Characterization of commercial RO membranes for the concentration of ammonia converted to ammonium sulfate from anaerobic digesters

Commercially available thin film composite (TFC) polyamide membrane (GE AG) and asymmetric cellulose acetate membrane (GE CE) have been characterized using different concentrations of ammonium sulfate, up to 40 g·l−1, in order to determine their suitability for the concentration of ammonium as ammonium sulfate from anaerobic digester digestate. At the manufacturer specified pressures of 1550 kPa and 2900 kPa, the solvent permeances for the GE AG and the GE CE membranes were found to be 2.1 × 10−5 and 7.4 × 10−6 l·m−2·h−1·Pa−1, respectively. These permeances were practically independent of feed concentration. At the same time, both membranes showed excellent solute rejections, which decreased slightly with feed concentration, but were generally greater than 0.99. Consequently, the lower pressure GE AG membranes were considered to be more suitable for the separation of ammonium sulfate, and they were further tested at pressures exceeding the manufacturer specified value. These experiments revealed a slight decrease in solvent permeance for up to 17%, as well as a slight decrease in solute rejection, which was 0.97 or greater. Given these results, both membranes studied in this research are suitable for future investigations; first using multiion aqueous solutions and then pre-treated anaerobic digester digestate. © 2015 Elsevier B.V. All rights reserved.

The ultimate goal of the research presented in this paper is to remove ammonia from anaerobic digesters in order to enhance biogas production, and at the same time, to concentrate the removed ammonia, converted into a stable form of ammonium sulfate. The latter could be used as a liquid fertilizer. In the first phase of this research, the results of which are presented in this paper, two commercially available RO membranes, GE AG and GE CE membranes, were tested with aqueous solutions of ammonium sulfate at different feed concentrations and pressures. For the investigated range of feed concentrations (up to 40 g·l−1) both membranes showed excellent solute rejections of 0.99 or greater at the manufacturer specified pressures of 1550 kPa and 2900 kPa for the GE AG and GE CE membranes, respectively. On the other hand, the lower pressure GE AG membrane was three times more permeable than the higher pressure GE CE membrane (2.1 × 10− 5 l·m− 2·h− 1·Pa−1 vs. 7.4 × 10−6 l·m−2·h− 1·Pa− 1). While both membranes were susceptible to compaction at pressures exceeding those specified by the manufacturer, the compaction effects were reversible for the GE AG membrane, but not for the GE CE membrane. As a result, the GE AG membrane is recommended for future studies with multi-ion aqueous solutions. The experimental results obtained in this study were interpreted in terms of a basic solution-diffusion model, which was found to be not entirely adequate to explain the observed results. In particular, solute permeance decreased with feed concentration at a given feed pressure, while solute rejection decreased with increased feed pressure at a given feed concentration. This suggests that in addition to concentrationdriven diffusion, some solute was also transported convectively along with the solvent. Nomenclature Am membrane area [m−2] C concentration of ammonium sulfate used for the calculation of osmotic pressure [mol·m−3] Cf feed concentration of ammonium sulfate [g·l−1] Cp i J Ns R Rs T V P H2 O P ðNH4 Þ2 SO4 ΔP Δπ Φ permeate stream concentration of ammonium sulfate [g·l−1] number of ions produced during dissociation [dimensionless] experimentally determined flux of solvent [l·m−2·h−1] flux of solute, ammonium sulfate [g·m−2·h−1] universal gas constant [J·mol−1·K−1] solute rejection [dimensionless] temperature [K] experimentally determined permeate flow rate [l·h−1] permeance of solvent, water [l·m−2·h−1·Pa−1] permeance of solute, ammonium sulfate [l·m−2·h−1] pressure differential across the membrane [Pa] osmotic pressure differential across the membrane [Pa] osmotic coefficient [dimensionless] Acknowledgments This research project was conducted in collaboration with CHFour Biogas Ltd., and was financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) (EGP2 #469657-14) Engage program.