Periodic Reporting for period 1 - TOQUAM (Topological Quantum Gas Microsope)
Reporting period: 2020-07-01 to 2022-06-30
The fundamental properties of matter are generally well described in terms of the underlying symmetries of the system. However, the discovery of the quantized Hall conductivity marked the beginning of a new classification of phases of matter and phase transitions of materials: Topological insulators (TIs). At present, band theory models describe the physics of non-interacting topological insulators; nevertheless, the rich and still not well-understood connection between topology and interactions represent a new paradigm in modern condensed matter physics. In our project, we have developed an experimental setup to bridge this gap.
In this action, we have built the first Cesium quantum gas microscope in the Hubbard regime. Thanks to a novel transport scheme and a high numerical aperture objective we were able to obtain fluorescence images of single Cs atoms. As a proof-of-concept, we reached the strongly correlated regime by controlling the atomic interactions of our atomic cloud after loading the system in a square optical lattice (383.5 nm periodicity). In this fully controllable environment we have probed the superfluid-to-Mott insulator transition. To reach single-site resolution at the short lattice wavelength, we implemented machine learning techniques to extract the lattice occupation, overcoming the optical resolution limit. Overall, the results associated with this action will allow observing the physics of phase transitions, edge states, and site-to-site density and spin correlations. Our research establishes the starting point to study fascinating, yet poorly understood, topological states of matter in the presence of interactions. These systems are prominent building blocks for fault-free topological quantum computing.
In this action, we have built the first Cesium quantum gas microscope in the Hubbard regime. Thanks to a novel transport scheme and a high numerical aperture objective we were able to obtain fluorescence images of single Cs atoms. As a proof-of-concept, we reached the strongly correlated regime by controlling the atomic interactions of our atomic cloud after loading the system in a square optical lattice (383.5 nm periodicity). In this fully controllable environment we have probed the superfluid-to-Mott insulator transition. To reach single-site resolution at the short lattice wavelength, we implemented machine learning techniques to extract the lattice occupation, overcoming the optical resolution limit. Overall, the results associated with this action will allow observing the physics of phase transitions, edge states, and site-to-site density and spin correlations. Our research establishes the starting point to study fascinating, yet poorly understood, topological states of matter in the presence of interactions. These systems are prominent building blocks for fault-free topological quantum computing.
In this project, we study Cesium atoms in the strongly-correlated Hubbard limit. This is achieved after loading an ultra cold atomic cloud into an optical lattice. This optical lattice allows mimicking physics similar to electrons moving in a real material. Here we can control perfectly the dynamics and initial state of the system. In particular, after loading the atomic cloud into this periodic array, we have observed a phase transition so-called: superfluid to Mott insulator. In short, by increasing the repulsive atomic interactions in the system, we have been able to stop the "flow" of atoms along the optical lattice and cross from a superfluid to an insulating state. It is worth mentioning that such a transition is a clear manifestation of the quantum properties of the system.
The lattice occupation in both phases were observed using fluorescence imaging of single atoms. To perform in situ fluorescence imaging we have overcome two main challenges:
1) Protect the quantum coherence of the system. After a long and fast distance transport, it was required to keep the system at a very low temperature ( < 50 nK).
2) Capture fluorescence photons while keeping the atoms in their final position, suppressing thermal tunneling events. This defines the fidelity of our detection.
After a delicate work where several technical noise sources have been identified and eliminated, we successfully reached low temperatures and preserve the quantum coherence of the system. The dominating noise sources include intensity noise from the optical lattice and magnetic field noise from the power supplies and external fields. Compared to other alkali atoms, these tasks were challenging to achieve due to the low energy scales required to control the Cesium atoms.
In order to perform fluorescence imaging, we have performed efficient laser cooling to prevent undesired dynamics of the cloud during the imaging process (thermal hopping). The cooling is implemented in a deep optical lattice. Some of the emitted photons are captured in a high-sensitivity camera during the cooling time.
In this approach, we have built an optical path to control the depth of our optical lattices in a regime where the atoms cannot escape and move while emitting photons. While this is a standard technique in current quantum gas microscopes, very little is known about implementing this technique in Cesium atoms. To fulfill this task, we have quantitatively studied the trap depth dependence vs. the molasses cooling efficiency and the imaging fidelity.
The lattice occupation in both phases were observed using fluorescence imaging of single atoms. To perform in situ fluorescence imaging we have overcome two main challenges:
1) Protect the quantum coherence of the system. After a long and fast distance transport, it was required to keep the system at a very low temperature ( < 50 nK).
2) Capture fluorescence photons while keeping the atoms in their final position, suppressing thermal tunneling events. This defines the fidelity of our detection.
After a delicate work where several technical noise sources have been identified and eliminated, we successfully reached low temperatures and preserve the quantum coherence of the system. The dominating noise sources include intensity noise from the optical lattice and magnetic field noise from the power supplies and external fields. Compared to other alkali atoms, these tasks were challenging to achieve due to the low energy scales required to control the Cesium atoms.
In order to perform fluorescence imaging, we have performed efficient laser cooling to prevent undesired dynamics of the cloud during the imaging process (thermal hopping). The cooling is implemented in a deep optical lattice. Some of the emitted photons are captured in a high-sensitivity camera during the cooling time.
In this approach, we have built an optical path to control the depth of our optical lattices in a regime where the atoms cannot escape and move while emitting photons. While this is a standard technique in current quantum gas microscopes, very little is known about implementing this technique in Cesium atoms. To fulfill this task, we have quantitatively studied the trap depth dependence vs. the molasses cooling efficiency and the imaging fidelity.
This project has pushed the limits of quantum gas microscopy in a unique setup in different ways:
From a technical point of view, we have built an experimental setup that provides large optical access and high flexibility to image and control single atoms. This is achieved by transporting the atomic cloud into a second chamber (43 cm in less than 30 ms). These features go beyond any transport scheme implemented in cold atom experiments and prevents for long cycling times that can limit the sampling of the many-body wave function.
Moreover, we have resolved with high fidelity single atoms in an optical lattice and we have explored the superfluid to Mott insulator using in situ density profiles. To date, our imaging resolution is almost two times larger compared to our short lattice spacing. Therefore, to reach single-site resolution we are currently developing machine learning architectures (convolutional neural networks) to overcome the diffraction limit. These results concludes the experimental apparatus's building stage, demonstrating the setup's robustness and its capabilities to explore more complex systems.
From a wider perspective, we are currently experiencing the second revolution in quantum mechanics. It is not a dream anymore; this revolution will surprise us with new technologies that could enormously impact our society in the following years. In order to develop these technologies for applications, we need to understand systems of many interacting quantum particles that we assemble in the lab. However, classical computers fail in many cases, therefore we need to build quantum simulators to help us advance our knowledge.
One important feature to study in quantum simulators is the interplay with topology. Understanding the role of topology in these systems could lead to the development of topological quantum computing or topological photonics. This project establishes a platform to study interacting topological insulators using ultracold quantum gases. The experimental setup and results developed during the project are the first building blocks to explore these exotic systems. While there are several challenges to overcome in the incoming years, the knowledge acquired in this project could lead to the development of new technologies in the European Union that will have a significant social-economic impact.
The worldwide competition to lead quantum technologies has begun, and the Topological Quantum gas microscope (TOQUAM) has the prospect of playing an essential role in this race.
From a technical point of view, we have built an experimental setup that provides large optical access and high flexibility to image and control single atoms. This is achieved by transporting the atomic cloud into a second chamber (43 cm in less than 30 ms). These features go beyond any transport scheme implemented in cold atom experiments and prevents for long cycling times that can limit the sampling of the many-body wave function.
Moreover, we have resolved with high fidelity single atoms in an optical lattice and we have explored the superfluid to Mott insulator using in situ density profiles. To date, our imaging resolution is almost two times larger compared to our short lattice spacing. Therefore, to reach single-site resolution we are currently developing machine learning architectures (convolutional neural networks) to overcome the diffraction limit. These results concludes the experimental apparatus's building stage, demonstrating the setup's robustness and its capabilities to explore more complex systems.
From a wider perspective, we are currently experiencing the second revolution in quantum mechanics. It is not a dream anymore; this revolution will surprise us with new technologies that could enormously impact our society in the following years. In order to develop these technologies for applications, we need to understand systems of many interacting quantum particles that we assemble in the lab. However, classical computers fail in many cases, therefore we need to build quantum simulators to help us advance our knowledge.
One important feature to study in quantum simulators is the interplay with topology. Understanding the role of topology in these systems could lead to the development of topological quantum computing or topological photonics. This project establishes a platform to study interacting topological insulators using ultracold quantum gases. The experimental setup and results developed during the project are the first building blocks to explore these exotic systems. While there are several challenges to overcome in the incoming years, the knowledge acquired in this project could lead to the development of new technologies in the European Union that will have a significant social-economic impact.
The worldwide competition to lead quantum technologies has begun, and the Topological Quantum gas microscope (TOQUAM) has the prospect of playing an essential role in this race.