Desalination 352 (2014) 92–102

Contents lists available at ScienceDirect

Desalination

j ourna l homepage: www.e lsev ie r .com/ locate /desa l

Performance enhancement of a single basin solar still with flow of waterfrom an air cooler on the cover

Aneesh Somwanshi a,⁎, Anil Kumar Tiwari b

a Disha Institute of Management and Technology, Raipur 492101, Indiab National Institute of Technology, Raipur 492001, India

H I G H L I G H T S

• Solar still output enhances by flowing water from an air cooler over glass cover.• The largest increase in annual distillate output is for climate of Jodhpur (56.5%).• The least increase in annual distillate is for climate of Chennai (41.3%).• Cost of additional distillate output with two stills is Rs 0.60/l for Jodhpur climate.

⁎ Corresponding author.E-mail address: aneeshsomwanshi@rediffmail.com (A

http://dx.doi.org/10.1016/j.desal.2014.08.0110011-9164/© 2014 Elsevier B.V. All rights reserved.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 4 December 2013Received in revised form 11 August 2014Accepted 12 August 2014Available online 30 August 2014

Keywords:Solar stillAir coolerAnnual yieldMass flow rate

The annual performance of a conventional solar still, with water from the tank of an air cooler, flowing over thecover, has been computed for four different climatic zones in Indian plains. It is seen that the annual yield can beincreased between 41.3% and 56.5%. The increase in annual efficiency is between 7.4% and 9.9%. Increase in yieldis the largest for hot and dry climate of Jodhpur and the least for warm and humid climate of Chennai. Moreoverthe distillate output increases slightly with increase in mass flow rate and tends to saturate around 0.075 kg/s.These figures are indicative of the viability of the concept. Cost of additional water produced by flow of waterfrom the desert cooler over the cover with two solar stills working together is between Rs. 0.60/l for hot anddry climate to Rs. 0.78/l for warm and humid climate. (U.S.D. 1.00 = Rs. 62.50)

© 2014 Elsevier B.V. All rights reserved.

1. Introduction

Use of solar still to produce freshwater from brackish or sea water isan important method of utilizing solar energy to produce fresh water.However the use of solar stills has been limited due to low productionper unit area; over the years attempts have been made to enhance theproduction rate and efficiency of solar stills. The main focus has beento increase the temperature difference between the water in the basinand the glass cover as this enhances the mass transfer between thewater in the basin and glass cover and thereby increases the productionof water in the solar still. This can be achieved either by increasing thetemperature of water in the basin by active or passive methods [1,2]or by decreasing the temperature of glass cover or a combination ofboth. One of the methods of reducing the temperature of glass cover isby continuous flow of water over the glass cover (Fig. 1) as was

. Somwanshi).

investigated by Tiwari and Rao [3]; they analyzed the transient perfor-mance of the solar still with water flow over the glass cover. It wasseen that the daily distillate output got doubled by lowering the tem-perature of the glass cover by continuous flow of water over it at uni-form velocity. Tiwari and Maduri [4] incorporated the flow of wastehot water in the basin along with water flow over the glass cover andobtained the increase in distillate output, commensurate with the in-crease in inlet water temperature in the basin. Lawrence et al. [5] vali-dated their model by incorporating the effects of water flow over thecover and heat capacity of water mass in the basin. They found an in-crease of 7 and 10% in efficiency of solar still due to water flow overthe glass cover in the cases with and without black dye present in thebasin of the solar still; moreover there is no change in efficiency ofsolar still at a flow rate of 1.5 m/s. Abu-Hijleh [6] theoretically investi-gated the effectiveness of film cooling under different operating charac-teristics; his results indicated that the proper use of the film coolingparameters can increase the still efficiency by 6% but poor combinationcan reduce still efficiency. Abu-Hijleh andMousa [7] extended the earli-er work and included the evaporation effect of water film flowing overthe glass cover; they obtained a 20% increase in the still efficiency.

Fig. 1. a: Schematic representation of a solar still coupled to a desert cooler. b: Over side view of the flowing water over the glass cover.

93A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

Dhiman andTiwari [8] presented an analyticalmodel of amultiwick typesolar still with water flowing over the glass cover and obtained 10% in-crease in distillate output. Janarthanan et al. [9] experimentally validatedthe model of a tilted wick type solar still with water flowing over theglass cover and concluded that the optimum flow rate of water flowingover the glass cover is 1.5 m/s. Ahmed and Alfaylakawi [10] experimen-tally evaluated the effect ofwind speed and the use ofwater sprinklers tocool the temperature of glass cover of the solar still; an increase in pro-duction rate by 15.7% and 31.8% was obtained, using water sprinklersat intervals of 20 min and 10 min respectively. Badran [11] appliedwater sprinkler and asphalt basin liner in a single slope solar still and ob-served experimentally a 51% increase in still productivity.

In the earlier analyseswithwater flow over the glass cover to reducethe temperature of glass cover, the initial temperature of water flowingover the glass cover is assumed to be either the ambient temperature ora temperature significantly higher than the wet bulb temperature. Inthe present communication the authors have presented an analyticalmodel of a conventional solar still with provision of flowing cooledwater (at wet bulb temperature) from the tank of an air cooler. Wateris collected in a tank placed below solar still and circulated back to theair cooler (Fig. 1a). Air cooler also knownas evaporative or desert cooler,is used to cool air. Cooling of air is accomplished by flow of air inducedby a fan in a direction normal to the vertical porous pad through whichwater flows from top to bottom; this water collects in a tank at the bot-tomof the pad and is pumped up to the top. Evaporation ofwater resultsin cooling of air as well as water. Cooled air is utilized for space condi-tioning whereas the coolness stored in tank water is not utilized; onecan use this coolness stored in tank water to cool extraneous objects(cover of stills) moreover it is seen that utilization of coolness of waterin the tank does not appreciably affect the temperature of air [16].

Present paper explores one such possibility in which cooled waterfrom the tank of the desert cooler is distributed through a commonpipe and circulated uniformly over the glass cover of a solar still(Fig. 1a) or an array of solar stills depending on the size of the coolerand its height above the solar still. In the present analysis the authorshave considered two stills working together; mass flow rate of waterflowing over the glass cover can be adjusted by a valve provided withan inlet pipe; water falling from the cover of the still is collected in atank provided at the bottom of still and circulated back to the air cooler.Authors have suggested two different strategies for summer and othermonths of the year. During summer months when cooling of air is re-quired the air cooler and solar still works simultaneously and the steadystate temperature of water in the cooler tank becomes nearly equal towet bulb temperature of the ambient air [12–16]. In other monthswhere cooling of air is not required and the cooler is not running thetank of the cooler is kept open to the atmosphere (in shade) for dayand night, and the temperature of water in the open tank is nearlyequal to the wet bulb temperature of air. As the steady state tempera-ture of water in the tank of the air cooler is very nearly equal to wetbulb temperature of ambient air for both the strategies, the temperatureof the glass cover should be lower than that in the case of using inletwater at ambient temperature, and consequently the rate of condensa-tion of water over the glass cover should increase; moreover the re-circulated water from the still to the air cooler does not appreciably af-fect the performance of the air cooler [16]. The presence of a water filmforms an intermediate layer of refractive index 1.33 between the air ofrefractive index 1.0 and glass with refractive index 1.5; this reducesthe intensity of the incident light which gets reflected and increasesthe transmission of solar radiation into the still. In the present paperthe authors have numerically computed and compared the annual

94 A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

distillate output for a conventional solar still with 2 cmdepth ofwater inthe basin, with and without the flow of water from the tank of an aircooler at wet bulb temperature of air and at ambient temperatureover the glass cover for four different climatic zones viz hot and dry,warm and humid, composite, and moderate climates in the Indianplains. Monthly average values of hourly metrological data [17] ofmajor cities like Jodhpur, Chennai, Delhi and Bangalore correspondingto different climatic zoneswere used for computation. To explore the ef-fects of parameters like depth of water in the basin of still and the de-pendence of mass flow rate of flowing water over the glass cover onyield, authors computed daily yield for two typical days in summerand winter (24/05/2012, 16/12/2012) for hot and dry climate of Jodh-pur. Moreover the authors have also computed the annual efficiencyof single slope solar still for different climatic zones with and withoutflow of water over the glass cover. Effect of flowing ordinary water(water at ambient temperature) over the glass cover to increase pro-duction of distilled water in the still was also investigated. Additionalcost of water produced by adopting the strategy suggested in the pres-ent paper was also computed.

2. Analysis

2.1. Determination of distillate output

The analysis by Tiwari and Rao [3] has been followed, except that thedependence of the saturation vapor pressure on temperature has notbeen approximated by a linear relation but by a better fitting exponen-tial relation. Accordingly the systems of equations were solved numeri-cally. Energy balance for various parts of solar still is shown in Fig. 2.The following assumptions have been made to write energy balanceequations:

• The solar distillation unit is vapor proof.• Heat capacities of glass and basinmaterials are negligible as comparedto that of water in the basin.

• The temperature gradients along the glass cover thickness and thatof themovingwater film over the glass cover have not been consid-ered.

• The areas of the glass cover and basin have been assumed to bethe same, and for this the inclination of the cover to the horizonis (10°0-15°0).

Fig. 2. Schematic diagram showing energy balance of various parts of solar still.

• Steady state temperature of water in the tank of the desert cooler isassumed to be equal to the wet bulb temperature of ambient air [16].

• Piping and pumping losses are negligible.• Thickness of water film is assumed to be very small.

Considering a thin strip of length dxover the glass cover (Fig. 1b), theenergy balance of the flowing water is given by [3],

Mf cw∂T f

∂t dxþ mf cw∂T f

∂x dx ¼ h02 Tg−T f

� �bdx−h2 T f−Ta

� �bdx− qef bdx;

ð1Þ

where, b is the breadth of the cover (normal to the direction of flow-xdirection), andMf, cw and mf are the mass of water film per unit lengthalong the direction of flow, specific heat of water and themass flow rateof water over the glass cover. Since the depth of the water film is verysmall the first term on the left hand side of Eq. (1) may be neglected.Thus Eq. (1) can be written as,

mf cw∂T f

∂x dx ¼ h02 Tg−T f

� �bdx−h2 T f−Ta

� �bdx− qef bdx; ð2Þ

where h ' 2 is the convective heat transfer coefficient from the glasscover to theflowingwaterfilm, and h2 is the sumof radiative heat trans-fer coefficient hrfa and convective heat transfer coefficient hcfa betweenthe water film and air.

h2 ¼ hrfa þ hcfa;

hrfa ¼ εwσT f þ 273

� �4− Ta þ 261ð Þ4

T f−Ta

� �264

375;

hcfa ¼ 2:8þ 3v:

[18] v is the velocity of air in (m/s).qef in Eq. (2) is the rate of evaporative heat transfer per unit area from

the film to air and is given by [18],

qef ¼ 0:013hcfa P f−γPa

� �;

mf ¼ bρulw is the mass flow rate of film water, b is the breadth of thesolar still, lw is the depth of the film and u is the flow velocity. Pf andPa are the saturated vapor pressures of air at water film temperatureand the saturated vapor pressure of air, given by [18],

P f ¼ exp 25:317−5144= T f þ 273� �n o

Pa ¼ exp 25:317−5144= Ta þ 273ð Þf g;

and γ is the relative humidity of air.Energy balance of the glass cover is as follows [18].

(a) With water film

h1w Tw−Tg

� �þ τ1S ¼ h02 Tg−T f

� �ð3aÞ

where h1w is the sum of radiative, convective and evaporativeheat transfer coefficient between the water in the basin and theglass cover, hrw, hcw and hew are the radiative, convective andevaporative heat transfer coefficients respectively and are givenby [18],

hrw ¼ εeffσ Tw þ 273ð Þ2 þ Tg þ 273� �2

� �Tw þ Tg þ 546

� �;

hcw ¼ 0:884 Tw−Tg

� �þ

Pw−Pg

� �Tw þ 273ð Þ

268:9� 103−Pw

� �24

351=3

hew ¼ 16:273� 10−3hcwPw−Pg

� �Tw−Tg

� � ;

Table 1Still parameters used for numerical computations.

Mw = 100 kg d = 0.1 m cw = 4190 J/kg KAg = Ab = 1 m2 αg = 0.05 [18] αb = 0.8 [18]hw = 135 W/m2K [5] h ' 2 = 135 W/m2K [5] αw(0.1 m) = 0.45 [18]εeff = 0.82 [18] εw = 0.95 [18] h = 0.75 mσ = 5.67 × 10−8 W/ m2K4 eg = 0.94 [18]

Table 3Average meteorological parameters [17] for hot and dry climate (Jodhpur); wet bulbtemperature computed by [21].

Month Ta (oC) S (W/m2) γ v (m/s) Twb (oC)

Jan 15.8 164.6 41.0 3.1 9.4Feb. 18.5 190.8 34.9 3.4 10.7Mar 24.5 230.8 27.1 3.3 13.8Apr. 29.0 266.7 28.7 3.7 17.2May 31.2 284.2 37.2 4.2 20.6June. 30.6 276.3 55.8 4.3 23.7July 28.9 233.3 69.6 3.6 24.6Aug. 28.0 220.8 72.1 3.1 24.1Sep 28.0 221.2 59.2 3.2 22.0Oct. 26.6 195.0 37.1 2.9 17.1Nov. 22.0 170.0 31.6 3.0 12.7Dec. 17.6 150.0 37.9 3.6 10.4

95A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

and τ1 in Eq. (3a) is the fraction of incident solar energy absorbedby the glass cover.

(b) Without water filmThe energy balance of the cover is

h1w Tw−Tg

� �þ τ01S ¼ h1g Tg−Ta

� �ð3bÞ

where h1g is the sum of radiative and convective heat transfercoefficients between the glass cover and the ambient,

h1g ¼ hrg þ hcg ;

hrg and hcg radiative and convective heat transfer coefficients andare given by [18],

hrg ¼ εwσTg þ 273

� �4− Ta þ 261ð Þ4

Tg−Ta

� �264

375

and hcg = 2.8 + 3ν and τ ' 1 in Eq. (3b) is the fraction of solarenergy absorbed by glass cover without flow of water over it.

Energy balance equation for water in the basin is [18],

τ2Sþ hw Tb−Twð Þ ¼ MCð ÞwdTw

dtþ h1w Tw−Tg

� �; ð4Þ

where Tb is the temperature of basin liner, Tg is the temperature of glasscover. τ2 is the fraction of solar energy absorbed by water mass; τ2 isreplaced by τ ' 2 when the glass cover is without a water film. hw isthe heat transfer coefficient between basin liner and water.

Table 2Meteorological data for two typical days in summer and winter (24/05/2012, 16/12/2012)for hot and dry climate of Jodhpur, monthly average of wind velocity for summer (4.2 m/s)and winter months (3.6 m/s) is considered for computation of daily yield.

Time Day-1 (summer/May) Day-2 (winter/December)

Ta (oC) γ S (W/m2) Ta (oC) γ S (W/m2)

1 29.2 32 0 12.8 64 02 28.4 32 0 12.0 67 03 27.9 34 0 11.3 70 04 27.4 34 0 10.9 72 05 27.3 34 0 10.7 73 06 27.6 33 166 11.0 60 1007 28.3 33 378 11.8 58 1088 29.6 30 586 13.3 53 2539 31.4 28 763 15.3 48 42310 33.6 25 901 17.7 40 55711 36.0 22 985 20.5 32 64012 38.3 18 1016 23.0 29 66813 40.0 17 985 24.9 25 64014 41.2 17 901 26.2 26 55715 41.6 17 763 26.7 25 42316 41.2 17 586 26.2 25 25317 40.2 18 378 25.1 26 10818 38.6 18 166 23.3 28 10019 36.7 20 0 21.3 31 020 34.9 23 0 19.2 35 021 33.3 24 0 17.4 41 022 31.9 27 0 15.8 45 023 30.7 28 0 14.5 50 024 29.9 30 0 13.6 55 0

Energy balance for the bottom of the basin is,

τ3S ¼ hw Tb−Twð Þ þ hb Tb−Tað Þ ð5Þ

where τ3 is the solar energy absorbed bybasin liner and τ3 is replaced byτ ' 3 when the glass cover is without a water film.

From Eqs. (3a), (3b) and (5),

Tg ¼ h1wTw þ h02T f þ τ1Sh1w þ h02ð Þ ; ð6aÞ

Tg without filmð Þ ¼ h1wTw þ h1gTa þ τ01S

h1w þ h1g� � ; ð6bÞ

Tb ¼ hwTw þ hbTa þ τ3Shw þ hbð Þ ð7aÞ

Tb without filmð Þ ¼ hwTw þ hbTa þ τ03Shw þ hbð Þ : ð7bÞ

Putting the above expressions for Tg and Tb in Eq. (4), one obtains,

dTw

dtþ A1Tw ¼ A2 ð8Þ

(with water film on cover)

and;dTw

dtþ A0

1Tw ¼ A02 ð9Þ

Table 4Averagemeteorological parameters [17] forwarmand humid climate (Chennai);wet bulbtemperature computed by [21].

Month Ta (oC) S (W/m2) γ v (m/s) Twb (oC)

Jan 25.1 205.4 66.5 3.9 20.7Feb. 26.1 245.4 64.0 3.5 21.1Mar. 27.5 276.7 63.0 3.5 22.3Apr. 28.3 280.0 70.1 3.5 24.0May 29.7 255.0 68.6 3.8 25.1June. 30.1 218.3 65.0 4.4 24.8July 29.5 197.1 66.4 4.2 24.5Aug. 29.4 200.0 66.7 4.1 24.5Sep. 28.5 208.7 71.7 3.0 24.5Oct. 27.0 184.2 76.9 2.8 23.9Nov. 26.1 169.2 74.2 3.6 22.6Dec. 25.5 176.7 68.2 4.2 21.3

Table 5Averagemeteorological parameters [17] for moderate climate (Bangalore); wet bulb tem-perature computed by [21].

Month Ta (oC) S (W/m2) γ v (m/s) Twb (oC)

Jan 22.4 223.3 61.3 2.2 17.5Feb. 25.0 252.5 51.2 2.2 18.2Mar. 27.5 273.3 46.2 2.0 19.4Apr. 27.2 265.8 61.1 1.8 21.6May 26.7 251.2 66.3 2.1 22.1June. 24.9 201.7 72.8 2.9 21.3July 24.3 187.5 73.3 2.9 20.9Aug. 24.5 186.2 71.5 2.6 20.8Sep. 25.1 209.6 67.8 2.0 20.8Oct. 24.1 192.9 70.9 1.7 20.3Nov. 22.8 187.5 69.3 1.9 18.9Dec. 21.9 197.5 66.5 2.2 17.8

96 A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

(without water film on cover) where,

A1 ¼ 1Mwcw

hw þ h1w−hwð Þ2

hw þ hbð Þ−h1wð Þ2

h1w þ h02ð Þ

" #;

A2 ¼ A2p þ A2qT f ;

A2p ¼ 1Mwcw

hwτ3Shw þ hbð Þ þ

h1wτ1Sh1w þ h02ð Þ þ

hwhbTa

hw þ hbð Þ þ τ2S� �

;

A2q ¼1

Mwcw

h1wh02

h1w þ h02ð Þ� �

;

A01 ¼ 1

Mwcwhw þ h1w−

hwð Þ2hw þ hbð Þ−

h1wð Þ2

h1w þ h1g� �

24

35

Table 6Average meteorological parameters [17] for composite climate (Delhi); wet bulb temper-ature computed by [21].

Month Ta (oC) S (W/m2) γ v (m/s) Twb (oC)

Jan 13.4 158.3 51.8 2.2 8.7Feb. 16.6 195.0 46.3 2.5 10.7Mar. 22.7 241.7 36.0 2.7 14.0Apr. 28.1 262.5 31.4 3.1 17.1May 28.1 267.5 38.2 3.2 20.8June. 31.7 252.9 53.3 3.2 24.2July 29.2 217.5 74.7 2.7 25.6Aug. 28.0 200.4 79.0 2.3 25.2Sep. 26.8 210.8 71.9 2.2 22.9Oct. 23.8 201.2 52.9 1.7 17.5Nov. 19.3 174.2 42.3 1.4 12.3Dec. 14.8 146.7 47.9 1.9 9.4

Fig. 3. Variations in daily yield at different mass flow rates for a typica

and,

A02 ¼ 1

Mwcw

hwτ03S

hw þ hbð Þ þh1wτ

01S

h1w þ h1g

� �þ hwhbTa

hw þ hbð Þ þ τ02Sþh1wh1gTa

h1w þ h1g

� �24

35:

It may be remembered that A and A ' refer to the cases with the filmand without film situations.

In a small interval of time A1, A1' , A2, A2' may be treated as constants

and the solution of Eqs. (8) and (9) may be written as

Tw ¼A2p þ A2qT f

� �A1

1− exp −A1tð Þ½ � þ Tw0 exp −A1tð Þ ð10Þ

and,

Tw without filmð Þ ¼ A02

A01

1− exp −A01t

� �� þ Tw0 exp −A01t

� �: ð11Þ

Tw0 in Eqs. (10) and (11) is the initial temperature of water in the basinat time t = 0.

The average water temperature in the basin is,

Tw ¼

Zt0

Tw without filmð Þdt

t

¼ A02

A01

1− 1− exp −A01t

� � �A0

1t

" #þ Tw0

1− exp −A01t

� � �A0

1t:

ð12Þ

By using Eq. (12) with a time step of 1 h the daily average tempera-ture of water in the basin can be obtained, the average temperature ofthe glass cover can be obtained by putting Tw equal to Tw in Eq. (6b).The hourly yield of solar still (without water film) is given by,

m•

ew ¼hew Tw−Tg

� �L

� 3600 ð13Þ

because Tw is a assumed to be constant over a period of 1 h (size of step)in computation.

Eq. (10) can be written as,

Tw ¼ A3 þ A4T f ð14Þ

l day in summer (24/05/2012) for hot and dry climate of Jodhpur.

Table 7Daily yield of solar still (ml/m2) at various depths of water in a still for summer and winter seasons for hot and dry climate of Jodhpur; σ is the percentage difference of output for depth0.1 m and 0.02 m (v = 4.2 m/s and v = 3.6 m/s for summer and winter).

Summer (24/05/2012) σ Winter (16/12/2012) σ

Depth- 0.02 m 0.05 m 0.1 m 0.02 m 0.05 m 0.1 mWithout water flow 4209 3929 3561 18.2 1332 1297 1199 11.1Flowing ordinary water. (mf = 0.001 kg/s) 5290 5020 4721 12.0 1759 1635 1498 17.7Flowing water from cooler. (mf = 0.075 kg/s) 6038 5830 5590 8.0 1989 1834 1663 19.6

97A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

where,

A3 ¼ A2p

A11− exp −A1tð Þ½ � þ Tw0 exp −A1tð Þ

and,

A4 ¼ A2q

A11− exp −A1tð Þ½ �:

A3 and A4 can be obtained by knowing initial temperature of water inbasin and by considering the time step of 1 h.

From Eqs. (14) and (6a) the temperature of glass cover is,

Tg ¼h1w A3 þ A4T f

� �þ h02T f þ τ1S

h1w þ h02ð Þ : ð15Þ

Putting this value of Tg, Eq. (2) can be solvedwith the known bound-ary conditions viz at x = 0, Tf = Tf0 and the average film temperaturecan be obtained as,

T f ¼T f1 þ T f0

2ð16Þ

here Tf1 obtained by solving Eq. (2) is the temperature of water film atthe exit (x = l). Putting the value of the average film temperature inEq. (15) the average value of glass temperature Tg can be obtained,and the water temperature can be obtained by Eq. (14). The hourly

Fig. 4.Hourly variations of daily yield for a typical day in summer (24/05/2012) for hot anddry climate of Jodhpur.

yield is given by,

m•

ew ¼hew Tw−Tg

� �L

� 3600; ð17Þ

where L is the latent heat of vaporization of water is given by [19],L = 2.4935 × 106[1 − 9.4779 × 10−4T + 1.3132 × 10−7T2 −4.7974 × 10−9T3]., for T ≤ 70 °C.

The daily yield Md is given by,

Md ¼X241

m•

ew: ð18Þ

Further,

Monthly yield Mm ¼ Mdn; ð19Þ

where n is the number of days in a month. The annual yield is given by

Ma ¼X121

Mm:

2.2. Fraction of solar energy absorbed by glass cover, water and basin

For normal incidence of light, the reflection coefficient at the inter-face of two media is given by [20],

R ¼ μ1−μ2ð Þ2μ1 þ μ2ð Þ2 ;

where μ1 and μ2 are refractive indices of the two media.

Fig. 5. Hourly variations of daily yield for a typical day in winter (16/12/2012) for hot anddry climate of Jodhpur.

Fig. 6. Hourly variation of temperature of water in basin for a typical day in summer(24/05/2012) for hot and dry climate of Jodhpur.

Fig. 8. Hourly variations of difference between water temperature in basin and averageglass temperature.

98 A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

Fraction of solar energy absorbed by the glass cover is,

τ1 ¼ αg 1−Rawð Þ 1−Rwg

� �with water film ð20Þ

and

τ01 ¼ αg 1−Rag

� �without water film ð21Þ

where Raw, Rwg and Rag are the reflection coefficient of air–water, water–glass and air–glass interfaces respectively and αg is the absorption coef-ficient of glass cover and is given by

αg ¼ 1− exp −βglg� �

:

lg is the thickness of the glass cover and βg is the attenuation factor ofglass.

Fraction of solar energy absorbed by water in the basin is,

τ2 ¼ Pαw þ P 1−αwð Þ 1−αbð Þ:P ¼ 1−αg

� �1−Rawð Þ 1−Rwg

� �1−Rgw

� �1−Rwað Þ 1−Rawð Þ; ð22Þ

Fig. 7.Hourly variation of average temperature of glass covers for a typical day in Summer(24/05/2012) for hot and dry climate of Jodhpur.

τ02 ¼ P0αw þ P0 1−αwð Þ 1−αbð Þ:and; P0 ¼ 1−αg

� �1−Rag

� �1−Rgw

� �1−Rwað Þ 1−Rawð Þ: ð23Þ

Here Rgw, Rwa, and Raw are the reflection coefficients of glass–water,water–air, and air–water interface respectively; αw and αb are theabsorption coefficients of water and basin.

Fraction of solar energy absorbed by the basin is,

τ3 ¼ P 1−αwð Þαb: ð24Þ

τ03 ¼ P0 1−αwð Þαb: ð25Þ

2.3. Annual efficiency, pumping power and electrical energy (kWh) required

Annual efficiency of the solar still is given by

ηa ¼MaLX121

S:t

: ð26Þ

Fig. 9.Monthly yield with and without water film for hot and dry climate (Jodhpur).

Fig. 10. Monthly yield with and without water film for warm and humid climate(Chennai). Fig. 12.Monthly yield with and without water film for composite climate (Delhi).

99A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

Pumping power required to pump water from still to cooler will be,

Pp ¼ mght

Wð Þ: ð27Þ

Here m, g, h and t are the mass of water (kg) required to pump, ac-celeration due to gravity (m/s2), water head (m) and time (s) neededto pump water.

Annual electrical energy (kWh) needed to pump water will be,

E kWhð Þ ¼ Ppn0t

1000: ð28Þ

n ' and t are number of days in a year and time in hours.

3. Numerical computation and discussion

Numerical computations have beenmade to obtain daily and annualyield of a still for three different cases viz. (i) bare glass cover (withoutflowing water over cover), (ii) glass cover with water flow over coverheaving mean ambient temperature (ordinary water), and (iii) glass

Fig. 11. Monthly yield with and without water film for moderate climate (Bangalore).

cover with flow of water at wet bulb temperature. The mean monthlyvalues of the meteorological parameters of major cities correspond-ing to different climatic zones used for computations are given inTables 3–6, and meteorological parameters for two days in summer andwinter (24/05/2012, 16/12/2012) are given in Table 2. Wet bulb temper-ature was computed by using values of ambient air temperature and rel-ative humidity [21]. Daily yieldwithwaterflowing fromdesert coolerwascomputed at five different mass flow rates viz 0.010 kg/s, 0.025 kg/s,0.050 kg/s, 0.075 kg/s and 0.085 kg/s and with flow of water at ambientyield was computed for mass flow rates of 0.010 kg/s, 0.025 kg/s,0.005 kg/s, 0.001 kg/s and 0.0001 kg/s. Still parameters used for com-putations are given in Table 1. For numerical computations initial tem-peratures of water and glass were assumed to be nearly equal to (i) theambient and (ii) nearly equal to the wet bulb temperature of ambientair; heat transfer coefficients and other temperature dependent con-stants were computed by use of initially guessed values, by consideringtime step of 1 h and length of still (1 m). Exit film temperature and av-erage film temperature were computed from Eqs. (2) and (16), averageglass temperature with and without filmwas computed from Eqs. (15),(6a) and (6b) and water temperature with and without film was com-puted from Eqs. (10) and (11). For computing daily yield the processwas repeated till the daily cycle repeated itself and for computing annu-al yield the process was repeated till the initial and final temperaturesconverged.Monthly and annual yields for different zoneswere comput-ed from Eqs. (13), (17), (18), and (19). By considering depth ofwater inbasin as 2 cmdaily yield at differentmassflow rates for a day in summerseason for hot and dry climate of Jodhpur is calculated and shown inFig. 3. It is seen that with flow of water at ambient over the glasscover yield increases with decrease in the mass flow rate, as waterflowing over the glass cover evaporates thus decreasing the averagetemperature of glass cover with a little effect on temperature of waterin the basin, and higher temperature differential enhances distillation;rate of evaporation is more with low mass flow rate thus increasesyield. From mass flow rate of 0.025 kg/s to 0.001 kg/s yield increasesby 6.1%, moreover yield saturates at 0.001 kg/s. With flow of water atwet bulb temperature (from cooler) there is very little increase inyield with increase in mass flow rate. As rate of evaporation of flowingwater is very less (almost negligible) there is a slight increase in tem-perature of water flowing over cover, which is more at a low massflow rate and thus decreases the temperature difference betweenwater in basin and average temperature of glass cover. Yield increasesonly by 1.8% from mass flow rate of 0.010 kg/s to 0.075 kg/s; it tendsto saturate atmassflow rate of 0.075 kg/s. The daily yield for two typicaldays in summer and winter for all the three cases with 0.02 m, 0.05 m,

Table 8Annual yield for various climatic zoneswith andwithout flowingwater over the glass cover. x is the percentage increase in annual yield byflowingwater from cooler and y is the percent-age increase in annual yield by flowing water at ambient temperature.

Without flow Flowing water at ambient Flowing water from cooler x y

Jodhpur 511 kg/m2 665 kg/m2 800 kg/m2 56.5% 30.1%Chennai 542 kg/m2 660 kg/m2 766 kg/m2 41.3% 21.8%Bangalore 459 kg/m2 590 kg/m2 705 kg/m2 54.6% 28.4%Delhi 462 kg/m2 589 kg/m2 715 kg/m2 54.8% 27.5%

100 A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

and 0.1mwater depths in basin is computed and shown in Table 7, andmass flow rate is taken as 0.075 kg/s (flowing water from cooler) and0.001 kg/s (flowing water at ambient). It is seen that for all the threecases the yield increases at lower water depths for summer as well aswinter. As the heat capacity of water in the basin is small for lowerdepth, the rise in temperature of water in basin is morewhich increasesproduction rate of still. The hourly variation of daily yield for summerand winter for Jodhpur climate keeping 2 cm water depth and massflow rate 0.075 kg/s (flowing water from cooler) and 0.001 kg/s(flowing water at ambient) is shown in Figs. 4 and 5. It is seen thatwith flow of water over the glass cover at ambient temperature theaverage temperature of glass cover reduces with some effect in temper-ature of water in basin but as the difference between temperature ofwater in the basin and average glass cover increases, thus increasesthe production rate of still. Average temperature of glass cover is signif-icantly reduced further more with water at wet bulb temperatureflowing over the glass cover. It is seen that the daily yield increases by43.4% and 49.3% in summer and winter respectively with flow ofwater from cooler, whereas increase in daily yield is about 25.7% and32% in summer and winter respectively with flow of water at ambientover the glass cover. The hourly variation of temperature of water inbasin and average temperature of glass cover of a summer day for hotand dry climate of Jodhpur considering all three cases is shown bycurves A, B and C in Figs. 6 and 7. It is clearly seen that with flow ofwater at lower temperature the temperature of glass cover and waterin basin is reduced. Reduction is largest with water flowing at wetbulb temperature. The largest reduction in average glass cover temper-ature is about 36.1 °C and for basin water it is around 18.6 °C. Thelargest reduction occurs in the afternoon; because higher ambient tem-perature and solar insolation in noon time increase the temperature ofglass cover and water in the basin for bare still (without film). Due to

Fig. 13. Maximum yield with flowing water from desert coolers for different climaticzones.

lower values of wet bulb temperature in noon hours the temperatureof flowing water over the cover is less, which increases the reductionin glass cover temperature and water temperature in the basin. Thehourly variation in difference between temperature of water in thebasin and average glass cover temperature is shown in Fig. 8. It isevident that the difference is largest with flow of water over cover atwet bulb temperature (curve A) as compared to the case of ambienttemperature water flow or bare still. There is corresponding effect onthe output.

By keeping the depth of water in basin as 2 cm and with mass flowrates 0.001 kg/s (flowing water at ambient) and 0.075 kg/s (flowingwater from cooler) annual yield is computed for all four climatic zonesand are shown in Figs. 9–12 and corresponding values of annual yieldis given in Table 8, here x and y are the percentage increase in annualyield. It is clearly seen that with flow of water from the cooler the in-crease in annual yield is maximum (56.5%) for hot and dry climate ofJodhpur due to low values of relative humidity and wet bulb tempera-ture. Increase in annual yield is minimum for warm and humid climateof Chennai (41.3%) as due to high humidity and wet bulb temperature.For composite climate of Delhi and moderate climate of Bangalore in-crease in annual yield is 54.8% and 54.6% respectively. With flow ofwater at ambient over the glass cover increase in annual yield is againmaximum of 30.1% for hot and dry climate of Jodhpur and minimumof 21.8% for warm and humid climate of Chennai and for Delhi and Ban-galore increase in yield is 27.5% and 29.4% respectively. For bare still(withoutfilm) as relative humidity is not going to affect distillate outputannual yield is maximum for warm and humid climate of Chennai andminimum for composite climate of Delhi. Monthly distillate outputwith flow of water from cooler over cover for different climatic zonesis shown in Fig. 13. For the climate of Jodhpur and Delhi yield is maxi-mum in the month of May and minimum in the month of January,

Fig. 14. Annual efficiency of solar still for various climatic zones.

Table 9Monthly shortage (-)/surplus (+) of distilled water produced by the system with flow ofwater from cooler over cover for specific demand of a family. Y is monthly yield in liters(l) and S is shortage/surplus (l) of water produced.

Jodhpur Chennai Bangalore Delhi

Y S Y S Y S Y S

Jan. 365.2 −134.8 627.6 +127.6 689.8 +189.8 329.8 −170.2Feb. 464.2 −35.8 747.9 +247.9 856.3 +356.4 457.6 −42.4Mar. 797.8 −202.2 956.8 −43.2 1103.2 +103.2 826.3 −173.7Apr. 1041.3 +41.3 944.3 −55.7 968.7 −31.3 1096.5 +96.5May. 1192.4 +192.4 921.6 −78.4 892.2 −107.8 1258.4 +258.4Jun. 1007.6 +307.6 786.5 +86.5 629.3 −70.7 1090.4 +390.4Jul. 770.7 +70.7 711.6 +11.6 568.9 −131.1 778.9 +78.9Aug. 694.6 −5.4 715.1 +15.1 576.3 −123.7 665.7 −34.3Sep. 731.5 +31.5 676.5 −23.5 665.1 −34.9 687.6 −12.4Oct. 722.8 +22.8 573.2 −126.8 607.1 −92.9 675.4 −24.6Nov. 527.2 −172.8 485.8 −214.2 546.6 −153.4 495.3 −204.7Dec. 378.3 −121.7 541.4 +41.4 591.1 +91.1 337.1 −162.9

(Note: These figures correspond to stills of area (m2) = average monthly demand/average monthly yield).

101A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

and for summer months (Mar., Apr., May) the total yield is significantlyhigher than that of winter months (Dec., Jan., Feb.). For Jodhpur andDelhi climate total yield for the summer months is about 2.5 to 2.8times the total yield for the winter months. For moderate climate ofBangalore and warm and humid climate of Chennai the maximumyield and minimum yield are in the month of March and November re-spectively, and the total yield for the summermonths is not significantlyhigher than the total yield for the winter months. For Bangalore andChennai climate the total yield for summer months is about 1.4 to 1.5times the total yield for winter months. Annual efficiency of still withand without flowing water over glass cover is computed from Eq. (26)and shown in Fig. 14. It is clearly seen thatwithflow of water from cool-er over the glass cover increase in annual efficiency is maximum about9.9% for Jodhpur climate and minimum of 7.4% for Chennai climate,and for Delhi and Bangalore annual efficiency increases by 8.7% and8.2% respectively. Monthly shortage (-) or surplus (+) of distilledwater produced by the systemwith flow of water over cover, for differ-ent climates is evaluated and shown in Table 9. For computingmonthlyshortage/surplus demandof drinkingwaterwas assumed to be same forall the climatic zones. Demand of drinking water for a family wasassumed to be 500 l/month for winter months (Dec., Jan., Feb.),1000 l/month for summer months (Mar., Apr., May) and 700 l/monthfor the remaining months of the year. Corresponding area of solar stillrequired to meet the above demand is calculated by dividing averagemonthly demand (constant) by average monthly yield for different cli-matic zones. Requirement of still area to meet the average demand isthe largest of 12.34 m2 for the climate of Bangalore and the least of10.87 m2 for the climate of Jodhpur. For the climate of Chennai andDelhi area required to meet the average demand is 11.35 m2 and12.17 m2 respectively.

4. Cost analysis

With the present concept stated in the paper, water from the aircooler can be utilized for cooling glass cover of more than one solarstill kept in line, depending on size and height of air cooler from thestill. While performing cost analysis two stills of area 1 m2 (each) areconsidered; moreover the cost also depends on the distance and theheight of the cooler from the solar still. In the present analysis the dis-tance and height of cooler from still are taken as 2 m. Fixed and runningcost of air cooler was not included due to the fact that desert cooler isutilized for conditioning of room during summer months. Duringother months when air conditioning of the room is not required, thetank containing water is open to the atmosphere in shade (removing

pads and a detachable arrangement provided in top) will be used as astorage tank.

Fixed cost (F) (U.S.D. 1.00 = Rs. 62.50)

a. Pipe 7 m (1.5″, PVC) + valve + fittings(Distance of cooler from still is 2 m,additional 3 m for other arrangements)

Rs. 1000/-(@100/m for pipe and Rs.300/- for fitting)

b. Pump (350 l/1.5 W)(Efficiency 80%)

Rs. 600/-(@200/- with two replacements)

c. Water storage tank(G.I. sheet 0.25 mm thickness)

Rs. 500/(2.2 × 0.5 × 0.25) m

d. d. Fitting and labor costTotal fixed cost (F)

Rs. 100/-Rs. 2200/-

Running cost (R)

Pumping power required for 2 m water head (height from cooler tostill is 2m) and for 0.075 kg/m2 smassflow ratewill be 1.5W(Eq. (27)),electrical energy consumed was calculated from Eq. (28); the annualrunning cost was calculated by considering the cost of electricity as Rs.4/kWh

a. Annual running cost (R)

Rs. 53/-

The useful life (n ' ') for pipes and pump (with two replacements)was taken as 10 years and rate of interest i taken as 6%; the capitalrecovery factor (C.R.F.) will be given by,

C:R:F ¼ i 1þ ið Þn00

1þ ið Þn00−1: ð29Þ

Annual fixed cost (A) will be given by,

A ¼ C:R:F � F ð30Þ

Annual fixed cost

Rs. 299/- Total annual cost Rs. 352/- (A + R)

The cost of additionalwater produced annually is computed for differ-ent climatic zones with two solar stills as follows: Rs. 0.60/l for Jodhpurclimate, Rs. 0.78/l for Chennai climate, and for Delhi and Bangalore it isRs. 0.69/l and Rs. 0.71/l respectively (U.S.D. 1.00 = Rs. 62.50).

5. Conclusions

1. With flow of water at wet bulb temperature over the glass cover, thetemperature of glass cover gets significantly reduced, also reducingthe temperature ofwater in thebasin. But the temperature differencebetween the water in basin and the average glass cover increases,increasing the distillate output.

2. With flow of water at wet bulb temperature (from cooler), there is aslight increase in distillate output with mass flow rate and the yieldtends to saturate at a mass flow rate of 0.075 kg/s. With flow ofwater at ambient temperature, the distillate output decreaseswith in-crease inmass flow rate and saturates at mass flow rate of 0.001 kg/s.

3. Increase in annual yield is in between 41.3% and 56.5% with flow ofwater from the desert cooler, and increase is in between 30.1% and21.8% with flow of water at ambient temperature. Increase in annualyield is the largest for hot and dry climate of Jodhpur and the least forwarm and humid climate of Chennai.

4. With two stills served by a desert cooler the cost of additional waterproduced by the solar still is the least (Rs. 0.60/l) for Jodhpur and thehighest (0.78/l) for Chennai; for Delhi and Bangalore it is 0.69/l and0.71/l respectively. (USD 1.00 = Rs. 60.00).

102 A. Somwanshi, A.K. Tiwari / Desalination 352 (2014) 92–102

Acknowledgment

The authors are grateful to Dr. Amrit Dixit of DIMAT Raipur and tothe reviewers for valuable suggestions.

References

[1] M.S. Sodha, A. Kumar, G.N. Tiwari, G.C. Pandey, Effects of dye on the performance ofa solar still, Appl. Energy 7 (1980) 147–162.

[2] G.N. Tiwari, H.N. Singh, R. Tripathi, Present status of solar distillation, Sol. Energy75-3 (2003) 367–373.

[3] G.N. Tiwari, V.S.V. Bapeshwara Rao, Transient performance of a single basin solarstill with water flowing over the glass cover, Desalination 49 (1984) 231–241.

[4] G.N. Tiwari, Maduri, H.P. Garg, Effect of water flow over the glass cover of a singlebasin solar still with an intermittent flow of waste hot water in the basin, EnergyConvers. Manag. 25-3 (1985) 315–322.

[5] S.A. Lawrence, S.P. Gupta, G.N. Tiwari, Effect of heat capacity on the performance ofsolar still with water flow over the glass cover, Energy Convers. Manag. 30-3 (1990)277–285.

[6] A.K. Abu-Hijleh Bassam, Enhanced solar still performance using water film coolingof the glass cover, Desalination 107 (1996) 235–244.

[7] A.K. Bassam Abu-Hijleh, Hasan A. Mousa, Water film cooling over the glass cover ofsolar still including evaporation effects, Energy 22-1 (1997) 43–47.

[8] N.K. Dhiman, G.N. Tiwari, Effect of water flowing over the glass cover of a multiwicksolar still, Energy Convers. Manag. 30-3 (1990) 245–250.

[9] B. Janarthanan, J. Chandrasekaran, S. Kumar, Performance of a floating cumtilted-wick type of solar still with the effect of water flowing over the glass cover,Desalination 190 (2006) 51–62.

[10] Husham M. Ahmed, Khalid A. Alfaylakawi, Productivity enhancement of conven-tional solar stills using water sprinklers and cooling fan, J. Adv. Sci. Eng. Res. 2-3(2012) 168–177.

[11] O.O. Badran, Experimental study of enhancement parameters on a single slope solarstill productivity, Desalination 209 (2009) 136–143.

[12] J.R. Watt, Evaporative Air Conditioning Handbook, The Industrial Press, New York,1986.

[13] M.S. Sodha, S.P. Singh, R.L. Sawhney, Evaluation of design patterns for directevaporative coolers, Build. Environ. 30 (1995) 287–291.

[14] J.M.Wu, X. Huang, H. Zang, Theoretical analysis on heat andmass transfer in a directevaporative cooler, Appl. Therm. Eng. 29 (2009) 980–984.

[15] R.L. Sawhney, S.P. Singh, N.K. Bansal, M.S. Sodha, Optimization of an evaporativecooler for space cooling, Int. Jr. of Housing Science and Its application, 5-11, 1987,pp. 225–231.

[16] M.S. Sodha, A. Somwanshi, Variation of water temperature in direction of flow:effect in the performance of a desert cooler, Jr. of Fundamental Renewable Energyand Application, 2, 2012.

[17] NASA surface metrology, solar energy, NASA Atmospheric Science and Data Center,U.S.A. https://eosweb.larc.nasa.gov/sse/RETScreen.

[18] G.N. Tiwari, A.K. Tiwari, Solar Distillation Practice for Water Desalination Systems,Anamya publishers, New Delhi, 2007.

[19] S. kumar, G.N. Tiwari, H.N. Singh, Annual performance of active distillation system,Desalination 127 (2000) 79–88.

[20] G.D. Rai, Non Conventional Energy Sources, Khanna Publishers, New Delhi, 2005.[21] Free calculator, converters, www.easycalcultaion.com.

Nomenclature

Ab: area of basin liner of solar still (m2)Ag: area of glass cover of solar still (m2)b: breadth of glass cover of solar still (m)h: height of lower vertical side of solar still (m)cw: specific heat of water (J/kgoC)d: depth of water in the basin (m)h1g: sum of radiative and convective heat transfer coefficients between glass cover andambient (W/m2oC)h1w: sumof radiative, convective and evaporative heat transfer coefficients betweenwaterin basin and glass cover (W/m2oC)h2: sum of radiative and convective heat transfer coefficients between film and air(W/m2oC)h ' 2: convective heat transfer coefficient between glass cover and water film (W/m2oC)hb: overall bottom heat loss coefficient from basin liner to ambient through bottominsulation (W/m2oC)hcg: convective heat transfer coefficient between glass cover and ambient (W/m2oC)

hcw: convective heat transfer coefficient between water in the basin and glass cover(W/m2oC)hcfa: convective heat transfer coefficient between water film and air (W/m2oC)hrfa: radiative heat transfer coefficient between water film and air (W/m2oC)hrw: radiative heat transfer coefficient between water in the basin and glass cover(W/m2oC)hew: evaporative heat transfer coefficient between water in the basin and glass cover(W/m2oC)hrg: radiative heat transfer coefficient between glass cover and ambient (W/m2oC)hw: heat transfer coefficient between basin liner and water (W/m2oC)l: length of glass cover of solar still (m)lw: depth of water film (m)lg: thickness of glass cover (m)L: latent heat of vaporization (J/kg)Mf:mass of water film per unit length (kg)m• f : mass flow rate of water film (kg/s)Mw:mass of water in the basin (kg)m• ew: hourly yield through solar still (kg/hm2)Md: daily yield through solar still (kg/m2)Mm: monthly yield through solar still (kg/m2)Ma: annual yield through solar still (kg/m2)n: number of days in monthn ': number of days in a yearn ' ': useful life in yearsPf: saturated vapor pressure at film temperature (N/m2)Pw: saturated vapor pressure at water temperature (N/m2)Pa: saturated vapor pressure at air temperature (N/m2)q•ef : rate of evaporative heat transfer from film to air per unit area (W/m2)q•rf : rate of radiative heat transfer from film to air per unit area (W/m2)q•cf : rate of convective heat transfer from film to air per unit area (W/m2)q•cg : rate of convective heat transfer from glass cover to film per unit area (W/m2)q•ew: rate of evaporative heat transfer from water to glass per unit area (W/m2)q•rw: rate of radiative heat transfer from water to glass per unit area (W/m2)q•cw: rate of convective heat transfer from water to glass per unit area (W/m2)q•bw: rate of convective heat transfer from basin to water per unit area (W/m2)q•ba: rate of convective heat transfer from basin to air per unit area (W/m2)R: reflection coefficientS: solar insolation on horizontal surface (W/m2)Tf: temperature of water film (oC)Ta: ambient temperature (oC)Tg: temperature of glass cover (oC)Tb: temperature of basin liner (oC)Tf0: inlet temperature of water film at x = 0 (oC)Tfl: exit temperature of water film at x = l (oC)Twb: wet bulb temperature of ambient air (oC)T f : average film temperature (oC)Tw0: initial temperature of water in basin at t = 0 (oC)Tw: average basin water temperature (oC)Tg0: initial temperature of glass cover at t = 0 (oC)Tg: average glass cover temperature (oC)u: velocity of water flowing over the glass cover (m/s)v: velocity of air (m/s)Pp: pumping power (kW)E: electric energy required (kWh)

Greek letters

εw: emmisivity of waterεg: emmisivity of glassεeff: effective emmisivityσ: Stefans–Boltzman's constant (W/m2K4)ρ: density of water (kg/m3)αw: absorption coefficient of waterαg: absorption coefficient of glassαb: absorption coefficient of basin linerτ1, τ ' 1: fraction of solar energy absorbed by glass cover with and without filmτ2, τ ' 2: fraction of solar energy absorbed by water in basin with and without filmτ3, τ ' 3: fraction of solar energy absorbed by basin liner with and without filmγ: relative humidity of ambient airηa: annual efficiency of solar stillμ: refractive indexβg: attenuation factor of glass