Pharmaceutical impacts on aerobic granular sludge morphology and potential implications for abiotic removal

Abstract

The goal of this study was to investigate abiotic pharmaceutical removal and abiotic pharmaceutical effects on aerobic granular sludge morphology. For 80 days, a pharmaceutical mixture containing approximately 150 μg/L each of diclofenac, erythromycin, and gemfibrozil was fed to an aerobic granular sludge sequencing batch reactor and granule characteristics were compared with those from a control reactor. Aqueous and solid phase pharmaceutical concentrations were monitored and staining was used to assess changes in biofilm structures. Solid phase pharmaceutical concentrations were elevated over the first 12 days of dosing; however, they then dropped, indicative of desorption. The lipid content in pharmaceutical-exposed granules declined by approximately half over the dosing period, though the relative concentrations of other key biofilm components (proteins, alpha-, and beta-polysaccharides) did not change. Batch experiments were conducted to try to find an explanation for the desorption observed, but reduced solid phase pharmaceutical concentrations could not be linked with the presence of common wastewater constituents such as ammonia or phosphate. Sorption of all three compounds was modeled best by the Henry isotherm, indicating that, even at 150 μg/L, granules' sorption site coverage was incomplete. Altogether, this study demonstrates that simplified batch systems may not accurately represent the complex abiotic processes occurring in flow-through, biotic systems.

Keywords: Adsorption; Aerobic granular sludge; EPS; Lipids; Pharmaceuticals.

Figure 1.

Effluent concentrations of ammonia and phosphate from the control and test reactors. Control effluent ammonia concentrations are not plotted past day 40 for reasons discussed in Section 3.1. Influent concentrations of both ammonia and phosphate were approximately 60 mg/L. Note that effluent phosphate concentrations peak above this value due to the activity of phosphate accumulating organisms, which release phosphate during the anaerobic phase of SBR operation.

Figure 2.

Batch adsorption data fit to (A) Henry and (B) Freundlich isotherms. Error bars all indicate standard deviation of triplicate samples and at times are smaller than sample points. Error bars in the x-direction indicate the standard deviation of C_e values for triplicates; those in the y-direction indicate the standard deviation of q_e values.

Figure 3.

Pharmaceutical concentrations in the aqueous (top row) and solid phase (bottom row) over time. Error bars represent the standard deviation of triplicate samples and are present on days 34, 56, and 80 for aqueous samples and days 34 and 80 for solid samples; points on these days are averages. Data in this image were originally published in (Bodle et al., 2023) and are shown here with permission from the authors.

Figure 4.

EPS component distributions in control and test granules over time. Green indicates proteins (fluorescein isothiocyanate stain); blue, lipids (nile red); red, alpha-PS (concanavalin A); pink, beta-PS (calcofluor white). All scale bars are 200 μm.

Figure 5.

Sorbed masses of DCF, ERY, and GEM in batch tests versus time (based on a mass balance from aqueous measurements). Error bars indicate the standard deviation of triplicate samples and are at times smaller than sample points.