Self-assembly is key to living cells, where it brings well-adjusted parts together into functional biological structures. In rarer, pathological cases, ill-fitting proteins also aggregate and form fibers involved in Alzheimer's and other diseases. While functional self-assembly is widely studied, the physical principles governing ill-fitting self-assembly remain largely unknown Our current understanding of self-assembly revolves around examples involving simple, relatively symmetrical particles. Here we consider the opposite limit of very complex, ill-fitting particles, whose aggregation generates geometrical frustration. To escape this frustration, our early theoretical results suggest that they tend to form fibrous aggregates. According to this putative “dimensional reduction” principle, collections of complex particles generically behave differently than collections of simple ones. Indeed, while increasing the attractive interactions in the latter typically induces a transition from a dilute (gas-like) phase to a dense (liquid or solid) phase, we propose that the former should generically present an additional, intermediate regime where fibers (or planes) form. We will investigate the effects of geometrical frustration on self-assembly, and establish how universal dimensional reduction actually is. We will thus develop new theoretical methods to tackle complex particles in elasticity and statistical mechanics, as well as experimentally probe colloidal and protein self-assembly using high-resolution 3D printing and X-ray scattering. Our work will reveal new organizational principles for matter, possibly as broadly applicable as the very concept of crystallization. Beyond providing a deeper understanding of biology and disease, these principles could provide guidelines for engineering objects at the nano- and microscale and lead to a better mastery of processes involved in drug manufacturing and protein crystallography.

The question of dissipation in quantum many-body systems is a subject of considerable interest, for fundamental reasons – associated with the understanding of how many-body quantum correlations survive in presence of decoherence – and for the development of realistic platforms for quantum technologies. In parallel to the question of how dissipation may harm quantum coherences, it has also been suggested to use dissipation in order to create interesting many-body systems – a concept that generalizes that of optical pumping. The idea of dissipative-state engineering is to introduce a controlled coupling to an environment that can induce correlations or symmetries, and can produce or stabilize quantum correlated states. Our project is to combine experimental and theoretical efforts, to explore the dissipative engineering of collective spin states that are relevant for quantum sensing and quantum simulation. The platform is an ultracold-atom quantum simulator that realizes the Hubbard Hamiltonian in an optical lattice. We will use strontium 87 atoms, a fermionic spin-9/2 species exhibiting a SU(N) spin symmetry – corresponding to an invariance of the system when permuting any two spin states within N spin states, where the number N can be controlled at will from 2 to 10. Based on prior theoretical work, two-body losses in SU(2) or SU(3) systems provide an opportunity to engineer highly-entangled generalized Dicke states, that are of interest to quantum sensing. Our project is to first experimentally verify this possibility, and then study its generalization for SU(N>3) systems, both from the experimental and the theoretical standpoint. For this we will make use of an exceptional tunability that is offered by the narrow lines of strontium atoms, that are of interest to the precision measurement community. In practice, we will use these transitions to engineer losses via photo-association. By simply tuning the magnetic field, these losses can be made either spin-sensitive or spin insensitive, which allows full control over both the strength and the spin selectivity of dissipation. We will also develop a new scheme to drive at will two-body or three-body losses. Due to the local character of losses and the anti-symmetric nature of the few-body wavefunction, those two- and three- body losses specifically target respectively SU(2) or SU(3) singlet states. These capacities will enable an investigation of dissipative quantum dynamics with a full control of the symmetry of the Hamiltonian, from SU(2) to SU(10), and of the symmetry of dissipation. When losses are low, we generally expect the system to be driven into stationary states that favour triplet correlations. We will study these states, and perform Ramsey sequences to characterize their metrological quality. Furthermore, we will also perform the first study of dynamics of SU(N) lattice gases in the regime of strong dissipation, where the quantum Zeno effect is at play, so that long-lived, strongly correlated many-body states should emerge, with similarities to those that arise in the quantum t-J model at low energy. In both regimes, we will study the robustness of dynamics and of the stationary or metastable states in presence of inhomogeneities, in order to assess the practical usefulness of the highly symmetric novel quantum many-body states that can spontaneously arise in these systems. Therefore, outcomes of this project can find applications to quantum sensing and quantum simulation, that would be directly relevant to alkaline-earth-like species that are currently at the core of optical clocks and atom interferometers.

Fascinating phenomena emerge from the appearance of far-from-equilibrium thermodynamic states, from active matter to protein folding dynamics, including non-trivial heat flows. Nevertheless, the complexity of these phenomena makes their fundamental understanding difficult. In this context, the FENNEC project aims at investigating experimentally and theoretically far-from-equilibrium nanothermodynamics through the careful engineering of colored heat baths of an optically levitated particle. First, it proposes to develop experimental and theoretical tools to generate and characterize out-of-equilibrium dynamics of the systems. Then, it addresses the stochastic energetics of the particle in controlled out-of-equilibrium states or during state-to-state transformations, introducing protocols that could improve and optimize such transformations. Finally, it proposes to take advantage of the out-of-equilibrium system dynamics to investigate and improve the efficiency of heat and particle transport. The FENNEC project will thus provide a unique experimental platform associated with innovative theoretical frameworks for studying far-from-equilibrium systems associated with colored heat baths. It will pave the way for developing efficient nanosystems for transport and sensing.

The project brings together experimentalists and theoreticians from Toulouse (LCAR UMR 5589, LPT UMR5152) and Orsay (LPTMS 8626). We propose a new approach to cold atom experiments, where a controlled complexity renders a single system very versatile. The complexity will be built progressively in the course of the project upon the interplay between different physical phenomena: tunnel effect, chaos, quantum localization and interaction effects. It will allow us not only to investigate many new regimes of fundamental physics (from disordered to strongly correlated systems) but also to create innovative tools to manipulate cold atoms. To this end, we will focus on a system composed of ultra-cold atoms trapped in a one-dimensional optical lattice, whose amplitude and phase are engineered in time. By time-modulating the lattice amplitude, the system will be driven periodically in a non-perturbative way, leading to the dressing of the lattice by chaos: regular islands will be embedded in a chaotic sea. The central physical effect that will be used is chaos-assisted tunneling (CAT) occurring in such system. CAT presents resonances allowing to tune over orders of magnitude the tunneling rate between two regular islands, over a short range of parameter (e.g. the frequency of lattice modulation). We will first demonstrate these strong variations and fully characterize the distribution of resonances, which have not been observed so far with cold atoms. Then we will build and control step by step the complexity by adding quantum localization effects and atom-atom interactions. These new ingredients have never been considered in the context of CAT. Their inclusion constitutes a theoretical and experimental challenge. The transport between regular islands can be tuned in our system from diffusive to localized, leading to new regimes of tunneling which will be characterized by their tunneling rate distributions. In particular we want to use CAT as a clear signature of multifractality, an intriguing characteristic of the localization transition which remains elusive experimentally. We also want to build a fundamental understanding of the effects of interactions on CAT. We will start from the non-linear regime associated with large number of atoms per lattice site to revisit e.g. the Josephson effect in the chaotic regime. We will then access the strongly-correlated regime by using the experimental apparatus with a 3D lattice. Based on these fundamental outcomes, we will develop new techniques to manipulate cold atoms in optical lattices. The dressing by chaos offers indeed fascinating possibilities for long-range hoppings in dynamical lattices. In the presence of interactions, CAT enables also to tune the ratio between hopping amplitude and interaction strength through resonances reminiscent of Feshbach resonances. These new tools, as they rely on the universal physics of CAT, are independent of the atomic species. They can also be easily transposed in 2D and 3D. They will allow controlling complex cold atoms systems in an unprecedented manner as well as reaching physics models and regimes that are currently inaccessible. The three partners include a cold atom experimental team, a theoretical team specialist of Anderson localization and working at the interface with experiments and a theoretical team expert in CAT and many-body effects. The partners bring key-skills to the project based on their complementary domains of expertise.

Our project deals with the creation and the study of a new type of quantum liquid droplets in yet dilute ultracold atomic samples of Potassium 39. Their existence emerges because of quantum fluctuations in a mixture of two Bose-Einstein condensates for which the overall mean-field energy cancels because of attractive interspecies interaction. The system offers a unique opportunity to quantitatively study beyond-mean field effects dominating the dynamics of the system. We plan to focus on the quasi one-dimensional (1D) geometry where the physics of droplets is most counter-intuitive. Key observations will be the self-trapped nature of droplets as well as their flat bulk region. Quantitative studies of the collective excitations will be performed. We will also approach the strongly interacting regime, which corresponds to lowering the density in 1D. Furthermore, we plan to demonstrate radio-frequency dressing as a novel method to control interaction in atomic Bose-Einstein condensates. It will permit not only the control of two-body interactions but also of three-body interactions, offering an alternative mechanism for the stabilization of dilute atomic droplets.