Fast Nondestructive Parallel Readout of Neutral Atom Registers in Optical Potentials

M. Martinez-Dorantes, W. Alt, J. Gallego, S. Ghosh, L. Ratschbacher, Y. Völzke, and D. Meschede

Phys. Rev. Lett. 119, 180503 – Published 30 October 2017

Abstract

We demonstrate the parallel and nondestructive readout of the hyperfine state for optically trapped ${}^{87}{Rb}$ atoms. The scheme is based on state-selective fluorescence imaging and achieves detection fidelities $> 98 %$ within 10 ms, while keeping 99% of the atoms trapped. For the readout of dense arrays of neutral atoms in optical lattices, where the fluorescence images of neighboring atoms overlap, we apply a novel image analysis technique using Bayesian inference to determine the internal state of multiple atoms. Our method is scalable to large neutral atom registers relevant for future quantum information processing tasks requiring fast and nondestructive readout and can also be used for the simultaneous readout of quantum information stored in internal qubit states and in the atoms’ positions.

Received 1 June 2017

DOI:https://doi.org/10.1103/PhysRevLett.119.180503

Physics Subject Headings (PhySH)

Atom optics Atomic & molecular processes in external fields Cooling & trapping Fine & hyperfine structure Light-matter interaction Optical pumping Optics & lasers Quantum information architectures & platforms Quantum memories Spontaneous emission Stark effect Strong electromagnetic field effects

Quantum Information, Science & TechnologyAtomic, Molecular & Optical

Authors & Affiliations

M. Martinez-Dorantes^*, W. Alt, J. Gallego, S. Ghosh, L. Ratschbacher, Y. Völzke, and D. Meschede

Institut für Angewandte Physik der Universität Bonn, Wegelerstrasse 8, 53115 Bonn, Germany

^*Corresponding author. mmdorantes@iap.uni-bonn.de

Parallel Low-Loss Measurement of Multiple Atomic Qubits

Minho Kwon, Matthew F. Ebert, Thad G. Walker, and M. Saffman

Phys. Rev. Lett. 119, 180504 (2017)

Article Text (Subscription Required)

Click to Expand

Supplemental Material (Subscription Required)

Click to Expand

References (Subscription Required)

Click to Expand

Issue

Vol. 119, Iss. 18 — 3 November 2017

Reuse & Permissions

Access Options

Author publication services for translation and copyediting assistance advertisement

Images

Figure 1
(a) Simplified level scheme of the ${}^{87}{Rb}$ $D 2$ transition. Only relevant levels are shown. The $σ^{-}$ -polarized state detection beam (SDB) drives the cycling transition $| B ⟩ \to | e ⟩$ , but polarization impurity can lead to off-resonant scattering via the state $F^{'} = 2$ into the dark states $| D ⟩$ (wavy arrow). (b) Simplified experimental setup. Neutral atoms are loaded into a 1D standing-wave optical trap. A six-beam optical molasses (not shown) is used for position detection imaging (PDI). A single SDB along the $z$ direction is used for state detection imaging. The retroreflector is used to increase the fraction of fluorescence photons imaged onto the electron-multiplying CCD (EMCCD) camera.
Reuse & Permissions
Figure 2
(a) Measured survival probability of atoms after scattering on average 1100 photons in a deep dipole trap (trap depth $U_{0} = 3.5 mK k_{B}$ ) as a function of the detuning of the state detection beam (SDB). The bottom axis shows the detuning $Δ_{SDB}$ with respect to the bare atomic transition $| B ⟩ \to | e ⟩$ ; the top axis gives the detuning $Δ_{1}$ with respect to the same transition ac-Stark shifted by the optical dipole trap. The red solid line is a guide to the eye. (b) Histogram of CCD camera counts for single atoms in the dark states $| D ⟩$ (black) and in the bright state $| B ⟩$ (red) using SDB illumination for 10 ms with a detuning of $Δ_{SDB} / 2 π = + 123 MHz$ , and an intensity $I = 1.9 I_{sat}$ . The solid line is a fit to the camera model (see text). The vertical blue dashed line represents the threshold used to distinguish between the two states for a simple threshold analysis. The insets show images obtained using position detection imaging (PDI) and state detection imaging of atoms in the $| B ⟩$ and $| D ⟩$ states.
Reuse & Permissions
Figure 3
Bayes’ method applied to state detection of a well-resolved atom. (a) Position detection imaging of a single atom in the dipole trap (left), state detection imaging of a bright atom (middle) and dark atom (right). (b) Vertically integrated counts for the images above. The vertical dashed line indicates the position of the atom determined from the position detection image. The plots at the right side show the calculated count distributions $P_{i} (c | S)$ for different pixel columns for a bright (red shading) and dark (black curve) atom. The stronger the bright and dark histogram of a column differ, the more information about the atomic state is contained in this column. (c) State determination using the Bayesian update algorithm for a single atom prepared in the bright (left) and dark states (right). See text for details.
Reuse & Permissions
Figure 4
State reconstruction in a one-dimensional register. (a) Schematic representation the Bayesian method with a shifting patch in one dimension, assuming nearest-neighbor light contamination. The grid represents the pixels of the sensor, the dots represent empty lattice sites, and the circles represent trapped atoms, where the red (black) filling depicts the estimated probability to be bright (dark). A patch is defined around three sites and the middle set of pixels is used to update the combined-state probabilities only for the atoms inside the patch. Then the patch is shifted, the left atom is excluded from the patch by marginalizing its probability, and a new site at the right is included. The shaded regions correspond to pixels that contain either no information or have already been used. (b) State-dependent imaging (SDI) of a one-dimensional register of neutral atoms (middle box) initialized in random states. The image is integrated along the vertical direction, and the integrated CCD counts are used to calculate the probability that a lattice site contains an atom in the bright state [ $P (B)$ ] using the Bayesian update algorithm. Position detection imaging (PDI) is used before (top box) and after (bottom box) state detection to verify that the atoms remain trapped in the same lattice site.
Reuse & Permissions
Figure 5
State reconstruction for atoms in a 2D lattice using Bayes’ method. The top left picture is a simulated image. The remaining schematics represent a shifting patch algorithm for the 2D image assuming nearest-neighbor contamination. In each step, the state of the atoms inside the patch is calculated using only the central pixels. The patch is shifted until the last pixel containing information is used.
Reuse & Permissions

Physical Review Letters