Schematic of the capacitively coupled plasma source, including pumps, gas supply, and power lines. The sample is inserted directly into the phase electrode so that the sample surface is homogeneously exposed to plasma. From the plasma source the sample can be directly transferred into a UHV chamber with TPD, LEED, and XPS.
(a) XP spectra of the carbon 1s level of the (i) untreated and (ii) plasma-treated sample. The untreated sample exhibits a large amount of carbon on the surface along with carbonate formed from reaction with atmospheric CO2. After exposing the surface to oxygen plasma the carbon content is below the XPS detection limit. (b) LEED image recorded after oxygen plasma treatment. Clearly a sharp 1×1 diffraction pattern of the PrO2(111) surface is observable which was impossible to achieve by other cleaning methods.
XP spectra of the Pr 3d core level: (a) untreated, (b) after oxygen plasma treatment (dashed lines represent the single contributions obtained by curve fitting) and after annealing under UHV conditions to (c) 405, (d) 580, (e) 680, and (f) 920 K, respectively. The labels x, y, and z denote the assignment to 3d 9 4f 3, 3d 9 4f 2, and 3d 9 4f 1 final state configurations, respectively.
(a) TPD spectrum of a Pr-oxide film after oxygen plasma treatment depicting the QMS-signal for molecular oxygen, carbon monoxide, and carbon dioxide. No desorbing carbon species, expected to form from residual impurities, is observed to be consistent with the XPS results. Desorption of O2 is observed at several temperatures and with sharp signals most likely corresponding to the transition from one oxide phase to the other. (b) LEED image recorded at 32 eV after heating a plasma-treated film for 30 min to 950 K. The diffraction pattern represents the cub-Pr2O3(111) surface. (c) Schematic view onto the cub-Pr2O3(111) plane showing atom positions (cut through bulk crystal). Highlighted areas mark identical structure units. Red and blue diamonds show the unit cells of cub-Pr2O3(111) and unreconstructed Si(111), respectively.
Si(LLL) XRD measurements close to (a) the Si(111) Bragg peak and (b) the Si(222) Bragg peak. (a) Fringes due to oxide and silicate are clearly visible while the praseodymia Bragg peaks are veiled from the wings of the strong Si(111) Bragg peak. The thickness of the praseodymia film and the silicate interface layer can be estimated from ΔL oxide film/d Si and ΔL silicate/d Si, respectively. (b) Praseodymia (222) Bragg peaks and oxide film fringes are well resolved while the kinematically forbidden Si(222) Bragg peak at L = 2 is very weak.
XRD analysis of the (222) Bragg peaks due to cubic praseodymia films: (a) untreated sample, (b) plasma-treated sample, and (c) after storage under ambient conditions for two weeks. The solid lines are Gaussian fitted to the Bragg peaks of Pr6O11(222), Si(222), PrO2−Δ(222), and PrO2(222). For comparison, Bragg peak positions expected from bulk data reported in literature are presented, too.
Structural data of the praseodymia films obtained from XRD. The fraction of the PrO2 and Pr6O11 phases are estimated from the ratio of the corresponding Bragg peak intensities neglecting effects due to the structure factor. The structure factors of both PrO2 and Pr6O11, however, do not differ drastically since the additional oxygen content of PrO2 (compared to Pr6O11) attributes only slightly to the structure form factor due to the small atomic form factor of O compared to Pr.
Article metrics loading...
Full text loading...