Arsenic Oxidation by Flavin-Derived Reactive Species under Oxic and Anoxic Conditions

The objective of this research is to see how the redox and pH conditions impact the oxidation of arsenic (As) in the presence of reduced and oxidized riboflavin species. Batch experiments using simple dark-brown glass reactors were conducted under anoxic (glovebox filled with 97-98% nitrogen and 2-3% hydrogen gases) and oxic (purified air introduced to the reactors through tubes) conditions and a range of pre-designed solution pH from 5.2 to 9.0. Variables collected in this research include the concentrations of arsenate (As(V)), arsenite (As(III)), total riboflavin, oxidized riboflavin and hydrogen peroxide (H2O2), all as a function of time. As(V) and As(III) concentrations in solution samples collected at different time points from 1 to 55 min were determined by hydrogen-generation atomic fluorescence spectrometry (HG-AFS, PS Analytical) with a detection limit of 0.5 μg/L. To measure As concentrations below this limit, anodic stripping voltammetry was performed using a 797 VA Computrace equipped with the TRACE Gold sensor (Metrohm, Application Bulletin 416/3). The total riboflavin concentrations were determined by measuring the fluorescence signal at the excitation/emission wavelength of 450/520 nm with a Flexstation-3 Multimode Reader (Molecular Devices) equipped with black/clear bottom 96-well microplates. The concentrations of oxidized riboflavin were measured by UV-vis spectrophotometry, with absorbance measured on a Flexstation-3 Multimode Reader (Molecular Devices). The detection limit of total riboflavin and oxidized riboflavin is 16 nmol/L. The concentrations of H2O2 were measured following the protocol of the commercial Fluorimetric Hydrogen Peroxide Assay Kit (Sigma-Aldrich). The fluorescence was measured at the excitation/emission wavelength of 540/590 nm, with a detection limit of 0.01 μmol/L. Additional details on the experimental setup for data collection can be found in the Supporting Information at https://doi.org/10.1021/acs.est.9b03188. The data in this research were collected in the Ecohydrology Research Group laboratories at the University of Waterloo. Funding for this work was provided by the Canada Excellence Research Chair in Ecohydrology,