Field demonstration for the solvent-based sampling method to perform compound-specific isotope analysis on gas-phase VOC.

The solvent-based sampling method for collecting gas-phase volatile organic compounds (VOCs) and conducting compound-specific isotope analysis (CSIA) was deployed during a controlled field study. The solvent-based method used methanol as a sink to accumulate petroleum hydrocarbons during the sampling of soil air and effluent gas. For each gaseous sample collected, carbon isotope analysis (δ13C) was conducted for a selection of five VOCs (benzene, toluene, o-xylene, cyclopentane and octane) emitted by a synthetic hydrocarbon source emplaced in the subsurface. The δ13C values obtained for gaseous VOCs (collected from soil gas and effluent gas) were compared to measurements obtained for the same VOCs present in the source material (none aqueous phase liquid - NAPL) and dissolved in groundwater to evaluate the reliability of the solvent-based sampling method in providing accurate isotope measurements. Since the NAPL source was composed of only 12 VOCs, potential bias related to the analytical procedure (such as co-elution) were avoided, hence emphasizing on field-related bias. This field evaluation demonstrated the capacity of the solvent-based method to produce precise and accurate δ13C measurements. The isotopic discrepancies between the gaseous and the NAPL values were < 1 ‰ for 39 out of the 41 comparison points, thus deemed not statistically different based on a common isotopic uncertainty error of ±0.5 ‰. Moreover, the current field study is the first field study to report δ13C measurements for up to five gas-phase VOCs obtained from the same sample, which appears to be of interest for VOC fate or forensic studies. The possibility to use several VOC isotopic measurements enabled by the sampling method would contribute to strengthen the connection assessment between gaseous VOCs and the suspected emitting source. Accordingly, the field results presented herein support the application of this sampling methodology to conduct CSIA assessment in the frame of VOC vapor studies.

Stable carbon and hydrogen isotope fractionation of volatile organic compounds caused by vapor-liquid equilibrium.

Several types of laboratory experiments were conducted to evaluate isotope fractionation caused by phase transfer process for a selection of common environmental contaminants. Carbon and hydrogen isotope fractionation caused by vaporization of non-aqueous phase liquid (NAPL), by volatilization from water and by dissolution into an organic solvent (tetraethylene glycol dimethylether or TGDE) under equilibrium conditions was investigated with closed system experimental setups to isolate the air-liquid partitioning process. A selection of aromatic, aliphatic and chlorinated compounds along with one fuel oxygenate (methyl tert-butyl ether or MTBE) were evaluated to determine isotope enrichment factor related to respective phase transfer process. During NAPL vaporization, the residual mass of aromatic compounds, aliphatic compounds and MTBE became progressively depleted in heavy carbon and hydrogen isotopes. In contrast, during volatilization from water, the residual mass of aromatic compounds and MTBE dissolved in the water became progressively enriched in heavy hydrogen isotopes, whereas no significant change in carbon isotope was observed, except for MTBE showing a significant depletion. For the air-TGDE partitioning process, most of the aromatic compounds tested led to no significant carbon (except ethylbenzene) or hydrogen (except toluene and o-xylene) isotope fractionation. In contrast, significant carbon isotope fractionation was observed for aliphatic and chlorinated compounds and hydrogen isotope fractionation for aliphatic compounds, and are comparable to progressive NAPL vaporization in direction and magnitude. The isotope fractionation factors determined in this study are key for interpreting the change in isotope ratios when assessing the fate of gas-phase VOCs present in the soil air or when gas-phase VOCs are sampled using TGDE as the sink matrix. The results of this study contribute to expand the list of common environmental contaminants that can be assessed by the compound-specific isotope analysis (CSIA) method deployed in the frame of gas-phase studies.

Optimization of the solvent-based dissolution method to sample volatile organic compound vapors for compound-specific isotope analysis.

The methodology of the solvent-based dissolution method used to sample gas phase volatile organic compounds (VOC) for compound-specific isotope analysis (CSIA) was optimized to lower the method detection limits for TCE and benzene. The sampling methodology previously evaluated by [1] consists in pulling the air through a solvent to dissolve and accumulate the gaseous VOC. After the sampling process, the solvent can then be treated similarly as groundwater samples to perform routine CSIA by diluting an aliquot of the solvent into water to reach the required concentration of the targeted contaminant. Among solvents tested, tetraethylene glycol dimethyl ether (TGDE) showed the best aptitude for the method. TGDE has a great affinity with TCE and benzene, hence efficiently dissolving the compounds during their transition through the solvent. The method detection limit for TCE (5±1µg/m3) and benzene (1.7±0.5µg/m3) is lower when using TGDE compared to methanol, which was previously used (385µg/m3 for TCE and 130µg/m3 for benzene) [2]. The method detection limit refers to the minimal gas phase concentration in ambient air required to load sufficient VOC mass into TGDE to perform δ13C analysis. Due to a different analytical procedure, the method detection limit associated with δ37Cl analysis was found to be 156±6µg/m3 for TCE. Furthermore, the experimental results validated the relationship between the gas phase TCE and the progressive accumulation of dissolved TCE in the solvent during the sampling process. Accordingly, based on the air-solvent partitioning coefficient, the sampling methodology (e.g. sampling rate, sampling duration, amount of solvent) and the final TCE concentration in the solvent, the concentration of TCE in the gas phase prevailing during the sampling event can be determined. Moreover, the possibility to analyse for TCE concentration in the solvent after sampling (or other targeted VOCs) allows the field deployment of the sampling method without the need to determine the initial gas phase TCE concentration. The simplified field deployment approach of the solvent-based dissolution method combined with the conventional analytical procedure used for groundwater samples substantially facilitates the application of CSIA to gas phase studies.

Documentation of time-scales for onset of natural attenuation in an aquifer treated by a crude-oil recovery system.

A pipeline transporting crude-oil broke in a nature reserve in 2009 and spilled 5100 m(3) of oil that partly reached the aquifer and formed progressively a floating oil lens. Groundwater monitoring started immediately after the spill and crude-oil recovery by dual pump-and-skim technology was operated after oil lens formation. This study aimed at documenting the implementation of redox-specific natural attenuation processes in the saturated zone and at assessing whether dissolved compounds were degraded. Seven targeted water sampling campaigns were done during four years in addition to a routine monitoring of hydrocarbon concentrations. Liquid oil reached the aquifer within 2.5 months, and anaerobic processes, from denitrification to reduction of sulfate, were observable after 8 months. Methanogenesis appeared on site after 28 months. Stable carbon isotope analyses after 16 months showed maximum shifts in δ(13)C of +4.9±0.22‰ for toluene, +2.4±0.19‰ for benzene and +0.9±0.51‰ for ethylbenzene, suggesting anaerobic degradation of these compounds in the source zone. Estimations of fluxes of inorganic carbon produced by biodegradation revealed that, in average, 60% of inorganic carbon production was attributable to sulfate reduction. This percentage tended to decrease with time while the production of carbon attributable to methanogenesis was increasing. Within the investigation time frame, mass balance estimations showed that biodegradation is a more efficient process for control of dissolved concentrations compared to pumping and filtration on an activated charcoal filter.

Solvent-based dissolution method to sample gas-phase volatile organic compounds for compound-specific isotope analysis.

An investigation was carried out to develop a simple and efficient method to collect vapour samples for compound specific isotope analysis (CSIA) by bubbling vapours through an organic solvent (methanol or ethanol). The compounds tested were benzene and trichloroethylene (TCE). The dissolution efficiency was tested for different air volume injections, using flow rates ranging from 25ml/min to 150ml/min and injection periods varying between 10 and 40min. Based on the results, complete mass recovery for benzene and TCE in both solvents was observed for the flow rates of 25 and 50ml/min. However, small mass loss was observed at increased flow rate. At 150ml/min, recovery was on average 80±17% for benzene and 84±10% for TCE, respectively in methanol and ethanol. The δ(13)C data measured for benzene and TCE dissolved in both solvents were reproducible and were stable independently of the volume of air injected (up to 6L) or the flow rate used. The stability of δ(13)C values hence underlines no isotopic fractionation due to compound-solvent interaction or mass loss. The development of a novel and simple field sampling technique undertaken in this study will facilitate the application of CSIA to diverse gas-phase volatile organic compound studies, such as atmospheric emissions, soil gas or vapour intrusion.

Analytical modelling of stable isotope fractionation of volatile organic compounds in the unsaturated zone.

Analytical models were developed that simulate stable isotope ratios of volatile organic compounds (VOCs) near a point source contamination in the unsaturated zone. The models describe diffusive transport of VOCs, biodegradation and source ageing. The mass transport is governed by Fick's law for diffusion. The equation for reactive transport of VOCs in the soil gas phase was solved for different source geometries and for different boundary conditions. Model results were compared to experimental data from a one-dimensional laboratory column and a radial-symmetric field experiment. The comparison yielded a satisfying agreement. The model results clearly illustrate the significant isotope fractionation by gas phase diffusion under transient state conditions. This leads to an initial depletion of heavy isotopes with increasing distance from the source. The isotope evolution of the source is governed by the combined effects of isotope fractionation due to vaporisation, diffusion and biodegradation. The net effect can lead to an enrichment or depletion of the heavy isotope in the remaining organic phase, depending on the compound and element considered. Finally, the isotope evolution of molecules migrating away from the source and undergoing degradation is governed by a combined degradation and diffusion isotope effect. This suggests that, in the unsaturated zone, the interpretation of biodegradation of VOC based on isotopic data must always be based on a model combining gas phase diffusion and degradation.

Carbon isotope fractionation during volatilization of petroleum hydrocarbons and diffusion across a porous medium: a column experiment.

The study focuses on the effect of volatilization, diffusion, and biodegradation on the isotope evolution of volatile organic compounds (VOCs) in a 1.06 m long column filled with alluvial sand. A liquid mixture of 10 VOCs was placed at one end of the column, and measurements of VOC vapor concentrations and compound-specific isotope ratios (delta(13)C) were performed at the source and along the column. Initially, the compounds became depleted in 13C by up to -4.8% per hundred along the column axis, until at 26 h, uniform isotope profiles were observed for most compounds, which is expected for steady-state diffusion. Subsequently, several compounds (n-pentane, benzene, n-hexane) became enriched in 13C throughout the column. For the same compounds, a significant decrease in the source vapor concentration and a gradual enrichment of 13C by up to 5.3% per hundred at the source over a period of 336 h was observed. This trend can be explained by a larger diffusive mass flux for molecules with light isotopes compared to those with a heavy isotope, which leads to a depletion of light isotopes in the source. The isotope evolution of the source followed closely a Rayleigh trend and the obtained isotope enrichment factor corresponded well to the ratio between the diffusion coefficients for heavy and light molecules as expected based on theory. In contrastto diffusion, biodegradation had generally only a small effect on the isotope profiles, which is expected because in a diffusion-controlled system the isotope shift per decrease of mass flux is smaller than in an advection-controlled system. These findings open interesting perspectives for monitoring source depletion with isotope and have implications for assessing biodegradation and source variability in the unsaturated zone based on isotopes.

Carbon isotope fractionation during diffusion and biodegradation of petroleum hydrocarbons in the unsaturated zone: field experiment at Vaerløse Airbase, Denmark, and modeling.

A field experiment was conducted in Denmark in order to evaluate the fate of 13 volatile organic compounds (VOCs) that were buried as an artificial fuel source in the unsaturated zone. Compound-specific isotope analysis showed distinct phases in the 13C/12C ratio evolution in VOC vapors within 3 m from the source over 114 days. At day 3 and to a lesser extent at day 6, the compounds were depleted in 13C by up to -5.7% per hundred with increasing distance from the source compared to the initial source values. This trend can be explained by faster outward diffusion of the molecules with 12C only compared to molecules with a 13C. Then, the isotope profile leveled out, and several compounds started to become enriched in 13C by up to 9.5% per hundred with increasing distance from the source, due to preferential removal of the molecules with 12C only, through biodegradation. Finally, as the amount of a compound diminished in the source, a 13C enrichment was also observed close to the source. The magnitude of isotope fractionation tended to be larger the smaller the mass of the molecule was. This study demonstrates that, in the unsaturated zone, carbon isotope ratios of hydrocarbons are affected by gas-phase diffusion in addition to biodegradation, which was confirmed using a numerical model. Gas-phase diffusion led to shifts in delta(13)C >1% per hundred during the initial days after the spill, and again during the final stages of source volatilization after >75% of a compound had been removed. In between, diffusion has less of an effect, and thus isotope data can be used as an indicator for hydrocarbon biodegradation.