Vacuum 129 (2016) 79e85

Contents lists available at ScienceDirect

Vacuum
journal homepage: www.elsevier.com/locate/vacuum

High strength metallurgical graphene for hydrogen storage

nanocomposites
Piotr Kula, Witold Szymanski, Lukasz Kolodziejczyk*, Radomir Atraszkiewicz,
Jacek Grabarczyk, Marian Clapa, Lukasz Kaczmarek, Anna Jedrzejczak, Piotr Niedzielski
Institute of Materials Science and Engineering, Lodz University of Technology, Stefanowskiego 1/15, 90-924, Lodz, Poland

a r t i c l e i n f o a b s t r a c t

Article history: This work presents the results of an investigation on the atomic hydrogen chemisorption of metallurgical
Received 23 March 2016 graphene enhanced by plasma discharge at low pressures in the range of 5e20 Pa. It was experimentally
Accepted 16 April 2016 revealed that hydrogenated graphene is not stable when in contact with humid oxygen or humid air.
Available online 19 April 2016
Moreover, thermal desorption experiments showed that hydrogen can be released from hydrogenated
graphene at relatively low temperatures. Additionally, a numerical simulations were performed of the
Keywords:
pressure changes inside the nanogaps of hydrogen storage nanocomposites under various durations of
Metallurgical graphene
the desorption cycle. Experimental fatigue resistance tests performed along with simultaneous electrical
Hydrogen storage
Sorption
resistance measurements of the graphene sheets were also conducted. Metallurgical graphene shows
Tensile test tensile fatigue performances that might be sufficient to resist the pressure pulses and strain that arise in
Electrical resistance the gaps of graphene-based hydrogen storage systems.
© 2016 Elsevier Ltd. All rights reserved.

1. Introduction the adsorption capacity of hydrogen strongly depends on the dis-

Graphene-based nanomaterials appear promising for hydrogen-
based fuel systems, as its theoretically estimated surface specific
area (~2630 m2/g) is much higher than that of other carbon
structures providing a favorable platform for surface chemistry
[1,2]. Other structural features of graphene (e.g., its tunable
porosity, light weight and high thermal and chemical stability)
could allow for the integration of a hydrogen-storage module into
flexible and light graphene-based devices.
Three mechanisms of hydrogen sorption in graphene-based
materials are considered: physisorption (interaction is caused by
van der Waals force), chemisorption (involving a chemical reaction
between the graphene and the hydrogen atom) and electro-
chemical sorption (an accumulation of ions or organic substances
on the interface metal-solution as a result of electrostatic in-
teractions and chemical adsorption). Recently, hydrogen storage in
carbon nanomaterials is mostly based on physisorption on the
surfaces and depends mainly on the electrostatic and dispersion
interactions [3e8]. Physisorption usually happens with hydrogen in
the molecular form. According to the work of Patchkovskii et al. [9],
* Corresponding author.
E-mail address: lukasz.kolodziejczyk@p.lodz.pl (L. Kolodziejczyk).
tances between the carbon planes within graphene. The theoretical
physical adsorption of a monolayer of H2 on a single side of the
graphene surface can lead to approximately 3 wt% of H2 (H/
C ¼ 0.18) [10]. However, from a thermodynamic viewpoint, it is
impossible for H2 molecules to penetrate between the graphene
layers of graphite. The best results in hydrogen storage are achieved
by maintaining the distance between the graphene planes in the
range of 4e6 Å. A maximum of ~3.3 wt% H2 is accomplished with
one monolayer of adsorption between two graphene layers sepa-
rated by 6 Å, for which up to 16 kJ/mol binding energy is obtained
[11,12]. The binding energy decreases for larger interlayer spacing
due to the second graphene layer. Therefore, it is important to build
pillared 3D constructions between graphene sheets to provide the
optimum interlayer spacing (e.g., Li dopants, in which interlayer
spacing of approximately 6 Å with Li:C ¼ 1:6 provides over 3 wt%
H2 adsorption). This interlayer spacing [9] or pore width [13] per-
mits H2 molecules to interact with both sides of the monolayer
graphene. Therefore, the spatial distribution of molecular hydrogen
adsorption on the graphene plane is delocalized. The H2 interaction
with graphene in bulk graphite is a localized phenomenon;
therefore, adsorption is generally restricted to the external gra-
phene sheet [14]. One of the main disadvantages of physisorption in
the process of hydrogen storage is the very weak interaction energy
~0.01e0.1 eV, and physisorption can only be observed for low

http://dx.doi.org/10.1016/j.vacuum.2016.04.017
0042-207X/© 2016 Elsevier Ltd. All rights reserved.
80 P. Kula et al. / Vacuum 129 (2016) 79e85
temperature and high pressure conditions, thereby significantly
limiting the potential application range.
Due to the presence of free electrons in the carbon atom,
reversible sorption of hydrogen could be more effective by chem-
isorption. A significant decrease of hydrogen volume and pressure
occurs during its chemisorption as the result of the change in the
chemical bond from HeH toward CeH. The chemisorption of mo-
lecular hydrogen on graphene presents a quantum barrier of
approximately ~1.5 eV [15] because it requires the dissociation of
H2. From a hydrogen storage point of view, the maximum achiev-
able gravimetric density in graphene through chemisorption is
8.3%, forming of a completely saturated graphene sheet, with 1:1 C
vs. H stoichiometry. Hydrogenated graphene (graphane) contains
covalently bonded hydrogen atoms to carbon atoms; therefore, the
sp3 CeH bonds on the basal plane are essentially different from the
physical H2 absorbed graphene structures. Furthermore, spectro-
scopic studies exhibit the appearance of sp3 CeH bonds, and the
hydrogen can be released upon heating or UV/laser irradiation,
resulting in a sudden increase of gas volume during the hydrogen
desorption step [16].
Various synthesis methods, such as plasma, CVD and liquid and
gas phase reactions, have been developed to obtain hydrogenated
graphene. Subrahmanyam et al. demonstrated that it could be used
as a chemical H2 storage material [17] reaching experimental
hydrogen concentrations of up to 7.4 wt% [18e20]. Goler et al. [21]
showed that by controlling the curvature of corrugated graphene
sheets, an appropriate material for hydrogen storage applications
could be created. Taking the above references into consideration, a
state-of-the-art graphene nanocomposite structure GraphRoll has
been proposed [22]. GraphRoll and other nanocomposite structures
based on graphene must provide good physical and mechanical
properties (close to the theoretical ones) to be considered efficient
and reliable materials for the chemical sorption/desorption of
hydrogen. It should show sufficient fatigue resistance, as it must be
resistant to the temperature and pressure pulsation. Therefore, new
technology for the growth of graphene on a liquid metallic matrix
was developed leading to the creation of high strength metallur-
gical graphene (HSMG) [23,24].
This work presents the results of an investigation of the atomic
hydrogen chemisorption on metallurgical graphene enhanced by
plasma discharge as well as a numerical simulation of the pressure
changes inside the nanogaps in hydrogen-storable nanocomposites
during the desorption cycles. Additionally, experimental fatigue
resistance tests combined with simultaneous resistance measure-
ments of graphene sheets were conducted to demonstrate the su-
perior mechanical properties of the metallurgical graphene
compared to graphene grown by CVD.
2. Materials and methods
The HSMG sheets were synthesized in an industrially scaled
thermochemical facility based on the process described by Kula and
Kolodziejczyk [23,25]. The multilayer copper/nickel composite
substrate (in this case 72% Cu, 28% Ni), was heated to the temper-
ature of 1200e1250  C in argon protective atmosphere at the
constant pressure of 100 kPa. Then the chamber was evacuated to
the pressure of 2 Pa and the mixture of acetylene, hydrogen and
ethylene in the proportions of 2:1:2, at the partial pressure of 3 kPa,
has been simultaneously introduced with argon into the process
chamber for 1 min. Afterwards, the plate was cooled to the tem-
perature of 1050  C and then to the room temperature at the
cooling speed of 0.5  C/min and 5e10  C/min, respectively, in argon
atmosphere at the constant pressure of 100 kPa. Prior to the me-
chanical and electrical property investigations, all of the samples
were subjected to structural analyses. The Raman spectra were
measured with a 100 objective at 532 nm with use of an inVia
Renishaw spectrometer. The laser power on the sample was kept
significantly below 1 mW to avoid possible laser-induced heating.
2.1. Sorption/desorption characteristics
The sorption and desorption of hydrogen on the surface of the
polycrystalline HSMG graphene was investigated in a vacuum/low
pressure chamber equipped with a vacuum pump and gas flow
system. The level of hydrogen chemisorption was assessed by the
measurement of the electrical resistance of the graphene sheet as a
result of its free electrons' (charge carriers) involvement in the CeH
bonds. A graphene sample was placed on an optical glass Schott
B270 Superwite substrate with 25  25  1 mm dimensions (Fig. 1).
Gold electrodes were deposited such that the active surface area of
the graphene was approximately ~1 cm2.
A sample of metallurgical graphene was placed in a vacuum
chamber from which air was evacuated to a pressure of 8$103 Pa.
Then, argon was introduced to purge the chamber of residual air.
The atomic hydrogen chemisorption process was performed in a
reactor filled with hydrogen at a pressure of 5e20 Pa. For the
dissociation of hydrogen, a glow discharge generated by a 50 Hz
alternating voltage between the electrodes at a height of 10 cm
above the surface of the graphene was used. The desorption process
took place in ambient air at a normal pressure or alternatively in the
reactor chamber by introducing oxygen with a small amount of
water vapor. The measured resistance value has been normalized.
The entire procedure was then repeated five times to obtain the
cyclic sorption/desorption characteristics. The hydrogenated gra-
phene specimens were also heated in a low-pressure hydrogen
environment to demonstrate the possibility of thermal hydrogen
desorption from hydrogenated graphene. For this purpose, the
hydrogenated graphene samples after the chemisorption process
were heated to a temperature of approximately 150  C. Both
chemisorption and subsequent thermal desorption processes were
carried out in the same hydrogen atmosphere at a pressure of 5 Pa.
The resistance of the graphene was measured continuously during
experiment.
2.2. Tensile tests
Cyclic tensile tests of the HSMG and CVD graphene
Fig. 1. HSMG graphene sample on the glass substrate that was used in the hydrogen
sorption/desorption test: 1 e gold electrodes; 2 e substrate; 3 e graphene. (For
interpretation of the references to color in this figure caption, the reader is referred to
the web version of this article.)
 P. Kula et al. / Vacuum 129 (2016) 79e85
(Supermarket, USA) sheets supported on a 100 mm thick low-
density polyethylene (LDPE) substrate were carried out using a
CETR UMT-2 universal test system (Bruker Corporation, USA)
equipped with a positioning system with a resolution of 0.5 mm and
a 2-axis load sensor (0.2e25 N) with a resolution of 1 mN. The size
of the LDPE substrate was 20  15 mm, whereas the size of the
transferred graphene sheets was 12  7 mm. Simultaneously, the
graphene sheet resistance was measured during each test. The
graphene samples were fixed with metal clamps and were elec-
trically isolated from the rest of the test system. The tensile test
setup was described in our previous paper [24]. The initial length of
the sample was 4 mm (free part of the graphene inserted between
the clamps). The course of the study was divided into stages, each of
which was to strain the sample by 0.250 mm and then return to the
initial strain value. At each stage, 20 of such cycles were carried out.
One stage of the fatigue study consisted of performing 20 cycles
with the amplitude of 0.250 mm. After completing the stage, the
sample was strained by a further 0.250 mm, which was the
beginning of the next cycle (loaded to pre-strain the sample). The
speed of linear movement during the stretching, 0.056 mm/s
remained constant for the whole test duration. During the study,
the change in the electrical resistance of the sample, caused by the
extension, was measured. The study was conducted until the
discontinuity of graphene was reached, which was equivalent to
the loss of its conductivity monitored by the gauge.
2.3. Computer simulation
The nanotube was used as the model for the pressure assess-
ment in the graphene-based nanocomposite gaps resulting from
molecular hydrogen recombination from hydrogenated graphene.
For the purpose of this study, a computer simulation in the ANSYS
CFX (Ansys Corporation, USA) environment was developed to
determine the flow rate and pressure gradient of hydrogen in the
nanotube during desorption from its walls. The first step in the
creation of the computational model was to build a discrete model
based on CAD geometry. The model consists of 32 million hex-
ahedral elements (Fig. 2). Additionally, at the walls of the nanotube,
the number of elements has been compacted to form a boundary
layer with a large number of finite elements.
A discrete model was then imported into the ANSYS CFX pre-
processor, assuming a transient analysis of total duration of 1, 10
and 100 s, at time steps of 0.01, 0.1 and 1 s, respectively. The model
Fig. 2. Carbon nanotube discrete model (nanotube length 100 mm, diameter 5 nm).
81
assumed that within the predetermined total process duration all
atoms desorb from the wall of the nanotube, recombine to H2
molecules and mix with the molecular hydrogen that fills the
nanotube at a pressure of 101.3 kPa. Because the purpose of the
analysis was to observe changes in velocity and pressure, no heat
exchange in the domain and laminar flow model (a ratio of the
diameter to the length of the nanotube, 1:20000) were assumed.
Other models that consider the exchange of heat can be found in
the literature; however, in those cases, chemical reactions were
implemented into the models, and heat had to be taken into ac-
count [26e29]. Under these assumptions, only two equations were
solved: the conservation of momentum and the conservation of
mass. It is assumed that the ends of the nanotube are completely
open and allow for a free flow of hydrogen while the cylindrical
inner-surface is completely filled with hydrogen. The hydrogen
mass flow rate for the various total analysis times was set to
9.89$1017 g/s, 9.89$1018 g/s and 9.89$1019 g/s, respectively.
3. Results and discussion
3.1. Hydrogen sorption-desorption characteristics of graphene
The atomic hydrogen formed in the process of dissociation is
combined through a chemical bond with the carbon atoms,
reducing the quantity of charge carriers in the graphene. The
reduction of the carriers' number, inversely proportional to the
resistance, was confirmed by the Hall effect experiments [unpub-
lished results]. This chemisorption effect is stable even for 12 h if
the hydrogenated graphene sample can be kept in argon. After
filling the chamber with atmospheric air, the chemical desorption
of the hydrogen bonded to the surface of graphene begins. This can
be seen as a decrease in the resistance of graphene to the initial
level (Fig. 3a).
In Fig. 3b, each sorption-desorption cycle from Fig. 3a was
shifted in time so that the common point of the timeline was the
starting point for the adsorption process. A sharp decrease in the
resistance during exposure to air is mainly the result of the inter-
action between the hydrogenated graphene and the oxygen and
water vapor from the air, a phenomenon described in the next
section.
The full reversibility of the chemical sorption and desorption of
hydrogen in HSMG graphene is evidenced by the same evolution of
the resistance curves (except for the first cycle) and an almost
perfect compliance of the resistance values after each desorption
process. The resistance does not return to the initial value after the
first cycle because the hydrogen adsorbed on the grain boundaries
and defects does not desorb under ambient conditions (atmo-
spheric pressure and room temperature).
3.2. Mechanism of atmospheric desorption of hydrogen
After the hydrogenation process of graphene, exposure to air
causes a rapid decrease in the graphene resistance (Fig. 3), which
may indicate a decrease in chemically adsorbed atomic hydrogen.
Therefore, the following experiment was designed to determine the
effect of the atmospheric composition on the dehydrogenation of
the analyzed graphene flakes. For that purpose, the selected at-
mospheres (O2, Ar, CO2, O2 þ H2O and CO2 þ H2O) were indepen-
dently introduced in a controlled way into the vacuum chamber. It
has been proven that the most intense increase in carriers is
observed for the O2 þ H2O atmosphere mixture. Fig. 4 shows the
hydrogen sorption cycle of HSMG (marked as 1) followed by the
introduction of the O2 þ H2O mixture during the desorption step
with a clearly visible rapid decrease in the resistance (marked as 3).
Therefore, it was assumed that the dehydrogenation mechanism
82 P. Kula et al. / Vacuum 129 (2016) 79e85
Fig. 3. (a) The resistance changes of a HSMG graphene sample subjected to five sorption/desorption cycles. (b) The cycles are shifted on the time axis to overlap each sorption/
desorption cycle at the beginning of the desorption process.
Fig. 4. Changes in the HSMG graphene sample resistance during the sorption in
hydrogen and desorption in the O2 þ H2O gas mixture: 1 e chemisorption during a low
discharge; 2 e graphene saturation with hydrogen; 3 e desorption in a humid oxygen
environment.
is as follows. First, the water vapor condenses on the surface of the
graphene, and the formation of hydrogen bonds between donor
(hydrogen bonded to the surface of the graphene) and acceptor
(oxygen in a molecule of water) are observed. Then, the dissolved
oxygen molecules in the condensed water vapor environment on
the surface of graphene dissociate according to the equation (1) to
form OH ions:
O2 þ H2 O þ 4e /4OH :
In the presence of ions, an intense graphene dehydrogenation
occurs, according to the equation (2), whereby the reaction is
exponential and cyclic due to the formation of water molecules,
which are the reaction medium capable of dissolving molecular
oxygen and regenerating OH ions.
OH þ H  CHSMG /H2 O þ CHSMG
It is worth emphasizing, that this reaction occurs only in the
case of hydrogen linked to the sp3 hybridized carbon of graphene
structure. This is because on its edge, carbon is sp2 hybridized, in
which the carbon forms two s bonds with two electrons of the sp2
orbitals of two adjacent carbon atoms and one s bond with the
electron orbital s of a hydrogen atom.
Hydrogenated graphene contains the sp3 hybridization of the
carbon atom, which forms three s bonds with three electron of the
sp2 orbitals of three adjacent carbon atoms and one s bond with
the electron of the s hydrogen atom orbital (Fig. 5). This arrange-
ment of hydrogen bonding and the formation of sp3 hybridized
carbon in the graphene structure leads to a binding energy Cgra-
phene-H ¼ 145 kJ/mol [30]. In the case of the interaction of hydrogen
with edge defects, a strong tendency to form chemical bonds Cgra-
phene-edge-H is observed; the binding energy that corresponds to the
CeH bonds energy in the aromatic compounds ranges from 300 to
460 kJ/mol.
This phenomenon explains why, in hydrogen sorption-
desorption cycles on the graphene flakes, the system practically
returns to the initial value of resistance. The small difference in the
initial graphene sheet resistance values between the first and
subsequent cycles (Fig. 3b) is related to the permanent interaction
of hydrogen with defects in the graphene (holes, grain boundaries
of polycrystalline graphene [24] and edges). In this case, the energy
of the hydrogen-carbon bonds of the graphene system is 145 kJ/
mol, in contrast to the hydrogen-carbon bond energies in aromatic
structures, which are at least two times higher [31e33]. Conse-
quently, in these conditions the hydrogen bonded to defects is not
desorbed.
Therefore, it is important to eliminate even trace amounts of
water vapor from tanks intended for the storage of graphene flakes
capable of reversible hydrogen bonding. This is a key issue in
eliminating the uncontrolled reduction of the hydrogenated gra-
phene structures and the formation of water molecules according
(1)
(2)
Fig. 5. A model of two possible types of hydrogen chemisorption.
 P. Kula et al. / Vacuum 129 (2016) 79e85
to the above-described mechanism.
3.3. Thermal desorption
Although the previously mentioned experiments show the
applicability of HSMG graphene for multi-cyclic chemical sorption/
desorption, the dehydrogenation of hydrogenated graphene by
oxidation does not have any practical importance for reversible
hydrogen storage systems. However, the simple test of heating
hydrogenated graphene in a hydrogen environment demonstrated
the thermal desorption of hydrogen, which was illustrated by a
decrease in the electrical resistance when the temperature
increased up to 150  C (Fig. 6).
This finding is fundamental for further consideration of low-
pressure graphene-based nanomaterials as reversible hydrogen
storage systems using the concept of hydrogen chemisorption.
Additionally, the relatively low temperature of thermal hydrogen
desorption is promising for future engineering applications.
3.4. Raman spectroscopy
Changes in the structure of graphene resulting from the hy-
drogenation process were analyzed by Raman spectroscopy. The
obtained spectra are presented in Fig. 7.
The first spectrum (Fig. 7a) is an example of the Raman spec-
trum of graphene before hydrogenation. Bands typical for graphene
are observed: D (~1350 cm1), G (~1580 cm1) and 2D
(~2690 cm1). In Fig. 7b, the spectrum of graphene after desorption
process is shown prior to re-hydrogenation. In addition to the
characteristic graphene bands, new peaks are visible: Dʹ
(~1620 cm1) and D þ Dʹ (~2950 cm1) [34]. They demonstrate the
presence of residual hydrogen (after the desorption process) at the
grain boundaries and defect areas of graphene. Fig. 7c presents the
spectrum of graphene obtained directly after the process of
hydrogen chemisorption. In addition to the basic bands typical for
hydrogenated graphene, D0 and D þ D0 peaks are visible with a
distinctly sharp shape [34,35]. The hydrogen chemisorption process
also exhibits a slight broadening of the 2D band and a decrease of
its height relative to the G peak, which is consistent with the
literature [34].
3.5. Simulation of hydrogen desorption inside the carbon nanotube
During hydrogen desorption completed at various times, the
viscosity of the gas and its density at predetermined pressures
result in a significant increase in pressure, particularly at the mid-
point of the nanotube length. This effect, for 1-s desorption from
Fig. 6. Resistivity changes during thermal hydrogen desorption process.
83
Fig. 7. Changes in Raman spectra of HSMG graphene due to its hydrogenation: a)
graphene on glass; b) graphene after the desorption process; c) graphene after the
chemisorption of hydrogen.
the wall results locally in a tenfold increase in pressure from the
reference level of 101.3 kPae1063.6 kPa (Fig. 8).
This is directly caused by the mobility of the molecules of des-
orbed hydrogen and the hydrogen initially filling the nanotube, in
which the nanotube diameter/length ratio is equal to 1:20000. The
system of equations of both conservation of mass and momentum
solved during the analysis is based on the gas properties, including,
i.e., a dynamic viscosity of 8.8,103 g/m,s and a molar mass of
2.016 g/mol. The values of other parameters were adopted in
accordance with the NASA format [36]. Comparing the results for 10
and 100 s of total hydrogen desorption time, an increase in pressure
of 130.2 kPa (128% increase, compared to reference level) and
18.9 kPa (19% increase), respectively, are observed. Referring to the
result of the desorption within 1 s, time mostly influences the
analysis of the flow phenomena for this geometry, which directly
translates into the behavior of wall nanotubes during the hydrogen
release process.
The results of the flow analysis were used as the boundary
conditions for the analysis carried out in the ANSYS Mechanical
software, in which the nanotube is defined as a shell material with a
thickness of 0.14 nm, Young's modulus E2D equal to 340 N/m
(E ¼ 1.05 TPa) [37e39] and the Poisson ratio n ¼ 0.22 [40e42].
The calculated relative strain for a nanotube with a diameter of
5 nm at pressure of 1063 kPa reached 1474.7%, which
Fig. 8. The changes of the maximum absolute pressure inside the carbon nanotube
model calculated for various times of the complete hydrogen desorption process.
84 P. Kula et al. / Vacuum 129 (2016) 79e85
Table 1
Simulation results of the strain of nanotube (in %) for different diameters and
pressures.
Pressure, kPa Diameter of the analyzed nanotube
∅ ¼ 5 nm ∅ ¼ 10 nm ∅ ¼ 20 nm
1 1.5 1.5 1.4
10 14.8 14.5 14.4
100 135.0 131.0 123.0
1063 1474.7 1450.6 1439.1
unquestionably would lead to the destruction of the nanotube
structures (Table 1). Moreover, also for nanotubes with diameters of
10 and 20 nm, calculated relative strains exceeded 1400% (at the
same pressure), which excludes the possibility of increasing the
nanotube diameter to prevent its rupture.
The most viable solution seems to be controlling the hydrogen
desorption duration, as it was presented based on the ANSYS CFX
calculations, in which extending the desorption time to 10 s yielded
relative pressures in the nanotube equal to 130.2 kPa with a relative
strain of 135%. These maximum hydrogen desorption rates may be
expected in future reversible hydrogen storage systems; therefore,
that level of necessary safety strain has been taken as the bench-
mark for experimental fatigue tensile tests for various types of
graphene. The results are presented in the next section.
3.6. Fatigue resistance tensile tests
For tensile test measurements, the relationship between the
relative elongation of the sample and its tensile strength was
determined by measuring changes in conductivity. Typical char-
acteristics of these relationships for CVD graphene (black line) and
HSMG graphene (blue line) are shown in Fig. 9. In addition, the
course of the curve showing a relative deformation of the sample
during the test is marked with a red line. The deformation values
are shown on the right axis of the graph.
The study showed significant differences in the tensile strengths
for both materials. The CVD graphene lost conductivity at a relative
elongation of approx. 40%, while the value for the HSMG graphene
was as high as 140%. For both materials, increased elongation
caused a gradual increase of resistance. This increase was propor-
tional to the elongation. More regular behavior was observed for
the HSMG graphene, in which extension led to a proportional in-
crease in the resistance; however, after reducing the strain it
returned to its previous value. Hence, on the chart regular cycles of
resistance changes corresponding to the amplitude distortion are
Fig. 9. (a) Changes in the resistances of the HSMG and CVD graphene samples as a function of cyclic loading; (b) zoomed area that shows the loss of conductivity for the CVD
graphene.
noted. This regularity occurs with relative deformations of up to
100%. Above this value, within the same range of strain, a pro-
gressive increase in the sample resistance is visible until the loss of
its continuity at a resistance of approximately 55 MOhm. In this
case, CVD graphene behaves with less stability. From the initial
stage of the test there is a large inhomogeneity of the resistance
changes and the obtained values are significantly higher than for
HSMG graphene. Hence, the lower resistance value (approx.
4 MOhm) was measured for the CVD graphene at its fracture point.
In earlier work [24] in the static tensile test conducted until the
graphene structure fractured, the CVD graphene lost electrical
contact at approximately ~16% elongation. In the literature, the
maximum elongation values are in the range from 1 to >10%
[43e45], while the metallurgical graphene maintains electrical
conductivity dependent upon the crystal structure defects, even at
an elongation of approximately ~33% [24]. In the case of fatigue
tests, obtained strains are approximately three times higher for
both the CVD and the HSMG graphene. Similar studies were con-
ducted by Xiao Li's group [43]. In that experiment, poly-
dimethylsiloxane (PDMS) was used as a substrate for the graphene
grids, which were subjected to varying loads over time. They
received an increase in resistance with increasing deformation,
which is consistent with the results presented in this paper. In the
case of a PDMS-graphene system the critical elongation was
15e20%. In the present study, CVD graphene loses continuity at a
relative deformation of approximately 40%, while HSMG graphene
can reach deformation of more than 100% before a loss in continuity
occurs. The discrepancy between the results of the cited work and
the presented experimental data may result from different sub-
strates and form of graphene (grid, plane), the quality of the
transfer and the fatigue test conditions.
It can be concluded that the HSMG graphene retains its high
strength at high strains, reaching 100% elongation, and in this
range, it is resistant to cyclic strain changes of fatigue nature.
Comparing this experimental result to the benchmark levels of
strain taken in the previous section, further research on high
strength metallurgical graphene as a promising material for
graphene-based hydrogen storage systems is relevant.
4. Conclusions
Sorption-desorption experiments confirmed significant chemi-
sorption of hydrogen on HSMG graphene during exposure to glow
discharge hydrogen plasma at low pressures in the range of
5e20 Pa. Hydrogenated graphene was shown to be unstable in
contact with humid oxygen or humid air. Thermal desorption
 P. Kula et al. / Vacuum 129 (2016) 79e85
experiments demonstrated that hydrogen is released from hydro-
genated graphene at relatively low temperatures (ca. 150  C). The
simulation results indicated that the thermal desorption of
hydrogen in gaps of the graphene-based pillared nanocomposites
may reach high levels exceeding the tensile strength of graphene.
An experimental test of high strength metallurgical graphene ex-
hibits a tensile fatigue performance that might resist pressure
pulses and strain that appears in the gaps of graphene-based
hydrogen storage systems.
Acknowledgments
The study is financed by The National Centre for Research and
Development as part of the program GRAF-TECH, project acronym
GraphRoll, project title: “Graphene nanocomposite for the revers-
ible hydrogen storage”, agreement no. GRAF-TECH/NCBR/07/24/
2013.
References
[1] V. Georgakilas, M. Otyepka, A.B. Bourlinos, V. Chandra, N. Kim, K.C. Kemp, et
al., Functionalization of graphene: covalent and non-covalent approaches,
derivatives and applications, Chem. Rev. 112 (2012) 6156e6214.
[2] V. Tozzini, V. Pellegrini, Prospects for hydrogen storage in graphene, Phys.
Chem. Chem. Phys. 15 (2013) 80e89.
[3] Y. Zhu, D.K. James, J.M. Tour, New routes to graphene, graphene oxide and
their related applications, Adv. Mater. 24 (2012) 4924e4955.
[4] C.N.R. Rao, A.K. Sood, K.S. Subrahmanyam, A. Govindaraj, Graphene: the new
two-dimensional nanomaterial, Angew. Chem. Int. Ed. Engl. 48 (2009)
7752e7777.
[5] C. Li, G. Shi, Three-dimensional graphene architectures, Nanoscale 4 (2012)
5549e5563.
[6] M. Pumera, Graphene-based nanomaterials for energy storage, Energy Envi-
ron. Sci. 4 (2011) 668e674.
[7] S.-Y. Lee, S.-J. Park, Comprehensive review on synthesis and adsorption be-
haviors of graphene-based materials, Carbon Lett. 13 (2012) 73e87.
[8] Y. Lu, Y.P. Feng, Adsorptions of hydrogen on graphene and other forms of
carbon structures: first principle calculations, Nanoscale 3 (2011) 2444.
[9] S. Patchkovskii, J.S. Tse, S.N. Yurchenko, L. Zhechkov, T. Heine, G. Seifert,
Graphene nanostructures as tunable storage media for molecular hydrogen,
Proc. Natl. Acad. Sci. U. S. A. 102 (2005) 10439e10444.
[10] W.-Q. Deng, X. Xu, W.A. Goddard, New alkali doped pillared carbon materials
designed to achieve practical reversible hydrogen storage for transportation,
Phys. Rev. Lett. 92 (2004) 166103.
[11] Y. Wang, C. Guan, K. Wang, C.X. Guo, C.M. Li, Nitrogen, hydrogen, carbon di-
oxide, and water vapor sorption properties of three-dimensional graphene,
J. Chem. Eng. Data 56 (2011) 642e645.
[12] W. Yuan, B. Li, L. Li, A green synthetic approach to graphene nanosheets for
hydrogen adsorption, Appl. Surf. Sci. 257 (2011) 10183e10187.
[13] R.S. Aga, C.L. Fu, M. Krcmar, J.R. Morris, Theoretical investigation of the effect
of graphite interlayer spacing on hydrogen absorption, Phys. Rev. B 76 (2007)
165404.
[14] J.S. Arellano, L.M. Molina, A. Rubio, J.A. Alonso, Density functional study of
adsorption of molecular hydrogen on graphene layers, J. Chem. Phys. 112
(2000) 8114.
[15] Y. Miura, H. Kasai, W. Dineo, H. Nakanishi, T. Sugimoto, First principles studies
for the dissociative adsorption of H[sub 2] on graphene, J. Appl. Phys. 93
(2003) 3395.
[16] P. Kula, The “self-lubrication” by hydrogen during dry friction of hardened
surface layers, Wear 201 (1996) 155e162.
[17] K.S. Subrahmanyam, P. Kumar, U. Maitra, A. Govindaraj, K.P.S.S. Hembram,
U.V. Waghmare, et al., Chemical storage of hydrogen in few-layer graphene,
Proc. Natl. Acad. Sci. U. S. A. 108 (2011) 2674e2677.
[18] Z. Yang, Y. Sun, L.B. Alemany, T.N. Narayanan, W.E. Billups, Birch reduction of
graphite. Edge and interior functionalization by hydrogen, J. Am. Chem. Soc.
134 (2012) 18689e18694.
[19] D.M. Chen, T. Ichikawa, H. Fujii, N. Ogita, M. Udagawa, Y. Kitano, et al., Unusual
hydrogen absorption properties in graphite mechanically milled under
various hydrogen pressures up to 6 MPa, J. Alloys Compd. 354 (2003) L5eL9.
85
[20] S. Orimo, T. Matsushima, H. Fujii, T. Fukunaga, G. Majer, Hydrogen desorption
property of mechanically prepared nanostructured graphite, J. Appl. Phys. 90
(2001) 1545.
[21] S. Goler, C. Coletti, V. Tozzini, V. Piazza, T. Mashoff, F. Beltram, et al., Influence
of graphene curvature on hydrogen adsorption: toward hydrogen storage
deviceS, J. Phys. Chem. C 117 (2013) 11506e11513.
[22] P. Kula, Ł. Kaczmarek, P. Zawadzki, Ł. Kołodziejczyk, W. Szyman  ski,
P. Niedzielski, et al., Functionality of graphene as a result of its heterogenic
growth on SiC nanoparticles on the basis of reversible hydrogen storage, Int. J.
Hydrogen Energy 39 (2014) 19662e19671.
[23] P. Kula, R. Pietrasik, K. Dybowski, R. Atraszkiewicz, W. Szymanski,
L. Kolodziejczyk, et al., Single and multilayer growth of graphene from the
liquid phase, Appl. Mech. Mater. 510 (2014) 8e12.
[24] P. Kula, W. Szymanski, L. Kolodziejczyk, R. Atraszkiewicz, K. Dybowski,
J. Grabarczyk, et al., High strength metallurgical graphene e mechanisms of
growth and properties, Arch. Metall. Mater. 60 (2015) 2535e2541.
[25] L. Kolodziejczyk, P. Kula, W. Szymanski, R. Atraszkiewicz, K. Dybowski,
R. Pietrasik, Frictional behaviour of polycrystalline graphene grown on liquid
metallic matrix, Tribol. Int. 93B (2016) 628e639.
[26] A.V. Avdeenkov, I.V. Bodrenko, D.G. Bessarabov, A.V. Bibikov, A.V. Nikolaev,
M.D. Taran, et al., Thermodynamical model for hydrogen storage capacity in
carbon nanostructures, Int. J. Hydrogen Energy 40 (2015) 4184e4193.
[27] T. Brestovi c, N. Jasminska, R. Pyszko, M. La zar, M. Puskar, Measurement of
boundary conditions for numerical solution of temperature fields of metal
hydride containers, Measurement 72 (2015) 52e60.
[28] S. Ishikawa, T. Yamabe, A theoretical deduction of the shape and size of
nanocarbons suitable for hydrogen storage, Appl. Phys. A 114 (2013)
1339e1346.
[29] F. Socorro, I d. la Nuez, M. Rodrıguez de Rivera, Calibration of isothermal heat
conduction calorimeters: case of flow calorimeters, Measurement 33 (2003)
241e250.
[30] M. Darvish Ganji, S.M. Hosseini-khah, Z. Amini-tabar, Theoretical insight into
hydrogen adsorption onto graphene: a first-principles B3LYP-D3 study, Phys.
Chem. Chem. Phys. 17 (2015) 2504e2511.
[31] W. Van Scheppingen, E. Dorrestijn, I. Arends, P. Mulder, H.-G. Korth, Car-
bonoxygen bond strength in diphenyl ether and phenyl vinyl ether: an
experimental and computational study, J. Phys. Chem. A 101 (1997)
5404e5411.
[32] X.-Q. Zhu, H.-R. Li, Q. Li, T. Ai, J.-Y. Lu, Y. Yang, et al., Determination of the C4-H
bond dissociation energies of NADH models and their radical cations in
acetonitrile, Chem. e A Eur. J. 9 (2003) 871e880.
[33] L.J.J. Laarhoven, P. Mulder, D.D.M. Wayner, Determination of bond dissociation
enthalpies in solution by photoacoustic calorimetry, Acc. Chem. Res. 32 (1999)
342e349.
[34] D.C. Elias, R.R. Nair, T.M.G. Mohiuddin, S.V. Morozov, P. Blake, M.P. Halsall, et
al., Control of graphene's properties by reversible hydrogenation: evidence for
graphane, Science (80-) 323 (2009) 610e613.
[35] Z. Luo, T. Yu, Z. Ni, S. Lim, H. Hu, J. Shang, et al., Electronic structures and
structural evolution of hydrogenated graphene probed by Raman spectros-
copy, J. Phys. Chem. C 115 (2011) 1422e1427.
[36] B.J. McBride, S. Gordon, M.A. Reno, Coefficients for calculating thermodynamic
and transport properties of individual species, NASA Tech. Memo. 4513 (1993)
98.
[37] E. Cadelano, P.L. Palla, S. Giordano, L. Colombo, Elastic properties of hydro-
́
genated graphene, Phys. Rev. B 82 (2010) 235414.
[38] C. Lee, X. Wei, J.W. Kysar, J. Hone, Measurement of the elastic properties and
intrinsic strength of monolayer graphene, Science (80-) 321 (2008) 385e388.
[39] G.-H. Lee, R.C. Cooper, S.J. An, S. Lee, A. van der Zande, N. Petrone, et al., High-
strength chemical-vapor-deposited graphene and grain boundaries, Science
340 (2013) 1073e1076.
[40] K.H. Michel, B. Verberck, Theory of the elastic constants of graphite and
graphene, Phys. Status Solidi 245 (2008) 2177e2180.
[41] V. Perebeinos, J. Tersoff, Valence force model for phonons in graphene and
carbon nanotubes, Phys. Rev. B 79 (2009) 241409.
[42] F. Scarpa, S. Adhikari, A. Srikantha Phani, Effective elastic mechanical prop-
erties of single layer graphene sheets, Nanotechnology (2009) 20, 065709.
[43] X. Li, R. Zhang, W. Yu, K. Wang, J. Wei, D. Wu, et al., Stretchable and highly
sensitive graphene-on-polymer strain sensors, Sci. Rep. 2 (2012) 870.
[44] H.H. Pe rez Garza, E.W. Kievit, G.F. Schneider, U. Staufer, Controlled, reversible,
and nondestructive generation of uniaxial extreme strains (>10%) in gra-
phene, Nano Lett. 14 (2014) 4107e4113.
[45] M.A. Bissett, M. Tsuji, H. Ago, Mechanical strain of chemically functionalized
chemical vapor deposition grown graphene, J. Phys. Chem. C 117 (2013)
3152e3159.