Graphenylene Monolayers Doped with Alkali or ... - ACS Publications

Subscriber access provided by McMaster University Library

Article

Graphenylene Monolayers Doped with Alkali or Alkaline Earth Metals: Promising Materials for Clean Energy Storage Tanveer Hussain, Marlies Hankel, and Debra J Searles J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.7b02191 • Publication Date (Web): 28 Jun 2017 Downloaded from http://pubs.acs.org on June 28, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry C is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Graphenylene Monolayers Doped with Alkali or Alkaline Earth Metals: Promising Materials for Clean Energy Storage Tanveer Hussain*a, Marlies Hankel*a and Debra J. Searles a,b a

Centre for Theoretical and Computational Molecular Science, Australian Institute

for Bioengineering and Nanotechnology, The University of Queensland, Brisbane, Qld 4072, Australia b

School of Chemistry and Molecular Biosciences, The University of Queensland,

Brisbane, Qld 4072, Australia [email protected] [email protected] +61 7 3346 3976 Abstract: Biphenylene carbon (BPC) functionalized with alkali and alkaline earth metals are considered for their potential as high capacity hydrogen storage materials. Using density functional theory calculations all the dopants are found to bind with the BPC substrate strongly enough to avoid metal clustering. The strong binding between the membrane and the dopant has been confirmed by consideration of their partial density of states. The positive charge of the metal dopants resulting from the donation of charges to the BPC sheet, polarizes the nearby H2 molecules and binds them through electrostatic and van der Waals interactions. Each dopant can adsorb multiple H2 molecules leading to a moderately high storage capacity (6.14 wt%) with adsorption energies (~ -0.30 eV/H2) suitable for ambient condition applications.

Introduction: Hydrogen (H2) provides a promising energy resource, and a clean substitute for fossil fuels. Its abundance, high energy and zero carbon content make H2 an attractive clean and sustainable energy carrier, but the gaseous nature and its small size hamper its use

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 26

in fuel cells.1-4 Efficient storage of hydrogen, has been the biggest challenge for transition to a hydrogen economy. Among various approaches, a material-based storage strategy for H2 could be of advantageous provided the material used is reusable, cost efficient and the operation (adsorption/desorption) is at, or close to, ambient conditions.5, 6 Very few materials have been identified that fulfill the above-mentioned criteria, however it has been found that metal decorated carbon-based nanostructures can anchor the H2 with suitable adsorption energies. Due to their large surface area and existence in various morphologies including nanotubes,7-10 fullerenes,11, 12 graphene, 13-15

and graphane16-19 they have found their promise in hydrogen storage applications.

Graphenylene or biphenylene carbon (BPC) is a 2D sp2-carbon membrane with uniform pores. It was first predicted to exist over 40 years ago by Balban 20, 21 and is a possible isomerisation product of graphyne. 22, 23 BPC has a unique structure which is composed of cyclohexatriene-like units (C6 rings) and cyclobutadiene-like units (C4 rings) and periodic pores with a diameter of 3.2 Å. BPC is essentially a form of [N] phenylene, which has already have been synthesized in zigzag, linear and helical forms24, however the BPC sheet has not yet been synthesized. Brunetto et al.25 recently used ab initio quantum molecular dynamics simulations to predict that BPC could be obtained from porous graphene 26 via dehydrogenation, proposing a possible synthetic pathway. Theoretical studies have also proposed BPC nanotubes and fullerenes would be stable.27, 28 BPC’s unique structure has seen it investigated for several different applications. Song et al.29 showed that if BPC is considered as a porous membrane for gas separation, it has a higher selectivity for H2 diffusion than other gases, with selectivities of 1012 for H2/CO, 1013 for H2/N2, 1014 for H2/CO2 and 1034 for H2/CH4 at room temperature. Theoretical studies also showed that BPC has a small band gap of around 0.8 eV using DFT tight binding calculations with the BLYP functional25 and 1.08 eV with a hybrid B3LYP functional.20 Liu et al.31 determined the band gap of BPC using a hybrid functional to be 0.48 eV and that it could be tuned over a range of 0.075 to 4.98 eV by hydrogenation and 0.024 to 4.87 eV by halogenations. Yu 32 and Hankel and Searles 33 investigated BPC as an anode material for lithium ion batteries. The studies showed that for monolayer and bilayer BPC, specific capacities of Li3C6 and Li2.5C6 could be achieved. Both in-plane and out-ofplane lithium diffusion on the monolayer show barriers that are less than 0.57 eV making it a promising material for a lithium ion battery anode.


2

Page 3 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Despite it being a carbon-based nanostructure that has promise for a number of applications, the application of BPC for H2 storage has not been considered extensively because H2 only binds weakly on it, like other carbon-based nanostructures. Yadav et al.34 investigated hydrogen storage on lithium doped twodimensional (2D) membranes including BPC. They found that up to four lithium could be adsorbed on a C64 unit of BPC with two-sided decoration. Each of the four lithium atoms could then adsorb up to two hydrogen molecules, giving a capacity of 2.6 wt%. Pan et al.35 investigated hydrogen storage on calcium doped 2D carbon membranes. For BPC (C64), they adsorbed two calcium atoms in the two large pores. These could then adsorb 12 hydrogen molecules each, six on either side of the membrane and giving a maximum capacity of 8.57 wt%. In the present study, we carry out a systematic study where the monolayer is functionalized with metal adatoms to enhance the adsorption of H2. The metal adatoms considered are alkali (Li, Na, K) and alkaline earth metals (Mg and Ca).

Computational method: In this project, first principles calculations based on density functional theory were performed as implemented in VASP code.36, 37 The projector augmented wave method (PAW) were used to consider the electron-ion interaction.38 For the treatment of exchange and correlation functional, generalized gradient approximation (GGA) of Perdew-Burke and Ernzerhof for the exchange-correlation was used.39 Previous studies showed that in case of these weakly interacting systems (H2 interaction with metal adatoms) corrections to van der Waals interaction through use of vdW-GGA resulted in more accurate adsorption energies than GGA only, which usually underestimates the values.16-18 Thus we have employed DFT-D2 method of Grimme to obtain more reliable adsorption energies.40,43 The Monkhorst-Pack scheme was used for the sampling of Brillouin zone with a 3×3×1 k-point mesh for structural optimization and a denser mesh of 7×7×1 for density of states calculations.41 All the structures were converged until the forces acting on each ion became less than 0.01 eV/ Å. The binding energies of metal adatoms on BPC monolayer were calculated using:

Eb= [E(nX@BPC) – E(BPC) – n E(X)] / n.


(1)

3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 26

Here the 1st, 2nd and 3rd terms on the right represent the total energies of BPC doped with metal adatoms, pure BPC and metal adatoms, respectively. {n=1,..,5, X=Li, Na, K, Mg, Ca}. The energy of the metal adatom is that of an isolated metal in the gas phase. The H2 adsorption energies were calculated using: Eb= [E (mH2@XnBPC) – E(XnBPC) – m E(H2)] / m.

(2)

In eq. (2) the 1st and 2nd terms on the right hand side of the equality represent the total energies of metal-doped BPC with m H2 or no H2 molecules respectively and the 3rd term is the total energy of m H2 molecules.

Results and discussion: Before studying its interaction with metal adatoms and H2 storage properties, a brief discussion on the structural and electronic properties of BPC is given. The calculated lattice parameter of 6.750 Å is in good agreement with the previous studies.31,32 As shown in fig. 1, the optimized structure of the 2×2 supercell used in this study has three types of rings: a C12 circular ring (A); a C6 hexagonal ring (B); and a C4 square ring (C). The carbon-carbon bond lengths in the hexagonal and square rings are found to be 1.365 Å and 1.472 Å, respectively, and these results are in accord with those from reference.32

C

B

A

Fig.1 The top view of the optimized BPC 2 x 2 supercell. Black balls represent the carbon atoms. The letters denote the circular ring (A), the hexagonal ring (B) and the square ring (C).


4

Page 5 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


It is well known that self-interaction in DFT calculations usually results in an underestimation of the calculated band gap. BPC hardly shows any band gap in our calculations, as shown in fig. 2, and this is consistent with previous results.29,31 Liu et al.31 also used a Heyd–Scuseria–Ernzerhof hybrid functional, and found that BPC has a small band gap which opens up considerably upon the hydrogenation and halogenation of BPC. Although the magnitudes of the band gaps were found to be affected by use of the more accurate hybrid functional, the qualitative changes due to hydrogenation and halogenation were similar. The partial density of states shown in the lower panel for fig. 2 indicate that the C(p) contributes to the DOS near the Fermi energy.

Fig. 2. Total and partial density of states of the pure BPC sheet with energies given relative to the Fermi energy.

We now consider the effects of alkali (Li, Na, K) and alkaline (Mg, Ca) metal adatoms on BPC. Due to the three types of rings in BPC, there exist several possible bridge (over bond) and hollow (over ring) binding sites for the metal adatoms as well as the obvious sites, which are directly above the C atoms. In order to obtain the most


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 26

stable configurations, the binding energies (Eb) of the dopants initially located on all bridge and hollow binding sites have been calculated. For all the stable BPC-adatom structures and the uniform distribution of adatoms over the sheet, the calculated Eb of the dopants should exceed their corresponding cohesive energies (Ec). If this is not the case then clustering of the dopants is the more likely outcome.19,42 In our calculations for the different possible positions for the metal adatoms we found that the metal prefers to sit above the six-membered ring. The calculated Eb of single Li, Na, K, Mg and Ca atoms above the C6 ring are -2.25 eV, -1.82 eV, -2.07 eV, -1.75 eV and -2.13 eV respectively. The corresponding cohesive energies (Ec), relative to the bulk metal, of Li, Na, K, Mg and Ca are -1.63 eV, -1.11 eV, -0.93 eV, -1.51 eV and -1.84 eV, respectively.32,42 The binding energy of Li can be compared to the results of Yadav et al.34 where LDA, GGA and vdW-DF2 functionals were considered.

Although the values are similar, our binding energy has a larger

magnitude than their GGA result (-2.225 eV) and smaller than their vdW-DF2 result (-1.859 eV), which is consistent with expectations. Large Eb to Ec ratios for all the dopants suggest that they would prefer to disperse atomically without being clustered on the BPC. Due to its large surface area, the BPC sheet is expected to be capable of binding a large number of adatoms, which should greatly enhance the H2 storage capacity. Considering this, we have introduced several adatoms on BPC and studied their binding with the sheet. The loading was increased by adding metal atoms one at a time to each of the empty rings, optimizing the structure and selecting the lowest energy structure for the addition of the next metal atom (for results of Li doping, see Table 1 in the supporting material). A loading of five metal atoms (10.42 mol% doping concentration) gave binding energies that are higher than their respective cohesive energies while ensuring that the distance between the dopants remained reasonably large. The spacing of the metal atoms reduced the possibility of cluster formation and provided appropriate space for anchoring a large number of H2 molecules, which could help achieve a higher H2 storage capacity. Therefore, loadings of five metal atoms were considered in further work. In the distribution of the five metal atoms we initially placed four atoms above six-membered rings far apart and either side of the membrane. The fifth metal atom was placed above the large pore. In the case of Li and Ca, metal atoms were located away from the sixmembered ring after structure optimisation, even though this was the minimum energy position when only one metal atom was considered.


6

Page 7 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


With five metal atoms, the value of Eb per dopant was been found to be -1.88 eV, 1.60 eV, -1.85 eV, -0.69 eV and -2.02 eV for Li, Na, K, Mg and Ca respectively, which shows that the metal binds strongly to the doped BPC monolayers, apart from Mg which has a low binding energy per dopant atom. The complete results are given in Table 1, and the top and side views of the optimized geometries are shown in fig. 3. Since the average binding energy of five Mg atoms to BPC is much weaker than the Mg cohesive energy (-1.51 eV), cluster formation is expected and we have excluded it from further consideration for hydrogenation.

Table 1: Calculated binding energies, distance of dopants to the BPC, distance between dopant atoms and Bader charge acquired by the dopants.

System

Binding energy

Average distance Minimum

per dopant

from dopant atom dopant-dopant atom (e-)

atom (eV)

to BPC (Å)a

distance (Å)

BPC-5Li

-1.88

2.17

3.70

+0.986

BPC-5Na

-1.60

2.66

4.40

+0.901

BPC-5K

-1.85

2.94

5.11

+0.784

BPC-5Mg -0.69

2.80

4.71

+1.16

BPC-5Ca

2.59

4.62

+1.20

a

-2.02

Charge

per dopant

Obtained by averaging the distance between each of the metal dopants and their

nearest carbon atom.


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 26

Fig. 3 Side (left) and top (right) views of the supercell for the optimized structures of (a) BPC-5Li, (b) BPC-5Na, (c) BPC-5K (d) BPC-5Mg and (e) BPC-5Ca. Black, blue, yellow, magenta, orange and cyan balls represent C, Li, Na, K, Mg and Ca atoms respectively.

Fig. 3 shows that the metal adatoms are located over the C6 rings on both sides of the BPC sheet. One metal adatom is located over or inside the large C12 ring. The ionic radii of Li and Mg are small compared to Na, K and Ca, and therefore to be in reasonable proximity to a C atom, their optimized position needs to be away from the centre of the C12 ring. The K ionic radii is the largest, which explains why it sits above the centre of the C12 ring. Ca and Na are of similar size and are found in the centre of the C12 ring and in the plane of the BPC sheet. Only for the case of Li and Ca are the adatoms located over the C4 rings. In the case of Li this might be explained by recognizing that the Li are small enough to move closer together to find geometry with a lower energy than our initial placement. Although the single Li atom prefers to be located on a six-membered ring, there is a local minimum energy structure where it is on a four-membered ring, which has a very similar binding energy. In the case of Ca, two of the adatoms have moved over the four-member rings. As the Ca is larger than Li this might be explained by space constraints. The alkali metals (Li, Na and K) have a lower charge than the alkaline earth metals (Mg and Ca), and this may explain why the alkali metals are able to lie above each other on the C4 and C6 rings. The Ca


8

Page 9 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


ions are slightly displaced above and below the C6 ring whereas the Mg ions are above and below different C4 rings. The top view, right panel, shows no visible distortion to the BPC sheet. The side view, left panel, shows that for Li, Na and Ca the fifth adatom is located within the plane of the sheet. For Ca this causes a larger distortion to the BPC sheet than for Li or Na. For Mg the fifth adatom is located well above the sheet plane at a similar distance to the adatoms over the rings. This would cause a larger repulsion between the Mg adatoms on one side of the sheet and could explain the smaller binding energy per Mg adatom.

The nature of the bonding of metal adatoms on BPC has been investigated by considering the density of states (DOS) which are shown in fig. 4. Here we show both the total (TDOS) and partial (PDOS) for Li, Na, K and Ca adatoms on the BPC sheet. It is evident from the TDOS and PDOS plots that all the doped systems exhibit metallic character with significant overlapping of the C(p) and X(s) orbitals (X=Li, Na, K and Ca). In all cases, the main contribution comes from the C(p) orbital with an energy lower than the Fermi energy (i.e. left of E-Ef=0 in the figure). For the Li adatom (fig. 4a), significant contributions from Li(s) appear at 0.75 and 1.2 eV on the right of the Fermi level, where they overlaps with the C(p) orbitals. For Na (fig. 4b) there is hardly any contribution from the Na orbitals in the valence band, however there is a sharp peak due to Na(s) at the Fermi level. Two of the Na(s) peaks hybridize with C(p) at 0.65 eV and 1.2 eV causing the binding of Na on the BPC monolayer. For the K adatom (fig. 4c), a large peak due to K(s) indicates hybridization with the C(p) orbitals at the Fermi level and explains the binding of the K to the BPC. For Ca (fig. 4d), one can see sharp peaks due to the Ca(p) orbital just below the Fermi energy and around 0.5 eV, and overlap with the C(p) orbitals suggest hybridization and a strong binding of Ca to BPC. The DOS analysis is complemented by COHP analysis and is given in fig-S1 in supplementary information.


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 26

Fig. 4. Total and partial density of states for BPC sheet bonded with alkali metal adatoms (a) Li, (b) Na, (c) K and (d) Ca. Energies are given relative to the Fermi energy.

Now we describe the hydrogenation of the stable metal (Li, Na, K, Ca) functionalized BPC sheets. Due to the smaller electronegativities of the metal dopants compared to the carbon, a substantial amount of charge will transfer from the dopants to the BPC sheet, leaving the dopants in a partially positive charge state as seen from the results in table 1 where the Bader charges on the metal atoms are given. The accumulation of this charge is vital to adsorption of the H2 molecules, and when H2 molecules are brought into the vicinity of the dopants the H2 molecules are polarized and are held to the substrate. As the adsorbed H2 has a partial charge, the next H2 will be situated at some distance from the first due to repulsion. We continued to introduce H2 until the system became saturated, which is evident from fact that further H2 molecules did not bind. We found that each functionalized system adsorbed 20 H2 molecules with adsorption energies between -0.182 eV and -0.257 eV. The adsorption energies and storage capacities are given in table 2.


10

Page 11 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


System

Average adsorption Average H2-

Storage

energy/H2 (eV)

capacity

dopant distance (Å)

C48Li5+20H2

-0.257

2.12

6.14 wt%

C48Na5+20H2

-0.228

2.29

5.51 wt%

C48K5+20H2

-0.182

2.35

4.93 wt%

C48Ca5+20H2

-0.204

2.30

4.90 wt%

Table 2: Calculated adsorption energies per H2 molecule and storage capacities for the maximum loading of 20 H2 molecules. Table 2 shows that high capacities can be achieved for the metal doped BPC sheet. For Li and Na they reach the target for 2017 of 5.5 wt% proposed by the DOE. Furthermore, the calculated binding energies per H2 molecule for maximum loading lies within the desired range for efficient and reversible H2 materials. Fig. 5 shows the structures of the metal-doped BPC sheets with the maximum loading of 20 H2 molecules. Here different configurations were considered before reaching to saturation in H2 adsorption. The hydrogenation process takes place in a step-wise manner by considering various configurations of single H2 on metal functionalized BPC sheet. The lowest energy configuration of single H2 adsorption is used an initial structure for the 2nd H2 adsorption. Here again several initial structures have been considered and the optimized one that corresponds to the lowest energy has been chosen for the adsorption of 3rd H2. This process continues unless the metal functionalized systems reaches the saturation, which is upon the adsorption of 20H2 molecules in each case. Upon hydrogenation the doped systems undergo some structural deformation as seen in the left images in Fig. 5, with the maximum displacement between carbon atoms in the direction perpendicular to the sheet being 0.48 Å, 0.31 Å, 0.33 Å and 1.0 Å with Li, Na, K and Ca doping, respectively. The deformation with Ca doping is higher than with the other dopants, yet even at a maximum H2 loading on the metallized BPC monolayers remain stable with no breakage of bonds. It is therefore expected that this distortion would not be


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 26

problematic for their application as reversible sensors or hydrogen storage materials. For Li, the H2 molecules are found to be very close to the Li adatoms while for the other cases they are distributed more evenly over the area of the BPC sheet. However, as can be seen in table 2 the average distance is still within the physisorption range.

Fig. 5 Side (left) and top (right) views of the optimized structures of (a) BPC-5Li, (b) BPC-5Na, (c) BPC-5K, and (d) BPC-5Ca loaded with maximum of 20 H2 molecules. Black, blue, yellow, magenta, cyan and green balls represent C, Li, Na, K, Ca and H atoms respectively.


12

Page 13 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Fig. 6 shows the change in binding energy per H2 molecule with increasing H2 loading. As can be expected the binding energy generally decreases with increased loading. This decrease is monotonic in all cases except for the change from 15 to 20 H2 molecules for Ca, where the change is very small and probably due to some H2 being placed a local minima. The global minimum is very difficult to find in such a complicated energy landscape.

Fig. 6: Adsorption energies of Li, Na, K and Ca doped BCP sheet with different H2 loadings.

When we compare our results to other work on similar systems

34,35

we find some

marked differences. Yadav et al.34 found that only four Li can adsorb on the BPC membrane. Also on each of these, only two H2 molecules can adsorb. We note that they used the vdW-DF2 functional to treat the van der Waals interactions whereas we used DFT-D2. They found that although the H2 adsorption energies for GGA alone and vdW-DF2 are very similar, with the GGA energies slightly stronger, those for LDA are also stronger. The dispersion correction by Grimme (D2) we employed in our study are known to overbind and would therefore produce slightly higher binding energies compared to vdW-DF2. This would result in a larger number of metal atoms being adsorbed as well as a larger number of H2 molecules. However, when we compare the binding energy for a single Li over the six-membered carbon ring, -2.25 eV with D2, with the value reported by Yadav et al.34 -1.859 eV with vdW-DF2, -


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 26

2.225 with GGA only and -2.688 eV with LDA, the adsorption energy reported here is just slightly stronger than the GGA value and significantly weaker than the LDA value. Considering that the adsorption energy for five lithium atoms found in this study, -1.88 eV, is still much stronger than the cohesive energy and the small differences between the vdW-DF2 and D2 energies, we can assume that even with the vdW-DF2 van der Waals corrections up to five lithium atoms should be able to be adsorbed on the BPC membrane. This also assumes that the lithium atoms are well dispersed on the membrane, which helps to avoid clustering. The larger lithium doping in our case will influence the amount of hydrogen that can be adsorbed. We are therefore confident that while the final capacity could be slightly smaller than the one reported here if the vdW-DF2 (or similar) corrections would have been used, that the difference would not be large. Pan et al.35 also used vdW-DF2, and decorated BPC (C64) with one calcium atom in each of the large round pores. This was due to the fact that they found the large pore to be the most stable adsorption site for calcium on BPC. They found that up to 12 H2 molecules can be adsorbed on each of the two calcium atoms, six per Ca on either side of the membrane leading to a capacity of 8.57 wt%. This is higher than the 4.9 wt% that we found, and is due to the larger calcium coverage that we obtained. We found that four H2 molecules could be adsorbed on each Ca, compared to five H2 per Ca in the study by Pan et al.35

Our rationale to use a larger metal coverage of the membrane was to provide multiple adsorption sites for hydrogen molecules, which are still well dispersed. Our results, those by Yadav et al.34 and Pan et al.35 indicate that BPC (C64) is a very promising material for hydrogen storage after metal decoration. Different metal decorations are possible all showing high capacity for hydrogen storage.

Conclusion: We studied the interaction of BPC sheet with alkali and alkaline earth metal dopants and further explored their hydrogen storage properties by means of periodic density functional theory. The most stable configurations and the maximum number of dopants over the monolayers were investigated. It was determined that five adatoms of Li, Na, K or Ca could be adsorbed to BPC with reasonably high binding energies


14

Page 15 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


exceeding their corresponding cohesive energies, ensuring a uniform distribution rather than metal clustering. The partial density of states confirmed the existence of a strong bonding between the metal adatoms and the BPC monolayer. The transfer of electric charge from the dopants to the BPC sheet leads to polarization of the H2 molecules when they are exposed to the dopants. Electrostatic and van der Waals interactions then result in adsorption of multiple H2 on the metal functionalized BPC sheets. The calculated adsorption energies of H2 were found to be in an ideal range for the practical mobile applications. Each dopant could adsorb a maximum of four H2 molecules resulting a significantly high H2 storage capacity. The results for the H2 adsorption energies indicated that light metal decorated BPC sheets could be considered as effective H2 storage materials.

Supporting Information The variation in vdW corrected binding energies upon the increased Li doping concentration in BPC monolayer is given in table 1 in supplementary information. Crystal Orbital Hamilton Population (COHP) analysis of BPC monolayer functionalized with Li, Na, K and Ca at 10.42% doping concentration is also given in fig-S1.

Acknowledgements This research was undertaken with the assistance of resources provided at the NCI National Facility at the Australian National University through the National Computational Merit Allocation Scheme supported by the Australian Government, support from the Queensland Cyber Infrastructure Foundation (QCIF) and the University of Queensland Research Computing Centre (RCC). TH is indebted to the University of Queensland for financial support through UQ postdoctoral fellowship.

References: 1.

Schlapbach, L.; Züttel, A. Hydrogen-Storage Materials for Mobile Applications. Nature 2001, 414, 353−358


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

2.

Page 16 of 26

J.A. Turner, J. A. Sustainable Hydrogen Production Science 2004, 305, 972974

3.

J. Alper, J. Water Splitting Goes au Naturel. Science 2003, 299, 1686-7

4.

T. Hussain, T.; Pathak, B.; Ramzan, M.; Maark, T. A.; Ahuja, R. Calcium Doped Graphane as a Hydrogen Storage Material. Appl. Phys. Lett. 2012, 100, 183902-5.

5.

Bhatia, S. K.; Myers, A. L. Optimum Conditions for Adsorptive Storage Langmuir 2006, 22, 1688-1700.

6.

Hussain, T.; Pathak, B.; Maark, T. A.; Araujo, C. M.; Scheicher, R. H.; Ahuja, R. Ab Initio Study of Lithium-Doped Graphane for Hydrogen Storage. Euro Phys. Lett. 2011, 96, 27013-16.

7.

Tekkaya, E. D.; Karatepe, N. Hydrogen Adsorption of Carbon Nanotubes Grown on Different Catalysts. Int. J. Hyd. Energy, 2015, 40, 7665-70.

8.

Park, S. J.; Lee, S. Y. Hydrogen Storage Behaviors of Platinum-Supported Multi-Walled Carbon Nanotubes. Int. J. Hyd. Energy, 2010, 35, 13048-54.

9.

Liu, C.; Chen, Y.; Wu, C. Z.; Xu, S. T.; Cheng, H. M. Hydrogen Storage in Carbon Nanotubes Revisited. Carbon 2010, 48, 452-55.

10.

Berseth, P. A.; Harter, A. G.; Zidan, R.; Blomqvist, A.; Araújo, C. M.; Scheicher, R. H.; Ahuja, R.; Jena. P. Carbon Nanomaterials as Catalysts for Hydrogen Uptake and Release in NaAlH4. Nano Lett. 2009, 9, 1501-1505.

11.

Li, M.; Li, Y.; Zhou, Z.; Shen, P.; Z. Chen, Z. Ca-Coated Boron Fullerenes and Nanotubes as Superior Hydrogen Storage Materials. Nano Lett. 2009, 5, 1944-48.

12.

Yoon, M.; Yang, S.; Hicke, C.; Wang, E.; Geohegan, D.; Zhang, Z. Calcium as the Superior Coating Metal in Functionalization of Carbon Fullerenes for High-Capacity Hydrogen Storage. Phys. Rev. Lett. 2008, 100, 206806-9.

13.

Liao, J. H.; Zhao, Y. J.; Yang, X. B. Controllable Hydrogen Adsorption and Desorption by Strain Modulation on Ti Decorated Defective Graphene. Int. J. Hyd. Energy 2015, 40, 12063-71.

14.

Valencia, H.; Gil, A.; Frapper, G. Trends in the Hydrogen Activation and Storage by Adsorbed 3d Transition Metal Atoms onto Graphene and Nanotube Surfaces: A DFT Study and Molecular Orbital Analysis. J. Phys. Chem. C 2015, 119, 5506-5522.


16

Page 17 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


15.

Lee, H.; Ihm, J.; Cohen, M. L.; Louie, S. G. Calcium-Decorated GrapheneBased Nanostructures for Hydrogen Storage. Nano Lett. 2010, 10, 793-798.

16.

Hussain, T.; De Sarkar, A.; Ahuja, R. Functionalization of Hydrogenated Graphene by Polylithiated Species for Efficient Hydrogen Storage. Int. J. Hyd. Energy 2014, 39, 2560-2566.

17.

Hussain, T.; Maark, T. A.; De Sarkar, A.; Ahuja, R. Polylithiated (OLi2) Functionalized Graphane as a Potential Hydrogen Storage Material. 2012, Appl. Phys. Lett. 101, 243902-5

18.

Hussain, T.; Islam, M. S.; Rao, G. S.; Panigrahi, P.; Gupta, D.; Ahuja, R. Hydrogen Storage Properties of Light Metal Adatoms (Li, Na) Decorated Fluorographene Monolayer. Nanotechnology 2015, 26, 275401-6.

19.

Hussain, T.; De Sarkar, A.; Ahuja, A. Strain Induced Lithium Functionalized Graphane as a High Capacity Hydrogen Storage Material. Appl. Phys. Lett. 2012, 101, 103907-103911.

20.

Balban, A. T.; Rentea, C. C.; Ciupitu, E. Estimation of the Relative Stability of Several Planar and Tridimensional Lattices for Elementary Carbon. Rev. Roum. Chim. 13, 231-247 (1968)

21.

Randić, M.; Balban, A. T.; Plavšić, D. Applying the Conjugated Circuits Method to Clar Structures of [n] Phenylenes for Determining Resonance Energies. Phys. Chem. Chem. Phys. 2011, 13, 20644-48.

22.

Baughman, R. H.; Eckhardt, H. Structure-Property Predictions for New Planar Forms of Carbon: Layered Phases Containing sp2 and sp Atoms J. Chem. Phys. 1987, 87, 6687-99.

23.

Ivanovskii, A.L. Graphynes and Graphdyines. Progress in Solid State Chemistry 2013, 41,1-19.

24.

Schulman, J. M.; Disch, R. L. A Theoretical Study of Large Planar [N] Phenylenes. J. Phys. Chem. A 2007, 111, 10010-10014.

25.

Brunetto, G.; Autreto, P. A. S.; Machado, L. D.; Santos, B. I.; dos Santos, R. P. B.; Galvão, D. S. Nonzero Gap Two-Dimensional Carbon Allotrope from Porous Graphene J. Phys. Chem. C 2012, 116, 12810-13.

26.

Bieri, M.; Treier, M.; Cai, J.; Aït-Mansour, K.; Ruffieux, P.; Gröning, O.; Kastler, M.; Rieger, R.; Feng, X.; Müllen, K.; Fasel, R. Porous Graphenes: Two-Dimensional Polymer Synthesis with Atomic Precision. Chem. Commun. 2009, 45, 6919-6921.


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

27.

Page 18 of 26

Koch, A. T.; Khoshaman, A. H.; Fan, H. D. E.; Sawatzky, G. A.; Nojeh, A. Graphenylene Nanotubes. J. Phys. Chem. Lett. 2015, 6, 3982-87.

28.

Paupitz, R.; Junkermeier, C. E.; van Duin, A. C. T.; Branicio, P. S. Fullerenes Generated from Porous Structures. Phys. Chem. Chem. Phys. 2014, 16, 2551525522.

29.

Song, Q.; Wang, B.; Deng, K.; Feng, X.; Wagner, M.; Gale, J. D.; Müllen, K.; Zhi, L. Graphenylene, a Unique Two-Dimensional Carbon Network with Nondelocalized Cyclohexatriene Units. J. Mater. Chem. C 2013, 1, 38-41.

30.

De La Pierre, M.; Karamanis, P.; Baima, J.; Orlando, R.; Pouchan, C.; Dovesi, R. Ab Initio Periodic Simulation of the Spectroscopic and Optical Properties of Novel Porous Graphene Phases. J. Phys. Chem. C 2013, 117, 2222-2229.

31.

Liu, W.; Miao, M.; Liu, J. Band gap Engineering of Graphenylene by Hydrogenation and Halogenation: a Density Functional Theory Study. RSC Adv. 2015, 5, 70766-70771.

32.

Yu, Y. X. Graphenylene: a Promising Anode Material for Lithium-ion Batteries with High Mobility and Storage. J. Mater. Chem. A 2013, 1,1355913566.

33.

Hankel, M.; Searles, D. J. Lithium Storage on Carbon Nitride, Graphenylene and Inorganic Graphenylene. Phys. Chem. Chem. Phys. 2016, 18, 1420514215.

34.

Yadav, S.; Tam, J.; Singh, C. V. A First Principles Study of Hydrogen Storage on Lithium Decorated Two Dimensional Carbon Allotropes. Int. J. Hydrogen Energy, 2015, 40, 6128-6136.

35.

Pan, R.; Fan, X.; Luo, Z.; An, Y. Calcium Decorated Two Dimensional Carbon Allotropes for Hydrogen Storage: A first-principles study. Comput. Mater. Sci. 2016, 124, 106-113.

36.

Kresse, G.; Hafner, J. Ab Initio Molecular Dynamics for Liquid Metals. Phys. Rev. B 1993, 47, 558-561.

37.

Kresse, G.; Hafner, J. Ab Initio Molecular-Dynamics Simulation of the Liquid-Metal–Amorphous-Semiconductor Transition in Germanium. Phys. Rev. B 1994, 49, 14251-14269

38.

Blochl, P. E. Projector Augmented-Wave Method. Phys. Rev. B 1994, 50, 17953-17979.


18

Page 19 of 26

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


39.

Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized Gradient Approximation Made Simple. Phys. Rev. Lett. 1996, 77, 3865-3868

40.

Grimme, S. Semiempirical GGA-Type Density Functional Constructed with a Long-Range Dispersion Correction. J. Comput. Chem. 2006, 27, 1787-1799.

41.

Monkhorst, H. J.; Pack, J. D. Special Points for Brillouin-Zone Integrations. Phys. Rev. B 1976, 13, 5188-5192.

42.

Chan, K. T.; Neaton, J. B.; Cohen, M. L. First-Principles Study of Metal Adatom Adsorption on Graphene. Phys. Rev. B 2008, 77, 235430-23541.

43.

Lee, K.; Murray, E. D.; Kong, L.; Lundqvist, B. I.; Langreth, D. C. HigherAccuracy Van der Waals Density Functional. Phys. Rev. B 2010, 82, 0181014.


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 20 of 26

Page 21 of 26


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Page 22 of 26

Page 23 of 26


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60



1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

C

B

A


Page 24 of 26

Page 25 of 26

Adsorption energy, eV

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


0.4

Li Na K Ca

0.35

0.3

0.25

0.2

0.15

0.1

0.05

0 5

10 15 Number of hydrogen molecules


20


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42

5Li-‐BPC 5Na-‐BPC

5K-‐BPC 5Mg-‐BPC 5Ca-‐BPC ACS Paragon Plus Environment Alkali and Alkaline Metal Func8onalized BPC Monolayers

Page 26 of 26

Graphenylene Monolayers Doped with Alkali or ... - ACS Publications

Recommend Documents