Large Volume Injection - Solvent Vent Conditions

[Travisdog2]

Hi,

I've been having a bit of play recently with 'larger' volume injections to try to get my head around all of the functionality of a new instrument. I have been using Agilent's solvent vent calculator to help build some basic vent parameters and these have been 'not unsuccessful' so far but I would be grateful for a bit of advice from folk with a bit more experience.

The solvent vent calculator suggests initial temperatures which are pretty much unobtainable without cryogenic cooling (which we don't have), so I initially used a start temperature of 40C which was OK, but leads to ridiculously long inlet cooling times (~30-40 minutes to get back down to 40C). Whilst I did attempt to address this by cooling the inlet during the run itself, but when the oven is force cooled at the end of a run the inlet heats up again as the pressure of the fans presumably forces hot air into the inlet, so it's a bit of a waste of time!

To address this, I selected a 50C start temperature and this has given more reasonable cycle times and some solid results for mid to late eluting analytes giving ~91-155% of expected peak area. However, my early eluters (BP of ~200C) have proved a bit more troublesome and it's clear that I'm losing analyte along with my solvent. I know that dropping initial inlet temperature is the obvious solution, but wondered if there might be an alternative way of achieving similar outcomes. I can reasonably change vent flow, vent pressure and vent time and wondered what you would suggest?

Kind Regards

TD2

[cjm]

Questions I would have include: liner type? wool/no wool? column type/dimensions? solvent type? detector type?

One parameter that comes to mind if you have no cryo is to incorporate a pressure pulse in the injection to reduce inlet residence time.

[Travisdog2]

Hi CJM,

Fair point, my apologies for the lack of detail here.

• Liner type is Ultra inert 4mm single taper splitless. Not ideal for LVI I know, but one of my analytes is very very sensitive to liner conditions and falls to bits with wool, with older liners or at anything above 185C.

• Column is a typical 30m DB5 0.25/0.25

• Solvent is IPA

• Detector is a single quad MSD

Will give a pressure pulse a proper try. I did try increasing pressure from 5PSI to 7.5PSI during the vent but it didn't seem to do all that much other cause a slight reduction in overall response except for my latest eluter (which had a slightly higher response albeit with a slight hint of fronting), but your suggestion might warrant further experimentation.

Thank you for your input

TD2

[Peter Apps]

What volume are you injecting, over what time period, and do you programme the inlet temperature.

The more you don't tell us the more we can't help you (Rod's Law).

[Travisdog2]

Ah Rod's Law...

As a long time forum lurker, I never thought I'd be guilty of failing to provide the information required, but I guess I am guilty as charged!

In terms of volume it's a comparatively small 'large volume'; just 5µL as I'm experimenting and trying to gain a bit of understanding of what does what. Solvent vent is followed by a ramp to 180C, held for 30s to prevent complete destruction my delicate early eluter, followed by a ballistic ramp to 300C+ to deal with my late eluters which need a bit of oomph - this follows my standard PTV conditions which are themselves more effective than the standard protocol for the sample type in question. Counterintuitively, faster injections rather than the recommended slow injections seem to be more effective.

As it turns out, I think I have found some good conditions which have bought back my early eluters without affecting later elaters and retaining a reasonably realistic starting temperature/cycle time. Over the course of today, I have tried each of the key parameters other than temperature (i.e. vent pressure, vent flow etc) and what seemed to work for me (at least for this comparatively small LVI) was a fast vent flow (500ml/min) for a short period. I haven't done the maths yet but I would guess that we're looking at around >90%+ theoretical peak area for all analytes (when compared to a more conventional 1µL PTV injection) which in my book isn't half bad!

TD2

[Peter Apps]

Presuming that you are doing a fast injection (which you don't tell us) then with an open liner you are hoping that most of the sample gets squirted onto the liner wall and doesn't run down past the column tip before it evaporates. For reliable PTV at any volume you uneed someting to catch the liquid sample as it is injected. You say that you can't use wool, but a fritted liner might work, or as a last resort a Uniliner that seals to the column.