5.11

Material Performance in Helium-Cooled Systems

R. Wright and J. Wright
Idaho National Laboratory, Idaho Falls, ID, USA

C. Cabet
Commissariat a l’Energie Atomique, Gif-sur-Yvette, France

ß 2012 Elsevier Ltd. All rights reserved.

5.11.1

Introduction

252

5.11.2
5.11.3
5.11.3.1
5.11.3.2
5.11.3.3
5.11.3.4
5.11.3.5
5.11.3.6
5.11.3.7
5.11.3.8
5.11.4
5.11.4.1
5.11.4.2

5.11.4.3
5.11.4.4
5.11.4.5
5.11.5
5.11.5.1
5.11.5.2
5.11.5.3
5.11.6
5.11.7
5.11.8
5.11.9
5.11.10
References

Experience with VHTR Systems
Comparison of IHX Concepts
Shell-and-Tube
Plate and Fin
Etched Plate
Microchannel Heat Exchangers
Plate-Stamped Heat Exchanger
Foam IHX
Capillary IHX
Ceramic IHX
Heat Exchanger Alloys
Regulatory Issues
Alloy 617 (52Ni–22Cr–13Co–9Mo)
Alloy 230 (57Ni–22Cr–14W–2Mo–La)
Alloy 800H (42Fe–33Ni–21Cr)
Alloy X (47Ni–22Cr–9Mo–18Fe)

Welding
Base Metal Preparation and Filler Metal Selection
Preheating, Interpass Temperatures, and Postweld Heat Treatment
Nontraditional Joining Methods
Control Rod Materials
Core Barrel Materials
Environmental Effects of VHTR Atmospheres on Materials
Aging Effects
Summary

252
254
254
255
255
256
256
257
257
257
257
259
259
262
263
264
264
265
265
265

266
269
269
273
275
276

Abbreviations
AGCNR
AVR
DLOC
GE
GMAW
GTAW
HE
HTGRs
HTR-10

Advanced gas-cooled nuclear reactor
Arbeitsgemeinschaft Versuchsreaktor
Depressurized loss of coolant
General Electric
Gas-metal-arc welding
Gas-tungsten-arc welding
Heat exchanger
High-temperature gas reactors
High-temperature reactor-10MWth
in China

HTTR

IHX
INL
NGNP
ODIN
ORNL
PBMR
PCHE
PCS

High-temperature engineering test
reactor in Japan
Intermediate heat exchanger
Idaho National Laboratory
Next generation nuclear plant
Online Data & Information Network
Oak Ridge National Laboratory
Pebble bed modular reactor
Printed circuit heat exchanger, Heatrics
Division Ltd.
Power conversion system

251


252

Material Performance in Helium-Cooled Systems

PFHE
PSHE

RCS
RCSS
RSS
SMAW
THTR
VHTR

Plate and fin heat exchanger
Plate stamped heat exchanger
Reactivity control system
Reactor control and shutdown system
Reserve shutdown system
Shielded metal arc welding
Thorium Hochtemperatur Reaktor
Very high-temperature reactor

5.11.1 Introduction
Over the past decade, there has been renewed interest in very high-temperature reactor (VHTR) technology. This type of reactor is of interest because
of a number of unique characteristics, including passive safety, electricity production on a more modest
scale compared to light water plants that might be
more compatible with the electrical distribution system in developing countries, and very high outlet
temperature that can be used for process heat or
hydrogen production. The relative value of electricity production or process heat applications varies
considerably with world economic conditions. Currently, it appears that steam for process heat and
hydrogen production will drive development of this
technology rather than electricity production.
There are currently two operating VHTR prototypes, the high-temperature engineering test reactor
(HTTR) in Japan and the high-temperature reactor
(HTR-10) in China. The HTTR is a 30 MWt (megawatts thermal output) prismatic core reactor and the
HTR-10 is a 10 MWt pebble bed prototype reactor.

Both the operating reactors are designed to investigate
electricity production with the VHTR technology;
however, each program has parallel activities to
develop process heat and hydrogen production as well.
The challenges for high-temperature materials are
not significantly different for either prismatic or pebble
bed reactor designs. Interest in specific applications for
VHTR technology is evolving rapidly. It appears that
the most significant immediate interest is in a reactor
with an outlet temperature on the order of 750  C with
a steam generator for either electricity generation or
process heat. This technology would use a relatively
mature conventional steam generator technology and
is expected to present lower technical risk. Higher
outlet temperatures using a heat exchanger between
the primary helium coolant and a secondary gas are
viewed to be higher risk development projects that
offer the opportunity for outlet temperatures from

850 to 950  C for hydrogen production by thermochemical processes or higher temperature process
heat for industrial applications. The material issues
associated with reactor internals are not affected significantly by the reactor outlet temperature; however,
heat exchangers operating at the higher outlet temperatures represent significantly different issues compared to steam generators. The focus of this chapter is
on higher outlet temperature systems because of the
development challenges.
The next generation nuclear plant (NGNP) being
developed in the United States is one particular
VHTR concept that is under very active development
and is typical of the development around the world.
This reactor is being developed to produce hydrogen

as well as electricity. Conceptual designs call for a gascooled reactor with an outlet temperature greater than
the 850  C required to efficiently operate the hydrogen
generation plant, with a maximum of 950  C. While the
design concepts are not yet final, it is highly probable
that helium will be the primary coolant in the reactor.
The primary material in the core will be graphite, and
the prime candidates for high-temperature metallic
components are the nickel-based alloys Alloy 617 or
Alloy 230. An artist’s representation of one concept
for the reactor and power conversion vessel and the
associated hydrogen generation plants is shown in
Figure 1. In this representation, a heat exchanger
carries most of the reactor thermal output to a secondary circuit that powers a turbine for electricity generation. An additional heat exchanger takes $10% of the
thermal energy of the reactor and diverts it as process
heat to the hydrogen production plant.
The most critical metallic component in the
VHTR system is the intermediate heat exchanger
(IHX). This heat exchanger will operate at a reactor
outlet temperature of up to 950  C. In addition, the
reactor system is intended to have a license period of
60 years. The combination of very high-temperature
operation and long duration of service restricts material choices for the heat exchanger to a small number
of coarse-grained solid-solution strengthened alloys
that provide stability and creep resistance and have
high chromium content for environmental resistance.

5.11.2 Experience with VHTR
Systems
Very early in the development of nuclear power for
electricity generation or process heat, the concept of an

inert gas-cooled, high-temperature reactor was explored.


Material Performance in Helium-Cooled Systems

253

Power
conversion unit
Intercooler

Generator

Turbine

Low-pressure
compressor
Primary
heat rejection
High-pressure
compressor
Recuperator

Commercial
power

Blower
Power for
electrolysis


Pump

Heat
exchanger
Pebble-bed or
prismatic reactor

Blower

Heat
exchanger

Hydrogen production
(electrolysis)

Hydrogen

Hydrogen production
(thermochemical)

Hydrogen

Heat
exchanger

Figure 1 An artist’s conception of a very high-temperature gas-cooled reactor and associated hydrogen production plants.

Table 1

Design characteristics of VHTRs that have been built and operated


Country of origin

Thermal power MWt
Net electric power MWe
Maximum core outlet temp ( C)
Helium pressure MPa
Steam temp ( C)
Reactor type
Vessel material
Date of operation

Dragon

AVR

Peach bottom

Ft. St. Vrain

THTR-300

HTTR

OECD/Britain

Germany

USA


USA

Germany

Japan

21.5
–
750
2.0
–
Sleeve
Steel
1964–1975

46
13
950
1.1
505
Pebble
Steel
1966

115
40
725
2.25
538
Sleeve

Steel
1967

842
330
775
4.8
538
Block
PCRVa
1979–1989

750
300
750
3.9
530
Pebble
PCRV
1985

30
10
950
4
Prism
Steel
1997

a

Prestressed concrete reactor vessel.
Source: Simon, R. A.; Capp, P. D. Operating experience with the dragon high temperature reactor experiment. In Proceedings of the
Conference on High Temperature Reactors, Petten, NL, Apr 22–24, 2002; pp 1–6.
Burnette, R. D.; Baldwin, N. L. Specialists Meeting on Coolant Chemistry, Plate-Out and Decontamination in Gas Cooled Reactors,
Juelich, FRG, Dec 1980; International Atomic Energy Agency, 1980; pp 132–137.
Shaw, E. N. History of the Dragon Project – Europe’s Nuclear Power Experiment; Pergamon: New York, 1983.
Ba¨umer, B.; et al. AVR – Experimental High-Temperature Reactor; 21 Years of Successful Operation for a Future Energy Technology;
Association of German Engineers (VDI), The Society for Energy Technologies: Du¨sseldorf, Germany, 1990.
Baumer, R.; Kalinowski, I. Energy 1991, 16(1/2), 59–70.
Brey, H. L. Energy 1991, 16(1/2), 47–58.
Fuller, C. H. Design Requirements, Operation and Maintenance of Gas-Cooled Reactors, San Diego, CA, Sept 21–23, 1988;
International Atomic Energy Agency, 1989; pp 55–61.

The Peach Bottom reactor in the United States and
the European Dragon project were among the first
to seriously address the technical issues associated
with high-temperature environmental interaction
between the cooling gas and metallic components.1–3
Proposals for a VHTR with an outlet temperature of
1000  C or above were put forward in the late 1970s.

The Arbeitsgemeinschaft Versuchsreaktor (AVR) was
the first experimental pebble bed reactor. A commercial demonstration scale pebble bed, the Thorium
Hochtemperatur Reaktor (THTR), was developed
based on AVR experience. A summary of important
design characteristics for gas-cooled VHTRs that
have been operated to date are given in Table 1.1–7


254


Material Performance in Helium-Cooled Systems

The HTTR in Japan is the only one of the reactors
listed in the table that is still in operation. The HTR-10
is not included in the table since there is no extensive
operating experience with this reactor as yet.
Operating experience with these reactors has
shown that the primary helium coolant tends to contain H2O, H2, N2 as well as carbon-containing compounds CO, CO2, and CH4 at concentrations of a few
parts per million. Impurities are introduced through
adsorption on the fuel, leaks into the coolant, and
lubricants from components like the helium circulators. In the reactors that are currently under consideration, the gas pressure is typically between 5 and
7 MPa. The coolant is circulated at high velocity,
reaching velocities over 100 m sÀ1 in some designs.

5.11.3 Comparison of IHX Concepts
The IHX design for the VHTR will be influenced
by a number of interrelated considerations, including
the required separation distance between the reactor
and the hydrogen production or other process heat
plant, the heat losses from the intermediate loop piping, the operating pressure, the working fluid in the
secondary loop, and the target efficiency of the hydrogen or process heat plant. The required separation
distance will affect the intermediate loop piping size,
the intermediate loop pumping requirements, and the
piping heat losses to the environment. The intermediate loop pressure is critical; a low pressure will produce
a high pressure differential between the primary and
secondary sides of the IHX and high stress on the IHX.
A high intermediate loop pressure will produce a high
pressure differential across the intermediate loop pipe
walls and within the hydrogen production or process

heat equipment. Pressure drops within the IHX affect
the pumping power requirements, which also depend
on the intermediate loop working fluid, and the fluid
temperature and pressure, and will have an effect on
the overall VHTR cycle efficiency.
The IHX may be arranged in parallel or in series
with the VHTR power conversion system (PCS). In a
serial arrangement, the total primary system flow
(reactor outlet gas) passes through the IHX. The
IHX receives gas of the highest possible temperature
for delivery to the hydrogen production process (with
slightly cooler gas going to the PCS), and must be
large enough to handle the full primary flow.
A parallel configuration splits the reactor outlet gas
flow, with only a portion entering the IHX for the
hydrogen or process heat plant, and the remainder of

primary flow going to a direct cycle power generation
turbine. This results in the smallest possible IHX and
the highest overall electrical power efficiency but
lower process heat efficiency because of the cooler
gas reaching that process.
Specific IHX designs under consideration include
countercurrent tube and shell, plate and fin, involute
heat exchangers, microchannel heat exchangers, and
the printed circuit heat exchanger (PCHE). The
design has a significant influence on the required
material properties. Tube-and-shell designs have the
advantage of technological maturity, use heavy gauge
materials, and are fabricated using conventional fusion

welding methods. For the most simple tube-and-shell
configuration, it has been estimated that 13 tons of
high-temperature alloy is required per megawatt of
heat transfer capability; helical designs can reduce
this value to about 1.2 tons MWÀ1. Compact heat
exchanger designs have the potential for greater heat
transfer efficiency; it is estimated that some of these
designs will require only 0.2 tons of alloy per MW. The
compact designs are much less technologically mature
and increase the demands on material performance.
Some compact designs have wall thicknesses of less
that 1mm which places a premium on corrosion resistance and have significant stress concentrations that
will lead to increased demand for creep resistance.
In addition, several of these design concepts require
diffusion bonding of multiple sheets of material or
brazing in complex geometries. Neither of these joining methods has been used yet in nuclear applications,
and nondestructive inspection methods have not been
well developed.
5.11.3.1

Shell-and-Tube

A shell-and-tube heat exchanger is the most common
type of heat exchanger. It consists of a number of
tubes (often finned) placed inside a volume (shell).
One of the fluids runs through the tubes while the
second fluid runs across and along the tubes to be
heated. In one variation of this concept, the heat
transport fluid will flow on the shell side, allowing
the tubes to contain the catalysts necessary for hydrogen production. In the simple configurations, the tube

axis is parallel to that of the shell. The VHTR IHXproposed design features the tubes arranged in a
helical configuration. This type of arrangement
increases efficiency because of increased surface
area and reduces the size, providing the potential to
decrease the cost of materials. Tube-and-shell heat
exchangers represent relatively mature technology


Material Performance in Helium-Cooled Systems

that has been widely commercialized in both nuclear
and fossil energy systems. A helical design was extensively tested for the AVR reactor program and a
similar system is in use in the HTTR in Japan.
5.11.3.2

Plate and Fin

The plate and fin heat exchanger (PFHE) transfers
heat between two fluids by directing flow through
baffles so that the fluids are separated by metal
plates with very large surface areas. The fluids spread
out over the plate, which facilitates the fastest possible transfer of heat. This design has a major advantage
over a conventional heat exchanger because the size

of the heat exchanger is less for a comparable heat
transfer capability. However, the candidate heat
exchanger materials have relatively low thermal conductivities and will reduce the efficiency of a finned
structure. Brazing is typically used to join the fins to
the plate. Brazed plate heat exchangers are used in
many industrial applications, although usually at low

or even cryogenic temperatures. Although brazed
products have been developed for high-temperature
aerospace applications, the strength and creep
properties of brazed joints in an IHX for a hightemperature reactor are of great concern.
The unit cell heat exchanger is a typical modular
plate-fin design that is being developed by Brayton
Energy. An example is shown in Figure 2. Many of
these individual unit cells would be grouped into
larger heat exchanger assemblies. Integration of the
modules within the vessel and with the interfacing
piping is critical. Offset fin plate heat exchangers
have very large heat transfer area density and effective countercurrent flow.
5.11.3.3

(a)

(b)
Figure 2 Unit cell heat exchanger (a) primary side plate,
(b) the unit cell showing countercurrent flow.

255

Etched Plate

Etched plate heat exchangers are diffusion-bonded,
highly compact heat exchangers that can achieve a
thermal effectiveness of over 98% in a single unit.
Compact heat exchangers are four to six times smaller
and lighter than conventional shell-and-tube heat
exchangers of the equivalent heat transfer capability

(Figure 3). The small size gives the compact diffusionbonded heat exchangers significant benefits over conventional heat exchangers across a range of industries.
They are well established in the oil production, petrochemical, and refining industries. In addition, they
are suitable for a range of corrosive and high-purity

Figure 3 The diffusion-bonded heat exchanger in the
foreground undertakes the same thermal duty, at the same
pressure drop, as the stack of three shell-and-tube
exchangers behind.


256

Material Performance in Helium-Cooled Systems

Hot channel

Cold channel

Figure 4 Printed circuit heat exchanger configuration for the model.

streams and are particularly advantageous when
space is limited and weight is critical.
The most widely commercialized etched plate
heat exchanger is a PCHE developed by Heatric
Division of Meggitt (UK) Ltd. PCHE consists of
metal plates on the surface of which millimeter-scale
semicircular fluid-flow channels are photochemically
milled, using a process analogous to that used for the
manufacture of electronic printed circuit boards.
The plates are then stacked and diffusion-bonded

together to fabricate a heat exchanger core shown
schematically in Figure 4. Heatric reports pressure
capability in excess of 70 MPa and the ability to
withstand temperatures ranging up to 900  C.
Note that the channels are straight in this schematic, but in reality they have a zigzag configuration.
Flow distributors can be integrated into plates or
welded outside the core, depending on the design.
The channel diameter, plate thickness, channel
angles, and other attributes can be varied, so each
PCHE is custom-built to fit a specified task. Channel
dimensions are generally between 3 and 0.2 mm and
the thickness of the web of material left after milling
is typically less than 1 mm.8 The current fabrication
limits are 1.5 m  0.6 m plates and 0.6 m stack height.
The diffusion-bonded blocks made from several hundred individual sheets are modular and multiple
blocks can be welded together to form larger units.
5.11.3.4

Microchannel Heat Exchangers

Microchannel heat exchangers, produced, for example, by Velocys, also feature a compact design similar
to the etched plate design; however, the manufacturing
process is somewhat different. They are constructed
from diffusion-bonded corrugated sheets rather than

30Њ

Figure 5 Plate-stamped heat exchanger concept.

etched plates. The layers of corrugated sheet form

many small-diameter channels that result in a high
surface area/volume ratio and a high heat transfer
coefficient.
5.11.3.5

Plate-Stamped Heat Exchanger

The plate-stamped heat exchanger (PSHE) concept
consists of a set of modules, each being composed of a
stacking of plates stamped with corrugated channels.
The plates are stacked in such a way as to cross the
channels of two consecutive plates and therefore to
allow the different channels to communicate through
the width of the plate as shown on the left in the
figure. A general view of a plate is shown in Figure 5.
Assembly of the plates into an IHX module is
accomplished by welding only on the edges of the
plates. No joining is performed in the active part of
the plates, which gives the module relatively good
flexibility. Therefore, this concept is thought to
accommodate the thermal stresses better than the
other concepts of plate IHXs. The location of the
welded joints is also favorable to inspection, even if


Material Performance in Helium-Cooled Systems

this remains a difficult question. The joining processes which seem to be the most relevant are laser
or electron beam welding due to the capability to
perform narrow-gap joints and to avoid the overlapping of the welds of two consecutive plates. It should

also be noted that the thickness of the PSHE plates
is the largest among the metallic plate types IHX
(1.5 mm), which means that it is the most favorable
concept with respect to corrosion life. These reasons
suggest that the PSHE concept may be the most
promising among the plate IHXs.
5.11.3.6

Foam IHX

The foam IHX concept is based on stacking plates
separated by metallic foam. The barrier between the
fluids is constituted by the separated plates and the
fluids flow through the foam (see Figure 6). It is a
new technology for heat exchanger application for
which very high efficiency has been claimed. Several
concerns have been identified regarding this type of
IHX concept. The pressure losses induced by the
foam are particularly high. Loss of small fragments
of the foam is hardly avoidable and the geometry of
the foam leads to an increased risk of clogging by
graphite dust.
5.11.3.7

Capillary IHX

A concept with thread tubes between two tube-plates
with external shell including bellows has been investigated. The diameter of the tubes is 2–3 mm. This kind
of heat exchangers is currently being developed on an
industrial scale. The small size of the tubes allows a

sharp reduction in size and mass, but some difficulties
arise at the same time, including the concern that
the vibration risk is increased so that the supporting
system needs to be very robust. The number of tubes

Figure 6 Foam heat exchanger concept.

257

reaches very high values, which increases the complexity of manufacturing, notably as assembly by
narrow gap welding is required. Demonstration of
the elements necessary for successful implementation of the technology is mainly based on technological feasibility tests like demonstration of individual
tube to tube-plate welds by laser techniques. The
results confirm the feasibility for limited thickness of
the plate (a small mock-up is shown in Figure 7).
5.11.3.8

Ceramic IHX

The development of IHXs made of ceramics is still at
the research stage. Ceramic heat exchangers under
development are either tubular or plate IHXs (mostly
PFHE for the ceramic plate IHXs). Tube-and-shell heat
exchangers based on SiC composite tubes have been
developed for fossil energy applications for example.
Joining the fiber-reinforced composite tubes to tube
sheets and accommodating thermal expansion are
the dominant technical challenges. Their resistance to
aggressive environment is remarkable and they can
operate at very high temperatures, >1000  C. Small

monolithic compact designs have been developed
from silicon carbide and silicon nitride through conventional ceramic forming and firing routes. In addition to technical issues, the cost of ceramic tubes of
sufficient size for a VHTR IHX remains problematic.
Table 2 provides a summary-level comparison of
the significant attributes of the different IHX concept
alternatives.

5.11.4 Heat Exchanger Alloys
The desire for higher temperature operation resulted
in the evolution of the materials under consideration,
from stainless steels to iron-based high-temperature


258

Material Performance in Helium-Cooled Systems

Figure 7 Capillary heat exchanger mock-up.

Table 2

Comparison of IHX concept alternatives

PCHE
PFHE
PSHE

Tubular IHX

Maturity


Stress behavior

Sensitivity to
corrosion

Compactness

Numerous developments
in conventional industry
Numerous developments
in conventional industry
Numerous developments
in conventional industry

High stress levels. 5 years
lifetime seems very challenging
High stress levels. 5 years
lifetime seems very challenging
Challenging but best stress
accommodation among
the plate IHXs
Limit of state of the art

Sensitive

26 MW mÀ3

Very sensitive


24 MW mÀ3

Sensitive

35 MW mÀ3

Better than plates
but still sensitive
Very sensitive (loss
of fragments risk)
Very sensitive

0.4 MW mÀ3

Foam IHX

Industrial components in
operation
R&D

No results

Capillary IHX

Industrial developments

No results

Ceramic IHX


R&D

Difficult design because of
fragile behavior

alloys to nickel-based alloys (see Chapter 2.08, Nickel
Alloys: Properties and Characteristics). An extensive
German program in the 1980s carried out exhaustive
studies of the corrosion behavior of the iron-based
Alloy 800H for control rods and nickel-based Alloy 617
for structural applications.9–12 The Japanese HTTR
program extensively studied Alloy X and developed a
variation known as XR with improved properties for
some applications, while retaining Alloy 800H for the
control rods.13 Compositions of these candidate alloys
are given in Table 3.13–16 Based on creep resistance
above 850  C, the leading candidate alloys for VHTRs
are Alloy 617 and Alloy 230.
A common characteristic of the alloys that have
been put in service in high-temperature gas-cooled

Resistant

Comparable to
other plate IHXs
Better than
classical tubular
IHX
Comparable to
other plate IHXs


reactors is that they rely primarily on the formation
of a tenacious chromia scale for long-term protection
from environmental interaction with the gas-cooling
environment.9,10,12,17 The alloys are also primarily
solid-solution strengthened with carbides on the
grain boundaries to stabilize the microstructure and
enhance the creep resistance. Sustaining such a protective surface oxide requires sufficient oxygen partial pressure. The primary coolant gas of choice for
VHTRs is helium. Although the helium is nominally
pure and thus considered to be inert, there are inevitably impurities at the parts per million by volume (ppm)
levels in the coolant in operating high-temperature
reactors. Although at low levels, the impurities can
significantly affect the performance of materials,


Material Performance in Helium-Cooled Systems

Table 3

259

Compositions of potential high-temperature alloys for VHTR (compositions in wt%)

Alloy

Ni

Fe

Cr


Co

Mo

Al

Alloy 617
UNS N06617
Alloy 230
UNS N06230
Alloy 800H
UNS N08810
Alloy X
UNS N06002

44.5

3

20–24

10–15

8–10

0.8–1.5

Bal


3

20–24

5

1–3

0.2–0.5

30–35

39.5

19–23

Bal

17–20

20.5–23

W

8–10

0.1

C


0.6

0.05–0.15 1

13–15

0.15–0.6
0.5–2.5

Ti

0.2–1

Si

Mn
1

0.05–0.15 0.25–0.75 0.3–1
0.15–0.6

0.05–0.1

0.03

0.05–0.15 <1

<1

Source: Incoloy Alloy 800H & 800HT, product sheet, Special Metals, 2004.

Inconel 230, UNS N06230, product sheet, Special Metals, 2004.
Inconel 617, UNS N06617, product sheet, Special Metals, 2005.
Tanaka, R.; Kondo, T. Nucl. Technol. 1984, 66(1), 75–87.

depending on the chemistry of the particular alloy,
the concentration of impurities, and the temperature
at which the alloy can be oxidized, carburized, or
decarburized. Several reviews of the behavior of
metallic alloys for control rods, core internals, and
heat exchangers in the reactor helium environment
are available.9–12,17
Regardless of the IHX design, material selection
for this component is critical. The material must be
available in the appropriate product forms – both
plate and sheet, weldable and suitable for use at
800  C or above. The majority of materials research
and development programs in support of hightemperature gas reactors (HTGRs) were conducted
from the 1960s to the early 1980s. The thrust of these
programs was to develop a database on materials for
application in steam-cycle and process-nuclear-heatbased HTGRs. Less work has been done on materials
with emphasis on direct and/or indirect gas-turbinebased HTGRs. The available material property data
were reviewed in detail, and an assessment of relevant factors was made including thermal expansion,
thermal conductivity, tensile, creep, fatigue, creep–
fatigue, and toughness properties for the candidate
alloys. Thermal aging effects on the mechanical
properties and performance of the alloys in helium
containing a wide range of impurity concentrations
are also considered.17 The assessment includes four
primary candidate alloys for the IHX: Alloy 617,
Alloy 230, Alloy 800H, and Alloy X.

5.11.4.1

Regulatory Issues

The IHX will form part of the pressure boundary for
the VHTR and material selection and design will be
subject to regulatory requirements. In the United
States, the design will be guided by Section III of

the ASME Boiler and Pressure Vessel (B&PV) Code.
This section specifies materials and design data for
components in nuclear systems. Subsection NH of
the Code, which specifies materials and design parameters for materials that will undergo inelastic deformation, includes only a very few materials. The
temperature limits for Subsection NH Code materials,
other than bolting, at 300 000 h are listed in Table 4.
The maximum temperatures at which fatigue curves
are provided are also listed. Note that of the materials
under consideration for VHTR service, only Alloy
800H is currently contained in the appropriate section of the ASME Code and the service temperature
is limited to 760  C. Any VHTR design that has an
intended IHX service above this temperature will
require extension of the Code to include additional
materials, or at a minimum, extension of the use of
Alloy 800H to higher temperatures. Other regulatory
systems are in use internationally; however, it is generally true that additional materials and expanded
databases are required before a new VHTR design
can be finalized and licensed.
5.11.4.2

Alloy 617 (52Ni–22Cr–13Co–9Mo)


Alloy 617, also designated as Inconel 617, UNS
N06617, or W. Nr. 2.4663a, was initially developed
for high-temperature applications above 800  C. It is
often considered for use in aircraft and land-based
gas turbines, chemical manufacturing components,
metallurgical processing facilities, and power generation structures. The alloy was also considered and
investigated for the HTGR programs in the United
States and Germany in the late 1970s and early 1980s.
The high Ni and Cr contents provide the alloy
with high resistance to a variety of reducing and
oxidizing environments. In addition, the Al also


260

Material Performance in Helium-Cooled Systems

forms the intermetallic compound g0 -(Ni3Al) over a
range of temperatures, which results in precipitation
strengthening on top of the solid-solution strengthening imparted by the Co and Mo. Strengthening is
also derived from M23C6, M6C, Ti(C, N), and other
precipitates when in appropriate sizes, distributions,
and volume fractions.
Table 4
Materials specified in NH for elevated temperature service in nuclear applications
NH code materials
(other than bolting)

304 stainless steels

(UNS S30400,
S30409)
316 stainless steel
(UNS S31600,
S31609)
Alloy 800H
(UNS N08810)
2 1/4Cr 1Mo steel,
annealed condition
(UNS K21590)
Grade 91 steel
(UNS K90901)c

Maximum
temperature ( C)
For stress allowables
S0, Smt, St, Sr up to
300 000 ha

For
fatigue
curves

816

704

816

704


760

760

593b

593

649

538

a

The primary stress limits are very low at 300 000 h and the
maximum temperature limit.
b
Temperatures up to 649  C (1200  F) are allowed up to 1000 h.
c
The specifications for Grade 91 steel covered by subsection
NH are SA-182 (forgings), SA-213 (small tube), SA-335
(small pipe), and SA-387 (plate). The forging size for SA-182 is not
to exceed 4540 kg.

Table 5
Investigator

Observations and predictions of which precipitates
form in Alloy 617 at given temperature ranges have not

been consistent. A comprehensive review of the precipitates in Alloy 617 has been performed by Ren.18,19
Additional reviews can be found in Natesan et al.20
However, it is clear from the reviews that the kinetics
of the precipitation and coarsening processes are
important in determining the effects of aging on
properties. The g0 intermetallic is generally too fine
to be observed in optical microscopy.
Other phases that have been identified include
CrMo(C, N) and TiN,21 M12C and a possible Laves
phase,22 and a Ni2(MoCr).23 A summary of observations is given in Table 5. The apparent trend is that
in the temperature range of interest to the VHTR
IHX, precipitates may form at initial exposure and
the alloy may become stronger, but most of the precipitates will be dissolved after long-term exposure,
and the alloy will depend on solid-solution strengthening in the long run. The most recent information
on precipitation in Alloy 617 upon aging is shown in
the T–T–T diagram in Figure 8. The influence of
aging on the mechanical properties of the alloys
under consideration for IHX applications is discussed
in the section on environmental effects.
The grain size also plays an important role in
the strength of the alloy. For general applications, a
grain size of $45 mm or coarser is typically preferred,
but it has been shown that creep strength increases
with increasing grain size, so microstructures of
100–200 mm grain size are often produced. A tradeoff exists, however, when fatigue is an issue, since
finer grain sizes are preferred for fatigue resistance.
In addition, for compact IHX, the thin sheet form

Prediction and observations of second-phase precipitates in Alloy 617
M23C6


M6C

g0

Stable
Form
wt < 5% persist to
Thermocalc®
prediction
T 800  C
T ! 780  C
T ¼ 650  C
(ORNL)
Observation in material aged for 10 000 h and less
T 1093  C
Not observed
Small wt% persist to
Mankins21
and
T ¼ 760  C
Kimball24
1000  C
1000  C
Not reported
Kihara25
Observed
Observed
550–1000  C
Kirchhofer22

Observation in material aged for much longer than 10 000 h at 482–871  C
Observed
Observed
Observed at 482, 538, &
Wu23
also
593  C, not at 704  C
for 43 000 h and
observed
longer, nor 870  C
eta-MC
after long time

m

Ti(C,N)

wt >10% 600 $
800  C

Not reported

Not observed

Not observed

Not observed
Not observed

Not reported

400–1000  C

Not observed

TiN observed


Material Performance in Helium-Cooled Systems

261

1100

Temperature (ЊC)

1000

Limited γЈ
formation

900
Ti (C, N) + M6C + M23C6 + γЈ

800

Fine M4C + y Æ M23C6 + γЈ start

700
γЈ start extended
γЈ end


600
500
Ti (C, N) + M4C
400
0.1

1

Ti (C, N) + M4C+ M20C4
10

100
Time (h)

1000

10 000

100 000

Ti(C, N) + Coarse M6C + M23C6 + γЈ

Ti (C, N) + M4C4 + M4C + γЈ5

Ti (C, N) + M4C + Coarse/Fine M6C + γЈ

Ti (C, N) + M4C4 + M4C + γЈ5

Ti (C, N) + M23C4 + M6 + γЈ+ Ni2/(Mo,Cr)

Figure 8 Time-temperature-transformation (T-T-T) diagram for precipitation of phases in Alloy 617 upon aging.

restricts development of large grain size. Whether the
grains will significantly coarsen after the dissolution
of certain grain boundary precipitates at long-term
exposure is not clear.
The existing mechanical property database for
Alloy 617 is extensive (Table 6). This alloy has adequate creep strength at temperatures above 870  C,
good cyclic oxidation and carburization resistance,
and good weldability. It also has lower thermal
expansion than most austenitic stainless steels and
high thermal conductivity relative to the other candidates. It retains toughness after long-time exposure
at elevated temperatures and does not form intermetallic or Laves phases that can cause embrittlement.
Preliminary testing described later indicates that
Alloy 617 has the best carburization resistance of
the four alloys.
During early development, Alloy 617 was systematically studied by Huntington Alloys, Inc. for
applications in gas turbines, nitric acid production
catalyst-grids, heat-treating baskets, Mo refinement
reduction boats, etc. When Alloy 617 was considered
for the HTGR, it was extensively investigated
by Huntington, Oak Ridge National Laboratory
(ORNL), and General Electric (GE). The Huntington
data were used to develop ASME B&PV Code qualification, including the 1980s draft Code Case for the
HTGR and applications covered by (nonnuclear)
Section I and Section VIII Division 1. Alloy 617 is

not currently qualified for use in ASME Code
Section III, although it is allowed in Section I and
Section VIII, Division 1 (nonnuclear service). Efforts

to gain the approval from the ASME Code committees
for nuclear service were stopped when interest in
VHTR technology waned in the 1990s.
Both the ORNL-HTGR and GE-HTGR studies
generated data from Alloy 617 that had been aged
and/or tested in simulated HTGR helium. The
helium impurities were the same as those considered
for the VHTR system but the concentrations were
different. Unfortunately, only processed data still
exist; all original test curves needed for certain modeling efforts are irretrievable. Alloy 617 was also
extensively investigated in Germany for its HTGR
and other programs. The data generated were collected in the Online Data & Information Network
(ODIN). Original test curves, if not all, are stored in
the ODIN. However, the strain measurements of creep
test curves were not all conducted with fine resolution
and may not all be ideal for constitutive equation
development. The aging effects on Alloy 617 are summarized in Ren and Swindeman.18 The development
in modeling creep behavior of Alloy 617 is summarized in Swindeman et al.24
It is believed that creep–fatigue will be the most
significant failure mechanism for materials in the
IHX. Creep–fatigue damage results from cyclic loads
superimposed on materials subjected to temperatures


262

Material Performance in Helium-Cooled Systems

Table 6


Summary of testing done on Alloy 617

Research organization

Number of heats

Number of samples

Test type

Temperature ( C)

Huntington alloys

13 þ 1 wire

ORNLa

4 plate þ 1 wire

GEa

1 plate þ 1 bar

Germany

Not specified

179
249

73
51
25
1
36
7
40
302
1947
29
261

25–1093
593–1093
24–871
593–871
24
RT after 871
750–1100
950
850, 950
RT-1000
500–1000
700–1000
<500–1000

Honeywell aerospace
Gen IV programb

Not specified

Not specified

Tensile
Creep
Tensile
Creep
Charpy
Tensile after creep
Creep
Creep–fatigue
Fatigue
Tensile
Creep
Creep crack growth
Low cycle fatigue
Not specified
Creep–fatigue

80

800, 1000

a

Some tests exposed to HTGR environment.
Air, pure helium and vacuum environments.

b

5.11.4.3


Alloy 230 (57Ni–22Cr–14W–2Mo–La)

Alloy 230, also designated as Haynes 230, UNS
N06230, or W. Nr. 2.4733, is a newer alloy than
Alloy 617. In addition to outstanding resistance to
oxidizing environments, Alloy 230 has good weldability and fabricability. It also has a lower thermal
expansion coefficient than Alloy 617; it appears that
thermal expansion has an inverse correlation with Ni
content. Alloy 230 has a higher tensile strength than
Alloy 617 up to 800  C, but above that the difference is

105

Cycles to failure (Nf)

and loads that will induce creep damage under monotonic loading. Recent creep–fatigue data for Alloy 617
and Alloy 230 are shown in Figures 9 and 10 for
tests that involved fully reversed cyclic loading at
total strain ranges of 0.3% and 1% with varying
hold time during the tensile portion of the cycle at
800 and 1000  C. The plots show the reduction in
cycles to failure with increasing tensile hold time
under creep loading conditions. It can be seen that
in general, increased hold time results in decreased
cycles to failure. At 800  C, the two alloys have
similar behavior; however, at 1000  C, Alloy 617
appears to have somewhat higher cycles to failure
compared to Alloy 230. A limited number of tests
have been carried out on specimens that contain

weldments and it has been found that the cycles to
failure in specimens containing a fusion weld is
reduced. In these specimens, the cracking is typically in the weld metal and not in the heat-affected
zone or at the weld–base metal interface.

Alloy 617 – 0.3% total strain
Alloy 617 – 1.0% total strain
Alloy 230 – 0.3% total strain
Alloy 230 – 1.0% total strain
Alloy 617CCA – 0.3% total strain
Alloy 617CCA – 1.0% total strain

104

1000

No hold
100
10

100

1000

Hold time (s)
Figure 9 Effect of hold time on cycles to failure in
creep–fatigue at 800  C.

insignificant. It appears that Alloy 617 has slightly
better creep properties than Alloy 230. Alloy 230

has a better thermal fatigue crack initiation resistance
but a worse thermal cycling resistance compared to
Alloy 617.
The Ni base and high Cr content impart resistance
to high-temperature corrosion in various environments, and oxidation resistance is further enhanced
by the microaddition of the rare earth element La.
Compared to Alloy 617, Alloy 230 has a high W
concentration which replaces much of the Co in


Material Performance in Helium-Cooled Systems

105

Cycles to failure (Nf)

No hold

Alloy 617– 0.3% total strain
Alloy 617–1.0% total strain
Alloy 617CCA – 0.3% total strain
Alloy 617CCA – 1.0% total strain
Alloy 230 – 0.3% total strain
Alloy 230 – 1.0% total strain

104

103

102

10

100
1000
Hold time (s)

10 000

Figure 10 Effect of hold time on cycles to failure in
creep–fatigue at 1000  C.

Alloy 617. The W and Mo in conjunction with C are
largely responsible for the strength of the alloy, and its
relatively high B content in comparison to that in Alloy
617 can be controlled to achieve optimized creep resistance. Usually, B acts as an electron donor; it can affect
the grain boundary energy and help improve ductility.
In Ni-based alloys, B also segregates to grain boundaries and helps to slow grain boundary diffusion, thus
reducing the creep process. On the other hand, excess
boron in a neutron field could also lead to embrittlement due to transmuted He, although irradiation is not
a factor for IHX applications.
In the solution-annealed condition in which this
alloy is typically supplied, the grain size is typically
>45 mm with large carbide precipitates rich in W,
presumably of the M6C type. After aging, Alloy 230
typically exhibits M6C and M23C6 precipitates. After
aging for 1000 h at 850  C, very small carbide precipitates rich in Cr and M23C6 were observed along
the grain boundaries. No grain coarsening was
observed.25 Creep strength is believed to be brought
about by solid-solution strengthening, low stacking
fault energy, and precipitation of M23C6 carbides on

glide dislocations.26,27 However, a negative impact
of M23C6 on room temperature ductility was also
reported. After aging at 871  C for 8000 h, the room
temperature tensile elongation of Alloy 230 decreased from $50% to 35%, with a precipitation of
M23C6 observed in microstructural examination, but
an additional 8000 h of aging did not further decrease

263

ductility.26 Significant microstructural changes were
also observed after thermal aging in air for 10 000 h
at temperatures ranging from 750 to 1050  C. After
the 750  C aging, coarser intergranular precipitation
of M23C6 and coarse and blocky intra- and intergranular precipitates of M6C were observed. After
the 850–1050  C aging, the M6C carbides were irregular in shape. After aging at 1050  C, the secondary
intragranular M23C6 appeared to have dissolved.
A decrease in toughness and ductility coincided with
the appearance of the intragranular M23C6 and reached
a minimum after the aging at 850  C. The toughness
and ductility recovered after the aging at 1050  C.28
There is less characterization of Alloy 230 compared to Alloy 617. The major known large-scale
study was tensile and creep tests by Haynes International. Creep times ranged from 15.3 to 28 391 h. Like
Alloy 617, Alloy 230 is not currently qualified for use
in ASME Code Section III, although it is allowed in
Section VIII, Division 1 (for nonnuclear service). At
present, the database for Alloy 230 is significantly
smaller than that for Alloy 617 and a much larger
effort is required to develop an Alloy 230 Code Case
for elevated temperature application. Some recent
data on environmental effects of exposure to prototypical VHTR chemistries are given in the following

sections and creep–fatigue properties are included
in Figures 9 and 10.
5.11.4.4

Alloy 800H (42Fe–33Ni–21Cr)

This alloy is the only iron-based alloy under consideration, although it has a solid-solution strengthened
austenitic structure like the other three alloys. Upon
aging, precipitates can form and somewhat reduce
the tensile and creep ductility. Alloy 800H has the
lowest creep rupture strength and the lowest resistance to oxidation of the four alloys. There is an
additional variant of this alloy, 800HT, that has a
composition similar to that of 800H, but has an additional specification for coarse grain size. The majority
of material that is currently available in this alloy
series is Alloy 800HT, which also meets the specification for Alloy 800H.
Among the four candidate materials, Alloy 800H
is the only one that is Code qualified for use in
nuclear systems, but only for temperatures up to
760  C and a maximum service time of 300 000 h.
Alloy 800H was the primary high-temperature alloy
used in the German HTGR programs and an enormous amount of data were obtained. However, only
very limited data from the German HTGR programs


264

Material Performance in Helium-Cooled Systems

are currently available on the mechanical properties
of this alloy beyond 800  C, especially in impure

helium environments.

5.11.5 Welding
All of the solid-solution alloys that have been mentioned are readily welded using conventional fusion
welding methods. Alloys 617 and 230 are described in
more detail later as prototypical of these materials.
Alloy 617 has excellent weldability. Alloy 617 filler
metal is used for gas-tungsten-arc (GTAW) and gasmetal-arc welding (GMAW). The composition of the
filler metal matches that of the base metal, and deposited weld metal is comparable to the wrought alloy in
strength and corrosion resistance.29 Alloy 230 is also
readily welded by GTAW and GMAW. Shielded
metal-arc welding (SMAW) and resistance welding
techniques can also be used. Submerged-arc welding

Stress (MPa)

Alloy X has the best oxidation resistance of the four
alloys, although its carburization resistance is the
worst. Above 700  C, Alloy X can form embrittling
phases that result in property degradation. The creep
rupture strength is not as good as Alloy 617 or 230.
The limitations of this alloy will be similar to the
draft code case for Alloy 617 in terms of grain size,
product form, and limitations on service time.
A limited database exists for Alloy X for conditions typical of a VHTR, but the high-temperature
scaling in Hastelloy X has been less than optimal. As a
result, a modified version, Alloy XR, has been developed in Japan; however, the United States has little
access to Alloy XR material, either for evaluation or
for ASME Code qualification. Japanese are currently
using Alloy XR in a heat exchanger in the HTTR at

temperatures of 850–950  C. The material is codified
in Japan for nuclear use, which would likely accelerate code acceptance in ASME. An extensive environmental database and HTGR experience exist.
However, the database may be limited to large grain
material, similar to the Alloy 617 draft code case.
Also, similar to the Alloy 617 draft code case, Alloy
XR may have issues with weldments that need to be
addressed. It is uncertain if this alloy is readily available as a commercial product.
Figures 11–14 compare the creep rupture
strength, oxidation behavior, carburization behavior,
and allowable stress for the four alloys, respectively.

Alloy X

60

Alloy 230

50

Alloy 617

40
30
20
10
0
1

1000
100

Time to rupture (h)

10

10 000

100 000

Figure 11 Creep rupture strength at 962  C in air.

Thickness (mm)

Alloy X (47Ni–22Cr–9Mo–18Fe)

Alloy 800H

70

20
10
0
−10
−20
−30
−40
−50
−60
−70
−80
−90


Alloy 800H

Alloy 617

Oxide scale

Alloy 230

Affected zone

Alloy X
Internal oxide

Figure 12 Schematic representation of isothermal
oxidation behavior after 800 h exposure at 950  C in helium
environment.

24
Normalized weight gain

5.11.4.5

80

Alloy X
19

Alloy 230
Alloy 617


14
9
4
−1

0

200

400

600

800

1000

1200

Exposure time (h)
Figure 13 Mass change as a function of time in H2–5.5%
CH4–4.5% CO2 carburizing environment at 1000  C.

is not recommended, as this process is characterized
by high heat input to the base metal and slow cooling
of the weld. These factors can increase weld restraint
and promote cracking. The as-welded properties of
these alloys are given in Table 7.30 The welds exhibit
room temperature strength that matches or is slightly



Material Performance in Helium-Cooled Systems

265

better than the base metal, but a considerable decrease
in ductility is observed at elevated temperatures, as
shown in Table 8.

alloy, HASTELLOY S alloy, or HASTELLOY W
alloy welding products may all be considered, depending upon the particular case.29,30

5.11.5.1 Base Metal Preparation and Filler
Metal Selection

5.11.5.2 Preheating, Interpass
Temperatures, and Postweld Heat
Treatment

Prior to any welding operation, the welding surface
and adjacent regions should be thoroughly cleaned
with an appropriate solvent. All greases, oils, corrosion products, and other foreign matter should be
completely removed. It is preferable, but not necessary, that the alloy be in the solution-annealed condition when welded.30
Alloys 617 and 230-W™ (AWS A5.14,
ERNiCrWMo-1) filler wire are recommended for
joining Alloy 617 and 230, respectively, by GTAW
or GMAW. The filler metals are not specifically
designed for nuclear application. For dissimilar
metal joining of Alloy 230 to nickel-, cobalt-, or

iron-based materials, 230-W filler wire, Alloy 556™

Preheat is not required, generally room temperature
(typical shop conditions) is adequate. Interpass temperature should be maintained below 93  C. Auxiliary
cooling methods may be used between weld passes, as
needed, providing that such methods do not introduce contaminants. Postweld heat treatment is not
generally required either. Table 9 shows the nominal
welding parameters based on welding conditions
used in the Haynes International laboratories and
should serve as a guide for performing typical
GTAW and GMAW operations on Alloy 230. All
processes used 230-W filler wire.30
5.11.5.3

Maximum allowable stress (MPa)

250
Alloy X
Alloy 230

200

Alloy 617
Alloy 800H
150
100

50

0

0

200

400

600

800

1000

Temperature ( ЊC)
Figure 14 Allowable stress for heat exchanger materials
for plate, sheet, and strip forms from the ASME boiler and
pressure vessel code Section VIII.

Table 7

Nontraditional Joining Methods

As noted earlier in the description of IHX designs,
several of the compact heat exchanger design concepts will require the joining of sheet product to be
either diffusion bonding or brazing. Diffusion bonding of these alloys is relatively well developed
because of applications in aerospace systems that
require this fabrication method. The etch plate compact design fabricated from austenitic stainless steel
has been commercialized for petrochemical applications, and limited diffusion bonding studies have
been completed using Alloy 617. Characterization
of diffusion-bonded stacks of sheet indicates that
mechanical properties comparable to base metal can

be achieved at room temperature. The details of diffusion bonding parameters are considered proprietary
by the IHX vendors, and it is not clear whether
temperatures sufficiently high to cause carbide dissolution and/or grain growth is a matter of concern.

Room-temperature tensile properties of joints in as-welded condition

Alloy

Specimen

Yield strength (0.2%
offset) (MPa)

Tensile strength
(MPa)

Elongation
(%)

Reduction of area
(%)

61729

GMAWa
GTAWb
GMAWc

510
542

490

761
823
785

43.3
37.3
48.2

42.0
38.3

23030
a

Alloy Filler Metal 617. Average of ten tests.
Alloy Filler Metal 617. Average of 17 tests.
Alloy 230-W filler wire.

b
c


266

Material Performance in Helium-Cooled Systems

Table 8


Tensile properties of 230 base and weld metals
23  C

GMAW deposit weld metal
Cold-rolled and 1232  C solution annealed (sheet)
Hot-rolled and 1232  C solution annealed (plate)
Vacuum investment castings (as-cast)

538  C

871  C

UTS

YS

EL

UTS

YS

EL

UTS

YS

EL


785
838
840
615

490
422
375
325

48.2
47.2
47.7
37.8

610
699
690
450

435
303
251
230

34.8
53.7
54.6
38.2


310
308
315
285

275
234
242
185

45.4
75.0
99.5
19.0

Source: HAYNESW 230W Alloy. Haynes International, Inc. Publication H-3000H, 2004.

Table 9

Weld parameters for Alloy 230

Welding method

GMAW

Configuration (mm)

Thickness > 2.3 1.1
dia. wire


Technique

Stringer bead or
slight weave
100–130a
18–21
4.3–4.8
12.7–19.1
203–356
Torch, 50
Ar-25% He

Current (A)
Voltage (V)
Feed rate (m minÀ1)
Stick-out (mm)
Travel speed (mm minÀ1)
Gas flow (l minÀ1)
Gas

GTAW
Auto

Manual

Square butt joints 1.0/1.6/3.2
thick, 1.6 electrode with
45  included shape
No filler metal added
50/80/120b

8.0/8.5/9.5

V or U groove, >3.6 thick, 3.6 dia.
wire, 3.6 electrode with 30 
included shape
Stringer bead interpass
T < 100  C
120 root, 140–150 fillb
11–14

10/12/12
Shield, 14.2 backing, 4.7
Argon

102–152
Shield, 14.2–16.5 backup, 4.7
Argon

a

DCEP, torch flow CFPH.
DCEN.
Source: HaynesW 230W Alloy, Haynes International High-Temperature Alloys.
b

Very little information on brazing these alloys is
available. A general concern is that low melting
point braze materials could result in poor elevated
temperature properties in structures fabricated by
these methods.


5.11.6 Control Rod Materials
The pebble bed modular reactor (PBMR) is the most
complete recent design for a VHTR. The reactor was
designed to operate at about 400 MWt and primarily
to produce electricity. Recent changes in the global
economic climate have caused reconsideration of the
design for a VHTR in South Africa; however, the
analysis that went into the design and selection of
materials for the control rods is illustrative of the
most recent analysis of these issues. A schematic of
the PBMR core is shown in Figure 15. The design
outlet gas temperature for the PBMR was 900  C; the
core was designed to be 11m high and 3.7 m in

diameter, and the annulus filled with about 452 000
60-mm-diameter fuel pebbles.31
The PBMR builds on the German experience of
the AVR and THTR; however, it will use a direct
cycle to produce power rather than a steam generator,
and it will have an annular core configuration with a
solid graphite central reflector. The annular core
produces several advantages: it shifts the peak power
radially outward, thus enabling significantly higher
output; it enhances the fuel safety margin; and, by
increasing the neutron flux in the outer graphite
reflector, it increases the effectiveness of the control
and shutdown systems.32
The reactor control and shutdown system (RCSS)
has two components: the reactivity control system

(RCS) and the reserve shutdown system (RSS). The
RCS consists of 12 control rods and 12 shutdown rods,
located in the outer reflector.33 They are evenly spaced
around the core and at a radial distance of about 70mm
from the inner surface of the reflector (see Figure 16).34
During normal operation, the control rods, which


Material Performance in Helium-Cooled Systems

Reactor pressure
Core barrel
Top
Side

Cold gas riser
Center
Pebble bed
Bottom
Inlet
Inl
Hot gas
Outlet

Figure 15 Schematic of the pebble bed modular reactor
(PBMR) annular pebble bed reactor. Reproduced from
Kriel, W. Material selection: High-temperature metallic
materials. Slides, Sept 21–22, 2005.

Core barrel


267

penetrate a maximum distance of 1.5m into the core,34
are used for minor reactivity adjustments to keep the
reactor critical, provide reactivity compensation for
xenon poisoning effects during load following effects,35
and allow for some excess reactivity so that the reactor
may continue operation for some time if no fuel is being
loaded.36 They are also used for hot shutdown purposes.33 The reactor power is actually adjusted by regulating the mass flow rate of the gas inside the primary
circuit rather than by adjusting the control rods.32,33
During scram, the additional 12 shutdown rods are
lowered to the bottom of the active core. In the event
of a loss of electrical power, insertion of the rods is by
gravity. The first set of control rods will drop, and later
the shutdown rods will drop, should the need arise.
The RSS consists of eight storage containers of
10-mm-diameter small absorber spheres containing
B4C that can be fed by gravity into eight channels in the
central reflector. The RSS serves as both the secondary
shutdown system and the cold shutdown system. It
must be activated in addition to the RCSS to bring
the PBMR to a cold shutdown condition (100  C).
The control rod design is similar to that of previous metal control rods. A schematic is shown in
Figure 17. A number of annular B4C rings are
encased between two tubes of Alloy 800H, to form a
section about a meter long. One unique feature is that

Reactor pressure
vessel


Side reflector
barrel
Reactor
inlet pipes
Annular core

Reactor
outlet pipe

Center reflector

Gas riser
channels

Small absorber
sphere channels
Control rod

Figure 16 Top view of the PBMR core, showing the location of the control rod channels, fuel pebbles, small absorber
sphere channels, and other features. Reproduced from PBMR. Data and boundary conditions to be used in VSOP, TINTE,
and MCNP PBMR 400 MW ( Th) reactor models.


268

Material Performance in Helium-Cooled Systems

Control rod drive
mechanism


RCS chain

Control rod segment

Control rod link

Secondary shock absorber

Figure 17 Schematic of the control rod assembly in the
PBMR. Reproduced from Broom, N.; Smit, K. PBMR Design
Methodology. Slides, Oak Ridge, TN, 12th April 2005.

the inner tube is much thinner than the outer tube.
These sections are mechanically linked to form an
articulated control rod several meters long. One difference from past designs (e.g., the AVR) is that the
control rod is suspended from the drive mechanism
by a chain, rather than a cable. A secondary shock
absorber is in place in the channel below the control
rod to protect it and the core structure in the event of
a chain failure. Additional shock absorbers within the
drive mechanism dampen the impact load on the
control rod drives during scram.33 A control rod
guide tube (not shown in Figure 16) connects the
control rod drive mechanism to the core structure to
guide the control rod into the core.37
During normal operation, the temperature of the
control rods is estimated to be from about 650 to
700  C,38,39 and the temperature resulting from a
depressurized loss of coolant (DLOC) event is estimated to be only 850  C.39 The end-of-life fast fluence

is reported as 2 Â 1022 (E > 0.1 MeV),38 and the thermal

fluence is reported as 5 Â 1021 n cmÀ2.39 The secondary
shock absorbers have an operating temperature of
900  C; during DLOC, they can be subjected to temperatures of up to 1100  C for short periods. Under
these conditions, the use of Alloy 800H was justified
by the PBMR program because the high-temperature
strength and creep resistance are sufficiently qualified for long-term normal operation at 700  C, and
limited operation above 850  C under abnormal
events can be tolerated according to available data.
The response of Alloy 800H to neutron irradiation
at the temperatures expected in the control rod sleeves
is not well characterized. The PBMR design notes that
the irradiation response has been characterized to high
levels of fast fluence at lower temperatures and Alloy
800H control rods have had extensive qualification and
service in previous German VHTR programs. Limited
data from older VHTR programs on irradiation effects
in Alloy 800H at temperatures above 600  C suggest
that helium embrittlement from (n, a) reactions associated with thermal neutrons is the predominant
degradation mechanism. Recent work has examined
property changes associated with irradiation to
1.45 dpa at temperatures of 580 and 660  C.40,41 Significant strengthening was observed along with a sharp
decrease in ductility. Material irradiated at 660  C and
subsequently tensile tested at 700  C showed tensile
elongation of less than 0.5%. The mechanisms of
embrittlement are not yet completely clear and irradiation experiments to higher fluence at higher temperature will likely be required to examine this issue.
The PBMR design includes specialized equipment
to remove and replace the control rods, as well as
storage for used control rods. The RCSS will be

inspected every 6 years during the scheduled maintenance outage, and repaired as necessary. These
outages are planned to last 30–50 days, depending on
the other maintenance scheduled, with the exception
of a 180-day shutdown after 24 years to replace the
core reflector.34,42
Eventually, it is hoped that a VHTR similar in
design to the PBMR reactor can run with an outlet
temperature of 1000  C or even higher. In this case,
carbon fiber-reinforced carbon composites (Cf/C)
or silicon carbide fiber-reinforced silicon carbide
composites (SiCf/SiC) must be considered for the
more challenging temperatures of the control rods.39
Experience with irradiation of SiCf/SiC composites
for nuclear fusion applications suggests that these
materials have superior resistance to property degradation from neutron irradiation as well as resistance
to higher temperatures and could potentially have


Material Performance in Helium-Cooled Systems

lifetimes comparable to the life of the plant. As with
ceramic IHXs, application of these advanced materials
in a nuclear system would require considerable further development and cooperation with appropriate
standards and regulatory organizations.

5.11.7 Core Barrel Materials
Another evidently important metallic internal structure shown in Figures 15 and 16 is the core barrel.
The function of the core barrel is to mechanically
contain the shape of the graphite blocks making up
the core and to channel the flow of the primary

coolant. Although schematics in Figures 15 and 16
are specific to a pebble bed design, the core barrel is
essentially identical in design and function for a
prismatic design as well. The temperatures, neutron
fluence, and mechanical loads on the core barrel are
moderate and the PBMR design, for example, proposed the use of Type 316 stainless steel for this
application. Alloy 800H is also a leading candidate
for the core barrel material. Although the demands
on the material are modest in terms of mechanical
loading and neutron irradiation, it is a very large
structure ($8 m high by 3.5 m diameter and 50 mm
in thickness) that will need to be fabricated on site by
welding in most cases. Both Type 316 stainless steel
and Alloy 800H are available in the required size and
are readily fabricated.

5.11.8 Environmental Effects of
VHTR Atmospheres on Materials
All the high-temperature reactor systems operated to
date had extensive gas cleanup systems associated with
the helium coolant flow. These systems are intended to
keep the total impurity levels in the helium below
typically 10 ppm. Particularly in the early reactors,
where the fuel was either not intended to contain the
fission products or was ineffective in this function,
the cleanup systems were also intended to capture
radionuclides.1,3 Capture of tritium that is produced
(at least in part) by transmutation of lithium impurities
in the graphite remains an important function of the
cleanup system. In the AVR and THTR reactors,

active control was maintained on the H2O and CO
concentrations to reduce oxidation of the graphite
reflectors, and the other impurities were routinely
found to reach acceptable steady-state levels without
active control.4–7,43,44 It has been noted that the

269

cleanup systems may play a secondary role in maintaining gas chemistry, with the massive amount of
graphite at high temperature present in all of the
reactor designs playing a dominant role.1
Molecular sieves are effective in capturing most of
the gaseous impurities; however, they have difficulty
capturing H2 and CO. To resolve this problem, the
gas stream is passed over a bed of CuO that oxidizes
the H2 to H2O and CO to CO2 upstream of the
molecular sieve where these gases are effectively
removed. The Peach Bottom plant attempted the
use of heated Ti getters for hydrogen and tritium;
however, these were not effective and oxidation of the
H2 prior to removal is now accepted practice.2 In a
typical plant, up to about 20% of the gas stream is
diverted to the cleanup system each hour.
Table 10 shows the impurity levels reported for
steady-state operation for several of the VHTRs.1,2,43,44
As shown in the table, at steady state, all of the reactors
for which operating data are available had similar levels
of impurities. Some caution should be exercised when
comparing the data for different plants, since, in some
cases, there are varying values reported in different

publications for the same plant. This may be associated
with conversion from partial pressure of impurities
(the preferred units for corrosion studies) to ppm by
volume (the typical units used for comparison of one
plant to another). Several plants have undergone extensive postmortem analysis of the core internals and heat
exchangers.1,2 There are reports of some oxidation and
at least one report of massive deposition of carbon on
the internals, as discussed in more detail in the following paragraphs; however, there have been no problems
with failure of components on the primary side associated with environmental effects.
There have been a large number of experimental
studies and modeling of the effect of VHTR
helium on the high-temperature alloys listed in
Table 3.10–12,16,17,45–54 Depending on the specific
proposed application, different model chemistries
have been developed, and the testing has focused on
these. Several of the model impurity chemistries are
shown in Table 11.45–52,54 Comparison of the values
in Table 11 with actual operating experience suggests that the model chemistries tend to have higher
impurity levels of some species than those found in
operating reactors. This is notable for H2 in particular.
It is not clear why these particular values were chosen;
however, it can be noted that several of the proposed
applications were for process heat for coal gasification, and there was concern that hydrogen would
diffuse from the process plant into the primary


270

Material Performance in Helium-Cooled Systems


Table 10

Impurities reported in the helium coolant during steady-state operation of VHTRs (in ppm)

Dragon
Peach bottom
Fort St. Vrain
AVR
THTR

H2O

H2

CO

CO2

CH4

O2

N2

0.1
0.5
1
0.15
<0.01


0.1
10
7
9
0.8

0.05
0.5
3
45
0.4

0.02
<0.05
1
0.25
0.2

0.1
1.0
0.1
1
0.1

0.1
–
–

0.05
0.5

–
22
0.1

Source: Simon, R. A.; Capp, P. D. Operating experience with the dragon high temperature reactor experiment. In Proceedings of the
Conference on High Temperature Reactors, Petten, NL, Apr 22–24, 2002; pp 1–6.
Burnette, R. D.; Baldwin, N. L. Specialists Meeting on Coolant Chemistry, Plate-Out and Decontamination in Gas Cooled Reactors,
Juelich, FRG, Dec 1980; International Atomic Energy Agency, 1980; pp 132–137.
Nieder, R. Specialists Meeting on Coolant Chemistry, Plate-Out and Decontamination in Gas Cooled Reactors, Juelich, FRG, Dec 1980;
International Atomic Energy Agency, 1980; pp 144–152.
Nieder, R.; Stroter, W. VGB Kraftwerstech. 1988, 68, 671–676.

Table 11
Model impurity chemistries used in environmental testing programs (compositions in ppm). HHT, PNP were
used for German nuclear process heat projects

HHT
PNP
AGCNR
JAERI B

H2O

H2

CO

0.75
0.75
1

0.5

250
250
200
100

20
7
20
50

CO2

CH4

0.1
1

25
10
10
2.5

O2

N2
5
<2.5
<2.5

<2.5

AGCNR was a German VHTR, and JAERI B composition was extensively studied in development of the HTTR.
Source: Nieder, R. Specialists Meeting on Coolant Chemistry, Plate-Out and Decontamination in Gas Cooled Reactors, Juelich, FRG, Dec
1980; International Atomic Energy Agency, 1980; pp 144–152.
Nieder, R.; Stroter, W. VGB Kraftwerstech. 1988, 68, 671–676.
Bates, H. G. A. Nucl. Technol. 1984, 66(2), 415–428.
Brenner, K. G. E.; Graham, L. W. Nucl. Technol. 1984, 66(2), 404–414.
Christ, H. J.; et al. Mater. Sci. Eng. 1987, 87, 161–168.
Christ, H. J.; et al. Oxid. Metals 1988, 30, 27–51.
Christ, H. J.; et al. Oxid. Metals 1988, 30, 1–26.
Fujioka, J.; et al. Nucl. Technol. 1984, 66(1), 175–185.
Inouye, H. Nucl. Technol. 1984, 66, 392–403.

coolant circuit.45 Note that N2, at concentrations
similar to those listed in Table 10, has never been
found to contribute significantly to environmental
interactions with nickel-based alloys.48,49
Interplay between the alloy surface, temperature,
and gas composition determines whether corrosive
oxidation, carburization, or decarburization occur.
The corrosion mechanisms of particular significance
to mechanical stability are carburization and decarburization. Carburization is associated with lowtemperature embrittlement, and decarburization is
linked to reduced creep rupture strength. Ideally, a
continuous self-healing, impermeable passivating oxide
layer is needed to establish the most corrosion-resistant
alloy. In the case of Alloy 617, the chromia layer
(Cr2O3) is the most important barrier to the effects
of corrosive reactor gases.
As noted earlier, of the existing materials, Alloy

617 is the leading candidate for use in the VHTR

heat exchangers because it has the highest creep
strength of the solid-solution alloys under consideration for temperatures above 850  C. Evaluation
of this alloy for VHTRs began in the early 1980s,
with the most comprehensive work done by Brenner
and Graham,46 Christ et al.,47–49 Graham,51 and
Quadakkers and Schuster.54 Alloy 230 is under consideration as an alternative to Alloy 617 because it
has equivalent creep properties and may suffer from
less internal oxidation.
Based upon the work of Quadakkers and others,
assessments of Alloy 617 stability at various gas concentrations and temperatures can be displayed graphically. Quadakkers used a diagram of the type
shown in Figure 18 to display the results of the
stability calculations for the nickel–chromium
alloy.47–49,54 Five conditions are represented within
the diagram: I – strong reduction (decarburization
without a surface oxide); II – decarburization (with


Material Performance in Helium-Cooled Systems
Ã
3COþ2Cr ¼ Cr2 O3 þ3C
PCO
CO pressure for this equilibrium

CrnCm

log ac

III

I

II

½IIIŠ

The measured steady-state conditions (AGCNR gas
composition) determine where the alloy sits within
the modified chromium diagram; hence, the steadystate carbon activity (acss ) and the steady-state oxygen
pressure (POss2 ) are tentatively calculated with the
following equations (provided the kinetics of methane splitting is low compared to the water vapor
dissociation on the alloy surface):

Cr2O3

IV
V

271

P*CO

Cr metal

ss1=2

log Po2
Figure 18 Modified chromium stability diagram at a given
temperature; Zone III is preferred for optimal chromia layer
protection against corrosion.


a porous oxide); III – stable external oxide (with
stable internal carbides); IV – mixed surface oxide
and carbide layers (with internal carburization); and
V – strong internal and external carburization. Zone III
was determined to be the area of highest stability; an
environment that is oxidizing and slightly carburizing.
For a heat exchanger operating at temperatures of
900  C and assuming standard gas composition,
called AGCNR helium and shown in Table 11,
Alloy 617 would be in zone III, for example.
The diagram describing alloy stability was based
on the most relevant species involved in the corrosion
process, namely chromium. Identifying which form –
Cr2O3 chromium carbide or chromium metal – is
most stable in a particular environment will determine the ultimate fate of the alloy.55 It is important to
note that the gas chemistries found in operating reactors and used in the previous test programs are not in
thermodynamic equilibrium. A steady-state gas composition is reached at any temperature based on
kinetic considerations. As is shown later, the concentrations of H2O and CO largely determine the partial
pressure of oxygen and the activity of carbon, respectively. The important features of the diagram are
critical carbon activation (acà ), critical partial pressure
of oxygen (POÃ 2 ), and the critical partial pressure of
Ã
carbon monoxide (PCO
). At a given temperature,
these parameters are calculated from the following
thermodynamic reactions:
acÃ

23Cr þ 6C ¼ Cr23 C6

metal À carbide equilibrium activity

½IŠ

POÃ 2

Cr2 O3 ¼ 2Cr þ 1:5O2
disassociation pressure of chromia

½IIŠ

ss
=PO2 Þ
acss CO ¼ Cþ0:5O2 acss ¼ K ðPCO

½IVŠ

POss2 H2 O¼H2 þ½OŠ POss2 ¼ðKPH2 O =PH2 Þ2
ð1ÀðPCH4 =PH2 O Þ1=100Þ

½VŠ

A chromium activity of 0.75 was assumed at the Alloy
617 surface.48,49
At very high temperatures, there is a critical temperature above which the oxide layer is unstable and
CO evolution will occur. This critical temperature
represents the maximum application temperature for
the alloy. The results of an experiment for Alloy 230
showing oxide instability and CO evolution are given
in Figure 19 for a specimen that was held at a constant

temperature of 900  C to establish a surface oxide before
rapidly increasing the temperature to 1000  C. The
chemical reaction that gives rise to CO evolution is
2Cr2 O3 þ Cr23 C6 ¼ 6CO þ 27Cr

½VIŠ

It is also significant to note the rapid rate at which
the reaction occurs, indicating that significant
changes in surface condition of this alloy can occur
in a few hours.
The reaction [VI] (also known as the microclimate
reaction) suggests that degradation of the protective
oxide on the surface of the IHX alloys under consideration can be suppressed by increasing the partial
pressure of CO. Figure 20 shows the upper temperature for oxide stability as a function of CO partial
pressure for both Alloy 617 and Alloy 230.
The onset of reaction [VI] occurs at a particular
temperature, TA, when the CO concentration is no
longer sufficient to drive the reaction from right to
left. It results in eventual total loss of either chromium
oxide or carbide, depending on concentration, then
in complete carburization or decarburization of the
alloy, depending on the gas composition. This degradation mechanism is considered in the stability diagrams developed using the approach of Quadakkers
Ã
, and it has been discussed extensively by
as PCO
Brenner and Graham,46 Christ et al.,47–49 and Graham.51


Material Performance in Helium-Cooled Systems


40

1000

Partial pressure (mbar)

35

T(°C)

800

CO

30
PCO(mbar)
PCH4(mbar)

25

600

P(CO)inlet

20

400

15


Temperature (°C)

272

CH4
200

10

5

0

5ϫ104

1ϫ105

0

2ϫ105

1.5ϫ105

Time (s)
Figure 19 Experimental determination of oxide instability (as evidenced by CO evolution) for Alloy 230 in He with 200 ppm
H2, 21 ppm CO, 19 ppm CH4, and 0.5 ppm H2O.

1275
1250

Fit a(Cr) = 0.72
TA In K

1225
1200
1175
Alloy 230
Alloy 617
Alloy 617 after(20)

1150
1125
0

1

2

3
P(CO) ln Pa

4

5

6

Figure 20 The critical temperature for oxide instability for Alloy 617 and Alloy 230 as a function of CO partial pressure.

At the temperature TA, the reaction will go to completion, and as a result, this temperature represents a

maximum use temperature for the particular alloy for
a given gas composition. Experimental results for this
reaction are shown for Alloys 617 and 230.
A shortcoming of the stability diagram approach
presented earlier is that it does not account for the
fact that Cr2O3 becomes volatile at a temperature
above about 950  C.56 There may be sufficient oxygen partial pressure to form the oxide as predicted
from the modified stability diagram; however, it will

not be protective because of the vapor pressure of
the oxide. Most laboratory studies have been at
very low flow rates to more closely approach thermodynamic equilibrium for fundamental studies
of corrosion mechanisms. It has been noted that
these conditions may not be representative of reactor systems, where very high gas velocities are
likely, for example, 75–100 m sÀ1 at the outlet of the
VHTR.48,49 With very low levels of impurities,
this increases the possibility that impurities will be
depleted during the experiments and may give rise


Material Performance in Helium-Cooled Systems

to anomalously low values for Cr2O3 vaporization in
test systems compared to reactor operation.46–49,51,52
Micrographs of cross-sections from Alloy 617 and
Alloy 230 plate material after exposure to heliumcontaining impurities that resulted in decarburizing
atmosphere at 1000  C are shown in Figure 21(a)
and 21(b), respectively. The microstructures are
largely as anticipated. Alloy 617 shows a relatively
thick chromium oxide scale with significant formation

of grain boundary aluminum oxides. Alloy 230 shows
less surface oxidation and notably reduced tendency
for formation of grain boundary oxides. A decarburization region is also apparent for both alloys; the
decarburized region is particularly notable in Alloy
230 shown in Figure 21. Internal oxidation may be
of particular concern in alloy selection for compact
heat exchangers where very thin material sections may
be encountered.
Carburization of Alloy 617 has been examined in
previous work and the results shown in Figure 22 for
this alloy after exposure to carburizing conditions at
1000  C are largely consistent with behavior reported
in the literature. For the conditions examined, here
there is little formation of a surface oxide scale.
Carbon uptake in the material is manifested by
increased grain boundary carbide precipitate volume
fraction. The behavior of Alloy 230 is markedly different from that of Alloy 617. While there is little
evidence of scale formation, there is a very large
volume fraction of carbide formation in the alloy
that heavily decorates both grain and twin boundaries. This large volume fraction of carbides was
found through the entire thickness of the $3 mm
thick coupon after 500 h at 1000  C. A micrograph
from the center of an Alloy 230 coupon is shown in
Figure 23.
To investigate the effect of environmental interaction on the mechanical properties of the heat
exchanger alloys, Alloy 617 specimens that were carburized at 900 (1000 h) and 1000  C (500 h) as well as
companion specimens that were oxidized at 900  C
(1000 h) were tested. Representative room temperature stress–strain curves for the materials exposed at
900  C are shown in Figure 24. It is clear from the
figure that the heavily carburized specimen (IN 617-2)

has higher flow strength compared to the oxidized
specimen (IN 617-7), but considerably reduced ductility. Room temperature tensile results for the material carburized at 1000  C are essentially identical to
those for material carburized at 900  C. Tensile testing the carburized Alloy 617 at 800  C showed
increased reduction in area to about 4%; the oxidized

273

(a)

(b)

50.00 µm

Figure 21 Optical micrographs of cross-sections through
(a) Alloy 617 and (b) Alloy 230 after 500 h at 1000  C under
decarburizing conditions.

material had greater than 50% reduction in area in
tension at 800  C. Tensile tests of Alloy 230 with
extensive carburization also indicated nil ductility at
room temperature and less than 1% tensile elongation at 800  C.

5.11.9 Aging Effects
In addition to decreased ductility from carburization,
aging will cause precipitation of carbides and other
phases in the temperature range of interest for heat
exchanger applications. The alloys with higher aluminum, such as Alloy 617, have a significant volume
fraction of Ni3Al (g0 ) formed at some aging temperatures. This phase increases the strength and results in
reduced ductility and impact properties. A T–T–T
diagram for Alloy 617 illustrating the region of



274

(a)

Material Performance in Helium-Cooled Systems

50.00 mm

50.00 µm

Figure 23 Optical micrograph of carbides in the center
of the Alloy 230 coupon after carburizing exposure at 1000  C.

(b)

50.00 mm

Figure 22 Optical micrographs of cross-sections
through (a) Alloy 617 and (b) Alloy 230 after 1000 h at
1000  C under carburizing conditions.

stability of the g0 is shown in Figure 8. The most
rapid precipitation of additional phases occurs at a
temperature of 750  C, which is somewhat below the
expected maximum operating temperature of the
IHX. The effect of thermal aging on impact properties
of Alloy 617 is shown in Figure 25 for an aging time of
1000 h. Absorption of energy from impact testing

drops considerably for these aging treatments compared to the solution-annealed material.
The room temperature tensile ductility of Alloy
800H has been found to be essentially unchanged by
aging in air at 800  C for 30 000 h. In contrast, Alloy X
showed a drop in room temperature tensile elongation from 45% to 10% for similar aging treatment.
While the ductility of Alloys 617 and X decreases
significantly after aging conditions, both retain substantial ductility. Addition of degradation in properties
from carburization in the VHTR atmosphere could, of
course, be a cause for further concern.

Stress (Mpa)

% strain vs stress (MPa)
900
850
800
750
700
650
600
550
500
450
400
350
300
250
200
150
100

50
0

IN617-7
IN617-2

0

5

10

15
% strain

20

25

30

Figure 24 Room temperature tensile stress–strain curves
for Alloy 617 that has been oxidized (IN617-7) and
carburized (IN617-2) at 900  C for 1000 h.

Microstructures of Alloys X and 800H after 500 h
exposure to oxidizing conditions illustrate the differences in precipitation between the two alloys upon
aging. The grain size of heat-treated Alloy X is 35 mm
and precipitation within the grains is heavy as shown
in Figure 26. Several carbide populations coexist: fine

dark carbides which form stringers at twin boundaries
and round grayish carbides are within the grains; large
round white carbides (Mo rich, M6C) are formed at
grain boundaries.
Alloy 800H microstructure shown in Figure 27
after oxidizing exposure at 750  C exhibits large
and regular grains and lesser amounts of quite


Material Performance in Helium-Cooled Systems

275

Impact toughness (J) at RT

350
617 unaged
617 1000 h aged

300
250
200
150
100
50
//

0
500


600

700
800
900
Aging temperature (°C)

1000

Figure 25 Room temperature impact energy for Alloy 617
after aging at various temperatures for 1000 h.

homogeneous precipitation compared to Alloy
X, except for some stringers of carbides along the
rolling direction.
The more pronounced precipitation in Alloy X
compared to Alloy 800H and the resulting decrease
in ductility for Alloy X would suggest that Alloy
800H might be the preferred alloy for use in this
temperature range. Note, however, that the oxidation
resistance of Alloy X is higher compared to Alloy
800X in terms of external as well as internal oxidation. For compact heat exchanger designs where thin
sections are a concern, the limited internal oxidation
might favor selection of Alloy X. Limited internal
oxidation is a particularly attractive attribute if it
can be shown that further reduction of ductility due
to carburization is not a significant probability.
Although aged materials retain ductility at
operating temperatures, embrittlement from aging
or environmental effects is an issue in some circumstances. Reduced ductility and impact properties of

any of the alloys are of concern during startup from
ambient temperature after an outage, for example,
because stresses from thermal gradients may be
large in this circumstance.

5.11.10 Summary
Although designs for very high-temperature nuclear
reactors are continuing, enough is known from system requirements and past experience to make reasonable material choices for the most highly
challenged systems, the heat exchanger and control
rod sleeves. Due to the high anticipated service temperature and the requirement for minimal change in

100 µm
Figure 26 Microstructure of Alloy X after 672 h at 750  C
under oxidizing conditions.

100 µm
Figure 27 Microstructure of Alloy 800H after 672 h
exposure at 750  C under oxidizing conditions.

material behavior for service lives up to 60 years, the
material choices are limited to face center cubic
solid-solution alloys. Compact heat exchanger designs
are especially demanding due to small section sizes.
The number of candidate alloys is further constrained by the necessity of selecting materials that
are sufficiently mature to be acceptable to regulators
for licensing the plant. The leading candidate alloy
for IHX application at outlet temperatures in excess
of 850  C is Alloy 617. Alloys 230, 800H, X, and XR
are also viable candidates for some applications.
Only Alloy 800H is currently in the US nuclear

design code and for temperatures limited to 760  C.
Additional material property characterization will be
required to allow application of Alloy 800H at higher
temperatures and to design with the other candidate