i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5

Available online at www.sciencedirect.com

journal homepage: www.elsevier.com/locate/he

Energy and exergy analyses of hydrogen production via

solar-boosted ocean thermal energy conversion and PEM
electrolysis

Pouria Ahmadi*, Ibrahim Dincer, Marc A. Rosen

Faculty of Engineering and Applied Science, University of Ontario Institute of Technology (UOIT), Oshawa, Ontario, Canada L1G8B8

article info abstract

Article history: Energy and exergy analyses are reported of hydrogen production via an ocean thermal
Received 10 August 2012 energy conversion (OTEC) system coupled with a solar-enhanced proton exchange
Received in revised form membrane (PEM) electrolyzer. This system is composed of a turbine, an evaporator,
29 October 2012 a condenser, a pump, a solar collector and a PEM electrolyzer. Electricity is generated in the
Accepted 4 November 2012 turbine, which is used by the PEM electrolyzer to produce hydrogen. A simulation program
Available online 17 December 2012 using Matlab software is developed to model the PEM electrolyzer and OTEC system. The
simulation model for the PEM electrolyzer used in this study is validated with experimental
Keywords: data from the literature. The amount of hydrogen produced, the exergy destruction of each
Energy component and the overall system, and the exergy efficiency of the system are calculated.
Exergy To better understand the effect of various parameters on system performance, a para-
Hydrogen production metric analysis is carried out. The energy and exergy efficiencies of the integrated OTEC
OTEC system are 3.6% and 22.7% respectively, and the exergy efficiency of the PEM electrolyzer is
Water electrolysis about 56.5% while the amount of hydrogen produced by it is 1.2 kg/h.
Solar energy Crown Copyright 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All
rights reserved.

1. Introduction competitive. Water electrolysis is a mature technology for

Hydrogen as an energy carrier can facilitate sustainable
energy systems. The development of sustainable carbon-
neutral energy sources has become one of the most signifi-
cant issues in the world today. Hydrogen can be produced
from various energy sources using methods like biomass
conversion, steam methane reforming and water splitting.
Hydrogen can be produced in a relatively environmentally
benign manner (depending on the source of the input energy)
via splitting water by photocatalysis, thermochemical cycles
and electrolysis. Currently, both thermochemical and photo-
catalysis hydrogen production are not economically
large scale hydrogen production. Hydrogen production by
proton exchange membrane (PEM) electrolysis has numerous
advantages, such as low environmental impact and easy
maintenance.
A large amount of solar energy is stored as heat in the
surface waters of the worlds oceans, providing a source of
renewable energy. Ocean thermal energy conversion (OTEC) is
a process for harnessing this renewable energy in which
a heat engine operates between the relatively warm ocean
surface, which is exposed to the sun, and the colder (about
5
C) water deeper in the ocean, to produce electricity. OTEC
usually incorporates a low-temperature Rankine cycle engine

* Corresponding author. Tel.: 1 289 600 9505.

E-mail addresses: pouryaahmadi81@gmail.com, Pouria.Ahmadi@uoit.ca (P. Ahmadi), Ibrahim.Dincer@uoit.ca (I. Dincer), Marc.
Rosen@uoit.ca (M.A. Rosen).
0360-3199/$ e see front matter Crown Copyright 2012, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.ijhydene.2012.11.025
1796 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5
which boils a working fluid such as ammonia to generate
a vapour which turns the turbine to generate electricity, and
then is condensed back into a liquid in a continuous process.
Eighty percent of the energy the Earth receives from the sun is
stored in the Earths oceans [1,2], and many regions of the
world have access to this OTEC resource. OTEC can produce
fuels by using its product electricity to create hydrogen, which
can be used in hydrogen fueled cars as well as in the devel-
opment of synthetic fuels. For a small city, millions of tons of
CO2 are generated annually through fossil fuel use while with
OTEC the value is zero, during the operation of devices. OTEC
has a potential to replace some fossil fuel use, perhaps via
OTEC ships travelling the seas of the world.
An OTEC system utilizes low-grade energy and has low
energy efficiency (about 3e5%). Therefore, achieving a high
electricity generating capacity with OTEC requires the use of
large quantities of seawater and a correspondingly large
amount of pumping energy. These factors impact negatively
on the cost-effectiveness of this technology and OTEC is not
commercially viable today. To improve the effectiveness and
economics of OTEC cycles, it is proposed that they being
integrated with industrial operations so that, apart from
generating electricity, they could be used for fresh water
production, air conditioning and refrigeration, cold water
agriculture, aquaculture and mariculture, and hydrogen
production [1]. Potential markets for OTEC have been identi-
fied, most of which are in Pacific Ocean, and about 50 coun-
tries are examining its implementation as a sustainable
source of energy and fresh water, including India, Korea,
Palau, Philippines, the U.S. and Papua New Guinea [3]. In 2001,
as a result of cooperation between Japan and India a 1-MW
OTEC plant was built in India [3], and others are planned to
be constructed in the near future [4].
Considerable research has been directed to the develop-
ment of OTEC recently. Uehara [5e8] conducted numerous
theoretical and experimental studies on the major compo-
nents of an OTEC plant, and showed that ammonia is a suit-
able working fluid for an OTEC plant employing a closed
organic Rankine cycle (ORC). The energy efficiency of the
Rankine cycle in an OTEC plant is usually limited to around 5%
due to the small temperature difference between surface
water and deep water of the ocean. So, to improve the effi-
ciency of OTEC, other thermodynamic cycles like the Kalina
cycle and the Uehara cycle that use an ammoniaewater
mixture as the working fluid are being considered [9]; they are
reported to have better energy efficiencies rather than
a Rankine cycle at the same temperature difference [9].
Increasing in the temperature difference between the hot
heat source and the cold heat sink can improve the efficiency
of OTEC plants, as can the integration of OTEC with other
energy technologies. Saitoh and Yamada [10] proposed
a conceptual design of a multiple Rankine cycle system using
both solar thermal energy and ocean thermal energy in order
to improve the cycle efficiency. In addition, Owens [11] opti-
mized OTEC plants using ammonia as the working fluid in
closed cycles. Uehara and Ikegamia [6] performed an optimi-
zation study using the power method for a closed-cycle OTEC
plant. Later, Uehara and Ikegamia [12] studied the perfor-
mance of OTEC using the Kalina cycle and proposed a new
OTEC cycle with an absorption and extraction process [13].
Rabas et al. [14] studied a non-condensable gas removal
system for a particular 10-MW hybrid power plant. More
recently, Kazim [15] proposed a hydrogen production cycle
using ocean thermal energy, and Ikegamia et al. [7,16]
proposed desalination using an integrated hybrid cycle and
an open OTEC cycle.
Energy analysis, which is based on the first law of ther-
modynamics, often does not provide a clear picture of ther-
modynamic efficiency and losses. Exergy is not subject to
a conservation law, but is destroyed due to irreversibilities
during any process [17]. Exergy analysis overcomes such
deficiencies and can help improve efficiency and sustain-
ability [18]. Exergy is a useful tool for determining the location,
type and true magnitude of exergy losses, which appear in the
form of either exergy destruction or waste exergy emission
[19]. Therefore, exergy methods can assist in developing
strategies and guidelines for more effective use of energy
resources and technologies. Sun et al. [20] report an optimi-
zation design and exergy analysis of an organic Rankine cycle
in ocean thermal energy conversion, and show that ammonia
is a good working fluid for an ORC cycle in OTEC from a net
power output point of view. Also, from an exergy viewpoint, it
has been shown that a larger scale ORC in OTEC can justify the
use of a heat exchanger with a higher effectiveness to
decrease the exhaust loss, which accounts for a large
proportion of the exergy loss of an ORC in an OTEC system [20].
In this paper, a novel OTEC system equipped with a solar
collector and PEM electrolysis to produce hydrogen is ther-
modynamically modelled and assessed with energy and
exergy analyses. The primary objective is to improve under-
standing of this system. The following specific tasks are
performed:
 Model and simulate (using Matlab software) the integrated
system.
 Validate each part of the model and simulation.
 Perform energy and exergy analyses of the integrated
system to determine the exergy destruction rate and energy
and exergy efficiencies of each component and the entire
system.
 Determine the sustainability index for the system and
quantify its variation with exergy destruction, and examine
the relation between exergy efficiency, exergy destruction
and sustainability index.
 Conduct a comprehensive parametric study to determine
the effect of major design parameters on system
performance.
2. Modelling and energy analysis
For thermodynamic modelling purposes, the integrated OTEC
system for hydrogen production considered here (Fig. 1) is
divided into three parts: flat plate solar collector, ocean
thermal energy conversion (OTEC) unit and PEM electrolyzer.
Fig. 1 shows a schematic diagram of an integrated OTEC
system equipped with a flat plate solar collector and PEM
electrolyzer. This integrated system uses the warm surface
seawater to evaporate a working fluid like ammonia or a Freon
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5 1797

Fig. 1 e Schematic of ocean thermal energy conversion (OTEC) with a flat plate solar collector and PEM electrolyzer for
hydrogen production.

refrigerant, which drives a turbine to produce electricity, _ p

mC _ pg
1  efF Ul AP =mC
0

which in turn is used to drive a PEM electrolyzer to produce FR (3)

Ul AP
hydrogen. After passing through the turbine, the vapour is
condensed in a heat exchanger that is cooled by cold deep here, F0 is collector efficiency factor which is around 0.914 for
seawater. The working fluid is then pumped back through the this case [21] and Ul is the overall collector loss coefficient
warm seawater heat exchanger, and the cycle is repeated obtained from [21]. In Eq. (2), radiation flux absorbed by the
continuously. absorber is calculated as:
Energy and exergy analyses are used to determine the
temperature profile in the plant, input and output enthalpy S saI (4)
and exergy flows, exergy destructions rates and energy and
exergy efficiencies. The relevant energy balances and gov- where sa is optical efficiency and I is solar radiation inten-
erning equations for the main sections of the plant shown in sity. The energy efficiency of the solar flat plate collector is
Fig. 1 are described in the following subsections. expressed as:

2.1. Energy analysis Q_ u

h (5)
IAP
2.1.1. Flat plate solar collector
As shown in Fig. 1, water enters the solar collector at point 2 More information about solar flat plate collector modeling is
and is heated by the collector. The useful heat gained by the detailed in Ref. [22].
working fluid can be written as:
2.1.2. Organic rankine cycle (ORC)
Q_ u mC
_ p T3  T2 (1) As shown in Fig. 1, the electricity production unit is based on
an organic Rankine cycle which is suitable for low-grade heat.
where T3, T2, Cp and m _ are the water outlet temperature, inlet
Fig. 2 shows the corresponding temperatureeentropy (TeS )
temperature, specific heat at constant pressure and mass flow
diagram of the ORC.
rate. The HotteleWhillier equation for the heat gained by the
The net power output of the system is expressible as:
flat plate collector considering heat losses from the collector is
 
calculated as [21]: W _G W
_ net W _ WS W
_ CS W
_ WF (6)

Q_ u AP FR S  Ul Tin  T0  (2) _ WS and W

_ G is the turbine generator power, W _ CS are
where W
where T0 is the ambient temperature and the FR is heat the warm and cold seawater pumping power, and W _ WF is
removal factor which is defined as: working fluid working power.
1798 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5

Fig. 2 e TeS diagram of the ocean thermal energy conversion (OTEC) for hydrogen production.

2.1.2.1. Turbine generator power. The turbine power output is here, LE is the length of the evaporator plate and lE is taken
calculated by writing the energy balance for a control volume from Ref. [23]. Also, Deq is the equivalent diameter, which is
around a turbine which is the product of working fluid mass calculated as:
_ F and the adiabatic heat loss between the evapo-
flow rate, m
Deq 2d (14)
rator and the condenser, as follows:
where d is the clearance.
_Gm
W _ F hT hG h5  h6 (7)

here, hT and hG are the turbine isentropic efficiency and 2.1.2.3. Cold seawater pumping power. The cold seawater
generator mechanical efficiency. pumping power is expressed as:

_
2.1.2.2. Warm seawater pumping power. The warm seawater _ CS mCS DHCW g
W (15)
hCSP
pumping power can be written as:
where hCSP is the cold seawater pump efficiency and DHCW is
_ WS DHWS g
_ WS m
W (8) the total pump head of the cold seawater piping, given as:
hWSP

where DHWS is the total pump head difference of the warm DHCS DHCS P DHCS C DHCS d (16)
seawater piping:
here DHCS P is the pump head of the cold seawater pipe [23]:
DHWS DHWS P DHWS E (9)
DHCS P DHCS SP DHCS B (17)
here, DHWS P is the pump head of the warm seawater pipe,
These two terms are similar to terms in Eqs. (11) and (12). Also,
which can be given as [23]:
DHCS C is the cold seawater pressure difference in the
DHWS P DHWS SP DHWS B (10) condenser, defined as:

where DHWS SP is the friction loss of the straight pipe and 2

VCS L
DHCS C lC  EC (18)
DHWS B is the bending loss on the warm seawater pipe, which 2g Deq C
are given as [23]:
where LC is the length of the evaporator plate and lC is taken
 1:85
LWS VWS from reference [23]. Also, Deq is the equivalent diameter given
DHWS SP 6:82 1:17  ; CWS 100 (11)
dWS CWS in Eq. (14) and DHCS d is the pressure difference caused by the
density difference between the warm seawater surface and
X 2
VWS cold deeper seawater, calculated as [23]:
DHWS B lm (12)
2g  
1 1
DHCS d LCS  rWS rCS LCS (19)
here, LWS is the length of the warm seawater pipe, dWS is the rCS 2
warm seawater inner pipe diameter and VWS is the velocity of
warm seawater inside the pipe. Also, DHWS E is the pressure 2.1.3. PEM electrolyzer
difference of warm seawater in the evaporator, expressible as: The schematic diagram of the PEM electrolyzer for H2
production is shown on the right side of Fig. 1. During elec-
2
VWS L trolysis, the required electricity and heat are both supplied to
DHWS E lE  E (13)
2g Deq W the electrolyzer to drive the electrochemical reactions. As
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5 1799

shown in Fig. 1, liquid water is fed to the PEM electrolyzer at where x is the distance in the membrane measured from the
ambient temperature. Liquid water enters a heat exchanger cathode-membrane interface and lx is the water content at
that brings it to the PEM electrolyzer temperature and it then a location x in the membrane. The value of lx can be
enters the electrolyzer. Leaving the cathode, the H2 produced calculated in terms of the water content at the membrane-
dissipates heat to the environment and cools to the reference electrode edges:
environment temperature. The oxygen gas produced at the
la  lc
anode is separated from the water and oxygen mixture and lx x lc (26)
D
then cooled to the reference environment temperature. The
remaining water is then returned to the water supply stream where D is the membrane thickness, and la and lc are the
for the next hydrogen production cycle. The overall PEM water contents at the anode-membrane and the cathode-
electrolysis reaction is simply water splitting, i.e., electricity membrane interfaces, respectively. The overall ohmic resis-
and heat are used to separate water into produce hydrogen tance can thus be expressed as [25]:
and oxygen. Hydrogen is stored in a tank for later usage.
ZD
dx
RPEM (27)
2.1.3.1. Thermochemical modeling of a PEM electrolyzer. sPEM lx
0
Energy and exergy analyses of a PEM electrolyzer can be per-
formed in conjunction the electrochemical modeling. The Based on Ohms law, the following equation can be written
total energy needed by the electrolyzer can be obtained for the ohmic overpotential:
noting:
Vohm;PEM JRPEM (28)

DH DG TDS (20) The activation overpotential, Vact, caused by a deviation of

net current from its equilibrium, and also an electron transfer
where DG is Gibbs free energy and TDS is the thermal energy reaction, must be differentiated from the concentration of the
requirement. The values of G, S, and H for hydrogen, oxygen oxidized and reduced species [25,26]. Then,
and water can be obtained from thermodynamic tables [24].  
The total energy need is the theoretical energy required for RT 1 J
Vact;i sinh ; i a; c (29)
H2O electrolysis without any losses. The catalyst used in PEM F 2J0;i
electrolysis provides an alternative path for the reaction with here, J0 is the exchange current density, which is an important
lower activation energy. The mass flow rate of hydrogen is parameter in calculating the activation overpotential. It
determined by [25]: characterizes the electrodes capabilities in the electro-
chemical reaction. A high exchange current density implies
J
N_ H2 ;out N_ H2 O;reacted (21) a high reactivity of the electrode, which results in a lower
2F
overpotential. The exchange current density for electrolysis
where J is the current density and F is the Faraday constant.
can be expressed as [26]:
The electric energy input rate to the electrolyzer can be
expressed as:  
Eact;i
J0;i Jref
i exp  ; i a; c (30)
RT
Eelectric JV (22)
where Jref is the pre-exponential factor and Eact;i is the acti-
where Eelectric is the electric energy input and Exelectric the i
vation energy for the anode and cathode. More details about
electric exergy input. Also, V is given as:
PEM electrolysis modeling can be found elsewhere [26, 27].
V V0 Vact;a Vact;c Vohm (23)

where V0 is the reversible potential, which is related to the

difference in free energy between reactants and products, and
V0 can be obtained by the Nernst equation as follows:
V0 1:229  8:5  104 TPEM  298 (24)
Also, Vact;a , Vact;c and Vohm are the activation overpotential of
the anode, the activation overpotential of the cathode, and the
ohmic overpotential of the electrolyte, respectively. Ohmic
overpotential in the proton exchange membrane is caused by
the resistance of the membrane to the hydrogen ions trans-
porting through it. The ionic resistance of the membrane
depends on the degree of humidification and thickness of the
membrane as well as the membrane temperature [26]. The
local ionic conductivity sx of the proton exchange
membrane can be expressed as [25,26]:
   exergy is important in combustion evaluation. Applying the
1 1
sPEM lx 0:5139lx  0:326exp 1268  (25)
303 T
3. Exergy analysis
Exergy analysis can help to develop strategies and guidelines
for more effective use of energy, and has been applied to
various thermal processes, including power generation, CHP
and trigeneration. Exergy can be divided into four compo-
nents: physical, chemical, kinetic and potential. For the
processes involved in this study, the latter two are neglected
since changes in elevation and speed are negligible [28,29].
Physical exergy is defined as the maximum work obtainable as
a system interacts with a reference environment at an equi-
librium state [21,22]. Chemical exergy is associated with the
departure of the chemical composition of a system from the
chemical equilibrium of a reference environment. Chemical
first and the second laws of thermodynamics, the following
exergy balance for each component is obtained:
1800 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5

Table 1 e Input data for the system simulation.

Parameter Value Parameter Value

Turbine isentropic efficiency, hT 0.80 Warm seawater mass flow rate (kg/s) 150
Generator mechanical efficiency, hG 0.90 Cold seawater mass flow rate (kg/s) 150
Working fluid pump isentropic efficiency, hWFP 0.78 Cold seawater pipe length (m) 1000
Seawater pumps isentropic efficiency, hP 0.80 Cold seawater pipe inner diameter (m) 0.70
Ambient temperature (
C) 25 Warm seawater pipe length (m) 50
Solar radiation incident on collector surface, I (W/m2) 500 Warm seawater pipe length (m) 0.70
Warm seawater temperature, TWSI (
C) 22 Solar collector effective area (m2) 5000
Cold seawater temperature at depth of 1000 m, TCSI (
C) 4 Electrolyzer working temperature (
C) 80

$ X X $ $
ExQ _ i exi
m _ e exe ExW ExD
m (31) W_ net
h (38)
i e _
Q EVP
where the subscripts e and i denote the inlet and outlet of
$
_ net is given in Eq. (6), and Q_ EVP is expressible as
where W
a control volume, ExD is the exergy destruction and other
terms are as follows [28e31]: Q_ EVP m
_ WF h1  h4 (39)
 
$ T0 _
ExQ 1  Qi (32) 3.1.2. Exergy efficiency
Ti The exergy efficiency is defined as the product exergy output
divided by the exergy input [23]. According to Nag and Dai
$
_
Exw W (33) [34,35], the exergy efficiency of the ORC power generation
cycle in an OTEC system is given as:
ex exph exch (34)
_ net
W _ net
W
J $ $ $ (40)
exph h  h0  T0 s  s0 (35) Exin Exin;WS Exin;CS
$ $
Here, ExQ and Exw respectively are the corresponding exergy $
   
_ in;WS hWS;in  h0  T0 sWS;in  s0
Exin;WS m (41)
rates associated with heat transfer and work across the
boundary of a control volume, T is the absolute temperature
$
   
and the subscript o refers to the reference environment _ in;CS hCS;in  h0  T0 sCS;in  s0
Exin;CS m (42)
conditions. The reference environment considered in our
here, the reference environment state is taken to be
study is T0 20
C and P0 1 bar. The specific chemical exergy
P0 1.01 bar and T0 298.15 K.
for gas mixtures is defined as follows [28,29]:
X X
exch xk exkch RT0 xk lnxk (36)

Further details about each term in this equation are given in 4. Results and discussion
Refs. [28e31].
To improve environmental sustainability, it is necessary 4.1. Modelling validation and simulation code results
not only to use sustainable or renewable sources of energy,
but also to utilize non-renewable sources like natural gas To model the integrated OTEC system for hydrogen produc-
more efficiently, while minimizing environmental damage. In tion, a simulation code using Matlab software is developed.
this way, society can reduce its use of limited resources and
extend their lifetimes. Here, a sustainability index SI is used to
relate exergy with environmental impact [30,32,33]:

1 Table 2 e Input parameters used to model PEM

SI (37) electrolysis [26,27].
DP
Parameter Value
where DP is a depletion number, which is defined as the exergy
destruction/input exergy [21,29]. This relation demonstrates PO2 (atm) 1.0
how reducing a systems environmental impact can be ach- PH2 (atm) 1.0
ieved by reducing its exergy destruction. TPEM (
C) 80
Eact,a (kJ/mol) 76
Eact,c (kJ/mol) 18
3.1. Efficiencies la 14
lc 10
3.1.1. Energy efficiency D (mm) 100
2
Jref
a (A/m ) 1.7  105
The energy efficiency of the OTEC system is defined as the net 2
Jref
c (A/m ) 4.6  103
power output of the system divided by input energy at evap-
F (C/mol) 96,486
orator, which can be expressed as:
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5 1801

Table 3 e Parameter values resulting from energy and

exergy analyses of the system.
Parameters Value

Net power output, W _ net (kW) 101.96

Fig. 3 e Comparison of present model with experimental
data.
Three main parts are first individually modelled including the
each exergy flow rate. Engineering equation solver (EES) is
linked to Matlab to calculate the properties of the different
working fluids (i.e., water and ammonia) such as pressure,
temperature, enthalpy and entropy. Several simplifying
assumptions are made here to render the analysis more
tractable, while retaining adequate accuracy to illustrate the
principal points of the study
 All processes operate at steady state.
 The thermodynamic cycle of the integrated system in Fig. 1
is an ideal saturated Rankine cycle using pure ammonia as
the working fluid.
 All the components are adiabatic.
 Pressure drops in ORC cycle are negligible.
 State 5 is saturated vapour.
 Heat losses from piping and other auxiliary components are
negligible.
To conduct the simulation, input data are required. For
each subsystem some reliable data are input to the simulation
code in order to determine the outputs. Table 1 lists the input
parameters for the OTEC system simulation. Also, Table 2 lists
the parameter used to simulate the PEM electrolyzer.
0.7
Exergy efficiency, J (%) 22.70
Energy efficiency, h (%) 3.60
Sustainability index, SI 1.29
$
Total exergy destruction rate, ExD;tot (kW) 42.12
Hydrogen production rate, m _ H2 (kg/h) 1.20
PEM electrolyzer exergy efficiency, JPEM (%) 56.34
Warm surface pump power, W _ WS (kW) 1.30
Cold surface pump power, W _ CS (kW) 3.13
Working fluid pump power, W _ WF (kW) 0.88
To ensure the accuracy and validity of the developed
computer simulation code, the PEM electrolyzer is validated
with experimental data from the literature. Specifically, the
electrochemical model is used to simulate experiments pub-
lished in the literature and the modeling results and experi-
mental data are compared. The electrolyte used in the
experiments [36,37] is Nafion, a polymer widely used as
electrolyte in fuel cells and electrolyzers. The thicknesses of
the electrolytes tested by Ioroi et al. [36] and Millet et al. [37]
were 50 mm and 178 mm, respectively. Platinum was used as
the electrode catalyst. The simulation code for the JeV char-
acteristics for PEM electrolysis are compared with experi-
mental data of Ioroi et al. [36] as shown in Fig. 3. The modeling
results agree well with the experimental data, supporting the
accuracy of the present model. It is found that the cell
potential increases rapidly when current density is less than
300 A/m2. When J exceeds 300 A/m2, the cell potential
increases slightly with J. To enhance the understanding of the
electrochemical performance of the PEM electrolyzer, ohmic
and activation overpotentials are examined and shown
individually in Fig. 4. This figure shows that the ohmic over-
potential is very small and increases slightly with current
density. This observation is attributable to the fact that the
membrane ionic conductivity (sPEM) is significantly high for
typical values of l as well as the operating temperature,
which leads to a lower RPEM (see Eqs. (25) and (27)), which in
turn, lowers the overall ohmic resistance and the ohmic
overpotential.

0.6
Ohmic activation
0.5 Cathode activation
Overpotential(V)

Anode activation
0.4

0.3

0.2

0.1

0
0 1000 2000 3000 4000 5000 6000 7000 8000
Current density, J (A/m2)
Fig. 5 e Exergy destruction rate and dimensionless exergy
Fig. 4 e Variation of electrolyzer overpotentials at various destruction ratio for each component of the ocean thermal
current densities. energy conversion (OTEC) system for hydrogen production.
1802 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5

0.36 60 1.6 60

Total exergy destruction rate (kW)

Total exergy destruction rate (kW)
0.34 1.55
55 55

Sustainability index, SI
Exergy efficiency

0.32 1.5
50 50

0.3 1.45

45
45
1.4
0.28

40
40 1.35
0.26

1.3 35
0.24 35 400 440 480 520 560 600
400 440 480 520 560 600
Solar radiation intensity (W/m2)
Solar radiation intensity (W/m2)
Fig. 8 e Variation with solar radiation intensity of the
Fig. 6 e Variation with solar radiation intensity of the OTEC
sustainability index and total exergy destruction rate.
exergy efficiency and total exergy destruction rate.

4.2. Exergy analysis results destruction ratio are higher in the condenser than in other
components, suggesting that it would likely be worthwhile to
Results of the thermodynamic modelling and exergy analysis focus improvement efforts on this component.
are presented here, including assessments of the effects of Since the solar radiation intensity changes during the day,
varying several design parameters on the cycle performance. the variation of system performance is investigated with solar
As already discussed, the inputs of the simulation program radiation. Fig. 6 shows the variation of exergy efficiency and
are transferred to the developed code in order to calculate the exergy destruction rate of the OTEC system for various values
outputs. The results of the simulation program are listed in of solar radiation intensity. Increasing solar radiation inten-
Table 3. The net power output is seen to be around 101 kW, sity results in an increase in the exergy efficiency of the OTEC
which leads to a hydrogen production rate of about 1.2 kg/h. In system, suggesting that during such a day, the solar intensity
addition, the exergy efficiency of the integrated OTEC system increases the inlet temperature entering the evaporator,
is much higher than the energy efficiency, mainly due to the which leads to an increase in enthalpy at point 3 entering the
fact that the work is produced using a low-grade (in terms of turbine to produce work. The higher the temperature differ-
a temperature near to that of the reference environment) heat ence between cold and hot surfaces the more the work. Since
at the ocean surface. the cold surface temperature is almost constant at the depth
The exergy analysis results are presented in Fig. 5 and of the ocean, an increase in evaporator inlet temperature
show that the highest exergy destruction occurs in the results in an increase in the output power. Also, Fig. 6 shows
condenser, which is mainly due to the temperature difference that an increase in solar radiation leads to a decrease in total
between two fluid streams passing through it, but also due to exergy destruction.
the pressure drop across the device. Another important result The effect of solar radiation intensity on exergy destruction
relates to the dimensionless exergy ratio, which is useful for rate for each component is shown in Fig. 7. As shown in this
prioritizing exergy losses in an intuitive manner. Both the figure, an increase in solar radiation intensity has major
exergy destruction rate and the dimensionless exergy effects on the exergy destruction rates of turbine and
condenser while the changes in exergy destruction rate for the

130 2

mH2(kg/hr) Wnet(kW)
1.9
120
1.8
mH2 (kg/hr)

110
1.7
Wnet (kW)

1.6
100

1.5
90
1.4

80 1.3
400 440 480 520 560 600

Solar radiation intensity (W/m2)

Fig. 7 e Variation with solar radiation intensity of the Fig. 9 e Variation with solar radiation intensity of the OTEC
exergy destruction rate of components. net power output and hydrogen production rate.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5 1803

0.28 1.4 50
ex

Total exergy destruction rate (kW)

0.26 1.36
SI 46

Sustainability index (SI)

0.24
1.32
Exergy efficiency

0.22
42
1.28
0.2
1.24
0.18 38

1.2
0.16
34
0.14 1.16

0.12 1.12 30
280 284 288 292 296 300 280 284 288 292 296 300
T 0 (K) T0 (K)

Fig. 10 e Variation with ambient temperature of the exergy Fig. 12 e Variation with ambient temperature of the total
efficiency and sustainability index. exergy destruction rate.

evaporator and the pump are not drastic. So, since the sustainability index, due to an increase in the fluid tempera-
decrease in the exergy destruction rate of the condenser is ture entering the evaporator which increases the work of the
much higher than that for the turbine, the total exergy turbine. Above 285 K the exergy efficiency decreases because
destruction rate decreases with an increase in solar radiation the exergy input to the system, which is the denominator of
intensity. Fig. 8 shows the effect of solar radiation intensity on Eq. (33), increases, so two different effects occur and the effect
both sustainability index (SI) and total exergy destruction rate. of an increase in inlet exergy values dominates so the exergy
The results are similar to those in Fig. 6, i.e., the overall exergy efficiency of the OTEC system decreases. The variation of
destruction of the cycle decreases and the sustainability index exergy destruction rate of each component of the OTEC with
increases with increasing solar radiation intensity. The exergy ambient temperature is shown in Fig. 11, which demonstrates
efficiency, exergy destruction and sustainability are thus the same trend as Fig. 10. An increase in T0 has major effects on
observed to be linked in such systems, supporting the utility of the exergy destruction rate for the turbine and the condenser,
exergy and environmental analyses. while for the evaporator and working fluid pump it is almost
Fig. 9 shows the effect of solar radiation intensity on constant. An increase in temperature for the condenser first
hydrogen production mass flow rate. As the solar intensity leads to an increase in exergy destruction rate and then above
increases the hydrogen production increases since the turbine 285 K the exergy destruction rate decreases. However, the
work increases and correspondingly the electrical input to exergy destruction rate always increases for the turbine. The
PEM electrolysis increases. summation of these two effects results in an increase in the
Another major parameter in an OTEC plant is ambient total exergy destruction rate first then it leads to a decrease in
temperature because it affects the surface temperature and the total exergy destruction as shown in Fig. 12.
also the value for inlet exergy at each point of the plant. Hence, To enhance the analysis, the effect of condenser temper-
this temperature is an important parameter. Fig. 10 shows the ature on the exergy efficiency of the OTEC system is also
variation of exergy efficiency and sustainability index with investigated (see Fig. 13). An increase in condenser tempera-
ambient temperature. As can be seen, an increase in ambient ture is seen to reduce the exergy efficiency of the system.
temperature first leads to an increase in exergy efficiency and Increasing condenser temperature leads to an increase in
enthalpy at point 2 and results in a reduction in turbine work,
30
Turbine Condenser
0.32 60
Exergy destruction rate (kW)

Evaporator Working fluid pump

Total exergy destruction rate (kW)

25
50
0.28
20
Exergy efficiency

40
0.24
15
30

0.2
10 20

0.16
5 10

0 0.12 0
280 284 288 292 296 300 5 7 9 11 13 15
T0 (K) Condenser temperature (0C)

Fig. 11 e Variation with ambient temperature of the exergy Fig. 13 e Variation with condenser temperature of the total
destruction rate of each component. exergy efficiency and exergy destruction rate.
1804 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5

which lowers the exergy efficiency. Since the exergy J0 exchange current density (A/m2)
destruction rate in the condenser is much higher than in other Jref
i pre-exponential factor (A/m2)
components, any changes in the input exergy flow to the L length (m)
condenser leads to an increase in its exergy destruction, so an m_ mass flow rate (kg/s)
increase in condenser temperature leads to an increase in N_ molar mass flow rate (mol/s)
exergy flow at point 2 keeping other parameters constant. P pressure (kPa)
Q_ heat transfer rate (kW)
R gas constant (kJ/kg K)
5. Conclusions RPEM proton exchange membrane resistance (U)
s specific entropy (kJ/kg K)
The comprehensive thermodynamic modeling and exergy S radiation flux (W/m2)
analysis of an integrated OTEC system for hydrogen produc- T temperature (
C or K)
tion has provided useful insights. A complete thermodynamic V velocity (m/s)
simulation is performed for an OTEC cycle equipped with a flat V0 reversible potential (V)
plate solar collector and PEM electrolysis for hydrogen Vact activation overpotential (V)
production. Also, an exergy analysis is conducted and the Vact;a anode activation overpotential (V)
results show that the exergy efficiency of the integrated OTEC Vact;c cathode activation overpotential (V)
W_ work rate (kW)
system is about 22%. Other significant conclusions follow:
W_G turbine generator power (kW)
W_ CS cold seawater pumping power (kW)
 An increase in solar radiation intensity decreases the total
W_ WF working fluid power (kW)
exergy destruction rate of the system and raises the exergy
W_ WS warm seawater pumping power (kW)
efficiency of the cycle and the hydrogen production rate.
 An increase in solar intensity results in a drastic reduction
in condenser exergy destruction rate, which ultimately Greek letters
leads to a reduction in total exergy destruction rate of the h energy efficiency
cycle. hT turbine efficiency
 An increase in ambient temperature has two different hG generator efficiency
effects on exergy efficiency: an increase in both exergy hWFP working fluid pump efficiency
efficiency and sustainability index below T0 292 K, and hCSP cold surface pump efficiency
a reduction of both exergy efficiency and SI above T0 292 K. hWSP warm surface pump efficiency
 Increasing the condenser temperature reduces the exergy la water content at the anode-membrane interface
efficiency of the cycle because this increase raises the (U1)
enthalpy of the turbine outlet which reduces the net power lc water content at the cathode-membrane interface
output. (U1)
lx water content at location x in the membrane (U1)
r density (kg/m3)
sPEM proton conductivity in PEM (s/m)
Acknowledgement sx local ionic PEM conductivity (s/m)
J exergy efficiency
The authors acknowledge the support provided by the Natural
Sciences and Engineering Research Council of Canada. Subscripts
act activation
Nomenclature G generator
CS cold surface
C condenser
A surface area (m2) D destruction
CP specific heat at constant pressure (kJ/kg K) E evaporator
d diameter (m) eq equivalent
Deq equivalent diameter (m) in inlet condition
DP depletion number net net power
Eact;i activation energy in cathode or anode (kJ) ohm ohmic
$
Ex exergy flow rate (kW) out outlet condition
$
ExD exergy destruction rate (kW) PEM polymer exchange membrane
$
Exelectric electric exergy input rate (kW) T turbine
F Faraday constant (C/mol) p pump
g gravitational acceleration (m/s2) WF working fluid
G Gibbs free energy (J/mol) WS warm surface
h specific enthalpy (kJ/kg)
I solar radiation intensity (W/m2) Superscripts
J current density (A/m2) $ rate
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 3 8 ( 2 0 1 3 ) 1 7 9 5 e1 8 0 5
references
[1] Tchanche BF, Lambrinos GR, Frangoudakis A, Papadakis G.
Low-grade heat conversion into power using organic rankine
cycles: a review of various applications. Renewable and
Sustainable Energy Reviews 2011;15:3963e79.
[2] Faizal M, Rafiuddin Ahmed M. On the ocean heat budget and
ocean thermal energy conversion. International Journal of
Energy Research 2011;35:1119e44.
[3] Meegahapola L, Udawatta L, Witharana S. The ocean thermal
energy conversion strategies and analysis of current
challenges. In: Proceeding of second international
conference on industrial and information systems (ICIIS
2007), Sri Lanka; 2007.
[4] Esteban M, Leary D. Current developments and future
prospects of offshore wind and ocean energy. Applied Energy
2012;90:128e36.
[5] Uehara H, Nakaoka T. OTEC using plate-type heat exchanger
(using ammonia as working fluid). Transactions of JSME
1984;50(453):1325e33.
[6] Uehara H, Ikegami Y. Optimization of a closed-cycle OTEC
system. Journal of Solar Energy Engineering 1990;112:247e56.
[7] Uehara H, Miyara A, Ikegami Y, Nakaoka T. Performance
analysis of an OTEC plant and a desalination plant using an
integrated hybrid cycle. Journal of Solar Energy Engineering
1996;118:115e22.
[8] Uehara H, Dilao CO, Nakaoka T. Conceptual design of ocean
thermal energy conversion power plants in the Philippines.
Solar Energy 1998;41(5):431e41.
[9] Yamada N, Hoshi A, Ikegami Y. Performance simulation of
ocean thermal energy conversion plant. Renewable Energy
2009:1752e8.
[10] Saitoh TS, Yamada N. Advanced multiple Rankine cycle
system including Uehara cycle for solar and ocean energy
utilization. In: Proceedings of forum on desalination using
renewable energy, Republic of Palau; 2003. p. 167e75.
[11] Owens WL. Optimization of closed-cycle OTEC plants.
ASMEeJSME Thermal Engineering Joint Conference 1980;2:
227e39.
[12] Uehara H, Ikegami Y, Parametric performance analysis of
OTEC using Kalina cycle. ASME joint solar engineering
conference; 1993. p. 203e7.
[13] Uehara U, Ikegami Y, Nishida T. Performance analysis of
OTEC system using a cycle with absorption and extraction
processes. Transactions of JSME 1998;64(624):384e9.
[14] Rabas TJ, Panchal CB, Stevens HC. Integration and
optimization of the gas removal system for hybrid-cycle
OTEC power plants. Journal of Solar Energy Engineering 1990;
112:19e28.
[15] Kazim A. Hydrogen production through an ocean thermal
energy conversion system operating at an optimum
temperature drop. Applied Thermal Engineering 2005;25:
2236e46.
[16] Ikegamia Y, Sasakib H, Goudab T, Uehara H. Experimental
study on a spray flash desalination. Desalination 2006;194:
81e9.
[17] Rosen MA, Dincer I. On exergy and environmental impact.
International Journal of Energy Research 1997;21:643e54.
1805
[18] Kanoglu M, Dincer I, Rosen MA. Understanding energy and
exergy efficiencies for improved energy management in
power plants. Energy Policy 2007;35:3967e78.
[19] Dincer I, Rosen MA. Exergy, energy, environment and
sustainable development. UK: Elsevier; 2007.
[20] Sun F, Ikegami Y, Jia B, Arima H. Optimization design and
exergy analysis of organic rankine cycle in ocean thermal
energy conversion. Applied Ocean Research 2012;35:38e46.
[21] Farahat S, Sarhaddi F, Ajam H. Exergetic optimization of flat
plate solar collectors. Renewable Energy 2009;34:1169e74.
[22] Sukhatme SP. Solar Energy. New York: McGraw-Hill; 1993.
[23] Nihous GC, Vega LA. Design of a 100 MW OTEC-hydrogen
plant ship. Marine Structure 1993;6:207e21.
[24] Cengel YA, Boles MA. In: Thermodynamics: an engineering
approach. 7th ed. McGraw-Hill; 2011.
[25] Esmaili P, Dincer I, Naterer GF. Energy and exergy analyses of
electrolytic hydrogen production with molybdenum-oxo
catalysts. International Journal of Hydrogen Energy 2012;
37(9):7365e72.
[26] Ni M, Leung MK, Leung DYC. Energy and exergy analysis of
hydrogen production by a proton exchange membrane (PEM)
electrolyzer plant. Energy Conversion and Management
2008;49:2748e56.
[27] Nieminem J, Dincer I, Naterer G. Comparative performance
analysis of PEM and solid oxide steam electrolysers.
International Journal of Hydrogen Energy
2010:10,842e10,850.
[28] Ahmadi P, Dincer I. Thermodynamic and
exergoenvironmental analyses, and multi-objective
optimization of a gas turbine power plant. Applied Thermal
Engineering 2011;31:2529e40.
[29] Ahmadi P, Dincer I, Rosen MA. Exergo-environmental analysis
of an integrated organic Rankine cycle for trigeneration.
Energy Conversion and Management 2012;64:447e53.
[30] Dincer I, Hussain MM, Al-Zaharnah I. Energy and exergy use
in the industrial sector of Saudi Arabia. Proceedings of the
Institution of Mechanical Engineers, Part A: Journal of Power
and Energy 2003;217(5):481e92.
[31] Bejan A, Tsatsaronis G, Moran M. Thermal design and
optimization. New York: Wiley; 1996.
[32] Ahmadi P, Rosen MA, Dincer I. Multi-objective exergy-based
optimization of a polygeneration energy system using an
evolutionary algorithm. Energy 2012;46:21e31.
[33] Ahmadi P, Rosen MA, Dincer I. Greenhouse gas emission and
exergo-environmental analyses of a trigeneration energy
system. International Journal of Greenhouse Gas Control
2011;5:1540e9.
[34] Nag PK, Gupta A. Exergy analysis of the Kalina cycle. Applied
Thermal Engineering 1998;18(6):427e39.
[35] Wang JF, Dai YP, Gao L. Exergy analyses and parametric
optimizations for different cogeneration power plants in
cement industry. Applied Thermal Engineering 2009;86(6):
941e8.
[36] Ioroi T, Yasuda K, Siroma Z, Fujiwara N, Miyazaki Y. Thin
film electrocatalyst layer for utilized regenerative polymer
electrolyte fuel cells. Journal of Power Sources 2002;112(2):
583e7.
[37] Millet P, Andolfatto F, Durand R. Design and performance of
a solid polymer electrolyte water electrolyzer. International
Journal of Hydrogen Energy 1996;21(2):87e93.