HT GC as alternative to lipid removal.

[MSCHemist]

Hi all

I was looking at a few app notes. With nonpolar columns that can go up to ~425 deg C and triglycerides coming off in the 300-400 C deg range I was wondering if there was a need to do lipid removal. Just inject the lipids and ramp up to 400 deg. I have a lot of flavor samples where the solvent is vegetable oil. I usually opt for headspace. Why not inject them directly?

[aidnai]

My guess is that 300-400C is hot enough that you aren't just volatilizing the lipids, you're likely 'browning the fat' as well. I bet you end up making a lot of tar.

[Consumer Products Guy]

We assay a lot of fragrance components here, aren't flavor components somewhat similar? What about keeping the inlet relatively low like under 250C and then the triglycerides won't even get to the column. Just change out the liner frequently.

[GOM]

Hi

It is a good thought but since you don't need it for actually analysing the oil itself, my feeling is that that the disadvantages of HT in this case out way the advantages of your current method e.g.

Cost of column
Degradation and loss of phase - need to be really careful to remove all oxygen from the carrier
Guard column maintenance
Inlet maintenance as Aidnai pointed out (and those septa and viton o rings don't last long if using high temperature split/splitless)
At lower inlet temperatures the triglycerides still have a vapour pressure and I think that you will get bleed into the column

Overall, I would stick with your current headspace method, which you know works

Having said that , if you have the luxury I would give it a try. Success or failure - you will still have learnt something

Regards

Ralph

[MSCHemist]

lipids still muck up the column. Sometimes I can't avoid injecting a small amount of lipids in. Flavor components have a huge range of volatilities from the low end stuff like acetaldehyde, dimethyl sulfide, trimethylamine, mid range stuff like limonene, linalool, and high range vanillin, ethyl vanillin, capsaicinoids and one of the highest I do is piperine from pepper (elutes 280-300 deg C). The problem with lipids is they follow the flavors most of which go to nonpolar solvents as well and are tough to get rid off without stripping the flavors as well. Headspace and SPME have limitations and introduce biases and since the flavors are very soluble in the lipids the sensitivity can be an issue.

[Ralph C]

Hi

Sorry, I thought that your headspace method was OK.

The only other thought that I had was to remove the triglycerides by saponification.

The reaction can be carried out on a small scale (<10ml) in a 20ml crimped PE headspace vial or a 10ml screw cap septum sealed tube from Sigma and the salting out stage to precipitate the soap will force the organics into the ethanol.

The disadvantages are
a) the reaction time would be in the order of 1 hour
b) at best we could expect a 98% conversion so there may be traces of oil still present
c) Would the caustic react with any of your flavour compounds?
d) there will be a large glycerol peak so (if you aren't already using one) it will require a polar Innowax column to reduce the tailing impact.

I miss tinkering around with method developments like this

Regards

Ralph
The chromatographer formerly known as GOM

[Consumer Products Guy]

Triglycerides are not that soluble in certain solvents like alcohol, especially when chilled. Perhaps you could exclude triglycerides that way.

[MSCHemist]

Yep esters are especially important in flavors. Some other compounds will react with the heat or base.

The gold standard for flavors is SAFE solvent assisted flavor evaporation which is a vacuum distillation and extraction but that apratatus is expensive tough to set up and very labor intensive and time cosuming. I've also run Liken Nickerson simultaneous distillation and extraction Two flasks one with boiling aqueous sample the other with boiling MeCl2 boil meet and extract then go back to their respective flasks over and over again transfering a lot of the flavors to the MeCl2 and excluding nonvolatiles the problem there is the boiling cooks the sample and it is labor intensive.

The best is SPME with a 2cm carb/dvb/pdms fiber. Some biases but works really well and not too labor intensive or expensive.

salting out into acetonitrile or methanol is a possibility. Lipids are partially soluble in ethanol.

I was just brainstorming with the HT. I am thinking about it though for my GC sugars method. A normal db 1/5 gets the disaccharides. A HT just a bit higher say 350-360 can get the trisaccharides like trimaltose.

[GOM]

Hi

I appreciate that you were just brainstorming with the HT - it was a good thought. My thoughts on saponification were along the same lines.

As you say, SPME is probably the best way in terms of cost/time despite the biases of partitioning into the headspace and from the headspace into the fibre phase.

From experience I also found that the carb/dvb/pdms fiber gave the best results for a broad range of odour/flavour analytes.

Out of interest, are you analysing your own company products or competitor products or both?

Regards

Ralph

[MSCHemist]

both sometimes duplications, sometimes for projects. We do reaction flavors so we like to know what the profile is. Sometimes we have off flavors and need to identify the culprit. We has an interesting one a while back in a sour cream flavor where the diacetyl and acetylpropionyl was from oxazoles with amino acids in the flavor.