The growth of hBN in small crystallites with the hexagonal planes oriented normal to the substrate interface was unexpected since one expects thin films to grow in a close-packed orientation to minimize the number of undercoordinated atoms. However, this result is less surprising if one considers that the conditions used to prepare this sample are only slightly different from those used to grow the metastable cubic phase.[3] Recent work has shown that small variations in substrate temperature, laser fluence, ion fluence, and gas pressure will produce BN thin films of hexagonal, cubic, or incoherent structure.[6] McKenzie has suggested that in-plane compressive stress due to ion bombardment during growth may orient the highly-compressible c-axis parallel to the surface.[10] Further work will be necessary to elucidate the relationship between substrate temperature, stress, and ion bombardment in order to test this hypothesis.
NEXAFS has been measured on the cubic, hexagonal and incoherent forms of boron nitride. B 1s and N1s absorption edge data measured from the hBN phase sample clearly shows both pi* and sigma* features, revealing the expected sp2 bonding, while the cBN sample shows only the sigma* manifold, indicative of sp3 bonding. Boron and nitrogen 1s photoabsorption was also measured from a thin film of iBN on Si. Direct comparison to the hBN and cBN NEXAFS spectra clearly indicated sp2 bonding in the iBN film because of the prominent pi* absorption feature. The angular dependence of the integrated intensity of the B 1s pi* absorption for the iBN film shows microcrystalline hexagonal layer planes which are oriented normal to the film surface, contrary to the usual growth mode of hBN. The observation of a preferred bond order in an apparently disordered film demonstrates the power of the NEXAFS technique to resolve structural features in microcrystalline samples.
The authors would like to thank C.A. Taylor II for help with film preparation and G.D. Waddill and R.H. Howell for helpful discussions about the data analysis. This work was conducted under the auspices of the US Department of Energy by the Lawrence Livermore National Laboratory under contract No. W-7405-ENG-48, and was performed at the National Synchrotron Light Source, Brookhaven National Laboratory and at Stanford Synchrotron Radiation Laboratory, which are both supported by the Department of Energy, Office of Basic Energy Sciences.