is 430520

8/3/2019 Is 430520

http://slidepdf.com/reader/full/is-430520 1/7

1. Introduction

Thermal desorption analysis (TDA) of hydrogen in high-

strength steels,1–6) titanium alloys, ceramics,7) and vitreous

silica glass7) has been attempted extensively over the past

few years, in order to clarify the mechanism of environ-

mental degradation of the mechanical properties. TDAenables us to separate the hydrogen trapping states in mate-

rials based on the peak temperatures of hydrogen desorbed

from materials during heating. These peak temperatures

are dependent on the metallurgical microstructures of

steels:8–10) martensite steel desorbs hydrogen at tempera-

tures below 200°C (peak 1), whereas, cold-drawn pearlite

steel desorbs not only peak 1 hydrogen but also hydrogen at

temperatures between 200°C and 500°C (peak 2).

Total hydrogen content in high-strength steels was mea-

sured and was correlated with environmental degradation of

the mechanical properties. Thereafter, it was reported that

the peak 1 hydrogen trapping state would affect environ-

mental degradation,11–13) whereas the peak 2 hydrogen trap-

ping state would not. However, experimental verification is

lacking. In addition, it is not yet established whether the

peak 2 hydrogen trapping state does not affect environmen-

tal degradation even though a large amount of peak 2 hy-

drogen is occluded into high-strength steels.

In such a background, high-strength steels occluding

only peak 1 hydrogen and only peak 2 hydrogen have been

prepared to investigate the relationship between each trap-

ping state of hydrogen and environmental degradation of

the mechanical properties. Based on the results, we intend to separate and extract the harmful and innocuous trapping

states of hydrogen in high-strength steels to the degrada-

tion, and then to clarify the trapping sites corresponding to

the two hydrogen trapping states. Furthermore, the mecha-

nism and model for harmful and innocuous states of hydro-

gen to the degradation are discussed.

2. Experimental

2.1. Materials

The high-strength steel employed in the present study is

eutectoid steel (JIS SWRS 82B) with the chemical compo-

sition shown in Table 1. One specimen occluding only peak

1 hydrogen was fabricated as follows.14) The specimen was

cold-drawn from 13 to 5 mm in diameter, then, after austen-

itizing at the temperature of 950°C for 15min, the speci-

ISIJ International, Vol. 43 (2003), No. 4, pp. 520–526

© 2003 ISIJ 520

Hydrogen in Trapping States Innocuous to Environmental

Degradation of High-strength Steels

Kenichi TAKAI and Ryu WATANUKI1)

Department of Mechanical Engineering, Faculty of Science and Technology, Sophia University, Kioi-cho, Chiyoda-ku, Tokyo 102-

8554 Japan. 1) Formerly at Department of Mechanical Engineering, Faculty of Science and Technology, Sophia

University, now at NEC.

(Received on August 31, 2002; accepted in final form on November 5, 2002 )

Hydrogen in trapping states innocuous to environmental degradation of the mechanical properties of high-

strength steels has been separated and extracted using thermal desorption analysis (TDA) and slow strain

rate test (SSRT). The high-strength steel occluding only hydrogen desorbed at low temperature (peak 1), as

determined by TDA, decreases in maximum stress and plastic elongation with increasing occlusion time of

peak 1 hydrogen. Thus the trapping state of peak 1 hydrogen is directly associated with environmental

degradation. The trap activation energy for peak 1 hydrogen is 23.4kJ/mol, so the peak 1 hydrogen corre-

sponds to weaker binding states and diffusible states at room temperature. In contrast, the high-strength

steel occluding only hydrogen desorbed at high temperature (peak 2), by TDA, maintains the maximum

stress and plastic elongation in spite of an increasing content of peak 2 hydrogen. This result indicates that

the peak 2 hydrogen trapping state is innocuous to environmental degradation, even though the steel oc-

cludes a large amount of peak 2 hydrogen. The trap activation energy for peak 2 hydrogen is 65.0 kJ/mol,

which indicates a stronger binding state and nondiffusibility at room temperature. The trap activation energy

for peak 2 hydrogen suggests that the driving force energy required for stress-induced diffusion during elas-

tic and plastic deformation, and the energy required for hydrogen dragging by dislocation mobility during

plastic deformation are lower than the binding energy between hydrogen and trapping sites. The peak 2 hy-

drogen, therefore, is believed to not accumulate in front of the crack tip and to not cause environmental

degradation in spite of being present in amounts as high as 2.9mass ppm.

KEY WORDS: hydrogen; high-strength steel; environmental degradation; cold-drawing; fracture; activation

energy; thermal desorption analysis.

8/3/2019 Is 430520


men was transformed isothermally in a lead bath at the tem-

perature of 350°C for 30min. This specimen is termed the

350°C-specimen. The specimen occluding only peak 2 hy-

drogen was fabricated as follows.14) An isothermal treat-

ment consisting of austenitizing at the temperature of

950°C for 15min and transforming in a lead bath at the

temperature of 550°C for 30min, then cold-drawing for 85% reduction (true strain: 1.91) was applied from the

specimen diameter of 13mm to the final specimen diameter

of 5 mm. Since this cold-drawn eutectoid steel occludes

both peak 1 and peak 2 hydrogen, the steel was heated to

200°C in order to remove only peak 1 hydrogen. This speci-

men occluding only peak 2 hydrogen is termed the

550°C–85%-specimen. The specimen preparation condi-

tions and tensile strengths are shown in Table 2.

The microstructures of these specimens were observed

using transmission electron microscope (TEM). Figure 1

shows bright-field images of 350°C-specimen and 550°C–

85%-specimen obtained by TEM. As shown in Fig. 1(a),

fine Fe3C (q ) is precipitated in ferrite (a ), that is, the metal-

lurgical microstructure of the 350°C-specimen is bainite.

The morphology of precipitated Fe3C in the 350°C-speci-

men is similar to that in tempered martensite. As shown in

Fig. 1(b), fine Fe3C is lengthened and arranged along the

cold-drawing direction, that is, the metallurgical mi-

crostructure of the 550°C–85%-specimen is cold-drawn

pearlite. In addition, dark-field images obtained by TEM in-

dicated that the Fe3C in the 350°C-specimen is single crys-

tal, while the Fe3C in the 550°C–85%-specimen is a fine-

grained nano-polycrystal due to cold-drawing, though the

images are omitted in the present paper.

2.2. Hydrogen Occlusion

The specimens were immersed in an aqueous solution of

20mass% NH4SCN kept at the temperature of 50°C under

various immersion times to occlude hydrogen. Hydrogen is

easily occluded in steels using this solution. The conditions

for hydrogen occlusion are the same as those standardized

by FIP (Fédération Internationale de la Précontrainte)15) for

the delayed fracture test of a prestressing steel.

2.3. TDA

The TDA apparatus consists of a gas chromatograph, a

thermal conductivity detector, a heating unit, and a

recorder. The gas chromatograph was calibrated with stan-

dard-mixture gas (Ar 50ppmH2) to quantify the hydrogen

content in the specimens. TDA was conducted in the tem-

perature range from 25 to 800°C and the sampling time of

the carrier gas to the gas chromatograph was at 5 min inter-

vals to obtain the hydrogen desorption profiles. The hydro-

gen content in the specimens of 5mm in diameter was mea-

sured immediately after occluding of hydrogen to prevent it

from desorbing. To calculate the trap activation energy of

hydrogen, the specimens were measured at various heating

rates such as 200, 300, 400, and 500°C/h immediately after

immersion in NH4SCN solution for 24 h. It is noted that

TDA enables us to identify the existing states of hydrogen

and calculate the trap activation energy of hydrogen from

the profiles.

2.4. Environmental Degradation Test

Environmental degradation of the mechanical properties

was evaluated as a decrease in maximum stress and plastic

elongation during the application of tensile stress, with a

slow strain rate test (SSRT). The 350°C-specimen and

550°C–85%-specimen for SSRT were processed to 3 mm

in diameter and 20mm in gauge length. The 350°C-speci-

men was mounted on the SSRT apparatus immediately after

occluding peak 1 hydrogen. The 550°C–85%-specimen

was mounted on the SSRT apparatus after occluding both

peak 1 and peak 2 hydrogen, and then removing only peak

1 hydrogen by heating to 200°C. The SSRT was conducted

at the strain rate of 5107/s in atmosphere. At this strain

rate, the specimens failed within approximately 48h. A

negligible amount of hydrogen was released within the test

time of 48h, as discussed later.

ISIJ International, Vol. 43 (2003), No. 4

521 © 2003 ISIJ

Table 1. Chemical compositions of specimen (mass%).

Table 2. Specimen preparation conditions and tensile strength.

Fig. 1. Transmission electron micrographs of high-strength

steels: (a) 350°C-specimen transformed isothermally at

350°C, and (b) 550°C–85%-specimen transformed

isothermally at 550°C and then cold-drawn with 85% re-

duction of area.

8/3/2019 Is 430520


3. Results

3.1. Separation of Hydrogen Trapping States in High-

strength Steels

Figure 2 shows hydrogen desorption profiles during con-

tinuous heating of the specimens measured using TDA at

the heating rate of 200°C/h immediately after immersion in

NH4SCN solution for 24h. The 350°C-specimen occludesonly the peak 1 hydrogen desorbed at the temperature of

120°C. Whereas, the 550°C–85%-specimen occludes both

peak 1 and peak 2 hydrogen desorbed at 120°C and 370°C,

respectively. This notable difference in hydrogen desorption

peaks is believed to be due to the trapping states of peak 1

and peak 2 hydrogen.

Figure 3 shows the contents of peak 1 and peak 2 hydro-

gen as a function of immersion time in NH4SCN solution.

The content of peak 1 hydrogen in the 350°C-specimen in-

creases with immersion time, and then gradually saturates

after 6h, as shown in Fig. 3(a). The contents of peak 1 and

peak 2 hydrogen in the 550°C–85%-specimen also increasewith immersion time, and then gradually saturate after 18 h,

as shown in Fig. 3(b). Additionally, the ratio of peak 1 and

peak 2 hydrogen during the hydrogen occlusion process re-

mains approximately constant. The 550°C–85%-specimen

can occlude as much as 3.9mass ppm peak 2 hydrogen.

3.2. Evaluation of Environmental Degradation

Figure 4 shows stress–strain curves of the 350°C–speci-

men occluding only peak 1 hydrogen and the 550°C–85%-

specimen occluding only peak 2 hydrogen under various


© 2003 ISIJ 522

Fig. 2. Hydrogen desorption profiles during continuous heating

of high-strength steels measured by TDA, after immer-sion in NH4SCN solution for 24h: (a) 350°C-specimen,

and (b) 550°C–85%-specimen immediately after hydro-

gen occlusion and after heating to 200°C to remove only

peak 1 hydrogen.

Fig. 3. Relationship between hydrogen content and immersion

time in NH4SCN solution for high-strength steels

determined by TDA: (a) 350°C-specimen and (b) 550°C–

85%-specimen.

Fig. 4. Stress–strain curves of high-strength steels under various

immersion times in NH4SCN solution: (a) 350°C-speci-

men occluding only peak 1 hydrogen, (b) 550°C–85%-

specimen occluding only peak 2 hydrogen.

8/3/2019 Is 430520


immersion times in NH4SCN solution, obtained by SSRT.

The maximum stress and strain of the 350°C-specimen de-

crease with increasing immersion time, namely, increasing

content of peak 1 hydrogen, as shown in Fig. 4(a). The de-

crease in maximum stress and strain saturates after 3 h of

immersion time. In contrast, stress–strain curves of the

550°C–85%-specimen are the same regardless of the im-

mersion time, namely, regardless of the content of peak 2

hydrogen, as shown in Fig. 4(b). However, the 550°C–85%-

specimen occluding both peak 1 and peak 2 hydrogen with-out heating to 200°C decreased in maximum stress and

strain with increasing immersion time. Hence, these results

suggest that trapping states of peak 1 hydrogen is harmful

while those of peak 2 hydrogen is innocuous to environ-

mental degradation.

Figure 5 shows the maximum stress and plastic elonga-

tion as a function of the occlusion times of peak 1 and peak

2 hydrogen. Concentrations in parenthesis show hydrogen

content of peak 1 and peak 2. The hydrogen concentrations

of only 18 h and 24 h are showed in Fig. 5, since hydrogen

distribution in the specimen probably is uniform at immer-

sion time of only 18h and 24h. The maximum stress and

plastic elongation initially decrease with increasing occlu-

sion time of peak 1 hydrogen, and then this decrease satu-

rates. Note that the maximum stress and plastic elongation

do not change even though the content of peak 2 hydrogen

increases to as high as 2.9 massppm.

Shown in Fig. 6 are fractographs of specimens occluding

only peak 1 and only peak 2 hydrogen, observed by scan-

ning electron micrograph. The 350°C-specimen occluding

0.8 mass ppm of only peak 1 hydrogen undergoes brittle

fracture, as shown in Fig. 6(a). In contrast, the 550°C–

85%-specimen occluding as much as 2.9 mass ppm of only

peak 2 hydrogen undergoes ductile fracture with elongation

and necking, as shown in Fig. 6 (b). This fractograph shows

the same fracture as that of the initial specimen without hy-

drogen.

These results indicate that the trapping state of peak 1hydrogen is harmful to environmental degradation of the

mechanical properties of high-strength steels, even though

the steel occludes a small amount of peak 1 hydrogen. On

the contrary, the trapping state of peak 2 hydrogen is in-

nocuous to the degradation, even though the steel occludes

a large amount of peak 2 hydrogen.

4. Discussion

4.1. Trapping States and Sites Corresponding to Peak

1 and Peak 2 Hydrogen

The mobility of peak 1 and peak 2 hydrogen in high-

strength steels has been measure by TDA. Figure 7 shows

desorption process of peak 1 and peak 2 hydrogen for spec-

imens kept in atmosphere under various holding times at

room temperature after immersion in NH4SCN solution for


523 © 2003 ISIJ

Fig. 5. Effects of immersion time in NH4SCN solution for high-

strength steels on (a) maximum stress, and (b) plastic

elongation. Concentration in parenthesis shows hydrogen

content of peak 1 and peak 2. 350°C-specimens occlud-

ing only peak 1 hydrogen decrease maximum stress and

plastic elongation with increasing immersion time,

whereas 550°C–85%-specimen occluding only peak 2

hydrogen does not decrease maximum stress and plastic

elongation with increasing immersion time.

Fig. 6. Fractographs of (a) 350°C-specimen occluding only peak

1 hydrogen and (b) 550°C–85%-specimen occluding only

peak 2 hydrogen. The specimen occluding peak 1 hydro-

gen of 0.8mass ppm shows brittle fracture, while the

specimen occluding peak 2 hydrogen of as high as

2.9 mass ppm shows ductile fracture.

8/3/2019 Is 430520


24 h. The content of peak 1 hydrogen gradually decreases

with holding time, as shown in Fig. 7(a). Hence, peak 1 hy-

drogen corresponds to weaker binding states and diffusible

states at room temperature.

In contrast, the content of peak 2 hydrogen initially de-

creases with holding time, then remains at 2.7 mass ppm

after 200 h, as shown in Fig. 7(b). The content of peak 2 hy-

drogen of 3.9mass ppm in the 550°C–85%-specimen im-

mediately after occluding hydrogen also decreased to

2.7 mass ppm after heating to 200°C in order to remove

only peak 1 hydrogen, then did not decrease with holding

time and remained at 2.7 mass ppm at room temperature

over 1 200 h, though the data are omitted in the present

paper. In addition, the content of peak 2 hydrogen below

2.7 mass ppm in 550°C–85%-specimen initially did not de-

crease with holding time, then remained approximatelyconstant. From these results, peak 2 hydrogen above

2.7 mass ppm consists of nondiffusible hydrogen and a part

of diffusible hydrogen. However, the peak 2 hydrogen in the

specimen used with SSRT corresponds to nondiffusible

states at room temperature. Additionally, these hydrogen

desorption processes show that a negligible amount of hy-

drogen is released within the SSRT time of 48 h.

To evaluate the hydrogen binding states qualitatively,

the trap activation energy of hydrogen was calculated from

the dependence of the peak temperatures on the heating

rates. The equation for trap activation energy ( E a) is as fol-

lows:16)

∂ ln(f /T p2)/∂(1/T p) E a/R.....................(1)

where f is the heating rate (K/s), T p is the peak temperature

(K), and R is the gas constant [8.13(J ·K 1 ·mol1)]. The

trap activation energy for peak 1 hydrogen is 23.4kJ/mol

and that for the peak 2 hydrogen is 65.0 kJ/mol. These val-

ues of the trap activation energy confirmed that trapping

state of peak 1 hydrogen is unstable whereas that of peak 2

hydrogen is stable.

The authors have recently investigated hydrogen trapping

sites corresponding to each desorption peak on the basis of

TDA and secondary ion mass spectrometry results concern-ing cold-drawn pure iron,8) eutectoid steel,14) and spheroidal

cast iron.17) The content of only peak 1 hydrogen increased

with the reduction degree of the cold-drawn area in pure

iron, whereas, the contents of both peak 1 and peak 2 hy-

drogen increased with the reduction degree of the cold-

drawn area in eutectoid steel. From these results, peak 2 hy-

drogen is occluded into the specimen when the following

condition is satisfied: the pearlite microstructure consisting

of lamellar Fe3C is cold-drawn. The trapping sites of peak 1

hydrogen should correspond to complex sites such as va-

cancies, vacancy clusters, strain field of dislocations, dislo-

cation cores, grain boundaries, and ferrite/Fe3C interfaces.This means that the trap activation energy obtained from

Eq. (1) is the average value of desorption from these sites

and from interstitial hydrogen in lattice without trapping.

On the other hand, the trapping sites of peak 2 hydrogen

should correspond to the strained interfaces between ferrite

and Fe3C and/or interface dislocations enclosed between

Fe3C lamellae due to cold-drawing.

4.2. Relationship between Hydrogen Trapping States

and Environmental Degradation

On the basis of the results that the trapping state of peak

2 hydrogen is innocuous while that of peak 1 hydrogen is

harmful to environmental degradation, the relationship be-

tween hydrogen trapping states and environmental degrada-

tion is discussed. Figure 8 shows a schematic illustration of

stress-induced hydrogen diffusion in front of a crack tip in

high-strength steels with/without cold-drawing. It is widely

recognized that the maximum stress region occurs around

the elastic-plastic boundary in front of the crack tip in the

case of the plane strain state. The stress-induced diffusion

of hydrogen is depicted by arrows in Fig. 8. The gray scale

in front of the crack tip indicates the stress level, where the

smallest, black region corresponds to a high stress level. In

the triaxial stress field in front of the crack tip, hydrogen ac-

cumulates in the region of the elongated lattice by means of elastic interaction. Furthermore, hydrogen is locally accu-

mulated by stress-induced diffusion, since dislocation and

microcracks occur in the maximum stress region.

Hydrogen trapped weakly in ferrite and along the inter-

faces of Fe3C is desorbed as peak 1 hydrogen, as shown in

Fig. 2(a), in the case of the 350°C-specimen. When tensile

stress is applied to the specimens during SSRT, a hydrogen

potential gradient corresponding to the stress gradient ap-

pears at the front of the crack tip as the initiation of a mi-

crocrack. Then a much higher local hydrogen content than

the average hydrogen content arises around the elastic-plas-

tic boundary. Thus, the diffusible peak 1 hydrogen in the

350°C-specimen is be believed to accumulate, causing en-

vironmental degradation such as decrease in maximum

stress and plastic elongation, in spite of hydrogen of

0.8 mass ppm, as shown in Fig. 6(a). The trap activation en-


© 2003 ISIJ 524

Fig. 7. Hydrogen release in atmosphere under various holding

times at room temperature for high-strength steels: (a)

350°C-specimen occluding only peak 1 hydrogen, and (b)

550°C–85%-specimen occluding both peak 1 and peak 2

hydrogen.

8/3/2019 Is 430520


ergy of 23.4kJ/mol for peak 1 hydrogen suggests that the

driving force energy required for stress-induced diffusion

during elastic and plastic deformation and that required for

hydrogen dragging by dislocation mobility during plastic

deformation are higher than binding energy between hydro-

gen and trapping sites. The trapping state of peak 1 hydro-

gen is, therefore, directly associated with environmental

degradation, that is, it is a harmful trapping state.

In contrast, hydrogen trapped strongly along the strained interfaces of Fe3C is desorbed as peak 2 hydrogen, as

shown in Fig. 2(b), in the case of the 550°C–85%-speci-

men. Though peak 2 hydrogen in the specimen immediately

after occluding hydrogen above 2.7mass ppm consists of

nondiffusible hydrogen and a part of diffusible hydrogen,

the content of peak 2 hydrogen in the specimen after heat-

ing to 200°C does not decrease with holding time and re-

mains at 2.7 massppm at room temperature over 1 200h,

that is, peak 2 hydrogen in the 550°C–85%-specimen for

environmental degradation test is in the only irreversibly

trapped state at room temperature. Hence, the nondiffusible

peak 2 hydrogen in the 550°C–85%-specimen is be be-

lieved to not accumulate in front of the crack tip and to not

cause environmental degradation in spite of a large amount

of hydrogen of as high as 2.9mass ppm, as shown in Fig.

6(b). The trap activation energy of 65.0 kJ/mol for peak 2

hydrogen suggests that the driving force energy required for

stress-induced diffusion during elastic and plastic deforma-

tion and that required for hydrogen dragging by dislocation

mobility during plastic deformation are lower than binding

energy between hydrogen and trapping sites. Therefore, the

trapping state of peak 2 hydrogen is not associated with en-

vironmental degradation, that is, it is the innocuous trap-

ping state, even though the steel occludes as much as of 2.9massppm of peak 2 hydrogen.

5. Conclusions

Hydrogen in trapping states innocuous to environmental

degradation of high-strength steels has been separated and

extracted using TDA and SSRT. The results obtained in the

present study can be summarized as follows.

(1) The high-strength steel occluding hydrogen des-

orbed at low temperature (peak 1), as determined by TDA,

decreases the maximum stress and plastic elongation with

increasing occlusion time of peak 1 hydrogen. The fracto-graph shows that peak 1 hydrogen causes brittle fracture of

high-strength steels. Thus the trapping state of peak 1 hy-

drogen is directly associated with environmental degrada-

tion of the mechanical properties, that is, it is harmful trap-

ping state.

(2) The high-strength steel occluding hydrogen des-

orbed at higher temperature (peak 2), as determined by

TDA, retains the maximum stress and plastic elongation in

spite of increasing content of peak 2 hydrogen. The fracto-

graph shows that peak 2 hydrogen as much as 2.9mass ppm

causes not brittle fracture but ductile fracture to high-

strength steels. Thus the trapping state of peak 2 hydrogen

is innocuous to environmental degradation, even though the

steel occludes a large amount of peak 2 hydrogen.

(3) The peak 1 hydrogen corresponds to weaker bind-

ing states and diffusible states at room temperature. The

trap activation energy for peak 1 hydrogen is 23.4 kJ/mol.

The trapping sites of peak 1 hydrogen should correspond to

complex sites such as vacancies, vacancy clusters, strain

field of dislocations, dislocation cores, grain boundaries,

ferrite/Fe3C interfaces, and interstitial hydrogen in lattice

without trapping.

(4) The peak 2 hydrogen corresponds to stronger bind-

ing states and nondiffusible states at room temperature. The

trap activation energy for peak 2 hydrogen is 65.0 kJ/mol.The trapping sites of peak 2 hydrogen should correspond to

the strained interfaces between ferrite and Fe3C and/or in-

terface dislocations enclosed between Fe3C lamellae due to

cold-drawing.

(5) Since the peak 2 hydrogen is nondiffusible at room

temperature and its trap activation energy is 65.0kJ/mol,

this hydrogen is believed to not accumulate in front of

crack tip and to not cause environmental degradation in

spite of being present in amounts as high as 2.9 mass ppm.

The trap activation energy of 65.0 kJ/mol for peak 2 hydro-

gen suggests that the driving force energy required for

stress-induced diffusion during elastic and plastic deforma-

tion and that required for hydrogen dragging by dislocation

mobility during plastic deformation are higher than binding

energy between hydrogen and trapping sites.


525 © 2003 ISIJ

Fig. 8. Schematic illustration of stress-induced hydrogen diffu-

sion in front of crack tip of high-strength steels

with/without cold-drawing: (a) 350°C-specimen occlud-

ing only peak 1 hydrogen, that is, only diffusible hydro-gen, and (b) 550°C–85%-specimen occluding only peak

2 hydrogen, that is, only nondiffusible hydrogen.

8/3/2019 Is 430520


Acknowledgments

This study was performed with the support of Special

Coordination Funds of the Ministry of Education, Culture,

Sports, Science and Technology of Japanese Government.

REFERENCES

1) M. Nagumo: ISIJ Int., 41 (2001), 590.2) M. Nagumo, K. Takai and N. Okuda: J. Alloys Compounds, 293–295

(1999), 310.

3) M. Nagumo, M. Nakamura and K. Takai: Metall. Mater. Trans. A,

32A (2001), 339.

4) D. G. Enos and J. R. Scully: Metall. Mater. Trans. A, 33A (2002),

1151.

5) K. Takai: Bull. Jpn. Inst. Met., 39 (2000), 239.

6) K. Takai: Corros. Eng., 49 (2000), 395.

7) K. Takai: Report on the Project “Function of Hydrogen on

Environmental Degradation of Structural Materials—IV”, ISIJ,

Tokyo, (2002), 130.

8) K. Takai, G. Yamauchi, M. Nakamura and M. Nagumo: J. Jpn. Inst.

Met., 62 (1998), 267.

9) K. Takai, J. Seki and Y. Homma: Tetsu-to-Hagané, 81 (1995), 1025.

10) K. Takai, Y. Homma K. Izutsu and M. Nagumo: J. Jpn. Inst. Met., 60

(1996), 1155.

11) N. Suzuki, N. Ishii and T. Miyagawa: Tetsu-to-Hagané, 82 (1996),

170.

12) T. Kushida, H. Matsumoto, N. Kuratomi, T. Tsumura, F. Nakasatoand T. Kudo: Tetsu-to-Hagané, 82 (1996), 297.

13) N. Suzuki, N. Ishii, T. Miyagawa and H. Harada: Tetsu-to-Hagané,

79 (1993), 227.

14) K. Takai and A. Nozue: J. Jpn. Inst. Met. , 64 (2000), 669.

15) Fédération Internationale de la Précontrainte: Report on Prestressing

Steel 5, (1980), Sep., 1.

16) W. Y. Choo and J. Y. Lee: Metall. Mater. Trans. A, 13 (1982), 135.

17) K. Takai, Y. Chiba, K. Noguchi and A. Nozue: Metall. Mater. Trans.

A, 33A (2002), 2659.


© 2003 ISIJ 526

is 430520

Documents