The diffusion coefficient and solubilities of several metals (Ag, Cu, Au, Pd, and Ti) have been investigated by bias temperature stress (BTS) of normalmetal/SiO2/normalSi (MOS) structures using the candidate metal as the electrode. Temperatures ranged from 250° to 600°C, and the electric fields ranged from +106 to −106 V/cm in ambients of vacuum (10−6 torr), nitrogen false(N2false) , and forming gas false(N2H2false) . In dry N2 false(<3 normalppm H2Ofalse) , the activation energy for silver diffusion is found to be 1.24 eV in the temperature range of 275°–365°C. The diffusion coefficient for silver at 300°C in silicon dioxide is 4.5×10−13 cm2/normals . Copper diffusion has an activation energy of 1.82 eV in the temperature range of 350°–450°C in a forming gas environment false(<3 normalppm H2Ofalse) and a diffusion coefficient in SiO2 at 450°C of 1.2×10−11 cm2/normals . A thermodynamic model to predict the activation energy of the solid solubility of these metals in SiO2 and an interstitial diffusion model, that includes both strain and electrostatic energies, which predicts the diffusion activation energy, have been developed. Diffusion coefficients are estimated from a closed form solution of the diffusion equation and the observed behavior of the metal in SiO2 .

This paper discusses the driving forces behind the continued strength of crystalline silicon technology. The history of silicon technology development is reviewed, and projections made as where to silicon technology is likely to go in the following 10 years. Next the barriers that have inhibited the emergence of competing technologies are discussed, along with the steps that need to be taken to surmount those barriers.

This paper addresses the issue of why concentrator systems have not gained a significant market share. The history of concentrator development is reviewed, and the status of existing concentrator efforts outlined. A critical look at the requirements to propel concentrators to a prominent market role in large‐scale power production is presented. Various concentrator and flat‐plate PV system approaches are compared by computing the expected cost of energy, and conclusions are drawn as to what the best course of action will be. Concentrator systems are projected to be the lowest‐cost, lowest‐risk PV option for medium and large PV power plants. Copyright © 2000 John Wiley & Sons, Ltd.

No abstract

For maximal performance solar cells should resemble semiconductor lasers, i.e., they should be constructed in the form of a double heterostructure. We have found rather good performance in SIPOS-crystalline silicon-SIPOS double heterostructure solar cells, where SIPOS≡SiOx. The processing of these solar cells gives insights into the truly outstanding performance of the n+-SIPOS: p-Si heterojunction which has a forward saturation current coefficient J0=10−14 A/cm2, or equivalently an ‘‘emitter Gummel number’’ Ge=3.3×1015 s/cm4. This suggests that crystalline silicon solar cells can be much more efficient than had been suspected.