Acetic acid by gas chromatography

[gogo]

Hi,

Is there anyone who has ever analyzed residual of acetic acid by gas chromatography? I’m trying to test a low level of acetic acid in dimethyladamantanamine hydrochloride using GC Agilent with FID detector. The column parameters are AT-624 30m, 0,53 mm, 3,0 um. The problem is with a baseline which goes up after several injections of the sample. The peak of acetic acid is observed only on the first chromatogram, on the next ones the baseline goes up and the peak disappears. The baseline comes back to normal after 3-4 injections of solvent (DMA). Looks like the sample absorbs on the stationary phase and injections of solvent helps to flush it out. Heating the column at high temperatures doesn’t help. What is more interesting, there is no problem with determination of other residual solvents like: ethanol, isopropanol, ethyl acetate, toluene.
Would anybody help please???? Thank you in advance.

Gogo

[chromatographer1]

If is extremely difficult to do an analytical test for low levels of acetic acid when in a matrix solution. The problem lies in chemistry and the equilibration of ionic state of the acid.

Don't try to do the impossible, use IC for acetic acid (there are methods already developed and vendors are glad to help you).

All you will end up doing is spending a lot of time and money demonstrating that your GC results are unreliable and the method non-validated.

Using procedures that swing the equilibration so the method is more reliable will tend to destroy your column and will make your GC much less useful for other analyses (if not useless).

best wishes,

Rodney George
consultant

[gogo]

Thank you for your comment. There is no option to validate the method on IC in my lab. More than that, the GC method was validated by other lab and everything worked out there.
I'll be trying to change the chromatographic conditions.
Anyway thank you for your advice.

[Peter Apps]

Since you are stuck with using GC have a look at these issues:

baseline drift that increases progressively with number of injections is probably carryover of something from the sample, not necessarily the target analyte. What is you temperature programme ?

If injecting solvent really does help to flush out the carryover from the column than your operating conditions must generate liquid solvent plugs in the column - this is poor practise. The only thing that the solvent flushes is the syringe - try increasing the number of wash cycles on the injector.

I presume that you are dissolving the sample in the solvent - rather than extracting the acid from the sample. This will inevitably lead to rapid build up of crud in the inlet. You might need to look at an extractive sample prep that gets the acid into solution and leaves the matrix behind. Do you have access to headspace sampling ?

Acetic acid will stick to any active sites in the system - you need deactivated inlet liners (with no glass wool) and as short a column as possible. Regular maintenance will be critical, budget to replace columns frquently.

Peter

[gogo]

Thank you, Peter.
My temp programme is: 40°C for 10 min, then raise to 130°C at a rate of 6°C/min, hold at 130°C for 5 min, then raise to 240°C at a rate of 35°C/min, hold at 240°C for 11 min. Ret time of acetic acid is around 22 min.
The extraction of acetic acid would be the best solution I guess. However, I analyze the other solvents (ethanol, isopropanol, acetonitrile, ethyl acetate and toluene) in the dimethyladamantanamine hydrochloride sample at once. So the perfect method would apply for all of them.
Yes, I have a headspace sampling available.
According to the method I dissolve the sample in N,N-dimethylacetamide.
As for number of wash cycles there are 10 set in the instr method (from two wash vials A, B). I'm thinking about increasing this number to 15 (this is max) and changing the injector temp from 140°C to around 200°C. I agree there is some contamination caused by the sample somewhere around the injector. Maybe if I get rid of that, the baseline will not go upwards and the peak will be viable on the chromatogram.

Your tips are very helpful.
Thank you again.

[gcguy]

I have found that using THF as the diluent for acetic acid determination gives a much better peak shape than other solvents. Use unstabilised THF as there are less peaks in it.

GCguy

[kalidassa]

Hai
we have done some product for the same in gc. we are using DMSO as solvent DB-Wax higher micron as Column. Tyr it but you need to filter the sample

KALIDASS

[gogo]

I need to check if my sample is easy dissolved in THF or DMSO.

Many thanks

[chromatographer1]

Let me add a comment that a validated method that works some of the time isn't really a validated method. It means that the validation was carried out without taking account of all the possible situations where the method WON'T work.

The variable that changed is probably water content. Using aprotonic solvents like THF and DMSO as previously suggested by others is helpful. Adding an additive strong acid to swing the equilibration of disassociation of the acetic acid is also helpful. But what a mess we have when we try to do acetic acid by GC and do we really ever know the results are accurate?

History keeps repeating itself with stories I hear from chemists trying to do an analysis the hard way rather than the easy way. I hope it all works out and the results are worthy of your time and sweat.

I started doing acetic acid determinations in 1980 and I believe now and forever that Ion chromatography is the easiest and best way to do it, unless you do it by NMR which for some matrices works great.

good luck and best wishes,

Rodney George
consultant

[Peter Apps]

If you can get around the problems of adsorptive activity, headspace provides a straightforward way of separating volatile analytes from involatile matrices. Does your sample dissolvent in water ?, if it does acidify with phosphoric acid to pH less than 2 and give it a shot. It's not going to be easy - in the end Rodney is right, trace levels of acetic acid are best done by other methods.

Peter

[mahdiebrahimi]

For short chain fatty acid (Acetic, propionic, butyric and valeric) I am using Quadrex 007 Series bonded phase fused silica capillary column (15m, 0.32mm ID, 0.25 µm film thickness). I adjust injector/detector temperature to 220/230 0C respectively. The column temperature is set at the range of 70°C - 150°C with temperature programming at the rate of 7°C/ min. My solvent is water and i added 25% metaphosphoric acid to solvent for fixation purpose. I am getting very good result until now. you can try this column and this method.

[Peter Apps]

Interesting method - how many injections can you do between inlet liner replacements ?

Peter

[gogo]

Just want to update you with my results.
I decreased the injector temperature from 140°C to 200°C (is re-valiation necessary?). Additionally, I increased the amount of glass wool in a liner to minimum. This was causing cumulation of the sample and baseline noise on the chromatograms. The chromatograms look ok now. However, the injection precision is not that great.
Now another problem appeared. As Rodney said, the results seem to be no accurate and the recovery fails at concentration around 0,05% - 0,3%.

[Peter Apps]

I doubt that going to lower levels is realistic - why not ask the lab that originally "validated" the method for advice ?

Peter

[sdegrace]

We have had very good success using reverse phase HPLC for acetic acid as well. Again, most vendors sell a column which can retain organic acids and will be happy to share methods. Acetic acid is cursed on GC, and especially with headspace. So-called "validated" methods are just ticking timebombs, as we have learned to our regret.

Stephen