The Evaluation of Proposals to the Final Call for Proposals Concerning Tmr Access to Large

Photon-Mediated Phenomena in Semiconductor Nanostructures

(Photon-Mediated Phenomena, HPRN-CT-2002-00298)

Start date: August 01, 2002

Duration: 48 months

Mid-Term Review Report

(August 01, 2002 – July 31, 2004)

Network co-ordinator: Prof Alex L. Ivanov

School of Physics and Astronomy

Cardiff University

Queen’s Buildings

5 The Parade

PO Box 913

Cardiff CF24 3YB

Wales, United Kingdom

Tel: +44 - 2920 - 875315

Fax: +44 – 2920 – 874056

Email:

Network home page:

Location of the mid-term review meeting: School of Physics and Astronomy, Cardiff

University, Conference Room WX/3.14

Date and timing of meeting: August 31, 2004

Please keep the text of this report to the minimum, using diagrams and tables wherever possible.

PART A - RESEARCH RESULTS

A.1Scientific Highlights

We have followed the main scientific aims formulated in our Network project:

Task 1. Optics of QDs embedded in a three-dimensional microcavity

(Lead team: Dortmund)

As we have already discussed in the first annual report, the experiments performed by the Network YR, Dr N. Le Thomas (Dortmund team), demonstrated the highest coupling efficiency of spontaneous emission (beta-factor) ever observed for semiconductor nanocrystals. This result was achieved by replacing spherical, colloidal CdSe nanocrystals by anisotropic CdSe nanorods. The highly polarized emission couples almost 100% into TE-modes of spherical microcavities while the TM-modes are suppressed. Dr N. Le Thomas also achieved an optical detection of single CdSe nanorods and analysed the temperature-dependent line shape.

Strong coupling of atoms or semiconductor quantum dots with photons in high-Q microsphere cavities are in the focus of current investigations. Such systems are discussed e.g. as sources of deterministic single photons and entangled photons. Semiconductor nanocrystals, as has been demonstrated in antibunching experiments are possible candidates of a solid state-based “atom”, however, their optical transition dipole moment was found to be much smaller with respect to epitaxially grown quantum dots. We studied the impact of anisotropy and crystal symmetry of semiconductor nanostructures for achieving high optical transition dipole moments by comparing nanorods and nanodots. Making use of the Bloch part of the excitonic wave function, i.e. the solid state nature of the “artificial atom”, we are able to transform the exciton ground state symmetry of a CdSe nanorod from an optically dark to optically bright states with high degree of linear polarisation. By measuring the photon lifetime, we show experimentally for R = 3µm spheres with a sufficiently low mode volume V ~ 200(/n)3, the existence of a high quality factor Q > 200 000, which should be allow to enter the strong coupling regime with emitters of ~3ns-1 spontaneous radiative decay rate. By attaching the spheres with nanocrystals having a high optical transition dipole moment, the line shape of some cavity modes exhibit a splitting of ~40 micro eV that we attributed to the strong coupling regime.

Within “Dortmund-Paderborn” cooperation, in order to develop tuneable, photochemically stable, and positioned nanoemitters in photonic structures we tested a new epitaxial technique, the hybrid growth of MBE using nanocrystals as colloidal seeds. The enormous potential of colloidal nanocrystals concerning tunability shall be combined with device-compatible techniques of Molecular Beam Epitaxy to grow monolithic, high-Q optical microcavities with nanocrystals as the active optical material. It is planned to test different nanocrystalline emitters by integrating them in compact, monolithic microcavities or by embedding them in micromechanically tuned microcavities with movable 2D-structured mirrors as an alternative approach. The optical properties we want to control are the positioning of nanoemitters in field maxima/minima, stable single dot emission, control of polarization degree, minimizing of decoherence, and transition dipole moment control by wave function engineering. A result of the above cooperation is a common Dortmund/Paderborn patent application.

The Network YR, Dr N. Nikolaev (Cardiff team) develops theoretical optics of single photonic dots, aiming to share the results with the Dortmund and Lund teams. The Grenoble team, involving support from other national grants, aims to fabricate solid state single photon sources based on CdTe and CdSe quantum dots in pillar microcavities. An experimental set-up has been developed for quantum optics measurements of these photonic devices at low temperature (photon correlation, entanglement, interference with two photons, etc.). This acquired knowledge will be shared with the Network partners.

Task 2. Optical properties and relaxation kinetics of MC polaritons

(Lead team: Grenoble)

The Grenoble team has performed a number of unique experiments on Bose-Einstein condensation of microcavity (MC) polaritons in CdTe-based nanostructures under nonresonant optical excitations. The Paderborn team fabricated and characterised some of the samples. Recently, the Grenoble team has developed a set-up which allows spectroscopic imaging in real- and k- spaces to investigate the optical properties of microcavity polaritons. Relaxation of polaritons along the dispersion curve has been studied as a function of the excitation density using a nonresonant excitation. In the low excitation regime, a marked bottleneck effect at high in-plane k wavevectors is clearly observed. For higher excitation, it is suppressed and at a given threshold, the far-field emission pattern consists of a sharply "speckled" ring. This could be the first observation of spontaneous coherence in a solid state system (international state-of-the-art). The relevant experimental results obtained recently by the Dortmund team, deal with the polariton broadening in energy and momentum space measured as a function of in-plane momentum: when optically exciting the lower polariton branch, the strong dispersion versus wavevector results in a directional emission on a ring.

The cooperation between the Cardiff and Cambridge teams has resulted in the suggestion and development of a new field in the optics of semiconductor microcavities – resonant acousto-optics of MC polaritons. Namely, in contrast with the conventional acousto-optics, which usually deals with weak, nonresonant acousto-optic nonlinearities due to the photoelastic effect, the proposed scheme exploits the excitonic states nearly resonant with the acoustic and optical fields simultaneously. As a result, the large-value, resonant acousto-optic nonlinearities can be realised. The resonant acousto-optic effect is particularly strong and well-defined for MC polaritons parametrically driven by a surface acoustic wave (SAW). If we realize the resonant acousto-optic effect at room temperature (ZnSe- or GaN-based microcavities) it will probably be a real breakthrough in the device acousto-optics.

The Cambridge team has developed and studied models of coherent exciton condensation in semiconductor microcavities. A new cooperative work of the Grenoble and Cardiff teams on the LO-phonon-mediated optical Stark effect for MC polaritons is in progress.

Task 3. Interface-photon-mediated interaction of self-assembled quantum dots (QDs)

(Lead team: Cardiff)

The Cardiff and Dortmund teams have studied, both theoretically and experimentally, the radiative corrections to the excitonic molecule (XX) state in GaAs-based microcavities. This work is extremely important for the project, because for the first time we have demonstrated the co-existence of the MC and interface, QW polaritons and the importance of the “hidden” optics associated with the evanescent light field of interface polaritons. Namely, we prove that the radiative corrections to the XX state, the Lamb shift (a real part of the energy) and radiative width (an imaginary part of the energy), are large, about 10%-30% of the molecule binding energy and definitely cannot be neglected. Furthermore, we demonstrate that the optics of excitonic molecules is dominated by the in-plane resonant dissociation of the molecules into outgoing 1-lambda-mode and 0-lambda-mode cavity polaritons. It is the latter decay channel, which deals with the short-wavelength MC polaritons invisible in standard optical experiments, that refers to the “hidden” optics of microcavities.

The Cardiff team is currently completing a theoretical study of the interface-photon-mediated interaction of self-assembled or surface-deposited QDs (the Network YR, PhD student C. Creatore is involved in this study). The conventional optics of interface (self-assembled) or surface-deposited (colloidal) quantum dots (QDs) deals with the pump and signal light associated with bulk photon modes. However, a long-distance coupling between the dipole-active electronic states (excitons) in QDs also occurs by means of in-plane propagating interface photons. The interface light is localized in the z-direction (the structure growth direction) and invisible at macroscopic distances from the nanostructure. Thus the interface photons contribute to the total optics of in-plane distributed QDs in a “hidden” way. Thus we develop the interface, quasi-two-dimensional optics and to describe how the QDs communicate via interface photons. The microscopic approach we use deals with in-plane randomly-distributed QDs (Poisson statistics) which are inhomogeneously broadened in energy (Gaussian statistics). The calculated eigen-spectrum of the Hamiltonian “bulk photons + QDs” allows us to classify the eigenstates in terms of the rapidly decaying modes (“radiative states”) and relatively weakly decaying modes (“interface photon-mediated QD states”). Furthermore, a very particular design of microcavities for the evanescent light field was proposed and developed. The first experiments on visualization of the interface light field associated with self-assembled InGaAs QDs have been performed by the Lund team

The Cardiff team also studies analytically and model numerically, partly in cooperation with the Cambridge team, relaxation and photoluminescence dynamics of long-lived indirect excitons in GaAs/AlGaAs coupled quantum wells.

Task 4. Disorder and decoherence effects in the optical response of photonic structures

(Lead team: Cambridge)

The Cambridge team currently deals with modelling of the crossover of a coherent polariton condensate to a semiconductor laser. Pair-breaking processes from non-equilibrium physics are found to be the dominant reason (rather than temperature or density constraints) for the suppression of coherence. Prediction of the shape of the angular emission of light from a coherent condensate of both excitons and separately polaritons in a trap is made. This is relevant to the emission spectra measured by the Grenoble group. The crossover from polariton Bose condensation to a coherent condensate driven by photon coupling is analysed. This establishes the regime where optical coupling is the dominant mechanism, which in practice covers the physical regime of parameters.

The Cambridge team has also studied the behavior of a system that consists of aphoton mode dipole coupled to a medium of two-level oscillatorsin a microcavity in the presence of decoherence. Two types of decoherence processes, which are analogous to magneticand nonmagnetic impurities in superconductors, have been analysed. In addition, modelling of electron and hole transport have been performed for optically-excited coupled quantum wells.

The Cardiff and Cambridge teams have also started a theoretical study on disorder-induced change of the wavevector-frequency boundary between the interface and bulk photon modes resonantly coupled with QW excitons.

Work of the Cambridge team on coherent excitons in microcavities is directly in response to experiments of Grenoble and Dortmund groups. Using real and k space imaging the Grenoble team has obtained the first evidence of localized polaritons in semiconductor microcavities (international state-of-the-art). The localization energy amounts to about 1 meV. A clear spectral blueshift can be observed with increasing excitation density, most probably as a result of Coulomb interaction between localized carriers.

Task 5. Spectroscopic methods for detection of interface light

(Lead team: Lund)

The Lund team has recently investigated random telegraph noise (RTN) in individual InP quantum dots in GaInP where the photoluminescence is modulated and discovered a new type of noise. The "old" type of RTN consisted of a modulation of the intensity of the photoluminescence. The newly discovered noise consists of a shift of the emission lines without any change of the total emission intensity. The Lund group has also developed a comprehensive theory of which correlation functions can occur for electrons and also for bosons. This is important for the calculation of few-particle effects in quantum dots and quantum wires. Furthermore the Lund team has investigated the effect of etching on individual quantum dots where clear effects of strain relief have been observed when the capping layer is removed from the dots. The strain effects on capped quantum wires have been studied where the capping layer strains the core of the wire. These studies have been perfomed on level of individual quantum wires (consisting of a GaAs core and a GaInP shell). All the above research works directly relate to the experimental search for the interface light field associated with excitons in QDs, QWells and QWires (see Task 3). The latter experiments are still in progress. Currently, the Lund group is particularly concentrated on the observation of the exciton-mediated interface light field and fast diffusions of excitons in quantum wires. Future plans involve more investigations of the possibility of finding an exciton crystal in a quantum wire, which has been predicted by A.L. Ivanov in 1993.

The Lund team has observed that the emission lines from single InP quantum dots shift with temperature, where the shift is dependent on the energy position of the lines in a systematic way. The Lund team has also calculated the basic electronic structure of most III-V quantum dots grown on most III-V substrates, as outlined in Task 5. In addition the team has etched out dots in predefined places in pillars. The magnetic field effects and the formation of Landau levels in individual InP quantum dots are planned to be investigated within “Lund-Grenoble” cooperation.

Task 6. Materials Characterisation

(Lead team: Crete)

The Paderborn team has grown and sent to FORTH for Cross-section Transmission Electron Microscopy (XTEM) three quantum dot samples (#1031, 1035, 1036) prepared by a Stranski-Krastanow process, during the period of Dr H. Ouacha (a Network YR, Crete team) secondment to Paderborn. These growths are part of the activities planned within Task 7 (Paderborn team) while the analyses are part of the activities within Task 6. The highlight of these results: (a) The QDs were formed in all the specimens. However the best well defined dots were observed in the specimen 1036, (b) The QDs were not periodic laterally or along the growth, (c) Extended defects were not observed between the QDs and the matrix, and (d) In the specimen 1035, ZnSe areas have the hexagonal structure.

In addition, the Crete group will perform a high-precision XTEM characterization of structures with self-assembled GaInAs QDs for the optical experiments of the Cardiff team on the long-distance photon-mediated interaction of QDs.

Task 7. Formation principles of II-IV Quantum Dots in microcavities

(Lead team: Paderborn)

The main scientific result of the Paderborn team was the deveplopment and MBE-preparation of ZnSe-based microcavities with CdZnSe-Multi-Quantum Wells, which have shown large Rabi-splitting at room temperature. We have demonstrated for the first time that polariton devices, based on II-VI microcavities, can potentially operate at room temperature (normal) conditions. The optimisation of the Bragg mirror processing was the most crucial pre-request to achieve this result extremely important for possible technological applications. The samples were optically characterized by the Grenoble group. The YR N.Rousseau is fully integrated into this research work. In collaboration with the Crete-Team cross-section Transmission Electron Microscopy (XTEM) studies have been performed, to understand in more detail the formation kinetics of self-organized grown CdSe quantum dots. In these experiments Dr. H. Ouacha (a Network YR, Crete team) was directly involved in the MBE sample preparation. The results of these experiments allow us to integrate CdSe-QD-layers in ZnSe-microcavity structures, according to the working plan.

In turn, the Grenoble team, using a new MBE process, has succeeded in obtaining a clear Stranski-Krastanov growth mode for self-assembled CdTe and CdSe QDs. For both types of QDs, no thermal activation of confined carriers is observed up to 150-200K. The Grenoble group has also developed the technology for a hybrid microcavity based on dielectric mirrors and ZnSe-based microcavities grown on GaAs substrates. This know-how will be shared with Network partners.

Some preliminary work on room-temperature polaritons in ZnSe-based microcavities parametrically driven by a surface acoustic wave is planned within “Paderborn-Cardiff” collaboration, in order to study the resonant acousto-optic effect proposed by the Cardiff and Canbridge teams.

As detailed above and discussed in Subsection B.3 (Breakdown of Tasks and Milestones), our PMP Network project has definitely already advanced the international state-of-the-art.

A.2Joint Publications and Patents

List of the PMP-Network publications

[1] N. Le Thomas, E. Herz, O. Schöps, and U. Woggon, “Spectroscopy of single CdSe nanorods: Fine structure and polarization properties”, submitted to Physical Review Letters.