Materials Transactions, Vol. 44, No. 6 (2003) pp.

1204 to 1208
#2003 The Japan Institute of Metals
EXPRESS REGULAR ARTICLE

Nondestructive Evaluation of Thermally Degraded 2.25Cr1Mo

Steel by Electrical Resistivity Measurement
Jai Won Byeon and Sook In Kwun*
Research Institute of Engineering and Technology, Korea University, Seoul, 136-701, Korea
An attempt was made to evaluate nondestructively the degradation of thermally aged 2.25Cr1Mo steel by electrical resistivity
measurement. Articial aging was performed to simulate the microstructural changes in 2.25Cr1Mo steel arising from long time exposure at
540
C. Microstructural parameter (amount of dissolved Mo in the matrix), mechanical property (ductile-brittle transition temperature) and
electrical resistivity were measured to investigate the relationship among these parameters. Both the amount of dissolved Mo and the electrical
resistivity decreased rapidly in the initial 1000 hours of aging and then changed little thereafter. On the other hand, the ductile-brittle transition
temperature (DBTT) increased rapidly in the initial stage of aging and then saturated afterward. Electrical resistivity was found to have linear
correlation with the amount of dissolved Mo and DBTT, respectively. Electrical resistivity was suggested as a potential nondestructive
evaluation parameter for assessing DBTT of the thermally degraded 2.25Cr1Mo steel.
(Received March 11, 2003; Accepted May 6, 2003)
Keywords: nondestructive evaluation, electrical resistivity, thermal degradation, 2.25chromium1molybdenum steel, ductile-brittle transition
temperature

1.

Introduction

The 2.25Cr1Mo steel has been widely used for elevated

temperature structure applications such as turbine rotors,
boilers and pressure vessels in fossil plant and petroleum
chemical facilities due to its superiority in high temperature
strength, oxidation resistance and hydrogen embrittlement
resistance at 300
C540
C. However, this steel is known to
be subject to aging degradation through carbide induced
brittleness brought on by precipitation of stable carbides,
temper embrittlement brought on by grain boundary segregation and softening of matrix after long exposure to high
temperatures.17)
In order to evaluate the microstructural degradation of Cr
Mo type steel arising from long time exposure to high
temperature, composition35) or morphology6) analysis of
carbides has been performed to identify the dierent kind of
carbides. Comparing the results of diraction pattern analysis
of several carbides with their composition, it was found that
carbides had distinct composition depending on the type of
carbide such as Cr rich M23 C6 , Mo rich M6 C, Mo2 C (M2 C),
Fe3 C (M3 C).35) Gope et al.6) and Watanabe and Shoji7)
reported an increase in M6 C carbide, which had been
previously reported by Baker and Nutting8) to be a stable
carbide, in thermally degraded 2.25Cr1Mo steel. The
increase of Mo rich M6 C in CrMo type of low alloy steel
is reported to cause irreversible carbide-induced embrittlement.1,2,7,9) In order to simulate the changes of carbides in
2.25Cr1Mo steel after prolonged on-site usage, Adbel-Latif
et al.3) carried out accelerated articial aging heat treatment.
Since it is dicult or practically impossible to obtain
specimens for mechanical testing without damaging the
structure in operation, many researchers have attempted to
evaluate the damage of structures nondestructively, for insitu monitoring.1013) Electrical resistivity,10,11) magnetic,12)
*Division

of Materials Science and Engineering, Korea University, Seoul,

136-701, Korea.

and ultrasonic13) methods are currently used for nondestructive evaluation of structures. Electrical resistivity method,
which was widely used for evaluating material properties like
precipitation behavior,1418) has recently been applied in the
eld of non-destructive evaluation of structure.10,11) Maeda
et al.11) reported a decrease in electrical resistivity in
thermally aged stainless steel. Although some researchers10,11) have reported changes of electrical resistivity in the
degraded structure, the correlation between electrical resistivity and microstructural changes in the thermally degraded
materials is lacking.
An attempt was made to evaluate nondestructively the
thermally degraded 2.25Cr1Mo steel by electrical resistivity
measurement in this investigation. Especially, the correlations among microstructural changes (depletion of solute
Mo), mechanical property (DBTT) and electrical resistivity
were investigated.
2.

Experimental

2.1 Material and heat treatment

A laboratory-made 15 mm thick hot rolled 2.25Cr1Mo
steel (ASTM A387-22-CL2) plate specimen tempered at
720
C for 1 hour after normalizing at 900
C was used, and its
chemical composition is given in Table 1. In order to
simulate the typical microstructure of the rotor steel in
service at 540
C for about 20 years, an accelerated isothermal
heat treatment3) was performed at 630
C for up to 4800
hours.

Table 1 Chemical composition of the 2.25Cr1Mo steel (ASTM A387-22CL2).

Element Fe

Cr

Mo

Si

Ni

Mn

Al

mass% bal. 0.138 2.27 0.97 0.142 0.17 0.46 0.007 0.014 0.004

Nondestructive Evaluation of Thermally Degraded 2.25Cr1Mo Steel by Electrical Resistivity Measurement

1205

2.2

Measurement of the amount of dissolved Mo in the

matrix and X-ray diraction analysis for electrolytically extracted carbides
In order to analyze the amount of solute element in the
matrix, electrolytic extraction technique was employed to
dissolve only the matrix, and then the solution was ICP
(inductively coupled plasma) analyzed. For electrolytic
extraction,4) a specimen in a solution of 90% methanol and
10% hydrochloric acid was subjected to a voltage of 3 volts.
After selectively dissolving the matrix, the remaining residue
(carbide) was separated for X-ray diraction analysis.
2.3 Measurement of DBTT
The V-notch Charpy impact test was carried out within the
temperature range of 140
C and 20
C using ASTM
standard specimens. The DBTT was obtained by tting
under the assumption that the change of absorbed energy with
temperature follows the eq. (1).19)


P1 P2 P1  P2
T  Ta

tanh
Y
1
2
2
Tb
where, Y, P1 , P2 , Ta , Tb , T are absorbed energy (J/m2 ), upper
shelf energy, lower shelf energy, DBTT (
C), constant, and
test temperature (
C), respectively.

where,
, V, I, A, L are electrical resistivity, measured

voltage, applied current, cross sectional area of the specimen,
and interval between voltage measurement leads, respectively. The electrical resistivity in this experiment was a
value averaged over ve measurements, and the error range
of the measurement was within 5n
m. These errors in
measurement seemed to have been generally caused by the
size of specimen and the change of temperature of contact
points.
3.

Results and Discussion

3.1 Microstructure and mechanical property

Figure 1 is FESEM (eld emission scanning electron
microscope) micrographs showing the morphologies of
carbides before and after aging. Compared to the as-received
specimen, the carbides in the aged specimen became
coarsened in size and reduced in number. The grain size of

Fig. 1 FESEM micrographs showing the morphology of carbides before

and after aging: (a) as-received, (b) 4200 h.

the as-received specimen was approximately 20 mm, and the

grains did not grow during aging.21) No grain boundary
segregation was detected in the articially aged specimen.21)
Figure 2 shows the change of the amount of dissolved Mo
in the matrix with aging time. As aging time increased the
amount of dissolved Mo in the matrix decreased rapidly for
about 1000 hours of aging and then changed little thereafter.
The total amount of Mo in the as-received specimen was
about 0.97% (Table 1). It is estimated that the 0.65% of Mo
(67% of the total amount of Mo) was dissolved in the matrix

0.70

Amount of Dissolved Mo(%)

2.4 Measurement of electrical resistivity

Electrical resistivity was measured using the direct current
four terminal potential method.10,11,20) Measurement was
carried out at room temperature consistently xed at 20
C. A
computer-controlled direct current source was used to supply
current 1 A to the specimen and the voltage was measured
to 1nV using nano-voltmeter. As the interval between voltage
output leads can lead an error in electrical resistivity
values,10) a xture with constant terminal interval was used.
A sheet-type specimen of 60 mm in length, 10 mm in width,
and 2 mm in thickness was used to measure electrical
resistivity. The thickness of the specimen was parallelabraded to within 2 mm  2 mm. Electrical resistivity was
calculated using eq. (2).
VA
2


I L

0.65

Our result
Wada & Biss(ref. 4)
2.25Cr-1Mo steel
500oC - 570oC, 100,000 hr

0.60
0.55
0.50
0.45
0.40
0.35
0.30
0.25
0.20
0

1000

2000

3000

4000

5000

Aging Time, t / h
Fig. 2 Change of the amount of dissolved Mo in the matrix with aging
time.

J. W. Byeon and S. I. Kwun

300
250
200
150
100
50
0
-140 -120 -100 -80 -60 -40 -20

Temperature, T / oC

-30

275

-32

270

-34

DBTT, T / oC

(b)

40

aged specimen was lower than that of the as-received

specimen at all temperatures and the impact-absorbed energy
curve in the aged specimen was shifted to higher temperature.
Figure 5 shows the change of DBTT and electrical
resistivity with aging time. The DBTT increased sharply
for initial about 1000 hours, then it changed little thereafter.
This change of DBTT is consistent with the results obtained
from the thermally degraded turbine rotor steel by Yu et al.10)
and Viswanathan and Gehl.22) In our previous research21) the
fracture surface of the specimen that had been impactfractured in liquid nitrogen after being aged for 4800 hours
showed complete intragranular cleavage fracture surface
indicating that the grain boundary had not been weakened
due to temper embrittlement caused by grain boundary
segregation. Consequently, the increase of DBTT in this
investigation seems to be mainly due to carbide-induced
embrittlement. The irreversible carbide-induced embrittlement in 2.25Cr1Mo steel has been reported to be related to
formation2,9) or coarsening7) of Mo rich M6 C carbides. On
the other hand, the irreversible embrittlement in the aged
2.25Cr1Mo1V steel has been reported to be related to
coarse M6 C carbide.1)

(d)

(c)

20

Fig. 4 Temperature dependence of the impact-absorbed energy of the

2.25Cr1Mo steel after various aging times.

M 23 C 6
M 6C
M 2C
M 3C

Intensity(arbitrary unit)

as-received
340 hr
730 hr
1500 hr
4800 hr

265

-36
260
-38

DBTT
Electrical resistivity

-40

255
250

-42
245

-44

(a)
240

-46
35

40

45

50

55

2
Fig. 3 X-ray diraction patterns for the electrolytically extracted carbides
after various aging times: (a) as-received, (b) 290 h, (c) 730 h, (d) 3700 h.

-48
0

1000

2000

3000

4000

Electrical Resistivity, / nm

while the remaining 0.32% of Mo was combined to the

carbides. As aging time increased the amount of Mo
dissolved in the matrix decreased to about 0.3% (46% of
the total amount of Mo), the remaining 0.67% is thought to be
combined to the carbides. The result of Wada and Biss,4) who
reported depletion of Mo in 2.25Cr1Mo steel used on site
for about 100000 hours at 500 to 570
C, was also plotted in
Fig. 2 for comparison. The fact that the 0.28% of Mo was
dissolved in the matrix reported by Wada and Biss4)
conrmed the similarity to present results.
Figure 3 shows the results of X-ray diraction analysis of
electrolytically extracted carbides. The peaks of M23 C6 ,
M3 C, M2 C, M6 C carbides are observed along with other
unidentiable peaks in the as-received specimen. There were
a large amount of M23 C6 and a small amount of M2 C and
M6 C carbides. The amount of M2 C changed little as aging
time increased. This contrasts with reports of extinction of
M2 C with aging time from electron diraction pattern
analysis by Gope et al.6) and with reports of increase in
M2 C from X-ray diraction analysis of extracted carbide by
Watanabe and Shoji.7) Although M3 C carbide existed in large
quantities in the as-received specimen, it rapidly became
extinct in the initial stage of the aging. Results of X-ray
diraction analysis of Watanabe and Shoji7) did not report
the existence of M3 C (main peak is located very close to the
main peak of M23 C6 ) in the undamaged 2.25Cr1Mo steel.
Gope et al.6) have reported that although M3 C was present in
large quantities in the undamaged material, it was undetectable in the aged material. According to the results of X-ray
diraction analysis of this experiment, M6 C existed even in
the as-received specimen albeit in small quantities, and M6 C
carbide rapidly increased in the aged specimen. Gope et al.6)
and Watanabe and Shoji7) have also reported an increase in
M6 C carbide with aging time. For Mo rich M6 C carbide to
increase, a supply of Mo was needed. From the results of
analysis of amount of solute Mo, it was inferred that Mo,
which had become depleted from the matrix, had contributed
to the increase in M6 C.
Figure 4 shows the change of Charpy impact-absorbed
energy with testing temperature. The absorbed energy of the

Absorbed Energy, E / 104J m-2

1206

235
5000

Aging Time, t / h
Fig. 5

Change of DBTT and electrical resistivity with aging time.

Nondestructive Evaluation of Thermally Degraded 2.25Cr1Mo Steel by Electrical Resistivity Measurement

3.2

223 69:8  %Mo

where,
and %Mo represent electrical resistivity (n
m) and

the amount (mass%) of Mo dissolved in the matrix,
respectively. The main microstructural changes in the
2.25Cr1Mo steel during aging are the coarsening of carbides
(Fig. 1), the depletion of Mo dissolved in the matrix (Fig. 2),
and the resultant increase in the amount of Mo rich M6 C
carbides. Conduction electron scattering resulting from the
particle (carbides) is only eective when the particle is
signicantly small.25) For example, the critical particle size
for conduction electron scattering in Al alloys with spherical

Electrical Resistivity, / nm

275
270
265

= 223 + 69.8 (%Mo)

: electrical resistivity
%Mo: amount of dissolved Mo

255
250
245
240

0.3

0.4

0.5

0.6

0.7

Amount of Dissolved Mo(%)

Fig. 6 Change of electrical resistivity with the amount of dissolved Mo in
the matrix.

DBTT = 75.5 - 0.449 ()

-32
-34
-36
-38
-40
-42
-44
-46
-48
235

240

245

250

255

260

265

270

275

Electrical Resistivity, / nm
Fig. 7 Correlation between electrical resistivity and DBTT.

GP zones has been reported to be approximately 1 nm.26)

Therefore, the conduction electron scattering eect arising
from carbide precipitates with micron order of size can be
neglected. With the aforementioned points in mind, it is
thought that the decrease of electrical resistivity in the
thermally aged 2.25Cr1Mo steel is mainly due to decrease
of conduction electron scattering arising from the solute Mo.
As shown in Fig. 5, the DBTT increased greatly in the
initial stage of aging while the electrical resistivity decreased
drastically. Figure 7 shows the correlation between electrical
resistivity and DBTT and a linear correlation between these
two can be seen. From the results of linear regression
analysis, the relationship can be represented as eq. (4).
DBTT 75:5  0:449 

Here, the DBTT is ductile-brittle transition temperature in
C

unit. The depletion of Mo dissolved in the matrix increases
the amount of Mo rich M6 C carbides and decreases the
electrical resistivity. As a result, the decrease of electrical
resistivity is seen to reect the increase of M6 C carbide. By
measuring electrical resistivity, the DBTT change arising
from carbide-induced embrittlement in the aged 2.25Cr1Mo
steel, could be estimated within 3
C of error. Therefore, the
electrical resistivity is proposed as a potential non-destructive evaluation parameter for assessing DBTT of the
thermally degraded 2.25Cr1Mo steel.
4.

260

235
0.2

-30

DBTT, T / oC

Change of electrical resistivity and correlation with

DBTT
As shown in Fig. 5, the electrical resistivity rapidly
decreased in the initial 1000 hours of aging and then changed
little thereafter. Electrical resistivity, which results from
conduction electron scattering, is aected by phonon
scattering, impurities, and lattice defect etc. In particular,
the defect of atomic unit size has been found to be a
predominant microstructural factor aecting electrical resistivity.23) In this respect, the reduction of electrical resistivity
with increasing aging time is thought to be closely related to
the depletion of Mo dissolved in the matrix. Electrical
resistivity has been reported to be linearly proportional to the
concentration of solute atom in the dilute alloy.20) The
resistivity of alloys increases with increasing electron
concentration (valence) of solute atom due to either dierent
atom size and/or local charge dierence.24) From the point of
view of atom size and valence, solute Mo (valence = 4,
atomic radius = 0.136 nm) dissolved in the Fe (valence =
2, atomic radius = 0.124 nm) matrix seems to scatter
conduction electrons eectively and hence increase the
electrical resistivity. Figure 6 shows the correlation between
the amount of Mo dissolved in the matrix and electrical
resistivity. It can be seen that the electrical resistivity is
linearly proportional to the amount of Mo dissolved in the
matrix. From the results of linear regression analysis, the
correlation between these two can be formulated as eq. (3).

1207

Summary and Conclusion

From the research to evaluate the thermally degraded

2.25Cr1Mo steel by electrical resistivity measurement
nondestructively, the following conclusions were drawn:
(1) The Mo was rapidly depleted in the initial 1000 hours
of aging and then hardly changed. The amount of M6 C
carbide increased by thermal aging. The DBTT increased
rapidly in initial about 1000 hours of aging and then changed
little afterward.
(2) The electrical resistivity decreased rapidly during
about 1000 hours of aging and then tended to level out. It
showed a linear correlation with the amount of Mo dissolved
in the matrix. The marked decrease of electrical resistivity in

1208

J. W. Byeon and S. I. Kwun

the initial stage of aging is thought to be mainly due to

decrease of conduction electron scattering arising from the
depletion of Mo dissolved in the matrix.
(3) The change of electrical resistivity showed a linear
correlation with that of DBTT of the thermally aged 2.25Cr
1Mo steel. Therefore, the DBTT change arising from
carbide-induced embrittlement of aged 2.25Cr1Mo steel,
could be evaluated nondestructively by measuring electrical
resistivity.
Acknowledgement
This work was supported by grant No. (R01-2002-00000018-0) from the Basic Research Program of the Korea
Science & Engineering Foundation.
REFERENCES
1) Z. Qu and K. H. Kuo: Metall. Trans. 12A (1981) 13331337.
2) N. S. Cheruvu: Metall. Trans. 20A (1989) 8797.
3) A. M. Adbel-Latif, J. M. Corbett and D. M. R. Taplin: Met. Sci. 16
(1982) 9096.
4) T. Wada and V. A. Biss: Metall. Trans. 14A (1983) 845855.
5) J. Pilling and N. Ridley: Metall. Trans. 13A (1982) 557563.
6) N. Gope, A. Chatterjee, T. Mukherjee and D. S. Sarma: Metall. Trans.
24A (1993) 315326.
7) Y. Watanabe and T. Shoji: Metall. Trans. 22A (1991) 20972106.
8) R. G. Baker and J. Nutting: J. Iron Steel Inst. 192 (1959) 257268.
9) R. Viswanathan and S. Gehl: JOM 44 (1992) 3442.

10) K. M. Yu, S. H. Nahm and Y. I. Kim: J. Mater. Sci. Lett. 18 (1999)

11751176.
11) N. Maeda, T. Goto, T. Kamimura, T. Naito, S. Kumano and Y. Nakao:
Int. J. Pres. Ves. & Piping 71 (1997) 712.
12) K. S. Ryu, S. H. Nahm, Y. B. Kim, K. M. Yu and D. Son: J. Magn.
Magn. Mater. 222 (2000) 128132.
13) M. Hirao, H. Ogi, N. Suzuki and T. Ohtani: Acta. Mater. 48 (2000)
517524.
14) S. Suzuki, K. Hirabayashi, H. Shibata, K. Mimura, M. Isshiki and Y.
Waseda: Scr. Mater., 48 (2003) 431435.
15) K. Matsumoto, S. Komatsu, M. Ikeda, B. Verlinden and P. Ratchev:
Mater. Trans., JIM 41 (2000) 12751281.
16) A. Yamamoto, K. Minami, U. Ishihara and H. Tsubakino: Mater.
Trans., JIM 39 (1998) 6974.
17) S. Hirosawa, T. Sato, J. Yokota and A. Kamio: Mater. Trans., JIM 39
(1998) 139146.
18) T. Ikeda, T. Tachikawa, M. Ikeda and S. Komatsu: Mater. Trans., JIM
38 (1997) 413419.
19) K. E. Stahlkopf, R. E. Smith, W. L. Server and R. A. Wullaert: ASTM
STP 601 (1975) 291307.
20) D. V. Morgan and K. Overshott: IEE Materials and Devices Series
Vol. 10-Electrical Resistivity Handbook, ed. G. T. Doys and T. Farrell,
Peter Peregrinus Ltd., London, United Kingdom, (1992) 13.
21) J. W. Byeon: Ph. D. thesis, Korea University, Korea, (2001).
22) R. Viswanathan and S. Gehl: J. Eng. Mater. Tech. 113 (1991) 263270.
23) D. D. Pollock: Electrical Conduction in Solids, (Am. Soc. Metals,
Ohio, USA, 1985).
24) R. E. Hummel: Electronic Properties of Materials-An Introduction for
Engineers, (Springer-Verlag, New York, USA, 1985) 78.
25) J. Miyake and M. E. Fine: Acta Metall. Mater. 40 (1992) 733741.
26) H. Herman, J. B. Cohen and M. E. Fine: Acta Metall. 11 (1963) 4356.