Gate-Controlled Spin Qubits in Confined Altermagnets
Hamed Vakili · 2026
Read Gate-Controlled Spin Qubits in Confined Altermagnets by Hamed Vakili (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-2606-24150.txt.
Introduction
Vakili proposes quantum bits built inside altermagnets, a recently recognised class of magnetic materials. Gate-defined spin qubits normally rely on spin-orbit coupling to steer spins electrically. Altermagnets offer another handle. Squeezing one into an elliptical quantum dot splits its low-energy states into a doublet. The two members carry opposite spin polarisation, forming a natural qubit.
Simulations over time show a phase-controlled quadrupolar gate drive performing standard single-qubit rotations. The drive wobbles the confinement shape around a fixed bias. The qubit splitting sits in the microwave range, while leakage to other states stays far away in energy, protecting operations. Vakili then pairs two dots. Their joint spectrum projects cleanly onto separate-dot states with a coupled two-qubit term. Wobbling the central barrier drives the pair close to a maximally entangled state.
Catalog record
| CREATOR | Hamed Vakili |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 31 KB |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2606.24150
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Gate-Controlled Spin Qubits in Confined Altermagnets Hamed Vakili Abstract We propose gate-defined spin qubits in electrostatically confined altermagnetic quantum dots. Elliptical confinement of the d d -wave altermagnetic structure produces a low-energy doublet with opposite spin polarization. For the range of parameters used here, the qubit states energy gap lies in the microwave range while the leakage gap remains in the meV range. Even without spin-orbit coupling, time-dependent simulations show that a phase-controlled quadrupolar gate drive about a fixed bias point implements X π / 2 X_{\pi/2} and X π X_{\pi} rotations by resonantly modulating the confinement anisotropy. We extend the study to two-qubits using a double quantum dot. We show that the double quantum dot spectrum can be cleanly projected onto isolated quantum dot product states with a nonzero nonlocal Pauli block in the effective logical two-qubit Hamiltonian. Resonant central-barrier modulation then drives the logical two-qubit component close to a maximally entangled state. These calculations show anisotropic altermagnetic quantum dots as a route to locally gate-controlled spin qubits without requiring spin-orbit coupling. Introduction —Gate-defined spin qubits are a mature solid-state platform because electrostatic gates can define the confinement potential, tune tunnel couplings, and control readout conditions in semiconductor nanostructures 41 ; 20 ; 75 ; 8 . The basic experimental toolbox was established through single-shot electrical spin readout, coherent exchange and singlet-triplet control, magnetic-resonance spin rotations, and electrically driven spin resonance in GaAs quantum dots (QD) 13 ; 53 ; 34 ; 51 . Silicon and donor devices then supplied long-coherence hosts compatible with isotopic purification, high-fidelity addressable control, and two-qubit logic 44 ; 56 ; 76 ; 81 ; 82 ; 78 ; 29 . More recently, spin qubits have advanced from few-qubit demonstrations to processor-level operation in Si and Ge platforms, including high-fidelity two-qubit gates, six-qubit control, germanium hole-spin processors, and foundry-compatible unit cells 22 ; 23 ; 25 ; 50 ; 80 ; 47 ; 54 ; 63 ; 68 ; 43 . Beyond digital quantum logic, gate-defined arrays have also been used as controllable artificial matter, for example in Fermi-Hubbard quantum simulation 26 ; 77 ; 10 ; 32 ; 12 . In most implementations the spin qubit gating is performed by a micromagnet field gradient, or spin-orbit-assisted driving. These approaches have enabled substantial progress, but they also introduce device- and material-specific constraints, such as stray fields from micromagnets and hyperfine noise in nuclear-spin-rich hosts such as GaAs. This motivates the search for materials and device designs in which a useful spin qubit can be generated without a net magnetic moment and without relying on spin-orbit coupling (SOC). Altermagnets provide such a route. They are collinear compensated magnets whose crystal symmetries permit momentum-dependent spin splitting 65 ; 66 ; 45 . Altermagnetic spin splitting has been reported experimentally in several systems 38 ; 59 , and related transport and hybrid-device consequences have been studied theoretically 52 ; 4 ; 18 ; 42 ; 3 ; 74 ; 64 ; 83 ; 11 ; 67 ; 73 . In a gate-defined quantum dot, the relevant electronic states are discrete bound states of the confinement potential rather than extended Bloch states labeled by a crystal momentum. The central question is therefore how the momentum-dependent d -wave spin splitting appears after the altermagnetic Hamiltonian is projected onto the QD bound-state spectrum, and whether the discrete bound states can be used as qubit logical states. Here we address this question with an atomistic real-space implementation of the minimal tight-binding model for a 2D d -wave altermagnet 61 . As a check on model dependence, we also repeated representative simulations using microscopic altermagnet models with explicit nonmagnetic sites 6 ; 69 and a reduced effective two-band description of the d -wave spin splitting 65 ; 61 . The same qualitative signatures appear in those calculations, so the effect is not an artifact of one tight-binding parametrization. The device concept is sketched in Fig. 1 : an elliptical gate defines a single altermagnetic dot for one-qubit operation, while a pair of such gates plus a tunable barrier defines a double dot for two-qubit operation. This elliptical confinement is an idealized representation of noncircular gate-defined dots, where electrostatic gates routinely reshape single-particle spectra, spin structure, and anisotropic response 36 ; 39 ; 2 ; 57 ; 40 ; 24 . A phase-controlled resonant quadrupolar gate drive supplies the single-qubit control: the carrier phase sets the equatorial rotation axis, so calibrated pulses give X_{\pi/2} , X_{\pi} , and phase-shifted Y -axis rotations. Combined with virtual Z rotations implemented as phase-frame updates of subsequent pulses, these operations provide arbitrary single-qubit control 46 ; 50 ; 80 ; 54 ; 68 . The two-electron configuration-interaction (CI) calculation then uses resonant barrier modulation to generate a Bell-like state in the localized two-qubit basis with low leakage. Within the scope of this model, the phase-controlled single-qubit rotations and entangling two-qubit dynamics give the ingredients required for universal gate-based quantum computation 8 ; 7 . Figure 1: Schematic device geometry. A two-dimensional d -wave altermagnetic layer is confined by electrostatic gates. For the single-qubit device, differential RF voltages modulate the quadrupolar confinement anisotropy around a fixed elliptical bias point. For the two-qubit device, two locally anisotropic gates define left and right dots, while a central barrier gate sets the static interdot coupling and supplies resonant barrier modulation. Hamiltonian and QD setup —The simulations are performed on a square atomistic lattice with spacing a and open boundaries. Each lattice point is one atomic site, and the two magnetic sublattices are separated in real space by the checkerboard label \eta_{i}=+1\penalty\ (-1) for A (B) sublattices. The operator c_{is}^{\dagger} creates an electron with spin s=\uparrow,\downarrow on atom i , and c_{i}=(c_{i\uparrow},c_{i\downarrow})^{T} . The one-electron Hamiltonian is H_{\mathrm{1e}}=H_{\mathrm{hop}}+H_{\mathrm{ex}}+H_{Z}+H_{V}. The spin-independent tight-binding part is written directly in second quantization as \displaystyle H_{\mathrm{hop}}={} \displaystyle\Delta_{ab}\sum_{\langle ij\rangle,s}\left(c_{is}^{\dagger}c_{js}+\mathrm{H.c.}\right) \displaystyle+t_{1}\sum_{\langle\langle ij\rangle\rangle,s}\left(c_{is}^{\dagger}c_{js}+\mathrm{H.c.}\right) \displaystyle+\sum_{i,s}\left[(t_{2}+\eta_{i}t_{AM})c_{is}^{\dagger}c_{i+2\hat{x},s}+\mathrm{H.c.}\right] \displaystyle+\sum_{i,s}\left[(t_{2}-\eta_{i}t_{AM})c_{is}^{\dagger}c_{i+2\hat{y},s}+\mathrm{H.c.}\right]-\mu\sum_{i,s}n_{is}. (1) Here \langle ij\rangle denotes cardinal nearest-neighbor bonds between opposite sublattices, \langle\langle ij\rangle\rangle denotes diagonal next-nearest-neighbor bonds within the same sublattice, and the t_{2},\penalty\ t_{AM} hoppings i\rightarrow i+2\hat{x} and i\rightarrow i+2\hat{y} also connect sites on the same sublattice. The \eta_{i}t_{AM} term gives opposite anisotropic same-sublattice hopping on the two sublattices and is the atomistic origin of the d -wave altermagnetic spin-splitting. The staggered exchange field is local on the atomistic lattice, H_{\mathrm{ex}}=\frac{J}{2}\sum_{i}\eta_{i}\,c_{i}^{\dagger}(\mathbf{L}\cdot\bm{\sigma})c_{i}, where \mathbf{L} is the altermagnetic spin axis, taken along \hat{z} unless stated otherwise. The in-plane magnetic field is included through H_{Z}=\frac{g\mu_{B}B_{x}}{2}\sum_{i}c_{i}^{\dagger}\sigma_{x}c_{i}. No spin-orbit coupling is included in the main-text Hamiltonian. This SOC-free model isolates the confinement-projected altermagnetic mechanism and leaves B_{x} as the only explicit spin-mixing perturbation. Single-dot confinement. —For a single QD we use \displaystyle H_{V}={} \displaystyle\sum_{i,s}V(\mathbf{r}_{i})c_{is}^{\dagger}c_{is},\penalty\ V(\mathbf{r})=\frac{1}{2}m^{\ast}\omega_{x}^{2}x^{2}+\frac{1}{2}m^{\ast}\omega_{y}^{2}y^{2}, (2) where the confinement is set by the energy scales \hbar\omega_{x} and \hbar\omega_{y} . The ellipticity parameter is \delta\equiv\frac{(\hbar\omega_{y})^{2}-(\hbar\omega_{x})^{2}}{(\hbar\omega_{y})^{2}+(\hbar\omega_{x})^{2}}. Thus \delta controls how strongly the dot distinguishes the x and y directions. In the \delta scans below, the root-mean-square confinement energy \{[(\hbar\omega_{x})^{2}+(\hbar\omega_{y})^{2}]/2\}^{1/2} is held fixed unless stated otherwise. The role of the elliptical confinement can be estimated by projecting the d -wave operator onto continuum anisotropic harmonic-oscillator envelope states. For an envelope state |n_{x},n_{y}\rangle , \left\langle k_{y}^{2}-k_{x}^{2}\right\rangle_{n_{x}n_{y}}\propto\left(n_{y}+\frac{1}{2}\right)\omega_{y}-\left(n_{x}+\frac{1}{2}\right)\omega_{x}. (3) Equation ( 3 ) is used only as an analytical guide; all spectra shown below are