In this paper, the removal effect of reverse osmosis (RO) on three common pharmaceuticals and personal care products (PPCPs), including ibuprofen (IBU), carbamazepine (CBZ), and triclosan (TCS), were compared under different process conditions, and the removal rate of PPCPs, membrane flux, and PPCPs membrane adsorption capacity were analyzed. The removal rate increased with the increase of the influent concentration and pre-membrane pressure, while pH influenced the removal effect of different PPCPs by affecting the electrostatic interaction between pollutants and membranes. It was also found that the dynamic adsorption of PPCPs on RO membranes under different conditions complied with the pseudo-first-order reaction kinetic adsorption model. The maximum stable adsorption capacity and the adsorption rate of PPCPs on membranes under various conditions were simulated based on the model. Moreover, through factorial design, the removal rates of RO on IBU, CBZ, and TCS could reach 98.93%, 97.47%, and 99.01%, respectively, under the optimal conditions (with an influent concentration of 500 μg/L, pre-membrane pressure of 16 bar and pH = 10). By optimizing the process of removing PPCPs with the RO membrane method, the optimal process conditions of removing IBU, CBZ, and TCS with the RO membrane method were obtained, which provided reference conditions and data support for the practical application of removing PPCPs with the RO membrane method.

Membrane biofilm reactors (MBfRs) have attracted more and more attention in the field of wastewater treatment due to their advantages of high mass transfer efficiency and low-carbon emissions. There are many factors affecting their nitrogen removal abilities, such as operation time, electron donor types, and operation modes. The operation time is directly related to the growth status of microorganisms, so it is very important to understand the effect of different operation times on microbial composition and community succession. In this study, two parallel H₂-based MBfRs were operated, and differences in microbial composition, community succession, and NO₃⁻-N removal efficiency were investigated on the 30th day and the 60th day of operation. The nitrogen removal efficiency of MBfRs with an operation time of 60 days was higher than that of MBfRs with an operation time of 30 days. Proteobacteria was the dominant phylum in both MBfRs; however, the composition of the microbial community was quite different. At the class level, the community composition of Proteobacteria was similar between the two MBfRs. Alphaproteobacteria was the dominant class in MBfR, and Betaproteobacteria and Gammaproteobacteria were also in high proportion. Combined with the analysis of microbial relative abundance and concentration, the similarity of microbial distribution in the MBfRs was very low on day 30 and day 60, and the phylogenetic relationships of the top 50 dominant universal bacteria and Proteobacteria were different. Although the microbial concentration decreased with the extension of the operation time, the microbial abundance and diversity of specific functional microorganisms increased further. Therefore, the operation time had a significant effect on microbial composition and community succession.