Die Gewinner der SPG Preise 2022

Exceptional Topological Insulators

Following the success of topological phases in solid-state systems, non-Hermitian physics has recently attracted a lot of interest. One of the reasons is that most experimental platforms are in fact either accidentally or tuneably lossy, such that their effective description involves a non-Hermitian Hamiltonian. Non-Hermiticity does not only give rise to new bandstructure features such as point gaps or exceptional points but also enriches the world of topological phases.

Here we introduce the exceptional topological insulator (ETI) realizing a surface anomaly — akin to the three-dimensional topological insulator — that can only exist within the topological bulk embedding [1]. Like the single surface Dirac electron, the exotic surface state of the ETI cannot be regularized in purely two dimensions. It covers the bulk energy point gap as a single sheet of complex eigenvalues or with a single exceptional point. Even though it does not require any symmetry to be stabilized, we explain how this non-Hermitian topological phase can also be inferred using symmetry-indicators of the bulk Hamiltonian [2].

The ETI can be induced universally in gapless solid-state systems and metamaterials, thereby setting a paradigm for non-Hermitian topological matter. For instance, the ETI phase emerges in a Weyl semimetal, when quasiparticles at the two Weyl nodes acquire finite but distinct lifetimes. Such a scenario is a natural result of strong electron-phonon interaction, paving the road for a future material discovery.

[1] M. M. Denner et al. "Exceptional Topological Insulators”, Nat. Commun. 12, 5681 (2021)
[2] P. M. Vecsei, M. M. Denner et al. “Symmetry indicators for inversion-symmetric non-Hermitian topological band structures”, Phys. Rev. B 103, L201114 (2021)

A strong and electrically tunable optical resonance in a two-dimensional semiconductor

Atomically thin transition metal dichalcogenides (TMDs), such as molybdenum disulfide (MoS₂), strongly interact with light. Their optical properties are governed by excitons – electrons and holes bound by Coulomb attraction – that remain stable at room temperature. The ability to additionally tune their transition energies is essential for various interesting opto-electronic applications based on light emission, detection, modulation and manipulation. This can in principle be achieved via the quantum-confined Stark effect with an electric field applied perpendicular to the sample plane.

In a single layer MoS₂, absorption can reach up to 100 %, but the optical transition cannot be electrically tuned in such devices, as the excitons have essentially no out-of-plane dipole moment due to their confinement to a single layer [1]. A large out-of-plane dipole moment requires a vertical separation of the electron and hole. We have created a novel structure that shows optical transitions with strong absorption and wide tunability in the visible range [2]. Our device consists of a double layer of MoS₂ sandwiched between an insulator and top and bottom electrodes made from the electrical conductor graphene. The main idea is to use transitions based on electrons and holes that reside in different MoS₂ layers – so-called interlayer excitons. The vertical separation between the electrons and the holes gives rise to a static electric dipole. By applying a voltage to the outer graphene layers, we generate an electric field that tunes the absorption of the two MoS₂ layers. By adjusting the voltage applied, we can then select the wavelengths at which the electron-hole pairs are formed in these layers.

This research paves the way for new approaches to developing opto-electronic devices by combining the strong light-matter interaction of excitons in TMD monolayers with the high tunability of interlayer excitons in external electric fields.

[1] J. G. Roch, N. Leisgang et al., Nano Letters 18, 1070–1074 (2018).
[2] N. Leisgang et al., Nature Nanotechnology 15, 901–907 (2020).

Identifying the fundamental working mechanisms in superconducting transistors

The development of metallic superconducting transistors could enable new devices such as cryogenic signal routers, multiplexers, and frequency tunable resonators, for which no semiconductor counterparts exist. Recent experiments with metallic nanowire devices suggested that superconductivity can be controlled by the application of electric fields. In such experiments, critical currents were tuned and eventually suppressed by relatively small voltages applied to nearby gate electrodes, at odds with current understanding of electrostatic screening in metals. We demonstrated that this effect is linked to gate currents below 100 fA at the onset of critical current suppression in our devices [1]. Employing novel device geometries, we disentangled the roles of electric field and electron-current flow. Our results showed that the suppression of superconductivity does not depend on the presence or absence of an electric field at the surface of the nanowire but requires a current of high-energy electrons [2]. The suppression is most efficient when electrons are injected into the nanowire, but similar results are obtained when electrons are passed between two remote electrodes at a distance d to the nanowire (with d in excess of 1 µm). In the latter case, high-energy electrons decay into phonons which propagate through the substrate and affect superconductivity in the nanowire by generating quasiparticles. We showed that this process involves a non-thermal phonon distribution, with marked differences from the loss of superconductivity due to Joule heating near the nanowire or an increase in the bath temperature.

[1] M. F. Ritter et al., Nature Communications 12, 1266 (2021).
[2] M. F. Ritter et al., Nature Electronics 5, 71 (2022).

Light-mediated strong coupling between a mechanical oscillator and atomic spins 1 meter apart

Engineering strong interactions between quantum systems is essential for many phenomena of quantum physics and technology. Typically, strong coupling relies on short-range forces or on placing the systems in high-quality electromagnetic resonators, which restricts the range of the coupling to small distances. At macroscopic distances, however, coherent bidirectional coupling has remained a challenge because it becomes increasingly difficult to isolate the systems from the environment.

To realize long-distance Hamiltonian interactions, we connected two systems by light in a loop geometry [1]. The systems can exchange photons through the loop, thereby realizing a bidirectional interaction. Moreover, the loop leads to an interference of quantum noise introduced by the light field. For any system that couples to the light twice and with opposite phase, quantum noise interferes destructively and associated decoherence is suppressed. In this way, the coupled systems are effectively closed to the environment, even though the light field mediates strong interactions between them. Since the coupling is mediated by light, it allows systems of different physical nature to be connected over macroscopic distances in a reconfigurable way.

In our experiments, we used a free-space laser beam to strongly couple a spin-polarized atomic ensemble and a micromechanical oscillator held in separate vacuum chambers [2]. The coupling is highly tunable and allows us to engineer beam-splitter and parametric-gain Hamiltonians, observing normal-mode splitting and two-mode thermal noise squeezing, respectively. Moreover, we switch from Hamiltonian to dissipative coupling by applying a phase shift to the light field within the loop. This high level of control in a modular setup demonstrates the versatility of light-mediated interactions. Our method gives access to a comprehensive toolbox for hybrid quantum systems and opens up a range of new opportunities for quantum control and coherent feedback networks.

[1] T. M. Karg, B. Gouraud, P. Treutlein, and K. Hammerer, Phys. Rev. A 99, 063829 (2019).
[2] T. M. Karg, B. Gouraud, C. T. Ngai, G.-L. Schmid, K. Hammerer, and P. Treutlein, Science, 369, 174 (2020).

Solving quantum chemistry problems with first generation digital quantum computers

Predicting reaction pathways, rates, and kinetics, as well as molecular structures and dynamics, is essential for supporting the design of new materials, drugs, catalysts etc. However, this requires solving the Schrödinger equation (SE), which governs the behavior of quantum mechanical systems and whose dimension grows exponentially with the number of particles. Its direct resolution is hence limited to the smallest molecules. Although there exists a plethora of methods aiming to find approximate solutions to the SE, a trade-off between accuracy and efficiency persists. Because of these difficulties, quantum computing appears today as an interesting tool for quantum chemistry. Indeed, quantum mechanical wavefunctions could be more efficiently prepared using qubits which are quantum mechanical system.

My PhD work has focused on the design of quantum algorithms for the resolution of the molecular SE. In particular, I have adapted the equation of motion (EOM) approach to a quantum algorithm for the calculation of electronic excited states. I have then worked on extending electronic structure quantum methods to less studied but equally important vibrational structure problems [1]. Finally, I focused on the simulation the difficult non-adiabatic quantum dynamics that couple nuclear and electronic motion in molecules. To achieve this, I worked in a rather unexplored grid encoding of the problem in the quantum computer and showed how to obtain the quantum circuit for simulating these dynamics for a model Hamiltonian with product formulas [2].

[1] P. J. Ollitrault, et. al. (2020). Hardware efficient quantum algorithms for vibrational structure calculations. Chemical Science, 11(26), 6842–6855.
[2] P. J. Ollitrault, et. al. (2021). Molecular quantum dynamics: A quantum computing perspective. Accounts of Chemical Research, 54(23), 4229–4238.