Microwave-Controlled Generation of Shaped Single Photons in Circuit Quantum Electrodynamics

Large-scale quantum information processors or quantum communication networks will require reliable exchange of information between spatially separated nodes. The links connecting these nodes can be established using traveling photons that need to be absorbed at the receiving node with high efficiency. This is achievable by shaping the temporal profile of the photons and absorbing them at the receiver by time reversing the emission process. Here, we demonstrate a scheme for creating shaped microwave photons using a superconducting transmon-type three-level system coupled to a transmission line resonator. In a second-order process induced by a modulated microwave drive, we controllably transfer a single excitation from the third level of the transmon to the resonator and shape the emitted photon. We reconstruct the density matrices of the created single-photon states and show that the photons are antibunched. We also create multipeaked photons with a controlled amplitude and phase. In contrast to similar existing schemes, the one we present here is based solely on microwave drives, enabling operation with fixed frequency transmons.

Popular Summary

Today’s communication networks span the entire globe and allow us to share and process information on a worldwide scale. Networks based on the principles of quantum mechanics have the potential to enable novel ways of exchanging and processing information, for example, by providing intrinsically secure communication channels or using computers harnessing the inherent parallelism of quantum information processing. Such quantum networks will likely consist of spatially separated nodes with quantum bits (qubits) exchanging information by emitting and absorbing photons. While photons are natural candidates for mobile quantum information carriers because of their weak interactions with the environment, interfacing photons with stationary qubits in an efficient way requires care. One method of engineering this interface is to change the emission of a photon to make its waveform symmetric in time. Then, this process can be reversed at the receiving qubit to absorb the photon with high efficiency.

We demonstrate the controlled emission of a microwave photon from an artificial atom realized by a superconducting circuit on a chip—an engineered quantum system known for its strong coupling between qubits and microwave photons. We shape the waveform of the photon by irradiating the atom with a tunable microwave pulse that triggers the release of the photon and regulates its amplitude and phase. We use this method to generate photon shapes symmetric in time. We are also able to create single-photon pulses with multiple peaks, where the phase of each peak can be controlled independently.

With precise control over the photon emission process, our experiment realizes an important step toward coherent quantum networks based on a scalable solid-state system. We expect that our results may be useful for quantum circuits based on three-dimensional cavities.

Figure 1

(a) Energy level diagram of the qubit-resonator system. The matrix elements of the drive $\mathrm{\Omega }$ and the Jaynes-Cummings coupling $g$ are indicated by solid lines and dotted lines, respectively, connecting the coupled bare states. The two second-order paths connecting the states $|f0⟩$ and $|g1⟩$ are indicated by arrows. (b) In the rotating frame of the drive, the states $|f0⟩$ and $|g1⟩$ are nearly resonant. The second-order effective coupling $\tilde{g}$ between them after adiabatic elimination of the intermediate states $|e0⟩$ and $|e1⟩$ is indicated by the solid blue line. The decay of the state $|g1⟩$ into $|g0⟩$ by photon emission is shown by the yellow arrow.

Figure 2

(a) Amplitude of the drive signal used to initialize the transmon and (b) transfer the excitation into the resonator to generate a symmetric photon shape. (c) The normalized voltage amplitude $|I+iQ|$ of shaped photons obtained for drive pulses with $T=20\mathrm{ns}$, ${\mathrm{\Omega }}_0/2\pi =680\mathrm{MHz}$ (blue triangles), $T=200\mathrm{ns}$, ${\mathrm{\Omega }}_0/2\pi =700\mathrm{MHz}$ (orange squares), and $T=500\mathrm{ns}$, ${\mathrm{\Omega }}_0/2\pi =600\mathrm{MHz}$ (green circles). The dashed lines show the simulated photon shapes scaled to normalize their peak values to unity. The measured traces are fitted to the simulation with only the time shifts and scaling factors as fit parameters. The inset shows both voltage quadratures $I$ and $Q$ of the photon pulses.

Figure 3

The real and imaginary parts of the measured density matrices $\rho$ of the symmetric temporal photon mode with fidelities $F=76\%$ and $F=86\%$ to the respective ideal states $|1⟩$ and $(|0⟩+|1⟩)/\sqrt{2}$. The wire frames show the single-photon Fock state $|1⟩$ (all density matrix elements equal to 1 or 0) and the ideal superposition state $(|0⟩+|1⟩)/\sqrt{2}$ (all density matrix elements equal to $1/2$ or 0). The ticks in the bars represent increments of 0.1.

Figure 4

(a) Amplitude of the drive signal used to initialize the transmon and (b) transfer the excitation into the resonator to generate a multipeaked photon shape. (c) The real part of the measured output field $a_{\text{out}}$ as a function of time for six-peak photons with phases of each of the peaks flipped by $\pi$ one by one. The bottom-most trace is a reference with all peaks having the same phase. (d) Averaged emitted power as a function of time for the same photon pulses as in (c).

Figure 5

(a) A false color microscope image of the chip showing the transmission line resonator coupled to the transmon circuit (rectangular structure in the middle) and the microwave control lines: the resonator input/output port, the transmon charge line (blue), and the unused flux line (gray). (b) Schematic of the measurements setup (see Appendix pp1). The control electronics is shown in blue, the measurement chain in yellow, and the transmon circuit coupled to the transmission line resonator (TLR) in orange.

Figure 6

(a) The measured voltage waveform of the photon superposition state $(|0⟩+|1⟩)/\sqrt{2}$ for the indicated length of the drive pulse $T$ at a fixed pulse amplitude ${\mathrm{\Omega }}_0/2\pi =420\mathrm{MHz}$ and (b) for the indicated amplitude of the pulse ${\mathrm{\Omega }}_0$ at a fixed pulse length $T=300\mathrm{ns}$. (c) Fourier transform of the measured photon pulse waveform obtained for ${\mathrm{\Omega }}_0/2\pi =600\mathrm{MHz}$ and $T=500\mathrm{ns}$ in a reference frame rotating at the resonator frequency ${\omega }_r$. The different curves correspond to the indicated values of the detuning between the drive frequency ${\omega }_d$ and the expected zero-amplitude limit of the $|f0⟩\to |g1⟩$ transition frequency ${\omega }_0=2{\omega }_q+\alpha -{\omega }_r$. The triangle marks the center frequency of the peak. (d) Population of the $|f⟩$ state after a 100 ns square pulse driving the $|f0⟩\to |g1⟩$ transition plotted versus detuning of the pulse from ${\omega }_0$ for the indicated pulse amplitudes (shown here in units of the full scale output amplitude of the AWG). The triangles mark the minima of the curves.