Formation of formaldehyde as an artifact peak in head space GC analysis resulting from decomposition of sample diluent DMSO: A GC-MS investigation with deuterated DMSO

Abstract

During our method development for residual formaldehyde detection in a drug substance, unusually high levels of formaldehyde were detected when using a mixed solvent of EtOH/DMSO (4:1, v/v) as sample diluent in headspace GC analysis (HS-GC). Initial investigation found that formaldehyde is used in the preparation for one of the starting materials of the drug substance. Nevertheless, there is neither other source of formaldehyde in the manufacturing process of the drug substance, nor would formaldehyde be generated during the process. In the ensuing root cause investigation, it was found that once the solvent DMSO is replaced by other solvent [e.g., N,N-dimethylformamide (DMF)], while keeping other method parameters unchanged in the HS-GC analysis, the level of formaldehyde in the same batch of the drug substance became undetectable (LOD: 3 ppm). All the evidence suggested that the observed formaldehyde in the HS-GC analysis might be due to the decomposition of DMSO, which could be facilitated by the presence of this particular drug substance. In other words, the presence of the drug substance (in the form of HCl salt) would cause a minor decomposition of DMSO to produce formaldehyde. To prove this hypothesis, a GC–MS experiment of the drug substance was conducted in which deuterated DMSO (DMSO-d6) was used in place of regular DMSO; the expected deuterated derivatization product, i.e., diethoxymethane-d2 (C2H5OCD2OC2H5), was observed in the HS-GC–MS analysis. Therefore, it became clear that this drug substance facilitates the minor decomposition of DMSO in the HS-GC analysis. In such a case, formaldehyde is an artifact peak, or ghost peak, rather than a true impurity of the drug substance. The false positive results of formaldehyde were also found in other four compounds (three drug substances and one reagent) which are all in the form of HCl or HBr salts, suggesting that generation of formaldehyde from DMSO could be a widely occurred phenomenon in HS-GC analysis of alkyl amines in the form of HCl or HBr salts, when DMSO-containing diluents are used during sample preparation.