Cation competition in ammonia recovery from reject water through electrodialysis and bipolar membranes

Abstract

Nutrient recovery has lately been a concerning topic regarding the environmental friendliness of it and the high availability of technologies. Ammonia is one of the main compounds in reject water that could be recovered and utilized further in the agricultural sector. Several methods have been found, including conventional electrodialysis, in which anion and cation exchange membranes are being used and, with the application of electrical current, there is production of clean and desalinated water, creating at the same time a concentrated solution. As a further evolution of electrodialysis, bipolar membranes could be added in the configuration, leading to acid and base production. However, ammonium is not the only cation included in reject water, but also Na+, K+, Mg2+ and Ca2+ are present and affect the overall performance electrodialysis. Thus, the competition between the cations needs to be investigated further regarding the operational parameters of each configuration.
This study investigated the cation competition in electrodialysis and bipolar membrane configuration regarding the ammonia removal efficiency and the overall energy consumption. The research questions were focused on the effect of enriched solutions with cations on ED and BPC to the efficiency parameters, to the impact of cation composition in the feed solution when NH4+, Na+, K+, Mg2+ and Ca2+ are included in an ED and finally, the effect of municipal reject water cation molar ratios in a combined ED and BPC configuration. The experiments included batch mode systems, with several mass and molar ratios of NH4+ applied, the above-mentioned parameters were measured. More specifically, BPC and ED configurations were tested with mass ratios of other cations in an enriched NH4+ solution, while molar ratios were tested in case of an ED configuration with NH4+, Na+, K+, Mg2+ and Ca2+ be present in the feed solution. Finally, the two configurations were tested in a sequence batch, with ED to be the pretreatment step and BPC the final stage. The phenomena that were also investigated were proton production from bipolar membranes and EC pattern on the diluate solution in this case.
In ED removal efficiency was presented as a linear curve on time while in BPC the same value took a logarithmic trend, which is attributed to proton production and finally competition. During BPC operation, there was constant production of H+ through water dissociation that led to the acidic environment in the diluate solution but also to stabilization of EC when H+ presence was dominant. In addition, in molar ratio experiments with the application of ED, removal efficiency was higher for more challenging reject waters compositions such as molar ratios between 0.30 and 0.60. Considering 75% removal efficiency as an effective case, percent demineralization was also calculated. For removal efficiency below the effective case, percent demineralization presented a minimum for molar ratio of 0.60, while for higher removal efficiency the overall trend was slightly different, having a more exponential shape. Finally, energy consumption in molar ratio experiments, for removal efficiency of 75% presented a gradual decreasing linear trend with the increase of molar ratio.
Based on the results occurred in batch experiments, a sequence batch of ED to concentrate the feed solution was established, by applying the more challenging molar ratios of 0.30, 0.45 and 0.60 and the concentrate was then fed to a BPC to explore the proton effect in a concentrated solution. The percent demineralization and removal efficiency remained stable during the experimental phase while transport number had a notable increase with the increase of molar ratio, remaining approximately the same in every individual batch. Moreover, energy consumption had an important increase with the decrease of molar ratio due to the high membrane resistance and the observed scaling effect.