Towards a Better Understanding of Activated Carbon-Based Amendments for In Situ Treatment of Petroleum Hydrocarbons in Anaerobic Groundwater Systems

Injected activated carbon (AC) particulate amendments for the in situ treatment of groundwater impacted by petroleum hydrocarbons (PHCs) is relatively new, and relies on a combination of AC sorption and biodegradation. Currently the performance of this technology remains unclear, primarily related to the long-term interplay between sorption and biodegradation and whether the presence of AC enhances the anaerobic biodegradation of PHCs. To address these uncertainties, this research investigated the sorption and anaerobic biodegradation (sulfate reducing and methanogenic) behaviour in microcosm experiments amended with AC and column experiments designed to mimic an AC permeable reactive barrier (PRB) over a period of 1 to 2 years. \nThe powdered AC (PAC) used in this research (WPC from Calgon Carbon Corporation) had a rough, irregular surface with potential macropore openings of 0.8 ± 0.3 µm, and variable particle sizes with an average diameter of 11.5 ± 4.4 µm. Sorption and desorption equilibrium experiments showed that the magnitude of single-solute (benzene [B], toluene [T], or o-xylene [X]) and multi-solute (BTX combined) sorption or desorption followed X > T > B and B > T > X, respectively. Due to competitive sorption, the magnitude of B, T, and X sorption in the multi-solute system was reduced relative to the single-solute systems. Sorption and desorption equilibrium conditions differed suggestive of hysteresis; however, this behaviour was not fully explored in this research. The best-fit single-solute Freundlich isotherm parameters for benzene, toluene and o-xylene were 36.1 ± 3.8, 0.484 ± 0.045, and 88.2 ± 7.7 for K_(f_i ) ([mg/g][L/mg]n) and 0.421 ± 0.044, 132 ± 20 and 0.371 ± 0.099 for n_(f_i ) (-), respectively. The improved simplified ideal adsorption solution (ISIAS) model was fit to the multi-solute sorption data and the competition factors (ai) were estimated for benzene, toluene and o-xylene as 1.42 ± 0.38, 1.43 ± 0.16 and 1.08 ± 0.08, respectively. Temporal sorption (up to 48 hours) and desorption (up to 720 hours) experiments showed that the time to reach sorption or desorption equilibrium for single-solute benzene and toluene was rapid (≤ 0.5 hours). \nSingle-solute (toluene-only) and multi-solute (BTX) abiotic and bioactive (including sulfate-limited [10-20 mg/L SO42-]) or sulfate amended [138-275 mg/L SO42-]) microcosms with and without PAC were constructed (in addition to starved controls without toluene, BTX or PAC). Aqueous and solid phase toluene or BTX concentrations from the single- and multi-solute PAC amended microcosms, respectively, were compared to the single- and multi-solute Freundlich or ISIAS model predictions. In general, both the single- and multi-solute sorption isotherm models were found to overestimate the measured solid phase concentrations in the microcosms. This disparity is presumably due to differences in mixing conditions and solution matrix chemistry, or competitive sorption by metabolites and biofilm formation in the PAC amended microcosms. In the multi-solute microcosm systems, the observed o-xylene solid phase concentrations deviated the most from the ISIAS model predicted solid concentrations followed by toluene and then benzene. \nIn the subset of single- and multi-solute sulfate-limited bioactive microcosms, evidence of methanogenesis coupled to a background substrate (other than toluene or BTX) was evident from the geochemical (i.e., CH4 production) and molecular (i.e., Methanomethylovorans, Methanosaeta and Methanobacterium enrichment) data. Methane production and methanogenic enrichment were consistently elevated in the bioactive microcosms with PAC, potentially supporting enhanced methanogenesis and archaeal growth in the presence of AC. Contrarily, in the subset of single- and multi-solute sulfate-amended bioactive microcosms, sulfate reduction was coupled to the oxidation of toluene or o-xylene (not benzene, which was recalcitrant) evidenced by the repetitive or preferential biodegradation of toluene followed by o-xylene in the multi-solute microcosm and supported by geochemical (i.e., SO42- reduction, and HS- and total inorganic carbon [TIC] formation) and molecular (i.e., enrichment of sulfate reducing bacteria, including Desulfosporosinus, Desulfoprunum and Desulfobacteraceae) data. In the single- and multi-solute sulfate-amended bioactive microcosms with PAC, the solid phase mass of toluene was repetitively reduced by ≥ 96% showing that PAC regeneration occurred during anaerobic biodegradation. Although anaerobic biodegradation of toluene and o-xylene were repetitively demonstrated, there was no substantial difference in the PHC, geochemical or molecular data collected between the single- or multi-solute sulfate amended bioactive microcosms with and without PAC indicating that the presence of PAC did not influence the anaerobic microbial activity. Additionally, the anaerobic biodegradation rate of toluene was not enhanced in the presence of PAC. Collectively, there were no discernible differences in the anaerobic biodegradation of toluene between the sulfate amended bioactive microcosms with and without PAC over the 1-year monitoring period. \nThree types of single-solute (toluene-only) and multi-solute (BTX) columns (37 cm long, 3.75 cm inner diameter) were constructed to represent PAC sorption alone, bioactivity alone, and PAC sorption with bioactivity. The columns were operated for approximately 2 years, with Year-1 serving as an acclimation period, and Year-2 used for high-resolution temporal monitoring. For the columns containing PAC, a 6-cm long PAC zone (0.5% wt/wt) was located near the middle of the column to mimic an AC PRB. During Year-2 for the multi-solute column with PAC sorption alone, the change in BTX concentration between the influent and effluent followed X > T > B given that o-xylene has the highest sorption capacity followed by toluene and then benzene. For the single-solute and multi-solute columns with PAC sorption alone the change in toluene concentration between the influent and effluent was greater for the single-solute column relative to the multi-solute column given that the sorption capacity of toluene was reduced in the multi-solute competitive system. For the multi-solute bioactive column without PAC sorption the change in concentration between the influent and effluent followed T > X > B due to the preferential biodegradation of toluene prior to o-xylene and the recalcitrance of benzene (as observed in the microcosm experiments). For the multi-solute bioactive column with PAC sorption the change in concentration between the influent and effluent followed T > X > B due to the preferential biodegradation of toluene and preferential sorption of o-xylene as toluene was biodegraded. \nThe solid phase concentration of toluene within the PAC zone of the single- and multi-solute bioactive columns was highest at the leading edge followed by a gradual reduction towards the end of the PAC zone. The gradient in the solid phase toluene concentration shows that toluene desorbed more due to biodegradation towards the end of the PAC zone, furthest from the influent where toluene was continually replenished. In the bioactive multi-solute column, the magnitude of solid phase BTX concentrations within the PAC zone followed X > T > B at the leading edge; however, directly downgradient the solid phase BTX concentrations within the PAC zone followed X > B > T due to the biodegradation of toluene which resulted in higher sorption of o-xylene and benzene. \nUsing the Year-2 data, an overall column mass balance was estimated by subtracting the cumulative effluent mass from the cumulative mass injected. The single- and multi-solute bioactive columns with PAC sorption yielded the largest removal of toluene mass compared to columns with either only PAC sorption or only bioactivity. For example, among the single-solute columns the mass removal of toluene was greatest for the bioactive column with PAC sorption (99.5% reduction), followed by the column with only PAC sorption (74.6% reduction) and the column with only bioactivity (44.4% reduction). The depletion in effluent toluene mass during Year-2 was consistent between the single- and multi-solute bioactive columns with PAC sorption; however, for the multi-solute column breakthrough of benzene and o-xylene occurred given the recalcitrance of benzene and the competitive inhibition of o-xylene during toluene biodegradation. \nAnaerobic biodegradation within the single- and multi-solute bioactive columns with or without PAC sorption was supported by changes in geochemical parameters that would be expected under sulfate reducing and methanogenic conditions (i.e., SO42- reduction, and HS-, TIC and CH4 formation). Like the microcosm systems, there was no difference in the magnitude of change of the geochemical parameters between columns with or without PAC, suggesting that the PAC zone in the bioactive columns did not influence the microbial activity. However, the PAC zone did influence the spatial distribution of anaerobic microbes. For the columns with only bioactivity the relative abundance of Methanosarcina and Methanomethylovorans were highest at the influent ends relative to the effluent ends. Whereas, for the bioactive columns with PAC sorption the relative abundance of Methanosaeta, Methanobacterium and Methanosarcina were highest within the PAC zone relative to outside of the PAC zone, in addition to being higher in abundance relative to all bacteria detected within the PAC zone (primarily Desulfosporosinus, Edwardsbacteria and Berkelbacteria). In the multi-solute bioactive column with PAC sorption, the abundance of Desulfosporosinus was also notably elevated at the leading edge of the PAC zones (coinciding with the location of the highest solid phase toluene concentration sorbed to the PAC). \nCompound specific isot

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.