Epitaxial growth of a single-domain hexagonal boron nitride monolayer

Following the rise of graphene, the two-dimensional (2D) material composed by carbon atoms arranged in a honeycomb lattice, the scientific community is dedicating increasing interest in other 2D atomic crystals, hexagonal boron nitride (h-BN) in particular. While isostructural to graphene, h-BN is a wide band gap electrical insulator which provides a superior insulating platform to build high-performance low-dimensional devices. One of the most promising methods to produce 2D crystals is by chemical vapor deposition (CVD) of the proper precursor onto a metal substrate. However, the hetero-elemental nature of h-BN renders the CVD growth of uniform layers a much more demanding task compared, e.g., to the case of graphene. In particular, it is still difficult to obtain large single crystalline h-BN domains because of the formation of rotated phases that give rise to grain boundaries and other 1D defects. We tackled this problem by tuning the growth conditions of h-BN on Ir(111) to enhance the quality of the resulting h-BN monolayer. For characterization, we performed High Resolution X-ray Photoelectron Spectroscopy (HR-XPS) and Photoelectron Diffraction (XPD) measurements at the SuperESCA beamline, and Spot Profile Analysis-LEED (SPA-LEED) experiments at the Surface Science Laboratory of Elettra.

We followed two different procedures to grow h-BN on Ir(111) by CVD using borazine (B3N3H6) as a precursor. The first method (High Temperature Growth (HTG)) consisted of borazine exposure with the sample at 1070 K. The second approach (Temperature-Programmed Growth (TPG)), instead, consisted of repeated cycles of borazine exposure at room temperature followed by annealing to 1270 K. Both approaches ensure the growth of an extended h-BN single layer that does not leave bare Ir regions, as proved by the measurements of the B 1s, N 1s, and Ir 4f7/2 spectra which did not change over the sample surface or by prolonging the growth time.

The B 1s and N 1s core level spectra shown in Fig.1a and b, respectively are similar for the HTG and TPG growth. Both spectra present two components which are due to the presence of a moiré periodic corrugation with h-BN regions differently interacting with the Ir substrate: B0, N0 represent weak interaction while B1, N1 are due to the strongly interacting regions. Major differences between HTG and TPG growth methods were found for the XPD measurements (Fig.1c-f). The diffraction patterns of the HTG preparation display a 6-fold symmetry consistent with the formation of antiparallel fcc and hcp domains (Fig.1g and h). The TPG procedure, instead, gives rise to 3-fold symmetric XPD patterns, indicating the prevalence of the fcc orientation. To determine the orientation of the h-BN layer with respect to the Ir(111) surface for the TPG growth, we exploited the 3-fold symmetry of the substrate and performed XPD measurements also of the Ir 4f7/2 core level. The excellent agreement between the experimental patterns and the multiple scattering simulations of the photoemission intensity, displayed in Fig. 1, further supports the interpretation of the results. The SPA-LEED measurements indicate that the h-BN layer produced with the TPG method displays an average domain size larger than 900 Å.

Figure 1. B 1s (a) and N 1s (b) core level spectra measured at 284 eV and 500 eV photon energy, respectively, together with the spectral contributions resulting from peak-fit analysis. (c-f) Stereographic projections of the integrated photoemission intensity modulations measured at 115 eV kinetic energy, as a function of emission angle for B 1s (green) and N 1s (blue) core level. The colored sectors are the experimental data, while the gray regions are the multiple-scattering calculations for a flat, free-standing h-BN layer performed with the EDAC code. fcc (g)  and hcp (h) adsorption configurations for h-BN on Ir(111). The drawing on the left represents a schematic of the h-BN lattice and the XPD experimental geometry. 

These findings lead us to conclude that the orientation of the h-BN layer is controlled by the growth temperature, which in turn results in a different interplay between the thermal energy and the binding energy difference between fcc and hcp seeds during the first stages of the h-BN growth. We expect that this method could be further exploited on other transition metal surfaces in order to control the h-BN orientation, thus opening new perspectives for the production of high-quality h-BN films. 


This research was conducted by the following research team:

Fabrizio Orlando1, Paolo Lacovig2, Luca Omiciuolo1, Nicoleta G. Apostol2,3, Rosanna Larciprete4, Alessandro Baraldi1,2,5 and Silvano Lizzit2
  
1 University of Trieste, Trieste, Italy.
2 Elettra Sincrotrone Trieste, Trieste, Italy.
3 National Institute of Materials Physics, Magurele-Ilfov, Romania.
4 CNR, Institute for Complex Systems, Roma, Italy.
5 IOM-CNR, Laboratorio TASC, Trieste, Italy
 

Contact persons:
Silvano Lizzit:

Reference

F. Orlando, P. Lacovig, L. Omiciuolo, N.G. Apostol, R. Larciprete, A. Baraldi and S. Lizzit, “Epitaxial Growth of a Single-Domain Hexagonal Boron Nitride Monolayer”, ACS Nano 8, 12063 (2014), doi:10.1021/nn5058968.

 

Last Updated on Wednesday, 15 April 2015 18:46