Measurement of cyclic volatile methylsiloxanes in the aquatic environment using low-density polyethylene passive sampling devices using an in-field calibration study--challenges and guidance.

Cyclic volatile methylsiloxanes (cVMS) are used in personal care products and are hydrophobic, volatile and persistent. Their environmental water concentrations are low and are difficult to detect using conventional sampling methods. This study shows the potential of passive sampling for cVMS. We used low-density polyethylene (LDPE) samplers and in-field calibration methods for octamethylcyclotetrasiloxane (D4) and decamethylcyclopentasiloxane (D5). (13)C-D4 and (13)C-D5, methyltris(trimethylsiloxy)silane (MT), tetrakis(trimethylsiloxy)silane (TK), and five deuterated polycyclic aromatic hydrocarbons (PAHs) were used as performance reference compounds (PRCs). Samplers were calibrated (7-d) using effluent at a treatment plant, with uptake of cVMS and losses of the PRCs measured at 12 time-points. Concentrations of D4 (53ngL(-1)) and D5 (1838ngL(-1)) were stable in the effluent. Uptake of D4 and loss of (13)C-D4 were isotropic and equilibrium was approached by 7-d. Two estimates of sampler uptake rate (Rs) were 2.1Ld(-1) and 2.5Ld(-1). The estimated log LDPE/water partition coefficient was 4.4. The uptake of D5 was slower (Rs=0.32Ld(-1)) and equilibrium was not reached. Offloading of (13)C-D5, MT and TK were slow, and isotropic behaviour was not demonstrated for D5. Offloading of PAHs followed the predicted pattern for LDPE. Uptake of cVMS appeared to be under membrane control, due to low diffusion coefficients in LDPE. Samplers can monitor time-weighted average concentrations of D4 for less than a week, and D5 for longer periods. LDPE samplers allow cVMS to be determined at lower concentrations than by spot sampling methods.