Suppression of differential light shifts in ground and metastable trapped-ion qubits

Drew Parks, Thomas Dellaert, Patrick McMillin, Conrad Roman, Andrei Derevianko, Wesley C. Campbell · 2026

Read Suppression of differential light shifts in ground and metastable trapped-ion qubits by Drew Parks, Thomas Dellaert, Patrick McMillin, Conrad Roman, Andrei Derevianko, Wesley C. Campbell (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-02543.txt.

Introduction

Parks and co-authors suppress a noise source in trapped-ion quantum bits. A magnetic field lets hyperfine clock qubits pick up a vector differential light shift. Polarization can tune that shift. The team measures a magic polarization condition in ytterbium-171 ions. It suppresses light shifts in two qubit types at once. Both the ground and metastable clock qubits benefit.

Calculations extend the result to common trapped ion species. The minimum bias fields needed are modest. They sit below field strengths already typical in labs. Methods for metastable qubit control in ytterbium-171 are also shown. State preparation and measurement infidelity reaches the ten-thousandths range. The work eases a practical burden for ion-based quantum computing.

Catalog record

CREATORDrew Parks, Thomas Dellaert, Patrick McMillin, Conrad Roman, Andrei Derevianko, Wesley C. Campbell
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE30 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.02543

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Suppression of differential light shifts in ground and metastable trapped-ion qubits Drew Parks, Thomas Dellaert, Patrick McMillin, Conrad Roman, Andrei Derevianko, Wesley C. Campbell Abstract In the presence of a magnetic field, hyperfine clock qubits can acquire a vector differential light shift that can be tuned via polarization to suppress the total differential light shift of high-power, off-resonant laser light. We experimentally measure this “magic” polarization condition, suppressing differential light shifts in both the S 1 / 2 2 {}^{2}\mathrm{S}_{1/2} ground and F o 7 / 2 2 {}^{2}\mathrm{F}_{7/2}^{o} metastable clock qubits of Yb + 171 {}^{171}\mathrm{Yb}^{+} . We present calculations of the minimum bias magnetic fields required to suppress differential light shifts in the ground state clock qubits of commonly trapped ion species, finding that they are below the strengths of fields already typically present in experiments. We further present methods for metastable clock-qubit control in Yb + 171 {}^{171}\mathrm{Yb}^{+} , demonstrating a state preparation and measurement infidelity of 2.9 − 1.5 + 3.0 × 10 − 4 2.9^{+3.0}_{-1.5}\times 10^{-4} ( − 35 ± 4 ​ dB -35\pm 4\,\mathrm{dB} ). In trapped-ion qubit systems, high-power off-resonant laser light is an often-used method for performing single- and two-qubit gates 3 ; 15 . However, this off-resonant laser light can induce undesirable differential light shifts between the qubit states. These light shifts can map gate-laser intensity noise to the qubit frequency, resulting in dephasing and reducing the effective coherence time. Further, in operational schemes that rely on the isolation of qubit subspaces such as the omg architecture 2 , these differential light shifts may cause detrimental crosstalk errors 28 , in which qubits not participating in the interaction experience phase shifts from the perturbing light. Magnetic-field-induced vector differential light shifts 17 have been shown to give rise to a “magic” polarization condition in neutral atoms that suppresses this source of decoherence 23 ; 8 . Recently, this magic condition was demonstrated in the ground state qubits of trapped ionic species {}^{133}\mathrm{Ba}^{+} 28 and {}^{171}\mathrm{Yb}^{+} 27 as well. In this paper, we explore extensions of magic polarization to ground state qubits of various commonly trapped ions, and present experimental observations of the magic polarization effect in both the ground and metastable m_{F}=0 clock qubits of {}^{171}\mathrm{Yb}^{+} in magnetic field environments typical of trapped-ion experiments. Theory We consider the clock qubit encoded in the nS_{1/2} ground state hyperfine manifolds characteristic of many commonly used trapped-ion species with non-zero nuclear spin I . This manifold consists of two hyperfine components. We focus on the two B-field-insensitive qubit clock states \ket{F}\equiv|F,m_{F}=0\rangle and \ket{F'}\equiv|F^{\prime},m^{\prime}_{F}=0\rangle , where F^{\prime} is the upper and F is the lower energy state. These are eigenstates for unperturbed Hamiltonian \hat{H}_{0} . For the positive hyperfine structure constant A_{nS_{1/2}} , F^{\prime}=I+1/2 and F=I-1/2 and the assignment is reversed otherwise. In the absence of applied fields, the qubit angular frequency \omega_{q}=E_{F^{\prime}}-E_{F}>0 ( \hbar\equiv 1 ). Below we review the derivation of the magic conditions, where both qubit levels experience identical perturbations by the laser field. This section is based on the earlier work on magic trapping of neutral atoms 19 ; 5 ; 9 ; 12 , which was recently extended to laser-noise insensitive operation of 133 Ba + qubit 28 . We also justify certain approximations made in the earlier work. Interaction of an ion with a weak laser field of frequency \omega_{L} , amplitude \mathcal{E}_{L} and polarization {\hat{\varepsilon}} can be parameterized by an optical potential operator \hat{U}=-\hat{\alpha}(\omega_{L},{\mbox{\boldmath${\hat{\varepsilon}}$}})\frac{\mathcal{E}^{2}_{L}}{4}+\mathcal{O}(\mathcal{E}^{4}_{L})\,, (1) where \hat{\alpha}(\omega_{L},{\mbox{\boldmath${\hat{\varepsilon}}$}}) is the operator of dynamic polarizability that can be expanded into irreducible tensor operators of rank K=0,1,2 , \hat{\alpha}(\omega_{L},{\mbox{\boldmath${\hat{\varepsilon}}$}})=\hat{\alpha}^{(0)}(\omega_{L})+\mathcal{A}\hat{\alpha}^{(1)}(\omega_{L})+\hat{\alpha}^{(2)}(\omega_{L})\,, (2) where \mathcal{A}\equiv-i\left({\mbox{\boldmath${\hat{\varepsilon}}$}}^{*}\times{\mbox{\boldmath${\hat{\varepsilon}}$}}\right)\cdot\mathbf{\hat{z}} is the photon’s helicity projection on the quantization axis \mathbf{\hat{z}} , and AC polarizabilities \hat{\alpha}^{(K)} are proportional to the conventional scalar, vector and tensor polarizabilities 19 . Importantly, since the laser field breaks the rotational invariance of atomic Hamiltonian \hat{H}_{0} , \hat{\alpha}(\omega_{L},{\mbox{\boldmath${\hat{\varepsilon}}$}}) can have both diagonal and off-diagonal matrix elements in the \{\ket{F},\ket{F'}\} basis. In an externally applied magnetic field, energies of qubit states can be obtained by diagonalizing the \hat{H}_{0}+\hat{U}+\hat{H}^{Z} matrix, where \hat{H}^{Z} is the Zeeman Hamiltonian. We will take the quantization axis along the laser propagation direction \mathbf{\hat{k}} . In our treatment (and in the experiment), the B-field is also directed along \mathbf{\hat{k}} . Then \hat{H}^{Z}=-\hat{\mu}_{z}B , where \hat{\mu}_{z} is the magnetic dipole operator’s z component. While the diagonal matrix elements of \hat{H}^{Z} vanish for our qubit states (there is no linear Zeeman shift for the M_{F}=0 states), it admixes the qubit basis states, leading to quadratic Zeeman shifts. At zero magnetic field, the off-diagonal matrix elements of the optical potential lead to \mathcal{O}(\mathcal{E}^{4}_{L}) corrections, which will be ignored for consistency with the omitted hyper-polarizability contributions in Eq. ( 1 ). If the laser field is off, eigen-energies repel each other exhibiting quadratic Zeeman shift. When both fields are on, the correction to the energy splitting can be estimated in the second-order perturbation theory \Delta\omega_{\mathrm{DLS}}\approx U_{FF}-U_{F^{\prime}F^{\prime}}+\frac{4}{\omega_{q}}\mathrm{Re}\!\left(H^{Z}_{FF^{\prime}}U_{F^{\prime}F}\right)\,, (3) where we omitted the quadratic Zeeman shift. This is the differential light shift (DLS) of interest. The DLS expression ( 3 ) contains B-field-free DLS and a Zeeman-optical interference term. We would like to find the conditions when the DLS vanishes, thereby eliminating the associated laser-induced decoherence. The DLS can be recast in terms of differential polarizability \Delta\alpha(\omega_{L},{\mbox{\boldmath${\hat{\varepsilon}}$}}) , so that \Delta\omega_{\mathrm{DLS}}=-\Delta\alpha(\omega_{L},{\mbox{\boldmath${\hat{\varepsilon}}$}})\,\mathcal{E}^{2}_{L}/4 . Here \Delta\alpha(\omega_{L},{\mbox{\boldmath${\hat{\varepsilon}}$}})\equiv\alpha_{F^{\prime}F^{\prime}}-\alpha_{FF}-\frac{4B\mu_{\mathrm{B}}}{\omega_{q}}\alpha_{F^{\prime}F}\,, (4) where we used \matrixelement{S_{1/2},F,M_F=0}{\hat{\mu}_z}{S_{1/2},F',M_{F'}=0}=\mu_{\mathrm{B}} , with \mu_{\mathrm{B}} being the Bohr magneton. We suppressed the \omega_{L} and {\hat{\varepsilon}} dependencies of polarizabilities on the r.h.s. of this equation for brevity. Since both qubit states are attached to the same electronic level, their polarizabilities are nearly identical, so that \alpha_{F^{\prime}F^{\prime}}\approx\alpha_{FF} . The small but important differences arise due to hyperfine interaction (HFI) admixtures to electronic wave functions and energies. The detailed formalism for computing HFI-mediated corrections \hat{\beta}^{(K)} to the AC polarizabilities can be found in Ref. 19 which generalizes the earlier static polarizability derivation 4 . Such HFI-mediated corrections are derived in third order of Floquet perturbation theory and involve matrix elements of HFI and two dipole operators along with energy denominators. Three types of diagrams appear in the calculations depending on the position of the HFI operator in the chain of three operators: top, center, bottom, see Fig. 1 of Ref. 19 . The third-order diagrams also include normalization corrections. We do not reproduce the lengthy formulae from Ref. 19 here, but these are the expressions that we use in our calculations. Note that simply including hyperfine splittings in the energy denominators while neglecting hyperfine corrections to the electronic wave functions can lead to qualitatively erroneous results 19 , because the two effects enter at the same order. In general, both the scalar ( K=0 ) and tensor ( K=2 ) HFI-mediated polarizabilites \hat{\beta}^{(K)} contribute to the \alpha_{F^{\prime}F^{\prime}}-\alpha_{FF} term in Eq. ( 4 ). For ground-state alkali-metal atoms, scalar contribution \hat{\beta}^{(0)} strongly dominates 5 . The same inequality \beta^{(0)}\gg\beta^{(2)} also holds for the alkali-like S_{1/2} ions considered here. A detailed analysis reveals that \beta^{(2)} is dominated by the center diagram which requires HFI matrix elements evaluated between the P_{J} intermediate states, while \beta^{(0)} is largely determined by the normalization (residual) diagram involving a much larger expectation value of HFI in the nS_{1/2} ground state. The normalization diagram

Read the complete text