The Winners of the SPS Awards 2010

The SPS award committee, presided by Prof. Hans Beck (Uni Neuchâtel) had this year again a hard job to choose among the 7 candidates the three ones winning the renowned awards. All the works submitted showed an excellent quality and were of high scientific level.
The three finalists had the opportunity to give a 15 min talk at the annual meeting in order to present their outstanding work in more detail. They are presented below.

(Laudationes written by Hans Beck, abstracts written by the respective authors)

Konstantinos G. Lagoudakis got a BSc in Physics, with grade A, at the University of Athens in his home country, and a MSc, with distinction, in Optics and Photonics at the Imperial College in London. For his PhD work at the Ecole Polytechnique Fédérale in Lausanne he has studied the fundamental properties of Bose-Einstein condensates of exciton polaritons.

Working on the hot topic of Bose-Einstein condensation of polaritons in microcavity systems, Konstantinos has shown his outstanding capabilities, both in mastering difficult experimental techniques and in developing a deep theoretical understanding. First, he has demonstrated the appearance of vortices in the condensate, which can be detected through a "fork-like dislocation" in the near field interferometry images of the condensate. Their existence proves the superfluid nature of the condensate. He then went one step further and realized the first clear observation of half-quantum vortices in this system with the help of polarization-resolved interferometry, real-space spectroscopy and phase imaging. This particular phenomenon, that had been predicted theoretically, is based on the fact that the polariton superfluid is characterized by a two-component order parameter, due to the spin of the condensing particles.

Attributing the General Physics prize to Konstantinos Lagoudakis the Swiss Physical Society honors an outstanding and brillant young physicist who has produced important new results at the forefront of his research field. His publications have been widely appreciated. He has already new experimental findings to present, namely the ultrafast apparition of vortices and the observation of Josephson oscillations in the polariton condensates.

Observation of half-quantum vortices in an exciton-polariton condensate

The elementary excitations of spinor superfluids are non standard vortical entities that carry fractional vorticity: the commonly known half quantum vortices (HQVs). Contrary to the most usual case of singly quantized vortices where the phase is rotating by 2π around the core, HQVs are characterized by a π rotation of the fluid phase and a π rotation of the fluid spin when circumventing the vortex core. They have been predicted theoretically in the late seventies by the pioneering work of Volovik and Mineev(1976) and Cross and Brinkman(1977) [1]. In the dilute atomic gas Bose Einstein Condensates community, spinor condensates are by now established in several laboratories. They constitute the ideal systems to create and investigate half quantum vortices but the required complexity to excite and capture them has not allowed for their direct observation so far. The only signs of these unconventional vortices have only been reported in high T_c superconductor grain boundaries [2]. Exciton polariton condensates in the solid state are alternative systems in which the observation of half quantum vortices was thought to be possible. The nature of the polariton quantum fluid topology, which depends on the local polarization splitting, can be either that of a scalar or that of a spinor fluid depending on whether the splitting is large or almost zero respectively. In this experimental work we have set out to go beyond the celebrated case of vortices in scalar polariton quantum fluids and we have managed to provide the first experimental evidence of half quantum vortices in a spinor quantum fluid by taking advantage of the spinor nature of polariton condensates.

[1] G. E. Volovik and V. P. Mineev JETP Lett. 24, 561-563 (1976)
[2] J. R. Kirtley et al. Phys. Rev. Lett. 76, 1336 (1996)

Erik van Heumen is Dutch. Having obtained his MSc at the University of Leiden he came to the Département de Physique de la Matière Condensée in Geneva for a PhD thesis, dedicated to a better quantitative understanding of high temperature superconductivity. He is now back in his home country, at the University of Amsterdam. Being project leader of a research project entitled "Superconductivity enters the iron age: testing quantitative theories of superconductivity in iron pnictide high T_c superconductors" he fully profits from his new insight into the phenomenon of superconductivity.

High temperature superconductivity is still a major challenge for theoretical physicists. Whereas most specialists in the field believe that the strong electronic correlations necessitate modelling that is different from what has been developed for low temperature materials, Erik has based his considerations on the well established strong coupling formalism that builds a link between the optical conductivity in the normal state, the electronic self-energy, the intensity of the bosonic coupling producing Cooper pairs and the superconducting critical temperature. Using his measured optical data for materials with doping levels spanning the range from underdoped to overdoped, he gets information about the "glue function" that provides the bosonic coupling leading to pairing. He then shows that the resulting critical temperatures follow the dome-shaped dependence on doping with values that are only a factor 2 to 3 above the experimental values.
Given that various other models are not consistent with the relation between optical spectra and transition temperature this is a major step forward in our understanding of high temperature superconductivity. Erik also shows that the frequency dependence of the "glue function" is not compatible with the traditional electron-phonon coupling.

Erik is not only an extremely skilled and powerful experimentalist, but his work also certifies his profound understanding of theoretical analysis and modelling. Let us hope that his deep insight will also lead to an identification of the "glue" that creates electron pairs in the high T_c materials !

Towards a quantitative understanding of the high T_c phenomenon

Conventional wisdom, based on observations in simple metals like lead, is that superconductivity arises from the interaction between electrons mediated by the vibrations of the crystalline lattice. For these materials the theoretical description of the electron-phonon interaction is known as the Migdal-Eliashberg theory. For a number of reasons the pairing mechanism of high T_c superconductors is believed not to be mediated by electron-phonon interaction: (i) The transition temperatures are extremely high. (ii) The electrons in these materials are strongly correlated resulting inter alia in strong magnetic fluctuations and a doping controlled transition into a Mott-Hubbard insulating state. (iii) The condensed pairs have d-wave symmetry. While Migdal- Elaiashberg can in principle be generalized to incorporate interactions mediated by magnetic fluctuations, it can not describe the Mott-Hubbard insulating state and related phenomena. The validity of a generalized Migdal-Eliashberg approach in part of the phase diagram should therefore be tested experimentally. This has been the main objective of this work. We have used a generalized Migdal-Eliashberg approach to analyze the optical spectra of several members of the cuprate family, spanning a range of dopings and critical temperatures [1,2,3]. The overdoped cuprates are thought to be close to Fermi liquids where this formalism should work reasonably well. We found that this is indeed the case, but the same formalism also works surprisingly well for the compounds around optimal doping where the highest critical temperatures are obtained. It works so well that if we analyze the optical spectrum at room temperature, it allows for a quantitative prediction of the optical spectrum at much lower temperatures. At the same time it allows us to extract the energy scales and strength of the interactions felt by the electrons in these materials. Using this information we can calculate the critical temperature expected for such interactions and we find that it is more than strong enough to explain the high Tc of these materials. Although we have not yet pinpointed the exact source of the interactions, our work shows that they originate from the correlated electron motions. Therefore this work puts stringent conditions on possible theoretical frameworks for the high T_c cuprates.

[1] E. van Heumen, “Towards a quantitative understanding of the high T_c phenomenon” Ph D thesis, Université de Genève (2009)
[2] E. van Heumen, E. Muhlethaler, A.B. Kuzmenko, H. Eisaki, W. Meevasana, M. Greven and D. van der Marel, Phys. Rev. B 79, 18451 (2009)
[3] E. van Heumen, W. Meevasana, A.B. Kuzmenko, H. Eisaki and D. van der Marel, New Journal of Physics, 11, 055067 (2009)

Sandra Foletti has studied physics at the two Federal Schools of Technology, two years in Lausanne and up to the MSc degree in Zürich. She did the first part of her experimental PhD work on spin qubits in GaAs double quantum dots at the Weizmann Institute of Science in Israel, before she moved to Harvard University for the second part of her thesis work.

Quantum computation and information processing – Sandra’s research field – is a challenging application of the basic laws of quantum mechanics. It aims at implementing secure information transfer and at providing methods for solving complex computational problems. It is well known that the solid state environment limits the necessary coherence time of the quantum bits that should do the work for the user. Sandra has shown that even in GaAs, where each atom bears a nuclear spin, this problem can be overcome. Her work focuses on the dynamic coupling and decoupling between the coupled spins of two electrons in a double quantum dot structure and the surrounding bath of nuclear spins. Her work has litterally revolutionized the field ! By decoupling the qubit spin from its environment she has succeeded in extending the coherence time by 3 orders of magnitude, reaching several hundreds of microseconds. Taking advantage of the hyperfine interaction between electrons and nuclei she has generated a magnetic field gradient which is needed to achieve full control over the individual two-electron spin qubit.This was a non-trivial experimental task that Sandra has mastered beautifully.

Therefore, Sandra’s work, although in principal being done in the framework of basic quantum physics, represents a striking breakthrough in the worldwide effort to use spin qubits in solid state systems with sufficiently long coherence time. This is an important step forward towards the application of fundamental physics to quantum computing, which is a promising future tool for the community of computer users. Sandra therefore fully merits the Swiss Physical Society prize for applied physics.

Universal quantum control of two-electron spin quantum bits using dynamic nuclear polarization

Spin based quantum bits have been considered as promising candidates for quantum computation due to their potential for scalability, miniaturization and control. In this work [1] we utilize a logical spin-based quantum bit where the two-level system is provided by the m_z = 0 singlet and triplet states of two electrons confined to a double quantum-dot structure. In comparison with the single electron spin qubit, this approach has the advantage that it is protected against uniform magnetic field fluctuations and more importantly, that all manipulations can be done electrostatically.
Universal control of the quantum bit requires arbitrary rotations around two axes of the Bloch sphere: one axis, demonstrated in earlier work, corresponds to rapid exchange of the two electrons. Here we demonstrate control over the second rotation axis whose physical implementation is a magnetic field gradient across the two dots. We take advantage of the hyperfine interaction that couples the electrons to approximately a million nuclei of the host GaAs lattice to create such a gradient.
To operate our two-electron spin qubit, we first generate a field gradient by dynamical nuclear polarization of the underlying lattice, employing pulse schemes that transfer angular momentum and thus magnetic moment to the nuclei. With such pumping sequences, we can achieve gradients in excess of 200 mT, that can be sustained for 30 min and beyond.
Once a nuclear gradient of the appropriate magnitude is created, the qubit can be manipulated. We demonstrate full quantum control of our qubit by reconstructing the evolution of the state of the qubit within the Bloch sphere through quantum state tomography. The typical manipulation times of our qubit are in the nanosecond range. In combination with coherence times of ≈ 100 μs [2], this allows for at least 10⁴ quantum operations which are required for fault tolerant error correction schemes.
Our gradient generation through the hyperfine interaction demonstrates that the nuclear spin environment may actually serve as a valuable resource for spin qubits. Moreover, it lays the foundation for future development of additional control schemes that are needed in order to prolong the inhomogeneous decoherence time T^∗₂ which ultimately limits the qubit fidelity [3].

[1] S. Foletti, H. Bluhm et. al., Nature Physics 5, 903 (2009)
[2] H. Bluhm, S. Foletti et al., arXiv:1005.2995
[3] H. Bluhm, S. Foletti et al., arXiv:1003.4031