Salt gradient lakes, which exhibit an increase in temperature with depth, occur naturally. A salt gradient solar pond is a body of saline water in which the salt concentration increases with depth, from a very low value at the surface to near saturation at the depth of usually 1–2 m (Tabor, 1981). The density gradient inhibits free convection, and the result is that solar radiation is trapped in the lower region. Solar ponds are wide-surfaced collectors in which the basic concept is to heat a large pond or lake of water in such a way as to suppress the heat losses that would occur if less dense heated water is allowed to rise to the surface of the pond and lose energy to the environment by convection and radiation (Sencan et al., 2007). As shown in Figure 10.16, this objective can be accomplished if a stagnant, highly transparent insulating zone is created in the upper part of the pond to contain the hot fluid in the lower part of the pond. In a non-convecting solar pond, part of the incident insolation is absorbed and converted to heat, which is stored in the lower regions of the pond. Solar ponds are both solar energy collectors and heat stores. A salt-gradient non-convecting solar pond consists of three zones (Norton, 1992; Hassairi et al., 2001):
1. Upper convecting zone (UCZ). This is a zone, typically 0.3 m thick, of almost constant low salinity, which is at close to ambient temperature. The UCZ is the result of evaporation, wind-induced mixing, and surface flushing. Usually this layer is kept as thin as possible by the use of wave-suppressing surface meshes or by placing wind-breaks near the pond.
2. Non-convecting zone (NCZ). In this zone, both salinity and temperature increase with depth. The vertical salt gradient in the NCZ inhibits convection and thus gives the thermal insulation effect. The temperature gradient is formed due to the absorption of solar insolation at the pond base.
3. Lower convecting zone (LCZ). This is a zone of almost constant, relatively high salinity (typically 20% by weight) at a high temperature. Heat is stored in the LCZ, which should be sized to supply energy continuously throughout the year. As the depth increases, the thermal capacity increases and annual variations of temperature decrease. Large depths, however, increase the required initial capital expenditure and exhibit longer start-up times.
In solar ponds, it is necessary to suppress natural convection. Many techniques have been considered for this purpose; the most common method used is a salt stratification. Salinity increases with depth in the NCZ until the LCZ is reached (see Figure 10.16). In the LCZ, solar radiation heats the high-salinity water, but because of its high relative density, hot, salty water cannot rise into the lower salinity layers, thus the heat is trapped and stored for use.
FIGURE 10.16 Schematic vertical section through a salt-gradient solar pond.
Chemically stable salts, as well as any natural brine, can be used in salt gradient solar ponds. A selected salt must be safe to handle, non-toxic, relatively cheap, and readily available, and its solubility should be temperature-dependent and should not reduce significantly the insolation transmission characteristics of water. Sodium and magnesium chlorides, though satisfying most criteria, have solubilities that are modestly temperature-dependent (Norton, 1992). Due to its low cost, sodium chloride remains the most popularly used salt.
Inorganic dirt brought by the wind can enter the pond, but generally the dirt causes no problem as it settles at the bottom. Various species of freshwater and saltwater algae grow under the conditions of temperature and salt concentration that exist in a stratified solar pond. Algae growth is undesirable because it reduces solar transmissivity. Most of these algae species are introduced by rainwater and air-borne dust. An effective way to prevent algae formation is to add copper sulfate at a concentration of about 1.5 mg/l.
The thermal efficiency of a solar pond depends on the stability of its salt gradient. The pond cannot function without the proper maintenance of the stratification. The salt gradient is maintained by:
1. Control of the overall salinity difference among the three convecting layers;
2. Reducing internal convection currents in the NCZ; and
3. Limiting the growth of the UCZ.
Additionally, the efficiency of a solar pond is limited by some intrinsic physical properties. The first thing that reduces the efficiency is the reflection losses at the surface of the pond. After penetrating the surface, in the first few centimeters of water, the insolation is rapidly attenuated by about 50%, since half the solar spectrum is in the infrared region, for which water is almost opaque. This is the reason the shallow ponds yield negligible temperature rise. Practical efficiency values for ponds of 1 m depth are of the order of 15–25% (Tabor, 1981). These figures are lower than for flat-plate collectors, but the lower cost, the built-in storage capability, and collection over large areas make solar ponds attractive under suitable environmental conditions. Generally, because the economics of solar ponds improve with size, large ponds are preferred.
10.7.1 Practical design considerations
In evaluating a particular site for a solar pond application, several factors need to be considered. The main ones are:
1. Since solar ponds are horizontal solar collectors, sites should be at low to moderate northern and southern latitudes, i.e., latitudes between ±40°.
2. Each potential site has to be evaluated for its geological soil characteristics because the underlying earth structure should be free of stresses, strains, and fissures, which could cause differential thermal expansions, resulting in earth movement if the structure is not homogeneous.
3. Since the thermal conductivity of soil increases greatly with moisture content, the water table of the prospective site should be at least a few meters below the bottom of the pond to minimize heat losses.
4. A source of cheap salt- or seawater should be available locally.
5. The site should be fairly flat to avoid moving large quantities of earth.
6. A cheap source of water must be available to make up for evaporation losses.
Generally, two types of leakages occur in solar ponds: leakage of the saline water from the bottom of the pond and leakage of heat into the ground. The loss of hot saline water is the most serious, since it results in the loss of heat and salt. Additionally, the solar pond must not pollute the aquifers, and any continuous drain of hot water lowers the pond’s storage capacity and effectiveness. Therefore, the selection of a liner for the pond is very important. Although it is possible to build a soil liner by compacting clay, in most cases, the permeability is unacceptable because the resultant loss of hot fluid to the soil increases thermal losses, requires replenishment of salt and water, and may present an environmental problem. All ponds constructed up to today have a plastic or elastomer liner, which is a reinforced polymer material 0.75–1.25 mm in thickness. The lining represents a considerable but not critical cost item that should be considered in cost analysis.
Evaporation is caused by insolation and wind action. The evaporation rate depends on the temperature of the UCZ and the humidity above the pond’s surface. The higher the temperature of the water in the UCZ and the lower the humidity of ambient air, the greater is the evaporation rate. Excessive evaporation results in a growth of the UCZ downward into the NCZ (Onwubiko, 1984). Evaporation can be counter-balanced by surface water washing, called surface flushing, which could compensate for evaporated water as well as reduce the temperature of the pond’s surface, especially during periods of high insolation. In fact, surface flushing is an essential process in maintaining the pond’s salt gradient. Its effect on the growth of UCZ is reduced if the velocity of the surface washing water is small. Surface temperature fluctuations will result in heat being transferred upward through the UCZ by convection, especially at night, and downward by conduction. The thickness of the UCZ varies with the intensity of the incident radiation (Norton, 1992).
Another method to reduce the evaporation rate is by reducing the wind velocity over the water’s surface by using windbreaks. The sheer forces of wind on a large area of water generate waves and surface drift. The kinetic energy transferred to the water is consumed partly by viscous losses and partly by mixing of the top surface water with the somewhat denser water just below the surface. When light to moderate winds exist, evaporation can be the dominant mechanism in a surface layer mixing. Under strong winds, however, evaporation becomes of secondary importance because wind-induced mixing can contribute significantly to the deepening of the UCZ (Elata and Levien, 1966). Another effect of wind is that it induces horizontal currents near the top surface of the pond, thus increasing convection in the UCZ. Wind mixing has been reduced by floating devices such as plastic pipes and plastic grids.
The pond is filled in layered sections, one after the other, each layer having a different salt concentration, as indicated above. Usually, these layers are built from the bottom upward, with the densest bottom layer filled first and subsequent lighter layers floated on the denser layer. Shortly after the stepwise filling process, the pond gradient smoothes itself, due to the diffusion and kinetic energy of liquid flow injected into the pond during the filling process (Tabor, 1981).
Salt slowly diffuses upward at an annual average rate of about 20 kg/m2 as a result of its concentration gradient (Norton, 1992). The diffusion rate depends on the ambient environmental conditions, type of salt, and temperature gradient. A combination of surface washing with freshwater and the injection of adequate density brines at the bottom of the pond are usually sufficient to maintain an almost stationary gradient.
A solar pond is usually constructed by flattening the site and building a retaining wall around the perimeter of the pond, not by digging out the earth; thus only a small fraction of the earth is moved, which reduces costs drastically. To avoid the use of wall supports, the earth walls thus built are tapered with a slope of 1 in 3, which gives an inclination of about 20° (Tabor, 1981). Preferred sites for solar ponds are near the sea, where saline water is locally available; otherwise, a large quantity of salt needs to be purchased. Sufficient quantities of low-salinity or fresh water are also required for the UCZ and for surface washing.
Thermal efficiency is defined as the ratio of total heat removed from the solar pond to the total amount of solar radiation which has fallen on the surface of the solar pond during a defined period. For this definition to be meaningful, the considered period should be long enough such that the stored or lost heat from the pond can be ignored compared with the received solar energy during that period. Numerous theoretical and experimental investigations have been carried out to determine the thermal efficiency of solar ponds. A transient computational analysis on the performance of solar ponds under varying conditions carried out by Wang and Akbarzadeh (1982) showed that a solar pond could provide heat at an efficiency of 15% and at an average temperature of about 87 °C or at 20% efficiency and an average temperature not higher than 65 °C.
10.7.2 Methods of heat extraction
Basically, two methods are used to extract heat accumulated to the bottom of the solar pond. The first uses a heat exchanger in the LCZ, which is in the form of a series of parallel pipes; the second is to use an external heat exchanger, which is supplied with hot saline water from the LCZ and returns the fluid to the other end of the pond at the same layer. For this purpose, horizontal nozzles that keep the velocity of efflux low are usually used. The same nozzles can also be used for filling the pond.
Heat has been successfully extracted from the lower convective zone of solar ponds for industrial process heating, space heating, and power generation (Andrews and Akbarzadeh, 2002; Rabl and Nielsen, 1975; Tabor and Doron, 1986).
The conventional method of heat extraction from a solar pond is to draw the heat from the LCZ only. This can be done using a heat exchanger located in the LCZ. A heat transfer fluid circulates in a closed cycle through the internal heat exchanger and transfers its thermal energy through an external heat exchanger. Figure 10.17 shows such a system for a heating application. The heat exchanger is usually a series of polyethylene tubes passing through the pond’s LCZ connected to a manifold of larger diameter pipes outside the pond. This method was applied to the solar pond at Pyramid Hill, Australia. Heat extraction can also be performed by pumping the hot brine from the top of the LCZ through an external heat exchanger and then returning the brine at a reduced temperature to the bottom of the LCZ (see Figure 10.18). The velocity of the brine being pumped needs to be regulated to prevent erosion of the gradient layer. In the salt-gradient solar pond at El Paso, USA, the heat is extracted from the pond using an external heat exchanger (Lu et al., 2004).
FIGURE 10.17 Heat extraction from a solar pond with an internal heat exchanger in the LCZ.
FIGURE 10.18 Heat extraction from a solar pond with an external heat exchanger using hot brine from the LCZ.
An alternative way of extracting heat from solar ponds was investigated theoretically with the aim of improving the overall energy efficiency by Andrews and Akbarzadeh (2005). In this method, heat is extracted from the NCZ as well as, or instead of, the LCZ. The theoretical analysis showed that the extraction of heat from the gradient layer can decrease heat losses to the surface and hence results in an increase in the overall energy efficiency.
A good survey of heat extraction methods from salinity-gradient solar ponds is given by Leblanc et al. (2011).
10.7.3 Transmission estimation
As was seen previously, when solar radiation falls on the surface of a solar pond, part of it is reflected at the water surface and part is absorbed at the bottom. Since water is a spectrally selective absorber, only shorter wavelengths reach the bottom of the pond. Because the absorption phenomena differ widely with wavelength, the absorption of solar energy in solutions with inorganic salts used in solar ponds can be represented by the sum of four exponential terms. The transmittance of water at depth x, τ(x), can be related to x by (Nielsen, 1976):
(10.20)
where the coefficients αi and bi are given in Table 10.5.
Table 10.5
Coefficients for Eq. (10.11)
It should be noted that Eq. (10.20) does not include the transmission in the infrared part of the spectrum (λ > 1.2 μm, row 5 in Table 10.5), since this part is of no interest in solar pond analysis. Additionally, detailed analysis of heat transfer in a solar pond is very complex; it includes effects of volumetric absorption and variation of conductivity and density with salinity. The interested reader is referred to the articles by Tsilingiris (1994) and Angeli et al. (2006).
EXAMPLE 10.3
Find the transmittance of a solar pond for a depth of 0.6 m.
Solution
From Eq. (10.20),
For more accuracy the fifth term in Table 10.5 can be considered which gives the same answer as the last term is equal to 5.25 × 10−60. This is why this last term is usually ignored even at shallow depths, except perhaps at the very few millimeters.
10.7.4 Experimental solar ponds
Two important experimental solar pond applications are the El Paso solar pond in Texas, USA, and the Pyramid Hill solar pond, in Victoria, Australia.
The El Paso solar pond is a 3000 m2 research, development and demonstration project operated by the University of Texas at El Paso. This project, initiated in 1983, has been operated intermittently since 1985 through the end of 2003, and became one of the longest operational solar pond projects in the world.
Through 16 years of operation and research, the El Paso solar pond project has provided valuable experiences and demonstrated various applications including desalination, waste brine management, industrial process heat production, and electricity generation (Leblanc et al., 2011).
The El Paso solar pond has a depth of about 3.25 m. The UCZ, NCZ, and LCZ are approximately 0.7 m, 1.2 m, and 1.35 m, respectively. The pond uses an aqueous solution of predominately sodium chloride (NaCl). The typical operating temperature of the pond ranged from 70 °C in winter to 90 °C in early fall. The highest temperature observed was 93 °C, and the maximum temperature difference between the LCZ and UCZ was above 70 °C.
The El Paso solar pond is a research, development and demonstration project operated by the University of Texas at El Paso and funded by the US Bureau of Reclamation and the State of Texas. This project is located on the property of Bruce Foods, Inc., a food canning company, and it was the first in the world to deliver industrial process heat to a commercial manufacturer, the first solar pond electric power generation facility in the US, and the nation’s first experimental solar-pond-powered desalination facility. The El Paso solar pond ceased its operation and was decommissioned at the end of 2003. In order to make salinity-gradient solar pond technology more reliable, productive and economic, a series of techniques have been developed. These techniques include: automated instrumentation system for solar pond monitoring; stability analysis strategy and high-temperature (60–90 °C) gradient maintenance methods; scanning injection technique for improved salinity gradient construction and maintenance; new liner technology; and improved heat extraction system (Lu et al., 2004). Additionally, different lining systems had been used, including flexible membrane liners and compacted clay/plastic buried liners. During the 16 operational years, the El Paso solar pond had experienced liner failures (Robbins et al., 1995; Lu and Swift, 1996), and three different liners have been used. Details of the liners used and the causes of failure after some years of operation are outlined by Leblanc et al. (2011).
In February 2000, Royal Melbourne Institute of Technology (RMIT) University in partnership with two Australian companies began the “Pyramid Hill Solar Pond Project” (Andrews and Akbarzadeh, 2002) to demonstrate and commercialize a solar pond system as an innovative, cost-effective method of capturing and storing solar energy for a range of applications, including heating, electricity generation and combined heat and power. The project was made possible through a grant under the Renewable Energy Commercialization program from the Australian Greenhouse Office. The 3000 m2 salt-gradient solar pond was constructed at Pyramid Salt’s facility in northern Victoria, Australia, hence its name. The initial stage of the project has focussed on industrial process heating, providing heat for use in high-grade salt production. The heat was used also for aquaculture, specially producing brine shrimps for a stock feed. Supply of heat for commercial salt production began in June 2001. The solar pond facility was also used to demonstrate inland desalting in December 2006.
The pond was designed with a depth of 2.3 m. The LCZ was designed to be 0.8 thick, the NCZ 1.2 m thick, and the UCZ 0.3 m thick. Pyramid Salt is an established commercial producer of salt from saline groundwater that is pumped to the surface as part of a salinity mitigation scheme. The solar pond has been integrated into the scheme where the surface of the pond is flushed with saline groundwater (about 3% salinity) and the overflow is used in the salt production process. The pond was positioned about 200 m from Pyramid Hill’s salt production plant to minimize heat losses.
The pond was lined with a 1 mm thick Nylex Millennium polypropylene liner. This particular lining was selected for its ability to withstand saturated brine at temperatures up to 100 °C and its resistance to ultraviolet (UV) radiation.
10.7.5 Applications
Solar ponds can be used to provide energy for many different types of applications. The smaller ponds have been used mainly for space heating and cooling and domestic hot water production, whereas the larger ponds are proposed for industrial process heat, electric power generation, and desalination.
Solar ponds are very attractive for space heating and cooling and domestic hot water production because of their intrinsic storage capabilities. To increase the economic viability, large solar ponds can be used for district heating and cooling, and such a system can offer also seasonal storage. However, no such project has been undertaken so far. Cooling is achieved with the use of absorption chillers, which require heat energy to operate (see Chapter 6, Section 6.4.2). For this purpose, temperatures of about 90 °C are required, which can easily be obtained from a solar pond with little fluctuation during the summer period.
Although many feasibility studies have been made for the generation of electric power from solar ponds, the only operational systems are in Israel (Tabor, 1981). These include a 1500 m2 pond used to operate a 6 kW Rankine cycle turbine-generator and a 7000 m2 pond producing 150 kW peak power. Both of these ponds operate at about 90 °C. A schematic of the power plant design, working with an organic fluid, is shown in Figure 10.19.
FIGURE 10.19 Schematic of a solar pond power generation system.
For power production in the multi-megawatt range, a solar pond of several square kilometers surface area is needed. However, this is not feasible economically, since excavation and preparation account for more than 40% of the total capital cost of the power-generating station (Tabor, 1981). So, it would appear logical to employ a natural lake and convert a shallow portion of it to a solar pond.
Another use of the output from a salt gradient solar pond is to operate a low-temperature distillation unit to desalt seawater, such as MSF (see Chapter 8, Section 8.4.1). Such systems operate at a top temperature of 70 °C, which can easily be obtained with a solar pond. This concept has applicability in desert areas near oceans. Solar pond coupled desalination also involves the use of the hot brine from the pond as a thermal source to evaporate the water to be desalted at low pressure in a multiple-effect boiling (MEB) evaporator. The low pressure is produced by vacuum pumps powered by the electricity produced by the organic Rankine cycle (ORC) engine.
Matz and Feist (1967) propose solar ponds as a solution to brine disposal at inland Electrodialysis (ED) plants as well as a source of thermal energy to heat the feed of an ED plant, which can increase its performance.
Desalination of brackish water appears to be an extremely useful application of solar pond technology, as discussed above. In addition to providing clean renewable energy to power the desalination processes, salinity-gradient solar ponds can utilize the waste brine. The El Paso solar pond project showed that the cost of water produced by a zero discharge desalination system—which combines a solar pond with membrane filtration, thermal desalination and brine concentrator—ranges from US$ 1.06/m3 for a 3800 m3/day desalination plant to US$ 0.92/m3 for a 75,000 m3/day desalination plant (Swift et al., 2002).
For industrial process heat at medium temperatures (50–90 °C), the levelized energy cost (LEC), which is an economic measure that is useful for comparing and ranking technology options because it is a cost that accounts for the purchase, financing, tax and operation of a power system over its lifetime, ranges from US$ 6.60 per GJ for a 1-hectare (ha) pond to US$ 1.30/GJ for a 100 ha pond at sites having similar climate conditions to El Paso, Texas (Leblanc et al., 2011). The unit cost of supplying industrial process heat from a salinity-gradient solar pond is less expensive than either natural gas or coal for even modest pond sizes. For generating electricity, maximum revenue is obtained by maximizing operating hours of the engine or in a base load mode of operation. This is due to the relative high cost of the ORC generation equipment. The production of base load electricity using solar ponds is more expensive than current electrical base load generation technologies. However, it can become more cost competitive when the pond size is larger, greater than about 100 ha, and if the impact of environmental costs associated with burning fossil fuels are considered.
Leave a Reply