The massive production of electronic products generates huge amount of electronic waste (e-waste) in the world, which gradually contaminates the environment. A range of chemical contaminants (e.g. toxic metals or flame retardants) commonly targeted in surveillance programs have been already identified as pollutants emitted from e-waste. This sadly suggests there may be other families of contaminants released by e-waste disposal and recycling activities.<br/>Non-targeted analysis, based on high-resolution mass spectrometry (HRMS) is a novel approach for the screening of a broader range of contaminants in environmental matrices. The resulting raw HRMS data are complex and contains both informative data and noise, and data processing is required to extract useful peaks while removing uninformative data. A range of data processing software / algorithms (both proprietary and open-source) are now available to track environmental contaminants. However, some studies suggest that various algorithms may result in different data interpretation.<br/>In this study, liquid chromatography coupled to hybrid quadrupole time of flight mass spectrometry (HPLC-QTOF-MS) was used for acquiring HRMS data in both positive and negative ionization modes for soil samples (including some soils for agricultural purposes) collected near an e-waste site, Agbogbloshie (Ghana) and a control site, Asesewa (Ghana). The data were then analyzed through the proprietary data processing software provided the instrument vendor and an open-source software. Different peak picking, peak alignment, and denoising algorithms and parameters were used to generate peak lists for subsequent identification, and the number of extracted features were compared. The quality of extracted features was then assessed by minimum consecutive data points that each peak consists of and signal-to-noise ration (S/N). The numbers of overlapping peaks obtained under different parameters between the two software were also assessed. Differences in tentatively identified peaks were observed when different parameters and databases were used, and suggestions will be discussed to limit the impact of data processing on the identification of unknown chemicals.

Comparison of various non-targeted data processing workflows for the identification of chemical contaminants in e-waste impacted soils

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.