Design of systems for hydrogen production based on the Cu–Cl thermochemical water decomposition...

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 en e r g y 3 6 ( 2 0 1 1 ) 1 1 3 0 9e1 1 3 2 0

Avai lab le a t www.sc iencedi rec t .com

journa l homepage : www.e lsev ier . com/ loca te /he

Design of systems for hydrogen production basedon the CueCl thermochemical water decompositioncycle: Configurations and performance

Mehmet F. Orhan, Ibrahim Dincer, Marc A. Rosen*

Faculty of Engineering and Applied Science, University of Ontario Institute of Technology, 2000 Simcoe Street North,

Oshawa, Ontario L1H 7K4, Canada

a r t i c l e i n f o

Article history:

Received 31 August 2010

Received in revised form

4 February 2011

Accepted 7 February 2011

Available online 1 April 2011

Keywords:

Hydrogen production

Thermochemical water

decomposition

Nuclear

Renewable energy

Thermodynamic analysis

Copperechlorine cycle

* Corresponding author.E-mail addresses: [email protected]

0360-3199/$ e see front matter Copyright ªdoi:10.1016/j.ijhydene.2011.02.034

a b s t r a c t

In this study, we analyze several Cu-Cl cycles by examining various design schemes for an

overall system and its components, in order to identify potential performance improve-

ments. The factors that determine the number and effective grouping of steps for new

design schemes are analyzed. A thermodynamic analysis and several parametric studies

are presented for various configurations. The energy efficiency is found to be 44% for the

five-step thermochemical process, 43% for the four-step process and 41% for the three-step

process, based on the lower heating value of hydrogen. Also, conclusions regarding

implementation of these new configurations are discussed and the potential benefits

ascertained.

Copyright ª 2011, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

1. Introduction All devices incur irreversibilities. Systems or devices

Concerns regarding global climate change are significant and

have resulted in extensive R&D on alternative, clean

energy sources.Whilemany of the available renewable energy

resources are limited due to their reliability, quality, quantity

and density, nuclear energy has the potential to contribute

a significant share of energy supply without contributing

significantly to climate change. Hydrogen production via

thermochemical water decomposition is a potential process

for direct utilization of nuclear thermal energy to increase

efficiency and thereby facilitate energy savings.

(M.F. Orhan), ibrahim.di2011, Hydrogen Energy P

designed to produce work via chemical reaction, such as

thermochemical cycles and power plants, invariably are also

subject to other losses associated with their operation.

Efficient use of energy is a significant contributor to any

sustainable plan for meeting growing energy demands. The

main objective of a plant designer is to determine the optimal

plant configuration and operating conditions to satisfy

requirements according to specified constraints (environ-

mental and otherwise). As one part of pre-commercialization

optimization, it is essential to evaluate alternative/developed

design schemes in terms of their energy and exergy

[email protected] (I. Dincer), [email protected] (M.A. Rosen).ublications, LLC. Published by Elsevier Ltd. All rights reserved.

mailto:[email protected]



http://www.sciencedirect.com

http://www.elsevier.com/locate/he

http://dx.doi.org/10.1016/j.ijhydene.2011.02.034



Fig. 1 e The CueCl thermochemical cycle for hydrogen

production.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 6 ( 2 0 1 1 ) 1 1 3 0 9e1 1 3 2 011310

efficiencies. The overall efficiencies of alternative design

schemes depend on parameters like operating temperature,

conversion efficiencies of subprocesses, and complexity of the

plant and its subsystems.

Thermochemical water splitting with a copperechlorine

(CueCl) cycle is a promising process that could be linked with

nuclear reactors to decompose water into its constituents,

oxygen and hydrogen, through intermediate copper and chlo-

rine compounds. The CueCl cycle is a hypothetical process

that has not yet been constructed. However, many studies of

the CueCl cycle are available in the literature. For example, Al-

Dabbagh and Lu have studied the design and reliability of

control systems for a CueCl thermochemical hydrogen

productionplant [1]. Equilibriumconversion in theCueCl cycle

multiphase processes has been studied by Daggupati et al. [2].

They perform a thermodynamic equilibrium analysis of indi-

vidual steps within the cycle. Recently, they studied solid

particle decomposition and hydrolysis reaction kinetics in

CueCl [3], andasolid conversionprocessduringhydrolysis and

decomposition of cupric chloride in the CueCl cycle [4]. One of

themost challenging steps in the thermochemical CueCl cycle

is the hydrolysis of CuCl2 into Cu2OCl2 and HCl while avoiding

the need for excess water and the undesired thermolysis

reaction, which yields CuCl and Cl2. Argonne National Labo-

ratory has designed a spray reactorwhere an aqueous solution

of CuCl2 is atomized into a heated zone, into which steam/Ar

are injected in co- or counter-current flow [5]. Also, an experi-

mental studyusing a spray reactorwith anultrasonic atomizer

has been carried out [6]. Jaber et al. [7] have studied heat

recovery from molten CuCl in the CueCl cycle. In that study,

they examined the convective heat transfer between molten

CuCl droplets and air in a counter-current spray flow heat

exchanger [7]. Ceramic carbon electrode-based anodes for use

in the CueCl cycle have been studied by Ranganathan and

Easton [8,9]; CCE materials, prepared using 3-aminopropyl tri-

methoxysilane, are investigated in that study. Wang et al. [10]

have compared sulphureiodine and copperechlorine ther-

mochemical hydrogen production cycles from the perspec-

tives of heat quantity, heat grade, thermal efficiency, related

engineering challenges, and hydrogen production cost [10].

Thermophysical properties of copper compounds in the CueCl

cycle have been studied by Zamfirescu et al. [11]. Furthermore,

the kinetics of the copper/hydrochloric acid reaction in the

CueCl cycle have been studied [12].

Although technical studies of the CueCl cycle have been

reported, there is a need for further developments and

improved design schemes to facilitate eventual commerciali-

zation. At present, there are different variations of the CueCl

cycle, which consist of different intermediate steps. The

numbers of steps and the methods of grouping have influ-

ences on the scale-up challenges and overall cycle efficiencies.

Therefore, it is important to compare the advantages and

disadvantages of different configurations of CueCl cycles. In

this study, we analyze the CueCl cycle by examining various

design schemes for an overall system and its components, in

order to identify potential performance improvements.

Several variations of the CueCl cycles with different numbers

of steps andmethods of grouping them are compared, and the

factors that determine the more effective number and

grouping of steps are analyzed. The main characteristics of

the cycles with different steps and numbers of steps are dis-

cussed. Also, conclusions regarding implementation of these

new configurations are presented and the potential benefits

ascertained. Finally, a thermodynamic analysis and several

parametric studies are presented for various configurations.

2. Description of systems

The CueCl cycle (Fig. 1) consists of a set of reactions to achieve

the overall splitting of water into its constituents, hydrogen

and oxygen. The overall net reaction is H2O(g)/H2(g)þ 1/

2O2(g). The CueCl cycle decomposes water into hydrogen and

oxygen through intermediate copper and chloride

compounds. These chemical reactions form a closed internal

loop that recycles all chemicals on a continuous basis, without

emitting any greenhouse gases.

Many studies (e.g. [1e13]) of the CueCl cycle have shown

that it offers a potentially attractive option for generating

hydrogen from nuclear energy. Compared with other

hydrogen production options, the thermochemical CueCl

cycle is expected to have a higher efficiency, to produce

hydrogen at a lower cost, and to have a smaller impact on the

environment by reducing airborne emissions, solid wastes

and energy requirements.

3. Analysis of configurations

The concept of thermochemical production of hydrogen from

water was first studied thermodynamically in the 1960s. The

decomposition, by the use of thermal energy, is limited by the

Carnot’s theorem. At standard temperature and pressure the

free energy and enthalpy changes for the direct splitting of

water are DG¼ 56.7 kcal/gmol and DH¼ 68.3 kCal/gmol,

respectively. It is not until around 4400 �C that the DG for the

reaction becomes negative. At such an extreme temperature

there are great problems with materials and separations

rendering the direct decomposition infeasible. The work

required for the one-step process can be reduced by increasing

the operating temperature; however, if anupper temperature of

1100 �C is imposed the reduction is modest. In a multi-step



Table 1 e Reactions in the three-step CueCl cycle.

Step Name Reaction Temperature range

S-i Hydrolysis 2CuCl2(aq)þH2O(g)

/ 2CuCl(l)þ 2HCl(g)

þ 1/2O2(g)þH2O(g)

400e600 �C

S-ii Electrolysis 4CuCl(s)þH2O(l)

/ 2Cu(s)þ 2CuCl2(aq)

20e80 �C

S-iii H2

production

2Cu(s)þ 2HCl(g)

/ 2CuCl(l)þH2(g)

430e475 �C

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 en e r g y 3 6 ( 2 0 1 1 ) 1 1 3 0 9e1 1 3 2 0 11311

process it is, in theory, possible to reduce thework requirement

to zero by operating reactions with positive entropy changes at

high temperatures and reactions with negative entropy

changes at low temperatures. Considering a two-step cycle it is

possible to calculate ideal values of DH and DS in that cycle.

Comparing thosevalueswith existing chemical products shows

that no two-step cycle is possible within an 1100 �C limit.

Consequently all possible cycles will have three or more steps.

Three different variations of the CueCl cycle are currently

under investigation: 3-step, 4-step and 5-step cycles [13].

3.1. Three-step CueCl cycle

The CueCl cycle illustrated in Fig. 2 consists of the threemajor

reactions shown inTable 1. It canbe found that step i is roughly

equivalent to the grouping of steps 1, 2 and 4 of the five-step

cycle, or the combinationof steps I and II of the four-step cycle.

3.2. Four-step CueCl cycle

A four-step CueCl cycle is shown in Fig. 3, with the main

reactions described in Table 4. It can be found from Fig. 3 that

P1

S-

S-

S-iii

HE

HE7

H2O

INPUT

HCl (g)

CuCl(l)

CuCl(s)

Cu(s)

P2

H2 (20ºC )

OUT

CuCl(l)

430ºC -475ºC

430ºC -475ºC

400ºC

HE1

HCI(g) production,Fluidized bed

Steam (4

HE6

Heat recoveryheat

H2

H2 production

Fig. 2 e Conceptual flow chart

step I is roughly equivalent to the combination of steps 1 and 4

in the five-step cycle. As shown below, in the case of 5-step

cycle, aqueous cupric chloride is first dried into a solid product

of cupric chloride particles, and then fed to the hydrolysis

reactor to produce copper oxychloride. The 4-step cycle

combines these processes by supplying aqueous cupric chlo-

ride into the hydrolysis chamber, such as spraying the

solution with co-flowing steam to produce the same copper

oxychloride product (see Table 2). The 4-step process has

an advantage of reducing complexity by eliminating

solids handling and thus requiring less equipment. However,

i

ii

3

water

CuCl2+ Water

O2 (20ºC )

PUT

400ºC

20ºC

20ºC20ºC

HE2

00ºC )O2

HE4

400ºC

500ºC

heat

Cu production

HE: Heat Exchanger

S: Step

P: Pump

HE5Heat recovery

400ºC

20ºC

of three-step CueCl cycle.



P1

S-I

HE3

S-II

S-III

S-IV

HE4

HE8

H2O

INPUT

Cu2OCl2(s)

HCl (g)

CuCl(l)

CuCl(s)

water

CuCl2+ Water

Cu(s)

P2

H2 (20ºC ) O2 (20ºC )

OUTPUT

500ºC

400ºC

500ºC

20ºC

20ºC

CuCl(s)

430ºC -475ºC

20ºC

430ºC -475ºC

400ºC

HE1

HE2


Steam (400ºC )

O2 production

HE5

400ºC

HE7

500ºC

Heat recoveryheat

heat

H2

Cu production

H2 production

HE: Heat Exchanger

S: Step

P: Pump

HE6Heat recovery

400ºC

20ºC

Fig. 3 e Conceptual flow chart of four-step CueCl cycle.

Table 2 e Reactions in the four-step CueCl cycle.

Step Name Reaction Conditions

S-I Hydrolysis 2CuCl2(aq)þ 2H2O(g)

/Cu2OCl2(s)

þ 2HCl(g)þH2O(g)

375e400 �C

S-II O2 production Cu2OCl2(s)/ 1/2O2(g)

þ 2CuCl(l)

500e530 �C

S-III Electrolyzer 4CuCl(s)þH2O(l)

/ 2Cu(s)þ 2CuCl2(aq)

20e80 �C

S-IV H2 production 2Cu(s)þ 2HCl(g)

/ 2CuCl(l)þH2(g)

430e475 �C


the 5-step process may be advantageous from the viewpoint

of energy and exergy efficiencies, because lower grade heat

can be used to remove the water in the drying process, rather

than high temperature heat in the hydrolysis reactor for the

unnecessary latent heat of vaporization of water.

3.3. Five-step CueCl cycle

A conceptual layout of a five-step CueCl cycle is illustrated in

Fig. 4. This case includes three thermochemical reactions and

one electrochemical reaction as shown in Table 3.

The cycle involves five steps:

1. HCl(g) production, using such equipment as a fluidized bed,

2. oxygen production,

3. copper (Cu) production,

4. drying, and

5. hydrogen production.

A chemical reaction takes place in each step, except drying

(see Table 3).

As illustrated in Fig. 4, only water and nuclear-derived heat

enter the cycle and only H2 and O2 are produced (with no

greenhouse gas emissions). Liquid water at ambient temper-

ature enters the cycle and passes through several heat

exchangers where it evaporates and increases in temperature

to 400 �C. Heat for this process is obtained from cooling the

hydrogen and oxygen gases before they exit the cycle. Steam



P1

S1

HE3

S2

S3

S5

HE4

HE8

H2O

S4

HE6

INPUT

Cu2OCl2(s)

HCl (g)

CuCl(l)

CuCl(s)

waterCuCl2

+ Water

Cu(s)

P2

CuCl2(s)

CuCl2(s) 400ºC

H2 (20ºC ) O2 (20ºC )

OUTPUT

500ºC

400ºC

500ºC

20ºC

20ºC

150ºC

CuCl(s)

430ºC -475ºC

20ºC

430ºC -475ºC

150ºC

400ºC

HE1

HE2


Steam (400ºC )

O2 production

HE5

150ºC

Flash dryer

HE7

500ºC

Heat recoveryheat

heat

heat

H2

Cu production

H2 production

HE: Heat Exchanger

S: Step

P: Pump

Fig. 4 e Conceptual flow chart of five-step CueCl cycle.


at 400 �C and solid copper chloride (CuCl2) at 400 �C from the

dryer enter the fluidized bed (S1), where a chemical reaction

occurs. The chemical formulation for this reaction is

2CuCl2(s)þH2O(g)/CuO * CuCl2(s)þ 2HCl(g) (1)

Table 3 e The five steps in the CueCl cycle with theircorresponding reactions.

Step Name Reaction Temperaturerange

S1 Hydrolysis 2CuCl2(s)þH2O(g)/Cu2OCl2(s)

þ 2HCl(g)

375e400 �C

S2 O2

production

Cu2OCl2(s)

/ 2CuCl(l)þ 1/2O2(g)

500e530 �C

S3 Electrolyzer 4CuCl(s)þH2O(l)

/ 2CuCl2(aq)þ 2Cu(s)

20e80 �C

S4 Dryer CuCl2(aq)/CuCl2(s)þH2O >100 �CS5 H2

production

2Cu(s)þ 2HCl(g)

/ 2CuCl(l)þH2(g)

430e475 �C

This reaction is endothermic and yields hydrochloric acid

gas (HCl) and Cu2OCl2. Hydrochloric acid gas is compressed

and Cu2OCl2 is transferred to another process step after its

temperature is increased to the oxygen production reaction

temperature of 500 �C.In the second (oxygen production) step (S2) an endo-

thermic chemical reaction takes place:

CuO * CuCl2(s)/ 2CuCl(l)þ 1/2O2(g) (2)

in which Cu2OCl2 is heated and O2 and copper monochloride

(CuCl) are produced. Liquid copper monochloride is solidified

by cooling it to 20 �C, after which it enters the third (copper

production) step (S3) together with the solid copper mono-

chloride from the fifth step. In the third process step solid

copper monochloride and water react endothermically at

20 �C as follows:

4CuCl(s)þH2O/ 2CuCl2(aq)þ 2Cu(s) (3)



S1 HE1 S2 S3 S5HE2 HE3

H2O(g)

S4HE4

Cu2OCl2(s)

HCl (g)

CuCl(l)

O2

(500ºC )

CuCl(s)

water

CuCl2

+ Water

Cu(s)Cu2OCl2(s)

P1

HCl (g)

CuCl2(s)

H2

(430ºC -475ºC )

O2

OUTPUT

400ºC 400ºC 500ºC

400ºC

500ºC 20ºC

20ºC

150ºC

20ºC

CuCl(s)

430ºC -475

20ºC

430ºC -475ºC

150ºC400ºC

OUTPUT

Cu-Cl Cycle

Fig. 5 e Variation of the five-step CueCl cycle for effective heat recovery within the cycle.


However in this reactionwateractsasacatalyst, anddoesnot

react with the other elements or compounds. Another specifi-

cation for this third reaction that differentiates this step from

others and makes it the most expensive, based on the price of

electricity, is that electrolysis occurs. In this reaction, solid

copper and a copper chlorideewater solution are produced. A

mixture of copper chloride andwater is transferred to thedryer

(S4), and solid copper enters the fifth step after its temperature

is increased to that step’s operating temperature. A physical

reaction takes place in the dryer as follows:

CuCl2(aq)/CuCl2(s)þH2O(l) (4)

In the fifth (hydrogen production) step (S5), hydrochloric gas

and copper enter, and are converted to gaseous hydrogen (H2)

and solid copper monochloride (CuCl). The reaction takes

place at 450 �C at steady state as follows:

2Cu(s)þ 2HCl(g)/ 2CuCl(l)þH2(g) (5)

An alternative layout of a five-step CueCl cycle is given in

Fig. 5. This layout has been designed for effective heat

recovery within the cycle. As explained above, there are five

main steps in the cycle; a chemical reaction takes place

in each step at various and high temperatures. These chem-

ical reactions form a closed internal loop that recycles all

chemicals on a continuous basis. The product of one step is

a reactant for another. However, since each step is at different

temperature, the product of a step needs to be cooled or

heated to next step’s temperature before entering it. Thus,

Table 4 e Reaction energy of the CueCl cycle steps and their aa pressure of 101 kPa.

Step Name Reaction heat (kJ/kmol H2) El

1 Fluidized bed 105,266

2 O2 production step 110,523

3 Cu production step e

4 Dryer 18,346

5 H2 production step �55,493

there are many heat recovery opportunities within the cycle.

The recovered energy as well as the energy that is released

from the exothermic reactions could be re-used in the cycle,

as illustrated in Fig. 5.

4. Performance analysis

The heat transfer for a chemical process, involving no work

interaction W, is determined from the energy balance_Ein � _Eout ¼ D _Esystem as applied to the system with W¼ 0. For

a steady state reaction process, the energy balance reduces to

Q ¼ Hp � HR ¼X

nP

�ho

f þ h� ho�P�X

nR

�ho

f þ h� ho�R

(6)

Here, the amount of product obtained in a chemical reaction is

called the yield, which is usually expressed in terms of grams

ormoles (molar yield). The effectiveness of a chemical process

can be measured by its fractional yield yf, calculated by

dividing the actual amount of the product obtained to the

stoichiometric (theoretical) amount. That is,

fractional yield�yf

�¼ actual yield

stoichiometric yield(7)

or in percentage form

percentage yield�yp

�¼ actual yield

stoichiometric yield100% (8)

The ideal or theoretical yieldof a chemical reaction is 100%, since

it is calculated based on the stoichiometry of the reaction.

However, 100% yield is impossible to achieve in actual cases

ssociated exergy values at the specified temperature and

ectricity (kJ/kmol H2) Exergy (kJ/kmol H2) Temp. (�C)

e 58,654 400

e 67,915 500

140,557 140,557 25

e 5421 150

e �32,620 450



Fig. 6 e Variation of reaction heat with reaction

temperature for the hydrolysis step in the CueCl cycle. Fig. 8 e Variation of reaction heat with reaction

temperature for the Cu production step in the CueCl cycle.


due to practical limitations. For example, in actual cases, the

reactions do not always take place stoichiometrically. Gener-

ally, one or more reactants are used in excess. Yields around

100% are called quantitative, above about 90% are called excel-

lent, above about 80% very good, above about 70% are called

good, below about 50% are called fair, and below about 40% are

called poor [14].

The overall energy efficiency of the CueCl cycle, he, can be

described as the fraction of energy supplied that is converted

to the energy content of H2 based on its lower heating value:

he ¼LHVH2

Q in þ Q loss

(9)

where LHVH2is the lower heating value, Q loss is the total heat

loss from theCueCl cycle andQ in is the total energyusedby the

process to produce a unit amount of product hydrogen, all per

kmol of hydrogen. The total energy demand of the CueCl cycle

is the sumof the reactionheats of thefivemain steps described

above. Note that, in this summation, the exothermic reaction

heat (i.e., the fifth step) is taken as negative, assuming this heat

canbeused for other endothermic reactions. The lowerheating

value of hydrogen is 240,000 kJ/kmolH2.

Fig. 7 e Variation of reaction heat with reaction

temperature for the O2 production step in the CueCl cycle.

Asmentioned earlier, we analyze a conceptual CueCl plant

which has not yet been built. Furthermore, many parameters

and specific details on quantity, capacity and material of

equipment (pumps, heat exchangers, compressors, fluidized

bed, evaporator, etc.) that are needed for such analyses are

unknown. Therefore, for its simplified analysis for overall

efficiency calculations we consider only the five main steps of

the CueCl cycle and assume that there are no heat losses in

these steps, individually. However, overall we assume that the

total heat losses (Qloss) from the CueCl cycle cover

a percentage of the total heat input (Qin) supplied to the cycle.

Theoverall exergyefficiencyof theCueCl cycle expressedas

hex ¼ exout

exin(10)

Using the exergy balance for the system, the exergy efficiency

can be written alternatively as

hex ¼ 1� exdestroyed

exin(11)

Fig. 9 e Variation of dryer inlet heat duty with drying

temperature.



Fig. 10 e Variation of dryer heat duty with inlet

temperature of the copper chloride solution.

Fig. 12 e Effect of yield on the overall efficiencies of the

CueCl cycles.


where exdestroyed is the total specific exergy destruction

including exergy losses and exin and exout are the input and

output specific exergies, all on amolar basis. Here, total exergy

needed to produce a unit amount of hydrogen is taken as

input and only exergy value of hydrogen is taken as output,

since it is the desired output. Detailed analyses of how the

input, output and destroyed exergy are obtained for each step

are given elsewhere [15e19]. For the overall cycle, we obtain

the total input and destroyed exergy of the cycle by adding the

input and destroyed exergy values of the individual steps.

Again, in the summation of input exergy, the exothermic

reaction (i.e., the fifth step) is taken as negative, assuming this

energy can be used for other endothermic reactions.

5. Results and discussion

The main steps of the CueCl cycle are evaluated individually

to determine how each process is influenced by varying key

operating and design parameters. A sensitivity analysis is also

performed in which one or more flowsheet parameters are

varied and the resulting effect on other variables studied, so as

to achieve a simple process optimization.

Fig. 11 e Variation of reaction heat of the H2 production

step with its temperature.

As discussed previously, three configurations currently

exist for the CueCl cycle (i.e., three-, four- and five-step

cycles). However, all of these combine one ormore steps in the

five-step CueCl cycle. Therefore, since the five-step CueCl

cycle is main design schematic, its five main reactors are

analyzed individually. The reaction energy of the main steps

in the CueCl cycle and their associated exergy values at

specified states are given in Table 4. The expected values of

the temperatures of themain cycle steps are also listed.When

we vary a step temperature in the analysis, the temperatures

for the other steps remain constant, at the value in Table 4. For

example, to investigate the relation between T1 (reaction

temperature of Step 1) and efficiencies (or some other vari-

ables), T2, T3, T4 and T5 (reaction temperature of Steps 2, 3, 4

and 5, respectively) are fixed at the values in Table 4.

The variation of the reaction heat (heat load) of HCl

production (Step 1) with reaction temperature is illustrated in

Fig. 6, for percentage yields of yp¼ 100% (complete reaction)

and yp¼ 80%. As explained earlier, an endothermic reaction

occurs in the fluidized bed. As reaction temperature increases,

the reaction heat for the HCl production step decreases, in

a nearly linear manner.

Fig. 13 e Variation of efficiencies of the CueCl cycle with

the reaction temperature of Step 1 (Hydrolysis Reactor).




the reaction temperature of Step 2 (Oxy-decomposition

Reactor).


the reaction temperature of Step 4 (Dryer).


The variation of the reaction heat of the oxy-decomposi-

tion reactor with reaction temperature is illustrated in Fig. 7.

An endothermic reaction takes place in this step. As reaction

temperature increases, the reaction heat for the O2 production

step increases.

The variation of the reaction heat for the Cu production

step with reaction temperature is illustrated in Fig. 8. An

endothermic reaction takes place in this step. As reaction

temperature increases, the reaction heat for the Cu produc-

tion step decreases.

The effects of varying the dryer temperature Tdryer and

inlet temperature of the CuC2/H2O mixture on the process

heat of the dryer are illustrated in Figs. 9 and 10, respectively.

The heat needed to evaporate water is seen to be approxi-

mately directly proportional to the evaporator temperature,

but to decrease approximately linearly as inlet temperature

increases. The evaporator inlet temperature is determined in

the previous step (copper production), where the reaction

temperature varies between 25 �C and 80 �C, respectively.


the reaction temperature of Step 3 (Electrolysis Reactor).

Although the amount of heat required for the drying step is

much higher than other steps in the cycle, it occurs at a lower

quality (a relatively low temperature, below 100 �C) and

therefore with heat that is more readily available. Thus, the

required heat can be obtained from low-grade “waste” or

recovered heat to improve the cycle efficiency. Spray drying is

an efficient method of water removal due to the relatively

large surface area available for heat and mass transfer,

provided the liquid atomizes into sufficiently small droplets

(on the order of a few hundredmicrons). Spray drying is a well

established industrial process that allows powder character-

istics and properties to be controlled and maintained nearly

constant throughout a drying operation.

The variation of the reaction heat for the H2 production

step, with reaction temperature, is illustrated in Fig. 11. An

exothermic reaction takes place in the H2 production step. As

reaction temperature increases, the reaction heat for the H2

production step decreases in a nearly linear manner. As the

percentage yield ( yp) decreases from 100% (complete reaction)

to 80%, less heat is released from the reaction.


the reaction temperature of Step 5 (H2 Production).



0 10 20 30 40 500

5

10

15

20

25

30

35

40

45

T0 (°C)

η((%

)

ηeηe

ηexηex


reference-environment temperature.


Most previous theoretical studies on the CueCl cycle

reported in the literature have assumed complete reaction for

each step in the cycle [15e25]. To obtain more realistic values,

we carry out an assessment inwhich a parametric study of the

impact on percentage yield ( yp) is determined for each step.

Fig. 12 illustrates the variation of overall efficiencies of the

CueCl cycle with percentage yield of the steps. As can be

observed from this figure, the energy efficiency of the cycle

approaches 45% and the exergy efficiency 10%, when all

reactions become fully stoichiometric. Based on the para-

metric study included in this figure, we assume a percentage

yield of 80% for the steps in the CueCl cycle involving

a chemical reaction and a 75% heat exchanger effectiveness

for heat management and handling within the cycle. In the

sensitivity analysis, the 80% percentage yield for each step is

an assumption based on expected values (typical for this kind

of chemical reaction). This value is mainly used to determine

the effect of reaction yield on the overall cycle efficiency, for

example, by howmuch the overall cycle efficiency decreases if

the reaction yield is 80% instead of 100%.

Figs. 13e17 show the variations of energy (he) and exergy

(hex) efficiencies of the CueCl cycle with the temperatures of

main reactions in the cycle. These results appear to be lower

than the values obtained in previous theoretical studies. This

difference is likely attributable to the fact that we assume

a complete reaction for each step in the previous calculations.

The overall energy efficiency of the cycle varies from 42% to

44% and exergy efficiency from 6% to 8% (see Figs. 13e17).

Table 5 e Grouping of different steps of the CueCl cycle.

Cu-ClCu-Cl

Five-step cycle Five-step cycle Four-step cyFour-step cy

S1:S1:

S4:S4:

S2:S2: S-II:S-II:

S-I:S-I:

S3:S3: S-III:S-III:

S5:S5: S-IV:S-IV:

The overall energy and exergy efficiencies of the CueCl

cycle vary with the reaction temperature of each step as well

as with the reference-environment temperature. The varia-

tion of energy and exergy efficiencies of the cycle with refer-

ence-environment temperature (T0) is illustrated in Fig. 18.

The exergy efficiency decreases with increasing T0, while the

energy efficiency remains constant.

Table 5 and 6 summarize a number of grouping methods,

and the corresponding major advantages and disadvantages

of different copper-chlorine cycle variations. The main

advantage resulting from reducing the number of steps is that

the challenges arising from processing of solid particles are

dramatically reduced. The mixing and handling of a homoge-

nous liquid or gas is usually easier than of a solid. Also, the

heat and mass transfer for solid phases (e.g. solid-solid or

fluid-solid) is more complex and unpredictable. For example,

any non-ideal mixing may result in an incomplete hydrolysis

reaction in the Cu-Cl cycle, whichmay further add difficulty to

the separation of product particles from reactant particles

[13]. Feeding of solid particles into a reactor, caking prevention

of solid deposition on the reactor wall and clogging of pipes or

channels are some other challenges that may be faced during

processing of solid particles. These challenges can be signifi-

cantly reduced in the four-step and three-step Cu-Cl cycles,

where the feed is liquid only.

Reducing the number of steps in the Cu-Cl cycle can also

yield some disadvantages. The main disadvantages are the

higher grade heat requirement, as shown in Fig. 19, and

a reduction in the yield of desirable products (and increased

yield of undesirable products), as shown in Fig. 20). These

disadvantages are more significant in the three-step Cu-Cl

cycles since they need higher heat grade than the four-step or

five-step cycles. Also, changing the five-step cycle to a three-

step cycle increases the heat intensity more significantly than

reducing the five-step cycle to a four-step cycle, because the

heat load of three reactions in the five-step cycle are combined

and added to a single reactor in the three-step cycle. This

phenomenon can make the selection of reactor material very

difficult from a practical engineering perspective. Also,

combining two or more reactions causes more complex

reactions with additional engineering challenges for efficient

separation of products. These disadvantages taken together

result in a reduction of the overall efficiency of the cycle.

The variation of energy and exergy efficiencies of the cycle

with number of steps is illustrated in Fig. 21. The energy effi-

ciency decreases as thenumber of steps in the cycle decreases,

while the exergy efficiency increases. The exergy destruction

cycle cycle

clecle Three-step cycleThree-step cycle

S-i:S-i:

S-ii:S-ii:

S-iii:S-iii:



0

5

10

15

320 340 360 380 400 420 440 460

Reaction temperature (°C)

Und

esir

able

pro

duct

s co

nver

sion

(%

)

Fig. 20 e Dependence of by-products on reaction

temperature (adapted from [13]).

400

500

600

700

800

900

1000

1100

1200

Five-step Four-step Three-step Two-step

Configuration

Max

imum

cyc

le te

mpe

ratu

re (

°C)

Fig. 19 e Variation of the maximum operation temperature

with number of steps for various CueCl cycles.

0

5

10

15

20

25

30

35

40

Three-step Four-step Five-step

Cu-Cl cycle

Eff

icie

ncy

(%)

Energy

Exergy

Fig. 21 e Variation of efficiencies of various CueCl cycles

assuming an overall heat loss of 30% of the total input

energy to the cycle and an 80% reaction efficiency for each

step.

Table 6 e Main advantages and disadvantages ofdecreasing the number of steps in the CueCl cycle.

Configuration Advantages Disadvantages

Five-step cycle Less challenges to process solid particles Less steps and hence equipment Reduced complexity Better reaction kinetics Homogenous reaction mixture

Higher heat grade and intensityMore equipment material challenges More undesirable side productsLower efficiencies

Four-step cycle

Three-step cycle

•

••••

••

••


for a chemical reaction in any step of the cycle is very signifi-

cant becauseof thedifference betweenchemical exergy values

of chemical compounds. Thus, fewer steps and hence fewer

chemical reactions appear to yield higher exergy efficiencies.

However the opposite behaviour is observed for the energy

efficiencies. Since the quantity and grade of energy needed for

theproductionof aunit ofhydrogen increasesas thenumberof

cycle steps decreases, the energy efficiency decreases.

6. Conclusions

Further developments and improved design schemes of the

CueCl thermochemical water decomposition cycle for

hydrogen production are needed that utilize the results of the

various technical studies that have been reported previously.

In this study, various design schemes for the overall CueCl

cycle and its components are examined in order to identify

potential performance improvements. The implications of

implementing these new configurations are discussed and

the potential benefits ascertained. Also, a thermodynamic

analysis and several parametric studies are presented for

various configurations. This information should assist

ongoing efforts to understand the thermodynamic losses in

the cycle, to improve efficiency and to facilitate eventual

commercialization. Future work is expected to focus on

detailed analyses and comparisons of these new configura-

tions of the CueCl cycle using advanced simulation and

optimization tools.

Acknowledgements

The authors gratefully acknowledge the support provided by

the Ontario Research Excellence Fund and the Natural

Sciences and Engineering Research Council of Canada.

Nomenclature

e molar energy, kJ/kmol

E energy, kJ

Ex exergy, kJ

h specific enthalpy, kJ/kg

h molar enthalpy, kJ/kmol

h0

molar enthalpy at reference state, kJ/kmol

h0

f molar enthalpy of formation, kJ/kmol

LHV lower heating value, kJ/kg fuel




n number of moles per mole hydrogen produced,

kmol/kmol H2

Q heat, kJ_Q heat transfer rate, kW

T temperature, K

yf fractional yield

yp percentage yield, %

Greek Letters

h efficiency

Subscripts

aq aqueous solution

e energy

ex exergy

f fractional

g gas

l liquid

p percentage

s solid

0 reference state

Superscripts

�(over bar) quantity per unit mole

_ (over dot) quantity per unit time

0 standard reference state

r e f e r e n c e s

[1] Al-Dabbagh AW, Lu L. Design and reliability assessment ofcontrol systems for a nuclear-based hydrogen productionplant with copperechlorine thermochemical cycle.International Journal of Hydrogen Energy 2010;35:966e77.

[2] Daggupati VN, Naterer GF, Gabriel KS, Gravelsins RJ,Wang ZL. Equilibrium conversion in CueCl cycle multiphaseprocesses of hydrogen production. Thermochimica Acta2009;496:117e23.

[3] Daggupati VN, Naterer GF, Gabriel KS, Gravelsins RJ, Wang ZL.Solid particle decomposition and hydrolysis reaction kineticsinCueCl thermochemical hydrogenproduction. InternationalJournal of Hydrogen Energy 2010;35:4877e82.

[4] Daggupati VN, Naterer GF, Gabriel KS. Diffusion of gaseousproducts through a particle surface layer in a fluidized bedreactor. International Journal of Heat and Mass Transfer2010;53:2449e58.

[5] Ferrandon MS, Lewis MA, Tatterson DF, Doizi D, Croize L,Dauvois V, et al. Hydrogen production by the CueClthermochemical cycle: investigation of the key step ofhydrolysing CuCl2 to Cu2OCl2 and HCl using a spray reactor.International Journal of Hydrogen Energy 2010;35:992e1000.

[6] Ferrandon MS, Lewis MA, Alvarez F, Shafirovich E. Hydrolysisof CuCl2 in the CueCl thermochemical cycle for hydrogenproduction: experimental studies using a spray reactor withan ultrasonic atomizer. International Journal of HydrogenEnergy 2010;35:1895e904.

[7] Jaber O, Naterer GF, Dincer I. Heat recovery frommolten CuClin the CueCl cycle of hydrogen production. InternationalJournal of Hydrogen Energy 2010;35:6140e51.

[8] Ranganathan S, Easton EB. Ceramic carbon electrode-basedanodes for use in the CueCl thermochemical cycle.International Journal of Hydrogen Energy 2010;35:4871e6.

[9] Ranganathan S, Easton EB. High performance ceramic carbonelectrode-based anodes for use in the CueCl

thermochemical cycle for hydrogen production.International Journal of Hydrogen Energy 2010;35:1001e7.

[10] Wang ZL, Naterer GF, Gabriel KS, Gravelsins R, Daggupati VN.Comparison of sulphureiodine and copperechlorinethermochemical hydrogen production cycles. InternationalJournal of Hydrogen Energy 2010;35:4820e30.

[11] Zamfirescu C, Dincer I, Naterer GF. Thermophysicalproperties of copper compounds in copperechlorinethermochemical water splitting cycles. International Journalof Hydrogen Energy 2010;35:4839e52.

[12] Zamfirescu C, Naterer GF, Dincer I. Kinetics study of thecopper/hydrochloric acid reaction for thermochemicalhydrogen production. International Journal of HydrogenEnergy 2010;35:4853e60.

[13] Wang ZL, Naterer GF, Gabriel KS, Gravelsins R, Daggupati VN.Comparison of different copperechlorine thermochemicalcycles for hydrogen production. International Journal ofHydrogen Energy 2009;34:3267e76.

[14] Vogel AI, Tatchell AR, Furnis BS, Hannaford AJ, Smith PWG.Vogel’s textbook of practical organic chemistry. 5th ed.Prentice Hall; 1996.

[15] Orhan MF, Dincer I, Rosen MA. The oxygen production stepof a copperechlorine thermochemical water decompositioncycle for hydrogen production: energy and exergy analyses.Chemical Engineering Science 2009;64:860e9.

[16] Orhan MF, Dincer I, Rosen MA. Energy and exergy analyses ofthe fluidized bed of a copperechlorine cycle for nuclear-based hydrogen production via thermochemical waterdecomposition. Chemical Engineering Research and Design2009;87:684e94.

[17] Orhan MF, Dincer I, Rosen MA. Thermodynamic analysisof the copper production step in a copperechlorine cyclefor hydrogen production. Thermochimica Acta 2008;480:22e9.

[18] Orhan MF, Dincer I, Rosen MF. Energy and exergyassessments of the hydrogen production step ofa copperechlorine thermochemical water splitting cycledriven by nuclear-based heat. International Journal ofHydrogen Energy 2008;33:6456e66.

[19] Orhan MF, Dincer I, Rosen MA. Energy and exergy analyses ofthe drying step of a copperechlorine thermochemical cyclefor hydrogen production. International Journal of Exergy2009;6:793e808.

[20] Orhan MF, Dincer I, Rosen MA. Efficiency analysis of a hybridcopperechlorine (CueCl) cycle for nuclear-based hydrogenproduction. Thermochimica Acta 2009;155:132e7.

[21] OrhanMF,Dincer I, RosenMA.An exergyecosteenergyemassanalysis of a hybrid copperechlorine thermochemical cyclefor hydrogen production. International Journal of HydrogenEnergy 2010;35:4831e8.

[22] Orhan MF, Dincer I, Naterer GF, Rosen MA. Coupling ofcopperechloride hybrid thermochemical water splittingcycle with a desalination plant for hydrogen production fromnuclear energy. International Journal of Hydrogen Energy2010;35:1560e74.

[23] Rosen MA, Naterer GF, Chukwu CC, Sadhankar R, Suppiah S.Nuclear-based hydrogen production with a thermochemicalcopperechlorine cycle and supercritical water reactor:equipment scale-up and process simulation. InternationalJournal of Energy Research, doi:10.1002/er.1702, in press.

[24] Balta MT, Dincer I, Hepbasli A. Geothermal-based hydrogenproduction using thermochemical and hybrid cycles:a review and analysis. International Journal of EnergyResearch 2010;34:757e75.

[25] Dincer I, Balta MT. Potential thermochemical and hybridcycles for nuclear-based hydrogen production. InternationalJournal of Energy Research 2011;35:123e37.

http://dx.doi.org/10.1002/er.1702



Design of systems for hydrogen production based on the Cu–Cl thermochemical water decomposition...

Documents

Transcript of Design of systems for hydrogen production based on the Cu–Cl thermochemical water decomposition...