Table of contents

The 1992 General Conference of the Condensed Matter Division of the European Physical Society held in Praha between April 6 and 9, 1992 was the 12th in this conference series. More than 600 participants from more than 25 countries took part in this meeting. The scientific talks were presented by invitation in plenary and three parallel sessions. In the poster sessions about 600 papers were presented. In addition, there was a plenary talk from one of the joint Hewlett-Packard prizewinners, Professor H Ibach. An exhibition of scientific instruments and books was also organized, this year with a pronounced presence of the Commission of the European Communities and an ESPRIT workshop was included in the programme of the Conference.

The proceedings contain all the accepted plenary and invited lectures whose authors complied with the deadline for submission of the manuscripts. The editors paid special attention to a prompt publication of the proceedings in order to enable all the conference participants and those who were not able to attend the meeting to benefit from the information presented during the conference. The invited talks were brought together under six broad headings in the proceedings and cover practically the entire field of condensed matter physics. The abstracts of the posters were published elsewhere, in Europhysics Conference Abstracts volume 16A.

On behalf of the Conference organizers I wish to thank all who contributed to the conference proceedings for their cooperation and assistance, generous contributions from Sponsors is greatly appreciated. I also thank the members of the International Programme and Advisory Committees (see PDF) for suggestions and preparation of the scientific programme of the conference. The immense work of members of the Organizing Committee (see PDF) is also acknowledged.

Mister Chairman, Distinguished Guests, Ladies and Gentlemen

I was honored by the invitation to address this audience. Let me first stress that we appreciate the fact the the European Physical Society elected Prague to be the site of this important meeting. Condensed matter physics and its companion, material science, are fields which impress by their rapid development and by the richness of scientific discoveries which translate unceasingly into the technological progress influencing vast areas of human life, too many to be enumerated. It is comforting to realize that the scientists of this country take part in this great international endeavor.

I believe here and now is the proper opportunity to acknowledge with deep gratitude the gentle yet generous support which the European Physical Society has offered over the past decades to aid our scientific community in its struggle against isolation and stagnation.

These bleak years are gone by now: in 1989, we have entered the area of revival and rebuilding of our national life. This is also true about our national science and scholarship. With certain pride I can state that, thanks to a joint effort of the scientists and scholars themselves, of the Government, and of the legislative bodies of this republic, the transformation of science and its institutions has already made an impressive progress, and is now in a very promising stage.

One of the basic aspects of the new ways of our scientific work should definitely be its integration into the international science. In this connection, one particularly inspiring feature of this conference is the active participation of the delegates of the European Community, and I should like to welcome all of them. Let me conclude by wishing this Conference full success, and all participants a pleasant stay in our country. Thank you.

Mr Minister, Distinguished Guests, Dear Colleagues

It is a great pleasure for me to briefly address you at the beginning of this very important conference, the Twelfth General Conference of the Condensed Matter Division. At present, the study of condensed matter represents a very active and important domain of physics. This is in its many branches that one finds today the majority of physicists engaged in basic research and its many connections with applied research are of utmost importance. It is already a matured domain in physics. Yet this is here that many recent surprises, always welcome as conducive to a better understanding and to new developments, have occurred, such as high critical temperatures in superconductivity, magnetic multilayer structures and fullerenes. We shall hear much about all that during this conference.

As a particle physicist I thank you for inviting me to be with you. As President of our Society, I would like to welcome you all on behalf of EPS and thank the Condensed Matter Division for all the works which it does to promote this branch of physics at the full European level. The success of this conference should bear further witness to its important role.

The European Physical Society has a dual structure. On the one hand it is a Society of Societies, 31 of them now that the Physical Societies of Albania, Croatia, Estonia and Lithuania have been admitted by Council 10 days ago. It thus plays a federative role. On the other hand it is a directly international society through the work of its divisions, interdivisional groups, committees and task forces, which all operate at the full European level. Ever since it was created 24 years ago, the EPS tried to ignore, as far as it was at all possible, the imposing barrier set in the middle of our continent. Difficulties spoke for themselves but good contacts and some collaboration could be established. Now that this barrier no longer exists we do our best to capitalize on our contacts to help as much as we can our colleagues in central and Eastern Europe. There are great expectations for more research at home and efficient collaborations with the Western part of Europe. Difficulties have to be overcome. Our East–West Coordination Committee is very active. The American Physical Society has followed with interest what we have been doing so far and wishes to associate itself with us in this endeavour. There will be an important meeting to that effect in Budapest next month.

We are presently discussing restructuring our Society. This should give to the divisions, which do most of the visible work through their organizing conferences and workshops, a larger executive role and it should provide them with a membership base 10 times as large as the present one. This Conference will thus also be for me the opportunity to discuss with some of you many topical issues such as the role of the speciality divisions in our Society, East–West partnership at the research group level, our role in the discussions about new International research facilities, and our connections at the division level with the American and Japanese Physical Societies.

But we are in Prague and I would like to spend the rest of my time praising the great scientific tradition of this city. Prague is indeed the oldest great cultural centre in central Europe. The university was founded in 1348 and the first major work in what was to become our field came out already in 1368, the "de motus" on scholastic kinematics of Johannes de Holandria. Prague was a cradle for modern astronomy. This started with Tadeas Hajek whose presence enticed Tycho Brahe to come here, followed by Johannes Kepler. We are then at the eve of the 17th century. In the 19th century, we find Christian Doppler and Ernst Mach. The brilliant tradition continues with C Strouhal, B Kucera, F Zaviska and J Heyrovsky, who was awarded the Nobel prize for his prominent work on polarography.

I very much enjoyed the fruitful contacts which I long had with Czechoslovakian colleagues in particle physics. This is indeed my third visit to Prague. As a CERN physicist I should say how happy we are to now have Czechoslovakia as a Member State. High energy Czechoslovak physicists already participate in three major experiments at CERN.

I left for the end a mention of the stay of Einstein in Prague in 1911–1912. This is where he held his first Professorship. This is where he started his prominent work on General Relativity. His paper "On the influence of gravity on the propagation of light" was written here. It is from Prague that he attended the first Solvay meeting and it is here that he started his relation with Ehrenfest. Einstein gave credit to "the concentration in the quiet rooms in Vinicna Street", which is within walking distance of where we are, and to the beauty of the view from the top of Letna Park. I can but conclude using Einstein's own words when he was enticing his frield Besso to visit him here: "The City of Prague is very fine, so beautiful that it is worth a long journey for itself".

On behalf of EPS I declare the Conference open.

Ladies and Gentlemen, Dear Guests, Dear Colleagues

On behalf of the main organizers of the Conference, who are the Physical Section of the Union of Czech Mathematicians and Physicists and The Institute of Physics of the Czechoslovak Academy of Science, I would like to express our delight to have you here: not only prominent physicists and those who have the chance to become prominent in due time, but also all others whose work makes an original contribution to the very broad and rich area called Condensed Matter Physics.

Each field of Physics surely has its characteristic style and features. The present experimental High Energy Physics, e.g., in attacking without any doubt the most fundamental problems concerning matter and the interactions that govern its world, tends to use big installations and large groups of people try to accomplish relatively few large-scale experiments addressing well defined questions and unifying thus the activities into one impressive flow. On the other hand the solid state or Condensed Matter Physics exhibits rather diferent features. Here many parallel strugglings do exist, an immense plurality in views and approaches, large number of apparently independent experiments, models and theoretical concepts which implies a big amount of contributions and discoveries whose importance is hardly predictable in advance. Nobody is usually able to anticipate when and at which place a breakthrough will appear and what its implication may be for other developments. In this way we may view the progress in the CMP rather as accomplished by many quasi-independent parallel steps loosely corresponding to a mosaic picture in which every stone—large or small—finds its irreplaceable position. Very like as in our life to whose present standard CMP has contributed so much. If you agree with this overall characteristics allow me to say a few words to illustrate by some examples about the Czechoslovak contribution to the development of CMP. During the Second World War the Czech Universities were closed and one of the consequences was that the post-war onset of modern CMP era was considerably delayed in Czechoslovakia compared to other developed European or Oversea countries. In spite of this the level based on former tradition of physics here was sufficient to react quickly to new stimuli from outside. Otherwise it is hard to explain that a one-year-stay of the late Zdeněk Matyáš with Sir Neville Francis Mott in Bristol, 1946–47, was sufficient to quickly initiate the Prague School of Solid State Theory, and to stimulate widespread activities in Solid State Physics in the early 50s, which has been appreciated everywhere. Let me mention some of the main protagonists: Jan Tauc, founder of the Czechoslovak Semiconductor Physics, recognized for his pioneering work on Photovoltaic Effects, Amorphous Semiconductors and Optical Properties of Semiconductors, since 1970 professor of the Brown University (USA), now present at the Conference; the late Zdeněk Trousil, outstanding technologist who independently discovered and realized zone melting method for rafination of very pure germanium; the solid state theorists Emil Antončik, Professor of the University at Aarhus (Denmark), and Bedřich Velický, both present at the Conference, the latter one as the Program Chairman; in magnetism Luboš Valenta fundamentally contributed to the theory of spontaneous magnetization in thin films and later stimulated the research in actinide magnetism that has been successfully developed in this country; Jan Kaczer in his systematic study of magnetic domains was first to observe the cylindrical domains that later become bubbles. And, of course, Zdeněk Frait with his perfect Ferromagnetic Resonance experiments in metals and metallic thin films. All three are present here. The modern computer-aided method for studying crystal structures were pushed forward by the late Alan Linek; the research of dielectric and ferroelectric crystal connected with the names of Václav Janovec, Jan Fousek, Jan Arend and Vladimir Dvořák, the last one being appreciated mostly by his original contribution to the theory of modulated (in particular incommensurable) structures and related phase transitions. And last but not least the contribution of František Kroupa to the theory of mechanical properties of solids. Most of these contributions were accomplished during the fifties and sixties. As the people I have mentioned were active in Prague, both in the Institutes of the Academy and at the Charles University, I have to fill out this incomplete enumeration stating that almost simultaneously similar activities appeared also in other cities, as Brno, Bratislava, Košice and a few others, both Czech and Slovak creating so the Czechoslovak tradition in this field. Let me add, that many physicists, who started work in Czechoslovakia as students or post doctoral fellows emigrated during the Communist era and developed into generally recognized personalities in CMP. We are glad that most of them grasped the opportunity to come and attend this conference. Besides Professors Tauc and Antončik, I am glad to welcome here Professor Arend (Switzerland), Dr Dorda (FRG), Professors Gerber, Jaroš and Mathon (UK), Professor Kunc (France) and others who may be present.

Czeckoslovak physicists were also active in the preparation of the European Physical Society. The representative of the Physics Section of our Union, Dr Plajner, was among those, who signed its charter and this act commenced good relations, and further fruitful cooperation between EPS and our Section which helped us in our isolation under the past Communist regime; and to feel, at least in a limited way, the European appurtenance. An important event along this line was the organization of the 6th EPS General Conference in Prague in 1984. Even more, we value the fact that this 12th GC of EPS Condensed Matter Division, entrusted to our hands still in the time of totality, may be held under entirely new external conditions. They are not without difficulties but with much more clear perspectives.

I wish that all of you who have honoured us by attending this conference not only gain the maximum benefit from the presented contributions and high level discussions, but also gain favourable impressions of our City, its cultural life and historical monuments and — if I may express it thus — useful personal and professional contacts with our physicists.

This presentation addresses the position of scientific and technological research in a rapidly changing Europe within a global context by examining the trends and looking at the motivations of the major actors involved. Particular attention is paid to EC cooperation schemes with Central and Eastern Europe, notably Czechoslovakia.

The cross sections for magnetic scattering of electromagnetic radiation were derived many years ago but are in general ~10⁻⁶ of the familiar cross sections for charge scattering. The brilliance of present and projected synchrotron sources now make this technique one of considerable promise. We discuss the cross sections and some of the recent results; emphasizing aspects such as the excellent wavevector resolution possible with synchrotron experiments. An important discovery of resonant magnetic scattering was made by Gibbs et al. in holmium. This occurs when the incident photon energy is tuned to certain absorption edges for the magnetic elements. A consequence is an enormous (>10⁵ in the case of actinides) enhancement of the intensity, making the technique of particular importance in the study of f element magnetism. We conclude with speculation on some future possibilities of magnetic x-ray scattering.

Thin films of amorphous silicon (a-Si:H) and its alloys are currently used in a range of devices which include photovoltaic cells, photoreceptors, page-width image sensors and thin-film transistors for active matrix liquid-crystal displays. In the latter alone it is estimated that some $4 billion will be invested in the present fiscal year in putting these into large volume production. This paper will review the present applications of a-Si:H and briefly describe a number of others which will have been proposed for this material. The advantages and disadvantages of a-Si:H, compared with its competitors, and the material and technological issues of current importance, will be discussed.

Progress in understanding the properties of the 2D soap froth and related systems is reviewed in outline. Computer simulations have been successful in modelling such "soft condensed matter", including its coarsening behaviour and response to stress.

We report recent progress on the development of components needed for high T_c SQUID magnetic sensors, and on some of the early applications which are appearing or can be foreseen. The SNS type edge junction, and the step edge and biepitaxial grain boundary junction technologies are considered and we include details of our own YBCO device based on steps which are Ar ion-milled in MgO substrates.

A review of applications prospects for these detectors will be given in the light of the available noise performance which for high T_c DC SQUIDs, at much less than 10⁻⁴ ϕ₀ Hz^−1/2, is now better than for commercial low T_c RF SQUIDs. Illustrations are given from clinical biomagnetism and from non-destructive evaluation of conducting materials.

We will discuss the polarized reflectivity spectra of Y₂Ba₄Cu_6+nO_14+n (n = 1, 2) between 6 K and 300 K. These substances are genuine untwinned substances because of thermally stable double CuO chains intead of single chains as in YBa₂Cu₃O₇. The light was polarized along the two plane directions a and b. We have found a strong contribution of the chain (E||b) to the optical conductivity and thus an anisotropy of about 1:3. A plateau in the reflectance below 60 meV has been assigned to phonon interactions.

The IR scattering rate of the charge carriers in the plane, σ^a(ω), is linear from 0.05 eV, up to 0.8 eV. In the superconducting state, plasma edges have developed between 11 meV and 21 meV, respectively, which we assume to be determined rather by the lowest phonon than by a gap. The unusual optical conductivity in the near IR may be based on a diffusive motion of magnetic polarons in an anti-ferromagnetic matrix. Good fits of the reflectivity are obtained by a two-fluid model pointing to a real space pairing (e.g. magnetic bipolarons) with a subsequent Bose condensation.

Some results for the ordering specific heats of the local pair superconductor obtained by a Monte Carlo method are presented. The λ-type anomaly agrees well with experiment. Furthermore, an explanation of the observed anomalous temperature dependences of the NMR Knight shift and T₁ of the superconducting cuprates is given on basis of a model of spinless holes, moving in an antiferromagnetic background of localized copper spins.

The nonlinear current-voltage characteristics of high T_c superconductors are analysed in order to determine which kind of transport process is dominant across their intrinsic interfaces. The CVC of ceramic samples as well as of samples with silver coated grains are discussed. It is shown that the CVC of grain boundaries are consistent with metallic conductance, instead of tunnelling conductance as the charge transport process in high temperature superconductors.

In this paper we review the anisotropic aspects of the critical current in YBCO. We make the distinction between the 2-dimensional and 3-dimensional contributions for this weakly anisotropic material.

Anomalous temperature dependences of the lower critical fields H_c1(T) for H||c measured in YBa₂Cu₃O_x by the extrapolation method, the superconducting gaps Δ_|| and Δ_⊥ measured in Bi₂Sr₂CaCu₂O_8+x by Raman spectroscopy in two different polarizations parallel and perpendicular to the CuO₂ sheets, and the scaling behavior of the Bi_1.7Pb_0.3Sr_1.9CaCu_1.90_8+x magnetoresistance arising from giant fluctuations in the decoupled quasi 2D CuO₂ superconducting planes are analyzed in terms of the models taking into account the layered structure of the 2212 and 123 cuprates with a very weak coupling between the CuO₂ planes. The importance of the proximity induced superconductivity in explaining unusual H_c1(T) and Δ_⊥(T) features is emphasized.

In high-T_c superconductors thermally activated depinning of flux lines occurs in a large temperature range below T_c. At low current densities this thermal depinning means flux diffusion whose exponential temperature dependence explains the irreversibility lines of various experiments in a natural way without assuming a melting transition of the flux-line lattice. Linear response expressions are given for the complex a.c. penetration depth, surface impedance, a.c. resistivity and a.c. susceptibility of films and cylinders which consider elastic pinning, viscous drag, and creep of the flux lines.

At low temperatures, large, temperature-independent magnetic relaxation effects are found in the high-temperature superconducting ceramics and also in the organic superconductors giving evidence that, at these temperatures, vortex motion involves mainly a form of macroscopic quantum tunneling. At higher temperatures, flux motion is thermally assisted. The measured tunneling rates at T = 6 mK and low magnetic induction are in excellent agreement with the T → 0 values given for single vortex tunneling by the quantum collective creep theory of Blatter, Geshkenbein and Vinokur. The effects of anisotropy and layering in the tunneling rates of single crystals are in very good agreement with the aforementioned theory.

We report on inelastic neutron scattering experiments carried out on single-crystals of the YBa₂Cu₃O_6+x system. The spin dynamics has been successfully investigated as a function of temperature in the metallic state over the whole doping range from the weakly doped (x = 0.45, 0.51) to the heavily doped (x = 0.69, 0.92) and the overdoped (x = 1) regimes. Dynamical AF-correlations remain in the metallic state but the in-plane correlation length decreases with doping whereas the AF-coupling between the two Cu(2) layers is not affected, except in the overdoped regime.

At low temperatures, in any superconducting samples, an energy gap has been discovered in the spin excitation spectrum, with a value E_G/kT_c ≃ 3.5 which becomes weaker close to the I-M transition. Moreover, in the heavily doped regime, a pseudo gap persists above T_c.

A quite unusual T-dependence of the spin excitation spectrum was found, especially in the weakly doped regime.

The thermodynamic stability of two widely investigated models for high T_c superconductivity, the Kondo-lattice and the extended three band Hubbard model, is analyzed in the strong local repulsion regime. Superconducting instabilities in these models are also found. For the first model superconductivity stabilizes the system against phase separation. For the second model the superconducting instabilities occur close to the phase separation region. In both models, therefore, phase separation and superconductivity appear to be related phenomena.

When doping the antiferromagnetically (AF) ordered high-T_c cuprates with holes magnetic polarons (spin-polarized clusters) are formed. At higher hole concentrations and due to cluster diffusion a phase separation is formed by establishing large fractal or percolate clusters. This percolation picture allows us to understand experimentally obtained resulted for phase separation as well as magnetic and conductive phase diagrams.

We report on calculations for the Emery model of the carriers in the copper oxygen planes of HTSC materials using the variational principle of Bogoljubov. A strong coupling ansatz optimized for small values of the ratio hopping constant to Coulomb correlation gives the lowest free energy in the neutral case and in the case of moderate doping. From the curves chemical potential vs. doping concentration large regions of thermodynamic instability and phase separation respectively are determined using a Maxwell construction. The temperature and parameter dependence of these coexistence regions with respect to para-, ferro- and antiferromagnetic order is given.

The exchange coupling between magnetic layers, across a non-magnetic spacer layer, is observed to oscillate as a function of spacer thickness. In earlier work it was shown that the oscillation periods are characteristic of the spacer and it is shown now that the amplitude and phase depend critically on the matching between spacer and ferromagnet. Detailed tight-binding calculations of the coupling across a chromium spacer, using five d orbitals, exhibit oscillations with components of several different periods. However an oversimplified matching is used, so that a fully quantitative comparison with experimental work on Fe/Cr/Fe sandwiches cannot yet be made.

The exchange interaction of Fe across interlayers X, X = Cr, Au, Ag and Al has been investigated by means of M(H) curves and lightscattering from spinwaves. For these investigations Fe/X/Fe structures were grown with as good epitaxy and crystallinity as possible, which is documented by in situ RHEED investigations. For the coupling across Cr and Au we find oscillations as a function of the interlayer thickness with a long and a short period. For Al interlayers only long periods can be clearly resolved and for Ag no oscillations have been observed. Apart from the oscillations between ferro- and antiferromagnetic type coupling we find also contributions of 90° type coupling which aligns the magnetic films perpendicular to each other. The temperature dependence of 90° coupling has been investigated in the case of Au interlayers.

Characteristic features of the magnetism in light-actinide intermetallic compounds, as large orbital moments originating from itinerant electron states and huge magnetic anisotropy, are discussed in relation to electron-band structure, and especially to the 5f-ligand hybridization. UTX compounds are an example, where the bonding anisotropy leads to a strong anisotropy of magnetic properties. Experiments on single crystals of UNiGa can distinguish how different types of magnetic coupling affect electrical resistance for various current directions. The effect of magnetic field on inter-site spinfluctuations in UNiAl is manifested in the enhancement of the λ-coefficient of specific heat.

The magnetic properties of the actinide monopnictides and monochalcogenides are governed by the properties of the usually trivalent actinide ion, by the interaction with the crystalline electric field, by isotropic and anisotropic exchange and by hybridization. In a few cases these factors can be separated so that a qualitative understanding becomes possible.

The interface properties are crucial for most of the unique properties of ferromagnetic multilayers. Typical examples and results of the influence of interface structural and electronic properties on the magnetic anisotropy and the interlayer exchange coupling effects will be reviewed for various Fe/M interfaces (M: Cr, Al, Ag, Au, Gd, Y). Emphasis will be put on interface investigations by ⁵⁷Fe-Conversion Electron Mössbauer Spectroscopy (CEMS) on magnetic interactions at Fe/Gd interfaces and on the recent studies of amorphous Fe films. Those films can be stabilized below a critical thinkness of about 2.3 nm at Fe/R interfaces (R: Gd, Y). Between Y layers amorphous Fe can be studied to temperatures well above the average Curie temperature of T_c = 80 K.

In this the 150th year after the reported discovery of magnetostriction in ferromagnetic materials by Joule, an overview is given of the present state of knowledge in this field. Soft and hard materials are considered, and their potential for application is discussed. The advent of "atomically engineered" magnetic materials by deposition processing is also considered.

Optical measurements in the far infrared performed at low temperatures on heavy fermion compounds have given surprising results and important information about the electronic structure, complementary to the one obtained by the other techniques (e.g. specific heat, magnetization).

The existence of a narrow gap a few meV below (above) the Fermi level has been observed as a common feature of heavy fermions, together with the evidence of a coherent hybridization of f- and conduction-electrons.

The implications of the hybridization gap on other physical properties such as the temperature dependence of the resistivity, the specific heat and the susceptibility will be discussed.

Low plasma frequencies, due to heavy effective masses, have been measured. The effect of alloying has also been studied and discussed.

The characteristic temperature dependence of the specific heat, of the static magnetic susceptibility, of the electric resistivity and of the thermopower of heavy fermion systems can be reproduced within a single theory, namely the self-consistent second order U-perturbation treatment of the periodic Anderson model. This approach, within which explicit calculations are possible in the limit of a large spatial dimension d for correlated lattice electrons, properly fulfils the Luttinger sum rules. Here we give an overview on the results obtained within this approach, in particular for the temperature dependence of actual measurable physical quantities, and show that an excellent qualitative reproduction of the typical experimental results is possible by simply using the d = ∞ result for d = 3.

Transport experiments on heavy fermion superconductors have been realized in order to investigate the nature of the order parameters and the link between the properties of normal (Neel temperature, Fermi liquid parameters) and superconducting phases. Furthermore, the heavy fermion compound CeCu₂Ge₂ was found to undergo a superconducting transition at very high pressure. At 100 kbar, T_c ≊ 0.64 K and H_c2(0) ≊ 20 kOe.

The basic theoretical methods for treating defects and, in particular, impurities in semiconductors are first reviewed. They split into two groups: the techniques appropriate to deep states (local density, empirical tight binding) or to shallow states (effective mass approximation). A discussion is given of how to handle the most difficult intermediate cases (corresponding to the shallow-deep instability) and of the possible coexistence of these two types of states.

The second part describes applications of these general arguments to specific cases. This begins with the well documented case of transition metal impurities followed by recent results obtained for rare earth impurities. A fairly detailed analysis is performed in the case of EL2 showing that this center must be a complex involving the antisite. The case of the DX donor in Ga_1-xAl_xAs compounds is also reviewed, with the arguments in favor of the two competing models.

First we review experimental evidence for the negative intra-defect Hubbard correlation energy U of DX centers. We show from investigations of photoconductivity and the emission kinetics of free carriers that the ground state of the Si-donor in Al_xGa_1-xAs binds two electrons. Thus it is negatively charged in agreement with the theoretical prediction by Chadi and Chang. As a consequence, all donors in these materials are always charged and produce a strongly fluctuating Coulomb potential, which manifests itself in the capture kinetics of photo-excited carriers and the apparent thermal activation energy of bound carriers. Additional interesting effects arise from the freedom in the spatial distribution of the doubly occupied donors among the random matrix of donors. In order to minimize the total Coulomb energy, short range dipoles are preferentially formed which tend to reduce ionized impurity scattering in thermal equilibrium. Experiments and models are reviewed.

Negative-donor centres (D⁻ centres) form readily in suitably doped multiple-quantum wells due to electron transfer from remote donors in the barriers to donor sites in the wells. We present a review of our magneto-optical spectroscopy of selectively doped GaAs-(Ga,Al)As quantum wells aiming at the study of confined D⁻ centres and at D⁻ doping of quantum wells.

Up to recently the observation of a defect metastability was considered to reflect a rearrangement of a defect complex, because a point defect was believed to be structurally stable. We summarize our recent theoretical studies which show that these assumptions are incorrect and that metastability of substitutional sp defects is an intrinsic property of III-V compounds. This intrinsic metastability may be observable for some centers but not for others. We compare the behavior of three defects in GaAs: the As antisite (a model for EL2), the Si impurity (a model for DX), and the Sb donor (a stable center).

The theory of the magnetooptical properties of highly excited quantum wells is reviewed with emphasis on recent experimental work in a collaboration led by the Chernogolovka group. Realistic numerical calculations strongly support the interpretation of the experimental magnetoluminescence spectra in terms of bound electron-hole pairs, but at the experimental plasma temperatures of ~40-50 K phase coherence and thus observation of the excitonic insulator state is not yet achieved. Important screening effects beyond the Hartree-Fock approximation exist even at magnetic fields of ~10T. At lower densities, evidence of dynamical screening of the electron-hole interaction is found.

In the presence of free carriers, which can be optically excited or of extrinsic origin, electron-electron interactions induce changes in the optical properties of semiconductors. Both such systems have been investigated with interband magnetooptics under well defined conditions achieved by working with thin layers and applying high magnetic fields. The quenching of the excitonic resonance and the pronounced Auger processes (leading to the observation of the energy-up conversion) in the case of modulation doped structures, and the energy structure of a high density electron-hole gas in undoped quantum wells are discussed.

The interaction between 2D electrons and photoexcited holes is of fundamental importance in determining the photoluminescence spectrum in the fractional quantum Hall regime. We present new measurements on a low-density electron system which reveal a distinct Fermi edge singularity at zero and low magnetic fields, indicating dynamical response of the Fermi sea to the presence of the hole. The photoluminescence intensity from this system is very sensitive to the effects of electron correlation, and hierarchies of FQH states are clearly identified. The photoluminescence energy, however, is quite insensitive to electron ground state in the FQH regime.

Two-dimensional electron systems exposed to low temperatures and high magnetic fields exhibit a rich spectrum of novel electronic phases. The highly correlated quantum liquids underlying the fractional quantum Hall effect (FQHE) dominate this region of phase space. Beyond the primitive Laughlin liquids, experiments have unraveled a multitude of higher order fractions whose origin and mutual relationships continue to be debated. The existence of the electron spin has led to the appearance of yet other electron liquids, some with even-denominator quantum numbers and novel properties. In the lowest Landau level, at half-filling, strong experimental features point towards the existence of yet another state; distinctly different from a FQHE liquid. At the highest fields and lowest electron concentration an insulating phase prevails as the temperature drops towards zero. This phase has been shown to engulf the v = 1/5 Laughlin liquid and is postulated to reflect the sought-after electron solid, pinned to residual impurities. Such an assignment remains controversial and alternative origins have been proposed. Recent experiments are expanding out of the plane elucidating electron correlation phenomena as the third dimension is added successively. This paper summarizes the highlights of the past decade and covers some recent developments. It focuses on experiments and stresses the progress in electronic transport measurements in this active field of research.

Theory shows that in a superlattice consisting of very thin alternating layers of Si and Ge the bulk signature of the Si conduction band wave function is altered and the selection rule for optical transitions peculiar to indirect gap materials is broken. Such structures are also predicted to give rise to novel nonlinear phenomena and to strong optical spectra in the far infrared. However, recent experimental data suggest that the optical spectra of Si-Ge cannot be accounted for in terms of the ideal (infinite and defect free) superlattice model. A more realistic theory is needed which will account for the effect upon optical properties of the finite dimensions, defects and electric fields peculiar to Si-Ge structures. We find that these effects alter the assessment of the key material parameters of Si-Ge.

A review is given of the most recent development in the field of modulation doping in the Si/SiGe heterosystem. Main topics are the successful implementation of high-quality, strain-relaxed SiGe buffers, the realization of n-type quantum well structures with unprecedentedly high electron mobilities, magnetotransport investigations of such structures, and their exploitation for test devices with very high transconductance values.

In the framework of a linear-response treatment we perform a theoretical investigation on the band-offset problem at strained Si/Ge interfaces where chemical and elastic effects can be studied independently. Our main finding is that the band offset is a bulk property, depending only upon the macroscopic strain present in the two materials far from the interface, and independent of any interface feature, such as abruptness, interface strain or buckling. Moreover, our results indicate that the strain variations affect only weakly the valence-band offset and that simple band-structure effects are responsible for a rather large strain-induced tunability of the offset between the topmost valence states.

The standardly used source molecules in the metalorganic vapour phase epitaxy (MOVPE) of III/V-semiconductor heterostructures (trialkyl group(III)-compounds and group(V)-hydrides) have severe drawbacks. The trialkyl group(III)-compounds are pyrophoric and the group(V)-hydrides are highly toxic substances. These properties lead to problems of the MOVPE process with respect to safety concerns as well as crystal purity (incorporation of O and C especially for Al-containing epitaxial layers). In this work, we describe new developments of the interdisciplinary research for alternative precursor molecules. The principle ways for changing the molecular structure of alternative group(III)- as well as group(V)-compounds are discussed in order to overcome the above mentioned problems. Special emphasis is put on the class of the hydride group(III) amine adducts, which are in particular suitable in combination with less toxic group(V)-compounds. MOVPE growth experiments are reported for heterostructures in the GaAs/(AlGa)As material system. The applicability of the alternative precursors is established by studying the structural, electrical and optical properties of GaAs, (AlGa)As bulk layers as well as of single and multi quantum well heterostructures by high-resolution x-ray diffraction, temperature-dependent Hall as well as photoluminescence investigations.

We applied an extension of the Internal Photoemission, the Internally Detected Photoexcitation Spectroscopy (IDEPS), to the direct observation of deep levels and band discontinuities at different semiconductor heterostructure interfaces. We studied Schottky barriers (Al/GaAs, Pd/n-GaAs, Ni-Au-Ge/n-GaAs) and heterojunctions (a-Ge/n-GaAs). The GaAs surface, before metal deposition, was treated in two different ways: chemical etching (CE samples) in air or argon ion sputtering (AIS samples) in vacuum. We found, for Al/GaAs, Pd/n-GaAs and Ni-Au-Ge/n-GaAs Schottky heterostructures, a barrier height of 0.78 eV, and for a-Ge/n-GaAs a valence band discontinuity of 0.42 eV. This results in a pinning of the Fermi-level at an average value of 0.76 ± 0.03 eV above the top of the GaAs valence band. This pinning is probably due to the presence of oxygen at the semiconductor surface in the CE samples. In the AIS samples, where the metal was deposited without vacuum breaking, the Φ_b results in a great value. Three onsets, due to transitions involving interfaces states, at 1.05 eV, 1.27 eV, and at 1.36 eV, were also evident in the Pd/n-GaAs heterostructure. Furthermore, we studied how a bombardment by hydrogen or hydrogen + nitrogen ions at low energy (~100 eV) of this Schottky barrier modifies its interfacial states. In the case of Ni-Au-Ge/n-GaAs heterostructure we also determined the value of the energy gap at the interface, demonstrating that the thermal treatments could reduce such a value.

This paper compares the evidence for dry etch damage from studies of conductance of nanometre-scale wires and the luminescence of nanometrescale patterned quantum wells. It presents a quantitative model for the effects of dry etch damage in conducting wires, which may account for the dead layers observed in some luminescence experiments. Taken with other models of exciton diffusion and intrinsic luminescence quenching due to phonon bottlenecking, we may now have a complete description of systematic variations in luminescence with size. Nonsystematic variations in luminescence data may be a result of point defects in the epitaxially-grown material: cathodoluminescence mapping and PL measurements on wires of different length point to such a mechanism.

We describe some of the physics peculiar to electron transport in quasi-one-dimensional systems in semiconductors, with particular reference to three systems in which it has been investigated: ultra-fine free-standing wires of doped GaAs, the split-gate high-electron-mobility transistor (with and without a back-gate to vary the carrier concentration), and vertical quantum pillars fabricated from multiple-barrier heterostructure material. They are all electron waveguide structures, characterized by ⩽0.1 μm feature sizes in two of the three spatial dimensions, and they all exhibit phenomena quite different from those seen in macroscopic semiconductor structures. We try to identify the subset of this physics that may be exploitable, and comment on the more general relevance of this new physics to devices.

We have studied wide quantum well structures with various spectroscopic tools. The effects originating from lateral quantization are found to be strongly dependent on the state under investigation and on the excited carrier density. While conventional low-intensity photoluminescence excitation spectroscopy reveals the quantization of the exciton's center-of-mass motion, two-photon absorption spectroscopy indicates that higher excitonic states are completely quantized by the lateral confinement. The energy positions are found to be in good agreement with one-dimensional (1D) subband calculations. Additionally, transitions between 1D subbands can be found under high excitation conditions. The singularity of the 1D-density of states is reflected by distinct peaks in the time resolved spectra of a dense electron-hole plasma, which is optically excited by high power laser pulses.

We present a theoretical study of the internal thermalization and the energy relaxation of highly photoexcited electrons and holes in bulk intrinsic and doped GaAs and InP for a wide range of excitation and doping densities. The theoretical approach consists of an Ensemble-Monte-Carlo analysis of time-resolved spectral luminescence and pump-and-probe absorption measurements recently performed at Aachen and Stuttgart. A detailed comparison between theory and spectroscopic data for different experimental scenarios provides the input for a critical discussion of our present understanding of hot-carrier physics and of the semiclassical transport model underlying conventional Monte-Carlo techniques.

Quantum well (QW) structures composed of a diamagnetic narrow gap IV-VI semiconductor (PbTe, PbSe) and a dilute magnetic compound (PbMnTe, PbMnSe, PbEuTe) are investigated by various magneto-optical methods as coherent anti-Stokes Raman scattering (CARS), interband-absorption and photoluminescence. The CARS experiments yield precise values of the spin splittings of valence and conduction bands as a function of magnetic field and temperature. By interband transmission energy gaps and effective masses are determined.

The data are analyzed in the framework of an envelope function approach taking into account the exchange interaction between free charge carriers and the magnetic moments of the paramagnetic ions in a molecular field approximation. A comparison of the spin splittings of the bands in the QWs with those observed in homogeneous materials yields information on the probabilities to find the mobile carriers in the diamagnetic and the semimagnetic layers, respectively. This information allows the determination of the type of the superlattice and of the band offsets at the interfaces.

The Heisenberg exchange interaction between mobile carriers and the localized magnetic ions opens new aspects for modelling low-dimensional magnetic systems in context with electronic confinement effects. The magnetic ions in dilute magnetic semiconductors (DMS) can be polarized by optical pumping with circularly polarized light via exchange interaction with spin-polarized carriers. The magnetization of optically-oriented Mn-ions were detected by a superconducting quantum interferometer (SQUID) in modulated hetero- or superlattice structures like CdTd/Cd_1-xMn_xTe, CdTd/Cd_1-xMn_xTe, Cd_1-xMn_xTe/Cd_1-yMn_yTe, ZnTe/Cd_1-xMn_xSe, Pb_1-xMn_xTe/PbTe and Pb_1-xEu_xTe/PbTe. In wide-gap modulated semi-conductor structures time-resolved studies of the photomagnetization (PM) were performed by other investigators, whereas in narrow-gap systems only our CW measurements are available. The fundamentals of optically induced magnetization depend critically on the exchange parameters which by themselves reflect essential properties of the crystal system and band-structure. In narrow gap systems like Pb_1-xMn_xTe/PbTe or Pb_1-xEu_xTe/PbTe a quite unique model system is available to investigate the exchange interaction in a rocksalt-structure and in a many-valley electronic bandstructure. Therefore, the dependence of photomagnetization and of spin-relaxation of optically pumped carriers of the dimensionality and on the penetration of wavefunctions into DMS barriers is also distinct in these systems. The different behaviour concerning mean-field approach or dynamic polarization of magnetic ions is critically reviewed in context of earlier and new studies of optically induced magnetization.

Quantum interference phenomena have been studied in superconducting Al loops with dimensions smaller than the temperature dependent coherence length ξ(T). Besides the classical Little-Parks (LP) effect, anomalous resistance oscillations have been observed whose amplitude shows an unusual magnetic field and current dependence at low flux quanta. The magnetic field (H) - critical temperature (T_c) phase boundary has been measured for different transport currents j_t. A clear field enhancement of T_c is observed at low current densities. The amplitude of the field enhancement of T_c is strongly influenced by the transport current through the loop. Higher j_t values induce a transition from the T_c(H) boundary with LP oscillations to a new regime without oscillations. This transition is governed by the interplay between the characteristic superconducting length scale ξ(T), the loop circumference 2πr and the de Broglie wavelength λ_D ∝ (1/j_t), tuned by the transport current.

A review is presented of the state of the art of conducting polymers. Special attention is given to the high electrical conductivity of highly ordered doped polyacetylene, to the attempts to reproduce σ-values above 10⁵ S/cm, and to efforts to correlate the electrical conductivity with other physical parameters (European Brite/Euram Project HICOPOL).

The dynamical aspects of the proton translocation mechanism in bacteriorhodopsin (BR) were under investigation using quasielastic incoherent neutron scattering (QINS). We performed experiments on BR-molecules in well-ordered two-dimensional lattices of purple membranes (PM), using a multi-chopper time-of-flight spectrometer. Spectra were obtained under dark conditions at 112 different scattering angles with momentum transfer Q from 0.1 to 1.8 Å⁻¹. The sample, consisting of oriented PM-stacks, exhibits QINS intensities and energy widths varying with sample orientation, temperature and hydration. The results from an analysis of the hydration dependence are consistent with the presence of fast proton diffusion at 100% relative humidity (r.h.). At 86% r.h. no translational diffusion is observed on the time scale of the experiment, although the (local diffusive) motion of water protons appears to be occurring preferentially parallel to the membrane plane.

Despite the large differences in mass, length and time scales, suspensions of spherical colloidal particles in a solvent share many features of simple atomic liquids, including structure, phase behaviour and even the transition to a glassy state at high concentrations. The analogies, but also the differences, will be illustrated by a few recent applications of the Statistical Mechanics formalism of simple liquids to sterically or charge-stabilized, concentrated colloidal suspensions. Examples include sedimentation equilibrium and non-linear screening of charged colloids by counterions.

Amphiphiles are organic molecules which spontaneously form two-dimensional phases, particularly at pre-existing interfaces. This is the physical basis for cell membranes and detergents, and is very important in colloidal systems. Recently available techniques have provided proof that monolayers of many amphiphiles display a rich variety of phases. Many of these phases display behaviour characteristic for low-dimensional (i.e. highly anisotropic) systems, and some are hexatic. In addition to the experimental conditions of surface pressure and temperature, the particular phase adopted by a monolayer depends on the chain length, the headgroup and its environment. The present paper shows that, although not all structures are possible, the same structures recur under suitable conditions with different headgroups and different environments, including the commensurate environment provided by a multilayer structure. The common factor limiting the choice of structure appears to be the aliphatic chains, and the particular structure which actually occurs appears to be determined by interaction of the environment with the headgroups. This idea is developed and shown to predict correctly the variation with chain length of two of the upright-to-tilted phase transitions observed in monolayers on the water surface. It is possible to talk about a bulk pressure within the monolayer, which is influenced by the interactions at the monolayer interfaces and which controls the molecular tilt.

The small angle neutron scattered intensity of aqueous solutions of 2-butoxyethanol and C₃E₆ (a nonionic surfactant) is presented. Their analysis shows the presence of spherical micelles and concentration fluctuations. The radius of the micelles corresponds to the length of the molecules of surfactant. The concentration fluctuations increase with temperature and are dominant when one approaches the demixtion curve.

We investigate mixtures of hard spheres with positive non-additivity of the diameters (σ₁₂ > (σ₁₁ + σ₂₂)/2), which undergo phase separation at high density. Due to the simplicity of the interaction, they appear as prototype systems for studying phase separation in liquid mixtures. We performed Monte Carlo simulations with the so called Gibbs ensemble technique which allows a direct determination of the coexisting phases. We compare the coexistence curves obtained from simulation with the predictions of a very simple first order perturbative theory (MIX1). For asymmetries and non-additivities in a physically interesting range, our results confirm the fair reliability of MIX1. Thus we have used the perturbative theory to study trends in the phase behaviour of such binary systems as function of non-additivity and asymmetry.

Various multilayers of Cd stearate were prepared by means of the Langmuir-Blodgett technique and investigated at T = T_s - ΔT (T_s, melting point of LB film). The x-ray scattering spectrum was recorded in several time intervals using an energy dispersive set up. For ΔT = 3 ... 9 K the observed intensity reduction of Bragg peaks, which corresponds to the multilayer periodicity, is recorded as a function of time. In addition new peaks appear. This is interpreted in terms of a kinetically driven phase separation of the metastable LB-phase. The time needed for phase disorder and the nucleation of the new phase decreases for smaller ΔT. The measured layer spacing of the new phase is larger than the length of a molecule but smaller than a LB bilayer. We suggest the new phase to be interpreted as a two dimensional array of cylindrical structure elements as known from bulk material. Obviously this phase transition needs the existence of an intermediate disordered state and is kinetically restricted at lower temperature.

Recently measured light scattering spectra for CaKNO₃ exhibit a cross over of the dynamics from that typical for a moderately supercooled liquid to one characteristic for a strongly supercooled state. It can be described by formulae derived within the mode coupling theory of dense liquids for the asymptotic behaviour of correlation functions near a glass transition singularity.

We describe briefly experiments which show that suspensions in a liquid of nearly-equal-sized hard colloidal spheres exhibit a glass transition at high particle concentrations.

An overview is presented of the method used to parallelize a set of total energy pseudopotential codes on a 64-node i860 Meiko Computing Surface and a 32-node Intel iPSC/860 Hypercube. These codes have been used to calculate the surface energies and relaxed structures of the 3 × 3, 5 × 5 and 7 × 7 Takayanagi reconstructions of the (111) surface of silicon. It is found that the 7 × 7 reconstruction has the lowest energy and that structural trends across the series of reconstructions can be related to the degree of charge transfer from the adatoms to the rest atoms.

The purpose of this paper is to investigate the utility of traditional many-body perturbation theory in highly correlated d-electron systems taking full account of the underlying one-particle structure. Starting from a zeroth order Greens function based on a self-consistent local-density calculation, we have here computed the dielectric matrix of ferromagnetic nickel within the RPA, thus demonstrating the feasibility of many-body calculations in realistic systems. From our results we obtain the optical absorption, the optical conductivity and the electron-energy-loss spectra. A deviation from experiment in our optical absorption spectrum is easily traced back to our local-density starting point. Using instead quasiparticle energies in the zeroth order Greens function results in much better agreement with experiment. The particle-hole interaction is relatively unimportant in optical absorption and the local-field effects (LFE) are small, especially at lower energies and small momentum transfers. From the dielectric matrix, we also obtain the GW approximation for the electronic self-energy and from there, the quasi-particle bandstructure which compares well with experiment. In particular we obtain the desired bandnarrowing. On the other hand, we obtain no satellite 6eV below the Fermi energy and our spin-splitting is too large, thus demonstrating the need for a theory beyond the GW approximation.

Although bulk III-V semiconductor alloys generally exhibit phase separation, vapor-phase epitaxial growth of Ga_0.5In_0.5P on GaAs (001) shows spontaneous ordering into a [111] oriented monolayer superlattice. This ordering is believed to be induced at the surface during growth. We have performed total-energy first-principles pseudopotential calculations for the surface layers of cation- and anion-terminated (001) surfaces of Ga_0.5In_0.5P on GaAs substrates and valence-forcefield model calculations for deeper layers. These calculations show that the relative stability of the various Ga/In arrangements is a strong function of the layer depth below the surface. At the cation-terminated surface a Ga/In atomic arrangement corresponding to the observed bulk ordering is preferred over other arrangements by 90 meV per surface atom. This preference is intimately tied to electronically driven surface reconstructions. For subsurface Ga/In layers the energy differences are generally small (<30 meV per atom), but at the fourth subsurface layer they play a critical role in controlling the three-dimensional stacking of the ordered (001) layers. Finally, thermodynamic calculations based on a configurational Hamiltonian whose interaction energies are fit to the total energy calculations show that the observed ordering can be explained as a thermodynamically stable surface phase at growth temperatures, which, depending on the atomic mobilities, may remain as a metastable bulk phase as the growth proceeds. On the basis of these results we propose a possible mechanism for the development of the ordered phase.

An ab-initio LCAO method is used to calculate the electronic properties of metals (Li, Na, K and Al), H and electronegative atoms (Cl and O) chemisorbed on GaAs (110). Results for different coverages are presented. The adsorption geometry, the corresponding chemisorption energy, the local density of states and the transfer of charge are obtained for each case. A discussion of the similarities and differences between the different cases is presented.

A study of quasi-one-dimensional Pb, Tl and mixed Pb-Tl overlayers on noble metal substrates using LEED, Auger electron spectroscopy and photoelectron spectroscopy is presented. A or p(6√2 × 2√2)R45°] Tl chain structure on Cu(100) shows a temperature dependent Fermi level bandgap and a structure consistent with a simple dimerization along the chains indicating the formation of a Peierls distortion or charge-density wave (CDW). In contrast a lower density [or c(4 × 4)] chain structure on Cu(100) and a higher density [or p(2 × 1)] chain structure on Ag(100) show no evidence for the formation of superstructures and no detectable bandgap. In mixed Pb-Tl layers on Cu(100) LEED indicates the formation of interpenetrating chains and the formation of a Tl chain phase with the same density as the structure but with the Peierls distortion suppressed. Preliminary conductivity measurements indicate that Tl chains on Cu(100) with a slightly lower density than the CDW structure may have entered the superconducting region of the 1D phase diagram with a transition temperature in the region of 220 K.

Ultrasmall (≲5 nm in lateral diameter) double-barrier tunnel junctions have been realized using metal particles (2-5 nm in diameter) sandwiched in between a metallic substrate and the metallic tip of a scanning tunneling microscope (STM). The particles were either grown by e-beam evaporation or prepared using colloid chemistry. Two electrical transport effects, in good agreement with the semi-classical theory of single-electron tunneling, have been found at room temperature: the Coulomb gap and the Coulomb staircase. The interpretation in terms of single-electron tunneling is further supported by liquid helium temperature STM measurements on identical samples.

Implanted in matter, the positive muon provides a novel form of magnetic resonance probe. Its resonance or precession frequencies determine internal magnetic or hyperfine fields; its relaxation functions characterize the spatial or temporal variation of these fields. These signals are displayed with huge sensitivity via the asymmetry in the muon radioactive decay.

This unique microscopic probe has the additional feature that it mimics proton behaviour. It can pick up an electron to form muonium, the light pseudo-isotope of hydrogen. These chemical aspects are well illustrated by studies of muons and muonium in H₂O ice. The examples chosen serve to describe the three major experimental techniques, muon spin rotation, relaxation and resonance, and also to introduce the use of muonium to model hydrogen behaviour in a variety of materials, metals, semiconductors and molecular materials.

The universal behavior of the electronic transport in solids is discussed. It is shown that these findings can be interpreted by a general quantum transport model. It is based on a generalization of the Bohr quantum condition, resulting in a reference length a_e = 7.25 nm and a reference electron density of 3 × 10¹⁵ m^-2, which limits the electron current. It is shown that the limitation is supported by experimental data. The priority of the formulation of the square of the electric charge as e² = αh is discussed. It leads to a new system of units. The formulations of the reference values of all electromagnetic quantities are given. It is shown that the data of the Aharonov Bohm effect of 1/f noise confirm the presented statements.

The luminescent properties of porous silicon films are reviewed and discussed in relation with the mechanisms proposed for light emission. It is shown that a good control of the light emission is obtained by the use of electrochemically oxidized porous layers. The transient behavior of the photoluminescence is characterized, leading to propose a model for the recombination mechanisms involved in the light emission phenomena. Finally, the electroluminescence which appears during the anodic oxidation of porous silicon and the first reports of EL from solid devices are described.

Samples of porous silicon, produced electrochemically with a spatially varying current density, have have studied by Raman scattering, by spectral photoluminescence (PL) and by photoluminescence topography. The PL maximum observed at about 1.7 eV shifts to higher energies (2.0 eV) for the strongest structuring, where the highest PL is also found. These results give evidence for an effective direct bandgap in this material, which is a condition for the high PL. For a different kind of sample, produced by reactive ion etching, a very low wavelength PL peak is observed at 550 nm (2.0 eV). A model of spherical crystalline silicon particles (of up to 706 silicon atoms), saturated at their surface with hydrogen, has been studied by density functional theory. The calculation shows that this nanocrystalline silicon has a direct gap which is substantially larger than the bulk (indirect) gap and is consistent with experimental observation.

Infrared transmission and Raman spectra of siloxene (Si₆O₃H₆) and porous Si have been analyzed in terms of the structural composition. Siloxene consists of Si₆-rings interconnected by oxygen and terminated by hydrogen, and corrugated Si(111)-planes saturated by OH and H. The pronounced photoluminescence of siloxene is mainly related to the Si₆-rings. Porous Si shows similar luminescence and almost identical vibrational properties as siloxene, which makes this class of materials a possible candidate for the origin of the luminescence in p-Si. Annealing at temperatures below 300 °C and moderate illumination of siloxene induces the transformation from the Si-plane-like structure to the Si-ring-like structure. Anneal temperatures above 300 °C and intense illumination of both siloxene and p-Si leads to an effusion of hydrogen and an irreversible oxidation of the samples.

The chromatographic procedure which led to isolation of the first pure samples of C₆₀ and C₇₀, together with recent advances in the separation of higher fullerenes, are described. At least three isomers of C₇₈ exist. Structural features contributing to the stability differences between fullerenes are discussed. ¹³C NMR, UV, IR, Raman and neutron scattering/diffraction measurements on C₆₀ and C₇₀ are summarized and reviewed.