Publications

An analysis is given of the way in which thickness fluctuations in a multilayer can affect the grazing incidence X-ray reflectivity pattern. It is shown that peak shifts, broadening and height variations can result. They alter in a predictable way the parameter values derived from a fit with an ideal lattice. A principle for designed Bragg reflectors is also proposed.

The effect of intermediate grounding and sintering environment (air or sintered powder) on the superconducting characteristics of Bi(Pb)-Sr-Ca-Cu-O hulk ceramics obtained by spray-frozen freeze drying technique was investigated by X-ray diffraction, electric and magnetic measurements. Intermediate grounding leads to higher scattering coefficients of the superconducting characteristics. This result does not depend on sintering environment. Improved results are obtained when sintering is performed by immersing pellets "in powder" rather than placing them "on powder". The powder used as sintering environment was pyrolized precursor powder calcinated in air for 100 h. (C) 1998 Elsevier Science Ltd. All rights reserved.

Modeling of the I(V, T) dependence in thin p-n semiconductor structures shows that various behaviors in its shape factor can be correlated with and explained through the voltage dependence of the carrier concentration product np and its components, at the zero field injection-point, for various conditions controlled by thickness and doping. A ''second exponential'' high injection range exp (qV/nkT), with a shape factor n as high as four appears to be a consequence of the departure from quasineutrality due to significant minority carrier space charge, favored by low doping and thickness. In thin structures, the presence of a second exponential range is associated with the presence of a quadratic range (I similar to V-2) at very high current densities.

Modeling of the I(V,T) dependence in thin film semiconductor structures suggests that experimental data, on temperature and voltage dependent activation energy and shape factor of the current, give significant information on the recombination process within the structure. Grounds are thus laid for further investigation on quantum well effects within such structures. (C) 1996 American Institute of Physics.

An anomalous high signal was observed across lead sulphide-ferroelectric ceramic heterostructures illuminated at room temperature with light having wavelengths between 0.8 and 3.1 mu m. The frequency characteristic of this signal is similar to a high-pass filter and depends on the value of the load resistance in series with the lead sulphide thin film, This signal is only two to three times smaller than the usual photoconductive signal generated by the lead sulphide thin film. Its nature is not pyroelectric because of its frequency dependence. The interface charge released from the PbS-ferroelectric interface during polarization reversal was integrated and a hysteresis loop was obtained. Electrons and hole concentrations were evaluated from this curve. The values obtained are: 4.4 x 10(16) cm(-3) for holes and 3.4 x 10(16) cm(-3) for electrons.

The paraconductivity of Li-doped YBa2Cu3O7-x was measured. We have found that the character of the fluctuation changes as lithium content increases: instead of a crossover from 2D to 3D behavior (D is the dimensionality), as we have observed for a low doping level, a double crossover to the 2D percolative and 3D percolative regimes becomes manifest by increasing the lithium content.

Starting from the microscopic current associated with the single-electron wave functions and using Wannier functions suitably generalized for nonperiodic heterostructures we construct mesoscopic (either discrete or continuous) conserved currents associated with states having components only in the considered band. This provides a rigorous basis to the approximate envelope function approaches. As an application we analyze general one-dimensional heterojunctions, mapping explicitly the microscopic asymptotic states onto mesoscopic (one-band envelope) ones. This proves that, apart from the effective masses and the band offsets, the connection rules are characterized in general by three parameters, the values of which are unconstrained, confirming the results of phenomenological analyses.

PbS/PbTiO3/Si heterostructures were prepared by successive depositions of a PbTiO3 thin film using the sol-gel method and of a PbS thin film using the coprecipitation method on a bulk silicon wafer. The current-voltage characteristic of the heterostructure is of a diode type. The heterostructure is sensitive over a broad range of wavelengths, from 0.35 to 3 mu m, when illuminated with continuous light or with modulated Light. Three sensitivity peaks are observed, at 2.7, 0.95, and 0.38 mu m. The measured signals are attributed to a photovoltaic effect which appears in the studied heterostructure.

To model high injection phenomena in P-N junction devices it is usually necessary to use numerical analysis. This is because the standard procedure of dividing the structure into neutral zones and depletion layer (regional approximation) fails. In a recent paper, Yue et al. [J. Appl. Phys. 77, 1611 (1995)] proposed an extension of the Shockley theory, retaining the form of the conventional diffusion current-only model for the conduction mechanism. Their solution was based on the resolution of the ambipolar transport equation in the base. It is shown here by a numerical simulation of the complete structure, within the framework of the drift-diffusion model, that both the exponential current dependence: J alpha exp(eVa/2kT) as well as the Yue et al. approximation are valid only in the limiting case of a strongly extrinsic short base diode. (C) 1996 American Institute of Physics.

The dependence of the carrier concentrations, of the resistivity and of the Hall coefficient of irradiated silicon on the neutron fluence has been investigated, starting from the supposition that the main phenomena induced by irradiation in the semiconductor bulk are shallow-donor removal and deep-centres creation. The free parameters of the model are the initial doping of the starting material, the permitted energy level values of the radiation-induced centres in the semiconductor band gap and their introduction rates. The influence of each parameter on the calculated dependences is studied in detail, for three cases: one deep acceptor-like centre, two deep accepters and one deep acceptor plus one deep donor-like centre. Each of the three cases is discussed in correspondence with different experimental results.