Sample: grape FLAVOUR

[Kaliane]

I need to analyse a sample of grape flavour... Do someone work with this kind of sample?
The amount of ester and methanol are so high that the linearity of my FID is not so good... I'm afraid of work with diluition next to 10 x ... it may result in a big error...
Dilute the sample (10x) is better than a low linearity in the calibration curve? What do you think?

Thank you!

I use a GC/FID, autosampler at split mode, column WAX (polyethylene glycol).

[Consumer Products Guy]

Well, I'm surprised that your grape flavor contains methanol (hazardous) and esters; I would've expected mainly methyl anthranilate, one of the few common flavors that contains nitrogen.

We assayed flavors and fragrances routinely by GC and GCMS, saw diluents like dipropylene glycol, not methanol or esters.

[Kaliane]

Well, I'm surprised that your grape flavor contains methanol (hazardous) and esters; I would've expected mainly methyl anthranilate, one of the few common flavors that contains nitrogen.

We assayed flavors and fragrances routinely by GC and GCMS, saw diluents like dipropylene glycol, not methanol or esters.

Hello,

I'm glad you talk about the methyl anthranilate... Do you think it's possible that the peak of methyl anthranilate overlaps the peak of ethyl acetate? Or the peak of methanol? I don't have any standard of methyl anthranilate to test...

[Kaliane]

Well, I'm surprised that your grape flavor contains methanol (hazardous) and esters; I would've expected mainly methyl anthranilate, one of the few common flavors that contains nitrogen.

We assayed flavors and fragrances routinely by GC and GCMS, saw diluents like dipropylene glycol, not methanol or esters.

I found this sentence:
"The hydrolysis products of methyl anthranilate are methyl alcohol and anthranilic acid".
I have a doubt.. before inject the sample in the GC, I destilled the sample.. Are this distillation able to hydrolysed the methyl anthranilate?
Maybe the methyl anthranilate was hydrolysed in the grape flavour?

Please, help me...

What do you think?

Source: 9.1 Metabolism/Metabolites
https://pubchem.ncbi.nlm.nih.gov/compou ... ochemistry

[AICMM]

Kaliane,

1) FID's are wonderfully linear so you give me pause with this comment. Do you mean you cannot see low amounts and high amounts in the same run?

2) Use the splitter as your dilution. Build curves appropriately and run them at the same split ratio (and I would use an IS if I were you) and you should be able to work both the high end and the low end of this problem.

Methanol in grape flavoring. That's a scary thought.

Best regards,

AICMM

[Consumer Products Guy]

I don't have any standard of methyl anthranilate to test...

I don't mean to be overly harsh, but how can you do this project without having any methyl anthranilate?????

What exactly is your sample? If grape flavor, why are you distilling at all? Or is the sample some sort of grape juice?

[Kaliane]

I don't have any standard of methyl anthranilate to test...

I don't mean to be overly harsh, but how can you do this project without having any methyl anthranilate?????

What exactly is your sample? If grape flavor, why are you distilling at all? Or is the sample some sort of grape juice?

The aim is to analyse some higher alcohols, like 2-methyl-1-propanol, 1-propanol, 2-methyl-1-butanol e 3-methyl-1-butanol, and also methanol and ethyl acetate. But I got surprised with the amount of methanol... so I'm trying to understand what is happening... maybe hydrolysis...

I got the sample with the name "grape flavour".. but I'm not 100% sure about the compostion... Maybe some substances were added in the sample... I'm going try to discover more infomation about the sample... thank you

[Kaliane]

Kaliane,

1) FID's are wonderfully linear so you give me pause with this comment. Do you mean you cannot see low amounts and high amounts in the same run?

2) Use the splitter as your dilution. Build curves appropriately and run them at the same split ratio (and I would use an IS if I were you) and you should be able to work both the high end and the low end of this problem.

Methanol in grape flavoring. That's a scary thought.

Best regards,

AICMM

Yes. You are right about the linearity of FID... I had a problem with the samples.. but now its everything ok...

I'm just worried about the methanol concentration in the sample... and what it means...

Thank you!

[Consumer Products Guy]

I'm just worried about the methanol concentration in the sample... and what it means...

Since grape flavor is consumable, and methanol is hazardous, I'd say that most likely your sample prep has caused methanol to form, likely from the methyl anthranilate.

I'd keep the sample preparation as simple as possible, and avoid any alcohols to eliminate any transesterification.

[rb6banjo]

Could it be

[rb6banjo]

Could that large peak be ethanol? You don't really mention how you determined that the large impurity is indeed methanol. In my samples, on wax phases, ethyl acetate comes out slightly before ethanol.

My Conditions are: 40 °C for 2 min., 40-230 °C at 10 °C/min., 230 °C for 14 min.
Alcoholic beverages are my samples.

Ethanol is also used as a carrier for many flavors because it's not hazardous when present in reasonable concentrations. In most flavorings, you don't need a lot of the "key components" (in this case methyl anthranilate) to get the grape flavor. Most of the flavoring is going to be some type of neutral solvent - like ethanol and water.

What I'm getting at is even if you hydrolyzed all of the methyl anthranilate, I'm pretty sure you wouldn't get much methanol in your flavoring. Certainly not percent-type concentrations.

[Gizmo]

rb6banjo makes a good point with regard to ethanol. That really sound like the answer or some flavor manufacturer is doing something very wrong.

Kovats RI on DB Wax for ethanol and methanol are fairly close (~925 and ~905) and its possible that with your operating conditions there could be some real overlap between the two, or the ethanol peak is large enough to overlap a wide RI range and being mistaken for methanol - ethanol will often be quite a large peak in flavor samples if they are not covered by a solvent delay or block.

If this is a point of concern for your analysis maybe running a much more dilute sample and comparing to EtoH and MeOH standards that were effectively resolved by your method might be helpful.