University of California Water Resources Center
Membrane Desalination of Agricultural Drainage Water: Water Recovery Enhancement and Brine Minimization
- Author(s): Cohen, Yoram
- et al.
Salinity of brackish groundwater in the San Joaquin Valley (SJV) is typically in the range of about 3,000 - 30,000 mg/L total dissolved solids (TDS) In recent years, there has been a growing interest in the potential use of membrane desalination technology for reducing the salinity SJV brackish water. Membrane desalination of SJV brackish water would have to be carried out at relatively high water recovery in order to reduce the volume of generated RO concentrate. However, at high water recoveries the concentration of mineral salt ions on the feed-side of the membrane may increase to levels that exceed the solubility limits of various sparingly water-soluble mineral salts (e.g., calcium sulfate, calcium carbonate, and barium sulfate). The ensuing crystallization of these minerals results in scale build-up that leads to permeate flux decline, shortening of membrane life, and thus a reduction in process efficiency and increased operational cost. Therefore, process strategies must be designed to enhance product water recovery, while reducing the potential for mineral salt scaling. Accordingly, the principal objective of the present study was to evaluate the feasibility of RO desalting of SJV brackish water.
The present project focused on a systematic evaluation of: (a) the recovery limits for RO desalination of SJV AD water that are imposed by mineral salt scaling, and (b) the integration of accelerated precipitation (AP) of mineral salts with RO desalting to mitigate scaling and enable high RO recovery. AP treatment would serve to demineralize and desupersaturate the RO primary or secondary feed with respect to mineral salt scalants. In the first phase of the project, a systematic analysis based on multi-electrolyte thermodynamic solubility calculations was carried to determine the recovery limits due to mineral scaling. Subsequently, a detailed theoretical analysis and laboratory bench-scale studies were carried out with field water samples to assess mineral scaling propensity for a number of specific SJV water sources. In the second phase of the study, the integration of accelerated precipitation with RO desalting was investigated as a potential approach to lowering source water scaling propensity to enable enhanced water product water recovery.
Analysis of historical water quality data and of recently obtained water field samples, from various locations in the San Joaquin Valley, demonstrated a significant variability of water salinity and scaling propensity with respect to calcite, gypsum, barite, and silica. The above analysis and experimental RO scaling tests suggested that the expected range of product water recovery by RO desalting across the SJV can be in the range of 50%-70% for most of the sites, with the exception of the ERR site for which a much higher recovery was estimated (in excess of 90%). The integration of accelerated precipitation with RO desalting was shown to be technically feasible for the range of brackish water quality in the San Joaquin Valley. In this process, the concentrate from primary RO (PRO) desalting would be treated by accelerated precipitation softening (i.e., chemical demineralization) or desupersaturated to lower the scaling propensity of this stream, followed by secondary RO (SRO) desalting. Overall recovery of up to ~90%-95% could be achieved at an estimated cost of $0.56 -$0.98 per m3 product water, with the ACP process accounting for about 15%-25% of the overall water production cost.
The present study demonstrated provided a framework for assessing RO recovery limits and thus identifying potential hurdles that should be addressed in pilot studies and ultimately in the design and implementation of large-scale RO desalting processes. It is expected that, the methods developed in the present study for scale characterization, evaluation of accelerated precipitation effectiveness and RO process performance analysis will significantly advance the knowledge base needed to arrive at optimal design and deployment of future RO brackish water desalination strategies for the San Joaquin Valley.