Chromatography Forum

I'm testing residual solvents and getting a shoulder in the methanol peak.
Method parameters:
agilent 7890
Diluent-Dimethylacetamide
Column-75m x 530um x 3.0um db-624
flow- 13ml/min nitrogen
splitless
column temp prog- 35c for 15min-25c/min to 150c for 11min
inlet-140c
det-220c

Hello

If you're using splitless mode what is your:
-purge flow (try to set it to 60-70ml/min)
-purge time (set it to 0.2-0.8 min)

Regards

Tomasz Kubowicz

purge flow is currently 40ml/ml
purge time is 0.5min

I will try to change them.


Thank you.

I forgot to mention that i'm testing with methanol 5 other solvents that have no shoulder problems (methanol is the first peak in the chromatogram)

Changing split flow/time did not help and the methanol peak still looks bad.
It looks like the peaks this picture:

http://www.chromacademy.com/troubleshoo ... -sys24.jpg

Does anyone have any ideas about solving this problem?

Hello

You have problem with inlet. I'd recommend:
1.Replace liner
2. Cut 5-10 cm column from inlet side
3.Make sure that cutting is correct (use magnifying glass)
4. Make sure that column length from column nut is correct (for Agilent GC S/SL inlet it should be 4-6mm)

PS. I'd increase inlet temperature. I think it should be higher than 140

Regards

Tomasz Kubowicz

Does anyone have any ideas about solving this problem?

I have the idea that if you provided details about how you are introducing the samples - headspace vs liquid injections for instance - you might get more answers.

Why do you have your inlet temperature below the boiling point of the solvent ?

Peter

This is a headspace method:
Incubation temp. is 120 c
injection volume is 800ul
I agree that the inlet temp. is low,
but this is a validated method that i recived from a manufacturer.

This is a headspace method:
Incubation temp. is 120 c
injection volume is 800ul
I agree that the inlet temp. is low,
but this is a validated method that i recived from a manufacturer.

Is it done with a 6-port valve injection, an automated syringe injection, or a balance pressure injection - in other words what hardware are you using ???

If a 6-port valve injection or a balance pressure injection, how is the transfer line from the headspacer connected to the GC inlet ?.

Is all the carrier gas coming from the headspacer or is the GC supplying some carrier gas direct to the inlet ? Do you know the flow rate through the transfer line ?

If a syringe injection, what is the injection speed and what is the syringe temperature ?

The more you tell us the more we can help.

Peter

It is a syringe injection using CTC/PAL system.
sample volume-800ul
incubation temp. 120c
incubation time 15min
syringe temp. 125c
fill speed-100ul/sec
injection speed-500ul/sec

You are probably getting some carrier flow changes as you inject. This distorts the starting bands for all the peaks, but the later eluting ones will be focussed by the stationary phase and subsequently come out as symmetrical peaks, while methanol is already in its way down the column and elutes with a shoulder.

If you can spare the loss of peak area a slit injection would almost certainly solve the shoulder problem. If detection limit is an issue then you could play with injection speed, inlet liner style, or starting a slow column temperature ramp immediately.

Apart from making the peak slightly less pretty the shoulder will have no effect on your results - the peak shape is still plenty good enough for repeatable integration.

Peter

Hi

My logic may well be be wrong but your flow is 13ml/min = 216uL/sec and your injection speed is 500uL /sec.

As Peter says you may well have a pressure pulse so try reducing your injection speed and/or use split injection - you appear to have plenty of sensitivity to spare.

I would rather see symmetrical peaks and know that the chromatography is probably right than put up with distorted peaks.

Regards

Ralph

As an aside, at 13ml/min you are way outside the best efficiency for nitrogen as a carrier. Your flow should be around 4ml/min to be at 1.5x u optimum = approx 25cm/sec. Then again, if it works, it works.

Thank you, I will try to use this information.

Another thing is that when using a 2mm direct liner the shoulder appears towards the end of the peak and when using a 4mm (with a conical end) splitless liner the shoulder appears at the front of the peak, if that means something.

Hi


Depending on the make and style
2mm liners hold approx. 0.245 mL or 245 µL of vapor
4mm liners hold approx. 0.972 mL or 972 µL of vapor

You are injecting 800uL very quickly in splitless mode.

Apart from backflash considerations (see http://www.agilent.com/cs/library/esemi ... bility.pdf) your observation for the 2mm liner would be consistent with a large pulse of sample forced down the column. followed by the liner volume being flushed onto the column. For the 4mm liner a smaller pulse of sample followed by the liner volume being flushed onto the column.

In both these cases I am guessing that the excess pressure from the fast injection volume and inability of the inlet system volume, flow and pressure to cope with it has led to an initial pulse of the sample finding the path of least resistance i.e. in this case down the column before the liner is flushed

Regards

Ralph

Edit

If your observations had been the other way around I would suspect back flash into your inlet line due to too fast an injection volume.

So use a 4mm liner and, as suggested before, try a slower injection speed and/or spllt injection.

Let us know how you get on.