Can internal standard method be used in headspace study?

[zimanli]

if I use internal standard method to do standard calibration curve, then what about my real headspace measurement? I think the partial pressure of one content is not definitely related to the its concentration. How do you do your headspace measurement?

[chromatographer1]

Generally I believe the best way to perform hs analysis is to do standard addition, not an internal std.

Stds based on a matrix not the identical to the actual sample will produce different linearity curves, with or without an internal std.

If you use the same sample (a solution of drug or other matrix) but spike portions of the sample at different levels to measure a residual volatile component you can show linearity of recovery and calculate the amount present before the spiking of the analyte quite readily.

If you have to perform the same analysis (for example: for a pharmaceutical product) repeatedly, you can perform a validation of the recovery, minimum detection quanity, reproducibility, and linearity for the matrix involved. An IStd might then not be required, but is advisable so that you can confirm that any particular analysis did execute as desired, thereby proving the result of that analysis. (In other words nothing went wrong, no leaks, detector or machine failure, column breakage, etc).

I hope this is helpful,

Rod

[zimanli]

Dear chromatographer1, thank you very much for your warm-hearted replies.

I am brand new in GC. Could you please elaborate the standard addition a little bit more. My understanding of standard addition is that I spike different amount of standard into my same sample and get a calibration curve. Is that right?

Thanks a lot!

[Rafael Chust]

I agree with chrom1.

Basically standard addition is add a known quantity of standard of one of your main peaks.

Passing the whole process, you will have an idea of total recovery of your HS.

[chromatographer1]

Let us assume you are dissolving your sample into a solvent at a known concentration. For discussion purposes: naproxen into DMF at 50mg/mL

You decide you wish to measure the amount of ethanol in the naproxen.

To do a std addition, you would prepare 3 solutions of ethanol in DMF at 0.1mg/mL, 0.25mg/mL, and 0.50mg/mL

You would add 100µL of naproxen solution (5mg or 5000µg) to each of three vials. You would add then 100µL of one of each of the three ethanol solutions to the three vials. To a 4th vial you would add the naproxen solution and 100µL of the DMF you used to make the ethanol solutions.

The amount of ethanol added then would be 0µg, 10µg, 25µg, and 50µg: or in other words, 0.0%, 0.2%, 0.5%, and 1.0%

Processing these three samples you would plot the areas of ethanol found against the amount of ethanol added. Placing a line through the points, you would determine the x-axis intercept of that line. That value indicates the amount of ethanol present in the unspiked solution.

I hope this explanation is clear. It is meant only to be an example of the std addition method. If you wanted an internal std you would make a solution of the internal std in DMF and use it to dissolve the naproxen.

And I hope your four points are linear and the areas of your internal std consistent, if not, then you do not have a good HS system for some reason and parameters or other changes must be made.

best wishes,

Rod

[Peter Apps]

Hi Zimanli

You can use internal standards in HS-GC as follows.

As with all other internal standard methods you need to select an internal standard which is as chemically similar to your analyte as possible, and which you can separate from the analyte, and all the other volatiles in the sample.

You then try a few quick and dirty experiments to find how much i.s. you need to spike into your sample to give a peak for the i.s. about the same size as your analyte peak.

Then set up a calibration using either analyte spiked into a matrix that matches the matrix of your sample, or by known additions to the samples as explained in previous posts, with each standard spiked with the amount of i.s. that you got from your quick and dirty tests.

Plot the ratio of the analyte peak area to i.s. peak area against the analyte concentration or added analyte concentration to get your calibration equation.

The internal standard will compensate for fluctuations in volumes, pressures and split ratios, and to a certain extent for sample equilibration temperature, and so you should see a better linear fit for the calibration, and better repeatability. If your repeatability of peak areas is around 4 % you will probably get a repeatability of peak area ratios of below 1%.

Peter

[zimanli]

Thank you so much!
you guys are kind and great.

ziman