Numerical study of a double-slope solar still coupled with capillary film condenser in south Algeria

Energy Conversion and Management 94 (2015) 245–252

Contents lists available at ScienceDirect

Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman

Numerical study of a double-slope solar still coupled with capillary film condenser in south Algeria Mohamed Mustapha Belhadj a,b,⇑, Hamza Bouguettaia b, Yacine Marif a, Moussa Zerrouki a a b

Unité de Recherche en Energies Renouvelable en Milieu Saharien (URER’MS), Centre de Développement des Energies Renouvelable (CDER), PO Box 478, RT Raggane, Adrar, Algeria Department of Physics, Laboratory of New and Renewable Energy in Arid Zones (LENREZA), Ouargla University 30000, Box 511, RT Ghardaïa, Ouargla, Algeria

a r t i c l e

i n f o

Article history: Received 9 August 2014 Accepted 23 January 2015

Keywords: Solar energy Solar distillation Condensation cell Capillary film Simulation Modeling

a b s t r a c t The effect of joining a condensation cell to a single-basin double slope solar still was investigated numerically. Direct solar radiation heated the saline water then evaporated. A fraction of the resulting vapor is condensed on the inner glass cover plate and the rest on the outer metal plate. Solar radiation, ambient temperature and the temperatures at different system components were monitored. The performance of the system was evaluated and compared to that of a conventional solar still under the same meteorological conditions. The proposed prototype functioned perfectly and its daily yield reached 7.15 kg m2 d1. Results show that the productivity of the present system was about 60% higher than that of the conventional and capillary film types. The contributions of the glass cover, metal plate and condenser plate are 43%, 18% and 39% of the total distillate yield respectively. It was noticed that the productivity of the capillary film solar still was sensitive to the mass flow of the feeding water. It was also found that the absorptivity coefficient and the diffusion gap have significant effect on distillate production of the system. Ó 2015 Elsevier Ltd. All rights reserved.

1. Introduction In some parts of the world, particularly in the Middle East and North Africa, the production of fresh water is often a serious problem. Algeria is considered a dry country due to the severe scarcity of its water resources. The demand for fresh water is increasing at 4–5% per annum. Water scarcity is aggravated by high population growth. The population is expected to reach 46 million by the year 2020 [1]. The demand for fresh water has largely exceeded the amount that fresh sources can meet. This led to an increasing interest in new desalination technologies in order to fulfill the fast growing socioeconomic needs. Conventional techniques for desalination can broadly be classified into thermal and membrane based categories. The former class of techniques includes Multi-Stage Flash (MSF) [2], Multi-Effect Desalination (MFD) [3] and Mechanical Vapor Compression (MVC) [4,5]; while the latter type comprises Reverse Osmosis (RO) [6] and Electro-Dialysis Reversal (EDR). These technologies used electrical power [7,8] which is an environmental concern because a large portion of this power is generated by coal or gas ⇑ Corresponding author at: Unité de Recherche en Energies Renouvelable en Milieu Saharien (URER’MS), Centre de Développement des Energies Renouvelable (CDER), PO Box 478, RT Raggane, Adrar, Algeria. Tel.: +213 49 96 51 68; fax: +213 49 96 04 92. E-mail address: [email protected] (M.M. Belhadj). http://dx.doi.org/10.1016/j.enconman.2015.01.069 0196-8904/Ó 2015 Elsevier Ltd. All rights reserved.

based power plants. They are not used in regions with low infrastructure either for the supply in decentralized regions due to their permanent need of qualified maintenance and electricity supply. The small scale example solar desalination is chosen as an alternative solution ever since this technique has proved useful for producing distilled water in large amounts. Most of the remote, arid regions in the south of the country are devoid of natural fresh water and depend entirely on underground water for drinking and other domestic uses. Numerous low-density population localities lack not only fresh water availability, but in most cases, electrical power grid connections as well. Favorably, these zones profit from an important renewable energy potential; the use of which in water desalination exhibits an interesting chance to offer a secure source of potable water. The integration of solar energy in desalination and purification of available brackish water seems to be a logical and attractive answer for supplying these small remote agglomerations with fresh water. For small-scale applications (from 5 to 100 m3/day water production), the cost of water production systems is much higher than for large-scale systems. The cost of water production for largescale can go up to US$ 30/m3 [7] for stations of smaller capacity. The desalination of brackish or seawater by solar distillation using basin-type solar stills is an operation largely used in the arid and semi-arid areas. These types have the advantage of being simple devices in terms of construction, operation and maintenance. They are cost free energy, environment friendly, usually more targeted

246

M.M. Belhadj et al. / Energy Conversion and Management 94 (2015) 245–252

Nomenclature a Cp Dcs dc E F Gg Gr h h hm hf L lr M _ m Nu P Pr R Ra Rc S T t U Ws

thermal diffusivity ðm2 s1 Þ 1 specific heat ðJ kg K1 Þ steam diffusivity ðm2 s1 Þ thickness of cell (m) thickness (m) fraction (dimensionless) global irradiation ðW m2 Þ Grashof number (dimensionless) length (m) coefficient of heat transfer ðW m2 K1 Þ coefficient of mass transfer ðms1 Þ 1 latent heat ðJ kg Þ length (m) width (m) mass (kg) rate of mass flow ðkg m2 s1 Þ Nusselt number (dimensionless) pressure (Pa) Prandtl number (dimensionless) 1 constant of perfect gases ðJ K1 mol Þ Rayleigh number (dimensionless) ratio (dimensionless) surface ðm2 Þ temperature (K) time (s) heat transfer coefficient ðW m2 K1 Þ wind speed ðms1 Þ

Greek symbols a absorptance (dimensionless) b coefficient of thermal dilatation ðK1 Þ e emissivity (dimensionless) k thermal conductivity ðW m1 K1 Þ l dynamic viscosity (Pa s)

towards poor communities and more importantly, they can run unattended for long periods of time in remote isolated sites. Recently, numerous researchers are interested in solar desalination. For example, the work on passive solar stills has been reviewed by Malik et al. [9]. Bechki et al. [10] carried out experiments in which they investigated the effect of partial intermittent shading on the performance of a simple basin solar still in south Algeria. Their series of tests consisted in lowering the glass temperature by an intermittent shading of the north glass cover. This procedure resulted in a further 12% enhancement in the daily distillate output. Tanaka et al. [11] constructed and tested a simple solar distiller coupled to a distiller with vertical multi-effect. The distance between the vertical cells was approximately 5 mm. The theoretical estimated daily increase in the total productivity of the distiller was about 15.4 kg m2 d1 for a number of cells equal to 10 and a space of 5 mm between the walls. Bouchekima et al. [12] presented the results of experiments carried out with a capillary film distiller using solar energy. Boukar and Harmim [13] reported performance evaluation of one-sided vertical solar still tested under desert climatic conditions of Algeria. Their study showed that still output varies between 0.5 and 2.5 kg m2 for the fabric surface sponges. The effect of the use of an internal condenser on the performance of a solar distiller was investigated experimentally by A.S. Ahmed [14]. Their results showed that the combination of an internal condenser with the solar distiller improved the performance of a distiller. Abu-Arabi et al. [15] displayed a wide literature survey and modeled a solar still with cooling water flowing between a double glass cover. Fath and Hosny [16] proposed a theoretical

m R h /

q r c cs g

kinematic viscosity ðm2 s1 Þ reflectance (dimensionless) angle inclination of glass cover (°) angle inclination of metal plate (°) density ðkg m3 Þ Stefan–Boltzmann constant ðW m2 K4 Þ surface azimuth angle (°) solar azimuth angle (°) efficiency (dimensionless)

Subscripts a ambient b basin liner ba from basin liner to ambient bw from basin liner to water saline c convective cs condenser plate csa from condenser to ambient dba from conductive basin to ambient e evaporative eff effective eq equivalent g glass cover ga from glass cover to ambient m mixture p condenser–evaporator metal plate pc from condenser–evaporator to condenser plate Ref reflect r radiative w brackish water wg from water to glass cover wp from water to condenser–evaporator plate sky sky

study of a single-sloped basin with enhanced evaporation and built-in additional condenser. The greenhouse solar still types are simple to construct and do not present great technical difficulties. However they present the disadvantage of having a somewhat poor yield and their production remains insufficient. In order to enhance this production, large-sized installations had to be carried out. The present work provides a study of a double slope still joined to a capillary film condenser cell. The meteorological conditions influencing the performance of this distiller were investigated. The performance of the present system was evaluated and compared with that of a conventional type.

2. System description and modeling Fig. 1 shows a sketch of the solar still used in this study. The system is constituted of coupling between conventional solar still (CSS) and a capillary film solar still (CFSS). The main components of the present system are: The absorber of the (CSS) was blackened on the surface to ensure maximum absorption of solar radiation for effective heating of the water. The base of this assembly was lagged with a thick polystyrene insulation. This base was covered with glass orientated in the south and a metal plate orientated in the north. A wick was homogenously placed on the surface of the metallic plate. The latter was tightly held with a wooden frame and covered with a glass cover which was sealed tightly by silicone sealant to prevent any vapor leakage.

247

M.M. Belhadj et al. / Energy Conversion and Management 94 (2015) 245–252

(c) Basin liner:

Feed reservoir Wick feed

M b Cpb

Sunshade

dT b ¼ F bw Sb Gg  Sb hcbw ðT b  T w Þ  Sb U dba ðT b  T a Þ dt

Such as U dba ¼ hcba þ hrba þ eq

Condenser plate Glass cover

ð4Þ

!1

Condenser-Evaporator plate

where

Distillate drain.1 Distillate drain.2

eq ¼

1 Eb kb

1 þ Eiso þ kiso

1 1 hcbw

.

The values of the solar absorption factor were computed as follows: Factor of absorption by glass cover

Brine trough Wood cover

ð3Þ

Saline water

Distillate drain cover

F gsky ¼ ag

Fig. 1. The used design components.

ð5Þ

Factor of absorption by water in basin

F wg ¼ aw sg The capillary film solar cell (CFSS) is tilted with an angle of 60° with regard to the first still. The latent heat of vaporization of the first device is reused to heat another quantity of saline water in the wick attached to the rear surface of the metal plate. Water vapor diffuses through the narrow humid air layer between the first and second metal plates and condenses on the opposite front surface [17,12]. Furthermore, this configuration can take advantage of the reflected radiation from the surface evaporation basin. A mathematical model was developed in order to simulate the performance of the proposed distiller (CSS and CFSS) under local meteorological conditions (with the orientation completely towards the south c ¼ 0 Þ and it was assumed that:    

ð6Þ

Factor of absorption by basin absorber

F bw ¼ ab sg sw

ð7Þ

(d) Metal plate cover:

M p Cpp

dT p ¼ F pw Sp Gg þ Sb ðhcwp þ hrwp ÞðT w  T p Þ dt _ wp Sb hfw ðTÞ  m _ pcs Sp hfp ðTÞ þm _ w Cpw Sp ðT a  T p Þ  Sp ðhcpcs þ hrpcs ÞðT p  T cs Þ þ m _ s Cpw Sp ðT p  T a Þ m

ð8Þ

Quantity absorbed by metal plate cover,

F pw ¼ ð1  aw  sw Þap sg

The device is air vapor tight. Heat-transfer surfaces are isothermal walls. No dry spots on the wick (completely wet). Constant rate of feeding water throughout tests.

ð9Þ

(e) Rate of feed is written as follows:

_w¼m _ wp þ m _s m

ð10Þ

(f) Condenser plate: Fig. 2 illustrates the still overall heat balance. The transient energy equations for the system components can be written as follows:

M cs Cpcs

dT cs _ pcs Sp hfp ðTÞ ¼ Sp ðhcpcs þ hrpcs ÞðT p  T cs Þ þ m dt  Sp hccsa ðT cs  T a Þ  Sp hrcsa ðT c s  T s kyÞ

ð11Þ

(a) Glass cover:

Mg Cpg

dT g ¼ F gsky Sg Gg þ Sb ðhcwg þ hrwg ÞðT w  T g Þ dt _ wg Sg hfg ðTÞ  Sg hcga ðT g  T a Þ  Sg hrga ðT g  T sky Þ þm ð1Þ

The correlation used to compute the specific latent heat of vaporization of water is [18]:

hf ðTÞ ¼ ð2503:94  2:4515TÞ  103

(b) Water in basin:

Mw Cpw

2.1. Correlations used in the conventional solar still

dT w ¼ F wg Sb Gg  Sg ðhcwg þ hrwg ÞðT w  T g Þ dt _ wg Sb hfw ðTÞ  m _ wp Sb hfw ðTÞ þ Sb hcbw ðT b  T w Þ m  Sp ðhcwp þ hrwp ÞðT w  T p Þ

ð2Þ

Heat loss through the top of the glass cover to the environment is predominantly by radiation (to sky) and convection (to ambient air). Wind speed influences the convective heat transfer from the top part and the wind coefficient of heat transfer is calculated from [19]:

Irradiative

Evaporator-condenser

Convective

Solar radiation Glass cover

ð12Þ

ℜ

Evaporative Condenser

Basin liner Fig. 2. Schematic diagram of CSS and CFSS.

248

M.M. Belhadj et al. / Energy Conversion and Management 94 (2015) 245–252

hcia ¼ 2:8 þ 3:0 ws ;

ws 6 5 m s1

ð13Þ R

With sky temperature given by [19]:

T sky ¼ T a  12

ð14Þ

ð16Þ

The ambient temperature is given by [21]:



p

14  TSV 12



þ 0:5ðT max þ T min Þ

ð17Þ

In addition, there is an internal heat radiation from hot water to each of the condensing surfaces. The internal radiative heat transfer coefficient is given by:

  hrwj ¼ rewj T 2w þ T 2j ðT w þ T j Þ

ð18Þ

1

ew

þ e1j  1



ð19Þ

C

J

For h 6 30 , corresponding to the aspect between basin and glass cover enclosure that contains air, Jakob [22] recommends the following simple correlations:

Nu ¼ 0:195Ra0:25 ; 104 < Ra < 4  105 Nu ¼ 0:068Ra0:33 ;

4  105 < Ra < 107

ð20Þ

where hcwj ¼ Nu dk . b

Consider an inclined metal plate that makes an angle / with the horizontal, as shown in Fig. 3. The Nusselt number can be determined from the vertical plate relations provided; and ðgÞ in the Rayleigh number relation is replaced by ðg cosð/ÞÞ for / 6 60 [22,34]. Both condensation and evaporation processes involve mass transfer. Consequently, relevant correlations are used to estimate the coefficients of internal convective and evaporative heat transfers from hot water to each of the condensing surfaces:

 1=3 T w ðPw  Pj Þ hcwj ¼ C ej ðT w  T j Þ þ 298; 000  Pw  1=3 k g q b C ej ¼ 0:075 db la

ð21Þ

hewj ¼ 16:273  103 hcwj where j ¼ g and p.

ðPw  Pj Þ ðT w  T j Þ

ð23Þ

U

K

D

2.2. Correlations used in the capillary film solar still Taking into account that the metal plate cover is inclined with an angle, the convective heat transfer coefficient from the cavity (CFSS) is calculated according to [23] by:

km dc

ð24Þ

dc is the distance between evaporator and condenser of the cavity (CFSS), and

ð25Þ

where 1:6

1780ðsinð1:8/ÞÞ Gr Pr cosð/Þ    1780 X 2 ¼ Max 0; 1  Gr Pr cosð/Þ "  # 1=3 Gr Pr cosð/Þ 1 X 3 ¼ Max 0; 5830

X1 ¼ 1 

ð26Þ ð27Þ ð28Þ

The rate of evaporation between metal plate and condenser plate can be given by [24]:

mpcs ¼ Dq hm

ð29Þ

Based on Chiltone–Colburn analogy, the relation between heat and mass transfer coefficients for air–water vapor mixture can be expressed with a good accuracy as [24]. The coefficient of mass transfer is written by the following formula:

hm ¼

ð22Þ

With db ¼ ðLw lr Þ=ð2Lw þ 2lr Þ, as the characteristic distance of the evaporator. From literature [23], C eg ¼ 0:884 (corresponding to h 6 30 of the glass cover), whereas, the computed value is C pe ¼ 0:7065 (corresponding to an angle / > 30 of the metal plate). The other physical properties of water such as (km ; bm ; am and lm ) were computed from temperature-dependent correlations as follows.

H

Fig. 3. Geometry of the solar distiller with the solar rays; ABCD = area of evaporator, AB = length of basin liner (Le), AD = width of basin (Lr), EP = length of glass cover, QE = length of metal plate, AP = height of basin (dv), EPQ angle of inclination of glass cover (h), EQP = angle inclination of metal plate (/), the angle between projected SQE ¼ ð90  ðh þ /ÞÞ, the angle of orientation AHU ¼ ðcs  cÞ, the zenith angle NHG = w.

Nu ¼ 1 þ 1:44 X 1 X 2 þ X 3



(

ψ

G

M

where

1

F

N

hcpcs ¼ Nu

ewj ¼ 

P A I

ð15Þ

  ¼ rewj T 2w þ T 2j ðT w þ T j Þ

T a ¼ 0:5ðT max  T min Þ cos

ϕ

Q B

  ¼ rei T 2i þ T 2sky ðT i þ T sky Þ

There is a heat transfer by radiation between the water basin and each condensation surface. The radiative heat transfer coefficient is estimated as:

hrwj

E

θz

The coefficient of irradiative heat transfer to the sky is given by [20]:

hrisky

S

hcpcs  ð23Þ qm Cpm Dacs

ð30Þ

For binary mixtures of water vapor and air, the diffusion coefficient is given by [25] as:

Dcs ¼ 0:187  108 T 2:072

ð31Þ

The above expression which evaluates the diffusion coefficient in (m2/s) is valid for 20 °C and 90 °C, with reference pressure and temperature values of 1 atm and 20 °C. The saturation vapor pressure inside the solar distiller was calculated by using a correlation reported by [26,35]. The densities of water vapor in the evaporator (CFSS) Dq were calculated using Eq. (29).

Dq ¼

  1 Pi Pj  R Ti Tj

ð32Þ

249

M.M. Belhadj et al. / Energy Conversion and Management 94 (2015) 245–252

The distillate yield ðmT Þ and the efficiency ðgÞ of the system in a time interval ðt s  tr Þ are calculated from:

mT ¼

Z

Specific heat

ts

_ ewg þ m _ ewp ÞSb hfw ðT w Þ þ m _ epcs Sb hfp ðT p Þdt ½ðm

ð33Þ

Density Conductivity

ð34Þ

Dynamic viscosity

tr

g ¼ R ts tr

Table 2 Dry air properties.

mT Gg Sb dt

100

Or,

Thermal diffusivity

aair ¼ 1:8343  105 þ 0:146  106 T Pr ¼ 0:7147 þ 0:254  103 T bair ¼ 1=ðT þ 273:15Þ

ð35Þ

where ðgg Þ; ðgp Þ and ðgcs Þ are the efficiencies of the glass cover, metal plate and condenser plate respectively. The parameters used in the simulation are given in Tables 2–4. The thermo-physical properties are presented in Table 1 and are assumed as constants in all calculations. The numerical study was carried out under the climatic and meteorological conditions of Adrar city; which is located in the south of Algeria and characterized by a longitude of 0.17°W, an altitude of 264 m and latitude of 27.53° [27]. In order to obtain an acceptable design, the shades impact of the tilted metal plate and the sides of the still were taken into account. The experiments were performed during the month of July. The mean solar intensity on a horizontal surface in this month exceeded 1000 W/m2. The model of Capderou [28] was used to compute the received direct and diffuse irradiation. The formula of Kasten [29] was applied to obtain the total incident radiation. Part of the solar radiation reaching the still is reflected or absorbed by the cover. Most of the radiation incident on the basin bottom is absorbed by the saline water; a small portion is reflected from water surface towards the metal plate. The amount of solar energy received by saline water in the evaporator and the amount reflected to the metal plate condenser are:

Geff ¼ ðaw sg þ ab sg sw ÞSeff Gg

ð36Þ

where aeff ¼ aw sg þ ab sg sw . ðaeff Þ is the effective fraction of radiation received on the evaporator basin.

GRef ¼ ð1  aeff Þap Sp Gg

ð37Þ

Table 1 Thermo-physical properties used in the present study. Parameters

CSS

Ew (m) Ep (m) Sb (m2) Sp (m2) qg (kg m3)

0.01 1

qb (kg m3) qp (kg m3)

7140 7864

kb (W m1 K1) kiso (W m1 K1) Cpg (kJ kg1 K1) Cpb (kJ kg1 K1) Cpp (kJ kg1 K1) mb (kg) mw (kg)

113 0.037 0.460 0.896

ag aw ap ab eg ew ep eb

0.0475 0.05

ws (m s1)

kair ¼ 0:0242 þ 7:57  105 T

lair ¼ 1:7176  105 þ 0:46  107 T

Prandtl number Thermal expansion

!

g g g g ¼ gg 1 þ p þ cs p 100 gg gp gg

Cpair ¼ 1055:05  0:3475T þ 6:25  104 T 2 qair ¼ 353=ðT þ 273:15Þ

CFSS 0.001 0.5

2800

Specific heat Density

Cpw ¼ 3958  52:3S þ 0:837T

Conductivity

kw ¼ 0:5536 þ 2:238  103 T  9:87  106 T 2

qw ¼ 1002:6  0:505  101 T  0:38  102 T 2

Kinematic viscosity

mw ¼ 17:199  104  0:3389  104 T 2 þ 0:2  106 T 3

Prandtl number

Pr ¼ 12:501  0:261T þ 1:577  103 T 2

Table 4 Properties of humid air. Dynamic viscosity Thermal expansion Specific heat

lm ¼ 1:718  105 þ 4:62  108 T bm ¼ 1:6578  103 T½0:362x  11 ; x ¼ P=P 0 P s P 1

Cpm ¼ ðCpa þ Cpw yÞ=ð1 þ yÞ; y ¼ 0:622

The effective area of saline water which is exposed directly to incident radiation could be computed from the solar azimuth angles, altitude, longitude and latitude of the site. This area is calculated by using the geometrical analysis given in Fig. 3.

Area of triangle; SNHG ¼ 0:5ðLw  Z HU ÞZ NG

ð38Þ

Area of rectangle; SABNU ¼ Lw Z AU

ð39Þ

Area of rectangle; SUHKD ¼ ðLr  Z UA ÞZ HU

ð40Þ

The effective area of the evaporator receiving solar radiation can be written as:

Seff ¼ Sb  jðSNHG þ SABNU þ SUHKD Þj

ð41Þ

But in case, Sb 6 jðSNHG þ SABNU þ SUHKD Þj, during sunrise and sunset. Sb , is the area of the absorber in the absence of shadow. A Fortran90 program was written to solve the above non-linear equations using the Runge–Kutta method [30]. The initial temperatures of the system components were assumed to be equal to ambient temperature T a . The flow chart of the program is presented in Fig. 4. 3. Model validation

0.390 10 0.5

0.80–0.95 0.95 0.88–0.90 0.955 0.93 0.95 2.0

Table 3 Brine water properties.

2.0

The model was validated by comparing simulation results with experimental data obtained at the LENERZA laboratory, Ouargla University, (South Algeria), in case of a vertical condensation cell. Fig. 5 shows simulation and experimental comparison of hourly variation of fresh water productivity for glass cover, metal plate, condensation plate and total production respectively. It was found that there is a good agreement between the simulation results and experimental data. The maximum deviations between theoretical and experimental analysis was less than 6.32% [31]. This deviation was expected, and that is due to uncertainties in the measured quantities of the experiments and the simplifying assumptions in the mathematical models in addition to the computational errors and the accuracy of calculations.

250

M.M. Belhadj et al. / Energy Conversion and Management 94 (2015) 245–252

10

Total : Theoretical Total : experimental

Cumulative productivity (kg m -2 h-1)

data: 06/08/2007

8

1000

100

800

80

600

60

400

40

200

20

0

0 4

6

8

10

12

14

16

18

Radiation reflect on plate metal ( W/m2 )

Fig. 4. Flow chart of the temperatures computation and distillate yield in FORTRAN90.

Solar radiation on basin liner ( W/m 2 )

Fig. 7 shows the variation in ambient temperature ðT a Þ and the temperatures of the glass cover, saline water, the metal plate and the condensation plate surface. The process of distillation is basically based on the difference of temperature between the heat transfer surfaces. At 12:00 am, the temperature difference between water and the glass cover ðT w  T g Þ and between water and the metal plate ðT w  T p Þ are 13 °C and 9 °C respectively. This increase in the temperature of the metal plate (condensation cell) will heat the slowly adhering water on the textile. The steam leaving the wick will reach the surface of the condenser. The effect of the absorption coefficient of the evaporator basin liner ðaeff Þ on the yield of the distiller is shown in Fig. 8. For instance, the productivity was increased from 21.91% (6.27 kg m2 h1) to 28.07% (8.03 kg m2 h1). These correspond to an absorption coefficient of 80–100% respectively. It is seen that the productivity is directly proportional to the magnitude of ðaeff Þ; if the absorption coefficient is increased, water productivity increases, and this is consistent with the results reported in literature. For this reason, it is recommended that the absorber should be painted in black to enhance the absorption of incoming solar radiation, which increases the water temperature and hence the distillate yield.

20

6

Local time (hr)

4

Fig. 6. Irradiation variation in basin liner and reflected radiation from the metal plate with time.

2 100

Tg Tw Tb Tp Tcs Ta

0 90 6

8

10

12

14

16

18

20

Local time (hr) Fig. 5. Comparison of total cumulative productivity calculated from the present model with experimental data (case of the angle / ¼ 90 ).

4. Results and discussion

80

Temperature (°C)

4

70 60 50 40

Fig. 6 displays the variation of the global irradiation with time. It could be readily seen that the irradiation on the basin is higher than that reflected on the metal plate. This latter part of the solar radiation increases the temperature of the plate, and that influences the evaporation–condensation process. This indicates that the direct use of ðGg Þ in the thermal equilibrium equations would influence the daily yield of the distilled water. These values correspond to a typical day under clear sky conditions.

30 20 4

6

8

10

12

14

16

18

20

Local time (hr) Fig. 7. Variations of the ambient temperature ðT a Þ, glass cover ðT g Þ, the saline water and the evaporator basin ðT b Þ, metal plate ðT p Þ and condenser surface ðT cs Þ.

251

M.M. Belhadj et al. / Energy Conversion and Management 94 (2015) 245–252

Fig. 11 shows the effect of the distance between the two opposing plates in the capillary film solar still. It is observed that the distillate productivity increases with the value of the ratio Rc ¼ dhc . The

αabsorptance = 0.80 αabsorptance = 0.85 αabsorptance = 0.90 αabsorptance = 0.95 αabsorptance = 1.00

8

6

productivity varies conversely with the distance ðdc Þ. The distilled mass decreased from 20% (3 kg d1 m2) to 16.67% (2.5 kg d1 m2) with increasing the distance from 0.02 m to 0.05 m respectively. This is attributed to the fact that the convective heat transfer coefficient inside the CFSS varies directly with the thickness ðdc Þ. These results are consistent with those reported by Tanaka et al. [11].

4

2

5. Cost estimation 0 4

6

8

10

12

14

16

18

20

Local time (hr) Fig. 8. Effect of absorption coefficient of the evaporator basin inner on the distillate yield.

In order to obtain a better yield in the CFSS part of the distiller, the saline water flow rate has an important influence on the productivity of fresh water, which decreases with increasing the flow rate. The feed rate of saline water to the wick was kept constant throughout the test. Fig. 9 shows the effect of the variation of the feeding flow saline water into the CFSS, the production of the single distillation cell decreases from 26.92% (3.3 kg m2 d1) to less than 21.21% (2.6 kg m2 d1) when the feed rate increases from 0.138 kg h1 to 0.972 kg h1. This result is consistent with the findings of Zerrouki et al. [32]. In this paper, the reported distillate production of the present still is the total contribution from water in the basin and the wick attached to the rear surface of the metal plate. Fig. 10 presents the cumulative yields of the CSS and the present still at reference values of the design, functioning and meteorological parameters. It could be seen that the productivity of the two parts was somewhat slow in the early hours of the day (9.00 am). This is mainly due to the fact that the different parts of the system were roughly at the same temperature. The distillation starts when the space inside the distiller is saturated with water vapor. The cumulative productivity of the CSS reaches 4.52 kg m2 d1 (with an efficiency of 38%), and 7.15 kg m2 of the present still (with an efficiency nearing 60%).

While designing the proposed solar still intended to function in remote isolated locations in Algerian Sahara, the main objective was to maintain the cost minimal. The total estimation cost of the present system is about 68.27$ [33]. It should be noted that prices of various construction materials were given according to the Algerian market. The minimum average daily productivity is estimated at 7.15 kg m2 d1, which is higher than traditional solar still production. The total productivity of the proposed system equals 25,025 l [29], assuming that the system operates 350 days

10

7 6 5 4 3 2 1 0 4

6

8

10

12

14

16

18

20

Local time (hr) Fig. 10. Distillate yield comparison between the conventional solar still (CSS) and capillary film solar still (CFSS).

4,0

mw = 0.5 kg mw = 1.5 kg mw = 2.5 kg mw = 3.5 kg

Cumulative productivity (kg m -2 h-1 )

Cumulative productivity (kg m -2 h-1)

8

-1

5

4

Present still Solar still

9

Cumulative productivity (kg m-2 h-1)

Cumulative productivity (kg m -2 h-1 )

10

3

2

1

R c = 33 R c = 16 R c = 11 R c = 10

3,5 3,0 2,5 2,0 1,5 1,0 0,5

0 0,0 4

6

8

10

12

14

16

18

Local time (hr) Fig. 9. Effect of feed flow rate on the yield of the CFSS cell.

20

4

6

8

10

12

14

16

18

Local time (hr) Fig. 11. Effect of the cell diffusion gap in the overall production rate.

20

252

M.M. Belhadj et al. / Energy Conversion and Management 94 (2015) 245–252

per year. The cost of one liter from the present solar still is 273.08/ 25025 = 0.01091 $/l, and the cost of distilled water is approximately US$10.91/m3. Taking the typical discount rate as 7%, the water cost becomes US$11.67/m3. The cost analysis shows clearly that such a system of proposed solar stills may be used to provide fresh water to residents in remote areas at reasonable costs. 6. Conclusions In the present work, a numerical model was developed for the functioning of a solar distiller with double non-symmetrical slope modified by the integration of a condensation cell cooled by capillary film of very thin water flow. The applicability of the model is shown for the set of the typical parameters corresponding to the site of Adrar, Algeria. The analysis of the results obtained from the numerical simulation on the effect of certain operational parameters; enables us to conclude the following: – The average daily output of present still is more than 7 kg m2 d1. – The distilled water production increases with the reduction in flow feed saline water. – The productivity varies conversely with the distance between the two plates of the condensing chamber. – The performance of the studied system was enhanced by 55% compared to the conventional type solar still. – The daily efficiency and estimated cost per liter of present still are about 60% and 11.67$/m3 respectively. Whereas, those of the conventional still are 38% and 14.41$/m3. The results show that the total water cost is lower because of higher productivity of the system.

References [1] Ahmed Kettab. Les ressources en eau en Algérie: Stratégies, enjeux et version. Desalination 2001;136:25–33. [2] Garci˘a-Rodri˘guez L, Gómez-Camacho C. Conditions for economical benefits of the use of solar energy in multi-stage flash distillation. Desalination 1999;125:133–8. [3] Raach H, Mitrovic J. Simulation of heat and mass transfer in a multi-effect distillation plant for seawater desalination. Desalination 2007;204:416–22. [4] Zimerman Z. Development of large capacity high efficiency mechanical vapor compression (MVC) units. Desalination 1994;96:51–8. [5] Al-Juwayhel F, El-Dessouky H, Ettouney H. Analysis of single-effect evaporator desalination systems combined with vapor compression heat pumps. Desalination 1997;114:253–75. [6] Ghermandi Andrea, Messalem R. Solar-driver desalination with reverse osmosis: the state of the art. Desalination Water Treat 2007;7:285–96. [7] Muller-Holst H, Engelhardt M, Herve M, Schokopf W. Solar thermal seawater desalination systems for decentralised use. Renew Energy 1998;14(14):311–8. [8] Prakash Narayan G, Sharqawy Mostafa H, Summers Edward K, Lienhard John H. Renew Sustain Energy Rev 2010;14:1187–201.

[9] S Malik MA, Tiwari GN, Kumar A, Sodha MS. Solar distillation. Oxford (UK): Pergamon Press; 1982. [10] Bechki D, Bouguettaia H, Blanco-Galvez J, Babay S, Bouchekima B, Boughali S, et al. Effect of partial intermittent shading on the performance of a simple basin solar still in south Algeria. Desalination 2010;260:65–9. [11] Tanaka H, Nosoko T, Nagata T. A highly productive basin-type-multiple-effect coupled solar still. Desalination 2000;130:279–93. [12] Bouchekima B, Gros B, Ouahes R, Diboun M. The performance of the capillary film solar still installed in South Algeria. Desalination 2001;137:31–8. [13] Boukar M, Harmim A. Parametric study of a vertical solar still under desert climatic conditions. Desalination 2004;168:21–8. [14] Ahmed ST. study of single-effect solar still with an internal condenser. Solar Wind Technol 1988;5(6):637–43. [15] Abu-Arabi M, Zurigat Y, Al-Hinai H, Al-Hiddabi S. Modeling and performance analysis of a solar desalination unit with double-glass cover cooling. Desalination 2002;143:173–82. [16] Fath HES, Hosny HM. Thermal performance of a single-sloped basin still with an inherent built-in additional condenser. Desalination 2002;142:19–27. [17] Ouahes RC, Le Goff PJ, hardy A. High-yield solar distillation of brackish water. Desalination 1987;67:43–52. [18] Radhwan Abulhaiy M. Transient analysis of a passive solar still for heating and humidifying greenhouses. Desalination 2004;16(1):89–97. [19] Kabeel AE, Khalil A, Omara ZM, Younes MM. Theoretical and experimental parametric study of modified stepped solar still. Desalination 2012;289:12–20. [20] Yadav YP, Yadav SK. Parametric studies on the transient performance of a high-temperature solar distillation system. Desalination 2004;170:251–62. [21] Reicosky DC, Winkelman LJ, Baker JM, Baker DG. accuracy of hourly air temperatures calculated from daily minima and maxima. Agric. For. Meteorol. 1989;46:193–209. [22] Cengel YA. Heat and mass transfer: a practical approach. New York: Mc Graw Hill; 2006 [chapter 9]. [23] Tiwari A Kr, Tiwari GN. Effect of water depths on heat and mass transfer in a passive solar still: in summer climatic condition. Desalination 2006;195:78–94. [24] Rahbar N, Esfahani JA. Productivity estimation of a single-slope solar still theoretical and numerical analysis. Energy 2013;49:289–97. [25] Sacadura JP. Initiation aux transferts thermiques. Ed Tech et doc Paris; l979 [in French]. [26] Tsilingiris PT. Modeling heat and mass transport phenomena at higher temperatures in solar distillation systems – the Chilton–Colburn analogy. Sol Energy 2010;84:308–17. [27] Harmim A, Merzouk M, Boukar M, Amar M. Mathematical modeling of a boxtype solar cooker employing an asymmetric compound parabolic concentrator. Sol Energy 2012;86:1673–82. [28] Capderou M. Theoretical and experimental models solar atlas of Algeria (in French) Tome 1 and 2. Algeria: University Publications Office; 1987. [29] Communy PH. Héliothermique, mémoire, France; 1957. p. 194–07. [30] Simos TE, Runge-Kutta Fehlberg A. method with phase-lag of order infinity for initial-value problems with oscillating solution. Comput Math Appl 1993;25(6):95–101. [31] Belhadj Mohammed Mustapha, Bouguettaia H, Marif Y, Zerrouki M, Bechki D, Benhammou M. Theoretical and experimental study of the solar still coupled to a vertical still with water film. J Eng Res Appl 2014:547–52. [32] Zerrouki M, Settou N, Marif Y, Belhadj MM. Simulation study of a capillary film solar still coupled with a conventional solar still in south Algeria. Energy Convers Manage 2014;85:112–9. [33] Kabeel AE, Omara ZM, Essa FA. Enhancement of modified solar still integrated with external condenser using nanofluids: an experimental approach. Energy Convers Manage 2014;78:493–8. [34] Madhlopa A, Johnstone C. Numerical study of a passive solar still with separate condenser. Renew Energy 2009;34:1668–77. [35] Tsilingiris PT. Thermophysical and transport properties of humid air at temperature range between 0 and 100 °C. Energy Convers Manage 2008;49:1098–110.