Singlet-only Always-on Gapless Exchange Qubits with Baseband Control
Nathan L. Foulk, Silas Hoffman, Katharina Laubscher, Sankar Das Sarma · 2025
Read Singlet-only Always-on Gapless Exchange Qubits with Baseband Control by Nathan L. Foulk, Silas Hoffman, Katharina Laubscher, Sankar Das Sarma (2025) 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-2501-18589.txt.
Introduction
Nathan L Foulk, Silas Hoffman, Katharina Laubscher, and Sankar Das Sarma propose a new kind of spin qubit. Their paper is from two thousand twenty-five. The design, called SAGE, stores one qubit in the spins of four electrons. It stays on at all times and can be run with baseband control alone.
Older exchange-only qubits leak and pick up errors from local magnetic fields. The SAGE space is shielded from those field shifts, and leaks are held down by the always-on exchange forces. When magnetic noise rules, the team finds far longer coherence and far fewer gate errors. Two-qubit gates need just one pulse between qubits, with a much simpler pulse shape.
Catalog record
| CREATOR | Nathan L. Foulk, Silas Hoffman, Katharina Laubscher, Sankar Das Sarma |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 39 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2501-18589
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Singlet-only Always-on Gapless Exchange Qubits with Baseband Control Nathan L. Foulk, Silas Hoffman, Katharina Laubscher, Sankar Das Sarma Abstract We propose a singlet-only always-on gapless exchange (SAGE) spin qubit that encodes a single qubit in the spins of four electrons while allowing universal baseband control. While conventional exchange-only qubits suffer from magnetic-field-gradient-induced leakage and coherent errors due to local nuclear environments and variations in the g g -factor, the SAGE qubit subspace is protected from coherent errors due to local magnetic field gradients and leakage out of the computational subspace is energetically suppressed due to the exchange interactions between electrons being always-on. Consequently, we find that when magnetic gradient noise dominates over charge noise, coherence times and single-qubit gate infidelities of the SAGE qubit improve by an order of magnitude compared to conventional exchange-only qubits. Moreover, using realistic parameters, two-qubit gates can be performed with a single interqubit exchange pulse with times comparable in duration to conventional exchange-only qubits but with a significantly simplified pulse sequence. Introduction. Electrons localized in gate-defined semiconducting quantum dots are a promising candidate for building a large-scale fault-tolerant quantum computer. The most studied implementation is the so-called Loss-DiVincenzo (LD) qubit in which quantum information is encoded in the spin of a single electron [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ] . Control of LD qubits typically necessitates ac electric fields (causing heating effects) and on-chip micromagnets. These experimental requisites have inhibited scale-up of LD qubits [ 8 , 9 , 10 , 11 , 12 ] . Alternatively, exchange-only (EO) qubits use three electrons localized to three quantum dots to form a single encoded qubit. EO qubits are so-named because there exist electrically controllable exchange interactions between the spins, H_{\textrm{exch}}=\frac{1}{4}\sum_{i<j}J_{ij}\boldsymbol{s}_{i}\cdot\boldsymbol{s}_{j} , where J_{ij} is the magnitude of the exchange interaction between spin i and j and \boldsymbol{s}_{j}=(s_{j}^{x},s_{j}^{y},s_{j}^{z}) is a vector of the generators of rotation of the j th spin about x , y , and z axis, respectively. These exchange interactions can be pulsed to access any point on the encoded qubit Bloch sphere. This all-electrical dc control and the absence of a micromagnet have drawn substantial experimental interest [ 13 , 14 , 15 , 16 , 17 , 18 , 19 , 20 ] . Although EO qubits are immune to global magnetic fields, coherence times and gate fidelities are limited by differences in local magnetic fields induced by nuclear magnetic environments or differences in the g -factor between dots. Specifically, these effective magnetic field gradients drive both coherent errors within the qubit subspace and cause leakage to noncomputational states. There exists a singlet-only [ 21 , 22 ] encoding of four electrons across four quantum dots which is immune to magnetic-field-gradient-induced coherent errors within the qubit subspace [ 23 , 24 , 25 , 26 , 27 , 28 ] while retaining the control advantages of conventional EO qubits. Four electrons coupled through the exchange interaction with Hamiltonian H_{\textrm{exch}} can be used to define an encoded qubit whose states lie in the S=S^{z}=0 subspace, where S is the total spin quantum number and S^{z} is the quantum number associated with the eigenvalue of the spin operator along the z -axis. The qubit states in the singlet-only encoding are \displaystyle\left|0\right\rangle=\left|S_{12}S_{34}\right\rangle, (1) \displaystyle\left|1\right\rangle=\frac{1}{\sqrt{3}}\Bigl(\left|T^{0}_{12}T^{0}_{34}\right\rangle-\left|T^{+}_{12}T^{-}_{34}\right\rangle-\left|T^{-}_{12}T^{+}_{34}\right\rangle\Bigl), (2) where \left|S_{ij}\right\rangle refers to a singlet spin state and \left|T_{ij}^{0}\right\rangle , \left|T_{ij}^{+}\right\rangle , and \left|T_{ij}^{-}\right\rangle refer to triplet spin 0 , 1 , and -1 states, respectively, on dots i and j . The four spins are coupled into pairs, and the qubit state is determined from the spin parity of these pairs. Without loss of generality, the spins of the first (second) encoding pair are labeled as 1 and 2 (3 and 4). Unfortunately, these singlet-only qubits are unprotected from leakage into noncomputational states driven by magnetic field gradients. Previous proposals for four-electron exchange qubits either did not address this problem [ 23 , 24 , 25 , 26 , 27 , 28 , 29 ] , required all-to-all connectivity that would be difficult to manufacture [ 30 , 31 , 32 , 33 , 34 ] , or relied on ac driving of a gapped qubit [ 19 , 21 ] , which reintroduces the ac-induced heating that prompted the switch to exchange qubits. Qubit definition and coherence times. We propose a singlet-only always-on gapless exchange-only (SAGE) spin qubit that provides protection from magnetic field gradients with scalable baseband control and suppressed leakage to noncomputational states. The SAGE qubit is implemented in a T-shape geometry as shown in Fig. 1 (a). The resultant encoded qubit Hamiltonian is [see also the Supplemental Material (SM) [ 35 ] ] H_{\textrm{q}}=\frac{J_{14}}{4}(\sqrt{3}\sigma^{x}+\sigma^{z})-\frac{J_{13}}{4}(\sqrt{3}\sigma^{x}-\sigma^{z})-\frac{J_{12}}{2}\sigma^{z}. (3) Any rotation in the x - z plane can be performed by decreasing the appropriate exchange couplings [see Fig. 1 (b)]. Furthermore, when J_{12}=J_{13}=J_{14}=J , the qubit states are degenerate [see Eq. ( 3 )]. This always-on, gapless operation improves on previous four-spin exchange qubits by energetically suppressing leakage while preserving dc control in the non-rotating frame. The spectral gap between qubit states and leakage states is maximized in the T-geometry, see SM [ 35 ] . SAGE qubit measurement is straightforward using Pauli spin blockade (PSB) [ 36 , 37 , 21 , 3 ] so that electrons 1 and 2 are pulsed diabatically toward the same dot. A charge measurement then determines the qubit state [ 37 ] since the two electrons will only occupy the same dot if they form a spin singlet. This charge measurement is reliable due to the singlet-triplet relaxation time being much longer than the charge relaxation time [ 38 ] . However, similar to PSB measurements in conventional EO devices, the measurement is sufficient only when the qubit state is unleaked. SAGE qubit initialization can be performed either by spin relaxation or by tunneling cold electrons into the singlet state from a neighboring reservoir [ 39 ] . (a) (a) (a) J_{13} J_{12} J_{14} 1 3 4 2 (b) y x z \vec{J_{12}} \vec{J_{14}} \vec{J_{13}} Figure 1: Qubit geometry and operation. (a) Schematic of the SAGE qubit layout in real space. The encoding pairs of spins are \{1,2\} and \{3,4\} . (b) Bloch sphere representation of the SAGE qubit and each exchange interaction’s rotation axis of control. Figure 2: Qubit idle coherence times. Average idle coherence time for several gapless exchange qubits (see main text for the definitions of the acronyms). For the SAGE and TriAGE qubits, each exchange coupling is set to 10 MHz, whereas all of the conventional EO couplings are set to zero. These decays are averaged over 2500 realizations of magnetic and exchange disorder. (a) Coherence times for a range of magnetic disorder strengths. Each time is extracted from a Gaussian fit of the average idle decay with \delta J=5\times 10^{-3} . (b) Coherence times for a range of relative exchange disorders with \delta h=50 kHz. Both \delta J=5\times 10^{-3} and \delta h=50 kHz reflect parameters that may be experimentally attainable. H_{C}=0 for conventional EO systems, since J=0 for each coupling and the noise is multiplicative. Figure 3: Single-qubit gate fidelities. (a) Randomized benchmarking results for several gapless exchange qubits. Each point represents an average of 100 gate