The role of intraparticle diffusion path length during electro-assisted regeneration of ion exchange resins: implications for selective adsorbent design and reverse osmosis pretreatment
2021
Abstract Inorganic membrane scaling is a major cause of process failure for reverse osmosis (RO). Pretreatment using selective ion exchangers can reduce membrane scaling by removing scale-forming ions such as calcium (Ca2+) and phosphate (P). Ca- and P-selective ion exchangers also enable calcium phosphate recovery via precipitation from spent regenerants enriched with both ions. However, conventional regeneration approaches require substantial chemical and energy inputs. In this study, we designed a RO pretreatment process using electro-assisted regeneration of a column train containing a hybrid anion exchanger (HAIX) and a weak acid cation exchanger (WAC) to selectively remove > 90% Ca2+ and P from synthetic secondary effluent. HAIX and WAC were regenerated using pH 11 catholyte and pH 3 anolyte produced via electrochemical water splitting. HAIX contains ferric oxide nanoparticles (FeOnp) in a parent strong base anion exchanger (SBA). By comparing P regeneration efficiency of HAIX and SBA, we identified that FeOnp dopants reduced intraparticle diffusivity by 85% from SBA to HAIX. However, HAIX exhibited faster column regeneration kinetics than SBA, potentially due to the shortened intraparticle diffusion path length caused by the surface-distributed P in saturated HAIX, which overcompensated for low intraparticle diffusivity. Regenerability enhancement caused by the reduced intraparticle diffusion path length was also confirmed by comparing WAC with shallow-shell WAC (SSWAC), which has surface-distributed carboxylate functional groups. We anticipate that this study will inform selective adsorbent design by identifying strategies to improve regenerability, facilitate process intensification by replacing chemical inputs with electricity, and enhance resource recovery from wastewater.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
37
References
5
Citations
NaN
KQI