🔬 mesoscale physics

Weight-four parity checks with silicon spin qubits

This paper demonstrates a silicon spin-qubit device utilizing coherent shuttling to achieve universal control and generate a five-qubit GHZ state, thereby enabling weight-four parity checks essential for advancing quantum error correction in sparse semiconductor arrays.

Original authors: Brennan Undseth, Nicola Meggiato, Yi-Hsien Wu, Sam R. Katiraee-Far, Larysa Tryputen, Sander L. de Snoo, Davide Degli Esposti, Giordano Scappucci, Eliška Greplová, Lieven M. K. Vandersypen

Published 2026-02-02

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Brennan Undseth, Nicola Meggiato, Yi-Hsien Wu, Sam R. Katiraee-Far, Larysa Tryputen, Sander L. de Snoo, Davide Degli Esposti, Giordano Scappucci, Eliška Greplová, Lieven M. K. Vandersypen

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to build a super-complex puzzle, but the pieces (called qubits) are tiny, fragile, and usually stuck in a crowded room where they accidentally bump into each other and ruin the puzzle. This is the problem with many current quantum computers: they are too crowded, and the "wires" needed to connect them cause too much interference.

This paper presents a clever new way to solve that problem using silicon spin qubits. Here is the simple breakdown of what the researchers did, using everyday analogies.

1. The "Bus" and the "Bus Stops"

Instead of cramming all the puzzle pieces next to each other, the researchers built a sparse array. Think of this like a quiet neighborhood with a few houses (the qubits) spaced far apart, connected by a single shuttling bus.

The Bus: A long, empty corridor where a single electron (the qubit) can travel.
The Bus Stops: Four specific spots along the bus where the electron can stop and talk to the people living in the houses (the data qubits).
The Driver: The researchers use a "mobile driver" (an ancilla qubit) that picks up a passenger, drives them to a house, lets them talk, and then drives them away.

This is a big deal because in a crowded room, you can't move without knocking things over. In this sparse neighborhood, the driver can move freely without disturbing the other houses.

2. The "Remote Control" Trick

Usually, to tune a quantum computer, you need to stick a sensor right next to every single piece to see if it's working. But in this sparse design, the houses are too far apart to have a sensor at every door.

The researchers invented a remote tuning method. Imagine you are trying to tune a radio in a house you can't enter. Instead of going inside, you send a messenger (the shuttle bus) to the house, ask it to do a little dance, and listen to the echo of the dance to figure out if the radio is tuned correctly.

They send an electron down the bus to a "bus stop" far away.
They check how the electron's "spin" (its internal compass) changed after the trip.
Based on that change, they can adjust the controls for that distant house without ever needing a sensor right next to it.

3. The "Four-Way Handshake" (Parity Checks)

To fix errors in quantum computers, you need to check if a group of qubits agrees with each other. This is called a parity check.

Think of it like a group of four friends holding hands. If one friend lets go (an error), the group knows something is wrong.
The researchers demonstrated a weight-four parity check. This means their "driver" qubit could visit four different "houses" in a row, shake hands with each one, and report back whether the group was "even" or "odd."
This is the first time this specific type of four-way check has been done with silicon spin qubits using this shuttling method.

4. The "Group Hug" (Entanglement)

The ultimate test of a quantum computer is creating entanglement, where particles become linked so that what happens to one instantly affects the others, no matter the distance.

The researchers used their bus system to link five qubits together into a single, giant "Group Hug" (called a GHZ state).
This is the largest group of linked silicon spin qubits ever created. It proves that the "bus" system works well enough to keep these fragile connections alive while moving them around.

5. Why This Matters (According to the Paper)

The paper claims this is a major step forward for two main reasons:

Scalability: Because the houses are far apart, they don't interfere with each other as much. This makes it easier to build a much larger computer later without the system getting messy.
Error Correction: They successfully demonstrated the specific type of "handshake" (parity check) needed to build a Surface Code, which is the gold standard for making quantum computers that can fix their own mistakes.

In summary: The team built a silicon quantum processor where qubits live in a sparse neighborhood connected by a bus. They proved they can drive a qubit around to check on four neighbors, fix errors, and link five qubits together in a giant entangled state—all without needing a sensor at every single door. This lays the groundwork for building larger, more reliable quantum computers in the future.

Technical Summary: Weight-four parity checks with silicon spin qubits

Problem Statement
For semiconductor spin qubits to serve as a viable platform for fault-tolerant quantum computing, their physical advantages—small size, long coherence times, and high-fidelity operations—must be compatible with quantum error correction (QEC). While previous demonstrations have utilized Toffoli-like primitives for conditional corrections, scalable QEC requires the repeated extraction of stabilizer measurements to infer error syndromes in both space and time. A standard benchmark, the surface code, relies on weight-four X- and Z-type parity checks, necessitating an ancillary qubit to interact with four data qubits and be measured without destroying the data states. Realizing this connectivity in densely populated quantum dot arrays is challenging due to crosstalk and the complexity of integrating readout. Conversely, sparse arrays utilizing spin shuttling offer a path forward by enabling long-range connectivity, eliminating residual exchange during idling, and easing control signal routing. However, integrating coherent spin shuttling into a multi-qubit setting while maintaining universal control and efficient measurement without local charge sensing across the entire array remains a significant hurdle.

Methodology
The authors present a silicon spin-qubit device comprising a sparse five-qubit array that leverages a shuttling bus to enable long-range connectivity. The device is fabricated on an isotopically purified $^{28}$ Si/SiGe heterostructure and features three distinct zones: a readout zone, a shuttling bus, and a row of four "bus stops" (quantum dots).

Device Architecture: The shuttling bus is defined by overlapping gates (C0–C14) allowing coherent transport of an ancilla qubit (A1). Four bus stops (BP1–BP4) host data qubits (D1–D4). A single-electron transistor (SET) charge sensor (S1) is located at one end of the bus but has a limited sensing range, effectively covering only the first bus stop.
Remote Tuning Protocol: To overcome the lack of local charge sensing for distant bus stops, the authors developed a remote tuning protocol. They utilize the spin physics of a shuttled ancilla to probe the electrostatic environment. By placing the ancilla in a superposition, shuttling it to a bus stop, and measuring the phase accumulation (via Ramsey interferometry), they reconstruct charge stability diagrams and infer virtual detuning and average chemical potentials. This allows for the virtualization of gate controls, compensating for electrostatic crosstalk and enabling modular calibration of the array without direct charge sensing at every node.
Control and Measurement: The system uses electric-dipole spin resonance (EDSR) driven by a microwave antenna for single-qubit gates and exchange interactions for two-qubit gates. A quantum non-demolition (QND) measurement framework is employed, utilizing parity-mode Pauli spin blockade (PSB) in the readout zone. This allows for the initialization and readout of the multi-qubit state with real-time feedback.
Parity Check Implementation: To perform weight-four parity checks, the ancilla qubit is shuttled to interact sequentially with the four data qubits. A decoupled controlled-Z (CZ) gate sequence is used to mitigate phase accumulation from the inhomogeneous magnetic field and shuttling pulses. The parity of the data qubits is extracted by measuring the ancilla state after the interaction sequence.

Key Contributions

Sparse Array with Shuttling: The demonstration of a silicon device where a mobile ancilla qubit can coherently interact with four isolated data qubits, forming a connectivity graph equivalent to a surface code stabilizer plaquette.
Remote Tuning without Local Sensing: The development of a protocol to populate, tune, and calibrate a sparse array using spin shuttling and spin physics, eliminating the need for local charge sensors at every qubit location.
Weight-Four Parity Checks: The successful implementation of X- and Z-type parity checks up to weight-four, a prerequisite for surface code error correction.
Multi-Qubit Entanglement: The generation of genuine entanglement across all qubit combinations in the array, specifically reporting the creation of a five-qubit Greenberger-Horne-Zeilinger (GHZ) state.

Results

Processor Characterization: The authors achieved universal control of the effective five-qubit processor. Single-qubit gate fidelities ( $F_{X90}$ ) were measured between 99.67% and 99.98%. Two-qubit gate fidelities ( $F_{2Q}$ ) ranged from 90.1% to 95.4%, limited primarily by incoherent noise in the exchange interaction strength.
Shuttling Fidelity: The 1.2 $\mu$ m round-trip shuttling path demonstrated a single-qubit fidelity of 97.7(1)% relative to an identity operation. The dominant error source was identified as the charge transition from the initialization zone to the bus.
Parity Check Accuracy: The weight-four Z-type and X-type parity checks achieved accuracies of 72.2(6)% and 67.0(7)%, respectively. These figures represent the probability that the ancilla measurement is correct and the data qubits remain in their intended prepared states.
Entanglement Verification: The fidelity of the five-qubit GHZ state was measured to be greater than 50%, constituting the first observation of genuine five-qubit entanglement in a gate-defined semiconductor spin platform. The fidelity for four-qubit GHZ states (including the ancilla) reached up to 76.6(1.6)%.
Error Budget: Analysis indicates that the majority of errors originate from infidelity in two-qubit exchange interactions. Dephasing during shuttling and idling accounts for approximately one-quarter of the total error budget.

Significance
The paper claims that this work opens immediate opportunities to pursue QEC experiments with spin qubits. By demonstrating that spin shuttling can enable long-range connectivity and modular calibration in a sparse architecture, the authors establish a viable alternative to the challenges of realizing dense qubit arrays. The protocols developed for remote tuning and QND measurement lay the groundwork for the scalable operation of sparse spin qubit arrays. Furthermore, the successful generation of a five-qubit GHZ state and the implementation of weight-four parity checks demonstrate that the necessary operations for a minimal [[4, 1, 2]] surface code are feasible in this platform. The authors note that while current two-qubit gate fidelities limit performance, the methods used are compatible with achieving error rates below 1%, suggesting a clear path toward fault-tolerant quantum memories using gate-defined semiconductor spins.