Research
Modelling carbyne C12-ring calcium decorated for hydrogen
storage
L.A. Desales Guzmána
J.H. Pacheco Sánchezb
G. García-Rosalesc
F.J. Isidro Ortegad
a División de Estudios de Posgrado e
Investigación, Instituto Tecnológico de Toluca, Metepec, 52149, Edo. Mex.,
México, Av. Tecnológico S/N, Col. Agrícola Bellavista Metepec 52149, Estado de
México, México, Tel: 527221128116, e-mail: luis.aud.int@gmail.com
b División de Estudios de Posgrado e
Investigación, Instituto Tecnológico de Toluca, Metepec, 52149, Edo. Mex.,
México,Tel: 527222087200 ext. 3211; e-mail:
jpachecos@toluca.tecnm.mx
c División de Estudios de Posgrado e
Investigación, Instituto Tecnológico de Toluca, Metepec, 52149, Edo. Mex.,
México,Tel: 527222526552; gegaromx@yahoo.com.mx
d División de Estudios de Posgrado e
Investigación, Instituto Tecnológico de Toluca, Metepec, 52149, Edo. Mex.,
México, Tel: 527224233048 fisidroo@toluca.tecnm.mx
Abstract
We computationally investigate the hydrogen storage properties of C12 carbyne structure decorated with calcium (Ca) atoms adsorbed on its
outer surface. The calculations are carried out on DFT-GGA-PW91 and DFT-GGA-PBE
levels of theory as implemented in the modeling and simulation Materials Studio
program. To account for the dispersion (van der Waals) interactions we also
carry out calculations using DFT-D method of Grimme. Dmol 3 is used to calculate, total energies, charge density HOMO-LUMO and
Mulliken population analysis. Based on these results, up to six H2 molecules per Ca atom can be physisorbed with an average energy of
0.1890 eV (PW91) and 0.3204 (DFT-D) per H2 molecule. This physisorption leads to 6.16 weight percentage (wt %)
for the gravimetric hydrogen storage capacity. According to these results, the
calcium-decorated carbyne C12 structure is a good candidate for hydrogen storage with application
to fuel cells.
Keywords: Carbyne; hydrogen storage; calcium decorated; Dmol3; physisorption
PACS: 88.30.R-; 88.20.fn
1. Introduction
Hydrogen has attracted much attention as a promising clean energy source that could
one day replace fossil fuels to combat global warming [1-5]. One of the biggest challenges to reach that goal is
the high-density hydrogen storage, to perform reversible hydrogen charging cycles at
ambient conditions for automotive onboard applications [2,4,5].
A desired storage system must have a high gravimetric density (4.5 to 6.5 wt %) at
moderate pressure, and should operate at delivery temperatures between -40 to 85∘C, goal suggested by US Department of Energy (DOE) for the year 2020
[3,6,7]. In the past few decades, the carbon-based
nanostructure materials, including nanotubes [8-11], fullerenes [8,11-12] and graphenes [8,13] were expected to be the promising candidate materials
for hydrogen storage because of their high surface/volume ratios and high degree of
reactivity between carbon and hydrogen [14]. However, physical sorption in these systems
generally requires cryogenic temperature because of the small binding energy of ∼0.05 eV, which makes it impractical for vehicular usage [15]. Desirable binding energy should lie in a range
of 0.2-0.4 eV per H2 for the ambient condition storage [2,5,14-19].
In recent years, several first theoretical studies have been made to improve the
hydrogen storage capacity of these carbon-based nanostructure materials by
decorating atoms that can be transition metals (TM), alkaline metals (AM) or
alkaline earth metals (AEM) [2,3,14-25]. Each TM (Sc, Ti) [18] or AEM (Ca) [2,15,21] atom can bind up to six hydrogen molecules per metal
atom with a binding energy of ∼0.2-0.6 eV [14,15,17].
However, a disadvantage shown by TM over AEM is that they have adsorption energies
much larger than required, so not all hydrogen molecules might be adsorbed with the
same energy since the first hydrogen molecules are strongly chemisorbed in atomic
form, and the subsequent are physisorbed in molecular form [2,3,14,15]. This means that the total storage capacity cannot be
carried out at a single operating temperature. On the other hand, the clustering of
metal atoms will surely occur, due to the relatively large cohesive energy of these
transition metals (≥4 eV per atom), which would further reduce the hydrogen storage capability
[3,26,27]. Metal has a relatively small cohesive energy (≤2 eV per atom), so Ca atoms have a lower tendency to cluster on the host
material once they are deposited. For instance, Ca-decorated graphene [20] and carbon nanotubes [19], the maximum hydrogen storage
capacity can be as high as 8 and 5wt%, respectively. However, the binding energies
of Ca on graphene and carbon nanotubes are only 0.632 and 0.88 eVCa-1, respectively, which are much lower than the Ca-Ca cluster cohesive
energy of 1.84 eV [14]. As a
result, the Ca atoms prefer to form clusters on the carbon nanostructure, which may
significantly reduce the hydrogen storage.
For this reason, the bonding energy of nanostructured decorated material must be
greater than the cohesive energy to ensure there is no clustering between metals. On
the other hand, Ca atom is expected to bind all hydrogen molecules reversibly by
physisorption as well as being an abundant and low-cost element compared to metals
such as Sc or Ti [3].
In this work, DFT calculations are performed in the carbyne C12-ring structure decorated with Ca using Biovia Materials Studio Dmol3 program to determine its capability for hydrogen storage. First, the
carbyne type molecule used here is related to polyyne which is a single chain of
carbon atoms with sp hybridization, formed with alternating single and triple bonds
(-C≡C-), proposed by Tammann in 1921, the most common case is acetylene
(H-C≡C-H) [28,29]. The C12-ring is a variation of the linear chains of carbyne, where a decrease in
the dimensions of the host materials is expected to increase the gravimetric
capacity as well as volumetric capacity of hydrogen following the density increase
of Ca atoms per volume or per C atom [15]. From this point of view, this kind of carbon may be
utilized as a host material because of its perfect one-dimensional geometry, and
consequently greater surface area.
We found a good behavior of C12-ring as an adsorbent material of hydrogen through a dopant metal agent
as calcium atom. The adsorption energy between C12-ring and the dopant agent corresponds to chemisorption; however,
according to the process of gas adsorption on a solid surface, we clearly found a
physisorption between the dopant-metal in carbyne and the hydrogen molecules. The
results show that up to six hydrogen molecules are physisorbed by the calcium atom
with a desired binding energy 0.2 eV for hydrogen storage at ambient conditions and
the hydrogen storage capability of 6.16 wt%. The HOMO-LUMO and Mulliken analysis
show the areas with the greatest presence of electronic charge in the molecule;
information that is compared with literature, making this molecule a suitable
candidate to meet the requirements established by DOE.
2. Calculation method
In this work, DFT calculations are performed in the carbyne C12-ring structure decorated with Ca using Biovia Materials Studio Dmol3 program [30,31] to determine its capability of hydrogen storage. To
calculate the adsorption energies, the generalized gradient approximation GGA-PW91
functional [32,33] for spin unrestricted was used.
The interaction energies between the hydrogen molecules with a unit of C12-ring are calculated by means of a set of double numerical bases (dnd),
which has the advantage of being equivalent to the analytical base Gaussian 6-31 G*.
All presented geometry optimizations are obtained until the maximum forces are lower
than 0.004 Ha/Å. Here the effect of van der Waals (vdW) interactions is included
explicitly by using the empirical correction scheme of Grimme (DFT-D) for periodic
systems [34]. Standard values of
the dispersion coefficients C6 (0.14, 1.75, and 10.80 J nm6 mol-1, for H, C, and Ca, respectively), vdW radii (1.001, 1.452, and 1.474 Å),
cutoff radius for pair interactions (30.0 Å), PBE global scaling factor S6 (0.75), and damping factor d (20.0) have been used. We report both
GGA-PW91 functional results and Grimme empirical correction scheme DFT-D method. In
addition, the total energies, HOMO-LUMO electronic charge density and Mulliken
population analysis are calculated.
3. Results and discussion
We performed the analysis of Ca-carbyne complex and found the most stable
configuration from C12 carbyne molecule and Ca atoms adsorbed on its outer surface. First,
carbyne molecule has been theoretically predicted to be a stable structure and it is
possibly found in two phases: (i) energetically favorable polyyne phase (-C≡C-)n with alternation of triple (1.24 Å) and single (1.34 Å) bonds by Peierls
distortion, called α-carbyne with an electronic configuration ((σ1s)2(σ1s*)2(σ2s)2(π2px))2(π2py)2(σ2pz)2)n, and (ii) cumulene-type (=C=C=)n double bonds, as β-carbyne
with an electronic configuration ((σ1s)2(σ1s*)2
(σ2s)2(σ2s*)2(π2px)2(π2py)2)n
[35,36]. We also found that carbyne
rings can be created using an even number of carbons (polyyne), while cumulene rings
can be designed with an odd carbons number. Figure
1(a) schematically shows the geometry optimization of carbyne C 12-ring, which corresponds to polyyne type used in this research. Second,
we studied the case of a single Ca atom adsorbed to the carbyne surface separated at
2.48 Å and we analyze the stability of Ca-carbyne complex by evaluating the binding
energy of Ca atom to the carbon ring from the following expression [6,15,16,23]:
Eb(Ca C12)=E(C12)+E(Ca)-E(Ca C12)
(1)
where E(CaC12), E(Ca) and E(C12), are the total energies of the CaC12 system of one Ca atom and one carbyne molecule, respectively.
The binding energy (Eb) of CaC 12 system shows a value of 2.95 eV and 2.23 eV with a GGA-PW91 and GGA-PBE
functionals respectively, indicating that the system is strongly chemisorbed.
Therefore, Ca atom will not tend to cluster if more calcium atoms are added to the
system since the bonding energy is greater than the cohesive energy of the Ca-Ca
bond (1.84 eV) [3,14,26,27]. Literature shows that typical chemisorption values
are above 155 kJ/mol (1.6 eV) [37], therefore the CaC12 system can be considered as a single unit which we call decorated
complex.
Figure 1 (b), shows how the Ca atom tend to
elongate the single bond of the carbyne molecule C12 to 2.210 Å. We found that the optimal position of Ca atom is at a bridge
site of the carbyne (C-C) bonds as at this position optimally binds to the adjacent
carbon atoms. In another position, with the Ca near a single carbon atom, the Ca
atom disturbs the π-system creating a high energy radical.
In addition, we analyze the binding mechanism of Ca atom to carbyne molecule in this
configuration.Figure 2 shows Total Density
of States (DOS) for pure carbyne molecule 12, and Fig. 3 shows the Partial
Density of States for the CaC12. The 4s orbitals of the Ca atom play the main role in binding of Ca to C12. This makes Ca atom to donate the s electrons due to their relatively
low ionization potentials which leads to delocalization of the orbitals (Fig. 3). Also, the 3d orbitals of Ca atom and a
small fraction of the 4p orbitals participate in bonding. There is a clear
indication of charge transfer from s orbitals of Ca atom to carbyne π-system, and
from carbyne π-system to the d orbitals of the Ca atom forming the donor-acceptor bond. Electron
transfer from Ca to carbyne results in a positive charge on Ca atom (Mulliken charge
+0.956e).
Once obtained CaC12 system in equilibrium, the next step is the adsorption analysis of H2 molecules on the decorated complex. The H2 molecules are added one by one over the Ca atom, reaching a maximum of
six H2 molecules adsorbed on it. We calculate the average energy of nH2 molecules adsorbed on CaC12 complex using the Eq. (2) [6,15,16,23]
Eave=[E(Ca C12)+nE(H2)-E(Ca C12-nH2)]n
(2)
where E(Ca C12) and E(H2) are the total energies of the complex CaC12 and of an isolated H2 molecule, respectively. The E(Ca C12-nH2) is the total energy of the system with n molecules adsorbed on the doped complex.
The adsorption energy of the n-th of H2 molecule is calculated using Eq. 3 [6,15,16,23].
Ead=E(Ca C12+(n-1)H2)+E(H2)-E(Ca C12)+nE(H2)
(3)
where E(Ca C12+(n-1)H2) is the total energy of the CaC12 system with (n-1) H2 molecules adsorbed on the decorated complex. The information of the
optimized structures of CaC12-n H2, for n=1-6 H2 molecules, is shown in Table
I.
TABLE I Total energy DFT-GGA-PW91 and DFT-GGA-PBE (Grimme) (Ha), Average
Energy (Eave) per H2 molecule (eV) and Adsorption Energy
(Ead) of the nth H2 molecule in a
CaC12 - nH2 system (eV), and weight percent
(%wt) of CaC12 - nH2 system.
System |
Total Energy
(Ha) |
Total Energy
(Ha) |
EaveandEadper H
2
|
EaveandEadper H
2
|
%wt |
GGA-PW91 |
GGA-PBE (Grimme) |
GGA-PW91 (eV) |
GGA-PBE-Grimme (eV) |
H2
|
-1.170096 |
-1.166029 |
- |
- |
- |
Ca |
-677.562931 |
-677.355188 |
- |
- |
- |
CaC12
|
-1134.496685 |
-1133.868382 |
- |
- |
- |
CaC 12- H2
|
-1135.67364 |
-1135.050485 |
(0.186)-(0.186) |
(0.437)-(0.437) |
1.0825 |
CaC 12-2 H2
|
-1136.850879 |
-1136.224976 |
(0.190)-(0.194) |
(0.333)-(0.230) |
2.1418 |
CaC 12-3 H2
|
-1138.02861 |
-1137.399998 |
(0.196)-(0.207) |
(0.304)-(0.244) |
3.1787 |
CaC 12-4 H2
|
-1139.204782 |
-1138.575493 |
(0.188)-(0.165) |
(0.292)-(0.257) |
4.1938 |
CaC 12-5 H2
|
-1140.381165 |
-1139.750244 |
(0.185)-(0.171) |
(0.281)-(0.237) |
5.1879 |
CaC 12-6 H2
|
-1141.558532 |
-1140.924903 |
(0.187)-(0.197) |
(0.273)-(0.234) |
6.1615 |
To examine the hydrogen adsorption on the doped complex, more than 10 different
configurations were simulated in the formation of H2 molecules (linear and circular), thereby obtaining a formation at the
time of reaching the six H2 molecules. It was observed that the H2 molecules tend to stay around the Ca atom, thereby obtaining a stable
system (CaC12-6H2). In Fig. 4, it is clearly
illustrated how the process of adding molecules to the doped complex is performed by
placing one by one up to a maximum of six H2 molecules; we show the first five H2 molecules tend to be adsorbed around the Ca atom and the sixth H2 molecule is adsorbed on top of the Ca atom.
The first H2 molecule is adsorbed with an energy of Eave=0.1866 eV/H2 and Ead=0.1867 eV/H2 (PW91) on the doped complex, the average distance between H2 molecules to Ca atom is 2.4101 Å and the bond distance among hydrogen
molecules changes from 0.774 Å to 0.776 Å, which is due to the interactions between
positively charged Ca atom and H2 molecules. When the third H2 molecule is adsorbed on
decorated complex, the adsorption energy is 0.1962 eV/H2, this value shows an increase of 5.14 % respect to the first H2 molecule that is adsorbed on the doped complex. The additional H2 molecules are maintained with an energy of 0.18 eV/H2. The hydrogen storage capacity obtained in this study is 6.1615 wt% as
seen in Table I, with an average energy of
0.1890 eV/H2 and adsorption energy of 0.1871 eV/H2 for the calculations with GGA-PW91 functional.
To account for the dispersion (van der Waals) interactions which are not included in
GGA-PW91 functional used, we also carry out calculations using the DFT-D method of
Grimme [34] with its default
force-field parameters as implemented in Dmol3, where Grimme’s empirical correction scheme for common density
functionals has been applied successfully to many chemical problems including
molecular hydrogen [2,3,15,14]. It can be seen that the first H2 molecule is adsorbed with an energy of Eave=0.4373 eV/H2 and Ead=0.4374eV/H2 on the doped complex calculated from Eqs. 2 and 3. The average distance
between H2 molecules to Ca atom is 2.4756 Å and the bond distance among hydrogen
molecules changes from 0.774 Å to 0.776 Å.
When the second H2 molecule is adsorbed on the doped complex, it presents a decrease of 31%
with respect to the first H2 molecule resulting in an energy of 0.3338 eV/H2. The following added molecules to doped complex presents an average
decrease of 5.49% respect to the second molecule. Having thus an average energy of
0.3204 eV/H2, and an average adsorption of 0.2736 eV/H2, higher energies than the calculations with GGA-PW91. The hydrogen
storage capacity obtained in this study is the same as for the results with PW91
(1.0825 to 6.1515) %wt (Table I). Thus, PW91
results can be taken as lower bounds for hydrogen storage energies whereas DFT-D
results can be taken as higher bounds since they present better energies in the
desirable range of 0.2 to 0.6 eV [14,15,17] for the hydrogen storage at ambient conditions as can be
seen in Fig. 5.
Compared with results in the literature, some researches [15,18] obtain 0.24 to 0.07 eV as the range of binding
energies for carbyne chains and 0.113 to 0.161 eV for the case of B80
decorated with calcium atoms, being in a range < 0:25 eV, making this system
(CaC12-6H2) a good candidate for hydrogen storage since it
has energies greater than 0.25 eV per H2 only for calculations with DFT-D
Grimme.
We performed the HOMO-LUMO spatial distribution analysis for the CaC12 and CaC12-6H2 systems shown in Fig. 6a,b,
respectively. Blue lobes display positive values and yellow lobes negative values of
the wave function. For CaC12system (Fig. 6
a) the distribution of electron density HOMO is on the single carbon
bonds, the Ca atom is placed in front of the positive lobe since in this region
there is an electron pair that can more easily yield or lose the carbyne molecule.
The LUMO distribution can be seen to be located on the triple bonds, place of lower
energy orbitals that are empty.
Calculations of energy difference HOMO-LUMO or gap (Δ) for the CaC 12-n H2 system, are summarized in Table II. It is observed that values of energy difference for the doped
complex with 1 to 3 H2 molecules give an average of 0.532 eV. Therefore,
a higher than the latter is required to adsorb on the doped complex. Otherwise, for
the doped complex with 4 to 6 hydrogen molecules, the average energy difference is
0.488 eV. This indicates that for the latter systems is relatively easier for
adsorbing molecules to the doped complex. Likewise, the average energy difference
HOMO-LUMO for CaC12-nH2system is 0.5103 eV.
Comparing this value with the binding energies of Table I, we have that the energy difference HOMO-LUMO is greater than
adsorption energy for the doped complex with one to six H2 molecules.
This confirms the stability of the CaC12 system, providing a good
physisorption alternative for hydrogen storage.
TABLE II HOMO-LUMO energy difference or gap (¢) of CaC12 and CaC12 - nH2, n =
1 - 6 system /in eV.
System |
HOMO(eV) |
LUMO(eV) |
( Δ) |
CaC12
|
-4.360 |
-4.124 |
0.236 |
CaC12-H2
|
-4.248 |
-3.705 |
0.543 |
CaC12-2
H2
|
-4.249 |
-3.714 |
0.535 |
CaC12-3
H2
|
-4.258 |
-3.740 |
0.518 |
Ca4C12-
H2
|
-4.265 |
-3.758 |
0.507 |
Ca5C12-
H2
|
-4.273 |
-3.786 |
0.487 |
CaC12-6 H2
|
-4.279 |
-3.807 |
0.472 |
Mulliken population analysis is summarized in Table
III, where we clearly observe the atomic charge +0.956e of Ca atom for
CaC12 system, coming out from electrons donation of 4s orbitals plus
a very small fraction of 4p and 3d orbitals of Ca atom, according to the graph PDOS
vs Energy. This means that there is a charge transfer of the very high Ca atom to
the nearest carbon atoms in the system. This calculation is in agreement with
results obtained by Cruz-Torres et al. [23] where it can be observed that the atomic charge
transfer of Ca atom into C12 system is +0.933e, having an overestimation
of 2.46%, due to a lower number of carbon atoms in our system.
TABLE III Mulliken population analysis of CaC12 and CaC12
- nH2 with n = (1 - 6) system.
System |
C (atom) |
Ca (atom) |
H (atom average) |
Ca C12
|
C1-0.399 |
0.956 |
- |
|
C2 -0.399 |
|
|
Ca
C12-H
2
|
C1- -0.405 |
1.072 |
-0.078 |
|
C2 0.042 |
|
|
Ca C12-
2H
2
|
C1-0.403 |
1.124 |
-0.078 |
|
C2 0.044 |
|
|
Ca C12-
3H
2
|
C1-0.409 |
1.113 |
-0.091 |
|
C2 0.041 |
|
|
Ca C12-
4H
2
|
C1-0.416 |
1.158 |
-0.090 |
|
C2 0.044 |
|
|
Ca C12- 5
H
2
|
C1-0.406 |
1.172 |
-0.090 |
|
C2 0.043 |
|
|
Ca C12- 6
H
2
|
C1-0.408 |
1.106 |
-0.081 |
|
C1-0.040 |
|
|
The whole CaC12-n H2 system values with n≤6 H2 molecules are shown in Table III,
where the average charge for Ca atom reaches a maximum value of +1.106e for six H2 molecules, while the average charge for hydrogen atoms is -0.081e,
increasing 15.69% with respect to the doped complex without hydrogen molecules. This
physisorption process for a maximum number of hydrogen molecules on carbyne doped
with calcium atom is stable as previously found [15], and easy for charge transfer in a mobile
application system.
Likewise, it can be observed that the atomic charge on carbon atoms 1 and 2 (Fig. 1b) is maintained at an average -0.4078e on
C1 atom, and 0.0423 on C2 atom. Reaching a maximum value when 5 H2 molecules are adsorbed in the CaC12 system, with a charge transfer value on Ca atom of +1.172e, with an
increment of 81% with respect to the +1.085e of CaC120-5H2 system [23], then the charge transfer decreases as the number
of hydrogens increases. We also observe a decrease of 6.79% among CaC120-5H2 and CaC120-6H2 reported previously [23], in agreement with our study 5.96% among CaC120-5H2 and CaC120-6H2 reported by us. Then the addition of the sixth hydrogen molecule in a
doped system proves a strong conservation charge transfer in the system.
Therefore, with the previous studies presented in this work, we can state that C12 carbyne molecule decorated with Ca atom can satisfactorily adsorb six H2 molecules by physisorption with an average energy of 0.1903 eV (PW91)
and 0.3204 (DFT-D) corresponding to 6.16 %wt, this storage capacity meets the
requirements established by US Department of Energy by the end of the year 2020.
4. Conclusions
In summary, we performed the analysis of Ca-carbyne complex and found the most stable
configuration from C12 carbyne molecule and Ca atom adsorbed on its outer surface, using
Materials Studio Dmol3 program. First, the carbyne molecule used in this work correspond to
polyyne type with alternation of triple (1.34 Å) and single (1.24 Å) bonds. Second,
we studied the case of a single Ca atom adsorbed to the carbyne surface, and we
calculate the binding energy of the CaC12 system, showing a value of 2.95 eV (PW91) and 2.23 eV(PBE) indicating
that the system is strongly chemisorbed.
The results show that up to six H2 molecules are physisorbed by the Ca atom with an average energy of
0.1890 eV per H2 (PW91). The first five H2 molecules tend to adsorb around the Ca atom and the sixth H2 molecule is adsorbed on top of the Ca atom.
To account for the dispersion (van der Waals) interactions which are not included in
GGA-PW91 functional used, we also carry out calculations using the DFT-D method of
Grimme, obtaining an average energy of 0.3204 eV per H2. Thus, PW91 results are taken as lower limits and DFT-D results are
taken as higher bounds since they present better energies in the desirable range of
0.2 to 0.6 eV for hydrogen storage at ambient conditions.
The hydrogen storage capacity obtained in this study 6.16 wt% agrees with objectives
established by the US Department of Energy (DOE), and satisfactorily meets the
target set for the year 2020.
Referencias
1 M. Kayanuma, U. Nagashima, H. Nishihara, T. Kyotani and H. Ogawa,
Chem. Phys. Lett. 495 (2010) 251-255.
[ Links ]
2 C. Li, J. Li, F. Wu, S. Li, J. Xia and L. Wang, J. Phys.
Chem 115 (2011) 23221-23225.
[ Links ]
3 S. Er, G. A. de Wijs and G. Brocks, J. Mater. Chem.
A 3 (2015) 7710-7714
[ Links ]
4 P. Jena, J. Phys. Chem. Lett. 2 (2011)
206-211.
[ Links ]
5 S. Niaz, T. Manzoor and A. H. Pandith, Renew. Sustain.
Energy Rev. 50 (2015) 457-469.
[ Links ]
6 F. J. Isidro-Ortega, J. H. Pacheco-Sánchez and L. A.
Desales-Guzmán, Int. J. Hydrogen Energy, 42 (2017)
30704-30717.
[ Links ]
7 Hydrogen storage | department of energy, Energy.gov. 2018
[Online]. Available, Available, https://energy.gov/eere/fuelcells/hydrogen -storage
[Accessed 22 February 2018].
[ Links ]
8 R. Ströbel, J. Garche, P. T. Moseley, L. Jörissen and G. Wolf,
J. Power Sources 159 (2006) 781-801.
[ Links ]
9 A. C. Dillon, K. M. Jones, T. A. Bekkedahl, C. H. Kiang, D. S.
Bethune and M. J. Heben, Nature 386 (1997)
377-379.
[ Links ]
10 C. Liu, Y. Y. Fan, M. Liu, H. T. Cong, H. M. Cheng and M. S.
Dresselhaus, Science 286 (1999) 1127.
[ Links ]
11 D. V. Schur et al., Int. J. Hydrogen
Energy 27 (2002) 1063-1069.
[ Links ]
12 Y. S. Nechaev, J. Nano Res. 12 (2010)
1-44.
[ Links ]
13 S. Patchkovskii, J. S. Tse, S. N. Yurchenko, L. Zhechkov, T.
Heine and G. Seifert, Proc. Natl. Acad. Sci. 102 (2005)
10439-10444.
[ Links ]
14 F. Li, C. Zhang, W. X. Ji and M. Zhao, Phys. Status
Solidi Basic Res. 252 (2015) 2072-2078.
[ Links ]
15 P. B. Sorokin, H. Lee, L. Y. Antipina, A. K. Singh and B. I.
Yakobson, Nano Lett. 11 (2011) 2660-2665.
[ Links ]
16 J. Wang, Y. Du and L. Sun, Int. J. Hydrogen
Energy 41 (2016) 5276-5283.
[ Links ]
17 H. Lee , J. Ihm, M. L. Cohen and S. G. Louie, Nano
Lett. 10 (2010) 793-798.
[ Links ]
18 G. Wu, J. Wang , X. Zhang and L. Zhu, J. Phys. Chem.
C 113 (2009) 7052-7057.
[ Links ]
19 H. Lee , J. Ihm , M. L. Cohen andS. G. Louie , Nano
Lett. 10 (2010) 793-798.
[ Links ]
20 K. Gopalsamy and V. Subramanian, Int. J. Hydrogen
Energy 39 (2014) 2549-2559.
[ Links ]
21 A. K. Singh andB. I. Yakobson , J. Mater. Sci.
47 (2012) 7356-7366.
[ Links ]
22 C. Ataca, E. Aktürk and S. Ciraci, Phys. Rev. B -
Condens. Matter Mater. Phys. 79 (2009) 1-4.
[ Links ]
23 A. Cruz-Torres, F. D. Castillo-Alvarado, J. Ortiz-López and J. S.
Arellano, Rev. Mex. Fis. 59 (2013) 126-134.
[ Links ]
24 X. L. Lei, G. Liu, M. S. Wu, B. Xu, C. Y. Ouyang and B. C. Pan,
Int. J. Hydrogen Energy 39 (2014)
2142-2148.
[ Links ]
25 M. Yoon, S. Yang, C. Hicke, E. Wang, D. Geohegan and Z. Zhang,
Phys. Rev. Lett. 100 (2008) 1-4.
[ Links ]
26 Q. Sun, Q. Wang, P. Jena and Y. Kawazoe, Am. Chem.
Soc. 127 (2005) 14582-14583.
[ Links ]
27 M. J. López, I. Cabria and J. A. Alonso, J. Phys. Chem.
C 118 (2014) 5081-5090.
[ Links ]
28 Y. H. Hu, Phys. Lett. Sect. A Gen. At. Solid State
Phys. 373 (2009) 3554-3557.
[ Links ]
29 M. Liu , V. I. Artyukhov, H. Lee , F. Xu andB. I. Yakobson ,
ACS Nano 7 (2013) 10075-10082.
[ Links ]
30 B. Delley, J. Chem. Phys. 92 (1990)
508-517.
[ Links ]
31 B. Delley, J. Chem. Phys. 113 (2000)
7756-7764.
[ Links ]
32 J. Perdew and Y. Wang, Physical Review B 45
(1992) 13244-13249.
[ Links ]
33 J. Perdew and W. Yue, Physical Review B 33
(1986) 8800-8802.
[ Links ]
34 B. R. Brooks et al., J. Comput.
Chem. 30 (2009) 1545-1614.
[ Links ]
35 A. M. Sladkov, V. I. Kasatochkin, Y. P. Kudryavtsev and V. V.
Korshak, Bull. Acad. Sci. USSR Div. Chem. Sci. 17 (1968)
2560-2565.
[ Links ]
36 R. B. Heimann, J. Kleiman and N. M. Salansky,
Nature 306 (1983) 164-167.
[ Links ]
37 P. Atkins, J. De Paula and R. Friedman, Physical
chemistry, Oxford University Press, (Oxford, 1st edn.
2014).
[ Links ]