The effect of tungsten trioxide thin films at ferroelectric-electrode boundaries on fatigue behaviour

A conventional thin film capacitor heterostructure, consisting of sol-gel deposited lead zirconium titanate (PZT) layers with sputtered platinum top and bottom electrodes, was subjected to fatiguing pulses at a variety of frequencies. The fatigue characteristics were compared to those of a similarly processed capacitor in which a ~20nm tungsten trioxide layer had been deposited, using pulsed laser deposition, between the ferroelectric and upper electrode. The expectation was that, because of its ability to accommodate considerable oxygen non-stoichiometry, tungsten trioxide (WO3) might act as an efficient sink for any oxygen vacancies flushed to the electrode-ferroelectric boundary layer during repetitive switching, and hence would improve the fatigue characteristics of the thin film capacitor. However, it was found that, in general, the addition of tungsten trioxide actually increases the rate of fatigue. It appears that any potential benefit from the WO3, in terms of absorbing oxygen vacancies, is far outweighed by it causing dramatically increased charge injection in the system.

Energy levels and radiative rates for transitions in Mg-like iron, cobalt and nickel

Energy levels and radiative rates for electric dipole (E1) transitions among the lowest 141 levels of the (IS2 2s(2) 2P(6)) 3l(2) , 3l3l', and 3l4l configurations of Fe XV, Co XVI, and Ni XVII are calculated through the CIV3 code using extensive configuration-interact ion (CI) wavefunctions. The important relativistic effects are included through the Breit-Pauli approximation. In order to keep the calculated energy splittings close to the experimental values, we have made small adjustments to the diagonal elements of the Hamiltonian matrices. The energy levels, including their orderings, are in excellent agreement with the available experimental results for all three ions. However, experimental energies are only available for a few levels. Since mixing among some levels is found to be very strong, it becomes difficult to identify these uniquely. Additionally, some discrepancies with other theoretical work (particularly for Ni XVII) are very large. Therefore, in order to confirm the level ordering as well as to assess the accuracy of energy levels and radiative rates, we have performed two other independent calculations using the GRASP and FAC codes. These codes are fully relativistic, but the CI in the calculations is limited to the basic (minimum) configurations only. This enables us to assess the importance of including elaborate Cl for moderately charged ions. Additionally, we report results for electric quadrupole (E2), magnetic dipole (MI), and magnetic quadrupole (M2) transitions, and list lifetimes for all levels. Comparisons are made with other available experimental and theoretical results, and the accuracy of the present results is assessed. (c) 2007 Elsevier Inc. All rights reserved.

A quantitative determination of reaction mechanisms from density functional theory calculations: Fischer-Tropsch synthesis on flat and stepped cobalt surfaces

We systematically investigated the mechanism of the C-1 + C-1 coupling reactions using density functional theory. The activation energies of C-1 + C-1 coupling and carbon hydrogenation reactions on both flat and stepped surfaces were calculated and analyzed. Moreover, the coverages of adsorbed C-1 species were estimated, and the reaction rates of all possible C-1 + C-1 coupling pathways were quantitatively evaluated. The results suggest that the reactions of CH2 + CH2 and CH3 + C at steps are most likely to be the key C-1 + C-1 coupling steps in FT synthesis on Co catalysts. The reactions of C-2 + C-1 and C-3 + C-1 coupling also were studied; the results demonstrate that in addition to the pathways of RCH + CH2 and RCH2 + C, the coupling of RC + C and RC + CH also may contribute to the chain growth after C-1. (C) 2008 Elsevier Inc. All rights reserved.