I appreciate comments made to my question, but the main problem here I
think is simply to little sample with "only" 5-10% injected on the
column in splitless mode.

I start with about 10 ul rat plasma, extract lipids, separate lipid
classes on TLC and make FAMEs of NEFAs and TGs. The final extract volume
is 50 ul. I inject 2.5 ul splitless and end up with a bit to small peaks
at present. 

5% of original samples is thus used which equals to about 5 ng carbon.
In the near future it will get worse. We are planning to look at VLDL,
LDL etc from the same 10 ul rat plasma...

My idea was to perform large volume sampling  (LVS) with a programmable
temperature vaporizer (PTV). It allows slow injections into a cold liner
where the solvent is continously evaporated and vented. When 50 ul has
been injected, temperature is increased rapidly and the analyte focus on
the column. With this method up to 50-100% of the sample may be injected
instead of the typical 1-5%.

Can't do this perhaps with very dirty extracts, but in my case I see no
noice at all around my peaks, just small peaks of analytes, so it should
work.

Further suggestions appreciated
Lars


> ----------
> From:         Eric Lichtfouse[SMTP:[log in to unmask]]
> Sent:         augusti 6, 1998 14:13
> To:   [log in to unmask]
> Subject:      Re: Improved sensitivity of (GCC) IRMS?
> 
> >Dear all,
> >
> >In adittion to Tullis question, can anyone suggest methods to improve
> >sensitivity (reduced sample amount demands) on my GCCIRMS compared to
> >the "standard performance"? I have a Delta plus connected to a GCC.
> >
> >               Lars Löfgren
> Dear Lars,
> first you may try to concentrate your sample in a V glass vial, then
> dilute
> it in a minimum of solvent (30ul). Second, purification of the sample
> e.g.
> by TLC could reduce the noise thus improving signal/noise ratios.
> Third,
> you may try to increase the GC resolution (lower temperature rate,
> longer
> column, optimizing He flow, thinner phase thickness, ultrastable 100%
> methylsiloxane phase, true on column injector...) again to reduce
> noise
> level.
> 
> Good luck,
> Eric LICHTFOUSE
> Laboratoire Sols et Environnement
> INRA-ENSAIA/INPL, BP 172
> 2, avenue de la foret de Haye
> F-54505 Vandoeuvre-les-Nancy Cedex, France
> Phone (33) 3 83 59 58 99 (-58 45, -58 65)
> Fax (33) 3 83 59 57 91
> [log in to unmask]
>