No abstract

No abstract

No abstract

Upon sputtering and subsequent annealing, the subsurface region of the ͑100͒ surface of FeAl undergoes a transition to a practically stable and ordered Fe 3 Al slab. It is capped by an Al top layer forming a rather complex interface to the bulk FeAl. This shows that routine surface preparation procedures with sputtering involved can lead to substantial restructuring extending deep into the surface. ͓S0163-1829͑96͒51732-7͔In substitutionally disordered binary alloys the phenomenon of segregation of one of the constituents to the surface usually leads to a layer-dependent deviation from the average bulk stoichiometry. 1,2 For chemically ordered alloys, however, this situation is rather rare and the corresponding surfaces in most cases exhibit bulk termination. Different elemental terminations of a surface can coexist in the case of chemically alternating layers. Exceptions result when the sample is prepared by ion bombardment with preferential sputtering of one of the constituents, in most cases depletion of the lighter elements. Subsequent annealing is generally believed to continuously restore the bulk stoichiometry as a consequence of diffusion, whereby nonstoichiometric phases may appear as intermediate structures. Frequently, the latter are substitutionally disordered or deviate from the ideal bulk in the top layer only. In the present paper, we report thatfor a certain annealing temperature-the surface can also exhibit an ordered structure that extends deep into the subsurface region forming an interface to the bulk. The formation of metastable epitaxial films different from the bulk both in structure and stoichiometry is a well-known phenomenon when deposited material is thermally made to react. A prominent example is the evolution of silicide films on a silicon substrate after thermal processing of a metallic deposit. However, to our knowledge there is no report, nor has anyone even seemed to take into account yet, that such a process can take place simply upon moderate sputtering of an ordered compound surface.In particular, we show that for the ͑100͒ surface of an ordered FeAl crystal an ordered and practically stable Fe 3 Al phase develops. It is capped by a pure Al top layer, which might be interpreted either as the usual termination of the Fe 3 Al͑100͒ surface or as a special feature of the interface to bulk FeAl. In any case the Al capping or termination confirms the trend for Al segregation as already found earlier for low index FeAl surfaces. 3 These findings-which are possibly not unique but, to our knowledge, up to now unobserved-are certainly relevant from the fundamental point of view and should be valuable information for scientists working in the field of multicomponent surfaces.We present a surface structure determination by quantitative low-energy electron diffraction ͑LEED͒ of a FeAl͑100͒ crystal surface prepared as described below. The sample's bulk stoichiometry was Fe 0.53 Al 0.47 with bulk impurities of 50 ppm oxygen and carbon. Using x-ray diffraction the lattice constant was determine...

The interaction of Si, Ti, and Mo atoms with pyrolytic graphite substrates has been studied for evaporated layers of about 100 nm and implanted ions with mean ranges between 2 and 4 nm. In the temperature range from room temperature to 1800 K the thermal diffusion of carbon into the evaporated layers has been studied by Rutherford backscattering spectroscopy while the temperature dependence of the carbide formation has been studied by x-ray photoelectron spectroscopy. For all three systems stable carbidic phases are predicted by equilibrium phase diagrams. For Ti the formation of TiC is already observed after room temperature implantation, while for Mo annealing to 1200 K is necessary for Mo2C formation. In the case of Si oxygen contamination due to the air transfer after implantation resulted in a mixed SiOxCy phase which only transformed into a SiC phase at temperatures above 900 K, where the oxygen was released. The temperature range of stability of the carbidic layers was found to be correlated to the melting temperature of the metal–carbide eutectic. Above this temperature the metal atoms rapidly dissolve in the graphite lattice.