Acta Metallurgica Sinica (English Letters) (2020) 33:759–773

https://doi.org/10.1007/s40195-020-01039-7

Review of Hydrogen Embrittlement in Metals: Hydrogen Diffusion,

Hydrogen Characterization, Hydrogen Embrittlement Mechanism
and Prevention
Xinfeng Li1,2 · Xianfeng Ma1 · Jin Zhang3 · Eiji Akiyama4 · Yanfei Wang5 · Xiaolong Song2

Received: 8 October 2019 / Revised: 1 February 2020 / Published online: 22 April 2020
© The Chinese Society for Metals (CSM) and Springer-Verlag GmbH Germany, part of Springer Nature 2020

Abstract
Hydrogen dissolved in metals as a result of internal and external hydrogen can affect the mechanical properties of the metals,
principally through the interactions between hydrogen and material defects. Multiple phenomena such as hydrogen dissolu-
tion, hydrogen diffusion, hydrogen redistribution and hydrogen interactions with vacancies, dislocations, grain boundaries
and other phase interfaces are involved in this process. Consequently, several hydrogen embrittlement (HE) mechanisms have
been successively proposed to explain the HE phenomena, with the hydrogen-enhanced decohesion mechanism, hydrogen-
enhanced localized plasticity mechanism and hydrogen-enhanced strain-induced vacancies being some of the most important.
Additionally, to reduce the risk of HE for engineering structural materials in service, surface treatments and microstructural
optimization of the alloys have been suggested. In this review, we report on the progress of the studies on HE in metals,
with a particular focus on steels. It focuses on four aspects: (1) hydrogen diffusion behavior; (2) hydrogen characterization
methods; (3) HE mechanisms; and (4) the prevention of HE. The strengths and weaknesses of the current HE mechanisms
and HE prevention methods are discussed, and specific research directions for further investigation of fundamental HE
mechanisms and methods for preventing HE failure are identified.

Keywords Hydrogen embrittlement · Hydrogen diffusion · Hydrogen embrittlement mechanism · Hydrogen embrittlement
prevention

1 Introduction

The degradation of mechanical properties, especially in plas-

Available online at http://link.sprin​ger.com/journ​al/40195​.
ticity, of materials is known as hydrogen embrittlement (HE).
* Xinfeng Li HE is relevant for many metallic materials, including low-
[email protected] alloy steels, precipitation hardening steels, superalloys and
* Yanfei Wang aluminum alloys. HE failure always occurs at low stress level
[email protected] with brittle fracture, often causing huge economic losses and
even catastrophe. For example, high-strength PSB1080 steels
1
Sino-French Institute of Nuclear Engineering used in wind foundation fractured prematurely owing to HE
and Technology, Sun Yat-sen University, Zhuhai 519082,
China [1], resulting in an economic loss of five million Yuan and
2 downtime for maintenance. The drawback has driven signifi-
State Key Laboratory for Mechanical Behavior of Materials,
Xi’an Jiaotong University, Xi’an 710049, China cant research effort among the industry and academia for the
3 investigation of HE using multi-scale experimental and theo-
Department of Physics and Astronomy, Washington State
University, Pullman, WA 99164, USA retical methods. It has been generally concluded that hydrogen
4 can decrease the macroscopic and microscopic tensile strength
Materials Design Division, Institute for Materials Research,
Tohoku University, Sendai 980‑8577, Japan [2–8], fatigue strength [9–11] and fracture toughness [12–16],
5 whereas its effect on fatigue crack propagation rate of steels is
School of Chemical Engineering and Technology, China
University of Mining and Technology, Xuzhou 221116, still disputed, depending on the frequency or stress ratio level
China [17]. In addition, cohesive zone modeling related to hydrogen

13
Vol.:(0123456789)
760 X. Li et al.

proposed by Serebrinsky et al. [18] has been used to evalu- H3 O+ + M → MHads + H2 O Volmer reaction (acid solution)
ate hydrogen-assisted crack initiation and propagation under (2)
static and cyclic loading, and obtained numerical simulation
H2 O + M + e → MHads + OH−
results are comparable to the experimental results determined
by Wang et al. [19], Olden et al. [20], Xing et al. [21] and Yu Volmer reaction (neutral or alkaline solution) (3)
et al. [22]. At the nanoscale, dislocations play an important
role in hydrogen-induced intergranular crack initiation and MHads + H3 O+ + e → M + H2 + H2 O Heyrovsky (acid solution)
propagation [23]. Hydrogen segregation suppresses disloca- (4)
tion emission at the crack tip with a ductile-to-brittle transi- MHads + H2 O + e → M + H2 + OH −
tion, whose process is connected to material states and loading
conditions through a kinetic model of hydrogen delivery to the
Heyrovsky (neutral or alkaline solution) (5)
crack tip region [24]. Molecular dynamics simulation results
[25–27] imply that hydrogen either has no effect on or pro- 2MHads → 2M + H2 Tafel reaction (6)
motes or suppresses dislocation motion. However, according
to in situ TEM observation results dislocations are dragged by MHads → MHabs (7)
hydrogen [28]. Despite a number of previous studies on the HE
of metals, many issues are yet to be clarified such as hydro- where M is the metal surface. MHads is the hydrogen
gen characterization, HE failure criteria and fundamental HE absorbed on the metal surface. MHabs is the hydrogen dis-
mechanisms. In this review, we discuss four aspects of the HE solved in the materials. Hydrogen atoms attached onto the
behavior of metals, particularly steels, namely (1) hydrogen metal surface are formed through the processes described by
diffusion behavior; (2) hydrogen characterization methods; (3) Eqs. (2) and (3). Some hydrogen atoms are released in the
HE mechanism; and (4) HE prevention. form of hydrogen molecules, as described by Eqs. (4)–(6).
For electrochemical hydrogen charging, the hydrogen con-
tent is determined by the hydrogen charging current den-
2 Hydrogen Diffusion Behavior
sity, hydrogen charging potential, hydrogen charging time,
the type of electrolyte, poisoning agent, hydrogen charging
Due to the small radius of hydrogen atoms, hydrogen enters
temperature and hydrogen charging method. With increas-
into steels in the form of atoms. High-pressure hydrogen gas,
ing KCN poisoning concentration, the hydrogen fugacity
electrochemical hydrogen charging and corrosion reactions are
on the pure iron surface increases, and the increase in the
the typical sources of hydrogen in metals.
overpotential also results in a high hydrogen fugacity under
For high-pressure hydrogen gas, the dissolution of hydro-
the same electrolyte [33], as shown in Fig. 1a. Compared
gen involves three steps [29]: The first step is physical absorp-
with gaseous hydrogen charging [34], the equivalent charg-
tion on a solid surface due to the van der Waals interactions
ing pressures caused by electrochemical charging are higher,
between the hydrogen gas and the surface, with a low absorp-
and their hydrogen concentration is proportional to square
tion energy of 3–5 kJ mol−1. Physical absorption is reversible
root of hydrogen pressure, as shown in Fig. 1b.
and it is easy to reach equilibrium. The second step is chemical
Some researchers [31, 35, 36] have found that hydrogen
absorption. Owing to short-range chemical interactions, chem-
concentration is proportional to the square root of the charg-
ical absorption occurs within single atom layer. In addition,
ing current density, as described by:
chemical absorption is irreversible and relatively slow because
of high hydrogen gas molecule dissociation energy of 4.47 eV √
CH = K i (8)
and high absorption energy of 40–160 kJ mol−1. The third
step is hydrogen dissolution. Following chemical absorption, where K is a constant and i is the current density.
hydrogen atoms diffuse toward the interior of materials owing Furthermore, hydrogen content in steels is scaled by the
to the effect of the hydrogen concentration gradient. For high- electro-migration factor 𝜑, which can be expressed as [37]:
pressure hydrogen gas, the dissolved hydrogen concentration
(CH) in a metal follows Sieverts’ law as [30]: Z ∗ e𝜌
𝜑= (9)
√ kT
CH = S PH2 (1) where Z∗ is the effective charge number. e is the electron
charge. 𝜌 is the resistivity of the alloys. k is the Boltzmann
where S is the solubility constant that depends on the type of constant. T is the temperature.
alloys and temperature [31] and PH2 is the hydrogen partial For an ideal crystal, hydrogen atoms are preferentially
pressure. located at BCC tetrahedral or FCC octahedral interstitial
For electrochemical hydrogen charging or corrosion reac- sites. At room temperature and 0.1 MPa hydrogen pressure,
tions, the hydrogen evolution reactions are given by [32]:

13
Review of Hydrogen Embrittlement in Metals: Hydrogen Diffusion, Hydrogen Characterization,… 761

Fig. 1  a Hydrogen fugacity (fH2) for pure iron in 0.1 M NaOH with different additions of KCN [33]; b total hydrogen concentration of the
3.5NiCrMoV specimens charged electrochemically with hydrogen plotted to fit the regression line for charging in gaseous hydrogen [34]

Table 1  Activation energies of various hydrogen traps in steels [3]

Trapping sites Activation

energy (kJ
­mol−1)

Reversible trapping sites

Interstitial sites in iron 4–8
Dislocation 26.4–26.8
Lath boundary 17.8–18.6
Austenite/martensite 22
Fig. 2  Hydrogen traps in the steels [39]: a interstitial sites; b surface Grain boundary 17.8–18.6
traps; c subsurface traps; d grain boundary traps; e dislocation traps; NbC (coherent) 39–48
f vacancy traps Irreversible trapping sites
Ferrite/cementite interface 66.3–68.4
Fe3C interface 84
the atomic ratio of dissolved hydrogen in BCC iron has a low Al2O3 interface 79
value of 2 × 10–8 according to Sieverts’ law [38]. However, in MnS interface 72
practice, the hydrogen content is always significantly greater NbC (incoherent) 63–68
than this value owing to the high number of hydrogen traps,
such as vacancies, dislocations, grain boundaries and surface
hydrogen traps, as schematically shown in Fig. 2.
On the basis of hydrogen trap activation energy hydrogen immigration. For stress-induced hydrogen diffu-
( Eb ), these traps are divided into reversible hydrogen sion, the dependence of the local hydrogen concentration (C )
traps ( Eb < 60 kJ mol−1) and irreversible hydrogen traps on the hydrostatic stress is expressed as:
( Eb > 60 kJ mol−1). Reversible hydrogen traps include inter- ( )
stitial sites, dislocations, lath boundaries, grain boundaries 𝜎h VH
C = C0 exp (10)
and coherent precipitates, whereas carbides, inclusions and RT
incoherent precipitates are irreversible hydrogen traps. It
was reported [40, 41] that the HE susceptibility of alloys is where C0 is the average hydrogen concentration. 𝜎h is the
correlated with the presence of reversible hydrogen traps. hydrostatic stress. VH is the hydrogen partial molar volume.
Table 1 lists the hydrogen traps in steels and their trap acti- Additionally, hydrogen can be immigrated through the
vation energies. movement of dislocations as dislocation velocity rate (V ∗)
Generally, hydrogen atoms are redistributed by stress- during tensile deformation is below the corresponding criti-
induced hydrogen diffusion and dislocation-induced cal value (VC), as described by:

13
762 X. Li et al.

V∗ =
𝜀̇ 3 Hydrogen Characterization Methods
𝜌Dis b (11)
The content of hydrogen in steels with the BCC struc-
D Eb ture is extremely low. Usually, HE failure of high-strength
Vc = (12) steels occurs due to the hydrogen concentration of only
RT 30b
several ppm. Moreover, hydrogen atoms are small and
where 𝜀̇ is the tensile strain rate. 𝜌Dis is the dislocation den- can easily move within the metals. Therefore, the accu-
sity. b is the magnitude of the Burgers vectors. D is the rate measurement of hydrogen content has been intensely
hydrogen diffusion coefficient. Eb is the activation energy investigated, but remains a difficult problem. To date, the
of dislocation with hydrogen. In most cases, hydrogen redis- prevailing hydrogen characterization methods include the
tribution is achieved through the synergetic effect of stress- glycerin method (GM), inert gas fusion heat conduction
induced hydrogen diffusion and dislocation immigration. method (IGFHCM), thermal desorption spectroscopy
During hydrogen charging, hydrogen traps near the sam- method (TDS), secondary ion mass spectroscopy (SIMS)
ple surface are first occupied by hydrogen atoms, follow- and atom probe tomography (APT). The former three
ing which the hydrogen atoms diffuse toward the interior methods are always used to measure the macroscopic
of the alloys, indicating that hydrogen diffusion parameters average hydrogen concentration, whereas the local micro-
are affected by the hydrogen traps. Hence, the relationship scopic hydrogen content is determined using the latter two
between hydrogen trap density ( N ) and hydrogen diffusion methods. The common principle of GM, IGGHCM and
behavior of AISI430 steel was determined as [42]: TDS methods is that hydrogen desorption occurs after the
C
( ) samples are heated to some extent. For example, GM was
DL
N= 0 −1 (13) used to measure the average hydrogen concentration of
3 Deff
X70 steel [12], and it was found that hydrogen concen-
where C0 is the hydrogen concentration near the surface. tration increased with an increase in hydrogen charging
DL is the lattice hydrogen diffusion coefficient. Deff is the current density. Hydrogen behavior in the alloys on the
apparent hydrogen diffusion coefficient. Considering the trap microscale has been characterized by the SIMS technique.
activation energy, the modified relationship is expressed as It was observed that hydrogen preferentially segregated at
follows [43]: the Al–(Fe–Al) interfaces [50]. Based on the combination
of EBSD and SIMS results, Oudriss et al. [51] concluded
that hydrogen diffusion along the general grain boundary
( )
DL N E 1
ln − 1 = ln T + b × (14)
Deff NL R T was faster than that along the special grain boundary. APT
is a powerful tool for the examination of the hydrogen dis-
where DL is the lattice hydrogen diffusion coefficient. Deff is tribution on the nanoscale. Recent research findings dem-
the apparent hydrogen diffusion coefficient. R is the gas con- onstrate that hydrogen mainly resides in residual austenite,
stant. T is the temperature. Table 2 summarizes the values whereas it is not available at the carbide/ferrite interfaces
of the apparent hydrogen diffusion coefficient and apparent [52]. Specifically, the hydrogen content in austenite phase
hydrogen concentration for different steels. (33.9 wppm) is three times higher than that in martensitic
phase (10.7 wppm) [53], as shown in Fig. 3a–c. It was also

Table 2  Hydrogen diffusion behavior parameter for some steels

Type of steels Grain size (μm) Microstructure Apparent hydrogen diffusion Apparent hydrogen
coefficient ­(m2 ­s−1) concentration (mol
­m−3)

Pure iron [44] – Ferrite 5.8 × 10–10 0.15

304 steels [45] – Austenite 7.37 × 10–16 32.51
SAF2205 [20] – Ferrite + Austenite 3.0 × 10–15 –
SAE1008 [46] 19 Ferritic + carbides 2.19 × 10–10 0.49
PSB1080 [47] 13 Martensite + bainite 4.43 × 10–11 12.21
300 M [48] – Martensite + austenite 9.6 × 10–12 –
PH17-4 [49] 27 Martensite + Cu-rich precipitates 2.18 × 10–12 1235
PH13-8Mo [49] 23 Martensite + NiAl precipitates 9.42 × 10–12 561

13
Review of Hydrogen Embrittlement in Metals: Hydrogen Diffusion, Hydrogen Characterization,… 763

found that carbides in the quenching–partitioning–temper- the grain boundary is of 18.6 kJ mol−1 and 17.8 kJ mol−1,
ing (QPT) steel act as effective non-diffusible hydrogen respectively, while it is approximately 70.2 kJ mol−1 for the
trapping sites [54], as shown in Fig. 3a–f. matrix/carbide interfaces [57]. The hydrogen micro-print
Compared to other hydrogen measurement methods, technique (HMT) proposed by Ovejero-Garcia et al. [58] is
the TDS method has a unique capability of analyzing the another key hydrogen characterization method that can be
activation energy of hydrogen traps. The activation energy employed to analyze the relationship between the hydrogen
is calculated using the following equation [56]: distribution and the microscale microstructure. The proce-
( ) dure of the HMT method consists of three steps [58]: (1) The
𝜕 ln T𝜙2 E sample surface is coated by an AgBr emulsion; (2) diffusible
1
P
=− a (15) hydrogen atoms react with A ­ g+ and generate Ag particles
𝜕T R
P
( Ag + H → Ag + H ); and (3) the unreacted AgBr is
+ 0 0 +

removed and Ag particles remain. As the samples are etched,

where 𝜙 is the heating rate. TP is the peak temperature. Ea is the relationship between hydrogen diffusion pathways and
the hydrogen trap activation energy. R(is the) gas constant. microstructural features of alloys can be elucidated. It was
After obtaining the dependence of ln T 2 on T1 , the trap
𝜙
demonstrated that hydrogen evolution in the steels preferen-
P p

activation energy of an alloy is estimated from the slope of tially occurs along the δ–γ interfaces, matrix/inclusion inter-
the fitting curve. Based on the TDS results, the hydrogen faces, grain boundaries and slip lines [58]. According to the
trap activation energy at the martensitic lath boundary and HMT results for the IN718 alloy, the hydrogen release rate
in the γ matrix is higher than that of the δ phases [59], as

Fig. 3  A–C hydrogen concentration distribution in austenite phase and martensitic phase of QPT steel [55]. A 3DAPT map of a combined atom
map of carbon and hydrogen of the as-charged specimen, where iso-concentration surface representing 2.5 at.% carbon is displayed in red. Car-
bon atoms and hydrogen atoms are represented by pink and green, respectively. The inserted map is the corresponding mass spectrum. B Atom
maps of iron, manganese, silicon, carbon and hydrogen of the selected blue rectangle in A. C Average compositions of carbon and hydrogen
along the marked cylinder in A. a–h: Hydrogen concentration distribution in matrix and carbides of QPT steel [6]. b, d, g are enlarged views
showing carbon and hydrogen atom distribution as indicated in a, c, e; f is the carbon content along the blue cylinder in e; i average composi-
tions of carbon and hydrogen along the blue cylinder in h. Carbon and hydrogen are represented by red and green, respectively

13
764 X. Li et al.

shown in Fig. 4. The similarities and the differences among (HEDE), hydrogen-enhanced localized plasticity mechanism
the GM, IGFHCM, TDS, SIMS, APT and HMT methods are (HELP) and hydrogen-enhanced strain-induced vacancies
described in Table 3. (HESIV) have been proposed to explain HE phenomena. It
is clear that the hydrogen pressure theory and HIPT theory
can explain irreversible HE, while reversible HE is mostly
4 Hydrogen Embrittlement Mechanism rationalized by the HEDE, HELP and HESIV mechanisms.
However, it is of consensus that no one HE mechanism is
Hydrogen damage of metals is divided into reversible and widely accepted for explaining all reversible HE phenomena.
irreversible HE. For reversible HE, hydrogen atoms migrate The hydrogen pressure theory first proposed by Zapffe
and then accumulate at the potential cracking locations, et al. [60] in 1941 states that hydrogen atoms preferentially
leading to the delayed fracture of the alloys. By contrast, in segregate at defect locations in the metal, such as micro-
irreversible HE, hydrogen atoms combine with each other voids and inclusions sites. Then, locally accumulated
to form hydrogen molecules at defect sites, generating high hydrogen atoms are combined into hydrogen molecules.
hydrogen gas pressure and hydrogen-induced cracking. After As time goes on, hydrogen atoms around these defects
a hydrogen removal treatment is conducted, the reversible continuously diffuse toward the defect sites and produce
hydrogen damage of steels is healed, but irreversible HE still a high hydrogen gas pressure. When the local hydrogen
remains. So far, many HE mechanisms including the hydro- gas pressure exceeds the critical strength of the mate-
gen pressure theory, hydrogen-induced phase transformation rial, hydrogen-induced cracking takes place. Typically,
(HIPT) theory, hydrogen-enhanced decohesion mechanism “fish eye” in the steels and hydrogen-induced cracking of

Fig. 4  a SE image of specimen immersed in Ag decoration solution with hydrogen charging; a’ magnified image of yellow rectangle area in a
[59]

Table 3  Similarities and differences of different hydrogen characterization methods

Method Temperature Sample scale Hydrogen concentration type Mark

Average hydrogen GM 45 °C mm-scale Diffusible hydrogen

concentration IGFHCM > Melting point mm-scale Diffusible and non-diffusible
hydrogen
TDS 600–1000 °C mm-scale Diffusible and non-diffusible Hydrogen trap activation
hydrogen energy
Local hydrogen SIMS Room temperature μm-scale Diffusible and non-diffusible Hydrogen and grain boundary
concentration hydrogen interactions
HMT Room temperature μm-scale Diffusible and non-diffusible Hydrogen and microstructure
hydrogen interactions
APT Low temperature nm-scale Diffusible and non-diffusible Hydrogen and precipitates
hydrogen interactions

Diffusible hydrogen means that hydrogen desorption occurs at 200 °C and non-diffusible hydrogen indicates that hydrogen desorption occurs at
approximately 600 °C

13
Review of Hydrogen Embrittlement in Metals: Hydrogen Diffusion, Hydrogen Characterization,… 765

pipeline steels in the H­ 2S environment are explained by ΔCPL is the hydrogen concentration induced by plastic strain.
the hydrogen pressure theory. It has been recently reported For high-strength steel, hydrogen distribution at the crack tip
[61] that the issue of “fish eye” is solved when the hydro- is dominated by the hydrostatic stress, even at 2.3% plastic
gen content in the melted steel is controlled to be below strain [68].
2 × 10–6 (wt%). For the HIPT theory, specific metals such The HEDE mechanism is supported by the observa-
as Zr, Nb, V and Ta combine easily with hydrogen to form tion of intergranular fracture and by theoretical calcula-
brittle hydrides due to their large bond energies. In light tions. In intergranular fracture, similar to the role of S, P
of hydrogen concentration of the alloys, the hydrides are and Bi impurity elements, hydrogen segregates at the grain
divided into spontaneously formed hydrides and stress- boundaries and reduces the cohesive interactions between
induced hydrides. At high hydrogen concentration, the metal atoms, resulting in intergranular fracture [23], as
hydrogen directly combines with specific metals such that shown in Fig. 6a. Using theoretical calculation, Wei et al.
spontaneously formed hydrides are obtained. For stress- [69] revealed that the cohesive energies of the [001]/22.6°,
induced hydrides, the initially low hydrogen concentration [011]/38.9° and [111]/60° grain boundaries decreased lin-
is redistributed due to the effect of stress gradient field. early with increasing hydrogen concentration, as shown in
Once the local hydrogen concentration reaches the solubil- Fig. 6b. However, the weakening of the interatomic interac-
ity of metals, hydrides are precipitated. Additionally, the tions in the metal by hydrogen has not been directly demon-
fracture failure mechanism of hydride-containing metals strated experimentally. In addition, a recent study indicated
had been expounded explicitly (Fig. 5a). This mechanism that slip bands in one or two directions are present on the
is composed of four steps [62]: (1) hydrogen diffusion grain boundary facets (Fig. 6c), beneath which severe plas-
and accumulation at the crack tip; (2) hydride formation tic deformation with dislocation cells is detected in Ni-201
and growth; (3) the occurrence of a crack along a specific (Fig. 6d) [23]. Similar phenomena were also observed in
cleavage plane within the hydrides; and (4) the crack is hydrogenated iron [70] and martensitic steel [71, 72]. Thus,
arrested at the matrix/hydride interfaces. Consequently, the role of plasticity in the HEDE mechanism should be
the repetition of the above procedures leads to crack prop- investigated further.
agation. In some circumstances, hydrogen atoms from the In 1972, through careful observation of hydrogen-assisted
decomposed hydrides diffuse toward the stress concentra- fracture surfaces, Beachem et al. [73] found tear ridges on
tion area again and generate fresh brittle hydrides, further brittle fracture surfaces and proposed the HELP mechanism
aggravating brittle cleavage fracture of hydrides [63]. (Fig. 5c). In this mechanism, hydrogen facilitates dislocation
In 1926, Pfeil et al. [65] introduced the HEDE mecha- proliferation and motion, causing local dislocation pileups
nism illustrated in Fig. 5b. They suggested that hydrogen with premature failure of the material. This mechanism has
reduces the cohesive metallic interatomic interactions so focused research attention on the interaction between the
that atomic separation is prone to occur under low ten- hydrogen atmosphere and dislocations, and is primarily uti-
sile stress. As the local hydrogen concentration increases, lized to understand plastic trace on the fracture surfaces such
the degree of reduction in the metallic interatomic forces as dimple fracture in hydrogenated alloys. Unlike the HEDE
increases, which is quantitatively described as [2]: mechanism, plasticity of the alloy plays a significant role in
hydrogen-assisted fracture in the HELP mechanism.
𝜎cH = 𝜎c0 − 𝛽C (16) Several phenomena provide evidence supporting the
where 𝜎cH is the locally critical cohesive stress at the spe- HELP mechanism (1) In situ TEM experiment. Repre-
cific hydrogen concentration C . 𝜎c0 is the critical cohesive sentative researchers such as Birnbaum, Robertson and
stress without hydrogen. 𝛽 is a related parameter for the loss Sofronis directly observed increased dislocation mobility
of critical cohesive stress due to hydrogen. For the HEDE in the presence of hydrogen. They found that hydrogen not
mechanism, potential hydrogen accumulation regions only induces the motion of stationary dislocations, but also
include: (1) the dislocation shielding region at the crack tip; increases the dislocation nucleation rate. Once hydrogen
(2) maximum hydrostatic stress sites; and (3) grain boundary gas is removed, dislocation motion occurs in the opposite
and phase boundary at the crack tip. According to the results direction. These results were dependent on the purity of the
reported by Wu et al. [66] and Gao et al. [67], hydrogen material and the hydrogen gas pressure, rather than on the
concentration near the crack tip (CT ) is composed of three crystal structure (BCC and FCC) and dislocation type (edge
contributions, as expressed by: dislocation, screw dislocation and mixed dislocation) [74,
75]. (2) Slip trace array and high dislocation density near
CT = C0 + ΔC𝜎 + ΔCPL (17) the fracture surface. Compared to those of the hydrogen-
uncharged samples, both slip spacing and slip step of the
where C0 is the hydrogen concentration in the matrix. ΔC𝜎
hydrogen-charged samples are reduced [76]. In addition, slip
is the hydrogen concentration caused by hydrostatic stress.
features and dislocation cells are observed on and beneath

13
766 X. Li et al.

Fig. 5  Schematic diagrams of HE mechanisms. a HIPT [64]: hydrogen-induced phase transformation theory; b HEDE [64]: hydrogen-enhanced
decohesion mechanism; c HELP [64]: hydrogen-enhanced localized plasticity mechanism; d NVC [5]: nanovoid coalescence mechanism; e
HEDE + HELP [5]: combined effect of hydrogen-enhanced decohesion mechanism and hydrogen-enhanced localized plasticity mechanism

the brittle fracture facets. (3) Softening effect. Hydrogen can for Al. (4) Elastic shielding theory. Hydrogen reduces the
decrease the yield strength of the material and the degree short-range interactions of dislocations enabling their easy
of reduction is related to the crystal type, material purity, migration. This shielding effect shows a dependence on the
strain rate and temperature [77, 78]. For instance, hydrogen- dislocation type and temperature. Further, hydrogen-induced
induced softening effect is obvious at low temperatures and a shielding effect on the edge dislocation is much stronger
low strain rate for single-crystal Fe, whereas it is negligible

13
Review of Hydrogen Embrittlement in Metals: Hydrogen Diffusion, Hydrogen Characterization,… 767

Fig. 6  a Hydrogen-induced intergranular fracture in Ni [23]; b dependence of grain boundary bonding energy on hydrogen concentration in Al
[69]; c slip traces on intergranular fracture of hydrogenated Ni-201 [23]; d dislocation cells beneath hydrogen-induced intergranular fracture of
Ni-201 [23]

than that for the screw dislocation. For steels, this effect is was of 20 MPa, increased to 40 MPa for hydrogenated Al
observed in the temperature range from 180 to 200 °C. and reached 120 MPa for the hydrogen–vacancy Al (Fig. 7b)
Although the HELP mechanism has been adopted to [28].
interpret HE behavior, some phenomena still cannot be In 2004, Nagumo et al. [80] suggested that the hydro-
explained by this mechanism. According to macroscopic gen–vacancy complex contributes to fracture failure of
tensile test results [79], hydrogen induces the Portevin–Le alloys and proposed the HESIV mechanism. This mecha-
Chatelier effect in IN718 alloys with the suppression of nism assumes that hydrogen accelerates the formation of
dislocation motion by hydrogen. In situ microscopic ten- strain-induced vacancies and stabilizes vacancy clusters,
sile tests validate that dislocations in pure Al are dragged as evidenced by the positron annihilation lifetime spec-
by hydrogen as well. Once hydrogen gas is no longer pro- trum [81] and molecular dynamics results [82]. As a result,
vided, the previously pinned dislocations recover and begin these vacancy clusters facilitate void initiation and growth
to move (Fig. 7a) [28]. Additionally, simulation results also with the occurrence of premature failure. Hou et al. [83]
indicate that hydrogen hinders dislocation mobility [24, 28]. established a predictive model for the hydrogen adatoms in
For example, the shear stress of the dislocations in pure Al nanovoids that enabled the prediction of hydrogen molecule

13
768 X. Li et al.

Fig. 7  a Dislocation configuration of hydrogen-uncharged and hydrogen-charged Al [28]; b stress–strain curve of Al, Al–H and Al–VaH com-
plex [28]. HU hydrogen-uncharged, HC hydrogen-charged, Al–VaH Al–hydrogen vacancy complex

formation in nanovoids. In recent years, the activation of material preparation processes, such as smelting, welding,
HEDE and HELP mechanisms has been determined by pickling and plating, whereas hydrogen generated dur-
fracture surface and microstructural evolution observation. ing service is external hydrogen and is due to corrosion,
During intergranular fracture or matrix/inclusion interface hydrogen gas and ­H2S gas acid environments. Accord-
decohesion, the HEDE mechanism is active, while HELP ingly, the prevention of HE can be considered based on two
mechanism activates when hydrogen-assisted cracking approaches. The first is the use of surface treatments that
initiates from the intersections between slip bands. Wang involve surface coating and surface modification treatments.
et al. [70] and Martin et al. [23, 84] observed a combination These methods are used to prevent external HE. The second
of brittle hydrogen-assisted fracture surface features with approach is the modification of the material microstructure
microstructural evolution beneath intergranular fracture, such as adding/eliminating the appropriate alloy elements
suggesting synergistic effect of HEDE and HELP mecha- and the optimization of the alloy microstructure.
nism (Fig. 5e). This joint mechanism was also established by
a statistical, physical-based and micro-mechanical model of 5.1 Surface Coating
hydrogen-induced intergranular fracture in steels. Moreover,
the activation of the HELP and HEDE mechanism depends As a metal surface is coated with a film, the hydrogen entry
on the local hydrogen concentration in low-carbon steel into the alloy is suppressed and the alloys exhibit high HE
[85, 86]. In 2012, the nanovoid coalescence (NVC) mecha- resistance. Surface blackening treatment means the forma-
nism was raised [5], as schematically shown in Fig. 5d. The tion of an oxide layer with the thickness of 1–3 μm on the
essence of NVC mechanism is the simultaneous effects of metal surface, which is suitable for improving the atmos-
the HEDE, HELP and HESIV mechanisms. It is important to pheric corrosion resistance of steels. Hydrogen permeation
note that the above-mentioned mechanisms are based on the test results indicate that hydrogen flux and hydrogen dif-
postmortem observation of the samples, which cannot pro- fusion efficiency of the blackening treatment samples are
vide direct evidence for the reduction in cohesive strength reduced in comparison with those of the non-treated sam-
by hydrogen in the HEDE mechanism and for the promotion ples, resulting in the enhancement of the stress corrosion
of dislocation motion by hydrogen in the HELP mechanism. delayed fracture time [88]. In addition, the surface coated
by Ni, Cd, Al and Al–Ni complex film can effectively sup-
press hydrogen infusion and reduce the HE susceptibility of
5 Prevention of Hydrogen Embrittlement alloys. Levchuk et al. [89] reported that the hydrogen flux
of a tempered martensitic steel with 1 μm Al ion implan-
Due to the effects of hydrogen concentration and stress gra- tation was reduced by a factor of 1000 times than that of
dient, hydrogen diffuses toward and accumulates at the stress non-implantation steel. At a high temperature of 800 °C, the
concentration region. As the local hydrogen concentration bonding strength between matrix and film was not affected.
reaches the critical value, which is still unknown, fracture Figueroa et al. [90] revealed that AISI 4340 steel with a
failure occurs as schematically shown in Fig. 8. Usually, Zn–Al film had higher HE resistance than that of the alloy
hydrogen sources are divided into internal hydrogen and coated by a Cd film. Zhou et al. [7] found that a Ni–graphene
external hydrogen. Internal hydrogen is produced in the complex film reduced the hydrogen diffusion coefficient

13
Review of Hydrogen Embrittlement in Metals: Hydrogen Diffusion, Hydrogen Characterization,… 769

Fig. 8  Schematic diagram of hydrogen atom evolution and HE failure [87]

effectively compared with Ni film. Furthermore, hard films

such as TiC, ­Al2O3and ­Si3N4 can dramatically reduce hydro-
gen diffusion behavior, as shown in Fig. 9. It was found
[91] that hydrogen diffusion coefficient of stainless steel was
reduced by five orders of magnitude after a TiN film with the
thickness of 1 μm was coated on the steel surface. It is worth
noting that hydrogen diffusion in the TiC and A ­ l2O3 films
is much slower than that in the austenite phase, suggesting
that these films may act as promising barrier for enhancing
the HE resistance of austenite steels.
Even though coating on the alloy surface can suppress
HE, some controversial results have been reported. Michler
et al. [87] reported premature fracture of Zn, Ni, Ti–DLC
and NiP films at low strain (Fig. 10a) and serious defects of
Fig. 9  Dependence of hydrogen diffusion coefficient of various films Cu film (Fig. 10b) and non-densely spherical structure of
on temperatures [88]

Fig. 10  a Ni film cracking at 8% strain (1 MPa hydrogen gas, − 50°) [87]; b Cu film defects, 1 micropores; 2 cracks; 3 voids [87]; c non-densely
spherical structure Al film [87]

13
770 X. Li et al.

Al film (Fig. 10c), and concluded that HE of stainless steel 5.3 Materials Microstructure Modification
was not affected by the coating of these films. Nishiguchi
et al. [92] found that an Al film can prevent hydrogen atom Alloy elements play a prominent role in HE of steels. The
infusion into the steels. However, the uncoordinated defor- reduction in HE can be obtained by reducing the amounts
mation between the matrix and the Al film led to premature of the C, Si, P, S elements or increasing the amounts of the
failure of the Al film. Thus, it is essential to optimize the Ni, Al, Mo elements. For example, it was found that HE of
film properties such as strength, plasticity, toughness and the Mn–B steel increases with an increase in the C element
bonding strength with the matrix to enable their use as criti- content. However, the susceptibility to HE remains constant
cal components for reducing HE susceptibility. for the C content greater than 0.3% [97]. Banerji et al. [98]
reported that the threshold stress intensity factor of 4340
5.2 Surface Modification steels was enhanced by a factor of five after the P element
content was reduced. The decrease in the Cr, Mn, Si content
Surface nitriding, carbonization treatment and peening treat- or increase in the Mo and Ti content can inhibit the segrega-
ment are promising approaches for enhancing HE resistance. tion of P. According to the results of first-principles calcula-
On the one hand, carbon and nitrogen elements can stabilize tions [99], Al blocks hydrogen diffusion compared to the Si
the austenite phase, impeding the transition from austenite element in BCC iron. Thus, bainitic steel with high Al ele-
to martensite of the unstable 301 and 304 stainless steels ment shows low HE sensibility [100]. The addition of the Cu
[93]. In addition, interstitial carbon and nitrogen atoms and Al elements to the Fe–Mn–C steel increases the stacking
reduce the lattice spacing generating compressive stress at fault energy and reduces the stress concentration at the grain
the surface. Due to the high austenite stability and compres- boundaries, enhancing the HE resistance of the alloy [101].
sive stress states generated by the nitriding/carbonization Additionally, Mo, V and Ti elements often combine with C
treatment, the treated alloys exhibit high resistance to HE to form carbides, which serve as irreversible hydrogen traps
[94]. Furthermore, the HE resistance of the alloys increases and thus enhance the HE resistance of the alloy.
distinctly when nitride the layer thickness is greater than HE of steels depends on their microstructures. Specifi-
the thickness of the hydrogen permeation layer [95]. On the cally, the martensitic structure shows the highest HE sus-
other hand, surface peening treatment not only induces com- ceptibility, followed by bainite, pearlite and austenite in
pressive stress, but also increases the density of the hydro- turn. Ninninga et al. [102] reported that fastener steels with
gen trap sites. During hydrogen charging, hydrogen trap- pearlitic microstructure exhibited lower HE susceptibility
ping sites near the surface are first occupied by hydrogen, compared to bainitic steel. It was found [103] that pearlitic
and then, hydrogen diffuses into the interior of the alloys. steel possessed higher HE resistance than martensitic steel,
Due to the peening layer barrier effect, hydrogen invasion while both of these types of steels have the same strength
layer in the steels is limited and the alloys exhibit low HE level. For austenite, the HE susceptibility of the alloys is
sensibility. For example, Takakuwa et al. [96] revealed that correlated with its content, shape and stability. Owing to
cavitation peening strongly enhanced the fatigue property low hydrogen diffusion and high hydrogen solubility in
of the hydrogen-charged 316 steel. Fracture life increased austenite, an increase in the austenite content leads to the
by a factor of three and fatigue crack propagation rate was reduction in HE of steels [104]. Compared to film-shaped
reduced by 75% compared to those of hydrogen-uncharged austenite, hydrogen-assisted cracking preferentially initiates
samples. Li et al. [4] found that shot peening reduced the from block residual austenite sites, and therefore, the steels
hydrogen permeation behavior of PSB1080 steel so that HE with film-shaped austenite show lower HE [105]. Based on
susceptibility of the alloy decreased. Fracture surface obser- the results obtained by Wang et al. [106], hydrogen-assisted
vation showed that shot peening treatments obstructed the cracking originates from the interfaces between the freshly
occurrence of intergranular fracture. However, another study formed martensite from austenite and original martensite,
[47] indicated that reversible HE decreased, but irrevers- indicating that the increased austenite stabilization, such as
ible HE increased with increasing shot peening pressure. To the addition of C element, can enhance HE resistance. In
summarize, unlike the coating treatment, it is not necessary addition, carbides always increase HE resistance of steels.
to consider the interface bonding behavior for the surface However, Lee et al. [107] reported that HE sensibility of
nitriding, carbonization and peening treatments. Moreover, martensitic steels firstly decreased and then increased with
the influence of novel surface treatments such as the surface increasing carbide content. Kang et al. [57] indicated that
nanocrystallization treatment and laser shot peening on HE sphere-shaped carbides inhibited crack initiation and propa-
of alloys need to be investigated. gation, while cracking was prone to occur for steels contain-
ing needle-shaped carbides in the presence of hydrogen. The
role of nanoscale precipitates on HE of steels has been con-
troversial. Due to the interactions between the dislocations

13
Review of Hydrogen Embrittlement in Metals: Hydrogen Diffusion, Hydrogen Characterization,… 771

and precipitates, the strength of the alloys increases, which recommended. For precipitation-strengthened steels, dislo-
corresponds to low resistance to HE. By contrast, precipi- cations interact with the stress–strain field generated by the
tates in the steels serve as reversible hydrogen traps and misfit between the matrix and the precipitates. In this case,
reduce the local hydrogen concentration accumulation, lead- dislocation motion is suppressed with increasing strength
ing to low HE susceptibility. For example, Shi et al. [108] level of steels, which corresponds to high HE. Meanwhile,
revealed that the pipeline steel with enriched Cu precipitates high-density precipitates act as hydrogen traps with the
exhibited excellent resistance to hydrogen-induced crack- reduction in the local hydrogen concentration. Based on this,
ing. Kimura et al. [109] reported that the critical hydrogen the presence of precipitates in steels can impede HE. Thus, it
concentration of the steel with 10 nm C ­ r2O3 precipitates appears that nanosized precipitates have a dual effect on the
was enhanced significantly in comparison with traditionally HE of precipitation-strengthened steels. However, it is still
manufactured martensitic steel. However, Li et al. [2] have necessary to determine the density of the nanosized precipi-
recently reported the higher HE sensibility of PH13-8Mo tates, at which the steel exhibits the lowest HE susceptibility.
steel primarily strengthened by coherent NiAl precipitates
compared to that of 20SiMn2CrNiMo steel. Finally, any Acknowledgements This work was financially supported by the
National Natural Science Foundation of China (No. 51505477),
types of inclusions such as MnS, O–Al–Si–Ca and TiN the Guangdong Provincial Key S&T Special Project (Nos.
inclusions are not beneficial for HE of steels because they 2017B020235001 and 2019B010943001) and the Guangdong Educa-
often act as hydrogen-assisted crack initiation sites. Thus, tion Department Fund (No. 2016KQNCX005). X. Li appreciates basic
this suggests that a reduction in the amount of inclusion can start-up fund of Sun-Yat Sen University (45000-18841218).
be used to enhance HE resistance of the alloys.

References
6 Conclusions and Outlook
[1] X. Li, Z. Jin, Z. Peng, L. Pei, X. Song, J Fail Anal Prev. 15, 295
With the aid of advanced equipment, the understanding (2015)
of HE mechanism has shifted from microscale (HEDE) to [2] X. Li, J. Zhang, Q. Fu, X. Song, S. Shen, Q. Li, Mater. Sci. Eng.
A 724, 518 (2018)
nanoscale (HELP, HESIV). However, some issues persist:
[3] X. Li, J. Zhang, S. Shen, Y. Wang, X. Song, Mater. Sci. Eng. A
(1) for the HEDE mechanism, experimental proof for the 682, 359 (2017)
quantitative relationship between the local hydrogen con- [4] X.F. Li, J. Zhang, M.M. Ma, X.L. Song, Int. J. Min. Met. Mater.
centration and atomic bonding interactions has been lacking, 23, 667 (2016)
[5] T. Neeraj, R. Srinivasan, J. Li, Acta Mater. 60, 5160 (2012)
although the data from theoretical calculations have been
[6] Z. Xu, L. Wei, T.Y. Hsu, Z. Shu, W. Li, X. Jin, Scr. Mater. 97, 21
available; (2) the role of plastic deformation on HE mech- (2015)
anism has been unclear. For example, do plastic features [7] P. Zhou, L. Wei, Z. Xu, L. Yu, C. Jian, J. Electrochem. Soc. 163,
such as tear ridges on grain boundaries contribute to brittle 160 (2016)
[8] T. Zhao, Z. Liu, X. Xu, Y. Li, C. Du, X. Liu, Corros. Sci. 157,
fracture or are the plastic features themselves the results of
146 (2019)
brittle fracture separation? (3) according to the postmortem [9] J.A. Ronevich, B.P. Somerday, C.W. San Marchi, Int. J. Fatigue
microstructural evolution characterization and fracture sur- 82, 497 (2016)
face observation, NVC mechanism and synergistic effect of [10] A. Alvaro, D. Wan, V. Olden, A. Barnoush, Eng. Fract. Mech.
219, 106641 (2019)
HEDE and HELP mechanism have been proposed. However,
[11] Y. Ogawa, H. Matsunaga, J. Yamabe, M. Yoshikawa, S. Mat-
no direct evidence supports these mechanisms; (4) all the suoka, Int. J. Hydrog. Energy 43, 20133 (2018)
existing HE mechanisms are valid only for certain materials [12] R. Wang, Corros. Sci. 51, 2803 (2009)
in certain applications and a uniform HE mechanism for all [13] E. Chatzidouros, A. Traidia, R. Devarapalli, D. Pantelis, T. Ste-
riotis, M. Jouiad, Int. J. Hydrog. Energy 43, 5747 (2018)
materials is yet to be developed.
[14] Y. Song, M. Chai, B. Yang, Z. Han, S. Ai, Y. Liu, G. Cheng, Y.
As coatings are utilized to prevent HE failure, coating Li, Materials 11, 1068 (2018)
defects, plasticity and bonding strength between the matrix [15] M.S. Bhuiyan, H. Toda, K. Shimizu, H. Su, K. Uesugi, A.
and coating should be considered. For peening treatments, Takeuchi, Y. Watanabe, Metall. Mater. Trans. A 49, 5368 (2018)
[16] S. Pallaspuro, H. Yu, A. Kisko, D. Porter, Z. Zhang, Mater. Sci.
surface residual compressive stress is recommended for
Eng. A 688, 190 (2017)
HE prevention, because hydrogen entry is suppressed by [17] J. Yamabe, M. Yoshikawa, H. Matsunaga, S. Matsuoka, Procedia
the compression of the crystal lattice. Furthermore, for Struct. Integr. 2, 525 (2016)
the strategy of preventing HE on austenite microstructure [18] S. Serebrinsky, E. Carter, M. Ortiz, J. Mech. Phys. Solids 52,
2403 (2004)
modification, austenite morphology and stabilization are
[19] Y. Wang, J. Gong, W. Jiang, Y. Jiang, J. Tang, Acta Metall. Sin.
important. In this case, film-shaped austenite should be 47, 594 (2011). (in Chinese)
preferentially selected. Meanwhile, an appropriate addition [20] V. Olden, C. Thaulow, R. Johnsen, Mater. Des. 29, 1934 (2008)
of austenite-stabilized elements such as C, N, Ni and Mn is [21] X. Xing, W. Chen, H. Zhang, Mater. Lett. 152, 86 (2015)

13
772 X. Li et al.

[22] M. Yu, X. Xing, H. Zhang, J. Zhao, R. Eadie, W. Chen, J. Been, [59] Z. Tarzimoghadam, M. Rohwerder, S.V. Merzlikin, A. Bashir,
G. Van Boven, R. Kania, Acta Mater. 96, 159 (2015) L. Yedra, S. Eswara, D. Ponge, D. Raabe, Acta Mater. 109, 69
[23] M.L. Martin, B.P. Somerday, R.O. Ritchie, P. Sofronis, I.M. Rob- (2016)
ertson, Acta Mater. 60, 2379 (2012) [60] C. Zapffe, C. Sims, Trans. AIME 145, 225 (1941)
[24] J. Song, W. Curtin, Nat. Mater. 12, 145 (2013) [61] X.C. Ren, W.Y. Zhu, J.X. Li, L.J. Qiao, B. Jiang, G. Chen, Y.H.
[25] J. Song, W.A. Curtin, Acta Mater. 68, 61 (2014) Cui, Acta Metall. Sin. 42, 153 (2006). (in Chinese)
[26] H. Yu, A. Cocks, E. Tarleton, J. Mech. Phys. Solids 123, 41 [62] C.D. Cann, E.E. Sexton, Acta Metall. 28, 1215 (1980)
(2019) [63] R. Dutton, K. Nuttall, M.P. Puls, L.A. Simpson, Metall. Trans.
[27] W. Xie, X. Liu, W. Chen, H. Zhang, Comput. Mater. Sci. 50, A 8, 1553 (1977)
3379 (2011) [64] S. P. Lynch. Progress Towards Understanding Mechanisms Of
[28] D. Xie, S. Li, M. Li, Z. Wang, P. Gumbsch, J. Sun, E. Ma, J. Li, Hydrogen Embrittlement And Stress CorrosionCracking[J]. Nace
Z. Shan, Nat. Commun. 7, 13341 (2016) International Corrosion Conference (2007)
[29] Q. Liu, A. Atrens, Corros. Rev. 31, 85 (2013) [65] L.B. Pfeil, The effect of occluded hydrogen on the tensile strength
[30] J.-G. Sezgin, C. Bosch, A. Montouchet, G. Perrin, K. Wolski, Int. of iron, in Proceedings of the Royal Society of London. Series A,
J. Hydrog. Energy 42, 15403 (2017) Containing Papers of a Mathematical and Physical Character,
[31] T.P. Perng, J.K. Wu, Mater. Lett. 57, 3437 (2003) vol. 112 (1926), pp. 182–195
[32] A. Lasia, J. Electrochem. Soc. 142, 3393 (1995) [66] S.D. Wu, L. Chen, M.Z. Liu, Acta Metall. Sin. 26, 10 (1990). (in
[33] L. Qian, A.D. Atrens, Z. Shi, K. Verbeken, A. Atrens, Corros. Chinese)
Sci. 87, 239 (2014) [67] H. Gao, W. Cao, C. Fang, E.R.D.L. Rios, Fatigue Fract. Eng.
[34] J. Venezuela, C. Tapia-Bastidas, Q. Zhou, T. Depover, K. Ver- Mater. 17, 1213 (2010)
beken, E. Gray, Q. Liu, L. Qian, M. Zhang, A. Atrens, Corros. [68] J. Lufrano, P. Sofronis, Acta Mater. 46, 1519 (1998)
Sci. 132, 90 (2017) [69] X. Wei, C. Dong, Z. Chen, K. Xiao, X. Li, RSC Adv. 6, 27282
[35] R.N. Iyer, H.W. Pickering, M. Zamanzadeh, J. Electrochem. Soc. (2016)
136, 2463 (1989) [70] W. Shuai, M.L. Martin, P. Sofronis, S. Ohnuki, N. Hashimoto,
[36] M. Devanathan, Z. Stachurski, J. Electrochem. Soc. 111, 619 I.M. Robertson, Acta Mater. 69, 275 (2014)
(1964) [71] A. Nagao, C.D. Smith, M. Dadfarnia, P. Sofronis, I.M. Robert-
[37] B. Chao, S.H. Chae, X. Zhang, K.H. Lu, J. Im, P.S. Ho, Acta son, Acta Mater. 60, 5182 (2012)
Mater. 55, 2805 (2007) [72] A. Nagao, M.L. Martin, M. Dadfarnia, P. Sofronis, I.M. Robert-
[38] K. Kiuchi, R.B. Mclellan, Acta Metall. 31, 961 (1983) son, Acta Mater. 74, 244 (2014)
[39] A. Pundt, R. Kirchheim, Annu. Rev. Mater. Res. 36, 555 (2006) [73] C.D. Beachem, Metall. Trans. 3, 441 (1972)
[40] G.M. Pressouyre, Metall. Trans. A 10, 1571 (1979) [74] I.M. Robertson, Eng. Fract. Mech. 68, 671 (2001)
[41] M. Nagumo, M. Nakamura, K. Takai, Metall. Mater. Trans. A [75] I.M. Robertson, P. Sofronis, A. Nagao, M.L. Martin, S. Wang,
32, 339 (2001) D.W. Gross, K.E. Nygren, Metall. Mater. Trans. A 46, 1085
[42] S.K. Yen, I.B. Huang, Mater. Chem. Phys. 80, 662 (2003) (2015)
[43] C.F. Dong, Z.Y. Liu, X.G. Li, Y.F. Cheng, Int. J. Hydrog. Energy [76] D.P. Abraham, C.J. Altstetter, Metall. Mater. Trans. A 26, 2859
34, 9879 (2009) (1995)
[44] H. Addach, P. Bercot, M. Rezrazi, M. Wery, Mater. Lett. 59, 1347 [77] H. Matsui, H. Kimura, A. Kimura, Strength Met. Alloys 2, 977
(2005) (1979)
[45] Y. Wang, X. Wang, J. Gong, L. Shen, W. Dong, Int. J. Hydrog. [78] J.P. Hirth, Metall. Trans. A 11, 861 (1980)
Energy 39, 13909 (2014) [79] X. Li, J. Zhang, E. Akiyama, Q. Fu, Q. Li, J. Mater. Sci. Technol.
[46] V. Ramunni, T.D.P. Coelho, P.V. de Miranda, Mater. Sci. Eng. A 35, 499 (2019)
435, 504 (2006) [80] M. Nagumo, Mater. Sci. Technol. 20, 940 (2004)
[47] X. Li, J. Zhang, Y. Wang, M. Ma, S. Shen, X. Song, Mater. Des. [81] K. Sakaki, T. Kawase, M. Hirato, M. Mizuno, H. Araki, Y. Shirai,
110, 602 (2016) M. Nagumo, Scr. Mater. 55, 1031 (2006)
[48] D. Figueroa, M. Robinson, Corros. Sci. 52, 1593 (2010) [82] M. Wen, L. Zhang, B. An, S. Fukuyama, K. Yokogawa, Phys.
[49] X. Li, J. Zhang, Q. Fu, E. Akiyama, X. Song, S. Shen, Mater. Sci. Rev. B 80, 94113 (2009)
Eng. A 742, 353 (2019) [83] J. Hou, X.S. Kong, X. Wu, J. Song, C. Liu, Nat. Mater. 18, 833
[50] J. Yamabe, T. Awane, S. Matsuoka, Int. J. Hydrog. Energy 40, (2019)
10329 (2015) [84] M.L. Martin, I.M. Robertson, P. Sofronis, Scr. Mater. 59, 3680
[51] A. Oudriss, S. Le Guernic, Z. Wang, B. Osman Hoch, J. Bouhat- (2011)
tate, E. Conforto, Z. Zhu, D.S. Li, X. Feaugas, Mater. Lett. 165, [85] M. Djukic, V.S. Zeravcic, G. Bakic, A. Sedmak, B. Rajicic, Pro-
217 (2016) cedia Mater. Sci. 3, 1167 (2014)
[52] X. Cheng, Z. Zhang, W. Liu, X. Wang, Prog. Nat. Sci. Mater. Int. [86] M. Djukic, V.S. Zeravcic, G. Bakic, A. Sedmak, B. Rajicic, Eng.
23, 446 (2013) Fail. Anal. 58, 485 (2015)
[53] X. Zhu, L.I. Wei, H. Zhao, L.I. Wang, X. Jin, Int. J. Hydrog. [87] T. Michler, Surf. Coat. Technol. 203, 1819 (2009)
Energy 39, 13031 (2014) [88] H.K.D.H. Bhadeshia, ISIJ Int. 56, 24 (2016)
[54] X. Zhu, W. Li, T. Hsu, S. Zhou, L. Wang, X. Jin, Scr. Mater. 97, [89] D. Levchuk, F. Koch, H. Maier, H. Bolt, J. Nucl. Mater. 328, 103
21 (2015) (2004)
[55] X. Zhu, W. Li, H. Zhao, L. Wang, X. Jin, Int. J. Hydrog. Energy [90] D. Figueroa, M.J. Robinson, Corros. Sci. 50, 1066 (2008)
39, 13031 (2014) [91] K. Saito, S. Inayoshi, Y. Ikeda, Y. Yang, S. Tsukahara, J. Vac.
[56] W.Y. Choo, J.Y. Lee, Metall. Trans. A 13, 135 (1982) Sci. Technol. A 13, 556 (1995)
[57] H.J. Kang, J.S. Yoo, J.I. Tae Park, S.T. Ahn, Kang. Mater. Sci. [92] K. Hiroharu, N. Hiroshi, F. Takumi, O. Tamiko, Y. Yoshihito,
Eng. A 543, 6 (2012) I. Takeshi, S. Masanori, S. Yoshiaki, Jpn. J. Appl. Phys. 57, 1
[58] J. Ovejero-García, J. Mater. Sci. 20, 2623 (1985) (2018)
[93] T. Michler, Surf. Coat. Technol. 202, 1688 (2008)

13
Review of Hydrogen Embrittlement in Metals: Hydrogen Diffusion, Hydrogen Characterization,… 773

[94] X. Liu, W. Xie, W. Chen, H. Zhang, J. Mater. Res. 26, 2735 [103] S.K. Ji, H.L. You, D.L. Lee, K.T. Park, S.L. Chong, Mater. Sci.
(2011) Eng. A 505, 105 (2009)
[95] T. Tsuchiyama, K. Tsuboi, S. Iwanaga, T. Masumura, A. [104] L.W. Tsay, M.Y. Chi, H.R. Chen, C. Chen, Mater. Sci. Eng. A
Macadre, N. Nakada, S. Takaki, Scr. Mater. 90–91, 14 (2014) 416, 155 (2006)
[96] O. Takakuwa, H. Soyama, Int. J. Hydrog. Energy 37, 5268 (2012) [105] X. Zhu, K. Zhang, W. Li, X. Jin, Mater. Sci. Eng. A 658, 400
[97] Y. Zhang, C. Zhou, W. Hui, H. Dong, J. Iron Steel Res. 26, 49 (2016)
(2014) [106] M. Wang, C.C. Tasan, M. Koyama, D. Ponge, D. Raabe, Metall.
[98] S.K. Banerji, C.J. Mcmahon, H.C. Feng, Metall. Trans. A 9, 237 Mater. Trans. A 46, 3797 (2001)
(1978) [107] J. Lee, T. Lee, Y.J. Kwon, D.J. Mun, J.Y. Yoo, S.L. Chong, Met.
[99] B. Han, Dissertation, Yanshan University (2010) Mater. Int. 22, 364 (2016)
[100] C. Zheng, L.V. Bo, F. Zhang, Z. Yan, R. Dan, L. Qian, Mater. [108] X. Shi, W. Yan, W. Wang, Y. Shan, K. Yang, Mater. Des. 92, 300
Sci. Eng. A 547, 99 (2012) (2016)
[101] T. Dieudonné, L. Marchetti, M. Wery, J. Chêne, C. Allely, P. [109] Y. Kimura, Y. Sakai, T. Hara, A. Belyakov, K. Tsuzaki, Scr.
Cugy, C.P. Scott, Corros. Sci. 82, 218 (2014) Mater. 49, 1111 (2003)
[102] T. Nanninga, Corros. Sci. 52, 1237 (2010)

13