Comprehensive certification of a testosterone calibration standard facilitating the investigation of charged aerosol detection for the quantification of impurities of related structure
journal contribution
posted on 2024-11-01, 04:25authored byStephen Davies, Ponhatai Kankaew, Gregory Tarrant, William Donald, Ronda GreavesRonda Greaves
Testosterone, an analyte of interest to medical testing laboratories, was purity assigned using a mass balance approach, to afford a certified value of 986 +/- 3 mg g(-1) with established metrological traceability to the SI unit for mass (kg). The gas chromatography with flame ionisation detection (GC-FID) analysis identified the thermal degradation of testosterone as a source of bias, and confirmed the other four impurities to be steroids of closely related structure through co-elution studies. Combined with measures of water, solvent and non-volatile residue the total purity was in agreement with that determined by the complementary quantitative NMR spectroscopy, providing a high level of confidence that the certified value was bias free. This sample was then used to investigate the applicability of charged aerosol detection to the assessment of high accuracy, low uncertainty organic purity determination of steroids. The relative peak areas of testosterone and impurities varied depending on the concentration of the test solution, precluding the direct conversion of peak areas to an accurate realization of the relative mass fraction. One impurity, androst-4,6-diene-17 beta-ol-3-one, being overestimated by 155% compared to the GC-FID result. Charged aerosol detection (CM)) calibration curves established for testosterone and each impurity confirmed the detector response to vary by up to 30% at low levels (0-20 mu g g(-1)), thereby challenging the view that the CAD response is independent of structure. Despite this, using the calibration curve of testosterone to quantify each impurity delivered a reasonable estimate of organic purity. The variations in detector response for each analyte were used to assign a conservative uncertainty to the calculated value, offering a way forward when CAD is to be used for a less well defined material. Finally, the candidate material demonstrated a propensity to absorb water, highlighting the need to monitor water absorption during on-going stability trials.