污染地块VOCs源衰减对室内蒸气入侵风险的影响

我国挥发性有机物(VOCs)污染地块普遍采用J&E模型预测蒸气入侵风险，该模型假定污染源含量在整个暴露周期内恒定，与其在地块中的客观变化规律不符.以某VOCs污染地块为例，采用J&E恒定源模型、 SD衰减源模型及RBCA衰减源模型，分别预测VOCs侵入建筑物室内的质量浓度及蒸气入侵风险.结果显示，J&E预测暴露期内的源含量及室内污染物质量浓度始终较高，SD和RBCA衰减源模型预测显示两者均呈指数下降.RBCA衰减源模型预测的源衰减更快，但其室内污染物质量浓度预测结果小于SD衰减源模型预测结果.SD模型中建筑物室内外压差是影响源衰减的关键参数，压差增大，污染物以对流方式侵入室内，源衰减速率增加.压差降低，污染物以扩散方式侵入室内，建筑物对源衰减的作用减弱，预测结果与J&E模型差异不明显. J&E模型预测的致癌风险和危害商最高，SD次之，RBCA最低.因此，J&E模型易高估暴露期内蒸气入侵风险，RBCA衰减源模型未考虑建筑物对源衰减的阻滞而低估风险，SD模型考虑了建筑物对源衰减的影响，更适用于评估实际场地的室内蒸气入侵风险.

Effects of Source Depletion on Vapor Intrusion Risk Assessment

The current regulatory site investigation employs the J&E model to predict vapor intrusion risk. However, the J&E model assumes that the source concentration is constant for a given exposure period, which is not consistent with the actual site source under depletion. In this study, we compared the differences between the J&E model (constant source), SD source depletion model, and RBCA source depletion model for predicting indoor concentration variation as well as the risk levels during the exposure period with a case study in Beijing. The results showed that the source and indoor air concentrations predicted by the SD and RBCA models showed exponential decreases, whereas those predicted by the J&E model maintained high concentrations throughout the exposure period, which greatly overestimated the risk. The RBCA predicted source depletion at the fastest rate, but the predicted indoor air concentrations were still lower than those of the SD model, which was related to the fact that the RBCA did not consider the effect of buildings on source depletion and did not follow mass conservation. Further, the sensitivity analysis showed that the pressure difference (dP) had the greatest influence on the source concentration in the SD model. For the calculated carcinogenic risk and hazard quotients, the J&E constant source model, the SD source depletion model, and the RBCA source depletion model were ranked in descending order. The results indicated that in general the J&E model was too conservative, the RBCA model may have underestimated risk, and the SD model was more suitable for quantifying vapor intrusion risk in reality.