Dear Orest, 


(1) I agree you should be looking for a different type of septum. What you
describe is unacceptable. Unfortunately, I can't suggest an alternative
right away.

Apart from finding septum material where it doesn't belong, a septum
degrading in this way means one will soon have a leak in the injector, which
even with EPC will cause problems with chromatography.

(2) What intrigues me even more is your choice of "guard" column. First of
all, there is no such thing as a guard column in GC. The term is borrowed
from HPLC where a pre-column's main purpose is indeed to protect the main
column.

The real function of a pre-column in GC is apparent from its proper name,
"retention gap". What a GC pre-column is supposed to achieve is a refocusing
of the injected sample from a widely dispersed vapour cloud into a narrow
band by exploiting the solvent effect.

For this to work in an optimal fashion the pre-column should be about 1 m
long (check), match the polarity of the actual column (?) and match the ID
of the actual column (does not check). The latter even more so if one uses
EPC to maintain constant column flow. EPC relies on the relation between
(fixed and uniform) column dimensions, temperature, known viscosity of the
carrier gas and column head pressure (of the carrier gas) to maintain
constant flow by increasing column head pressure in line with the
temperature programme (and the known fixed column dimensions). This becomes
really critical the higher the temperature gets (such as for late eluting
high FW hydrocarbons).

I am not saying this mis-match between pre-column and actual column is the
(only) cause of the problem you describe but it certainly does not help.

My first suggestion would be to connect your column directly to the injector
(with a fresh septum) to see if under these conditions the problem can be
reproduced or disappears.

If it disappears but you would still like to use a retention gap, use one
with an ID of 0.25mm. Search the Varian website for Chrompack retention
gaps. Chrompack used to be the only company selling retention gaps deactived
with a nano-film of stationary phase offering a choice of apolar (DB-1 to
DB-5 equivalent) and medium polar (DB-1701 equivalent) deactived
pre-columns. Nano-film deactivated pre-columns have a longer life span then
TMS deactived ones since they less prone to develop active sites after a bit
of (ab)use.

If the problem persists, don't call me, call Thermo ;-).


Best regards,

Wolfram




> -----Original Message-----
> From: Stable Isotope Geochemistry 
> [mailto:[log in to unmask]] On Behalf Of Orest E. Kawka
> Sent: 21 March 2007 02:10
> To: [log in to unmask]
> Subject: Baseline change (tail?) problem: high MW alkanes, 
> on-column inj
> 
> We have recently switched to an on-column injections on our 
> Thermo Trace GC Ultra - Delta V Plus measuring hydrogen 
> isotopes. We have encountered many problems (including septum 
> material ending up in the 0.53 mmID pre-column) but have 
> reached a point where the biggest problem now appears to be a 
> baseline change or peak tail after the compounds of interest 
> (> n-C34 n-alkanes). It really appears more like the baseline 
> stays slightly elevated for a while after a peak comes out 
> and then drops relatively quickly. It is not a long sloping 
> tail. We thought of some type of degradation/decomposition, 
> but can't determine where/what it could be. We would 
> appreciate any input/suggestions. Here are our conditions:
> 
> GC CONDITIONS:
> Oven temperature: 100C and 120C, programmed up to 325C Cold 
> inlet Guard pre-column deactivated 0.53 mm ID (0.75-1.5 m in 
> length) Analytical column HP-5MS 30m x 0.25 mm ID (0.25um 
> film) Using a press-fit connector to connect above Flow rate 
> 1.0 - 1.3 mL/min Syringe 80 mm needle cone tip, (currently 
> using airgap above sample to ensure full emplacement on 
> column) Injection Volume 1.0 uL  Conc (200 - 300 ng/uL)
> 
> H/D analysis - relatively new reactor
> Ar background with GC (backflush off; open-split on): 120-150 mV
> H2 Background with same conditions: 400-800mV (oven temp ~110C - 325C)
> 
> By the way, any suggestions on how to keep the septum 
> material (Thermo on-column translucent septa with teflon 
> face) from getting into the guard column would also be quite 
> helpful. We found pieces of septum material in the guard 
> column and in the press-fit connector.
> 
> Thanks,
> Orest
>