Li Shao-nan, Sun Yang, Yang Ting, Huangpu Wei-guo
Institute of Pesticide and Environmental Toxicology, College of Agriculture and Biotechnology, Zhejiang University, Hangzhou 310029, China.
Sci Total Environ. 2007 Dec 15;388(1-3):206-13. doi: 10.1016/j.scitotenv.2007.08.006. Epub 2007 Sep 10.
Mobility of two commonly used hydrophobic termiticides, chlorpyrifos and fenvalerate, was carried out by soil thin layer chromatograph using 3 field soils and 7 artificial soil as absorbent phases. Mobility factors (R(f)) were measured, and single- and multi-variable linear regression equations were then established. The result indicated that chlorpyrifos removed faster than fenvalerate in both field and artificial soils. In field soils negative correlation coefficients (r) was found between R(f) and organic matter (OM) content, pH, cationic exchange capacity (CEC), and clay content. It was noticeable, however, that correlation coefficient (r) derived from single-variable equations were not a reliable criterion for evaluation of relative importance of individual soil parameter in R(f) determination. One could see, in multi-variable regression, a functional superposition of OM and CEC in pesticide/soil interaction, and the influence of soil pH was overwhelmed by joint action of the other 3 parameters. R(f) of the two hydrophobic termiticides could therefore be predicted with adequate accuracy by either of the combinations of the two parameter: 1) OM content and clay content, 2) CEC and clay content. Introducing field soil properties into equations established from artificial soils one could see that the four-variable equations, which toke sphagnum as the only source of CEC, gave better prediction of field soil R(f). In spite of that these equations were different in two points with those from field soils: Firstly the parameter of soil pH could not be deprived, and secondly, R(f) of chlorpyrifos was positively correlated with the level of clay content.
采用土壤薄层色谱法，以3种田间土壤和7种人工土壤为吸附相，对两种常用的疏水型杀虫剂毒死蜱和氰戊菊酯的迁移性进行了研究。测定了迁移因子（R(f)），并建立了单变量和多变量线性回归方程。结果表明，在田间土壤和人工土壤中毒死蜱的去除速度均比氰戊菊酯快。在田间土壤中，R(f)与有机质（OM）含量、pH值、阳离子交换容量（CEC）和粘土含量之间呈负相关。然而，值得注意的是，单变量方程得出的相关系数（r）并不是评估单个土壤参数在R(f)测定中相对重要性的可靠标准。在多变量回归中可以看出，OM和CEC在农药/土壤相互作用中存在功能叠加，土壤pH值的影响被其他3个参数的联合作用所掩盖。因此，两种疏水型杀虫剂的R(f)可以通过以下两种参数组合之一进行较为准确的预测：1）OM含量和粘土含量，2）CEC和粘土含量。将田间土壤性质引入由人工土壤建立的方程中可以看出，以泥炭藓作为CEC唯一来源的四变量方程对田间土壤R(f)的预测效果更好。尽管这些方程与田间土壤方程在两点上有所不同：首先，土壤pH值参数不能被忽略；其次，毒死蜱的R(f)与粘土含量水平呈正相关。
