageLink to this message  By jaap de zeeuw on Thursday, September 23, 1999 - 03:31 am:

You also may consider to use 0.53 mm columns There are several highly selective columns for ethylene analysis available. If you want I can sent you a full presentation with many appls on impurities in ethylene analysis.
Jaap de zeeuw

Top of pagePrevious messageNext messageBottom of pageLink to this message  By W SAYERS on Saturday, September 25, 1999 - 08:03 am:

Previous advise to measure flow rate was correct. However, when trying to tranfer methods to new instruments and or columns it is more helpful to match the linear velocity through the column rather than flow rates. By doing so you eliminate
differences due to column dimensions etc.
There is a good chart of optimum linear velocities with different carrier gases in the J&W caTALOG

Posting is currently disabled in this topic. Contact your discussion moderator for more information.