Menthol via GC-FID

[brett]

I'm having some extraction issues with a particular product I'm testing. The sample is anhydrous and made up of 20% ZnO, and probably 60% petroleum jelly. The problem is that the analyte I'm trying to quantitate is menthol at 0.2%. This is the heart of the problem, as I have to weigh 1 gram of sample in 5 mL of solvent to get it to a decent final concentration. I've been trying to extract the menthol using methanol and heat (50C) but am getting inconsistent recoveries. Is there a better way to do this?

Somewhat related question: Is chloroform (as a sample solvent) compatible with a wax/PEG GC column?

[rb6banjo]

In my experience, those solvents behave well on wax phases. It should not be a problem for you.

It sounds like a good ol' "Vick's Vapo-Rub" matrix. My Mom used a ton of that stuff on us when we were kids. Can you slurry it up in water or does it just sit there? I would do that and try headspace SPME to determine it. I know that I can do menthol in cough drops by dissolving the drop(s) in a known amount of water and then do the menthol by headspace/SPME.

I took 526 mg of a cough drop and dissolved it in 20 mL of water. I then analyzed 1 mL of that solution by HS-SPME to get:

https://flic.kr/p/owVPpF

According to the label, it is 7 mg/dose. A typical dose is 3.8 g (mass of a drop). That's 1,842 µg/g or 48 µg/mL in my 20 mL of solution. That gives a massive response in my mass spectrometer and on one of my FID detectors. I would have to dilute it more for quantitative analysis. Perhaps this technique is too sensitive. 0.2% comes in at about 2,000 ppm in the petrolatum. The Signal-to-Noise on this "48 µg/mL" (if you take the dose as true) is 6,323:1 for the FID. On the ZB-5ms phase, the menthol peak fronts quite a bit. A wax is likely a better choice.

You might be able to do the menthol in the petrolatum directly using headspace analysis as the menthol likely won't partition into the headspace readily (it'll be pretty soluble in the matrix) but you might get enough for good quantitation.

I used:

SPME Fiber: Carboxen/PDMS/Divinylbenzene-2cm (Supelco, Inc.)
Preheat: 60 °C for 15 minutes in a dry-bath heater.
Extraction: Headspace (no stirring) at 60 °C for 30 minutes.

This is just a GC method that I have been using recently, it has no specific relevance to this analysis:

Column: Phenomenex ZB-5ms (30 m x 0.32 mm x 0.25 µm)
Injector: 240 °C, Helium carrier at 25 cm/s (methane at 40 °C), split ratio of 5:1
Oven: 40 °C for 2 min., 40-240 °C at 7.1 °C/min., 240 °C for 4.83 min.
Detector: FID at 240 °C, 30:300:25 H2:Air:He makeup gas

[Peter Apps]

What is the problem with just dissolving the sample - co-elution of menthol with hydrocarbons from the petrolatum ? A change of column might help, depending on the MW range of the hydrocarbons. You might also try dissolving the sample in hexane and then partitioning vs methanol, then analyse the methanol.

Peter

[brett]

The sample does not dissolve in water at all. Headspace analysis sounds kinda interesting but we don't have the hardware for that.

My thinking was that I want to dissolve the menthol without dissolving the petroleum jelly because I don't know how much junk my column can take. The chromatogram of the sample is below:

Oven program: 80C initial
10 degrees/min to 230C, hold for 25 min



The menthol peak looks very usable but I'm just wondering how it'll affect the life of the column since it's such a dirty sample. Maybe I'll try the hexane/methanol partitioning if this doesn't work out...

[Consumer Products Guy]

Typically I dissolve the sample in DMF and use trimethylsilyl derivatization on non-polar column for menthol.

[Peter Apps]

This may be one of those analyses where living with the junk might be more effective than doing a sample cleanup.

If you have a non-polar column (looks like you do) then a build up of hydrocarbons from the jelly will cause only a baseline rise which you can autozero out at the start of each run and maybe a bit of retention shift. You have tons of resolution to spare, so your menthol peak will always be recognisable even if it shifts a bit. An overnight bake out every now and then will clean the system out.

It's not elegant, but it might be the way to go.

Peter

[AICMM]

brett,

I do not believe you will get consistent results until you completely dissolve the matrix. Menthol is very soluble in hexane and, therefore, probably very soluble in
petroleum jelly. Until you get those two dissolved I don't expect you will get the precision you desire. (I suspect this is why CPG says completely dissolve in DMF.) If
it were me, the next thing I would consider is a backflush configuration to hold the pet. jelly while the menthol goes to analytical and then bake the pre-column clean.

Best regards,

AICMM

[rb6banjo]

What if you slurry the matrix with good solvent for the menthol and a good enough solvent to at least suspend the petrolatum - say, methanol? Then, pass the slurry through a solid-phase extraction cartridge (C8 or C18, whatever's needed to retain the petrolatum). The menthol should pass through with the methanol and the petrolatum should remain on the stationary phase in the SPE tube.

[Consumer Products Guy]

You could try something like DMF or methanol and warm to dissolve/extract, then chill to kick out the petrolatum.

[willemdm]

hi,

maybe you can use a GPC purification to seperate the petroleum jelly from the smaller molecules like menthol.

[MSCHemist]

I think you need to do a SPE cleanup of some sort. Perhaps dissolve the same in ethanol (with a few ml of acetone first) that dissolves all but the most nonpolar compounds and run it through a C-18 SPE column and hopefully the petroleum should bind and stay on the SPE. That prep works well with capsaicin from oleoresins.

basic protocol
dissolve in ethanol
warm 80 deg in closed container
cool
apply 5ml to Sep-Pak C18 (I use the type that attaches to a 10ml luer syringe)
Wash with 5ml of ethanol 3 times
concentrate to whatever you'd like or I completely dry it then redissolve in MeCl2 to exclude carbs and salts.

[brett]

So many different ideas!

I see how the SPE would work but I think the time and cost of analysis wouldn't be too practical in our case.

A weird side note is that when I use 100% chloroform as the solvent, I get a consistent ~70% recovery (I tested the linearity from 50% to 150% and got an R2 > 0.999). When I dissolved in chloroform first, then extracted with an equal part menthol, I get much closer to 100% recovery.

I still have more testing to do but as my thanks, I'll be sure to let you all know how it turns out!

[MSCHemist]

C18 carts are ~ $2 a piece and I buy ethanol accross the street at the liquor store (Rectified spirits $40 for 1.75 L 96% ethanol the rest water). It doesn't take too long to do though for large numbers of sample you may want to use regular carts and a vacuum manifold.

[brett]

The GC method was validated as is. I guess we'll see how long the column lasts before I have to cut it. Seems to be running fine so far.

And, I did manage to obtain a few sample C18 cartridges but haven't had the chance to try them out yet. Thanks for all your input!