ࡱ> ?CNA@G Y}bjbjَ  Wy]4x>TTTTTTT-' 3 ?$h\ cTTTTTcTT>T.TTTT*@(>u8^RECONSTRUCTION AND OPERATION OF THE EL PASO SOLAR POND WITH A GEOSYNTHETIC CLAY LINER SYSTEM Huanmin Lu, Reseach Specialist Mechanical and Industrial Engineering Department University of Texas at El Paso 500 W. University Dr., El Paso, Texas 79968 Andrew H.P. Swift, Interim Dean College of Engineering The University of Texas at El Paso 500 W. University Dr., El Paso, Texas 79968 ABSTRACT After the original XR-5 membrane liner failed in 1992, the El Paso Solar Pond was reconstructed and operated with a geosynthetic clay liner (GCL) system. It is about 3000 m2 in surface area, and about 3.2 meters deep with a 15( side-wall slope. The new heat extraction system includes 6-inch rubber hoses and two redesigned polypropylene diffusers. A new automated instrumentation system was developed for monitoring pond status. It uses a newly developed scanner combined with a computer for both control and data logging. The salinity gradient was established using a new method, by injecting fresh water into brine through a newly designed PVC bar shaped diffuser, which scans automatically within preset regions. After two months, the pond bottom reached 80(C and heat extraction began. The performance of the GCL system, characterized by its hydraulic conductivity, has been monitored, and to our knowledge, generates the first full scale, elevated temperature data for a GCL system. Preliminary hydraulic conductivity data indicate values comparable with other clay liner systems. 1. INTRODUCTION After seven years of operation and a complete failure of the XR-5 membrane liner, the El Paso Solar Pond has been reconstructed and operated successfully with a geosynthetic clay liner (GCL) system in 1995. 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 U.S. Bureau of Reclamation and the State of Texas. The project, located on the property of Bruce Foods, Inc., was initiated in 1983 in cooperation with the U.S. Bureau of Reclamation. The El Paso Solar Pond was converted from a 3355 square meter (at the water line of 3.3 m) water pond which had been previously used for fire protection. An XR-5 liner, a PVC coated polyester fabric of 1.0 kg/m2 finished weight, was installed in 1984 as part of a double lining system with an existing hypalon liner beneath forming a secondary containment (Reid, et al., 1989). Since 1985, the El Paso Solar Pond has been intermittently operated for seven years. In 1985 the El Paso Solar Pond became the first in the world to deliver industrial process heat to a commercial manufacturer; in 1986 the first solar pond electric power generating facility in the United States; and in 1987 the nations first experimental solar pond powered water desalting facility (Reid, et al., 1992; and Swift, et al., 1990 and 1992). The El Paso pond sustained record breaking near-boiling temperatures, and the testing of new gradient establishment and management methods successfully demonstrated the feasibility of the periodic pond concept. Also, new clarity and stability control strategies were developed and helped identify an optimum stability margin for maintaining a high performance solar pond (Lu and Sandoval, 1993; and Xu, et al., 1991 and 1993). After six years of operation, the El Paso pond operation ceased due to a failure of the XR-5 liner. The XR-5 liner had deteriorated much more quickly than the anticipated lifetime of twenty years. Over 100 holes were identified on the XR-5 liner lower side walls. The XR-5 liner had lost much of its elasticity and strength near the storage zone in the pond where the liner was exposed to high temperature (Robbins, et al., 1995). This event has led to rethinking of membrane liners for solar ponds, especially on a large scale. Compacted clay liners (CCL) have been used in Israel and Mexico for solar ponds, but the characteristics and cost of CCL are very site specific, and can vary depending upon the experience of the installers (Daniel, 1993). After an investigation of various types of lining systems, including CCL and membrane liners, two separate lining systems - a flexible polypropylene geomembrane and a geosynthetic clay liner (GCL) were selected and designed specifically for the El Paso Solar Pond by the U.S. Bureau of Reclamation and researchers at the University of Texas at El Paso (Lichtwardt and Remmers, 1995). The geosynthetic clay liner used at the El Paso pond is Gundseal(, which consists of a layer of sodium bentonite clay glued to 30-mil polypropylene and is manufactured by Gundle Lining Systems, Inc. The GCL system was installed in May 1994. After hydration of the clay liner, which is required for application in solar ponds, heavy brine with some dry salt was pumped into the pond from the evaporation ponds, and the salinity gradient was constructed at the end of March 1995. Two months later, the bottom temperature of the pond reached 80(C, and heat extraction was started. This paper describes the reconstruction of the El Paso Solar Pond and its first year of operation. The preliminary hydraulic conductivity data of the GCL system, which characterizes the performance of the liner, is also presented. 2. CONSTRUCTION AND INSTALLATION OF GCL SYSTEM The El Paso Solar Pond was reconstructed after the removal of the failed XR-5 liner. The new El Paso Solar Pond has a bottom area of 24.4 m by 36.6 m (80 ft by 120 ft) with a side wall slope of 15(; its surface area at 3.2 meters depth is about 3000 m2. The pond itself was re-excavated and compacted before the installation of the liner system. Five thousand tons of dirt were moved into the pond and well compacted to change the slope of the walls from 30( to 15(. The pond was constructed so that leakage rates through the liners could be investigated. The bottom of the pond slopes to the northeast corner where a drainage pipe was installed and connected to a sump outside the pond. The entire pond was covered with a 30-mil, sealed, polypropylene liner which provided the secondary containment and formed the basis for the experimental containment facility. The primary lining system consisted of the GCL on the bottom and a 40-mil polypropylene liner on the sides. The GCL was installed with the clay side down to minimize contact between the bentonite and the sodium chloride brine. The lining system on the sides and bottom were connected separately to the sump so that leakage rates could be determined independently through the sides and bottom. The seams were overlapped 30-50 cm (12-20 inches) and the 30 lb/ft2 overburden was installed as recommended by the manufacturer. About 15 cm of sand were placed on the top of the Gundseal( to ensure the liner always stays in position and has a good contact with the sand beneath it during hydration. The installation of the two liner systems was completed in mid-1994 and the prehydration of the GCL was finished in late 1994. For more details see Lichtwardt and Remmers (1995) and Robbins, et al., (1995). 3. INSTRUMENTATION SYSTEMS Instrumentation and data monitoring are essential for successful solar pond operation. The pond status, which includes the temperature and salinity distributions in the pond, stability status and clarity status, need to be monitored on a regular basis. At the El Paso Solar Pond, the stability of the pond is indicated by the stability margin number (SMN) defined as the ratio of the measured stability coefficient to the calculated stability coefficient required to satisfy the dynamic stability criterion, and the pond clarity status is monitored by measuring turbidity and pH. During the six-year operation of the El Paso solar pond from May 1986 through April 1992, several measurement methods and instruments were investigated and tested, and a semi-automated integrated instrumentation system which uses scanner technology combined with data logging facilities was developed and used. In addition, the procedures of data collection and analysis were developed, tested and refined (Lu, 1994). Based on the previous experiences, an improved automated instrumentation system was developed and is used at the new El Paso Solar Pond as the main monitoring system. This system consists of a newly developed drum scanner, a sample pump, Dynatrol density meter, pH probe, turbidimeter, a shell and tube heat exchanger, and a computer. Two sets of T-type thermocouples and a small extraction diffuser are mounted on the drum scanner. The scanner and sample pump are mounted on the deck of a newly constructed instrumentation tower, which is located at the south part of the pond with its base 61 cm (2 feet) away from the south toe of the pond wall. The density meter, pH probe, turbidimeter and the heat exchanger are all mounted in the same wooden box located on the south bank of the pond. The computer, which is used for both control and data logging capabilities, is housed in the instrumentation room. With the integrated instrumentation system, the temperature, salinity, and the water quality at selected depths of the pond can be measured simultaneously. Measurement for the entire pond can be completed in about three hours. This allows for a near real time view of the pond status and enables both better gradient establishment and high thermal performance operation of the pond. The instrumentation tower is 3.7-m (12-feet) tall. Its 3 ( 3 ( 0.3-m (10 ( 10 ( 1-foot) thick concrete slab foundation was cast in place on an extra 30-mil polypropylene material after the installation of the 30-mil polypropylene. Four 15-cm (6-inch) columns and two 30-cm (12-inch) columns were cast into the concrete slab to support the 2.4 ( 3.7 m (8 ( 12 feet) deck. The 15-cm columns, made of 6-inch CPVC pipe filled with concrete and reinforcement bars, are also used for support and guidance of both the injection diffuser and the heat extraction diffuser. The two 30-cm columns, made of concrete with reinforcement bars and wrapped with 40-mil polypropylene, are used to provide additional rigidity. A 1.2-m (4-feet) wide, 18-m (60-feet) long bridge, which connects the deck to the south bank of the pond, provides ready access to the instrumentation tower. The drum scanner, the so-called third-generation scanner at the El Paso Solar Pond, utilizes a stepping motor and is controlled by a computer. The rubber driving drum of the scanner, 9.5 cm (3.7 inches) in diameter and 3.2 cm (1.3 inches) in width, was mounted directly to the shaft of the stepping motor. To improve rigidity, a coated metal, curved strip from a tape measure was used as the vehicle to move the sensor head, which carries two thermocouples and suction diffuser, up and down. Two 3.8 cm (1.5 inches) diameter rubber press drums were used and adjusted to produce enough friction to drive the metal strip without any slip. A HURST SAS 4004-002 stepping motor was selected as the driving motor. A coupling connects the stepping motor to the shaft of the driving drum. The motor was covered to protect the motor from corrosion by the environment. To initialize the starting position at the beginning of each scan and to ensure scanning within a given range, a limit switch was used. The limit switch will initialize the starting position each time the scanner is turned on; it also provides safety protection in case the strip moves out of computer control. The limit switch will shut the scanner down once it is pressed by the sensor head. The scanning range depends on the length of the strip and can reach 5 m (16 feet). The position accuracy of this scanner for a 3 m (10 foot) scan is (0.08 cm (0.03 inch). This scanner offers the following advantages over the previous scanners: it is relatively easy to use, delivers high spatial resolution and accuracy, is more reliable due to the instruments self position calibration and scanning range protection, provides continuity in pond data collection, is easy to assemble and maintain and is small in size and cost effective. For more details about the scanner, see Zhang (1993). The sensor head was made of a 7.6 cm (3 inch) diameter polypropylene rod. Its length is about 12.5 cm (4.9 inches). A semi-circle, 0.2 cm (0.079 inch) inlet gap was cut on the middle of the sensor head which serves as the extraction diffuser through which the brine samples are withdrawn and flow through the external devices such as the Dynatrol density meter, pH probe, and turbidimeter to measure salinity, pH and turbidity. For measuring the temperature distribution in the pond, two thermocouples are mounted on the two opposite sides of the sensor head via two 1 cm (0.4 inch) diameter polypropylene rods, which are attached on the sensor head and extend outward. The two thermocouples are about 20 cm (7.9 inches) apart and have been aligned to the level of the diffuser inlet gap. The two thermocouples provide redundancy. In order to overcome the buoyancy of the sensor head in heavy brine and let the scanner move downward more smoothly, a counterweight of length 20 cm (7.9 inches) and diameter 5 cm (2 inches) CPVC pipe filled with lead shot, was attached at the bottom of the sensor head. (See Figure 1.)  Fig. 1 Scanner and Sensor Head A wind prevention pipe (not shown in Fig. 1) was mounted on the scanner base and extended into the water surface to prevent wind effects on the exposed metal strip of the tape measure. The sample pump used in this integrated system is a Cole-Parmer peristaltic pump with adjustable speed. The Dynatrol density meter is a Model CL-10 HY density cell with integral temperature compensation. It is designed for measurement of density, specific gravity, or percentage of solids at process conditions, and its response is immediate and continuous. The pH measurement instruments in this integral system consists of a Cole-Parmer Model 27003 pH electrode and a Model 5654-60 pH transmitter. The Model 27003 electrode is an in-line pH/ATC (Automatic Temperature Compensation) combination electrode with a flat surface which does not allow the electrode to collect suspended solids, since the fluid flow across the surface of the electrode provides a cleaning action that extends the life of the electrode and improves performance. The turbidimeter is a HACH Model 1720C turbidimeter. It is a continuous-reading nephelometric turbidimeter designed for low-range turbidity monitoring. Since the turbidimeter requires the flow-through brine samples to be at a temperature less than 40(C, a shell and tube heat exchanger is utilized in the measurement system to cool the hot brine samples before they reach the instruments. For more information about these instruments, see Lu (1994). Besides the integrated instrument system, a string scanner system is used as a temperature measurement back-up system. It consists of a bare-junction, T-type thermocouple scanned vertically through the depth of the pond on a weighted nylon string hung from and wrapped around a drum attached to a DC motor. The temperature profiles are plotted by a chart recorder connected to the thermocouple. 4. HEAT EXTRACTION SYSTEM Heat in the solar pond is extracted by withdrawing the hot brine pumped from the storage zone by means of a diffuser (extraction diffuser) mounted in the storage zone, passed through an external heat exchanger, then returned to the bottom of the pond through another diffuser (return diffuser). The method has been used at the El Paso pond for several years and has been shown to be effective. The extraction diffuser can be moved to the height of maximum temperature in the storage zone and the return diffuser can be placed below it. This method allows placement for both the extraction and return diffusers near the point of use, reducing pipe cost. Also, this method insures that the cooler brine is returned to the bottom, reducing ground losses, and that the piping can be easily removed for inspection and repair. The rebuilt El Paso Solar Pond heat extraction system has been redesigned and reconstructed. In the old system, the diffusers and pipes were all made of steel. After several years of operation, they all indicated selective rusting. Also, the free ions of iron, which were introduced into the pond brine from the steel diffusers and pipes, were suspected of causing a clarity problem. This was observed in 1991 when a black layer appeared in the pond and the pond brine became very turbid (Abou-Chakra, 1992). Therefore, both extraction and return diffusers are now made of 1.9 cm (3/4 inch) polypropylene plate, and two pieces of 15 cm (6-inch) diameter rubber hose are used to connect the diffusers to the external piping system, which is still constructed of steel pipe. Both suction and return diffusers are double-plate diffusers. The suction diffuser is mounted under the deck of the instrumentation tower, among the four columns and 20 cm below the lower boundary (see Figure 2). The lower plate of the suction diffuser is circular, 76 cm (30 inches) in diameter, and the upper plate is a square of 102 by 102 cm (40 by 40 inches). The two plates are spaced at 15 cm (6 inches) apart. The opening of the diffuser is covered with stainless steel screen to prevent the piping system from sucking in debris. Four guide devices are mounted at each corner of the upper plate, against the columns to ensure the level position of the diffuser and prevent vibration. The vertical position of the diffuser can be easily adjusted by cable attached to the diffuser.  EMBED Word.Picture.6  Fig. 2 Schematic of Suction Diffuser The return diffuser is placed at the pond bottom about 15 m (50 feet) away from the instrumentation tower, on a gravel bed. The gravel bed is about 10 cm (4 inches) thick, and below the gravel lies a piece of 10-mil polypropylene which covers the sand and prevents it from being washed away by the brine exiting the diffuser. Both upper and lower plates of the diffuser are circular, 122 cm (48 inches) in diameter. The gap between the two plates is also 15 cm (6 inches). The maximum withdrawal flow rate for this design is 2.3 m3/min. (600 gpm) and at this flow rate the exiting velocity is less than 7 cm/sec. 5. SALINITY GRADIENT ESTABLISHMENT The salinity gradient was built in late March, 1995, with a new method, developed at The University of Texas at El Paso, that uses a scanning fresh water injection. The procedure consists of partially filling the pond with saturated brine and injecting fresh water in a scanning step by step fashion through a diffuser that is immersed within the existing solution. This method had been successfully implemented previously at the El Paso Solar Pond on a small scale (Xu, et al., 1992). During the gradient establishment, several additional new techniques were used. A newly designed bar shaped diffuser, instead of the traditional circular diffuser, was used as the injection diffuser. Also, a scanning injection technique was utilized. During each step of injection, the diffuser continuously moved up and down within a preset region, instead of sitting at a fixed position as it had previously. The movement of the diffuser was automatically controlled by a data logger, the Digistrip II, and relay circuitry. With these new techniques, the salinity gradient was built with great ease, was less labor-intensive and less time-consuming. Most importantly, the achieved salinity profile was much smoother and matched well with the desired profile. Figure 3 shows the schematic of the injection method. The reason that this method worked so well, is that critical mixing Froude numbers are not required, as in the fixed point method. As long as a sufficient mixing Froude number was used, the minimum critical value being 18 (Liao, 1987), the process was successful.  Fig. 3 Schematic of Injection Method The newly designed injection diffuser was made of two pieces of 10 cm (4 inch) diameter PVC pipe and two pieces of 7.6 cm (3 inch) diameter PVC pipe, and was assembled as a pipe within a pipe to ensure uniform velocity at the exit slots along the diffuser. The two pieces had equal length and were glued to a tee which connected with the fresh water line through a 7.6 cm (3 inch) rubber hose. On each piece of the 10 cm diameter PVC pipe there were 14 slots 3.8 cm (1.5 inches) long and 0.32 cm (1/8 inch) wide, spaced 2.5 cm (1 inch) from each other (see Figure 4). The total open area of the 28 slots was 34 cm2. Fresh city water was used as the water source for the gradient establishment. The designed flow rate was 0.57 m3/min. (150 gpm), and the exiting velocity of injected fluid would be 2.8 m/sec. The injection diffuser was mounted on a specially made guiding device, which can easily slide up and down along the two north columns of the instrumentation tower, so that the diffuser can travel vertically through the depth of the pond. The movement of the injection diffuser was driven by a DC motor through a drum-cable system mounted under the decking of the instrumentation tower. A 20 k( precision potentiometer was employed as a position feed-back system. The position of the injection diffuser was displayed on the Digistrip II data logger. Both upper and lower limits of each injection region could be set on the Digistrip. The scanning position, range and direction of the injection diffuser were controlled by the Digistrip and a relay circuitry. When the diffuser moved in one direction and reached the preset limit, the Digistrip would send an electric signal to the relay circuitry and then change the movement of the diffuser to the opposite direction automatically. After one injection step was completed, the upper and lower limits could easily be reset on the Digistrip for the next injection region.  Fig. 4 Schematic of Injection Diffuser The desired salinity profile is shown in Figure 5. According to this profile, the total depth of the pond was 320 cm (10.5 feet), the surface zone, or upper convective zone (UCZ) was 50 cm (1.64 feet) thick, the bottom zone, or lower convective zone (LCZ) was 120 cm (3.94 feet) thick, and the gradient zone, or non-convective zone (NCZ) was 150 cm (4.92 feet) thick. The total quantity of salt required for building the pond was 990 metric tons, which is equivalent to a salt content in the brine of about 3160 m3 at a density of 1.200.  Fig. 5 The Desired Salinity Profile Brine with some dry salt was pumped into the pond from evaporation ponds beginning in late 1994. Because of the difficulty of dissolving crystallized salt, it took several months to transfer enough brine to obtain the required amount of salt in the pond. The pond was fully saturated with a density of 1.206 in late March 1995, and the pond level was 217 cm, only 2 cm higher than the desired value. A step-by-step scanning injection procedure was planned for gradient establishment. The entire injection procedure contained 30 steps. The lower limit of the injection region moved upward 5 cm (2 inches) after each step. The upper limit was determined by the injection region of each individual step. The injection region was 20 cm (8 inches) for all but the last four injection steps. The required volume of fresh water and the injection time for each step were computed based on the injection procedure by a computer program. It took four days to build the gradient. The establishment was started on the morning of March 28, 1995, and ended on the evening of March 31, 1995. On each day, the injection lasted about 10 hours, starting in the morning and stopping in the evening. After the injection process was completed in the evening of March 31, the pond level was 273 cm, and fresh water was added onto the pond surface through a floating diffuser made of polypropylene plate. The fresh water addition on the pond surface lasted about two days and was stopped when the pond level reached 324 cm. During the injection process the injection result was checked daily (each morning) by measuring the salinity distribution of the pond and comparing it with the projected profile. If there was a significant difference between the two salinity profiles, the days injection plan would be modified to correct the difference. Figure 6 shows the day-by-day development of the salinity distribution of the pond during the gradient establishment. As shown in this figure, the daily injection results fit the projected salinity profile very well. Very little deviation from the planned injection procedure was required.  Fig. 6 Salinity Profile Development during Gradient Establishment 6. TEMPERATURE DEVELOPMENT AND POND OPERATION A transient temperature gradient became established immediately after the salinity gradient was established. Figure 7 shows the upper and lower zone temperature development of the pond during the first two months after salinity gradient establishment. It can be seen that the bottom temperature of the pond increased at an average rate of about 1C per day. The bottom temperature reached 80C in early June 1995, and heat extraction was then started. The temperature in the lower convective zone was maintained at between 70 to 80C through mid-December 1995. The clarity of pond brine was controlled by lowering the pH level in the surface and gradient zones. The pH value in these two zones was maintained at a level between 4 and 5. The pH level in the bottom zone was higher than 5, because of the GCL system.  Fig. 7 Temperature Development in the Pond UCZ, LCZ - upper and lower convective zones, respectively 7. PRELIMINARY PERFORMANCE OF GCL SYSTEM The performance of the GCL is characterized by its hydraulic conductivity. At the El Paso Solar Pond, the hydraulic conductivity of the GCL was determined by measuring flow through the liner as a function of time and gathered by the secondary liner. Any water that permeated the GCL was collected in the sump and intermittently pumped out through a totalizing flow meter. The data was recorded on a daily basis until late in June 1995, when the submersible pump in the sump failed due to high temperature. Since then, the brine collected in the sump has been pumped out periodically through another pump outside of the sump. The hydraulic conductivity of the GCL has been measured to be on the order of 110-6 cm/sec. This is comparable to the compacted clay/plastic lining system installed in Israel (Lichtwardt, et al., 1995). It appears the permeability of the GCL is affected by temperature, but more data is needed before definitive conclusions can be drawn. 8. CONCLUSIONS The liner is an important issue for utilizing salinity-gradient solar pond technology, and the failure of the XR-5 liner demonstrates the need for advanced liners designed specifically for solar ponds. A geosynthetic clay liner (GCL) system designed specifically for solar pond applications has been installed and its performance has been monitored at the El Paso Solar Pond. The GCL system features several advantages: self-sealing capacity, quick and relatively easy installation, and cost effectiveness. This lining technology has been recently used in landfills and other applications but never tested at elevated temperatures. The data gathered at the El Paso Solar Pond should be valuable for both solar pond technology and environmental control projects that use GCL liners, especially ones where temperatures may become elevated. The scanning technique for salinity gradient establishment, an automated instrumentation system for solar pond monitoring, and other techniques for pond operation and maintenance have been developed and implemented at the El Paso Solar Pond. These techniques are expected to greatly contribute to the commercialization of salinity-gradient solar pond technology. 9. ACKNOWLEDGMENTS The authors would like to thank the U. S. Bureau of Reclamation, the State of Texas, and the Center for Environmental Resource Management at The University of Texas at El Paso for their support of this project. We also thank Mr. Herb Hein, Mr. Dean Brown and the faculty, staff and students at The University of Texas at El Paso for their significant contributions to this project. References Abou-Chakra, F. N., 1992, Analyses of the Sources, Factors, and Treatment Methods Affecting Turbidity at the El Paso Solar Pond, Master Thesis, The University of Texas at El Paso, El Paso, Texas. Daniel, D., ed., 1993, Geotechnical Practice for Waste Disposal, Chapman & Hall, London. Liao, Y., 1987, Gradient Stability and Injection Analysis for the El Paso Solar Pond, Master Thesis, The University of Texas at El Paso, El Paso, Texas. Lichtwardt, M. A., and Remmers, H., 1995, Advanced Solar Pond Liner Technologies, Joint Fall Meeting of the Texas Sections of APS and AAPT, and SPS Zone 13, Lubbock, Texas. Lu, H., and Sandoval, J., 1993, Experiences of Clarity Monitoring and Maintenance at the El Paso Solar Pond, Proceedings of the 3rd International Conference on Progress in Solar Ponds, El Paso, Texas. Lu, H, 1994, Monitoring and Data Analysis for Solar Pond Operation, Master Thesis, The University of Texas at El Paso, El Paso, Texas. Reid, R. L., et al, 1989, Design, Construction, and Initial Operation of a 3355 m2 Solar Pond in El Paso, Journal of Solar Energy Engineering, Trans. of the ASME, Vol. 3, No. 4. Reid, R. L., Swift, A.H.P., et al., 1992, El Paso Solar Pond Test Project, Phase 1, Final Report, U. S. Bureau of Reclamation Report No. R-92-07. Robbins, M. C., Lu, H., and Swift, A.H.P., 1995, Investigation of the Suitability of a Geosynthetic Clay Liner System for the El Paso Solar Pond, Proceedings of the 1995 Annual Conference of American Solar Energy Society, Minneapolis, Minnesota. Swift, A.H.P., et al., 1990, Demonstration of Solar Pond Coupled Desalting Using a Multistage, Falling Film, Flash Evaporator, Proceedings of the 2nd International Conference on Progress in Solar Ponds, Rome, Italy. Swift, A.H.P., et al., 1992, El Paso Solar Pond Test Project, Phase II, Final Report, Research and Laboratory Services Division, U. S. Department of Interior. Xu, H., Golding, P., and Swift, A.H.P., 1991, Method for Monitoring Stability within a Solar Pond Salinity Gradient, Solar Engineering - 1991, Mancini, et al., ed., ASME Press, NY. Xu, H., Golding, P., and Swift, A.H.P., 1992, Use of Injection Method in Salinity Gradient Establishment, Solar Engineering - 1992, Stine, et al., ed., ASME Press, NY. Xu, H., et al.,1993, Operation Experience with the El Paso Solar Pond, Proceedings of the 3rd International Conference on Progress in Solar Ponds, El Paso, Texas. Zhang, X., 1993, Design of a Computer Controlled Scanner and Data Acquisition System for Solar Pond Performance Monitoring, Master Thesis, The University of Texas at El Paso, El Paso, Texas. :;pq yz~fg()))))))**)***11777==HHHHHHHHCKDKQjf4 9CJUjf4 9OJQJUVmHj9CJUj9CJU9CJ j9CJOJQJ j9CJOJQJ j9CJOJQJ9H*OJQJ j9CJOJQJ9CJH*OJQJ9CJOJQJ67]^}~3V $$7]^}~3V ##((,,-,q3r377778888==^?_?`?z?{?BBEEHHHH,I-IKKKKKQQQQRRYYYYYYY[[\\5\6\__dddde##((,,-,q3r377778888==^?_?$_?`?z?{?BBEEHHHH,I-IKKKKKQQQQRRYYYYY$QQQQTTTTVVYYYY[[[\\\]]dddj(9jJ2 """)))000___UUUMMMBBB999 ```( +9%I;/]E:IS+!!lYQGj2ga1aS {Cg..&YQFh.IRj#SjJu3lJA7e ,N, QdoV YC6r3w_GqC-}zn#&s[ Rp$L   !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\]^_`abcdefghijklmnopqrstuvwxyz{|}~()Root Entry  F,7 22>u8Data WordDocument  ObjectPool`d>u8 22>u8_879144447 F`d>u8`>u8PIC LMETA CompObjb !"#$%&'()*+,-./0123456789:;<=>?@ABCDEFGHIJKLNLG$D1   FMicrosoft Word 6.0 Picture MSWordDocWord.Picture.6 Oh+'0  $ H l   DhG$D&   -&WordMicrosoft Word   D-Times New Roman- &* A-#--#;;-- -- &&---gF;----m---- e----ue----te---e--l__---HV;----J----H'T----J)9---T'--'L---I}(---- jEj--- jEj--- $^:$--- $^:$--$--!!!--jIj--\---&----%---ll----p*ph--&j---H----JC---4.O--.O-- !<4C---- %=5---- $Q6/$2G--\-2 m Column of7**A**'5-"2 instrumentation* *A%*%**'#-2 4tower*<%'-- <<t----- %1!---- $D+)!0---2 Cable7%*%'-- ;;!----- %[S---- $o\T@Lb---2  6*&2 &2 C  rubber hose***%** %'-- >>s----- %III---- $]6I--n-2 Guide<**%'--|---- 00L----- %q------e -2 v  Brine out7*%**'-V - -B( r-r'-'--vh-- --- %88---- $r--\-2 Screen.%%%*' & -]3##"&54632ObjInfo ObjectPool`>u8`>u8WordDocument KSummaryInformation(ܥe3 eYKVnnnn  @!eT4nnn(nnnn Column of instrumentation tower Cable 6 rubber hose Guide Brine out Screen XX 0 _# B&:&& &U &M&e- M&N&&&'&w 0&@&f ?&)LM& MN&)&I &ZN&J&J&j&j&hg&{ji&&jyx&"N& @&6 ml&, nwv&] &U n& lll&smj&3,'&+&?&G6 d{z6i TUS+ T((((((Y&y W"6a`6{URTCQ( LY&&D6 6 `USTCR(V&7)6 6 G^F]F-(JQ&# &}6 H& S3R2& HS1R0(v&N|&&6 6 SW.VR(S \WWXY u Y]abuDY]a  $%+,<=CDNOVWXYZZXxK@NormalY]a "A@"Default Paragraph Font #*;BMUXYY Y YYg@HP LaserJet 5P/5MPLPT1:HPPCL5GHP LaserJet 5P/5MP Dk&ә@6X&HP LaserJet 5P/5MP Dk&ә@6X& @Times New Roman Symbol &ArialRoman PS"hMM!4Word DevelopmentWord Development(C:\MSOFFICE\WINWORD\TEMPLATE\NORMAL.DOTWord DevelopmentWord Development@k%@@k%@Microsoft Word 6.026n{DuqVJ4H2Aph6B"}3%6Z\H"BMXR ӕ$sVo<gX .YLg:}ߚ=VCq8SeFMjLZc"LeN?Pppiiiwwwfxю7Tvpxd?2}Dx#$_`,19م>w؁V!0ȳypQit|45'  --  --' -- --' -- --'-- --'-- --'-- -- $777H-777--'-- -7 7Times New Roman-7:7:~7~:Y7Y:575:7:777ww--"System-'--- ---'--- 7---'---   2-K-- QG-KHKNKNKHN-TJ-NKNQNQNKR-XN-RORURUROV-\R-VSVYVYVSY-_U-YVY\Y\YV]-cY-]Z]`]`]Za-g]-a^adada^d-j`-dadgdgdah-nd-hehkhkhel-rh-lilololio-uk-olororols-yo-spsvsvspv-|r-vsvyvyvsz-v-zwz}z}zw~-z-~{~~~{-}-~~----------------------------------1p-4s.m5p8s2m8p;s5m<p?s9m@oCr=lCtFw@qGvJyDsK{N~HxNQK~RUOVYSY\V]`Zad^dgahkeloiorlsvpvysz}w~{~---'--- 7---'--- ---'--- ---'---    2 #50 2 100 2 w150 2 R200 2 .250 2 300---'--- ---'---  2 '1.05002 g1.10002 1.15002 1.2000---'--- ---'--- o 2 tSpecific Gravity---'--- ---'--- J Times New RomanB- 2 Height (cm)----'--- --  H---'--- I---'--- I-LSO-- UK-OLOROLOR 2 VInitial profile---'--- I---'--- I-LSO--  RL 2 VDuring Injection---'--- I---'--- --I I $ILFI-NE@OOIOIOAAGAG:2  Injection 2 -Diffuser2 <---'--- ---'- -'- --'  '  ' --  --'$D+82   &WordMicrosoft Word  G D-Times New Romanw- & - !k"H-Z*PTimes New Roman -2 7PG 3.81 cm"""4-'--_--&[&%[--[%----[$-- & &# U-- $##SS- %##SS---- $   $' ' // $? ? GG---5E5--22D--22- & &U-- $%S%S- %%S%S---- $!! $'!'!// $?!?!GG---55--2=2--22B- &  & & [&k  [--[ k ----[ l -- & &m U-- $l l SS- %l l SS---- $ $''// $??GG---5J 5--22K--22- & &qU-- $jSjS- %jSjS---- $nrrn $n'r'r/n/ $n?r?rGnG---55--2Q2--22L- &  & -"--{B"B--'--H(H--B--Bc--Bc--~--~--H--H'--H'--------HB-- !_(-0-2 =G 2.54 cm"""4-'- !y----%Gl---- $\:=--r -2 1G Tee3%%'- !m%"----%y---- $(---*-2 :*G Slot%"-'- !n[Z----%+s---- $-- &c-2 3cG PVC pipe%0-""-'--E -- & -D@      ``iݜ*o?H6A^BbzbO/(fvEE>j(9jJ2 """)))000___UUUMMMBBB999 ```( +9%I;/]E:IS+!!lYQGj2ga1aS {Cg..&YQFh.IRj#SjJu3lJA7e ,N, QdoV YC6r3w_GqC-}zn#&s[ Rp$L 6n{DuqVJ4H2Aph6B"}3%6Z\H"BMXR ӕ$sVo<gX .YLg:}ߚ=VCq8SeFMjLZc"LeN?Pppiiiwwwfxю7Tvpxd?2}Dx#$_`,19م>w؁V!0ȳypQit|45'  --  --' -- --' -- --'-- --'-- --'-- -- $999-999-99Times New Roman-9<9<9<k9k<L9L<.9.<9<999ii--"System-'--- ---'--- 9---'---   4-'9-- *<$6*9-<'6.91<+6194<.6497<1679:<46:9=<76=9@<:6@9C<=6CKFN@HFYI\CVIeLhFbLqOtInO|RLyRUOUXRX[U[^X^a[ad^dgagjdknhnqkqtntwqwztz}w}z}---'--- 9---'--- ---'--- ---'---    2 %50 2 100 2 150 2 d200 2 E250 2 '300 2 350---'--- ---'---  2 )1.00002 Y1.05002 1.10002 1.15002 1.2000---'--- ---'--- p 2 uSpecific Gravity---'--- ---'--- I Times New Roman- 2 Height (cm)----'--- ---'--- ---'  '  ' --  --' dnD@ 7    ``iݜ*o?H6A^BbzbO/(fvEE>j(9jJ2 """)))000___UUUMMMBBB999 ```( +9%I;/]E:IS+!!lYQGj2ga1aS {Cg..&YQFh.IRj#SjJu3lJA7e ,N, QdoV YC6r3w_GqC-}zn#&s[ Rp$L 6n{DuqVJ4H2Aph6B"}3%6Z\H"BMXR ӕ$sVo<gX .YLg:}ߚ=VCq8SeFMjLZc"LeN?Pppiiiwwwfxю7Tvpxd?2}Dx#$_`,19م>w؁V!0ȳypQit|45'  --  --' -- --' -- --'-- --'-- --'-- -- $999mE-999--'-- -9 9Times New Roman-9<9<9<k9k<L9L<.9.<9<999ii--"System-'--- ---'--- 9---'---   4-'9-- *<$6*9-<'6.91<+6194<.6497<1679:<46:9=<76=9@<:6@9C<=6CKFN@HFYI\CVIeLhFbLqOtInO|RLyRUOUXRX[U[^X^a[ad^dgagjdknhnqkqtntwqwztz}w}z}^-dZ-^[^a^^^^a-g]-a^adaaaad-j`-dadgddddg-mc-gdgjggggk-qg-khknkkkkn-tj-nknqnnnnq-wm-qnqtqqqqt-zp-tqtwttttw-}s-wtwzwwwwz-v-zwz}zzzz}-y-}z}}}}}-|-}-----------      !"#$%&'*+,-./0-------------------------------[$X[^[X^$[^a^[a$^ada^d$adgdag$dgjgdk$hknkhn$knqnkq$nqtqnt$qtwtqw$twzwtz$wz}zw}$z}}z$}}$$$$$$$$$$$$$$$$$$$$$RUOUXRX[U[^X^a[ad^dgagjdknhnqkqtntwqwztz}w}z}I $FLLFL $IOOIO $LRRLR $OUUOU $RXXRX $U[[U[ $X^^X^ $[aa[a $^dd^d $aggag $djjdk $hnnhn $kqqkq $nttnt $qwwqw $tzztz $w}}w} $zz $}} $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $ $*>-0D&:-*>';*>-A*>-;*>'A.>-4D*:-.>+;.>1A.>1;.>+A1>-7D-:-1>.;1>4A1>4;1>.A4>-:D0:-4>1;4>7A4>7;4>1A7>-=D3:-7>4;7>:A