Evaluation of Silicone-Based Passive Samplers for Monitoring Organic Aquatic Pollutants

Abstract

Pollution of waterways with anthropogenic organic chemicals can pose a risk to the environment and human health. Monitoring the concentrations of such chemicals is indispensable in the assessment of potential risks and in the evaluation of sources and fates, which is crucial to risk management. Monitoring of anthropogenic organic pollutants remains a challenge to regulators. For example, environmentally relevant concentrations of many nonpolar organic anthropogenic chemicals are often in the lower nanogram range. Hence, a substantial concentration of the analytes of interest is required in order to detect these at ultratrace concentrations. Sampling, transport and extraction of large volumes of water pose substantial challenges and are very expensive. Furthermore, snap shot samples may be not representative and can miss point in time or sporadic events. High volume active sampling systems on the other hand can require a great amount of maintenance and are limited by financial as well as logistic issues, such as access to energy to power the system. Passive sampling however, has the potential to exceed active and grab sampling methods, as the sampling devices can be deployed over extended periods of time, they do not require maintenance or electricity and they are relatively cheap.

In this study, silicon rubber (SR) was evaluated as a tool for the monitoring of anthropogenic organic chemicals in water. SR passive samplers were calibrated for a broad range of chemicals (i.e. organic pesticides and PAHs), ranging in log from 2.61–6.90. Samplers were calibrated in different environmental flow scenarios, confirming that flow is an important factor in affecting the exchange kinetics of chemicals. The results demonstrated that water side resistance is the rate limiting resistance for the uptake of chemicals into SR passive samplers under most ambient conditions. In addition to the sampling rate––where possible––silicon rubber to water partition coefficients were determined for some more polar chemicals, which approached equilibrium with the sampler during the experiments.

After these initial laboratory experiments, SR passive samplers were deployed at different sites in South East Queensland to evaluate the performance of the SR material and to identify advantages and limitations of the methods. In the first field study, SR passive samplers were used for longterm monitoring of pesticides after a point source contamination. This study showed the good reproducibility and excellent sensitivity of the SR passive samplers.

In a second field study, the efficiency of the SR passive samplers to predict pesticide occurrence in water was compared with the traditional way of determining pesticide occurrence by collecting and analysing fish from the study sites. The latter still remains a valid method for the detection of deregistered organochlorine pesticides. Nevertheless, the SR passive samplers showed properties that allowed partitioning of polar compounds into the samplers as well as the accumulation of more hydrophobic chemicals, and together with the fish sampling proved to be a sensitive monitoring tool to determine pesticide concentrations in water.

The application of performance reference compounds (PRCs) on SR passive samplers was evaluated and compared to the use of passive flow monitors (PFMs). The PRC approach was shown to have certain drawbacks, mainly due to variations in the amount of PRC loaded on to the samplers as well as the limited availability of sufficient compounds that are suitable as PRCs. In this study, PFMs proved to be a valid replacement of the PRC technique for obtaining information about the flow conditions during sampler deployment. The loss of gypsum from the PFM was used to construct a preliminary model to predict the volume of water cleared by the SR passive samplers during the exposure. The model could then be successfully validated by using three deployments at different sites and under varying environmental conditions during the exposure.

Furthermore, a method was developed to extract SR passive samplers in order to assess genotoxicity of the samples in the umuC bioassay. A suitable method was established to eliminate blank toxicity. The application of a positive control showed promising results, however SR samplers deployed in a preliminary field study and subjected to the umuC assay did not yield positive results, as they were substantially below the detection limit of the assay.

This project has shown that silicone rubber is a promising passive sampling material. It can be used for the detection of a wide range of organic chemicals with octanol–water partition coefficients ranging from less than 1,000 until up to 10,000,000. Once calibrated, the SR passive samplers could be used to predict concentrations of these chemicals in water and can now be used as a valid tool for environmental monitoring.