|M.Sc Student||Ophek Liron|
|Subject||Improving Energy Efficiency and Simulation Accuracy of a|
Single Step SWRO Process Combining Boron
and TDS Rejection
|Department||Department of Civil and Environmental Engineering||Supervisor||Professor Ori Lahav|
|Full Thesis text|
The first part of this work focused on a new operational approach, which has the potential to substantially cut down on the energy and cost demand associated with seawater reverse osmosis desalination, without changing the currently-installed infrastructure. The approach comprises acidification/decarbonation of the feed seawater followed by high-pH single RO pass using high-flux membranes. Since the limitation imposed by CaCO3(s) precipitation is overcome, the recovery ratio can be significantly increased. Results obtained from operating a pilot RO system with newly available low-energy RO membranes revealed that following acidification and decarbonation a recovery of 56% could be practically attained along with effluent TDS and boron concentrations of 375 and 0.3 mg/l, respectively (feed water pH adjusted to pH9.53). The specific energy consumption of the latter scenario was calculated to be ~10% lower than that associated with "conventional" SWRO operation. Two further scenarios were theoretically considered, under which the limiting operational parameter became Mg(OH)2(s) and BaSO4(s) precipitation. It was concluded that despite the fact that higher recovery ratios could be obtained, the high pressure required in these scenarios made them less appealing from both the SEC and cost standpoints. The normalized cost of the suggested approach was found to be ~0.07 $/m3 cheaper than the currently-used SWRO approach for obtaining product water characterized by TDS<500 and B<0.5 mg/l.
The second part of the work focused on the effect of feed water temperature variations on reverse osmosis membrane filtration performance. Temperature dependent SWRO process design is still limited by inadequate data, particularly for boric acid. In this work generic procedures were evaluated for determining membrane-specific correlations for permeability constants as a function of temperature, and the constants were assessed thereafter within the Solution-Diffusion RO simulation model. The developed procedure comprised a calibration step and the extraction of boron, salt and water permeabilities across the desired temperature range. It is shown that each of the permeability constants has to be calculated using an appropriate mathematical method (all in accordance with solution diffusion and film layer models). For the seawater membrane tested in the work (SWC5-4040, Hydranautics) an exponential type equation described best the temperature dependence of the permeability of the salt (ions), while the temperature effect on boric acid permeability followed a linear correlation. Once extracted and fitted with mathematical terms the correlations were embedded in a membrane transport model (WATRO). Validation experiments were conducted at 18, 24 and 31 oC and compared to WATRO simulations in which three temperature-dependent literature permeability terms were used. The updated WATRO simulation results matched the boron permeate experimental results much better, particularly at the edges of the temperature range. With respect to TDS all the examined temperature terms produced practically identical results. Membrane-specific temperature permeability correlations, particularly with respect to PB(OH)3, can contribute to optimal design and operation and reduce both energy requirements and water product cost.