One Year Operation of a Salinity Gradient Solar Pond in Northern Cyprus

One Year Operation of a Salinity Gradient Solar Pond in Northern Cyprus- Experimental Investigations and CFD Simulation

¹Soudabeh Gorjinezhad, ¹Sultan Kadyrov,  ²Mohammad Askari, ¹Negar Zare Pakzad, ¹Mehdi Amouei Torkmahalleh, ³Goodarz Ahmadi

1-      Chemical Engineering Program, Middle East Technical University Northern Cyprus Campus, Guzelyurt, Mersin 10, Turkey

2-      Mechanical Engineering Program, Middle East Technical University Northern Cyprus Campus, Guzelyurt, Mersin 10, Turkey

3-      Mechanical Engineering Department, Clarkson University, Potsdam, NY, 13699- 5725

1. Introduction

Today, renewable energy sources gain importance day by day. It is crucial to develop devices and processes to supply energy from non-polluting and renewable energy sources for sustainable development of the world. Solar pond is an example of such devices that basically collects solar energy and stores it as thermal energy for a long period. Studies indicate that the temperature in solar ponds may in general reach up to 70-80°C implying that the thermal energy from solar ponds can be useful to applications with low grade energy demand [1]. One of the most important advantages of solar pond over other renewable energy sources, such as solar collectors, is its lower cost of investment [1]. Solar pond is environmentally friendly in particular when it is used for electricity generation. The heat obtained from solar pond can be converted into electric power even at low temperatures [2]. In this regard, organic Rankine cycle engines are generally run using the temperature difference in a solar pond. For applications where the organic fluid fails to operate due to low temperature difference, thermoelectric generators can be a good candidate to replace organic Rankine cycle engines for power generation [3].

Solar Ponds normally consist of three different salinity layers. The first layer, known as the upper convective zone (UCZ), is located at the top of the pond, and contains the least salinity level. The second layer, whose salinity level increases with depth, is called non-convective zone (NCZ). This layer is responsible to act as an insulator to prevent heat from escaping to the UCZ, maintaining higher temperature at deeper zones. The last layer made of a saturated salt solution, is responsible for energy storage, and is known as the lower convective zone (LCZ) [5]. The working principles of solar ponds are quite simple. In the absence of salinity gradient, the solar radiations reaching a pond are mostly absorbed by the lower water levels and will cause the water to heat up. The heated water is then risen up to the pond surface due to buoyancy effect and loses its thermal energy to the atmosphere. Thus, the main reason behind imposing salinity gradient to solar ponds is to create a density gradient restricting such buoyancy driven natural convection and as a result to trap the thermal energy at the bottom of the pond [4].

Northern Cyprus is enriched in solar energy in particular during summer. The performance of a solar pond in this region has not been evaluated yet. To take advantage of solar energy in Northern Cyprus, a salinity gradient solar pond has been constructed and operated at Middle East Technical University Northern Cyprus Campus (METU NCC) located at Guzelyurt, Northern Cyprus (Figure 1). This study is primarily concerned with describing the development, validation and use of CFD for computer modeling of flow and thermal analysis of the experimental salinity gradient solar pond at METU NCC. The pond has started operating since Oct. 2014 and experimental results are being recorded accordingly. The CFD commercial software considered to be utilized is ANSYS FLUENT and the challenge is to visualize the transient effect of temperature distribution, salinity concentration, velocity field and the degree of influence of wind speed on the overall stability and performance of the solar pond over time. 

Figure  SEQ Figure \* ARABIC 1. Constructed Solar Pond at Middle East Technical University (Northern Cyprus)

2. Materials and Experiment

A cylindrical salinity gradient solar pond of 61 cm diameter, 55 cm height and 1.2 cm thickness installed and operated since October 8^th, 2014 at Middle East Technical University Northern Cyprus Campus (METUNCC) located at Guzelyurt, Northern Cyprus. Three inlet ports were installed to carefully add the solutions to the pond to create the three salinity layers. Also, these inlets were used to add proper amount of solutions to each zone during the experiments to compensate for evaporation and sampling losses as well as surface washing. The pond has been divided in three zones. The bottom zone (LCZ) consisted of 75 liters of saturated salt solution (concentration C). The middle zone (NCZ) consisted of three equally divided sub-layers, each 15 liters in volume, with 0.75C, 0.5C and 0.25C concentrations. The top zone (UCZ) consisted of 15 liters of fresh water. The rest of the pond was left empty. The pond was equipped with three sampling valves to withdraw samples from each layer to monitor the salt concentrations using conductivity measurements. Six thermometers (2 in each zone) were installed to monitor the temperature variations at 9 a.m., 1 p.m., 5 p.m., and 10 p.m. every day. Temperature recording at other time of the day was performed as needed. The bottom and sides of the pond were insulated with a 2.5cm thick thermal insulator. The inner surfaces of the pond were painted black. Excess (undissolved) salt was added to the bottom of the pond, to ensure saturation of the bottom layer. The ambient temperature and solar radiation were monitored using a sun tracker (Kipp and Zonen) placed next to the solar pond. Wind velocity and relative humidity were also monitored throughout the experiment.

3. Computational Scheme and Modeling

The geometry and meshing of the solar pond were constructed by ANSYS Workbench Design Modeler and Meshing environment. FLUENT version 14.5 was used as the CFD solver. A two-dimensional computational domain is considered for investigating the transient behavior of the pond representing a cross-section of the solar pond. Accordingly, the rectangular domain of consideration is that of the existing pond dimensions with 61 (cm) in width and 50 (cm) in height. The lower wall of the rectangle is assumed to be the ground and upper boundary to be the top water level open to atmosphere. Vertical walls and the bottom surface of the pond are set to be impermeable and thermally insulated with the corresponding thermal conductivity value of the insulation material. Also the top surface, due to contact with ambient air, is taken to be at ambient temperature with an average wind speed according to local meteorological data. The salinity concentration of the LCZ is kept at saturated level, while the top surface is set to zero concentration, initially.

The modeling of the pond is divided into two phases. Firstly, a simple model is going to be developed in which it is assumed that the concentration level remains the same due to the present regular injection of brine solution to the system. Accordingly, the thermofluid properties will be assumed to be primarily a function of temperature and will be evaluated for each layer at its initial concentration. Thus, in the first phase, the concentration will not be considered as a modeling parameter and it is aimed to primarily study the variation of the temperature using the most simplified model possible. In the second phase, the model will be enhanced by defining the concentration as a varying parameter to the system. In this regard, the variation of properties with concentration can be taken into account and the significance of salt diffusion and its influence on the stability of the pond can be investigated. In addition, the results between both cases and the experimental results will be compared to justify the computational complexity added to the model.

4. Results and Discussion

Figure 2 shows the average monthly solar pond temperature variations from October 2014 to February 2015) as well as the average monthly ambient temperature and irradiance. As can be seen, the pond has been functioning as the temperature differences among the three layers have been established.  The pond temperature variations are well correlated with the ambient temperature variations. The temperature at all layers decreased from October 2014 to end of December 2015, but then beginning January 2015, it increased as solar irradiance increased.

Figure 1. Average monthly temperature variations of the solar pond

References

[1.]       Sukhatme, S., & Nayak, J. (2008). Solar energy: Principles of thermal collection and storage (3rd ed.). New Delhi: Tata McGraw-Hill.

[2.]       Akbarzadeh, A., Bernad, F., Casas, S., Gibert, O., Cortina, J. L., & Valderrama, C. (2013). Salinity gradient solar pond: Validation and simulation model. Solar Energy, 98, 366-374.

[3.]       Singh, B., Gomes, J., Tan, L., Date, A., & Akbarzadeh, A. (2012). Small Scale Power Generation using Low Grade Heat from Solar Pond. Procedia Engineering, 50-56

[4.]       Mehdizadeh, M., & Ahmadi, G. (2014). Two-dimensional Computer Simulation of Salinity Gradient Solar Pond Operation. Proceedings of ASME 2014 Fluids Engineering Summer Meeting, FEDSM2014.

[5.]       Jaefarzadeh, M. (2005). Thermal behavior of a large salinity-gradient solar pond in the city of mashhad. Salinity Gradient Solar Pond. Retrieved October 14, 2014, from http://profdoc.um.ac.ir/articles/a/203017.pdf