metals

Article
Charpy Impact Properties of Hydrogen-Exposed 316L
Stainless Steel at Ambient and
Cryogenic Temperatures
Le Thanh Hung Nguyen , Jae-Sik Hwang , Myung-Sung Kim, Jeong-Hyeon Kim,
Seul-Kee Kim and Jae-Myung Lee *
Department of Naval Architecture and Ocean Engineering, Pusan National University, 30, Jangjeon-Dong,
Geumjeong-Gu, Busan 609-735, Korea; [email protected] (L.T.H.N.); [email protected] (J.S.H.);
[email protected] (M.S.K.); [email protected] (J.H.K.); [email protected] (S.K.K.)
* Correspondence: [email protected]; Tel.: +82-51-510-2342




Received: 14 May 2019; Accepted: 27 May 2019; Published: 29 May 2019


Abstract: 316L stainless steel is a promising material candidate for a hydrogen containment system.
However, when in contact with hydrogen, the material could be degraded by hydrogen embrittlement
(HE). Moreover, the mechanism and the effect of HE on 316L stainless steel have not been clearly
studied. This study investigated the effect of hydrogen exposure on the impact toughness of 316L
stainless steel to understand the relation between hydrogen charging time and fracture toughness at
ambient and cryogenic temperatures. In this study, 316L stainless steel specimens were exposed to
hydrogen in different durations. Charpy V-notch (CVN) impact tests were conducted at ambient and
low temperatures to study the effect of HE on the impact properties and fracture toughness of 316L
stainless steel under the tested temperatures. Hydrogen analysis and scanning electron microscopy
(SEM) were conducted to find the effect of charging time on the hydrogen concentration and surface
morphology, respectively. The result indicated that exposure to hydrogen decreased the absorbed
energy and ductility of 316L stainless steel at all tested temperatures but not much difference was
found among the pre-charging times. Another academic insight is that low temperatures diminished
the absorbed energy by lowering the ductility of 316L stainless steel.

Keywords: cryogenic temperature; hydrogen embrittlement; impact load; charpy impact test

1. Introduction
The Marine Environment Protection Committee (MEPC) of the International Maritime
Organization (IMO) regulations met for its 72nd session with the aim to dramatically reduce the
greenhouse gas emissions from ships by at least 50% by 2050 compared to 2008 [1]. The new regulations
lead to the high demand for new eco-friendly fuels for marine ships and vessels with low greenhouse
gas emissions. Among the alternative energies, liquid hydrogen (LH2 ) has been of great concern
because it has zero carbon dioxide emissions in the exhaust gas and a higher energy-to-weight ratio in
comparison with conventional fuels, like natural gas or gasoline. Despite these advantages, hydrogen
can dissolve into materials and cause hydrogen embrittlement (HE) in hydrogen containers because of
its small size [2]. Furthermore, a low temperature of up to −253 ◦ C of liquid hydrogen could make the
materials used for LH2 vessels become brittle. Therefore, the effect of cryogenic temperature and HE
on the working capability of materials used for hydrogen containers must be understood. Figure 1
illustrates a hydrogen container.

Metals 2019, 9, 625; doi:10.3390/met9060625 www.mdpi.com/journal/metals

Metals 2019,
Metals 9, x625
2019, 9, FOR PEER REVIEW 22of
of 14
14

Figure 1. Schematic diagram of a hydrogen container.

Figure 1. Schematic diagram of a hydrogen container.
For the transportation and storage of LH2 , 316L stainless steel is considered one of the most
attractivethe
For transportation
material candidates andforstorage of LH2, 316L stainless
liquid hydrogen-containing steel because
vessels is considered one resistance
of its high of the most to
attractive material candidates for liquid hydrogen-containing vessels
HE [3] and good mechanical properties at low temperatures [4]. Understanding the effect of cryogenic because of its high resistance to
HE [3] and good
temperature andmechanical
HE on the properties
performance at low temperatures
of 316L stainless [4].steel Understanding
is very important the effect of cryogenic
in selecting 316L
temperature and HE on the performance of 316L stainless steel
stainless steel as a material candidate for containing liquid hydrogen. However, the mechanism of is very important in selecting 316L
HE
stainless steel as asteel
on 316L stainless material
has not candidate
yet beenfor containing
clearly understood liquid[5].hydrogen. However, the mechanism of
HE on 316L stainless
Former steel has notthe
studies investigated yetinfluence
been clearlyof HE understood [5].
on the mechanical properties and microstructure
of 316L stainless steel. Fukuyama et al. (2004) conducted tensilethe
Former studies investigated the influence of HE on testsmechanical
at 10–70 MPa properties
of hydrogen and
microstructure
atmosphere and at ambient temperature for 316L stainless steel. They realized that the impact of
of 316L stainless steel. Fukuyama et al. (2004) conducted tensile tests at 10–70 MPa of
hydrogen atmosphere and at ambient temperature for 316L
hydrogen on the tensile performance of the material was negligible. Hydrogen was only distributed stainless steel. They realized that the
impact
in the thinof hydrogen
outer layer on andthe tensile performance
its concentration was ofnotthe uniform
material along
was negligible. Hydrogenbecause
with the specimen was only of
distributed in the thin outer layer and its concentration was not
its low diffusivity [6]. Kanezaki et al. (2008) cathodically charged 316L stainless steel in a H2 SO4 uniform along with the specimen
because
solution of (pH its=low 3.5)diffusivity
at 27 A/m2[6]. , 50Kanezaki
◦ C for 672eth.al. (2008)
They foundcathodically
that hydrogen charged was316L
onlystainless
distributed steelinin
thea
H 2SO4 solution (pH = 3.5) at 27 A/m2, 50 °C for 672 h. They found that hydrogen was only distributed
thin outer layer with a thickness of approximately 100–200 µm after cathodic charging and the high
in the thinconcentration
hydrogen outer layer with wasa only
thickness of approximately
distributed on the 100 µm 100–200
outer μm layer after
[7]. cathodic
The highcharging and the
nickel content of
high hydrogen concentration was only distributed on the 100
316L stainless steel promotes a better stability of the austenite phase, which plays an important μm outer layer [7]. The high nickel
role
content of 316L
in resistance againststainless steel promotesfracture
hydrogen-assisted a better [8].stability of thelattice
In the cubic austenite phase, which
of materials, plays an
the presence of
important role in resistance against hydrogen-assisted fracture [8].
hydrogen in the matrix causes several changes, such as defects in transformation and phase formation. In the cubic lattice of materials,
the
Forpresence
austenitic ofstainless
hydrogensteels in thelikematrix
316Lcauses several
stainless steel,changes,
HE leads suchtoas thedefects
phaseintransformation
transformationfrom and
phase formation.
austenitic For austenitic
γ (face-centered cubic) tostainless
ε (hexagonalsteelsclose-packed)
like 316L stainless steel, HE leads
and α (body-centered to the
cubic) phase
as proven
transformation from austenitic γ (face-centered cubic) to ε (hexagonal
by several studies [9–11]. The diffusion rate of hydrogen in the γ-austenite phase is lower than that in close-packed) and α (body-
centered cubic) phase;
the martensite as proven by several
therefore, studies
the effect of [9–11].
HE in the Theformer
diffusion rateisoflower.
phase hydrogen in the γ-austenite
Furthermore, HE also
phase is lower than that in the martensite phase; therefore,
changes the surface morphology and microstructure of hydrogen-exposed specimens [12]. the effect of HE in the former phase is
lower. Furthermore, HE also changes the surface morphology
Although some studies have reported the effect of HE on the performance of 316L stainless steel and microstructure of hydrogen-
exposed specimens
in the cryogenic [12].
environment, they are different from those reported by this study. Former tensile tests
Although some studies
were conducted to study the effect have reported
of HE on thetheeffect of HE on properties
mechanical the performance of 316Linstainless
of materials the ambientsteel
in the cryogenic environment, they are different from those reported
and cryogenic temperatures [13–16]. However, fracture toughness is also an important parameter in by this study. Former tensile
tests were the
evaluating conducted
resistance to of
study the effectmaterial
a pre-cracked of HE on underthe an mechanical
applied load. properties
The Charpy of materials in the
V-notch (CVN)
ambient
impact test and[17] cryogenic
is utilized temperatures
to estimate the [13–16].
absorbed However,
energyfracture toughness
during fracture is also
under a highan strain
importantrate.
parameter in evaluating the resistance of a pre-cracked material
This test is widely applied in the industry to measure the material’s fracture toughness because under an applied load. The Charpy
of its
V-notch
low cost,(CVN)simplicity impactandtest [17] is utilized
popularity to estimate
[18]. Former studiesthe absorbed
concluded energy
that during
hydrogen fracturedecreased
exposure under a
high strain rate. This test is widely applied in the industry
the impact toughness of steels [19,20]; however, others found that the effect of hydrogen exposure to measure the material’s fracture
toughness because of its low cost, simplicity and popularity [18].
on the impact toughness of steels was negligible [5,21]. Nevertheless, the effect of HE on the impact Former studies concluded that
hydrogen
behavior of exposure
steels atdecreased
the cryogenic the impact toughness
temperature wasofnot steels [19,20];
clearly however, others
investigated. Thus, found that the
the cryogenic
effect of hydrogen exposure on the impact toughness of steels was negligible [5,21]. Nevertheless, the
Metals 2019, 9, x FOR PEER REVIEW 3 of 14
Metals 2019, 9, 625 3 of 14
effect of HE on the impact behavior of steels at the cryogenic temperature was not clearly
investigated. Thus, the cryogenic impact performance of 316L stainless steel and the effect of HE on
impact
this performance
material must beof 316L stainless
understood steel 316L
to select and the effect of
stainless HEasona this
steel material
candidate formust be understood to
hydrogen-containing
select
vessels.316L stainless steel as a candidate for hydrogen-containing vessels.
This study
This study revealed
revealed thethe effect
effect of
of hydrogen
hydrogen exposure
exposure on on the
the fracture
fracture toughness
toughness of of 316L
316L stainless
stainless
steel, including the impact resistance of 316L stainless steel under low temperature.
steel, including the impact resistance of 316L stainless steel under low temperature. Various Various durations of
cathodic hydrogen charging were used to cause HE in the 316L stainless steel specimens.
durations of cathodic hydrogen charging were used to cause HE in the 316L stainless steel specimens. After charging,
zinc electroplating
After charging, zincwas performed towas
electroplating prevent hydrogen
performed from being
to prevent desorbed
hydrogen fromoutbeing
of thedesorbed
specimens outand
of
maintain HE. The CVN impact test at the ambient and low temperatures
the specimens and maintain HE. The CVN impact test at the ambient and low temperatures was was conducted to determine
the fractureto
conducted toughness
determine andthethefracture
ductilitytoughness
of the specimens
and theafter beingof
ductility degraded by HE. Meanwhile,
the specimens after being
the hydrogen amount in each case was measured to study the
degraded by HE. Meanwhile, the hydrogen amount in each case was measured to study relationship between the hydrogen
the
concentration and the absorbed energy of the specimens. The microstructure
relationship between the hydrogen concentration and the absorbed energy of the specimens. of the fracture surfaces
The
was investigatedofbythe
microstructure a scanning
fractureelectron
surfacesmicroscope to understand
was investigated by a the characteristic
scanning of the
electron fracture and
microscope to
ascertain the CVN impact test result.
understand the characteristic of the fracture and ascertain the CVN impact test result.

2. Materials and Methods

2. Materials and Methods
2.1. Material
2.1. Material
The 316L stainless steel exhibited high resistance to HE, showing excellent mechanical performance
The 316L stainless steel exhibited high resistance to HE, showing excellent mechanical
at low temperatures. This study used 316L stainless steel made based on the Japanese standard. Table 1
performance at low temperatures. This study used 316L stainless steel made based on the Japanese
shows the composition of 316L stainless steel.
standard. Table 1 shows the composition of 316L stainless steel.
Table 1. Composition of 316L stainless steel.
Table 1. Composition of 316L stainless steel.
Chemical Elements (wt. %)
Specimen Chemical Elements (wt. %)
Specimen C Si S P Mn Mo Ni Cr Fe
C Si S P Mn Mo Ni Cr Fe
316L 316L 0.020 0.020 0.56 0.56 0.003
0.003 0.028
0.028 1.33
1.33 2.1
2.1 10.19
10.19 16.4 16.4 Bal. Bal.

Specimens
2.2. Specimens
specimens were
The specimens were made based on ASTM E23—16b: Standard Standard Test
Test Methods
Methods for Notched
Notched Bar
Impact Testing of Metallic Materials [17]. V-notched specimens are widely used to test the absorbed
because they
energy of metals because they are easy to prepare and the CVN impact test results can be achieved
quickly and cheaply. Each specimen had a specific dimension of 55 mm mm × × 10 mm ×× 10
10 mm
mm in
in length,
length,
width and
and thickness, respectively. The V-notch was 2 mm in depth and 45 degree in V-angle.notch
thickness, respectively. The V-notch was 2 mm in depth and 45 degree in V-angle. The The
radiusradius
notch was 0.25
wasmm.
0.25Figure 2 illustrates
mm. Figure the dimension
2 illustrates of eachofV-notched
the dimension specimen.
each V-notched specimen.

Figure 2.
Figure (a) Photograph
2. (a) Photograph and
and (b)
(b) dimension
dimension of
of the
the V-notched
V-notched specimen
specimen (unit: mm).
(unit: mm).
Metals 2019, 9, 625 4 of 14
Metals 2019, 9, x FOR PEER REVIEW 4 of 14

2.3. Cathodic Hydrogen Charging and Zn Electroplating

2.3. Cathodic Hydrogen Charging and Zn Electroplating
The two predominant methods used to generate HE in specimens are electrochemical charging
The two predominant methods used to generate HE in specimens are electrochemical charging
and hydrogen gas charging. Hydrogen gas charging involves exposing the specimen to hydrogen
and hydrogen gas charging. Hydrogen gas charging involves exposing the specimen to hydrogen
atmosphere, while electrochemical hydrogen charging, or cathodic hydrogen charging, is a method by
atmosphere, while electrochemical hydrogen charging, or cathodic hydrogen charging, is a method
which a specimen is immersed in an electrolytic solution. Electricity is then used to generate water
by which a specimen is immersed in an electrolytic solution. Electricity is then used to generate water
electrolysis. Cathodic hydrogen charging was used herein to generate HE in the 316L stainless steel
electrolysis. Cathodic hydrogen charging was used herein to generate HE in the 316L stainless steel
specimens because this method was simple to perform. Cathodic hydrogen charging was conducted
specimens because this method was simple to perform. Cathodic hydrogen charging was conducted
based on the ISO 16573-2015: Steel—Measurement method for the evaluation of HE resistance of
based on the ISO 16573-2015: Steel—Measurement method for the evaluation of HE resistance of
high-strengthsteels
high-strength steels[22].
[22].The
Thegenerated
generatedhydrogen
hydrogendiffused
diffusedintointothe
thespecimen
specimensurface,
surface,which
whichplayed
played
a role as a cathode. A platinum mesh was used as an anode. The notched specimens were totally
a role as a cathode. A platinum mesh was used as an anode. The notched specimens were totally
immersedin
immersed inthe
theelectrolyte
electrolytesolution
solutioncomposed
composedofof3% 3%NaCl + 0.3%
NaCl+ 0.3% NH NH 4 SCN. The NaCl molecules
4SCN. The NaCl molecules in
in the H2 O solvent were dissolved to+ Na and + −
Cl ions, which acted as electrolytes in the solution,
the H2O solvent were dissolved to Na and Cl− ions, which acted as electrolytes in the solution, while
+ ions from recombination to H during hydrogen charging [23]. The
while NH 4 SCN prevented the H
NH4SCN prevented the H+ ions from recombination to H2 during 2 hydrogen charging [23]. The
charging’s current density was kept constant at 20 A/m 2 using the potentiostat mode of a WBCS3000S
charging’s current density was kept constant at 20 A/m using the potentiostat mode of a WBCS3000S
2
StandardType
Standard Type Battery
BatteryCycler
Cycler(WonATech,
(WonATech, Seoul,
Seoul, Korea).
Korea). TheThe charging
charging durations
durations were
were 12
12 h,
h, 24
24hhand
and
48 h. Hydrogen was liberated at the cathode surface (specimen) and might diffuse
48 h. Hydrogen was liberated at the cathode surface (specimen) and might diffuse into the specimen.into the specimen.
The amount
The amount of ofliberated
liberatedhydrogen
hydrogen depended
dependedon the
on current density.
the current A larger
density. A current
larger density
current liberated
density
more hydrogen
liberated on the specimen
more hydrogen surface. Figure
on the specimen surface.3 illustrates the schematic
Figure 3 illustrates diagram of
the schematic hydrogen
diagram of
charging for
hydrogen the V-notched
charging specimens.specimens.
for the V-notched

Figure 3. Schematic diagram of the cathodic hydrogen charging.

Figure 3. Schematic diagram of the cathodic hydrogen charging.
After the hydrogen exposure, each specimen was slightly polished by a grit sanding sponge after
being charged
After to removeexposure,
the hydrogen the little dirt
eachonspecimen
the specimen
was surface. Next, zincby
slightly polished electroplating
a grit sanding was conducted
sponge after
to prevent hydrogen from being desorbed. The thin Zn layer acted as a
being charged to remove the little dirt on the specimen surface. Next, zinc electroplatingbarrier that prevents hydrogen
was
from diffusing
conducted out during
to prevent the loading
hydrogen fromtest. Thedesorbed.
being schematicThe diagram
thin of
Znthe electroplating
layer acted as awas the same
barrier that
as the hydrogen charging, except for the current density and the plating
prevents hydrogen from diffusing out during the loading test. The schematic diagram of thesolution. In this procedure, the
applied current density was 300 A/m 2 . The plating solution included Zn2+ ions, which were referenced
electroplating was the same as the hydrogen charging, except for the current density and the plating
from the In
solution. ISO 16573-2015
this procedure, standard [22]. current
the applied Each specimen
densitywaswasinitially
300 A/melectroplated
2. The plating for approximately
solution included8
min.
Zn The which
2+ ions, specimen
were was then rotated
referenced fromand thecontinued to be electroplated
ISO 16573-2015 standard [22].again
Each in approximately
specimen 8 min
was initially
to ensure that the Zn layer homogeneously covered the specimen. The
electroplated for approximately 8 min. The specimen was then rotated and continued to be electroplating times could be
slightly adjusted
electroplated to make
again sure that the8Zn
in approximately layers
min sufficiently
to ensure covered
that the the specimens
Zn layer homogeneouslywith homogeneous
covered the
thicknesses. Figure 4 illustrates the specimens before (a) and after (b) electroplating.
specimen. The electroplating times could be slightly adjusted to make sure that the Zn layers
sufficiently covered the specimens with homogeneous thicknesses. Figure 4 illustrates the specimens
before (a) and after (b) electroplating.
Metals 2019, 9, 625 5 of 14
Metals 2019, 9, x FOR PEER REVIEW 5 of 14

Figure 4. Charpy
Charpy V-notch
V-notch (CVN) impact specimens before (a) and after (b) electroplating.

2.4. Hydrogen
2.4. Hydrogen Concentration
Concentration Measurement
Measurement
After being
After being hydrogen
hydrogen pre-charged,
pre-charged, the
the hydrogen
hydrogen concentration
concentration on
on the
the surface of each
surface of each specimen
specimen
was
was measured
measured by by aa ONH-2000
ONH-2000hydrogen
hydrogenanalyzer
analyzer(ELTRA
(ELTRAGmbH,
GmbH,Haan,
Haan,Germany)
Germany)for forthe
the1212
h,h,
2424h
and 48 h charged and uncharged specimens. For this measurement, each sample has 1
h and 48 h charged and uncharged specimens. For this measurement, each sample has 1 g in weight g in weight was
cut from
was the specimen
cut from the specimensurface. The The
surface. samples werewere
samples thenthen
investigated for hydrogen
investigated for hydrogenconcentration
concentration for
approximately
for approximately 2.5 2.5
minmininside the the
inside dual range
dual thermal
range conductivity
thermal cellcell
conductivity of the hydrogen
of the hydrogen analyzer. At
analyzer.
least five samples were analyzed for each duration of hydrogen exposure to ascertain
At least five samples were analyzed for each duration of hydrogen exposure to ascertain the the repeatability
of the hydrogen
repeatability concentration
of the results. The hydrogen
hydrogen concentration results. concentration
The hydrogenwas used to unveil
concentration wasthe usedrelationship
to unveil
between the absorbed energy and the hydrogen concentration for each CVN specimen.
the relationship between the absorbed energy and the hydrogen concentration for each CVN
specimen.
2.5. CVN Impact Test
2.5. CVN
AfterImpact
beingTest
electroplated, the specimens were prepared by controlling the desired temperature
conditions. Then, the Charpy impact
After being electroplated, tests were
the specimens conducted.
were preparedThe CVN impact
by controlling thetest demonstrated
desired the
temperature
absorbed energy during the material fracture, which also indicated work needed to make
conditions. Then, the Charpy impact tests were conducted. The CVN impact test demonstrated the the material
fracture atenergy
absorbed the experimental
during the temperature of thewhich
material fracture, CVN also
impact test. The
indicated absorbed
work neededenergy was
to make calculated
the material
as follows:
fracture at the experimental temperature of the CVN impact test. The absorbed energy was calculated
as follows: ECVN = MR(cosβ − cosα)g, (1)

where ECVN is the absorbed energy; M E isCVN

the= mass of the
MR(cosβ hammer; R is the specimen length; β is the
− cosα)g, (1)
angle after the impact; α is the falling angle before the impact; and g is the gravitational acceleration.
where
The ECVN is the
absorbed absorbed
energy of theenergy;
specimens M iswasthe recorded
mass of the hammer;the
to estimate R istoughness
the specimen length; β of
and ductility is the
the
angle
impactafter
tests.the impact;
Ductile α is the
fracture hasfalling
high angle
absorbed before the impact;
energy in the CVN and impact
g is thetest
gravitational acceleration.
as well as better impact
The absorbed energy of the specimens was recorded to estimate the
toughness, while brittle fracture has low absorbed energy in the CVN impact test. Therefore,toughness and ductility of the
the
impact tests. Ductile
ductile fracture fracture has
was preferred high applications.
in most absorbed energy Aside in from
the CVN impact testused
the specimens as well as better
for the CVN
impact toughness, while brittle fracture has low absorbed energy in the CVN
impact test at the ambient temperature, the others were pre-cooled in the desired low temperature forimpact test. Therefore,
the ductile
40–50 min to fracture
ensurewastheirpreferred
thermal in most applications.
equilibrium. To lowerAside from the specimens
the temperature to −50 ◦ Cused
andfor the◦CVN
−125 C for
impact test at the ambient temperature, the others were pre-cooled in the desired
the low-temperature impact test, liquefied nitrogen (LN2 ) was injected to an insulated chamber and low temperature
for
the 40–50 minrate
inlet flow to ensure their was
of nitrogen thermal equilibrium.
automatically To lower
adjusted the temperature
to maintain the desiredto −50 °C and −125
temperature °C
by the
for the low-temperature impact test, liquefied nitrogen ◦ (LN 2) was injected to an insulated chamber
two thermocouples connected to a computer. For −196 C, the specimens were directly immersed to
and the
liquid inlet flow
nitrogen rate of nitrogen
at atmospheric was in
pressure automatically adjusted
another insulated to maintain
chamber. the desiredfor
After pre-cooling temperature
40–50 min,
by the two thermocouples connected to a computer. For −196 °C, the
the CVN impact test was conducted within 5 s for each specimen since the time that each specimen specimens were directly
immersed
was broughtto liquid
out of nitrogen at atmospheric
the insulated chamber [17]. pressure
Each incaseanother insulatedatchamber.
was repeated least threeAfter
timespre-cooling
to ensure
for 40–50 min, the CVN impact test was conducted within 5 s for each
the reliability of the testing results. Figure 5 presents the experimental procedure of the CVNspecimen since the time that
impact
each specimen was brought out of the insulated chamber [17]. Each case was repeated at least three
times to ensure the reliability of the testing results. Figure 5 presents the experimental procedure of
Metals 2019, 9, 625 6 of 14
Metals 2019, 9, x FOR PEER REVIEW 6 of 14

test.CVN
the Tableimpact
2 illustrates the experimental
test. Table 2 illustrates scenario of the hydrogen
the experimental scenariopre-charging, zinc pre-charging,
of the hydrogen electroplating zinc
and
CVN impact tests.
electroplating and CVN impact tests.

Figure 5. Experimental procedure of the CVN impact test.

Table 2. Experimental scenario.

Table 2. Experimental scenario.
Electroplating Temperature of
Charging Charging
Current Current
ChargingCharging
Electroplating Current Temperature of CVN
No No 2) Current Density CVN Impact Test
Density
Density (A/m ) (A/m Duration
Duration (h) (h) Density (A/m
(A/m) ) Impact Test (°C)
2 2 2 (◦ C)
1 1
0 0
0 0
0 0
2 2 12 12
2525
3 3 20 20 24 24 300 300
4 4 48 48
5 5 0 0 0 0 0 0
6 12
6 12 −50
7 20 24 300 −50
7 8 20 24 48 300
8 48
9 0 0 0
9 10 0 0 12 0
−125
10 11 20 12 24 300
12 48 −125
11 20 24 300
12 13 0 48 0 0
13 14 0 0 12 0 −196
15 20 24 300
14 12
16 48 −196
15 20 24 300
16 48
2.6. Scanning Electron Microscopy (SEM)
For eachElectron
2.6. Scanning case of the CVN impact
Microscopy (SEM)test, a Supra 25 scanning electron microscope (Carl Zeiss AG,
Oberkochen, Germany) was used to analyze the microstructure in the middle of the fracture surfaces
For each case of the CVN impact test, a Supra 25 scanning electron microscope (Carl Zeiss AG,
after the impact test. This step focuses on the effect of the hydrogen charging time on the surface
Oberkochen, Germany) was used to analyze the microstructure in the middle of the fracture surfaces
morphology (e.g., tortuousness, depth and size of dimples) on the specimen’s fracture surface after the
after the impact test. This step focuses on the effect of the hydrogen charging time on the surface
CVN impact test. ImageJ software was employed to measure the dimple size. The absorbed energy and
morphology (e.g., tortuousness, depth and size of dimples) on the specimen’s fracture surface after
microstructure combination could unveil the impact properties of the specimens. The microstructure
the CVN impact test. ImageJ software was employed to measure the dimple size. The absorbed
energy and microstructure combination could unveil the impact properties of the specimens. The
microstructure of the ductile fracture mainly has a tortuous appearance with large and deep dimples.
Metals 2019,
Metals 2019,9,
9,625
x FOR PEER REVIEW 77 of
of 14
14
Metals 2019, 9, x FOR PEER REVIEW 7 of 14
In contrast, the brittle fracture microstructure mainly contains small spherical dimples and cleavage
In contrast,
facets.
of the ductilethe brittlemainly
fracture fracturehas
microstructure mainly contains
a tortuous appearance small
with large spherical
and dimples
deep dimples. Inand cleavage
contrast, the
facets.
brittle fracture microstructure mainly contains small spherical dimples and cleavage facets.
3. Results and Discussion
3. Results
3. Results and Discussion
and Discussion
3.1. Hydrogen Concentration in the Specimens
3.1. Hydrogen Concentration in the Specimens
Figure 6 illustrates the hydrogen concentration at the surface of the uncharged and 12 h, 24 h
and Figure illustratesspecimens.
6 illustrates
48 h pre-charged the hydrogen Thisconcentration
chart shows that at the
thesurface
hydrogen of the uncharged and
concentration in the h, 24 hh
12charged
and 48 h
h pre-charged
pre-charged specimens.
specimens. This chart shows that the hydrogen
specimens was higher than that in the uncharged specimens. The average hydrogen concentration in concentration in the
the charged
charged
specimens
specimens
the uncharged was higher
samples than
than
wasthat3.52inwt. uncharged
the ppm,
uncharged specimens.
whichspecimens. The average
was approximate to thathydrogen concentration
in the previous in
research
the
the uncharged
uncharged samples
samples was
was 3.52
3.52 wt.wt. ppm,
ppm, which
which waswas approximate
approximate to to
that
[24]. In comparison with the non-exposed samples, the results for the 12 h, 24 h and 48 h charged that
in in
the the previous
previous research
research [24].
[24].
In
samplesIn comparison
comparison with
are 5.79, with
the
10.1 the non-exposed
non-exposed
and 8.74 samples,
wt. ppm, samples,
which the results
thecorrespond
results for the for
12 h,the
to 164%, 24 12 h,and
h and
287% 2448h248%,
hand 48
charged h samples
charged
respectively.
samples
are
The5.79, are and
10.1
hydrogen 5.79, 10.1wt.
8.74
concentrationandppm,8.74 wt. ppm,
which which to
correspond
dramatically increasedcorrespond
164%, 287%
and reached to and
164%, 287%
248%,
a saturation and
point248%,
respectively.
at 24Thehrespectively.
ofhydrogen
charging.
The hydrogen
concentration concentration dramatically increased and reached
After 24 h of charging, the hydrogen concentration remained relatively closed to theAfter
dramatically increased and reached a saturation a saturation
point at 24 h point
of at 24
charging. h of charging.
24 h
saturated
After
of 24
charging, h of
the charging,
hydrogen the hydrogen
concentration concentration
remained remained
relatively closed relatively
to the
concentration of hydrogen. Besides, the errors of hydrogen concentration were relatively large. This closed
saturated to the saturated
concentration of
concentration
hydrogen.
result confirmed of hydrogen.
Besides, thethe errors
high Besides,
resistance the
of hydrogen errors
of 316L of hydrogen
concentration
stainless were
steel concentration
to relatively
HE because were
large. relatively
This
hydrogen result
hardly large. This
confirmed
diffused
result
the highconfirmed
resistance theof high
316L resistance
stainless of 316L
steel to stainless
HE because steel to
hydrogenHE
into this material and was homogeneously distributed on the thin outer layer of the hydrogen-because
hardly hydrogen
diffused hardly
into this diffused
material
into
and
exposedthis homogeneously
was material and
specimens. Thewas nexthomogeneously
distributed
sectionon explains
the thindistributed
outer layeronof
the effect ofthethethin
the outer layer
hydrogen-exposed
hydrogen ofspecimens.
charging the
time hydrogen-
andThe
the
exposed
next sectionspecimens.
explains The
the next
effect section
of the explains
hydrogen the
charging effect
time
temperature of the CVN impact test on the absorbed energy of 316L stainless steel. of the
and hydrogen
the temperature charging
of the time
CVN and the
impact
temperature
test of the CVN
on the absorbed energyimpact teststainless
of 316L on the absorbed
steel. energy of 316L stainless steel.
12
12
(ppm)

10
(ppm)

10
8
concentration

8
concentration

6
6
4
Hydrogen

4
Hydrogen

2
2
0
0 12 24 36 48
0
0 12 Charging
24 time (h) 36 48
Charging time (h)
Figure 6. Hydrogen concentration at the uncharged and 12 h, 24 h and 48 h charged specimens.
Figure
Figure 6. Hydrogen
6. Hydrogen concentration
concentration at the
at the uncharged
uncharged andand
12 12 h, 24
h, 24 h and
h and 48 48 h charged
h charged specimens.
specimens.
3.2. CVN
3.2. CVN Impact
Impact Absorbed
Absorbed Energy
Energy
3.2. CVN Impact Absorbed Energy
In aa CVN
In CVN impact
impact test,
test, the
the ductile
ductile fracture
fracture has
has higher
higher absorbed
absorbed energy
energy and
and higher
higher plastic
plastic
In a CVN
deformation thanimpact
the test,fracture.
brittle the ductile fracture
Figure 7 hasthehigher
shows absorbed energy and
temperature-dependent higher
absorbed plastic
energy of
deformation than the brittle fracture. Figure 7 shows the temperature-dependent absorbed energy of
deformation
the CVN thantest
impact thefor
brittle
the fracture.
12 h, 24 hFigure
and 48 7hshows the temperature-dependent
hydrogen charged and uncharged absorbed
316L energy
stainless of
steel
the CVN impact test for the 12 h, 24 h and 48 h hydrogen charged and uncharged 316L stainless steel
the
at 25
25CVN
◦°C
impact test
(ambient for the 12 h,−50
temperature), 24 h and 48 h◦ hydrogen charged and uncharged 316L stainless steel
at C (ambient temperature), −50 ◦°C,
C, −125
−125 °C Cand
and−196
−196°C,◦ C,respectively.
respectively.
at 25 °C (ambient temperature), −50 °C, −125 °C and −196 °C, respectively.
350
350
300
300
250
(J) (J)

250
energy

200
energy

200
150
Absorbed

150 0h
Absorbed

100 12h
0h
100 24h
12h
50 48h
24h
50 48h
0
-200 -150 -100 -50 0 50
0
-200 -150 Temperature
-100 (o-50
C) 0 50
Temperature (oC)
Figure 7. Temperature-dependent absorbed energy behavior of 316L stainless steel for uncharged, 12
hFigure 7. Temperature-dependent
charged, absorbedspecimens
24 h charged and 48 h charged energy behavior of 316L
(the error barsstainless
indicatesteel for uncharged,
the incertitude 12
of the
Figure 7.
h charged,Temperature-dependent absorbed energy behavior of 316L stainless steel for uncharged,
24 h charged and 48 h charged specimens (the error bars indicate the incertitude of the12
absorbed energy).
habsorbed
charged,energy).
24 h charged and 48 h charged specimens (the error bars indicate the incertitude of the
absorbed energy).
Metals 2019, 9, 625 8 of 14
Metals 2019, 9, x FOR PEER REVIEW 8 of 14

Overall,the
Overall, thetested
tested316L
316Lstainless
stainlesssteel’s
steel’sabsorbed
absorbedenergy
energygradually
graduallydecreased
decreasedwhen whendecreasing
decreasing
thetemperature
the temperatureofofthe theCVN
CVNimpact
impacttest.
test.Specifically,
Specifically,inincomparison
comparisonwithwith2525◦°C, theaverage
C, the averageabsorbed
absorbed
energy of the uncharged specimens at −50 ◦ °C, −125 ◦ °C and −196
energy of the uncharged specimens at −50 C, −125 C and −196 C are 76.0%, 69.4% and 61.0% ◦°C are 76.0%, 69.4% and 61.0%
respectively.These
respectively. Theseratios
ratiosfor
forthe
the12
12h,h,24
24hhandand48 48hhcharged
chargedspecimens
specimenswerewereapproximate
approximatetotothosethoseofof
theuncharged
the unchargedspecimens.
specimens.Therefore,
Therefore,the theductility
ductilityof ofimpact
impacttests
testsdecreased
decreasedgradually
graduallyfrom fromambient
ambient
temperatureto
temperature −196◦°C.
to−196 The decrease
C. The decrease in inthe
theabsorbed
absorbed energy
energyas aswell
wellasasthe
theductility
ductilitywaswascaused
causedby by
loweringthe
lowering thetemperature
temperatureofofimpactimpacttests,
tests,which
whichresulted
resultedininthethemore
morebrittle
brittlefracture
fractureofof316L
316Lstainless
stainless
steel [25,26].
steel [25,26].
Figure88shows
Figure showsthe thecharging
chargingtime-dependent
time-dependentabsorbed absorbedenergy
energybehavior
behaviorofof316L
316Lstainless
stainlesssteel
steelatat
2525◦ C
°C(ambient
(ambienttemperature),
temperature), −50 ◦ C,
−50 °C, −125
−125 ◦ C°Candand
−196 ◦ C,°C,
−196 respectively.
respectively. At the
At the same same temperature
temperature of theof
CVN impact test, the absorbed energy of the pre-charged specimens relatively dropped in comparisonin
the CVN impact test, the absorbed energy of the pre-charged specimens relatively dropped
tocomparison to that of the
that of the uncharged uncharged
specimens. specimens.
For For the test
the CVN impact CVN impact test
conducted conducted
at the ambientattemperature,
the ambient
temperature,
the drop in the the drop energies
absorbed in the absorbed
at the 12 h, energies
24 h and at48the 12 h, 24specimens
h charged h and 48corresponded
h charged specimens
to 16.6%,
corresponded to 16.6%, 14.2% and 12.6%, respectively, when compared
14.2% and 12.6%, respectively, when compared to those in the uncharged specimens. This result also to those in the uncharged
specimens.
indicates thatThis result also
hydrogen indicatesslightly
pre-charging that hydrogen
decreased pre-charging slightly
the ductility decreased
of the notched the ductility
specimens at theof
the notched specimens
ambient temperature. at the ambient temperature.

400
25 oC
350 - 50 oC
- 125 oC
300
- 196 oC
Absorbed Energy (J)

250

200

150

100

50

0
0 12 24 36 48
Charging Time (h)

◦ (ambient
Figure8.8.Charging
Figure Chargingtime-dependent
time-dependentabsorbed
absorbedenergy
energybehavior
behaviorofof316L
316Lstainless
stainlesssteel
steelatat2525 C
°C (ambient
◦ ◦ ◦
temperature),−50
temperature), C, −125
−50 °C, −125 °C and −196
C and −196 °CC (the
(the error
error bars
bars indicate
indicatethe
theincertitude
incertitudeofofthetheabsorbed
absorbed
energy).
energy).

For the CVN impact test conducted at −50 ◦ C, the absorbed energy for the pre-charged specimens
For the CVN impact test conducted at −50 °C, the absorbed energy for the pre-charged specimens
was almost alike to the uncharged specimens. Specifically, the dropped ratio for the CVN impact test
was almost alike to the uncharged specimens. Specifically, the dropped ratio for the CVN impact test
for the 12 h and 48 h pre-charged specimens is approximately 0.23% and 1.7% respectively, compared
for the 12 h and 48 h pre-charged specimens is approximately 0.23% and 1.7% respectively, compared
to the uncharged specimens. The absorbed energy for the 24 h pre-charged specimen increased by
to the uncharged specimens. The absorbed energy for the 24 h pre-charged specimen increased by
0.45%. Therefore, the effect of hydrogen pre-charging at −50 ◦ C was negligible.
0.45%. Therefore, the effect of hydrogen pre-charging at −50 °C was negligible.
At −125 ◦ C, the absorbed energy also exhibited a slight decrease with the pre-charging time
At −125 °C, the absorbed energy also exhibited a slight decrease with the pre-charging time of
of hydrogen. The drop in the absorbed energy is 7.9%, 11.3% and 9.7% for the 12 h, 24 h and 48 h
hydrogen. The drop in the absorbed energy is 7.9%, 11.3% and 9.7% for the 12 h, 24 h and 48 h pre-
pre-charged specimens, respectively. Finally, at −196 ◦ C, the drop in the absorbed energy for the 12 h,
charged specimens, respectively. Finally, at −196 °C, the drop in the absorbed energy for the 12 h, 24
24 h and 48 h pre-charged specimens is 12.9%, 9.2% and 10.0%, respectively. Overall, most of the
h and 48 h pre-charged specimens is 12.9%, 9.2% and 10.0%, respectively. Overall, most of the
hydrogen pre-charged specimens exhibited a drop in the absorbed energy, indicating that hydrogen
hydrogen pre-charged specimens exhibited a drop in the absorbed energy, indicating that hydrogen
pre-charging decreased the ductility of 316L stainless steel in the CVN impact tests at the ambient
pre-charging decreased the ductility of 316L stainless steel in the CVN impact tests at the ambient
temperature, −50 ◦ C, −125 ◦ C and −196 ◦ C. Former studies also revealed that hydrogen exposure
temperature, −50 °C, −125 °C and −196 °C. Former studies also revealed that hydrogen exposure
changes the fracture mode of materials from ductile to brittle [2,27] and therefore decreased the
changes the fracture mode of materials from ductile to brittle [2,27] and therefore decreased the
absorbed energy in the CVN impact test of metallic materials. In Figures 7 and 8, the exposure to
absorbed energy in the CVN impact test of metallic materials. In Figure 7 and Figure 8, the exposure
hydrogen decreased the absorbed energy in the CVN impact tests but the drop in absorbed energy
to hydrogen decreased the absorbed energy in the CVN impact tests but the drop in absorbed energy
remained
remainedstable
stablefrom
from2424hhtoto48
48hhofofcharging,
charging,which
whichreflected
reflectedaasimilar
similarbehavior
behaviorwith
withhydrogen
hydrogen
concentration in 24 h charged and 48 h charged specimens. A former study also revealed that when
concentration in 24 h charged and 48 h charged specimens. A former study also revealed that when
increasing the hydrogen outgassing time to low carbon stainless steel, not much difference in
hardness of the specimens was observed [28].
Metals 2019, 9, 625 9 of 14

increasing the hydrogen outgassing time to low carbon stainless steel, not much difference in hardness
Metals 2019, 9, x FOR PEER REVIEW 9 of 14
of the specimens was observed [28].

3.3.
3.3. Fracture
Fracture Surface
Surface Morphology
Morphology
Figure
Figure 99 illustrates
illustratesthe thespecimens
specimensunderunderCVNCVNimpact
impacttests. In In
tests. Figure
Figure 9, cracks were
9, cracks werefound on the
found on
V-notched
the V-notched specimen
specimenafterafter
the the
CVN CVNimpact test.test.
impact TheThe
appearance
appearance of the cracks
of the mostly
cracks mostlydepended
depended on
the
on the temperature of the impact tests. The presence of plastic deformation increased in the order of
temperature of the impact tests. The presence of plastic deformation increased in the order of
−196
−196 °C,
◦ C,−125
−125°C,
◦ C,−50
−50°C◦ Cand
andambient
ambient temperature.
temperature. For CVN
For CVN impact
impacttests conducted
tests conducted at at
−196 °C,◦all
−196 V-
C, all
notched specimens
V-notched specimens were
wereseparated
separatedintointo2 2pieces
piecesandandshowed
showed veryvery little plastic deformation.
little plastic deformation. The The
specimens in CVN impact tests conducted at −125 °C
◦ and −50 °C
◦ were partly
specimens in CVN impact tests conducted at −125 C and −50 C were partly separated by the impact separated by the impact
tests
tests and
and showed
showed relatively little plastic
relatively little plastic deformation.
deformation. Meanwhile,
Meanwhile, impact
impact tests
tests conducted
conducted at at ambient
ambient
temperature showed extensive plastic deformation and all specimens were not
temperature showed extensive plastic deformation and all specimens were not separated. Further, the separated. Further,
the differences
differences in appearance
in the the appearance of impacted
of impacted specimens
specimens according
according to charging
to charging timestimes
werewere not clear.
not clear.

Figure 9.
Figure Photograph of
9. Photograph of tested
tested specimens.
specimens.

Scanning electron
Scanning electronmicroscopy
microscopy waswas conducted
conducted afterafter
the CVN impact
the CVN test to test
impact investigate the fracture
to investigate the
surface of the specimens. The brittle fracture has little plastic deformation
fracture surface of the specimens. The brittle fracture has little plastic deformation and mainly and mainly contains small
sphericalsmall
contains dimples, flat facets
spherical dimples,and flat
cleavage
facets facets on the fracture
and cleavage facets onsurfaces. For surfaces.
the fracture ductile fracture, the
For ductile
fracture surface mainly includes a tortuous appearance with deep and
fracture, the fracture surface mainly includes a tortuous appearance with deep and large dimples. large dimples. The ductile
fracture
The has fracture
ductile more plastic deformation
has more than the brittle
plastic deformation thanfracture. Therefore,
the brittle fracture.theTherefore,
absorbed energy in the
the absorbed
CVN impact
energy in thetest
CVN of impact
ductile test
fracture was higher
of ductile fracturethanwas that of brittle
higher than fracture.
that of brittle fracture.
Figure 10 shows the fracture surfaces for the impact
Figure 10 shows the fracture surfaces for the impact test conducted test conducted at ambient
at ambient temperature
temperature for the
for
(a) uncharged, (b) 12 h charged, (c) 24 h charged and (d) 48 h charged specimens
the (a) uncharged, (b) 12 h charged, (c) 24 h charged and (d) 48 h charged specimens after the CVN after the CVN impact
test. Astest.
impact shown As in this figure,
shown in thisthe fracture
figure, thesurfaces
fracturehave a tortuous
surfaces have aappearance and mainlyand
tortuous appearance covered
mainly by
large dimples. The cleavage fracture was not observed in the SEM
covered by large dimples. The cleavage fracture was not observed in the SEM images at ambient images at ambient temperature
and the largeand
temperature dimples
the largeweredimples
relatively deep.
were Those were
relatively deep.the signs
Those of plastic
were deformation.
the signs Therefore, at
of plastic deformation.
ambient temperature, the type of impact fracture was mainly ductile.
Therefore, at ambient temperature, the type of impact fracture was mainly ductile. For different For different charging times,
the morphology of the uncharged specimen was the roughest
charging times, the morphology of the uncharged specimen was the roughest appearance byappearance by included many tortuous
and large many
included dimples, the depth
tortuous and of those
large dimples
dimples, was
the bigger
depth than the
of those other charged
dimples was bigger specimens,
than thewhichother
confirms the negligible drop in absorbed energy of charged specimens
charged specimens, which confirms the negligible drop in absorbed energy of charged specimens at ambient temperature. Besides,at
the dimples on the fracture surface of 12 h charged specimen were
ambient temperature. Besides, the dimples on the fracture surface of 12 h charged specimen wererelatively smaller and shallower
than the other
relatively specimens
smaller at ambient
and shallower thantemperature, confirmsatthat
the other specimens the absorbed
ambient energy
temperature, of 12 h charged
confirms that the
specimen was the smallest among the charging time. Therefore, the
absorbed energy of 12 h charged specimen was the smallest among the charging time. Therefore, effect charging times at 12 htheon
the surface
effect morphology
charging times at 12 as hwell as deformation
on the of CVN impact
surface morphology as welltests at ambient temperature
as deformation of CVN impact wastests
the
largest, followed by 24 h and 48 h charging time. Besides, there were negligible
at ambient temperature was the largest, followed by 24 h and 48 h charging time. Besides, there were differences between the
deformation as well as absorbed energy between 24 h charged specimen
negligible differences between the deformation as well as absorbed energy between 24 h charged and 48 h charged specimen at
ambient temperature.
specimen and 48 h charged specimen at ambient temperature.
Metals 2019, 9, 625 10 of 14
Metals 2019, 9, x FOR PEER REVIEW 10 of 14
Metals 2019, 9, x FOR PEER REVIEW 10 of 14

Figure
Figure 10.10.Scanning
Scanningelectron
electronmicroscope
microscope (SEM)
(SEM) images
imagesobtained
obtainedfrom
fromthetheimpact
impacttest at at
test thethe
ambient
ambient
Figure 10.
temperature: Scanning
(a) electron
uncharged, microscope
(b) 12 h (SEM)
charged, (c)images
24 h obtained
charged andfrom
(d) the
48 h impact
charged.
temperature: (a) uncharged, (b) 12 h charged, (c) 24 h charged and (d) 48 h charged. test at the ambient
temperature: (a) uncharged, (b) 12 h charged, (c) 24 h charged and (d) 48 h charged.
Figure1111shows
Figure showsthe themicrostructure
microstructure analysis
analysis of of the
the fracture
fracturesurface
surfaceininthe theCVN
CVN impact
impact testtest
at at
−50◦ Figure
°C. The 11 shows
fracture the
surfaces microstructure
at −50 °C
◦ areanalysis
primarily of the fracture
covered by surface
relatively
−50 C. The fracture surfaces at −50 C are primarily covered by relatively large dimples. However, in the
large CVN
dimples. impact test
However, at
−50
thethe °C. The
fracture
fracture fracture surfaces
surfaceatatthis
surface at −50
thistemperature °C
temperature was are primarily
was less covered
less circuitous
circuitousthan by
thanthat relatively
thatininthe large
theambient
ambientdimples.
temperature,However,
temperature, thethe
the fracture
dimples in surface
Figure at
11 this
are temperature
relatively was
smaller less
and circuitous
shallower than
than that
the in the ambient
dimples
dimples in Figure 11 are relatively smaller and shallower than the dimples in Figure 10 and an increased in temperature,
Figure 10 and the
an
dimples in
increased Figure of
presence 11thearesmall
relatively smaller
spherical dimplesandandshallower
flat than
facets was the dimplesThus,
observed. in Figure
this 10 and an
appearance
presence of the small spherical dimples and flat facets was observed. Thus, this appearance of the
increased
of presence
the fracture of the
surfaces small spherical
indicated that thedimples and flat
deformation facets
was less was observed.
ductile Thus, this
in the impact testsappearance
at −50 °C
fracture surfaces indicated that the deformation was less ductile in the impact tests at −50 ◦ C than that
of thethat
than fracture
at thesurfaces
ambientindicated
temperaturethat the
anddeformation
led to a drop was
in less ductile in energy
the absorbed the impactfromtests the at −50 °C
ambient
at the ambient temperature and led to a drop in the absorbed energy from the ambient temperature
than that at to
temperature the−50
ambient temperature
°C in Figure and led
7. Moreover, thetoappearance
a drop in the absorbed
of the fractureenergy
surfaces from the ambient
in Figure 11 are
to −50 ◦ C in Figure 7. Moreover, the appearance of the fracture surfaces in Figure 11 are made quite
temperature
made to −50 °C confirming
quite analogous, in Figure 7. that
Moreover,
the impactthe appearance
properties ofofspecimens
the fracture at surfaces
−50 °C are in Figure
roughly 11 are
alike.
analogous, confirming that the impact properties of specimens at −50 ◦ C are roughly alike. Overall,
made quite
Overall, the analogous, confirming
effect of hydrogen that the impact
pre-charging on theproperties
CVN impact of specimens
properties at of
−50316L
°C are roughly
stainless alike.
steel at
theOverall,
effect ofthe
hydrogen at −50 ◦ C was
−50 °C was effect ofpre-charging
negligible. hydrogen on the CVN
The droppre-charging
in the absorbed onimpact
the CVN
energy
properties
impact
from
of 316L stainless
properties
the ambient of 316Lsteel
temperature stainless
to −50 °Csteel
was at
negligible.
caused byThe
−50 °C was the drop in the
negligible.
decrease The absorbed
drop
in the energy
in the
temperature offrom
absorbed the ambient
the energy
CVN from the
impact temperature
test. to −50 ◦ Cto
ambient temperature was
−50caused
°C wasby
thecaused
decrease in the
by the temperature
decrease of the CVNof
in the temperature impact
the CVNtest.impact test.

Figure 11. Scanning electron microscope (SEM) images obtained from the impact test at −50 °C: (a)
Figure
Figure Scanning
11.11.
uncharged, (b) 12 h electron
Scanning electronmicroscope
charged, microscope
(c) (SEM)
andimages
(SEM)
24 h charged images obtained
(d) 48 hobtained fromthe
charged.from theimpact
impacttest
testatat−50 °C:◦ C:
−50 (a)(a)
uncharged,
uncharged, (b)(b)
1212
hh charged,(c)
charged, (c)2424hhcharged
chargedand
and (d)
(d) 48
48 h charged.
charged.
Metals 2019, 9, 625 11 of 14
Metals 2019, 9, x FOR PEER REVIEW 11 of 14

Figure
Figure 12 12 illustrates
illustrates the
themicrostructure
microstructureimages imagesofofthe thefracture
fracturesurface
surfaceinin the
the CVN CVN impact
impact test
test at
at −125 ◦ C of 316L stainless steel according to the hydrogen charging time. As shown in this figure,
−125 °C of 316L stainless steel according to the hydrogen charging time. As shown in this figure, the
the fracture
fracture surfaces
surfaces werewere mainly
mainly covered
covered byby smallspherical
small sphericaldimples.
dimples.The Thepresence
presenceof of the
the small
small and
and
spherical dimple fracture was more frequent here than in the ambient temperature and −50 ◦ C, making
spherical dimple fracture was more frequent here than in the ambient temperature and −50 °C,
the average
making size of dimples
the average at −125 ◦at
size of dimples C −125
smaller °C than at the
smaller thanambient temperature
at the ambient and −50 and
temperature
◦ C and the
−50 °C
spherical dimples are
and the spherical relatively
dimples are shallow.
relativelyThus, this appearance
shallow. Thus, this of the fractureofsurfaces
appearance indicated
the fracture that
surfaces
the CVN impact behavior at −125 ◦ C was more brittle than at the ambient temperature and −50 ◦ C,
indicated that the CVN impact behavior at −125 °C was more brittle than at the ambient temperature
confirming at −125 ◦ C was smaller than at −50 ◦ C. Besides, there
and −50 °C,that the average
confirming thatabsorbed
the average energy
absorbed energy at −125 °C was smaller than at −50 °C.
were differences in the fracture surface according
Besides, there were differences in the fracture surface according toto the charging times. The surface
the charging morphology
times. The surface of
the uncharged specimen was rougher than the charged specimens. Specifically,
morphology of the uncharged specimen was rougher than the charged specimens. Specifically, the the size of dimples
of the
size of uncharged
dimples of the specimen was specimen
uncharged relatively waslarger and deeper,
relatively largerwhich confirms
and deeper, the confirms
which drop in absorbed
the drop
energy of charged specimens at −125 ◦ C. In addition, the fracture surfaces of the charged specimens in
in absorbed energy of charged specimens at −125 °C. In addition, the fracture surfaces of the charged
Figure 12 are
specimens inroughly
Figure alike,
12 arewhich mainly
roughly covered
alike, which by mainly
small spherical
covereddimples.
by small Therefore,
sphericalthedimples.
surface
morphology
Therefore, theand the average
surface size ofand
morphology dimples confirmed
the average size that hydrogen
of dimples pre-charging
confirmed degradedpre-
that hydrogen the
impact toughness and surface morphology of the CVN impact test conducted at −125 ◦ C but the
charging degraded the impact toughness and surface morphology of the CVN impact test conducted
differences
at −125 °C butamong the chargedamong
the differences specimens were negligible.
the charged specimens were negligible.

◦ C:
Figure 12. Scanning
Figure 12. Scanning electron
electron microscope
microscope (SEM)
(SEM) images
images obtained
obtained from
from the
the impact
impact test at −125
test at −125 °C: (a)
uncharged,
uncharged, (b)
(b) 12
12 hh charged,
charged, (c)
(c) 24
24 hh charged
charged and
and (d)
(d) 48
48 hh charged.
charged.

Figure
Figure 13 13 illustrates
illustrates the
the microstructure
microstructure images
images on
on the
the fracture
fracture surface
surface in
in the
the CVN
CVN impact
impact test
test at
at
−196 ◦ C. The appearance of the fracture surfaces in this figure is dominated by small and spherical
−196 °C. The appearance of the fracture surfaces in this figure is dominated by small and spherical
dimples,
dimples, which
which are are smaller and more
smaller and more circular
circular than
than those
those in
in the
the fracture
fracture surface
surface of
of the
the impact
impact tests
tests
conducted −50 ◦ −125 ◦
conducted at the ambient temperature, −50 °C and −125 °C. Besides, the deep and large dimples are
at the ambient temperature, C and C. Besides, the deep and large dimples are
◦ C was the least ductile fracture
rarely
rarely observed
observed herein,
herein, confirming
confirming that
that the
the type
type of
of fracture
fracture atat −196
−196 °C was the least ductile fracture
among
among thethe tested
tested temperatures. That is,
temperatures. That is, this
this was
was the
the reason
reason whywhy the
the absorbed
absorbed energy
energy inin the
the CVN
CVN
impact test at −196 ◦ C was the smallest among the tested temperatures.
impact test at −196 °C was the smallest among the tested temperatures.
Metals 2019, 9, 625 12 of 14
Metals 2019, 9, x FOR PEER REVIEW 12 of 14

Figure
Figure 13.
13. Scanning
Scanning electron
electron microscope
microscope (SEM)
(SEM) images
images obtained
obtained from
from the
the impact
impact test
test at
at −196
−196 °C:
◦ C: (a)
(a)
uncharged, (b) 12 h charged, (c) 24 h charged and (d) 48 h charged.
uncharged, (b) 12 h charged, (c) 24 h charged and (d) 48 h charged.

Among
Among the the SEM
SEM images
images in
in Figure 13, the uncharged
uncharged specimen’s
specimen’s size
size and
and depth
depth were largest
among
among thethe fracture
fracture surfaces
surfaces obtained
obtained from the impact test at −196 ◦ C, confirming
−196 °C, confirming that
that the
the type
type of
fracture of the uncharged specimen at −196 ◦ C was
−196 °C was the
the most
most ductile
ductile fracture.
fracture. Besides,
Besides, the
the differences
differences
among
among the
the fracture
fracture surfaces
surfaces the charged specimens
specimens were not noticeable. That That is,
is, they
they were covered
by
by similar shallow and small spherical dimples, confirming
confirming that hydrogen exposure
exposure decreased
decreased the
ductility and the absorbed energy of the charged specimens in the CVN impact test at −196 ◦ C.
−196 °C.

4. Conclusions
4. Conclusions
The CVN
The CVNimpact behavior
impact of the
behavior ofhydrogen-exposed 316L stainless
the hydrogen-exposed steel at thesteel
316L stainless ambient temperature,
at the ambient
−50 ◦ C, −125 ◦ C and −196 ◦ C was studied herein. The following conclusions can be drawn from this
temperature, −50 °C, −125 °C and −196 °C was studied herein. The following conclusions can be
study: from this study:
drawn
• Exposure
Exposure to to hydrogen
hydrogen increased
increased thethe hydrogen
hydrogen concentration
concentration of of the
the samples
samples collected
collected atat the
the
specimen surface. After 24 h of charging, the hydrogen concentration in the charged
specimen surface. After 24 h of charging, the hydrogen concentration in the charged specimens specimens
reached
reached aa saturation point.
saturation point.
 Hydrogen charging resulted in a slight reduction in the absorbed energy and ductility of 316L
• Hydrogen charging resulted in a slight reduction in the absorbed energy and ductility of 316L
stainless steel at most of the tested specimens. The drop in absorbed energy varied from 0.23%
stainless steel at most of the tested specimens. The drop in absorbed energy varied from 0.23% to
to 16.6%.
16.6%.
 The surface morphology of the uncharged specimens was more ductile than that of the pre-
• The surface morphology of the uncharged specimens was more ductile than that of the pre-charged
charged specimens impacted at ambient temperature,◦−125 °C and −196 °C. While the differences
specimens impacted at ambient temperature, −125 C and −196 ◦ C. While the differences for
for specimens impacted at −50 °C were negligible.
specimens impacted at −50 ◦ C were negligible.
 Another academic insight obtained herein is that low temperature decreased the ductility of the
• Another academic insight obtained herein is that low temperature decreased the ductility of the
V-notched specimens in the CVN impact test. The loss of ductility caused by the ductile to brittle
V-notched specimens in the CVN impact test. The loss of ductility caused by the ductile to brittle
transformation was attributed to the lowering of the temperature in the CVN impact tests.
transformation was attributed to the lowering of the temperature in the CVN impact tests.
Therefore, the impact properties of 316L stainless steel have a high resistance against HE and
this material
Therefore,canthe
beimpact
a possible candidate
properties as material
of 316L stainlessforsteel
the have
hydrogen containment
a high system.HE
resistance against However,
and this
using 316L
material canstainless steel candidate
be a possible at low temperature
as materialshould
for thebe carefullycontainment
hydrogen considered system.
becauseHowever,
of the losses in
using
ductility and fracture resistance caused by this low temperature. Former studies investigated
316L stainless steel at low temperature should be carefully considered because of the losses in ductility the
effect of HE resistance
and fracture on the mechanical
caused by properties of materials.
this low temperature. A standard
Former studies for the absorbed
investigated energy
the effect of
of HE
hydrogen-exposed materials of
on the mechanical properties in materials.
specific temperature
A standard for ranges must be energy
the absorbed surveyed and created. The
of hydrogen-exposed
results of this
materials study could
in specific contribute
temperature to the
ranges research
must database
be surveyed andfor hydrogen
created. Thetanks,
resultshydrogen pipelines
of this study could
and fuel cell vehicles containing 316L stainless steel.
Metals 2019, 9, 625 13 of 14

contribute to the research database for hydrogen tanks, hydrogen pipelines and fuel cell vehicles
containing 316L stainless steel.

Author Contributions: L.T.H.N. wrote the paper, designed the testing program and performed CVN impact
tests. J.S.H. coordinated the sample acquisition, specimen machining, measured hydrogen concentration. M.S.K.
prepared the experimental testing facility and analyzed impact test results. J.H.K. optimized experimental setup
for microstructure analysis and contributed reviewing the original manuscript. S.K.K. optimized experimental
setup for CVN impact tests and contributed reviewing the original manuscript. J.M.L. supervised the experiments
and contributed reviewing the original manuscript.
Funding: This work was supported by the National Research Foundation of Korea (NRF) grant funded by the
Ministry of Science and ICT (MSIT) (No. 2018R1A2B6007403). This work was also supported by the National
Research Foundation of Korea (NRF) grant funded by the Korea government (MSIT) through GCRC-SOP (No.
2011-0030013).
Conflicts of Interest: The authors declare no conflicts of interest.

References
1. Marine Environment Protection Committee (MEPC), 72nd session, 9–13 April 2018. Available online:
http://www.imo.org/en/MediaCentre/MeetingSummaries/MEPC/Pages/MEPC-72nd-session.aspx (accessed
on 14 April 2019).
2. Dwivedi, S.K.; Vishwakarma, M. Hydrogen embrittlement in different materials: A review. Int. J. Hydrogen
Energy 2018, 43, 21603–21616. [CrossRef]
3. Michler, T.; Lee, Y.; Gangloff, R.P.; Naumann, J. Influence of macro segregation on hydrogen environment
embrittlement of SUS 316L stainless steel. Int. J. Hydrogen Energy 2009, 34, 3201–3209. [CrossRef]
4. Czarkowski, P.; Krawczynska, A.T.; Slesinski, R.; Brynk, T.; Budniak, J.; Lewandowska, M.; Kurzydlowski, K.J.
Low temperature mechanical properties of 316L type stainless steel after hydrostatic extrusion. Fusion Eng.
Des. 2011, 86, 2517–2521. [CrossRef]
5. Nagumo, M. Fundamentals of Hydrogen Embrittlement; Springer: Berlin, Germany, 2016; ISBN 9789811001611.
6. Fukuyama, S.; Zhang, L.; Yokogawa, K. Development of Materials Testing Equipment in High Pressure
Hydrogen and Hydrogen Environment Embrittlement of Austenitic Stainless Steels. Jpn. Inst. Met. 2004, 68,
62–65. [CrossRef]
7. Kanezaki, T.; Narazaki, C.; Mine, Y.; Matsuoka, S.; Murakami, Y. Effects of hydrogen on fatigue crack growth
behavior of austenitic stainless steels. Int. J. Hydrogen Energy 2008, 33, 2604–2619. [CrossRef]
8. Weber, S.; Martin, M.; Theisen, W. Lean-alloyed austenitic stainless steel with high resistance against hydrogen
environment embrittlement. Mater. Sci. Eng. A 2011, 528, 7688–7695. [CrossRef]
9. Sugiyama, S.; Ohkubo, H.; Takenaka, M.; Ohsawa, K.; Ansari, M.I.; Tsukuda, N.; Kuramoto, E. The effect of
electrical hydrogen charging on the strength of 316 stainless steel. J. Nucl. Mater. 2000, 283–287, 863–867.
[CrossRef]
10. Rozenak, P.; Eliezer, D. Phase changes related to hydrogen-induced cracking in austenitic stainless steel.
Acta Metall. 1987, 35, 2329–2340. [CrossRef]
11. Rozenak, P.; Robertson, I.M.; Birnbaum, H.K. HVEM studies of the effects of hydrogen on the deformation
and fracture of AISI type 316 austenitic stainless steel. Acta Metall. Mater. 1990, 38, 2031–2040. [CrossRef]
12. Garber, R.; Bernstein, I.M.; Thompson, A.W. Effect of hydrogen on ductile fracture of spheroidized steel. Scr.
Metall. 1976, 10, 341–345. [CrossRef]
13. Venezuela, J.; Tapia-Bastidas, C.; Zhou, Q.; Depover, T.; Verbeken, K.; Gray, E.; Liu, Q.; Liu, Q.; Zhang, M.;
Atrens, A. Determination of the equivalent hydrogen fugacity during electrochemical charging of 3.5NiCrMoV
steel. Corros. Sci. 2018, 132, 90–106. [CrossRef]
14. Matsuo, T.; Yamabe, J.; Matsuoka, S. Effects of hydrogen on tensile properties and fracture surface
morphologies of Type 316L stainless steel. Int. J. Hydrogen Energy 2014, 39, 3542–3551. [CrossRef]
15. Yamabe, J.; Takakuwa, O.; Matsunaga, H.; Itoga, H.; Matsuoka, S. Hydrogen diffusivity and tensile-ductility
loss of solution-treated austenitic stainless steels with external and internal hydrogen. Int. J. Hydrogen Energy
2017, 42, 13289–13299. [CrossRef]
16. Momotani, Y.; Shibata, A.; Terada, D.; Tsuji, N. Effect of strain rate on hydrogen embrittlement in low-carbon
martensitic steel. Int. J. Hydrogen Energy 2017, 42, 3371–3379. [CrossRef]
Metals 2019, 9, 625 14 of 14

17. ASTM Standard Test Methods for Notched Bar Impact Testing of Metallic Materials; ASTM International: West
Conshohocken, PA, USA, 2016.
18. Rossoll, A.; Berdin, C.; Forget, P.; Prioul, C.; Marini, B. Mechanical aspects of the Charpy impact test. Nucl.
Eng. Des. 1999, 188, 217–229. [CrossRef]
19. Fassina, P.; Bolzoni, F.; Fumagalli, G.; Lazzari, L.; Vergani, L.; Sciuccati, A. Influence of hydrogen and low
temperature on pipeline steels mechanical behaviour. Procedia Eng. 2011, 10, 3226–3234. [CrossRef]
20. Mori, K.; Lee, E.W.; Frazier, W.E.; Niji, K.; Battel, G.; Tran, A.; Iriarte, E.; Perez, O.; Ruiz, H.; Choi, T.; et al.
Effect of Tempering and Baking on the Charpy Impact Energy of Hydrogen-Charged 4340 Steel. J. Mater.
Eng. Perform. 2015, 24, 329–337. [CrossRef]
21. Rosenberg, G.; Sinaiova, I. Evaluation of hydrogen induced damage of steels by different test methods. Mater.
Sci. Eng. A 2017, 682, 410–422. [CrossRef]
22. ISO 16573:2015. Steel—Measurement Method for the Evaluation of Hydrogen Embrittlement Resistance of High
Strength Steels; International Organization for Standardization: Geneva, Switzerland, 2015.
23. Jeon, H.H.; Lee, S.M.; Han, J.; Park, I.J.; Lee, Y.K. The effect of Zn coating layers on the hydrogen embrittlement
of hot-dip galvanized twinning-induced plasticity steel. Corros. Sci. 2016, 111, 267–274. [CrossRef]
24. Brass, A.M.; Chêne, J. Hydrogen uptake in 316L stainless steel: Consequences on the tensile properties.
Corros. Sci. 2006, 48, 3222–3242. [CrossRef]
25. Byun, T.S.; Lach, T.G. Mechanical Properties of 304L and 316L Austenitic Stainless Steels after Thermal Aging for
1500 Hours; U.S. Department of Energy/Office of Nuclear Energy: District of Columbia, WA, USA, 2016;
pp. 1–19.
26. Kim, J.H.; Choi, S.W.; Park, D.H.; Lee, J.M. Charpy impact properties of stainless steel weldment in liquefied
natural gas pipelines: Effect of low temperatures. Mater. Des. 2015, 65, 914–922. [CrossRef]
27. Djukic, M.B.; Zeravcic, V.S.; Bakic, G.; Sedmak, A.; Rajicic, B. Hydrogen Embrittlement of Low Carbon
Structural Steel. Procedia Mater. Sci. 2014, 3, 1167–1172. [CrossRef]
28. Godoi, W.; Kuromoto, N.K.; Guimaraes, A.S.; Lepienski, C.M. Effect of the hydrogen outgassing time on the
hardness of austenitic stainless steels welds. Mater. Sci. Eng. A 2003, 354, 251–256. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).