A photon–photon quantum gate based on Rydberg interactions

Abstract

Different types of interactions in quantum systems are currently being investigated regarding their suitability for future quantum technologies. One promising candidate is the strong and long-range interaction between Rydberg states of neutral atoms, for which first applications in the fields of quantum computation [1] and quantum simulations [2][3][4] have been reported in recent years. Using electromagnetically induced transparency, these Rydberg interactions can be mapped to light [5][6][7][8][9][10][11][12][13][14][15], thus creating an effective interaction between photons. Here we extend the range of applicability of these interactions to the field of quantum communication and quantum networking [16] by realising a photon-photon quantum gate [17,18]. To achieve this, a photonic control qubit is stored in a Rydberg quantum memory in an ultracold atomic gas. This qubit interacts with a propagating photonic target qubit coupled to a Rydberg state to generate a conditional π phase shift. Finally, the control photon is retrieved. We measure controlled-NOT truth tables with fidelities 0.66(9) and 0.70(8) and an entangling-gate fidelity of 0.637(45). The demonstration of a photon-photon quantum gate is the hallmark for having achieved full quantum control of one photon over another. The level of control achieved here is an important step on one hand for exploring novel many-body states of photons and on the other hand for applications in quantum communication and quantum networking. A prominent example is deterministic photonic Bell-state detection which is crucial for quantum repeaters.Optical technologies serve as today's standard for distributing information in the Internet since photons offer high speed and large bandwidth. Because of these benefits, future quantum technologies will probably rely on photonic qubits to transfer quantum states between distant nodes. However, ambitions to use photons for processing rather than only transmitting qubits are hampered by the fact that photons hardly interact with each other. A solution to this problem is offered by a Rydberg polariton [6][7][8][9][10][11][12][13][14][15]. This intriguing quasiparticle -composed of a photonic component and an atomic Rydberg excitation -is obtained when a photon enters a medium in which electromagnetically induced transparency (EIT) couples the photon to a Rydberg state.The key idea is that the atomic Rydberg components create a strong long-range interaction between two Rydberg polaritons. Many schemes for implementing a photonphoton gate based on Rydberg interactions have been a (Control qubit) b (Target qubit) |gñ |gñ |g ñ |69 |67 ñ ñ |e |e R R ñ ñ |e |e L L ñ ñ Δ Δ HFS HFS |Lñ |Lñ |Rñ |Rñ Δ t Δ t ' Control Rydberg coupling Target coupling Ground state coupling FIG. 1: Atomic level schemes. a, Level scheme for EIT storage of the control qubit. The initial population (green) is prepared in state |g . |L polarisation is stored in a Rydberg state |69 , |R polarisation in a ground state |g ′ . b, Level scheme for the target qubit. |L pol...

Dark-time decay of the retrieval efficiency of light stored as a Rydberg excitation in a noninteracting ultracold gas

“…This is a big improvement over the 66(2)% which we reported in Ref. [16] for a measurement for t = 4.5 µs, ̺ 0 = 2 × 10 12 cm −3 , and storage in state 69S. The much lower atomic density in the present measurement is crucial for this improvement [30].…”

“…In addition, we overlap the retrieved light with a local oscillator and measure a visibility above 90%. This is a considerable improvement over the 66% visibility which we previously measured at a peak density of 2 × 10 12 cm −3 [16]. If the dipole trap with a wavelength of 1064 nm is left on during the experiment, the 1/e decay time is reduced to 13 µs.…”

“…This would increase the time-averaged data acquisition rate by an order of magnitude when choosing the separation between repetitions to 100 µs as in Ref. [16].…”

Stationary Light in Atomic Media

“…This corresponds to a cross phase modulation so strong as to give a π phase shift at the single photon level. This feat has recently been achieved using a single atom in an optical resonator and an ensemble of Rydberg atoms . Theoretical work has suggested a way around needing a π phase shift .…”

Two photon conditional phase gate based on Rydberg slow light polaritons

“…intermediate p-level caused by the interaction, • β sc accounts for scattering into additional polariton channels, • β wp includes the impact of the finite length of the wave-packets on (a) the losses from p-state during propagation and (b) the inhomogeneous interactioninduced phase-shift • β at describes the distortion of the phase shift due to spatially varying atomic density, • β Ry denotes losses due to the finite lifetime of the Rydberg state, whereas • β tr depicts finite beam waist effects leading to scattering of photons to other transversal photonic modes. At this point it worth mentioning the family of phase-gate schemes in which at least one photon [58,57,27,68,69,70] is stored as a collective atomic excitation. This type of schemes is superior compared to counter-propagating scheme when it comes to the β sc and β wp in equation Eq.…”