A comparison of membrane softening on three South Florida groundwaters

Abstract

This paper compares and contrasts the performance of one commercially-available softening membrane on surficial aquifer water at three different South Florida locations and the performance of two additional commercially-available softening membranes at one of the three South Florida locations. Data from pilot studies performed at each location along with established membrane performance equations were also used to develop a model which predicts the performance of membranes under differing operating conditions. In summary, the main findings of this investigation are: 1) The permeate water mass transfer coefficient was essentially the same for all three pilot locations and for the three membranes at the same location. This means that the pumping energy required to produce a certain permeate flow per a given amount of membrane area is similar for a full-scale membrane plant at each location. 2) The permeate salt concentration is a variable dependent on source water concentration and the salt mass transfer coefficient. For calcium, the mass transfer coefficient was different for only one of the membranes tested, so for the other two membranes the differences in source water calcium concentration governed the permeate calcium concentration. Other salts, such as chloride, had significantly different mass transfer coefficients, so the source water concentrations and the mass transfer coefficient both impacted the permeate salt concentration greatly. Also, as the permeate water flux is reduced for all the membranes, the permeate salt concentration increases since the salt flux does not vary with the variation in permeate water flux. 3) Membrane performance at different operating conditions than the piloted condition can be predicted using the established membrane performance equations, giving the full-scale membrane facility designer flexibility in the design of the membrane process.