Volatilisation of crop protection chemicals from crop and soil surfaces under controlled conditions—prediction of volatile losses from physico-chemical properties

Abstract

Volatilisation of crop protection chemicals from soil and crop surfaces is one of a number of processes that may contribute to their dissipation in the environment. Therefore, information on the potential of a chemical to volatilise from these surfaces is required by international and national registration authorities. This paper reports the results of more than 190 experiments, which were carried out with 80 different crop protection chemicals under controlled conditions (laboratory and/or greenhouse) according to the BBA guideline. Percent loss values observed during 24 h after application are reported for 123 soil and 71 crop volatility studies. Generally, volatile losses from crop surfaces were found to be greater than from soil surfaces under comparable experimental conditions. It has been previously proposed that volatile losses from soil surfaces, from crops, and from aqueous systems can be estimated from physico-chemical parameters. The data are therefore analysed to determine whether a correlation exists between volatilisation and physico-chemical parameters, such as vapour pressure, Henry’s law constant, water/air and soil/air distribution coefficients. It was found that these parameters can be used to make reasonable predictions of volatile losses from crop and soil surfaces, which can be expected for crop protection chemicals under controlled conditions. Vapour pressure was the best predictor of losses from soil and crops. The use of the soil/air distribution coefficient is an alternative for predicting/estimating the volatility potential of a chemical from soil. Based on direct measurements, no noticeable volatility can be expected from compounds with a vapour pressure below 10 −3 Pa from soil and 10 −4 Pa from crops, this is fully confirmed by indirect measurements. A tiered volatility testing scheme including appropriate trigger values is proposed.