ww.sciencedirect.com

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 1 1 8 1 0e1 1 8 1 7

Available online at w

ScienceDirect

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

Graphene coating as a protective barrier againsthydrogen embrittlement

Tae-Heum Nam a, Jung-Hun Lee a, Seok-Ryul Choi a, Ji-Beom Yoo a,b,Jung-Gu Kim a,*

a School of Advanced Materials Science and Engineering, Sungkyunkwan University, 300, Chunchun-dong,

Jangan-gu, Suwon 440746, Republic of Koreab SKKU Advanced Institute of Nanotechnology (SAINT), Sungkyunkwan University, 300, Chunchun-dong,

Jangan-gu, Suwon 440746, Republic of Korea

a r t i c l e i n f o

Article history:

Received 20 March 2014

Received in revised form

19 May 2014

Accepted 20 May 2014

Available online 14 June 2014

Keywords:

Graphene

Protective barrier

Hydrogen embrittlement

Slow strain rate tests

* Corresponding author. Tel.: þ82 31 290 736E-mail address: kimjg@skku.edu (J.-G. Kim

http://dx.doi.org/10.1016/j.ijhydene.2014.05.10360-3199/Copyright © 2014, Hydrogen Energ

a b s t r a c t

The applicability of a graphene coating as a protective barrier against hydrogen embrit-

tlement was studied. To simulate the hydrogen embrittlement, complex environment of

tensile stress with simultaneous hydrogen charging was applied. The strain at fracture,

ductility and ultimate tensile strength of graphene-coated copper under the charged

condition were preserved above 95% comparing uncharged bare copper. After hydrogen

charging for 12 h, the hydrogen content in graphene-coated copper was lower than that in

bare copper. Using attenuated total reflectance infrared spectroscopy and Raman spec-

troscopy, it was verified that graphene can interrupt the hydrogen penetration by the

formation of CeH sp3 bonds. Unfortunately, it induced a distortion of graphene structure,

which increased the defects in the graphene. Nevertheless, the graphene coating is ex-

pected to decrease the hydrogen embrittlement susceptibility of metal substrate.

Copyright © 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights

reserved.

Introduction

The effect of hydrogen on themechanical properties ofmetals

is known as hydrogen embrittlement (HE), which can cause

catastrophic failures [1]. HE has been an important issue in

various applications, such as reactor vessels in nuclear plants

[2,3], high-pressure gaseous hydrogen (HPGH2) storage tanks

[4,5], and pipelines for natural gas and petroleum [6,7]. There

are two mechanisms of HE in metals. One is that the atomic

hydrogen is attracted to crack tips, and reduces the fracture

energy while encouraging cleavage-like failure [8,9]. The other

mechanism involves atomic hydrogen enhancing themobility

of dislocations through an elastic shielding effect, causing

0; fax: þ82 31 290 7371.).

32y Publications, LLC. Publ

locally reduced shear strength, and eventually enhancing the

local plasticity [10]. Although there is a difference in reaction

process of hydrogen, it is distinct that HE is induced by

interaction between internal hydrogen and external tensile

stress. Therefore, in order to reduce the potential risk of

hydrogen embrittlement, or to protect metal from hydrogen

permeation, several methods have been proposed, including

heat treatment [11], neon/helium glow discharge [12], in-

hibitors of hydrogen permeation [13], and protective barriers

of zirconium dioxide [14]. However, the limitations of these

methods include bulky thickness, toxic substance usage,

relatively low inhibition efficiency, and detrimental effects on

the matrix materials. Recently, we report new methods for

eliminating the hydrogen inside low alloy steel by

ished by Elsevier Ltd. All rights reserved.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 1 1 8 1 0e1 1 8 1 7 11811

electrochemical discharge [15] and electrical field [16]. Devel-

opment of new method, which can fundamentally protect

material from hydrogen permeation, is still required.

Graphene possesses various unique properties that are

suitable for protective barriers of metal against hydrogen

permeation. Graphene is impermeable to gas molecules [17].

Hydrogen is also adsorbed on graphene surfaces as sp3 CeH

bonds [18e20], which enables the viability of graphene for

hydrogen storage [21,22]. Additionally, graphene is chemically

stable in ambient atmosphere at up to 400 �C [23], and can

protect the underlying metal substrate from corrosion [24,25]

and oxidation [26]. Also, growth techniques for large-area

graphene film are being developed, which will enable the

commercial applicability of graphene [27]. However, graphene

has not been studied as a protective barrier against HE.

Especially, under applied stress and hydrogen charge simu-

lating HE condition, the effect of graphene on the mechanical

properties of bulky metal has not also been investigated.

In this study, the applicability of graphene as a protective

barrier against HE is investigated. Graphenewas coated onto a

copper substrate by chemical vapor deposition (CVD). The

mechanical properties of specimens were assessed by slow

strain rate tests (SSRTs) under tensile stress with simulta-

neous hydrogen charging. In order to quantitatively evaluate

the susceptibility of specimen to the HE, the hydrogen

embrittlement ratio (HER, %) was calculated. The actual pro-

tective efficiency of graphene for hydrogen permeation was

evaluated by measuring the content of hydrogen inside of a

specimen using gas chromatography (GC). The effect of gra-

phene on the hydrogen evolution reaction on the specimen

surfacewas investigated by cathodic polarization tests. Lastly,

a protective mechanism of graphene against HE is proposed

and verified by attenuated total reflectance infrared spec-

troscopy (ATR-IR) and Raman spectroscopy.

Material and methods

Graphene preparation

For the synthesis of graphene by chemical vapor deposition

(CVD), copper foil (70 mm, purity 99.9%) was placed into a

quartz tube. The specimen was annealed at 1050 �C under H2.

A gas mixture of CH4 (20 sccm) and H2 (10 sccm) was flowed

into the quartz tube for the growth of graphene under 10 Torr.

After 30 min, the chamber was cooled down to room tem-

perature under H2.

Hydrogen charging

Cathodic charging is used to introduce hydrogen into a spec-

imen. The use of so-called poisons in the electrolyte is applied

to hinder the H2 formation and to consequently improve the

hydrogen absorption. Arsenic-based poisons such as As2O3

are used in sulfuric acid during hydrogen charging [28]. Before

the experiment, nitrogen gaswas bubbled into the solution for

2 h to exclude the influence of oxygen. To charge the

hydrogen, the galvanostatic test is performed at �1.0 mA/cm2

for 12 h. A solution of 0.5 M sulfuric acid (H2SO4) þ 250 mg/L

arsenic trioxide (As2O3) (pH ¼ 1.0) was used to control the

hydrogen formation and to ease the percolation and diffusion

of the hydrogen atoms into the specimen.

Slow strain rate tests (SSRTs)

To identify the effect of graphene as a hydrogen diffusion

barrier on the mechanical properties of a specimen, slow

strain rate tests (SSRTs) are conducted at a constant strain rate

of 1.0 � 10�6/s. This test is extensively used to evaluate the

resistance of a material to HE and stress corrosion cracking

[29,30]. A schematic picture of the test setup was presented

(Fig. 1a). Cylindrical tensile specimens are fabricated accord-

ing to NACE Standard TM0177 and have threaded ends with a

diameter of 6.35mmand a gauge length of 25.4mm. Except for

the gauge length, the specimens are coated with insulating

lacquer to make the exposed surface areas identical. The

specimen is connected to pull-rods, and the load and elon-

gation are monitored continuously by a load cell and a linear

variable differential transformer (LVDT) until fracture occurs.

During the SSRT, hydrogen is charged until failure occurs. By

applying complex conditions of hydrogen charge and tensile

stress, the susceptibility of the metal to HE can be evaluated.

Hydrogen content measurement

Hydrogen is electrochemically charged for 12 h, and then the

amount of hydrogen accumulated inside the specimen is

measured by gas chromatography in the temperature range of

50e500 �C. The hydrogen charged specimens are introduced to

a thermostat-controlled furnace and held under inert argon

purge. A constant heating rate of 10 �C/min is applied, and the

amount of hydrogen evolved is measured.

Cathodic polarization

An EG&G PAR Model 273A potentiostat was used for electro-

chemical polarization tests which were conducted using a

conventional three-electrode system. The test specimen was

used as the working electrode, a saturated calomel electrode

(SCE) was used as the reference electrode, and graphite was

used as an auxiliary electrode. A solution of 0.5 M sulfuric acid

(H2SO4) þ 250 mg/L arsenic trioxide (As2O3) (pH ¼ 1.0) was

used. Before the experiment, nitrogen gas was bubbled into

the solution for 2 h to exclude the influence of oxygen.

Cathodic polarization tests were performed at a scanning rate

of 0.166 mVSCE from þ50 mVOCP to �1200 mVSCE.

Characterization of graphene

The hydrogen charged and uncharged graphene was trans-

ferred onto Si wafer (SiO2(300 nm)/Si) by the polymethyl

methacrylate (PMMA) for Raman spectroscopy and atomic

force microscopy (AFM) analyses. First, the PMMA was coated

on graphene/Cu samples as a supporting layer. The copper foil

was etched by copper etchant (SigmaeAldrich Korea Ltd.,

3e4% of HCl, 30% of FeCl3, 66e67% of water) and the PMMA/

graphene transferred onto the wafer. Finally, the PMMA was

removed by acetone. Attenuated total reflectance infrared

(ATR-IR) spectra were recorded with an IFS 66/S FTIR spec-

trometer (Bruker) equipped with a Harrick Scientific

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 1 1 8 1 0e1 1 8 1 711812

horizontal reflection Ge-attenuated total reflection accessory

and the spectrum was collected for 64 scans. Raman spec-

troscopy and mapping (WiTec, Alpha 300 M) with an excita-

tion wavelength of 532 nm were used to measure the

crystallinity and binding with hydrogen.

Results and Discussion

Slow strain rate tests (SSRTs)

The stressestrain curve measured in the SSRT is shown

(Fig. 1b). The behavior of charged bare copper is distinctly

different, but that of graphene-coated copper is similar to that

of uncharged bare copper. Under hydrogen charging, all of

mechanical properties of graphene-coated copper are higher

than those of bare copper. Detailed results of SSRT, such as

time to fracture, ductility and ultimate tensile strength, are

presented (Table S1). The strain at failure is used for assessing

the HE resistance, and is presented (Fig. 1c). It was clearly

Fig. 1 e The results of specimen by slow strain rate tests. (a) sc

presented and it is composed of (1) a saturated calomel referen

tensile specimens (with a diameter of 6.35 mm and a gauge len

corrosion cell containing an electrolyte. (b) the strain-stress cur

embrittlement ratio (HER, %).

declined as a result of the adverse effect of the hydrogen

accumulated inside the metal. The fracture strain of the

hydrogen-charged copper (45.5%) decreased compared to that

of uncharged copper (50.6%). However, the fracture strain of

hydrogen-charged graphene-coated copper (49.1%) is similar

to that of uncharged copper. In the study of graphene fracture,

monolayer graphene canbe elongated at the strain of 25%until

fracture [31]. From these results, it is supposed that hydrogen

permeation can be effectively blocked by graphene coating

until the strain of 25%. After that the cracked graphene coating

reduces the surface area of underlying copper exposed to

hydrogen, which may decrease the amount of hydrogen

entering. Therefore, the strain at fracture of graphene-coated

copper is slightly lower than that of uncharged bare copper.

In order to quantitatively evaluate the susceptibility of a

specimen to the HE, the hydrogen embrittlement ratio (HER,

%) is calculated by the following Eq. (1) [32]:

Hydrogeneembrittlement ratio ðHER;%Þ¼�

3O� 3H

3O

��100 (1)

hematic of hydrogen embrittlement test equipment is

ce electrode, (2) a graphite counter electrode, (3) cylindrical

gth of 25.4 mm), (4) an SSRT extending specimen, and (5) a

ves (c) the strain (%) at failure and (d) hydrogen

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 1 1 8 1 0e1 1 8 1 7 11813

where 3o is the strain of the uncharged specimen at failure,

and 3H is the strain of the hydrogen-charged specimen at

failure. The hydrogen embrittlement ratio (HER) is presented

(Fig. 1d), where a higher HER signifies higher susceptibility of

the metal to HE. The HER of hydrogen-charged graphene-

coated copper (3.0%) is much lower than that of hydrogen-

charged copper (10.1%). In other words, the susceptibility of

copper to HE is reduced by using graphene coating. Mean-

while, fracture type of all specimens is not brittle. Due to the

low diffusivity of hydrogen in face-centered cubic (FCC) crys-

tal structure, hydrogen can be only trapped near the charging

surface in case of room temperature charging [33]. Therefore,

fracture type of copper is not changed from ductile to brittle

until HER of 30% due to the low susceptibility of copper to HE

[32]. However, it is obvious that the susceptibility of copper is

distinctly declined due to the graphene barrier. The results of

fracture strain and HER confirm that graphene acts as an

excellent hydrogen permeation barrier under tensile stress.

Hydrogen content measurement

The actual protective efficiency of graphene for hydrogen

permeation is evaluated by measuring the amount of

hydrogen inside of the specimen. Hydrogen is electrochemi-

cally charged for 12 h, and then the amount of hydrogen

accumulated in the specimen is measured by gas chroma-

tography in the temperature range from 50 to 500 �C. The

accumulated hydrogen is easily released by heat treatment,

which is generally used for eliminating hydrogen from ma-

terials [34]. In a recent study on graphene as a hydrogen

storage material, the hydrogen absorbed by the graphene was

dehydrogenated in the temperature range from 200 to 500 �C[18,35].

The hydrogen content released with temperature is pre-

sented (Fig. 2a). In the whole temperature range, the hydrogen

content of bare copper was higher than that of graphene-

coated copper. The maximum peak of the hydrogen content

appeared at temperatures of 200 and 450 �C, corresponding to

the diffusional (reversibly) and trapped (irreversibly) hydrogen

Fig. 2 e Hydrogen is electrochemically charged for 12 h, and then

is measured by gas chromatography in the temperature range

temperature and (b) the total amount of hydrogen released.

in the metal, respectively. The absorbed hydrogen diffuses

toward the interior of the material and a portion of hydrogen

was retained in various preferential locations. For example,

the lattice defects (grain boundaries, vacancies, dislocations)

and precipitates (carbide) provide a variety of trapping sites

[36]. The diffusional hydrogen is easily released by relatively

low thermal energy, but comparatively high thermal energy is

required to release the trapped hydrogen from the material.

Therefore, the diffusional hydrogen is released at relatively

low temperature, but the trapped hydrogen can be evolved

after the critical temperature reached. The total amount of

hydrogen released from graphene-coated copper (0.25 mmol/g)

was much lower than that from bare copper (0.59 mmol/g)

(Fig. 2b). Unfortunately, the protective efficiency of the

monolayer graphene is not perfect, but it is expected that the

efficiencymay be enhanced bymultiple layers of graphene, or

optimization of graphene growth. The results of hydrogen

measurement confirm that graphene is a proper barrier for

hydrogen penetration, and effectively declines the suscepti-

bility of copper to HE.

Cathodic polarization

Hydrogen can be electrochemically generated during the

cathodic reaction of corrosion, electroplating, and cathodic

protection, which increase the potential risk of HE. Hydrogen

is produced by cathodic reduction reactions of water and acid

by reactions:

H2Oþ e�/HþOH� (2)

Hþ þ e�/H (3)

Most of the hydrogen atoms are recombined as the mo-

lecular hydrogen (H2), but a portion of atomic hydrogen can

diffuse into the lattice and retained in the material. These

reactions can be occurred below the equilibrium potential ðEOHÞ

given by the Nernst Equation [28,37]:

EOHðmVSHEÞ ¼ �59 pH� 29:5 log PH2

(4)

the amount of hydrogen accumulated inside the specimen

from 50 to 500 �C. (a) the released hydrogen content with

Fig. 3 e The effect of graphene on the hydrogen evolution

reaction is investigated by cathodic polarization tests. (a)

the measured cathodic polarization curve and (b)

schematic cathodic polarization curve of hydrogen

evolution reaction.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 1 1 8 1 0e1 1 8 1 711814

At pH ¼ 1, corresponding to the electrolyte pH in the pre-

sent study, the hydrogen equilibrium potential is �59 mVSHE

(�300 mVSCE) for a hydrogen partial pressure PH2of 1 atm. The

amount of hydrogen (N) produced per second per unit area

(cm2) is given by Faraday's law:

N ¼ ired;H�F (5)

ired;H ¼ iO;H exp�� 2:3

�E� EO

H

��bred;H

�(6)

where ired,H is the current density induced by cathodic

reduction reaction, iO,H is the exchange current density (cur-

rent density at equilibrium potential Ecorr), bred,H is the Tafel

constant (slope of cathodic polarization curve), E is any po-

tential lower than EOH, and F is the Faraday constant

(96487 Coulomb/mole). This equation indicates that hydrogen

production at the specimen surface occurs at higher rates as

the current density (ired,H) increases and the potential (E) be-

comes progressively less than EOH.

The measured cathodic polarization curve and the sche-

matic cathodic reduction behavior in a deaerated acid solution

(pH ¼ 1) is shown (Fig. 3). Based on the cathodic polarization

curves (Fig. 3b), the hydrogen reduction reactiondominated the

cathodic reactions below the potential of �300 mVSCE. In the

whole potential range (Fig. 3a), the hydrogen production rate

(current density) on graphene-coated copper is much lower

than that on bare copper. In a recent study on graphene as

corrosion-inhibiting coating, it is reported that graphene acts

primarily as an inhibitor of the cathodic reactions [25]. Also, it

was revealed that a graphene monolayer forms a charge

transfer barrier between the solution and the graphene mono-

layer on the copper surface, which suppresses the corrosion

reaction [24]. The charge transfer resistance of specimen is

evaluated by electrochemical impedance spectroscopy (Fig. S1

and Table S2). The charge transfer resistance of graphene

coated-copper is much higher than that of bare copper. From

these results, it is confirmed that graphene acts as a charge

transfer barrier for the hydrogen evolution reaction. Therefore,

the graphene coating can improve the resistance against HE.

Characterization of graphene

In a recent study on graphene as a hydrogen storage material,

it was revealed that hydrogen is absorbed by graphene as CeH

sp3 bonds, which enables the hydrogen storage [18e20]. As a

hydrogen barrier, it may be proposed that graphene interrupts

hydrogen permeation by the interaction with hydrogen (for-

mation of CeH sp3 bonds). This proposed barrier mechanism

of graphene was verified through the CeH stretching bands of

ATR-IR spectra and the bands of Raman spectra. The IR

spectra of the hydrogenated (hydrogen charged) graphene

clearly show aliphatic CeH stretching bands in the region of

2850e2950 cm�1, but they disappear for the dehydrogenated

graphene [18]. ATR-IR spectra of pristine graphene and

hydrogen-charged graphene were shown (Fig. 4a). There was

no CeH stretching band (region of 2750e3000 cm�1) for pris-

tine graphene, but it was distinctly appeared for hydrogen-

charged graphene.

The pristine graphene synthesized on copper foil by

chemical vapor deposition is predominantly single-layer

graphene, which is confirmed by the ratio of the G band peak

(1580 cm�1) and the 2D band peak (2680 cm�1) in Fig. 4b. The D

peak represents the disorder in the graphene and the D0 peakoccurs due to the presence of defects in the graphene [38e40].

The D peak (1350 cm�1), D0 peak (1620 cm�1), and a combina-

tionmode of these peaks (Dþ D0 lie at 2950 cm�1) appeared for

hydrogen-charged graphene. When hydrogen absorbed onto

graphene, geometrical deformation of the hexagonal struc-

ture was induced, which increase disorders and defects in the

graphene [38,41]. During hydrogen charging, most of the

hydrogen exists as a molecule and is physisorbed above the

graphene. However, a portion of the hydrogen is retained as

atomic hydrogen, and absorbed hydrogen on top of the gra-

phene breaks p bonds and produces additional s bonds, which

induces transition from C]C to CeH and CeC bonds by

forming sp3 hybridization bonds [41e43]. In order to under-

stand why the atomic hydrogen breaks p bonds on the gra-

phene sheet and produces additional s bonds, we simply

calculated the bonding energy (BE) by comparing which is

more stable between graphene and hydrogenated graphene.

Fig. 4 e Hydrogen adsorption on graphene was verified through the CeH stretching bands of ATR-IR spectra and the bands

of Raman spectra. (a) the region of CeH stretching bands (2750e3000 cm¡1 ATR-IR spectra) (b) the Raman spectra of the

pristine and hydrogen-absorbed graphene and (c) the change of energy depending on reaction processes.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 1 1 8 1 0e1 1 8 1 7 11815

For the simple comparison, graphene was assumed that the

hexagonal system consists of three CeC bonds (s bond, BE

3.596 eV) and three C]C bonds (s bond and p bond, BE

6.239 eV). The BE of hydrogenated graphene system consists of

six CeC bonds and three CeH bonds (BE 4.280 eV) [44]. The

change of free energy is calculated by:

DG¼X

BEðgrapheneÞ�X

BEðhydrogenated�grapheneÞ (7)

DG¼ ð3�3:596þ3�6:239Þ� ð6�3:596þ3�4:280Þ ¼�4:911eV

(8)

Based on this result, the hydrogenated graphene is ener-

getically favorable and thus atomic hydrogen can be chemi-

cally absorbed on graphene. Recent studies on adsorption of

hydrogen onto graphene have reported that absorbed

hydrogen forms hydrogen clusters [45,46] and cover gra-

phene up to about 75% of surface [46,47]. The energy barrier

depending on reaction process (penetration through gra-

phene or absorption on graphene) is schematically shown

(Fig. 4c). Passing through a center of hexagonal structure in

graphene, above 2.89 eV is required because there is a

repulsive force induced by strong electron cloud of graphene

[48,49]. It is much larger than the energy barrier of CeH

forming (0.18 eV) [50]. Hence, hydrogen penetration through

graphene is much harder than CeH sp3 formation. From

these experimental results and theoretical calculation, it can

be proposed that the charged hydrogen is combined with

graphene as CeH bonds, which interrupts the hydrogen

penetration into copper.

Conclusion

In this study, the applicability of graphene as a protective

barrier against HE was verified. To simulate the hydrogen

embrittlement (HE), complex condition of tensile stress with

simultaneous hydrogen charging was applied. The suscepti-

bility of copper to HE is drastically depleted due to graphene

protective coating, but protection effect was disappeared after

graphene have been strained above 25%. The actual protective

efficiency of graphene for hydrogen permeation is evaluated

bymeasuring the amount of hydrogen inside of the specimen.

The hydrogen permeation is sufficiently suppressed, but not

perfect due to the defects of graphene. Through the charac-

terization of hydrogen absorbed graphene, it was identified

that graphene can effectively protect the hydrogen penetra-

tion by formation of CeH bonds.

i n t e rn a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 1 1 8 1 0e1 1 8 1 711816

Acknowledgments

This work was supported by the National Research Founda-

tion of Korea (NRF) grant funded by the Korea government

(MEST) (No.NRF-2012R1A2A2A03046671).

Appendix A. Supplementary data

Supplementary data related to this article can be found at

http://dx.doi.org/10.1016/j.ijhydene.2014.05.132.

r e f e r e n c e s

[1] Johnson WH. On some remarkable changes produced in ironand steels by the action of hydrogen acids. Proc R Soc Lond1875;23:168e75.

[2] Harries DR, Broomfield GH. Hydrogen embrittlement of steelpressure vessels in pressurized water reactor systems. J NuclMater 1963;9:327e38.

[3] Lucas GE. An evolution of understanding of reactor pressurevessel steel embrittlement. J Nucl Mater 2010;407:59e69.

[4] Zheng J, Liu X, Xu P, Liu P, Zhao Y, Yang J. Development ofhigh pressure gaseous hydrogen storage technologies. Int JHydrogen Energy 2012;37:1048e57.

[5] Takasawa K, Wada Y, Ishigaki R, Kayano R. Effects of grainsize on hydrogen environment embrittlement of highstrength low alloy steel in 45 MPa gaseous hydrogen. MaterTrans 2010;51:347e53.

[6] Capelle J, Gilgert J, Dmytrakh I, Pluvinage G. Sensitivity ofpipeline steel API X52 to hydrogen embrittlement. Int JHydrogen Energy 2008;33:7630e41.

[7] Hadianfard MJ. Failure in a high pressure feeding line of anoil refinery due to hydrogen effect. Eng Fail Anal2010;17:873e81.

[8] Troiano AR. The role of hydrogen and other interstitials inthe mechanical behavior of metals. Trans ASM1960;52:54e80.

[9] Song J, Curtin WA. Atomic mechanism and prediction ofhydrogen embrittlement in iron. Nat Mater 2013;12:145e51.

[10] Birnbaum HK, Sofronis P. Hydrogen-enhanced localizedplasticity-mechanism for hydrogen-related fracture. MaterSci Eng A 1994;176:191e202.

[11] Kim SJ, Jung HG, Kim KY. Effect of post-weld heat treatmenton hydrogen-assisted cracking behavior of high-strengthprocess pipe steel in a sour environment. Scr Met2012;67:895e8.

[12] Yamauchi Y, Takeda K, Nobuta Y, Hino T. Hydrogen andhelium removal retained in stainless steel by neon glowdischarge. J Nucl Mater 2009;390-391:1048e50.

[13] Elboujdaîni M, Sastri VS, Perumareddi JR. Studies oninhibition of hydrogen-induced cracking of linepipe steels.Corrosion 2006;62:29e34.

[14] Voloshchuk I, Zakroczymski T. Hydrogen entry andabsorption in ZrO2 coated iron studied by electrochemicalpenetration and desorption techniques. Int J HydrogenEnergy 2012;37:1826e35.

[15] Nam TH, Kim JG, Choi YS. Electrochemical hydrogendischarge of high-strength low alloy steel for high-pressuregaseous hydrogen storage tank: effect of dischargingtemperature. Int J Hydrogen Energy 2013;38:999e1003.

[16] Park JS, Nam TH, Kim JS, Kim JG. Effect of electrotransporttreatment on susceptibility of high-strength low alloy steelto hydrogen embrittlement. Int J Hydrogen Energy2013;38:12509e15.

[17] Bunch JS, Verbridge SS, Alden JS, Zande AM, Parpia JM,Craighead HG, et al. Impermeable atomic membranes fromgraphene sheets. Nano Lett 2008;8:2458e62.

[18] Subrahmanyam KS, Kumar P, Maitra U, Govindaraj A,Hembram KPSS, Waghmare UV, et al. Chemical storage ofhydrogen in few-layer graphene. Proc Natl Acad Sci USA2011;108:2674e7.

[19] Balog R, Jørgensen B, Nilsson L, Andersen M, Rienks E,Bianchi M, et al. Bandgap opening in graphene inducedby patterned hydrogen adsorption. Nat Mater2010;9:315e9.

[20] Andersen M, Hornekær L, Hammer B. Graphene on metalsurface and its hydrogen adsorption: a meta-GGA functionalstudy. Phys Rev B 2012;86:085405.

[21] Tozzini V, Pellegrini V. Reversible hydrogen storage bycontrolled buckling of graphene layers. J Phys Chem C2011;115:25523e8.

[22] Tozzini V, Pellegrini V. Prospects for hydrogen storage ingraphene. Phys Chem Chem Phys 2013;15:80e9.

[23] Liu L, Ryu S, Tomasik MR, Stolyarova E, Jung N,Hybertsen MS, Steigerwald ML. Graphene oxidation:thickness-dependent etching and strong chemical doping.Nano Lett 2008;8:1965e70.

[24] Prasai D, Tuberquia JC, Harl RR, Jennings GK, Bolotin KI.Graphene: corrosion-inhibiting coating. ACS Nano2012;6:1102e8.

[25] Kirkland NT, Schiller T, Medhekar N, Birbilis N. Exploringgraphene as a corrosion protection barrier. Corros Sci2012;56:1e4.

[26] Chen SS, Brown L, Levendorf M, Cai WW, Ju SY,Edgeworth J,Li X, et al. Oxidation resistance of graphene-coated Cu andCu/Ni alloy. ACS Nano 2011;5:1321e7.

[27] Bae SK, Kim HK, Lee YB, Xu XF, Park JS, Zheng Y. Roll-to-rollproduction of 30-in. graphene films for transparentelectrodes. Nat Nanotechnol 2010;5:574e8.

[28] ASTM G129. Standard practice for slow strain rate testing toevaluate the susceptibility of metallic materials toenvironmentally assisted cracking. Pennsylvania: ASTMInternational; 2013.

[29] Cramer SD, Covino BS. ASM handbook volume 13A:corrosion: fundamentals, testing, and protection. Ohio: ASMInternational; 2003.

[30] Lee CG, Wei X, Kysar JW, Hone J. Measurement of the elasticproperties and intrinsic strength of monolayer graphene.Science 2008;321:385e8.

[31] Panagopoulos CN, Zacharopoulos N. Cathodic hydrogencharging and mechanical properties of copper. J Mater Sci1994;29:3843e6.

[32] Gibala R, Hehemann RF. Hydrogen embrittlement and stresscorrosion cracking. Ohio: ASM International; 1985.

[33] Gesnouin C, Hazarabedian A, Bruzzoni P, Ovejero-García J,Bilmes P, Llorente C. Effect of post-weld heat treatment onthe microstructure and hydrogen permeation of 13CrNiMosteels. Corros Sci 2004;46:1633e47.

[34] Ryu SM, Han MY, Maultzsch J, Heinz TF, Kim P,Steigerwald ML, et al. Reversible basal plane hydrogenationof graphene. Nano Lett 2008;8:4597e602.

[35] Pressouyre GM. A classification of hydrogen traps in steel.Metal Trans A 1979;10:1571e3.

[36] Jones Jones DA. Principles and prevention of corrosion. 2nded. New Jersey: Prentice Hall; 1996.

[37] Stansbury EE, Buchanan RA. Fundamentals ofelectrochemical corrosion. Ohio: ASM International; 2000.

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n en e r g y 3 9 ( 2 0 1 4 ) 1 1 8 1 0e1 1 8 1 7 11817

[38] Elias DC, Nair RR, Mohiuddin TMG, Morozov SV, Blake P,Halsall MP, et al. Control of graphene's properties byreversible hydrogenation: evidence for graphene. Science2009;323:610e3.

[39] Ferrari AC. Raman spectroscopy of graphene and graphite:disorder, electronephonon coupling, doping andnonadiabatic effects. Solid State Commun 2007;143:47e57.

[40] Ferrari AC, Robertson J. Interpretation of Raman spectra ofdisordered and amorphous carbon. Phys Rev B2000;61:14095e107.

[41] Boukhvalov DW, Katsnelson MI. Tuning the gap in bilayergraphene using chemical functionalization: densityfunctional calculations. Phys Rev B 2008;78:085413.

[42] Arellano JS, Molina LM, Rubio A, Alonso JA. Densityfunctional study of adsorption of molecular hydrogen ongraphene layers. J Chem Phys 2000;112:8114e9.

[43] Casolo S, Lovvik OM, Martinazzo R, Tantardini GF.Understanding adsorption of hydrogen atoms on graphene. JChem Phys 2009;130:054704.

[44] Whitten KW, Davis RE, Peck ML. General chemistry.Philadelphia: Saunders College Publishing; 1999.

[45] Hornekaer L, Sljivancanin Z, Xu W, Otero R, Rauls E,Stensgaard I, et al. Metastable structures and recombinationpathways for atomic hydrogen on the graphite (0001)surface. Phys Rev Lett 2006;96:156104.

[46] Hornekær L, Rauls E, Xu W, Sljivancanin Z, Otero R,Stensgaard I, et al. Clustering of chemisorbed H(D) atoms onthe graphite (0001) surface due to preferential sticking. PhysRev Lett 2006;97:186102.

[47] Balog R, Jørgensen B, Wells J, Lægsgaard E, Hofmann P,Besenbacher F, et al. Atomic hydrogen adsorbate structureon graphene. J Am Chem Soc 2009;131:8744e5.

[48] Miao M, Nardelli MB, Wang Q, Liu Y. First principles study ofthe permeability of graphene to hydrogen atoms. Phys ChemChem Phys 2013;15:16132e7.

[49] Nechaev YS, Veziroglu TN. Thermodynamic aspects of thegraphene/graphane/hydrogen systems: relevance to thehydrogen on-board storage problem. Adv Mater Phys Chem2013;3:255e80.

[50] Miura Y, Kasai H, Di~no WA, Nakanishi H, Sugimoto T.Effective pathway for hydrogen atom adsorption ongraphene. J Phys Soc Jpn 2003;72:995e7.