Solar Energy 142 (2017) 278–287

Contents lists available at ScienceDirect

Solar Energy
journal homepage: www.elsevier.com/locate/solener

Analysis of heat and mass transferring mechanism of multi-stage

stacked-tray solar seawater desalination still and experimental research
on its performance
Zhili Chen a,b, Jingtang Peng a, Guanyi Chen b,⇑, Lian Hou c, Tao Yu a, Yang Yao a, Hongfei Zheng d
a
Department of Architectural Program and Environmental Engineering, Logistical Engineering University, Chongqing 401311, China
b
Enrionmental Science and Engineering School, Tianjin University, Tianjin 300072, China
c
Xi’an High Technology Research Institute, Xi’an 710025, China
d
School of Mechanical and Vehicular Engineering, Beijing Institute of Technology, Beijing 100081, China

a r t i c l e i n f o a b s t r a c t

Article history: In this paper, a multi-stage stacked-tray solar seawater desalination still was designed and used to assess
Received 5 April 2016 heat and mass transfer mechanisms and test water production performance in both transient and steady
Received in revised form 7 December 2016 states. A mathematical model of heat and mass transfer was developed and used to calculate the heat
Accepted 13 December 2016
transfer velocity equation at each stage, the heat and mass transfer equation at the highest stage, and
determine the performance coefficient of the still. The running of the still only needs the solar energy.
It is showed that the water production rate became stable after 3 h and higher temperatures resulted
Keywords:
in higher water production rates. Both the performance coefficient in steady state and performance coef-
Desalination
Multi-stage stacked-tray still
ficient were above 1 when the temperature above 70 °C. Under the practical weather, the smaller the sea-
Solar energy water depth was, the bigger the accumulative water production and performance coefficient were. The
Heat and mass transfer total production was 8.1 kg/m2
d and the performance coefficient was 1.12 when the depth of seawater
was 2 cm. The good agreement between the model predictions and experimental data shows the validity
of the model.
Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction Agrawal designed a type of suspended carrier able to float on the

Among various types of solar energy desalination stills, the
tray-type still is widely used due to its simple structure, conve-
nient construction from raw materials, and easy maintenance, par-
ticularly in remote island areas with limited energy and freshwater
resources and technical expertise. The conventional tray-type solar
energy still has some disadvantages, including no recycling of
latent heat from condensation, lower heat transfer coefficient, high
thermal capacity, and a relatively low driving force during the
evaporation. There has been extensive research into enhancing
the heat and mass transfer properties of tray-type stills.
Murugavel et al. enhanced the water production rate of tray-
type stills by utilizing core materials derived from cotton (light
cotton, jute fibre cloth, and 2 mm-thick sponge) and water-
absorptive materials (washed natural rocks and quartz sands).
Their experimental results showed that black light cotton pro-
duced the most water (Murugavel et al., 2008). Srivastava and
⇑ Corresponding author.
E-mail addresses: 15692886295@163.com, chen@tju.edu.cn (G. Chen).
surface of water and covered with a black cloth on the top, so as
to enhance the absorption of solar energy and increase the evapo-
ration area. Simulation results showed that the water production
per day for these stills in fine weather was 68% higher compared
with conventional tray-type stills (Srivastava and Agrawal, 2013).
Khalifa et al. found that the water depth played an important role
in water production from desalination stills. When the depth of
seawater decreased from 10 cm to 1 cm, the water production
was observed to increase by 48% (Khalifa and Hamood, 2009). In
another study, researchers designed a grid-fin type of absorption
tray covered with black sponge cloth to increase the evaporation
area and found that the water production per day was 48% higher
than that of a conventional tray-type still (Srivastava and Agrawal,
2013). Hiroshi Tanaka added an external reflector to a traditional
tray-type desalination still, and, compared with a traditional
desalination still, daily water production increased by 29%, 43%,
and 67% at glass cover inclination angles of 10°, 30°, and 50°
respectively (Tanaka, 2010). Velmurugan et al. used a small-scale
solar pond to evaluate the impact of salinity on water production
and internal temperature distribution and found that the water

http://dx.doi.org/10.1016/j.solener.2016.12.028
0038-092X/Ó 2016 Elsevier Ltd. All rights reserved.
 Z. Chen et al. / Solar Energy 142 (2017) 278–287 279

production rate of the still with the solar pond was 20% higher than
that without the pond (Velmurugan and SritharTiwari, 2007).
These previous studies primarily focused on single-effect tray-
type stills, which can directly release latent heat to the outside
environment after the condensation of water vapor during evapo-
ration. Since the quantity of solar radiation per unit is fixed, the
fraction of solar energy that can be used to heat the seawater is
limited. Thus, it is important to determine how to most effectively
harness solar energy for desalination. Multi-effect structures can
recycle the latent heat released during condensation to heat sea-
water, so as to harness a greater fraction of solar energy and
enhance the water production rate.
Suneja and Tiwari (2013, 1999a,b, 1998) designed single-effect,
two-effect, three-effect and multi-effect trayed desalination still
and drew from the relevant experiments a conclusion that the
water yield of the newly designed still is much higher than that
of conventional ones. The increase of effects helps enhance the
water yield, but when the number of effect is more than seven,
the increase of effects has a limited impact on the water yield.
The depth of seawater in each effect has an impact on the water
yield. For example, the relatively small amount of seawater in
the secondary effect in a two-effect still is helpful to increase water
yield. Tiwari et al. (2015), Kumar et al. (2014), Tiwari and Sahota Fig. 1. The structure and operation of a solar multi-stage stacked-tray still. 1—Solar
(2017) also studied those stills driven by different types of solar light; 2—Solar-energy heat collector; 3—Bracket; 4—Insulation layer; 5—Shell; 6—
Leveling pipes; 7—Rain water collecting trough; 8—Glass cover plate; 9—Freshwater
energy collectors and made an analysis of their corresponding per-
collecting trough; 10—A tray with a reinforced condensation surface; 11—Fresh-
formances. El-Sebaii constructed a three-effect device and devel- water collecting trough; 12—Freshwater output pipe; 13—Strong brine output pipe;
oped a water production prediction model (El-Sebaii, 2005). 14—Freshwater tank; 15—Heated water storage tank.
Schwarzer et al. designed a five-to-seven-effect desalination unit
(Schwarzer et al., 2009). Rajaseenivasan et al. designed a two-
effect dual-slope tray desalination device and studied the impact
with a reinforced condensation surface at the top of the tank were
of seawater quality on the water production rate. Their results
interconnected through leveling pipes. The freshwater collecting
showed that with the same available surface area and operating
trough at the back of each tray was used to collect freshwater. A
parameters the water production of a two-effect tray-type still
glass cover plate at the top of the tank consisted of a brim equipped
was 85% higher than that of a single-effect still (Rajaseenivasan
with a rain water collecting trough.
and Murugavel, 2013). Panchal added black granite fillers to a
two-effect tray still to reduce the thermal capacity of the seawater
and studied the impact of different sizes of fillers on the average 2.2. Still operation
water production. They found that the application of fillers of
30 mm, 20 mm and 10 mm resulted in an in-crease of 11%, 7%, The solar-energy heat collector was used to heat the raw water
and 3 in water production, respectively (Panchal, 2015). Hiroshi at the bottom of a heated water tank. The produced water vapor
Tanaka added vertical multi-effect evaporation units to the flank was condensed into freshwater at the back of the first-stage tray,
of a two-effect tray structure and reflectors to the outside of the and the heat released during the condensation was used to heat
still in order to increase the energy input (Tanaka, 2009; Tanaka, raw water at the top of the first-stage tray into water vapor, which
2015). was then condensed into freshwater at the back of the second-
The modification of tray-type stills has been shown to enhance stage tray. The heat released from this stage was used to heat
the desalination efficiency and reduce costs; however the high the raw water at the top of the second-stage tray, finally heating
thermal inertia, small condensation area, high resistance to con- the raw water at the top-stage tray. A transparent glass cover plate
vective mass transfer, and low water production rate still hinder set above the top-stage tray formed a distillation chamber similar
its applicability and effectiveness. In this study, a multi-effect to a conventional solar still. Water vapor was condensed at the
stacked-tray desalination still was designed and a theoretical back of the glass cover plate and converted into freshwater. Fresh-
model for heat and mass transfer was developed. water condensed at each stage was collected and delivered to users
through collecting troughs. After a period of operation, there may
appear fouling in the device. To solve this problem, the inflow vol-
2. The structure and operation of a solar multi-stage stacked- ume can be raised to wash the water tray at each stage and the
tray still cleaning agent can be used when necessary. According to the
actual situation, the glass cover shall be removed for thorough
2.1. Still structure cleaning every half a year to one year.

The structure and operating mechanism of a solar multi-stage

stacked-tray still are illustrated in Fig. 1. The still consisted of a
solar-energy heat collector, a heated water storage tank, water
trays with reinforced condensation surfaces, a freshwater collect-
ing trough, leveling pipes, a heated water output valve, and a fresh-
water output valve. The solar-energy heat collector was
constructed of several tubular vacuum tubes connected with the
heated water storage tank at one end. Multi-stage stacked trays
3. Mathematical model
In order to determine the heat performance of the still and how
to best improve water production, a mathematical model was
established for each component and the whole system. This model
allowed for prediction of still operating performance in different
structures and conditions.
280 Z. Chen et al. / Solar Energy 142 (2017) 278–287

3.1. Heat collecting efficiency of vacuum-tube heat collector
Vacuum-tube heat collectors may be adopted to collect solar
energy and transfer the heat into seawater and produce water
vapor. The solar radiation energy collected is I
g. The heat collect-
ing efficiency is given by Zheng et al. (2012):
Tm  Ta ðT m  T a Þ2
g ¼ 0:803  2:01  0:0034
I I
where Tm is the average temperature of the heat collectors (K); Ta is
the environmental temperature (K), and I is the solar irradiance
(W/m2).
3.2. Energy and mass transfer analysis
heat transfer, and evaporative heat transfer rates, respectively
between the sea surface on stage i  1 and the stacked-tray on stage
i, qr,i, qc,i, and qe,i mean the radiation heat transfer rate, convective
heat transfer rate, and evaporation heat transfer rate respectively
between sea surface on stage i and the stacked-tray on stage i + 1.
3.3. Heat transfer equation
ð1Þ
The heat transfer rate equation at all stages was determined
using the following steps:
(1) qr,i is the radiation heat transfer rate between seawater
surface of the stage i and the stacked-tray of the stage i + 1
(W/m2):

After solar energy accumulates in the vacuum-tube collectors, qr;i ¼ 0:9
rðT 4i  T 4iþ1 Þ ð4Þ
the heated water storage tank of the multi-stage stacked-tray solar
seawater desalination causes the water to evaporate. The process where r is the Stefan-Boltzmann constant (5.67  108W/(m2 K4)).
of the energy and mass transfer in the still is shown in Fig. 2. (2) qc,i is the convective heat transfer rate between sea surface
The following assumptions were applied: of stage i and the stacked-tray of stage i + 1(W/m2):
qc;i ¼ hc ðT i  T iþ1 Þ ð5Þ
(1) The average temperature of the heat collectors is the same as
where hc is the convective heat transfer coefficient [W/(m2 K)].
that of the first seawater stage.
(2) The temperature of the stacked-trays at each stage is the kf
same as that of the seawater stacked.
hc ¼ 0:2
Ra0:26 ð6Þ
Xi
(3) The masses of seawater stacked by stacked-trays of each
stage are the same. where kf is the thermal conductivity of vapor [W/(m2 K)], xi is the
distance between seawater surface of the stage i and the stacked-
Thus, the energy balance equation of the first stage seawater is: tray of the stage i + 1 (0.04 m), and Ra is the Rayleigh number calcu-
lated by:
dT 1
g
I ¼ m0 C p þ qr;1 þ qc;1 þ qe;1 ð2Þ gbx31 qf
ds Ra ¼ DT 0 ð7Þ
where m0 is the mass of the seawater per unit area (kg), Cp is the
lf af
specific heat at constant pressure of seawater (J/(kg K)), T1 is the where qf is the density of vapor at the average temperature of
temperature of the first stage seawater (K); qr,1, qc,1, and qe,1 are the evaporation surface (sea surface of each stage) and condensa-
the radiation heat transfer, convective heat transfer, and evapora- tion surface (the bottom surface of stacked-tray) (kg/m3); lf is
tive heat transfer rates, respectively, between the first stage seawa- the kinetic viscosity [kg/(m
s)]; af is the thermal diffusion coeffi-
ter surface and the second stage stacked-tray. cient (m2/s); g is the acceleration due to gravity (9.8 m/s2); and b
The energy balance for seawater on stage i can be obtained by: is the thermal expansion coefficient (K1). In Eq. (7), DT is deter-
dT i mined by:
qr;i1 þ qc;i1 þ qe;i1 ¼ mC p þ qr;i þ qc;i þ qe;i ði ¼ 2; 3; in 1Þ !
ds
0 Pi  Piþ1
ð3Þ DT ¼ ðT i  T iþ1 Þ þ Ma
PT
Ti ð8Þ
M a M w
 Pi
where m is the mass of the seawater per unit area on each stacked-
tray (kg); Ti is the temperature of seawater (stacked-tray) on stage i where pi and pi+1 are the vapor pressure at the sea surface (evapo-
(K); qr,i1, qc,i1 and qe,i1 are the radiation heat transfer, convective ration surface) of stage i and stacked-tray of stage i + 1 (condensa-
tion surface) respectively (Pa); Ma is the molar mass of air,
(28.97 g/mol), Mw is the molar mass of vapor (18.015 g/mol); and
PT is the atmospheric pressure (101,325 Pa).
qc, g (3) qe,i is the evaporative heat transfer rate between the sea sur-
qr , g face of stage i and the stacked-tray of stage i + 1(W/m2):
qe;i ¼ 16:273  103 hc ðPi  Piþ1 Þ ð9Þ

Parameters in the heat transfer equations are summarized in

qc qe qr Table 1.

3.4. The performance of solar still at the top stage

qe, i qc, i qr , i

qk2 The top stage solar energy distillation of this still is similar to
qe, 2 qc, 2 qr , 2 the single-effect tray-type still. Thus, the equation of the heat
and mass transfer inside the still can be obtained by a similar
qe, 1 qc, 1 qr , 1 qk1 method.
Fresh water production by the still can be expressed as the
QI amount of fresh water produced in one day per unit area. Dunkle
presented the first analysis of heat exchange in the distillation still
Fig. 2. Schematic of energy and mass transfer in the still. (Dunkle, 1961). The energy transfer process in the still is illustrated
 Z. Chen et al. / Solar Energy 142 (2017) 278–287 281

Table 1
Variation of characteristic parameters of vapor with temperature.

Parameters Symbols Expressions

Water vapor pressure Pi Exp[25.317–5144/Ti]
Specific heat at constant pressure Cp 999.2 + 0.1434  (Ti  273.15) + 1.101  104(Ti  273.15)2  6.7581  108(Ti  273.15)3
Thermal conductivity kf 0.0244 + 0.7673  104(Ti  273.15)
Thermal expansion coefficient b 1/Ti
Density qf 353.44/Ti
Kinetic viscosity lf 1.718  105 + 4.620  108(Ti  273.15)
Thermal diffusivity af kf/(qf
Cp)

in Fig. 2. The heat balance equation of top stage of the seawater and The convective heat transfer coefficient (hc ) is a measure of the
the bottom stacked-tray is as follows: natural convection between heated horizontal plates. The empiri-
cal formula of the heat transfer coefficient is obtained experimen-
dT w
aw
s
I þ qe;i þ qr;i þ qc;i ¼ qe þ qr þ qc þ qk þ C p;w
Mw
ð10Þ tally. The empirical formula given in Dunkle (1961) is adopted to
dt
calculate the convective heat transfer coefficient at the top stage.
where I is the solar radiation intensity; s is the transmissivity of " #13
glassic to the sun; aw is the absorption rate of water in the trays ðpw  pg Þ
and the bottom surface to the sun; C p;w is the heat capacity of sea- hc ¼ 8:84  104 ðT w  T g Þ þ Tw ð15Þ
268:9  103  pw
water; Mw is the mass of seawater in the most upper stage tray, and
qe ; qr ; qc are the evaporative, radiative and convective heat transfer Evaporative heat transfer and convective heat transfer occur
rates, respectively, from the surface to the cover plate; and qk is the simultaneously. Generally qe ¼ he ðpw  pg Þ. Dunkle provided a rel-
heat transfer from the still surface. This equation shows that a part atively simple calculation:
of aw
s
I was absorbed by water and bottom tray among the radi-
ðpw  pg Þ
ation sI transmitted through the top cover. In addition to energy qe ¼ 16:273  103 qc ð16Þ
transfer by evaporation, convection and radiation (qe ; qc and qr ðT w  T g Þ
respectively) to the top cover, a fraction of solar radiation (qk ) or
was absorbed by seawater and the bottom surface and lost to the
environment. The remaining energy heated the bottom surface qe ¼ 16:273  103 hc ðpw  pg Þ ð17Þ
and seawater. In Eq. (10), the water temperature can be taken as
Since salt exists in seawater, pw needs to be modified according
the bottom surface temperature as a whole. For the glassic top
to Henry’s law:
cover, the heat balance equation is given as:
pw;s ¼ pw ð1  xÞ ð18Þ
dT g
qe þ qc þ qr þ ag
I ¼ qga þ C p;g
Mg ð11Þ
dt x is the mole percentage of sodium chloride in each stage of sink.
where C p;g is the specific heat of the glassic cover plate; Mg is the Therefore, the evaporation of water can be expressed as:
total mass of the cover plate; qg;a is the heat exchange between qe
me ¼ ð19Þ
cover plate and outside atmosphere. More specifically, hfg
ð1  aw Þs
I
ag should also be included by energy absorbed by
where hfg is the latent heat of water vaporization.
top cover.
The heat loss (qk ) at the bottom of the tray and the distillation
All energy reflected from bottom tray is lost as it travels
still is difficult to determine, because the heat release coefficient
through the top cover, so the secondary reflection and secondary
varies with operating conditions.
absorption at the top cover are negligible, thus simplifying the
The coefficient can be approximately computed by the follow-
analysis. The cover heat capacity, M g , is very small compared to
ing formula (Cooper and Dunkle, 1981):
the bottom heat capacity, M w . If the variation in the stored heat
with the temperature of the top cover T g is also assumed negligi- qk ¼ hk ðT w  T a Þ ð20Þ
ble, the second on the right side of Eq. (11) may be omitted.
The value of hk should be determined using the most upper
The heat balance equation for the entire whole distillation still
stage and the operating conditions of the desalination still.
can be obtained by combining Eqs. (10) and (11).
Heat dissipation at the top cover includes both radiation and
dT w dT g convection. Radiation between the top cover and the sky is calcu-
ag I þ aw
s
I þ qe;i þ qc;i þ qr;i ¼ qga þ Mw
C p;w þ M g
C p;g
lated by:
dt dt
ð12Þ h i
qga;r ¼ eg r T 4g  T 4s ð21Þ
It is difficult to solve Eqs. (10)–(12), as there are many factors
which impact. The top cover can be approximated as a parallel where T s is the temperature of the sky (K). The convective dissipa-
plate with a small dip in the water surface. The value of the angular tion of the top cover is related to the wind speed (v ):
coefficient between them takes 1, and the areas of the two sides
hga;c ¼ 5:7 þ 3:8v ð22Þ
are equal. Dunkle simplified the equations as follows (Zheng
et al., 2012). Taken together, these relationships give the following equation:
 
qr ¼ 0:9r T 4w  T 4g ð13Þ qga ¼ qga;r þ qga;c
  4 
ew ¼ 0:96; eg ¼ 0:94 can be taken as approximate values in this case. ¼ e4g r T 4g  T 4a  12 þ ð5:7 þ 3:8v ÞðT g  T a Þ ð23Þ
The convective heat transfer rate at the bottom of the plate and
glass cover was determined by By summing all heat transfer terms, a relationship between the
water temperature of the whole system and the variation of evap-
qc ¼ hc ðT w  T g Þ ð14Þ oration heat transfer with time can be obtained so long as varia-
282 Z. Chen et al. / Solar Energy 142 (2017) 278–287
tions in solar radiation and atmospheric temperature with time are
known. These relationships can then be used to calculate the fresh
water production.
4. Experimental research on the performance of multi-stage
stacked-tray solar seawater desalination still
4.1. Experiment
The experimental set up of the multi-stage stacked-tray solar
seawater desalination still at steady state is shown in Fig. 3. The
heat in this test is produced by an electric heater which simulates
a solar collector. Each tray is made of a 0.75 mm thick stainless
steel plate, with an operating area of 0.40  0.55 m2. The outer
periphery of the still and the bottom are insulated with polystyr-
ene foam board. The bottom of each tray is folded and pressed into
a triangle wave structure, with a wave length of 2 cm. The fresh
water evaporating on the bottom tray exits through a V-shaped
stainless steel trough under the tray bottom. The average
cavity height from seawater surface to the back of the top tray is
4.5 cm.
The temperature of brine water in each tray was determined
using a WMZ-03 temperature indicator using thermocouple
signal. The electric power consumed by the electric heating appa-
ratus was determined by a voltmeter and current meter. To pro-
duce stable electrical power, a pressure regulator was used in the
circuit.
The electric heater simulated a heat source at the center of
water in the bottom tray. The temperature and distilled water pro-
duction were measured every 20 min, and the distilled water pro-
duction was determined using a 500 ml measuring cylinder. The
variation in environmental temperature and relative humidity
were relatively small (13.5–28.0 °C and 50–70%, respectively).
4.2. The steady state performance of the still
Since each stacked-tray of the still uses a multi-trough to
enhance heat and mass transfer, forming a thin seawater layer
on each stacked-tray, the total amount of seawater stored in the
still is only several kilograms and can be quickly heated. However,
the heat is transferred by the evaporation of seawater between
each tray and the heat transfer rate is limited; therefore, the sys-
tem does not easily arrive at steady state. Fresh water production
started after approximately 20 min and reached a steady state flux
after 120–150 min. The variation of the temperature of water in
each tray with time after the power was turned on is shown in
Fig. 4A (power supply of 2.2 A  220 V and stage 2). After running
for 180 min, the testing temperature was observed to remain
stable. The variation in water production rate is shown in Fig. 4B.
The water production rate reached a steady state 180 min under
the same conditions.
Glass cover plate
Freshwater collecting trough
Electric heater
220V
Fig. 3. Steady state experimental diagram of multi-stage basin solar seawater desalination.
4.3. The relationship between water production rate and steady state
operating temperature
Taking the water temperature of bottom tray as the operating
temperature of the still, the total water production per unit time
(water production rate) can be calculated for the still. The relation-
ship between the water production rate and operating tempera-
ture is shown in Fig. 4C.
The operating temperature was found to strongly influence
water production. When the operating temperature increases, the
water production rate increases rapidly due to the increased tem-
perature differences between evaporation and condensation states.
However, the system productivity is reduced when the tempera-
ture becomes too high and the temperature should be maintained
below 100 °C to maintain overall efficiency and reduce heat
dissipation.
4.4. Relationship between water production rate and heating power
The variation in water production rate of the still with heating
power is shown in Fig. 4D. The water production rate increases lin-
early when the heating power increases. However, the heating
power cannot be too big, otherwise it will cause a large amount
of water boiling or heat dissipation, thereby reducing the perfor-
mance coefficient of the still.
5. Experimental test in actual weather conditions
5.1. Experimental still
The experimental still was equipped with a vacuum tube solar
collector, as shown in Fig. 5A. Solar collector consists of 7 vacuum
tubes, and the collector area is 0.9m2. The local solar altitude angle
was considered during installation, and an optimal tilt angle of 30°
was chosen. One end of the vacuum tube was inserted into the bot-
tom water of the seawater desalination still’s heat storage tank
which was used for heating the seawater at the bottom of the still.
5.2. Condition of receiving the solar radiation energy by the still
The vacuum tube solar collector system provides thermal
energy for the desalination still, and thus, the operating conditions
of the solar collector system will directly affect the production effi-
ciency of the total system. The variation in solar radiation energy at
a 30° inclined plane with run time, as shown in Fig. 5B.
5.3. Desalination still operating temperature
An experimental test is made on the variation of temperature of
water and glassic roof in each tray with runtime when the seawa-
ter depth in each tray is 3 cm, which can be seen in Fig. 5C.
Evaporation tray
Freshwater collecting trough
condensation tray
 Z. Chen et al. / Solar Energy 142 (2017) 278–287 283

100
T1

2
90

water production rate/kg/ 20min/m

T2 0.4
working temperature/ ° C 80 T3 P=420W
70 Tg 0.3
60
50 0.2
40
30 0.1
20
10 0.0
0 20 40 60 80 100 120 140 160 180 200 220 240 40 80 120 160 200
time/min time/min
(A) Variation of temperature of water in each (B) Variation of water production rate with
tray with runtime runtime

1.2
1.2
1.1 1.1
1.0 1.0
0.9

rate/kg/h/m2
0.9
0.8
rate/kg/h/m 2

0.7 0.8
0.6 0.7
0.5
0.6
0.4
0.3 0.5
0.2 0.4
0.1
0.0 0.3
60 70 80 90 100 200 300 400 500
temperature/ ° C heating power/P

(C) Variation of water production rate with (D) Relation between water production rate
operation temperature and heating power
Fig. 4. Steady state performance of the still.

Statistical results indicate that the seawater temperature in
each tray reached a maximum value at about 13:30–15:30 and
then decreased. The trend in the temperature of water and glassic
roof in each tray was: (T1) > (T2) > (T3) > Tg. Before 10:00, the sea-
water temperature in the third tray (T3) was higher than that of
the second tray (T2); however, after 10:00, T2 was higher than T3.
These results may be attributed to the third tray receiving solar
radiation energy directly through the top glassic roof (i.e., the sea-
water of the third tray was heated by solar radiation). The energy
required for heating up the second tray is mainly derived from the
heat of condensation of the former stream, while evaporation and
condensation require more time. Heating water in the second tray
takes a longer amount of time compared with the first and third
tray, and before 10:00, the water temperature is lower than that
of the third tray. The water temperature in each tray decreased
slowly due to the insulating properties of the still, which is still
about 32 °C until 7 o’clock the next day. It is also good for the water
production of the system.
5.4. Effect of sea water depth on water production
As shown in Fig. 6A–D, the relationship between water produc-
tion and seawater depths has been evaluated.
Large differences in water production rate were observed
among different trays of the desalination still, which can be attrib-
uted to variation in heating time. The heat for producing water in
the first tray comes from condensation heat which is released from
the steam produced by heating the water in the bottom of the heat
storage tank through the solar collector condensing in the conden-
sation surface of the first tray. More energy may be obtained over
shorter periods of time, resulting in increased and earlier water
production in the first tray. The water production rate decreased
rapidly when the solar radiation was reduced. The heat to produce
water in the second tray derives from the condensation heat
released from the steam of the first tray, which is relatively small
and requires longer periods of time. This delays water production
in the second tray and reduces the production rate. The heat to
produce water in the third tray derives from solar radiation
through the top glassic roof and to a lesser extent from steam
released from the first tray. The water production rate in the third
tray was considerably higher than that of the second try. Since the
solar radiation through the top glassic roof is less than that
received from the solar collector and there is a time lag in trans-
mitting the condensation heat from the first tray to the second,
the water production rate of the first tray is higher than that of
the third one at above 16:00; However, the reduction of the solar
radiation received from the solar collector is more significant than
that of the solar radiation through the top glassic roof and the solar
radiation gradually decreased during the afternoon while the sea-
water in the third tray received the heat released from the conden-
sation of steam in the second one. Therefore, the heat received by
the seawater in the third tray was greater than that received by the
seawater in the first tray after 16:00. Thus, the water production
rate in the third tray was greater than that in the first one. The
284 Z. Chen et al. / Solar Energy 142 (2017) 278–287

(A) Photo of experimental equipment of multi-tray basin

Solar Seawater desalination
4.0
day 1 T1

3.5 day 2 100 T2

day 3 T3
90
T4
3.0
80
radiation/MJ/m /h
2

temperature/°C

2.5 70
60
2.0
50

1.5 40
30
1.0
20
0.5 10
8:30 9:30 10:30 11:30 12:30 13:30 14:30 15:30 16:30 17:30 -- 8:30 9:30 10:30 11: 30 12:30 13:30 14:30 15:30 16:30 17 :30 18:30 --

time/t time/t
(B) Variation of solar energy radiation on unit (C) Variation of temperature of water and
area with runtime glassic roof in each tray with runtime
Fig. 5. Testing results in actual weather condition.

trends in cumulative water production and the water production in (10) Gas leakage of the still is ignored and a good sealing perfor-
each tray were similar across different seawater depths. However, mance is considered.
the cumulative water production and the water production in each (11) The influence of inclination of glass cover on evaporation
tray were greater at a depth of 3 cm. The differences in the varia- area is ignored.
tion of water production among different trays were also observed. (12) The environmental temperature is assumed as follows:
 
t  14
5.5. Comparative analysis of water productions under simulation and T a ðtÞ ¼ T a þ T aR cos p: ð24Þ
12
actual conditions
In the formula, Ta(t) is the hourly environmental tempera-
In order to validate the mathematical model developed in this ture, K; T a is the daily mean environmental temperature, K;
paper, MATLAB is used for calculation to figure out the simulation TaR is the difference between the highest and lowest temper-
result and predict still’s water production, which are analyzed and ature, K.
compared with experimental results.
Assuming the following base conditions when calculating the Values taken: T a = 300 K; TaR = 9 K.
simulation:
(13) Solar irradiance is assumed as follows:
(1) The upward heat of still’s shell is ignored.  
t  ta
(2) The heat lost outside from still’s shell is ignored. IðtÞ ¼ Imax sin p ðta < t < tb Þ ð25Þ
(3) The heat lost by tube is ignored. tb  ta
(4) The heat lost by air conduction is ignored.
(5) Set the temperature for each stage the same as that of In the formula, I(t) is the hourly solar irradiance, W/m2; Imax is
seawater. the maximum value of solar irradiance, W/m2; ta,tb are time of sun
(6) The sensible heat taken away by fresh water is ignored. rise and sunset respectively.
(7) The heat stored when heating each stage is ignored. Values taken: Imax = 800 W/m2; ta = 6.5 o’clock; tb = 19 o’clock.
(8) The influence of scaling on heat and mass transfer is ignored. Simulation calculation is performed using the model by finite
(9) The influence of condensation gas on condensation is not difference method to figure out the variation of water yield for
considered. each stage with time, and the results are shown in Fig. 7A–D. As
 Z. Chen et al. / Solar Energy 142 (2017) 278–287 285

the first stage the first stage

0.40 the second stage 0.35 the second stage
the third stage the third stage
0.35 0.30

0.30
0.25
0.25
rate/kg/h/m2

0.20

rate/kg/h/m2
0.20
0.15
0.15
0.10
0.10

0.05 0.05

0.00 0.00
9:30 11:30 13:30 15:30 17:30 19:30 9:30 11:30 13:30 15:30 17:30 19:30
time/t time/t
(A) Relation of water production rate with (B) Relation of water production rate and
runtime in each tray (seawater depth of 3cm) runtime in each tray (seawater depth of 4cm)

7
the first stage 6
6 the second stage
the first stage
the third stage
5 the second stage
5 total
the third stage
4 total
4
rate/kg/m2

3
rate/kg/m

2
2

1 1

0 0

8:30 10:30 12:30 14:30 16:30 18:30 20:30 22:30 8:30 10:30 12:30 14:30 16:30 18:30 20:30 22:30
time/t time/t

(C) Relation of accumulative water production (D) Relation of accumulative water production
with runtime (seawater depth of 3cm) with runtime (seawater depth of 4cm)
Fig. 6. The impact of seawater depth on water production.

shown in the figures, the heat and mass transfer model of the still
performs better in simulation calculation of the still’s operating sit-
uation and its water yield. The total water productions are 6.97 kg/
m2 and 5.25 kg/m2 respectively when the seawater depths are
3 cm and 4 cm, which are close to the experimental results and
the error rates are only 3.25% and 3.35%. The main reasons of errors
are as follows: (1) in the model, a part of physical model is modi-
fied and parts of heat capacity and dissipating heat are ignored,
which causes that the simulation value is higher than the experi-
mental value of water yield at the initial stage of operation. (2)
The weather and environmental factors have certain influence on
experimental results. Further study will be made to solve the prob-
lems mentioned above.
5.6. The still’s performance coefficient and the efficiency of heat and
mass transfer at all stages
5.6.1. The performance coefficient of the still
The performance coefficient is an important index for measur-
ing still performance. Larger performance coefficients indicate bet-
ter overall system performance. The performance coefficient can be
calculated by the following formula:
me hfg
PR ¼ ð26Þ
Q
where me is the total fresh water production; hfg is the latent heat of
water vaporization; and Q is the total energy input of desalination
still. The total amount of fresh water production of the still (me )
at solar radiation Q at seawater depths of 2, 3, and 4 cm is tested.
The experimentally-determined performance coefficients are given
in Table 2.
These results show that seawater depth impacts both water
production and the performance coefficient of the still, with cumu-
lative water production and the coefficient increasing as depth
decreases. However very low water depths negatively impact per-
formance, as there needs to be sufficient water for evaporation to
occur during still operation. As a result, under the condition of pro-
viding enough water for evaporation, it is advantageous to improve
cumulative water production and the performance coefficient of
the still when reducing the depth of the sea as far as possible.
5.6.2. The efficiency of heat and mass transfer at all stages
According to the calculation formula of heat and mass transfer
efficiency as shown in Formula (26), the energy input at each stage
can be figured out;
286 Z. Chen et al. / Solar Energy 142 (2017) 278–287

first stage(experimental) first stage(experimental)

second stage(experimental) 0.35 second stage(experimental)
0.40 third stage(experimental)
third stage(experimental)
first stage(simulation) first stage(simulation)
second stage(simulation) 0.30 second stage(simulation)
0.35
third stage(simulation) third stage(simulation)

0.30 0.25
rate/kg/h/m2

0.25

rate/kg/h/m 2
0.20
0.20
0.15
0.15
0.10
0.10

0.05
0.05

0.00 0.00

9:30 11:30 13:30 15:30 17:30 19:30 8:30 9:30 10:3011:3012:3013:3014:3015:3016:3017:3018:3019:30

time/t time/t

(A) Comparison of experimental and (B) Comparison of experimental and simulated

simulated values of water yield at different values of water yield at different stages
stages (seawater depth of 3cm) (seawater depth of 4cm)

first stage(experimental)
first stage(experimental) second stage(experimental)
7.0 second stage(experimental) 6.5 third stage(experimental)
third stage(experimental) total(experimental)
6.5 total(experimental)
6.0 first stage(simulation)
6.0 first stage(simulation) 5.5 second stage(simulation)
second stage(simulation) third stage(simulation)
5.5 third stage(simulation) 5.0 total(simulation)
total(simulation)
5.0 4.5
4.5 4.0
rate/kg/m2

4.0
rate/kg/m2

3.5
3.5 3.0
3.0
2.5
2.5
2.0
2.0
1.5
1.5
1.0
1.0
0.5 0.5
0.0 0.0
-0.5 -0.5
8:30 10:30 12:30 14:30 16:30 18:30 20:30 22:30 8:30 10:30 12:30 14:30 16:30 18:30 20:30 22:30
time/t time/t

(C) Comparison of experimental and (D) Comparison of experimental and simulated

simulated values of accumulated water yield values of accumulated water yield
(seawater depth of 3cm) (seawater depth of 4cm)

Fig. 7. Comparison of experimental and simulated values of water yield (seawater depth of 3–4 cm).

Table 2
Performance coefficients of the device.
The depth of the sea/cm
Solar radiation energy/MJ m
Accumulative water production/kg
The still’s performance coefficient
Q 1 ¼ ISg
Q 2 ¼ qC2 þ qr2 þ qe2
Q 3 ¼ qC3 þ qr3 þ qe3 þ IS2 s
In above formulas, Q1 is the energy input at the primary stage, MJ; I
is the solar radiation volume MJ/m2; S is the vacuum-tube heat col-
2 3 4 lecting area, m2; S2 is the heat collecting are of the glass cover at the
2
19.37 19.95 19.61
top stage, m2; Q2 is the energy input at the secondary stage, MJ; qc
7.29 6.56 5.12 is the convection heat exchange, MJ; qr is the radiation heat
1.12 0.98 0.78 exchange, MJ; qe is the evaporation heat exchange, MJ. In terms of
a tray-typed still, Dunkle figures out corresponding relationship for-
mulas (see Formula (13)–(17)). According to the temperature at
each stage and the radiation density, the heat and mass transfer
ð27Þ
efficiency can be worked out (see Fig. 8).
As shown in Fig. 8, the efficiency of heat and mass transfer at
ð28Þ
each stage is different from each other. The efficiency at the sec-
ondary stage is the highest while the heat and mass transfer effi-
ð29Þ
ciency at the primary stage is lowest and the efficiencies at the
 Z. Chen et al. / Solar Energy 142 (2017) 278–287
1.6
1.4 the first efficiency
the second efficiency
1.2 the third efficiency
1.0
efficiency
0.8
0.6
0.4
0.2
0.0
-0.2
8:30 9:30 10:30 11:30 12:30 13:30 14:30 15:30 16:30 17:30
time
Fig. 8. The heat and mass transfer efficiency at each stage.
secondary and third stage increase as the time goes on. The effi-
ciency at the secondary effect saw a fast increase since 9:30 and
reached as high as 1.55 at 17:30.
The reason why the efficiency at the primary stage is the lowest
is that its energy input is totally from solar radiation, that is direct
energy used for increasing seawater temperature while for the sec-
ondary and third stage, the latent heat can be used so that only a
small proportion of energy input is needed to increase the seawa-
ter temperature. The proportions at the secondary and third stage
are different from each other. Starting from 9:30, the efficiency at
the third stage is lower than that at the secondary stage because a
part of energy input is from sunshine. As the density of sunshine
decreases, the heat and mass transfer efficiencies at the secondary
and third stage increase due to the use of latent heat. Thus, the full
use of latent can effectively improve the still’s performance.
6. Conclusions
(1) A multi-stage stacked-tray solar seawater desalination still
is designed and its running only needs the solar energy. A
mathematical model of the heat and mass transferring pro-
cess was developed for each component and for the whole
system to determine heat transfer velocity equation at each
stage, the heat and mass transfer equation at the highest
stage, and the performance coefficient.
(2) The results of transient and steady state experimental test
showed that the water production rate became stable 3 h
later, that the higher the temperature was, the bigger the
water production rate and the performance coefficient in
steady state were, and that the performance coefficient
was above 1 when the temperature was above 70 °C.
(3) The experiment in actual weather conditions indicates that
when the amount of solar radiation is 19–20 MJ/(D - m2),
the water depth is 2 cm, 4 cm, and 3 cm, the total
output of fresh water is 8.1 kg/d/m2, 5.69 kg/d/m2, and
7.29 kg/d/m2, and the performance coefficient is 1.12, 0.78,
and 0.98, respectively. The heat and mass transfer efficiency
287
at the primary stage is lowest while the efficiency at the sec-
ondary stage is the highest at 1.55 at 17:30. The smaller the
seawater depth is, the bigger the accumulative water pro-
duction and performance coefficient are.
Acknowledgment
This work is supported by the National Natural Science Founda-
tion of China (No. 51106177) and the Basic and Frontier Research
Project of Chongqing (cstc2015jcyjBX0059).
References
Cooper, P.I., Dunkle, R.V., 1981. A non-linear flat-plate collector model. Sol. Energy
26 (2), 133–140.
Dunkle, R.V., 1961. Solar water distillation: the roof type still and a multiple effect
diffusion still. International Development in Heat Transfer, Part V. University of
Colorado. 895-902.
El-Sebaii, A.A., 2005. Thermal performance of a triple-basin solar still. Desalination
174, 23–37.
Khalifa, A.J.N., Hamood, A.M., 2009. On the verification of the effect of water depth
on the performance of basin type solar stills. Sol. Energy 83 (8), 1312–1321.
Kumar, Shiv, Dubey, Aseem, Tiwari, G.N., 2014. A solar still augmented with an
evacuated tube collector in forced mode. Desalination 347, 15–24.
Murugavel, K. Kalidasa, Chockalingam, Kn.K.S.K., Srithar, K., 2008. An experimental
study on single basin double slope simulation solar still with thin layer of water
in the basin. Desalination 220, 687–693.
Panchal, Hitesh N., 2015. Enhancement of distillate output of double basin solar still
with vacuum tubes. J. King Saud Univ.-Eng. Sci. 27, 170–175.
Rajaseenivasan, T., Murugavel, K. Kalidasa, 2013. Theoretical and experimental
investigation on double basin double slope solar still. Desalination 319, 25–32.
Schwarzer, K., Vieira da Silva, E., Hoffschmidt, B., Schwarzer, T., 2009. A new solar
desalination system with heat recovery for decentralised drinking water
production. Desalination 248, 204–211.
Srivastava, Pankaj K., Agrawal, S.K., 2013. Experimental and theoretical analysis of
single sloped basin still consisting of multiple thermal inertia floating porous
absorbers. Desalination 311, 198–205.
Srivastava, Pankaj K., Agrawal, S.K., 2013. Winter and summer performance of
single sloped basin type solar still integrated with extended porous fins.
Desalination 319, 73–78.
Suneja, Sangeeta, Tiwari, G.N., 1998. Optimization of number effects for higher yield
from an inverted absorber solar still using the Runge-Kutta method.
Desalination 120, 197–209.
Suneja, Sangeeta, Tiwari, G.N., 1999a. Parametric study of an inverted absorber
triple effect solar still. Energy Convers. Manage. 40, 1871–1884.
Suneja, Sangeeta, Tiwari, G.N., 1999b. Effect of water depth on the performance of
an inverted absorber double basin solar still. Energy Convers. Manage. 40,
1885–1897.
Suneja, Sangeeta, Tiwari, G.N., 2013. Effect of water depth on the performance of an
inverted absorber double basin solar still. Desalination 311, 198–205.
Tanaka, Hiroshi, 2009. Experimental study of vertical multiple-effect diffusion solar
still coupled with a flat plate reflector. Desalination 249, 34–40.
Tanaka, H., 2010. Monthly optimum inclination of glass cover and external reflector
of a basin type solar still with internal and external reflector. Sol. Energy 84
(11), 1959–1966.
Tanaka, Hiroshi, 2015. Theoretical analysis of solar thermal collector and flat plate
bottom reflector with a gap between them. Energy Rep. 1, 80–88.
Tiwari, G.N., Sahota, Lovedeep, 2017. Review on the energy and economic
efficiencies of passive and active solar distillation systems. Desalination 401,
151–179.
Tiwari, G.N., Yadav, J.K., Singh, D.B., et al., 2015. Exergoeconomic and
enviroeconomic analyses of partially covered photovoltaic flat plate collector
active solar distillation system. Desalination 367, 186–196.
Velmurugan, V., SritharTiwari, K., 2007. Solar stills integrated with a mini solar
pond-analytical simulation and experimental validation. Desalination 216,
232–241.
Zheng, Hongfei, Zheng, Zihang, Wang, Haijiang, Chen, Zhili, 2012. Experimental
study of tubular solar desalination unit. Trans. Beijing Inst. Technol. 32, 1086–
1090.