You are on page 1of 10

Desalination 445 (2018) 95–104

Contents lists available at ScienceDirect

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

Performance analysis of a heat pump driven humidification- T


dehumidification desalination system

Saeed Dehghani , Abhijit Date, Aliakbar Akbarzadeh
Energy Conversion and Renewable Energy Group, School of Automotive and Mechanical Engineering, RMIT University, Melbourne 3083, VIC, Australia

A R T I C LE I N FO A B S T R A C T

Keywords: In this paper, a humidification-dehumidification desalination HDH cycle with direct contact dehumidifier is
Humidification-dehumidification coupled with a heat pump. The heat pump is introduced in order to simultaneously supply the cooling and
HDH heating loads of the HDH cycle. Next, by implementing a mathematical model, the performance of the system is
Desalination investigated under different working conditions. To incorporate the humidifier and dehumidifier effectiveness
Heat pump
into the theoretical modeling, the ε-NTU correlation of the heat and mass transfer device is utilized. Performance
Direct contact dehumidifier
parameters such as specific electrical energy consumption (SEEC), recovery ratio (RR) and coefficient of per-
formance (COP) are employed for evaluating the system performance. The fully coupled condition of the HDH
system with no extra cooling or heating scenarios is investigated. It is shown that a given saline water and
freshwater temperatures, a fully coupled HDH system with the heat pump without adding the extra cooler can be
achieved by alternating mass flow rate ratio of either seawater to freshwater or seawater to dry air.

1. Introduction sources of the process from fossil fuels with renewable ones, especially
solar energy [14–16]. In other studies, an alternative low-grade heat
Fresh water is one of the vital needs for societies all around the source such as industrial waste heat or geothermal [3,17–19] were
world. Desalination is the process of producing fresh water from saline considered for supplying the thermal energy demand of the cycle. To
or impure water. Various types of desalination methods have been make the dehumidification process in non-direct HDH cycle cost-ef-
proposed and developed by researchers and can be categorized into fective, an efficient low-cost method can be proposed to condense water
membrane-based or thermal based methods. vapor out of the air stream. With a large fraction of the air/vapor
The most widely commercialized desalination processes based on mixture being non-condensable, direct contact condensation is con-
evaporation technique are multi-stage flash distillation, multiple effect siderably more effective than film condensation.
distillation and vapor compression. The second category of desalination There have been several studies done in this regard. Bharathan et al.
processes uses membrane technologies including reverse osmosis, [20] initially introduced a direct contact condenser approach to en-
electrodialysis and membrane distillation [1]. Humidification-dehumi- hance the heat transfer rate in presence of non-condensable gas.
dification (HDH) desalination has recently attracted researcher's at- Klausner et al. [21] fabricated a laboratory scale direct contact con-
tention [2–5]. denser to study the variation of temperature, humidity and condensa-
The main advantage of the HDH cycle is its small scale application tion rate through the condenser system. They evaluated their result by
which makes is a competitive alternative among other desalination considering a finite volume method for analyzing the direct contact
technology for remote area water supply. The cycle can be operated packing condenser. Yi Li et al. [22] studied the performance char-
reliably when it is assisted with a heat pump. Moreover, the compo- acteristics of HDH desalination with a DC dehumidifier desalination
nents of the HDH system are usually affordably available on the market process utilizing heat and mass transfer analysis. They also studied the
and the maintenance of the system in not complicated in terms of labor performance of a DC dehumidifier using a laboratory scale setup [23]
and tools [6,7]. under different conditions. Alnaimat et al. [24–26] performed transient
There have been numerous works on non-direct condensers for analysis of the HDH with DC dehumidifier using a one-dimensional
dehumidification process in the HDH cycle [8–13]. Moreover, in this numerical solution [24] as well as a transient dynamic response with
area, several studies have been performed to replace the thermal energy the same method [26]. They further examined the operation of their


Corresponding author.
E-mail address: Saeed.dehghani@rmit.edu.au (S. Dehghani).

https://doi.org/10.1016/j.desal.2018.07.033
Received 1 April 2018; Received in revised form 31 July 2018; Accepted 31 July 2018
0011-9164/ © 2018 Elsevier B.V. All rights reserved.
S. Dehghani et al. Desalination 445 (2018) 95–104

Nomenclature Subscripts

Symbol a air
da dry air
T temperature (°C) dw distilled water
ṁ flow rate (kg/s) b bottom
Q̇ heat rate (kW) t top
Ḣ enthalpy rate (kW) sw Seawater, saline water
mr Water to air mass flow rate ratio (–) fw fresh water
h specific enthalpy (kJ/kg) max maximum
RR recovery ratio (–) m middle
cp specific heat capacity at constant pressure (kJ/kg·K) br brine
SEEC specific electrical energy consumption h humidifier
COP coefficient of performance d dehumidifier
f average slope of the saturated air enthalpy versus tem- in input
perature (kJ/kg·K) out output
NTU number of transferred units 1, 2, 3, 4 flow states
Me Merkel number com compressor
a specific area (m2/m3) con condenser
H packing height (m) eva evaporator
CR heat capacity rate ratio ref refrigerant
PR pressure ratio (–)
Ẇ work (kW) Acronyms

Greek letters CAOW closed-air open-water system


CWOA cold-water open-air
ω absolute humidity of dry air or humidity ratio (kgw/kga) HDH humidification-dehumidification
φ relative humidity (–) HME heat and mass exchanger
ε effectiveness (–) DC direct contact
Δ difference or change
η efficiency (–)

solar driven HDH setup under various design and operating conditions seawater. They reported maximum productivity of 12.38 kg/kW h.
[25]. Shafii et al. [38] experimentally studied the performance of an open air-
Eslamimanesh and Hatamipour [27,28] conducted a theoretical heated HDH system with a heat pump to heat air in the condenser and
analysis for the open-air open-water HDH cycle to study the effect of cool it down after humidification at the evaporator. They obtained
working parameters on water production rate as well as performed an highest yield and GOR of 2.79 kg/h and 2.08, respectively.
economic study of the system. Niroomand et al. [29] investigated A lack of knowledge for the HDH with direct contact and heat pump
freshwater production, efficiency and effects of various parameters in- to provide the energy requirements can be clearly observed from the
cluding air flow rate, conditions of inlet cold and hot water and velocity literature. Therefore, in this study a comprehensive theoretical analysis
and diameter of droplets on the performance of the open-air HDH cycle of a heat pump driven HDH cycle with a DC dehumidifier was per-
with DC dehumidifier system. Mehrgoo and Amidpour investigated the formed in order to investigate the effect of operational parameters on
optimum water production rate utilizing constructal design theory for a the specific energy consumption and water production of the system.
fixed-size HDH system [30]. Ettouney [31] introduced different types of
dehumidifier including vapor compression, desiccant air drying and
2. System description
membrane air drying.
He et al. [32] studied a direct-contact HDH desalination system
Fig. 1 shows the overview of the heat pump assisted HDH with DC
through a thermodynamic based mathematical model. They applied
dehumidifier. It utilizes two heat and mass exchanger devices for pro-
plate heat exchangers to recover waste heat for thermal energy con-
ducing distilled water. These are the humidifier and dehumidifier.
sumption/requirements of the system. In another study, He et al. [33]
There is simultaneous heat and mass transfer in heat and mass exchange
used a mechanical compression heat pump to study the performance of
(HME) devices due to temperature and concentration gradients be-
the water-heated HDH with a non-direct dehumidifier. They reported
tween water and air. These HME devices consist of packing fill that
maximum GOR of 5 and water production of 85 kg/h. Lawal [34,35]
provides high surface area for effective heat and mass transfer. To si-
compared the performance of the heat pump assisted air-heated and
multaneously provide the heating demand in the humidifier and
water-heated HDH with non-direct dehumidifier desalination. By ap-
cooling demand in the dehumidifier a heat pump is considered. It is
plying a mathematical model they investigated the influence of oper-
used in order to cool down the fresh water after dehumidification
ating parameters on the performance of the system. They reported
process by transferring its gained heat to the working fluid of the heat
maximum recovery ratio and GOR of 4% and 10 for the system, re-
pump. Also, the heat pump supplies the heating demand of the saline
spectively. Queiroz et al. [36] coupled a vapor compression heat pump
directly to an adiabatic humidifier for purified water production by water at the same time in the high-temperature side by consuming
electrical power in the compressor.
recirculation air. Xu et al. [37] experimentally studied a heat pump
assisted water-heated HDH with non-direct dehumidifier system. They In the humidifier saline water is sprayed over packing fill after it is
heated in the high temperature side of the heat pump. This is done to
added an evaporator after the dehumidifier for extra cooling of the air
and a solar collector after the condenser for additional heating of feed increase the vapor content of the air. Then, the hot moist air is trans-
ferred to the dehumidifier in which cold fresh water is sprayed over it.

96
S. Dehghani et al. Desalination 445 (2018) 95–104

• Humidifier:
ṁ sw + ṁ da ωa, b = ṁ br + ṁ da ωa, t (1)

ṁ sw hsw, t + ṁ da ha, b = ṁ br hbr + ṁ da ha, t (2)


The heating load that needs to be added to the saline water stream
can be found out as follows:
̇ = ṁ sw cp, sw (Tsw, t − Tbr )
Qin (3)

• Dehumidifier:
ṁ da ωa, t = ṁ dw + ṁ da ωa, b (4)

ṁ fw hfw, t + ṁ da ha, t = ṁ fw hfw, b + ṁ da ha, b + ṁ dw hdw (5)


It should be noted that the dry mass flow of air is constant through
the humidifier and dehumidifier. Also, enthalpy of the humid air is
considered a binary mixture of dry air and water vapor, in other words:
ha = hda + ωhv.
The heat that needs to be rejected for cooling down the fresh water
coming out from the dehumidifier can be calculated as follow:
̇ = ṁ fw cp, fw (Tfw, b − Tfw, t )
Qout (6)
In order to obtain outlet stream conditions in the humidifier and
dehumidifier, the effectiveness equation needs to be defined.
Effectiveness compares the actual thermal energy transferred versus
ideal thermal energy transferred from each stream and is defined as
actual enthalpy rate variation to the maximum possible enthalpy rate
variation, in other words ε = ΔḢ /ΔHmax
̇ [40]. Therefore, the effec-
tiveness of humidifier and dehumidifier would be denoted as follows
[40]:

⎛ Ḣa, t − Ḣa, b ̇ , t − Hbr


Hsw ̇ ⎞
εh = max ⎜ ideal ,
Ḣ − H ̇ , t − Ḣ ideal br ⎟
̇a, b Hsw (7)
Fig. 1. Overview of the heat pump driven HDH with the direct contact dehu- ⎝ a, t ⎠
midifier.
⎛ Ḣa, t − Ḣa, b + Ḣdw ̇ , b − Hfw
Hfw ̇ ,t ⎞
εd = max ⎜ , ideal
̇ ̇ ideal ̇ ̇ ̇ ⎟
This causes condensation of the vapor in the humid air. ⎝ Ha, t − H a, b + Hdw H fw, b − Hfw, t ⎠ (8)
Dehumidification of the humid air will happen by cooling it down to its
dew point temperature along the dehumidifier column, which results in
forming droplets of distilled water on both packing surface and the
freshwater stream. The distilled water, along with fresh cooling water,
will be collected together at the bottom of the dehumidifier. The dis-
tilled water will be separated and collected while the fresh water will be
cooled down through a heat exchanger in the low-temperature side of
the heat pump to be sprayed again.
There are different configurations of HDH cycles based on which
stream is heated as well as which stream is open or closed. In closed air/
water cycles, one fluid is circulated in a closed loop between humidifier
and dehumidifier while the other fluid is in an open loop. For instance,
closed-air open-water (CAOW) water-heated cycle is demonstrated in
Fig. 1. Different types of the HDH cycle can be considered for different
applications such as close-air open-water air-heated cycle, close-water
open-air (CWOA) water-heated and close-water open-air cycle (CWOA)
air-heated [6,10,39].

3. Mathematical modeling

3.1. HDH cycle

In order to obtain the theoretical performance of the system, a


thermodynamic analysis as well as heat and mass transfer analysis is
performed. Fig. 2 shows the schematic diagram of the system with the
variables used in the mathematical modeling.
Applying mass and energy balance on system components will result Fig. 2. Schematic diagram of the heat pump driven HDH with DC dehumidifier
in following governing equations: system.

97
S. Dehghani et al. Desalination 445 (2018) 95–104

In both humidifier and dehumidifier, the ideal outlet air enthalpy NTUd = Med fd / cpfw if mrd cp, fw < fd (16)
occurs when the outlet air is fully saturated at the water inlet tem-
perature, and the ideal outlet seawater enthalpy is when its temperature NTUd = Med mrd if mrd cp, fw > fd (17)
is equivalent to the inlet air temperature.
Analytical expressions for the effectiveness of a heat and mass ex- Med = 2.049mrd(−0.779) Hd0.632 (18)
changer as a function of the parameters like heat capacities and heat
and mass transfer characteristics are generally not available unless ṁ fw cp, fw ṁ da fd ⎞
considering some approximations or by using numerical solutions. CRd = min ⎛⎜ , ⎟

Numerical solution can be achieved by applying the differential ele- ⎝ ṁ da fd ṁ fw cp, fw ⎠ (19)
ment method on the HME device and solving the differential equations For both humidifier and dehumidifier, packing height of 2.25 m is
numerically. considered for guaranteeing higher values of HME effectiveness for a
Merkel used the differential control volume method for cooling variety of working conditions. In order to calculate the cross-sectional
towers and derived the governing equations in the form of three dif- area of the HME devices, it is assumed that water flux over the packing
ferential equations [41]. Jaber and Webb [42], by considering some material has a value between 0.8 and 4.2 kg/sm2 [43,44].
approximations, solved these equations and presented the analytical
correlation for the effectiveness of cooling towers. In this work, the
3.2. Heat pump
performance of a cooling tower is considered similar to a humidifier.
Therefore, by implementing the same method used by Jaber and Webb
A mechanical compression water to water heat pump is applied/
[42] the effectiveness correlation of the humidifier can be obtained as
added to the HDH cycle in order to simultaneously cool down fresh
follows:
warm water coming from the dehumidifier and heat up the saline water
1 − exp(−NTUh (1 − CR h)) feeding to the humidifier. Temperature profiles of low-temperature side
εh =
1 − CRh exp(−NTUh (1 − CR h)) (9) (evaporator) and high-temperature side (condenser) are shown in
Fig. 3. For evaporator, a terminal temperature difference of 10 °C is
where NTUh is the number of transfer units in humidification process.
considered and for condenser, pinch temperature difference was greater
NTUh = Meh fh / cpsw if mrh cp, sw < fh (10) than 5 °C. As the temperature of the seawater at the pinch point is not
known prior, the temperature of the refrigerant at pinch point is con-
NTUh = Meh mrh if mrh cp, sw > fh (11) sidered 5 °C higher than the temperature of sprayed seawater to guar-
antee the required pinch temperature difference.
where f is the average slope of the saturated air enthalpy versus tem-
Fig. 4 shows the T-s diagram of an ideal vapor compression cycle
perature. This considers the slope as an effective specific heat for the
heat pump. Refrigerant R134-a is used in this study as the heat pump
moist air stream. Also, Me is the Merkel number which is a function of
working fluid.
packing characteristics [41]:
Each component of the heat pump will now be analyzed.
KVa
Me =
ṁ w (12)
3.2.1. Compressor
Packing fill manufacturers provide this correlation. In this study a The refrigerant entering the compressor as saturated vapor is com-
polypropylene packing material from Brentwood Industries with model pressed to a superheated vapor. Its pressure and temperature are in-
number: CF1200 MA and a specific area of 226 m2/m3 was used with creased during this process. The increase in temperature allows for
the following correlation provided by the manufacturer: heating up the seawater in the condenser. The isentropic efficiency of
the compressor is considered to be 80%.
Meh = 2.049mrh(−0.779) Hh0.632 (13)
̇ = ṁ ref (h2 − h1)/ ηcom
Wcom (20)
Moreover, heat capacity rate ratio, CRh, defined by Jaber and Webb
[42] is as follows. The pressure ratio (PR) of the compressor is outlet pressure divided
by the inlet pressure (PR = P2 / P1).
ṁ sw cp, sw ṁ da fh ⎞
CRh = min ⎛⎜ , ⎟
⎝ ṁ da fh ṁ sw cp, sw ⎠ (14) 3.2.2. Condenser
The ε-NTU correlation of the DC dehumidifier can be obtained by The condenser is considered in a way that can fully provide the heat
applying the same method used in a humidifier as follows: load required by the HDH cycle for heating up seawater for the target
temperature.
1 − exp(−NTUh (1 − CR d))
εd = ̇ = ṁ ref (h2 − h3) = Qin
̇
1 − CRd exp(−NTUh (1 − CR d)) (15) Qcon (21)

Fig. 3. Temperature profiles of the refrigerant along the a) evaporator and b) condenser.

98
S. Dehghani et al. Desalination 445 (2018) 95–104

7
6.5
6

Recovery ratio, RR (%)


5.5
5
4.5
4
3.5 Tfw,t=10 ºC
Tfw,t=20 ºC
3 Tfw,t=30 ºC
2.5
Fig. 4. Temperature versus entropy diagram of the heat pump with R-134a 2
refrigerant. 1 1.5 2 2.5 3 3.5 4 4.5
Mass flow rate ratio of seawater to air, mr h (-)
7 Fig. 7. Effect of freshwater top temperature on recovery ratio at different mass
6.5 flow rate ratios of seawater to dry air of humidifier at Tsw,t = 70 °C, mrsw/fw = 1
(–).
6
Recovery ratio, RR (%)

5.5
550
5
4.5 mrsw/fw=0.8 (-)
500
4 mrsw/fw=1.0 (-)
mrsw/fw=1.2 (-)

SEEC (kW.hr/m )
mrsw/fw=0.8 (-)

3
3.5
mrsw/fw=1.0 (-) 450
3 mrsw/fw=1.2 (-)
2.5 400
2
1 1.5 2 2.5 3 3.5 4 4.5
Mass flow rate ratio of seawater to air, mr h (-) 350

Fig. 5. Effect of seawater to fresh water mass flow rates on recovery ratio for
300
various mass flow rate ratios of seawater to dry air of humidifier at 1 1.5 2 2.5 3 3.5 4 4.5
Tsw,t = 70 °C, Tfw,t = 20 °C. Mass flow rate ratio of seawater to air, mr h (-)

7 Fig. 8. Specific electrical energy consumption of the system versus mass flow
rate ratio of seawater to dry air of humidifier at different seawater to fresh
water mass flow rate ratios at Tsw,t = 70 °C, Tfw,t = 20 °C.
6
Recovery ratio, RR (%)

5 600
Tsw,t=70 °C Tsw,t=70 ºC
550
4 Tsw,t=60 °C Tsw,t=60 ºC
Tsw,t=50 ºC 500 Tsw,t=50 ºC
SEEC (kW.hr/m )
3

3
450

2 400

1 350
1 1.5 2 2.5 3 3.5 4 4.5
300
Mass flow rate ratio of seawater to air, mr h (-)
250
Fig. 6. Effect of seawater top temperature on recovery ratio at different mass 1 1.5 2 2.5 3 3.5 4 4.5
flow rate ratios of seawater to dry air of humidifier at Tfw,t = 20 °C, mrsw/fw = 1
Mass flow rate ratio of seawater to air, mr h (-)
(–).
Fig. 9. Specific electrical energy consumption of the system versus mass flow
rate ratio of seawater to dry air of the humidifier at different seawater top
3.2.3. Expansion valve
temperature at Tfw,t = 20 °C, mrsw/fw = 1 (–).
After condensation of the refrigerant by transferring heat to the
seawater stream, it goes through an isenthalpic expansion valve for
further cooling caused by pressure drop. by decreasing the mass flow rate of fresh water, dry air mass flow rate
or adding an extra cooler.
h3 = h4 (22)
̇ = ṁ ref (h1 − h4 )
Qeva (23)

3.2.4. Evaporator ̇
Qcooler = (ṁ fw − ṁ fw,max ) × cpfw × (Tfw, b − Tfw, t ) (24)
The evaporator is used to cool the warmed fresh water after dehu-
midification. Based on the cycle working temperatures there might be a where ṁfw,max is the maximum flow rate of fresh water that can be
need of extra cooler capability as the evaporator may not provide all the cooled down to its target temperature without using an extra cooler.
cooling demand of the fresh water stream. This can be provided either COP is defined for both cooling at the low temperature side and

99
S. Dehghani et al. Desalination 445 (2018) 95–104

600 heating at the high temperature side:


Tfw,t=10 ºC
550
̇
Qcon
Tfw,t=20 ºC COPheating =
Tfw,t=30 ºC ̇
Wcom (25)
SEEC (kW.hr/m )
3

500
̇
Qeva
COPcooling =
450
̇
Wcom (26)
Considering the above equations, a system of non-linear equations
400
can be solved numerically with the following assumptions:

• Cycles operate at steady state and steady flow conditions.


350

• Humidifier and dehumidifier are adiabatic.


• Pumping and fan powers are negligible in comparison to the total
300
1 1.5 2 2.5 3 3.5 4 4.5
Mass flow rate ratio of seawater to air, mr h (-) thermal energy input of the system [10,43].

Fig. 10. Specific electrical energy consumption of the system versus mass flow
• Kinetic and potential energy terms are excluded in the energy bal-
ance.
rate ratio of seawater to dry air of the humidifier at different freshwater top
temperatures at Tsw,t = 70 °C, mrsw/fw = 1 (–). • The distilled water condensed in the dehumidifier is assumed to
leave at a temperature which is the average of the humid air tem-
peratures at inlet and outlet of the dehumidifier [10].
Table 1
Maximum GOR values of the HDH system at different working conditions.
• The air circulating throughout the system is fully saturated [27,45].

Tsw,t (°C) Tfw,t (°C) mrsw/fw (–) GOR (–) 4. Metrics of the system

70 20 0.8 0.73
To evaluate the system performance at various working condition,
1 0.67
1.2 0.61 different non-dimensional parameters are utilized. Mass flow rate ratio
70 10 1 0.67 of liquid to gas in HME plays a major role in the device effectiveness.
20 0.50 Thus, the mass flow rate of seawater to dry air is defined as follows:
30 0.34
70 20 1 0.76 ṁ sw
mrh =
60 0.67 ṁ da (27)
50 0.54
The mass flow rate of seawater to fresh water can be used for ad-
justing of cooling and heating loads in order to improve the system
7 10 performance in terms of energy consumption and water production. It is
6.5 defined as:
5
6 ṁ sw
Recovery ratio, RR (%)

mrsw / fw =
Cooling load (kJ/kg)

5.5 0 ṁ fw (28)
5 -5 The recovery ratio is the ratio of the amount of water produced to
4.5
the feed saline water in a certain time span as follows:
Cooling load -10
4
Cooler load ṁ dw
3.5 -15 RR = × 100 (%)
RR(-) ṁ sw (29)
3
-20 Gain output ratio (GOR) indicates heat recovery in the system and is
2.5
defined as the ratio of the latent heat of evaporation of the distillate
2 -25
1 1.5 2 2.5 3 3.5 4 4.5 water produced to the total heat input to the system from the heat
Mass flow rate ratio of seawater to air, mr h (-) source.
ṁ dw hfg ṁ dw hfg hfg
Fig. 11. Cooling load of the HDH system at various mass flow rate ratios of GOR = = = RR
seawater to dry air at the working condition of Tfw,t = 20 °C, Tsw,t = 70 °C, mrsw/ ̇
Qin ṁ sw cp, sw (Tsw, t − Tsw, b ) cp, sw ΔTsw (30)
fw = 1.
Specific electrical energy consumption is the ratio of the compressor
work to produced distilled water.
Table 2
Fully coupled condition of the HDH system driven by the heat pump. Wċ
SEEC = (kW h/m3)
ṁ dw (31)
Tsw,t (°C) Tsw,t (°C) mrsw/ Optimum Fully Cooling Modified
fw (–) mrh (–) coupled load (kJ/ mrsw/fw (–) For incorporating accurate and reliable thermodynamic properties
kg)
of moist air, water and seawater, ASHRAE handbook [46] as well as
70 20 0.8 1.8 ✗ 20 0.93 Engineering Equation Solver (EES) software [47] was used. EES soft-
1 2.4 ✗ 8.7 1.06 ware has a complete library of thermophysical properties for substances
1.2 3.1 ✓ −7.4 1.15 including moist air, using the formulation presented by Hyland and
70 10 1 2.1 ✗ 50 1.36 Wexler [48], as well as water, using the formulation of IAPWS (Inter-
20 2.4 ✗ 8.7 1.06
30 3.1 ✓ −34 0.79
national Association for Properties of Water and Steam) [49].
70 20 1 2.4 ✗ 8.7 1.06 The obtained system of nonlinear equations was solved utilizing
60 1.8 ✓ −7.3 0.94 Engineering Equation Solver (EES) software, which calculates moist air
50 1.5 ✓ −17 0.84 and water properties using built-in functions. These functions have
been previously defined in the software and evaluate the thermo-
physical properties of various substances based on a database in the

100
S. Dehghani et al. Desalination 445 (2018) 95–104

Fig. 12. Flow chart of the solution procedure to meet zero cooling demand.

9 9
8 8
7 7
Pressure ratio, PR (-)
Pressure ratio, PR (-)

6 6
5 5
4 4
3 3
2 2
1 1
0 0
45 50 55 60 65 70 75 5 10 15 20 25 30 35
Top seawater temeprature, Tfw,t (°C) Top fresh water temeprature, Tfw,t (°C)
a) b)
Fig. 13. Effect of a) top seawater temperature b) top fresh water temperatures on the compressor pressure ratio.

software. EES is a numerical solver, using an iterative procedure for 5. Results and discussion
solving the system of equations. EES automatically identifies and
groups equations that are solved simultaneously. The convergence of The mathematical modeling of the heat pump assisted HDH system
the numerical solution is verified by two methods: (i) ‘Relative equation with direct contact dehumidifier was performed by solving a system of
residual’ which is the difference between the left-hand and right-hand non-liner equations for given working conditions. The unknown tem-
sides of an equation divided by the magnitude of the left-hand side of peratures of the cycle as well as distilled water production rate were
the equation; and (ii) ‘change in variables changing in each iteration. found. Then by knowing the terminal temperature difference in the
The calculations are converged if the relative equation residuals are less evaporator and the pinch temperature difference in condenser, com-
than a certain value (for example 10−6) or if the variable change is less pressor work and thermal loads of the heat pump are determined.
than 10−9. Both relative equation residuals and change in variables are Fig. 5 demonstrates the effect of mass flow rate ratio of seawater to
adjustable for desired precision. Besides, there are two stopping criteria dry air on recovery ratio. As shown, increasing the fresh water rate
consisting of (i) ‘number of iteration’ and (ii) ‘elapsed time’ that can be while the mass flow rate of seawater is constant will result in higher
set for obtaining results with higher accuracy. EES software is widely distilled water production rate. Also, this will shift the maximum value
used by the scientific community for thermodynamic system analysis of the recovery ratio to lower mass flow rate ratios of seawater to dry
[50,51]. air which practically means higher air flow rate at a constant seawater
flow rate. Considering seawater flow rate constant, increasing the fresh

101
S. Dehghani et al. Desalination 445 (2018) 95–104

600 Table 3
Tfw,t=10 ºC Comparison of the numerical model presented in this work with experimental
550 Tfw,t=20 ºC data from the literature.
Tfw,t=30 ºC
Ta,b (°C) Tfw,b (°C) Packing specification
SEEC (kW.hr/m )
3

500
mrd (–) mrd (–)
450
1.0 1.4 1.8 1.0 1.4 1.8 a (m2/m3) Hd (m)
400
Ref [23] 28 27 25 33 30 28 267 0.3
This study 28 25 23 33 31 29 226 0.3
350 Difference (%) 0 7 8 0 3 3 – –

300
1 1.5 2 2.5 3 3.5 4 4.5
saline water temperature will result in higher temperature and hu-
Mass flow rate ratio of seawater to air, mr h (-)
midity ratio of air leaving the humidifier. This allows provide higher
Fig. 14. Heating and cooling coefficient of performance of the heat pump under distillation rate in the dehumidifier.
system working condition of Tsw,t = 70 °C, Tfw,t = 30 °C and modified mrsw/ The effect of top freshwater temperature on recovery ratio at var-
fw = 1.06. ious mass flow rate ratios of seawater to air is presented in Fig. 7. Lower
top temperatures of fresh water result in higher distilled water pro-
1 duction rate. The maximum value of the recovery ratio grows from
nearly 5% to 7% while the temperature is decreased from 30 °C to 10 °C.
0.95
By decreasing the top freshwater temperature, the dehumidification
0.9 process will be improved as the heat transfer rate between water and air
increases which causes higher mass transfer rate between the two
Effectiveness, ε

0.85
streams. Therefore, having lower top freshwater temperature results in
0.8 lower humid air dew point which means that the air leaving the de-
0.75 humidifier has a lower humidity ratio.
Specific electrical energy consumption is defined to measure the
0.7 ε h (-) energy consumption of the system per cubic meter of water produced.
0.65 ε d (-) In Fig. 8, specific energy consumption of the system versus mass flow
rate ratio of seawater to dry air at different mass flow rate ratios of
0.6 seawater to fresh water is shown. By reducing the ratio of seawater to
1 1.5 2 2.5 3 3.5 4 4.5
mr h (-) or mr d (-) fresh water flow rate the minimum value of specific energy consump-
tion of the system decreases and is shifted to higher mass flow rate of
Fig. 15. Effectiveness of humidifier and dehumidifier in relation to mass flow air. This is because lower heating loads are required at lower mass flow
rate ratio of water to air at the working condition of Tsw,t = 70 °C, Tfw,t = 20 °C rate ratios of either seawater to dry air or seawater to freshwater.
and modified mrsw/fw = 1.05. In Fig. 9 the variation of SEEC versus mass flow rate ratio of sea-
water to air in the dehumidifier for different top temperature of sea-
70 water is depicted. As shown, increasing the top brine temperature re-
Tat sulted in higher optimum value of SEEC and this occurred at higher
60 mass flow rate ratio of seawater to dry air. This is because the higher
Tfwb
temperature of top saline water requires more compressor work, even
though distillation rate is higher.
50
It can be seen that the SEEC of the system based on the working
T (°C)

condition varies approximately between 260 and 370 kW h/m3.


40 Considering electivity cost of 0.05 $/kW h the water production cost
will be between 13 and 18.5 $/m3. Lawal et al. [35] reported the SEEC
Tbr of a water heated HDH system with a non-direct about 9 $/m3 while
30
Tab considering electricity cost of 0.05 $/kW h. The difference in SEEC is
mainly because of absence of energy recovery in the proposed HDHs
20 system. The specific energy consumption for a single effect MVC system
1 1.5 2 2.5 3 3.5 4 4.5
Mass flow rate ratio of seawater to air, mr h (-) without brine recirculation is reported [52] to be 9.8 kW h/m which
results in water production cost of about 5 $/m3.
Fig. 16. Temperature of the cycle versus mass flow rate ratio of seawater to dry Fig. 10 demonstrates the effect of top temperature of the fresh water
air of the humidifier at the working condition of Tsw,t = 70 °C, Tfw,t = 20 °C and on the SEEC of the system at various liquid to gas flow rate ratios.
modified mrsw/fw = 1.05. Reduction of fresh water temperature slightly increases the minimum
value of SEEC and shifts it to lower mass flow rate ratios of liquid to gas
water flow rate will provide more cooling load to the dehumidifier that in the humidifier. With lower top freshwater temperature the rate of
eventually results in higher distillation rate. Also, this will allow the distillation is higher, however larger compressor work is also needed
system to work at a higher circulation rate of air throughout the HDH which slightly increases the SEEC of the system.
system. Table 1 shows the value of maximum GOR of the system at the
Fig. 6 illustrates the effect of the seawater top temperature on the optimum mass flow rate ratios of seawater to dry air (mrh). Increasing
recovery ratio of the desalination system for various values of mass flow both the top seawater and freshwater temperatures increases the GOR
rate ratio of seawater to dry air of the humidifier. As shown, by in- of the HDH system as there will be more water production and conse-
creasing the seawater top temperature from 50 °C to 70 °C the max- quently more heat recovery. In addition, increasing the mass flow rate
imum value of recovery ratio rises from 3% to 6%. Increasing the top of the fresh water at a constant flow rate of seawater will improve the

102
S. Dehghani et al. Desalination 445 (2018) 95–104

system GOR. numerical model of this work. This is done for the working conditions
Fig. 11 demonstrates the cooling load variation of the HDH system of Ta,t = 37 °C, Tfw,t = 20 °C, φa,t = φa,b = 1 and Hd = 0.3 m with the
against mass flow rate ratio at the working condition of Tfw,t = 20 °C, same Merkel number. The comparison between this numerical model
Tsw,t = 70 °C and mrsw/fw = 1. In the graph, when the cooling load is and the work of Yi Li is shown in Table 3. A maximum difference of 8%
greater than zero, an extra cooler is needed to let the system work in can be seen for outlet air temperature at mrd = 1.8. It should be men-
optimum mass flow rate ratio (mrh) where the RR reaches its maximum. tioned that there is a 15% difference between the specific surface area
The red line shows the cooling load that needs to be provided by an of the packing fill in these studies.
extra cooler. To avoid the use of the extra cooler two different scenarios
can be considered. First is to let the system work at a mass flow rate 6. Conclusion
ratio of seawater to air that meets zero extra cooling load i.e. the range
on the graph below mrh of 2 and greater than mrh of 3.2. The second The performance study of the heat pump assisted HDH system with
scenario is to adjust the mass flow rate ratio of the seawater to fresh- direct contact dehumidifier is investigated utilizing the mathematical
water to meet this requirement. The second scenario is discussed in modeling. It can be concluded that when the system is working in a
Table 2. condition that yields higher water production rate, the corresponding
In this table, for given conditions of top seawater and freshwater SEEC of the system is higher as well. The optimum RR of approximately
temperatures (Tsw,t, Tfw,t) as well as mass flow rate ratios (mrsw/fw), 5.8% and SEEC of 335.4 kW·h/m3 can be achieved under system
optimum mrh is found. Then, the cooling load required for the dehu- working conditions of Tsw,t = 70 °C, Tfw,t = 20 °C and mrsw/fw = 1.05.
midifier at the given working condition is found and compared with the These results can be considerably improved under optimum different
required load of the evaporator. If the dehumidifier cooling load de- working conditions of the system. A fully coupled heat pump HDH cycle
mand is higher than the evaporator load supply then the heat pump is can be achieved only in some cases. For other working conditions, two
not fully coupled and an extra cooling load is needed. Otherwise, the methods can be put in place to ensure coupled behaviour and no ad-
heat pump is fully coupled and there is no need for extra cooling. ditional cooling demand. Firstly, by varying the mass flow rate ratio of
Moreover, the extra cooling load can be utilized for higher dehumidifier seawater to air and secondly by changing the mass flow rate ratio of
performance. In this table, positive values of cooling load represent the seawater to fresh water.
extra cooling load demand of the dehumidifier, while the negative
values are excess cooling load in the evaporator. Furthermore, by References
knowing the cooling load values at a specific working condition, a
modified mass flow rate ratio of seawater to fresh water can be pro- [1] L. García-Rodríguez, Seawater desalination driven by renewable energies: a review,
posed to meet either zero cooling demand or supply. A flow chart of the Desalination 143 (2) (2002) 103–113.
[2] K. Srithar, T. Rajaseenivasan, Performance analysis on a solar bubble column hu-
solution procedure is presented in Fig. 12. As an example, when the midification dehumidification desalination system, Process. Saf. Environ. Prot. 105
system is working at conditions of Tfw,t = 20 °C, Tsw,t = 70 °C and mrsw/ (2017) 41–50.
[3] W.F. He, X.K. Zhang, D. Han, L. Gao, Performance analysis of a water-power
fw = 1 it is not fully coupled and extra cooling load of 8.7 kJ/kg is re-
combined system with air-heated humidification dehumidification process, Energy
quired. By modifying the mass flow rate ratio of seawater to fresh water 130 (2017) 218–227.
to 1.06 the HDH system can be fully coupled with heat pump. [4] A. Giwa, N. Akther, A.A. Housani, S. Haris, S.W. Hasan, Recent advances in hu-
Fig. 13 depicts the variation of compressor pressure ratio against the midification dehumidification (HDH) desalination processes: improved designs and
productivity, Renew. Sust. Energ. Rev. 57 (2016) 929–944.
top saline and freshwater temperatures. By increasing the temperature [5] S. Dehghani, A. Date, A. Akbarzadeh, 7 - humidification-dehumidification desali-
difference in the low-temperature side and high-temperature side of the nation cycle A2, in: Veera Gnaneswar Gude (Ed.), Emerging Technologies for
heat pump the pressure ratio is rising. For having either higher tem- Sustainable Desalination Handbook, Butterworth-Heinemann, 2018, pp. 227–254.
[6] G.P. Narayan, M.H. Sharqawy, E.K. Summers, J.H. Lienhard, S.M. Zubair,
peratures in condenser or lower temperatures in evaporator greater
M.A. Antar, The potential of solar-driven humidification–dehumidification desali-
pressure ratio of the compressor is needed. nation for small-scale decentralized water production, Renew. Sust. Energ. Rev. 14
Fig. 14 shows the coefficient of performance of heating and cooling (4) (2010) 1187–1201.
at different seawater to air mass flow rate ratios. The COP of heating [7] A.E. Kabeel, T.A. Elmaaty, E.M.S. El-Said, Economic analysis of a small-scale hybrid
air HDH–SSF (humidification and dehumidification–water flashing evaporation)
stays constant as the heating load is considered to be fully supplied by desalination plant, Energy 53 (2013) 306–311.
the heat pump while the COP of cooling varies with mass flow rate ratio [8] M. Farid, A.W. Al-Hajaj, Solar desalination with a humidification-dehumidification
as the cooling load changes due to changes of the bottom fresh water cycle, Desalination 106 (1) (1996) 427–429.
[9] H. Müller-Holst, Solar thermal desalination using the multiple effect humidification
temperature. The highest COP of cooling corresponds to maximum RR (MEH)-method, in: L. Rizzuti, H.M. Ettouney, A. Cipollina (Eds.), Solar Desalination
of the HDH system at the given working condition. for the 21st Century: A Review of Modern Technologies and Researches on
Variation in the effectiveness of the humidifier and dehumidifier Desalination Coupled to Renewable Energies, Springer Netherlands, Dordrecht,
2007, pp. 215–225.
devices with mass flow rate ratio of water to air is shown in Fig. 15. As [10] G.P. Narayan, M.H. Sharqawy, V.J.H. Lienhard, S.M. Zubair, Thermodynamic
demonstrated, the humidifier effectiveness decreases from 1 to 0.75 analysis of humidification dehumidification desalination cycles, Desalin. Water
with increasing the mass flow rate ratio, while the effectiveness of Treat. 16 (1–3) (2010) 339–353.
[11] G. Yuan, Z. Wang, H. Li, X. Li, Experimental study of a solar desalination system
dehumidifier increases with the same variation of mass flow rate ratios. based on humidification–dehumidification process, Desalination 277 (1) (2011)
The maximum recovery ratio occurs at mass flow rate ratio of nearly 92–98.
2.5. At this point the dehumidifier effectiveness is larger than humidi- [12] A.E. Kabeel, M.H. Hamed, Z.M. Omara, S.W. Sharshir, Experimental study of a
humidification-dehumidification solar technique by natural and forced air circu-
fier effectiveness.
lation, Energy 68 (2014) 218–228.
In Fig. 16 variation of the temperatures of air, brine and fresh water [13] W.F. He, L.N. Xu, D. Han, Parametric analysis of an air-heated humidification-de-
with mass flow rate ratios are shown. Increasing the mass flow rate humidification (HDH) desalination system with waste heat recovery, Desalination
ratio causes the air temperature after humidification to rise and the air 398 (2016) 30–38.
[14] M.M. Farid, S. Parekh, J.R. Selman, S. Al-Hallaj, Solar desalination with a humi-
temperature after dehumidification to decrease. When the bottom dification-dehumidification cycle: mathematical modeling of the unit, Desalination
temperature of fresh water reaches a maximum, the highest distilled 151 (2) (2003) 153–164.
water production rate is achieved. [15] H. Müller-Holst, M. Engelhardt, M. Herve, W. Schölkopf, Solarthermal seawater
desalination systems for decentralised use, Renew. Energy 14 (1) (1998) 311–318.
To validate the mathematical model of the proposed HDH desali- [16] C. Yamalı, İ. Solmus, A solar desalination system using humidification–dehumidi-
nation system the dehumidifier studied by Yi Li [23] is considered. This fication process: experimental study and comparison with the theoretical results,
dehumidifier has the same height as the one studied here with various Desalination 220 (1) (2008) 538–551.
[17] A.E. Kabeel, E.M.S. El-Said, A hybrid solar desalination system of air humidifica-
water to air mass flow rate ratios. The outlet air temperature and outlet tion–dehumidification and water flashing evaporation: part I. A numerical
fresh water temperature measured by Yi Li [23] are used to validate the

103
S. Dehghani et al. Desalination 445 (2018) 95–104

investigation, Desalination 320 (2013) 56–72. Desalination 443 (2018) 11–25.


[18] N.A.S. Elminshawy, F.R. Siddiqui, M.F. Addas, Development of an active solar [36] L.A.L. Queiroz, Carvalho MdO, J.B. Laurindo, V. Silveira Júnior, Water purification
humidification-dehumidification (HDH) desalination system integrated with geo- system using a heat pump, Appl. Therm. Eng. 31 (16) (2011) 3354–3357.
thermal energy, Energy Convers. Manag. 126 (2016) 608–621. [37] H. Xu, Y. Zhao, T. Jia, Y.J. Dai, Experimental investigation on a solar assisted heat
[19] H.T.A. El-Dessouky, Humidification-dehumidification desalination process using pump desalination system with humidification-dehumidification, Desalination 437
waste heat from a gas turbine, Desalination 71 (1) (1989) 19–33. (2018) 89–99.
[20] D. Bharathan, B.K. Parsons, J.A. Althof, Institute SER, Direct-contact Condensers for [38] M.B. Shafii, H. Jafargholi, M. Faegh, Experimental investigation of heat recovery in
Open-cycle OTEC Applications: Model Validation With Fresh Water Experiments for a humidification-dehumidification desalination system via a heat pump,
Structured Packings, Solar Energy Research Institute, 1988. Desalination 437 (2018) 81–88.
[21] J.F. Klausner, Y. Li, R. Mei, Evaporative heat and mass transfer for the diffusion [39] G. Prakash Narayan, M.G. St. John, S.M. Zubair, J.H. Lienhard V, Thermal design of
driven desalination process, Heat Mass Transf. 42 (6) (2005) 528. the humidification dehumidification desalination system: an experimental in-
[22] Y. Li, J.F. Klausner, R. Mei, Performance characteristics of the diffusion driven vestigation, Int. J. Heat Mass Transf. 58 (1–2) (2013) 740–748.
desalination process, Desalination 196 (1) (2006) 188–209. [40] G.P. Narayan, K.H. Mistry, M.H. Sharqawy, S.M. Zubair, J.H. Lienhard, Energy ef-
[23] Y. Li, J.F. Klausner, R. Mei, J. Knight, Direct contact condensation in packed beds, fectiveness of simultaneous heat and mass exchange devices, Front. Heat Mass
Int. J. Heat Mass Transf. 49 (25–26) (2006) 4751–4761. Transfer 1 (2) (2010).
[24] F. Alnaimat, J.F. Klausner, R. Mei, Transient analysis of direct contact evaporation [41] F. Merkel, Verdunstungskühlung, VDI Forschungsarbeiten, 275, 1925 (Berlin).
and condensation within packed beds, Int. J. Heat Mass Transf. 54 (15) (2011) [42] H. Jaber, R.L. Webb, Design of cooling towers by the effectiveness-NTU method, J.
3381–3393. Heat Transf. 111 (4) (1989) 837–843.
[25] F. Alnaimat, J.F. Klausner, Solar diffusion driven desalination for decentralized [43] M.H. Sharqawy, M.A. Antar, S.M. Zubair, A.M. Elbashir, Optimum thermal design of
water production, Desalination 289 (2012) 35–44. humidification dehumidification desalination systems, Desalination 349 (2014)
[26] F. Alnaimat, J.F. Klausner, R. Mei, Transient dynamic response of solar diffusion 10–21.
driven desalination, Appl. Therm. Eng. 51 (1) (2013) 520–528. [44] J.C. Kloppers, D.G. Kröger, A critical investigation into the heat and mass transfer
[27] A. Eslamimanesh, M.S. Hatamipour, Mathematical modeling of a direct contact analysis of counterflow wet-cooling towers, Int. J. Heat Mass Transf. 48 (3) (2005)
humidification–dehumidification desalination process, Desalination 237 (1) (2009) 765–777.
296–304. [45] N.K. Nawayseh, M.M. Farid, A.A. Omar, A. Sabirin, Solar desalination based on
[28] A. Eslamimanesh, M.S. Hatamipour, Economical study of a small-scale direct con- humidification process—II. Computer simulation, Energy Convers. Manag. 40 (13)
tact humidification–dehumidification desalination plant, Desalination 250 (1) (1999) 1441–1461.
(2010) 203–207. [46] N/A, ASHRAE Handbook - Fundamentals (SI Edition), American Society of Heating,
[29] N. Niroomand, M. Zamen, M. Amidpour, Theoretical investigation of using a direct Refrigerating and Air-Conditioning Engineers, Inc., 2009.
contact dehumidifier in humidification–dehumidification desalination unit based [47] S.A. Klein, F.L. Alvarado, Engineering Equation Solver Version 8.400, http://www.
on an open air cycle, Desalin. Water Treat. 54 (2) (2015) 305–315. fchart.com/ees/.
[30] M. Mehrgoo, M. Amidpour, Constructal design and optimization of a direct contact [48] R.W. Hyland, A. Wexler, Formulations for the thermodynamic properties of the
humidification–dehumidification desalination unit, Desalination 293 (2012) 69–77. saturated phases of H2O from 173.15 K to 473.15 K, ASHRAE Trans. (Part 2A)
[31] H. Ettouney, Design and analysis of humidification dehumidification desalination (1983).
process, Desalination 183 (1) (2005) 341–352. [49] A. Pruss, W. Wagner, The IAPWS formulation 1995 for the thermodynamic prop-
[32] W.F. He, L. Huang, J.R. Xia, W.P. Zhu, D. Han, Y.K. Wu, Parametric analysis of a erties of ordinary water substance for general and scientific use, J. Phys. Chem. Ref.
humidification dehumidification desalination system using a direct-contact dehu- Data 31 (2) (2002) 387–535.
midifier, Int. J. Therm. Sci. 120 (2017) 31–40. [50] G. Prakash Narayan, V.J.H. Lienhard, S.M. Zubair, Entropy generation minimiza-
[33] W.F. He, D. Han, C. Ji, Investigation on humidification dehumidification desali- tion of combined heat and mass transfer devices, Int. J. Therm. Sci. 49 (10) (2010)
nation system coupled with heat pump, Desalination 436 (2018) 152–160. 2057–2066.
[34] D. Lawal, M. Antar, A. Khalifa, S. Zubair, F. Al-Sulaiman, Humidification-dehumi- [51] R. Zmeureanu, Wu X. Yu, Energy and exergy performance of residential heating
dification desalination system operated by a heat pump, Energy Convers. Manag. systems with separate mechanical ventilation, Energy 32 (3) (2007) 187–195.
161 (2018) 128–140. [52] M.A. Jamil, S.M. Zubair, On thermoeconomic analysis of a single-effect mechanical
[35] D.U. Lawal, S.M. Zubair, M.A. Antar, Exergo-economic analysis of humidification- vapor compression desalination system, Desalination 420 (2017) 292–307.
dehumidification (HDH) desalination systems driven by heat pump (HP),

104