Chromatography Forum

Please forgive me if this is a dumb question. I get analytical standard grade material from Sigma Aldrich which the Certificate of Analysis reports an assay of 95.6% by HPLC. My assigned method is a gas chromatography weight percent method. Is it proper/recommended/industry standard practice, to weigh up my standard material and use the HPLC generated assay from the supplier in order to correct the weight for assay for a method which is being run on a GC??
Example: HPLC CofA shows 95.6% assay by HPLC, so I weigh up 1.0 gram for my standard to be run on a GC method. Since the 1.0 gram is only 95.6% then the actual weight of standard material corrected for assay is 0.956 grams. Is using the 0.956 grams as the standard weight accurate when I run samples as weight percent analysis?

*Please consider that UV HPLC is only responsive to certain functional groups or molecular structure - so I don't see how using HPLC by area percent report is ever appropriate to know the actual mass and true purity of a compound. What if there are other impurities which don't absorb light?

It should be ok. The assay is telling you the purity of the standard material you are using, and that can be determined by many different methods, but if it is accurate it should carry over into other testing methods.

Using that purity you can also weigh up 1.046g of material to have exactly 1.000g of the pure substance, if that makes the later calculations easier to work with.

It should be ok. The assay is telling you the purity of the standard material you are using, and that can be determined by many different methods, but if it is accurate it should carry over into other testing methods.

Using that purity you can also weigh up 1.046g of material to have exactly 1.000g of the pure substance, if that makes the later calculations easier to work with.

I should have emphasized that the HPLC assay of the analytical standard material was by area percent calculation at the given wavelength. Is it safe to assume that HPLC response factors are like Thermal Conductivity Detectors response factors, in that they are very similar across most compounds, such that the area is proportional to the weight in the area percent CofA HPLC value?

*My weight percent method however is run on a Flame Ionization Detector on the GC where area percent calculation can't accurately be used due wide variety of response factors based upon the number of carbon to hydrogen bonds per molecule. Thus, if I ran an area percent on GC FID the assay will be different than an HPLC area percent.

So bottom line is if HPLC analysis response factors are close to the same across the analyte and impurities, then that translates to actual weight on a balance of in my example of 0.956 grams of analyte?

**The main issue is that I know how much of the weight of the analytical standard material is due to the analyte for calibrating on the GC FID.

It should be ok. The assay is telling you the purity of the standard material you are using, and that can be determined by many different methods, but if it is accurate it should carry over into other testing methods.

Using that purity you can also weigh up 1.046g of material to have exactly 1.000g of the pure substance, if that makes the later calculations easier to work with.

I should have emphasized that the HPLC assay of the analytical standard material was by area percent calculation at the given wavelength. Is it safe to assume that HPLC response factors are like Thermal Conductivity Detectors response factors, in that they are very similar across most compounds, such that the area is proportional to the weight in the area percent CofA HPLC value?

*My weight percent method however is run on a Flame Ionization Detector on the GC where area percent calculation can't accurately be used due wide variety of response factors based upon the number of carbon to hydrogen bonds per molecule. Thus, if I ran an area percent on GC FID the assay will be different than an HPLC area percent.

So bottom line is if HPLC analysis response factors are close to the same across the analyte and impurities, then that translates to actual weight on a balance of in my example of 0.956 grams of analyte?

**The main issue is that I know how much of the weight of the analytical standard material is due to the analyte for calibrating on the GC FID.

If it is HPLC/UV then the response will vary depending on the UV absorbance characteristics of each analyte and impurity, some other detectors will have either better or worse response uniformity just like with GC detectors.

Some companies also include secondary test that include extinction coefficient, melting point/boiling point purity checks, or DSC testing. If you are looking at an organic compound and there is a trace metal then you would need more than the HPLC method to detect that by Area Percent though usually that would be in the ppb or ppm level and not percentages, one would hope.

We had a confusing incident related to a purity by HPLC C of A just yesterday that this reminded me of. C of A indicated 95% by HPLC as did the product specification sheet. But the department that was relying on it had some strange results in the finished product. Turns out the producer of the C of A and spec sheet was referring to the fact that the peak of interest in the chromatogram made up at least 95% of the total peak areas produced.

We later were able to determine via conversation with the vendor that the product itself ranges from 70 to 75% of the material in question dissolved in water, a component that reliably was not included in the total area count of the chromatogram.

We had a confusing incident related to a purity by HPLC C of A just yesterday that this reminded me of. C of A indicated 95% by HPLC as did the product specification sheet. But the department that was relying on it had some strange results in the finished product. Turns out the producer of the C of A and spec sheet was referring to the fact that the peak of interest in the chromatogram made up at least 95% of the total peak areas produced.

We later were able to determine via conversation with the vendor that the product itself ranges from 70 to 75% of the material in question dissolved in water, a component that reliably was not included in the total area count of the chromatogram.

That definitely should have had a solution concentration listed on the C of A or on the bottle itself.