Glycerol for GC-FID Troubleshooting

[Erin2]

Trying to make a calibration curve of glycerol on GC-FID for client reporting purposes. Have a method that works on LC but with GC, results are very inconsistent. One injection (200 ug/mL) ranges from 3-100 ug/mL detected (starts very low, goes up and back down again, there is carryover in rinses afterword). Currently injecting glycerol in DI H2O, also tried Glycerol in 10% MeOH:DI H2O.

Have been weighing out glycerol into water as syringe spiking seems impossible.

Any suggestions?

Much appreciated.

[Consumer Products Guy]

Trying to make a calibration curve of glycerol on GC-FID for client reporting purposes. Have a method that works on LC but with GC, results are very inconsistent.... Currently injecting glycerol in DI H2O, also tried Glycerol in 10% MeOH:DI H2O.

At my workplace, we assayed consumer products for glycerol (and related materials such as propylene glycol, butylene glycol, sorbitol etc.) by GC-FID every week. Sought-for levels could be 0.05% all the way up to 30%.

However, we dissolved samples in N,N-dimethylformamide and (after filtering, if necessary) mixing a portion with BSTFA derivatizing agent directly in an autosampler vial.

We pretty much used this as our guide https://aocs.onlinelibrary.wiley.com/do ... BF02540797

[Peter Apps]

Glycerol sticks to all the surfaces in the column and the injector, and depending on your injection volume, inlet temperature, split ratio, column dimensions and flow rates (none of which you tell us) you are almost certainly getting injector falshback from the high volume of vapour you get from injecting water.

If I were you I would use the LC method. Why do you want to do it by GC?

[DR]

...you are almost certainly getting injector flashback from the high volume of vapour you get from injecting water.

If I were you I would use the LC method. Why do you want to do it by GC?

If you must use GC, can you dilute in anything but water? (flashback was my first thought too)

[Consumer Products Guy]

There is a USP procedure for adulterants ethylene glycol and diethylene glycol in glycerin by GC which used water as solvent. We did this procedure at times, had to keep injection volumes line 0.5 microliters.

I actually thought the USP procedure was pretty dumb, as it was concerned with such adulterants at levels as low as 0.1%, and no one would adulterate glycerin at such low levels, just wouldn't be cost-effective. Making such procedure with limit of maybe 1% or so would've made the monograph much simpler and still find such adulterants.

[Jake]

I agree with all responses here - I wouldn't try analyzing for by glycerin GC underivatized.

There is a USP procedure for adulterants ethylene glycol and diethylene glycol in glycerin by GC which used water as solvent. We did this procedure at times, had to keep injection volumes line 0.5 microliters.

I actually thought the USP procedure was pretty dumb, as it was concerned with such adulterants at levels as low as 0.1%, and no one would adulterate glycerin at such low levels, just wouldn't be cost-effective. Making such procedure with limit of maybe 1% or so would've made the monograph much simpler and still find such adulterants.

There is a myriad of ways that diethylene glycol can make its way into the glycerin supply chain - some intentional, some accidental. Based on the large number of deaths associated with diethylene glycol-contaminated glycerin and its high toxicity, the FDA set a detection limit of 0.1%. Seems reasonable to me.

[Consumer Products Guy]

There is a myriad of ways that diethylene glycol can make its way into the glycerin supply chain - some intentional, some accidental. Based on the large number of deaths associated with diethylene glycol-contaminated glycerin and its high toxicity, the FDA set a detection limit of 0.1%. Seems reasonable to me.

Our products were topical, not ingested or injected.

One supplier wanting us to purchase their products and then sell under our name actually intentionally used diethylene glycol in it. I was asked to do reverse engineering of the product so company could estimate its materials cost, and found a then-unknown peak after derivatization. Our purchased GCMS library did not have a match, I used my rudimentary knowledge of MS fragmentation to guess at a molecular weight, and guessed diethylene glycol. We purchased some diethylene glycol from Aldrich, and when it arrived we confirmed diethylene glycol from MS pattern and retention time (and of course added the MS pattern to our custom "library".

At first the supplier denied presence of diethylene glycol, claiming it was a GC artifact so were were wrong because he "was a PhD", but I pressed the facts upon my boss who backed me up, and the supplier sheepishly admitted the did use diethylene glycol, but "any ingestion would just cause illness but not death". Needless to say - we NEVER did business with that company.

[DR]

absolutely repugnant.

[Jake]

There is a myriad of ways that diethylene glycol can make its way into the glycerin supply chain - some intentional, some accidental. Based on the large number of deaths associated with diethylene glycol-contaminated glycerin and its high toxicity, the FDA set a detection limit of 0.1%. Seems reasonable to me.

Our products were topical, not ingested or injected.

One supplier wanting us to purchase their products and then sell under our name actually intentionally used diethylene glycol in it. I was asked to do reverse engineering of the product so company could estimate its materials cost, and found a then-unknown peak after derivatization. Our purchased GCMS library did not have a match, I used my rudimentary knowledge of MS fragmentation to guess at a molecular weight, and guessed diethylene glycol. We purchased some diethylene glycol from Aldrich, and when it arrived we confirmed diethylene glycol from MS pattern and retention time (and of course added the MS pattern to our custom "library".

At first the supplier denied presence of diethylene glycol, claiming it was a GC artifact so were were wrong because he "was a PhD", but I pressed the facts upon my boss who backed me up, and the supplier sheepishly admitted the did use diethylene glycol, but "any ingestion would just cause illness but not death". Needless to say - we NEVER did business with that company.

Thanks for the info. Egads... Not even sure what to say about all that. That's really disgraceful.

And regarding the library - yeah, I would strongly prefer that the FDA/NIST ensure that specific toxic compounds being assayed for in standardized 'official' methods have them in the NIST MS library. Was your purchased library a NIST library? If so, and if it's still not there, I can help address this serious shortcoming. I work primarily in lipids, and while we dutifully buy each current NIST release, there's still a lot of basic lipid stuff still missing. But they're not toxic compounds called out in specific assays.

[Consumer Products Guy]

And regarding the library - yeah, I would strongly prefer that the FDA/NIST ensure that specific toxic compounds being assayed for in standardized 'official' methods have them in the NIST MS library. Was your purchased library a NIST library? If so, and if it's still not there, I can help address this serious shortcoming. I work primarily in lipids, and while we dutifully buy each current NIST release, there's still a lot of basic lipid stuff still missing. But they're not toxic compounds called out in specific assays.

"If" I remember correctly, our first GCMS (HP5971A) was installed the first weeks of 1990; the Hewlett-Packard/Agilent engineer put it together and gave us a short introduction. I had never seen a computer mouse before, asked what that was. This ran off a version of Windows before Windows 3, and we had purchased the NBS (later NIST) library that was available then. That's the version that did not have a spectrum for diethylene glycol or TMS-derivatized diethylene glycol. I'm sure later versions would've had these.

Anyway, early-on we decided we needed to make our own custom library for searching, and even injected about 2000 individual fragrance components to enter their spectra and retention times for our own use. Many of the components used in consumer products were not in the commercial spectral libraries.