Stability, electronic structure and magnetic properties of vacancy and nonmetallic atom-doped buckled arsenene: first-principles study

Abstract

Using first-principles calculations, we systematically explore the influence of vacancies and a series of substitutional nonmetallic atoms, such as H, F, B, N, P, C, Si, O and S, on the geometrical structure, electronic structure and magnetic properties of buckled arsenene. The calculations show that compared to graphene and silicene, vacancies are more easily formed in buckled arsenene, and vacancy doped buckled arsenene is thermo-dynamically stable at room temperatures. Moreover, all the substitutionally doped buckled arsenene samples with nonmetallic atoms are stable. Remarkably, due to the formation of one nonbonding p electron of dopant C and Si or a neighboring As atom around O, Si and vacancies, a doping C, Si, O and S atom and a vacancy induce a magnetic moment of 1.0 μ_B in buckled arsenene. Furthermore, it is found that the magnetic coupling between the moments induced by two C, Si, O and S are long-range anti-ferromagnetic, and the calculated DOS and the spin density distribution show that the p–p hybridization interaction involving polarized electrons is responsible for the magnetic coupling. Our results demonstrate that the magnetism of buckled arsenene can be effectively engineered by vacancies and the substitutional doping of some nonmetallic atoms.

---

1. Introduction

In recent years, two-dimensional (2D) layered materials, including graphene,^1–3 hexagonal boron nitride (h-BN),⁴ transition metal dichalcogenides (TMDs),^5–11 and black phosphorus,^12–16 have drawn tremendous research interest in condensed matter physics and materials research. It has been demonstrated that the monolayer form of black phosphorus, black phosphorene, possesses a higher current on/off ratio compared to graphene and larger carrier mobility than MoS₂, which render black phosphorene promising for application in nanoelectronics.^17–19 Furthermore, it was theoretically predicted that layered blue phosphorus is nearly as stable as black phosphorus, the most stable phosphorus allotrope, and the weak interlayer interaction should allow the mechanical exfoliation of blue phosphorus in analogy to the black allotrope.^20,21 More recently, arsenene, which is monolayer of arsenic, was theoretically proposed as a new 2D material of group V element and two types of honeycomb structures, buckled and puckered, were found to be stable.^22–25 Contrast to phosphorene, buckled arsenene is little more energetically favorable than the puckered one.²² Moreover, the most stable and the most common As allotrope is gray arsenic, which crystallize in a rhombohedrally stacked layered structure at ambient condition, similar to blue phosphorus.²⁶ Due to the weak bonding between layers, it might be possible to fabricate buckled arsenene by exfoliation from gray arsenic, as in cases of graphene and phosphorene.^22–25 In particular, it was predicted that buckled arsenene is a semiconductor with a band gap of about 1.52–2.49 eV and has a high carrier mobility of about 10³ cm² V⁻¹ s⁻¹.^{22–25,27,28} More importantly, recent theoretical works showed that the pristine buckled arsenene can be converted to a 2D quantum spin Hall (QSH) insulator by applying a critical value of tensile strain of 11.14% and hydrogenated arsenene is a QSH insulator with a band-gap as large as 193 meV.^29,30 These novel properties make buckled arsenene promising for future application in nanoelectronics.

As known, a spintronic device requires generation, transportation and detection of tunable spin currents, which can ideally be done by using a magnetic semiconductor with high carrier mobility. However, 2D semiconducting materials are intrinsically nonmagnetic.^6,9,28,31 Therefore, various approaches have been used to induce and tune the magnetic properties of 2D semiconducting materials for future application in spintronic devices.^32–45 The substitutional doping has been proved to be one of the most effective approaches.^32–40 Moreover, it was experimentally demonstrated that the substitutional doping of graphene, boron nitride and TMDs may be achieved by filling the vacancies created by the electron beam with substitutional atoms.^4,5,46,47 In this respect, it has been reported theoretically that the substitutional doping of nonmetallic atoms can induce the magnetism in 2D semiconducting materials.^{5,28,31–34,40,44,45,48} For example, the substitutional doping of numerous nonmetallic atoms in the MoS₂, SnS₂ and ReS₂ monolayer are stable under appropriate experimental conditions and H, B, N, P, As or F doping can induce a magnetic moment.^5,32,34 Likewise, the substitution of nonmetallic C, Si, O, S and Se atoms can produce magnetic moment in black phosphorene.^33,48 In addition, it is known that vacancy defects are inevitable in any material as suggested by the second law of thermodynamics and can significantly affect the magnetic properties of some 2D materials, such as graphene,³⁶ h-BN,^40,41 and TMDs.^42,43 Very recently, it was theoretically reported that the substitution of C and Ge atoms can produce magnetic moment in buckled arsenene and the semi-hydrogenated buckled arsenene is a magnetic material.^28,49 However, the influence of substitutional nonmetallic H, F, P, Si and S atoms on the electronic structure and magnetic properties of buckled arsenene and the magnetic coupling between the moments induced by two substitutional nonmetallic atoms have not been reported yet. In this paper, by using first-principles calculations, we systematically investigate the structural characteristics and corresponding electronic structure and magnetic properties of vacancy doped and substitutionally doped buckled arsenene with a series of nonmetallic atoms, such as H, F, B, N, P, C, Si, O and S.

2. Computational details

Our calculations were performed using the density functional theory implemented in Vienna Ab initio Simulation Package (VASP).^50,51 The generalized gradient approximation (GGA) proposed by Perdew, Burke and Ernzerhof (PBE) was adopted to describe the exchange–correlation energy.⁵² The ion–electron interaction was treated with the projector-augmented wave (PAW) method.⁵³ All atomic positions and lattice constants were optimized using the conjugate gradient method, until the force on each atom was less than 0.02 eV Å⁻¹. The energy cutoff for the plane wave basis was set to 450 eV and the Brillouin zone was sampled by a 4 × 4 × 1 Monkhorst–Pack k-point mesh. A vacuum space of 20 Å was adopted in the direction perpendicular to the plane of buckled arsenene to eliminate the interaction between periodic replicas. The vacancy and nonmetallic atoms doped buckled arsenene were modeled with a 5 × 5 or 6 × 6 supercell, which consist of 50 and 72 atoms in total, respectively, as shown in Fig. 1. Moreover, to examine the stability of doped buckled arsenene, ab initio molecular dynamics (MD) simulation at 300 K, where the canonical ensemble was used and the simulation time is limited to 3 ps with a time step of 1 fs, was performed with the VASP.

Fig. 1 Top view of structure of the buckled arsenene 5 × 5 (a) and 6 × 6 (b) supercells. Green and red balls represent As and doping atoms, respectively. The numbers label doping atoms and the neighboring As atoms around doping atom (or vacancy) in order to reference below.

The formation energy of a vacancy or substitutional dopant D in buckled arsenene is defined as

where E_tot[As₄₉D] is the total energy of the supercell with one vacancy or substitutional dopant D, and E_tot[As₅₀] is the total energy of the stoichiometric supercell. μ_i is the chemical potentials of atom species i (host As atoms or substitutional atoms X) and n_i is the number of atoms of species i that have been added (n_i > 0) to or removed from (n_i < 0) the supercell when the vacancy or substitutional doping is created. We use the energy of an atom in pristine buckled arsenene as the chemical potential of As atom. The chemical potential of substitutional atom X is taken as the energy of an atom in bulk X/diatomic X₂.

3. Results and discussion

Before studying the effects of vacancy and substitutional dopant, let us here investigate the stoichiometric buckled arsenene. The result shows that the optimized lattice constant, the buckling height, As–As bond length and bond angle of buckled arsenene are 3.609 Å, 1.397 Å, 2.509 Å and 91.99°, respectively. Moreover, buckled arsenene is an indirect semiconductor with a band gap of 1.597 eV and the conduction band minimum and the valence band maximum occur along the Γ–M direction and at the Γ point, respectively. These results are consistent with other calculations,^{22–25,27,28} indicating reliability of our computational methodology. As shown in Fig. 2(a), each As atom in buckled arsenene is covalently bonded with three neighbors by sharing its three p electrons, while the remaining two p electrons of the As atom form a lone-pair of electrons. As a result, the five valence electrons of each As atom are all paired, and thus stoichiometric buckled arsenene is intrinsically nonmagnetic. Correspondingly, the energy difference between the spin polarized and non-spin polarized state for stoichiometric buckled arsenene is zero, as listed in Table 1.

Fig. 2 Differential electron density for stoichiometric buckled arsenene, vacant and H, F, B, N, P, C, Si, O and S doped buckled arsenene. The yellow and light blue indicates the electron accumulation and depletion, respectively. Green and red balls represent As and doping atoms, respectively. The numbers label doping atom and the neighboring As atoms around doping atom (or vacancy) in order for reference below.

Table 1 The energy difference (ΔE_Spin) between the spin polarized and non-spin polarized state, the formation energy (E^f) of a vacancy or substitutional dopant, the distance between Y₁ and Y₂ atoms (d_Y1–Y2), and the magnetic moment of the supercell (M) for stoichiometric buckled arsenene As₅₀, a vacancy doped buckled arsenene As₄₉ and X doped buckled arsenene As₄₉X (X = H, B, F, N, P, C, Si, O and S)

System	ΔESpin (meV)	E^f (eV)	dX0–As1 (Å)	dX0–As2 (Å)	dX0–As3 (Å)	dAs2–As3 (Å)	dAs1–As3 (Å)	MSup (μB)
As50	0.00	—	2.509	2.509	2.509	3.609	3.609	0.0
As49	−333.40	2.15	—	—	—	2.81	3.55	1.0
As49H	−0.05	−1.04	3.24	1.53	3.23	3.61	2.98	0.0
As49F	324.52	−1.04	3.41	1.83	3.41	3.54	2.83	0.0
As49B	−0.05	1.43	2.06	2.06	2.06	3.48	2.48	0.0
As49N	−0.04	1.82	2.00	2.00	2.00	3.15	3.15	0.0
As49P	−0.05	0.10	2.40	2.40	2.40	3.50	2.50	0.0
As49C	−130.89	3.06	1.99	1.99	1.99	3.32	3.32	1.0
As49Si	−170.22	1.08	2.40	2.40	2.40	3.78	3.78	1.0
As49O	−144.14	−0.20	3.19	1.89	1.89	3.08	3.81	1.0
As49S	−112.66	−0.22	3.14	2.31	2.31	3.44	3.89	1.0

---

Next we investigate As vacancy (V_As) doped buckled arsenene system As₄₉, in which an As atom at site 0 in 5 × 5 supercell is removed, as shown in Fig. 1(a). As listed in Table 1, the formation energy of V_As in buckled arsenene is about 2.1 eV, which is much lower than that in graphene and that in silicene,^54,55 indicating that single vacancy is much easily formed in buckled arsenene, as compared with graphene and silicene. To access the stability of V_As doped buckled arsenene, MD simulations are performed for V_As doped system As₄₉. The movies in the ESI† show that after simulating 3000 steps, the atoms are not fluctuate significantly and the structure of the As vacancy doped system does not spontaneously disintegrate, confirming that the V_As doped buckled arsenene is thermo-dynamically stable at room temperatures. As shown in Fig. 3(a), V_As breaks the 3-fold symmetry of buckled arsenene. More interestingly, the energy difference between the spin-polarized and spin-unpolarized states indicates that the ground state of V_As doped system is magnetic and the magnetic moment induced by an As vacancy is 1.0 μ_B, as listed in Table 1. These results are different from those reported by ref. 28, where V_As doped buckled arsenene maintains the 3-fold symmetry and V_As don't induce the magnetism in buckled arsenene. The charge density difference in Fig. 2(b) clearly shows that the electrons accumulate in the middle region between As₂ and As₃ atom around V_As, which indicate that the As₂ and As₃ atoms are bonded with each other. As a result, each of As₂ and As₃ atoms around V_As still has three covalent bonds with its neighbors similar to As atom in stoichiometric buckled arsenene, while As₁ atom around V_As is bonded with its two neighbors, which leads to the formation of one nonbonding p electron of As₁ atom around V_As. Accordingly, the magnetic moment of 1.0 μ_B is induced by an As vacancy and it mainly arise from the localized nonbonding p electrons of As₁ atom around V_As. Indeed, the density of states (DOS) in Fig. 4(a) and the spin density distribution in Fig. 5(a) demonstrate that the spin polarization mainly arise from the p orbitals of As₁ atom around V_As. Corresponding to the formation of the bond between As₂ and As₃ atoms, the distances between As₂ and As₃ atoms, d_As2–As3, obviously decreases, as listed in Table 1.

Fig. 3 Top view of optimized structures of vacant and the doping site in vacancy, H, F, B, N, P, C, Si, O and S doped buckled arsenene. Green and red balls represent As and doping atoms, respectively. The numbers label doping atom and the neighboring As atoms around doping atom (or vacancy) in order for reference below.

Fig. 4 Total DOS of vacant, C, Si, O and S doped buckled arsenene and corresponding partial DOS of the p orbitals of doping X atom and the neighboring As atoms around doping atom (or vacancy). The Fermi level is indicated by the vertical dashed line.

Fig. 5 Spin density distribution of the relaxed vacant, C, Si, O and S doped buckled arsenene. The yellow and light blue isosurfaces correspond to the majority- and minority-spin densities. Green and red balls represent As and doping X atoms, respectively.

And then we study the nonmetallic atom doped buckled arsenene systems As₄₉X (X = H, F, B, N, P, C, Si, O and S), in which a doping atom X substitutes As atom at site 0 in 5 × 5 supercell, as shown in Fig. 1(a). It can be seen from the Table 1 that the formation energy of the various doping atoms in doped buckled arsenene are negative or little positive value. Combined with the experimental report that substitutional doping of some 2D materials may be achieved by filling the vacancies with impurity atoms, it appears that substitutional doping of nonmetallic atom in buckled arsenene is likely to form under appropriate experimental conditions. Furthermore, we examine the stability of the doped buckled arsenene by performing MD simulations for the nonmetallic atom doped systems As₄₉X, where simulation time is limited to 3 ps due to the large supercell size. As already reported by Tománek et al. in the ref. 20 and 21, the simulation time of about 1–2 ps can indicate the propensity to structural changes for large monolayer phosphorus supercells. It can be seen from the movies of the MD simulations in the ESI† that the doping atom and its neighboring As atoms only vibrate around the equilibrium position and the doped systems do not display significant structural changes at the end of the MD simulation, confirming that all of substitutionally doped buckled arsenene are stable at 300 K.

In order to investigate the influence of substitutional nonmetallic atoms on the magnetic properties of buckled arsenene, the energy difference between the spin polarized and non-spin polarized state for doped buckled arsenene systems As₄₉X (X = H, F, B, N, P, C, Si, O and S), i.e., ΔE_Spin = E_sp − E_nsp, are calculated and the calculated values are listed in Table 1. Table 1 show that ΔE_Spin of F, H, B, N and P doped buckled arsenene are large positive values or almost zero, suggesting that their ground states are nonmagnetic. In contrast, ΔE_Spin of C, Si, O and S doped systems are large negative values, as listed in Table 1, which indicate that the substitutional doping of C, Si, O and S can induce magnetism and stability of magnetic state is large. As listed in Table 1, the magnetic moments induced by a C, Si, O and S atoms are all 1.0 μ_B. Differently, the calculations by Li et al. revealed that the substitutional doping of O don't induce the magnetic moment in buckled arsenene.²⁸

We now turn to discuss the reason why the substitutional doping of H, F, B, N and P do not induced the magnetism in buckled arsenene. Fig. 3 shows the optimized atomic structure of the doping site in doped buckled arsenene. In addition, Table 1 list the distance between Y₁ and Y₂ atoms, d_X0–P4(5), which is used to quantify the local structure deformation caused by substitutional doping. It can be seen from Fig. 3(b) and (c) that the substitutional doping of H and F break the 3-fold symmetry of the buckled arsenene with the dopant H and F locating closer to nearest As₂ atom. In quantitative, Table 1 show that for H and F doped system, d_X0–As2 decreases about 39% and 27%, while both of d_X0–As1 and d_X0–As3 increase about 29% and 36%, respectively, in comparison with those of stoichiometric buckled arsenene. Moreover, the distance between As₁ and As₃ around H and F, d_As1–As3, greatly decreases, as listed in Table 1. Corresponding to the local structure deformation, the charge density difference in Fig. 2(c) and (d) shows that dopant H (F) only is bonded with the nearest As₂ atom and As₁ and As₃ atom around H (F) is covalently bonded with each other, which is different from stoichiometric buckled arsenene. As a result, an valence electron of dopant H and F is saturated due to the formation of the bond between dopant H (F) and As₂ atom, and other six unsaturated electrons of F are also paired each other. Similar to As atom in stoichiometric buckled arsenene, the five valence electrons of each As atom around dopant H (F) are all paired. Therefore the substitutional doping of F and H cannot induce the magnetism in doped buckled arsenene. In contrast to F and H doped systems, although the distance between B and its three nearest As atoms obviously shortens, the structure of B doped buckled arsenene maintains the 3-fold symmetry, as shown in Fig. 3(d). The charge density difference in Fig. 2(e) shows that B atom also is covalently bonded with its three nearest As atoms. Consequently the three valence electrons of doping B atom are all saturated due to the covalent interaction with the electrons of three nearest As atoms, and hence the doping of B don't induce the magnetism in B doped buckled arsenene. Similar to B doped system, dopant N and P is also covalently bonded with its three nearest As atoms, respectively, as shown in Fig. 2(f) and (g). Just as As atom in buckled arsenene, three valence electrons of N and P atom are saturated by bonding with three nearest As atoms and the remaining two are paired each other, thus both of N and P doped buckled arsenene are nonmagnetic.

In the following, we discuss the mechanism about origin of magnetism in C, Si, O and S doped systems. It can be seen from Table 1 that the structure of dopant C and Si site is almost same with that of N and P site, respectively, which indicate that the structures of C and Si doped buckled arsenene maintain the 3-fold symmetry, as shown in Fig. 3(g) and (h). Correspondingly, the charge density difference in Fig. 2(h) and (i) show that dopant C and Si is covalently bonded with its three nearest As atoms, respectively, which is similar to the bonding configuration in N and P doped systems. Among the four valence electrons of dopant C and Si, the three electrons are saturated due to the interaction with the electrons of three nearest As atoms, while one remaining p electron of C (Si) is nonbonding and unpaired, thus a doping C (Si) atom inducing the magnetic moment of 1.0 μ_B. Fig. 4(b) and (c) show that the p states of dopant C (Si) and the 4p states of the nearest As and the second neighboring As atoms around dopant overlap near the Fermi level, which suggest that the p orbitals of dopant C (Si) hybridize with the 4p orbitals of the nearest and the second neighboring As atoms around C (Si). Therefore the unpaired electron induced by dopant C (Si) not only occupy the p orbitals of C (Si), but also partially occupy the 4p orbitals of the three nearest and the six second neighboring As atoms around C (Si). According to proposition by Shen et al.,⁵⁶ the spin of p electrons of As atoms around dopant tend to align parallel to the moment of the dopant under the p–p hybridization interaction. Thus main part of the magnetic moments induced by C (Si) come from dopant C (Si) and three nearest As and six second neighboring As atoms around C (Si) also provide a partial contribution. Indeed, the spin density distribution in Fig. 5(b) and (c) reveal that most of the spin density are localized on the dopant C (Si) and rest spin density on three nearest As and six second neighboring As atoms around the dopant. In contrast to C and Si doped system, Fig. 3(i) and (j) show that the doping of O and S break the 3-fold symmetry of the buckled arsenene with O (S) locating farther from As₁ atom and closer to As₂ and As₃ atoms, which is different from the previous theoretical result that O doped buckled arsenene maintains the 3-fold symmetry.²⁸ In quantitative, d_O–As1 and d_S–As1 increase about 27% and 25%, while d_O–As2(d_O–As3) and d_S–As2(d_S–As3) decrease about 25% and 8%, respectively, as listed in Table 1. Corresponding to the structure deformation caused by O (S), the charge density difference in Fig. 2(j) and (k) show that the dopant O (S) is only bonded with the nearest As₂ and As₃ atoms, while it is not bonded with the As₁ atom around dopant, which result in formation of one nonbonding electron of As₁ atom. Consequently, two valence electrons of O (S) are saturated by bonding with As₂ and As₃ atoms and other four are paired each other, while one nonbonding electron of As₁ atom is unpaired, which produce the magnetic moment of 1.0 μ_B in the O and S doped systems. Just as Shen et al. propose in ref. 56, the spatially extended p states of the dopant and host As atom are able to extend the p–p interaction to a large range, the DOS in Fig. 4(d) and (e) reveal that the p states of As₁ atom not only hybridizes with the p states of dopant O (S) near the Fermi level, but also hybridizes with the p states of some neighboring As atoms around As₁ atom near the Fermi level, which indicate that the nonbonding p electron of As₁ atom not only occupy the p orbitals of As₁, but also partially occupy p orbitals of some neighboring As atoms around As₁ and p orbitals of the dopant O and S. Furthermore, the spin of p electron of the dopant and As atoms around As₁ atom align parallel to magnetic moment of As₁ atom through the p–p hybridization interaction.⁵⁶ Therefore, main part of the magnetic moments induced by O come from As₁ atom, and the dopant O and some neighboring As atoms around As₁, such as As₅, and As₆, also provide some contribution, which also can be drawn from the calculated spin density distribution in the relaxed As₄₉O and As₄₉S, as shown in Fig. 5(d) and (e).

Finally, we study the magnetic coupling between the magnetic moments induced by two dopants, which substitute a pair of As atoms in a 6 × 6 supercell. As shown in Fig. 1(b), we investigate two positional configurations of the dopants in the supercell, in which two dopants in the supercell are located at 0 and 1 sites, and 0 and 2 sites, respectively, and the two configurations are labeled as (0, 1) and (0, 2), respectively. For each configuration of doped systems As₇₀X₂ (X = C, Si, O and S), ferromagnetic (FM) and antiferromagnetic (AFM) calculations are performed by specifying parallel and anti-parallel alignment of the moments induced by two dopants in the supercell, respectively. The calculations show that for all configurations of As₇₀X₂ (X = C, Si, O and S), the value of magnetic moments induced by each dopant X in the FM and AFM state are all 1.0 μ_B and the corresponding magnetic moments distribution are almost the same as that for As₄₉X (X = C, Si, O and S), as shown in Fig. 5 and 6. Table 2 list the energy difference between FM and AFM states, i.e. ΔE_m = E_FM − E_AFM, the relaxed distance between the two dopants in the supercell, and the magnetic moment of the supercell in AFM state for each configuration of doped systems. As listed in Table 2, for configuration (0, 2) of C and Si doped systems and configuration (0, 1) of O and S doped systems, the magnetic moments induced by two dopants favor AFM coupling, while for configuration (0, 1) of C and Si doped systems and configuration (0, 2) of O and S doped systems, the magnetic coupling between the moments induced by two dopants is very weak. The long-range AFM coupling between the magnetic moments induced by dopants can be explained by analyzing the calculated DOS and spin density distribution. The spin density in Fig. 5 shows that some As atoms around dopant X are polarized to different degrees and the spin alignment of these As atoms and dopant parallel to each other under the p–p interaction. As a result, the p electrons localized around the As atoms between two dopants X in the supercell are polarized to different degrees, as shown in Fig. 6, and these polarized p electrons are able to effectively mediate indirect magnetic coupling between the magnetic moments induced by two dopants.^56,57 On the other hand, DOS in Fig. 4 show that for C and Si (or O and S) doped systems, the majority-spin p states of the doping atoms (or the As₁ atom) near the Fermi level are completely occupied, whereas the higher energy minority-spin p states of the doping atoms (or the As₁ atom) are empty. Consequently, when the spin alignment of two doping atoms (or two As₁ atom) are antiparallel, the spin-conserving hopping for electrons from the p orbitals of one doping atom (or one As₁ atom) to the p orbitals of other neighboring doping atom (or other neighboring As₁ atom) can lower energy of the doped system due to the strong intra-atomic exchange interaction between the electrons in p level, while this hopping is not allowed when the spins are parallel to each other,⁵⁸ which result in an AFM coupling between magnetic moments induced by two dopants in C, Si, O and S doped buckled arsenene.

Fig. 6 The spin density distribution of AFM ground state for the configuration (0, 2) of relaxed As₇₀X₂ (X = C, Si) and the configuration (0, 1) of relaxed As₇₀X₂ (X = O, S). The yellow and light blue iso-surfaces correspond to the majority- and minority-spin densities. Green and red balls represent As and doping X atoms, respectively.

Table 2 The relaxation distance (d) between the two dopants, the energy difference (ΔE_m) between FM and AFM states, and the magnetic moment of the supercell (M_Sup) in AFM state for X doped buckled arsenene As₇₀X₂ (X = C, Si, O and S)

System	Configuration (0, i)	d (Å)	ΔEm (meV)	MSup (μB)
As70C2	(0, 1)	6.22	1.73	0.0
	(0, 2)	7.14	7.21	0.0
As70Si2	(0, 1)	6.26	2.11	0.0
	(0, 2)	7.24	26.34	0.0
As70O2	(0, 1)	6.20	30.48	0.0
	(0, 2)	7.06	0.63	0.0
As70S2	(0, 1)	6.24	12.18	0.0
	(0, 2)	7.15	0.36	0.0

---

4. Conclusions

In summary, substitutional doping of nonmetallic atom in buckled arsenene are possible under appropriate experimental conditions and the vacancy and nonmetal atom doped buckled arsenene are thermo-dynamically stable at room temperature. Among the investigated nonmetallic atoms, the substitutional doping of H, F, B, N and P can not produce magnetism in buckled arsenene due to the saturation or pairing of valence electron of dopant and its neighboring As atoms. In contrast, it is found that a vacancy and a doping C, Si, O and S can induce the magnetic moments of 1.0 μ_B in buckled arsenene, which mainly come from one nonbonding valence electron of C and Si or a neighboring As atom around O, S and vacancy. Furthermore, the magnetic coupling between magnetic moments induced by two C, Si, O and S is found to be long-range AFM. By analyzing the calculated DOS and the spin density distribution, it is proposed that the AFM coupling can be attributed to the p–p hybridization interaction involving polarized electrons.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (11174104) and Special Program for Applied Research on Super Computation of the NSFC-Guangdong Joint Fund (the second phase) (nsfc2015_179). The authors acknowledge the computational support provided by High Performance Computing Center of Jilin University, China.

References

1. A. K. Geim and K. S. Novoselov, Nat. Mater., 2007, 6, 183.
2. R. R. Nair, P. Blake, A. N. Grigorenko, K. S. Novoselov, T. J. Booth, T. Stauber, N. M. R. Peres and A. K. Geim, Science, 2008, 320, 1308.
3. M. Y. Han, B. Özyilmaz, Y. B. Zhang and P. Kim, Phys. Rev. Lett., 2007, 98, 206805; L. Y. Jiao, L. Zhang, X. R. Wang, G. Diankov and H. J. Dai, Nature, 2009, 458, 877; X. L. Li, X. R. Wang, L. Zhang, S. Lee and H. J. Dai, Science, 2008, 319, 1229; F. Schwierz, Nat. Nanotechnol., 2010, 5, 487.
4. J. C. Meyer, S. Kurasch, H. J. Park, V. Skakalova, D. Künzel, A. Groβ, A. Chuvilin, G. Algara-Siller, S. Roth, T. Iwasaki, U. Starke, J. H. Smet and U. Kaiser, Nat. Mater., 2011, 10, 209–215.
5. H. P. Komsa, J. Kotakoski, S. Kurasch, O. Lehtinen, U. Kaiser and A. V. Krasheninnikov, Phys. Rev. Lett., 2012, 109, 035503.
6. S. Lebègue, T. Björkman, M. Klintenberg, R. M. Nieminen and O. Eriksson, Phys. Rev. X, 2013, 3, 031002.
7. H. Shi, H. Pan, Y. W. Zhang and B. I. Yakobson, Phys. Rev. B: Condens. Matter Mater. Phys., 2013, 87, 155304.
8. B. Radisavljevic and A. Kis, Nat. Mater., 2013, 12, 815–820.
9. C. Ataca, H. Şahin and S. Ciraci, J. Phys. Chem. C, 2012, 116, 8983–8999.
10. K. F. Mak, C. Lee, J. Hone, J. Shan and T. F. Heinz, Phys. Rev. Lett., 2010, 105, 136805.
11. Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman and M. S. Strano, Nat. Nanotechnol., 2012, 7, 699–712.
12. L. Li, Y. Yu, G. J. Ye, Q. Ge, X. Ou, H. Wu, D. Feng, X. H. Chen and Y. Zhang, Nat. Nanotechnol., 2014, 9, 372–377.
13. E. S. Reich, Nature, 2014, 506, 19.
14. W. Lu, H. Nan and J. Hong, et al., Nano Res., 2014, 7, 853–859.
15. Y. Liu, F. Xu, Z. Zhang, E. S. Penev and B. I. Yakobson, Nano Lett., 2014, 14, 6782–6786.
16. A. Castellanos-Gomez, L. Vicarelli, E. Prada, J. O. Island, K. L. Narasimha-Acharya, S. I. Blanter, D. J. Groenendijk, M. Buscema, G. A. Steele, J. V. Alvarez, H. W. Zandbergen, J. J. Palacios and H. S. J. van der Zant, 2D Mater., 2014, 1, 025001.
17. S. Das, W. Zhang, M. Demarteau, A. Hoffmann, M. Dubey and A. Roelofs, Nano Lett., 2014, 14, 5733–5739.
18. J. Qiao, X. Kong, Z.-X. Hu, F. Yang and W. Ji, Nat. Commun., 2014, 5, 4475.
19. H. Liu, A. T. Neal, Z. Zhu, Z. Luo, X. Xu, D. Tománek and P. D. Ye, ACS Nano, 2014, 8, 4033–4041.
20. Z. Zhu and D. Tománek, Phys. Rev. Lett., 2014, 112, 176802.
21. J. Guan, Z. Zhu and D. Tománek, Phys. Rev. Lett., 2014, 113, 046804.
22. C. Kamal and M. Ezawa, Phys. Rev. B: Condens. Matter Mater. Phys., 2015, 91, 085423.
23. Z. Zhu, J. Guan and D. Tománek, Phys. Rev. B: Condens. Matter Mater. Phys., 2015, 91, 161404.
24. L. Kou, Y. Ma, X. Tan, T. Frauenheim, A. Du and S. Smith, J. Phys. Chem. C, 2015, 119, 6918–6922.
25. S. Zhang, Z. Yan and Y. Li, Angew. Chem., Int. Ed., 2015, 54, 3112–3115.
26. R. boča, P. Hajko, L. Benco, I. Benkovský and D. Faktor, Czech J. Phys., 1993, 43, 813.
27. Y. Wang and Y. Ding, Nano. Res. Lett., 2015, 10, 1–10.
28. Z. Li, W. Xu, Y. Yu, H. Du, K. Zhen, W. Wang, L. Luo, H. Qiu and X. Yang, J. Mater. Chem. C, 2016, 4, 362.
29. Y. P. Wang, C. W. Zhang, W. X. Ji, R. W. Zhang, P. Li, P. J. Wang, M. J. Ren, X. L. Chen and M. Yuan, J. Phys. D: Appl. Phys., 2016, 49, 055305.
30. Y. P. Wang, W. X. Ji, C. W. Zhang, P. Li, F. Li, M. J. Ren, X. L. Chen, M. Yuan and P. J. Wang, Sci. Rep., 2016, 6, 20342.
31. H. Y. Guo, Y. Zhao, N. Lu, E. Kan, X. C. Zeng, X. J. Wu and J. L. Yang, J. Phys. Chem. C, 2012, 116, 11336–11342.
32. Q. Yue, S. Chang, S. Qin and J. Li, Phys. Lett. A, 2013, 377, 1362–1367.
33. H. L. Zheng, J. M. Zhang, B. S. Yang, X. B. Du and Y. Yan, Phys. Chem. Chem. Phys., 2015, 17, 16341.
34. D. Çakır, H. Sahin and F. M. Peeters, Phys. Chem. Chem. Phys., 2014, 16, 16771–16779.
35. B. S. Yang, H. L. Zheng, R. L. Han, X. B. Du and Y. Yan, RSC Adv., 2014, 4, 54335–54343.
36. O. V. Yazyev and L. Helm, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 75, 125408.
37. Y. Zhou, Q. Su, Z. Wang, H. Deng and X. Zu, Phys. Chem. Chem. Phys., 2013, 15, 18464–18470.
38. J. M. Zhang, H. L. Zheng, R. L. Han, X. B. Du and Y. Yan, J. Alloys Compd., 2015, 647, 75–81.
39. C. Zhao, Z. Xu, H. Wang, J. Wei, W. Wang, X. Bai and E. Wang, Adv. Funct. Mater., 2014, 24, 5985–5992.
40. R.-F. Liu and C. Cheng, Phys. Rev. B: Condens. Matter Mater. Phys., 2007, 76, 014405.
41. B. Ouyang and J. Song, Appl. Phys. Lett., 2013, 103, 102401.
42. H. L. Zheng, B. S. Yang, D. Wang, R. L. Han, X. B. Du and Y. Yan, Appl. Phys. Lett., 2014, 104, 132403.
43. Y. Ma, Y. Dai, M. Guo, C. Niu, J. Lu and B. Huang, Phys. Chem. Chem. Phys., 2011, 13, 15546–15553.
44. Y. Wang, C. Zhang, W. Ji and P. Wang, Appl. Phys. Express, 2015, 8, 065202.
45. H. Chen, Y. Li, L. Huang and J. Li, J. Phys. Chem. C, 2015, 119, 29148–29156.
46. J. A. Rodríguez-Manzo, O. Cretu and F. Banhart, ACS Nano, 2010, 4, 3422–3428.
47. X. Wei, M. S. Wang, Y. Bando and D. Golberg, ACS Nano, 2011, 5, 2916–2922.
48. Y. Tan, J. Wang, Y. Zhu, A. Zuo and K. Jin, Acta Phys. Sin., 2015, 63, 207301.
49. S. Zhang, Y. Hu, Z. Hu, B. Cai and H. Zeng, Appl. Phys. Lett., 2015, 107, 022102.
50. G. Kresse and J. Hafner, Phys. Rev. B: Condens. Matter Mater. Phys., 1993, 47, 558; G. Kresse and J. Hafner, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 49, 14251.
51. G. Kresse and J. Furthmüller, Comput. Mater. Sci., 1996, 6, 15; G. Kresse and J. Furthmüller, Phys. Rev. B: Condens. Matter Mater. Phys., 1996, 54, 11169.
52. J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77, 3865.
53. P. E. Blöchl, Phys. Rev. B: Condens. Matter Mater. Phys., 1994, 50, 17953.
54. F. Banhart, J. Kotakoski and A. V. Krasheninnikov, ACS Nano, 2011, 5, 26.
55. J. Gao, J. Zhang, H. Liu, Q. Zhang and J. Zhao, Nanoscale, 2013, 5, 9785.
56. L. Shen, R. Q. Wu, H. Pan, G. W. Peng, M. Yang, Z. D. Sha and Y. P. Feng, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 78, 073306.
57. L. Yu, Z. Wang, M. Guo, D. Liu, Y. Dai and B. Huang, Chem. Phys. Lett., 2010, 487, 251.
58. P. Dev and P. Zhang, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 81, 085207.

---

Footnote

† Electronic supplementary information (ESI) available: This part gives the MD movies of As₄₉ and As₄₉X (X = H, F, B, N, P, C, Si, O and S) at 300 K, respectively. See DOI: 10.1039/c6ra00032k

---