Skip to main content

Programmable Atomic Large-Scale Quantum Simulation

Periodic Reporting for period 1 - PASQuanS (Programmable Atomic Large-Scale Quantum Simulation)

Reporting period: 2018-10-01 to 2020-03-31

PASQuanS aims at performing a transformative step for quantum simulation towards programmable analogue simulators, addressing questions in fundamental science, materials development, quantum chemistry and real-world problems of importance in industry. PASQuanS builds on the achievements of the most advanced quantum simulation platforms, based on neutral atoms in optical lattices or arrays of optical tweezers, interacting via collisional or Rydberg-state-mediated interactions, and trapped ions.

By scaling up these platforms towards several hundreds of atoms/ions, by improving control methods and making these simulators fully programmable, PASQuanS will push these platforms far beyond both the state-of-the-art and the reach of classical computation. Full programmability will make it possible to address quantum annealing or optimization problems much sooner than digital quantum computation. PASQuanS will demonstrate a quantum advantage for non-trivial problems, paving the way towards practical and industrial applications.

PASQuanS unites five experimental groups with complementary methods to achieve the technological goals, connected with five theoretical teams focusing on certification, control techniques and applications of the programmable platforms, and five industrial partners in charge of the key developments of enabling technologies and possible commercial spin-offs of the project. PASQuanS will result in modular building blocks for a future generation of quantum simulators. Possible end-users of these simulators, major industrial actors, are associated with the consortium. They are engaged in a dialogue on quantum simulation, and help to identify and implement key applications where quantum simulation provides a competitive advantage.
During the reporting period, the PASQuanS consortium has progressed in five interconnected directions.

We have prepared the development of the next generation of atomic-based programmable quantum simulators with enhanced capabilities (size, degree of control and programmability). In particular, we made progress in the fidelity of preparation and the coherence of manipulation of the platforms. Furthermore:
• The fermionic platform based on atoms in optical lattices and tweezers demonstrated new spin and momentum detection techniques, critical for readout of states
• The Rydberg platform demonstrated optical trapping of Rydberg atoms, at room temperature and in a cryogenic environment, laying foundations for future scaling up of these platforms.
• Tweezer arrays have been extended to 3D, nearing 100 atoms (starting from ca.50 atoms).
• Ion trap platform now manipulate chains as large as 50 ions (starting from ca. 20 ions).
• Open-loop parameter optimization methods were applied to the generation and manipulation of Schrödinger cat states with up 20 Rydberg atoms in arrays.
• Technological innovations were developed that will become commercially available products from the industry partners. They include a versatile turn-key laser system for atom trapping, high power, single mode & single frequency laser (130W) systems with well-controlled noise and coherence properties, and a versatile phase stabilized frequency offset locking system for diode lasers.

We have worked towards the validation of the platforms and developed means to certify quantum advantage on them. First, we demonstrated quantum simulation on programmable large-scale simulators in numerically accessible regimes, including: demonstration of a symmetry-protected topological phase of matter; emergence of a phase transition with up to twelve fermionic particles, reaching the limits of numerical capabilities; development of new tensor network techniques for two dimensions validate large-scale quantum simulators; and application of a tree tensor network approach to the study of lattice gauge theories in two dimensions, with assessment of their experimental implementation on Rydberg quantum simulators. Second, we developed new tools for the validation of large scale-scale quantum simulators, in particular using local randomized measurements to obtain the second-order Rényi entropy, and for a cross-platform verification of experimental quantum devices. The protocols were applied experimentally in chains of up to 20 ions.

We have also explored quantum simulation in regimes where classical simulations are not possible and investigated applications to scientific problems and quantum advantage. In particular, we have assessed the threshold for quantum advantage by pushing classical simulations of models that can be implemented on the experimental platforms. Also, in a very important work towards assessing a practical quantum advantage, we have performed classical numerical simulations to determine the hardware requirements for quantum advantage for Hubbard models, specifically taking into account experimental imperfections. We have compared the requirements at the quantum advantage point for digital and analogue simulation, understanding regimes where analogue quantum simulators will have a practical quantum advantage over both classical computing and digital quantum computing.

We have developed new control architectures and techniques for quantum simulation, in particular devising a scheme to dissipatively prepare entangled spin-symmetric states of fermions in lattices, exploiting Fermi statistics, and designing variational ansatz methods to simulate models that go beyond current experimental feasibility.

Finally, and importantly, we have engaged in a dialogue with industry to identify key applications where quantum simulation provides a competitive advantage, and to develop specific use-cases. Examples are related to the smart charging of electrical vehicle fleet that can be mapped onto a combinatorial optimization problem and that could be implemented on, e.g. Rydberg quantum simulator; the applicability of quantum inspired tensor network methods to optimise Radiotherapy Plans for Cancer Treatment; the development of quantum-inspired Machine Learning on high-energy physics data.

Besides these scientific developments, PASQuanS is committed to dissemination, communication and outreach activities in many forms: participation to scientific conferences and schools, interviews in the media, press releases about scientific achievements or events. We maintain an active website (www.pasquans.eu) and Twitter account (@pasquans).
All of the developments detailed in 1.2 be they conceptual, experimental or technological, are beyond the state-of-the-art. They indicate that the platforms are now able to provide a practical quantum advantage. The dialog initiated with end-users has already led to the exploration of real-life and industrial use-cases based on a quantum simulation approach. We expect the first proof-of-principle demonstrations in the second phase of the project. These will pave the way towards the transfer of the building blocks to industrial partners, for industry-driven production of quantum simulators expected at the end of the Flagship.

In addition, PASQuanS contributes to the strategic objectives of the Flagship and helps expanding the European leadership and excellence in quantum technologies.It is deeply involved in the training of a new generation of quantum engineers that will form the required highly skilled workforce in Europe in order to maintain and develop the leadership of Europe in the field of Quantum Technologies.