TSINGHUA SCIENCE A N D T E C H N O L O G Y

ISSN 1007-0214 23/23 pp270-276

V o l u m e 10, N u m b e r 2, April 2005

Thermochemical Water Splitting for Hydrogen Production

Utilizing Nuclear Heat from an HTGR*

W U Xinxin ONUKI Kaoru 1

Institute of Nuclear and N e w E n e r g y Technology, Tsinghua University, Beijing 100084, China;

t J a p a n A t o m i c E n e r g y Research Institute, Ibaraki 311-1394, Japan

Abstract: A very promising technology to achieve a carbon free energy system is to produce hydrogen
from water, rather than from fossil fuels. Iodine-sulfur (IS) thermochemical water decomposition is one
promising process. The IS process can be used to efficiently produce hydrogen using the high temperature
gas-cooled reactor (HTGR) as the energy source supplying gas at 1000°C. This paper describes that dem-
onstration experiment for hydrogen production was carried out by an IS process at a laboratory scale. The
results confirmed the feasibility of the closed-loop operation for recycling all the reactants besides the water,
H , and O2. Then the membrane technology was developed to enhance the decomposition efficiency. The
2

maximum attainable one-pass conversion rate of HI exceeds 90% by membrane technology, whereas the
equilibrium rate is about 20%.

Key words: high temperature gas-cooled reactor; hydrogen production; thermochemical; water splitting

Among the various kinds of nuclear reactors, the

Introduction
The oil crisis in the 1970s emphasized the need for
new energy sources which are now even more urgent
due to the global environmental issues widely recog-
nized in the world in recent years. Nuclear and hydro-
gen energy sources are promising candidates for new
primary and secondary energy sources . [1]
Hydrogen is a flexible energy carrier which can be
stored in a variety of ways and transported over long
distances with low transportation costs. Hydrogen can
be used as a fuel in a wide variety of industrial applica-
tions and can be transformed into electricity in fuel
cells. Hydrogen is very clean in the sense that water is
the only product after burning. Therefore, it will play
an important role in future clean energy systems .
Received: 2003-08-02; revised: 2004-02-27
* Supported by the Ministry of Culture, Sport, Science, and Tech-
nology of the Japanese Government
* * To whom correspondence should be addressed.
E-mail: xinxin@inet.tsinghua.edu.cn; Tel: 86-10-62789752
high temperature gas-cooled reactor (HTGR) provides
especially high-temperature heat at about 1000°C with
inherent safety. HTGRs will be used not only to supply
a large amount of future electricity generation but also
to supply energy in non-electrical fields. One potential
application of the HTGR is for hydrogen production,
which is regarded as one of the leading nuclear heat
utilization applications in non-electrical fields. There
are several potential hydrogen production processes
that can use nuclear heat. The final goal of the hydro-
gen production system using HTGR is to produce hy-
drogen without CO2 emissions. Hydrogen production
from water, rather than from fossil fuels, will result in
a carbon free energy system. With certain endothermic
[2] and exothermic chemical reactions, water can be split
at lower temperatures than that required for direct
thermal decomposition. However, the reactions must
be carefully selected so that the process results in water
splitting. The process works like a chemical engine to
produce hydrogen while absorbing high-temperature
 WU Xinxin ( ^ ^ 4 ^ ) et al: Thermochemical
heat and discharging low-temperature waste heat. The
iodine-sulfur (IS) thermochemical water decomposi-
tion process is one promising m e t h o d [3]
that can effi-
ciently and cleanly produce hydrogen with the high
temperature gas-cooled reactor as the heat source sup-
plying gas at 1000°C. The system offers a novel and
promising means for large-scale hydrogen production
from water.
1 Iodine-Sulfur Process Principles
The IS process consists of three chemical reactions .
xI +S02+2H20=2HL+H S04
2 2
2HI=I +H 2 2
2H S0 =2H 0+2S0 +0
2 4 2 2 2 (3)
The principles for the IS process are illustrated in
Fig. 1. The so-called Bunsen reaction, Eq. (1), is an
exothermic sulfur dioxide ( S 0 ) gas absorbing reaction,
2
which proceeds spontaneously in the liquid phase at
20-100°C. Gaseous sulfur dioxide reacts with iodine
and water producing an aqueous solution of hydriodic
0.5O 7
f (3) H S 0
1000
800
600 -
400 -
200
(1) Bunsen reaction
Since the thermochemical process is an energy con-
version process, the energy conversion efficiency is an
important factor. The efficiency is usually defined as
the ratio of the higher heating value of hydrogen to the
net external heat input to the process and is referred to
as the "thermal efficiency" . [5]
Based on the higher heating value of hydrogen, the
IS process thermal efficiency, assuming maximum heat
recovery, should be more than 4 5 % for optimum oper-
ating conditions, which is expected to be competitive
with the conventional water-splitting cycle .
R & D on the IS process has been carried out inten-
sively at the Japan Atomic Energy Research Institute
Water Splitting for Hydrogen Production 271
acid and sulfuric acid. The two kinds of acids are sepa-
rated by liquid-liquid phase separation in the presence
of excess iodine. The hydrogen iodide (HI) decomposi-
tion reaction, Eq. (2) , produces hydrogen with a low
endothermic heat of reaction at 300-500°C in the gas
phase. It can also be carried out in the liquid phase.
The sulfuric acid ( H S 0 ) decomposition reaction, Eq.
2 4
(3), is an endothermic oxygen producing reaction,
which proceeds in two stages. First, gaseous H S 0 2 4
decomposes spontaneously into H 0 and S 0 2 3 at 400-
500 °C and then SO3 decomposes into S 0 2 and 0 2 at
[4]
about 800°C in the presence of a solid catalyst. By car-
(1) rying out these three reactions sequentially, the net re-
(2) sult is that the water is decomposed into hydrogen and
oxygen. The process has the attractive characteristics
that all the chemicals circulate in the process as fluids
and the sulfuric acid decomposition proceeds
stoichiometrically with a high conversion ratio and
large entropy change at temperatures suitable for utiliz-
ing the nuclear heat supplied by an HTGR.
2 4 decomposition
(endothermic)
H 0 (Raw material)
2
Fig. 1 IS process concept
(JAERI) since the beginning of the 1980s. The IS
process research has focused on three fields : a) de- [6]
velopment of a closed-loop system, b) improvement of
the process to attain high thermal efficiency, and c)
development of construction materials.
Closed-loop operation means that the process mate-
rials other than water, hydrogen, and oxygen circulate
in the process without loss, which is an important ad-
vantage of the thermochemical cycle. The aim of the
research on the closed-loop operation was to demon-
[1]
strate the chemical feasibility of continuous hydrogen
production in a laboratory-scale IS process. The
closed-loop tests sought to produce hydrogen and
 272 Tsinghua Science and Technology, April 2 0 0 5 , 10(2): 2 7 0 - 2 7 6

oxygen with the stoichiometric ratio of the water de- Most of the research concentrated on the environment
composition reaction and to maintain a stable process for the sulfuric acid decomposition step to develop cor-
solution in composition . The process was improved
[7]
rosion resistant materials for the very corrosive
by modifying the HI processing scheme using mem- environment^.
brane technology . In this scheme, electrodialysis was
[8]

used to concentrate the HI^ solution from the Bunsen 2 Laboratory-Scale Tests
reaction step with a membrane reactor equipped with
A simplified flowsheet of the IS process is shown in
hydrogen permselective membranes that were used to
Fig. 2 with a schematic flowsheet of the laboratory-
enhance the single pass conversion of HI. Since the IS
scale apparatus given in Fig. 3. The system was de-
process uses rather corrosive chemicals such as sulfu-
signed for hydrogen production with a capacity of 1-10
ric acid, hydroiodic acid, and iodine, the structural ma-
L/h and was made of quartz, glass, and Teflon.
terials for the scaled-up plant must be carefully chosen.
[ Ί Γ ~l I 1
1 I I |Q OUT I
2 I I I

[ I I \ I I I H out I ι

f
2

! Η,Ο ι ι FS Υ- I I I H , 0 IN Ι H [ ΠΓΙ I

I so ,o j -
2 2 ir ι ii ^— 1
ρ -ig I
!bal , Π !! * s II h \ V I
ι g-| " J L Ο h,so i 4 ι ^ΓΎ\ nso 2 ^-K Γ ^ Λ I

! t ! | « I S phase | | \ \\ Π g R H 2 ]

Τ ο ^ g η f ^ ι .phase , υ - : § »- •
ι ι a ι I ρ as 11 ι ι ι ι ι J 1

ι I _J Μ Μ ^ Γ Μ HI ι

H S 0 decomposition section
2 4 Bunsen reaction section HI decomposition section
Fig. 2 Simplified IS process flowsheet

0 2 H 0 2
H
2

I II I t Π Π χ τ η
ö n
0
ί ι 1 ι—"~ I
ΗΞ Ω Ι 1 ^ Ο JL ΓΗ
T3 εΛ s ^ ·* _ ΟΛ
<υ\/ -a Χ ^ η V

Β
Τ
I
if
11 II
S
X
A, I
VI J
Bunsen reactor^f^ J , ί
Vir
^ ( H S 0 phase)
2 4 „
(HI phase)
X ^
Liquid-liguid separator
Fig. 3 Schematic flowsheet of the laboratory-scale demonstration apparatus

The process was divided into three sections, the with high iodine concentrations, the two kinds of acids,
Bunsen reaction, the H S 0 decomposition, and the HI
2 4 H S 0 and HI, produced from the S 0 , I , and H 0
2 4 2 2 2

decomposition. In the Bunsen reaction step, Eq.(l) were spontaneously separated into two liquid phases.
 WU Xinxin ( ^ ^ 4 ^ ) et al: Thermochemical Water Splitting for Hydrogen Production 273

The light phase was mainly composed of H S 0 2 4 and

H 0 (the sulfuric acid phase). The heavy phase was
2

mainly composed of polyhydroiodic acid ( H Q , I , and 2

H 0 (the HI phase). The two solutions were then sepa-

2

rated in the separator. The H S 0 2 4 phase was then fed

to the H S 0 2 4 decomposition process where oxygen
was produced by concentrating, vaporizing, and de-
composing the H S 0 2 4 with Pt or F e 0 3 catalysts. The
2

50
products were then purified in the purification step. 20 30
t/h
The HI phase was fed to the decomposition section Fig. 4 Production rates of H and 0 2 2 products
with the residual sulfuric acid in the HI phase com-
pletely separated in the HI^ purifier. The reverse
Bunsen reaction then occured with the sulfur com-
pounds separated as gaseous S 0 2 and recycled to the
Bunsen reactor. The purified H I - I aqueous solution 2

was fed to the HI^ distillation column with azeotropic

hydriodic acid (HIaz) and I as the separated compo-
2

nents in the column. The H I ^ solution was fed to the

HI decomposer, Eq. (2), for decomposition to H and I 2 2

using Pt or an active carbon catalyst. With the excep-

tion of the hydrogen and oxygen, all the products in the Fig. 5 Fluctuations in the I 2 composition in the
sulfuric acid phase
H S0
2 4 and HI decomposition sections were reused in
the Bunsen reaction section as the reactant materials. of the process control [10]
. A n important example was
Every operation was carried out at atmospheric pres- the distillation of the HI^ solution and the related
sure. The products, hydrogen and oxygen, were trans- transport of process solutions. Figure 6 shows a simpli-
ferred by nitrogen carrier gas from the reactor to the fied flowsheet around the distillation column. Two
analysis section, where the composition and flow rates kinds of solutions flow out of the distillation column,
were measured. The production rates were then com- I solution from the bottom and HI solution from the
2

puted from those measurements.
The demonstration experiments were repeated sev-
eral times. In the longest run, the hydrogen production
was maintained for 48 h at the rate of 1 L/h with con-
tinuous and stable operation. The amounts of hydrogen
and oxygen produced are shown in Fig. 4
fluctuations of the iodine concentration in the sulfuric
acid phase extracted from the liquid phase separator
are shown in Fig. 5 . The hydrogen and oxygen were
[7]
top. If the flow rates of these solutions changed due to
fluctuations of the operating conditions, the changes
would have affected the composition of the liquid-
liquid phase-separated solution, which might have
completely disrupted the entire process due to the
[ 7 ]
and the
HI solution

produced at a 2-to-l ratio equal to the stoichiometric

ratio for water decomposition. The composition fluc-
tuations in the process solution were controlled within
Liquid-liquid
±3%. separator
For a stable production rate, the temperatures and Bottom solution (I ) 2

fluid transport rates must be stable. Since in the IS

H S 0 phase
process, every reactant except water is recycled into 2 4

HI phase
the process, the feedback effect of fluctuations in the
Recycle HI solution
operating conditions is very important. One result of Fig. 6 Simplified flowsheet around the distillation
the demonstration test was improved understanding column
 274 Tsinghua Science and Technology, April 2 0 0 5 , 10(2): 2 7 0 - 2 7 6

termination of the phase separation process. These dis- membrane reactor was used to enhance the conversion
ruptions were avoided by considering the effects of the ratio of the HI decomposition reaction. The membrane
fluctuating flow rates on the phase separation behavior reactor uses a novel membrane to separate the hydro-
as shown in Fig. 7. gen from the H -HI-I -H 0 gaseous mixture at 300-
2 2 2

500 °C. Because of the thermal and chemical stability

Disappearance of Disappearance of of ceramics, a ceramic hydrogen permselective mem-
phase separation phase separation
brane was developed using chemical vapor deposi-[11]

tion of silica onto a porous alumina. The membranes

A Bottom solution
had a H t o HI permeance ratio higher than 150, with
\ X Recycle HI solution
2
(l2)

the H permeance on the order of 10~ mol/ 7

Δ\ /
2
(HI, H 0 )
(Pa · m · s). The H permeance was stable throughout
2

a 24-h separating a H -HI-I -H 0 gaseous mixture at

2 2 2

Precipitation Precipitation 450 °C. A preliminary analysis of the membrane reactor

of I, of I 7
based on these figures suggested that the maximum at-
-100 -50 0 50 100 150
tainable one-pass conversion rate of HI exceeded 90%,
Flow rate fluctuations (%)
whereas the equilibrium rate was about 20% . [12]

Fig. 7 Effect of the fluctuation o f t h e flow r a t e s o n t h e

11
i o d i n e c o n c e n t r a t i o n (f )
u of the phase-separated sulfu- HI H 9

ric a c i d

The results of the demonstration experiments suc-

cessfully confirmed the feasibility of the closed-loop
operation for the IS process.

3 Process Improvements HI solution

V

The HI decomposition section produced hydrogen by

I,
purification of the HI^ solution from the Bunsen reac-
Distilling ΗΙ solution
tion, distillation of the purified HI* solution, decompo-
Λ

Fig. 8 Hl x processing scheme using membrane

sition of the hydrogen iodide, and finally separation of technology
the hydrogen from the reaction mixture. The process
must include an efficient scheme for the distillation of
HI from the HI* solution and for the HI decomposition. 4 Materials Development
For the HI distillation, the azeotropic conditions for the
Special structural materials were needed since the
HI-I -H 0 solution in conventional distillation col-
2 2
process fluid presented very corrosive environments. A
umns for distilling the HI-H 0 mixture would require a
2
preliminary screening of commercial materials to se-
large amount of energy. For the HI decomposition, the
lect materials for the corrosive IS process environment
low equilibrium conversion ratio of the HI decomposi-
focused on Fe-Si alloys and ceramics since boiling sul-
tion reaction requires a large amount of energy input to
furic acid is the most severe environment in the IS
recycle the excess amount HI. As a result, energy con-
process . [13]

sumption in the HI decomposition section would lower

With the Fe-Si alloys, the Si content was found to
the process efficiency.
control the corrosion rate. The critical Si content for
An alternative scheme was then developed using a
passivation was 9 wt.%-10 wt.% in boiling 95 wt.%
membrane separation technique which is illustrated in
sulfuric acid. In alloys with higher Si content, the cor-
Fig. 8. Electrodialysis was used to concentrate the HI*
rosion rate decreased sharply. The corrosion resistance
solution to a composition higher than azeotropi, with
was due to the formation of a compact passive film
the feasibility verified at room temperature and rela-
composed of S i 0 and SiO. However, alloys with high
2
tively low iodine concentrations. Pure HI was obtained
Si content are brittle, and thus, a lower Si content is
by distilling the concentrated HI solution. The
 WU Xinxin ( ^ ^ 4 ^ ) et al: Thermochemical
desired to improve ductility for reliable service. There-
fore, a hybrid was developed with a ductile metal sub-
strate and a corrosion resistant surface layer. The mate-
rial was fabricated using chemical vapor deposition
(CVD) with Fe-3 wt.% Si as the substrate and SiCl [9]
as the reactant gas to form a Si-enriched composition-
ally graded layer on the substrate surface. The Si con-
tent on the top surface was 14 wt.% with a layer thick-
ness at about 70 pm. An immersion test in boiling
95wt.% sulfuric acid for 300 h showed no detectable
mass change and demonstrated good corrosion resis-
tance of the surface layer. However, microscopic
analysis of the surface identified corrosion grooves in
micro-cracks formed during the CVD treatment. The
CVD conditions were then modified to prevent crack
formation. In addition to the Fe-Si alloy study, the cor-
rosion behaviors of Si containing ceramics were also
studied in the same environment to search for another
candidate material. The tested ceramics were A1 0 ,
Z r 0 , SiC, S13N4, and SiSiC, with the tests focusing on
2
the effects of the environment on the chemical stability
and mechanical properties of the materials. In the first
four materials, binding components were added during
fabrication while with SiSiC, the Si worked as the
binder. In 1000 h corrosion tests in boiling 95wt.%
sulfuric acid, no significant mass loss of the ceramics
was detected except for the Z r 0 that was severely cor-
2
roded. However, scan electron microscope pictures
showed dissolution of the binding components on the
surfaces of the A1 0 , SiC, and Si N test specimens.
2 3 3 4
Therefore, the bending strengths of these ceramics
would change with increased immersion time. Only the
SiC showed that bending strength increased due to the
formation of a protective surface layer composed of
amorphous silica. The bending strength of SiSiC did
not change. These results confirmed the excellent cor-
rosion resistance of SiC and SiSiC, and identified the
mechanism for the corrosion resistance. These ceram-
ics could be used in various kinds of equipment such
as valves and pumps and also as lining materials for
the reactors and pipes. Other ceramic materials includ-
ing S13N4 are unsuitable for the IS process due to their
mass loss and/or decreased mechanical strength.
The S O 3 decomposition reaction requires the highest
temperature environment in the IS process. The corro-
sion behaviors of candidate materials for this environ-
ment are currently under careful consideration,
Water Splitting for Hydrogen Production 275
including changes in important mechanical properties
such as the creep behavior.
5 Conclusions
4
Hydrogen will play an important role in future clean
energy systems. Hydrogen production from water,
rather than from fossil fuels, will result in a carbon free
energy system. The iodine-sulfur thermochemical wa-
ter decomposition process is one promising method
that can efficiently and cleanly produce hydrogen with
a high temperature gas-cooled reactor. The IS process
research has focused on three fields, including closed-
loop operation, process improvements, and construc-
tion materials.
1) The demonstration experiments of hydrogen pro-
duction were carried out with an IS process in the labo-
ratory. The experimental results successfully confirm
the feasibility of the closed-loop operation for the IS
2 3
process.
2) In order to increase the IS process efficiency for
hydrogen production, the process should include an ef-
ficient scheme for the distillation of HI from the HI*
solution and for the HI decomposition. An alternative
scheme was developed using a membrane separation
technique. A preliminary analysis of the membrane
reactor shows that the maximum attainable one-pass
conversion rate of HI exceeds 90%), whereas the equi-
librium rate is about 20% . [12]
3) Special structural materials were investigated for
boiling sulfuric acid, the most severe environment in
the IS process. The corrosion behavior of candidate
materials for this environment are currently under care-
ful consideration including changes in important me-
chanical properties such as the creep behavior.
References
[1] O n u k i K , N a k a j i m a H , I o k a I. I S p r o c e s s for t h e r m o c h e m i -
cal h y d r o g e n p r o d u c t i o n . JAERI Review, 1 9 9 4 , 6: 1-33
[2] F u n k J E , R e i n s t r o m R M . E n e r g y r e q u i r e m e n t s in t h e p r o -
d u c t i o n o f h y d r o g e n f r o m w a t e r . I&EC Process Design
Development, 1966, 5(3): 336-342.
[3] N o r m a n J H , B e s e n b r u c h G E , B r o w n L C, A l l e n C L .
Thermochemical water-splitting cycle, bench-scale investi-
gations and process engineering. Final Report on IS proc-
ess. G A - A 1 6 7 1 3 , 1 9 8 2 .
[4] O n u k i K , N a k a j i m a H , F u t a k a w a M , I o k a I, S h i m i z u S.
 276 Tsinghua Science and Technology, April 2 0 0 5 , 10(2): 2 7 0 - 2 7 6

T h e r m o c h e m i c a l w a t e r - s p l i t t i n g for h y d r o g e n p r o d u c t i o n . [9] F u t a k a w a M , O n u k i K, Ioka I, Nakajima H, S h i m i z u S.

In: 8 t h
International Topical Meeting on Nuclear Reactor C o r r o s i o n t e s t of c o m p o s i t i o n a l l y g r a d e d F e - S i alloy in
Thermal-Hydraulics. Kyoto, Japan, Sept. 30 -Oct.
th
4 ,
t h
b o i l i n g sulfuric acid. Corrosion Engineering, 1997, 46:
1997, 3 : 1 8 0 3 - 1 8 0 8 811-819.

[5] S a t o S. T h e r m o c h e m i c a l H y d r o g e n P r o d u c t i o n . Oxford: [10] S a k u r a i M , N a k a j i m a H , O n u k i K, I k e n o y a K, S h i m i z i S.

P e r g a m o n Press, 1979: 81-114. P r e l i m i n a r y p r o c e s s a n a l y s i s for t h e c l o s e d c y c l e o p e r a t i o n

[6] S h i m i z i S, N a k a j i m a H , K u b o S, O n u k i K. A s t u d y o n t h e o f t h e iodine-sulfur t h e r m o c h e m i c a l h y d r o g e n p r o d u c t i o n

t h e r m o c h e m i c a l I S p r o c e s s . In: 1 s t
Information Exchange p r o c e s s . Int. J. Hydrogen Energy, 1999, 24: 603-612.

M e e t i n g o n N u c l e a r P r o d u c t i o n of H y d r o g e n . F r a n c e : P a r i s [11] H w a n g G J, O n u k i K, S h i m i z u S. S e p a r a t i o n of h y d r o -
( O E C D ) , Oct. 2 - 3 , 2 0 0 0 .
n d r d
gen from a H - H 0 - H I 2 2 g a s e o u s m i x t u r e u s i n g a silica

[7] N a k a j i m a H , S a k u r a i M , I k e n o y a K, H w a n g G J, O n u k i K, m e m b r a n e s . AIChE. J. , 2 0 0 0 , 4 6 ( 1 ) : 9 2 - 9 8 .

S h i m i z u S. A s t u d y o n a c l o s e d - c y c l e h y d r o g e n p r o d u c t i o n [12] H w a n g G J, O n u k i K , S h i m i z u S. S i m u l a t i o n s t u d y o n t h e
b y t h e r m o c h e m i c a l w a t e r - s p l i t t i n g I S p r o c e s s . In: 7 t h
Inter- c a t a l y t i c d e c o m p o s i t i o n of h y d r o g e n i o d i d e in a m e m b r a n e
national Conference on Nuclear Engineering. Tokyo, April r e a c t o r w i t h a silica m e m b r a n e for t h e t h e r m o c h e m i c a l w a -
19-23,1999: ICONE-7104. t e r - s p l i t t i n g I S p r o c e s s . J. Membrane Scis., 2 0 0 1 , 194:

[8] H w a n g G J, O n u k i K, S h i m i z i S, O h y a H . Hydrogen 207-215.

separation in H - H 0 - H I
2 2 gaseous mixture using silica [13] N i s h i y a m a N , F u t a k a w a M , I o k a I, O n u k i K, S h i m i z u S.
membrane prepared by chemical vapor deposition. J. C o r r o s i o n r e s i s t a n c e e v a l u a t i o n o f brittle m a t e r i a l s in b o i l -
Membrane Sei., 1 9 9 9 , 162: 8 3 - 9 0 . i n g sulfuric acid. J. Soc. Mat. Sei, Japan, 1999, 4 8 : 7 4 6 -
7 5 2 . (in J a p a n e s e )