Thermal technologies are not only alternative desalting methods but also can be used to treat RO concentrate. Thermal processes include VC evaporation, MD, and freeze concentration. Because the feedwater salinity
has no significant impact on the efficiency of thermal processes, thermal processes can be applied to a wide salinity range from brackish water, seawater to brine TDS greater than 300 g/L.
4.4.1 Vapor compression (VC)
In VC systems, the mechanical (mechanical VC or MVC) or thermal (thermal VC or TVC) compression of the vapor provides the heat for evaporation. The process compresses the vapor generated within the unit itself.The mechanical compressor is usually electrically or diesel driven. Thermal compression uses high-pressure steam. Compression raises the pressure and temperature of the vapor so that it can be returned to the evaporator and used as a heat source.
MVC evaporation is considered the most thermodynamically efficient process of thermal desalination processes . A compressor is the driving force for this heat transfer and provides the energy required for separating the solution and overcoming dynamic pressure losses and other irreversibilities (Fig. 10) . The vapor generated from the solution is pumped to the higher pressure level required on the condensing side.
In this system, the distillation step is carried out at close to ambient temperatures. The advantages of MVC include simple pretreatment, and more durable and less maintenance as compared to membrane process . During the last decade, intensive research and development have been carried out to improve the MVC process further by increasing the unit capacities and reducing energy. The MVC is typically limited in size to 3000m3/day (0.8 mgd) due to the constraints of the mechanical compressors. The MVC units with capacities up to 10,000 m3/day (2.5 mgd) are under development . TVC systems can be employed for significantly larger installations [20,000m3/day (5.3mgd)].
VC process is well established and is used for seawater desalination as well as treating RO concentrate (i.e. brine concentrator application) in a near-ZLD application [14,19,78,79]. For example, brine concentrators (VC evaporators operating with seed recycle) are used in Australia to treat RO concentrate from cooling tower blowdown to achieve ZLD in power plants. Scaling is still an issue in VC process, and another disadvantage of the thermal technology is high energy consumption.
Membrane distillation (MD) is an emerging separation process that combines simultaneous mass and heat transfer through a hydrophobic microporous membrane [15,80,81]. The driving force for mass transfer in the process is vapor pressure difference across the membrane. A feed solution at elevated temperature is in contact with one side of the membrane and colder water is in direct contact with the opposite side of the membrane (Fig. 11); it is mainly the temperature difference between the liquids and to some extent their solute concentration, which typically
Heated Saline Feed
Figure 11 Schematic of air-gap membrane distillation process.
results in vapor pressure depression . Significant water flux can be produced when the temperature difference is above 20 °C between the feed (e.g., 40 °C) and permeate (e.g., 20 °C) streams. By applying vacuum to the permeate side, the water flux can increase by up to 85% over traditional MD process . The main difference and potential advantage of MD over RO is that the former uses the feed's vapor pressure as the driving force to penetrate the membrane instead of the high hydraulic pressures necessary for RO. This eliminates the need for the high-pressure pump and reduces the fouling and scaling problems associated with the pressure-induced concentration gradients at the membrane surface.
During the treatment of a RO concentrate with high silica concentration, MD could reduce the volume of RO concentrate by 60%, achieving an overall water recovery of 90% through RO-MD . Scaling occurred on MD membrane surface at high recovery as determined by the saturation indices of mineral scalants. However, the scalants formed in treating RO concentrate did not clog membrane pores and could be removed almost completely by chemical cleaning .
The energy source for feed heating and/or for a vacuum system to sweep away the vapor may be low-grade thermal energy such as supplied by low-pressure steam, waste heat, as well as solar or geothermal energy. A variety of configurations can be used to induce the vapor through the membrane and to condense the penetrant gas. The common method is that the feedwater directly contacts the membrane. Condensation is typically achieved via four-process configurations: air-gap membrane distillation, direct-contact membrane distillation, sweep-gas membrane distillation, and vacuum membrane distillation. Even though MD is frequently cited as a promising desalination technology, no significant commercial operations of this technology exist at this time.
The freezing process is based on the natural phenomenon that occurs when ice forms in a saline solution: the resulting individual ice crystals are made up of essentially pure water . Because ice crystals have great regularity and symmetry, they cannot accommodate other atoms or molecules without very severe local strain, practically every solute in the water is rejected by the advancing surface of the growing ice crystal. The "hypertonic" solution at the surface slows down the freezing part of the liquid-solid molecular exchange, by decreasing the availability of water molecules. The result is a lowering of the temperature at which the freezing and melting processes balance; that is, a depression of the freezing point . For example, the freezing point of seawater is - 1.9 °C . As the concentration of the hypertonic solution increases, the freezing point is continually lowered. Freeze desalination depends on the insolubility of salts in ice crystals; the crystals can be separated from the ice-brine slurry, washed, and melted to yield freshwater.
Most of the freeze desalination technologies fall into three categories: direct, indirect, or through the use of a secondary refrigerant. Indirect freeze concentration plants have been used to concentrate beers, coffees, and various juices in many countries around the world . In general, freeze desalting has five basic operations: (1) precooling the feed stream; (2) partial freezing of the feed stream; (3) separation of the ice-brine mixture; (4) melting the associated ice; and (5) heat rejection. Most freezing processes pump the heat removed by crystallization to the melting ice. This is the lowest available temperature lift, and therefore the least amount of work. Heat rejection is required by all freeze desalting technologies; however, because crystallization occurs below ambient temperature, internal heat must be pumped away from the system through a secondary refrigerant. This requires additional work that other desalination processes do not encounter. If this additional work was not required, freezing would probably be the lowest energy intensive desalination process . High capital and energy requirements that are only marginally competitive with other processes have forced the freeze process to be discontinued. •••-
Fig. 12 illustrates a representative change in both the ice and concentrate TDS over time  with freeze desalting a RO concentrate. In all, six experiments were conducted under the same conditions to provide preliminary indications of the system's flux rate. In order to create an equal measure of production, the volumetric change in concentrate over time was calculated as gallon per square foot of freezing surface area per minute -similar to that for RO. The flux of the freezing process was calculated at an average 13 gfd, comparable to that of typical RO system. However, as seen in Fig. 12, salt rejection ranged from 66% to 73%; significantly lower than that of most RO processes. In addition, while a 1.7-fold increase in concentrate TDS was observed over time, product-water TDS showed a concomitant increasing TDS. For all experiments, product-ice TDS increased from 1320 to 2150 mg/L at experiment end. This finding is significant for two reasons: (1) TDS of the ice removed from the system was never below 500 mg/L - a typical guideline for reclaimed water and (2) the removal of TDS was variable anc tracked the concentrate TDS. Therefore, as recovery of the product increases, so wouIc its salt content increase. Problems associated with freeze desalting include incomplete separation of salts from the ice slurry, fouling of the freezing surface (i.e., ice platting onto the freezing surfaces), and handling of the ice residuals. -•
TDS Rejection a-a-A-a
Concentrate (feed) TDS Product (ice) TDS
Concentrate (feed) TDS Product (ice) TDS
0 50 100 150 200 250 300 350 400 450 Time (min)
Figure 12 Mechanical freeze desalting performance data from 76 L batch experiment. Product ice double-rinsed with deionized water prior to sampling. Flux = 6.1 x10~6 m/s .
Was this article helpful?