Desalination 341 (2014) 50–60

Contents lists available at ScienceDirect

Desalination
journal homepage: www.elsevier.com/locate/desal

A hybrid solar desalination system of air humidification,

dehumidification and water flashing evaporation: Part II.
Experimental investigation
A.E. Kabeel ⁎, Emad M.S. El-Said
Mechanical Power Engineering Department, Faculty of Engineering, Tanta University, Tanta, Egypt

H I G H L I G H T S

• An experimental investigation of a solar hybrid desalination system was presented.

• The maximum productivity of the system reached 41.8 kg/day, according to operating conditions.
• The performance ratio of SSF unit is varied between 0.32 and 1.4.

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

Article history: This paper describes laboratory experiments with a hybrid solar desalination system consisting of a
Received 10 November 2013 humidification–dehumidification unit and single stage flashing evaporation unit configured by a (Al2O3/H2O)
Received in revised form 17 February 2014 nano-fluid solar water heater under the climatological conditions of Tanta City, Egypt. The laboratory work has
Accepted 25 February 2014
been carried out in actual thermal environment in August 2013. The validation has an emphasis on the main pa-
Available online 17 March 2014
rameters with impact on the water production capacity. These are the solar radiation, feed water mass flow rate,
Keywords:
inlet cooling water temperature, mass flow rate of cooling water and nano-particle volume fraction. The main
Humidification–dehumidification conclusion is that good agreement is obtained between simulated and measured variations of water production
Flashing desalination and performance ratio values for variations of these parameters. The performance ratio of SSF unit is varied be-
Hybrid tween 0.32 and 1.4 and flashing ranges between 3 and 9 °C. The maximum productivity of the system reached
Nano-fluid 41.8 kg/day according to test and operation conditions. The solar water heater efficiency is affected by the
nano-particle volume fraction. The humidifier efficiency is affected by increasing water mass flow rate more
than the increasing air mass flow rate and reaches about 98%. Finally, collector's efficiencies are about 55% and
56% for solar water heater and solar air heater respectively.
© 2014 Elsevier B.V. All rights reserved.

1. Introduction
Water is essential for the survival of all living things. It is essential for
agricultural and industrial growth, as well as for supporting growing pop-
ulations who require a safe drinking water supply. In fact, 97% of all water
in our planet exists in seas and oceans, and 2% in glaciers and ice caps. The
rest exists in lakes, rivers and underground reservoirs. Natural resources
cannot satisfy the growing demand for low-salinity water with industrial
development, together with the increasing worldwide demand for sup-
plies of safe drinking water. This has forced mankind to search for other
sources of water. In addition, the rapid reduction of subterranean aquifers
and the increasing salinity of these nonrenewable sources will continue to
exacerbate the international water shortage problems in many areas of
⁎ Corresponding author. Tel.: +20 1001543587.
E-mail addresses: kabeel6@hotmail.com (A.E. Kabeel), emad_mech@hotmail.com
(E.M.S. El-Said).
the world. Desalination has already become an acceptable solution for
shortages in conventional water resources. This is now acknowledged
by reputable institutions such as the World Bank. Seawater desalination
is being applied at 58% of installed capacity worldwide, followed by brack-
ish water desalination accounting for 23% of installed capacity. Due to the
limited fossil fuel resources, it is expected that their price will continue to
rise dramatically in the future. On the other hand climate change obliges
humanity to react accordingly. Renewable energy is the favorable alterna-
tive to fossil fuels. Solar energy can be used for saline water desalination
by producing the thermal energy required to drive the phase-change pro-
cesses by using direct solar collection systems. Garg et al. [1] presented
an experimental design and computer simulation of multi-effect humidi-
fication–dehumidification solar desalination and the developed model
which is useful in the estimation of the distillation plant output and opti-
mized various components of the system like, solar water heater, humid-
ification chamber, and condensation chamber. Dai and Zhang [2]
experimentally conducted a solar desalination unit with humidification

http://dx.doi.org/10.1016/j.desal.2014.02.035
0011-9164/© 2014 Elsevier B.V. All rights reserved.
 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60
and dehumidification. The performance of the system was strongly de-
pendent on the temperature of inlet salt water to the humidifier, the
mass flow rate of salt water, and the mass flow rate of the process air.
The optimum rotation speed for the fan corresponds to an optimum
mass flow rate of air with respect to both thermal efficiency and water
production. The unit worked perfectly and the thermal efficiency was
above 80%. Amer et al. [3] theoretically and experimentally investigated
humidification–dehumidification desalination system. The system is
based on an open cycle for water and a closed cycle for air stream. The
air is circulated by either natural or forced circulation. Detailed experi-
ments have been carried out at various operating conditions and using
several packing materials. The heat and mass transfer coefficients have
been experimentally obtained and fitted in forms of empirical correla-
tions. The system productivity increases with the increase in the mass
flow rate of water through the unit. Water temperature at condenser
exit increases linearly with water temperature at humidifier inlet and it
decreases as water flow rate increases. A maximum productivity of
5.8 l/h has been obtained using wooden slate packing and with forced
air circulation. Nafey et al. [4,5] numerically and experimentally investi-
gated a humidification–dehumidification desalination process using
solar energy under different environmental and operating conditions.
The comparison between theoretical and experimental results illustrated
that the mathematical model is in good agreement with the experimental
results. The productivity of the unit is strongly influenced by the air flow
rate, cooling water flow rate and total solar energy incident through the
day. The obtained results indicate that the solar water collector area
strongly affects the system productivity, more so than the solar air collec-
tor area. Ben Amara et al. [6] experimentally and theoretically studied a
pad humidifier used in a multi-stage solar desalination process in opera-
tion. The experimental operation of the humidifier is presented in terms
of temperature, relative and absolute humidity, and the quantity of evap-
orated water for different climatic and working conditions. Dai et al. [7]
mathematically and experimentally characterized that a solar desalina-
tion unit with humidification and dehumidification by reusing some of
somewhat concentrated saline water after evaporation, recovering con-
densation heat, and forced air flow was expected to produce more fresh-
water. Parametric analysis was conducted in order to optimize the unit
performance and to study the effect of some of the operating conditions
such as flow rates, temperatures of feed water, and air and cooling
water. The daily solar productivity corresponding to unit square meter
of collector area is about 6 kg/day/m2 with 20 MJ solar energy input a
day under given conditions. El-Shazly et al. [8] took another way with hu-
midification–dehumidification desalination method to enhance mass and
heat transfer rates, and improve both process productivity and product
quality by using pulsating liquid flow. An experimental investigation
was performed in humidification–dehumidification desalination unit
consisting of the main components (humidifier, dehumidifier, and solar
water heater). The results showed that the unit productivity has been in-
creased by increasing the off time i.e. decreasing the frequency of pulsed
water flow up to certain levels, a frequency of 20/60 on/off time was
found to have the highest productivity of the unit. Increasing the ampli-
tude of water pulsation (water flow per pulse) was found to increase
the unit productivity as well. Nafey et al. [9] theoretically and experimen-
tally investigated a small unit for water desalination by solar energy and a
flash evaporation process. The system consists of a solar water heater (flat
plate solar collector) working as a brine heater and a vertical flash unit
that is attached with a condenser/pre-heater unit. The average accumula-
tive productivity of the system in November, December and January
ranged between 1.04 and 1.45 kg/day/m2. The average summer produc-
tivity ranged between 5.44 and 7 kg/day/m2 in July and August and 4.2
to 5 kg/day/m2 in June. Baig et al. [10] investigated the effect of various
operating conditions on the performance ratio, brine temperature and sa-
linity as it leaves the last flash stage in a once-through the multi-stage
flash (MSF) distillation system. The numerical results obtained with the
published data on similar plants were compared. The results show that
both analytical solutions and experimental/field analysis are required to
51
identify the most influential parameters that affect the performance and
set proper plans for performance optimization. The accurate estimate of
the variables related to the brine heater, selecting proper number of
stages and the stage-to-stage temperature drop is of crucial importance.
The thermal properties dependent on the operating conditions may affect
the accuracy of numerical results. The salinity of the feed seawater has a
significant effect on the plant characteristics. Junjie et al. [11] experimen-
tally studied the heat and mass transfer properties of static/circulatory
flash evaporation, i.e., non-equilibrium fraction (NEF), evaporated mass
and heat transfer coefficient. The heat transfer coefficient was redefined
as average heat flux released from unit volume of water film under unit
superheat. Results suggested that this coefficient was a time-dependent
function and a peak value existed in its evolution versus time. Abutayeh
[12] experimentally and theoretically simulated a new desalination pro-
cess based on solar flash desalination process under a hydrostatically
sustained vacuum and analyzed the flash temperature effect on system
variables. The system consists of a saline water tank, a concentrated
brine tank, and a freshwater tank placed on ground level plus an evapora-
tor and a condenser located several meters above the ground. The vacu-
um is passively created and then maintained by the hydrostatic balance
between pressure inside the elevated flash chamber and outdoor atmo-
spheric pressure. The proposed desalination unit does not require high
temperatures to perform the flash operation due to the vacuumed
flash chamber. In addition, the vacuum is naturally produced by the hy-
drostatic forces without the need for vacuum pumps, which extends the
efficiency of the unit. Saad et al. [13] proposed and designed a new de-
salination system for converting seawater into freshwater by utilizing
the waste heat of internal combustion engines. The desalination process
is based on the evaporation of seawater under a very low pressure (vac-
uum). The low pressure is achieved by using the suction side of a com-
pressor rather than a commonly used vacuum pump. The evaporated
water is then condensed to obtain freshwater. The effects of operational
variables such as evaporator temperature, condenser temperature, vac-
uum pressure, and flow rate of both evaporator and condenser on the
yield of freshwater are experimentally investigated. It is found that de-
creasing the vacuum pressure causes a significant increase in the yield
of freshwater. It is also found that decreasing the condenser tempera-
ture, or increasing the evaporator temperature both lead to an increase
in the yield of freshwater. Moreover, increasing the condenser flow rate
tends to increase the yield of freshwater. The same trend is attained by
increasing the evaporator flow rate.
The solar collector is a convenient and common heater to be used as
heat source for many applications such as domestic water heater and
desalination purposes. However, the effectiveness of the present solar
collector for low-capacity desalination units is low due to some reasons
such as limited thermal conductivity of this working fluid and ineffi-
ciency and cost of solar radiation concentrators. Several years ago,
nano-fluid has been found to be an attractive heat transport fluid. It
has exhibited a significant potential for heat transfer augmentation rel-
ative to the conventional pure fluids. It has been expected to be suitable
for the solar water heating systems without severe problems in pipes
and with little or no penalty in pressure drop [14]. Yousefi et al. [15] ex-
perimentally investigated the effect of Al2O3/water nano-fluid, as work-
ing fluid, on the efficiency of a flat-plate solar collector. The weight
fraction of nano-particles was 0.2% and 0.4%, and the particle dimension
was 15 nm. The mass flow rate of nano-fluid varied from 1 to 3 l/min.
The results showed that in comparison with water, absorption medium
using the nano-fluids as working fluid increases the efficiency. For
0.2 wt.%, the increased efficiency was 28.3%.
The current work aims to experimentally study a pilot small-scale
hybrid air humidification and dehumidification–water flashing evapo-
ration (HDH–SSF) desalination system driven by solar thermal energy.
The proposed system was designed, manufactured and tested to insure;
- studying the influence and performance of the different system op-
erating conditions under actual climatological conditions,
52 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60

- performing a comparison between the experimental results and the-

oretical results computed from previous work. Flash chamber Humidifier Dehumidifier
- studying the effect of nano-fluid as a working fluid for solar water
loop on the enhancement of system production. Mixing Tank

Product tank
- studying the possible factors and parameters that have an effect on
the system production. Helical heat exchanger

2. Experimental setup and instrumentation

2.1. Experimental setup description

Fig. 1 illustrates a schematic diagram of the experimental setup. Fig. 2

shows a photograph of the experimental setup. The present system has
been mainly conceived on two ideas of air humidification and dehumidi-
fication processes and flash evaporation of saline water. The main system
components are the humidifier, dehumidifier (water cooled exchanger), Solar water heater
air heater (flat plate collector), water heater (flat plate collector) and
flashing evaporation unit. The HDH system consists of two loops, one Solar air heater Air blower
for water heating and the other for air heating with an interface of the hu-
midifying cooling tower. In the water loop, the water is delivered from the
mixing tank by pump (P4) and split into two main lines, a feeding line and Fig. 2. HDH–SSF experimental set-up photo.
a bypass line which is controlled by valve (V15). The heated saline water
is sprayed at the top of the humidifier. The water falls down to the bottom using the sub-atmospheric pressure. The extracted water vapor on flash-
of the humidifier where it is pumped to mixing tank to reheat. In the air ing chamber is flowed to the condenser. The saline cooled water is fed to
loop, air is drawn from the atmosphere using a centrifugal blower the flashing unit condenser to condense the water vapor and exits. The
(installed on inlet of the humidifier) through the air solar heater. The desalinated water in a condenser is collected from the bottom tray and
hot air flows through the humidifier and carries the evaporated water flowed to a product tank. The flashing evaporation depends on pressure
to the dehumidifier, where the air is cooled and dehumidified. The SSF reduction. So, the condenser and flashing chamber are vacuumed by
system consists of flashing chamber and condenser. A part of saline using a vacuum pump.
water in mixing tank (MT) is flowed to helical heat exchanger (HHEx) Then, the saline water exits from a flashing unit condenser and is
to backup water inside the closed loop of saline water flow between the mixed with the rejected brine water from the humidifier. The saline
flashing chamber and heat exchanger (HHEx) through a valve (SV1) water that exits from the dehumidifier is drained because its temperature
while the rest is drained using a valve (V5). The hot saline water from is low. The detailed descriptions of the unit main components are as
heat exchanger is pumped to flashing chamber to evaporate by flashing follows.

Drain MT To the atmosphere

V5
T10
T9

Drier SV2
T8 2
Vacuum pressure H2 T13
gauge V13 V19
PS P3 H3
T14
H
V14
SV1 C1
F4
C2
P4
FC V11
S1 T2
SWH

V4 LI
T12
T4 F3
V8
V18
TK1
T1 T3 F6
Drainage
F1 P5
V3
T5
T6
V1 V2 V10
EH HHEx V9 F5
P1 P2

V17
T7 T11
SAH
H1 F2
V6
V7 P6
Blower
Bleeding air V12

F Flow meter SWH Solar Water Heater LI V16

T Thermocouple SAH Solar Air Heater
S Pyrometer HHEX Helical Heat Exchanger
H Thermo hygrometer H Humidifier
V Valve FC Flashing Chamber
P Pump C1 Condenser V15
LI Level indicator C2 Dehumidifier LI
MT Mixing Tank Sea water
TK Tank Air Tk2 Tk3
EH Electrical Heater Desalinated water
SV Solenoid Valve

Fig. 1. HDH–SSF experimental set-up schematic diagram.

 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60
2.1.1. Packed-bed humidifier
An induced fan cooling tower is used as humidifier with direct air–
water contact. The tower shell has a circular cross section with inner di-
ameter of 45 cm, made of 0.7 mm thick steel sheet with a total height of
1.96 m and thermally insulated. Water is distributed over the packing sur-
face at the top of the humidifier by 4 aluminum swirl injectors. The pad
humidifier unit consists of reaching rings, made of PVC material and it
forms the wetted surface of the humidifier. The size of each ring is 5 cm
in diameter and 5 cm in length. They were mounted above a metal grid
with a diameter of 45 cm to reach a height of 0.8 m. This grid made of gal-
vanized steel of 10 mm thickness was constructed by welding. The humid
air that exits the humidifier top carries with it entrained water droplets
due to water splashing. These droplets are separated from the humid air
by disengaging surfaces called droplet eliminator before reaching the out-
let. At the top of the packing bed, there is a water sprayer, which sprays
water to the packing bed, while at the bottom there is a water storage
sump, where the water is collected as it drains down the packing bed.
Thus, the water flows downward, while the air passes in a cross flow di-
rection through the openings of the packing bed. The air is humidified as it
comes in contact with the wetted surface of the bed.
2.1.2. Dehumidifier
In this study counter flow water-cooled shell and multi-pass tube heat
exchanger are used as dehumidifier to condense the water vapor in
humid air from the humidifier. The shell is the dehumidifier's outermost
body and contains the heat exchanger tubes. The shell has a cubic cross
section with dimensions of 400 × 400 mm, made of 0.7 mm thick steel
sheet with a total height of 90 cm. The cooling water tubes are supported
Double glass cover
Corrugated steel sheet absorber
A Wooden frame
Air outlet
= 60°
A
Air inlet
Plywood back base
30mm Fiber glass insulation
Aluminum flat sheet reflector
Rubber seal
1015×100 mm slots
12 mm
63 mm
153 mm
133 mm
80 mm
Air inlet
Section A-A
Fig. 3. The structure of the air heater.
53
within the dehumidifier by the tube support sheets to serve as baffle
plates that provide the desired flow path of the condensing vapor to min-
imize direct impingement on the cooling water tubes. The baffle plates
also provide support that help prevent sagging of the tubes. At the end
of the dehumidification unit air has to pass through a metal screen, with
hole diameter of 1 mm and hole spacing of 2 mm, arranged in-line. This
screen is used for increasing the amount of water condensed from exit
air by obstructing the humid air that passes. At the bottom of the shell,
where the condensate collects, an outlet is installed.
2.1.3. Flat-plate solar air heater
Solar air heaters are cheap and most widely used collection devices be-
cause of their inherent simplicity. So, a double pass corrugated absorber
solar air collector is used in this study as a device to heat the air by
collecting and absorbing the solar radiation. In the present work the
double-pass flat plate solar air heater was designed, and manufactured
in order to use as air heater as shown in Fig. 3. The heater is 126 cm
wide, 126 cm long and 18.6 cm high. It consists of two glass covers having
a thickness of 3 mm, insulated container and matt black painted steel V-
groove absorber plate. The distance between the glass covers is 10 mm,
the insulation thickness is 5 cm and the thermal conductivity of the insu-
lation material is 0.045 W/mK. The frame of the solar air heater that was
made of wood sheets of 5 cm thickness was painted with a matt black dye
with a piece of 10 mm plywood to the back of the frame. Aluminum flash-
ing is used as the heat reflection plate. A matt black painted steel absorber
plate (394 cm × 100 cm) with 0.75 mm thickness was placed horizontally
on the centerline of the container. The collector is supported on steel
frame to support and control the solar collector angle. The steel frame
can be tilted at an angle from 0° to 90°.
2.1.4. Flat-plate solar water heater
Water is heated by means of flat plate solar collector as schematically
shown in Fig. 3. The collector is supported on steel frame to support and
control the solar collector angle. The steel frame can be tilted at an
angle from 0° to 90°. The change atilt angle mechanism is achieved man-
ually. The flat plate solar water heater is constructed from many parts as
follows:
• Absorber plate; 0.5 mm thickness steel sheets are used in manufactur-
ing absorber plates (fins). Solar glazing cover; single pane 3 mm solar
glass is patterned to reduce reflection and tempered to maximize
strength and durability with high solar transmittance.
• Casings; all casings are made of wood construction with 50 mm thick-
ness except back base is made of plywood with 8 mm thickness. Tub-
ing grid; considering commercial availability, alloy 122 copper tubing
with an inside diameter of 11.4 mm and an outer diameter of 12.7 mm
was used for nine risers and inside diameter of 20.6 mm and an outer
diameter of 22.2 mm was used for headers with optimal flow distribu-
tion. Back plate; the back plate is made of aluminum sheet with dis-
tinctive antirust performance, toughness and reliability. Aluminum
foil; the aluminum foil acts as a barrier against out-gassing, attached
to the insulation.
According to the above selection characteristics, Table 1 summarized
the specifications of the tested nano-fluid as-received and some repre-
sentative data. The nano-particle material that is used for the investiga-
tion of the present experimental work is procured from a China based
company (Qinhuangdao Taiji Ring Nano-Products Ltd.).
2.1.5. Flash evaporation unit
Flashing chamber construction; the flashing chamber shell should be
strong enough to withstand the operating pressures (lower than the at-
mosphere pressure) and should have the following aspects:
• enough vapor disengagement surface area for the brine to flash with a
reasonable vapor velocity to avoid brine droplets carry over with
vapor.
54 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60

Table 1 The tubes are arranged in five groups with inlet and outlet manifolds con-
Specifications of tested nano-fluid. nected to the main inlet and outlet manifolds.
Parameters Value/type

Nano-particle material 2.1.6. Helical coil heat exchanger description

Nano-particle material Al2O3 A helical coil heat exchanger consists of a helical coil fabricated out of
Product model (according AL-01 a copper tube that is fitted in the annular portion of two concentric cyl-
to data sheet from supplier)
Purity of nano-particle material 99.9%
inders with diameters of 325 mm and 275 mm respectively. The two
Appearance White cylinders are constructed with 1.5 mm thick steel sheet and coated to
Nano-particle size (diameter), 30 × 10−9 prevent outside and inside corrosion and thermally insulated. The fluids
dnp (m) flow inside the coil and the annulus, with heat transfer taking place
Specific surface area (m2/g) 160
across the coil wall. The dimensions of both cylinders are determined
Volume density (g/cm3) 0.916
Density, ρnp (kg/m3) 3910 by the velocity of the fluid in the annulus needed to meet heat transfer
Specific heat, Cp,np (kJ/kg) 880 requirements. Fig. 5 shows schematic cutaway view of the helical coil
Thermal conductivity, Knp (W/mK) 36 heat exchanger.
Crystal form γ
Characteristics Insoluble
Particle shape Sphere 2.1.7. Tanks and reservoirs
There are four tanks used in the test-rig with design and construc-
Nano-fluid
Base fluid Distilled water tion details as shown in Table 2. The reservoir tank (TK1) has one elec-
Dispersion condition Agitation tric heater, with a capacity of 3 kW. The heater was horizontally placed
Dispersant agent Sodium Dodecyl Benzene at the bottom of the tank. These heaters were used to heat the nano-
Sulfonate (SDBS) 1.0 wt.% fluid inside the tank to a required temperature before starting the ex-
periments and to investigate the effect of the initial feed water temper-
ature on the system productivity. Freshwater product tank has been
• enough height between the flashing brine surface and demister to divided into two halves, one for HDH unit product and the other for
separate brine droplets entrained with vapor by gravity before SSF unit.
reaching the demisters,
• enough area to house the demisters of adequate area to keep vapor 2.1.8. Instrumentation
velocity through the demister. The measured variables in the experiment include inlet and outlet
air and water temperatures in each unit component, inlet and outlet rel-
The flash chamber is a 0.05 m3 cubical tank with cross section of 0.2 m ative humidity in the humidifier, ambient temperature, air velocity and
width, 0.5 m length and 0.5 m height. The flash chamber is constructed solar irradiation incident on the field collector's plane. The air and water
with 2 mm thick steel sheet and coated to prevent outside environmental temperatures of all points (except T6, T10 and T12) were measured
corrosion effects and to reduce thermal losses to ambient temperature using thermocouple K type. The outlet of both relative humidity and
and thermally insulated. The brine depth is controlled to be not exceeding air temperatures in the humidifier, solar air heater inlet and dehumidi-
about 0.1 m in the flash chamber and should be kept as low as possible to fier outlet was measured using a digital temperature thermometer hu-
avoid high non-equilibrium losses, brine entrainment, and unit flooding. midity hygrometer, HTC-2. The air flow rate was calculated from the
The hot feed water enters the flash chamber through the feed water air velocity, measured by a hand held vane anemometer, S-AM81 at
pipe. Fig. 4 shows a schematic of the flashing chamber with attached con- the solar air heater inlet. A pyranometer was used to measure the
denser unit. The flash unit is a condenser cubical tank with cross section of total solar irradiation placed in a horizontal plane adjacent to the collec-
0.1 m width, 0.2 m length and 0.6 m height. The flash chamber is con- tor. All sensors were calibrated before using to determine probe
structed with 2 mm thick coated steel. The copper alloy tube is used.

Saline water Nano-fluid inlet

Vapor to condenser
Demister

Outer shell

Inner shell
Nanofluid tube coil

Brine splash 6 cm
plate
Inner support
Saline water 4 cm 55mm
inlet
5 cm

Brine outlet Nano-fluid outlet Saline water inlet

Fig. 4. Flashing evaporation unit schematic diagram. Fig. 5. Helical coil heat exchanger schematic diagram.
 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60 55

Table 2 this matter happens the vacuum inside the flashing chamber will be
Tank technical specification data used for the simulation. lost.
Technical specifications and operation data Value
2.2. System operation
Water mixing tank (MT) and cooling water tank (TK3)
Size, m 0.35 ID × 0.5
Material Steel sheets The detailed operation procedures of the experimental system ac-
Thickness, mm 1.5 cording to Fig. 1 can be explained in some steps: (1) fill the system sa-
Reservoir tank (TK1) line water loops until it reaches the operating levels; (2) set and
Size, m 0.15 × 0.15 × 0.25 adjust the test case parameters (temperatures, flow rates); (3) operate
Material Steel sheets the solar water heater loop by running pump (P1) and turn-on electrical
Thickness, mm 3
heater (EH) until water temperature in tank (TK1) reaches the required
Freshwater product tank (TK2) value; (4) set and adjust the open and close times of solenoid valve
Size, m 0.3 ID × 0.8 (SV1) according to flow rates of back up water; (5) run the vacuum
Material Steel sheets
pump (P3) until the pressure inside the flashing chamber reaches the
Thickness, mm 1.5
required value; (6) run the feeding pump (P2) to circulate the saline
water inside flashing chamber; (7) adjust the brine pool height using
sensibility. The water salinity was measured in the reef tank using a hy- level indicator and water bleeding through valve (V4) and control
drometer. A graduation pressure gage was used to measure the vacuum valve (V8); (8) run the air blower; (9) run the feeding pump (P4) to
pressure inside the flashing chamber. A floating flow meter (rotameter) feed the saline water to humidifier sprayers; (10) run the circulation
was used to measure the flow rate of water in the individual tube. Flow pump (P5) to pump the saline water mixing tank; and (11) run the
rate of product freshwater was carried out by knowing the water height cooling water pump (P6) to pump the saline water to condenser (C1)
using level indicator and stop watch. The detailed technical specifica- and dehumidifier (C2).
tions of sensors and probes used in experimental setup are presented The main steps of nano-fluid preparation and production are carried
in Table 3. out as follows:

1. purchasing nano-particle powders.

2.1.9. Control and safety devices
2. mixing the required volume of the powder (Al2O3) in the chemical
Three different sizes and two different types for manual valves were
measuring flask with base fluid (distilled water). Since the purity of
used to control the air flow from humidifier outlet to the dehumidifier
the nano-fluid is important, attempts were made to mix the particles
inlet and air flow from the air blower outlet to the humidifier inlet.
directly with water with no additives and then using high speed me-
Two way brass solenoid valve is used to control the feeding water
chanical stirrer to disperse it. This solution is allowed to fully react at
flow from mixing tank to helical heat exchanger and to prevent the
80 °C for 30 min. It may be mentioned here that the metallic oxide
back flow from vacuum pump to SSF condenser when vacuum pump
nano-powders like Al2O3 are not chemically reactive in atmosphere;
shuts down. The controlling and adjusting of flow rates will performed
however they do tend to form loose micro-sized agglomerates in at-
by using bleeding return method for solar air heater loop, solar water
mosphere and in fluid suspension over time.
heater loop, feeding of flashing chamber, feeding of humidifier and
3. after making the proper mixture the flask was kept under ultrasonic
cooling water (for C1 and C2) as shown in Fig. 1. An air drier filter was
vibration for 12 h to break and evenly disperse the nano-particle
used to dry and filter the suctioned air by vacuum pump and connect
powder in a base fluid. After this no sedimentation was observed
the SSF unit condenser outlet and vacuum pump inlet. This drier filter
for the fluid for about next 12 h.
protects the vacuum pump from damage that may have occurred by
4. in order to achieve good suspension, the prepared Al2O3 nano-fluid
water vapor and/or air contaminations. The detailed technical specifica-
was added with anionic surfactant, SDBS, with mass fraction of
tions of control and safety devices used in experimental setup are pre-
wt.1%.
sented in Table 4.

2.3. Data collection

2.1.10. Control scheme
There are three control loops pertaining to the system. The first loop
After steady operation of the system data collection was performed
is a feedback (closed) control loop to control water temperature on the
using direct reading in 1 h time step period. Testing was conducted at
reservoir tank (TK1) by controlling the power supply of an electrical
steady-state conditions, which were monitored and controlled by the
heater by using digital display analog temperature controller switch.
abovementioned system. Steady-state conditions were reached in ap-
The second loop is an open control loop to control back up feeding
proximately 30 min. The system operation starts at 9:00 AM and the
water flow rate from mixing tank to helical heat exchanger by using a
data were recorded every 1 h during the day from 10:00 AM until
twin analog twin timer switch. The third loop is an open control loop
18:00 PM for each parameter during the period of 12–25 August 2013,
to control suctioned air flow from the SSF unit condenser to vacuum
in Tanta City (30° 47′ N, 31°E), as follows: (1) the weather condition
pump. This loop is investigated to prevent the back flow from vacuum
(ambient temperature, humidity, wind speed and solar radiation); (2)
pump to SSF condenser when vacuum pump shuts down because if
the temperature values at points (T1) to (T13); (3) the humidity values
at points (H1), (H2) and (H3); and (4) the freshwater level that was col-
Table 3
Technical specifications of sensors and probes. lected in product tank (TK2).

Sensor/type Reference Description Accuracy Table 4

K-type thermocouple Zhejiang Temperature ±1% Technical specifications of control and safety devices.
HTC-2 (10%–99% RH) Selectech Humidity and ±1%
Type Reference Description Size
and (−50 to +70 °C) temperature
TM-401 Tenmars Air velocity ±2% PVC gate valve (V6 and V7) Local Air flow 4 in.
DS 120 MGE Solar radiation ±1% PVC gate valve (V19) Local Air flow 2 in.
SP206316 (0–76 in. Hg) Videojet Pressure ±3% Brass ball valve, YT-Q003 Baltic Water flow 1/2 in.
LZM-15G (0.5–4 LPM), LZM-25G Zhejiang Water flow rate ±1.5% Solenoid valve, UD-8 Uni-D Water/air flow 3/8 in.
(2–20 LPM) Drier filter, DCL032 Danfoss Air flow 3/8 in.
56 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60

3. Results and discussion

Φa,co,o % ṁ dis;hdh kg/day ṁ dis;ssf kg/day

A number of tests were performed covering various combinations of

30.81

24.35
30.81
29.2
30.2

30.8

26.5
30.7
operating conditions. The results are presented in this section. Theoret-
ical results are also presented which are obtained using a simulation
model presented in an earlier paper by Kabeel and El-Said [16]. The pres-
ent hybrid (HDH–SSF) system was installed and mounted on the roof of

10.98
10.7
10.9
10.1

11.4
9.1
10.8
10.9
the technical workshops of Tanta University, Tanta, Egypt. Experimental
data involves solar radiation, wind speed, relative humidity, ambient
temperatures, product water mass flow rates and thermocouple read-

68.7
70.7

68.2
69.6
67.1
69.5
ings. The theoretical results were based on the measured heat source
70

65 and ambient temperature profiles from the experiments.

Φa,ev,o %

The (HDH–SSF) desalination processes have limitations that must be

87.4
86.9
89.6

89.9
91.9
considered prior to system operation for particular conditions such as:
92
88
85

• Feed water mass flow rate and air velocity of HDH unit; feed water
Ta,co,o °C

mass flow rate and air velocity of HDH unit must not exceed the
30.7
30.3

31.3
31.1

30.8
33.6
31
32

flooding point of the humidifier packed bed. If higher than the air ve-
Ta,ev,o °C

locity is operated, the greater the resistance that will be encountered

34.8

34.4
35.7

35.5

34.9
35.8

34.8

by the down-flowing water and the higher the pressure drop across
36

the packing. High pressure will crush and damage the packing in the
Ta,ev,in °C

tower. Too high air velocity will lead to flooding whereby the water
is “held” in the pockets, or void spaces, between the packing and
51

53

52

51
56
52

54

50

does not drain down through the packing and the operation be-
Tnf,sc,o °C

comes difficult to carry out. Flooding results in a high pressure drop.

• Nano-particle volume concentration; nano-particle volume concen-
83

85

87

84
83
84

86

85

tration must not exceed the sedimentation limit value which de-
Tnf,sc,in °C

pends on nano-particle material and base fluid. The sedimentation

effects of nano-particles on the heat transfer surfaces decrease the
75

77

79

76
75
76

77

77

thermal conductivity of nano-fluids and increase the pressure drop

Tw,ev,in °C

and pumping power consumption.

• Inlet cooling water temperature; when the inlet cooling water tem-
31
31

33

31
31
31

31

32

perature (Tcw,in) of SSF condenser decreases as low as 10 °C, its

mass flow rate would be decreased. This can decrease the velocity
Tw,fc,in °C

of the cooling saline water to a velocity lower than the acceptable

45
44

51

45
43
44

46

49

minimum. It can increase the specific volume to unacceptable limits

and unstable operation.
Tcw,in °C

• Feed water temperature of HDH or SSF unit; feed water temperature of

20

20
20
20

20
20
20

25

HDH or SSF unit must not exceed 70 °C to avoid scale formation. In-
ζ % ṁ cw;hdh kg/s ṁ cw;ssf kg/s

creasing the temperature of feeding water in the HDH process in-

creases the solubility of the different salts which causes some
0.025

0.045
0.045
0.045

0.045
0.045
0.045

0.045

difficulties in desalination operation and increases salt concentration

in the product water.
We have studied the influence of different parameters on the desa-
0.06

lination unit performances. The quantities studied are as follows:

0.1

0.1

0.1
0.1
0.1
0.1

0.1

• feed water mass flow rate of HDH, ṁ fw;ev = 0.7–1 kg/s.

0

1
0

0
0

• air mass flow rate, ṁ a = 0.025–0.03 kg/s.

Tam °C ṁ fw;ev kg/s ṁ a kg/s ṁ fw;fu kg/s

• feed water mass flow rate of SSF, ṁ fw;fu = 0.05–0.085 kg/s.

• cooling water mass flow rate of HDH, ṁ cw;co = 0.06–0.1 kg/s.
0.085

0.085
0.085

0.085

0.085
0.085

0.085
0.05

• cooling water mass flow rate of SSF, ṁ cw;fu = 0.025–0.045 kg/s.

Results of some measured average values for various days.

• cooling water inlet temperature of HDH and SSF, Tcw,in = 20–25 °C.
0.025

0.03
0.03

0.03
0.03

• nano-particle volume fraction, ζ = 0–1%.

0.03

0.03

0.03

In addition, there are some important operating parameters that

must be considered as follows:
0.7

• brine pool height inside flashing chamber of SSF unit, Hpoolfu = 10 cm.
1
1

1
1
1

• vacuum pressure inside flashing chamber of SSF unit, pvac,fc = 0.04–

32.5

0.07 bar.
35

33
33
33
35

36

32

• supply water salinity, x = 3000 ppm.

S W/m2

A measurement uncertainty analysis was conducted using stan-

928

970

952
925
933
920

950

930

dard deviation method to calculate the uncertainty of directly mea-

20/08/2013

24/08/2013
13/08/2013
14/08/2013
17/08/2013
18/08/2013
19/08/2013

21/08/2013

sured values within a given level of confidence (taken as 90%). For

quantities calculated from measured quantities, the method of propa-
Table 5

Days

gation of errors was used, where the total uncertainty was calculated
as the combination of uncertainties of individual components (taken
 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60 57

1300 40 100
Ambient temperature
1200
35 98
1100
1000 96
30

Ambient temperature, (°C).

Humidification efficiency ηev, (%)

Solar radiation, (W/m2).

900
Solar radiation 94
800 25

700 92
13/08/2013 20
600 14/08/2013
90 m fw ,ev = 1 kg/s. and m a = 0.03 kg/s.
500 17/08/2013 15
18/08/2013 m fw ,ev = 1 kg/s. and m a = 0.025 kg/s.
400 88
19/08/2013 10
300 m fw ,ev = 0.7 kg/s. and m a = 0.03 kg/s.
20/08/2013 86
200 21/08/2013 m fw , fu = 0.085 kg/s. T cw ,in , co = 20°C.
5
100 24/08/2013 84 m cw , co = 0.1 kg/s. T cw ,in , fu = 20°C.
0 0 m cw , fu = 0.045 kg/s. = 0%.
9 10 11 12 13 14 15 16 17 18 82
m swh = 0.014 kg/s.
Time (hr).
80
9 10 11 12 13 14 15 16 17 18
Fig. 6. Solar radiation and ambient temperatures for eight tests (each line style represents
Time (hr).
one test day).

Fig. 8. Time variation of efficiency of the humidification.

from Holman and Gajda [17]). The mean value of the fractional percent-
age error (Fr) between the theoretical and the experimental results for
daily water production was about 6% due to the uncertainty of the mea- of 5.7% and 4.3% for HDH and SSF units respectively. The theoretical pro-
suring process and the limited precision of the experimental instruments ductivity is deduced under the steady state condition, the change of sur-
(see Appendix A). Table 5 gives an example of the experimental results rounding conditions, the system losses, and the error in measurement.
obtained for various days of August. The results obtained from the present In addition, when the solar intensity is increased, the output obtained
investigation are compared with the previous theoretical results to vali- from the theory and experimental studies increases.
date the mathematical model of the system developed [16]. The ratio of actual to maximum vapor content difference in evapora-
The measured values of the average solar radiation and ambient tor chamber is defined as the humidifier efficiency and is given by:
temperature for Tanta City (30° 47′ N, 31°E), Egypt are given in Fig. 6.
Average relative humidity of the ambient air is between 15 and 40%
ωa;ev;o −ωa;ev;in
and wind speed almost constant during the experimental study is be- ηev ¼ 100 ð1Þ
ωa;ev;s −ωa;ev;in
tween 1.5 and 3 m/s in that period of time. The tilt angle of both types
of solar collector is 45° for all test cases.
The theoretical accumulative productivity and the experimental ac- where ωa,ev,in and ωa,ev,o are respectively the humidity of air at the inlet
cumulative productivity of the system are evaluated as shown in Fig. 7. and the outlet of the humidifier. ωa,ev,s is the saturation humidity corre-
The approximated agreement between the experimental and theoreti- sponding to the actual humidification process. This definition of the ef-
cal productivity and measurement results is good. The shift between ficiency is also used by El-Agouz [18] and Orfi et al. [19]. Fig. 8 presents
the productivity calculation and predicted results is about an average the time variation of the humidification efficiency of the humidifier
(ηev) for a typical day of August in Tanta at different air and feed
water mass flow rates. The humidifier efficiency increases with the in-
m fw , ev = 1 kg/s. ma = 0.03 kg/s. m cw , fu = 0.045 kg/s.
crease of feed water mass flow rate at the same air mass flow rate.
m fw , fu = 0.085 kg/s. T cw ,in , co = 20°C. m swh = 0.014 kg/s.
m cw , co = 0.1 kg/s. T cw ,in , fu = 20°C. = 0%.
50 70
HDH (Theo.), Kabeel and El-Said [17]
45 HDH (Exp.) = 0 %.
60
Solar water heater efficiency ηswh, (%)

SSF (Theo), Kabeel and El-Said [17]

40 = 1 %.
Accumulative productivity, (Lit)

SSF (Exp.)
50
35 HDH-SSF (Theo.), Kabeel and El-Said [17]
HDH-SSF (Exp.)
30 40
25
30
20

15 m fw , ev = 1 kg/s. ma = 0.03 kg/s.

20
m fw , fu = 0.085 kg/s. T cw ,in , co = 20°C.
10
m cw , co = 0.1 kg/s. T cw ,in , fu = 20°C.
10
5 m cw , fu = 0.045 kg/s. m swh = 0.014 kg/s.

0 0
9 10 11 12 13 14 15 16 17 18 9 10 11 12 13 14 15 16 17 18
Time (hr). Time (hr).

Fig. 7. Accumulative productivity through the day (13/8/2013). Fig. 9. Time variation of efficiency of the solar water heater.
58 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60

100 60
time 15:00 PM, the solar air heater efficiency (ηsah) reaches the min-
Outlet air temp.
90 imum value due to the decrease in temperature difference to solar
50 radiation intensity ratio. On the other hand, it is illustrated that the
Solar air heater efficiency ηsah, (%)

80
outlet air temperature and solar air heater efficiency are increased

Outlet air temperature, (°C).

m fw , ev = 1 kg/s.
70 with the decrease in the air mass flow rate.
m fw , fu = 0.085 kg/s. 40 The performance of desalination system is a measure of its efficiency
60 m cw , co = 0.1 kg/s. T cw ,in , co = 20°C. for producing water [20]. So; the system performance ratio can be esti-
m cw , fu = 0.045 kg/s. T cw ,in , fu = 20°C. 30
mated as follows:
50
m swh = 0.014 kg/s. = 0%.
40 ṁdis;ssf λfg
20 PRssf ¼ ð2Þ
30 SAswh

20
m a = 0.025 kg/s.
10
m a = 0.03 kg/s.
0 0
9 10 11 12 13 14
Time (hr).
Fig. 10. Time variation of efficiency of the solar air heater and outlet air temperature.
On the other hand, the humidifier efficiency is less affected with the
increase of air mass flow rate at the same feed water mass flow rate. An
explanation of these findings is as follows: increasing the feed water
mass flow rate increases the absolute humidity of the air leaving the hu-
midifier more than increasing of air mass flow rate. The humidifier effi-
ciency reaches 98% for ṁfw;ev = 1 kg/s and ṁa = 0.03 kg/s through all
day hours.
Variations of solar water heater efficiency (ηswh) with different
nano-particle volume fractions through day time are presented in
Fig. 9. It shows that with increasing nano-particle volume fraction of
constant working fluid mass flow rate the solar collector efficiency
(ηswh) will increase. An explanation of these findings is as follows: in-
creasing the nano-particle volume fraction increases the thermal con-
ductivity of the working fluid of solar collector that increases the heat
transfer coefficient inside solar water heater tubes, fluid outlet temper-
ature and heat gained.
The variation in solar air heater efficiency (η sah ) and outlet air
temperature for different air mass flow rates at 0.03 kg/s and
0.025 kg/s is displayed in Fig. 10. It is indicated that the efficiency
is decreased and increased with the increase and decrease in solar in-
tensity respectively. When solar intensity increased the collector
losses and the inlet air temperature (ambient) of solar air heater
are also increased. So, the temperature difference through the solar
air heater is decreased, which decreases the efficiency (η sah ). At
ηsah 10 where S is the solar intensity in W/m2, Aswh is the collector heat transfer
area in m2, λfg is the latent heat of vaporization stage and ṁ dis;ssf is the
rate of total distillate water productivity of SSF unit. The stage tempera-
ture drop is equal to the difference between feed water temperature
15 16 17 18
(Tw,fc,in) and that of the wasted blow down brine (Tw,fc,o) and is known
as the flashing range (ΔTstg). Figs. 11 and 12 show the hourly variations
in the system performance ratio (PR) and the flashing range (ΔTstg) re-
spectively. The figures show that PR ranges between 0.32 and 1.4 and
flashing ranges between 3 and 9 °C.
The comparison between the experimental performance ratio and
flashing range measurement and theoretical predicted is evaluated. The
approximated agreement between the experimental and theoretical re-
sults is good.
4. Conclusion
In this paper, an experimental investigation of a solar hybrid desalina-
tion system consisting of HDH unit and single stage flashing evaporation
unit was presented. This work was carried out on a solar desalination pro-
totype installed in the Tanta University, Tanta, Egypt. The experimental
results show that freshwater production increases with solar radiation.
The main conclusions are summarized as follows:
1. The productivity of the system increases with the increase of the
water temperature and air mass flow rate.
2. The solar water heater efficiency is affected by the nano-particle vol-
ume fraction.
3. The humidifier efficiency is affected by increasing water mass flow rate
more than the increasing air mass flow rate and reaches about 98%.
4. The system performance of the unit was affected by the values of the
inlet cooling water temperature and feed water mass flow rate.
5. The performance ratio of SSF unit is varied between 0.32 and 1.4 and
flashing ranges between 3 and 9 °C.

1.6 12
m fw , ev = 1 kg/s. ΔTstg. (Exp.)
1.4 ΔTstg. (Theo), Kabeel and El-Said [17]
m fw , fu = 0.085 kg/s. ma = 0.3 kg/s. 10
1.2 m cw ,co = 0.06 kg/s. T cw ,in , co = 20°C.
Flashing range, ΔTstg. (°C).

m cw , fu = 0.045 kg/s. T cw ,in , fu = 20°C.

Performance ratio, PR

8
1.0 m swh = 0.014 kg/s.
= 0%.
0.8 6
m fw , ev = 1 kg/s.
0.6 m fw , fu = 0.085 kg/s. ma = 0.025 kg/s.
4
m cw , co = 0.1 kg/s. T cw ,in , co = 20°C.
0.4
m cw , fu = 0.045 kg/s. T cw ,in , fu = 20°C.
2 m swh = 0.014 kg/s.
0.2 PR (Exp.)
= 0%.
PR (Theo), Kabeel and El-Said [17]
0 0
10 11 12 13 14 15 16 17 18 10 11 12 13 14 15 16 17 18
Time (hr). Time (hr).

Fig. 11. Time variation of the SSF unit performance ratio (19/8/2013). Fig. 12. Time variation of the SSF unit flashing range (14/8/2013).
 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60 59

6. Collector's efficiencies are about 55% and 56% for solar water heater PVC polyvinyl chloride
and solar air heater respectively.
7. The experiment measured agreed well with the experimental calcula- Appendix AA.1. Uncertainty analysis
tion productivity. Within the studied ranges, the maximum productiv-
ity of the system reached 41.8 kg/day according to test and operation The uncertainty in the result on the basis of the uncertainties in the
conditions. laboratory measurements is calculated from the following formula by
Holman and Gajda [17]:
Nomenclatures vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u n  2
Latin symbols uX ∂R
WR ¼ t  Wi : ðA:1Þ
A area, m2 i¼1
∂X i
D diameter, m
X water salinity, ppm
For a result given as:
S actual absorbing radiation solar energy, W/m2
ṁ mass flow rate, kg/s R ¼ RðX 1 ; X 2 ; …; X n Þ ðA:2Þ
P pressure, N/m2
T temperature, °C where:
PR performance ratio, dimensionless
H height, m WR uncertainty in the result,
Fr fractional error, dimensionless X1, X2,…, Xn independent variables, and
V volume, m3 W1, W2,…, Wn uncertainty in the corresponding variables.

Greek symbols The uncertainty analysis in this work comprises uncertainty in calcu-
β collector tilt angle, degree lating the volume of the product (distillate) water. It is a function of
ω humidity ratio, kgwater vapor/kgdry air three variables in the form of:
ζ nano-particle volume fraction, dimensionless  
η efficiency, dimensionless V dis ¼ f Hdis;ssf ; Hdis;hdh ; ddis;tank;in : ðA:3Þ
λ latent heat, kJ/kg
θ air solar heater absorber groove angle, degree
This volume is given as:
Φ relative humidity, dimensionless
V dis ¼ Hdis;ssf  Adis;tank;in þ H dis;hdh  Adis;tank;in ðA:4Þ
Subscripts
stg stage where:
a air
w water Hdis,ssf freshwater height of SSF unit (m water)
am ambient Adis,tank,in feeding tube cross-sectional area (m2)
o out Hdis,hdh freshwater height of HDH unit (m water).
in in
s saturation
The uncertainty in product water volume is given by:
fg difference between the saturated vapor and saturated liquid
values of the same property 2 !2 !2 31=2
∂V dis ∂V dis
cw cooling water 6  W Hdis;ssf þ  W Hdis;hdh 7
6 ∂H ∂Hdis;hdh 7
dis distillate 6 dis;ssf
!2 7
W V dis ¼6 7 : ðA:5Þ
ev humidifier 6 ∂V 7
4 þ2 dis
 W Adis;tank;in 5
co dehumidifier ∂Adis;tank;in
swh solar water heater
sah solar air heater
The following partial derivatives can be obtained from Eq. (A.4):
fu flashing unit
bp brine pool ∂V dis
fw feed water ¼ Adis;tank;in ðA:6Þ
∂Hdis;ssf
fc flashing chamber
vac vacuum
pool brine pool
∂V dis
ssf single stage flashing ¼ Adis;tank;in ðA:7Þ
∂Hdis;hdh
hdh humidification and dehumidification
tank product tank
V volume, m3
∂V dis
¼ Hdis;ssf þ H dis;hdh : ðA:8Þ
∂Adis;tank;in
Superscripts
c calculated
m measured From Eqs. (A.6) to (A.8), substituting Eq. (A.5) and using Eq. (A.4), the
volume of the product water uncertainty (W V dis ) can be determined as:
Abbreviation 8h i2 h i2 91=2
>
< Adis;tank;in  W H þ  >
=
MSF Multi-Stage Flashing Adis;tank;in W H dis;hdh þ
NEA Non-Equilibrium Allowance W V dis ¼ h dis;ssf
 i2 : ðA:9Þ
>
: 2 Hdis;ssf þ H dis;hdh  W A >
;
SDBS Sodium Dodecyl Benzene Sulfonate dis;tank;in
60 A.E. Kabeel, E.M.S. El-Said / Desalination 341 (2014) 50–60

A sample calculation of uncertainty in Eq. (A.9) is presented as [2] Y.J. Dai, H.F. Zhang, Experimental investigation of a solar desalination unit with hu-
midification and dehumidification, Desalination 130 (2000) 169–175.
follows; [3] E.H. Amer, H. Kotb, G.H. Mostafa, A.R. El-Ghalban, Theoretical and experimental in-
  vestigation of humidification and dehumidification desalination unit, Desalination
H dis;ssf ¼ 0:0411 m water W Hdis;ssf ¼ 0:001 m 249 (2009) 949–959.
[4] A.S. Nafey, H.E.S. Fath, S.O. El-Helaby, A.M. Soliman, Solar desalination using humid-
ification dehumidification processes: Part I, A numerical investigation, Energy Con-
version and Management 45 (2004) 1243–1261.
  [5] A.S. Nafey, H.E.S. Fath, S.O. El-Helaby, A. Soliman, Solar desalination using humidifi-
H dis;hdh ¼ 0:0167 m water W Hdis;ssf ¼ 0:001 m cation–dehumidification processes. Part II. An experimental investigation, Energy
Convers. Manag. 45 (2004) 1263–1277.
[6] M. Ben Amara, I. Houcine, A. Guizani, M. Mgalej, Theoretical and experimental study
  of a pad humidifier used in a seawater desalination process, Desalination 168
−2 2 −4 2
Adis;tank;in ¼ 7:065  10 m W Adis;tank;in ¼ 4:71  10 m : (2004) 1–12.
[7] Y.J. Dai, R.Z. Wang, H.F. Zhangb, Parametric analysis to improve the performance of a
solar desalination unit with humidification and dehumidification, Desalination 142
Therefore, from Eq. (A.4) the volume of the product water Vdis = 4.1 (2002) 107–118.
[8] A.H. El-Shazly, A.A. Al-Zahrani, Y.A. Alhamed, S.A. Nosier, Productivity intensification
× 10−3 m3, and its uncertainty, Eq. (A.9), W V dis = 3.8503 × 10−5 m3 of humidification–dehumidification desalination unit by using pulsed water flow
and regime, Desalination 293 (2012) 53–60.
[9] A.S. Nafey, M.A. Mohamad, S.O. El-Helaby, M.A. Sharaf, Theoretical and experimental
W V dis study of a small unit for solar desalination using flashing process, Energy Convers.
Uncertainty% ¼ ¼ 0:94%: Manag. 48 (2007) 528–538.
V dis [10] H. Baig, M.A. Antar, S.M. Zubair, Performance characteristics of a once-through
multi-stage flash distillation process, Desalin. Water Treat. 13 (2010) 174–185.
[11] Y. Junjie, Z. Dan, C. Daotong, W. Guifang, L. Luning, Experimental study on static/cir-
culatory flash evaporation, Int. J. Heat Mass Transf. 53 (2010) 5528–5535.
A.2. Fractional error [12] M. Abutayeh, D. Yogi Goswami, Solar flash desalination under hydrostatically
sustained vacuum, J. Sol. Energy Eng. 131 (2009) 1–7.
The measured experimental daily water production includes some er- [13] M. Saad, M. Ahmed, V.M. Morcos, Performance analysis of a vacuum desalination
system, Proceedings of the ASME International Mechanical Engineering Congress
rors due to the uncertainty of the measuring process and the limited pre- & Exposition, Denver, Colorado, USA, 2011.
cision of the experimental instruments. For examination of the validity of [14] W. Daungthongsuk, S. Wongwises, A critical review of convective heat transfer of
the experimental work; the following fractional error (Fr) equation is nano-fluids, Renew. Sustain. Energy Rev. 11 (2007) 797–817.
[15] T. Yousefi, F. Veysi, E. Shojaeizadeh, S. Zinadini, An experimental investigation on
used:
the effect of Al2O3–H2O nano-fluid on the efficiency of flat-plate solar collectors,
Renew. Energy 39 (2012) 293–298.
ṁ cdis −ṁ mdis [16] A.E. Kabeel, M.S. Emad, A. El-Said, hybrid solar desalination system of air humidifi-
%Fr ¼ 100 : ðA:10Þ cation–dehumidification and water flashing evaporation: part I. A numerical inves-
ṁ cdis
tigation, Desalination 320 (2013) 56–72.
[17] J.P. Holman, W.J. Gajda, Experimental Methods for Engineers, McGraw-Hill, 1978.
According to Eq. (A.10) the average (Fr) percentage for measuring [18] S.A. El-Agouz, A new process of desalination by air passing through seawater based
the daily water production was about 4–6%. on humidification–dehumidification process, Energy 35 (12) (2010) 5108–5114.
[19] J. Orfi, M. Laplante, H. Marmouch, N. Galanis, B. Benhamou, S. Ben Nasrallah, C.T.
Nguyen, Experimental and theoretical study of a humidification–dehumidification
References water desalination system using solar energy, Desalination 168 (2004) 151–159.
[20] H.T. El-Dessouky, H.M. Ettouney, Fundamentals of Salt Water Desalination, 1st edi-
[1] H.P. Garg, R.S. Adhikari, K. Rakesh, Experimental design and computer simulation of tion Elsevier, 2002
multi-effect humidification–dehumidification solar distillation, Desalination 153
(2002) 81–86.