Chromatography Forum

Hi people much smarter than me!
My company has brought an HS GC FID for internal monitoring of residual ethanol in product after extraction.
I only have basic knowledge so learning as i go.

Ive got a method and have done a calibration curve based on %v/v ethanol dilutions, however the previous reporting they have had via external independant accredited labs have been %w/w.
The sample is too thick to pipette and i have to weigh the sample into the headspace vial.
Are my sample results %v/v, %w/v or %w/w.


Many thanks

Several questions come up here:

Are you working in a "regulated" environment (e.g., GLP, cGMP) -- this is likely to be true for pharma, food, or environmental analysis -- and did the outside lab run a validated method? If both conditions are true, then you have to run the method exactly as written, and you report whatever the method specifies.

If you can't pipette the original sample, and you don't know it's density, then you can only measure in %w/w. As far as the calibration is concerned, you can use the densities of ethanol and water to convert back and forth from weight to volume (assuming the % ethanol is sufficiently low to not materially change the density of the final solution).

Several questions come up here:

Are you working in a "regulated" environment (e.g., GLP, cGMP) -- this is likely to be true for pharma, food, or environmental analysis -- and did the outside lab run a validated method? If both conditions are true, then you have to run the method exactly as written, and you report whatever the method specifies.

If you can't pipette the original sample, and you don't know it's density, then you can only measure in %w/w. As far as the calibration is concerned, you can use the densities of ethanol and water to convert back and forth from weight to volume (assuming the % ethanol is sufficiently low to not materially change the density of the final solution).

Hi
I am working unregulated, its just in house results for information and production purposes, we send samples to accredited labs for CoA purposes and for tests unavailable in house. We do not have access to their method as its intellectual property and i cant find a suitable one online without purchasing access to official methods.
I am assuming i take the sample result and multiply it by the density of ethanol, as used in the calibration

Depending upon what levels of ethanol you are dealing with, I might be able to help, as our company was familiar with products containing ethanol and we validated our own assays as per cGMP.

So share that information if you can. We dealt primarily with thick products so we weighed the samples, and typically reported the results in %V/V because that is how the alcohol industry reports (like with your beer or wine).

Depending upon what levels of ethanol you are dealing with, I might be able to help, as our company was familiar with products containing ethanol and we validated our own assays as per cGMP.

So share that information if you can. We dealt primarily with thick products so we weighed the samples, and typically reported the results in %V/V because that is how the alcohol industry reports (like with your beer or wine).

Thank you, i am working in the range of 0.2-3% so low level. I have been multiplying the result im getting by ethanol density to report as %w/w

I am working in the range of 0.2-3% so low level.

That's NOT low level for ethanol determination.

If me, I'd weigh sample into a vial or jar, immediately/accurately dispense a fixed amount of water already containing known amount of internal standard (such as n-propyl alcohol or acetonitrile), cap immediately, shake well, and inject 0.5 microliters (no more, water expands a ton) on wide bore polar capillary, using FID.

So personally, I wouldn't do headspace.

I am working in the range of 0.2-3% so low level.

That's NOT low level for ethanol determination.

If me, I'd weigh sample into a vial or jar, immediately/accurately dispense a fixed amount of water already containing known amount of internal standard (such as n-propyl alcohol or acetonitrile), cap immediately, shake well, and inject 0.5 microliters (no more, water expands a ton) on wide bore polar capillary, using FID.

So personally, I wouldn't do headspace.

Thanks but i cant add water, it doesnt mix with seafood oils, the headspace is working fine i just need to get my head around some of the calculations

I was thinking the same thing about the water mixing with the oil.

If you were going to do it as a direct injection you would want to use a solvent like carbon disulfide or methylene chloride so they would not create a solvent peak that overwhelmed the detector. But then you would also have to make the run longer with a high temperature backout to make sure the oils did not build up on the column.

Thank James, will probably get one of the team to dry out some inhouse oil and use that for calibrations, then i can do it by weight and dont have to do any converting

If you are doing headspace then the matrix for your standards has to be the same as it is for the samples. Then, you make your standards by adding known amounts of ethanol to known amounts of matrix, using either weights or volumes and calcualting from one to the other using density.