5890 Dual Fid GC low recoveries

[Tert-butyllithium]

We have a 5890 GC with two columns and two FIDs to perform DRO/RRO analysis. The back column has always worked but the front column tends to be about 1/4 the recoveries making our detection limit on it not great. The front column was just replaced with a new restek rt-5 30m column. The detectors have been cleaned out and new ferrels in both the detector and the injector. The jet has been sonicated and cleaned, and reassembled. In the injector, the gold seal has been replaced along with the liner and the septum. There was a leak in the weldment and that was also replaced. The issue is the recoveries are still low. The pressures out of the splitless vent is 60ml/min and the septum purge is 2.9ml/min. We tried a lower total flow, but there was nothing at 30, very little at 40, more at 50 and the most at 60. 70 was too high and we saw a loss. Out of the detector with all of the gases on we see a flow 186ml/min. I am not sure what to do next and I really need a little guidance.

[osp001]

Ever tried matching injector + column #1 with detector #2, and vice-versa, to see if it's pre-detector or post-detector?

[Tert-butyllithium]

I'll have to try that tomorrow. We thought it might be the injector tower and we switched those and saw no change.

[AICMM]

Tert-butyllithium,

osp001's idea is a really good one.

I assume you are running splitless/split? If so, I would venture that your split valve is not working and you are always running split. However, I am disturbed by your flows comment. Should have something on the order of 300 mL/min from a functioning FID.

Best regards,

AICMM

[Peter Apps]

In addition to the previous suggestions. Your results for "recovery" at different total flows do not make any sense unless there is a serious problem in the front inlet. You need to leak check the whole system with a leak seeker before you do anything else.

What do you mean by "recoveries". ? Even if the front inlet-column-detector has problems it will not affect recoveries since standards and samples / spikes will be equally affected. Do you mean that peak areas on the front detector are smaller than those on the back ?

Peter