It is strange that you added the formic acid solution into the sample. If you made it in water, it is not suitable for GC. Somrthing is missing here.

What I know of FFA analysis and GC says you use HCl to drop the pH. It might be that the formic acid is overloading the GC column and therefore interfering with your analysis.

You should use HCl. Did you controlled the pH after acid addition?

Hello all! Thanks for the responses.

@Narong Chamkasem... Although uncommon, I have come across methods for VFA quantification with aqueous solutions. We use a split ratio of 50:1 to introduce as little water as possible into the column.

@Sigurður Smárason... Would you be able to explain what you mean by oveloading? How would I know that that is happening.

@Grzegorz Boczkaj... Would you know if HCl would be compatible with the column I mentioned? If so, I can definitely give that a try. Do you have a recommended concentration? The pH is usually between 2 - 3 after adjusting. I did not aim for a specific pH but just wanted it to get below the pKa value of all the VFAs.

@Wojciech Pietroń... it is 2% formic acid although I tried 5% and 10% too. It is prepared with HPLC grade formic acid.

Probably, you will not get any serious problems. But always such injection will cause some impurities in the injetor liner as well as can degrade the first part of the GC column. In such case if the peaks will start to tail you should replace the liner and cut first 5-10 cm of the GC column.

That is why, I would sugest to use a static hedspace or HS/SPME technique or LLE/DLLME.

regards,

G

Well formic acid is volatile and so will evaporate with the rest of your sample and transfer to the GC column. Even a 1% addition is a quite high concentration compared to your fatty acids. Depending on your starting conditions either the formic acid is going to pass through the column or adsorb into the beginning of it. Since a column has limited capacity to adsorb analytes and you have effectively added something in relatively high concentration to the mix it will compete with your analytes. In most cases this only degrades the chromatography (peak broadening). If this is happening along with lower response at higher % formic acid then it would be a good indication for column overloading.

I thank you should used acetic acid . Small amount of the acid is give the best separation and good tailing factor .

Acetic Acid is a VFA.

I suggest to acidify the sample with HCl then extract VFA to organic solvent to get clean sample. you can try with toluene, ethyl acetate or CHCl3.

I agree with what Jingkui suggested, but I want to get something clear: do you do your dilutions with your acidified solvent too? This comes from the fact that I'm not sure about your analytical procedure. By diluting your samples with the same acidified solvent you will keep constant the FA concentration. Formic acid may also be covering active sites on the surface of your liner.

I want to add something: I assume formic acid is used to drop pH, to around a value of 3, in order to protonate the acids in your sample and turn them more volatile/less active. Therefore, it makes sense that analyte reponses increase as the content of FA is raised. You should also check your peak shapes, see how they change as you increase the FA concentration. When you work with active compounds (like organic acids) you should also use deactivated liners (even if you do split injections) to get better peak shapes and get rid of an additional factor that may be affecting peak areas. Finally, consider using an internal standard for your method, it should also help you improve repeatability and make your analysis more robust.

You do realize that formic acid is a volatile fatty acid, right? So you are adding 2%, 5%, or 10% of a VFA to your sample to perform VFA analysis? This is what Dr. Chamkasem was alluding to - something is not quite right. You should be using a mineral acid (HCl) for sample preparation and dropping your pH significantly; you need to get the sample quite acidic to get extraction efficiency. This is what you are seeing with the increasing acid concentration - the analysis is telling you that you are not acidic enough. I am not certain where you got the procedure for using FA for VFA analysis but the standard methods that I am familiar with all require the use of mineral acid.

Anahita,

check the Henderson-Hasselbalch equation - this tells you what pH is needed to obtain more than 95% of your acids in non ionic form - regarding to its pKa. At this conditions you need to perform the extraction.

regards,

G

Thanks for the advice everyone. To answer some of the questions:

I have seen this method used in few papers (E.g.Article Enhanced solid-state anaerobic digestion of corn stover by a...

So far, formic acid has worked well for us. I totally understand the point of overloading the column and will keep that in mind during sample prep. I guess if I can maintain the total concentration to 2% in both the standards and samples, I can use this method. I noticed that at higher formic acid concentrations, it is distorting the acetic and propionic acid peaks. So, I will keep it to 2%.

The column I use (StabilWax DA) can take aqueous samples in very low volumes. So, I am currently testing out 1 uL with 50:1 split injection and 0.5 uL with 30:1 split injection. I also use deactivated liner as suggested. I am also testing out some internal standards.

Would you be able to suggest any other internal standard as my VFA mix contains formic, acetic, propionic, isobutyric, butyric, isovaleric, valeric, isocaproic, caproic and heptanoic acids. It is a commercially available mix.

I have tried (1) crotonic acid (2) 4-methylvaleric acid (3) 2-ethylbutyric acid. The first two were coleluting and cannot be used. The 2-ethylbutyric acid is separated "better" but still bit of overlap with valeric acid.

I suggest acidifying your samples with either HCl or H2SO4 to pH2, add some saturated NaCl and extract into a volatile organic solvent such as MTBE. Only a very small amount of strong mineral acids is needed to completely protonate the -COOH groups of the VFAs needed to facilitate extraction. The pH should be at least 2 units below the pKa values of your VFAs and so a pH of 2 or a little below is fine. Adding a small amount of saturated NaCl helps "salt out" or displace the VFAs from the aqueous phase into the organic phase to enable more complete extraction. Since ethyl acetate often contains detectable amounts amounts of acetic acid (AA), which can also be generated by hydrolysis of ethyl acetate under the acidic extraction conditions, I suggest extracting the VFAs into MTBE. For improved quantitation, 4-methyl valeric acid can be used as an internal standard as long as it's not inherently present in your samples and baseline resolution can be achieved.