Scaling the neutral-atom Rydberg gate quantum computer by collective encoding in holmium atoms

Abstract

We discuss a method for scaling a neutral atom Rydberg gate quantum processor to a large number of qubits. Limits are derived showing that the number of qubits that can be directly connected by entangling gates with errors at the 10 −3 level using long range Rydberg interactions between sites in an optical lattice, without mechanical motion or swap chains, is about 500 in two dimensions and 7500 in three dimensions. A scaling factor of 60 at a smaller number of sites can be obtained using collective register encoding in the hyperfine ground states of the rare earth atom Holmium. We present a detailed analysis of operation of the 60 qubit register in Holmium. Combining a lattice of multi-qubit ensembles with collective encoding results in a feasible design for a 1000 qubit fully connected quantum processor.

Efficient Grover search with Rydberg blockade

“…ensembles of multi-level atoms [10,11]. In this paper, we show that also in the more conventional encoding with one qubit per atom, the simultaneous blockade among many atoms offers interesting prospects for unique multi-bit quantum gates.…”

Fidelity of a Rydberg-blockade quantum gate from simulated quantum process tomography

“…III show that E ∼ (Bτ ) −2/3 . At room temperature the Rydberg lifetime scales as τ ∼ n 2 with n the principal quantum number and in the heavy alkali atoms Rb and Cs the van der Waals Blockade interaction scales as [34] B ∼ n 12 . Thus we expect the gate error to scale as E ∼ n −28/3 so that choosing large n should give arbitrarily small errors.…”

Precision Measurement of Transition Matrix Elements via Light Shift Cancellation

“…Accurate AMO theory can also be indispensable to the design and interpretation of experiments, where direct measurement of all relevant parameters is infeasible. More complicated atoms, such as Er [14], Dy [15], and Ho [16] have recently become of interest, and development of new theoretical methods must be supported by the existence of high-precision experimental benchmarks.…”