Modifying the geometric and electronic structure of hexagonal boron nitride on Ir(111) by Cs adsorption and intercalation

Epitaxial hexagonal boron nitride on Ir(111) is significantly modified by adsorption and intercalation of alkali-metal atoms. Regarding geometry, intercalation lifts the two-dimensional layer from its substrate and reduces the characteristic corrugation imprinted by direct contact with the metal substrate. Moreover, the presence of charged species in close proximity to the hexagonal boron nitride (hBN) layer strongly shifts the electronic structure (valence bands and core levels). We used scanning tunneling microscopy, low-energy electron diffraction, x-ray photoelectron spectroscopy (XPS), and the x-ray standing wave technique to study changes in the atomic structure induced by Cs adsorption and intercalation. Depending on the preparation, the alkali-metal atoms can be found on top and underneath the hexagonal boron nitride in ordered and disordered arrangements. Adsorbed Cs does not change the morphology of hBN/Ir(111) significantly, whereas an intercalated layer of Cs decouples the two-dimensional sheet and irons out its corrugation. XPS and angle-resolved photoelectron spectroscopy reveal a shift of the electronic states to higher binding energies, which increases with increasing density of the adsorbed and intercalated Cs. In the densest phase, Cs both intercalates and adsorbs on hBN and shifts the electronic states of hexagonal boron nitride by 3.56 eV. As this shift is not sufficient to move the conduction band below the Fermi energy, the electronic band gap must be larger than 5.85 eV.

Figure 1

(a)–(c) The $\beta$ phase: (a) inverse contrast 157.5 eV LEED pattern, (b) 50 $\times$ 50 ${\mathrm{nm}}^2$ STM image (bias voltage of $-2.0$ V, tunneling current of $3.1\times {10}^{-11}\mathrm{A}$), and (c) structural model. (d)–(f) hBN/Ir(111): (d) inverse contrast 157.5 eV LEED pattern, (e) 10 $\times$ 10 ${\mathrm{nm}}^2$ STM image (bias voltage of 0.068 V, tunneling current of $1.7\times {10}^{-10}$ A), and (f) structural model. (g)–(i) The $\alpha$ phase: (g) 50 $\times$ 50 ${\mathrm{nm}}^2$ STM image (bias voltage of 3.0 V, tunneling current of $1.0\times {10}^{-10}$ A), (h) 10 $\times$ 10 ${\mathrm{nm}}^2$ STM image (bias voltage of 2.0 V, tunneling current of $5.1\times {10}^{-11}$ A), and (i) structural model.

Figure 2

(a) Normalized Cs $3d_{5/2}$ PE spectra measured on the $\alpha$ and $\beta$ phases at large ($\theta ={84}^{\circ }$, gray) and small ($\theta ={36}^{\circ }$, blue) emission angles, with 900 eV soft x rays. The inset is the sketch of the two measurement configurations at large (gray) and small (blue) emission angles. (b) Comparison of PE N $1s$ spectra measured on pristine hBN/Ir(111) (black), the $\alpha$ phase (blue), and the $\beta$ phase (red). Two components are fitted to the spectra, representing the strongly and weakly bonded areas of hBN. N $1s$ XPS spectra are measured with 575 eV soft x rays.

Figure 3

(a) XSW results for N $1s$ and B $1s$ and HBE and LBE components of Cs $3d_{5/2}$, with measured data given as circles and fitted curves illustrated by solid lines. (b) Relative position of adsorbed and intercalated Cs, hBN, and the first layer of iridium atoms in the $\beta$ phase; the sphere sizes indicate the van der Waals radius, ionic radius or nn distance, and relative height extracted from XSW results.

Figure 4

(a) ARPES results on the $\alpha$ phase. (b) PE intensity mapping around the $\mathrm{\Gamma }$ point, with the horizontal axis as the momentum along the $\mathrm{\Gamma }K$ direction, measured on (left to right) bare hBN/Ir(111), the $\alpha$ phase, and the $\beta$ phase. (c) PE intensity at the $K$ point versus binding energy measured on bare hBN/Ir(111) (black), the $\alpha$ phase (red), and the $\beta$ phase (blue). (d) PE intensity at the $\mathrm{\Gamma }$ point versus binding energy measured on bare hBN/Ir(111) (black), the $\alpha$ phase (red), and the $\beta$ phase (blue).