Antimonene Oxides: Emerging Tunable Direct ... - ACS Publications

Letter pubs.acs.org/NanoLett

Antimonene Oxides: Emerging Tunable Direct Bandgap Semiconductor and Novel Topological Insulator Shengli Zhang,† Wenhan Zhou,† Yandong Ma,‡ Jianping Ji,† Bo Cai,† Shengyuan A. Yang,§ Zhen Zhu,∥ Zhongfang Chen,⊥ and Haibo Zeng*,† †

Key Laboratory of Advanced Display Materials and Devices, Ministry of Industry and Information Technology, College of Material Science and Engineering, Nanjing University of Science and Technology, Nanjing 210094, China ‡ Wilhelm-Ostwald-Institut für Physikalische und Theoretische Chemie, Universität Leipzig, Linnéstrasse 2, 04103 Leipzig, Germany § Research Laboratory for Quantum Materials, Singapore University of Technology and Design, Singapore 487372, Singapore ∥ Materials Department, University of California, Santa Barbara, California 93106, United States ⊥ Department of Chemistry, Institute for Functional Nanomaterials, University of Puerto Rico, Rio Piedras, San Juan PR 00931 S Supporting Information *

ABSTRACT: Highly stable antimonene, as the cousin of phosphorene from group-VA, has opened up exciting realms in the two-dimensional (2D) materials family. However, pristine antimonene is an indirect band gap semiconductor, which greatly restricts its applications for optoelectronics devices. Identifying suitable materials, both responsive to incident photons and efficient for carrier transfer, is urgently needed for ultrathin devices. Herein, by means of firstprinciples computations we found that it is rather feasible to realize a new class of 2D materials with a direct bandgap and high carrier mobility, namely antimonene oxides with different content of oxygen. Moreover, these tunable direct bandgaps cover a wide range from 0 to 2.28 eV, which are crucial for solar cell and photodetector applications. Especially, the antimonene oxide (18Sb− 18O) is a 2D topological insulator with a sizable global bandgap of 177 meV, which has a nontrivial Z2 topological invariant in the bulk and the topological states on the edge. Our findings not only introduce new vitality into 2D group-VA materials family and enrich available candidate materials in this field but also highlight the potential of these 2D semiconductors as appealing ultrathin materials for future flexible electronics and optoelectronics devices. KEYWORDS: Antimonene oxide, 2D Semiconductor, tunable direct bandgap, carrier mobility, topological insulator monolayer), and a high hole mobility above 10 000 cm2 V−1 S−1.10−13 In 2014, Li et al. successfully fabricated the first fieldeffect transistor (FETs) based on micrometer-sized flakes of few-layer black phosphorus.10 However, one issue that impedes the practical applications of phosphorene is the difficulty in sample synthesis. The 2D phosphorene can be exfoliated from bulk black phosphorus via Scotch tape method or liquid exfoliation method, while 3D black P is obtained from red P under high pressure (10 kbar), high temperature (1000 °C) conditions. The direct synthesis of atomically thin phosphorene is still a great challenge. The other more serious problem is that the exfoliated monolayer and few-layer phosphorene degrade very fast upon air exposure, limiting their practical applications.14−16 In the last two years, as the cousins of phosphorene, the semiconducting group-VA nanosheets (arsenene, antimonene, and bismuthene) have triggered intense research interests due

D

eveloping the next generation optoelectronic devices strongly demands for advanced semiconductors with high sensitivity to incident photons and excellent transport property to approach.1−3 Two-dimensional (2D) materials are promising candidates owing to their unique properties such as ultrathin nature, transparent, flexibility, strong interactions with lights and high carrier mobility.4,5 Meanwhile, the existing quantum confinement along the vertical direction would generate a number of fascinating electronic and optical properties. Recently, a family of 2D crystals, derived from the group-VA layered materials (P, As, Sb, Bi), has emerged with increasing research interests owing to their significant fundamental band gaps. Importantly, distinct from zero band gap group-IVA (graphene, silicene, germanene, and stanene),6 group-IIIA (borophene), and monoelement hafnene nanosheets,7−9 the semiconducting behavior of group-VA 2D materials can render them as potential candidates for next-generation electronics and optoelectronics devices Among these family members, the first candidate under extensive exploration is phosphorene, which exhibits a direct and tunable bandgap (from 0.3 eV in its bulk to 2.0 eV in a © 2017 American Chemical Society

Received: January 21, 2017 Revised: April 20, 2017 Published: May 1, 2017 3434

DOI: 10.1021/acs.nanolett.7b00297 Nano Lett. 2017, 17, 3434−3440

Letter

Nano Letters to their versatile properties.17−49 An intriguing example is antimonene predicted by Zhang et al. in 2015.17 Available theoretical studies revealed many peculiar properties, such as thermoelectric response, strain-modified inversion of conduction bands, and perpendicular electric field-induced 2D topological character.17−35 In contrast to puckered armchairlike phosphorene, the most stable antimonene holds buckled honeycomb structure with much stronger spin−orbital coupling (SOC), which brings exotic fundamental properties for photonics and spintronics. Excitingly, the theoretically predicted monolayer and few-layer antimonene were just prepared experimentally by mechanical exfoliation, liquid exfoliation, plasma-assisted process, and vapor deposition techniques and the fine microstructure of the buckled antimonene has been well characterized (Figure 1a).20−25

Interestingly, all these antimonene oxides hold tunable direct bandgaps, which cover a wide range from 0 to 2.28 eV. Simultaneously, the effective electron masses of all antimonene oxides are close to that of monolayer and few-layer black phosphorene (meΓX = 0.15−0.18mo), nearly four times smaller than that of monolayer MoS2 (me = 0.60mo). This suggests that all antimonene oxide systems, similar to 2D black phosphorus, may hold higher electron mobility than monolayer MoS2. Excitingly, antimonene oxide (18Sb−18O) is a 2D topological insulator with a sizable global band gap of 177 me, and characterized by the nontrivial Z2 topological invariant and the topological edge states. Thus, our predictions not only introduce new vitality into 2D group-VA materials family, enriching available candidate materials in this field but also highlight the potential of these 2D semiconductors as appealing ultrathin materials for future flexibility electronics and optoelectronics devices. The outermost shell of Sb atom has five valence electrons with 5s 2 5p 3 configuration. In the pristine honeycomb antimonene, each Sb atom forms sp3 hybridization to produce three bonds with adjacent Sb atoms, leaving the nonbonding lone pair electrons. When connected with oxygen atoms, the lone pair electrons of Sb can be donated to the oxygen atom forming a dative bond, in terms of the Lewis structure. Consequently, both Sb and O atoms fulfill the octet rules (Figure 1) and stable antimonene oxides can be expected. Following the above rationale, we are proposing the family of antimonene oxides with different content of oxygen by means of first-principles computations. We adopt one supercell with 18 Sb atoms. All these antimonene oxides are fully relaxed. Clearly, the dangling oxygen motifs, shown in Figure 1d, are composed of one O atom bonding with only one Sb atom. All of the SbO bonds are perpendicular to the 2D antimonene plane in order to minimize Coulomb repulsion. Significantly, the Sb lone-pair is captured by the more electronegative O atom, giving rise to an excess of negative charge localized on the O atom (Figure S1). Note that besides the different oxygen concentrations, there are also many different configurations for any specific oxygen concentration. To survey such a complicated situation, we consider representative oxygen concentrations, and examined models with different O arrangements. Our computations show that for each O concentration their energies are very close and their band gaps are nearly same (Figure S2). Thus, in the following discussions, we only present one typical configuration for each oxygen concentration. All the optimized structural parameters of antimonene oxides are shown in Table S1. Clearly, lattice constant a increases with the increase of oxygen concentration. Interestingly, the average SbO bond length does not change significantly at different oxygen concentrations. Almost all the SbO bond lengths in antimonene oxides are 1.70 Å, except that the SbO bond lengths are slightly shorter (1.68 Å) in fully oxidized antimonene. Meanwhile, the standard SbO bond lengths in the conventional binary oxides of antimony (Sb2O3, Sb2O4, and Sb2O5),50 approximately 2.00 Å, are calculated in Figure S3 and Table S2. So, one could see that the SbO bond in 2D antimonene oxide is strong, polar, and short compared with the conventional binary oxides of antimony. For the Sb−Sb bond lengths in antimonene oxide, they increase slightly from 2.76 to 2.93 Å with the increase of oxygen concentration, and the oxidized antimonene still keeps the significantly buckled structure (Figure S1). This phenomenon is mainly attributed

Figure 1. Two-dimensional pristine antimonene with its band structure, and antimonene oxides (a) TEM image of a successfully synthesized ultrathin antimonene. (b) Electronic band structures of pristine honeycomb antimonene with an indirect band gap calculated at the PBE and HSE06 level. (c) Schematics of the hybridization process between antimonene and oxygen. (d) A sketch map of antimonene oxide.

With considerable band gap as predicted and experimentally proven high stability, 2D antimonene is expected to be a rather promising and competitive candidate for electronic and optoelectronic applications. However, pristine antimonene monolayer is an indirect band gap semiconductor, which significantly restricts its applications in optoelectronics such as light-emitting diode (LED) or photovoltaic devices, because extra phonon momenta are required to assist the transition and hence result in a relatively lower efficiency of energy conversion compared with the direct band gap semiconductors.17 On the other hand, the carrier mobility of antimonene nanosheet is lower than that of phosphorene.18 Thus, it is highly desirable to modulate the properties of 2D antimonene to achieve appropriate band gap and higher carrier mobility. Herein, by means of first-principles computations, we demonstrate the feasibility to realize a new class of 2D materials, antimonene oxides with different content of oxygen. 3435


Letter

Nano Letters

Figure 2. Electronic structure evolution of antimonene oxides with oxygen ratio. (a) The electronic band structure of antimonene oxides with the different oxygen concentration. An emerging class of direct bandgap antimonene oxides with the VBM and CBM both located at the Γ highsymmetry point. (b) The PDOS of antimonene oxides. (c) Isosurfaces of electron localization function of antimonene oxides at the level 0.85, and the red-colored regions indicate accumulation of electrons.

To understand the origin of these direct bandgaps in antimonene oxides, we further calculated partial densities of states (PDOS) of antimonene oxides as well as compared with these of pristine antimonene (Figures 2b and S4). Two representatives are given in Figure 2b. The PDOS spectra of pristine amtimonene suggest that the 5p states of Sb dominate the electronic states near Fermi level coupled with small amount of 5s states. The sp3 hybridization have been observed in pristine antimonene, and thus all Sb atoms are the equivalent coordination atoms. For antimonene oxides, the calculated PDOS analysis shows that their VBM and CBM are contributed by the 5p states of Sb, whereas with increasing oxygen concentration O 2p states also participate in the VBM and CBM states, due to the formation of the dative bond between O and Sb. The donation of the Sb lone pair electrons to the antibonding SbO bonds. The electron localization function has been taken into account to study the disposition of valence electrons or electron pairs in antimonene and their oxides (Figure 2c). The nature of the isosurfaces for the lone pairs and the SbO bond attractor is indicative of an ionic type of environment around oxygen and supportive of the backbonding stabilization of the SbO bond. Especially, this local quantum entrapment and densification of the core and bonding electrons play an important role, which narrows the bandgap and increases the polarization of the monolayer with increasing the number of SbO bonds. Thus, by oxidation we can artificially control the yield of chemical modification to obtain the desired semiconducting bandgap for potential applications in the high efficient light emitter and photovoltaic devices. The 2D semiconductors are beneficial for electronic and optoelectronic applications partially due to their observed high

to a significant difference in Pauling electronegativity of Sb (2.05) and O (3.44). How do such oxidations affect the electronic properties of antimonene? To address this issue, we computed the electronic band structures of 2D antimonene oxides as well as their density of states (DOS) (Figure 2). The pristine antimonene is an indirect band gap semiconductor (band gap of 1.75 eV at the PBE level, see Figure 1b; 2.28 eV at the HSE06 level) and its valence band maximum (VBM) locates at Γ high-symmetry point, whereas the conduction band minimum (CBM) is between M and Γ Brillouin zone. Interestingly, surface oxidation results in an emerging class of direct band gap antimonene oxides with the VBM and CBM both located at the Γ high-symmetry point (Figure 2a). Note that it is the introduction of oxygen atoms that promotes the adjustment of energy level, especially for the CBM level. Here, the combination of HSE06 and SOC effect has been taken into account in the electronic band structures of antimonene oxides. Our computing results show that the differences of band gaps between PBE and HSE levels are about 0.5 eV for antimonene and its partial oxides (18Sb−nO, n < 18). For 18Sb−18O, we found that it remains similar electronic band structure using the PBE and HSE06 methods. Besides, spin−orbital interaction plays an important role on the electronic properties of antimonene.51 For 18Sb−18O, when considering SOC, it opened a tiny band gap about 0.2 eV, compared with its semimetal properties without SOC. Thus, through the above analysis, these tunable direct band gap 2D antimonene oxide semiconductors cover a wide range of gaps from 0 to 2.28 eV, which are beneficial for broadband photoresponse. 3436


Letter

Nano Letters mobility of carriers.52−56 Antimonene is semiconducting with an indirect band gap. At the bandgap wavevector, which is Γ high-symmetry point, the valence band and conduction band disperse rather strongly, indicating very small effective masses. Then, what will be the carrier mobilities of the antimonene oxides? Thus, we estimated the carrier effective masses of antimonene oxides in comparison with other typical 2D semiconductors, such as MoS2, and phosphorene (Figure 3).

antimonene experiences an indirect-to-direct band gap transition at a relatively small critical strain17 and can transfer to a 2D topological insulator at large strain, we may expect that oxidation at a large coverage could covert antimonene into 2D TIs. The 2D topological insulator character as expected above was confirmed by our computations for the fully oxygenated antimonene, namely 18Sb−18O (Figure 4a,e). In the absence

Figure 3. Comparison of key parameters of antimonene oxides with other classical 2D semiconductors. (a) Electronic band structures for antimonene oxide (18Sb−O), black phosphorene, and monolayer MoS2, together with fitted effective masses. (b) Changes in the effective carrier mass of antimonene oxides. The variation of the effective electron masses is similar to that of the black phosphorus with the number of layers.

For antimonene oxide with a low oxygenation ratio, namely 18SbO, the effective electron and hole masses of antimonene oxide are me = 0.17mo and mh = 0.31mo, respectively. Herein, mo is the free-electron mass. The effective electron mass of 18SbO is close to those of monolayer and few-layer phosphorene (meΓX = 0.15−0.18mo), nearly four times smaller than that of monolayer MoS2 (me = 0.60mo) and TiS3 (me = 0.41mo).12,52−56 Interestingly, even with increasing oxygen concentration, the effective electron mass of all antimonene oxide maintains very small. This suggests that all antimonene oxide systems, similar to black phosphorus, may hold higher electron mobility than monolayer MoS2. These outstanding features about carrier effective masses, mobilities and wide band-gaps can open their possibilities for the future applications in high-speed ultrathin transistors and photodetectors. The 2D topology insulators (TIs) are considered as more promising materials than 3D TIs for spin transport applications because the edge states in the former are more robust against scattering than surface states in the latter.30,57−62 However, the existing 2D quantum spin hall (QSH) insulators have small bandgaps of a few millielectronvolts and consequently can only operate at very low temperatures. Because of the large atomic number of Sb, antimonene can have a much enhanced SOC effect. It has been demonstrated that in antimonene, band inversion takes place in the vicinity of the Γ-point when a biaxial tensile strain larger than 14.5% is applied, leading to six tilted Dirac cones in the Brillouin zone.47 SOC effect opens up a topologically nontrivial band gap at the Dirac points with a bulk bandgap up to 270 meV at the tensile strain of 18%, exhibiting the features of 2D topologically insulators. However, large tensile strain from 14.5% to 18% is quite challenging to realize in practice.30 Oxidation expands the lattice parameters of antimonene (Table S2), structurally leading to the same effect as the applied tensile strain (Figure S5). Considering that the pristine

Figure 4. Topological insulator phase of 18Sb−18O antimonene oxide. (a) Electronic band structures of the 18Sb−18O without (blue lines) and with (red lines) SOC at the PBE level. (b) Electronic band structures of 18Sb−18O nanoribbon at the PBE level. The horizontal dash lines in (a,b) indicate the Fermi level. Schematic diagrams of the orbital evaluations at the Γ point for (c) 18Sb−O and (d) 18Sb−18O under (I) chemical bonding, (II) crystal field effect, and (III) SOC in sequence.

of SOC, 18Sb−18O is a gapless semiconductor, its VBM and CBM degenerate at Γ-point, and the Fermi level locates exactly at the degenerate point. When the SOC effect is included, the degenerate point at Γ point is lifted, consequently the valence and conduction bands are well separated with a sizable global band gap of 177 meV, leading to an insulating state. Such a SOC-induced transition from gapless semiconductor to insulator is also observed in many other 2D topological insulator (TI) materials, such as SiTe and H′-MX2 thin films.63,64 These special band features therefore strongly indicate the existence of nontrivial topological state in 18Sb− 18O. To confirm the nontrivial topological states of 18Sb−18O, we first checked the topological Z2 invariant by using the parity criterion proposed by Fu et al.65 Here, the parity eigenvalues 3437


Letter

Nano Letters are calculated for each pair of Kramers’ degenerate occupied bands at the four time-reversal invariant momenta in the 2D Brillouin zone, that is, one Γ- and three M-points. The calculated topological Z2 invariant for 18Sb−18O is 1, indicating that 18Sb−18O is a 2D TI. Note that its nontrivial band gap is quite sizable, about 177 meV, enough to ensure the quantum spin Hall effect being observable at room temperature. To further confirm the nontrivial band topology in 18Sb− 18O, we then investigated the topological edge states of 18Sb− 18O. To simulate the edge states, we built a zigzag nanoribbon of 18Sb−18O with a width of 86 Å (Figure 4f). The calculated edge band structure (Figure 4b) clearly shows the topological states across the bulk bandgap. By checking the spatial distribution of their wave functions, we can confirm that these states are localized on the edges of the nanoribbon. Therefore, 18Sb−18O indeed is a 2D TI. These gapless nontrivial edge states, which are robustness against scattering, are important for applications in spintronics. To further understand the origin of the topological phase transition as well as the band gap dependence on the oxygenation ratio (i.e, x-dependent band gap in 18Sb−xO), we examined the orbital evolutions at the Γ-point for 18Sb−O and 18Sb−18O, as shown in Figure 4c,d, respectively. For both systems, the band edges are dominated by the Sb-5s and Sb5px,y states due to the chemical bonding effect. After including the crystal field effect, these states are split into bonding and antibonding states, which we denote as |s±> and |p±x,y> (Figure 4c,d). Here the subscript ± denotes the parity. For 18Sb−O, without including SOC, the bands near the Fermi level are contributed by the |s−> and |p+x,y> orbitals, and the |s−> orbital locates above the |p+x,y> orbital (Figure 4c). For the 18Sb−xO, increasing the oxygen increases the bond length between the Sb atoms, and weakens the interaction between the Sb atoms, thus decreasing the splitting between the bonding and antibonding states. Consequently, with increasing x in 18Sb−xO, the |s−> orbital would shift toward to the |p+x,y> orbital, which well explains why the energy gap of 18Sb−xO can be continuously tuned by changing the x value. When moving from 18Sb−O to 18Sb−18O, the |s−> orbital shifts below the |p+x,y> orbital (Figure 4d). Because |s−> and |p+x,y> orbitals exhibit different parities, such a band inversion is a nontrivial band inversion, which indicates a topological phase transition. In the inverted band structure, |s−> is occupied, and the degenerate |p+x,y> is half occupied, thereby the system becomes a gapless semiconductor. When including SOC in stage (Figure 4d), the degeneracy of the half-occupied |p+x,y> is lifted and a band gap is opened. Therefore, the nontrivial band topology of 18Sb− 18O arises from the crystal field effect. To further assess the stability of the 18Sb−18O nanosheet, we have computed its phonon spectra along high symmetric lines, as shown in Figure 5a. Clearly, there are no soft phonon modes, indicating that the 18Sb−18O nanosheet is dynamically stable. Furthermore, we have carried out first-principles molecular dynamics (MD) to examine the thermal stability of the 18Sb−18O monolayer. We used a 4 × 4 supercell and carried out MD simulations for 18Sb−18O monolayer at 300 K. Our calculated results show that the nanosheets do not collapse throughout a 3 ps MD simulation at 300 K in Figure 5b, indicating a good thermal stability of the 18Sb−18O monolayer. Besides, we also computed the average formation energy. All the formation energy values are negative and decrease with increasing oxygen concentration (Figure S6),

Figure 5. Stability of 18Sb−18O antimonene oxide. (a) Phonon band dispersions of 18Sb−18O monolayer. (b) Selected snapshots of 18Sb−18O nanosheet in MD simulations at 300 K.

which indicate that the oxide formation is energetically favored in the presence of O2, and thermodynamically oxides with more oxygen concentrations are more favorable. Herein, we discuss some possible experimental methods for the synthesis of antimonene oxides. Quite recently, pure antimonene have been fabricated experimentally in our and other groups.20−25 Moreover, O-antimonene systems are similar to H-graphene because both of them are one-atom thick sheets terminated by one atom. Thus, we can take the hydrogenation of graphene as an analogy and propose some possible methods for the synthesis of antimonene oxides with concentration controllability and selective-area surface functionalization. Two aspects should be taken into account to realize antimonene oxides. (1) Surface group concentration controllability: (a) One method to synthesize graphane (fully hydrogenated graphene) from graphene has been carried out in a cold hydrogen plasma experiments.66 Via a 2 h exposure to hydrogen plasma, fully hydrogenated graphene (graphane) was prepared. In this way, graphene can be hydrogenated in different extent by shortening the treatment durations and lowering the plasma power.66−68 (b) Utilizing electron radiation to activate hydrogen adsorbents on graphene is another available route to fabricate H-functionalized graphene.69 Radiation power and durations remarkably affect the hydrogenation of the final products and both of these two parameters can be adjusted easily. (c) Via STM technology, hydrogen atoms can be added to the graphene lattice more accurately. Through adjusting the flux of hydrogen and exposure time, graphene can be hydrogenated to different extents. (2) Surface selective-area controllability: Selective-area H-graphene patterns have been fabricated using a photoresist mask in a standard photolithography process or a PMMA mask in an e-beam lithography process.70,71 Especially, it should be emphasized to modify the antimonene lattice with activated oxygen at low temperature. Therefore, electron radiation with oxidation coatings, ultraviolet radiation in air, or cold oxygen plasma are all expected to be possible routes to realize the expected O-antimonene systems with controlled O distribution and concentration. To summarize, we have presented first-principles evidence toward the realization of a new class of 2D antimonene oxide materials. Our DFT computations demonstrated that antimonene oxides with different content of oxygen, hold tunable direct bandgaps, which cover a wide range from 0 to 2.28 eV. We observed the small effective electron mass of all antimonene oxides with any oxygen concentration, indicating a high carrier 3438


Letter

Nano Letters

(51572128, 21403109), NSFC-RGC (5151101197), Natural Science Foundation of Jiangsu Province (BK20140769), the Fundamental Research Funds for the Central Universities (No. 30916015106), the Fundamental Research Funds for the Central Universities, and PAPD of Jiangsu Higher Education Institutions, Singapore MOE Academic Research Fund Tier 1 (SUTD-T1-2015004), and in the United States by Department of Defense (Grant W911NF-15-1-0650) and NSF (Grant EPS1002410). We also acknowledge Computer Network Information Center (Supercomputing center) of Chinese Academy of Sciences (CAS) for allocation of computing resource.

mobility. More interestingly, the antimonene oxide (18Sb− 18O) is a 2D topological insulator with a sizable global bandgap of 177 meV, characterized by the nontrivial Z2 topological invariant and the topological edge states. Our findings not only arouse new vitality into 2D group-VA materials family and enrich available candidate materials in this field but also highlight the potential of these 2D semiconductors as appealing ultrathin materials for future flexible electronics, optoelectronics, and photovoltatics devices. Methods. The structural optimizations and electronic structure calculations are performed in the context of density functional theory as implemented in VASP code.72 Exchangecorrelation energies are taken into account by the generalized gradient approximation (GGA) using the Perdew−Burke− Ernzerh function.73 The wave functions are constructed using a projected augmented wave approach with plane wave cutoff energy of 520 eV. The effect of SOC is included selfconsistently in the electronic structure calculations. The atomic positions and cell parameters are fully optimized using conjugate gradient method without imposing any symmetry. After the optimization process, the maximum residual force on each atom is less than 0.001 eV/Å. The total energies are converged within 10−6 eV/cell. A large vacuum space of 20 Å is set along the c-axis, the direction perpendicular to the surface, to avoid any interaction between the layer and its periodic images. The Brillouin zone integration is sampled using a set of 13 × 13 × 1 Monkhorst−Pack k-points. Carrier effective masses of antimonene oxides were calculated at the PBE level of theory using the CASTEP code.74 In addition, the stability of all these antimonene oxides can be quantified by calculating the average formation energy, defined as Ef = [ESbmOn − (ESbm + n/2EO2]/(m + n), where m, n are the number of Sb or O atoms in the supercell, and ESbmOn, ESbm, and EO2 are the total energies of the antimonene oxides, antimonene, and the O2 (triplet) molecule, respectively. We have performed first-principle MD simulations within the NVT ensemble with the time step of 1 fs at 300 K. A Nose−Hoover thermostat with the Noose Q ratio parameter of 1 was used to keep the temperature constant. The 4 × 4 supercell of 18Sb− 18O monolayer was equilibrated for 3 ps with 3000 steps.

■

■

(1) Wang, Q. H.; Kalantar-Zadeh, K.; Kis, A.; Coleman, J. N.; Strano, M. S. Nat. Nanotechnol. 2012, 7, 699−712. (2) Lopez-Sanchez, O.; Lembke, D.; Kayci, M.; Radenovic, A.; Kis, A. Nat. Nanotechnol. 2013, 8, 497−501. (3) Xia, F.; Wang, H.; Jia, Y. Nat. Commun. 2014, 5, 4458. (4) Akinwande, D.; Petrone, N.; Hone, J. Nat. Commun. 2014, 5, 5678. (5) Xia, F.; Wang, H.; Xiao, D.; Dubey, M.; Ramasubramaniam, A. Nat. Photonics 2014, 8, 899−907. (6) Zhao, J.; Liu, H.; Yu, Z.; Quhe, R.; Zhou, S.; Wang, Y.; Liu, C.; Zhong, H.; Han, N.; Lu, J.; Yao, Y.; Wu, K. Prog. Mater. Sci. 2016, 83, 24−151. (7) Mannix, A. J.; Zhou, X. F.; Kiraly, B.; Wood, J. D.; Alducin, D.; Myers, B. D.; Liu, X. L.; Fisher, L. B.; Santiago, U.; Guest, R. J.; et al. Science 2015, 350, 1513−1516. (8) Tai, G.; Hu, T.; Zhou, Y.; Wang, X.; Kong, J.; Zeng, T.; You, Y. C.; Wang, Q. Angew. Chem., Int. Ed. 2015, 54, 15473−15477. (9) Li, L. F.; Wang, Y. L.; Xie, S. Y.; Li, X. B.; Wang, Y. Q.; Wu, R. T.; Sun, H. B.; Zhang, S. B.; Gao, H. J. Nano Lett. 2013, 13, 4671−4674. (10) Li, L.; Yu, Y.; Ye, G. J.; Ge, Q.; Ou, X.; Wu, H.; Feng, D. L.; Chen, H. L.; Zhang, Y. B. Nat. Nanotechnol. 2014, 9, 372−377. (11) Fei, R.; Yang, L. Nano Lett. 2014, 14, 2884−2889. (12) Qiao, J.; Kong, X.; Hu, Z. X.; Yang, F.; Ji, W. Nat. Commun. 2014, 5, 4475−4482. (13) Liu, Q.; Zhang, X.; Abdalla, L. B.; Fazzio, A.; Zunger, A. Nano Lett. 2015, 15, 1222−1228. (14) Koenig, S. P.; Doganov, R. A.; Schmidt, H.; Castro Neto, A. H.; Ö zyilmaz, B. Appl. Phys. Lett. 2014, 104, 103106−103109. (15) Castellanos-Gomez, A.; Vicarelli, L.; Prada, E.; Island, J. O.; Narasimha-Acharya, K. L.; Blanter, S. I.; Groenendijk, D. J.; Buscema, M.; Steele, G. A.; Alvarez, J. V.; et al. 2D Mater. 2014, 1, 025001− 025020. (16) Chen, X.; Wu, Y.; Wu, Z.; Han, Y.; Xu, S.; Wang, L.; Ye, W.; Han, T.; He, Y.; Cai, Y.; Wang, N. Nat. Commun. 2015, 6, 7315. (17) Zhang, S.; Yan, Z.; Li, Y.; Chen, Z.; Zeng, H. Angew. Chem., Int. Ed. 2015, 54, 3112−3115. (18) Zhang, S.; Xie, M.; Li, F.; Yan, Z.; Li, Y.; Kan, E.; Liu, W.; Chen, Z. F.; Zeng, H. B. Angew. Chem. 2016, 128, 1698−1701. (19) Wang, G.; Pandey, R.; Karna, S. P. ACS Appl. Mater. Interfaces 2015, 7, 11490−11496. (20) Tsai, H. S.; Chen, C. W.; Hsiao, C. H.; Ouyang, H.; Liang, J. H. Chem. Commun. 2016, 52, 8409−8412. (21) Wu, X.; Shao, Y.; Liu, H.; Feng, Z.; Wang, Y.; Sun, J.; Liu, C.; Wang, J.; Liu, Z.; Zhu, S.; Wang, Y.; Du, S.; Shi, Y.; Ibrahim, K.; Gao, H. Adv. Mater. 2017, 29, 1605407−1605412. (22) Ares, P.; Aguilar-Galindo, F.; Rodríguez-San-Miguel, D.; Aldave, D. A.; Díaz-Tendero, S.; Alcamí, M.; Martín, F.; Gómez-Herrero, J.; Zamora, F. Adv. Mater. 2016, 28, 6332−6336. (23) Gibaja, C.; Rodriguez-San-Miguel, D.; Ares, P.; Gómez-Herrero, J.; Varela, M.; Gillen, R.; Maultzsch, J.; Hauke, F.; Hirsch, A.; Abellán, G.; Zamora, F. Angew. Chem., Int. Ed. 2016, 55, 14470−14470. (24) Ares, P.; Zamora, F.; Gomez-Herrero, J. ACS Photonics 2017, 4, 600−605. (25) Ji, J.; Song, X.; Liu, J.; Yan, Z.; Huo, C.; Zhang, S.; Su, M.; Liao, L.; Wang, W.; Ni, Z.; Hao, Y.; Zeng, H. Nat. Commun. 2016, 7, 13352.

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.nanolett.7b00297. Additional tables and figures (PDF)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Zhen Zhu: 0000-0002-2982-4607 Zhongfang Chen: 0000-0002-1445-9184 Haibo Zeng: 0000-0002-0281-3617 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was financially supported by the National Basic Research Program of China (2014CB931702), NSFC 3439


Letter

Nano Letters

(61) Zhou, J. J.; Feng, W.; Liu, C. C.; Guan, S.; Yao, Y. Nano Lett. 2014, 14, 4767−4771. (62) Ma, Y.; Dai, Y.; Kou, L.; Frauenheim, T.; Heine, T. Nano Lett. 2015, 15, 1083−1089. (63) Ma, Y.; Kou, L.; Dai, Y.; Heine, T. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 94, 201104−201109. (64) Ma, Y.; Kou, L.; Dai, Y.; Heine, T. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 93, 235451−235456. (65) Schmiedl, T.; Seifert, U. Phys. Rev. Lett. 2007, 98, 106803− 106806. (66) Elias, D. C.; Nair, R. R.; Mohiuddin, T. M. G.; Morozov, S. V.; Blake, P.; Halsall, M. P.; Ferrari, A. C.; Boukhvalov, D. W.; Katsnelson, M. I.; Geim, A. K.; Novoselov, K. S. Science 2009, 323, 610−613. (67) Burgess, J. S.; Matis, B. R.; Robinson, J. T.; Bulat, F. A.; Perkins, F. K.; Houston, B. H.; Baldwin, J. W. Carbon 2011, 49, 4420−4426. (68) Luo, Z.; Yu, T.; Kim, K. J.; Ni, Z.; You, Y.; Lim, S.; Shen, Z.; Wang, S.; Lin, J. ACS Nano 2009, 3, 1781−1788. (69) Jones, J. D.; Mahajan, K. K.; Williams, W. H.; Ecton, P. A.; Mo, Y.; Perez, J. M. Carbon 2010, 48, 2335−2340. (70) Sun, Z.; Pint, C. L.; Marcano, D. C.; Zhang, C.; Yao, J.; Ruan, G.; Yan, Z.; Hauge, R. H.; Tour, J. M. Nat. Commun. 2011, 2, 559. (71) Lee, W. H.; Suk, J. W.; Chou, H.; Lee, J.; Hao, Y.; Wu, Y.; Piner, R.; Akinwande, D.; Kim, K.; Ruoff, R. S. Nano Lett. 2012, 12, 2374− 2378. (72) Kresse, G.; Furthmüller, J. Comput. Mater. Sci. 1996, 6, 15−50. (73) Perdew, J. P.; Burke, L.; Ernzerhof, M. Phys. Rev. Lett. 1996, 77, 3865−3868. (74) Refson, K.; Tulip, P. R.; Clark, S. J. Phys. Rev. B: Condens. Matter Mater. Phys. 2006, 73, 155114−155125.

(26) Pizzi, G.; Gibertini, M.; Dib, E.; Marzari, N.; Iannaccone, G.; Fiori, G. Nat. Commun. 2016, 7, 12585. (27) Kamal, C.; Ezawa, M. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 91, 085423−085432. (28) Aktürk, O. Ü .; Ö zçelik, V. O.; Ciraci, S. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 91, 235446−235455. (29) Akturk, O. U.; Akturk, E.; Ciraci, S. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 93, 035450−035458. (30) Zhao, M.; Zhang, X.; Li, L. Sci. Rep. 2015, 5, 16108−16114. (31) Zhou, J.; Sun, Q.; Wang, Q.; Kawazoe, Y.; Jena, P. Nanoscale 2016, 8, 11202−11209. (32) Zhang, S.; Xie, M.; Cai, B.; Zhang, H.; Ma, Y.; Chen, Z.; Zhu, Z.; Hu, Z.; Zeng, Z. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 93, 245303−245309. (33) Sun, Q.; Dai, Y.; Ma, Y.; Yin, N.; Wei, W.; Yu, L.; Huang, B. 2D Mater. 2016, 3, 035017−035027. (34) Singh, D.; Gupta, S. K.; Sonvane, Y.; Lukačević, I. J. Mater. Chem. C 2016, 4, 6386−6390. (35) Kou, L.; Ma, Y.; Tan, X.; Frauenheim, T.; Du, A.; Smith, S. J. Phys. Chem. C 2015, 119, 6918−6922. (36) Zhu, Z.; Guan, J.; Tománek, D. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 91, 161404−161408. (37) Zhang, S.; Hu, Y.; Hu, Z.; Cai, B.; Zeng, H. Appl. Phys. Lett. 2015, 107, 022102. (38) Zhang, H.; Ma, Y.; Chen, Z. Nanoscale 2015, 7, 19152−19159. (39) Zeraati, M.; Allaei, S. M. V.; Sarsari, I. A.; Pourfath, M.; Donadio, D. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 93, 085424−085429. (40) Nie, Y.; Rahman, M.; Wang, D.; Wang, C.; Guo, G. Sci. Rep. 2016, 5, 17980−17985. (41) Ö zçelik, V. O.; Aktürk, O. Ü .; Durgun, E.; Ciraci, S. Phys. Rev. B: Condens. Matter Mater. Phys. 2015, 92, 125420−125427. (42) Lee, J.; Tian, W. C.; Wang, W. L.; Yao, D. X. Sci. Rep. 2015, 5, 11512−11527. (43) Zhang, Q.; Schwingenschlögl, U. Phys. Rev. B: Condens. Matter Mater. Phys. 2016, 93, 045312−045317. (44) Wang, Y.; Ji, W.; Zhang, C.; Li, P.; Li, F.; Ren, M.; Chen, X.; Yuan, M.; Wang, P. Sci. Rep. 2016, 6, 20342−20349. (45) Zhao, J.; Li, Y.; Ma, J. Nanoscale 2016, 8, 9657−9666. (46) Wang, D.; Chen, L.; Shi, C.; Wang, X.; Cui, G.; Zhang, P.; Chen, Y. Sci. Rep. 2016, 6, 28487−28493. (47) Lu, Y.; Zhou, D.; Chang, G.; Guan, S.; Chen, W.; Jiang, Y.; Jiang, J.; Lin, H.; Wang, X.; Yang, S.; et al. Npj. Comput. Mater. 2016, 2, 16011−26032. (48) Liu, M. Y.; Huang, Y.; Chen, Q. Y.; Cao, C.; He, Y. Sci. Rep. 2016, 6, 29114−29126. (49) Tsai, H.; Wang, S.; Hsiao, C.; Chen, C.; Ouyang, H.; Chueh, Y.; Kuo, H.; Liang, J. Chem. Mater. 2016, 28, 425−429. (50) Allen, J. P.; Carey, J. J.; Walsh, A.; Scanlon, D. O.; Watson, G. W. J. Phys. Chem. C 2013, 117, 14759−14769. (51) Rudenko, A. N.; Katsnelson, M. I.; Roldán, R. Phys. Rev. B: Condens. Matter Mater. Phys. 2017, 95, 081407−081412. (52) Ramasubramaniam, A. Phys. Rev. B: Condens. Matter Mater. Phys. 2012, 86, 115409−115414. (53) Dai, J.; Zeng, X. C. Angew. Chem. 2015, 127, 7682−7686. (54) Fei, R.; Yang, Li. Nano Lett. 2014, 14, 2884−2889. (55) Cai, Y.; Zhang, G.; Zhang, Y. J. Am. Chem. Soc. 2014, 136, 6269−6275. (56) Liu, H.; Neal, A. T.; Zhu, Z.; Xu, X.; Tománek, D.; Ye, P. D.; Luo, Z. ACS Nano 2014, 8, 4033−4041. (57) Xu, Y.; Yan, B.; Zhang, H. J.; Wang, J.; Xu, G.; Tang, P.; Duan, W.; Zhang, S. Phys. Rev. Lett. 2013, 111, 136804−136808. (58) Song, Z.; Liu, C. C.; Yang, J.; Han, J.; Ye, M.; Fu, B.; Yang, Y.; Niu, Q.; Lu, J.; Yao, Y. NPG Asia Mater. 2014, 6, e147−e153. (59) Zhang, H.; Liu, C. X.; Qi, X. L.; Dai, X.; Fang, Z.; Zhang, S. C. Nat. Phys. 2009, 5, 438−442. (60) Liu, Q.; Zhang, X.; Abdalla, L. B.; Fazzio, A.; Zunger, A. Nano Lett. 2015, 15, 1222−1228. 3440
