FIG 5 - uploaded by Lorenza Viola
Content may be subject to copyright.
Spin-locking experiments. (a) Spin-locking sequence applied simultaneously on two qubits with a shared source of engineered noise. (b) Bloch sphere sketch of the sequence in (a) for qubit 1 and qubit 2. The initial π 2
Source publication
Noise that exhibits significant temporal and spatial correlations across multiple qubits can be especially harmful to both fault-tolerant quantum computation and quantum-enhanced metrology. However, a complete spectral characterization of the noise environment of even a two-qubit system has not been reported thus far. We propose and experimentally...
Contexts in source publication
Context 1
... if both dressed qubits are sensing the same frequency component of a non-vanishing spectrum (Ω 1 ≈ Ω 2 ). This insight is confirmed by the experimental data shown in Fig. 5(c), in which K zz is only significant around Ω 1 = Ω 2 = |∆ c |, in a frequency region that narrows with the duration t over which the spin-locking drive is applied. This follows from a key feature of the qubit evolution during the spinlocking drive: namely, the terms in the ME that contain the cross-spectra oscillate at ±|Ω 1 − Ω 2 | ...
Context 2
... take values of χ j and Γ 1,j = 1/T (j) 1 given in Table I. In addition, we take the bare cavity drive detuning to be ∆ c = ∆ Fig. 5(d), displays strong quantitative agreement with experimental observations. Though the phenomenological decay rates γ ↑ j and γ ↓ j are not entirely negligible, they lead to decay on a timescale 100 µs, while the dynamics due to photon shot noise occur on a shorter timescale, 50 µs. This confirms that our engineered noise source ...
Context 3
... finally discuss an additional intriguing feature in the results of the experiment. While the correlation starts out at zero [see Fig. 5(e)] when we initiate the spin-locking drive, and then quickly rises to a maximum of about 0.13, K zz ultimately decays over several tens of µs to a negative value of about −0.03. Numerical simulations predict that, despite the presence of intrinsic qubit decay sources (non-zero Γ 1,j , γ ↑ j and γ ↓ j ), the system subsequently reaches ...
Context 4
... produce the experimental data needed to reconstruct these spectra, we apply the two-qubit spin-locking sequence illustrated in Fig. 2(a) and Fig. 5(a)-(b), by letting Ω 1 ≈ Ω 2 ≈ Ω ≡ (Ω 1 + Ω 2 )/2 to have both qubits sample the spectrum at ω = Ω, thereby maximizing the sensitivity to the noise spatial correlations. We perform a total of 26 spin-locking experiments, between which the Rabi frequency Ω/2π is swept through 26 values uniformly distributed to probe the Lorentzian peak from ...
Context 5
... if both dressed qubits are sensing the same frequency component of a non-vanishing spectrum (Ω 1 ≈ Ω 2 ). This insight is confirmed by the experimental data shown in Fig. 5(c), in which K zz is only significant around Ω 1 = Ω 2 = |∆ c |, in a frequency region that narrows with the duration t over which the spin-locking drive is applied. This follows from a key feature of the qubit evolution during the spinlocking drive: namely, the terms in the ME that contain the cross-spectra oscillate at ±|Ω 1 − Ω 2 | ...
Context 6
... take values of χ j and Γ 1,j = 1/T (j) 1 given in Table I. In addition, we take the bare cavity drive detuning to be ∆ c = ∆ Fig. 5(d), displays strong quantitative agreement with experimental observations. Though the phenomenological decay rates γ ↑ j and γ ↓ j are not entirely negligible, they lead to decay on a timescale 100 µs, while the dynamics due to photon shot noise occur on a shorter timescale, 50 µs. This confirms that our engineered noise source ...
Context 7
... finally discuss an additional intriguing feature in the results of the experiment. While the correlation starts out at zero [see Fig. 5(e)] when we initiate the spin-locking drive, and then quickly rises to a maximum of about 0.13, K zz ultimately decays over several tens of µs to a negative value of about −0.03. Numerical simulations predict that, despite the presence of intrinsic qubit decay sources (non-zero Γ 1,j , γ ↑ j and γ ↓ j ), the system subsequently reaches ...
Context 8
... produce the experimental data needed to reconstruct these spectra, we apply the two-qubit spin-locking sequence illustrated in Fig. 2(a) and Fig. 5(a)-(b), by letting Ω 1 ≈ Ω 2 ≈ Ω ≡ (Ω 1 + Ω 2 )/2 to have both qubits sample the spectrum at ω = Ω, thereby maximizing the sensitivity to the noise spatial correlations. We perform a total of 26 spin-locking experiments, between which the Rabi frequency Ω/2π is swept through 26 values uniformly distributed to probe the Lorentzian peak from ...
Context 9
... if both dressed qubits are sensing the same frequency component of a non-vanishing spectrum (Ω 1 ≈ Ω 2 ). This insight is confirmed by the experimental data shown in Fig. 5(c), in which K zz is only significant around Ω 1 = Ω 2 = |∆ c |, in a frequency region that narrows with the duration t over which the spin-locking drive is applied. This follows from a key feature of the qubit evolution during the spinlocking drive: namely, the terms in the ME that contain the cross-spectra oscillate at ±|Ω 1 − Ω 2 | ...
Context 10
... as an effective decay channel for the other, with the coupling between them mediated by the noise. measurements of the Stark-shifted resonator transmission peak when both qubits are in their ground state. This yields ∆ c /2π ≈ −2.03 MHz; the remaining parameters are then obtained by fitting the solution of the ME to the experimental data shown in Fig. 5(c Fig. 5(d), displays strong quantitative agreement with experimental observations. Though the phenomenological decay rates γ ↑ j and γ ↓ j are not entirely negligible, they lead to decay on a timescale 100 µs, while the dynamics due to photon shot noise occur on a shorter timescale, 50 µs. This confirms that our engineered noise source ...
Context 11
... decay channel for the other, with the coupling between them mediated by the noise. measurements of the Stark-shifted resonator transmission peak when both qubits are in their ground state. This yields ∆ c /2π ≈ −2.03 MHz; the remaining parameters are then obtained by fitting the solution of the ME to the experimental data shown in Fig. 5(c Fig. 5(d), displays strong quantitative agreement with experimental observations. Though the phenomenological decay rates γ ↑ j and γ ↓ j are not entirely negligible, they lead to decay on a timescale 100 µs, while the dynamics due to photon shot noise occur on a shorter timescale, 50 µs. This confirms that our engineered noise source ...
Context 12
... finally discuss an additional intriguing feature in the results of the experiment. While the correlation starts out at zero [see Fig. 5(e)] when we initiate the spin-locking drive, and then quickly rises to a maximum of about 0.13, K zz ultimately decays over several tens of µs to a negative value of about −0.03. Numerical simulations predict that, despite the presence of intrinsic qubit decay sources (non-zero Γ 1,j , γ ↑ j and γ ↓ j ), the system subsequently reaches ...
Context 13
... produce the experimental data needed to reconstruct these spectra, we apply the two-qubit spin-locking sequence illustrated in Fig. 2(a) and Fig. 5(a)-(b), by letting Ω 1 ≈ Ω 2 ≈ Ω ≡ (Ω 1 + Ω 2 )/2 to have both qubits sample the spectrum at ω = Ω, thereby maximizing the sensitivity to the noise spatial correlations. We perform a total of 26 spin-locking experiments, between which the Rabi frequency Ω/2π is swept through 26 values uniformly distributed to probe the Lorentzian peak ...
