Chromatography Forum

I have been given the task of trying to analyze Polyethylene glycol 3350 for Acetic acid by the current USP <467> residual solvents method. I have stated repeatedly that Acetic acid will not work the via headspace because of it's high K value in water and DMF.

1) am I correct about the K value of acetic acid?

2) even if I could get it into headspace, wouldn't the activity of the compound pose serious problems for analysis?

3) what sorts of alternatives are out there?

4) Chromatographer1, I spoke to you a while back (3 years) about the analysis of DMF and you forwarded me a paper you did on the rapid analysis of Residual solvents. What dealings have you had with acetic acid analysis?

Hi
Have not not tried it myself only encountered it by accident, but a few suggestions perhaps:

2) even if I could get it into headspace, wouldn't the activity of the compound pose serious problems for analysis?

Often problems with active solvents relate to their also high bp, around 118°C for the acid and similar for pyridine and butanols that analysts time from time tend to have problem with when taking an USP/EP method into use. By increasing temperature in valv/loop/transferline above the bp of these solvents and the sample solvent (DMF water etc) you hopefully get past that problem.

As for the column, the DB-624 type column in USP is proberbly a better choice to start with.

The problem is as you say to get the acetic acid into gas phase. Hopefully someone has a good experiance from that, but instinctively, dissolve sample/standard in diluted HCl? pushes the equlibrum in your way.

Charles

After much frustration, what you will find is that acetic acid does not chromatograph well by H/S on a DB-624 (or any G43) column. What I would suggest is that you develop and validate a separate method.

try a direct-injection method
1) use a highly deactivated injection port liner. Siltek 4-mm w/ wool (Restek) works great.
2) use a column suitable for chromatographing free organic acids. HP-Innowax (Agilent) is my favorite.
3) use ethyl acetate as a diluent for your sample. I can't explain why, but the ethyl acetate has a very positive effect on the analysis. The analysis does not work as well with any other diluent. Even if you need to use something with an intermediate polarity to improve solubility, try to make the concentration or ethyl acetate as high as possible in the final analytical solution.

I think you will be very impressed with the results. We have validated methods for a few compounds using this combination no problems. Good Luck.

I performed acetic acid on a polymer encapsulated API at a previous company. We dissolved about 200 mg of sample in 10 mL of DMF and did a direct injection, 1 ul. I think, if my oldtimers is not to bad, that we were able to see 0.05%. Whe had to modify the parameters of the GC from our current residual solvent to enhance the acetic acid peak over the other residual solvents that we were looking for.

I can not remember any other parameters or column information. Sorry.

From memory, FID is not very sensitive to acetic acid - just add that information to the list of potential woes.

If you have to use GC, there are some methods for volatile short chain fatty acids that you might want to explore, they often use special columns for underivatized acids.

I suspect that an alternative technique ( IC, HPLC ) might be easier.

Please keep having fun

At the back of my mind I have a vague recollection of small acids being headspaced by derivatization to esters. I can't remember any of the details though - BF3 and methanol in excess perhaps

Peter

My suggestion for doing acetic acid by GC is don't... particularly not with HS. we've gone to splitting out acetic acid and doing it by an LC method, and even with that extra validation chore I would never go back to doing it by GC unless absolutely force put.

Stephen

I can officially confirm that the USP method with HS does not work well ! I will try the method proposed by Steve by direct injection and evaluate the situation...

That forum will be very helpful..thank you everyone !

Jimmy, Québec

sdegrace, do you have a hplc method to propose for acetic acid ? what is the LDM ? Mobile phase ? Column ? Wavelengh ?

I would be interested to try that option...

Acetic acid (HoAc) by GC (HS or direct injection) is very difficult and almost impossible in some matrices.

The best methodology would be ion chromatography. The methods on the web are easily found through vendors.

But as long as you keep the pH more acidic than the pKa of acetic acid you should be able to measure HoAc by conductivity, refractive index or even UV (210nm or less preferably). Of course 100% water solvent and mobile phase would be preferable on a reverse phase column.

IC is proven and is the best technique I can suggest.

best wishes,

Rodney George
consultant

Thank You Rodney !

The problem is that we don't have ionic chromatograph but I will try the UV method as you suggested. Thank You very much for your time !!

Thanks to that forum, I have interesting solutions !

Hi Jimmy,

You can do HOAc by reverse phase with a column with decent retention for polar compounds, e.g., one with an embedded polar group. You need to read close to 200 nm but you can do PDA detection. I think the Atlantis dC18 from Waters would probably be good. We did it by UPLC and I found a HSS T3 (used 100 mm x 2.1 mm, 1.8 um packing) worked well. I think the Shield RP-18 would probably good too in terms of Acquity phases.

Method in a nutshell:

Mobile A: 20 mM phosphate, pH 2.75
Mobile B: Methanol
PDA Wavelength: 210 nm

The method runs isocratic 100% A at 0.5 mL/min for 1.5 min then goes into a cleaning phase for about the same amount of time at 90% B, then requilibrates for about 30 sec and back into it. Cleaning and requilibration during next injection delay.

I could do more to get this method exactly the way I want it, but it works well anyway. This method was used to do trifluoroactic acid at the same time as HOAc, BTW. YMMV depending on your specific samples. Lot of manufacturers sell columns that can be used to analyze carboxylic acids and if you ask them they'll obligingly give you part numbers and share methods.

Hope this helps,

Stephen

There is an EPA method for determination of halo acetic acids in drinking water by extraction and derivitization. It is then ran by GC with an ECD.

It is a bit of a process, but here it is:
http://www.epa.gov/safewater/methods/pd ... t552_3.pdf

I would do it by LC, given the choice, though.

Hi Stephen,

This method sounds great but in order to use It I would need to show some references. Was this method or anything similar to it published at all? Sources would be greatly appreciated.

Thanks

Shannon

Hi Jimmy,

You can do HOAc by reverse phase with a column with decent retention for polar compounds, e.g., one with an embedded polar group. You need to read close to 200 nm but you can do PDA detection. I think the Atlantis dC18 from Waters would probably be good. We did it by UPLC and I found a HSS T3 (used 100 mm x 2.1 mm, 1.8 um packing) worked well. I think the Shield RP-18 would probably good too in terms of Acquity phases.

Method in a nutshell:

Mobile A: 20 mM phosphate, pH 2.75
Mobile B: Methanol
PDA Wavelength: 210 nm

The method runs isocratic 100% A at 0.5 mL/min for 1.5 min then goes into a cleaning phase for about the same amount of time at 90% B, then requilibrates for about 30 sec and back into it. Cleaning and requilibration during next injection delay.

I could do more to get this method exactly the way I want it, but it works well anyway. This method was used to do trifluoroactic acid at the same time as HOAc, BTW. YMMV depending on your specific samples. Lot of manufacturers sell columns that can be used to analyze carboxylic acids and if you ask them they'll obligingly give you part numbers and share methods.

Hope this helps,

Stephen

You can try HP-FFAP, this column is a good choice for acid analysis.
I've done the similar test a while back, just used this column with direct injection, HOAc showed a nice peak.