Modeling quantum volume using randomized benchmarking of Room-Temperature NV center quantum registers
At a Glance
Section titled “At a Glance”| Metadata | Details |
|---|---|
| Publication Date | 2025-12-29 |
| Journal | npj Quantum Information |
| Authors | Tom Philipp Jäger, MinSik Kwon, M. Keller, Rouven Maier, Nicholas Bronn |
| Institutions | Max Planck Institute for Solid State Research, IBM Research - Thomas J. Watson Research Center |
| Analysis | Full AI Review Included |
Document Analysis: Room-Temperature NV Center Quantum Register Benchmarking
Section titled “Document Analysis: Room-Temperature NV Center Quantum Register Benchmarking”Executive Summary
Section titled “Executive Summary”- Platform Validated: A four-qubit quantum register based on a single Nitrogen-Vacancy (NV) center in diamond, operating at room temperature, achieves a Quantum Volume (VQ) of 8, the first such benchmark for ambient-condition NV systems.
- Architecture: One electron spin (ancilla) + three strongly coupled nuclear spins (14N, 13C1, 13C2) enabling all-to-all connectivity through mediated gates.
- Gate Performance: Single-qubit error per gate (EPG) as low as 1.0 × 10-3 (13C2), with two-qubit EPG of 2.3 × 10-2 to 4.7 × 10-2 depending on qubit pair.
- Benchmarking Method: Randomized benchmarking (RB) used to extract per-gate errors, initialization fidelity (~98%), and readout fidelity; results fed into a noise model for Quantum Volume simulation.
- Limiting Factors: Spin-lattice relaxation (T1 = 2.22 ms) limits single-qubit operations; electron spin T2 = 429 µs limits two-qubit entangling gates.
- Strategic Insight: The paper explicitly identifies that cryogenic operation could unlock significantly higher VQ, a route requiring isotopically engineered, low-strain diamond substrates.
- Core Goal: Provide a unifying, architecture-agnostic benchmarking protocol for NV centers to compare against superconducting, trapped-ion, and cold-atom quantum processors.
Technical Specifications
Section titled “Technical Specifications”| Parameter | Value | Unit | Context |
|---|---|---|---|
| Quantum Volume (VQ) | 8 | — | 2m with m = 3 (width = depth) |
| Quantum Volume Heavy-Output Threshold | > 0.66 (2σ) | — | Minimum 100 trials; 10,000 simulated instances |
| Electron Spin T1 | 2.22 | ms | Room-temperature relaxation limit |
| Electron Spin T2 (Hahn Echo) | 429 | µs | Decoherence of central spin |
| Electron Spin T2* | 12.85 | µs | Free-induction decay |
| Electron Spin Rabi Period | 1.10 | µs | Microwave π-pulse |
| 14N Hyperfine Coupling (Azz) | 2.19 | MHz | Splits ODMR into three resonances |
| 13C1 Hyperfine Coupling (Azz) | 414 | kHz | First strongly coupled carbon |
| 13C2 Hyperfine Coupling (Azz) | 90 | kHz | Second strongly coupled carbon |
| 14N Rabi Period | 55.13 | µs | Slowest nuclear drive |
| 13C1 Rabi Period | 29.69 | µs | RF-driven oscillation |
| 13C2 Rabi Period | 30.28 | µs | RF-driven oscillation |
| 14N Rabi Contrast | 0.89 | — | Qubit control fidelity metric |
| 13C1 Rabi Contrast | 0.91 | — | Qubit control fidelity metric |
| 13C2 Rabi Contrast | 0.96 | — | Highest control fidelity |
| Nuclear Spin T1 | > 100 | s | From prior reference [30] |
| Nuclear Spin T2 | ≈ 1.5 × T1,e | — | Limited by electron flip events |
| 1-Qubit EPG (14N) | 4.4 ± 0.2 × 10-3 | — | Single-qubit gate error |
| 1-Qubit EPG (13C1) | 1.6 ± 0.3 × 10-3 | — | Single-qubit gate error |
| 1-Qubit EPG (13C2) | 1.0 ± 0.5 × 10-3 | — | Best single-qubit performance |
| 2-Qubit EPG (N-C1) | 23 ± 1 × 10-3 | — | Mediated CNOT |
| 2-Qubit EPG (N-C2) | 47 ± 4 × 10-3 | — | Worst pair (slow N + weak C2) |
| 2-Qubit EPG (C1-C2) | 24 ± 4 × 10-3 | — | Mediated CNOT |
| Initialization Fidelity | ~98 | % | Via post-selection |
| Fluorescence Count Rate | 800 | kcounts/s | With solid immersion lens (SIL) |
| Static Magnetic Field | ~0.6 | T | Permanent magnet Zeeman splitting |
| Green Laser (Repump) | 532 | nm | Off-resonant NV- initialization |
| Red Laser (Resonant) | 637 | nm | Zero-phonon line enhancement |
| Yellow Laser (Charge State) | 594 | nm | NV0 detection / post-selection |
| MW Pulse Generator | Anritsu MG3702xA | — | Combined with IQ-Mixer + AWG |
| AWG (RF up to 100 MHz) | Keysight Agilent M8190A | — | Arbitrary waveform generation |
| ODMR Frequency Discard Threshold | 15 | kHz | Heating-induced drift rejection |
| RF Frequency Excitation Limit | 100 | MHz | AWG bandwidth ceiling |
Key Methodologies
Section titled “Key Methodologies”-
Experimental Setup
- Single NV center addressed via oil-immersion objective and solid immersion lens (SIL).
- Drive / readout delivered via coplanar waveguide (CPW) gold strip on diamond surface; 50 Ω terminated off-chip to reduce heating.
- External magnetic field of ~0.6 T applied via permanent magnet on positioner.
-
Charge & Spin State Initialization
- 532 nm laser resets to NV-.
- 637 nm laser enhances electron spin initialization fidelity (near zero-phonon line).
- 594 nm laser post-selects against NV0.
- SWAP gate transfers initialized electron state to nuclear spin; post-selection discards mis-initialized runs for ~98% fidelity.
-
QND Single-Shot Readout (SSR)
- Nuclear spin projected onto electron spin via selective MW pulse.
- Repeated N = 1000 electron readouts per experimental run; maximum-likelihood histogram classification with adjustable photon-count threshold.
-
Randomized Benchmarking (RB)
- Random Clifford sequences (N gates) followed by inverse unitary (U-1) to recover initial state.
- Exponential decay of return probability yields per-gate error (EPG), independent of SPAM errors.
- Native gate set: single-qubit X/Y rotations; controlled-NOT via electron-mediated three-pulse sequence (CnNOTe / CeNOTn).
-
Quantum Volume Protocol
- Square circuits of width = depth = m constructed from SU(4) unitaries on random qubit pairs.
- Heavy-output probability threshold > 2/3 (with 2σ confidence) determines success.
- Full noise model derived from RB data (Qiskit-based simulator, 10,000 instances, seed 42).
-
Coherence Characterization
- T1, T2 (Hahn echo), and T2* measured via standard pulse sequences on the electron spin.
- T2* oscillation frequencies used to extract 13C1 and 13C2 Azz couplings.
6CCVD Solutions & Capabilities
Section titled “6CCVD Solutions & Capabilities”Applicable Materials
Section titled “Applicable Materials”| Diamond Component | Recommended 6CCVD Material | Rationale |
|---|---|---|
| NV center host | Optical Grade Single Crystal Diamond (SCD), low N concentration | Reduces parasitic spin bath; maximizes T2. |
| Isotopic engineering | Isotopically Purified 12C SCD (< 10 ppm 13C) | Eliminates weakly coupled nuclear spin bath decoherence; extends coherence times. |
| Cryogenic-grade substrates | SCD substrates up to 10 mm thick | Required for future low-temperature NV experiments targeting VQ > 8. |
| Coplanar waveguide (CPW) | MPCVD-grown SCD with sub-nm Ra polish on demand | Enables uniform metallization and low-loss microwave transmission. |
Customization Potential
Section titled “Customization Potential”- Metalization: The study uses a gold strip forming a CPW antenna. 6CCVD offers in-house Au, Pt, Pd, Ti, W, and Cu metalization, enabling direct fabrication of custom CPW, microwave stripline, or bond-pad geometries on your diamond.
- Laser Cutting & Dicing: Need a specific sample geometry for the SIL or cavity coupling? 6CCVD provides laser cutting for SCD/PCD down to sub-100 µm features without chipping.
- Thickness Control: NV depth and fluorescence yield depend on precise layer thickness. 6CCVD supplies SCD from 0.1 µm to 500 µm, allowing shallow NV implantation or bulk crystal selection.
- Surface Roughness: Spin coherence and optical collection efficiency both benefit from ultra-low surface roughness. 6CCVD delivers Ra < 1 nm on SCD and Ra < 5 nm on inch-size PCD.
Engineering Support
Section titled “Engineering Support”- PhD-Level Consultation: Our in-house PhD engineering team can advise on defect engineering (NV, SiV, GeV centers), isotopic enrichment strategy, and sample geometry optimization for quantum networking and quantum computing platforms.
- Cryogenic Roadmap: As the paper explicitly notes, VQ is limited at room temperature by T1 and T2. 6CCVD can supply low-strain, high-purity SCD substrates qualified for cryogenic (4 K - 300 mK) experimentation.
- Noise-Model Integration: For groups building local noise models such as the one in this work, we can provide material-level T1/T2 characterization data and bulk defect metrics to seed your simulations.
Call to Action
Section titled “Call to Action”For custom specifications or material consultation, visit 6ccvd.com or contact our engineering team directly.
View Original Abstract
Abstract Accurately estimating the performance of quantum hardware is crucial for comparing different platforms and predicting the performance and feasibility of quantum algorithms and applications. In this paper, we tackle the problem of benchmarking a quantum register based on the NV center in diamond operating at room temperature. We define the connectivity map as well as single-qubit performance. Thanks to an all-to-all connectivity, the 2 and 3-qubit gates performance is promising and competitive among other platforms. We experimentally calibrate the error model for the register and use it to estimate the quantum volume, a metric used for quantifying the quantum computational capabilities of the register, of 8. Our results pave the way towards the unification of different architectures of quantum hardware and the evaluation of the joint metrics.