1

HYDROGEN PRODUCTION BY HIGH TEMPERATURE ELECTROLYSIS OF

WATER VAPOUR AND NUCLEAR REACTORS

Jean-Pierre Py (AREVA NP) Alain Capitaine (EDF)

Abstract


This paper presents hydrogen production by a nuclear reactor (High Temperature Reactor,
HTR or Pressurized Water Reactor, PWR) coupled to a High Temperature Electrolyser
(HTE) plant. With respect to the coupling of a HTR with a HTE plant, EDF and AREVA NP
had previously selected a combined cycle HTR scheme to convert the reactor heat into
electricity. In that case, the steam required for the electrolyser plant is provided either directly
from the steam turbine cycle or from a heat exchanger connected with such cycle.
Hydrogen efficiency production is valued using high temperature electrolysis. Electrolysis
production of hydrogen can be performed with significantly higher thermal efficiencies by
operating in the steam phase than in the water phase. The electrolysis performance is
assessed with solid oxide and solid proton electrolysis cells.
The efficiency from the three operating conditions (endo-thermal, auto-thermal and thermo-
neutral) of a high temperature electrolysis process is evaluated. The technical difficulties to
use the gases enthalpy to heat the water are analyzed, taking into account efficiency and
technological challenges.
EDF and AREVA NP have performed an analysis to select an optimized process giving
consideration to plant efficiency, plant operation, investment and production costs. The paper
provides pathways and identifies R&D actions to reach hydrogen production costs
competitive with those of other hydrogen production processes.

Introduction

The challenging target of large hydrogen production with a competitive cost can be solved
with nuclear energy. This paper describes how an electrolysis process could be linked with
the secondary part of a nuclear reactor.

In the following part the efficiency for the three operating conditions (endo-thermal, auto-
thermal and thermo-neutral) and two possible electrolyser technology are assessed (solid
oxide and solid proton electrolysis cells). Finally, results of an economical study are provided
for different variables and hypotheses related to electrolyser, allowing comparison of
hydrogen production costs obtained by high temperature electrolysis and other processes.


1. INTEREST OF HIGH TEMPERATURE ELECTROLYSIS (HTE)

1.1 With respect to Steam Methane Reforming (SMR)

Steam methane reforming (SMR) is the most currently used and commercialized process to
produce large quantities of hydrogen. In this process, natural gas reacts with steam in the
presence of a nickel catalyst in the reactor. The reaction product is a mixture of hydrogen
and carbon monoxide. In order to obtain pure hydrogen, purification downstream of the SMR
plant is necessary to remove undesired compounds such as carbon dioxide.
Based on higher heating value (HHV) of hydrogen, SMR efficiencies are in the range of 65 to
85 percent. Hydrogen production cost is directly proportional to gas price, very unpredictable.

1.2 With respect to alkaline water electrolysis (AWE)

2

Electrolysis of water is the most widely used means to produce hydrogen of high purity.
Should electricity be provided by renewable energies (hydro, wind, solar) or nuclear power,
hydrogen could be obtained without Green House Gas (GHG) emission.
Conventional electrolysis, or alkaline water electrolysis, is a mature technology since
decades and has efficiencies in the range of 70 to 80 percent. The working pressure of
electrolysis is between normal pressure and medium pressure (10 to 30 bar presently).
Alkaline water electrolysis is striving to obtain higher efficiency by increasing operating
temperature or pressure. Higher pressure electrolysis is thought to be advantageous for
future application because many processes which use hydrogen are operated at high
pressure; therefore, additional compression energy is saved. The specific consumption of
electricity of the process of electrolysis is reduced at higher pressure. Energy requirements
of conventional electrolysers vary according to commercial available electrolysis systems
between 4.3 and 4.9 kWh
el
/Nm
3
H
2
.

Besides alkaline water electrolysis another development is in progress. The proton exchange
membrane (PEM) is used as an electrolyte instead of aqueous solutions of alkaline
electrolysis. The membrane electrolysis uses a proton conducting membrane. This solid
polymer electrolyte water electrolysis can be operated at higher current densities due to
volume reduction compared to cells with potassium hydroxide-electrolyte. Typical operation
temperatures are 50 to 150 C, preferably below 100C for acceptable lifetime, which is a
limiting for this technology. The requirement of electricity should be reduced; values below 4
kWh
el
/ Nm
3
H
2
are envisioned. PEM is considered a promising method. However, high
component cost still remains the major drawback of this technology.


2. PRINCIPLES OF HIGH TEMPERATURE ELECTROLYSIS (HTE)

Another principle of electrolysis being considered is the high temperature electrolysis (HTE).
The key to the HTE process is operation at high temperature so as to reduce the
electrochemical over voltage imposed at the electrodes. Operating temperatures in the range
of 800 to 1000 C would offer the advantage of smaller specific electricity requirement than
conventional electrolysis; values ranging from 2,6 to 3 kWh
el
/ Nm
3
H
2
are expected. To
achieve high temperature in the process, water is provided as high temperature superheated
steam. The necessary electricity input is reduced corresponding to the variation of cell
voltage versus temperature (figure 1). The heat for steam production to supply the cells must
be added. Current density acceptable by the HTE cell is a function of the temperature of
operation and materials used for the Membrane Electrode Assembly (MEA).
The upper temperature limit for economical HTE is set by the materials of the HTE cell.

Ion O
2-
conduction electrolysis

The electrolyte presently considered for this technology is of the same type as the electrolyte
used for Solid Oxide Fuel Cells (SOFC), using zirconia stabilized with yttrium oxide (YSZ).
Presently the target operating temperature of O
2-
conduction HTE cells is 850C. Figure 2a
illustrates the operation of O
2-
electrolysis.

Hydrogen is produced at the cathode with a fraction of steam flow not dissociated; therefore
it must be separated from the residual steam.





3
Proton H
+
conduction electrolysis

The electrolyte for this technology must meet conductivity requirements at the presently
envisaged operating temperature of H
+
conduction HTE cells of 500C. A joint research R &
D program between AREVA and CNRS/IEM is dedicated at the development and
characterization of appropriate EMA able to operate at such temperature without any
conductivity defect in time. (see paper n . presented at this conference). Figure 2 b
illustrates the operation of H
+
electrolysis.
One of the most attractive benefits of this type of electrolysis is to obtain pure hydrogen at
the cathode. An other advantage of the proton HTE is that it allows to use the HTR steam
cycle as heat source. Materials having satisfactory conduction are being developed and still
require important development work to reach industrial scale of fabrication.

Operating modes of an electrolyser

An electrolyser operating mode is characterized by the inlet steam and gas outlet
temperatures. The three possible operating modes are :

The thermo-neutral mode

In this mode the energy input to the electrolyser exactly matches the energy necessary to
split the water molecule at the selected operating temperature. Thus, the gas outlet
temperatures are the same as the steam inlet temperature. The main interest of this mode is
to operate the electrolyser under a homogeneous internal temperature, resulting in less
constraints within the thin ceramic Membrane Electrode Assembly (MEA).

The auto-thermal mode

In this mode the energy input to the electrolyser exceeds the energy necessary to split the
water molecule at the selected operating temperature. Thus, the gas outlet temperatures are
higher than the steam inlet temperature. The electric power supplied to the electrolyser
provides the excess energy released by Joule effect.

The endo-thermal mode

In this mode the energy input to the electrolyser is less than the energy necessary to split the
water molecule at the selected operating temperature. Thus, the gas outlet temperatures are
lower than the steam inlet temperature.


3. COUPLING OF HTE WITH A NUCLEAR REACTOR

3.1 Heat supply and associated temperature and pressure

For a nuclear-powered HTE plant, a nuclear reactor is to provide the heat to produce high
temperature steam and electricity for the electrolysis.

3.2 Where could the heat be extracted from ?

For a given type of reactor the heat necessary for steam production to supply the HTE cells
can be extracted from different points. The flow rate, temperature and pressure of the heat
carrying fluid must be adapted to the selected HTE operating conditions. The heat
transported by the gases produced by the HTE cells is first recuperated to raise the

4
temperature of the feed water from ambient to boiling conditions, then to superheat the
steam up to operating temperature of the HTE cells.

3.2.1 Heat extraction in the case of a PWR :

From the main steam header at 280 C (figure 3)

Heat is extracted in the form of steam from the main steam header of a PWR at 64 bar and
280 C. The steam condenses in the primary side of a steam transformer at the same
temperature. The water fed at 5 bar on the secondary side, the corresponding saturation
temperature of which is 152 C, is evaporated at the same temperature. Only two percent
(i.e. 28 kg/s) of the main steam flow rate is sufficient to provide the required heat to produce
22.5 N m
3
H
2
/s. The required heat transfer surface necessary to produce such a steam flow
rate would be of the order of 130 m
2
. The steam transformer operates as a condenser on the
primary side and as steam generator on the secondary side.

From the low pressure turbine stage at 180 C (figure 4)

The heat is extracted in the form of steam from the low pressure turbine stage of a PWR at
10 bar and 180 C. The steam condenses in the primary side of a steam transformer at the
same temperature. The water is evaporated on the secondary side at the same pressure and
saturation temperature as above (5 bar, 152 C). The required heat transfer surface would be
more than four times larger in this case than in the previous one.

3.2.2 Heat extraction in the case of the ANTARES HTR

In the case of the ANTARES combined cycle HTR developed by AREVA, the heat could
either be extracted from the secondary gas loop or from the tertiary steam turbine cycle (see
figure 5). Previous studies have shown that the later solution is preferable for an overall
efficiency of the plant.

Two designs were explored, both allowing to bleed steam from the steam turbine cycle of the
HTR ANTARES combined cycle :

Design A : The steam feeding the electrolyser is directly extracted from the steam
turbine cycle of the combined cycle HTR. The gases produced by the electrolyser
exchange there latent heat with the feed water from the steam turbine cycle to preserve
the efficiency.

Design B : Heat is extracted from the steam turbine cycle of the nuclear plant through a
specific heat exchanger supplying saturated steam on the secondary side. This steam is
then superheated within the electrolysis plant by recovering the heat of the electrolysers
outlet gases. The main advantage of this design is to separate the specific water used
into the electrolyser from the water used into the steam turbine cycle.




Design A :

In this design the steam quality is considered sufficient to directly feed the electrolyser. The
remaining available steam drives the turbine which produces the electricity supplying the
electrolyser. For a given reactor outlet temperature, the hydrogen efficiency production

5
depends upon the three possible operating modes of the electrolyser (endo-thermal, thermo-
neutral and auto-thermal).

However, efficiency is not the unique target to select a design or an operating mode. Many
other criteria must be taken into account such as the temperature, the electrolyte material,
the ancillary facilities in operation, the investment cost, etc

The scheme on figure 6 shows the design arrangement with steam extraction upstream the
medium pressure turbine.

The hydrogen produced is directed towards dedicated storages, away from the plant; the
oxygen can either be directly released to the atmosphere or used in industrial processes.
The excess of water used into the electrolyser is recycled into the steam turbine cycle, after
degassing to prevent incondensable gas ingress to prevent corrosion.
The electrolysis efficiency depends upon many parameters. The electrolysis efficiency is the
ratio between (Hydrogen energy produced per second) / (Primary Energy required to
produce hydrogen). The efficiency varies according to the electrolyser operating mode. The
highest efficiency is obtained with auto-thermal mode; however, the thermo-neutral mode is
preferable because the inlet steam and outlet gas temperatures being the same the thermal
constraints to the electrolyser components are reduced.

Design B :

In this design the superheated steam feeding the electrolyser is produced by heat
exchangers recovering the heat of the gases produced. The preferable electrolyser operation
mode being the thermo-neutral mode, the optimal coupling scheme with the ANTARES HTR
is obtained by extraction of steam from two different points in the HTR steam cycle :



Upstream the high pressure steam turbine, to provide the heat necessary to evaporate
the water at low pressure (5 bar) to be electrolysed;
Upstream the low pressure steam turbine, to provide the heat necessary to superheat
up to 500 C the steam to be electrolysed.

For a plant producing 81000 t of hydrogen per year, equipped with HTE having specific
electricity consumption of 3.2 kWh/Nm
3
H
2
and operating in the thermo-neutral mode, the
required energy for the electrolysis plant are 257 MWe as electricity and 53.7 MWth as heat
to vaporize the water under 5 bar at 152C and 5.3 MWth as heat to superheat the steam.


4. RESULTS OF EDF AND AREVA NP STUDIES

4.1 EDF study

In EDF study, the steam quality is considered sufficient to directly feed the electrolyser. This
design is quite interesting because it uses the demineraliser facility of the nuclear plant to
produce the water to be electrolysed. The steam production for the electrolysers does not
require any exchanger because the steam is directly extracted from the steam turbine cycle.
This solution is beneficial in terms of efficiency (no heat exchanger required) and investment
cost. This design allows to reach a steam temperature close to 560C; higher temperatures
could be obtained by extracting heat at the outlet of the Intermediate Heat Exchanger of the
HTR through a dedicated heat exchanger.
The selected operating mode of the electrolyser is the thermo-neutral mode for efficiency
reason and operation of the electrolyser under a homogeneous internal temperature,
resulting in less constraints within the thin ceramic Membrane Electrode Assembly (MEA).

6


4.2 AREVA NP study

The whole interest of the auto-thermal mode is to allow superheating of the steam up to
500C with the gas produced by the electrolyser. The external heat (via steam produced by a
HTR or PWR reactor) can thus be supplied at temperatures lower than 200 C and can be
used to evaporate the water into steam to be dissociated in the electrolyser. The
thermodynamic study of the cycle has allowed validating the hypothesis of operation in the
auto-thermal mode. Optimisation of the sets of heat exchangers (Ts and heat transfer
areas) has lead to the general scheme in figure 7. Results of the study of the different modes
of operation of the electrolysers are summarized in the following table :

Mode of operation
Auto-thermal Thermo-neutral Endo-thermal
Extraction points on
the steam cycle
One : High pressure Two : High pressure,
low pressure
Two : High pressure,
low pressure
Electric power requi-
red for the HTE
261 MWe 257 MWe 253 MWe
Thermal power
required for the HTE
53.7 MWth 53.7 MWth
+ 5.3 MWth
53.7 MWth
+ 10.57 MWth
HTR net electric
output
~ 271.5 MWe ~ 268.75 MWe ~ 266.9 MWe
HTR plant electric
efficiency
1

45.25 % 44.79 % 44.45 %
Hydrogen plant
global efficiency
2
41,60 % 41,83 % 42,16 %


5. ASSESSMENT OF HYDROGEN PRODUCTION COSTS

5.1 Steam Methane Reforming (SMR)

Hydrogen production capacities, assuming stationary use, are mostly in the range of over 45
000 Nm
3
-H
2
/hour. The specific investment costs (i.e. capital investment cost (in ) divided by
hydrogen plant capacity (in kW), are in the range of 210 to 260 / kW. Operation and
maintenance (O&M) cost are comprised between one and two / GJ. SMR technology has
the least-expensive investment cost in comparison to other fossil fuelled hydrogen production
methods.

5.2 Alkaline water electrolysis (AWE)

Hydrogen production cost by AWE has been calculated with the following assumptions :






Annual hydrogen production capacity : 638 600 x 10
6
Nm
3

Capacity factor : 90 %
Plant lifetime : 20 years
Capital cost : 500 / kW

1
Electric efficiency defined as the ratio of the HTR net electric output to the reactor thermal power, i.e.
286,5/600
2
H2 plant global efficiency defined as the ratio of : [H2 production flow rate (kg/s) x H
2
LHR (120 MJ/kg)] by
[the electric power requirement of the electrolyser (MWe) x HTR plant electric efficiency]

7
Annual operating and maintenance costs :



o Fixed costs : 10 % of the initial plantinvestment
o Variable costs : 0.02 / Nm
3

Specific electricity consumption : 4.7 kWh / Nm
3
H
2

Electricity cost : variable (base case : 35 / MWh) ; escalation rate : 1 % / yr)
Discount rate : 8 % / yr

Hydrogen production cost
Delivery pressure : 200 bar
Operation
Investment
Electricity
Consumption
Compression
0,00
0,50
1,00
1,50
2,00
2,50
3,00
3,50
100
production cost
(/kg)

In this case electricity consumption represents about 70 % of the hydrogen production costs.

5.3 High Temperature Electrolysis (HTE)

Hydrogen production cost by HTE H
+
coupled with an ANTARES HTR module of 600 MWth
has been calculated with the following assumptions :

Annual hydrogen production capacity : 638 600 x 10
6
Nm
3








Capacity factor : 90 %
Plant lifetime : 20 years
Capital cost : variable according to HTE cells cost, not yet manufactured on an industrial
scale
Annual operating and maintenance costs :
o Fixed costs : 10 % of the initial plant investment
o Variable costs : 0
Specific electricity consumption : 3.2 kWh / Nm
3
H
2

Electricity cost : 35 / MWh
Discount rate : 8 % / yr

In the evaluation of HTE for hydrogen production, the most sensitive parameters are the cost
and lifetime duration of the HTE cells.

HTE cell costs can presently be assessed only by comparison with fuel cells costs which
have not yet reached an industrial production stage. It is expected that when produced on
large scale fuel cell costs should drop by a factor of 10 at least from present costs for R&D
applications (20 000 /kW). For the purpose of this study a range of 2000 to 200 /kW was
assumed for electrolysis cells, corresponding to a range of 11 000 to 1000 / electrolysis cell.


8
0
1
2
3
4
5
0 1 2 3 4 5 6 7 8 9 10 11 12
HTE cell cost (k)
P
r
o
d
u
c
t
i
o
n

c
o
s
t

(

/
k
g
)
HTE cell lif e time : 5 yrs
HTE cell lif e time :20 years

Hydrogen production cost
as a function of HTE cell cost and life duration


5.4 Comparison of HTE efficiencies

The following figures shows the HTE efficiencies obtained for O
2-
and H
+
electrolysis in the
three operating modes :


Proton (H
+
) electrolysis efficiency




PROTONIC ELECTROLYSIS EFFICIENCY
0.395
0.4
0.405
0.41
0.415
0.42
0.425
0.43
0.435
allothermal thermoneutral autothermal
MODE
V
A
L
U
E
EHT efficiency
VHTR efficiency

















9


Oxide (O
2-
) electrolysis efficiency





OXIDE ELECTROLYSIS EFFICIENCY
0.39
0.395
0.4
0.405
0.41
0.415
0.42
0.425
0.43
0.435
allothermal thermoneutral autothermal
MODE
E
F
F
I
C
I
E
N
C
Y
EHT efficiency
VHTR efficiency
















6 CONCLUSIONS

HTE technology



This study has shown that hydrogen production by H
+
electrolysis seems more attractive
than by O
2-
electrolysis because its operating range is lower (500C to 600C versus
850c and above) and it produces pure hydrogen at the cathode. However, the
competitiveness of H
+
HTE with respect to other processes remains to be confirmed.

Coupling of a HTE plant with a nuclear reactor

o The electrical part of the total energy supplied is predominant (260 MW
el
versus 60
MW
th
of heat ). The heat is merely used to vaporize the water at low temperature (54
MW
th
at 152 C) and possibly to superheat the steam (6 MW
th
at 500 C).
o The coupling of a HTE plant with a PWR is possible in the auto-thermal operating
mode (vaporization of water only), or in the other modes, provided that steam
superheating is performed by a small auxiliary heat source (e.g. an electric boiler).

Competitiveness

Although it features a lower specific electricity consumption than AWE, HTE is
conceivable only with a very cheap cost of electricity such as from nuclear power plants,
which absence of Green House Gases reinforce the interest.

10
Figure 1 : Electrical energy required to dissociate water versus temperature





H (kJ/mol)
Temperature (C)
=Electricity
= Heat

H
+
conduction
Low Temp. O
2-
conduction
Process Heat
300
TS
250
200
G
150
0 20 40 60 80 100 400 700 1000

11
Figure 2a : Principle of High Temperature Electrolysis with O
2-
conduction
























2 -
2O +

2
2H
-
2

O+ 4 e
Cathode
2H

-
+ 4 e
2 O 2O

2 -

Anode

2 O

External
Circuit

I

-
e

-
e

2 H

O 2 H


Figure 2b : Principle of High Temperature Electrolysis with H
+
conduction



Anode
2H
2

O O
2
+ 4 e
-
+ 4H
+
Cathode
4 H
+
+ 4 e
-
2H
2

External
Circuit
O
2
I
H
2
O
e
-
e
-

H
2


12
Figure 3

Principle of coupling of a HTE plant with a PWR

Heat extraction from the main steam header



EHT
plant
PWR plant
steam
280C X
v
= 0
38 kg/s X
v
= 1
Steam Steam
Transformer Transformer
Reheaters Reheaters
LP LP
HP HP
SG SG
Condenser Condenser
280 C
64 bar
1608 kg/s
1574 kg/s
152C
5 bar
28 kg/s
226C
68 bar
152C
5 bar
28 kg/s
water
Hydrogen Water Steam Oxygen+steam

13
Figure 4

Principle of coupling of a HTE plant with a PWR

Heat extraction from the low pressure turbine

Condenser
SG
HP
LP
Reheaters
Steam
Transformer
Condenser
SG
HP
LP
Reheaters
Steam
Transformer
280C
64 bar
1608 kg/s
1581 kg
180C
10 bar - 29kg/s
152C
5 bar - 28 kg/s
EHT
plant
PWR plant
152C
5 bar
28 kg/s

14
Figure 5

Possible heat extraction points from the ANTARES HTR


High Temp.
Process Heat
~550 to 800C
Primar
y Loop


600 MWt
Reactor
Gas
Cycle
IHX
Circulator
Gas
turbine
S.G.
He
Med. Temp.
Process Heat
~250 to 550C
Condenser
Steam
Cycle
He or N
2
/He
Water/steam
Low Temp.
Process Heat
~30 to 250C
Generator

15
Figure 6

Steam extraction from the ANTARES HTR upstream the medium pressure
turbine




A
B
A
B
Q+
Racteur 600MW
850C
800C
565C
circuit primaire Hlium
circuit secondaire He-N2
circuit tertiaire eau
IHX
Chaudire
circuit de refroidissement Electrolyseur

Steam extraction

16
Figure 7 : Flow sheet of HTE coupled with a HTR




Water
25C
5 bar

EHT

261 MW
l
H
2,
2 kg/s

560C
H
2
O, 28 kg/s
500C

O
2
+H
2
O
560C
500C

500C

H
2
70C
5 bar

Reheater

Cooler Reheater Superheater

Superheater

O
2
70C
5 bar


Evaporater
54 MW
th 152C

152C


Nuclear reactor