repeatability of GC injections

Discussions about GC and other "gas phase" separation techniques.

11 posts Page 1 of 1
Over Christmas I got dragged into supporting someone with GC (I normally stick firmly to LC). We have a method for derivatised sugars using phenyl-beta-glucoside as internal standard. The method runs on an Agilent 6890N GC with 5973MSD as the detector. We're using a ZB 5HT column, 30m*250u at constant flow of 0.9mL/min, ramping from 80-380 degrees. The injection is 1uL, splitless, 240degrees, 8.09psi, 95.5mL total flow, 91.3mL purge-flow, and gas-saver was "off" according to the text file with the results; in the method I think it said 20mL/min after 2 min. We're using Restek splitless single-taper gooseneck with-wool inlet liners.

The internal standard peak area varies quite a bit on repeated injections from a single vial, in a bad case from peak area 100,000 to 170,000. Is this normal?
Some compounds, in some cases, seem to behave similarly to the internal standard, and give better calibrated results than raw peak areas, but in other cases I get the impression (from too few replicates) that late-eluting compounds behave independently, or perhaps even opposite to the internal standard. The internal standard elutes quite early, and the most variable offender is the last peak to elute, almost at the end of the gradient.

These are the things I'm worrying about:

(1) What is the normal repeatability of multiple injections in GC?
(2) If ours are worse, is there anything obviously wrong in our method?
(3) When repeat injections suggest that the internal standard is behaving differently to the things we're measuring, surely there's no justification in using it? But if anything is varying a lot, and we can't compensate using an internal standard, surely we're utterly sunk: we just have unusable results?

Help! Any comments greatly appreciated!
Deriatized sugars can be a pain on a good day. And depending on the method of derivitization, the samples may not be all that stable to start with. I have seen TMS derivatives change over time from the kenetically favored products to the thermodynamically favored products. So, depending on the derivitization method used some variability is expected.

Aditionally, condition of the GC inlet liner and septum can make a significant difference. These need to be changed on a regular basis - with the septum being changed somewhere under 100 injections per septum for the typical septum and the liner beign changed as needed by the samples - sometimes as few as 20 injections on a liner.

Fouling of the mass spec can result in changes across a sequence of samples. This is particularly noticeable in CI mode.

The results are not as good as might be expected for easier compunds. To me the primary suspects, without further information, are GC inlet maintanace and sample derivitization.
I've used TMS derivatives of many compounds, several GCs, for like 30 years. Usually our GC injections are pretty reproducible, occasionally though we just re-inject the sequence if standard injections before and after the samples are not consistent.

For cGMP work with internal standard, it's the ratio of the internal standard to the sought-for compund that needs to be consistent. So how reproducible during your sequence is the ratio of your components since you use internal standard?
Hi Both,

Thanks for your replies.

Don, how long does it take for equilibration? The samples were derivatised (not by me) with MOX/MSTFA and left several hours before running.
Consumer-Products and Don, yes, absolutely: to get a better idea of true repeatability we are running just the highest-level standard, which has been in a -20 freezer since just before Christmas, repeatedly today. We're hoping that this sample is now so old it must surely be in the thermodynamically-favoured state, and we'll check the ratio of target to internal standard in the runs and report back.
The septum and liner were new before the last sequence. We're using EI in scan mode, but quantifying on extracted ions.

The reason for re-running today is that last time, the replicate runs were separated by 15-or-so other runs on other standards and samples. Today there will be just the one sample run many times.
From your description you have a problem :shock:
I can't comment on the sample type but the variation you are getting from multiple injections from a single vial using an internal standard method are unusable. I would hope to see better than 1%RSD reproducability on replicates although method and other issues could mean 3-5%. You are no where near based on the IS area count.
Do you have any historical "good" data to compare to ? Has anything changed in the method/GC recently ?

Whilst your method may not be the most ideal if it ran OK before it should be OK now.
Do you have the column test solution from the manufacturers to test and see what you get ?
You might be advised to swap liner and septa, check the column flow, leak check the system, run any checks that the GC offers to confirm it is running OK. Change the AS syringe
Then try running either the column test standard or your analytical standard , does it compare to the reference chromatogram. Do you get reproducability ?

Poor reproducability usually points to problems with the injection. Does the sampler take up the sample OK ?
Check the simple things first, more complex issues like blocked spilt lines, split line filters etc. are next
In the work we were doing with sugars and other compounds with TMS derivitization, we never looked for equilibration. My recollection was that there were several chromatographic peaks that could be attributed to each of the sugars we were looking at. We summed up signals for all and used the total area. We were not using MOX and had various anomers present. This was more than 10- 15 years ago and looking at the literature, it appears that this was not being done by anyone when the group I was in came up with the technique we used. I don’t know that we ever looked for our mixture to equilibrate – as we were silylating dried, ground plant material – which has other things present. The technique was passed on to me by others who had determined optimum conditions.

Taking a quick look around – I do note a review on analysis of sugars which does mention some of the chemistry going on – and should lead to other, helpful work: http://144.206.159.178/ft/553/20378/367667.pdf
Looking at the 10 injections from a single vial that we did yesterday (which were after 2 injections of a different sample that my colleague uses as a column-test), we have RSD values of 3.1% to 12.6% for our 8 peaks of interest. The ratio of compound peak area to internal standard peak area is slightly better, RSD 2.3% to 10%, but this really isn't a huge improvement.

CE instruments, I'm afraid we're not as far as having reliable historical data for comparison... Is it possible for injection problems to affect different peaks differently (as implied by the fact that the internal standard hasn't corrected the variability particularly well)?

Drat. I wish I knew anything about GC.
Injection related issues will affect different compunds differently. Look at your runs in order and note if the value for a compund trends across the 10 injections. If so , it is likely that inlet history affects the result. Some compunds decompose in the inlet and are sensitive to residue from past injections. In other cases compunds are sensitive to active sites in the inlet and after a sequence of runs, the residue burries the active sites and the signal intensith for the compund increases during the run sequence. The most common is discrimination agrivated by changing flow through a worn out septum - but you avoided that with inlet maintanance before the sequence.

Also, did you run a blank after the sequence of samples? How did it look - flat baselines? rolling baseline? lots of noise?

As far as your wish - you are learnign already! Welcome to the party!
Thanks again... yes, there is a slight trend towards increasing peak areas in all our compounds. I had put this down to evaporation from the vial after it's been punctured several times, but it is perhaps a bit too big for that...
I can see a need for systematic tests after changing individual bits, which is going to take time. It's got to be done!
Hello,
I have a Biodiesel All in One GC, and I have a problem with repeatability in EN 14103, FAME content. Back ALS in accordance with AC application have 250mkl syringe and injection volume is 2.5 mkl. We have different results for FAME, from the same vial. And, the Area of ISTD is not the same.
We replaced the inlet liner, septum and the situacion is the same, there is no change.
From analysis to analysis we have the smaller values of FAME.
Is the syringe size and injection volume may be a problem(too big syringe)? Why we have a different results?
Thanks in advance.
VesnaJN wrote:
Back ALS in accordance with AC application have 250mkl syringe and injection volume is 2.5 mkl. We have different results for FAME, from the same vial. And, the Area of ISTD is not the same.

Is the syringe size and injection volume may be a problem(too big syringe)?


Did you mis-type in your post?

Personally, I would NEVER use such a large syringe like 250 microliters for an injection volume of 2.5 microliters. For 2.5 microliter injection I would personally use a 5 microliter or 10 microliter syringe.
11 posts Page 1 of 1

Who is online

In total there are 2 users online :: 1 registered, 0 hidden and 1 guest (based on users active over the past 5 minutes)
Most users ever online was 1117 on Mon Jan 31, 2022 2:50 pm

Users browsing this forum: boulderco and 1 guest

Latest Blog Posts from Separation Science

Separation Science offers free learning from the experts covering methods, applications, webinars, eSeminars, videos, tutorials for users of liquid chromatography, gas chromatography, mass spectrometry, sample preparation and related analytical techniques.

Subscribe to our eNewsletter with daily, weekly or monthly updates: Food & Beverage, Environmental, (Bio)Pharmaceutical, Bioclinical, Liquid Chromatography, Gas Chromatography and Mass Spectrometry.

Liquid Chromatography

Gas Chromatography

Mass Spectrometry