Nanomaterial Texture-Based Machine Learning of Ciprofloxacin Adsorption on Nanoporous Carbon.
Journal:
International journal of molecular sciences
PMID:
39519248
Abstract
Drug substances in water bodies and groundwater have become a significant threat to the surrounding environment. This study focuses on the ability of the nanoporous carbon materials to remove ciprofloxacin from aqueous solutions under specific experimental conditions and on the development of the mathematical model that would allow describing the molecular interactions of the adsorption process and calculating the adsorption capacity of the material. Thus, based on the adsorption measurements of the 87 carbon materials, it was found that, depending on the porosity and pore size distribution, adsorption capacity values varied between 55 and 495 mg g. For a more detailed analysis of the effects of different carbon textures and pores characteristics, a Quantitative nano-Structure-Property Relationship (QnSPR) was developed to describe and predict the ability of a nanoporous carbon material to remove ciprofloxacin from aqueous solutions. The adsorption capacity of potential nanoporous carbon-based adsorbents for the removal of ciprofloxacin was shown to be sufficiently accurately described by a three-parameter multi-linear QnSPR equation (R = 0.70). This description was achieved only with parameters describing the texture of the carbon material such as specific surface area () and pore size fractions of 1.1-1.2 nm (VN21.1-1.2) and 3.3-3.4 nm (VN23.3-3.4) for pores.