Composition du Jury

-         Prof. Gerhard Rempe, Max Planck Institüt für Quantenoptik

-         Prof. Xavier Marie, Laboratoire de Physique et Chimie des Nano-Objets

-         Prof. Thomas Coudreau, Laboratoire Matériaux et Phénomènes Quantiques

-         Dr. Audrey Cottet, Laboratoire Pierre Aigrain

-         Prof. Agnès Maître, Institut des Nanosciences de Paris

-         Prof. Philippe Grangier, Laboratoire Charles Fabry de l’Institut d’Optique

How can we make single photons interact? This question represents a major challenge for the development of quantum technologies. In this presentation, I will show that efficient light-matter interfaces can be developed in the solid state to address this issue, using semiconductor quantum dots embedded in optimized microcavity structures. This led to a number of achievements in the last decade, including :

- The engineering of an effective photon-photon interaction, leading to an optical nonlinearity at the single-photon limit.

- The demonstration of an efficient spin-photon interaction, using the spin of a confined semiconductor hole as a stationary quantum bit.

- The development of near-optimal sources of pure and indistinguishable single photons for quantum optics.

The perspectives for realizing deterministic quantum gates and fundamental quantum experiments, based on a new generation of spin-based devices, will also be discussed.

This HDR focuses on fundamental aspects of the nonlinear interaction between light and matter in active –III-V semiconductor– photonic crystals. The combination of large nonlinear optical susceptibilities with high quality factors and small optical mode volumes of micro/nano optical resonators leads to a wealth of complex physical phenomena, even at a single cavity level: optical bistability, neuron-like spiking in the form of optical excitability, self pulsing dynamics... Their observation in novel micro and nano resonators has been a challenge over the past fifteen years. My work after my PhD thesis, carried out at LPN (now C2N) from 2004, has been devoted to this quest.

Beyond the single cavity limit, photonic crystal platforms offer new opportunities to study light matter interaction in arrays of coupled cavities. The interplay between photon tunneling and nonlinearity in multi-well optical potentials –or photonic molecules– is at the heart of many recent developments in quantum and nonlinear optics. As a central part of this HDR report I will focus on a key phenomenon taking place in double well potentials: the spontaneous mirror-symmetry breaking, which we have demonstrated for the first time at the nanoscale using two coupled nanolasers. The system undergoes a bifurcation from delocalized to localized states in the wells, which are mirror images of each other. I will show that, in order to achieve this goal, a thorough optimization of the coupled cavity system is needed in terms of both photon collection and, more importantly, the control of inter-cavity coupling. For the latter we have proposed an original photonic crystal barrier engineering design that enables not only a large tuning of the coupling strength, but also the control over its sign.

Unexpectedly, we found that the photon number at the bifurcation is about ∼ 150, which could even be further reduced by increasing the spontaneous β-factor of the nanolasers. We can thus predict such transitions with few intracavity photons in future nanolaser devices. Ultimately, signatures of quantum correlations could also be expected in future nonlinear coupled nanocavity devices close to symmetry breaking transitions. 

Lors de l’exposé je présenterai les travaux récemment effectués en vue de réaliser une source laser dans la filière IV-IV, en particulier avec le germanium. L’absence d’une source optique efficace compatible avec les technologies CMOS est le principal verrou à lever pour permettre d’unifier les technologies de la micro- électronique avec les technologies de la photonique dans une seule et même filière de matériaux IV-IV. Basiquement le silicium et le germanium sont pénalisés, pour l’émission de lumière, par la nature indirecte de leur structure de bande. Le cas du germanium est assez particulier, sa bande de conduction en vallée L (bord de zone) se trouve à seulement 140meV au-dessous de la bande de conduction du centre de zone. Je présenterai des éléments de modélisation de la structure de bande et expérimentaux montrant que cette barrière peut être abaissée grâce à l’application de contraintes mécaniques en tension au Ge. Le régime de bande interdite directe dans le germanium peut d’ailleurs être atteint pour des niveaux de déformation biaxiale au-delà de 1.7%. Au C2N-IEF en collaboration avec le C2N-LPN nous avons développé une technologie basée sur le transfert de contraintes depuis des films ‘stressor’ de nitrure de silicium vers des résonateurs optiques à haut facteur de qualité en germanium, de type μ-disques, en vue de réaliser un dispositif laser. Une approche originale, basée sur la bi-encapsulation des couches de germanium par les films ‘stressor’, a permis d’atteindre le régime de bande interdite directe dans les μ-cavités de Ge.

Différents groupes de recherche ont montré par ailleurs que l’alliage du germanium avec l’étain permet également d’atteindre un régime de bande interdite directe pour des compositions typiquement de 7% d’étain lorsque le matériau est relaxé. En combinant l’ingénierie des contraintes, l’alliage du Ge avec l’étain et le confinement électronique dans des héterostructures SiGeSn/GeSn, de nombreuses perspectives sont actuellement ouvertes pour réaliser des sources laser performantes et compatibles CMOS. Ces sources devraient permettre de couvrir une gamme spectrale qui s’étend aux frontières entre le proche et le moyen infrarouge permettant ainsi de renforcer l’utilisation des technologies silicium pour les interconnexions optiques et les senseurs biochimiques.

The evolution of the transistor, from its first conception in the 1920s to its ultimate manifestation in today’s integrated circuits, tells in part the history of research in the device physics community. While research in the latter part of the 20th century often focused on developing new devices with novel materials, the approaching end of Moore’s law has ushered in a new emphasis on novel computing paradigms that push this community to focus on integrating new devices within what were once considered outlandish architectures. The research presented in this habilitation thesis defense follows along this main evolution in device physics. Starting with research exploring a novel silicon MOSFET, the Schottky barrier MOSFET, that in the 1990s was proposed as an alternative to conventional CMOS, my investigations of this device ranged from understanding the fundamental electronic transport over a wide range of temperatures. The most interesting results, nevertheless, are at low temperatures where transport involving single or small clusters of atoms modulate the transport allowing for spectroscopy, and observations of quantum interference. The discovery of high temperature superconductivity in 1986 in novel ceramic oxides ushered in a flurry of research on these and related materials. One remarkable discovery was colossal magnetoresistance in magnetic oxides in which the interplay between electronic, magnetic and structural degrees of freedom continues to give rise to research surprises. In this context, I explored the quantum corrections that arise in the model LaSrMnO3 manganite thin film and show how one can obtain the phase coherence length. Finally, as the community has migrated from exploring devices for possible applications towards novel architectures, the final part of defense will explore how novel computing paradigms. A first approach explores the use of conventional C-element circuit in a stochastic computing setting to demonstrate Bayesian inference. Next, I explore an analogy between second order phase transitions and overfitting in the extreme learning algorithm. A third project presents how a Schottky barrier nanowire dressed with redox molecules can be used as a memristor. Finally, I conclude with my research perspectives where I plan to exploit my knowledge of device physics and statistical mechanics to realize bio-inspired hardware.