Evidence of a superdense nonmagnetic co layer at the top of a thin co film on Si(111): A polarized neutron reflectivity study

A thin cobalt film (25 nm) was deposited on a HF-etched Si(111) substrate in high vacuum by ebeam evaporation. Upon exposure to air the film surface got oxidized. We have carried out polarized neutron reflectivity (PNR) experiment on this cobalt film using the experimental set up in DHRUVA, Bhabha Atomic Research Centre. Fig.1 shows the PNR results. The topmost layer (~2 nm) is cobalt oxide. Interestingly, the PNR data indicate the presence of a high density nonmagnetic material at the oxide/Co interface. The Co layer below this high density layer has magnetic moment as expected for Co. X-ray reflectivity (XRR) experiment (Fig.2) confirms the high density of this nonmagnetic layer, which is ~4 nm thick. The density of this layer is ~1.4 times that of underlying normal Co. This nonmagnetic layer cannot be a cobalt oxide layer, as the density of an oxide layer would be even smaller compared to normal Co. Rutherford backscattering spectrometry experiment has ruled out the presence of any element of higher mass than Co. XRD results have shown, in addition to the main Co peak, a tiny peak corresponding to ~13% smaller planar spacing. This is consistent with ~ 1.4 times density enhancement. All these results confirm that this high density layer is a Co layer with a density of about 1.4 times that of normal Co. To our knowledge, there is no experimental report on such a high density Co at ambient pressure. A DFT calculation predicts a nonmagnetic fcc Co phase with a density of about 1.4 times the normal Co density. Our observed high density layer is likely to be such a superdense nonmagnetic Co layer. How could such a superdense Co layer form? Our Co film is grainy. During the growth of a grainy film, atoms could diffuse into grain boundaries producing a compressive stress, which may be the reason for the formation of the superdense Co layer.