5.21

Graphite

J. Fachinger
Furnaces Nuclear Applications Grenoble ZU Hanau Research and Development, Hanau, Germany

ß 2012 Elsevier Ltd. All rights reserved.

5.21.1

Introduction

540

5.21.2
5.21.2.1
5.21.2.2
5.21.2.3
5.21.2.4
5.21.2.5
5.21.3
5.21.4
5.21.4.1
5.21.4.1.1
5.21.4.2
5.21.4.2.1
5.21.4.2.2
5.21.5
5.21.5.1
5.21.5.2

5.21.5.3
5.21.5.4
5.21.6
5.21.6.1
5.21.6.2
5.21.6.3
5.21.6.4
5.21.7
5.21.8
References
Appendix 1

Amounts of i-Graphite and Its Origin
Russia
United Kingdom
France
United States
Others
Retrieval of i-Graphite
Graphite Properties
Physical Properties
Wigner energy
Chemical Properties
Oxidation in gaseous phases
Graphite reactions with liquids
Graphite Radioactivity
Formation of 3H
Formation of 14C
Formation of 36Cl
Diffusion of Radionuclides in Graphite

Graphite Treatment for Disposal or Recycling
Waste Packages and Encapsulation
Thermal Treatment
The Russian ‘Self-Propagating High-Temperature Synthesis SHS’
Recycling of i-Graphite
Final Disposal
Summary

540
540
541
541
541
542
542
542
542
543
543
543
545
545
546
546
547
549
550
550
551
552

553
553
556
557
558

Amounts of Irradiated Graphite in Different Countries

Abbreviations
AGR
AM-1
AMB
AVR
b
BEPO
CP1
EDF
EL2
FRJ-1

Advanced gas-cooled reactor
Prototype of RBMK
Aтoм Mиpный Бoльшoй
Allgemeiner Versuchsreaktor (a small
HTR prototype reactor in Germany)
Barn (10À24 cmÀ2)
British experimental pile ‘0’
Chicago pile-1
E´lectricite´ de France SA
Graphite moderated test pile in

France
Research Reactor Ju¨lich 1

GLEEP

Graphite low energy experimental
pile
HTGR
High-temperature gas-cooled reactor
HTR
High-temperature reactor
IAEA
International Atomic Energy Agency
MAGNOX Magnesium alloy graphite moderated
gas cooled uranium oxide reactor
RBMK
Reaktor Bolschoi Moschtschnosti
Kanalny
THTR
Thorium high-temperature reactor
UNGG
Uranium naturel graphite gaz reactor
WAGR
Windscale’s advanced gas-cooled
reactor

539


540


Graphite

5.21.1 Introduction
Graphite has been used in nuclear technology since
the birth of this technology. The first artificial nuclear
reactor, the Chicago Pile-1, consisted of a pile of
uranium and graphite.
It was the fundament for future developments in
the different graphite-moderated nuclear power
reactors such as the Uranium Naturel Graphite Gaz
reactors (UNGG) in France, Magnox and advanced
gas-cooled reactors (AGR) in the United Kingdom,
or RBMK in Russia. The culmination of this development was the high-temperature reactor, for example, the Fort Saint Vrain reactor in the United States
or the Thorium-Hochtemperaturreaktor (THTR) in
Germany.
Worldwide, more than 230 000 tons of irradiated
graphite (i-graphite) exist, which will eventually
need to be managed as radioactive waste.1 The major
part of i-graphite is still in operational or shutdown
nuclear power reactors. Actually, the reactor cores
have been removed from Fort Saint Vrain, GLEEP,
and BEPO. The removal of the core from the Allgemeiner Versuchsreaktor (AVR) is proposed for the next
few years.
Smaller quantities of i-graphite result additionally from operation, in the form of graphite sleeves,
which have to be replaced during operation, and
from different kinds of research reactors. The total
amount of i-graphite is assumed to be in the range of
220 000–250 000 tons. A more detailed overview of
the amount of i-graphite is given in Section 5.21.2.

Graphite changes its properties during irradiation
in a nuclear reactor. Most important for the treatment and disposal of i-graphite is the possibility
of storing energy in the form of structural defects,
the so-called Wigner energy. The entire Wigner
energy could be released rapidly after an initial
local release. Furthermore, the graphite will be contaminated by radionuclides. They result from the
activation of 13C and impurities in the graphite
matrix as well as from the depletion of fission products released from the fuel elements. This is
described in Section 5.21.5.
Various treatment methods have been developed
or are under development to decontaminate i-graphite
or to optimize the disposal volume and behavior,
respectively. Section 5.21.6 gives a short overview
of all these developments. A major issue is the establishment of a close graphite cycle, which is essential
for the future development of graphite-moderated
nuclear reactors.

The last section is dedicated to the final disposal
options for i-graphite and the behavior of i-graphite
under different disposal conditions.

5.21.2 Amounts of i-Graphite and
Its Origin
Generally, one can distinguish between four different
types of nuclear reactors that utilize graphite as neutron moderator and reflector.
 Aircooled graphite piles with a low power density
as test facilities, prototypes, and first-generation
plutonium production reactors.
 Carbon-dioxide-cooled reactors (Magnox and
UNGG) for electricity supply and/or plutonium

production.
 Helium-cooled high-temperature reactors for
electricity generation and process heat generation.
 Graphite-moderated water-cooled reactors for plutonium production and/or electricity generation.
The last feature, electricity generation, has been
optimized especially in the Russian RBMK reactors.

5.21.2.1

Russia

The main sources of i-graphite are the RBMK nuclear
power plants as well as high-capacity plutonium production reactors. Five RBMK power plants with 11
reactors are still in operation in Russia.2,3 The original
license foresaw a lifetime of 30 years. However, lifetime extensions are already licensed or envisaged.
Therefore, the first shutdowns are expected in 2013,
with a replacement program starting in 2015.3
The amount of graphite from these reactors is
given in Table A.1. An important fact about this
graphite-moderated water-cooled reactor type is
that a helium–nitrogen mixture gives the graphite
moderator a protective atmosphere, which will
have an important impact on the generation of 14C.4
Four more graphite-moderated power reactors,
Bilibino-1–4, are in operation. These dual purpose
reactors for electricity and heat contain 133 tons of
graphite each.3
The AMB-1 and -2 in Beloyarsk, and the AM-1,
prototypes of the RMBK reactors, were shutdown in
1981, 1989, and 2004 respectively. The fuel has been

removed from the reactor core and stored in cooling
basins. The reactor units with the graphite core are
under safe storage conditions of IAEA Stage I, under
surveillance; further dismantling is planned.


Graphite

Besides these graphite-moderated power plants,
13 high-capacity plutonium power reactors had
been in operation in Russia. All of them were shutdown between 1987 and 2008. The reactor units with
the graphite core are under safe storage conditions of
IAEA Stage II. The first dismantling concepts proposed the transformation of the reactor shafts into a
final radwaste repository. However, this disposal concept is not in accordance with the new Russian waste
disposal regulations.5 Therefore, a further 21 000 tons
of graphite internals of the reactor units, as well as
8000 tons of graphite rings at on-site storage facilities,
have to be managed as radioactive waste in Russia.
An additional source of i-graphite are research
reactors and other critical assemblies. The determination of the exact number of these reactors and
assemblies is complicated. However, it can be assumed
that more than 110 such facilities exist in Russia.
Furthermore, it was not possible to figure out which
moderator had been used in these facilities and therefore the amount of i-graphite could not be evaluated.
5.21.2.2

The first UNGG were built and operated by CEA
in Marcoule as plutonium production reactors for the
French nuclear deterrent forces. The main characteristics of these reactors are given in Table A.3. While
the G1 was still air cooled, all other UNGG used CO2

as cooling gas. The graphite bricks, used as moderator
as well as shielding for the internal walls of the reactor,
were mounted to a horizontal reactor core. The
decommissioning has achieved IAEA level 2.
Six more UNGG reactors have been built and
operated by EDF for electricity production. The
main characteristics of these reactors are also given
in Table A.3. All of them are under decommissioning. With the exception of Chinon A1 the design
was changed from a horizontal to a vertical shaft for
improved fuel handling. However, the fuel cartridges had to be protected by graphite sleeves to
withstand the mechanical forces of the upper fuel
cartridges. These graphite sleeves are temporarily
stored in silos except those from Buggy, which have
been disposed at the final disposal sites at Manche
and Aube.

United Kingdom

The United Kingdom has the largest amount of
i-graphite that has to be managed as radioactive
waste1,6 because most of the British nuclear power
plants are gas-cooled graphite-moderated reactors as
opposed to those in other countries, which utilize
water-moderated reactors as an alternative. The
Magnox reactor type was utilized after test reactor
and prototype development in the late 1950s. The
name was derived from the fuel cladding made from a
magnesium–aluminum alloy. The last Magnox reactor was commissioned in 1971. The next generation
of gas-cooled reactors were the AGR, commissioned
between 1976 and 1989. Both reactor types were

graphite-moderated and cooled with CO2. Totally
about 80 000 tons of graphite have to be handled as
radioactive waste now or in the next two decades after
the shutdown of the still operational AGR reactors.
An overview of the amount of graphite in the different
UK reactors is given in Table A.2 in the Appendix.
5.21.2.3

541

France

The first graphite-moderated reactor in France was
the pile EL2 built at Saclay in 1952.7,8 It was an
experimental reactor like the EL3 built in 1957.
The total mass of graphite in these two reactors was
109 tons. The operation of these reactors ended in
1965 and 1979, respectively.

5.21.2.4

United States

The development of nuclear reactors started in the
United States with the graphite-moderated pile 1
(CP1) in Chicago. The largest amounts of i-graphite
in the United States are from the plutonium production reactors at the Hanford site. The Hanford
B-Reactor was the first large-scale plutonium production reactor in the world. The reactor was graphite
moderated and water cooled. It consisted of an
8.5 Â 11 m horizontal tube and contained 1100 tons

of graphite.9 All reactors at the Hanford site were
intended for plutonium production. In all, nine plutonium production reactors were operated at the Hanford site. The reactors were shutdown between 1964
and 1971 after an average life span, except the lastbuilt N Reactor (1963), a dual purpose facility for civil
electricity generation (shutdown in 1987). Most of the
reactors have been entombed after defueling and act
as interim storage for the graphite moderator and
structural materials to allow the decay of radioactive
material until dismantling is possible with a low dose
risk. Reactors exclusively for civil application were
the HTGR at Peach Bottom and Fort Saint Vrain
with a graphite block as moderator. Both were shutdown after a relatively short operational time of 7 and
13 years, respectively. An overview of the amount of
graphite in the different reactors is given in Table A.4
in the Appendix.


542

Graphite

5.21.2.5

Others

Table A.5 gives an overview of graphite-moderated
reactors in other countries. Most of them are RBMK
reactors in countries of the former Soviet Union or
Magnox reactors. High-temperature reactors have
been constructed in China, Germany, and Japan.
Two graphite-moderated high-temperature reactors,

the AVR and the THTR, were operated in Germany.
About 1000 metric tons of i-graphite and irradiated
carbon bricks have to be managed as radioactive
waste. Furthermore, about 1 million irradiated fuel
pebbles were produced during the operational time
of these reactors. They consist mainly of a graphite
matrix that contains about 10 000 so-called coated
particles (see Chapter 3.06, TRISO Fuel Production and Chapter 3.07, TRISO-Coated Particle
Fuel Performance). These particles safely enclose
the nuclear fuel and the major part of the fission
products. The AVR is under dismantling to the
green field. Therefore the reactor vessel with the
graphite core will be pulled out as one piece and
transferred to an interim storage facility. It will stay
there for about 80 years before further treatment.
The development in Japan is based on a graphite
block core similar to that in the United States. The
Chinese HTR-10 is a pebble HTR like the one in
Germany.

whole reactor core was flooded before cutting the
graphite internals. This procedure provides two advantages. First, the water acts as shielding, which minimizes the dose rate of the employees, and second, the
water prevents the formation of graphite dust. However, water purification requires additional effort.
A third approach is being used for dismantling the
AVR in Ju¨lich, Germany. The whole reactor vessel,
including the graphite internals, built as one part, has
been lifted out of the reactor and transferred to an
interim storage. Before lifting, the reactor core was
filled with light concrete to consolidate the internals
and to reduce impacts, in case the vessel falls down

during the lifting procedure. Further dismantling will
be performed after the decay of the main part of
g-emitters, especially 60Co. This method represents
a way between complete dismantling and long-term
safe storage and enables fast cleanup of a site.
The choice of the best applicable retrieval method
depends on several site-specific facts and therefore
there is no ‘best procedure.’ They include the mechanical and physical properties of the graphite, the dose
rate of the graphite, and the surrounding structures, as
well as the specific side constructions, which determine
the space available for the installation of equipment.
A detailed overview of retrieval and available procedures and tools is under compilation by an expert team
of the European Carbowaste project.10

5.21.3 Retrieval of i-Graphite
5.21.4 Graphite Properties
The retrieval of i-graphite is based on two main
principles, dry and wet retrieval.10 Dry retrieval has
been chosen for the decommissioning of WAGR and
GLEEP. GLEEP was a low-energy and low-radiation
test reactor. The resulting total activity was so low
that the graphite could be removed manually without
shielding. Only protective overalls and gloves were
required to avoid incorporation. The graphite should
be treated in an industrial incinerator that is licensed
for the discharge of small amounts of radioactivity. It
was noted that the graphite blocks showed only small
effects after treatment at 1400 K for 3 h in air. Less then
2% of the graphite was lost during the process. However, about 87% of 3H and 63% of 14C were released
from the graphite.

The activity and dose rate were so high that a
manual retrieval of the graphite was not acceptable
for the WAGR. A remote removal system was developed for the retrieval of the graphite stack.
Wet retrieval was utilized for dismantling the hightemperature reactor at Fort Saint Vrain. Therefore, the

5.21.4.1

Physical Properties

The properties of graphite are related to its polycrystalline structure (see Chapter 2.10, Graphite: Properties and Characteristics). Graphite crystallites are
built by graphite planes, which are loosely bound by
van der Waal’s forces. The single planes consist of
carbon six rings with a sp2 electron configuration of
the carbon atoms and a dislocated p-system on both
sides of the plane. Therefore the properties of the
crystallites are anisotropic with respect to the orientation parallel or vertical to the planes. Irradiation
induces damages in these graphite crystallites, which
lead to anisotropic effects in the crystallites. A good
example is a radiation-induced expansion in one
direction and shrinkage in the other direction. Therefore the macroscopic changes can be anisotropic, if
the crystallites have an overall preferred orientation
direction, or isotropic, if the crystallites are randomly
distributed. This depends on the shape of the crystallites as well as on the production process. For example,


Graphite

extruded graphite shrinks parallel to the extrusion
direction and expands perpendicular to the extrusion direction at temperatures below 300  C and
shrinks in both directions at higher temperatures.

More isotropic molded graphite initially exhibits
shrinkage in all directions under all irradiation conditions. The irradiation-induced shrinkage proceeds to a
point of maximal density. Further irradiation causes an
expansion to the original density and beyond. Besides
being important for reactor operation, this effect is
also a key issue for waste management. It affects the
mechanical stability, which has a large influence on the
retrieval of graphite piles from the reactor for decommissioning. Furthermore, the density and porosity may
influence the radionuclide release in intermediate storage and especially in final disposal.
Another important parameter for disposal is the
reduction of the thermal conductivity. Small amounts
of fast neutrons will reduce thermal conductivity
and can be decreased further by 2 orders of magnitude to 2 W mÀ1 KÀ1, depending on neutron dose
and irradiation temperature.
Another very important property for disposal is
porosity, which allows the penetration of aqueous
phases into the graphite matrix and therefore an
undisturbed transport of radionuclides through the
graphite matrix.
5.21.4.1.1 Wigner energy

The Wigner effect is named after its discoverer Eugene
Paul Wigner. This effect describes the displacement of
atoms in a solid caused by neutron irradiation, which
can occur in any solid. However, it has a special importance for solid moderator materials such as graphite.
An atom can be moved from its position in a
crystal lattice by collision with neutrons, if they
have energies above 25 eV. Therefore, high-energy
neutron, for example, 1 MeV, causes cascades of
damages with about 900 displacements in a graphite

moderator. Not all of the displacements lead to lattice
defects because the displaced atoms could also
fill lattice vacancies. Atoms that cannot be placed in
lattice vacancies remain as interstitial atoms between
the lattice planes and therefore they are associated
with a higher energy.11 When such an atom has
sufficient thermal energy, it is able to move to normal
lattice position and release excess energy if the position energy is higher than the energy required for the
return to a lattice place. If such a process has been
initiated, the hole-stored Wigner energy can be
released immediately and heat up a graphite pile.
This was the cause of the Windscale fire accident.12

543

Wigner energy can cause the following problems
related to the management of i-graphite:
1. Initiation of an uncontrolled release of Wigner
energy during retrieval.
2. Sawing or cutting of the graphite core can lead
to a local heat increase, which may lead to an
uncontrolled release of Wigner energy. Therefore
such operations should be performed with sufficient cooling.
3. Release of the Wigner energy during final disposal.
4. The temperature of the final disposal sites for lowand medium-level wastes is normally strictly limited. These limits depend on corrosion processes or
microbial degradation, and higher temperatures
may disturb the integrity of the disposal or increase
the reaction rates of release mechanisms. Therefore,
the Wigner energy should be dissipated before storage by annealing the graphite at temperatures above
250  C or it has to be demonstrated that the disposal

site will not be affected by such an energy release.
This has been tried by NIREX but they concluded
that Wigner energy is not adequately understood to
guarantee that a release of Wigner energy cannot
affect the safety of a disposal site.13
Despite this potential risk, only low amounts of
i-graphite have considerable amounts of Wigner
energy. They are related to reactors operated at low
temperatures, for example, reflectors of material test
reactors. High reactor operation temperatures, for
example, achieved in an AGR would directly cause
the annealing of the graphite.13
5.21.4.2

Chemical Properties

‘Burning of radioactive graphite’ has been in public
discussion since the accident at Chernobyl. But graphite has an extremely low chemical reactivity, which
explains its geochemical stability, proved by the presence of natural graphite ores in the earth’s crust.
Graphite needs extremely powerful oxidation agents
to convert it into the thermodynamic-favored carbon
dioxide. This also allows the utilization of graphite
under extreme conditions in industrial processes, for
example, as electrode in arc melting at temperatures up
to 3000  C or its use as fire extinguisher.
5.21.4.2.1 Oxidation in gaseous phases

Graphite can react with gases such as O2, CO2, H2O,
and H2 at elevated temperatures and the temperature
depends on the perfection of the graphite’s crystal

structure14 and therefore on the amount of impurity.


544

Graphite

Generally, heterogeneous reactions involving a
porous solid and a gas can be controlled by one or
more of three idealized steps:
1. Mass transport of the reacting gas from gas stream
to the exterior graphite surface.
2. Mass transport of the reacting gas from the exterior surface to an active site and mass transport of
the products in the opposite direction.
3. Chemical reaction at the active sites.
The variation of the reaction rate with temperature for gas–carbon reactions can be divided into
three main zones (Figure 1).
In the low-temperature zone (zone I), the reaction
is controlled by the chemical reactivity of the solid
(step 3). There will be almost no concentration gradient of reacting gases throughout the whole volume
of the sample because of low reaction rate, and this
provides uniform access to the interior surface of
porous materials. For graphite – oxygen reaction,
the upper limit for temperature will be 500  C, and
for graphite–steam system, it will be 850  C.15
In the intermediate-temperature zone (zone II),
step 2 becomes important. The diffusion of reactants
in pores will influence the oxidation rate of material.
At higher temperatures, the concentration gradient of
the reacting gas becomes steeper within graphite and

the gas concentration becomes zero at a distance R
nearer the surface. The activation energy Ea in this
zone amounts to half of the true activation energy Et.
For graphite–steam reaction, this temperature region is
characterized by a temperature range of 850–1350  C
and graphite oxygen reaction, by 500–900  C.
In the high-temperature zone (>900  C for graphite oxygen and >1250–1400  C for graphite–water
steam) – zone III – the concentration of the reacting
gas is low at the exterior of the solid and the rate is

controlled by step 1. As bulk gas-transfer processes
have low activation energies, the apparent activation
energy for gas–carbon reactions in zone III is low.
The reactions occurring in the gas–graphite
system are
Reaction with oxygen
ðÁr H : standard enthalpy of reaction at 25 CÞ
½1Š

½2Š

Ln (reaction rate)

II

a

Ár H ¼ À283:0 kJmolÀ1

COðgÞ þ 1=2O2 ðgÞ!CO2 ðgÞ


½3Š

Reaction with carbon dioxide
Boudouard reaction :
CðsÞ þ CO2 ðgÞ ! 2CO

Ár H ¼ þ 172:5 kJ molÀ1
½4Š

The equilibrium can be shifted with increasing
CO pressure16,17 or in the presence of a catalyst.
Reaction with water
CðsÞþH2OðgÞ!COðgÞþH2ðgÞ

Ár H ¼þ131:3 kJ molÀ1
½5Š

COðgÞ þ H2 OðgÞ ! CO2 ðgÞ þ H2 ðgÞ
Ár H ¼ À213:7 kJ molÀ1

½6Š

The hydrogen and CO2 produced can then react
with carbon

C þ H2ðgÞ ! CH4 ðgÞ

b


Ár H ¼ À110:5 kJmolÀ1

CðsÞ þ 1=2O2 ðgÞ!COðgÞ

C þ CO2ðgÞ ! 2COðgÞ

III

Ár H ¼ À394:5 kJmolÀ1

CðsÞ þ O2 ðgÞ ! CO2 ðgÞ

Ár H ¼ þ172:5 kJ molÀ1

½4Š

Ár H ¼ À71:81 kJ molÀ1 ½7Š

The presence of hydrogen can shift reaction [5]
left [4].
Reaction with hydrogen

I
1/ T

Figure 1 Ideal reaction zones in graphite: I – reaction
rate is controlled by chemical reactivity of the sample;
II – reaction rate is controlled by diffusion in pores;
III – reaction rate is controlled by gas transport to the
exterior surface of the sample; a and b are transition zones.


C þ H2ðgÞ ! CH4 ðgÞ Ár H ¼ À71:81 kJ mol

½7Š

The mechanism and kinetics of these reactions are
described by Walker15.
The approximate relative rates of gas–carbon
reactions at 800  C and 0.1 atm. are given in Table 1.


Graphite

Table 1
Approximate relative rates of gas–carbon
reactions at 800  C and 0.1 atm. pressure
Reaction

Relative rate

C–O2
C–H2O
C–CO2
C–H2

1 Â 105
3
1
3 Â 10À3


In the literature, there are a number of investigations of nuclear graphite reactivity in different oxidation conditions. Results of oxidation of HTR-10
nuclear graphite IG-1118 exhibited three regimes:
400–600  C with an activation energy of 158.56 kJ
molÀ1, 600–800  C, at which the activation energy
was 72.05 kJ molÀ1, and the ‘third-zone’, over 800  C
regime with a very low oxidation energy. The comparison of reactivity of the two types of graphite used
in HTR in oxygen and air at 650–900  C (regime II)
leads to the conclusion that there is no difference in
the behavior of matrix graphite (A3-27) and standard
graphite V483T during oxidation.19 At the same
time, at a lower temperature (400  C, regime I) matrix
graphite is more reactive with respect to air. For the
temperature range 350–520  C, the activation energy
Ea for A3-3 graphite matrix amounts to 110 kJ molÀ1.20
The oxidation in air and moisture has to be considered for dismantling and interim storage, whereas
the reaction with humidity and aquatic phases is
important for final disposal.
Several investigations into virgin and irradiated
graphite have been carried out, mostly in air at
ambient pressure. A comprehensive review was
made by Stairmand,21 who concluded that significant
graphite oxidation can be excluded at temperatures
below 350  C.
However, graphite oxidation in air can occur in
high irradiation fields. Duwe et al.22 showed the consumption of oxygen and the production of carbon
dioxide during the interim storage of HTR fuel pebbles in sealed cans. But the dose of the irradiation
field from freshly irradiated fuel pebbles is normally
not relevant to the interim storage or final disposal of
i-graphite.
5.21.4.2.2 Graphite reactions with liquids


Graphite does not react with alkaline and acidic
solutions if no oxidizing agent is present. Dissolved
oxidants such as nitric acid, ozone, hypochlorides,
and hydrogen peroxide attack graphite to different
degrees.23–26 An important factor is the surface area,
which depends mainly on pore volume and pore size.

545

The reaction with oxidizing agents, for example,
concentrated nitric acid, leads finally to the evolution
of carbon dioxide:
C þ 4HNO3 ! 2H2 O þ 4NO2 þ CO2
However, different stable intermediate reaction
products can be formed: graphitic oxide (C7H2O4),
mellitic acid (C6(CO2H)6), and hydrocyanic acid
(HCN). The yield of these products and carbon
dioxide depends on the reaction conditions and the
nature of the graphite material.
Contact of i-graphite with aqueous phases during
interim storage or final disposal cannot be excluded
with an absolute certainty. In such a case, the oxidation of i-graphite depends mainly on the irradiationinduced production of highly reactive species by
radiolysis of the aqueous phase and the accessible
graphite surface. Corrosion experiments with A3-3
graphite show that the corrosion rate of graphite is
increased in final repository relevant aqueous phases
by external g-irradiation with a dose rate of $2 kGy
hÀ1.27 The obtained corrosion rates are in the range
from 10À5 down to 10À7 g mÀ2 day. High chloride

concentrations accelerate the graphite corrosion
probably by the formation of hypochlorides. This
clearly indicates that irradiation-induced corrosion
processes are relevant to the final disposal of graphite.
However, this is an extremely high dose rate not
relevant to the disposal of i-graphite. The first attempt
to determine the relation between the dose rate and
the corrosion rate was made in the framework of the
European RAPHAEL project.28 However, the low
number of performed measurements and the scattering of the obtained data did not allow the derivation of
a validated data set for such a correlation.

5.21.5 Graphite Radioactivity
The utilization of graphite in a reactor leads to two
different types of radioactive contamination in the
graphite material, the contaminants being
 Activated impurities in the bulk graphite material
 Radioactive isotopes occurring in the reactor circuit
The activation products are more or less homogenously distributed in the graphite, depending on
the original location of the impurities, as well as on
the possibility of their migrating in the graphite by
thermal gradients induced by the reactor conditions
and repulse effects during the activation process
itself. The radioactive isotopes from the reactor


546

Graphite


circuit are located (adsorbed) primarily at the graphite surface and migration into the bulk material
requires a transport force, which could be a thermal
gradient. The depleted isotopes have different
origins:
 Activation products of the coolant.
 Impurities in the coolant.
 Corrosion products from steel constructions of the
reactor distributed in the coolant and activated in
the reactor core.
 Release of fission products from fuel elements with
a cladding failure.
These different sources of the radioactive contamination indicate that the activities of i-graphite depend
on the reactor type, the type of the utilized virgin
graphite material, and the operational conditions of
the reactor. Therefore, even i-graphite of similar reactor types can show different contamination levels and
different isotope ratios and a detailed characterization
of i-graphite is required before retrieval from a specific reactor in addition to calculated radionuclide
inventories. A good example of such an approach was
the compilation of the so-called ‘Aktivita¨tsatlas des
AVR’29 which was calculated and validated on the
basis of radiochemical analysis of i-graphite from different locations in the reactor core.
However, a detailed consideration of the different
i-graphite materials from different reactor types
and different graphite types will not be helpful. Furthermore, the amount of detailed information would
definitely be out of the scope of this review and not
indicate the significant general problems of the waste
management of i-graphite.
The dose rate, one of the key parameters for the
retrieval and interim storage of i-graphite, depends
mainly on the 60Co activity. 60Co has a half-life of

5.3 years. The main source of 60Co in i-graphite is the
abrasion of fine metal parts from the pebble loop
system, which has been built up in the pipes by
neutron activation. Therefore, waiting for some
decay periods can be helpful to reduce the dose per
person for the workers at dismantling. Another
important parameter for retrieval is the release of
radionuclides into air. This could occur in the form
of contaminated dust, which can be handled by adequate exhausting methods. More problematic is the
release of tritium as gaseous component. However, it
must be ensured that information specific to the
reactor is retained during this period.
For final disposal, 14C and 36Cl have been identified as key nuclides with respect to the long-term

safety, due to their long half-life, mobility, and biocompatibility.
5.21.5.1

Formation of 3H

The radionuclide 3H, tritium has a half-life of
12.3 years. The contribution of radioactivity in nuclear
graphite resulting from tritium is significant.29–31 It is
produced by the following reactions:
 Fission reactions of uranium impurities in
the graphite and fuel cladding failure, such as
235
U(n,f) 3H reactions.
 Lithium impurities in the graphite via 6Li(n,a)
3
H reactions.

 3He (n,p) 3H in HTR reactors, which utilize
helium as coolant.
 10B (n,2a) 3H reactions in absorber rods (negligible
for designs without core rods).
The chemical properties of tritium are essentially
the same as those of ordinary hydrogen. Tritium
generated from lithium impurities is produced
mostly in graphite bulk. The release of tritium is
controlled by its diffusion out of the grain boundaries
and into the pore system.
5.21.5.2

Formation of 14C

Three routes, shown in Table 2, have to be considered for the formation of 14C. In the reactor core
materials, nitrogen is present only as an impurity,
whereas carbon and oxygen are in some cases major
constituent elements of the coolant, moderator, or
fuel. In spite of this fact, the 14N activation reaction
is usually more important for 14C production due to
its large cross-section. Therefore, the location and the
chemical form of nitrogen are important for the location of the formed 14C. Nitrogen levels vary widely
from 10 to 100 ppm in different reactor graphite
types30 and sometimes they are not known very precisely. A comprehensive study of 14C has been carried
out by Marsden et al.31 Calculations showed that
about 70% of the 14C originates from nitrogen impurities with an assumed amount of 50 ppm by weight.

Table 2

Activation reactions generation 14C


Reaction

Abundance of isotope
in natural element (%)

Capture crosssection (barn)

14

99.63
1.07
0.04

1.81
0.0009
0.235

N(n,p)14C
C(n,g)14C
17
O(n,a)14C
13


Graphite

Another source of 14C is the oxygen pathway from
the coolant. A birth ratio of 14C has been calculated
for an AGR from 17O:14N:13C as 25:21:1, assuming

50 ppm nitrogen.
Besides the level, the location of nitrogen impurity
in reactor core materials is an important parameter.
The nitrogen content of graphite is reduced during
manufacture by several high-temperature treatment
steps. However, the different heating and cooling processes cause the formation of cracks and closed pores,
which could be filled with air. Therefore, the absorption of nitrogen on graphite surfaces as well as the
nitrogen diffusion in the graphite matrix is one of the
major parameters for the local distribution of 14C in
i-graphite. Takahashi et al. reported that the nitrogen
content in graphite depends on the surface area of the
graphite and decreases from the surface to a depth of
about 30 nm32 and that 14C produced from nitrogen
will remain at its original position. This is in agreement
with the 14C distribution of HTR fuel pebbles from
the German AVR33 (Figure 2) and with the observation of a preferential release of 14C by surface oxidation
of i-graphite from the German high-temperature reactor AVR in a nitrogen/steam atmosphere.36
Takahashi et al.32 reported that the kinetic energy
of the formed 14C atom is about 470 kJ molÀ1, which
is in the range of covalent carbon bonding energies
and therefore the 14C atom will be attached to nodes
of the carbon lattice. They suggested that 14C will not
be released by radiolytic oxidation of the graphite.

µCi g–1(C)

6

547


However, this is also a surface-related reaction and a
similar release should be assumed as observed for
thermal surface oxidation. This is in contradiction
to Finn reporting a backscattering energy of 40 keV
($4 Â 106 kJ molÀ1) for 14C formation.37 This would
be significantly above any chemical bonding energy
and would lead to movements in the lattice and the
formation of new species. This high backscattering
energy as well as the large number of displacements
of carbon atoms during irradiation should lead to
a more homogenous distribution of 14C. However, the
displacements are in the range of 1–2 mm so that 500
displacements in one direction would be required for
a transport of 1 mm. So generally, it can be assumed
that 14C produced by activation of 13C is more or less
homogenously distributed as opposed to 14C generated
from 14N which is located in near-surface areas.
However, it cannot be concluded in general that
the 14C in i-graphite at the end of the reactor life is
generated mainly by nitrogen activation. Surface oxidation of i-graphite irradiated in carbon dioxide during reactor operation could reduce surface-bound
14
C. This reaction, as well as low amounts of nitrogen
impurities, could result in the remaining 14C inventory
being generated mainly by activation of 13C. Activation calculation for the Bugey 1 reactor shows that
the 13C activation leads to 96% of the measured
inventory 14C in i-graphite and only 4% of the
inventory must be addressed to nitrogen activation.38
This would also be in agreement with the results
obtained by Pichon,39 which show a fast release of
about 0.1% of the 14C inventory followed by a negligible leaching phase (Figure 3). A possible explanation could be a covalent bonding of 14C resulting

from 13C activation in the graphite matrix and leachable 14C fraction from nitrogen activation loosely
absorbed at the surface.

4

5.21.5.3

2

10

20

30

mm

Figure 2 Distribution of 14C in an high-temperature
reactor (HTR) pebble fuel element. Adapted from
Schmidt, P. Alternativen zur Verminderung der C-14-Emission
bei der Wiederaufarbeitung von HTR-Brennelementen;
Forschungszentrum Ju¨lich: Ju¨lich, 1979.

Formation of 36Cl

The dominant formation of 36Cl is by the neutron
activation of 39K (2.2 barn), the main stable natural
chlorine isotope with an occurrence of about 75%.
Chlorine itself is used for the removal of metal impurities in graphite by the formation of volatile chlorides. However, residual amounts of chlorine remain
in the graphite. Therefore, new cleaning methods for

nuclear graphite avoid the utilization of chlorine.
Furthermore 36Cl can be built by n,a-reaction of
39
K (4.3 mbarn) or from 34S via an n,g-reaction to
35
S (2.3 barn) followed by a b-decay to 35Cl. But these
reactions have no significant relevance.


548

Graphite

Cumulative released
fraction of 14C (%)

0.08

N Њ8 – Lime water

0.07

N Њ10 – Lime water

N Њ9 – Pure water

0.06
0.05
0.04
0.03

0.02
0.01
0
0

50

100

150

200

250
300
Time (days)

350

400

450

500

Figure 3 Leaching behavior of 14C from French G2. Adapted from Pichon, C.; Guy, C.; Comte, J. Cl-36 and C-14 behaviour
in UNGG graphite during leaching experiments, 2008.

Fraction of data less than concentration


1
0.9
0.8
0.7
0.6
0.5
0.4
0.3
Measured by NAA
Inferred from 36Cl

0.2
0.1
0
0

0.2

0.4

0.6

0.8

1

1.2

Initial chlorine concentration (ppm)
Figure 4 Initial chlorine concentration in Oldbury moderator graphite as measured by NAA and as inferred from

36
Cl activation. Adapted from Brown, F.; Palmer, J.; Wood, P. Derivation of a radionuclide inventory for irradiated
graphite-chlorine-36 inventory determination. In IAEA Technical Committee Meeting on Nuclear Graphite Waste
Management, Manchester, UK, 1999.

An investigation of core graphite from the Oldbury reactor shows the correlation of the initial chlorine impurities and the 36Cl inventory (Figure 4).40
Furthermore, the investigation shows chlorine loss
during irradiation. This is explained by the release
of chlorine from open pores and an activation of
chlorine in the closed pores. However, radiolytic
oxidation during operation will open the closed
pores by graphite oxidation, which results in an additional release path for 36Cl (see Figure 5).
Leaching experiments with French i-graphite
from G2 showed that a major amount of 80–85%

will be leached from the graphite in 1 month.
A further 5–10% will be leached in a period of
about 1½ years (Figure 6).39 This is in agreement
with the proposed chloride form of 36Cl located at
water-accessible surfaces and its high solubility.
A small part of 5–10% 36Cl remained in the graphite.
This could be explained by 36Cl located in graphite
areas that are not in contact with the leaching media,
for example, closed pores, or by covalent bonds
between the chloride and the carbon atoms of the
graphite lattice. Further investigations are required
to clarify the nature of the nonleachable 36Cl.


Graphite


Closed porosity

35Cl

Open porosity

Oxidation

Activation

549

Primary circuit

Release

35Cl

35Cl

Activation
Oxidation

36Cl

Release

36Cl


36Cl

Figure 5 Schematic of activation and release of chlorine in graphite.

Cumulative released fraction of
36Cl (%)

100
90
80
70
N Њ5 – Lime water
N Њ10 – Lime water
N Њ8 – Lime water

60
50

N Њ6 – Pure water
N Њ9 – Pure water
N Њ2 – Pure water

40
0

100

200

300


400

500

Time (days)
Figure 6 Leaching behavior of 36Cl from French G2. Adapted from Pichon, C.; Guy, C.; Comte, J. Cl-36 and C-14 behaviour
in UNGG graphite during leaching experiments, 2008.

5.21.5.4 Diffusion of Radionuclides in
Graphite
Diffusion in polycrystalline graphite is a complex
topic strongly related to the structure of the graphite.
The three general diffusion types, listed in the order
of increasing diffusion velocity, are:
 Volume diffusion by movements of atoms due to
the presence of lattice defects or exchange of lattice positions.41
 Diffusion along grain boundaries.
 Pore diffusion.
All the three different diffusion types can occur in
graphite: Volume diffusion in the graphite crystallites,
grain boundary diffusion at the micropores between the
crystallites, and pore diffusion in the pores between
the graphite particles. Self-diffusion of carbon in graphite occurs at temperatures about 2000  C42 and may be
important for central zones of HTR fuel elements.

Diffusion of fission products in graphite has
been studied intensively with respect to radionuclide release from HTR fuel elements. All these
processes become effective at higher temperatures
and can be neglected at temperature ranges relevant

to retrieval, interim storage, and final disposal.
However, they might be interesting for decontamination processes, especially for tritium. Table 3
shows some diffusion coefficients measured for
A3-3 graphite from HTR fuel pebbles, pitch coke
(AS1-500), and petrol coke (AL2-500) after
irradiation.43,44
The diffusion and release processes of radionuclides in i-graphite depend strongly on the nature of
the graphite and especially on the anisotropy of the
graphite.45,46
Tritium can be released from graphite more or
less completely by thermal treatment under inert
atmosphere at temperatures in the order of
1300  C.36


550

Graphite

Table 3

Diffusion coefficients of tritium nuclear graphite


Type of
graphite

Temperature ( C)

Diffusion coefficient,

D0 (sÀ1)

A3a
A3a
A3a
A3b
AS1-500b
AL2-500b

800
850
900
1000
1050
1025

1.72 Â 10À9
9.09 Â 10À9
6.89 Â 10À8
8.18 Â 10À9
9.83 Â 10À11
1.83 Â 10À10

a

Irradiation at temperatures < 100  C.
Irradiation at temperatures $500  C.

b


5.21.6 Graphite Treatment for
Disposal or Recycling
5.21.6.1 Waste Packages and
Encapsulation
Containers or drums are used as a packaging option
for i-graphite, mainly for safe handling in the operational phase of waste management and not as a barrier for long-term safety aspects. No special designs
of containers or drums have been made for i-graphite
and standards from common waste management are
applied. Therefore, this aspect is not covered in this
chapter. However, it must be mentioned that graphite can act as a noble metal and accelerate the
galvanic corrosion of stainless steel containers and
measures should be implemented to isolate the
graphite from the container or container materials
with a guaranteed lifetime until final disposal
should be used.
Another problem may arise while filling the waste
container, especially if more or less rectangular
graphite bricks are filled into drums. The disposal
costs depend on the classification and volume of the
radioactive waste and not on the weight. Therefore
different methods have been developed to achieve a
high packing density but all methods will generate
secondary wastes in the form of graphite dust.
Bach et al. compared grinding, plasma cutting, jet
cutting, wire sawing, and hydraulic breaking of
graphite, especially with respect to the related release
of graphite dust.
The encapsulation aims at a safe enclosure of the
waste by retardation or immobilization of radionuclides to avoid a release into the environment or at
least to reduce the release to an unobjectionable

level. Generally two options exist for encapsulation.
Embedding in a matrix material.
Impregnation of the graphite to fill the open pores.

The reference waste package concept for graphite
waste envisages the embedding of i-graphite in
cement pastes. The cement will establish an alkaline
environment in the pore water, which will reduce the
solubility of many key nuclides. Especially 14C will
form insoluble carbonates if it is oxidized to CO2 by
radiolysis processes. Further, the different cement
phases combined with a large pore surface area will
be able to absorb radionuclides or build insoluble
secondary phases. On the other hand, the porous
structure will not hinder the contact between aquatic
phases and the waste and therefore a radionuclide
release cannot be excluded, especially for 36Cl, which
shows no significant retardation by cement.
Alternative embedding materials could be glass,
polymers or resins, bitumen, low-melting metals, or
ceramics. The organic materials would all result in a
dense waste package that protects the graphite from
leaching. However, the production process and the
handling are related to a potential fire hazard. Furthermore, the long-term stability could be less than
the half-life of the key nuclides due to radiolysis or
ageing processes and therefore water exclusion cannot be guaranteed for the final disposal time scale.
Low-melting metals may have sufficient corrosion
stability, which has not been sufficiently determined
for disposal conditions. However, their own toxicity
may create a problem. For example, the license for

the German low-level waste underground disposal
site Konrad allows only the disposal of 72 Mg tin,
539 Mg zinc and 33 400 Mg lead due to the water
protection law of Lower Saxony.
The vitrification of graphite will result in a wellknown product similar to vitrified high-level waste.
Besides the known problems of the final disposal of
high-level waste such as fracturing, the graphite may
react with the glass melt and form dispersed gas
pebbles (bubbles?), which is known from the embedding of coated particles from HTR fuel.
The closing of the open pore system of graphite
has been successfully tested by impregnation with
bitumen, epoxy resins, and tar. Therefore, the graphite has to be evacuated and then immersed in bitumen
or resin under high pressure at elevated temperatures
to obtain a sufficient fluidity. Leaching tests with such
an impregnated material have proved the high immobilization potential of this procedure. However, this
would lead to problems similar to those already
described for materials used as embedding.
A new approach to fill the pore system of i-graphite
is a process that can be classified between embedding
and impregnation. It foresees the granulation of the


Graphite

i-graphite and a high-temperature compaction after
mixing with glass in an amount equal to the pore volume. First attempts show a density of about 2.2 cm3 gÀ1,
which is close to the theoretical density, and that the
glass does not increase the total volume. Furthermore,
this method would lead to volume-optimized waste
packages because the produced block dimensions can

be adjusted to the waste container dimensions. However,
the proposed good leaching resistance and mechanical
properties are yet to be demonstrated.
5.21.6.2

551

Centre, Ju¨lich (former KFA Ju¨lich). This development was related to the reprocessing of HTR fuel
pebbles. Another process, based on inductive heating,
has been developed by Westinghouse for graphite
fuel compacts.
However, the incineration of graphite would
result in a total release of 14C as CO2 together with
the bulk 12CO2, which may causes local increases of
the 14C activity in the surrounding area of the incineration plant. Therefore no public acceptance could be
achieved for such a graphite treatment option, even if
the released 14C activity would be negligible in comparison with the natural 14C amounts. The trapping of
CO2 is no alternative. Solidification of the CO2 as
insoluble calcium carbonate from 1.2 tons of graphite
($0.7 m3) would produce 10 tons CaCO3 (3.7 m3).
However, such a process has the advantage that the
14
C has been transferred into a defined species and will
have a more or less homogenous distribution.
An advanced thermal treatment method has been
developed first at the Research Centre, Ju¨lich. It was
shown that the majority of the carbon 14C inventory
could be removed from the AVR reflector and fuel
graphite and graphite from the thermal column of the
research reactor FRJ-1 by partial oxidation.34 The AVR

Graphite was irradiated at a high temperature in an
inert helium atmosphere and the other graphite at room
temperature in an air atmosphere. The thermal treatment process for 14C separation was performed in
nitrogen or argon plus 2% oxygen or humidified nitrogen or argon. First examinations by Podruzhina showed
a 14C release of about 70% with a total graphite oxidation in the range of 20 to 30%.34 This results were

Thermal Treatment

The most effective volume reduction would be the
complete oxidation of the i-graphite with a small ash
residue which contains the nonvolatile radionuclides.
Volatile radionuclides like tritium, 36Cl, or 137Cs may
cause some problems but could be trapped from the
off-gas and solidified for final disposal. (Tritium in
the form of HTO could be used for the production of
cement paste used as embedding material for radioactive waste.) Another problem is the incineration of
nuclear graphite due to its chemical purity. The high
thermal conductivity will conduct the heat from the
surface into the bulk material, which inhibits incineration. The poor combustibility of graphite was
shown by the first attempt of CEA, utilizing a coal
stove. Therefore, the material must be crushed before
incineration. Milling can be performed technically
without dust release but requires great effort. The
burning itself could be performed in furnaces or
in fluidized bed reactors. The burning of crushed
graphite has been demonstrated at the Research
100

Release rate (%)


80

60

14

C release rate; nitrogen + steam
Total carbon release rate; nitrogen + steam
C release rate; nitrogen + 2% oxygen
Total carbon release rate; nitrogen + 2% oxygen
14

40

20

0
0

Figure 7

50

100

150
200
Time (min)

250


300

350

C release and total carbon oxidation by thermal treatment of Allgemeiner Versuchsreaktor graphite at 1230  C.

14


552

Graphite

confirmed by Jansen.35 Higher release rates were
obtained by Florjan.36 Up to 60% of 14C will be released
within the first 60 min followed by a slower release of
20–30% in the next 2–7 h (Figure 7). The best 14C
release rates have been obtained at temperatures of
about 1200  C, whereas the separation of 12C and 14C
is better at lower temperatures (Tables 4 and 5). But the
release rates of Florjan could not be repeated until now.
Furthermore, these results show the different 14C
release behavior of the different graphite types under
similar treatment conditions. The best results have been
obtained with AVR graphite. This is explained by the
inhomogenous distribution of 14C with higher 14C concentrations on the surface and the existence of more
reactive 14C containing species. This indicates that the
irradiation conditions have an important influence on
this process and that further investigations will show

whether this process can be applied to CO2-cooled
reactors or RMBK reactors. However, this process
could be an alternative waste treatment option only
when 5% of the graphite materials have to be oxidized
and captured as CaCO3, if sufficient decontamination
factors can be achieved with these graphite types.
The removal of 14C from graphite has been considered the main problem in decontaminating graphite. However, the separation of radionuclides other
than 14C has to be managed, which can be performed
by different methods.
Table 4

5.21.6.3 The Russian ‘Self-Propagating
High-Temperature Synthesis SHS’
Graphite is homogenously mixed aluminum and titanium dioxide. The amounts are related to the following reaction:
3C þ 4Al þ 3 TiO2 ! 2 Al2 O3 þ 3TiC
The exothermic reaction is self-propagating and
only an initial start is required. The formed stable

14

C release by thermal treatment in N2 + 2% O2

Sample

R7

Origin

AVR


Treatment time (h)
Temperature ( C)
Total carbon release (%)
14
C release (%)
Separation factor

Table 5

An option would be the complete incineration of
the residual graphite, which would result in two more
waste streams. The residual 14C including the CO2
stream could be released in the environment if sufficient decontamination rates can be achieved or
fixated as CaCO3, which can be sent to a surface
disposal site. The second waste stream will be a
very small quantity of high active ashes and filter
dust which must be disposed as high-level waste
after an appropriate fixation. A typical volume reduction would be in the range of 160 for an incineration
process for nuclear graphite.1
A second option would be direct disposal in a nearsurface disposal site. However, this would require a
sufficient reduction of the 36Cl inventory (see Chapter
1.06, The Effects of Helium in Irradiated Structural
Alloys), which has not been investigated yet.

R8

K8

K9


M3

M4

FRJ2

Reflector graphite

Fuel graphite

3
900
2.85
61.0
21

3
900
2.69
62.2
23

3
1230
2.94
78.8
27

3
900

1.87
43.2
23

3
1230
2.32
64.4
28

K5

K6

M2

MS2

3
1230
4.04
79.8
20

Thermal column

14

C release by thermal treatment in N2 + steam


Sample

R6

Origin of graphite

AVR

Treatment time (h)
Temperature ( C)
Total carbon release (%)
14
C release (%)
Separation factor

R10

FRJ2

Reflector graphite

Fuel graphite

7
900
0.85
41.0
48

7

900
1.48
70.0
47

7
900
1.55
45.0
29

Thermal column
7
1280
5.40
92.6
17

7
900
4.12
69.8
17

7
900
0.02
49.0
2250



Graphite

titanium carbide contains 14C and the other radionuclides incorporated into the corundum and titanium carbide lattice. Additional confining additives
can be added to the reaction mixture, for example,
zirconium, which build even more stable crystalline
phases with selected radionuclides such as uranium
and plutonium. Furthermore, additives are used to
improve the final product quality and to minimize the
volatilization of radionuclides.
Therefore, this process is also suitable for graphite
contaminated with actinides from the Russian production reactors. The process requires a carefully controlled regime to minimize the radionuclide release.
5.21.6.4

Recycling of i-Graphite

The reuse of i-graphite may open a waste management route that has the potential to reduce the
amount of i-graphite for disposal.
The easiest attempt would be the direct use of
i-graphite without any treatment for the production of new materials for the nuclear industry.
Generally, it is known that used graphite can be
recycled as additive material for graphite production.
However, this cannot be done with i-graphite in
the existing graphite production facilities. Even lowcontaminated graphite must be handled in a closed
manufacturing unit to avoid the release of contaminated graphite dust. Furthermore, the amount of
i-graphite suitable for direct reuse is small in comparison with the total amount of i-graphite. Therefore
reuse will be associated with decontamination of
i-graphite. The success of the decontamination
depends on the achievable decontamination factors,
especially of 14C.

In principle, two methods could be proposed for
decontamination:
The wet method of leaching the graphite with
suitable decontamination agents such as mineral
acids or alkaline solutions.
Decontamination by thermal treatment in steam
or diluted oxygen atmosphere.
Both options are under investigation in the European Carbowaste project.47 At the moment, there
are not enough results from the leaching process to
evaluate the feasibility of this method, whereas the
thermal treatment mentioned in Section 5.21.6.2
has already proved its potential to remove 14C from
the i-graphite matrix. Figure 8 shows potential
routes to obtain feedstock material for graphite recycling. Two options can be considered after thermal

553

14

C decontamination. The first option is the total
oxidation of the residual graphite to find out if the
remaining 14C amount in the graphite would allow a
free release of the off-gas into the atmosphere. The
expected residues are in the range of a few kilograms
per Mg graphite. The alternative is treatment with
graphite cleaning methods known from the graphite
industry to remove the residual nonvolatile radionuclides to a level that can be handled in further
production steps. Potential products could be silicon
carbide, waste additives, and feedstock material for
new graphite materials in the nuclear industry.

The production of materials will definitely be
cheaper if fresh feedstock materials are used, but
the benefit will be the reduced waste volume (see
next section).
Another interesting aspect of the separation of the
14
C is the option to replace 14C production as tracer
material for scientific purposes by irradiation of nitrates.

5.21.7 Final Disposal
Figure 9 shows the general routes for radioactive
waste classification in different countries. Among
the European Union states, the Belgian and French
schemes are very similar and are closely related to
the EU classification scheme, which is based on the
general IAEA recommendations. These schemes formally recognize the lifetimes of the predominant
radionuclides within waste packages, and segregate
low- and intermediate-level waste into short-lived
and long-lived categories, on the basis of whether
the half-lives of these nuclides are less than or greater
than 30 years respectively.
Generally i-graphite can be assumed to be low- or
medium-level radioactive waste by these regulations,
whereby the classification for final disposal of
i-graphite is determined mainly by the inventory
of long-living radionuclides 14C and 36Cl. The high
bio-compatibility and the good solubility of 36Cl if it
occurs as chloride and therefore its high mobility
require larger efforts to provide a safe enclosure
from the environment.

The French surface disposal site Centre de l’Aube
has a total limit for the disposal of 0.4 TBq for 36Cl.
Figure 10 shows a calculation of the 36Cl inventory of
the stack of Bugey 1 plus some measured data. The
graphite core of the reactor Bugey 1 would have a 36Cl
inventory of about 0.1 TBq, assuming an average level of
50 Bq 36Cl gÀ1, which is probably too low as shown in
Figure 10. Measured values for Bugey 1 reveal an


554

Graphite

i-graphite
Disposal

Solidification

14

C depleted
i-graphite

Partial oxidation in
steam or diluted oxygen

14
C enriched
off-gas in the

form of CO or CO2

14

C separation

14

C products

Option 1
Total oxidation

Option 2

Free release of the off-gas
(depending on the
residual 14C inventory)
Reconversion of CO and
CO2 off-gas to carbon

Solidification and disposal

Further graphite cleaning by high temperature
treatment may be in presence of halogens as
decontamination agent

Other carbon-based products
(e.g., SiC, lamp black, waste
additives)

Additive for new graphite
products
Figure 8 Process scheme of a potential process for thermal treatment of i-graphite.

average of about 200 Bq gÀ1. Furthermore 36Cl is easily
leached, which has been discussed in Section 5.21.5.3.
Therefore, a near-surface disposal of graphite is not an
acceptable waste management option in France.
The situation in the United Kingdom is similar. An
estimate of the total 36Cl inventory is given by David
Lever.48 It is in the range of 23 TBq for the British
i-graphite. This 36Cl inventory will not allow the
near-surface disposal at the Drigg site if the release
could not be excluded over geological time scales.

A further aspect of UK reactors is the release of
C if it is in the form of methane. Figure 11 shows a
risk analysis of such a release if all the 14C is released
as methane. The assumption will not be true for
i-graphite; however, no quantitative results that give
a clear figure about the relation between 14CO2, 14CO,
and 14CH4 in the long-term release of 14C under
disposal conditions are available. However, there is
some evidence that organic 14C compounds cannot be
neglected. Leaching of HTR fuel spheres shows a
14


Belgium


France
VLLW < 100
Bq g-1

Cal A – low
concentrations
short half-lives
(Criteria X and Y)

LLW
Short-lived halflives < 30 years;
activity between 100
and 105 Bq g-1
ILW
Short-lived halflives < 30 years;
activity between
105 and 108 Bq g-1

European Union

Transition waste

IAEA

United Kingdom

EW – Exempt waste

VLLW – less than 400
kBq of b/g

activity per 0.1 m3
material

LILW-SL

LILW-SL

Short-lived half-lives
< 30 years

Short-lived half-lives
< 30 years

LLW
Long-lived halflives > 30 years;
Cal B – medium or activity between 100
LILW-LL
long half-lives in
and 105 Bq g-1
relatively high
Long-lived half-lives
ILW
concentrations
> 30 years
Long-lived halfpower < 20 W m-3
lives > 30 years;

LILW-LL
Long-lived half-lives
> 30 years


activity between
105 and 108 Bq g-1
Cal C – substantial
HLW;
amounts of b- and
activity between 108
a-emitters
and 1010 Bq g-1
power > 20 W m-3

HLW

HLW

Figure 9 Comparison of radioactive waste classification schemes.

Spent nuclear fuel
Waste with negligible
heat generation

LLW – < 4
GBq t-1 of
a and
<12 GBq t-1
of b/g
activity

Landfill/free disposal
Surface disposal

Geological disposal

High-level waste (HLW)
similar to European
definitions; arises mainly
from manufacture of nuclear
weapons
Transuranic waste (TRU):
radioactive waste containing
more than 3.7 ´ 103 Bq g–1
(100 nCi g−1) of a-emitting
transuranic isotopes with
half-lives > 20 years
nuclear weapons

ILW >4
GBq t-1 of
a or >12
GBq t-1 of
b/g activity, no
heating
consideration
in storage
HLW as ILW and with
cooling in storage
facilities

United States

Uranium mill tailings

Naturally occurring
radioactive material

Heat-generating waste

Low-level radioactive
waste (LLW): by definition:
everything else

Graphite

Generic disposal routes

Germany

555


556

Graphite
1.00E + 03

Bq g de 36Cl

1.00E + 02

Measures
C2J6
C1K7

B319
C4H2
B8J0
C2L4 = C2G6
ABlB
C3F9
D9Jl = A619
Risque = 50.00%
Risque = 2.50%
Risque = 0.10%
Initial centre
Initial + 2 sygma

1.00E + 01

1.00E + 00

36Cl
measurements

Activation
calculation results
for each channel
where the
samples come
from

1.00E – 01

11


13
15
17
19
21
Height (m) in the graphite stack (cooling gas flow direction)

23

Figure 10 Activation calculation results on BU1 stack with data assimilation method: 36Cl.

1E – 02
1E – 03
1E – 04

Total
Magnox spheres

1E – 05
1E – 06
1E – 07

Uranium spheres
Stainless steel spheres
Carbon steel spheres

1E – 08

Organic degradation


1E – 09
1E – 10

Radiolysis organics
Release from graphite
Risk target

1E – 11
1E – 12
1E – 13
1E + 00
1E + 01 1E + 02
1E + 03 1E + 04
1E + 05
1E + 06
Years postclosure (postclosure starts calendar year 2150)

Figure 11 Radiological risk versus time for 14CH4 by contributory sources. Adapted from Lever, D. Graphite Wastes:
Disposal Issues; Manchester, UK, 2006.

higher release of organic 14C than 14CO2, but as dissolved organic species and not as gaseous species.49
In Germany, radioactive waste is divided into two
classes: waste with and without significant heat development. Deep underground disposal is planned for
both types. The former Konrad iron mine has been
designated as the disposal site for the nonheat developing waste and is proposed to be ready for waste disposal
in 2013. The graphite from the reactor core of the AVR
and THTR clearly belongs to the category of non heat
developing waste and therefore could be disposed of at
this site. However, the 14C inventory of about

2.9 Â 1014 Bq for the ceramic core interior from the
AVR will claim a major share of the licensed

C inventory (4 Â 1014 Bq) of this disposal site and
will limit the disposal of other radioactive waste. Furthermore, the actual interim storage stage will extend
beyond the proposed operational time of this disposal
site and therefore alternatives are required.
14

5.21.8 Summary
A general solution for the management of i-graphite
has not been established yet. Only France has an ambitious final disposal plan for its i-graphite, with the
proposal of the near-surface underground disposal
site at a depth of about 200 m. Other countries like


Graphite

the United Kingdom have not made a final decision for
a reference waste management route until now.
Three main challenges have been identified for
the waste management of i-graphite.
The first is the retrieval of the major amounts of
i-graphite from the reactor cores. Some experience is
available from the decommissioning of the BEPO,
GLEEP, and Fort Saint Vrain reactors. However, no
general methodology can be recommended because
the retrieval depends on many site-specific factors.
A major concern is the need for more data on
i-graphite. This includes data on property changes of

the structure and mechanical properties due to irradiation and the radionuclide inventory, as well as fundamental data concerning the behavior of i-graphite
during treatment procedures, and disposal behavior.
Future research should focus also on the speciation of
the chemical form of radionuclides because the chemical form determines the long-term release behavior
under final disposal conditions. Furthermore, the localization of the key nuclides 14C and 36Cl at a nano-scale is
a major challenge because a near-surface distribution
and a homogenous distribution in the bulk would lead to
completely different release characteristics.
Another challenge is the development of a safe waste
management route. Generally, two principal methodologies could be utilized: the decontamination of
i-graphite by chemical or thermal treatment to obtain
a carbonaceous material for further use in nuclear technology or the final disposal of i-graphite. The most
advanced plan for a final disposal has been achieved
by France, which is planning an underground disposal
site for low-level radioactive waste containing longliving radionuclides, especially 36Cl, 14C, and radium.
Therefore, i-graphite will be grouted in drums or containers and disposed of afterward. This could be
assumed as the actual reference concept. Other conditioning methods, which ensure the safe long-term
enclosure of 36Cl and 14C as the Russian RSH method
or a long-term stable sealing of the graphite pore system, for example, with glass,50 may be alternatives to
enable a near-surface disposal of i-graphite from reactor cores.

4.
5.

6.

7.

8.


9.

10.
11.

12.

13.

14.
15.

16.
17.

18.
19.
20.

21.
22.

23.

References
1.

2.
3.


Marsden, B.; Wickham, A. Characterization, Treatment and
Conditioning of Radioactive Graphite from Decommissioning
of Nuclear Reactors; IAEA: Vieanna, 2006.
Energoatom Concern OJSC > Nuclear Power Plants.
Labyntseva, M. Russian Federation Country Report;
Marseilles, France, 2008.

24.
25.
26.
27.

557

RBMK – Wikipedia.
Russian Standards Catalog. GOST/R/50996/96///
Collection/storage/treatment/and/burial/of/radioactive/
waste/Terms/and/definitions/GOST/R/50996/96///Sbor/
khranenie/pererabotka/i/zakhoronenie/radioaktivnykh/
otkhodov/Terminy/i/opredeleniia/.
Holt, G. The decommissioning of commercial magnox
gas cooled reactor power stations in the United Kingdom
(Magnox Electric plc. Berkeley Centre, Berkeley
(United Kingdom)); IAEA-TECDOC–1043, pp 71–83.
Mary Byrd Davis: Natural uranium graphite gas reactors
(UNGG), Nuclear France: Materials and Sites,
WISE-Paris, 2002.
Bastien, D. (CEA Centre d’Etudes Nucleaires de Saclay,
91 - Gif-sur-Yvette (France) )HTGR Knowledge Base:
Twenty-nine years of French experience in operating

gas-cooled reactors; />aws/htgr/abstracts/abst_iwggcr19_16.html
Hanford Cultural and Historic Resources Program. Hanford
Site Historic District: History of the Plutonium Production
Facilities, 1943–1990; Battelle: Columbus, OH, 2003.
EC Project, Carbowaste, Contract FP7-211333,
Workpackage 1 & 2.
Wigner effect – Wikipedia, the free encyclopedia; http://
en.wikipedia.org/wiki/Wigner_energy, accessed on
Mar 1, 2010.
Windscale fire – Wikipedia, the free encyclopedia; http://
en.wikipedia.org/wiki/Windscale_fire, accessed on
Mar 1, 2010.
Wise, M. Management of UKAEA Graphite Liabilities. In
Proceedings of the IAEA Technical Committee Meeting on
Nuclear Graphite Waste Management; a.
org/inisnkm/nkm/aws/htgr/abstracts/abst_manchester.
html, 1999.
Laine, N.; Vastola, F.; Walker, P. J. Phys. Chem. 1963, 67,
2030.
Delle, W.; Koizlik, K.; Nickel, H. Graphitische Werkstoffe
fu¨r den Einsatz in Kernreaktoren. Teil 2: Polykristaliner
Graphit und Brennelementmatrix; Thiemig AG: Mu¨nchen,
1983.
Hedden, K.; Lo¨we, A. Carbon 1967, 5, 339.
Hedden, K.; Wicke, E. About some influences on the
reactivity of carbon, In Third Biennial Carbon Conference
1957, S249–256.
Xiaowei, L.; Jean-Charles, R.; Suyuan, Y. Nucl. Eng. Des.
2004, 227, S281.
Moormann, R.; Hinssen, H. K.; Ku¨hn, K. Nucl. Eng. Des.

2004, 227, 281.
Katscher, W.; Moormann, R.; Hinssen, H. K.; Stauch, B.
ersuche zur Graphitkorrosion in Luft bei Temperaturen
unterhalb 873 K, Vorschlag zur Modifizierung des
Verbrennungs-Head-Ends fu¨r HTR-Brennelemente,
Ju¨lich, 1984.
Stairmand, J. Graphite Oxidation – A Literature Survey;
AEA Technology Report, AEA-FUS-83, 1990.
Duwe, R.; Bru¨cher, H.; Fachinger, J. R&D on Intermediate
Storage and Final Disposal of Spent HTR Fuel; IAEA:
Vienna, 1997.
Gomez-Serrano, V.; Acedo-Ramos, M. A.;
Lopez-Peinado, A. J. Thermochim. Acta 1995, 254,
249.
Donnet, J. B.; Schutz, A. Carbon 1967, 5(2),
113–125.
Donnet, J. B.; Ehrburger, P. Carbon 1973, 11(4),
309–316.
Gmelin Handbuch der Annorganischen Chemie, Verlag
Chemie GmbH, 1966.
Fachinger, J.; den Exter, M.; Grambow, B.; Holgerson, S.;
Landesmann, C.; Titov, M.; Podruhzina, T. Behaviour of


558

28.
29.
30.


31.
32.

33.

34.

35.

36.
37.

Graphite
spent HTR fuel elements in aquatic phases of repository
host rock formations. In Proceedings of the 2nd
International Topical Meeting on High Temperature
Reactor Technology, Bejing, China, 2004.
EC Projekt ‘‘RAPHAEL’’; EC Contract 516508 (FI6O);
Internal report D-B3.5.
Aktivita¨tsatlas des AVR, Ju¨lich: WTI.
Bush, R.; Smith, G.; White, I. F. Carbon-14 Waste
Management; Office for Official Publications of the
European Communities: Luxemburg, 1984.
Marsden, B.; Wickham, A.; Hopkinson, K. The chemical
form of Carbon-14 within graphite, Serco Assurance, 2002.
Takahashi, R.; Toyahara, M.; Maruki, S.; Ueda, H.;
Yamamoto, T. Investigation of morphology and impurity of
nuclear-grade graphite, and leaching mechanism of
carbon-14. In Proceedings of IAEA Technical Meeting,
Manchester, UK, on Nuclear Graphite Waste

Management, published on CD-ROM IAEA-NGWM/CD
01-00120, October, 1999.
Schmidt, P. Alternativen zur Verminderung der
C-14-Emission bei der Wiederaufarbeitung von
HTR-Brennelementen; Forschungszentrum Ju¨lich: Ju¨lich,
1979.
Podruzhina, T. Graphite as radioactive waste: Corrosion
behaviour under final repository conditions and thermal
treatment, Forschungszentrum Ju¨lich (Diss. RWTH
Aachen 2004), Berichte des Forschungszentrum Ju¨lich
4166, 2004.
R. Jansen, Thermische Dekontamination von aktiviertem
Reaktorgraphit – Planung, Konstruktion und Test eines
Ofensystems. Forschungszentrum Ju¨lich (Dipl. RWTH
Aachen 2005), 2005.
Florjan, M. Dekontamination von Nuekleargraphit durch
thermische Behandlung (im Druck), RWTH Aachen 2009.
Finn, R.; Ache, J.; Wolf, A. J. Phys. Chem. 1969, 73(11),
S3928–S3933.

38.
39.
40.

41.
42.
43.
44.

45.

46.
47.
48.

49.

50.

Poncet, B. Personal communication.
Pichon, C.; Guy, C.; Comte, J. Personal communication.
Brown, F.; Palmer, J.; Wood, P. Derivation of a
radionuclide inventory for irradiated graphite-chlorine-36
inventory determination. In IAEA Technical Committee
Meeting on Nuclear Graphite Waste Management,
Manchester, UK 1999.
Mrowec, S. Acta Cryst. 1981, A37, 141–142.
Sach, R.; Williams, W. Carbon 1974, S425.
Fischer, P. G. Verhalten von Tritium in Reaktorgraphiten;
Forschungszentrum Ju¨lich GmbH: Ju¨lich, 1975.
Malka, V. J.; Raitz von Frentz, R. Adsorption und
Desorption von Tritium an Graphit; Forschungszentrum
Ju¨lich GmbH: Ju¨lich, 1978.
Ashida, K.; Watanabe, K. J. Nucl. Mater. 1991,
183, 89–95.
Saeki, M. J. Nucl. Mater. 1985, 131, 32.
EC Project, Carbowaste, Contract FP7–211333,
Workpackage 4.
Lever, D. Graphite Wastes: Disposal Issues, Presentation
at the University Manchester, UK, 2006; .
gov.uk/aboutus/meetings/iacs/nusac/031006/

presentation1.pdf.
Fachinger, J.; Zhang, Z. X.; Brodda, B. G. Graphite
corrosion and hydrogen release from HTR fuel elements
in Q-brine. In Proceedings of the International
Conference on Radioactive Waste Management and
Environmental Remediation, 5th, Berlin, Sept 3–7, 1995;
Vol. 1, pp 637–640.
Fachinger, J.; Hrovat, M.; Grosse, K.; Seemann, R.
Impermeable graphite: A new development for
embedding radioactive waste and an alternative
option of managing irradiated graphite. In Proceedings of
the Waste Management Symposium 2010, 10027,
Phoenix, 2010.

Appendix 1 Amounts of Irradiated
Graphite in Different Countries

Table A.1

Graphite-moderated reactors in Russia

Reactor

Graphite amount (Mg)

Scheduled
shutdown

Kursk 1
Kursk 2

Kursk 3
Kursk 4
Leningrad 1
Leningrad 2
Leningrad 3
Leningrad 4
Smolensk 1
Smolensk 2
Smolensk 3
Total

2000
2000
2000
2000
2638
1798
1798
1798
2158
1798
1798
19 988

2021
2024
2013
2015
2018
2020

2009 + 20 years
2011 + 20 years
2013
2020


Graphite

Table A.2

559

Graphite-moderated reactors in United Kingdom

Location

Reactor

Type

Graphite in reactor (tons)

Graphite total (tons)

Dungeness
Dungeness
Hartlepool
Hartlepool
Heysham
Heysham

Heysham
Heysham
Hunterston
Hunterston
Hinkley Point
Hinkley Point
Torness
Torness
Bradwell
Bradwell
Calder Hall
Calder Hall
Calder Hall
Calder Hall
Chapelcross
Chapelcross
Chapelcross
Chapelcross
Dungeness
Dungeness
Hinkley Point
Hinkley Point
Oldbury
Oldbury
Sizewell
Sizewell
Wylfa
Wylfa
Berkeley
Berkeley

Hunterston
Hunterston
Trawsfynydd
Trawsfynydd
Windscale
Winfrith
Windscale
Windscale
Harwell
Harwell
Total

B-1
B-2
1
2
Unit I-1
Unit I-2
Unit II-1
Unit II-2
B1
B2
B1
B2
1
2
Unit 1
Unit 2
Unit 1
Unit 2

Unit 3
Unit 4
Unit 1
Unit 2
Unit 3
Unit 4
A1
A2
A1
A2
Unit 1
Unit 2
A1
A2
A1
A2
Unit 1
Unit 2
A1
A2
Unit 1
Unit 2
WAGR
Dragon
Pile 1
Pile 2
BEPO
GLEEP

AGR

AGR
AGR
AGR
AGR
AGR
AGR
AGR
AGR
AGR
AGR
AGR
AGR
AGR
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox

Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
Magnox
AGR
HTR
Air cooled
Air cooled
Air cooled
Air cooled

850
850
1360
1360
1520
1520
1520
1520
970
970
970
970
1520
1520

1810
1810
1164
1164
1164
1164
1164
1164
1164
1164
2150
2150
2210
3310
2061
2061
2237
2237
3470
3470
1938
1938
1780
1780
1900
1900
285
40
<2000
2000

766
505
65 080

ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
1931
1931
1630
1630
1630
1630
1630
1630
1630
1630
2237
2237

2457
2457
2090
2090
2240
2240
3740
3740
1650
1650
2150
2150
1980
1980
285
40
<2000
2000
766
505
!77 000


560

Graphite

Table A.3

Graphite-moderated reactors in France


Location

Reactor

Type

Graphite in reactor (tons)

Graphite total (tons)

Marcoule
Marcoule
Marcoule
Loyettes
Avoine
Avoine
Avoine
Orleans
Orleans
Total

G1
G2
G3
Bugey 1
Chinon A1
Chinon A2
Chinon A3
St. Laurent A1

St. Laurent A2

Air cooled
UNGG
UNGG
UNGG
UNGG
UNGG
UNGG
UNGG
UNGG

1200
1207
1207
2039
1050
2200
2530
2572
2440
16 445

1200
1207
1207
3600
1060
2500
4000

4240
4100
23 114

Table A.4

Graphite-moderated reactors in United States

Location

Reactor

Type

Graphite in reactor (tons)

Graphite total (tons)

Platteville, CO
Peach Bottom, PA
Hanford
Hanford
Hanford
Hanford
Hanford
Hanford
Hanford
Hanford
Hanford
Savannah River

Savannah River
Oak Ridge
Brookhaven
Chicago
Pacific North West Labs
Pacific North West Labs
Argonne National Laboratory

Fort Saint Vrain
Peach Bottom
B
D
F
DR
H
C
KW
KE
N
SR-305
SP
8 GR (X-10)
BGRR
CP1
HTLTR
HTR
CP2

HTGR
HTGR

LWGR
LWGR
LWGR
LWGR
LWGR
LWGR
LWGR
LWGR
LWGR
Test

ND
ND
1080
1080
1080
1080
1080
1080
1080
1080
1080
ND
ND
ND
440
ND
ND
ND
ND


ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND
ND

Air cooled
Test
Test
Air cooled
United States
Graphite pile


Graphite


Table A.5

Germany
Belgium
Italy
Japan
North Korea
Lithuania
Spain
Ukraine

China

561

Graphite-moderated reactors in other countries
Location

Reactor

Type

Graphite in reactor
(tons)

Graphite total
(tons)

Shutdown

date

Juelich
Uentrop
Mol
Borgo Sabotino
Tokai
Oarai
Nyongbyon
Visaginas
Visaginas
Hospitalet de
l’Infant
Chernobyl
Chernobyl
Chernobyl
Chernobyl
INET

AVR
THTR 300
BR-1
Latina
Tokai 1
HTTR
Nyongbyon 1
Ignalina 1
Ignalina 2
Vandellos


HTGR
HTGR
Air cooled
Magnox
Magnox
HTTR
Magnox
LWGR
LWGR
Magnox

225
300
472
2065
920
ND
ND
1700
1700
2440

ND
300
472
ND
1600
ND
ND
2000

2000
ND

1988
1989 S

Unit 1
Unit 2
Unit 3
Unit 4
Tsinghua
HTR-10
ND
ND
ND

LWGR
LWGR
LWGR
LWGR
HTR

1700
1700
1700
<1700
111

2000
2000

2000
<2000
2000

1996 S
1991 S
2000 S
1986 S

Air cooled
LWGR
LWGR

ND
ND
ND

ND
ND
ND

Baotou
Jiuquian
Guangyuan

1987 S
1998 S

1990 S