Measurement-Device-Independent Verification of a Quantum Memory

In this Letter we report an experiment that verifies an atomic-ensemble quantum memory via a measurement-device-independent scheme. A single photon generated via Rydberg blockade in one atomic ensemble is stored in another atomic ensemble via electromagnetically induced transparency. After storage for a long duration, this photon is retrieved and interfered with a second photon to perform a joint Bell-state measurement (BSM). The quantum state for each photon is chosen based on a quantum random number generator, respectively, in each run. By evaluating correlations between the random states and BSM results, we certify that our memory is genuinely entanglement preserving.

Figure 1

Experimental setup and energy level scheme (gray shaded area). MOT-A and B are two laser-cooled ${}^{87}\mathrm{Rb}$ atomic ensembles, serving as the single-photon source and the quantum memory, respectively. MOT-A is further loaded into a small dipole trap to confine its dimension. For both MOT-A and B, a slight bias magnetic field is applied along the $y$ direction to remove the degeneracy of Zeeman sublevels, and all atoms are prepared in the ground state $|g⟩$ initially. In MOT-A, only one atom is excited to Rydberg state $|r⟩$ by beam pump I and II, forming a Rydberg superatom. It is soon retrieved into a signal photon by the read beam. The photon is dynamically switched to two paths with or without the quantum memory via a Pockels cell. In each path, an encoder consisted of two Pockels cells controlled by a QRNG encodes a polarization qubit onto this photon. The first photon is sent into MOT-B to store. This is achieved by mapping it onto a spin wave on $|s⟩$ with the help of the control beam. After a duration of storage, it is retrieved and interferes with the second photon in the BSM device. In MOT-B, two polarization modes of a photon are mapped onto two space modes before storing and recombined after retrieving. Two space modes are actively phase locked by adjusting the PZT. Single-photon counting module (SPCM).

Figure 2

Characterization of photon distortion during storage. (a) The temporal waveform of the photon before (blue) and after (orange) storage. The counts of poststorage are amplified by a factor of 4.5. (b) A Mach-Zehnder interferometer to detect homogeneity between stored and unstored photons. One arm is the EIT quantum memory, and the other arm is a 36 m fiber delay line. In the delay line arm, a half-wave plate (HWP) surrounded by two quarter-wave plates (QWPs) is inserted. The relative phase between two arms will vary along with the angle $\delta \phi$ of the HWP. (c) The oscillation of counts in port $L$ along with the angle $\delta \phi$ of the HWP.

Figure 3

MDI measurement and simulation results. (a) MDI verification results of the quantum memory. The blue circles are the tested results. The error bars represent 1 standard deviation. The black dashed line is the depolarizing simulation. The green shadow is the 1 standard deviation region of the tomography simulation. Pink shading indicates the entanglement-breaking regime. (b) Noise strength as a function of $t$. The black dashed line is $p$ in the depolarizing model. The green points indicate the overall noise strength $1-{\chi }_{00}$ in the tomography model.