Computational Investigation on the Electronic Structure and

Feb 18, 2019 - Using the state-of-the-art theoretical method, we have investigated the electronic and optical properties of a thiophene-based covalent...
1 downloads 0 Views 2MB Size
This is an open access article published under an ACS AuthorChoice License, which permits copying and redistribution of the article or any adaptations for non-commercial purposes.

Article Cite This: ACS Omega 2019, 4, 3556−3564

http://pubs.acs.org/journal/acsodf

Computational Investigation on the Electronic Structure and Functionalities of a Thiophene-Based Covalent Triazine Framework Biswajit Ball, Chandrima Chakravarty, Bikash Mandal, and Pranab Sarkar* Department of Chemistry, Visva-Bharati University, Santiniketan 731235, India

Downloaded via 193.56.75.48 on February 18, 2019 at 15:50:36 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Using the state-of-the-art theoretical method, we have investigated the electronic and optical properties of a thiophene-based covalent triazine framework (TBCTF). We have found that TBCTF is a direct band gap semiconductor. Our calculations reveal that constitutional isomerism is a tool for band gap tuning. The variation of band gap can be achieved by the bilayer TBCTF formation and further can be tuned by the z-axial strain. We have designed a new two-dimensional van der Waals heterostructure g-ZnO/TBCTF, which shows type-II band alignment, ensuring effective separation of photogenerated electron−hole pairs. This composite system also exhibits enhanced absorption in the visible range compared to that of individual g-ZnO and TBCTF monolayers. Therefore, this composite system may find potential application in photovoltaic devices. We have also investigated the hydrogen adsorption ability of TBCTF through Li atom doping. Our results indicate that the calculated hydrogen adsorption energies lie in the range, which is suitable for reversible hydrogen storage under ambient conditions. Therefore, the lithium-doped TBCTF may be a potential candidate for the hydrogen storage material. increases, the band gap decreases.28 However, the electronic properties of these COFs remain unaltered when deposited on graphene, indicating the robustness of its inherent properties. Zhou et al.25 have predicted that COF-5 shows effective carrier separation of photogenerated electrons and holes, suggesting a potential candidate for photovoltaic devices. They have also studied the optical properties of TP-COF and NiPc-PBBA COF, indicating their applicability in optoelectronic devices. Yang et al.29 have reported 10 new COFs, which show strong optical response in the visible and near infrared region. Liang et al.30 have investigated the electronic properties of some COFs and found that the variation of linkage conformation does not change the electronic properties significantly. They have also suggested that the presence of these COFs on the hBN substrate causes reduction of band gap because of the substrate polarization effect.30 Wang et al.31 have suggested that the reason behind the flat band characteristics of 2D boroxine-linked COFs can be attributed to the little aromaticity of the boroxine ring, which cannot effectively serve as an electron-transferring bridge. They have also showed that the band gaps of these COFs can be tuned by increasing the phenyl chains. Er et al.26 have explained the reason behind the excellent carrier mobility and photoconductivity along the vertical direction of an experimentally synthesized DA-COF. Very recently, Chakravarty et al.32 have explored the multifunctional application of an azine-linked COF in the field of nanoelectronics to nitroexplosive detection and conductance switching. The band gap tuning may be achieved through composite formation, which may lead to good photovoltaic activity of COFs. Recently, the 2D graphitic ZnO33,34 (g-ZnO)

1. INTRODUCTION Covalent organic frameworks (COFs) are an emerging class of two- or three-dimensional (2D/3D) crystalline porous materials.1−5 COFs are usually composed of lightweight elements,6 such as B, C, N, H, O, and Si, resulting in low mass density. The building subunits of these materials are linked by strong covalent bonds, providing high thermal stability. Because of large surface area and permanent porosity, COFs are emerged as potential candidates for storage of gases, such as hydrogen, carbon dioxide, methane, ammonia, and so forth.7−9 Researchers are eager about the interesting properties of COFs, such as mechanical properties, chemical sensing, photoelectricity, semiconductors, optoelectronics, and so forth.8,10−18 COFs are synthesized via reversible covalent-bond forming reactions such as condensations.1 Various synthetic methods are employed to develop COFs, namely, solvothermal synthesis,3 microwave synthesis,19 ionothermal synthesis,20 and so forth. Over the past few years, experimental studies mainly shed light on the rational design and controlled synthesis of COFs to explore its various applications in gas storage/separation, catalysis, chemical sensing, photoelectricity, energy storage, energy conversion, and so forth.21−24 Nowadays, many interesting electronic properties of COFs are revealed by several researchers computationally and theoretically.25,26 Yang and Pushpa27 have reported how the variations of the X4Y unit can tune the optical and electronic properties of (X4Y) (O2B−C6H4−BO2)3 COF (where X = C/ Si and Y = C, Si, Ge, Sn, and Pb). They have also demonstrated that the absorption region can be shifted from UV to visible by proper choice of X and Y. Zhu and Meunier28 have studied some single-layer COFs and have shown the tuning of band gap by incorporating carbon chains or phenyl rings into the side chains of HHTP-DPB-COF, COF-1, and CTF-1. As the number of carbon chains or the phenyl rings © 2019 American Chemical Society

Received: December 13, 2018 Accepted: January 25, 2019 Published: February 18, 2019 3556

DOI: 10.1021/acsomega.8b03488 ACS Omega 2019, 4, 3556−3564

ACS Omega

Article

images. The convergence criteria for total energy calculations in the self-consistent field iteration step were set to 1 × 10−6 eV, and the atomic positions in the structure were relaxed until the force on each atom becomes less than 0.01 eV/Å. To consider the weak vdW interaction in the case of the g-ZnO/ COF heterostructure, lithium adsorption, and hydrogen adsorption, the Grimmes DFT-D2 (PBE-D2) dispersion corrected method53 was applied. The Bader charge density analysis54 was performed to calculate the charges on the atoms. The optical properties were calculated using the frequencydependent complex dielectric function: ε(ω) = ε1(ω) + iε2(ω). The imaginary part of the dielectric function ε2(ω) was obtained with the help of first-order time-dependent perturbation theory. In the long wavelength limit, the imaginary part is given by,

monolayer has attracted the attention of researchers because of its exciting electronic properties. Wang et al.34 have proposed a MoS2/g-ZnO van der Waals (vdW) heterostructure, which forms a typical type-II band alignment that causes the effective separation of electron−hole pairs. Niu et al.33 have suggested that the g-ZnO-based vdW heterostructure, g-ZnO/blue phosphorous (BP), shows an effective charge carrier separation and large built-in electric field that ensures the photogenerated electrons to migrate easily from g-ZnO to BP. Previously, no COF composite has been studied, so, it would be interesting to study a way of band gap tuning of COF by means of composite formation. COFs possess high surface area and tunable pore size. Therefore, many researchers focus on their use as a hydrogen storage material. For mobile applications, a lightweight hydrogen storing material is desired and the adsorption energy should lie in between physisorption and chemisorption energy, where hydrogen is mainly adsorbed in molecular form.35−37 Mendoza-Cortes et al.38 have synthesized a COF, namely, COF-301-PdCl2 that shows superior hydrogen storage under ambient conditions. As suggested by earlier studies,37,39−45 simple vdW surfaces cannot hold hydrogen strongly and the hydrogen adsorption enthalpies can be significantly increased by the presence of charged sites. To search for better hydrogen storage materials, alkali and alkaline earth metal-decorated porous organic frameworks have attracted the attention of researchers. Gao et al.44 have shown that Ca-intercalated COF1 shows better hydrogen storage capacity at 300 K and 20 bar compared to bare COF-1, under same condition. Choi et al.46 have carried out density functional theory (DFT) calculations on some 3D COFs (COF-102, COF-103, COF-105, and COF-108) and found that the lithium- and magnesiumdecorated COFs are suitable candidates for hydrogen storage media. Very recently, Huang et al.47 have reported a thiophenebased covalent triazine framework (TBCTF) for metal-free, visible-light promoted selective oxidation of alcohols into corresponding aldehydes and ketones. In this article, we have investigated the detail electronic and optical properties of this COF to show its applicability in various fields. Constitutional isomers of TBCTF are used with the aim of band gap engineering. We have modeled bilayer COF and investigated how the electronic properties are affected under z-axial strain. A vdW heterostructure, g-ZnO/COF, is constructed to explore its photovoltaic activity. Finally, we have introduced lithium atoms on this COF and investigated its hydrogen adsorption ability.

ε2(ω) =

2e 2π ∑ |⟨ψ CB|u ⃗· r |⃗ ψKVB⟩|2 δ(EKCB − EKVB − ω) Ωε0 K ,CB,VB K (1)

where ω is the frequency of electromagnetic radiation. Ω denotes the volume of the unit cell, and ε0 is the free-space dielectric constant. CB and VB represent the conduction and valence bands, respectively. The polarization vector of the electric field of electromagnetic radiation and the position vector is represented by u⃗ and r⃗, respectively. The real part of the dielectric function, ε1(ω), is evaluated from the imaginary part with the help of Kramers−Kronig relation. After determining the real and imaginary parts of dielectric function, various optical properties, such as absorption coefficient [α(ω)] and reflectivity [R(ω)], can be obtained with the help of the following equations α(ω) =

2 ω[ ε12(ω) + ε2 2(ω) − ε1(ω)]1/2

ÅÄÅ ÑÉ2 ÅÅ ε1(ω) + iε2(ω) − 1 ÑÑÑ Å ÑÑ R(ω) = ÅÅÅ Ñ ÅÅ ε1(ω) + iε2(ω) + 1 ÑÑÑ ÅÇ ÑÖ

(2)

(3)

3. RESULTS AND DISCUSSION The optimized geometric structure of the 2D covalent triazine framework, TBCTF, is shown in Figure 1. The optimized lattice constant for TBCTF is 13.515 Å. As shown in Figure 1, the unit cell of TBCTF consists of two building blocks, namely, thiophene and triazine. We now focus on the electronic

2. COMPUTATIONAL DETAILS All of the electronic structure calculations were performed based on DFT by using Vienna ab initio simulation package.48−50 For all elemental constituents, projector augmented wave potentials have been taken into account. The plane-wave cutoff energy was set to 400 eV. The generalized gradient approximation (GGA) developed by Perdew, Burke, and Ernzerhof (PBE) has been utilized for treating the exchange correlation functional Exc[ρ].51 For structural optimizations and total energy calculations, the Brillouin zone sampling was carried out by using 2 × 2 × 1 Monkhorst−Pack (MP) k-point grids, whereas electronic structure calculations were performed with 7 × 7 × 1 MP grids.52 A large vacuum space of 20 Å was added along zdirection in order to avoid interactions between periodic

Figure 1. The optimized geometric structure of 2 × 2 × 1 supercell of a TBCTF. The unit cell is represented within the box. 3557

DOI: 10.1021/acsomega.8b03488 ACS Omega 2019, 4, 3556−3564

ACS Omega

Article

linking change as in the case of TBCTF1 is carried out and in addition to that the linking of middle thiophene moiety to one of the triazine ring is changed from C-2 to C-3. Now, we have studied how the electronic properties change with respect to change in the linking position of the building blocks. We have plotted the band structure and band decomposed charge density corresponding to the VBM and CBM states of both TBCTF1 and TBCTF2 in Figure 3. A closer look of Figure 3a,d indicates that the flat band characteristics of VBM and CBM states still remain in TBCTF1 as in the case of TBCTF, but in the case of TBCTF2, CBM becomes slightly dispersive. It may be noted that the curvature of Dirac-like cone at Kpoint above CBM, which is present in TBCTF, disappears as we go from TBCTF to TBCTF1 and TBCTF2. From the electronic band structure, we have found that both TBCTF1 and TBCTF2 are direct band gap semiconductors at Γ point with the band gap of 2.50 and 2.724 eV, respectively. The band gap increases as we go from TBCTF to TBCTF1 and TBCTF2. In TBCTF, the thiophene unit is attached with the triazine unit via C-2, which is more nucleophilic in character than C-3. However, in TBCTF1, the thiophene unit is attached with the triazine unit via C-3. Therefore, the electron-donating effect of thiophene is greater in the case of TBCTF than that in TBCTF1. In TBCTF2, the number of linkage of thiophene with triazine via less nucleophilic carbon center (C-3) increases. This may be the reason for band gap increase from TBCTF to TBCTF1 to TBCTF2. Therefore, the band gap of TBCTF can be effectively tuned by simply changing the position of sulfur atoms. Similar results were previously reported by Gutzler.56 As we obtain a tool for band gap engineering through constitutional isomerization, we turn our attention toward the optical properties of TBCTF, TBCTF1, and TBCTF2. We have considered both the parallel and perpendicular polarization for the calculation of optical properties. Our calculated optical band gap is in good agreement with the experimentally reported optical band gap of 2.47 eV.47 From the reflectivity spectra (Figure 4b,d), we have found that all of the COFs have very low reflectivity specially along the perpendicular polarization. Lower reflectivity along perpendicular direction may be due to large number of pores along that direction.27 The magnitude of reflectivity at zero energy, R(0), is much lower

structure of TBCTF. From the electronic band structure, plotted in Figure 2a, we have found that TBCTF is a direct

Figure 2. (a) The band structure plot of TBCTF. The Fermi level is indicated by the red dashed line. Charge density distribution of (b) CBM and (c) VBM states of TBCTF. The isovalue of 0.0017 e/Å3 is used.

band gap semiconductor with the band gap of 2.385 eV. A closer inspection of the band structure reveals that both the valence band maximum (VBM) and conduction band minimum (CBM) possess flat band characteristic. Similar results were previously observed by Er et al.26 and Wang et al.31 These flat band characteristic is an indication of localized wave function corresponding to the VBM and CBM states. To explain the flat band characteristic, we have plotted the band decomposed charge density of CBM and VBM states in Figure 2b,c, respectively. From Figure 2b,c, we notice that the charge density corresponding to CBM and VBM is highly localized and there is a node passing through the cross-conjugated position55 of TBCTF. Two constitutional isomers are made by changing the position of sulfur atoms within the thiophene moiety of TBCTF. We have modeled two systems, namely, TBCTF1 and TBCTF2, shown in the inset of Figure 3a,d. In TBCTF1, the linking of two terminal thiophene moieties with the triazine moiety is changed from C-2 to C-3. In TBCTF2, the same

Figure 3. (a) The band structure of TBCTF1; the inset figure indicates the unit cell of TBCTF1. The band decomposed charge density corresponding to (b) VBM and (c) CBM states of TBCTF1. (d) The band structure plot of TBCTF2; the inset figure represents the unit cell of TBCTF2. The band decomposed charge density corresponding to (e) VBM and (f) CBM states of TBCTF2. The isovalue of 0.0017 e/Å3 is used. 3558

DOI: 10.1021/acsomega.8b03488 ACS Omega 2019, 4, 3556−3564

ACS Omega

Article

Figure 4. The optical property of TBCTF, TBCTF1 and TBCTF2: (a) absorption coefficient [α(ω)] and (b) reflectivity [R(ω)] along parallel direction. (c) Absorption coefficient [α(ω)] and (d) reflectivity [R(ω)] along perpendicular direction.

Figure 5. (a) The top and side views of bilayer TBCTF in AB stacking mode. (b) The band structure of bilayer TBCTF. The Fermi level is indicated by the red dashed line. The variation of (c) energy and (d) band gap with respect to the z-axis strain.

than that of previously reported COFs.27 Such low reflectivity is very much useful for optoelectronic devices, such as solar cells and LEDs. Recently, Jiang et al.57 have reported that multilayer formation is an efficient way to promote the transport of charge carriers, which leads to the reduction of band gap. Inspired by this idea, we have explored the electronic properties of bilayer TBCTF. COF layers are stacked via weak vdW interaction and may take up various stacking patterns. We have considered two stacking patterns, namely, AA and AB stacking. We have designed AA stacking pattern by superimposing one COF layer over another by maintaining some distance and AB stacking is modeled from AA stacking by displacing the second layer with respect to the first layer by (1/2a1, 1/2a2) in the basal plane. In order to explore which stacking pattern is energetically more favorable, we have

calculated the binding energy, Eb, between two layers using the following equation E b = E bilayer TBCTF − 2Emonolayer TBCTF

(4)

where Ebilayer TBCTF and Emonolayer TBCTF are total energies of bilayer and monolayer TBCTF, respectively. A negative value of the binding energy implies that the formation of bilayer TBCTF is energetically favorable. The calculated equilibrium interlayer distances for AA stacking and AB stacking are 3.65 and 3.39 Å, respectively. The calculated binding energy for AB stacking is found to be 0.09 eV more negative than that of AA stacking. Therefore, AB stacking is energetically more favorable than AA stacking, and we now proceed to discuss electronic properties of bilayer TBCTF considering AB stacking (Figure 5a). From Figure 5b, we have found that bilayer TBCTF is also a direct band gap semiconductor at Γ point and the band gap of TBCTF decreases from 2.385 to 2.173 eV because of bilayer 3559

DOI: 10.1021/acsomega.8b03488 ACS Omega 2019, 4, 3556−3564

ACS Omega

Article

Figure 6. (a) Geometric structure of 4 × 4 × 1 supercell of g-ZnO monolayer. (b) The unit cell of TBCTF. The (c) top and (d) side views of gZnO/TBCTF composite (e) The projected band structure of g-ZnO/TBCTF composite. The magenta dashed line indicates the Fermi level. Bands dominated by g-ZnO is shown by red colour and green colour indicates bands mainly contributed by TBCTF. The band decomposed charge density of (f) CBM and (g) VBM states of the composite. The isovalue of 0.0017 e/Å3 is used.

Figure 7. The VBM (green) and CBM (red) energy levels of functionalized and nonfunctionalized composite systems (bold) and that of isolated monolayers. The VBM of g-ZnO is set to zero energy.

formation. We have noticed that the band structure of bilayer TBCTF looks like the superposition of band structure of two individual monolayers with the shift of Fermi energy. The reason behind the band gap decrease may be due to the interlayer coupling-induced charge transfer as pointed out by Jiang et al.57 For better understanding, we have plotted charge density difference of bilayer TBCTF (Figure S1), which shows that there is a redistribution of charge density between two adjacent layers. This redistribution of charge density induces charge transfer between TBCTF layers. The electronic band structure of AA-stacked bilayer TBCTF is also plotted (Figure S2), and the calculated band gap is 1.997 eV, which is slightly less than that of AB-stacked bilayer. Similar results were previously reported by Jiang et al.57 The vertical strain is proven to be an effective way for tuning the electronic properties of the vdW heterostructure through changing the interaction between adjacent layers.58−60 Inspired by this idea, we have applied strain along z-axis to modulate the electronic properties of bilayer TBCTF. The vertical strain along z-axis can be defined as ε = d0 − d, where d0 and d are the equilibrium and strained distances between two successive layers, respectively. As shown in Figure 5c, the total energy increases slowly with the increase of interlayer distance and the same increases steeply in the case of compressive strain. As the distance between two layers increases (decreases), band gap increases (decreases), as shown in Figure 5d. Next, we have investigated the electronic properties of the gZnO/TBCTF vdW heterostructure for band gap engineering and photovoltaic applications. A 4 × 4 × 1 supercell of the gZnO monolayer is shown in Figure 6a, which is similar to that of graphene. The optimized lattice constant of the g-ZnO

monolayer is calculated to be 3.33 Å, which is consistent with previous experimental and theoretical results.33,34 The heterostructure is constructed by stacking the unit cell of TBCTF over 4 × 4 × 1 supercell of the g-ZnO monolayer. The lattice mismatch58 for such heterostructure is only 1.5%. The top and side views of the optimized g-ZnO/TBCTF heterostructure are shown in Figure 6c,d, respectively. The calculated equilibrium interlayer distance between g-ZnO and TBCTF is 3.16 Å. The binding energy of the heterostructure is calculated with the help of the following equation E b = Eg ‐ ZnO/TBCTF − Eg ‐ ZnO − E TBCTF

(5)

where Eg‑ZnO/TBCTF, Eg‑ZnO, and ETBCTF are total energies of gZnO/TBCTF composite, g-ZnO, and TBCTF, respectively. The binding energy of the heterostructure is −6.11 eV, which suggests that the formation of g-ZnO/TBCTF is energetically favorable. The projected band structure of the heterostructure is shown in Figure 6e. The band gap decrease from 2.385 eV (in TBCTF) to 1.223 eV (in g-ZnO/TBCTF heterostructure) is due to the insertion of bands of g-ZnO into the band gap region of TBCTF. The interlayer coupling induces charge transfer (0.064e) from g-ZnO to TBCTF. We have found that the VBM is localized on g-ZnO, whereas the CBM is contributed by the states arising from TBCTF as shown in the band decomposed charge density (Figure 6f,g). This is also consistent with the projected band structure. Therefore, we can say that g-ZnO/TBCTF forms type-II band alignment, which is useful for effective separation of electron−hole pairs. Thus, g-ZnO/TBCTF may be a potential candidate for photovoltaic applications. The efficiency of a solar cell depends upon two factors: (i) the spatial separation of charge carriers, 3560

DOI: 10.1021/acsomega.8b03488 ACS Omega 2019, 4, 3556−3564

ACS Omega

Article

Figure 8. The absorption coefficient of TBCTF, g-ZnO, g-ZnO/TBCTF, cyano functionalized g-ZnO/TBCTF, nitro functionalized g-ZnO/ TBCTF along (a) parallel and (b) perpendicular directions.

have calculated the lithium binding energy using the following equation

which reduces the rate of recombination, and (ii) the rate of electron transfer from the CBM of the donor to the CBM of the acceptor. The electron transfer rate depends on the energy difference between CBM of the donor and acceptor. Larger the energy difference, larger is the rate of electron transfer and hence higher efficiencies. As explored in many previous studies,61−66 this energy gap can be tuned by changing the size/shape of quantum dots or by introducing functional groups. It is well known that the interior of the pore can be functionalized as revealed by many previous studies.67−69 With this motivation, we have investigated the effect of functionalization on TBCTF to engineer the energy gap between the CBM of g-ZnO and TBCTF. The middle thiophene moiety of the TBCTF unit cell is functionalized with (i) one cyano (−CN) group and (ii) one nitro (−NO2) group. In Figure 7, we have plotted the energy level of VBM and CBM of the composite systems along with those of isolated monolayers.64,70−72 From this figure, we have noticed that both (−CN and −NO2) functionalization increases the energy gap between CBM of g-ZnO and TBCTF. This energy gap is a measure of electron injection rate from the CBM of the donor to the CBM of the acceptor. This energy gap further increases with the increase of number of −CN/−NO2 groups (Figure S3). Thus, we can conclude that −CN/−NO2-functionalized g-ZnO/TBCTF composite systems show better photovoltaic activity than the nonfunctionalized one. Moreover, we have functionalized TBCTF with one methoxy group (−OCH3) and found that the composite system shows type-I band alignment, where both VBM and CBM are contributed by gZnO. However, when we functionalize TBCTF with one cyano (−CN) and one nitro (−NO2) group, the composite system shows type-II band alignment with better activity than that of the nonfunctionalized one. To have applicability in photovoltaic devices, the material should absorb as much as UV/visible light. Thus, we have investigated the optical properties of the g-ZnO/TBCTF composite. From Figure 8a,b, we can see that there occurs enhanced absorption in the case of g-ZnO/TBCTF composite and also in −CN/−NO2-functionalized g-ZnO/TBCTF composite compared to individual components along both parallel and perpendicular polarization. Thus, we strongly believe that the composite system, g-ZnO/TBCTF and the −CN/−NO2-functionalized g-ZnO/TBCTF composite may find potential applications in photovoltaic devices. Recent experimental and theoretical studies reveal that COFs are potential candidates for hydrogen storage media. Therefore, we have investigated the hydrogen adsorption ability of TBCTF through lithium atom doping. First of all, the most favorable binding site for lithium atoms on TBCTF is determined. Two different sites are possible, one is above the triazine ring and the other is above the thiophene ring. We

E b = E Li ‐ TBCTF − E TBCTF − nE Li

(6)

where ELi‑TBCTF, ETBCTF, and ELi are total energies of lithium decorated TBCTF, TBCTF, and lithium atom, respectively. n is the number of lithium atom. The calculated binding energy (per lithium atom) on triazine and thiophene moieties are −1.377 and −1.96 eV, respectively. Thus, we can say that the lithium atom preferably adsorb on above and below the thiophene ring. The binding energy of Li on the thiophene moiety is even more negative than the cohesive energy of metallic lithium (−1.63 eV).43 The optimized structure of LiTBCTF is shown in Figure 9a. The optimized lattice constant

Figure 9. The optimized unit cell of (a) lithium doped TBCTF (LiTBCTF) and (b−d) hydrogen molecule decorated Li-TBCTF [(H2)n−Li−TBCTF], where n = 6, 12, 18.

of Li-TBCTF is 13.62 Å, which is slightly greater than that of bare TBCTF. Thus, optimized cell constant undergoes slight expansion on Li adsorption. Lithium atoms adsorbed on the upper side of TBCTF are 2.62 Å away from the sulfur atom of thiophene, whereas the corresponding distance for the lower side is found to be 2.29 Å. Two different distances are found as sulfur atoms of thiophene moieties move slightly downward from the basal plane. The calculated distance between two Li atoms adsorb on the opposite side of a thiophene ring is 3.66 Å. The Bader charge analysis reveals that each Li atom on the upper surface carries 0.86e positive charge, whereas each on the lower surface carries 0.82e positive charge, suggesting a charge transfer from Li to TBCTF. Thus, one can expect that partial positive charge generated on Li atoms can polarize the molecular hydrogen and bind it through ion-quadrupole and ion-induced dipole interactions as revealed by many previous studies.37,39−44 Next, we turn our focus on the hydrogen adsorption ability of Li-functionalized TBCTF by introducing H2 molecules on the top of Li atoms in a stepwise manner. At first, one H2 molecule is placed on each Li atom. After geometry optimization, we have observed that Li atoms attract H2 molecules toward itself and also H2 molecules are slightly tilted toward TBCTF, indicating that H2 molecules also 3561

DOI: 10.1021/acsomega.8b03488 ACS Omega 2019, 4, 3556−3564

ACS Omega

Article

reveals that TBCTF is a direct band gap semiconductor, which may find application in semiconductor-based electronic devices. The band gap of TBCTF can be engineered through constitutional isomerism and bilayer TBCTF formation. We have proposed a new 2D g-ZnO/TBCTF vdW heterostructure. Our computational study suggests that the synthesis of the heterostructure is energetically feasible. The heterostructure is also a direct band gap semiconductor with type II band alignment, indicating the spatial separation of photogenerated electrons and holes. The VBM is localized on g-ZnO and the CBM is dominated by TBCTF in the heterostructure. This type of spatial separation of charge carriers is suitable for application in photovoltaic devices. Furthermore, the −CN/− NO2-functionalized g-ZnO/TBCTF shows better photovoltaic activity than the nonfunctionalized one. Optical property calculations show that the composite systems absorb more strongly in the visible region than the individual components. Therefore, our proposed composite systems g-ZnO/TBCTF and −CN/−NO2-functionalized g-ZnO/TBCTF may find its applicability as a light harvesting material. Moreover, we have systematically investigated the hydrogen adsorption ability of TBCTF through the doping of Li atoms. We have found that Li atoms adsorb preferably on the thiophene ring. Each Li atom can adsorb a maximum of three hydrogen molecules around it, leading to the theoretical hydrogen storage capacity of 7.40 wt %. The hydrogen adsorption energy indicates that Li-TBCTF can be used as a reversible hydrogen storage material under ambient conditions. Thus, we strongly believe that the g-ZnO/TBCTF (functionalized and nonfunctionalized) heterostructure may find potential applications in photovoltaic devices, and the Li-doped TBCTF can find its applicability as a hydrogen storage material.

interact with TBCTF. The optimized geometry of Li-TBCTF with one H2 molecule per Li is shown in Figure 9b. The shortest Li−H distance is found to be 1.90 Å. We have calculated the hydrogen adsorption energy of the Li-doped TBCTF by means of the following equation Ead = (E(H 2)n −Li ‐ TBCTF − E Li ‐ TBCTF − nE H2)/n

(7)

where E(H2)n−Li‑TBCTF, ELi‑TBCTF, and EH2 are total energies of LiTBCTF containing n number of hydrogen molecules, LiTBCTF, and hydrogen molecule, respectively. Our calculated results for hydrogen adsorption energies, shortest Li−H distances, and H−H bond lengths are listed in Table 1. Table 1. The Calculated Hydrogen Adsorption Energies per Hydrogen Molecule (Ead), Shortest Li−H Distances (dLi−H) and H−H Bond Lengths (dH−H) of Hydrogenated LiTBCTF Ead (eV/H2 molecule) systems

GGA + vdW

GGA

dLi−H (Å)

dH−H (Å)

(H2)6−Li-TBCTF (H2)12−Li-TBCTF (H2)18−Li-TBCTF

−0.35 −0.33 −0.26

−0.16 −0.15 −0.11

1.90 1.93 1.96

0.771 0.767 0.761

Using PBE-D2 (GGA−PBE) method, the adsorption energy per hydrogen molecule is calculated to be −0.35 eV (−0.16 eV). The H−H bond length is 0.771 Å, which is slightly longer than that found in pure H2 (bond length = 0.75 Å).37 Next, we have introduced two H2 molecules per Li atom, and the corresponding optimized geometry is shown in Figure 9c. In this case, the binding energy per hydrogen molecule is −0.33 eV (−0.15 eV), which is less than that of former one. The shortest Li−H distance and the H−H bond length are 1.93 and 0.767 Å, respectively. Furthermore, we have decorated Li-TBCTF with three hydrogen molecules per Li atom, and the corresponding optimized geometry is presented in Figure 9d. Here, hydrogen molecules are found to be adsorbed with the binding energy of −0.26 eV (−0.11 eV). The H−H bond length gets slight expansion. Hydrogen adsorption interaction by Li-TBCTF can be more clearly explored by non-covalent isosurfaces,73,74 where the blue region indicates strong attractive attractions (Figure S4). However, the decrease of binding energy with the increase of H2 molecule concentration indicates that only a limited number of H2 molecules can be adsorbed.43 We have also decorated Li-TBCTF with four hydrogen molecules per Li and found that H2 molecules get far away from Li centers. The repulsion between the adsorbed hydrogen molecules may be the reason for this.37 Therefore, each Li atom in Li-TBCTF can hold maximum three hydrogen molecules, which leads to the maximum storage capacity37 of 7.40 wt %. For reversible hydrogen storage under ambient conditions, the hydrogen binding energy should lie in the range of −0.2 to −0.6 eV as suggested by many previous studies.39,42,75 Our computed binding energies also lie in this range, suggesting that Lidecorated TBCTF may find application as hydrogen storage devices.



ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsomega.8b03488.



Charge density difference plot of AB-stacked bilayer TBCTF, band structure of bilayer TBCTF in the AA stacking mode, position of VBM and CBM energy levels of difunctionalized and nonfunctionalized g-ZnO/ TBCTF composite systems, and noncovalent isosurface plot of hydrogen molecule-decorated Li-TBCTF (PDF)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Pranab Sarkar: 0000-0003-0109-6748 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS Financial support from DST NanoMission, Government of India, New Delhi, through Research Grant SR/NM/NS-1005/ 2016 is gratefully acknowledged. B.B. is grateful to UGC, New Delhi for the award of Junior research Fellowship (JRF) [UGC-Ref. no. 245/(CSIR-UGC NET DEC. 2016). C.C. is grateful to CSIR, New Delhi, for the Award of Senior Research

4. CONCLUSIONS In summary, we have investigated the electronic and optical properties of recently synthesized TBCTF COF based on firstprinciples calculations. Our electronic structure calculation 3562

DOI: 10.1021/acsomega.8b03488 ACS Omega 2019, 4, 3556−3564

ACS Omega

Article

(19) Campbell, N. L.; Clowes, R.; Ritchie, L. K.; Cooper, A. I. Rapid microwave synthesis and purification of porous covalent organic frameworks. Chem. Mater. 2009, 21, 204−206. (20) Kuhn, P.; Antonietti, M.; Thomas, A. Porous, covalent triazinebased frameworks prepared by ionothermal synthesis. Angew. Chem., Int. Ed. 2008, 47, 3450−3453. (21) Lin, S.; Diercks, C. S.; Zhang, Y.-B.; Kornienko, N.; Nichols, E. M.; Zhao, Y.; Paris, A. R.; Kim, D.; Yang, P.; Yaghi, O. M.; Chang, C. J. Covalent organic frameworks comprising cobalt porphyrins for catalytic CO2 reduction in water. Science 2015, 349, 1208−1213. (22) Xu, H.; Gao, J.; Jiang, D. Stable, crystalline, porous, covalent organic frameworks as a platform for chiral organocatalysts. Nat. Chem. 2015, 7, 905. (23) Stegbauer, L.; Schwinghammer, K.; Lotsch, B. V. A hydrazonebased covalent organic framework for photocatalytic hydrogen production. Chem. Sci. 2014, 5, 2789−2793. (24) Aiyappa, H. B.; Thote, J.; Shinde, D. B.; Banerjee, R.; Kurungot, S. Cobalt-modified covalent organic framework as a robust water oxidation electrocatalyst. Chem. Mater. 2016, 28, 4375−4379. (25) Zhou, Y.; Wang, Z.; Yang, P.; Zu, X.; Gao, F. Electronic and optical properties of two-dimensional covalent organic frameworks. J. Mater. Chem. 2012, 22, 16964−16970. (26) Er, D.; Dong, L.; Shenoy, V. B. Mechanisms for engineering highly anisotropic conductivity in a layered covalent-organic framework. J. Phys. Chem. C 2015, 120, 174−178. (27) Yang, L.-M.; Pushpa, R. Tuning electronic and optical properties of a new class of covalent organic frameworks. J. Mater. Chem. C 2014, 2, 2404−2416. (28) Zhu, P.; Meunier, V. Electronic properties of two-dimensional covalent organic frameworks. J. Chem. Phys. 2012, 137, 244703. (29) Yang, L.-M.; Dornfeld, M.; Hui, P.-M.; Frauenheim, T.; Ganz, E. Ten new predicted covalent organic frameworks with strong optical response in the visible and near infrared. J. Chem. Phys. 2015, 142, 244706. (30) Liang, L.; Zhu, P.; Meunier, V. Electronic, structural, and substrate effect properties of single-layer covalent organic frameworks. J. Chem. Phys. 2015, 142, 184708. (31) Wang, R.-N.; Zhang, X.-R.; Wang, S.-F.; Fu, G.-S.; Wang, J.-L. Flatbands in 2D boroxine-linked covalent organic frameworks. Phys. Chem. Chem. Phys. 2016, 18, 1258−1264. (32) Chakravarty, C.; Mandal, B.; Sarkar, P. Multifunctionalities of an azine-linked covalent organic framework: from nanoelectronics to nitroexplosive detection and conductance switching. J. Phys. Chem. C 2018, 122, 3245−3255. (33) Niu, X.; Li, Y.; Shu, H.; Yao, X.; Wang, J. Efficient carrier separation in graphitic zinc oxide and blue phosphorus van der Waals heterostructure. J. Phys. Chem. C 2017, 121, 3648−3653. (34) Wang, S.; Ren, C.; Tian, H.; Yu, J.; Sun, M. MoS2/ZnO van der Waals heterostructure as a high-efficiency water splitting photocatalyst: a first-principles study. Phys. Chem. Chem. Phys. 2018, 20, 13394−13399. (35) Lochan, R. C.; Head-Gordon, M. Computational studies of molecular hydrogen binding affinities: the role of dispersion forces, electrostatics, and orbital interactions. Phys. Chem. Chem. Phys. 2006, 8, 1357−1370. (36) Bhatia, S. K.; Myers, A. L. Optimum conditions for adsorptive storage. Langmuir 2006, 22, 1688−1700. (37) Srinivasu, K.; Ghosh, S. K. Hydrogen adsorption in lithium decorated conjugated microporous polymers: a DFT investigation. RSC Adv. 2014, 4, 4170−4176. (38) Mendoza-Cortes, J. L.; Goddard, W. A., III; Furukawa, H.; Yaghi, O. M. A Covalent Organic Framework that Exceeds the DOE 2015 Volumetric Target for H2 Uptake at 298 K. J. Phys. Chem. Lett. 2012, 3, 2671−2675. (39) Lee, H.; Ihm, J.; Cohen, M. L.; Louie, S. G. Calcium-decorated graphene-based nanostructures for hydrogen storage. Nano Lett. 2010, 10, 793−798. (40) Hussain, T.; Hankel, M.; Searles, D. J. Graphenylene monolayers doped with alkali or alkaline earth metals: promising

Fellowship (SRF) [CSIR Award no. 09/202(0051)/2015EMR-I].



REFERENCES

(1) Cote, A. P.; Benin, A. I.; Ockwig, N. W.; O’keeffe, M.; Matzger, A. J.; Yaghi, O. M. Porous, crystalline, covalent organic frameworks. Science 2005, 310, 1166−1170. (2) Spitler, E. L.; Dichtel, W. R. Lewis acid-catalysed formation of two-dimensional phthalocyanine covalent organic frameworks. Nat. Chem. 2010, 2, 672. (3) Uribe-Romo, F. J.; Hunt, J. R.; Furukawa, H.; Klöck, C.; O’Keeffe, M.; Yaghi, O. M. A crystalline imine-linked 3-D porous covalent organic framework. J. Am. Chem. Soc. 2009, 131, 4570−4571. (4) El-Kaderi, H. M.; Hunt, J. R.; Mendoza-Cortes, J. L.; Cote, A. P.; Taylor, R. E.; O’Keeffe, M.; Yaghi, O. M. Designed synthesis of 3D covalent organic frameworks. Science 2007, 316, 268−272. (5) Lin, G.; Ding, H.; Yuan, D.; Wang, B.; Wang, C. A pyrene-based, fluorescent three-dimensional covalent organic framework. J. Am. Chem. Soc. 2016, 138, 3302−3305. (6) Feng, X.; Ding, X.; Jiang, D. Covalent organic frameworks. Chem. Soc. Rev. 2012, 41, 6010−6022. (7) Han, S. S.; Furukawa, H.; Yaghi, O. M.; Goddard, W. A., III Covalent organic frameworks as exceptional hydrogen storage materials. J. Am. Chem. Soc. 2008, 130, 11580−11581. (8) Hug, S.; Stegbauer, L.; Oh, H.; Hirscher, M.; Lotsch, B. V. Nitrogen-rich covalent triazine frameworks as high-performance platforms for selective carbon capture and storage. Chem. Mater. 2015, 27, 8001−8010. (9) Doonan, C. J.; Tranchemontagne, D. J.; Glover, T. G.; Hunt, J. R.; Yaghi, O. M. Exceptional ammonia uptake by a covalent organic framework. Nat. Chem. 2010, 2, 235. (10) Ding, S.-Y.; Dong, M.; Wang, Y.-W.; Chen, Y.-T.; Wang, H.-Z.; Su, C.-Y.; Wang, W. Thioether-Based Fluorescent Covalent Organic Framework for Selective Detection and Facile Removal of Mercury(II). J. Am. Chem. Soc. 2016, 138, 3031−3037. (11) Li, Z.; Zhang, Y.; Xia, H.; Mu, Y.; Liu, X. A robust and luminescent covalent organic framework as a highly sensitive and selective sensor for the detection of Cu2+ ions. Chem. Commun. 2016, 52, 6613−6616. (12) Dalapati, S.; Jin, S.; Gao, J.; Xu, Y.; Nagai, A.; Jiang, D. An azine-linked covalent organic framework. J. Am. Chem. Soc. 2013, 135, 17310−17313. (13) Das, G.; Biswal, B. P.; Kandambeth, S.; Venkatesh, V.; Kaur, G.; Addicoat, M.; Heine, T.; Verma, S.; Banerjee, R. Chemical sensing in two dimensional porous covalent organic nanosheets. Chem. Sci. 2015, 6, 3931−3939. (14) Chen, L.; Furukawa, K.; Gao, J.; Nagai, A.; Nakamura, T.; Dong, Y.; Jiang, D. Photoelectric covalent organic frameworks: converting open lattices into ordered donor−acceptor heterojunctions. J. Am. Chem. Soc. 2014, 136, 9806−9809. (15) Wan, S.; Guo, J.; Kim, J.; Ihee, H.; Jiang, D. A photoconductive covalent organic framework: self-condensed arene cubes composed of eclipsed 2D polypyrene sheets for photocurrent generation. Angew. Chem. 2009, 121, 5547−5550. (16) Dogru, M.; Handloser, M.; Auras, F.; Kunz, T.; Medina, D.; Hartschuh, A.; Knochel, P.; Bein, T. A Photoconductive thienothiophene-based covalent organic framework showing charge transfer towards included fullerene. Angew. Chem. 2013, 125, 2992−2996. (17) Ding, X.; Guo, J.; Feng, X.; Honsho, Y.; Guo, J.; Seki, S.; Maitarad, P.; Saeki, A.; Nagase, S.; Jiang, D. Synthesis of metallophthalocyanine covalent organic frameworks that exhibit high carrier mobility and photoconductivity. Angew. Chem., Int. Ed. 2010, 50, 1289−1293. (18) Medina, D. D.; Petrus, M. L.; Jumabekov, A. N.; Margraf, J. T.; Weinberger, S.; Rotter, J. M.; Clark, T.; Bein, T. Directional chargecarrier transport in oriented benzodithiophene covalent organic framework thin films. ACS Nano 2017, 11, 2706−2713. 3563

DOI: 10.1021/acsomega.8b03488 ACS Omega 2019, 4, 3556−3564

ACS Omega

Article

materials for clean energy storage. J. Phys. Chem. C 2017, 121, 14393−14400. (41) Srinivasu, K.; Ghosh, S. K. Graphyne and graphdiyne: promising materials for nanoelectronics and energy storage applications. J. Phys. Chem. C 2012, 116, 5951−5956. (42) Wu, M.; Wang, Q.; Sun, Q.; Jena, P. Functionalized graphitic carbon nitride for efficient energy storage. J. Phys. Chem. C 2013, 117, 6055−6059. (43) Hussain, T.; Hankel, M.; Searles, D. J. Computational Evaluation of Lithium-Functionalized Carbon Nitride (g-C6N8) Monolayer as an Efficient Hydrogen Storage Material. J. Phys. Chem. C 2016, 120, 25180−25188. (44) Gao, F.; Ding, Z.; Meng, S. Three-dimensional metalintercalated covalent organic frameworks for near-ambient energy storage. Sci. Rep. 2013, 3, 1882. (45) Srinivasu, K.; Chandrakumar, K. R. S.; Ghosh, S. K. Quantum chemical studies on hydrogen adsorption in carbon-based model systems: role of charged surface and the electronic induction effect. Phys. Chem. Chem. Phys. 2008, 10, 5832−5839. (46) Choi, Y. J.; Lee, J. W.; Choi, J. H.; Kang, J. K. Ideal metaldecorated three dimensional covalent organic frameworks for reversible hydrogen storage. Appl. Phys. Lett. 2008, 92, 173102. (47) Huang, W.; Ma, B. C.; Lu, H.; Li, R.; Wang, L.; Landfester, K.; Zhang, K. A. I. Visible-light-promoted selective oxidation of alcohols using a covalent triazine framework. ACS Catal. 2017, 7, 5438−5442. (48) Kresse, G.; Joubert, D. From ultrasoft pseudopotentials to the projector augmented-wave method. Phys. Rev. B: Condens. Matter Mater. Phys. 1999, 59, 1758. (49) Blöchl, P. E. Projector augmented-wave method. Phys. Rev. B: Condens. Matter Mater. Phys. 1994, 50, 17953. (50) Kresse, G.; Furthmüller, J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane-wave basis set. Comput. Mater. Sci. 1996, 6, 15−50. (51) Perdew, J. P.; Burke, K.; Ernzerhof, M. Generalized gradient approximation made simple. Phys. Rev. Lett. 1996, 77, 3865. (52) Monkhorst, H. J.; Pack, J. D. Special points for brillouin-zone integrations. Phys. Rev. B: Solid State 1976, 13, 5188. (53) Grimme, S. Semiempirical GGA-type density functional constructed with a long-range dispersion correction. J. Comput. Chem. 2006, 27, 1787−1799. (54) Bader, R. F. Atoms in Molecules: A Quantum Theory; International Series of Monographs on Chemistry; Clarendon Press, 1990; Vol. 22. (55) Gutzler, R.; Perepichka, D. F. π-Electron Conjugation in Two Dimensions. J. Am. Chem. Soc. 2013, 135, 16585−16594. (56) Gutzler, R. Band-structure engineering in conjugated 2D polymers. Phys. Chem. Chem. Phys. 2016, 18, 29092−29100. (57) Jiang, X.; Wang, P.; Zhao, J. 2D covalent triazine framework: a new class of organic photocatalyst for water splitting. J. Mater. Chem. A 2015, 3, 7750−7758. (58) Guan, Z.; Lian, C.-S.; Hu, S.; Ni, S.; Li, J.; Duan, W. Tunable Structural, Electronic, and Optical Properties of Layered TwoDimensional C2N and MoS2 van der Waals Heterostructure as Photovoltaic Material. J. Phys. Chem. C 2017, 121, 3654−3660. (59) Li, X.; Jia, G.; Du, J.; Song, X.; Xia, C.; Wei, Z.; Li, J. Type-II InSe/MoSe2(WSe2) van der Waals heterostructures: vertical strain and electric field effects. J. Mater. Chem. C 2018, 6, 10010−10019. (60) Wang, D.; Han, D.; Liu, L.; Niu, L. Structure and electronic properties of C2N/graphene predicted by first-principles calculations. RSC Adv. 2016, 6, 28484−28488. (61) Chakravarty, C.; Mandal, B.; Sarkar, P. Porous GrapheneFullerene Nanocomposites: A New Composite for Solar Cell and Optoelectronic Applications. J. Phys. Chem. C 2018, 122, 15835− 15842. (62) Mandal, B.; Sarkar, S.; Sarkar, P. Theoretical studies on understanding the feasibility of porphyrin-sensitized graphene quantum dot solar cell. J. Phys. Chem. C 2015, 119, 3400−3407. (63) Chakravarty, C.; Ghosh, P.; Mandal, B.; Sarkar, P. Understanding the Electronic Structure of Graphene Quantum Dot-

Fullerene Nanohybrids for Photovoltaic Applications. Z. Phys. Chem. 2016, 230, 777−790. (64) Sarkar, S.; Rajbanshi, B.; Sarkar, P. Understanding the electronic structure of CdSe quantum dot-fullerene (C60) hybrid nanostructure for photovoltaic applications. J. Appl. Phys. 2014, 116, 114303. (65) Rajbanshi, B.; Sarkar, P. Optimizing the Photovoltaic Properties of CdTe Quantum Dot-Porphyrin Nanocomposites: A Theoretical Study. J. Phys. Chem. C 2016, 120, 17878−17886. (66) Pramanik, A.; Sarkar, S.; Pal, S.; Sarkar, P. Pentacene-fullerene bulk-heterojunction solar cell: A computational study. Phys. Lett. A 2015, 379, 1036−1042. (67) Huang, N.; Krishna, R.; Jiang, D. Tailor-made pore surface engineering in covalent organic frameworks: systematic functionalization for performance screening. J. Am. Chem. Soc. 2015, 137, 7079− 7082. (68) Nagai, A.; Guo, Z.; Feng, X.; Jin, S.; Chen, X.; Ding, X.; Jiang, D. Pore surface engineering in covalent organic frameworks. Nat. Commun. 2011, 2, 536. (69) Lohse, M. S.; Stassin, T.; Naudin, G.; Wuttke, S.; Ameloot, R.; De Vos, D.; Medina, D. D.; Bein, T. Sequential pore wall modification in a covalent organic framework for application in lactic acid adsorption. Chem. Mater. 2016, 28, 626−631. (70) Sarkar, S.; Pal, S.; Sarkar, P. Electronic structure and band gap engineering of CdTe semiconductor nanowires. J. Mater. Chem. 2012, 22, 10716−10724. (71) Saha, S.; Sarkar, S.; Pal, S.; Sarkar, P. Tuning the energy levels of ZnO/ZnS core/shell nanowires to design an efficient nanowirebased dye-sensitized solar cell. J. Phys. Chem. C 2013, 117, 15890− 15900. (72) Rajbanshi, B.; Sarkar, S.; Sarkar, P. Band gap engineering of graphene-CdTe quantum dot hybrid nanostructures. J. Mater. Chem. C 2014, 2, 8967−8975. (73) Johnson, E. R.; Keinan, S.; Mori-Sánchez, P.; Contreras-García, J.; Cohen, A. J.; Yang, W. Revealing noncovalent interactions. J. Am. Chem. Soc. 2010, 132, 6498−6506. (74) Otero-de-la-Roza, A.; Johnson, E. R.; Contreras-García, J. Revealing non-covalent interactions in solids: NCI plots revisited. Phys. Chem. Chem. Phys. 2012, 14, 12165−12172. (75) Kim, Y.-H.; Zhao, Y.; Williamson, A.; Heben, M. J.; Zhang, S. Nondissociative adsorption of H2molecules in light-element-doped fullerenes. Phys. Rev. Lett. 2006, 96, 016102.

3564

DOI: 10.1021/acsomega.8b03488 ACS Omega 2019, 4, 3556−3564