ARTICLE IN PRESS

International Journal of Pressure Vessels and Piping 85 (2008) 22–29

www.elsevier.com/locate/ijpvp

Acquisition of long-term creep data and knowledge

for new applications$
Koichi Yagi
National Institute for Materials Science, Tsukuba, Japan

Abstract

This paper introduces the activities being conducted by the National Institute for Materials Science (NIMS) to acquire long-term
creep-test data on heat-resistant steels and alloys, as well as some study results on the creep strength of heat-resistant steels and alloys,
obtained through the NIMS Creep Data Sheet project. The results emphasize the importance of data acquisition in conducting long-term
creep tests and the need for a comprehensive understanding of changes in metallurgical microstructure, creep deformation, damage
formation and growth, and creep rupture life, and the relationships among these factors. The importance of establishing international
cooperation in data and information sharing, in order to prepare for the globalization of society, is stressed.
r 2007 Elsevier Ltd. All rights reserved.

Keywords: Heat-resistant steels and alloys; Creep rupture strength; Microstructure; Creep deformation; Life prediction; Aged plants; High-Cr ferritic
steels; Allowable stress; NIMS Creep Data Sheet

1. Introduction
Approximately 80 years have passed since creep problems
were first taken into consideration in the design of high-
temperature plants because of rises in operating temperature.
Approximately 30–50 years have passed since the time when
heat-resistant steels and alloys were standardized in Japan
and other countries, and their creep-test data were obtained
and made publicly available. Substantial amounts of creep-
test data on heat-resistant steels and alloys have now been
accumulated globally. However, our present rate of acquisi-
tion of long-term creep-test data on these materials and the
amount of accumulated knowledge that we have are still
insufficient for improving our accuracy of evaluating the lives
of plants used for a long time. They are also insufficient for
ensuring the safe use of the newly developed high-tempera-
ture heat-resistant materials and thus the alleviation of global
environmental problems. Various organizations and institu-
tions in many countries have made strenuous efforts to obtain
$ view of the results obtained thus far from the Data Sheet
This article appeared in its original form in Creep & Fracture in High
Temperature Components: Design & Life Assessment Issues, 2005.
Lancaster, PA: DEStech Publications, Inc.
E-mail address: yagi.koichi@nims.go.jp
these long-term creep-test data and are still continuing their
efforts. However, because of the lack of sufficient funds and
manpower for inputting information into their data acquisi-
tion activities, and because of other various problems, such as
limitations in facility maintenance and insufficient resources,
the activities of these organizations have either declined or
been suspended. It is extremely important that we obtain
long-term creep-test data on heat-resistant materials. The
accumulation of knowledge on these materials is indispen-
sable for securing the safety and improving the reliability of
high-temperature plants. In particular, these long-term creep-
test data form the basis of the drive to secure the safety and
improve the reliability of high-temperature plants. Efforts to
obtain these data must be continued.
This paper describes past activities to obtain long-term
creep-test data and gives an introduction to the present
data acquisition activities, in particular, the present status
of the Creep Data Sheet project of the National Institute
for Materials Science (NIMS). The paper also emphasizes
the importance of acquiring long-term creep-test data in
project. I also offer my own opinion on the need for future
international cooperation in acquiring long-term creep-test
data on heat-resistant materials.

0308-0161/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.ijpvp.2007.06.001
 ARTICLE IN PRESS
K. Yagi / International Journal of Pressure Vessels and Piping 85 (2008) 22–29
2. Need to acquire long-term creep-test data
An understanding of the creep-strength characteristics of
structural materials is inevitable for securing the safety of
thermal power plants and petrochemical plants that are
operated under high temperatures. These thermal power
plants and petrochemical plants are used for long periods
of time—approximately 20–30 years. These plants are
generally designed on the basis of a creep rupture strength
of 100,000 or 200,000 h. However, as it is difficult to
acquire a large number of long-term creep-test data, the
allowable creep strength for plant design is obtained by
extrapolation from short-term creep-test data. The inten-
tion is to review the allowable creep strength when the
long-term creep-test data become available in the future, if
there are any problems. The extrapolation method is used
to obtain the allowable creep strength by accelerating the
rupture life of the material of concern by increasing the test
temperature. Although many extrapolation methods of
obtaining the allowable creep strength have been proposed,
there are as yet no perfect ones. Therefore, the rupture life
of the material concerned, as predicted by the extrapola-
tion method, must still be verified by long-term creep tests.
The number of aged thermal power plants and petro-
chemical plants, not only in Japan but also globally, has
recently increased. Thus, it is necessary to improve the
reliability of the method of predicting the remaining lives
of these plants. To do this, it is necessary to clarify the
changes in metallurgical microstructure during a long
period of creep progress, as well as the creep damage and
fracture mechanisms and their effects on material life.
The creep strength of a metal is a very sensitive
characteristic of metallurgical microstructure. The creep-
strength characteristics of a metal vary greatly, depending
upon slight changes in chemical composition of the metal
and the conditions of rolling or heat treatment. The extent
of the effects of these elements is not yet completely known,
and there are many unknown factors that may affect these
characteristics. For these reasons, the creep-strength
characteristics of engineering materials should be examined
on the basis of the test data on these materials.
3. Trends in data acquisition activities
3.1. The global situation
Most of the heat-resistant steels that are presently used
in high-temperature plants were developed before, or
during, World War II. These steel materials were inter-
nationally standardized after the war. The method used at
present to determine the creep-related allowable strength of
steels was decided upon in 1950 by the standards of the
American Society of Mechanical Engineers (ASME), a
body that strongly influenced the design of high-tempera-
ture plants.
As a result, long-term creep tests were conducted in
various countries and the test results were made public as
23
creep data sheets over a period from the 1960s to the 1970s.
The representative ones are the ASTM data series, the
British Steelmakers Creep Committe (BCCC) high-tem-
perature strength data series, and the long-term data series
of the Iron and Steel Institute of Germany [1]. Creep data
acquisition activities were most active during the 1970s.
After that period, systematic data acquisition activities
declined. Some organizations and institutions are indepen-
dently conducting creep tests to acquire the necessary data,
but those data are not widely publicly available.
In Japan, it became necessary among industries during
the 1960s to certify that domestically developed heat-
resistant steels and alloys had met international standards.
For this purpose, the facilities and equipment necessary for
conducting long-term creep tests of heat-resistant steels
and alloys were built at the National Research Institute for
Metals (NRIM). The NRIM Creep Data Sheets project
began in 1966. The project continues under the auspices of
the NIMS, the renamed NRIM. NIMS is one of the
world’s representative research institutes conducting large-
scale long-term creep tests of heat-resistant steels and
alloys.
3.2. New developments in Europe
In Europe, the European Creep Collaborative Commit-
tee (ECCC) was established in 1992 to promote research on
the creep of heat-resistant materials for the whole of
Europe. The purpose of the establishment of the ECCC
was to acquire, collate, and analyze creep data on the steels
and alloys to be used in high-temperature plants, with the
cooperation of various organizations and institutes. ECCC
presently has 14 member countries. The objectives of
ECCC are to acquire creep data, to set up standards for
creep strength, and to reflect the acquired data in the design
values of European Standards.
As mentioned above, it is necessary to input a large
amount of resources to conduct long-term creep tests.
Thus, the dispersal of creep tests over various locations, as
done by the ECCC, is a good idea for creep-test data
acquisition under international cooperation. However, if
we compare the creep-test data that are obtained by
various organizations, it is evident that there are deviations
in the data obtained by different organizations. The factors
influencing creep tests are load and temperature. Provided
that the traceability of load and temperature is observed,
there should not be deviations from organization to
organization. However, in reality, there are variations in
data from organization to organization. In other words,
when data obtained by several organizations are being used
for analyses, the uncertainty of the deviation between
organizations is included in the analyses.
Although there are some problems involved in such
cooperative data acquisition, this is a good method of
efficiently acquiring data under international cooperation.
It is very important that we learn from the method used by
the ECCC.
 ARTICLE IN PRESS
24 K. Yagi / International Journal of Pressure Vessels and Piping 85 (2008) 22–29
4. History of the NIMS Creep Data Sheet project
4.1. Data Sheet project
The Creep Data Sheet project of NIMS (formerly NRIM
when the project was started) was begun in 1966 to acquire
the data on 100,000 h strength necessary for the design of
high-temperature plants, not by the extrapolation method
but by conducting 100,000 h long-term creep tests. These
were made readily available to the public in order to
contribute to the appropriate use of high-temperature
metallic materials and thus secure the safety of high-
temperature plants, improve plant reliability, and develop
new steels and alloys for high-temperature use.
Test materials for data sheet were initially selected under
the conditions that they were (1) domestic materials, (2)
materials already sold on the market, (3) materials that
were not patented and were manufactured by more than
one company, (4) materials that had been previously
manufactured, and (5) standard materials generally used in
thermal power plants and petrochemical plants. When the
Creep Data Sheet project was initiated, material sampling
was examined by the Creep Committee established within
the Iron and Steel Institute of Japan. Since 1987, material
sampling has been examined by the Structural Material
Data Sheet Committee and the Creep Sub-committee,
which were set up in NIMS. The sub-committee consists of
researchers and engineers belonging to plant and material
companies.
All materials included in the NIMS Creep Data Sheet
project, shown in Table 1, are domestic materials. As these
test materials are standardized materials, the NIMS Creep
Data Sheets are internationally acknowledged as standard
reference data for internationally recognized materials.
4.2. Results obtained until now
Table 1 also lists the number of Creep Data Sheets that
were obtained by conducting creep tests and had been
published as at the end of March 2005. The NIMS Data
Sheets are prepared for each material type by sampling
several heats from one material. The total number of heats
sampled so far is 366. The first edition of the Creep Data
Sheets is published when creep rupture data up to
Table 1
Materials for which NIMS creep data sheets have been published
Type of materials Number of materials Number of heats Publications
Carbon steels 3 19
Ferritic creep-resistant steels 31 200
Austenitic creep-resistant steels 16 94
Superalloys 16 53
Long-term creep rupture data, No. 50
Total 66 366
approximately 10,000 h had been obtained by creep tests.
Edition ‘‘A’’ is published when creep rupture data from
30,000 to 50,000 h had been obtained. Edition ‘‘B,’’ which
would be the final edition, is published when creep rupture
data over 100,000 h had been obtained. For some types of
materials, Edition ‘‘A’’ may be the final edition of the creep
rupture data.
The Creep Data Sheet (No. 36B) for ‘‘Quenched and
Tempered 2.25Cr–1Mo Steel Plates,’’ which was published
in September 2003, is the final creep data sheet of those
originally planned for 44 materials. Including this data
sheet, a total of 34 volumes of Edition ‘‘B’’ have been
published. When Edition ‘‘B’’ was published, there were
some creep-test specimens that had not yet ruptured. Thus,
creep tests on these materials continued and, as a result,
146 test specimens ruptured after the issuance of Edition
‘‘B.’’ Creep Data Sheet No. 50 was published in March
2004 as a special edition to include the data of these
specimens. The first stage of the NIMS Creep Data Sheet
project which was begun in 1966 was terminated by the
issue of Creep Data Sheet No. 50.
4.3. Measures to account for changing situations
Japan’s technologies have shifted in the energy-saving
direction because of the occurrence of two oil-shortage
shocks in the 1970s. In the field of high-temperature plants,
the increase in operating temperatures and vessel pressures
progressed, and the development of high-strength heat-
resistant steels was sought. In particular, as the modified
9Cr–1Mo steel developed in the United States had superior
characteristics, the application of this steel in high-
temperature plants and the development of high-Cr ferrite
steels were eagerly conducted. In the 1990s, global
environmental problems became a big issue; further
increases in the operating temperatures of high-tempera-
ture plants were taking place, and there was a demand for
the development of various types of heat-resistant steels
and alloys.
In view of these situations, the NIMS Creep Data Sheet
project included newly developed materials in order to
meet social demands. Table 2 lists those newly included
materials whose creep tests have been conducted since
1988. Some of these materials are quite new, and their use
Total
First edition Second edition Final edition Special edition
3 3 2 8
21 18 17 56
15 13 9 37
10 9 6 25
1 1
49 43 34 127
 ARTICLE IN PRESS
K. Yagi / International Journal of Pressure Vessels and Piping 85 (2008) 22–29 25

Table 2 5. Major results of studies conducted as part of the NIMS

List of materials subjected to creep testing after 1988 Creep Data Sheets project
Materials Product form Sampling
year 5.1. Complicated creep behavior of engineering materials

2.25Cr–1.6W–V–Nb steel Tube, pipe 2002 When a load is applied to a specimen of metal under
2.25Cr–1Mo–0.25V steel Forging 2003
high-temperature conditions, deformation related to a time
9Cr–1Mo–V–Nb steel Tube, pipe, plate 1988–2002
9Cr–2Mo steel Tube, plate 1988, 1990 factor (called creep deformation) occurs. Creep deforma-
9Cr–0.5Mo–1.8W–V–Nb steel Tube, pipe 1995, 1996 tion progresses from the transition creep range, in which
11Cr–0.4Mo–2W–Cu–V–Nb steel Tube, pipe 2000, 2001 the creep deformation speed decreases after loading, to the
18Cr–8Ni–Cu–Nb–N steel Tube 2002 acceleration creep range, after going through a constant
18Cr–12Ni–Mo–low C–Middle N Plate 1990, 1995
steel
creep range; then the specimen of metal finally ruptures.
18Cr–12Ni–Nb–Ti steel Tube 2002 However, the creep deformation behavior of engineering
25Cr–20Ni–Nb–N steel Tube 2002 materials is not as simple as in the textbooks. Fig. 1 shows
IN 738–LC Casting 1997 the relationship between creep rate and time of austenite
Hastelloy X Bar, plate 2003 stainless steel SUS 316. Under high-stress conditions, this
Inconel 718 Bar, plate 2000, 2001
material exhibits the creep deformation behavior described
in the textbooks. However, under low-stress and such long-
term test conditions, a minimum value of creep rate
periods are still relatively short. Thus, their chemical appears in addition to the minimum creep rate value. When
compositions and heat treatment conditions may change engineering materials are to be used, it is necessary to
within the standard values set. For this reason, the heat clarify the relationships between the intricate factors that
sampling timings for especially important materials among affect creep rupture life, such as hardening and recovery,
these newly included materials may be shifted. As no more metallurgical microstructure, and damage and cracking, to
data sheets may be issued in future under the set rules, the use these materials most effectively. Section 5.2 describes
question of how to publish and provide supplemental creep some of the results of studies conducted through the NIMS
data in the form of creep data books (not sheets for each Creep Data Sheet project and emphasizes the importance
individual material) is presently being investigated in the of acquiring long-term creep-test data.
NIMS [2].
5.2. Need to understand creep deformation characteristics to
4.4. Data publication through the internet predict creep rupture life

The NIMS Creep Data Sheets are printed in book form The relationship between the stress and rupture time of
and distributed to industries, universities, and libraries. 1Cr–0.5Mo steel and 2.25Cr–1Mo steel shows a reverse
The number of different data sheets distributed is 470 sigmoidal curve. The creep deformation curves of these
within Japan and 220 overseas. As one of the NIMS materials have two minimum values in relation to creep
Materials Databases, the Structural Materials Database rate and time. Thus, the creep curve was divided into two
has been published on an Internet web page since April parts (the first and second halves) for analysis, and those
2003. The Structural Materials Database consists of parameters that control creep deformation behavior were
published NIMS Creep Data Sheets, Fatigue Data Sheets,
Corrosion Data Sheets, and Space Use Materials Strength
Data Sheets. These data sheets are exhibited in PDF form.
Fact databases on creep and fatigue are also exhibited so
that users can select the desired data and prepare figures.
The total number of users registered to subscribe to the
NIMS Material Database was 17,800 as at the end of
March 2005: 1990 were subscribed to the Structural
Materials Database; of these, 1680 were domestic users
and 310 overseas subscribers. Approximately 80–90% of
the subscribers to the Structural Materials Database were
researchers and engineers employed by private enterprise.
The number of subscribers who register through the
Internet and the number of Internet log-ons are increasing.
There are inquiries about the NIMS Material Database,
not only from users employed by large enterprises but also
from small enterprises that are not registered yet to
subscribe to the data sheets. Fig. 1. Creep deformation behavior of SUS 316 steel.

26 K. Yagi / International Journal of Pressure Vessels and Piping 85 (2008) 22–29
500
1Cr-0.5Mo steel
300
(JIS STBA22)
Stress (MPa)
100
80
60
40
Type I
} Calculated
Type II 923K
20
100 101 102 103 104
Time to rupture ( h )
Fig. 2. Comparison of experimental creep rupture data with predicted
creep rupture life of 1Cr–0.5Mo steel. Creep rupture data signified by the
solid symbol were used for the analysis.
determined. When the creep rupture lives of these materials
were analyzed using these parameters, their high-stress
short-term rupture lives could be predicted from the first
half of the creep deformation curve. Their low-stress long-
term rupture lives could be predicted from the second half
(see Fig. 2) [3]. From these analytical results, it was found
that differences in the factors that dominate the creep
deformation behavior affect the creep rupture life. The
reason why the deformation behavior changes in the
middle of the creep curve is that the pinning effect of
dislocation is reduced by enlargement of the fine pre-
cipitated carbides [4]. As described above, it is necessary to
understand the relationship between creep rupture life and
creep deformation in order to estimate creep rupture life. In
addition, it is important to have a good understanding of
the relationship between metallurgical microstructure and
creep deformation behavior.
5.3. Variation in the creep rupture life of a steel can be
reduced by knowing which chemical compositions are present
in the material in small amounts
The creep rupture strength of a steel is affected by small
amounts of chemical compositions and by the heat
treatment conditions. Even if heat-resistant steels contain
a specified amount of chemical compositions, their strength
varies from heat to heat. In other words, it is understood
that creep-strength characteristics have a deviation. Fig. 3
shows the results of creep rupture tests of SUS 304 steel
(18Cr–8Ni steel). When the creep rupture tests were
conducted under the same temperature and stress condi-
tions, the difference between the shortest creep rupture life
and the longest creep rupture life was quite large—more
than 10 times. However, there are practically no variations
in the relationship between stress and time to rupture of
test specimens obtained from the same heat. It can be seen
that the variation in creep rupture life of the same heat is
ARTICLE IN PRESS
500
Amount of Mo
18Cr–8Ni (tube) Rich Poor
300 600°C
650°C
700°C
750°C
Stress (MPa)
773K
100
Middle N+
823K 80
high Nb+Ta
60
40
873K
20
105 106 101 102 103 104 105 106
Time to rupture (h)
Fig. 3. Creep rupture strength of 18Cr–8Ni steel (SUS 304).
small under the same test conditions. From this fact, if
factors affecting the variation in creep rupture life for
different heats could be found, the creep rupture life of the
steel could be estimated with greater accuracy. If we focus
on the chemical compositions of the various heats shown in
the figure, we can see that there are some heats that contain
small amounts of Mo and some that contain larger
amounts of Mo. The heats that contain large amounts of
Mo have relatively high strength. However, even among
those with small Mo there is one with high strength: its
Nb+Ta content is relatively high. For creep tests under
low temperatures and over short periods of time, the main
factor affecting the variation in strength of heats is the fine
precipitation of Nb carbide. This effect diminishes when
the temperature rises and the Nb carbide becomes larger.
Under these conditions, the creep rupture life of a heat
containing large amounts of Mo becomes longer. Under a
higher temperature and over a longer test period, the
amount of effective nitrogen becomes a more important
factor affecting variations in the rupture lives of heats [5].
From the above fact, if the factors controlling the creep
rupture strength of heat-resistant steels can be found, the
creep rupture life will have no heat-to-heat variation, and
the creep rupture life can be estimated as more accurate
values. This knowledge can then be utilized to estimate the
remaining life of the structural components in a high-
temperature plant once the chemical compositions of the
materials used in the plant are determined.
5.4. Need for long-term tests to determine the creep rupture
properties of high-Cr steels
As a result of the accumulation of long-term creep-test
data on high-Cr steels, we find that the creep rupture
strength abruptly decreases after 10,000 h of test at 6001C.
Fig. 4 shows the results of tests on a Modified 9Cr–1Mo
steel. The broken line in the figure shows the rupture time,
as predicted by the Larson–Miller parameter (LMP)
method using all data included in the figure. The creep
 ARTICLE IN PRESS
K. Yagi / International Journal of Pressure Vessels and Piping 85 (2008) 22–29 27

500 500
400 0.2% proof stress Mod.9Cr–1Mo steel
Mod.9Cr–1Mo steel 2
300
300
500°C
550°C
200
Stress / MPa

100 550°C

Stress / MPa
600°C
100
90 80
80
70 650°C
60
50 600°C
60
40 All data
30
Predicted 725°C 700°C 40 Proposed 725°C
20 method
700°C 650°C
101 102 103 104 105
20
Time to rupture / h 101 102 103 104 105 106
Fig. 4. Stress vs. time to rupture in Mod. 9Cr–1Mo steel, and creep Time to rupture / h
rupture life as predicted by the Larson–Miller parameter method using all
Fig. 5. Result of creep rupture life prediction for Mod. 9Cr–1Mo steel by
data.
the region-partitioning method. Creep rupture data indicated by large
circles were used for life prediction.
rupture life of the steel when tested more than 10,000 h at
temperatures of 600 and 6501C has not been predicted. The
metallurgical microstructure of the test specimens was
examined after the creep test. The examination result
showed that the metallurgical microstructure of those test
specimens that were tested under high stress for a short
period of time had homogeneously recovered. However, in
the specimens that were tested under low stress for a long
period of time, only the metallurgical microstructure near
the prior austenitic grain boundary had recovered,
indicating heterogeneous recovery. We can assume that
this heterogeneously recovered metallurgical microstruc-
ture could be the cause of the abrupt decrease in creep
rupture strength after a long period of test.
From the above phenomena, we can suggest that the
prediction of the creep rupture life of this type of steels
should be conducted by using only those data located
Fig. 6. Creep rupture strengths of 10 types of ferritic heat-resistant steels.
within regions showing the same metallurgical microstruc-
ture, and that the boundary of these regions should be one-
half of 0.2% proof stress [6]. Fig. 5 shows creep rupture life results of creep rupture tests on 10 types of ferrite heat-
as predicted by this method proposed by Kimura. The resisting steels; these results were clarified by using LMP
predicted creep rupture life agrees well with the test results. values. Although the variation in LMP values is large on
As this method is extremely simple and easy to use, it has the high stress side, the values converge to a certain level on
been adopted to review the allowable stress of high-Cr the low stress side. When the convergence level was
ferrite steels for use in thermal power plants in Japan. It is examined, it was found that this strength level agreed with
necessary to further examine this method by acquiring the strength level of 0.03% Mo carbon steel. We can
long-term creep-test data. assume that when ferrite heat-resistant steels are exposed to
high temperatures for a long period of time, their
5.5. Ultra-long-period creep rupture strength metallurgical microstructure changes and their strength
reaches a certain level, regardless of the proportions of
It is important to understand metallurgical microstruc- their original metal elements. This strength level is the
ture in order to evaluate the long-term creep strength inherent creep strength [7].
properties of practical-use heat-resistant steels. We can Fig. 7 schematically shows the conditions under which
assume that metallurgical microstructure change will the stress of the steel reaches the inherent creep strength
finally cease and will reach an equilibrium condition. The after the steel has been tested for a long period of time. By
question arises as to what the creep strength will be when using the concept of inherent creep strength, it is possible
the metallurgical microstructure becomes stable and there to subtract the lower limit value of the stress level and to
is no more change in microstructure. Fig. 6 shows the predict what value the life of the steel reaches. However,
 ARTICLE IN PRESS
28 K. Yagi / International Journal of Pressure Vessels and Piping 85 (2008) 22–29
Creep rupture strength property
of ferriticcreep resistant steels
Maximum level of
Inherent Creep Strength
for ferriticsteel
Stress, log σ
Decrease in creep
strength due to
Solid solution
microstructural
change
strengthening
Creep strength
of pure Iron Common Inherent
(α –Fe) Creep Strength
Time to rupture, log tR
Fig. 7. Schematic representation of the factors governing the inherent
creep strength and long-term creep strength properties of ferritic heat-
resistant steels.
most practical-use heat-resistant steels are not used at the
inherent creep strength but in the transitional condition,
wherein their creep strength decreases toward the inherent
creep strength. For this reason, it is important to determine
the time at which the creep strength starts to decline and to
improve the accuracy of estimation of the rate of decrease
in strength, in view of the prediction of the rupture life.
As mentioned above, to understand and evaluate the
long-term creep strength of practical-use heat-resistant
steels, it is important to accumulate knowledge about
changes in metallurgical microstructure and creep defor-
mation on the basis of long-term creep-test data, and also
on the formation and growth of creep damage (although
this information is not covered by this paper). The source
of accumulation of this knowledge is the acquisition of
long-term creep-test data. For this reason, data acquisition
through long-term creep tests can be seen as an activity
that becomes the foundation for securing the safety and
improving the reliability of high-temperature plants.
6. Future plan for the NIMS Creep Data Sheet
6.1. Contribution to remaining life evaluation of aged plants
As the number of high-temperature plants that have
been used for a long time is increasing, there is increasing
demand to secure the safety of these old plants and operate
them appropriately. Not only creep-strength data but also
information on metallurgical microstructure and creep
damage is needed to evaluate deterioration with age to steel
materials used long term, as well as to predict the
remaining lives of those plants. NIMS systematically
observes the metallurgical microstructure of creep-tested
materials, clarifies material information related to metal-
lurgical microstructure change, and publishes the informa-
tion as Metallurgical Atlas, one of the NIMS Creep Data
Sheets.
Table 3 lists the Metallographic Atlas that had been
published as of March 2005 and one that will be published
during 2006. These albums show the metallurgical micro-
structures of heat-resistant steels crept short (300 h) to long
term (100,000 h) under different temperatures, as observed
by optical microscopy. They also include photographs
taken under a transmission electron microscope, time–-
temperature–precipitation diagram, and creep fracture
mechanism map. They may be used to predict the
remaining lives of structural components in high-tempera-
ture plants. In addition to these albums, NIMS plans to
publish creep-deformation data sheets and welded-joint
creep-data sheets. Our aim is to help build a safe society by
improving the accuracy of remaining life evaluation.
6.2. Reflection in standards
NIMS is conducting its Creep Data Sheet activities to
acquire long-term creep-test data, but is not directly
involved in the use of these data sheets. However, after
the accidental failure of the reheater pipe of a thermal
power plant in June 2004 in Japan, it was decided to
reexamine the allowable stresses on heat-resistant steels,
and industries requested NIMS to be the center of the
reexamination activities. I consider that it is important for
NIMS to experience this kind of role. I further consider
that it would be important to open as many as possible of
the data possessed by NIMS to the public, so as to share
them with various organizations and institutes, publicly
discuss about them while NIMS takes on its central role,
and propose methods of their use.
In Japan, there are no organizations similar to the
ECCC. I think it is necessary for NIMS to learn about the
activities of the ECCC, to accumulate creep-test data under
the cooperation of private enterprises, to propose more
accurate methods of evaluating long-term strength, to
reflect in the decision of allowable stress, and to perform its
duties as the central organization for these activities.
7. Building international cooperation
A creep data book for heat-resistant steels and heat-
resistant alloys was published by Springer-Verlag with the
cooperation of 25 researchers and engineers in Japan and
Europe in June 2004 [8]. This book is the result of efforts
made by the ECCC, the German Creep Committee, and
researchers and engineers in Japan. Past international
cooperation has been limited to the exchange of published
materials, participation in committees convened in other
countries, and the joint holding of international confer-
ences. The publication of this book was achieved through
various detailed, step-by-step discussions from the plan-
ning stage. Although it took more than 5 years from the
planning stage to completion, it is believed that this project
was the starting point for precious interchange between
creep experts in Europe and Japan. It is expected that a
cooperative relationship in the accumulation of data and
knowledge will be developed by this kind of cooperative
 ARTICLE IN PRESS
K. Yagi / International Journal of Pressure Vessels and Piping 85 (2008) 22–29
Table 3
Publication of NIMS Creep Data Sheets––Metallographic Atlas of long-term creep materials
No. Materials
M-1 18Cr–8Ni steel
M-2 18Cr–12Ni–Mo steel
M-3 18Cr–10Ni–Ti steel
M-4 2.25Cr–1Mo steels
(M-5) 18Cr–12Ni–Nb steel
As of March 31, 2005.
work through increased understanding of the background
of each participant.
8. Acquisition of creep-test data in a global society
It is considered that industrial materials of practical use
will be used not only domestically but also internationally.
Under this situation, it is important that we internationally
share data and information related to these materials. On
the other hand, large amounts of money and human
resources are required to operate and maintain the test
equipment and facilities needed to acquire creep-test data.
It is impossible for a single organization to acquire such
large amounts of money and human resources. For this
reason, the international division of data acquisition work
and cost sharing is necessary. As each organization or
group has its own history and limitations, it is not an easy
task to share data and information beyond each individual
organization’s walls. The procedures adopted by the ECCC
are a good suggestion for how to deal with these problems.
As acquired data and information are precious assets for
each organization or group, it would be difficult to
completely open them to the public. However, it would
be possible to disclose information about what kind of
analytical methods were used and how the evaluation was
done, and then to open the final evaluation results to the
public. Fortunately, there are not many differences in
the basic chemical compositions of heat-resistant steels
throughout the world. It would be possible to eventually
solve these problems and find a future direction through
discussions on the use of these evaluation methods and
through the use of the final results as references. Coopera-
tion can occur if these kinds of procedures are followed. It
would be necessary to conduct a mutual exchange of
personnel and information that could be released for this
purpose. This kind of cooperation has already been carried
out. We need to start acting to build a concrete foundation
for cooperation by keeping in mind the future sharing of
data and information.
9. Conclusion
An understanding of the creep-strength properties of
used structural materials is necessary to secure the safety of
29
Designation Date of publication
JIS SUS 304HTB March 31, 1999
JIS SUS 316HTB March 31, 2003
JIS SUS 321HTB JIS STBA24 March 31, 2004
JIS SCMV4NT ASTM A542/A542M March 31, 2005
JIS SUS 347HTB (March 31, 2006)
high-temperature plants. Acquisition of creep-test data is
important as the foundation of an understanding of this
issue. This paper has described the global status of creep-
test data acquisition and the present status of the Creep
Data Sheets project being carried out by NIMS—the
world’s largest-scale data acquisition activity—and has
introduced the research results obtained through these
activities. The importance of the acquisition of long-term
creep-test data and the accumulation of comprehensive
knowledge on metallurgical microstructure change, creep
deformation behavior, creep damage formation and
growth, and creep fracture is also described. A future
method of international cooperation is proposed. I also
mention that using the procedures followed by ECCC
would be a good reference model. Finally, I would like to
stress again that the acquisition of long-term creep-test
data is most important.
References
[1] Yagi K, Merckling G, Kern TU. Creep properties of heat resistant
steels and superalloys. In: Yagi K, Merckling G, Kern TU, Irie H,
Warlimont H, editors. Landolt-Boernstein: Numerical data and
functional relationships in science and technology-New Series, VIII/
2B. Berlin: Springer; 2004. p. 2–3.
[2] Kimura K, Tabuchi M, Yamazaki M. NIMS creep data sheet
project—38 years of history and future prospects. Bull Iron Steel Inst
Japan 2004;9(11):816–20.
[3] Kushima H, Kimura K, Yagi K, Tanaka C, Maruyama K.
Characterization of creep deformation behaviour of Cr-Mo steel. In:
Proceedings of Proceedings of JIMIS-7, 1993. p. 609–16.
[4] Kushima H, Kimura K, Abe F, Yagi K, Irie H, Maruyama K. Effect
of microstructural change on creep deformation behaviour and long-
term creep strength of 1Cr-0.5Mo steel. Tetsu-to-Hagane 2000;86(2):
131–7.
[5] Miyazaki H, Tanaka H, Murata M, Abe F. Effect of minor elements
on the long-term creep rupture time of SUS 304 HTB stainless steel.
J Japan Inst Met 2002;66(12):1278–86.
[6] Kimura K, Kushima H, Abe F. Improvement of creep life prediction
of high Cr ferritic creep resistant steels by region partitioning method
of stress vs. Time to rupture diagram. J Soc Mater Sci Japan
2003;52(1):57–62.
[7] Kimura K, Kushima H, Yagi K, Tanaka C. Fundamental properties
of long-term creep strength for ferritic heat resistant steels. Tetsu-to-
Hagane 1991;77(5):667–74.
[8] Yagi K, Merckling G, Kern TU, Irie H, Warlimont H, editors.
Landolt-Boernstein: Numerical data and functional relationships in
science and technology-New Series, VIII/2B, Berlin: Springer; 2004.