Residual Solvents by FID

[Savepuck60]

Hello,
I am validating the USP 467 method, new to our lab.

Using Thermo GC and Headspace equipment along with Chromeleon software.

Our question is in regards to the final sample calculation.

We analyzed a calibration curve, 5-5000 ug/ml in DMF. We placed 1ml of each standard in the headspace vials.

For our sample, a dry crystal, we weighed 50mg into 1ml DMF into the headspace vial.

Our on column concentrations of our standards would be expressed in ug, as for example, the 100ug/ml standard, using 1ml, would be 100ug.

We then are taking the on column concentration of each analyte in the sample and dividing by the exact weight (i.e. 0.053mg).

So for example, an on column result of ethyl acetate of 800 ug/ml (ug) would be divided by 0.053g to give us a results of 15,094 ug/g.

We would like to verify that this calculation is correct. Thank you

[rb6banjo]

If you plotted 5-5,000 µg (not µg/mL) on the x-axis of the calibration plot, then treating all things equal when preparing your samples in 1 mL of DMF, you determine that the sample contains 100 µg of analyte, your final answer would be:

100 µg/0.053 g = 1,887 µg/g

If you plot µg Analyte (x-axis) vs. Peak area (cts, y-axis) the units of your response factor are cts/µg. The response divided for the unknown divided by the response factor gives µg (cts/(cts/µg) = µg). You know that the mass of the sample in the DMF was 53 mg. µg/g = 100 µg/0.053 g = 1,887 µg/g.

The DMF is just a spectator here. You must make sure that all calibration standards were prepared just like the sample. You can't add 53 mg of sample to 2 mL of DMF. It all falls apart that way.

[Savepuck60]

Thank you for your response.

Yes both samples and standards are final volume one ml in the 22ml headspace vial.

Our working standards were prepared in DMF from a custom mix of 10,000ug/ml.

Our samples are prepared by weighing 50mg crystal in the headspace vial and then adding one ml DMF and processing.

Our on column sample result is calculated from our calibration curve, in ug/ml and expressed as ug as one ml of each standard was placed in each headspace vial.

Our sample result is expressed as ug and then converted to ug/g after dividing by our sample weight.

[rb6banjo]

I don't see a problem with what you're doing because you are using a 1 mL "sample size" for your calibration. If you plot Concentration (µg/mL) on the x-axis and response on the y-axis, the extrapolated result (in µg/mL) is the same as µg - because you'd have to multiply by 1 mL before dividing by your sample mass in grams.

[Savepuck60]

Hello,

Our plant shutdown for upgrades, we are back up.

A follow up question if I may...

We analyzed a few samples, 50mg each dissolved into one ml of DMF.

We are being asked to report in percent. For example acetone sample value was 663.68 ug on column and the sample weight 0.0566g.

We divided the value by the weight and obtained 11,725 ug/g.

How would we convert this value to percent? Thank you

[Consumer Products Guy]

I am validating the USP 467 method, new to our lab.

I've been retired almost a decade, but back then our QA agreed with USP that USP methods were considered validated and did not need to be validated by the user.

Yes, our QA department asked for some type of verification at the locations which would use the USP methods.

Understand that I personally felt many USP methods were awkward or poorly written, but that was my opinion...

[rb6banjo]

If you measured 663.68 µg of acetone in the sample and you started with 56.6 mg of sample:

663.68 µg x 1 mg/1,000 µg = 0.66368 mg

0.66368/56.6 x 100 = 1.17% acetone in the sample.

% is a dimensionless parameter. It's the fraction of - in this case - the impurity relative to the mass of the sample.

[Savepuck60]

We now understand, thank you very much!

Hopefully you will take some of our vitamins and enjoy good health!