Sorption of Sulfonamide Pharmaceutical Antibiotics on Whole Soils and Particle‐Size Fractions

ABSTRACT Residues of pharmaceutical antibiotics are found in the environment, whose fate and effects are governed by sorption. Thus, the extent and mechanisms of the soil sorption of p ‐aminobenzoic acid and five sulfonamide antibiotics (sulfanilamide, sulfadimidine, sulfadiazine, sulfadimethoxine, and sulfapyridine) were investigated using topsoils of fertilized and unfertilized Chernozem and their organic–mineral particle‐size fractions. Freundlich adsorption coefficients ( K f ) ranged from 0.5 to 6.5. Adsorption increased with aromaticity and electronegativity of functional groups attached to the sulfonyl–phenylamine core. Adsorption to soil and particle‐size fractions increased in the sequence: coarse silt < whole soil < medium silt < sand < clay < fine silt and was influenced by pH. Sorption nonlinearity (1/ n ≤ 0.76) indicated specific interactions with functional groups of soil organic matter (SOM). Phenolic and carboxylic groups, N‐heterocyclic compounds, and lignin decomposition products were tentatively assigned as preferred binding sites using statistical analysis of pyrolysis–mass spectra and adsorption coefficients. Adsorption of sulfonamides to mineral soil colloids was weaker and resulted in a stronger desorption from clay‐size fractions. Moreover, steric accessibility of organic–mineral complexes in clay‐size fractions was significantly reduced. With a quantitative structure–property relationship (QSPR) model, combining the organic carbon concentration, the sulfonamides' chromatographic capacity factor ( k ′), and nondissociated species concentration ( CF a ), distribution coefficients ( K d ) were estimated with a cross‐validated regression coefficient Q 2 = 0.71. Modeling and molecular mechanics calculations of antibiotic–SOM complexes showed preferred site‐specific sorption via hydrogen bonds and van der Waals interactions. Distinct chemical structural properties, such as aromaticity and van der Waals volumes, correlated with the sorption data.