Skip to content

A universal set of geometric quantum gates through dressed-states assisted shortcuts to adiabaticity

FieldValue
JournalJournal of Physics B Atomic Molecular and Optical Physics
Published2026-02-02
AuthorsMaría E. Rus, Alejandro Ferrón, Omar Osenda, Sergio S. Gómez
DOI10.1088/1361-6455/ae407b

Numerical simulations produced single-qubit geometric-gate fidelity above 99.9% under intermediate systematic errors. The dressed-state shortcut protocol cancels its residual dynamical phase and extends to CS and CNOT gates with a nearby 13C nuclear spin.

Fast geometric control for an NV-center qubit

Section titled “Fast geometric control for an NV-center qubit”

The model uses the two lowest states of the NV-center spin-1 ground manifold as a qubit under a sufficiently strong static magnetic field (B_z). A microwave field supplies the Rabi frequency (\Omega_R(t)) and detuning (\Delta(t)=\omega_e-\omega(t)) in the rotating-wave approximation.

Control pulses move the state along an orange-slice path on the Bloch sphere. The z-rotation follows ABCDA, while the x-rotation follows BCDAB.

The x- and z-rotations form a universal single-qubit gate set. Their (\pi/2) choices produce the NOT and S gates used in the simulations. The orange-slice path uses four equal quarter intervals, giving a total evolution time (T=4\tau).

Ordinary adiabatic geometric gates require slow evolution, which increases operation time and decoherence exposure. SATD adds a corrective Hamiltonian so the initial and target adiabatic states connect at a chosen speed.

Unlike the original SATD work centered on STIRAP-like systems, this proposal applies the method to a directly driven two-level system.

SATD applies two successive unitary transformations. The first moves into the adiabatic frame, and the second selects dressed states whose Hamiltonian is diagonal in the initial-state basis.

Selecting dressed states that coincide with adiabatic states at the start and end makes the dressed-state evolution trivial. A corrective Hamiltonian then returns the system to the Schrödinger picture with modified detuning and Rabi pulses.

The corrected pulses depend on (g_x(t)), which governs state transfer, and (g_z(t)), which can be designed for phase cancellation.

The dressed-state freedom distinguishes SATD from Transitionless Quantum Driving. TQD is the special case (V(t)=1) throughout the evolution.

The unmodified shortcut separates the total phase into geometric and dynamical terms. For the orange-slice path, (\gamma_g=\pi-(\phi_2-\phi_1)) and (\gamma_d=I_{\phi_1}-I_{\phi_2}).

The authors choose a correction (g_z(t)) that symmetrizes the dressed-state energy around (t=2\tau). Equal energy integrals cancel (\gamma_d) and leave a purely geometric operation.

The correction changes the pulse profiles through (P=\sqrt{1+g_z/\Omega}). Because (g_z) is proportional to the square of (\dot{\theta}), the modification vanishes in the adiabatic limit.

The corrected pulse opens the original Bloch trajectory, yet the designed dressed-state energy symmetry removes the net dynamical contribution.

Pulse amplitude is a central design constraint because STA corrections often require larger microwave fields. The authors define (\eta=\Delta_0/\Omega_0) and (x=\tau\Omega_0), then compare original and corrected maxima with (R).

(R=1) means that a corrected maximum equals its original value. The analysis shows that (R) scales with (x), allowing shorter operation times without larger amplitudes by tuning (\Omega_0).

For (\eta=1), (x=2), and (\Omega_0/2\pi=3) MHz, the correction changes pulse shapes but leaves their maximum amplitudes unchanged. The ratio (|g_z/\Omega|) reaches its minimum at (\eta=2) and falls as (1/x^2).

The original azimuth changes abruptly at (T/2). Since the corrected pulses depend on this phase, the authors replace it with a smooth interpolation of characteristic time (\sigma).

The paper reports that (\sigma=5) to 10 ns is readily achievable for typical microwave Rabi amplitudes of 10 to 20 MHz. Phase smoothing caused no significant fidelity decay for (\eta\geq1) in the appendix calculation.

Two static control errors are modeled: (\delta) shifts the qubit frequency, and (\epsilon) changes the driving strength. Simulations use (\Omega_0/2\pi=3) MHz, (\tau=2/\Omega_0), and (\eta=0.5,1,2).

The NOT gate retains higher fidelity than the S gate as (\delta) grows. Its population inversion depends on the Rabi frequency, whereas the S gate depends on geometric-phase accumulation.

The S gate proves more robust to (\epsilon) than the NOT gate. Differences between (\eta) values remain around (10^{-3}) at the largest driving deviations.

For (\eta=2), fidelity decay with (\delta) is less pronounced than at the other tested values. Combined errors (\delta) and (\epsilon), each up to 15%, leave S and NOT fidelity above 0.96.

Decoherence uses a Lindblad model with (\kappa_1=1/T_1) for decay and (\kappa_2=1/T_\phi) for pure dephasing. Typical room-temperature NV values are (T_1\sim1) to 100 ms and (T_\phi\sim0.1) to 1 ms.

The paper notes that isotopically enriched diamond can increase both coherence times. With (\kappa_1=5\times10^{-4},\mu\mathrm{s}^{-1}), relaxation has much less effect than dephasing.

The fidelity of both gates falls linearly with (\kappa_2), but the S gate falls faster because dephasing suppresses its required relative phase. The tested fidelities remain above 0.99 under (\delta=\epsilon=0.05).

The conclusion reports fidelity above 99.4% under strong environmental decoherence. For decoherence, average fidelity integrates over 1001 initial states uniformly distributed across ([0,2\pi]).

Two-qubit control with a carbon-13 nucleus

Section titled “Two-qubit control with a carbon-13 nucleus”

The two-qubit model couples an NV electron spin to a nearby (^{13}\mathrm{C}) nucleus with (I=1/2). The four-state basis combines electron spin projections with nuclear spin down or up.

The microwave field drives only the electron spin. The Hamiltonian separates into two uncoupled (2\times2) blocks, and one block reproduces the single-qubit dynamics.

When (A_{\mathrm{hf}}\gg\Omega_R), hyperfine coupling suppresses transitions in the controlled block. The resulting unitary applies the single-qubit gate to one block and the identity to the other.

At (A_{\mathrm{hf}}/2\pi=130) MHz, a coupling reported as experimentally feasible in prior work, CS and CNOT fidelity reach around 99.8% under deviations below 5%.

CS gate infidelity decreases as (A_{\mathrm{hf}}/2\pi) increases through 25, 45, 80, and 150 MHz. The controlled block becomes less sensitive to the microwave transition when the hyperfine coupling dominates.

At zero systematic error, the fixed hyperfine coupling keeps two-qubit fidelity below the ideal single-qubit values. The paper attributes this limit to experimental constraints on the coupling.

Limits for experiment and material selection

Section titled “Limits for experiment and material selection”

Every performance result comes from numerical simulation. The paper does not report device fabrication, microwave pulse measurements, or an experimental gate demonstration.

Only the closed orange-slice trajectory is developed in detail. The authors state that the protocol could extend to other paths in principle, while open-path extensions remain future work.

Engineering design must balance faster operation, correction amplitude, and microwave phase bandwidth. Favorable (\eta) and (x) values reduce the correction, while the phase transition time is constrained by control electronics.

The host-material guidance is limited. The paper says that isotopically enriched diamond increases (T_1) and (T_\phi), but gives no defect density, isotope fraction, or sample geometry.

Longer coherence times reduce the modeled decoherence loss. Host selection therefore affects the short gate’s phase fidelity, although the simulations do not include fabrication-induced variation.

The modeled application requires a diamond host for an NV-center electron spin, with a nearby 13C nucleus for the two-qubit proposal. The diamond materials catalog lists MPCVD-grown single-crystal and polycrystalline diamond in plates and wafers up to 125 mm. The paper does not specify defect implantation or annealing requirements.

Custom fabrication support is available through the engineering support page for assessing those unresolved device requirements.

View Original Abstract

Abstract Geometric quantum computation relies on the geometric phase that arises in adiabatic cyclic evolutions of non-degenerate quantum systems, enabling the design of robust quantum gates. However, the adiabatic condition requires long evolution times, making the system vulnerable to decoherence. In this work, we propose a scheme to realize fast and high-fidelity geometric quantum gates by applying the superadiabatic transitionless driving protocol within the dressed-state framework. We analyze the implementation of single-qubit gates in a two-level system driven by a microwave field, focusing in particular on the nitrogen-vacancy center in diamond. We show how the dynamical phase can be canceled to obtain purely geometric operations. The robustness of the gates is assessed under systematic errors and environmental decoherence, demonstrating high fidelity even in regimes with strong fluctuations. Finally, we extend the protocol to construct nontrivial two-qubit gates, highlighting its feasibility for scalable quantum information processing.