Evaporation is a commonly used extract preparation technique in chemical characterization testing which can greatly impact the observed extractables. Although ideally only solvent is removed and the analytes remain, in practice the analytes evaporate as well, and these losses must be considered when addressing the appropriateness of evaporation as an extract preparation method. This issue is exacerbated when using evaporation before non-targeted analysis (NTA), where the identities of the analytes are not known prior to analysis. Herein, a means of evaluating the extent of evaporative loss in NTA is presented. To achieve this, a model for evaporative loss was adopted and verified experimentally, then applied to a relevant chemical space to determine the impact of experimental parameter selection. The model was used to predict recovery for a range of chemicals under a variety of experimental conditions (volume change, solvent, and temperature). The test solvents were selected to include a range of relevant properties, including molecular weight, density, and vapor pressure. The root-mean-square error of 70 evaporation recovery conditions was 12%. Excess losses, beyond the scope of this model, were found when samples were evaporated to dryness. Also, high recovery was demonstrated for select chemicals (with high air-solvent partition coefficients) at low concentration and with large volume changes. Overall, good agreement was found between the evaporation recovery model and experimental measurements, suggesting that the model can be used to predict the effect on a NTA.
Read full abstract