Tunable and sizable band gap in silicene by surface adsorption

Abstract

Opening a sizable band gap without degrading its high carrier mobility is as vital for silicene as for graphene to its application as a high-performance field effect transistor (FET). Our density functional theory calculations predict that a band gap is opened in silicene by single-side adsorption of alkali atom as a result of sublattice or bond symmetry breaking. The band gap size is controllable by changing the adsorption coverage, with an impressive maximum band gap up to 0.50 eV. The ab initio quantum transport simulation of a bottom-gated FET based on a sodium-covered silicene reveals a transport gap, which is consistent with the band gap, and the resulting on/off current ratio is up to 10(8). Therefore, a way is paved for silicene as the channel of a high-performance FET.

Figure 1. Structures of the Na-covered m × m silicene supercell.

(a–e) The adsorption coverage is N = 3.1%, 5.6%, 12.5%, 16.7%, and 50.0%, respectively, for m = 4, 3, 2, , and 1. The rhombi plotted in black line represent the supercells at different coverages. (f) Top and side views of an Na-covered silicene supercell over the h-BN sheet with N = 12.5%. d₀: the silicene buckling; d₁: the distance between Na and the top-surface silicene; d₂: the distance between h-BN and the bottom-surface of silicene.

Figure 2. (a) Adsorption energy, (b) Mulliken charge transferred from the adatom to silicene, (c) band gap, and (d) effective mass of electrons of the AM-covered silicene as a function of the coverage N.

Lines connecting symbols are guides to the eye.

Figure 3. Band structures of the Na-covered silicene at different coverages.

The Dirac point in (b) and (d) is folded to the Γ point due to the reduction of the first Brillouin zone. The red dashed-line in (c) stands for the band structure of the Na-covered silicene with an h-BN buffer layer. Inset of (e): Orbital of the conduction band bottom at the Γ point. The Fermi level is set to zero.

Figure 4. Kekulé structures of AM-covered silicene.

(a–b) Electrostatic potential distributions in Li and Cs-covered silicene with N = 16.7%. The rhombi plotted in black line represent the supercells. (c) Schemetic model of the AMSi₆ supercell. t₁ and t₂ are hopping parameters. Top and bottom silicene sublattices are colored yellow and blue, respectively. The numbers are labeling of the atom sites. (d) The k-space of the AMSi₆ supercell. The Brillouin zones of the pure silicene and the AMSi₆ monolayer are colored purple and green respectively. The K and K' Dirac points of silicene can reach each other via , (the unit vectors of the reciprocal lattice of the AMSi₆ monolayer) or their linear combination.

Figure 5. Schematic model of the FET based on the Na-covered silicene.

The channel is 113.8 Å long, and the electrodes are composed of semi-infinite silicene.

Figure 6. Device performance of the Na-covered (N = 50%) silicene FET.

The bias voltage is set to 0.1 V. (a) Transmission spectra with V_g = 0, −18 and −30 V, respectively. The vertical dashed-line indicates the bias window. The details in the bias window are provided in the left insets. The right insets are sketches illustrating the relative positions of E_f and the gap manipulated by V_g. (b) Transmission eigenstates (at E = E_f and k-point = (0, 1/3)) of the on-state (V_g = 0 V) and off-state (V_g = −30 V). The isovalue is 0.2 a.u. (c) Transfer characteristic in linear (left axis) and logarithmic scales (right axis).