- Email: [email protected]
Improvement of solar still performance by covering absorber with blackened layers of sponge
Author’s Accepted Manuscript Improvement of Solar Still Performanceby Covering Absorber with Blackened Layers of Sponge M.H. Sellami, T. Belkis, M.L. Ali Ouar, S.D. Meddour, H. Bouguettaia, K. Loudiyi www.elsevier.com/locate/gsd
PII: DOI: Reference:
S2352-801X(17)30053-X http://dx.doi.org/10.1016/j.gsd.2017.05.004 GSD52
To appear in: Groundwater for Sustainable Development Received date: 1 November 2015 Revised date: 8 June 2016 Accepted date: 29 May 2017 Cite this article as: M.H. Sellami, T. Belkis, M.L. Ali Ouar, S.D. Meddour, H. Bouguettaia and K. Loudiyi, Improvement of Solar Still Performanceby Covering Absorber with Blackened Layers of Sponge, Groundwater for Sustainable Development, http://dx.doi.org/10.1016/j.gsd.2017.05.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Improvement of Solar Still Performanceby Covering Absorber with Blackened Layers of Sponge M.H. Sellamia*, T. Belkisa, M.L. Ali Ouar a, S.D. Meddour a , H. Bouguettaiab and K. Loudiyic a
Process Engineering Laboratory (PEL) Email: [email protected] b Laboratory of New and Renewable Energy in Arid Zones (LENREZA), Ouargla University, 30000 Algeria. c Renewable Energies Laboratory (REL), Al Akhawayne University, Ifrane, Morocco. * Corresponding author
Abstract Algeria has been listed among top countries affected by a shortage of fresh drinkable water. Solar desalination can be used to produce fresh water from brackish water to supply isolated, low-density, population areas located in southern Algeria where solar energy and underground saline water are abundant. This article aims to improve the yield of a solar still by improving absorber performance through the use of an added inner heat storage system. To do this, we tested covering the absorber surface with layers of blackened sponge. The resultant heat storage is used to keep the operating temperature of the absorber high enough to produce distilled water when solar irradiance is low or during night time. Four small-scale solar powered distillation pilot units were set up and operated.The experiments carried out in the “Process Engineering Laboratory of Ouargla University” studied the effect of sponge thickness on the productivity of the solar still. The results obtained showed that a 0.5cm sponge thickness increased the yield by 57.77 % i.e. 58%, relative to the baseline case (i.e. with no blackened sponge added).In contrast, asponge thickness of 1.0cm resulted in a yield improvement of only 23.03 %, whereas a sponge thickness of 1.5cm resulted in a decreased yield of 29.95 % i.e. 30% (relative to the baseline case). Keywords: Desalination; Solar energy; Single slope basin still;Sponge; Storage medium.
Nomenclature Symbols Basin area ( m 2 ) Ab Ag
Glass-cover area ( m 2 )
Ca
Specific heat of ambient air ( J / kg.K )
Cw
Average specific heat of brackish water ( 4190J / kg.K ) E Sponge thickness (m). € DZD Euro Algerian Dinar Solar radiation at time (t ) , ( W / m 2 ) I (t ) Grashoff number Gr
h1 h2 hb
Total heat transfer coefficient between brackish water surface and glass cover ( W / m 2 .K ) Total heat transfer coefficient between glass cover and ambient air ( W / m 2 .K ) Convective heat transfer coefficient throughout basin liner ( W / m 2 .K ) 1
hca
Convective heat transfer coefficient between glass-cover and the ambient ( W / m 2 .K )
hcwb hcwg
hcws
Convective heat transfer coefficient between brackish water and the basin plate ( W / m 2 .K )
Convective heat transfer coefficient between brackish water surface and glass cover ( W / m 2 .K ) Convective heat transfer coefficient between brackish water and the sponge layer ( W / m 2 .K ) Evaporative heat transfer coefficient between brackish water and glass cover ( W / m 2 .K ) hew hra
Irradiative heat transfer coefficient between glass-cover and the ambient ( W / m 2 .K )
hrwg
Irradiative heat transfer coefficient between brackish water surface and glass cover ( W / m 2 .K )
hsp
Convective heat transfer coefficient throughout sponge layer ( W / m 2 .K )
L Lv
water depth in basin (m) Latent heat of water vaporization ( J / kg) Mass of brackish water in basin (kg) Nusselt Number. Partial water vapour pressure at inner of glass cover temperature ( N / m 2 ) .
Mw Nu Pg Pw Ra t Ta
Tb Tg
Partial water vapour pressure at water surface temperature ( N / m 2 ) . Rayleigh Number. Time (s ) . Ambient temperature (K ) Temperature of the base (K ) Glass-cover temperature (K )
Ts
Tsp Temperature of Sky (K ) Sponge temperature (K) Brackish water temperature (K ) Tw ∆T (Tw – Tg): Temperature difference between brackish water and glass-cover (°C). v Wind velocity ( m / s ). X Insulation, basin and glass-cover thickness m Greek Absorptivity Glass-cover tilt angle. Emissivity Transmissivity Stephan-Boltzmann constant 5,67.10 8W .m. 2 K 4
Density of brackish water
)
Subscript
b Exp
Basin Experimental 2
g
w i S sp
Glass-cover Water Insulation Sky Sponge
1. Introduction Unlike other distillation methods, solar stills use environmentally-friendly solar energy to remove salts from saline or brackish water. Solar desalination is a good way of supporting the needs of small communities for fresh water produced with clean and cheap energy. Solar distillation is particularly useful for arid areas such as southern Algeria where solar energy and brackish water are abundant. These regions depend on groundwater for drinking, but unfortunately this water is often too salty to be drinkable. Instead, brackish groundwater with high salinity (3 g/l concentration) is often encountered [1]. Indeed, sufficient access to water for drinking and industrial uses is an increasing problem in Algeria. In the Saharan region, an area close to two million km2, groundwater is available in large quantity: about 60 × 103 billion m3or approximately 76% of Algerian reserves. This water is difficult to use and is not renewable. Only4 to 5 billion m3are exploited annually because most of the water is brackish. Salinity levels are variable and some groundwaters have salinity levels of up to 8 g/l. This highly exceeds the maximum 550 ppm levelallowed for human consumption [2-3].In northern Algeria, groundwater reserves are estimated to be 1.8 billion m3.If these waters could be used; they could be importantin improving development and living conditions in the region. Overall, Algeria receives an average of 65 billion m3/yof rain, of which only 3 billionm3/y supply groundwater reserves, 47 billion m3/y evaporates or transpires into the atmosphere, and 15 billion m3/y provides base flow to surface waters [1-4].Algeria currently ranks 14th worldwide among countries suffering from lack of water. If effective solutions are not found to provide a greater supply of freshwater, Algeria’s situation is predicted to get worse. By the year 2025 [2], Algeria is predicted to become the 6th country most affected by a lack of freshwater resources. The location for our experiments is Ouargla city (31.95°N latitude, 5.40°E longitude), located in southern Algeria at 141 m above the sea level. It has a high solar irradiance of about 2260 kWh/(m2.year) and 3400 sunshine hours per year [2-3]. Consequently, solar desalination technology is advantageous there. Extensive research has been conducted on different theoretical and experimental methods to improve and enhance productivity of solar still units [5-6]. Several authors have studied the effect of water level on absorber performance. The latter has been shown to be an important factor affecting basin-still productivity because thin films lead to high heat absorption. Still performance was shown to increase with thinner saline water films because of greater heat and mass transfer effectiveness [7-8and 9]. Other investigators, Farshchi et al [10], Dashtban and Tabrizi [11], studied and set up a cascade solar still fitted with weirs(weir-type solar still): paraffin wax was used to provide high heat energy storage beneath the absorber plate, keeping operating temperatures high enough to produce distilled water overnight or during a lack of sunshine. Theweir-type solar still is based on providing asuitable distribution of the feed water on a large evaporation surface witha high residence time. Generally, the blackened absorber is step shaped: each step has a horizontal and a vertical surface. The feed water flows slowly from the upper step to a lower step while crossing baffles or vertical fins placed on each step. These metallic fins increase the absorber area andthe water residence time in each step and therefore improve the heat and mass transfer. Also, the low flow of feed water maintains the depth of water in a thin film which leads to 3
good mass and heat transfer between the absorber and the condenser. These are usually very close to each other. The water temperature and salt concentration increasesas the water travels from top to bottom. Hence, the evaporation rate becomes more important in the last step and salt deposits may occur. Sellami et al [3] carried out an experimental study by laminating the basin with layers of blackened alluvial sand as a heat storage medium. Theauthors studied the effect of sand mass and particle diameter. The results revealed that for a fixed mass of sand on the absorbing layer, there was a significant improvement in output when fine sand particles were used. Due to thestorage medium and to photocatalytic effects, the yield recorded for a fixed particle diameter (0.08 mm) increases with an increase in sand mass to an optimum value of 2.268 kg of sand/m2 absorber surface. The improvement recorded was 43.28% compared with that of conventional still. For the same reason, Sakthivel et al [12] proposed and carried out an experimental study using vertical jute-cloth with using the latent heat of condensation accumulated between the glass-cover and the basin. This attempt results in an enhancement of solar still efficiency by 8% and a cumulative yield by 20% compared with a conventional still. Other authors have used absorber/evaporator materials such as charcoal cloth to enhance absorber capacity and hence to improve the productivity of tilted wick-type solar stills [13]. Abd El Kawi and Naim [14] used charcoal particles as absorber material with the aim of improving distillation yields. The system’s thermal inertia was reduced because of the capillary action by the charcoal which was partially immersed in water and because of its black color and surface roughness. Indeed, the presence of charcoal led to a remarkable reduction in start-up time for the distillation process. Rai et al [15]developed and carried out an experimental study on the effects of brackish water salinity and dye on the performance of a single basin-type solar still connected to a solar collector. The thermo-siphon and forced circulation modes were also investigated. The best resultswere obtained in the case of a coupled still under forced circulation using a film distillation process. Sayigh and El-Salam [16] tested and compared seven solar stills with various slopes and thicknesses of glass cover. The water trays in the stills were also covered with different heat absorbing materials such as sand, straw, stones, and charcoal. The researchers found that the best performing unit had a 20° glass-cover tiltand used black stones. This setup resulted in the highest output yield and reached 45% of distiller’s internal efficiency. Likewise, Akash et al [17] conducted experiments with different type of absorbent materials namely: dye, ink and black rubber mats. These experiments increased the yield of the solar still in the range of 35 % to 60 % relative to the base case. Abu-lhijleh and Rababa’h [18] placed sponge cubes over the saline water surface in the basin to increase the absorber surface area for the evaporation. This setup caused an increase in output by 18 %. Velmurugan et al. [19] studied and carried out a single-basin solar still using wick, fin and sponge. The use of sponge, wick and fin resulted in 15.3, 29 and 45.5% improvements in the yield, respectively. Using sponge material in the basin is not new. The sponge is an organic polymer compound in foam form; many researchers have usedsponges in solar stills withoutspecifying their chemical nature. There are several types of sponge (polyethylene, polyurethane…).Their physical characteristics may be similar but not identical. As previously cited [18-19], using pieces of blackened sponge increases the exchange surface and the absorption rate of radiation leading to a reduction of reflection; the pieces used are free to move in the heat absorbing water. However, in our case we used blackened layers (sheets) of polyurethane foam glued to the bottom of the basin. So, our design differs from earlier ones, because the bonding layer also contributes to heat storage. In addition, meteorological conditions and operating parameters vary from one place to another: the conditions applying to a still in Ouargla (Algeria) differ from those elsewhere. 4
2. Heat and mass transfer correlations. In 1961, Dunkle [20] first proposed a group of complete heat and mass transfer correlations based on a modified Grashoff number (Gr) to express the operating processes of basin type solar stills. Based on free convective heat transfer experiments of air in an enclosure, Zeshaoo [21] proposed an empirical relation between Nusselt (Nu) and Rayleigh (Ra) dimensionless numbers: ) For
; the heat transfer coefficient is expressed by: )
Since then, many other researchers [22-23] have studied heat and mass transfer. Based on experimental results, they proposed empirical correlations for solar stills. They made approximating assumptions when establishing their correlations, which is why their correlations may not be widely applicable. For example Adhikari’s correlation [24] considered the effect of the characteristic space between the absorber and condenser surfaces, but he ignored the fact that the relationship between the evaporative and free convective heat transfer coefficient would change with temperature. In Dunkle’s correlation [20], the characteristic space between the evaporation and condensation surfaces does not appear in calculating the free convective heat transfer coefficient. In summary, the theoretical correlations obtained by various researchers show considerable difference because of variations in their experimental conditions. 2.1 Heat and mass transfer in the “witness” still The “witness” unit (a conventional still unit without any sponge layer) differs from the units with sponge layers. In those units, the heat absorbing sponge is glued to the basin surface and completely immersed in brackish water. Therefore, the sponge is in thermal communication with the water and also with the basin plate. Before conducting energy balances on the various still components, we make the following assumptions: (1) steady-state operation; (2) constant solar irradiance over the time period within which the energy balance is made; (3) low thermal resistance of solar still materials; (4) the temperature gradient across the glass-cover is neglected, and (5) the initial temperature of each part of the solar still equals the initial ambient temperature. Thus, the energy balances made for instantaneous conditions are applicable from our following discussion. 2.1.1 Glass-cover Irradiative and convective heat transfer mechanisms and the coefficients connected to the external glass-cover and the ambient air are given by [25-26]. 2.1.1.1 Radiation Irradiative heat transfer coefficient is related to sky temperature a through the following equation: 5
)
)
)
The sky temperature (Ts) is expressed as a function of the ambient temperature (Ta ) according to the following equation: ) Ta and Ts are expressed in Kelvin degrees. 2.1.1.2 Convection Convective heat transfer coefficient is related to wind velocity (v) through the relations: 5,7 3,8v; hcga 0 ,8 6,15v ;
v 5m / s
(5)
v 5m / s
2.1.2Brackish water Irradiative, convective and evaporative heat transfer mechanisms between the brackish water and the glass-cover and the appropriate coefficients are given respectively by [26-27] as: (
)
)
) )
[( (
)(
)]
⁄
) )
)
2.1.3 Energy balance Energies balances have been given for essential still‘s components: 2.1.3.1 The glass-cover (condenser) Energies balances for the glass-cover are given in form of heat transfer coefficients by: (
)
(
)
)
with: ) and: ) 3.3.1.2 Brackish water 6
The differential equation describing the brackish water temperature as a function of time is given by: )
)
)
(
(
))
)
The depth of brackish water is connected with its mass (Mw) by the relation: ) 2.3.1.3 Absorber plate The energy balance for the basin is given by: )
)
)
(
)
))
Generally for a brackish water depth of L=0.03 m, the coefficient: hcwb = 100 W/(m2.K); and for: L ≥ 0.03 m, the coefficient: hcwb = 135 W/(m2.K) [28]. In our study we’ll use 0.005 m ≤ L≤ 0.10 m; so the average value of (hcwb) in this case is : 118 W/(m2.K). The convective heat transfer coefficient between the base and the ambient air is given by: hb = 0.80 W/(m2.K) [28]. Table.1 2.2 Heat and mass transfer in test stills (with a sponge layer). In the case of our modified still, a new material (sponge) has been added to the witness still..).Therefore, all the previous equations remain unchanged except those for the brackish water and the absorber plate, equations (12) and (14). The layer of sponge is filled with heat absorbing water. This means that they have almost the same temperature. Equations (12) and (14) become respectively: ) )
)
) (
( (
( )
)
(
(
))
))
) )
3. Experimental procedure 3.1. Construction of solar stills Four identical solar still prototypes were constructed and assembled, one of which was used as a witness unit (i.e. a reference unit without a sponge) while the parameters under investigation were applied to the other three units. The four units operate under recorded meteorological conditions, namely ambient temperature, solar irradiance and wind velocity. Fig.1 shows a cross section schematic of the single-slope basin solar still used in these experiments. The solar still support was made of 40 mm thick wood. Its basin (absorber) is a tray (480 x 370 x 30 mm) made of 3 mm thick galvanized metal. The absorbers of both stills were blackened on the surface to ensure maximum absorption of solar irradiance for effective heating of the brackish water. The base of each assembly was further glued with a 30 mm thick polystyrene insulation. The 3 mm thick removable glass cover of the stills was placed such that it makes an angle of 30° with the horizontal which is recommended for the Ouargla region. The glass cover was sealed tightly with silicone sealant to prevent any vapor leakage. 7
Local groundwater (3 g/l concentration) was supplied to each still with a low flow (0.8-1 kg/h) and an adjustable float was used to provide and maintain the desired water level in the still. Generally, in our experimental study, the yield of distillate was between 3 and 5 kg/(d.m2) During experiments the brackish water level is fixed at 0.5 cm. Distilled water is collected and then conducted out of the enclosure by plastic tubing along the lower edge of the glass. To enhance the output of the solar still by increasing the heat absorbing surface, almost homogeneous blackened layers of sponge with three different thickness (0.5 cm; 1.0 cm; and 1.5 cm with an uncertainty of ± 0.1cm) were cut and placed to cover the entire surface of the absorber of each test still (stills under study). The layers are blackened with mattepaint andsealed tightly by silicone sealant. Fig.1 3.2. Experimental procedure The stills were installed and tested at Ouargla University, southern Algeria, with the long axes of the stills facing south-north direction for maximum solar irradiance. All runs started at 9.00 AMand terminated at 5.00PMlocal time. During operations, the measurements of solar irradiance, temperatures of inner surface of the glass cover and that of brackish water in the basin were made regularly. The ambient temperature, wind velocity and distillate amount were also monitored. The raw data values were recorded at regular intervals throughout the duration of the tests. The experiment consisted of studying the effect of the sponge thickness on the average output of the still. Prior to starting the tests the following steps were performed: ithree pieces of sponge (thicknesses of approximately:0.5 cm, 1.0cm and 1.5 cm ± 0.1 cm) were cut using a layers of 2.0 cm thickness previously bought; iithe glass cover is removed from the distiller; and iiieach piece was blackened with matte black paint and homogeneously glued to the absorber plate with silicone to prevent it from floating. For good adhesion, the glass cover was also glued with silicone and left overnight before taking the measures.The results of each experimental were normalized: a productivity factor was calculated, which is simply the ratio of the yields between the test and the conventional (witness) units. This experimental practice was carried out on the four units throughout the entire run, thus giving a better indication of the units’ output and overcoming the problem of fluctuations/non-symmetry during the day. For all units, the measuring process was repeated 5 times during 5 consecutive days. The average value was then taken for each unit. 4. Results and discussion Typical measured solar irradiance and ambient temperature vs. local time are presented in Figure 2. It is observed that the temperature follows the same trend as the solar irradiance. They increase in the first half of the day until a maximum value is reached between 01:30 and 02:30 PM, before they start to decrease in the afternoon. The maximum values for solar irradiance is 905 W/m2 recorded at 02:00 PM with 39°C recorded for the ambient temperature. Fig.2
8
As displayed in Fig.3, at the beginning of experiment (i.e. between 09:00 AMand 10:00 AM) the productivity of the witness still exceeds that of all other test units because these latter begin to store energy in the sponge layer. However, starting at11:00 AM, the test unit with a thickness of 0.5 cm of sponge exceeds practically all other stills until the end of the experiment. This is explained by the fact that the heat storing capacity of 0.5 cm of sponge is lower than upper thickness; so the units with 0.5 cm of sponge start distilling before other test units while the other one’s are storing heat. At04:00 PM the amount of distilled water production by the 05 cm unit is lower than that of the unit with1.0 cm spongethickness; but after this time it becomes the best again. This is due to the large area of sponge pore and the inner heat storage systemwhich keeps the brackish water temperature high enough to produce distillate throughout the experiment.The productivity of 1.0 cm sponge thickness unit doesn’t exceed the witness still unit until 12:00 PM because its sponge thickness needs a longer storage time before saturation. The productivity of the unit with the greater spongethickness (1.5 cm) doesn’t exceed that of the witness unituntil 04:00 PM; as its sponge thickness needs more storage time than other sponge thickness units. After this time, and after heat storage saturation, this unit (i.e. 1.5 cm) begins its most effective distillation from close to sunset until later in the evening but its yield, as we shall see later, stays pretty much the same. . Fig.3 The effect of sponge thickness on hourly productivity and on total amount of distillate is displayed in Figure 3 and Figure 4.The sponge used is organic foam with very low thermal conductivity and significant heat capacity. Thus, it constitutes an important thermal reservoir. The roughness and porous structure of the sponge tinted with black matte painting generate a large exchange surface area and capillary action on the water.It can be seen from Fig.3 and Fig.4 that the yield’s enhancement is inversely proportional to thesponge thickness. The experimental results showed that the daily cumulative amount (outputs between 09:00 AM to 05:00 PM) of distilled water produced by the witness still was 3.048 kg/m2. In contrast, the units with sponge thicknesses of 1.5 cm, 1.0 cm and 0.5 cm had respective yieldsof: 2.135 kg/m2, 3.750 kg/m2 and 4.809 kg/m2.This result indicates an increase in two of the yields and a decrease in the other. The best improvement is recorded with the unit with 0.5 cm of sponge thickness. The yield decrease is observed with the unit still equipped with a 1.5 cm sponge. The water level in the basin is limited to 0.5 cm; so, the sponge of 0.5 cm thick is totally submerged in brackish water which is not the case for the two other test stills. In those, the water is distributed in a very thin film in the partially aerated pores which results in a larger exchange surface and increases the heat transfer coefficient by convection between the brackish water and the glass-cover. This analysis is valid for sponge thicknesses exceeding 0.5 cm. Nevertheless, in the test unit with the 1.5 cm thick sponge, the heat capacity of the storage medium becomes very important and the sponge needs large amounts of heat and more time to saturate before beginning distillation. As a result, the unit does not become an effective still until after sunset. The optimal sponge thickness is not exactly 0.5 cm but its thickness should be close to that value (0.5 ± 0.1 cm). Cutting sponge horizontally in thin and homogeneous sheets is difficult given its elasticity and its roughness. Because of this, and because of the limited number of the test unit (three), the number of experiments was limited. Moreover, it is almost impossible to have a sponge thinner than 0.5cm. The cut of a sheet greater than1.5cm thickwas not considered useful since the 1.5cm thick layer decreased the yield by 0.913 kg/(m2.day) compared to the witness unit: most of the solar irradiance absorbed was stored by the thick sponge layer and was not converted into evaporation heat.However, after sunset, nocturnal distillation was observed in all units.This was due to the heat storage of units with sponges 9
but also to the overall thermal inertia (i.e. it was true also for the witness still). Nocturnal yield is measured in the next morning for each unit before starting experiments.The mean values of nocturnal distillation yields recorded for the four units are: 0.201, 0.314, 0.505 and 0.097, kg/m 2 for the units with sponge thickness of: 0.5, 1.0, 1.5 cm and the witness respectively. These values of nocturnal distillation were not taken in consideration in the calculation of the total cumulative yield or in the calculation of unit efficiency andyield improvement. This is because we do not know the duration of nocturnal distillation (and it is hard to correlate with the inexistent solar radiation). However, the mean nocturnal yield value (from 05:00 PM to 09:00 AM) for each unit has been added to its daily cumulative yield (i.e. from 09:00 AM to 05:00 PM) when estimating the water production cost in paragraph 5. So, the total yields after adding the nocturnal yields become: 5.010, 4.064, 2.64 and 3.145 kg/m2 for the units with sponge thickness of: 0.5, 1.0, 1.5 cm and the witness respectively. Fig.4 The temperature difference (∆T=Tw –Tg), as a function of time, between the basin water and the glass cover of stills with different sponge thicknesses and witness, is displayed in Figure 5. The curves have the same trend. The temperature difference increases in the morning hours to reach a maximum value between 01:00 and 02:00 PM. After that, it starts to decrease again. At atmospheric pressure, water evaporation occurs even at low temperatures, below the boiling point of 100 ° C. For evaporation from the basin, the water molecules that escape to constitute humidity (steam) must be suppliedby heatfrom the basin or from the layer of sponge. The pressure of the airsteam mixture within the distiller remains constant and is always equal to the outside pressure (atmospheric pressure) because there is a bond and a contact with the water tank supply via the level adjuster or the overflow hole. The condensation of vapor on the inner surface of glass cover results in a slight and temporary decrease in the internal pressure. This is rapidly compensated by water evaporation and the process is repeated as long as there is energy in the basin, and as long as there is a temperature difference between the basin and the condenser. The heat storage medium has a dual role. First, it absorbs the excess heat that causes internal overheating of distiller which would cause a performance decrease. Secondly, it feeds the liquid water molecules with energy especially in low solar-intensity conditions or overnight. Thus, the heat storage layer serves to keep a continual temperature difference between the brackish water and the glass-cover. Generally, to have distillation, the basin water temperature must always be higher than that of the glass cover for all units. However, we observethatbetween 09:00 AMand 10:00 AM; the water temperature for units with 1.0 and 1.5 cm sponge thickness was lower than that of their glass cover. This phenomenon is explained by the fact that a thick layer of sponge (storage system) absorbs solar energy and avoids the increase in water temperature. Thus the glass cover temperature becomes higher. The value of the temperature difference between the glass coverand the absorber is directly proportional to the quantity of distilled water. Figure 5 clarifies the results obtained fromFigures 3 and 4. The experiments which have been carried out over 5 days allowed us to have an average for each presented parameter. We must point out that during these 5 days the classification and the ranking of recorded improvements remain unchanged despite the nocturnal yields recorded. So, from these results we can say and confirm that it is unnecessary to have sheets of thicknesses with more than 1.5 cm since the best result is closer to a 0.5 ± 0.1cm thickness. Fig.5
10
5. Estimatedcalculation of ameliorated still and water production costs 5.1 Sponge layercost We will use the price of the optimum still (i.e. with a 0.5cmlayer) which produced a total of 5.01kg/m2 of distillate. Thesponge layer cost includesthe cost of gluing and painting. The cost is estimated in Algerian Dinars (DZD) and converted into euro currency using the current: (1 € = 118 DZD). The construction cost of a conventional still (i.e. our witness still) is estimated to be 6810 DZD/m2 i.e. (57.72 €/m2). In contrast, the cost of our most efficient sponge unit still is 7410 DZD/m2 i.e. (62.8 €/m2). So, improving the quality of the still by adding 0.5 cm sponge layer costs 8.81% more. However, it will be seen below (paragraph 5.2) that the difference between the price of our improved still and that of the conventional unit remains negligible, especially compared to the benefits obtained. 5.2 Distilled water productioncost The lifetimeof a singleslope basin still that well constructed is estimated at 20 years. So the estimated constructionprice becomes (62.8/20) €/year.m2. Generally, a conventional solar still does not require much maintenance and care (in our case one day every 6 months). So, the cost of maintenance is estimated at 10% per year of the construction cost. Thus, the total cost of the best unit becomes 3.46€/year. Finally, the cost price of distilled water is given in (DZD/kg) by: [Total unit cost (DZD/year) /yield (kg/day)/363 (day/year)]+ [the cost price of brackish water (DZD/kg)]. Therefore, the cost of distilled water becomes: (408/5.010/363)+0.31 = 0.54 DZD/kg or 4.58 10-3 €/kg. However, the cost of distilled water produced through other energy means is closer to13 DZD/kg (or 0.11 €/kg). In other words, producing distilled water using a still with sponge layers is 24 times cheaper than that produced through other means such as electricity generated by fossil energy. The best improvement obtained with our blackened sponge still (without taking into account nocturnal distillation) is 57.77%. It becomes 59.3% after adding nocturnal distillation amount. The equivalent gain in distilled water output is about 1.8 kg/ (m2.day) i.e. 653.4 kg/ (m2.year).This implies that our improved still has a production equivalent to 1.6 times that of a conventional unit. Given the concentrations of salts in our brackish water, we need to mix two volumes of distilled water with one volume of brackish water to obtain three volumes of usable fresh water; so, with 59.3% of improvement, an additional of 980 kg/ m2.year of drinkable water is produced. The average daily consumption of drinkable water is (2-3) kg/person; thus, the previous supplement of 980 kg/ (m2.year) corresponds approximately to the annual drinking water consumption for one person. So, each square meter can supply fresh water for one more person. From our study, the optimum still can produce more than 5 kg/ (m2.day). Therefore, it can supply almost 3 people. From this latest result we can deduce that a square meter of distillation unit can supply 3 persons with drinking water. In large scale, a town of 300,000 inhabitants, such as Ouargla city, can be supplied with drinking water by the use of a surface area of 1000m x100m of solar distillation units with a brackish water flow of about 75 x 104 kg / day a cost not exceeding 2500€/day. 6. Conclusion With ever increasing population and rapid industrialization growth, there is a great demand for freshwater or drinking water. Arid zones and remote areas in southern Algeria communities are rich in solar energy and groundwater. These regions present a favorable advantage for solar desalination. 11
These regionsdepend on underground saline water for drinking, but thegroundwater is often too salty and is not recommended for human consumption.Small-scale brackish water solar distillation can make a considerable contribution in providing fresh water supplies for rural communities. Four basin-type solar stills were constructed and experimentally tested under the climatic conditions of the arid zone of southern Algeria in Ouargla University. The idea was to study the possibility of converting brackish water to soft water using a heat storage system. Our objective was to increase the yield of a solar still by improving its absorber performance through the use of an inner heat storage system.This was obtained by coating the entire horizontal absorber surface with a layer of blackened sponge.The latter is a porous material with a large specific surface whichleads to high mass and heat transfer effectiveness. The layer plays the role of a heat storage medium which keeps the operating temperature of the still high enough to produce more distilled water even in the case of cloudy weather or atnight. The experiments consisted of studying the effect of the sponge thickness on the productivity of the solar stills. Results obtained, showed that: sponge layers thicknesses of 0.5 cm and 1.0 cm increased the still yield by of 57.77 % and 23.03 %, respectively. However, sponge thickness of 1.5 cm decreased the yield by 29.95 % compared with a conventional basin still. A simple economic analysis shows that solar distillation using a thin blackened layer of sponge as heat storage medium in absorber is efficient and produces a much improved yield of distilled waterat very lowcost compared to other techniques. In summary, the development of any region is generally based on the existence of two essential factors: water and energy. These two factors (abundant brackish water and solar energy) are characteristics of the Ouargla region. From an industrial viewpoint, brackish water is not recommended because of the resulting corrosion or clogging can degrade the equipment. Distilled or fresh water are needed also for industrial uses. According to our study, large-scale solar distillation of groundwater can contribute to the development of this region given the estimated price of a kilogram of distilled water supplied by the cleanest and cheapest energy possible: solar energy.
7. References [1] The world Bank-Bank-Netherlands water partnership, seawater and brackish water desalination in the Middle east, North Africa and Central Asia, Final report Annex1, Algeria (2004). [2] B. Bouchekima; A small solar desalination plant for the production of drinking water in remote arid areas of southern Algeria, Desalination. 156 (2003) 353-354. [3] M.H. Sellami, H. Bouguettaia, D. Bechki, M. Zeroual, S. Kachi, S. Boughali, B. Bouchekima and H. Mahcene, Effect of absorber coating on the performance of a solarstill in the region of Ouargla (Algeria), Desalination and Water Treatment. 21 (2013) 1-8. [4] D. Bechki, H. Bouguettaia, J. Blanco-Galvez, S. Babay, B. Bouchekima, S. Boughali, H. Mahcene; Effect of partial intermittent shading on the performance of a simple basin solar Still in South Algeria, Desalination. 260 (2010) 65-69. [5] M.A. Samee,U.K. Mirza,T.Majeed and N. Ahmad; Design and performance of a simple basin solar still, Renewable Sustain Energy Rev. 11 (2007) 543-549. [6] A.A. Youssif and A.F. Ismail; Theoretical and experimental investigation of a novel multistage evacuated solar still, Journal of Solar Energy. 127 (2005) 381-385. [7] A. Safwat, M. Abdelkader, and A. Abdelmotalip; Parameters affecting solar still productivity, Energy Conversion Management 41 (2000) 1791-1809. [8] R. Dev, S.A Abdul-Wahab and GN Tiwari; Performance study of the inverted absorber solar still with water depth and total dissolved solid, Desalination. 88 (2011) 252-264. 12
[9] S. Aboul-Enein, A. A. El-sebaii and E. El Bialy; Investigation of a single basin solar still with deep basins, RenewableEnergy. 14 (1998) 299-305. [10] F.F. Tabrizi, M. Dashtbanand H. Moghaddam; Experimental investigation of a weir-type cascade solarstill with built-in latent heat thermal Energy storage system, Desalination.260 (2010) 248-253. [11] M. Dashtbanand F.F. Tabrizi; Thermal analysis of weir-type cascade solar still integrated with PCMStorage, Desalination. 279 (2011) 415-422. [12] M. Sakthivel,S. Shunmugasandaram and T. Alwarsamy; An experimental study on a regenerative solar still with energy storage medium-jute cloth. Desalination 264 (2010) 24-31. [13] J.T. Mahdi,B. E Smith and A. O. Sharif; An experimental wick-type solar still system: Design and construction, Desalination 267 (2011) 233-238. [14] M.A Abd El Kawi and M.M. Naim; Non-conventional solar still part1. Non-conventional solar still with charcoal particles as absorber medium, Desalination 153 (2002) 55-64. [15] S.N. Rai, D.K. Dutt and G.N Tiwari; Some experimental studies of a single basin solar still, Energy conversion Management. 30 (1990) 149-153. [16] A.A.M. Sayighand E.M.A. El-Salam; Optimum design of a single slope solar still in Riyadh, Saudi Arabia, Revue d’Heliotechnique.1 (1977) 40-44. [17] B.A. Akash,M.S Mohsen, O. Osta, and Y. Elayan.; Experimental evaluation of a single-basin solar still using different absorbing materials, Journal of Renewable energy. 14 (1998) 307-310. [18] B. Abu-lhijlehand H.M. Rababa’h ; experimental study of solar still with sponge cubes in basin, Journal of Energy Conversion & management. 44 (2003) 1411-1418. [19] V.Velmurugan , M.Gopalkrishnan and R.Raghu. Single basin solar still with fin for enhancing productivity. Energy Conversion and Management. 49 (2008) 2602-2608. [20] R.V. Dunkle, Solar water distillation, The roof type still and a multiple effect diffusion still, International Developments in Heat Transfer, Part V, University of Colorado (1961) 895. [21] C. Zeshaoo, Natural convection heat Transfer across air Layers at various angles of Inclination, Engineering thermo-physics 20 (1984) 211. [22] MA.S. Malik, G.N. Tiwari, A. Kumar and M.S. Sodha, Solar distillation, Oxford, UK, Pergamon Press (1982) 8–17. [23] A.T. Shawaqfeh and M.M. Farid, New development in the theory of heat and mass transfer in solar stills, Solar Energy 35 (1995) 527. [24] R.S. Adhikari, and A. Kumar, Estimation of mass-transfer rates in solar stills, International Journal of Energy Research 44 (1990)737. [25] W.C. Swinbank, Long-Wave Radiation from Clear Skies, Quarterly Journal of the RoyalMeteorological Society, 89 (1963) 339. [26] O.O.Badran and M.A.K. Mazen, Evaluating of thermal performance of a single slope solar still, Heat Mass Transfer.43 (2007) 985-995. [27] R.V. Dunkle, International Developments in Heat Transfer, Conference, Part 5, Denver (196l) 895. [28] A.K. Singh, Optimization of Orientation for Higher Yield of Solar Still for Given Location, Energy conversion and Management 36 (1995)175-187.
13
Glass Cover Wood support
Distilled water
Saline Water
Sponge layer Insolation
Absorber
Fig.1 Schematic diagram of solar still
Fig.2. Typical measured solar irradiance and ambient temperature time history curves
14
Saline water entrance
Fig.3. Effect of sponge thickness on the distilled water productivity
Fig.4. Effect of sponge thickness on the hourly cumulative amount of distillate
15
Fig.5. Curves of temperature difference between glass cover and basin water for all units under study
Table.1. Thermo-physical and configuration parameters of solar still parts
Glass-cover Saline Water Basin
Insulation
Blackened Sponge
Highlights - Solution to the problem of fresh water shortage in sunny remote areas. - Improvements in solar stills using cheap and available materials. - More efficient and longer distillation times. - Production of distilled water at low cost. - Encouraging results.
16
PII: DOI: Reference:
S2352-801X(17)30053-X http://dx.doi.org/10.1016/j.gsd.2017.05.004 GSD52
To appear in: Groundwater for Sustainable Development Received date: 1 November 2015 Revised date: 8 June 2016 Accepted date: 29 May 2017 Cite this article as: M.H. Sellami, T. Belkis, M.L. Ali Ouar, S.D. Meddour, H. Bouguettaia and K. Loudiyi, Improvement of Solar Still Performanceby Covering Absorber with Blackened Layers of Sponge, Groundwater for Sustainable Development, http://dx.doi.org/10.1016/j.gsd.2017.05.004 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting galley proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Improvement of Solar Still Performanceby Covering Absorber with Blackened Layers of Sponge M.H. Sellamia*, T. Belkisa, M.L. Ali Ouar a, S.D. Meddour a , H. Bouguettaiab and K. Loudiyic a
Process Engineering Laboratory (PEL) Email: [email protected] b Laboratory of New and Renewable Energy in Arid Zones (LENREZA), Ouargla University, 30000 Algeria. c Renewable Energies Laboratory (REL), Al Akhawayne University, Ifrane, Morocco. * Corresponding author
Abstract Algeria has been listed among top countries affected by a shortage of fresh drinkable water. Solar desalination can be used to produce fresh water from brackish water to supply isolated, low-density, population areas located in southern Algeria where solar energy and underground saline water are abundant. This article aims to improve the yield of a solar still by improving absorber performance through the use of an added inner heat storage system. To do this, we tested covering the absorber surface with layers of blackened sponge. The resultant heat storage is used to keep the operating temperature of the absorber high enough to produce distilled water when solar irradiance is low or during night time. Four small-scale solar powered distillation pilot units were set up and operated.The experiments carried out in the “Process Engineering Laboratory of Ouargla University” studied the effect of sponge thickness on the productivity of the solar still. The results obtained showed that a 0.5cm sponge thickness increased the yield by 57.77 % i.e. 58%, relative to the baseline case (i.e. with no blackened sponge added).In contrast, asponge thickness of 1.0cm resulted in a yield improvement of only 23.03 %, whereas a sponge thickness of 1.5cm resulted in a decreased yield of 29.95 % i.e. 30% (relative to the baseline case). Keywords: Desalination; Solar energy; Single slope basin still;Sponge; Storage medium.
Nomenclature Symbols Basin area ( m 2 ) Ab Ag
Glass-cover area ( m 2 )
Ca
Specific heat of ambient air ( J / kg.K )
Cw
Average specific heat of brackish water ( 4190J / kg.K ) E Sponge thickness (m). € DZD Euro Algerian Dinar Solar radiation at time (t ) , ( W / m 2 ) I (t ) Grashoff number Gr
h1 h2 hb
Total heat transfer coefficient between brackish water surface and glass cover ( W / m 2 .K ) Total heat transfer coefficient between glass cover and ambient air ( W / m 2 .K ) Convective heat transfer coefficient throughout basin liner ( W / m 2 .K ) 1
hca
Convective heat transfer coefficient between glass-cover and the ambient ( W / m 2 .K )
hcwb hcwg
hcws
Convective heat transfer coefficient between brackish water and the basin plate ( W / m 2 .K )
Convective heat transfer coefficient between brackish water surface and glass cover ( W / m 2 .K ) Convective heat transfer coefficient between brackish water and the sponge layer ( W / m 2 .K ) Evaporative heat transfer coefficient between brackish water and glass cover ( W / m 2 .K ) hew hra
Irradiative heat transfer coefficient between glass-cover and the ambient ( W / m 2 .K )
hrwg
Irradiative heat transfer coefficient between brackish water surface and glass cover ( W / m 2 .K )
hsp
Convective heat transfer coefficient throughout sponge layer ( W / m 2 .K )
L Lv
water depth in basin (m) Latent heat of water vaporization ( J / kg) Mass of brackish water in basin (kg) Nusselt Number. Partial water vapour pressure at inner of glass cover temperature ( N / m 2 ) .
Mw Nu Pg Pw Ra t Ta
Tb Tg
Partial water vapour pressure at water surface temperature ( N / m 2 ) . Rayleigh Number. Time (s ) . Ambient temperature (K ) Temperature of the base (K ) Glass-cover temperature (K )
Ts
Tsp Temperature of Sky (K ) Sponge temperature (K) Brackish water temperature (K ) Tw ∆T (Tw – Tg): Temperature difference between brackish water and glass-cover (°C). v Wind velocity ( m / s ). X Insulation, basin and glass-cover thickness m Greek Absorptivity Glass-cover tilt angle. Emissivity Transmissivity Stephan-Boltzmann constant 5,67.10 8W .m. 2 K 4
Density of brackish water
)
Subscript
b Exp
Basin Experimental 2
g
w i S sp
Glass-cover Water Insulation Sky Sponge
1. Introduction Unlike other distillation methods, solar stills use environmentally-friendly solar energy to remove salts from saline or brackish water. Solar desalination is a good way of supporting the needs of small communities for fresh water produced with clean and cheap energy. Solar distillation is particularly useful for arid areas such as southern Algeria where solar energy and brackish water are abundant. These regions depend on groundwater for drinking, but unfortunately this water is often too salty to be drinkable. Instead, brackish groundwater with high salinity (3 g/l concentration) is often encountered [1]. Indeed, sufficient access to water for drinking and industrial uses is an increasing problem in Algeria. In the Saharan region, an area close to two million km2, groundwater is available in large quantity: about 60 × 103 billion m3or approximately 76% of Algerian reserves. This water is difficult to use and is not renewable. Only4 to 5 billion m3are exploited annually because most of the water is brackish. Salinity levels are variable and some groundwaters have salinity levels of up to 8 g/l. This highly exceeds the maximum 550 ppm levelallowed for human consumption [2-3].In northern Algeria, groundwater reserves are estimated to be 1.8 billion m3.If these waters could be used; they could be importantin improving development and living conditions in the region. Overall, Algeria receives an average of 65 billion m3/yof rain, of which only 3 billionm3/y supply groundwater reserves, 47 billion m3/y evaporates or transpires into the atmosphere, and 15 billion m3/y provides base flow to surface waters [1-4].Algeria currently ranks 14th worldwide among countries suffering from lack of water. If effective solutions are not found to provide a greater supply of freshwater, Algeria’s situation is predicted to get worse. By the year 2025 [2], Algeria is predicted to become the 6th country most affected by a lack of freshwater resources. The location for our experiments is Ouargla city (31.95°N latitude, 5.40°E longitude), located in southern Algeria at 141 m above the sea level. It has a high solar irradiance of about 2260 kWh/(m2.year) and 3400 sunshine hours per year [2-3]. Consequently, solar desalination technology is advantageous there. Extensive research has been conducted on different theoretical and experimental methods to improve and enhance productivity of solar still units [5-6]. Several authors have studied the effect of water level on absorber performance. The latter has been shown to be an important factor affecting basin-still productivity because thin films lead to high heat absorption. Still performance was shown to increase with thinner saline water films because of greater heat and mass transfer effectiveness [7-8and 9]. Other investigators, Farshchi et al [10], Dashtban and Tabrizi [11], studied and set up a cascade solar still fitted with weirs(weir-type solar still): paraffin wax was used to provide high heat energy storage beneath the absorber plate, keeping operating temperatures high enough to produce distilled water overnight or during a lack of sunshine. Theweir-type solar still is based on providing asuitable distribution of the feed water on a large evaporation surface witha high residence time. Generally, the blackened absorber is step shaped: each step has a horizontal and a vertical surface. The feed water flows slowly from the upper step to a lower step while crossing baffles or vertical fins placed on each step. These metallic fins increase the absorber area andthe water residence time in each step and therefore improve the heat and mass transfer. Also, the low flow of feed water maintains the depth of water in a thin film which leads to 3
good mass and heat transfer between the absorber and the condenser. These are usually very close to each other. The water temperature and salt concentration increasesas the water travels from top to bottom. Hence, the evaporation rate becomes more important in the last step and salt deposits may occur. Sellami et al [3] carried out an experimental study by laminating the basin with layers of blackened alluvial sand as a heat storage medium. Theauthors studied the effect of sand mass and particle diameter. The results revealed that for a fixed mass of sand on the absorbing layer, there was a significant improvement in output when fine sand particles were used. Due to thestorage medium and to photocatalytic effects, the yield recorded for a fixed particle diameter (0.08 mm) increases with an increase in sand mass to an optimum value of 2.268 kg of sand/m2 absorber surface. The improvement recorded was 43.28% compared with that of conventional still. For the same reason, Sakthivel et al [12] proposed and carried out an experimental study using vertical jute-cloth with using the latent heat of condensation accumulated between the glass-cover and the basin. This attempt results in an enhancement of solar still efficiency by 8% and a cumulative yield by 20% compared with a conventional still. Other authors have used absorber/evaporator materials such as charcoal cloth to enhance absorber capacity and hence to improve the productivity of tilted wick-type solar stills [13]. Abd El Kawi and Naim [14] used charcoal particles as absorber material with the aim of improving distillation yields. The system’s thermal inertia was reduced because of the capillary action by the charcoal which was partially immersed in water and because of its black color and surface roughness. Indeed, the presence of charcoal led to a remarkable reduction in start-up time for the distillation process. Rai et al [15]developed and carried out an experimental study on the effects of brackish water salinity and dye on the performance of a single basin-type solar still connected to a solar collector. The thermo-siphon and forced circulation modes were also investigated. The best resultswere obtained in the case of a coupled still under forced circulation using a film distillation process. Sayigh and El-Salam [16] tested and compared seven solar stills with various slopes and thicknesses of glass cover. The water trays in the stills were also covered with different heat absorbing materials such as sand, straw, stones, and charcoal. The researchers found that the best performing unit had a 20° glass-cover tiltand used black stones. This setup resulted in the highest output yield and reached 45% of distiller’s internal efficiency. Likewise, Akash et al [17] conducted experiments with different type of absorbent materials namely: dye, ink and black rubber mats. These experiments increased the yield of the solar still in the range of 35 % to 60 % relative to the base case. Abu-lhijleh and Rababa’h [18] placed sponge cubes over the saline water surface in the basin to increase the absorber surface area for the evaporation. This setup caused an increase in output by 18 %. Velmurugan et al. [19] studied and carried out a single-basin solar still using wick, fin and sponge. The use of sponge, wick and fin resulted in 15.3, 29 and 45.5% improvements in the yield, respectively. Using sponge material in the basin is not new. The sponge is an organic polymer compound in foam form; many researchers have usedsponges in solar stills withoutspecifying their chemical nature. There are several types of sponge (polyethylene, polyurethane…).Their physical characteristics may be similar but not identical. As previously cited [18-19], using pieces of blackened sponge increases the exchange surface and the absorption rate of radiation leading to a reduction of reflection; the pieces used are free to move in the heat absorbing water. However, in our case we used blackened layers (sheets) of polyurethane foam glued to the bottom of the basin. So, our design differs from earlier ones, because the bonding layer also contributes to heat storage. In addition, meteorological conditions and operating parameters vary from one place to another: the conditions applying to a still in Ouargla (Algeria) differ from those elsewhere. 4
2. Heat and mass transfer correlations. In 1961, Dunkle [20] first proposed a group of complete heat and mass transfer correlations based on a modified Grashoff number (Gr) to express the operating processes of basin type solar stills. Based on free convective heat transfer experiments of air in an enclosure, Zeshaoo [21] proposed an empirical relation between Nusselt (Nu) and Rayleigh (Ra) dimensionless numbers: ) For
; the heat transfer coefficient is expressed by: )
Since then, many other researchers [22-23] have studied heat and mass transfer. Based on experimental results, they proposed empirical correlations for solar stills. They made approximating assumptions when establishing their correlations, which is why their correlations may not be widely applicable. For example Adhikari’s correlation [24] considered the effect of the characteristic space between the absorber and condenser surfaces, but he ignored the fact that the relationship between the evaporative and free convective heat transfer coefficient would change with temperature. In Dunkle’s correlation [20], the characteristic space between the evaporation and condensation surfaces does not appear in calculating the free convective heat transfer coefficient. In summary, the theoretical correlations obtained by various researchers show considerable difference because of variations in their experimental conditions. 2.1 Heat and mass transfer in the “witness” still The “witness” unit (a conventional still unit without any sponge layer) differs from the units with sponge layers. In those units, the heat absorbing sponge is glued to the basin surface and completely immersed in brackish water. Therefore, the sponge is in thermal communication with the water and also with the basin plate. Before conducting energy balances on the various still components, we make the following assumptions: (1) steady-state operation; (2) constant solar irradiance over the time period within which the energy balance is made; (3) low thermal resistance of solar still materials; (4) the temperature gradient across the glass-cover is neglected, and (5) the initial temperature of each part of the solar still equals the initial ambient temperature. Thus, the energy balances made for instantaneous conditions are applicable from our following discussion. 2.1.1 Glass-cover Irradiative and convective heat transfer mechanisms and the coefficients connected to the external glass-cover and the ambient air are given by [25-26]. 2.1.1.1 Radiation Irradiative heat transfer coefficient is related to sky temperature a through the following equation: 5
)
)
)
The sky temperature (Ts) is expressed as a function of the ambient temperature (Ta ) according to the following equation: ) Ta and Ts are expressed in Kelvin degrees. 2.1.1.2 Convection Convective heat transfer coefficient is related to wind velocity (v) through the relations: 5,7 3,8v; hcga 0 ,8 6,15v ;
v 5m / s
(5)
v 5m / s
2.1.2Brackish water Irradiative, convective and evaporative heat transfer mechanisms between the brackish water and the glass-cover and the appropriate coefficients are given respectively by [26-27] as: (
)
)
) )
[( (
)(
)]
⁄
) )
)
2.1.3 Energy balance Energies balances have been given for essential still‘s components: 2.1.3.1 The glass-cover (condenser) Energies balances for the glass-cover are given in form of heat transfer coefficients by: (
)
(
)
)
with: ) and: ) 3.3.1.2 Brackish water 6
The differential equation describing the brackish water temperature as a function of time is given by: )
)
)
(
(
))
)
The depth of brackish water is connected with its mass (Mw) by the relation: ) 2.3.1.3 Absorber plate The energy balance for the basin is given by: )
)
)
(
)
))
Generally for a brackish water depth of L=0.03 m, the coefficient: hcwb = 100 W/(m2.K); and for: L ≥ 0.03 m, the coefficient: hcwb = 135 W/(m2.K) [28]. In our study we’ll use 0.005 m ≤ L≤ 0.10 m; so the average value of (hcwb) in this case is : 118 W/(m2.K). The convective heat transfer coefficient between the base and the ambient air is given by: hb = 0.80 W/(m2.K) [28]. Table.1 2.2 Heat and mass transfer in test stills (with a sponge layer). In the case of our modified still, a new material (sponge) has been added to the witness still..).Therefore, all the previous equations remain unchanged except those for the brackish water and the absorber plate, equations (12) and (14). The layer of sponge is filled with heat absorbing water. This means that they have almost the same temperature. Equations (12) and (14) become respectively: ) )
)
) (
( (
( )
)
(
(
))
))
) )
3. Experimental procedure 3.1. Construction of solar stills Four identical solar still prototypes were constructed and assembled, one of which was used as a witness unit (i.e. a reference unit without a sponge) while the parameters under investigation were applied to the other three units. The four units operate under recorded meteorological conditions, namely ambient temperature, solar irradiance and wind velocity. Fig.1 shows a cross section schematic of the single-slope basin solar still used in these experiments. The solar still support was made of 40 mm thick wood. Its basin (absorber) is a tray (480 x 370 x 30 mm) made of 3 mm thick galvanized metal. The absorbers of both stills were blackened on the surface to ensure maximum absorption of solar irradiance for effective heating of the brackish water. The base of each assembly was further glued with a 30 mm thick polystyrene insulation. The 3 mm thick removable glass cover of the stills was placed such that it makes an angle of 30° with the horizontal which is recommended for the Ouargla region. The glass cover was sealed tightly with silicone sealant to prevent any vapor leakage. 7
Local groundwater (3 g/l concentration) was supplied to each still with a low flow (0.8-1 kg/h) and an adjustable float was used to provide and maintain the desired water level in the still. Generally, in our experimental study, the yield of distillate was between 3 and 5 kg/(d.m2) During experiments the brackish water level is fixed at 0.5 cm. Distilled water is collected and then conducted out of the enclosure by plastic tubing along the lower edge of the glass. To enhance the output of the solar still by increasing the heat absorbing surface, almost homogeneous blackened layers of sponge with three different thickness (0.5 cm; 1.0 cm; and 1.5 cm with an uncertainty of ± 0.1cm) were cut and placed to cover the entire surface of the absorber of each test still (stills under study). The layers are blackened with mattepaint andsealed tightly by silicone sealant. Fig.1 3.2. Experimental procedure The stills were installed and tested at Ouargla University, southern Algeria, with the long axes of the stills facing south-north direction for maximum solar irradiance. All runs started at 9.00 AMand terminated at 5.00PMlocal time. During operations, the measurements of solar irradiance, temperatures of inner surface of the glass cover and that of brackish water in the basin were made regularly. The ambient temperature, wind velocity and distillate amount were also monitored. The raw data values were recorded at regular intervals throughout the duration of the tests. The experiment consisted of studying the effect of the sponge thickness on the average output of the still. Prior to starting the tests the following steps were performed: ithree pieces of sponge (thicknesses of approximately:0.5 cm, 1.0cm and 1.5 cm ± 0.1 cm) were cut using a layers of 2.0 cm thickness previously bought; iithe glass cover is removed from the distiller; and iiieach piece was blackened with matte black paint and homogeneously glued to the absorber plate with silicone to prevent it from floating. For good adhesion, the glass cover was also glued with silicone and left overnight before taking the measures.The results of each experimental were normalized: a productivity factor was calculated, which is simply the ratio of the yields between the test and the conventional (witness) units. This experimental practice was carried out on the four units throughout the entire run, thus giving a better indication of the units’ output and overcoming the problem of fluctuations/non-symmetry during the day. For all units, the measuring process was repeated 5 times during 5 consecutive days. The average value was then taken for each unit. 4. Results and discussion Typical measured solar irradiance and ambient temperature vs. local time are presented in Figure 2. It is observed that the temperature follows the same trend as the solar irradiance. They increase in the first half of the day until a maximum value is reached between 01:30 and 02:30 PM, before they start to decrease in the afternoon. The maximum values for solar irradiance is 905 W/m2 recorded at 02:00 PM with 39°C recorded for the ambient temperature. Fig.2
8
As displayed in Fig.3, at the beginning of experiment (i.e. between 09:00 AMand 10:00 AM) the productivity of the witness still exceeds that of all other test units because these latter begin to store energy in the sponge layer. However, starting at11:00 AM, the test unit with a thickness of 0.5 cm of sponge exceeds practically all other stills until the end of the experiment. This is explained by the fact that the heat storing capacity of 0.5 cm of sponge is lower than upper thickness; so the units with 0.5 cm of sponge start distilling before other test units while the other one’s are storing heat. At04:00 PM the amount of distilled water production by the 05 cm unit is lower than that of the unit with1.0 cm spongethickness; but after this time it becomes the best again. This is due to the large area of sponge pore and the inner heat storage systemwhich keeps the brackish water temperature high enough to produce distillate throughout the experiment.The productivity of 1.0 cm sponge thickness unit doesn’t exceed the witness still unit until 12:00 PM because its sponge thickness needs a longer storage time before saturation. The productivity of the unit with the greater spongethickness (1.5 cm) doesn’t exceed that of the witness unituntil 04:00 PM; as its sponge thickness needs more storage time than other sponge thickness units. After this time, and after heat storage saturation, this unit (i.e. 1.5 cm) begins its most effective distillation from close to sunset until later in the evening but its yield, as we shall see later, stays pretty much the same. . Fig.3 The effect of sponge thickness on hourly productivity and on total amount of distillate is displayed in Figure 3 and Figure 4.The sponge used is organic foam with very low thermal conductivity and significant heat capacity. Thus, it constitutes an important thermal reservoir. The roughness and porous structure of the sponge tinted with black matte painting generate a large exchange surface area and capillary action on the water.It can be seen from Fig.3 and Fig.4 that the yield’s enhancement is inversely proportional to thesponge thickness. The experimental results showed that the daily cumulative amount (outputs between 09:00 AM to 05:00 PM) of distilled water produced by the witness still was 3.048 kg/m2. In contrast, the units with sponge thicknesses of 1.5 cm, 1.0 cm and 0.5 cm had respective yieldsof: 2.135 kg/m2, 3.750 kg/m2 and 4.809 kg/m2.This result indicates an increase in two of the yields and a decrease in the other. The best improvement is recorded with the unit with 0.5 cm of sponge thickness. The yield decrease is observed with the unit still equipped with a 1.5 cm sponge. The water level in the basin is limited to 0.5 cm; so, the sponge of 0.5 cm thick is totally submerged in brackish water which is not the case for the two other test stills. In those, the water is distributed in a very thin film in the partially aerated pores which results in a larger exchange surface and increases the heat transfer coefficient by convection between the brackish water and the glass-cover. This analysis is valid for sponge thicknesses exceeding 0.5 cm. Nevertheless, in the test unit with the 1.5 cm thick sponge, the heat capacity of the storage medium becomes very important and the sponge needs large amounts of heat and more time to saturate before beginning distillation. As a result, the unit does not become an effective still until after sunset. The optimal sponge thickness is not exactly 0.5 cm but its thickness should be close to that value (0.5 ± 0.1 cm). Cutting sponge horizontally in thin and homogeneous sheets is difficult given its elasticity and its roughness. Because of this, and because of the limited number of the test unit (three), the number of experiments was limited. Moreover, it is almost impossible to have a sponge thinner than 0.5cm. The cut of a sheet greater than1.5cm thickwas not considered useful since the 1.5cm thick layer decreased the yield by 0.913 kg/(m2.day) compared to the witness unit: most of the solar irradiance absorbed was stored by the thick sponge layer and was not converted into evaporation heat.However, after sunset, nocturnal distillation was observed in all units.This was due to the heat storage of units with sponges 9
but also to the overall thermal inertia (i.e. it was true also for the witness still). Nocturnal yield is measured in the next morning for each unit before starting experiments.The mean values of nocturnal distillation yields recorded for the four units are: 0.201, 0.314, 0.505 and 0.097, kg/m 2 for the units with sponge thickness of: 0.5, 1.0, 1.5 cm and the witness respectively. These values of nocturnal distillation were not taken in consideration in the calculation of the total cumulative yield or in the calculation of unit efficiency andyield improvement. This is because we do not know the duration of nocturnal distillation (and it is hard to correlate with the inexistent solar radiation). However, the mean nocturnal yield value (from 05:00 PM to 09:00 AM) for each unit has been added to its daily cumulative yield (i.e. from 09:00 AM to 05:00 PM) when estimating the water production cost in paragraph 5. So, the total yields after adding the nocturnal yields become: 5.010, 4.064, 2.64 and 3.145 kg/m2 for the units with sponge thickness of: 0.5, 1.0, 1.5 cm and the witness respectively. Fig.4 The temperature difference (∆T=Tw –Tg), as a function of time, between the basin water and the glass cover of stills with different sponge thicknesses and witness, is displayed in Figure 5. The curves have the same trend. The temperature difference increases in the morning hours to reach a maximum value between 01:00 and 02:00 PM. After that, it starts to decrease again. At atmospheric pressure, water evaporation occurs even at low temperatures, below the boiling point of 100 ° C. For evaporation from the basin, the water molecules that escape to constitute humidity (steam) must be suppliedby heatfrom the basin or from the layer of sponge. The pressure of the airsteam mixture within the distiller remains constant and is always equal to the outside pressure (atmospheric pressure) because there is a bond and a contact with the water tank supply via the level adjuster or the overflow hole. The condensation of vapor on the inner surface of glass cover results in a slight and temporary decrease in the internal pressure. This is rapidly compensated by water evaporation and the process is repeated as long as there is energy in the basin, and as long as there is a temperature difference between the basin and the condenser. The heat storage medium has a dual role. First, it absorbs the excess heat that causes internal overheating of distiller which would cause a performance decrease. Secondly, it feeds the liquid water molecules with energy especially in low solar-intensity conditions or overnight. Thus, the heat storage layer serves to keep a continual temperature difference between the brackish water and the glass-cover. Generally, to have distillation, the basin water temperature must always be higher than that of the glass cover for all units. However, we observethatbetween 09:00 AMand 10:00 AM; the water temperature for units with 1.0 and 1.5 cm sponge thickness was lower than that of their glass cover. This phenomenon is explained by the fact that a thick layer of sponge (storage system) absorbs solar energy and avoids the increase in water temperature. Thus the glass cover temperature becomes higher. The value of the temperature difference between the glass coverand the absorber is directly proportional to the quantity of distilled water. Figure 5 clarifies the results obtained fromFigures 3 and 4. The experiments which have been carried out over 5 days allowed us to have an average for each presented parameter. We must point out that during these 5 days the classification and the ranking of recorded improvements remain unchanged despite the nocturnal yields recorded. So, from these results we can say and confirm that it is unnecessary to have sheets of thicknesses with more than 1.5 cm since the best result is closer to a 0.5 ± 0.1cm thickness. Fig.5
10
5. Estimatedcalculation of ameliorated still and water production costs 5.1 Sponge layercost We will use the price of the optimum still (i.e. with a 0.5cmlayer) which produced a total of 5.01kg/m2 of distillate. Thesponge layer cost includesthe cost of gluing and painting. The cost is estimated in Algerian Dinars (DZD) and converted into euro currency using the current: (1 € = 118 DZD). The construction cost of a conventional still (i.e. our witness still) is estimated to be 6810 DZD/m2 i.e. (57.72 €/m2). In contrast, the cost of our most efficient sponge unit still is 7410 DZD/m2 i.e. (62.8 €/m2). So, improving the quality of the still by adding 0.5 cm sponge layer costs 8.81% more. However, it will be seen below (paragraph 5.2) that the difference between the price of our improved still and that of the conventional unit remains negligible, especially compared to the benefits obtained. 5.2 Distilled water productioncost The lifetimeof a singleslope basin still that well constructed is estimated at 20 years. So the estimated constructionprice becomes (62.8/20) €/year.m2. Generally, a conventional solar still does not require much maintenance and care (in our case one day every 6 months). So, the cost of maintenance is estimated at 10% per year of the construction cost. Thus, the total cost of the best unit becomes 3.46€/year. Finally, the cost price of distilled water is given in (DZD/kg) by: [Total unit cost (DZD/year) /yield (kg/day)/363 (day/year)]+ [the cost price of brackish water (DZD/kg)]. Therefore, the cost of distilled water becomes: (408/5.010/363)+0.31 = 0.54 DZD/kg or 4.58 10-3 €/kg. However, the cost of distilled water produced through other energy means is closer to13 DZD/kg (or 0.11 €/kg). In other words, producing distilled water using a still with sponge layers is 24 times cheaper than that produced through other means such as electricity generated by fossil energy. The best improvement obtained with our blackened sponge still (without taking into account nocturnal distillation) is 57.77%. It becomes 59.3% after adding nocturnal distillation amount. The equivalent gain in distilled water output is about 1.8 kg/ (m2.day) i.e. 653.4 kg/ (m2.year).This implies that our improved still has a production equivalent to 1.6 times that of a conventional unit. Given the concentrations of salts in our brackish water, we need to mix two volumes of distilled water with one volume of brackish water to obtain three volumes of usable fresh water; so, with 59.3% of improvement, an additional of 980 kg/ m2.year of drinkable water is produced. The average daily consumption of drinkable water is (2-3) kg/person; thus, the previous supplement of 980 kg/ (m2.year) corresponds approximately to the annual drinking water consumption for one person. So, each square meter can supply fresh water for one more person. From our study, the optimum still can produce more than 5 kg/ (m2.day). Therefore, it can supply almost 3 people. From this latest result we can deduce that a square meter of distillation unit can supply 3 persons with drinking water. In large scale, a town of 300,000 inhabitants, such as Ouargla city, can be supplied with drinking water by the use of a surface area of 1000m x100m of solar distillation units with a brackish water flow of about 75 x 104 kg / day a cost not exceeding 2500€/day. 6. Conclusion With ever increasing population and rapid industrialization growth, there is a great demand for freshwater or drinking water. Arid zones and remote areas in southern Algeria communities are rich in solar energy and groundwater. These regions present a favorable advantage for solar desalination. 11
These regionsdepend on underground saline water for drinking, but thegroundwater is often too salty and is not recommended for human consumption.Small-scale brackish water solar distillation can make a considerable contribution in providing fresh water supplies for rural communities. Four basin-type solar stills were constructed and experimentally tested under the climatic conditions of the arid zone of southern Algeria in Ouargla University. The idea was to study the possibility of converting brackish water to soft water using a heat storage system. Our objective was to increase the yield of a solar still by improving its absorber performance through the use of an inner heat storage system.This was obtained by coating the entire horizontal absorber surface with a layer of blackened sponge.The latter is a porous material with a large specific surface whichleads to high mass and heat transfer effectiveness. The layer plays the role of a heat storage medium which keeps the operating temperature of the still high enough to produce more distilled water even in the case of cloudy weather or atnight. The experiments consisted of studying the effect of the sponge thickness on the productivity of the solar stills. Results obtained, showed that: sponge layers thicknesses of 0.5 cm and 1.0 cm increased the still yield by of 57.77 % and 23.03 %, respectively. However, sponge thickness of 1.5 cm decreased the yield by 29.95 % compared with a conventional basin still. A simple economic analysis shows that solar distillation using a thin blackened layer of sponge as heat storage medium in absorber is efficient and produces a much improved yield of distilled waterat very lowcost compared to other techniques. In summary, the development of any region is generally based on the existence of two essential factors: water and energy. These two factors (abundant brackish water and solar energy) are characteristics of the Ouargla region. From an industrial viewpoint, brackish water is not recommended because of the resulting corrosion or clogging can degrade the equipment. Distilled or fresh water are needed also for industrial uses. According to our study, large-scale solar distillation of groundwater can contribute to the development of this region given the estimated price of a kilogram of distilled water supplied by the cleanest and cheapest energy possible: solar energy.
7. References [1] The world Bank-Bank-Netherlands water partnership, seawater and brackish water desalination in the Middle east, North Africa and Central Asia, Final report Annex1, Algeria (2004). [2] B. Bouchekima; A small solar desalination plant for the production of drinking water in remote arid areas of southern Algeria, Desalination. 156 (2003) 353-354. [3] M.H. Sellami, H. Bouguettaia, D. Bechki, M. Zeroual, S. Kachi, S. Boughali, B. Bouchekima and H. Mahcene, Effect of absorber coating on the performance of a solarstill in the region of Ouargla (Algeria), Desalination and Water Treatment. 21 (2013) 1-8. [4] D. Bechki, H. Bouguettaia, J. Blanco-Galvez, S. Babay, B. Bouchekima, S. Boughali, H. Mahcene; Effect of partial intermittent shading on the performance of a simple basin solar Still in South Algeria, Desalination. 260 (2010) 65-69. [5] M.A. Samee,U.K. Mirza,T.Majeed and N. Ahmad; Design and performance of a simple basin solar still, Renewable Sustain Energy Rev. 11 (2007) 543-549. [6] A.A. Youssif and A.F. Ismail; Theoretical and experimental investigation of a novel multistage evacuated solar still, Journal of Solar Energy. 127 (2005) 381-385. [7] A. Safwat, M. Abdelkader, and A. Abdelmotalip; Parameters affecting solar still productivity, Energy Conversion Management 41 (2000) 1791-1809. [8] R. Dev, S.A Abdul-Wahab and GN Tiwari; Performance study of the inverted absorber solar still with water depth and total dissolved solid, Desalination. 88 (2011) 252-264. 12
[9] S. Aboul-Enein, A. A. El-sebaii and E. El Bialy; Investigation of a single basin solar still with deep basins, RenewableEnergy. 14 (1998) 299-305. [10] F.F. Tabrizi, M. Dashtbanand H. Moghaddam; Experimental investigation of a weir-type cascade solarstill with built-in latent heat thermal Energy storage system, Desalination.260 (2010) 248-253. [11] M. Dashtbanand F.F. Tabrizi; Thermal analysis of weir-type cascade solar still integrated with PCMStorage, Desalination. 279 (2011) 415-422. [12] M. Sakthivel,S. Shunmugasandaram and T. Alwarsamy; An experimental study on a regenerative solar still with energy storage medium-jute cloth. Desalination 264 (2010) 24-31. [13] J.T. Mahdi,B. E Smith and A. O. Sharif; An experimental wick-type solar still system: Design and construction, Desalination 267 (2011) 233-238. [14] M.A Abd El Kawi and M.M. Naim; Non-conventional solar still part1. Non-conventional solar still with charcoal particles as absorber medium, Desalination 153 (2002) 55-64. [15] S.N. Rai, D.K. Dutt and G.N Tiwari; Some experimental studies of a single basin solar still, Energy conversion Management. 30 (1990) 149-153. [16] A.A.M. Sayighand E.M.A. El-Salam; Optimum design of a single slope solar still in Riyadh, Saudi Arabia, Revue d’Heliotechnique.1 (1977) 40-44. [17] B.A. Akash,M.S Mohsen, O. Osta, and Y. Elayan.; Experimental evaluation of a single-basin solar still using different absorbing materials, Journal of Renewable energy. 14 (1998) 307-310. [18] B. Abu-lhijlehand H.M. Rababa’h ; experimental study of solar still with sponge cubes in basin, Journal of Energy Conversion & management. 44 (2003) 1411-1418. [19] V.Velmurugan , M.Gopalkrishnan and R.Raghu. Single basin solar still with fin for enhancing productivity. Energy Conversion and Management. 49 (2008) 2602-2608. [20] R.V. Dunkle, Solar water distillation, The roof type still and a multiple effect diffusion still, International Developments in Heat Transfer, Part V, University of Colorado (1961) 895. [21] C. Zeshaoo, Natural convection heat Transfer across air Layers at various angles of Inclination, Engineering thermo-physics 20 (1984) 211. [22] MA.S. Malik, G.N. Tiwari, A. Kumar and M.S. Sodha, Solar distillation, Oxford, UK, Pergamon Press (1982) 8–17. [23] A.T. Shawaqfeh and M.M. Farid, New development in the theory of heat and mass transfer in solar stills, Solar Energy 35 (1995) 527. [24] R.S. Adhikari, and A. Kumar, Estimation of mass-transfer rates in solar stills, International Journal of Energy Research 44 (1990)737. [25] W.C. Swinbank, Long-Wave Radiation from Clear Skies, Quarterly Journal of the RoyalMeteorological Society, 89 (1963) 339. [26] O.O.Badran and M.A.K. Mazen, Evaluating of thermal performance of a single slope solar still, Heat Mass Transfer.43 (2007) 985-995. [27] R.V. Dunkle, International Developments in Heat Transfer, Conference, Part 5, Denver (196l) 895. [28] A.K. Singh, Optimization of Orientation for Higher Yield of Solar Still for Given Location, Energy conversion and Management 36 (1995)175-187.
13
Glass Cover Wood support
Distilled water
Saline Water
Sponge layer Insolation
Absorber
Fig.1 Schematic diagram of solar still
Fig.2. Typical measured solar irradiance and ambient temperature time history curves
14
Saline water entrance
Fig.3. Effect of sponge thickness on the distilled water productivity
Fig.4. Effect of sponge thickness on the hourly cumulative amount of distillate
15
Fig.5. Curves of temperature difference between glass cover and basin water for all units under study
Table.1. Thermo-physical and configuration parameters of solar still parts
Glass-cover Saline Water Basin
Insulation
Blackened Sponge
Highlights - Solution to the problem of fresh water shortage in sunny remote areas. - Improvements in solar stills using cheap and available materials. - More efficient and longer distillation times. - Production of distilled water at low cost. - Encouraging results.
16