Chromatography Forum

I'm running EP fish oils on a Varian 3800 with a 1079 injector, and getting inconsistent area counts.

Here are my conditions:
Varian 3800
1 ul injection
200 fold split
250 degrees injector
270 degrees detector
Range 12
Attenuation 1

2.0 ml/min flow rate, with constant flow mode enabled
Wax column 30 m x 0.25 mm x 0.25 um film
170 degrees C hold for 2 minutes, to 240 degrees at 2.9 degrees/min. Hold for 4 minutes.
2 mm internal diameter insert, no glass wool.
injection speed 5 seems to work, but not consistently.

According to the restek backflash calculator (http://www.restek.com/info_calcs.asp)
at 1.0 ul injection with isooctane and a 54 mm insert, I'm at the upper limit of the capacity of the insert.
I then lowered the injection volume to 0.5 ul, but that was too low. However, when I did a 3 rep at 0.8 ul, the area counts were all over. For one peak, of the same standard under the same conditions, I got:

14743
25122.9
57813.3

I currently have the injection mode set at User, and just reset the septum purge flow to 5 ml/min.

I thought I had the issue resolved by shooting at 0.8 ul, within the capacity of the liner, but that did not work.

Any and all suggestions are quite welcome at this point!

RMHPLC Guy
Have you tried using deactivated quartz wool in the liner?
WK

Tried glass wool, with inconsistent results.

Is quartz wool better than glass wool?

With a split ratio of 200:1 the vapour capacity of the inlet liner is not an issue.

Octane has a higher boiling point than most GC injection solvents - can you use hexane instead ?

With the very rapid flow through the inlet, and a relatively high boiling solvent I suspect that you are not getting complete evaporation of the sample. You could dilute your samples and decrease the split ratio, put a small plug of glass wool in the liner to aid evaporation, use a fritted liner for the same purpose, increase the inlet temperature, and/or use pre- and post injection dwells (hot needle times). Check also that the column is inserted to the correct height in the inlet.

Peter

Hi
I'm not sure. So long as its deactivated nicely.
Are you using a FID? Are the gas flows constant and set correctly?
Have you tried manual injections or with a syringe where you can see the volume of liquid withdrawn.
WK

Method calls specifically for isooctane, so hexane is not really an option.

How important is the wool in the insert for injection to injection consistentcy?

Most split runs I have done I have used some wool, so this is a bit more experimental for me at this point.

Hi RMHPLCGUY,
I always use a tightly packed liner - not loosely packed as some recommend.
I would try packing in different ways to experiment.
But you may get breakdown of some components.
Perkin Elmer recommend tight packing for PSS and CAP injectors.
I have also a 3800 and usually pack the 3.4mm id liner reasonably tightly.
I'm hooked up to a Saturn 2000 GC-MS.
WK

WK,

Do you use the 3.4 mm id liner for mega bore columns, or for the smaller diameter columns (0.25 mm) like I'm using?

Thanks in advance?

Hi
Sample vapour should not be an issue if liner is not too small, 2ul at 270°C and about 8psi would give about 400ul for isooctane.

AS for glass wool. Yes sometimes it does matter. Silianized works fine generally but sometimes phosfor acid treated wool can be better for precision for acidic compounds like FFA, barbiturates etcetera. Alternitively a very clean glass wool like Supelcos "pesticide grade" can be needed.

On one application we got very poor RSD (about 8%) for methylesters of:
Caproic acid => Oleic acid

exchanging the glass wool from silianized to "pesticide grade" wool reduced the RSD to <1,0%

In the context of split injections the function of glass wool is to spread the liquid sample over a solid surface rather than having it as gas-born droplets. This provides better heat transfer to the liquid by conduction, and ensures that all the sample is vapor before it gets to the split point at the top of the column. The drawback of glass wool is that it presents extra surface area with active sites, so you need to use as little as you can. Konrad Grob has published extensively on the vagaries of vaporizing injectors.

If you cannot change to hexane, which of the method parameters can you change ?

Peter

Hi
I use the 3.4mm id for a 0.25mm id column.
I guess packing the liner might also reduce the sample amount you're able to inject.
WK