alcohol/acetic acid GC analysis in fermentation broth

[doublelost]

Hey, everybody,

I am going to use GC to do fermentation products (ethanol, n-butanol and acetic acid) analysis. But I was puzzled about how to pretreat the samples before shooting them into GC to get rid of those salts (like NH4Cl, KH2PO4, NaCl, MgSO4...) and sugars (fructose). Otherwise the salts and sugars might ruin the column...
My colleages suggested to use headspace but I don't think it will work for my samples, because generally the product concentration in the fermentation broth are quite low (e.g. ethanol conc. < 1 g/L). Also I read in one paper using n-butanol for ethanol extraction but obvious this is not workable either, since n-butanol is one of my products. Right now distillation is the only method that I could think of to separate the products from the fermentation broth; however, it makes the analysis work too complicated and moreover, error could be very large because if there is water remaining in the broth, it will affect the ethanol/butanol concentration results...

So...anyone got any idea for a better pretreatment method?

I do appreciate your reply.
Thank you!

Natalie Shen
PhD student, research asistant
Iowa State University
Ames, IA, 50011

[Peter Apps]

Hi Natalie

I would definitely go for headspace as the first option, or SPME if you can find a fiber that is sufficiently non-selective over the analytes that you need to detect.

Peter

[chromatographer1]

You will find it difficult to analyze your broth for acetic acid by GC. Acetic acid is best analyzed by ion chromatography, not GC. You has to have special column and a very inert sampling system to measure trace amounts of acetic acid.

Ethanol and butanol are much easier and headspace is the best procedure. Be sure to do standard addition for any accurate measurement on one sample of broth, divided up into 3 or more portions. I would suggest you add known amounts to two portions and then do headspace on the 3 vials. Don't forget to add the same volume of blank solvent to the third vial that you added as a spiked solvent to the first two vials.

I would suggest you purchase a precolumn to trap the salts and sugars so you could do the analysis more simply but they will most likely react with the acetic acid and you will fail to measure accurately any content of acid you could have.

best wishes,

Rod

[Beetle Juice]

Hey, hey
Headspace should be ideal for the alcohols just make sure you add an internal standard of somesort to your samples and standards, deutrated ethanol or isopropanol would be particulary elegant if they seperate well on your column. The purpose of adding this is to measure the difference in extraction efficency between your samples, which have high salt concentrations giving larger responses due to salting out effects and your standards which are prepared in a pure water matrix. The limit of a method such as this if run in SIM on a GC/MSD system would be around 10-100ug/L providing you a leak free system
As for acetic acid by GC you need a polar column preferably acid modified otherwise you will get dodgy peak shapes, by headspace is a nightmare becuase the compound likes to break down on any active sites. Its best done by HPLC or IC, but can be done by sample dilution (1:10) and filtration then by direct aquoues injection (pulsed splitless) 3ul injection into an inlet at low temperatures 180*C on a FFAP column lod around 10-50mg/L. Has worked

[WillyOne]

Do not become too crazy with salt content.
We analyze wines and liquors injecting directly in a split injector. Well not so easy, we add an Internal standar and we fill the glass insert with deactivated glass wool. The column is 530 microns. I do not have the full details at hand (here is midnight).
When the chromatogram degradates we change the glass wool and continue without problems

[WillyOne]

Excuseme only for alcohol contents. Acetic acid by potentiometry where we read total acids expressed as acetic or enzymatic analysis

[Peter Apps]

Hey, hey
Headspace should be ideal for the alcohols just make sure you add an internal standard of somesort to your samples and standards, deutrated ethanol or isopropanol would be particulary elegant if they seperate well on your column.

Compared to the target analytes, deuterated standards have higher molecular weights, and therefore different partition co-efficients between liquid and gas phase. The only perfectly equivalent ISs for headspace are optical isomers, which do not exist for small alcohols.

Peter

[Karen01]

Compared to the target analytes, deuterated standards have higher molecular weights, and therefore different partition co-efficients between liquid and gas phase. The only perfectly equivalent ISs for headspace are optical isomers, which do not exist for small alcohols.

While In some case the matrix effect on partition may be large, sometime it can be small enough to be negligible for practical purposes, so much so that an IS or even just external standards can be enough...

Spiking experiments will tell you what makes sense to do.

- Karen

[Peter Apps]

Since we are talking about fermentation broths here, which can vary from consomme to chunky vegetable depending on how well the bugs grow, and which always have nutrients and extraneous waste products swimming around as well, the matrix effect is very likely to be substantial, and variable.

I am more optimistic than Rod about acetic acid by GC. The lower limits will be higher than for the alcohols, and the repeatability not as good, but much will depend on the concentrations that you need to get down to. You will need to lower the pH with a strong acid, and that might give some issues with hydrolysis of other components.

Peter

[doublelost]

I do appreciate for your replies.
By the way, is guard column gonna help?
Thanks!

Natalie Shen

[Peter Apps]

A guard column will not make much (any ?) difference with headspace. It will help protect the column if you do liquid injections, but crud in the retention gap will have a worse effect on results than will crud in the inlet liner, and inlet liners are easier and quicker to change.

Peter

[chromatographer1]

I have seen issues with acetic acid which give confidence to the unaware user of a method and then terrible results emerge without recognition which cause regulatory people to give all kinds of grief to unsuspecting chemists that not all possibilities had been addressed in measuring acetic acid by those who 'validated' the method.

These all relate to chemistry, to ionized species which are matrix and time affected, and which have great bearing to the results. Then you add the immaculate condition and inert surface chemistry that is required for low levels and reproducibility that regulatory concerns desire, it becomes a can of dead worms which no one wishes to open knowingly.

I have been there and watched superb chemists work months on a method and who came to the conclusion, it isn't worth the risk.

Better to use IC or other means than to get into the legal issues with governments and gaol time.

Now if you are only doing tests which have no bearing on human health or medical issues then full speed ahead, and damn the torpedoes.

But I prefer to retire on the beach than in a jail.

just my two cents worth.

Rod

[Peter Apps]

Natalie

What concentrations of acetic acid are you expecting ?

Do you have access to liquid chromatography for these samples ?

Peter

[doublelost]

Well, I know that HPLC is a better quantification method for my samples but we just have GC in our lab. So my advisor wanted to make me work on GC firstly.
I am expecting the acetic acid concentration would be around 2 - 4 g/L.
Thanks!

Natalie

What concentrations of acetic acid are you expecting ?

Do you have access to liquid chromatography for these samples ?

Peter

[Karen01]

Since we are talking about fermentation broths here, which can vary from consomme to chunky vegetable depending on how well the bugs grow,

You centrifuge the sample and filter the supernate and then test that by headspace.

I am more optimistic than Rod about acetic acid by GC. The lower limits will be higher than for the alcohols, and the repeatability not as good, but much will depend on the concentrations that you need to get down to. You will need to lower the pH with a strong acid, and that might give some issues with hydrolysis of other components.

I know from experience that works and that hydrolysis is not an issue, at least if the HS oven is in the 60-70C range and you don't have to equilbrate too long (which depends in part on the volume and vial size you use.

- Karen