High-fidelity collisional quantum gates with fermionic atoms
Abstract
Quantum simulations of electronic structure and strongly correlated quantum phases are widely regarded as among the most promising applications of quantum computing. These computations naturally benefit from native fermionic encodings, which intrinsically restrict the Hilbert space to physical states consistent with fermionic statistics and conservation laws like particle number and magnetization independent of gate errors. While ultracold atoms in optical lattices are established as powerful analog simulators of strongly correlated fermionic matter, neutral-atom platforms have concurrently emerged as versatile, scalable architectures for spin-based digital quantum computation. Unifying these capabilities requires high-fidelity gates that preserve motional degrees of freedom of fermionic atoms, paving the way for a new generation of programmable fermionic quantum processors. Here we demonstrate collisional entangling gates with fidelities up to 99.75(6)% and Bell state lifetimes exceeding 10\,s, realized via controlled interactions of fermionic atoms in an optical superlattice. Using quantum gas microscopy, we microscopically characterize spin-exchange and pair-tunneling gates, and realize a robust, composite pair-exchange gate, a fundamental primitive for quantum chemistry simulations. Our results establish controlled collisions in optical lattices as a competitive and complementary approach to high entangling gate fidelities in neutral-atom quantum computers. When embedded within a fermionic architecture, this capability enables the preparation of complex quantum states and advanced readout protocols for a new class of scalable analog-digital hybrid quantum simulators. Combined with local addressing, these gates mark a crucial step towards a fully digital fermionic quantum computer based on the controlled motion and entanglement of fermionic neutral atoms.
Turn this paper into a full lesson
ArcXiv compiles a staged curriculum from this paper: 8-12 lessons across beginner → advanced, synthesised section guides, visuals, flashcards, a quiz, exercises, and on-demand deep dives per section. Grounded in the abstract, never invented.