Aqueous naphthenic acids and polycyclic aromatic hydrocarbons in a meso-scale spill tank affected by diluted bitumen analyzed directly by membrane introduction mass spectrometry

Abstract

With the increasing use of unconventional, heavy crude oils there is growing interest in potential impacts of a diluted bitumen (DB) spill in marine and freshwater environments. DB has the potential to release several toxic, trace organic contaminants to the water column. Here, the aqueous concentrations and compositions of two classes of organic contaminants, naphthenic acids (NAs) and polycyclic aromatic hydrocarbons (PAHs), are followed over 8 weeks after a simulated spill of DB (10 L) into a freshwater mesocosm (1200 L) with river sediment (2.4 kg). These complex samples contain biogenic dissolved organic matter, inorganic ions, petroleum contaminants, suspended sediments, and oil droplets. We report the first use of condensed phase membrane introduction mass spectrometry (CP-MIMS) as a direct sampling platform in a complex multi-phase mesocosm spill tank study to measure trace aqueous phase contaminants with little to no sample preparation (dilution and/or pH adjustment). CP-MIMS provides complementary strengths to conventional analytical approaches (e.g., gas- or liquid chromatography mass spectrometry) by allowing the entire sample series to be screened quickly. Trace NAs are measured as carboxylates ([M-H]-) using electrospray ionization and PAHs are detected as radical cations (M+•) using liquid electron ionization coupled to a triple quadrupole mass spectrometer. The DB-affected mesocosm exhibits NA concentrations from 0.3 to 1.2 mg/L, which rise quickly over the first 2 – 5 days , then decrease slowly over the remainder of the study period. The NA profile (measured as the full scan in negative-electrospray ionization at nominal mass resolution) shifts to lower m/z with weathering, a process followed by principal component analysis of the normalized mass spectra. We couple CP-MIMS with high-resolution mass spectrometry to follow changes in molecular speciation over time, which reveals a concomitant shift from classical ‘O2’ naphthenic acids to more oxidized analogues. Concentrations of PAHs and alkylated analogues (C1 – C4) in the DB-affected water range from 0 to 5 μg/L. Changes in PAH concentrations depend on ring number and degree of alkylation, with small and/or lightly alkylated (C0 – C2) PAH concentrations rising to a maximum in the first 4 – 8 days (100 – 200 h) before slowly decaying over the remainder of the study period. Larger and heavily alkylated (C3 – C4) PAH concentrations generally rise slower, with some species remaining below the detection limit throughout the study period (e.g., C20H12 class including benzo[a]pyrene). In contrast, a control mesocosm (without oil) exhibited NA concentrations below 0.05 mg/L and PAHs were below detection limit. Capitalizing on the rapid analytical workflow of CP-MIMS, we also investigate the impacts of sample filtration at the time of sampling (on NA and PAH data) and sample storage time (on NA data only).