Table of contents

Recent developments in optical spectroscopy applied to surfaces and interfaces are discussed in this short review. It is shown that, by exploiting the underlying physics of the various techniques, submonolayer resolution can be obtained using photons as both probe and signal. Recent examples of the use of reflection difference spectroscopy, reflection anisotropy, spectroscopic ellipsometry, Raman spectroscopy and optical second-harmonic generation are described, with particular emphasis on ultra-high vacuum studies of systems which have been characterised using conventional surface probes.

The lattice dynamics and elastic moduli of diamond are reinvestigated using a method based upon an ab initio valence force field obtained for the neopentane molecule. The calculated phonon dispersion relations are in very good agreement with experiment particularly with respect to the LA, LO, TO vibrational modes. The results demonstrate the transferability of force fields from ab initio calculations on suitably chosen molecular clusters to condensed phases, and also show that a five-parameter valence force field model is sufficient to reproduce most of the features of the lattice dynamics of diamond. Excellent agreement was found between calculated and experimental bulk moduli indicating adequate modelling of the bond stretching interactions. Deviations of the transverse acoustical vibration modes at low frequencies and the remaining elastic moduli are associated with bond angle interaction terms, which may be due to the omission of a longer range interaction force constant.

Using a spherical expansion of an ab initio N₂-N₂ potential, the authors have calculated harmonic lattice vibration frequencies for solid alpha - and gamma -nitrogen, in good agreement with experiment and time-dependent Hartree calculations. Previous dynamics calculations with an (isotropic) atom-atom model fitted to the same ab initio potential yielded libron frequencies that were considerably too high. The authors conclude, therefore, that the atom-atom model does not do justice to the accurate anisotropy of intermolecular potentials found by ab initio calculations. The upward shifts of the lattice frequencies caused by the anharmonicity in the potential agree well with the results from Green function calculations based on model potentials.

An important problem in quantum transport is to understand the role of dissipative processes. In this paper the author assume a model in which phase-breaking and dissipation are caused by the interaction of electrons with a reservoir of oscillators through a delta potential. In this model the self-energy is a delta function in space, leading to a kinetic equation with a simple physical interpretation. A novel treatment of the contacts is used to introduce the external current into the kinetic equation. One specializing to linear response the author obtains an integral equation that looks like the Buttiker formula (1961) extended to a continuous distribution of probes. The author show that this equation can be reduced to the usual Buttiker formula which involves only the actual physical probes. Dissipation modifies the transmission coefficients, and the author presents explicit expressions derived from this model. Also, in a homogeneous medium the integral equation reduces to the diffusion equation, it the electrochemical potential is assumed to vary slowly. This paper serves to establish a bridge between the quantum kinetic approach which rigorously accounts for the exclusion principle and the one-particle approach which is intuitively appealing.

The authors present the results of a calculation of rho (T) for thin wires at low temperatures. The calculation includes the contributions arising from normal electron-electron scattering and from normal electron-phonon scattering, two terms that are absent for bulk samples. Applying the results to thin wires of potassium, yields quantitative agreement with the hitherto unexplained rho (T) data of Rowlands (1978), Yu (1984), Zhao and co-workers (1988).

Amorphous films of GaAs have been prepared by radio-frequency sputtering in an argon plus hydrogen plasma. The optical gap and refractive index have been determined as a function of hydrogen content. Vibrational modes in the infrared have also been identified. The temperature dependence of the DC electrical conductivity has been measured for films with and without hydrogen and as a function of annealing. Co-sputtering of Al and GaAs has been used to produce a-GaAlAs films with various concentrations of Al. The electrical and optical properties of these films have also been measured. An interesting result is that the optical gap decreases with increasing Al content, in contrast to c-GaAlAs for which the gap increases. The authors attribute this result to the presence of 'wrong bonds'.

¹⁸¹Ta spin-relaxation measurements have been carried out in hydrides of the intermetallic glass Zr₂Ni((H)/(M)=0.83 and 1.5) using the perturbed angular correlation (PAC) technique. The absence of a broad distribution in the activation energy of the hydrogen hopping rate is confirmed. Moreover, the jump rate is found to be independent of H concentration. The results derived from PAC experiments suggest the existence of hydrogen-induced demixing in the amorphous hydrided alloy, with the H atoms located in tetrahedral Zr₄ sites.

The pressure dependences of the Raman-active graphite layer shearing modes in highly oriented pyrolytic graphite (HOPG) and FeCl₃-HOPG acceptor intercalation compounds for stage indices n=1, 2 and 3 have been measured in diamond anvil cell by use of a Spex-1403 Ramalog system. The results of high-pressure Raman spectra show that both the interior and neighbouring modes exhibit a frequency shift upwards with increasing pressure, which has been attributed to an in-plane lattice contraction. In stage-3 FeCl₃-HOPG, the pressure-induced staging transition from stage 3 to stage 4 has been found from the change in intensity ratio of the interior to neighbouring modes at 8.39 GPa. This finding confirmed firstly that the pressure-induced staging transitions are not just confined to the alkali-metal graphite donor intercalations but are valid for the acceptor graphite intercalation compounds.

Cathodoluminescence from undoped and Ge-doped high-purity synthetic quartz is examined over the temperature range 40 to 300 K. The undoped quartz exhibits the well known blue ( approximately 470 nm) luminescence attributed to the decay of self-trapped excitons. Ge-doped quartz shows green ( approximately 570 nm) luminescence at low temperatures; it is proposed that this emission occurs as a result of exciton self-trapping occurring at Ge substitutional sites. Line features which are considered as resulting from molecular oxygen ions are observed in both undoped and doped material. Evidence is presented that emission at 380 nm relates to an impurity incorporated during growth.

The angle-resolved secondary-electron emission (ARSEE) spectra of highly oriented pyrolytic graphite have been measured in the 7-70 eV electron kinetic energy range. A peak observed at approximately=59 eV with K/sub ///=1.35-1.65 AA^-1 was tentatively explained within a semiempirical approach based on a free electron parabola with an inner potential of -16.2 eV (relative to the vacuum level) which crosses the M point at 59 eV. From the crossing of the free electron parabola with the zone centre peaks at 16 eV and 28.5 eV were independently described. These results suggest that the empty density of states for large kinetic energies are well described by a free electron parabola. In the energy range below 30 eV good agreement was obtained with previously published ARSEE experiments. XPS and AES measurements were performed to verify that the 59 eV peak did no originate from an impurity level. After argon ion bombardment the fine structure observed in the ARSEE spectra of HOPG was completely removed.

The equation of state and the electronic bandstructure of ice have been calculated in the recently proposed high-pressure phase XI (anti-fluorite structure) by using the local-density approximation with the augmented-plane-wave method. The calculated pressure of metallization 1.76 TPa is in agreement with the value of 1.0-1.5 TPa conjectured by Besson (1986), but it is 2.5 times higher than that estimated by the classical Herzfeld theory (1927) and Thomas-Fermi-Dirac model. It has been also shown that the ionic model proposed for the ice XI phase is not adequate even for ultra-high pressure from the viewpoint of the bandstructure calculation.