This research addressed the problem of intrusion of hazardous chemical vapors from subsurface sources to surface and subsurface structures. Volatile organic compounds (VOCs) are commonly found entrapped as non-aqueous phase liquids (NAPLs) in the soil pores or dissolved in groundwater at industrial waste sites, refineries, and Department of Energy (DOE) and Department of Defense (DoD) complexes. Vapors emitted from these contaminant sources readily disperse into the atmosphere, into air-filled void spaces within the soil, and migrate below surface structures, leading to the intrusion of contaminant vapors into indoor air. Although it has been recognized for some time that vapor intrusion (VI) is a potential exposure pathway, a complete understanding of the basic mechanisms and methods to characterize this pathway have been lacking. Existing approaches to managing the VI pathway have often neglected the complex factors associated with the atmosphere-subsurface boundary, a dynamic water table at the unsaturated-saturated zone interface, subsurface heterogeneity, and NAPL source zone conditions. These may significantly influence VI behavior, suggesting a need to evaluate whether existing knowledge is directly transferable to VI studies.

The primary objective of this project was to improve understanding of factors that contribute to the uncertainty and variability of measured vapor concentrations in subsurface buildings and structures resulting from subsurface VOC sources. Mechanisms controlling vapor generation and subsequent migration through the subsurface in naturally heterogeneous subsurface under various physical and climatic conditions were investigated using laboratory and modeling studies. The viability of using shallow surface geophysical methods for the identification of VI pathways was investigated.

The basic premise of this project was that it is not feasible to control and quantify all the climatic and hydrogeologic factors that contribute to VI in field settings to get accurate data. Also, findings from site-specific case studies cannot be generalized for all sites with diverse climate and geology. Experiments were conducted in test systems varying from small bench scale to intermediate scale. The bench-scale experiments were designed to obtain data for the improvement of fundamental process understanding. Experiments conducted in a 16-ft long, highly instrumented intermediate-scale tank integrated vapor generation, transport, and intrusion under the influence of rainfall, water table fluctuation, and NAPL volatilization. Concurrently with the experiments, research-level modeling tools were developed to interpret data and to obtain new insights for improving conceptual understanding and to provide the basis for the development of more comprehensive models. Even though it was not within the scope of this project, the development of a comprehensive model that can be used for predictions was initiated. The research models tested for their ability to capture the responses observed in the experiments were used to conduct a limited set of hypothetical scenario simulations with the goal of demonstrating the practical implications of the research findings that will lead to guideline development and better monitoring of VI sites.

The results clearly demonstrated that numerical modeling based on improved conceptual models and observations in the intermediate-scale test tanks were informative to understand the processes that govern VI in complex systems. This research established that subsurface VI pathways are dynamic and complex, sometimes contributing to counter-intuitive cause-effect relationships. If these relationships are not well understood, it will lead to deficient monitoring strategies, wrong interpretation of monitoring data, poor decision-making, and risk assessment. The complexity can be attributed to the interaction of climate and geologic heterogeneity. The naturally heterogeneous soil profile controls the rates of wetting front propagation and transient soil moisture distribution during and after a precipitation event. The propagating front triggers the movement of the vapor in the soil pores partitioned from the contaminated soil-water. The rate at which the vapor eventually moves to the building depends on the relative permeability of the air phase that directly depends on the dynamics of the spatial distribution of the soil moisture. The time-scales and the strength of the vapor signals observed in indoor air will depend on the interplay of the intensity and duration of the rainfall event and the site-specific subsurface heterogeneity. Even though the specific scenarios of pavements around the building was not investigated, some of the tank studies where soil at the land surface was kept at high saturation resulting in capping the soil, resulted in the air that potentially carry the vapor from subsurface sources preferentially moving through the subsurface to the building. This suggests that the urban infrastructure associated with the building will have an effect on the VI pathway development.

Another finding of this work that has important practical implications is that water table fluctuation imparts very complex transport behavior within the capillary fringe that has significant effects on vapor loading from the groundwater plumes that defied simplified models used in screening. The results demonstrate the important role capillary fringe plays in loading of contaminant mass from the groundwater plume to the VI pathway in the unsaturated zone. The capillary fringe is a critical interface between the saturated zone containing the dissolved contaminant and the unsaturated zone with potential VI pathways. As diffusion through the capillary fringe creates rate-limited conditions, the primary mechanism that contributes to temporal variability is the fluctuation of the water table, but not necessarily the concentrations in the groundwater plume.

Trapped NAPL sources in the unsaturated zone are capable of loading significant mass into the unsaturated zone, but the loading rate is a strong function of the moisture distribution within and in the vicinity of the source, suggesting again the importance of climate factors in vapor generation. Intermediate-scale experiments suggested that climate and weather dependent transient thermal boundary conditions at the soil surface would not have any significant effects on subsurface vapor transport. Shallow surface geophysical methods may have the potential to provide tracking of soil moisture distributions in field soils, but further development is needed.

The findings from this study have implications for monitoring VI sites and buildings. Sampling strategies need to factor in the transients associated with climate and weather. This has to be done in conjunction with subsurface characterization methods to identify potential preferential pathways that are dynamic.

At the time this project was initiated, little was known about the practical effect of weather and climate factors affecting VI and indoor air quality. The new knowledge gained has led to improved conceptual models of the pathway, giving insights into how these factors influence VI. Benefits to DoD include improved management guidance for VI sites, ultimately leading to reduction in risk to subsurface structures and homes plagued by indoor air pollution issues, and more informed regulatory decision making on contaminant cleanup levels. The research also has provided the foundation to develop comprehensive models that after field validation can be used for prediction and monitoring and remediation design.