The local molar salt and water fluxes Js,n and Jw,n transported into the diluate channel are modeled as in Fidaleo and Moresi :equation(10)Js,n=tsjD,nF+LsCc,m,n−Cd,m,nequation(11)Jw,n=twjD,nF−Lwπc,m,n−πd,m,nwhere Ls is the overall salt permeability (in m/s), Lw is the overall water permeability (in mol/bar-m2-s), and πm,n is the local ACT-132577 at the membrane surface . Finally, the gross power density PD,g is given by:equation(12)PD,g=?stack2RLwlwhere w is the stack width and l is the stack length. In our analysis, the load resistance RL is continuously optimized with respect to the gross power density .
2.2. Net power density model
2.3. Modified cost model
When back-end blending as opposed to front-end blending is implemented, additional cost reductions are realized from recycling a portion of the already-pretreated diluate stream. To assess the impact of back-end blending on the levelized cost of electricity, we model the LCOE as in Weiner et al.