The silicon-cerium oxide interface is studied using x-ray photoelectron spectroscopy. The oxidation and reduction of species at the interface are examined as a function of annealing temperature both in vacuum and oxygen ambient, in order to determine their relative stabilities. By depositing a very thin CeO 2 film ͑ϳ30 Å͒, the cerium and silicon core level peaks can be monitored simultaneously. The presence of characteristic chemical shifts of the Si 2p peak gives information about any SiO x layer that may form at the interface. The oxidation state of the cerium can be probed from three different areas of the spectrum. From this information we can infer the oxidation state of both the silicon and the cerium. For the first time a complete picture of the interface is obtained. The implications of these findings on the utility of CeO 2 in device applications are discussed.

No abstract

An electrostatic force microscope was used to write and image localized dots of charge in a double barrier CeO 2 /Si/CeO 2 /Si͑111͒ structure. By applying a relatively large tip voltage and reducing the tip to sample separation to 3-5 nm, charge dots 60-200 nm full width at half maximum of both positive and negative charge have been written. The total stored charge is found to be Q ϭϮ(20-200)e per charge dot. These dots of charge are shown to be stable over periods of time greater than 24 h, with an initial charge decay time constant of ϳ9.5 h followed by a period of much slower decay with Ͼ24 h. The dependence of dot size and total stored charge on various writing parameters such as tip writing bias, tip to sample separation, and write time is examined. © 1999 American Institute of Physics. ͓S0003-6951͑99͒04035-8͔ Cerium oxide (CeO 2 ) is an insulating material with a lattice mismatch of only 0.35% to silicon ͑Si͒ and an energy band gap of ϳ5.5 eV. This attractive set of properties has the potential to lead to a fully functional silicon heterojunction technology. A significant amount of work has been done examining the growth and characterization of CeO 2 crystals on Si, 1-5 and the growth of single crystal Si on to CeO 2 /Si heterostructures 6 has been recently reported. Based on these promising results, a silicon resonant tunneling diode, an improved silicon-on-insulator technology, and stacked silicon electronics have all been proposed. A valuable and interesting addition to this array of technologies would be the capacity for integrated electrostatic data storage.In this letter, we describe the localized charging and subsequent imaging of a double barrier CeO 2 /Si/CeO 2 /Si͑111͒ structure by electrostatic force microscopy ͑EFM͒.7-9 The controllable writing of both positive and negative localized dots of charge with long lifetimes is described and it is further shown that these charge dots may be rewritten and replaced by charge of the opposite sign through the application of an opposite charging bias. A simple analysis is presented to quantify the total amount of charge stored in each charge dot. The time evolution of these charge dots is studied, and charge decay time constants are extracted. Finally, a study is presented of various writing parameters such as tip bias, tip to sample separation, and write time on the size and total stored charge of the resultant charge dots.Samples were produced from commercially available 3 in. Si͑111͒ wafers, n-type with 3.0-4.3 ⍀ cm resistivity. After being subjected to a standard acetone, isopropyl alcohol, deionized water degrease in ultrasound, the wafer was etched in 50:1 HF solution until hydrophobic, rinsed in deionized water, and immediately introduced into vacuum. Electron beam evaporation was used to deposit material from an undoped Si charge and a 99.99% CeO 2 charge to grow the structures. Initially, a 200 Å Si buffer layer was grown and examined by ͑RHEED͒ reflection high-energy electron diffraction to assure the characteristic (7ϫ7) reconstruction was appar...

The anneal behavior of layers implanted with dopants from column III (B, Al, Ga, and Tl) and column V (As, Sb, and Bi) in silicon substrates has been investigated. The ranges of implant conditions were energy 20–50 keV, dose 1013–1015/cm2, and substrate temperature 23°–500°C. Hall-effect and sheet resistivity measurements were used to determine the effective number of carriers/cm2 (Ns)eff and the effective mobility μeff. Analysis of nonuniform distributions of carrier densities and mobilities on these measurements shows that the values of (Ns)eff and μeff can be misleading unless the effect of the depth distributions is allowed for. These distributions have been determined in some cases by the use of layer removal techniques combined with Hall-effect and sheet resistivity measurements. We find in well-annealed implanted samples that the dependence of the mobility on carrier density follows that determined for bulk silicon. In many cases deviation from this relation can be accounted for on the basis of compensation. In the case of aluminum we suggest that this compensation may be accounted for by the presence of interstitial aluminum atoms acting as donors. We have found that interstitial thallium can behave as a donor. The anneal behavior of the implanted layer is influenced by ion species, dose, and substrate temperature. The carrier concentration measured in implantations of column III elements did not exceed the limits of thermal equilibrium solubility as is found for column V elements. In the former case, enhanced diffusion effects are observed. From the known substitutional behavior of column V elements, it is suggested that the anneal behavior in the 600°–800°C range is due to the dissociation of radiation damage complexes.

Hall-effect and sheet-resistivity measurements have been made on silicon samples implanted with Sb, Ga, and As ions at energies between 20 and 75 keV. These measurements determine the weighted average of the number Ns of carriers/cm2 and the carrier mobility in the implanted layer. A combination of Hall measurements and layer-removal techniques was used in some cases to obtain a more accurate value of the number of carriers/cm2 and the depth dependence of the carrier concentration and mobility.For Sb implantations both temperature and dose affect the anneal characteristics. Silicon samples implanted with Sb at room temperature exhibited n-type behavior following anneal at 300 °C, with little increase in Ns up to about 550 °C anneal temperatures. A 600 °C 10-minute anneal produced an order-of-magnitude increase in Ns. This change is associated with reordering of the amorphous layer created during room-temperature implantations. This amorphous layer is not produced in implantations made at temperatures above 450 °C. In low-dose (<1014/cm2) Sb implantations at 500 °C, Ns increased by a factor of 2 to 3 during anneal to 800 °C. In high-dose (>5 × 1014/cm2) Sb implantations, the carrier concentration exceeded the limit set by thermal equilibrium solubility of Sb in silicon. Under these conditions, annealing caused a decrease in Ns toward the value associated with the solubility.Such supersaturation effects were not observed in Ga and As implantations at 500 °C. Annealing to temperatures of 800–900 °C produced a one-to-two order-of-magnitude increase in the number of carriers/cm2. In Ga implantations annealed to 800–900 °C, the number of carriers/cm2 increased approximately linearly with increasing dose and then leveled off at a value near that expected from thermal solubility.The Rutherford-scattering data in the preceding paper indicates that the difference in implantation behavior between various ion species is due to differences in the relative number of ions on substitutional sites.