Estimating the oxygenated zone beneath building foundations for petroleum vapor intrusion assessment

Intercomparison of in-well, depth profile and online soil gas measurements for estimating petroleum natural source zone depletion rates: Will in-well gas samples suffice?

Natural Source Zone Deletion (NSZD) is a viable long-term management option for sites impacted by petroleum hydrocarbon fuels. NSZD rate estimation methods for petroleum mass losses often use soil gas gradients of oxygen, carbon dioxide, methane or vapour concentrations through the vadose zone. Seeking greater efficiencies, we investigated if existing short-screened wells are reliable for representative sampling of soil gases in a vadose zone undergoing NSZD. At gasoline, diesel, aviation-gasoline, crude oil and background locations, we compared soil gas compositions from three methods: (i) multilevel samplers (VZ) on three occasions over 18 months (three per location); (ii) oxygen sensors (GP) at multiple depths and tracked over 3 years (one per location); and (iii) short-screened monitoring wells (MW) across the water table (five per location) and sampled as per VZ. GP sensors showed seasonal and other variations, yielding variable NSZD estimates. Oxygen depth profiles from GP and VZ measurements showed good correspondence despite fluctuations. Oxygen, carbon dioxide and methane concentrations from the deepest sampling ports of VZ samplers correlated strongly with concentrations from adjacent MW installations, indicating in-well sampling in the field can provide reliable data. The advantages and limitations of using MW data in accelerating NSZD field assessments are outlined.

Soil gas gradient method for estimating natural source zone depletion rates of LNAPL and specific chemicals of concern

This paper presents a simplified approach for the soil gas gradient method for estimating natural source zone depletion (NSZD) rates of specific contaminants of concern (COCs) at sites contaminated by light non-aqueous phase liquids (LNAPL). Traditional approaches to quantify COC-specific NSZD rates often rely on numerical or analytical reaction-transport models that require detailed site-specific data. In contrast, the proposed method employs simple analytical solutions, making it more accessible to practitioners. Specifically, it requires only the maximum soil gas concentration, the effective diffusion coefficient, and the diffusive reaction length calculated from vertical soil gas concentration profiles. The simplified approach was validated against a reactive transport numerical model reported in the literature, showing consistent results within the same order of magnitude for BTEX NSZD rates at a gasoline spill site in South Carolina. Further validation using a larger dataset involved comparing NSZD rate estimates for benzene and total petroleum hydrocarbons (TPH) against those obtained using BioVapor, utilizing empirical soil gas data from the USEPA Petroleum Vapor Intrusion Database. Results demonstrated a strong correlation between NSZD rates and maximum soil gas concentrations, allowing the development of a rapid screening approach based only on the measured soil gas concentrations and literature values for diffusion coefficients and diffusive reaction lengths. This approach aligned well with previous modeling studies and was consistent with literature values for TPH NSZD rates. Overall, both the simplified and screening approaches offer practical, easy-to-use tools for evaluating temporal variability in natural attenuation rates, supporting baseline assessments and ongoing performance evaluations of remediation at LNAPL sites.

Transient migration behavior of VOC vapor in layered unsaturated soils subjected to multiple time-dependent point pollution sources: Analytical study

Predicting the migration behavior of volatile organic compounds (VOCs) vapor is essential for the remediation of subsurface contamination such as soil vapor extraction. Previous analytical prediction models of VOCs migration are mostly limited to constant-concentration nonpoint sources in homogeneous soil. Thus, this study presents a novel analytical model for two-dimensional transport of VOCs vapor subjected to multiple time-dependent point sources involving transient diffusion, sorption and degradation in layered unsaturated soils. Two representative time-dependent sources, i.e., an instantaneous source and a finite pulse source, are used to describe the short-term and long-term leakage. Results reveal that soil heterogeneity can cause pollution accumulation, especially in low-diffusivity capillary fringe. The assumption of an equivalent plane source from multiple point sources would significantly overestimate the vapor concentration and the contaminated range. The previous single point source model is no longer inapplicable when the relative distance and/or the release interval between sources is small, giving a strong interaction between multiple sources. Moreover, a faster vapor degradation rate or a higher groundwater level will reduce the area of vapor plume linearly. Hence, close attention should be paid to the time-variation characteristics of multiple sources, the vapor degradation and the groundwater level fluctuation in practice to facilitate soil remediation. The proposed model is a promising tool for addressing the above issue.

Insights into vapour intrusion phenomena: Current outlook and preferential pathway scenario

Vapour intrusion (VI) is the phenomenon by which volatile organic compounds (VOCs) migrate from the subsurface source through the soil and enter into the overlying buildings, affecting the indoor air quality and ultimately causing health hazards to the occupants. Health risk assessments associated with hydrocarbon contaminated sites and recommendations of site closure are often made by quantifying the VI risks using mathematical models known as 'vapour intrusion models' (VIM). In order to predict the health risk, various factors such as the lithological and geochemical conditions of the subsurface, environmental conditions, building operational conditions etc. are commonly evaluated using VIMs. Use of these models can overlook the role of preferential pathways like highly permeable subsurface layers and utility lines which act as the path of least resistance for vapour transport, which can increase the VI risks. The extensive networks of utility lines and sanitary sewer systems in urban areas can significantly exacerbate the uncertainty of VI investigations. The backfill materials like sand and gravel surrounding the utility lines can allow the vapours to easily pass through due to their high porosity as compared to natural formations. Hence, failure to understand the role of preferential pathways on the fate and transport of VOC in the vadose zone can result in more conservative predictions of indoor air vapour concentrations and wrong clean up approaches. This comprehensive review outlines the vapour transport mechanisms, factors influencing VI, VIMs and the role of preferential pathways in predicting indoor air vapour concentrations.

Refinement of the gradient method for the estimation of natural source zone depletion at petroleum contaminated sites

Rates of natural source zone depletion (NSZD) are increasingly being used to aid remedial decision making and light non-aqueous phase liquid (LNAPL) longevity estimates at petroleum release sites. Current NSZD estimate methods, based on analyses of carbon dioxide (CO₂) and oxygen (O₂) soil-gas concentration gradients ("gradient method") assume linear concentration profiles with depth. This assumption can underestimate the concentration gradients especially above LNAPL sources that are typically characterized by curvilinear or semi-curvilinear O₂ and CO₂ concentration profiles. In this work, we proposed a new method that relies on calculating the O₂ and CO₂ concentration gradient using a first-order reaction model. The method requires an estimate of the diffusive reaction length that can be easily derived from soil-gas concentration data. A simple step-by-step guide for applying the new method is provided. Nomographs were also developed to facilitate method application. Application of the nomographs using field data from published literature showed that NSZD rates could be underestimated by nearly an order of magnitude if reactivity in the vadose zone is not accounted for. The new method helps refine NSZD rates estimation and improve risk-based decision making at certain petroleum contaminated sites.

Numerical study of building pressure cycling to generate sub-foundation aerobic barrier for mitigating petroleum vapor intrusion

In this study, the role of building pressure cycling (BPC) method in generating a subslab aerobic barrier at petroleum contaminated sites was examined numerically. The numerical model was first validated with field observations and then used to simulate BPC applications in petroleum vapor intrusion scenarios. The results indicated that, after a long-term BPC operation (60 days), a subslab aerobic barrier could be generated with an adequate air injection rate (10 L/min in this study). The effects on hydrocarbon soil gas concentration profiles are expected to last for weeks even after the BPC system is turned off. Moreover, our investigations showed that the performances of the BPC application are virtually independent of hydrocarbon's reaction rate constant. The simulated sub-foundation aerobic conditions expected during BPC were comparable to those observed in a field study where a subsurface pipe system at the same air injection rate was used to create a subslab aerobic barrier. Thus, BPC application can represent an interesting alternative approach to the subsurface delivery systems as it is expected to achieve similar performance but with lower installation costs.

Modeling capillary fringe effect on petroleum vapor intrusion from groundwater contamination

At contaminated sites, indoor inhalation of volatile organic compounds from groundwater contamination, known as vapor intrusion (VI), is an important exposure pathway to determine groundwater cleanup level. Based on empirical analysis, US EPA concluded that there is a low probability for vapors from fuel hydrocarbons dissolved in groundwater to induce indoor concentrations that exceed risk-based standards, and recommended 6 feet vertical building-source separation distance as the risk screening tool for such cases. In this study, we examine this recommendation by performing numerical modeling to investigate the detailed effects of the capillary fringe on petroleum vapor biodegradation and attenuation. First, the numerical model is validated by comparison with laboratory data and field measurements in US EPA's database. Then the verified model is used to simulate two scenarios involving the capillary fringe effect, one with a groundwater source at various depth and the other with a soil gas source located above the groundwater level. For a groundwater contaminant source, the capillary fringe plays a significant role in VI by controlling the soil moisture content and oxygen availability, thus affecting the soil gas concentration biodegradation and attenuation. Specifically, the capillary fringe effect can significantly decrease the indoor air concentration by decreasing upward diffusion rates of hydrocarbon, increasing the thickness of the aerobic zone, and enhancing aerobic biodegradation. As a result, it is highly unlikely for sources located at groundwater level to induce unacceptable vapor intrusion risks, supporting US EPA's recommendation. Moreover, the simulations suggest that the vertical smear zone of residual light non-aqueous liquid contamination, induced by temporal fluctuations of groundwater level, may lead to a potential threat to indoor air quality for a short vertical source-building separation distance, and thus requires more attention. The sensitivity test of the numerical model also indicates that it is the vertical separation distance between building foundation and the top of the smear zone instead of the smear zone thickness that should be given more attention during the investigation.

Investigating the Role of Soil Texture in Petroleum Vapor Intrusion

In this work we investigate the role of soil texture in petroleum vapor intrusion (PVI) by performing numerical modeling, analytical calculations, and statistical analysis of the USEPA's PVI database. Numerical simulations were conducted for three kinds of soil (sand, sandy loam, and clay), and the results indicate that the maximum attenuations of vapor concentrations from source to indoor air were observed when the clay soil is below the building. In the anaerobic zone, the normalized soil gas concentration profiles were observed to be similar and independent of soil type, whereas in the aerobic zone, a more significant attenuation was observed in finer grained soils. Such a finding is consistent with the statistical results of the USEPA's PVI database, which indicate that in the near-source zone, the soil gas concentration in coarse-grained soil tends to be lower than that in fine-grained soil, possibly caused by a weaker source due to mass loss by volatilization, whereas at a distance away from the source, the measured soil gas concentrations in fine-grained soils become much lower because of aerobic biodegradation with a shorter diffusive reaction length. Thus, 3 and 5 m are proposed as soil-type-dependent vertical screening distances for fine and coarse-grained soils, respectively. It should be noted that the validity of these screening distances is examined only for relatively homogeneous soils, and they may not be applicable for cases involving layered soil systems, where the availability of O in the subfoundation should be evaluated with subslab or multidepth samples to confirm the presence of aerobic biodegradation.

Vapor intrusion risk of fuel ether oxygenates methyl tert-butyl ether (MTBE), tert-amyl methyl ether (TAME) and ethyl tert-butyl ether (ETBE): A modeling study

Vapor intrusion of synthetic fuel additives represents a critical yet still neglected problem at sites contaminated by petroleum fuel releases. This study used an advanced numerical model to investigate the vapor intrusion potential of fuel ether oxygenates methyl tert-butyl ether (MTBE), tert-amyl methyl ether (TAME), and ethyl tert-butyl ether (ETBE). Simulated indoor air concentration of these compounds can exceed USEPA indoor air screening level for MTBE (110μg/m³). Our results also reveal that MTBE has much higher chance to cause vapor intrusion problems than TAME and ETBE. This study supports the statements made by USEPA in the Petroleum Vapor Intrusion (PVI) Guidance that the vertical screening criteria for petroleum hydrocarbons may not provide sufficient protectiveness for fuel additives, and ether oxygenates in particular. In addition to adverse impacts on human health, ether oxygenate vapor intrusion may also cause aesthetic problems (i.e., odour and flavour). Overall, this study points out that ether oxygenates can cause vapor intrusion problems. We recommend that USEPA consider including the field measurement data of synthetic fuel additives in the existing PVI database and possibly revising the PVI Guidance as necessary.

A two-dimensional analytical model of vapor intrusion involving vertical heterogeneity

In this work, we present an analytical chlorinated vapor intrusion (CVI) model that can estimate source-to-indoor air concentration attenuation by simulating two-dimensional (2-D) vapor concentration profile in vertically heterogeneous soils overlying a homogenous vapor source. The analytical solution describing the 2-D soil gas transport was obtained by applying a modified Schwarz-Christoffel mapping method. A partial field validation showed that the developed model provides results (especially in terms of indoor emission rates) in line with the measured data from a case involving a building overlying a layered soil. In further testing, it was found that the new analytical model can very closely replicate the results of three-dimensional (3-D) numerical models at steady state in scenarios involving layered soils overlying homogenous groundwater sources. By contrast, by adopting a two-layer approach (capillary fringe and vadose zone) as employed in the EPA implementation of the Johnson and Ettinger model, the spatially and temporally averaged indoor concentrations in the case of groundwater sources can be higher than the ones estimated by the numerical model up to two orders of magnitude. In short, the model proposed in this work can represent an easy-to-use tool that can simulate the subsurface soil gas concentration in layered soils overlying a homogenous vapor source while keeping the simplicity of an analytical approach that requires much less computational effort.

Analytical model for the design of in situ horizontal permeable reactive barriers (HPRBs) for the mitigation of chlorinated solvent vapors in the unsaturated zone

In this work we introduce a 1-D analytical solution that can be used for the design of horizontal permeable reactive barriers (HPRBs) as a vapor mitigation system at sites contaminated by chlorinated solvents. The developed model incorporates a transient diffusion-dominated transport with a second-order reaction rate constant. Furthermore, the model accounts for the HPRB lifetime as a function of the oxidant consumption by reaction with upward vapors and its progressive dissolution and leaching by infiltrating water. Simulation results by this new model closely replicate previous lab-scale tests carried out on trichloroethylene (TCE) using a HPRB containing a mixture of potassium permanganate, water and sand. In view of field applications, design criteria, in terms of the minimum HPRB thickness required to attenuate vapors at acceptable risk-based levels and the expected HPRB lifetime, are determined from site-specific conditions such as vapor source concentration, water infiltration rate and HPRB mixture. The results clearly show the field-scale feasibility of this alternative vapor mitigation system for the treatment of chlorinated solvents. Depending on the oxidation kinetic of the target contaminant, a 1m thick HPRB can ensure an attenuation of vapor concentrations of orders of magnitude up to 20years, even for vapor source concentrations up to 10g/m³. A demonstrative application for representative contaminated site conditions also shows the feasibility of this mitigation system from an economical point of view with capital costs potentially somewhat lower than those of other remediation options, such as soil vapor extraction systems. Overall, based on the experimental and theoretical evaluation thus far, field-scale tests are warranted to verify the potential and cost-effectiveness of HPRBs for vapor mitigation control under various conditions of application.

Comparison between PVI2D and Abreu-Johnson's Model for Petroleum Vapor Intrusion Assessment

Recently, we have developed a two-dimensional analytical petroleum vapor intrusion model, PVI2D (petroleum vapor intrusion, two-dimensional), which can help users to easily visualize soil gas concentration profiles and indoor concentrations as a function of site-specific conditions such as source strength and depth, reaction rate constant, soil characteristics, and building features. In this study, we made a full comparison of the results returned by PVI2D and those obtained using Abreu and Johnson's three-dimensional numerical model (AJM). These comparisons, examined as a function of the source strength, source depth, and reaction rate constant, show that PVI2D can provide similar soil gas concentration profiles and source-to-indoor air attenuation factors (within one order of magnitude difference) as those by the AJM. The differences between the two models can be ascribed to some simplifying assumptions used in PVI2D and to some numerical limitations of the AJM in simulating strictly piecewise aerobic biodegradation and no-flux boundary conditions. Overall, the obtained results show that for cases involving homogenous source and soil, PVI2D can represent a valid alternative to more rigorous three-dimensional numerical models.

An Excel®-based visualization tool of 2-D soil gas concentration profiles in petroleum vapor intrusion

In this study we present a petroleum vapor intrusion tool implemented in Microsoft^® Excel^® using Visual Basic for Applications (VBA) and integrated within a graphical interface. The latter helps users easily visualize two-dimensional soil gas concentration profiles and indoor concentrations as a function of site-specific conditions such as source strength and depth, biodegradation reaction rate constant, soil characteristics and building features. This tool is based on a two-dimensional explicit analytical model that combines steady-state diffusion-dominated vapor transport in a homogeneous soil with a piecewise first-order aerobic biodegradation model, in which rate is limited by oxygen availability. As recommended in the recently released United States Environmental Protection Agency's final Petroleum Vapor Intrusion guidance, a sensitivity analysis and a simplified Monte Carlo uncertainty analysis are also included in the spreadsheet.