Chromatography Forum

Can anybody tell me what are the advantages/disadvantages of using a volatiles inlet for headspace analysis on an Agilent 6890N GC compared with a split/splitless inlet? I'd also be interested to know if anybody is running both headspace and liquid sampler on the same split/splitless inlet. Agilent say that they can plumb the headspace into the S/Sl inlet and still allow for the use of the ALS. Are there any issues with this configuration? Also, we are using Empower CDS, will this cause any issues with Empower?

I'm not quite sure what you mean by a volatiles inlet. But I am using an Agilent 6890N GC with split/splitless inlet for both headspace and liquid injections.
It probably depends on the headspace sampler you have. I have two perkin elmer headspace samplers (HS40XL and Turbomatrix) connected via the split/splitless inlet. To perform liquid injections is just a matter of undoing the transfer line and inlet adapter, and screwing on the septum nut to the inlet. The Agilent ALS then goes over the inlet for liquid injections.

We are also running Empower CDS - not a problem for the GC control, however, the HS is controlled by its own software and/or firmware. This does not cause a problem for empower as the GC is configured as for manual injections (autosampler disabled).

I'm not quite sure what you mean by a volatiles inlet.

A volatiles inlet is an inert low volume inlet recommended by Agilent for headspace analysis.

The problem is that it can only be used with helium or hydrogen as carrier gas. We have several methods that require the use of nitrogen carrier gas and therefore need to use the split/splitless inlet. We are running an Agilent G1888 headspace sampler which we do not have Empower control over. Agilent have told us that they can plumb the headspace transfer line so that it is always connected to the split/splitless inlet and that for liquid injections we would simply have to place the injector tower onto the inlet. Not sure how this would work!

We may want to use the volatiles inlet for future development in which case we would disconnect the transfer line from the front split/splitless inlet and connect to the back volatiles inlet.

I also have a G1888 connected via a "volatiles interface". I was concerned about the very short distance between the tip of the transfer line and the end of the column, since it might compromise mixing of the sample with the carrier gas. Also I picked up pressure fluctuations during injection from the G1888 which did not bode well for reproducible split ratios. I now thread the column up into the end of the transfer line so that it is in a region of high linear gas flow, with no possibility of dead volumes (the "trickle flow" of the interface is supposed to flush out the potential dead volumes in the intrnal channels). I have also disconnected the 6890 EPC, and just connected a needle valve to the split outlet of the interface, and plugged the trickle flow. Carrier gas comes from the G1888, also with mechanical gas control.

This is part of a suite of pretty drastic modifications to the G1888 for specialist applications.

There are two ways of connecting the G1888 transfer line to a split/splitless inlet. You can connect it with a compression fitting to the carrier gas feed line close to the inlet. This has the major disadvantage of exposing the sample to cool, raw stainless steel. There has been a recent thread on exactly this problem. Or you can get a needle that screws onto the end of the transfer line that sticks through the septum of the inlet. This maintains line temperature, and is deactivated, but is ridiculously expensive. To change from HS to liquid injections you just pull the needle out and mount the autoinjector. If you go with the needle you need enough spce between the needle tip and the colum to allow the sample plug to mix with the carrier gas.

What peak area repeatabilities do you get from the G1888 ?

Peter

We have Agilent coming out in the next day or so to further clarify the options available to us. Any modifications and subsequent qualification will be carried out by Agilent. I am keen to hear the opinions/experiences of others with similar configurations so that we can make the best decision for us and not just the easiest option for Agilent!

What peak area repeatabilities do you get from the G1888 ?

This is a new instrument which is in the process of being qualified. From the limited amount of work that I have done so far I have seen peak area reproducilibities between 1.0 and 1.5%.