i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 3 3 7 1 e3 3 7 9

Available online at www.sciencedirect.com

ScienceDirect

journal homepage: www.elsevier.com/locate/he

Effect of strain rate on hydrogen embrittlement in

low-carbon martensitic steel

Yuji Momotani a,*, Akinobu Shibata a,b, Daisuke Terada c,

Nobuhiro Tsuji a,b
a
Department of Materials Science and Engineering, Kyoto University, Yoshida-honmachi, Sakyo-ku, Kyoto 606-8501,
Japan
b
Elements Strategy Initiative for Structural Materials (ESISM), Kyoto University, Yoshida-honmachi, Sakyo-ku,
Kyoto 606-8501, Japan
c
Department of Mechanical Sciences and Engineering, Chiba Institute of Technology, 2-17-1 Tsudamuma,
Narashino, Chiba 275-0016, Japan

article info abstract

Article history: This study investigated the effect of strain rate on hydrogen embrittlement behavior in a
Received 5 August 2016 low-carbon martensitic steel. Elongation of the hydrogen-charged specimen decreased
Received in revised form significantly with decreasing the strain rate. The characteristics of the hydrogen-related
22 September 2016 fracture behavior also changed with the strain rate. Hydrogen micro-print technique and
Accepted 28 September 2016 electron backscattering diffraction analysis revealed that the deformation at a lower strain
Available online 20 October 2016 rate facilitated hydrogen to accumulate mainly on prior austenite grain boundaries. This
hydrogen accumulation led to the formation of micro-cracks along prior austenite grain
Keywords: boundaries and brittle fracture on or in the vicinity of prior austenite grain boundaries. On
Hydrogen embrittlement the other hand, in the case of a higher strain rate, micro-cracks formed mainly inside prior
Martensitic steel austenite grains and transgranular fracture occurred. This is presumably because there
Strain rate was not enough time for hydrogen to accumulate on prior austenite grain boundaries
Electron backscattering diffraction during tensile test.
Hydrogen micro-print technique © 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.

for application of high strength martensitic steels in

Introduction hydrogen-energy society as infrastructures. Low and medium
carbon high strength steels used usually have lath martensite
High strength martensitic steels have been used in various structures. The lath martensite structure consists of several
industrial applications, such as automobiles, constructions, structural units with different length scales, i.e., lath, block,
tools etc. However, it is well known that high strength packet, and prior austenite grain [6e9]. The lath is a single
martensitic steels are highly sensitive to hydrogen embrit- crystal of martensite with a thickness of about 0.2 mm. The
tlement [1e5]. Because a certain amount of hydrogen is block consists of many laths having nearly identical crystal
inevitably introduced to the steels during fabrication pro- orientation. The packet consists of several blocks with the
cesses, it is desired to overcome hydrogen embrittlement of same habit plane. It is thus important to characterize
high strength martensitic steels. In addition, this is important

* Corresponding author. Fax: þ81 75 753 4978.

E-mail address: yuji.momotani@tsujilab.mtl.kyoto-u.ac.jp (Y. Momotani).
http://dx.doi.org/10.1016/j.ijhydene.2016.09.188
0360-3199/© 2016 Hydrogen Energy Publications LLC. Published by Elsevier Ltd. All rights reserved.
3372 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 3 3 7 1 e3 3 7 9

hydrogen-related fracture in high strength martensitic steels tensile test conditions (strain rates). Sheet-type tensile test
from a viewpoint of lath martensitic microstructure, in order specimens with a gauge length of 10 mm, width of 5 mm, and
to improve hydrogen embrittlement properties. thickness of 1 mm were cut from the heat-treated sheets
Takai et al. [10] investigated local hydrogen distribution in 1 mm thick by spark wire-cutting machine. After the surfaces
lath martensitic steels by the use of secondary ion mass were polished with a 0.3 mm diamond suspension, hydrogen
spectroscopy, and reported that prior austenite grain bound- was introduced into the tensile test specimens by electro-
aries were main hydrogen trapping site. Shibata et al. [11,12] chemical charging with a current density of 1.0 A m2 for
characterized hydrogen-related crack propagation behavior 86.4 ks in a 3% NaCl aqueous solution containing 3 g L1
in martensitic steels by electron backscattering diffraction NH4SCN at room temperature. The hydrogen content in the
(EBSD) analysis. They found that the micro-cracks formed on specimen was measured by means of thermal desorption
or in the vicinity of prior austenite grain boundaries and spectroscopy (TDS) at a heating rate of 100 K h1. Fig. 1 shows
propagated along certain {011}M crystallographic planes. Kim typical desorption profiles of the hydrogen-charged specimen
et al. [13] and Nagao et al. [14] observed microstructures just and the uncharged specimen obtained by TDS measurement.
beneath the quasi-cleavage fracture surface in high strength We conducted three TDS measurements, and the average
steels by transmission electron microscopy. Their results diffusible hydrogen content, defined as the amount of
suggested that quasi-cleavage fracture surfaces were parallel hydrogen desorbed at temperature from room temperature to
to lath boundaries. It is well known that hydrogen also in- 573 K in the TDS analysis, in the uncharged and the hydrogen-
duces intergranular fracture. In the case of martensitic steel, charged specimens were 0.00 wt.ppm and 4.12 (±0.25) wt.ppm,
hydrogen-related intergranular fracture occurs at prior respectively. All of the tensile tests and the TDS measure-
austenite grain boundaries [15e19]. ments started 2.4 ks after finishing the hydrogen charging
On the other hand, it is well known that hydrogen process to keep the same hydrogen content in each hydrogen-
embrittlement behavior depends on not only microstructure charged specimen.
but also deformation conditions [20e24]. Previous studies re- Microstructures were observed by optical microscopy and
ported that the susceptibility to hydrogen embrittlement scanning electron microscopy (SEM, FEI: XL30S-FEG). For the
increased with decreasing the strain rate at any temperatures. microstructural observations, the specimens were polished
Although several attempts have been conducted, the reason electrolytically in a solution of 10% HClO4 þ 90% CH3COOH,
why hydrogen embrittlement behavior changes depending on and then etched in a solution of 3% HNO3 þ 97% C2H5OH.
the strain rate is still unclear. We considered that hydrogen Fracture surfaces of the specimens after the tensile tests were
accumulation behavior has a large influence on the strain rate observed by SEM. Crystallographic features of the hydrogen-
sensitivity of hydrogen embrittlement. Then, the present related cracks were analyzed by using EBSD.
study conducted uniaxial tensile tests at various strain rates Hydrogen accumulation behavior during the tensile test
in a low-carbon martensitic steel and discussed effect of was evaluated by means of hydrogen micro-print technique.
strain rate on hydrogen embrittlement from the viewpoint of The hydrogen micro-print technique is a method to reveal
hydrogen accumulation behavior. local hydrogen distribution using a reaction between Agþ and
hydrogen emitted from the specimen [25e28]. The sheet-type
tensile test specimens were polished electrolytically, and then
Experimental procedure one side of the surface was coated with an acid proof tape.
After the hydrogen charging, the acid proof tape was removed
An Fe-0.2C (wt.%) binary alloy was used in the present study. and the specimen was covered with a liquid emulsion con-
This is a simple model alloy for obtaining high strength taining AgBr particles (Ilford L-4) and gelatin diluted by 10%
martensitic structure. The chemical composition of the 0.2C
steel is shown in Table 1. A cast ingot of the steel was cold-
rolled to about 1.5 mm thickness and then austenitized at
1323 K for 1.8 ks in vacuum, followed by iced-brine quench-
ing and sub-zero cooling in liquid nitrogen. After the heat
treatment, both sides of the specimens were mechanically
ground until the final thickness of 1 mm was reached to
remove decarburized layers formed during the heat
treatment.
Hydrogen embrittlement behavior of the specimens was
evaluated by uniaxial tensile tests at various strain rates
ranging from 8.3  106 s1 to 8.3  101 s1 at room temper-
ature in air. At least two tests were performed for each of the

Table 1 e Chemical composition of the alloy used in the

present study (wt.%).
Fig. 1 e Typical desorption profiles of the hydrogen-
C Si Mn P S O N Fe
charged specimen and the uncharged specimen obtained
0.21 <0.02 <0.02 <0.005 0.0005 0.0009 0.0008 Bal. by TDS measurement.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 3 3 7 1 e3 3 7 9 3373

NaSO2 aqueous solution by using a wire loop method. During

the tensile test, hydrogen diffuses through the microstructure
and reduces the Agþ to Ag particle by the following reaction,

AgBr þ H ¼ Ag þ Hþ þ Br. (1)

After the tensile tests, the specimens were dipped into

HCHO aqueous solution for 3 s to harden gelatin, and then
immersed in a 15% Na2S2O3 aqueous solution diluted by 10%
NaSO2 aqueous solution for 10 min to eliminate the remaining
AgBr particles which had not reacted with hydrogen. All of the
procedures in the hydrogen micro-print technique were car-
ried out in a dark room to prevent AgBr particles from reacting
with light.

Results & discussion

Fig. 2 shows (a) an optical microscopy image and (b) a corre-

sponding EBSD orientation map of the as-quenched specimen.
In the EBSD orientation map, block boundaries, packet
boundaries, and prior austenite grain boundaries identified by
an orientation analysis are represented as black lines, yellow
lines, and black broken lines, respectively. The as-quenched
specimen showed fully lath martensite structures consisting
of blocks and packets inside each prior austenite grain. The
mean block, packet, and prior austenite grain sizes are 1.9 mm,
14 mm, 76 mm, respectively.
Fig. 3 shows engineering stressestrain curves of (a) the
uncharged specimens and (b) the hydrogen-charged speci-
mens tensile-tested at various strain rates. Fig. 4 summarizes
the ultimate tensile strength and the total elongation of the
uncharged specimens and the hydrogen-charged specimens
as a function of the strain rate. The uncharged specimens
tensile-tested at all strain rates exhibited a large amount of
post uniform elongation after reaching the tensile strength
between 1400 and 1550 MPa. Total elongation of the un-
charged specimens decreased from 12.7% to 7.5% with Fig. 2 e (a) Optical microscopy image and (b) corresponding
decreasing the strain rate. Compared with the uncharged EBSD orientation map of the as-quenched specimen. In the
specimens, the ductility of the hydrogen-charged specimens EBSD orientation map of (b), block boundaries, packet
was much smaller and significantly depended on the strain boundaries, and prior austenite grain boundaries
rate. The total elongation of the hydrogen-charged specimens crystallographically identified by an orientation analysis
decreased with decreasing the strain rate from 8.2% at are represented as black lines, yellow lines, and black
8.3  101 s1 to 0.3% at 8.3  106 s1. In particular, the broken lines, respectively.
hydrogen-charged specimens fractured within elastic defor-
mation range when the strain rates were 8.3  105 s1 and
8.3  106 s1. We confirmed from the TDS measurement that SEM images of the fracture surfaces of the hydrogen-charged
a small amount of hydrogen (0.46 wt.ppm) was desorbed specimens tensile-tested at different strain rates: (a), (b)
during the tensile test at a strain rate of 8.3  106 s1. The 8.3  106 s1, and (c), (d) 8.3  101 s1. As shown in Fig. 6 (a),
period of the tensile test was the longest when the strain the macroscopic fracture surface after the tensile test at a
rate was 8.3  106 s1. Thus, we can ignore the effect of strain rate of 8.3  106 s1 was characterized by intergranular
hydrogen desorption during the tensile test on the mechan- fracture. It should be noted, however, that serrated markings
ical properties. Consequently it can be concluded that the can be observed on some flat fracture surfaces in the high-
hydrogen embrittlement was enhanced with decreasing the magnification SEM image of Fig. 6 (b). On the other hand, the
strain rate. fracture surface of the hydrogen-charged specimen after the
Fig. 5 shows SEM images of the fracture surfaces of the tensile test at a strain rate of 8.3  101 s1 consisted of mainly
uncharged specimens tensile-tested at strain rates of (a) dimple patterns (Fig. 6 (c)) and a small amount of intergranular
8.3  106 s1 and (b) 8.3  101 s1. The fracture surface of the fracture surface (Fig. 6 (d)). Area fractions of intergranular
uncharged specimens at all strain rates consisted of ductile fracture surfaces with smooth facets, intergranular fracture
fracture surfaces covered by dimple patterns. Fig. 6 shows surfaces with serrated markings, and typical ductile fracture
3374 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 3 3 7 1 e3 3 7 9

Fig. 4 e Ultimate tensile strength and the total elongation

of the uncharged specimens and the hydrogen-charged
specimens tensile-tested plotted as a function of the strain
rate.

Fig. 3 e Engineering stressestrain curves of (a) the crystallographic plane and seems to coincide well with the
uncharged specimens and (b) the hydrogen-charged extrapolated traces of the prior austenite grain boundaries.
specimens tensile-tested at various strain rates. This results suggest that the fracture surface with smooth
facet was generated by intergranular fracture on prior
austenite grain boundary. On the other hand, it is clear that
surfaces covered by dimple patterns were measured from the fracture surface of the lower part in Fig. 8 (d) is close to but
pixel numbers of each surface in SEM images. Fig. 7 summa- not on the prior austenite grain boundaries. In addition, the
rizes the area fractions of each characteristic fracture surface fracture surface is nearly parallel to the traces of {011} planes
as a function of the strain rate. It is found from Fig. 7 that the (red dotted lines). This indicates that the fracture was not an
fraction of the macroscopic intergranular fracture surfaces, exact intergranular fracture at prior austenite grain bound-
including both surfaces with smooth facets and those with aries, but the fracture proceeded parallel to {011} planes in a
serrated markings, increased with decreasing the strain rate. transgranular manner in the vicinity of prior austenite grain
This corresponds well with the decrease in the tensile boundaries. The hydrogen-related fracture along {011} planes
ductility shown in Fig. 4. have been reported previously [11,12]. Therefore, we conclude
Fig. 8 (a) and (b) show SEM images around the fracture that the intergranular fracture surface with serrated markings
surfaces of the hydrogen-charged specimen after the tensile was originated from quasi-cleavage fracture in the vicinity of
test at a strain rate of 8.3  106 s1. In Fig. 8, broad surfaces of prior austenite grain boundaries.
the sheet-type tensile test specimen were observed, and the Fig. 9 shows SEM images of the areas approximately 500 mm
tensile direction was parallel to the horizontal direction of the away from the fracture surfaces in the hydrogen-charged
images. The fracture surface in Fig. 8 (a) consists of smooth specimens tensile-tested at strain rates of (a) 8.3  106 s1
facets, while that in Fig. 8 (b) is uneven. This suggests that and (c) 8.3  101 s1. Micro-cracks are observed in these SEM
fracture surfaces of Fig. 8 (a) and (b) presumably correspond to images. The relationship between the micro-cracks and mi-
the intergranular fracture surface with smooth facets and the crostructures of the martensite matrix was characterized by
intergranular fracture surface with serrated markings, EBSD. Fig. 9 (b) and (d) are EBSD orientation maps of the area
respectively. Fig. 8 (c) and (d) present EBSD orientation maps of (a) and (c), respectively. Block boundaries, packet boundaries,
the same areas of Fig. 8 (a) and (b). Block boundaries, packet and prior austenite grain boundaries identified through an
boundaries, and prior austenite grain boundaries identified orientation analysis are drawn in black lines, yellow lines, and
through an orientation analysis are drawn in black lines, black broken lines, respectively. At a lower strain rate of
yellow lines, and black broken lines, respectively. As shown in 8.3  106 s1 (Fig. 9 (a) and (b)), the observed micro-cracks
Fig. 8 (c), the smooth facet is not parallel to a certain formed on or in the vicinity of prior austenite grain
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 3 3 7 1 e3 3 7 9 3375

Fig. 7 e Area fractions of characteristic fracture surfaces in

the hydrogen-charged specimens as a function of the
strain rate.

boundaries. As shown in Fig. 9 (c) and (d), on the other hand,

some micro-cracks formed inside prior austenite grains at a
higher strain rate of 8.3  101 s1. In contrast, it should be
noted that clear micro-cracks were not observed in the un-
charged specimens after the tensile tests at all strain rates.
As described above, it was found that the hydrogen-related
fracture behavior changed greatly depending on the strain
rate. This is presumably because hydrogen accumulation
behavior during the tensile tests was different with the strain
Fig. 5 e SEM images showing the fracture surfaces of the rate. In the following, hydrogen accumulation behavior during
uncharged specimens after tensile test at different strain the tensile tests were analyzed through the hydrogen micro-
rates: (a) 8.3 £ 10¡6 s¡1, (b) 8.3 £ 10¡1 s¡1. print technique. Fig. 10 shows SEM images and correspond-
ing EBSD orientation maps of the specimen tensile-tested at a
strain rate of 8.3  106 s1 after hydrogen micro-print treat-
ment. Two types of distinctive distributions of Ag particles are

Fig. 6 e SEM images showing the fracture surfaces of the hydrogen-charged specimens after tensile test at different strain
rates: (a), (b) 8.3 £ 10¡6 s¡1, (c), (d) 8.3 £ 10¡1 s¡1.
3376 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 3 3 7 1 e3 3 7 9

Fig. 8 e SEM images ((a), (b)) and corresponding EBSD orientation maps ((c), (d)) around fracture surfaces after the tensile test
at a strain rate of 8.3 £ 10¡6 s¡1. The observation direction is parallel to the normal direction of the sheet-type tensile test
specimen. The gray regions (left hand sides) in (a), (b) are Ni layers plated. Block boundaries, packet boundaries, and prior
austenite grain boundaries determined by orientation analysis are drawn in black thin lines, yellow lines, and black broken
lines, and the traces of {011} planes near the fracture surfaces are indicated by red broken lines, respectively.

observed in the same specimen as shown in Fig. 10 (a) and (c). corresponding EBSD orientation map (Fig. 10 (d)) reveals that
In Fig. 10 (a) and (b), Ag particles seem precipitating along the Ag particles appeared preferentially along the prior
various boundaries in the lath martensite structure, such as austenite grain boundaries. These results suggest that
lath boundaries, block boundaries, packet boundaries, and hydrogen might be trapped mainly at dislocations after the
prior austenite grain boundaries. On the other hand, a simple hydrogen charging and the hydrogen tended to accumulate on
network of Ag particles is observed in Fig. 10 (c). The prior austenite grain boundaries at least in some regions,
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 3 3 7 1 e3 3 7 9 3377

Fig. 9 e SEM images and EBSD orientation maps of the same areas showing micro-cracks in the hydrogen-charged
specimens after the tensile test at strain rates of (a), (b) 8.3 £ 10¡6 s¡1, and (c), (d) 8.3 £ 10¡1 s¡1. The observation direction
is parallel to the normal direction of the sheet-type tensile test specimen. Block boundaries, packet boundaries, and prior
austenite grain boundaries identified through an orientation analysis are drawn by black lines, yellow lines, and
black broken lines, respectively in the EBSD maps ((b), (d)). The positions of micro-cracks are represented by white lines in
(b) and (d).

though the reason why hydrogen accumulation behavior was hydrogen to accumulate on prior austenite grain boundaries
different in the identical hydrogen-charged specimen is un- (or other sites) during the tensile test at a higher strain rate
clear. The accumulation of hydrogen along the prior austenite (8.3  101 s1). Although there is still a possibility that Ag
grain boundaries would lead to the intergranular fracture particles did not have enough time to react with hydrogen
frequently observed in this specimens (Fig. 6 (b)). because the period of the tensile test was too short (0.2 s), the
Fig. 11 shows an SEM image of the specimen tensile-tested hydrogen accumulation behavior shown in Fig. 11 is consis-
at a strain rate of 8.3  101 s1 after hydrogen micro-print tent with the formation site of micro-cracks (Fig. 9 (c) and (d)).
treatment. In contrast to Fig. 10, Ag particles are randomly On the basis of the present experimental results, it was
distributed in the microstructure, which indicates that found that hydrogen accumulated preferentially at prior
hydrogen did not accumulate on specific site in the micro- austenite grain boundaries during the tensile test at a lower
structure. The period of time up to fracture in the tensile test strain rate, which led to the formation of micro-cracks around
at a strain rate of 8.3  101 s1 was 0.2 s, which is much prior austenite boundaries and intergranular fracture. In the
shorter than that at a strain rate of 8.3  106 s1 (4.68 ks). case of a higher strain rate, hydrogen was considered to exist
Accordingly, we consider that there was not enough time for uniformly in the lath martensite microstructure. Therefore
3378 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 3 3 7 1 e3 3 7 9

Fig. 10 e SEM images and corresponding EBSD orientation maps of the specimen tensile-tested at a strain rate of
8.3 £ 10¡6 s¡1 after hydrogen micro-print treatment. Block boundaries, packet boundaries, and prior austenite grain
boundaries identified through orientation analysis are drawn by black lines, yellow lines, and black broken lines,
respectively, in the EBSD maps ((b), (d)).

decrease of elongation compared with the uncharged speci-

mens. These are probably the reason why the hydrogen
embrittlement (intergranular fracture) was suppressed with
increasing the strain rate.

Summary

In the present study, tensile tests of the hydrogen-charged

and uncharged specimens of 0.2C martensitic steel were
conducted at various strain rates, and the effect of strain rate
on hydrogen embrittlement behavior was studied. The main
results obtained are as follows:

1. The total elongation of the hydrogen-charged specimens

decreased significantly with decreasing the strain rate in
the uniaxial tensile test at room temperature. Three types
Fig. 11 e SEM image of the specimen tensile-tested at a of characteristic fracture surfaces, i.e., intergranular frac-
strain rate of 8.3 £ 10¡1 s¡1 after hydrogen micro-print ture surfaces with smooth facets, intergranular fracture
treatment. surfaces with serrated markings, and typical ductile frac-
ture surfaces (dimple patterns) were observed in the
micro-cracks formed mainly inside prior austenite grains and hydrogen-charged specimens. The fraction of the macro-
transgranular fracture occurred. Even when the strain rate scopic intergranular fracture surfaces, including both sur-
was high, the hydrogen distributed rather randomly facili- faces with smooth facets and surfaces with serrated
tated the transgranular fracture, resulting in the slight markings, increased with decreasing the strain rate.
 i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y 4 2 ( 2 0 1 7 ) 3 3 7 1 e3 3 7 9 3379

2. At a lower strain rate of 8.3  106 s1, almost all micro- [9] Kitahara H, Ueji R, Tsuji N, Minamino Y. Crystallographic
cracks formed around prior austenite grain boundaries, features of lath martensite in low-carbon steel. Acta Mater
and brittle fracture occurred on or in the vicinity of prior 2006;54:1279e88.
[10] Takai K, Seki J, Homma Y. Observation of trapping sites of
austenite grain boundaries. At a higher strain rate of
hydrogen and deuterium in high-strength steels by using
8.3  101 s1, in contrast, micro-cracks mainly formed secondary ion mass spectrometry. Mater Trans JIM
inside prior austenite grains, and ductile fracture occurred 1995;36:1134e9.
within the prior austenite grains. [11] Shibata A, Takahashi H, Tsuji N. Microstructural and
3. Hydrogen micro-print technique and EBSD analysis indi- crystallographic features of hydrogen-related crack
cated that hydrogen preferentially accumulated on prior propagation in low carbon martensitic steel. ISIJ Int
2012;52:208e12.
austenite grain boundaries during the tensile test at a
[12] Shibata A, Murata T, Takahashi H, Matsuoka T, Tsuji N.
lower strain rate (8.3  106 s1). On the other hand, at a
Characterization of hydrogen-related fracture behavior in
higher strain rate (8.3  101 s1), hydrogen existed uni- as-quenched low-carbon martensitic steel and tempered
formly in the lath martensite microstructure probably medium-carbon martensitic steel. Metall Mater Trans A
because there was not enough time for hydrogen to accu- 2015;46A:5685e96.
mulate on prior austenite grain boundaries during the [13] Kim YH, Morris Jr JW. The nature of quasicleavage fracture in
tensile test. tempered 5.5Ni steel after hydrogen charging. Metall Trans A
1983;14A:1883e8.
[14] Nagao A, Smith CD, Dadfarnia M, Sofronis P, Robertson IM.
The role of hydrogen in hydrogen embrittlement fracture of
lath martensitic steel. Acta Mater 2012;60:5182e9.
Acknowledgements [15] Banerji SK, McMahon Jr CJ, Feng HC. Intergranular fracture in
4340-type steels: effects of impurities and hydrogen. Metall
Trans A 1978;9A:237e47.
This study was financially supported by the Grant-in-Aid for [16] Craig BD, Krauss G. The structure of tempered martensite
Scientific Research (B) (No. 15H04158), and the Elements and its susceptibility to hydrogen stress cracking. Metall
Strategy Initiative for Structural Materials (ESISM), all through Trans A 1980;11A:1799e808.
the Ministry of Education, Culture, Sports, Science and Tech- [17] Wang M, Akiyama E, Tsuzaki K. Effect of hydrogen and stress
nology (MEXT), Japan. YM was supported by the JSPS Fellow- concentration on the notch tensile strength of AISI 4135
ship (No. 26∙2927). NT and AS were also supported by the ISIJ steel. Mater Sci Eng A 2005;A398:37e46.
[18] Wang M, Akiyama E, Tsuzaki K. Effect of hydrogen on the
Research Promotion Grant. The authors greatly appreciate all
fracture behavior of high strength steel during slow strain
the supports. rate test. Corros Sci 2007;49:4081e97.
[19] Wang G, Yan Y, Li J, Huang J, Su Y, Qiao L. Hydrogen
embrittlement assessment of ultra-high strength steel
references 30CrMnSiNi2. Corros Sci 2013;77:273e80.
[20] Ohnishi T, Higashi K, Inoue N, Nakatani Y. Effect of
temperature and strain rate on hydrogen embrittlement of
[1] Li S, Zhang Z, Akiyama E, Tsuzaki K, Zhang B. Evaluation of an Al-8%Mg alloy. J Jpn Inst Met 1981;45:972e6.
susceptibility of high strength steels to delayed fracture by [21] Brown JT, Baldwin Jr WM. Hydrogen embrittlement of steels.
using cyclic corrosion test and slow strain rate test. Corros Trans AIME 1954;200:298e303.
Sci 2010;52:1660e7. [22] Bernstein IM. The role of hydrogen in the embrittlement of
[2] Lee SJ, Ronevich JA, Krauss G, Matlock DK. Hydrogen iron and steel. Mater Sci Eng 1970;6:1e19.
embrittlement of hardened low-carbon sheet steel. ISIJ Int [23] Toh T, Baldwin Jr WM. In: Robertson WD, editor. Stress
2010;50:294e301. corrosion cracking and embrittlement. New York: John Wiley
[3] Michler T, Naumann J. Microstructural aspects upon and Sons; 1956. p. 176.
hydrogen environment embrittlement of various bcc steels. [24] Gray HR, Nelson HG, Johnson RE, Mcpherson WB, Howard FS,
Int J Hydrogen Energy 2010;35:821e32. Swisher JH. Potential structural material problems in a
[4] Takasawa K, Ikeda R, Ishikawa N, Ishigaki R. Effects of grain hydrogen energy system. Int J Hydrogen Energy
size and dislocation density on the susceptibility to high- 1978;3:105e18.
pressure hydrogen environment embrittlement of high- [25] Ovejero-Garcia J. Hydrogen microprint technique in the
strength low-alloy steels. Int J Hydrogen Energy study of hydrogen in steels. J Mater Sci 1985;80:2623e9.
2012;37:2669e75. [26] Ronevich JA, Speer JG, Krauss G, Matlock DK. Improvement
[5] Zhu X, Li W, Zhao H, Wang L, Jin X. Hydrogen trapping sites of the hydrogen microprint technique on AHSS steels. Metall
and hydrogen-induced cracking in high strength quenching Microstruct Anal 2012;1:79e84.
& partitioning (Q&P) treated steel. Int J Hydrogen Energy [27] Saito H, Miyazawa K, Ishida Y. Tritium transmission electron
2014;39:13031e40. microscopic autoradiography of hydrogen trapping sites at
[6] Marder AR, Krauss G. The morphology of martensite in iron- interfaces in an austenitic stainless steel SUS316L. J Jpn Inst
carbon alloys. Trans ASM 1967;60:651e60. Met 1991;55:366e75.
[7] Marder JM, Marder AR. The morphology of iron-nickel [28] Kuramoto S, Ichitani K, Nagao A, Kanno M. Effect of gelatin
massive martensite. Trans ASM 1969;62:1e10. hardening on hydrogen visualization in steels by
[8] Morito S, Huang X, Furuhara T, Maki T, Hansen N. The hydrogen microprint technique. TETSU-TO-HAGANE
morphology and crystallography of lath martensite in alloy 2000;86:17e23.
steels. Acta Mater 2006;54:5323e31.