��Ⱦ�����������ֵ��L�U�u��(Ӣ�İ�) ��(ji��)�x

Chapter 1 Screening Models of Vapor Intrusion 1.1 Introduction Vapor intrusion (VI) models simulate the transport of contaminant soil gas from subsurface sources into the buildings of concern at contaminated sites�� concentrating on the prediction of the contaminant concentration attenuation (i.e.�� concentration reductions) relative to a subsurface concentration during the soil gas transport process and the migration into a building[1]�� as shown in Figure 1.1. There are two kinds of VI models. One refers to simple risk screening tools�� usually one-dimensional (1D) analytical models[2]. The other includes the numerical models�� mostly multidimensional and used to study the influences of environmental factors in specific VI scenarios. In the first kind�� the prediction may not be so accurate due to the over-simplification of the complicated reality. In contrast�� numerical models can simulate more sophisticated cases but require significant computational effort and specialized knowledge�� limiting their use by ordinary site investigators. Nonetheless�� the differences are not absolute�� and predictions by screening models can still be reliable if most essential factors are included in specific VI scenarios. On the other hand�� numerical models have become more accessible to ordinary investigators due to the more user-friendly interface of modeling software and the advanced computational capacity of current personal computers. This chapter aims to overview the first kind of VI models with analytical mathematical solutions or VI screening models. In the classic review of VI models by Yao et al.[1]�� screening models are considered partially based on the previous work on pesticide transport in soil[3-8] and radon vapor intrusion[9-12]. Usually�� the soil gas transport part of VI screening models can be viewed as built upon the previous pesticide models[3-6]�� while the process of entry into the building benefited greatly from earlier studies of radon VI[9-13]. Though the subject has changed from radon to volatile organic compounds (VOCs) such as chlorinated solvents and petroleum products�� the general governing equation still works. The earliest pesticide models were developed in the late 1960s[14��15] and have been used to study the fate and transport of pesticides after the injection into the soil. The processes simulated by pesticide models usually include leaching�� volatilization�� biodegradation�� and so on. The pesticide models also consider factors like surface runoff�� evapotranspiration�� absorption�� or drainage in site-specific cases. Obviously�� only a part of these elements are relevant to VI and thus included in VI simulations. Radon intrusion models were first introduced in the 1970s[9�� 11-13��16��17]. The radon soil gas is believed to enter the indoor space through the foundation cracks or permeable walls�� mainly through advection caused by indoor-outdoor pressure differences. Such a pathway is also applied in VI involving VOCs�� the source of which�� however�� is located separately from the target building instead of uniformly distributed in the subfoundation soil as the radon source. This difference is why molecular diffusion due to concentration gradients plays a more important role in VI involving VOCs than radon intrusion. The first VI screening model was the classic Johnson-Ettinger (J-E) model introduced in 1991[18]�� which is also the most popular one. A complete VI screening model usually describes the whole process from the subsurface source (e.g.�� the contaminated groundwater) to the building of concern�� including soil gas transport�� entry into the building�� and the calculation of indoor air concentration�� as concluded by Yao et al.[1]. For the soil gas transport part�� the major differences among screening models are the simplifications of the same general governing equation of convection and diffusion[1]. Then�� the screening models employ different assumptions of the entry route through the building foundation�� usually involving a dirty floor�� a permeable slab or a slab crack. At last�� with the calculated contaminant entry rate into the building�� the indoor air concentration can be predicted by assuming the indoor space as a continuous-stirred tank. In other words�� temporal or spatial variations in contaminant indoor air concentration were not considered in the calculation of indoor concentration in a steady-state. In some cases�� an empirical subslab-to-indoor air attenuation factor was used as an alternative to estimating the indoor air concentration with the subslab soil gas concentration. Contaminant vapor transport in the soil involves advection�� diffusion�� and other mass change terms�� such as biodegradation and absorption/desorption�� the latter of which only plays a role in transient cases[1]. Currently�� it is generally agreed that diffusion dominates soil gas transport in the vadose zone[19�� 20]�� as confirmed by comprehensive numerical simulations of three-dimensional