The San Joaquin Valley has the highest agricultural output in America, but its future is in jeopardy. The salinity of its already shallow water table is increasing to the point that areas of land must be retired from use. By using desalination, irrigable water is produced and the volume of brine is decreased. However, the high salt content of the feed water can result in mineral surface scaling on the desalination membranes if it reaches saturated concentrations, which decrease plant productivity. Though, by adding antiscalant compounds, this threshold can be artificially lifted—twice as much for gypsum, and as much as sixty times more for barite. McCool et. al performed tests on water samples at each location to determine the concentration of these salts, and the effect that these concentrations would have on a full-scale desalination plant. The authors conclude that while desalination plants are theoretically feasible in the San Joaquin Valley, they must be carefully tailored to each location and regularly monitored.—Erin Partlan
McCool, B., Rahardianto, A., Faria, J., Kovac, K., Lara, D., Cohen, Y., 2010. Feasibility of reverse osmosis desalination of brackish agricultural drainage water in the San Joaquin Valley, Desalination 261, 240—250.
McCool et al. used two approaches to estimate the feasibility of desalination plants at five locations in the San Joaquin Valley. First, water samples were taken and tested to determine the theoretical feasibility based on known saturation values of various salts. The saturation limit is dependent on the fraction of water that is attempted to be removed through desalination as well as the pH of the feed water. The authors also used software to compute the theoretical recovery limits at each location based on the water samples. Secondly, other recent work has preliminarily identified that certain combinations of salts can lead to lowered saturation points. For this reason, bench-scale plate-and-frame reverse osmosis (PFRO) was performed for each location.
From the water samples, the various sites showed great spatial and temporal variability as salinity could differ by up to an order of magnitude between sites and over time. It was found that calcite was above saturation and that gypsum was near saturation at each site, though again, each varied greatly throughout the year. Since the feed water salt concentrations greatly determine specific design aspects for a desalination plant, it is necessary to tailor each plant for a particular location as there is too much variability in the San Joaquin Valley for one design to work throughout. In addition, traditional reverse osmosis plants are not typically able to handle large variations in salinity and the temporal changes experienced in the San Joaquin Valley may be problematic.
In using software to compute the theoretical recovery limits, highs and lows from the water samples were used. Also, the saturation limits used in modeling were slightly different from actual data, in particular for calcite as it was assumed that the concentration of this salt could be negated with changes to the pH of the feed water. Without this adjustment, calcite would be the limiting factor to the productivity of a desalination plant. By shifting the pH down to 6.0, gypsum becomes the limiting scalant. However, the recovery rates are still very limited. The authors suggest that feed-back process control will be required for any plant in the San Joaquin Valley.
In bench-scale testing, the reverse osmosis process was utilized for each location and under various conditions—at natural pH, at a lowered pH, and with or without the addition of antiscalants. In some tests (reduced pH with the addition of antiscalants), no scaling was observed and the decrease in outflow over time is consistent with known changes due to compacting of the membrane. Without the antiscalants, however, the decrease in outflow was much greater with visual scaling on the membrane at both natural and reduced pH. In fact, scaling occurred at a greater magnitude at the reduced pH, speculated to be a result of decreased bicarbonate, which is a large factor in preventing gypsum scaling. Again, the authors conclude that while desalination plants are theoretically feasible in the San Joaquin Valley, they must be carefully tailored to each location and regularly monitored.