Impacts of compound properties and sediment characteristics on the sorption behaviour of pharmaceuticals in aquatic systems

Abstract

Sorption is a key factor in determining the persistence, attenuation and bioavailability of sediment-associated contaminants. However, our understanding of the sorption behaviour of pharmaceuticals in sediments is poor. In this study, we investigated the sorption behaviour of a diverse set of pharmaceuticals in a range sediment types. Sorption affinity of pharmaceuticals for all sediments was found to increase in the order mefenamic acid < cimetidine < atenolol < amitriptyline < diltiazem. Comparison of the experimental observations with predictions from an existing model for estimating sorption revealed the model worked poorly for the study pharmaceuticals. Multiple linear regression analysis was therefore used to develop new models for estimating sorption of individual pharmaceuticals based on sediment properties. The analyses indicated that sorption is related to properties such as Log Dow of a compound in the the sediment (lipophilicity corrected for the sediment pH), cation exchange capacity, clay%, organic carbon content and exchangeable Ca 2+ , although, with the exception of atenolol, robust relationships between sediment properties and sorption were not obtained. Overall, the results demonstrate how complex the processes are that drive the sorption of pharmaceuticals in sediments and highlight the need for generation of further experimental data and further model development work.

Factors affecting the dissipation of pharmaceuticals in freshwater sediments

“…Silt percentage ( R 2 = 0.461, p < 0.05) was selected as the only descriptor for cimetidine. The result observed in the present study for diltiazem and cimetidine regarding the involvement of clay and silt in the final regression models was expected because in our previous study (Al‐Khazrajy and Boxall ) we found that the sorption affinity of the compounds is highly dependent on the log D OW (diltiazem) and the organic carbon percentage and clay percentage (cimetidine), so the identification of these parameters may be a reflection of the fact that they provide information on the bioavailability of the molecules to the microbes. For ranitidine the first descriptor chosen by the model was microbial activity ( R 2 = 0.631, p < 0.01), but when organic carbon percentage was included, the fit improved ( R 2 = 0.869, p < 0.001).…”

“…Factors such as the organic carbon content of the matrix, pH, and the level of microbial activity have been shown to be important parameters determining degradation rates of ionizable compounds (Kah et al ; Xu et al ). The adsorption coefficient ( K d ) was also included (obtained from Al‐Khazrajy and Boxall [] except ranitidine) because adsorption may modify the bioavailability of chemicals (Maqueda et al ). Each pharmaceutical and sediment was considered individually.…”

Determination of pharmaceuticals in freshwater sediments using ultrasonic-assisted extraction with SPE clean-up and HPLC-DAD or LC-ESI-MS/MS detection

“…4,5 Partitioning data for sediment/water systems have been previously reported. 37 Diltiazem and Amitriptyline exhibit moderately strong adsorption to sediment while atenolol, mefenamic and cimetidine show weak affinity to sediment. No partitioning coefficient values were available in the literature for ranitidine in sediment or soil.…”

Processes of distribution of pharmaceuticals in surface freshwaters: implications for risk assessment

“…Fourteen articles referring to suspended solids and sediment partitioning were identified, with 70 compounds studied (Löffler et al 2005;Stein et al 2008;Ra et al 2008;Yamamoto et al 2009;Lin et al 2010;Maskaoui and Zhou 2010;Githinji et al 2011;Lahti and Oikari 2011;Zhou and Broodbank 2014;Martínez-Hernández et al 2014;Paul et al 2014;Svahn and Bjorklund 2015;Li et al 2015;Al-Khazrajy and Boxall 2016). Of the compounds studied, 4 were metabolites and 8 were transformation products; 20 were acidic, 38 basic, 7 neutral, and 3 were zwitterions across a pH range of 3.5 to 10 ( Table 3, Table 4).…”