PSQA Implementation Roadmap: Technical Milestones and First Steps
February 4, 2026
28 min read
Tushar Agrawal
Quantum ComputingPSQAImplementationTechnical RoadmapSilicon CarbideResearch PlanStartup GuideTechnology 2026
A detailed technical roadmap for implementing the Phonon-mediated Silicon Carbide Quantum Architecture, from first experiments to million-qubit systems.
PSQA Implementation Roadmap
From First Experiment to Million Qubits
This document provides a detailed technical roadmap for implementing PSQA (Phonon-mediated Silicon Carbide Quantum Architecture). Unlike high-level business plans, this focuses on the actual experiments, equipment, and milestones needed to build a room-temperature quantum computer.
---
Phase 1: Foundation Experiments (Months 1-18)
1.1 Initial Lab Setup
MINIMUM VIABLE QUANTUM LAB
═══════════════════════════════════════════════════════════════════════════════
Equipment List (Total: ~$500k-1M for Phase 1):
OPTICAL TABLE & ISOLATION
─────────────────────────────────────────────────────────────────────────────
│ Item │ Specification │ Cost │ Vendor │
├───────────────────────────┼─────────────────────────┼──────────┼──────────┤
│ Optical Table │ 4' × 8' × 12" granite │ $15,000 │ Newport │
│ Vibration Isolation Legs │ Pneumatic, <1 Hz │ $8,000 │ Newport │
│ Laser Enclosure │ Class 4 compliant │ $5,000 │ Thorlabs │
│ Climate Control │ ±0.5°C, <40% RH │ $10,000 │ Various │
└───────────────────────────────────────────────────────────────────────────┘
LASER SYSTEM
─────────────────────────────────────────────────────────────────────────────
│ Item │ Specification │ Cost │ Vendor │
├───────────────────────────┼─────────────────────────┼──────────┼──────────┤
│ Excitation Laser │ 730 nm, tunable ±5nm │ $40,000 │ Toptica │
│ │ CW, 100 mW, <1 MHz lw │ │ │
│ Repump Laser │ 905 nm, 50 mW │ $15,000 │ Thorlabs │
│ AOM (2×) │ 200 MHz, <20 ns switch │ $8,000 │ AA Opto │
│ Wavelength Meter │ ±2 pm accuracy │ $20,000 │ Bristol │
└───────────────────────────────────────────────────────────────────────────┘
DETECTION SYSTEM
─────────────────────────────────────────────────────────────────────────────
│ Item │ Specification │ Cost │ Vendor │
├───────────────────────────┼─────────────────────────┼──────────┼──────────┤
│ Single Photon Detector │ SNSPD or APD, >70% QE │ $50,000 │ ID Quant │
│ Time Tagger │ <50 ps resolution │ $15,000 │ Swabian │
│ Spectrometer │ 850-950 nm, 0.01 nm res │ $25,000 │ Princeton│
│ CCD Camera │ Scientific grade │ $10,000 │ Andor │
└───────────────────────────────────────────────────────────────────────────┘
MICROSCOPY
─────────────────────────────────────────────────────────────────────────────
│ Item │ Specification │ Cost │ Vendor │
├───────────────────────────┼─────────────────────────┼──────────┼──────────┤
│ Confocal Microscope Head │ Home-built or Attocube │ $50,000 │ Custom │
│ Objective (2×) │ 100×, NA 0.9, NIR │ $6,000 │ Olympus │
│ XYZ Piezo Stage │ 100 μm range, 1 nm res │ $15,000 │ PI │
│ Coarse Positioner │ 10 mm range │ $5,000 │ Newport │
└───────────────────────────────────────────────────────────────────────────┘
MICROWAVE SYSTEM
─────────────────────────────────────────────────────────────────────────────
│ Item │ Specification │ Cost │ Vendor │
├───────────────────────────┼─────────────────────────┼──────────┼──────────┤
│ Signal Generator │ DC-6 GHz, <1 Hz phase │ $30,000 │ R&S │
│ IQ Modulator │ DC-6 GHz, >30 dB │ $5,000 │ Marki │
│ RF Amplifier │ 30 dB, 1-6 GHz │ $3,000 │ Mini-Cir │
│ MW Switch │ <10 ns, 50 dB isolation │ $2,000 │ Mini-Cir │
│ AWG │ 1 GS/s, 14-bit │ $40,000 │ Keysight │
└───────────────────────────────────────────────────────────────────────────┘
CONTROL ELECTRONICS
─────────────────────────────────────────────────────────────────────────────
│ Item │ Specification │ Cost │ Vendor │
├───────────────────────────┼─────────────────────────┼──────────┼──────────┤
│ FPGA (Timing Control) │ Artix-7 or better │ $3,000 │ Xilinx │
│ DAC Card │ 16-bit, 8 channels │ $5,000 │ NI │
│ ADC Card │ 16-bit, 4 channels │ $5,000 │ NI │
│ Digital I/O │ 32 channels, 100 MHz │ $2,000 │ NI │
└───────────────────────────────────────────────────────────────────────────┘
SAMPLES & MATERIALS
─────────────────────────────────────────────────────────────────────────────
│ Item │ Specification │ Cost │ Vendor │
├───────────────────────────┼─────────────────────────┼──────────┼──────────┤
│ 4H-SiC Wafers (10×) │ n-type, 6° off-axis │ $5,000 │ Cree │
│ Isotopically Pure SiC │ >99.9% ²⁸Si, ¹²C │ $20,000 │ IsoFlex │
│ Sample Holders │ Custom machined │ $2,000 │ Local │
│ Permanent Magnets │ NdFeB, various sizes │ $500 │ K&J │
│ 3-axis Helmholtz Coils │ ±100 G, <0.01 G res │ $5,000 │ Custom │
└───────────────────────────────────────────────────────────────────────────┘
TOTAL PHASE 1 EQUIPMENT: ~$450,000 - $600,000
1.2 First Experiments: Single V_Si Characterization
EXPERIMENT 1: FIND AND CHARACTERIZE SINGLE V_Si DEFECTS
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 1-6
Objective: Locate individual V_Si defects and measure their basic properties
Setup Diagram:
─────────────
┌─────────────────────────────────────────────────────────────────┐
│ CONFOCAL MICROSCOPE │
│ │
│ Laser (730 nm) ─────┐ │
│ ▼ │
│ ┌───────────┐ │
│ │ AOM │ ◄── TTL from FPGA │
│ └─────┬─────┘ │
│ │ │
│ ▼ │
│ ┌───────────┐ │
│ │ Dichroic │────────────► Filter + Detector │
│ │ (800nm) │ (emission) │
│ └─────┬─────┘ │
│ │ │
│ ▼ │
│ ┌───────────┐ │
│ │ Objective │ │
│ │ 100× │ │
│ └─────┬─────┘ │
│ │ │
│ ▼ │
│ ╔═══════════╗ │
│ ║ 4H-SiC ║ on XYZ piezo stage │
│ ║ Sample ║ │
│ ╚═══════════╝ │
│ │
└─────────────────────────────────────────────────────────────────┘
Experimental Protocol:
──────────────────────
Step 1: Sample Preparation
├── Start with commercial 4H-SiC wafer (as-grown V_Si density ~10¹⁴/cm³)
├── Cleave to 5×5 mm pieces
├── Clean: Acetone → IPA → Piranha → DI water
└── Mount on sample holder with silver paint (thermal contact)
Step 2: Coarse Scan (Locating Emission)
├── Excitation: 730 nm, 100 μW (below saturation)
├── Scan area: 50 μm × 50 μm
├── Pixel dwell: 10 ms
├── Detection: >850 nm bandpass, count rate
└── Expected: Bright spots from V_Si ensembles
Step 3: Fine Scan (Single Defect Isolation)
├── Zoom to low-density region
├── Scan area: 5 μm × 5 μm
├── Look for isolated spots with count rate ~10-50 kcps
└── Verify single emitter via antibunching (g²(0) < 0.5)
Step 4: Photon Correlation (Single Emitter Proof)
├── Set laser on single spot
├── Split emission 50/50 to two detectors
├── Record coincidences vs delay time τ
├── Fit g²(τ) = 1 - a·exp(-|τ|/τ_life)
└── SUCCESS if g²(0) < 0.5 (proves single emitter!)
Expected Results:
─────────────────
┌────────────────────────────────────────────────────────────────────┐
│ CONFOCAL SCAN │ ANTIBUNCHING (Single V_Si) │
│ │ │
│ · · · · │ g²(τ) │
│ · │ ↑ │
│ · · ● │ 1 │ ──────────────────── │
│ ◄─single! │ │ ╱ ╲ │
│ · · · │ │ ╱ ╲ │
│ │ 0 │───────●─────────────►τ │
│ · · │ │ g²(0)<0.5 │
│ │ └────────────────────── │
└────────────────────────────────────────────────────────────────────┘
Milestone 1.2 Success Criteria:
───────────────────────────────
□ Locate >10 single V_Si defects
□ Measure g²(0) < 0.3 for isolated defects
□ Determine optical lifetime τ = 5-7 ns
□ Measure saturation count rate > 50 kcps
□ Identify V1 vs V2 line via spectroscopy
1.3 Spin Readout Experiments
EXPERIMENT 2: OPTICALLY DETECTED MAGNETIC RESONANCE (ODMR)
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 4-9
Objective: Detect and control V_Si electron spin via optical methods
Additional Equipment Needed:
────────────────────────────
- Microwave antenna (PCB loop, ~2mm diameter)
- Permanent magnet on 3-axis mount
- RF amplifier connected to signal generator
ODMR Principle:
───────────────
When microwave frequency matches spin transition:
OFF-RESONANCE ON-RESONANCE (MW = ω₀)
───────────── ─────────────────────
Optical excitation Optical excitation
│ │
▼ ▼
┌─────────┐ ┌─────────┐
│ Excited │ │ Excited │
└────┬────┘ └────┬────┘
│ │
│ Fluorescence │ Less fluorescence!
│ (bright) │ (dark)
▼ ▼
┌─────────┐ ┌─────────┐
│ ms = ±½ │ ─────MW───────► │ ms = ±³⁄₂│
└─────────┘ └─────────┘
The ±³⁄₂ state has lower fluorescence rate!
This contrast lets us detect spin state.
ODMR Protocol:
──────────────
Time sequence:
Laser: ████████████████████████████████████████████████████████
(Continuous illumination during scan)
Microwave: ░░░░░░░░█████░░░░░░░░█████░░░░░░░░█████░░░░░░░░█████░░░░
(Frequency stepped each cycle)
Detection: Count photons at each MW frequency
Plot: counts vs MW frequency
Expected ODMR Spectrum:
───────────────────────
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ Fluorescence │
│ (counts) │
│ ↑ │
│ │ ──────────────────────────────────────── │
│ 100% │ │
│ │ │
│ │ ╲ ╱ │
│ │ ╲ ╱ │
│ 95% │ ╲──────────╱ │
│ │ ↑ │
│ │ Zero-field │
│ │ splitting │
│ │ 2D ≈ 70 MHz │
│ │ │
│ └──────────────────────────────────────────────────────► │
│ 0 35 MHz 70 MHz 105 MHz │
│ MW Frequency (at B=0) │
│ │
└──────────────────────────────────────────────────────────────────────┘
With magnetic field (B ≠ 0), the dip splits due to Zeeman effect:
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ Fluorescence │
│ ↑ │
│ │ ────────────────────────────────────────────── │
│ 100% │ │
│ │ ╲ ╱╲ ╱ │
│ │ ╲──────────╱ ╲──────────╱ │
│ 95% │ ↑ ↑ │
│ │ ω₀ - γB ω₀ + γB │
│ │ │
│ └──────────────────────────────────────────────────────► │
│ MW Frequency │
│ │
│ Splitting = 2γB where γ = 28 MHz/mT (electron gyromagnetic ratio) │
│ │
└──────────────────────────────────────────────────────────────────────┘
Milestone 1.3 Success Criteria:
───────────────────────────────
□ Observe ODMR contrast > 3% on single V_Si
□ Measure zero-field splitting 2D = 70 ± 2 MHz
□ Demonstrate Zeeman splitting with applied B field
□ Achieve ODMR linewidth < 5 MHz (indicates good coherence)
□ Map multiple V_Si defects, verify consistent properties
1.4 Coherent Spin Control
EXPERIMENT 3: RABI OSCILLATIONS AND COHERENCE MEASUREMENT
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 7-12
Objective: Demonstrate coherent control of single V_Si spin
Pulsed ODMR Setup:
──────────────────
The key is precise timing between laser, microwave, and detection:
┌─────────────────────────────────────────────────────────────────────────┐
│ │
│ FPGA ───────┬───────────────┬───────────────┬─────────────── │
│ │ │ │ │
│ ▼ ▼ ▼ │
│ ┌────────┐ ┌────────┐ ┌────────┐ │
│ │ Laser │ │ MW │ │ Photon │ │
│ │ AOM │ │ Switch │ │Counter │ │
│ └────────┘ └────────┘ │ Gate │ │
│ └────────┘ │
│ │
│ Timing precision required: < 10 ns │
│ Typical pulse lengths: 100 ns - 10 μs │
│ │
└─────────────────────────────────────────────────────────────────────────┘
Rabi Oscillation Protocol:
──────────────────────────
Pulse sequence:
Time: |←─ Init ─→|←── MW pulse ──→|←─ Readout ─→|
Laser: ██████████████░░░░░░░░░░░░░░░██████████████
| 1 μs | (laser off) | 300 ns |
Microwave: ░░░░░░░░░░░░░███████████████░░░░░░░░░░░░░░░
| | variable τ | |
Detection: ░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█████████████
| | count here |
Repeat for different τ (MW pulse duration):
τ = 0 ───► 100% in |0⟩ (bright)
τ = τ_π/2 ───► 50% |0⟩ + 50% |1⟩ (intermediate)
τ = τ_π ───► 100% in |1⟩ (dark)
τ = τ_3π/2 ───► 50% |0⟩ + 50% |1⟩ (intermediate)
τ = τ_2π ───► 100% in |0⟩ (bright again!)
Expected Rabi Oscillations:
───────────────────────────
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ Population │
│ in |0⟩ │
│ ↑ │
│ │ ● │
│ 100% │ ╲ ╱╲ ╱╲ ╱╲ │
│ │ ╲ ╱ ╲ ╱ ╲ ╱ ╲ │
│ │ ╲ ╱ ╲ ╱ ╲ ╱ ╲ │
│ 50% │------●--●------●--●------●--●------●--- │
│ │ ╲╱ ╲╱ ╲╱ ╲╱ │
│ │ ● ● ● ● │
│ 0% │ │
│ └──────────────────────────────────────────────────────► │
│ 0 τ_π 2τ_π 3τ_π 4τ_π 5τ_π 6τ_π │
│ MW Pulse Duration │
│ │
│ Rabi frequency Ω = γB₁ where B₁ is MW field amplitude │
│ Typical Ω ~ 10-50 MHz → τ_π ~ 10-50 ns │
│ │
└──────────────────────────────────────────────────────────────────────┘
Ramsey Experiment (T2* Measurement):
────────────────────────────────────
Pulse sequence:
Time: |←Init→|←π/2→|←── Free evolution τ ──→|←π/2→|←Readout→|
Laser: ████████░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░██████████
Microwave: ░░░░░░░░█░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░█░░░░░░░░░░░░░
| | | |
| π/2 π/2 |
Expected decay (T2*):
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ Signal │
│ (contrast) │
│ ↑ │
│ │ ● │
│ 100% │ ● │
│ │ ● │
│ │ ●● │
│ 50% │ ●●● │
│ │ ●●●●● │
│ │ ●●●●●●●●●●●●●●●●●●●●●● │
│ 0% │ │
│ └──────────────────────────────────────────────────────► │
│ 0 T2* 2T2* 3T2* 4T2* 5T2* │
│ Free Evolution Time τ │
│ │
│ Typical T2* ~ 10-50 μs for V_Si in natural abundance SiC │
│ With isotopic purification: T2* ~ 100-500 μs │
│ │
└──────────────────────────────────────────────────────────────────────┘
Milestone 1.4 Success Criteria:
───────────────────────────────
□ Observe Rabi oscillations with > 5 periods visible
□ Achieve π pulse fidelity > 95%
□ Measure T2* > 10 μs at room temperature
□ Demonstrate Hahn echo, measure T2 > 100 μs
□ Implement basic dynamical decoupling, extend to T2 > 500 μs
---
Phase 2: Phononic Coupling (Months 18-36)
2.1 Phononic Crystal Fabrication
PHONONIC CRYSTAL DESIGN AND FABRICATION
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 12-24
Design Parameters:
──────────────────
Target phonon frequency: ω_ph ~ 1-5 GHz (matches typical MW transitions)
For SiC (speed of sound v ~ 13,000 m/s):
Wavelength λ = v/f = 13,000 / 5×10⁹ = 2.6 μm at 5 GHz
Phononic crystal parameters:
- Lattice constant a ~ λ/2 ~ 1.3 μm
- Hole diameter d ~ 0.6a ~ 0.8 μm
- Slab thickness t ~ 0.5a ~ 0.65 μm
Design Layout:
──────────────
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ PHONONIC CRYSTAL UNIT CELL CAVITY DESIGN │
│ │
│ ┌─────┐ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │
│ │ │ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │
│ a │ ○ │ ← hole ○ ○ ○ ○ C ○ ○ ○ ○ │
│ │ d= │ diameter ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │
│ │0.6a │ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │
│ └─────┘ │
│ C = Cavity (missing holes) │
│ a = 1.3 μm confines phonons! │
│ │
│ │
│ WAVEGUIDE DESIGN: │
│ │
│ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │
│ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │
│ ○ ○ ○ ○ C ═══════════════════ C ○ ○ ○ ○ │
│ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │
│ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │
│ │
│ C = Cavity with V_Si qubit │
│ ═ = Waveguide (line defect, one row of holes removed) │
│ │
└──────────────────────────────────────────────────────────────────────┘
Fabrication Process:
────────────────────
Step 1: Prepare 4H-SiC-on-Insulator (SiCOI)
├── Bond SiC to SiO₂/Si handle wafer (smart-cut or grinding)
├── Thin to target thickness t ~ 650 nm
└── CMP to <1 nm RMS roughness
Step 2: E-beam Lithography
├── Spin PMMA resist (200 nm)
├── E-beam write pattern (dose ~1500 μC/cm²)
├── Develop in MIBK:IPA (1:3)
└── Inspect with SEM
Step 3: Dry Etch
├── ICP-RIE with SF₆/O₂ chemistry
├── Etch rate ~100 nm/min
├── Selectivity to PMMA >5:1
└── Vertical sidewalls critical!
Step 4: Undercut Release
├── HF vapor etch to remove buried oxide
├── Creates suspended membrane
└── Critical point dry to avoid stiction
Step 5: V_Si Creation (FIB or Laser)
├── Use He⁺ FIB for single-defect precision
├── Implant C⁺ at cavity centers
├── Anneal at 900°C, 30 min, Ar ambient
└── PL scan to verify V_Si location
Equipment Required:
───────────────────
│ Equipment │ Specification │ Access │
├────────────────────────┼─────────────────────────┼────────────────┤
│ E-beam Lithography │ <10 nm resolution │ Shared fab │
│ ICP-RIE │ SiC capability │ Shared fab │
│ Focused Ion Beam │ He⁺ or Ga⁺, <10 nm │ Shared fab │
│ SEM │ <5 nm resolution │ Shared fab │
│ AFM │ <1 nm Z resolution │ In-house │
Estimated fab cost per wafer: $5,000-10,000
Typical yield: 10-50% (improving with process optimization)
2.2 Phonon-Spin Coupling Measurement
EXPERIMENT 4: MEASURING SPIN-PHONON COUPLING
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 20-30
Objective: Quantify coupling strength between V_Si spin and phononic mode
Measurement Approach:
─────────────────────
The spin-phonon coupling appears as a frequency shift in ODMR when
the phononic cavity is driven:
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ EXPERIMENTAL SETUP: │
│ │
│ ┌───────────────────────────────────────────────────────────┐ │
│ │ 4H-SiC Chip with Phononic Crystal │ │
│ │ │ │
│ │ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │ │
│ │ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │ │
│ │ ○ ○ ○ ○ ● ○ ○ ○ ○ ○ ○ ○ ○ ○ ● ○ ○ ○ │ │
│ │ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │ │
│ │ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ ○ │ │
│ │ ↑ ↑ │ │
│ │ V_Si 1 V_Si 2 │ │
│ │ │ │
│ │ ────────────────────────────────────────────────── │ │
│ │ ↑ Piezo transducer (drives phonons) │ │
│ │ │ │
│ └───────────────────────────────────────────────────────────┘ │
│ │
│ Optical access from top (confocal microscope) │
│ Microwave antenna nearby (for spin control) │
│ Piezo transducer at chip edge (for phonon injection) │
│ │
└──────────────────────────────────────────────────────────────────────┘
Measurement Protocol:
─────────────────────
Method 1: AC Stark Shift
1. Drive phononic cavity at frequency ω_cav with amplitude n_ph
2. Measure ODMR while cavity is driven
3. Observe frequency shift: Δω = g²n_ph/Δ
Where:
- g = single-phonon coupling strength (our target parameter)
- n_ph = phonon number
- Δ = detuning between spin and phonon
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ ODMR WITH PHONON DRIVE: │
│ │
│ Fluorescence │
│ ↑ │
│ │ ──────────────────────────────────────── │
│ 100% │ │
│ │ │
│ │ No drive │ With drive │
│ │ ╲ │ ╲ │
│ 95% │ ╲───────│─────────╲─────── │
│ │ │ ↑ │
│ │ │ Frequency shift Δω │
│ │ │ │
│ └─────────────────────────────────────────────► │
│ MW Frequency │
│ │
│ From shift magnitude: g = √(Δω · Δ / n_ph) │
│ Expected: g/2π ~ 10-100 kHz for V_Si in phononic cavity │
│ │
└──────────────────────────────────────────────────────────────────────┘
Method 2: Phonon-Mediated Rabi Oscillations
If spin transition is resonant with phonon mode (ω_spin = ω_cav):
- Spin ↔ Phonon exchange occurs at rate g
- Vacuum Rabi splitting visible in spectroscopy
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ AVOIDED CROSSING (Strong Coupling Regime): │
│ │
│ Spin transition │
│ frequency │
│ ↑ │
│ │ ╱ │
│ │ ╱ │
│ │ ╱ ← Upper polariton │
│ │ ● │
│ │ ╱ ╲ 2g = vacuum Rabi splitting │
│ │ ╱ ● │
│ │ ╱ ╲ ← Lower polariton │
│ │ ╱ ╲ │
│ │ ╱ ╲ │
│ └──────────────────────────────────────────► │
│ Phonon cavity frequency │
│ (tuned via strain/temp) │
│ │
│ Splitting 2g/2π ~ 20-200 kHz proves strong coupling! │
│ │
└──────────────────────────────────────────────────────────────────────┘
Milestone 2.2 Success Criteria:
───────────────────────────────
□ Fabricate phononic crystal with Q > 10⁵ at GHz frequency
□ Measure phonon-induced ODMR shift
□ Extract coupling g/2π > 10 kHz
□ Demonstrate strong coupling regime (g > κ, γ)
□ Show phonon-mediated spin manipulation
2.3 Two-Qubit Gate via Phonon Exchange
EXPERIMENT 5: PHONON-MEDIATED ENTANGLEMENT
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 28-36
Objective: Demonstrate entanglement between two V_Si spins via phonon bus
This is the KEY MILESTONE that validates the PSQA architecture!
Gate Mechanism:
───────────────
Two qubits coupled to the same phononic mode exchange excitations:
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ EFFECTIVE INTERACTION: │
│ │
│ Qubit 1 Phonon Qubit 2 │
│ │
│ │0⟩,│1⟩ │n⟩ │0⟩,│1⟩ │
│ │ │ │ │
│ │ g₁ │ g₂ │ │
│ └─────────────────┴───────────────────┘ │
│ │
│ If both qubits coupled to same mode with strengths g₁, g₂: │
│ │
│ H_eff = J (σ₁⁺σ₂⁻ + σ₁⁻σ₂⁺) = J (X₁X₂ + Y₁Y₂) / 2 │
│ │
│ Where J = g₁g₂/Δ (dispersive regime, Δ = detuning) │
│ │
│ This is an iSWAP-type interaction! │
│ │
│ After time τ = π/(2J), get √iSWAP gate │
│ After time τ = π/J, get full iSWAP gate │
│ │
└──────────────────────────────────────────────────────────────────────┘
Experimental Sequence for Bell State:
─────────────────────────────────────
Prepare |00⟩ ──► Apply H to Q1 ──► Apply √iSWAP ──► Measure both
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ PULSE SEQUENCE: │
│ │
│ Time: │←─ Init ─→│←─ H₁ ─→│←── √iSWAP ──→│←─ Readout ─→│ │
│ │
│ Laser: ████████████░░░░░░░░░░░░░░░░░░░░░░░░░░░██████████████ │
│ │ Initialize│ │ Read both │ │
│ │ both to |0⟩ │ │ │
│ │
│ MW₁: ░░░░░░░░░░░░██░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░░ │
│ │ │π/2│ │ │
│ │ │ │ Hadamard on Q1 │ │
│ │
│ Phonon:░░░░░░░░░░░░░░░░███████████████████░░░░░░░░░░░░░░░░░░ │
│ │ │ Evolution │ │ │
│ │ │ t = π/(2J) │ │ │
│ │
│ │
│ STATE EVOLUTION: │
│ │
│ |00⟩ ──H₁──► (|0⟩+|1⟩)|0⟩/√2 ──√iSWAP──► (|00⟩+i|11⟩)/√2 │
│ │
│ This is a Bell state! (maximally entangled) │
│ │
└──────────────────────────────────────────────────────────────────────┘
Bell State Verification:
────────────────────────
To verify entanglement, measure correlations in multiple bases:
1. ZZ basis: Measure both qubits, expect perfect correlation
P(00) + P(11) = 1, P(01) = P(10) = 0
2. XX basis: Apply H to both before measurement
Same correlations in X basis
3. Calculate Bell state fidelity:
F = ⟨Ψ|ρ|Ψ⟩ where |Ψ⟩ = (|00⟩+i|11⟩)/√2
Target: F > 0.9 (proves quantum entanglement!)
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ EXPECTED RESULTS (CORRELATION HISTOGRAMS): │
│ │
│ ZZ Measurement: XX Measurement: │
│ │
│ Counts Counts │
│ ↑ ↑ │
│ █████ █████ │
│ █████ █████ █████ █████ │
│ █████ █████ █████ █████ │
│ █████ █████ █████ █████ │
│ └─────┴─────┴─────┴────► └─────┴─────┴─────┴────► │
│ 00 01 10 11 ++ +- -+ -- │
│ │
│ Correlated! Correlated! │
│ (classical would (proves quantum │
│ allow this) entanglement!) │
│ │
└──────────────────────────────────────────────────────────────────────┘
Milestone 2.3 Success Criteria:
───────────────────────────────
□ Demonstrate controlled √iSWAP gate between two V_Si
□ Gate time < 10 μs
□ Gate fidelity > 90% (via process tomography)
□ Generate Bell state with fidelity > 85%
□ Violate Bell inequality (CHSH > 2)
---
Phase 3: Scaling and Error Correction (Months 36-72)
3.1 Multi-Qubit Arrays
SCALING TO 100 QUBITS
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 36-48
Architecture Design:
────────────────────
10×10 array with phononic bus connections:
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ ║ ║ ║ ║ ║ ║ ║ ║ ║ ║ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ ║ ║ ║ ║ ║ ║ ║ ║ ║ ║ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ ║ ║ ║ ║ ║ ║ ║ ║ ║ ║ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ ║ ║ ║ ║ ║ ║ ║ ║ ║ ║ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ ║ ║ ║ ║ ║ ║ ║ ║ ║ ║ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ ║ ║ ║ ║ ║ ║ ║ ║ ║ ║ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ ║ ║ ║ ║ ║ ║ ║ ║ ║ ║ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ ║ ║ ║ ║ ║ ║ ║ ║ ║ ║ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ ║ ║ ║ ║ ║ ║ ║ ║ ║ ║ │
│ ●═══●═══●═══●═══●═══●═══●═══●═══●═══● │
│ │
│ ● = V_Si qubit in phononic cavity │
│ ═ = Phononic waveguide (horizontal connections) │
│ ║ = Phononic waveguide (vertical connections) │
│ │
│ Qubit spacing: ~20 μm │
│ Chip size: ~200 μm × 200 μm (excluding I/O) │
│ Total chip: ~1 mm × 1 mm with control electronics │
│ │
└──────────────────────────────────────────────────────────────────────┘
Control Infrastructure:
───────────────────────
For 100 qubits, need efficient multiplexing:
│ Function │ Approach │ Channel Count │
├────────────────────┼─────────────────────────────┼───────────────┤
│ Optical address │ Scanning confocal + AOM │ 1 laser │
│ MW control │ Shared antenna + freq mux │ 10 frequencies│
│ Phonon control │ Shared transducers │ 10 per chip │
│ Readout │ Scanning + SNSPD array │ 1-4 detectors │
Key insight: Don't need individual wiring for each qubit!
Use frequency and spatial multiplexing.
3.2 Error Correction Implementation
SURFACE CODE ON PSQA
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 48-60
Implementation Strategy:
────────────────────────
Use 7×7 = 49 physical qubits for distance-3 surface code:
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ SURFACE CODE LAYOUT: │
│ │
│ D ─── Z ─── D ─── Z ─── D ─── Z ─── D │
│ │ │ │ │ │ │ │ │
│ X ─── D ─── X ─── D ─── X ─── D ─── X │
│ │ │ │ │ │ │ │ │
│ D ─── Z ─── D ─── Z ─── D ─── Z ─── D │
│ │ │ │ │ │ │ │ │
│ X ─── D ─── X ─── D ─── X ─── D ─── X │
│ │ │ │ │ │ │ │ │
│ D ─── Z ─── D ─── Z ─── D ─── Z ─── D │
│ │ │ │ │ │ │ │ │
│ X ─── D ─── X ─── D ─── X ─── D ─── X │
│ │ │ │ │ │ │ │ │
│ D ─── Z ─── D ─── Z ─── D ─── Z ─── D │
│ │
│ D = Data qubit (25 total) │
│ Z = Z-type stabilizer qubit (12 total) │
│ X = X-type stabilizer qubit (12 total) │
│ │
│ Total: 49 physical qubits = 1 logical qubit (distance 3) │
│ │
└──────────────────────────────────────────────────────────────────────┘
Syndrome Extraction Cycle:
──────────────────────────
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ TIME STEPS IN ONE SYNDROME CYCLE: │
│ │
│ Step 1: Initialize ancilla (all Z and X) to |0⟩ │
│ │ │
│ Step 2: Hadamard on all X-ancillas │
│ │ │
│ Step 3-6: CNOT pattern (see below) │
│ │ │
│ Step 7: Hadamard on all X-ancillas │
│ │ │
│ Step 8: Measure all ancillas │
│ │
│ │
│ CNOT PATTERN (for Z stabilizer checking DDDD): │
│ │
│ D D D D │
│ ╲ ╱ ╲ ╱ │
│ ╲ ╱ ╲ ╱ │
│ Z ═══► Z (after 4 CNOTs) │
│ ╱ ╲ ╱ ╲ │
│ ╱ ╲ ╱ ╲ │
│ D D D D │
│ │
│ Z measures the product Z₁Z₂Z₃Z₄ of its 4 data neighbors │
│ If any D flipped: Z measurement flips from +1 to -1 │
│ │
│ │
│ CYCLE TIMING: │
│ ├── Init: 500 ns │
│ ├── Hadamard: 50 ns │
│ ├── CNOT × 4: 4 × 5 μs = 20 μs │
│ ├── Hadamard: 50 ns │
│ └── Measure: 500 ns │
│ │
│ Total cycle: ~25 μs │
│ Correction rate: 40 kHz │
│ │
└──────────────────────────────────────────────────────────────────────┘
Error Thresholds:
─────────────────
For surface code to work:
- Physical error rate p < 1% (threshold varies by implementation)
- PSQA target: p ~ 0.1-0.5%
With p = 0.1% and distance d:
- d = 3: Logical error ~10⁻⁴
- d = 5: Logical error ~10⁻⁶
- d = 7: Logical error ~10⁻⁸
- d = 15: Logical error ~10⁻¹⁶
For 1M physical qubits → ~10,000 logical qubits at d=7
Milestone 3.2 Success Criteria:
───────────────────────────────
□ Implement syndrome measurement cycle
□ Achieve cycle fidelity > 99%
□ Demonstrate error detection (inject errors, see syndromes)
□ Show error correction extending logical qubit lifetime
□ Operate continuously for >1000 cycles
---
Phase 4: Million-Qubit System (Months 72-120)
4.1 Multi-Chip Optical Interconnect
OPTICAL NETWORKING BETWEEN CHIPS
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 60-84
Design:
───────
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ CHIP-TO-CHIP ENTANGLEMENT VIA OPTICAL LINK: │
│ │
│ ┌─────────────┐ ┌─────────────┐ │
│ │ CHIP A │ │ CHIP B │ │
│ │ │ │ │ │
│ │ ●────────┼──── Optical fiber ────────┼────────● │ │
│ │ V_Si A │ (920 nm) │ V_Si B │ │
│ │ │ │ │ │
│ └─────────────┘ └─────────────┘ │
│ │
│ Protocol: Barrett-Kok entanglement (heralded) │
│ │
│ 1. Excite both V_Si with resonant laser │
│ 2. Emitted photons collected into fiber │
│ 3. Photons interfere at 50/50 beamsplitter │
│ 4. Single detector click heralds Bell state |Ψ⁺⟩ or |Ψ⁻⟩ │
│ 5. Local operations correct to |Φ⁺⟩ = (|00⟩+|11⟩)/√2 │
│ │
│ │
│ INTERFERENCE SETUP: │
│ │
│ Chip A Chip B │
│ │ │ │
│ │ Photon A Photon B │ │
│ │ │ │ │ │
│ │ │ ┌───────┐ │ │ │
│ │ └────────►│ 50/50 │◄─────────┘ │ │
│ │ │ BS │ │ │
│ │ └───┬───┘ │ │
│ │ │ │ │
│ │ ┌─────┴─────┐ │ │
│ │ ▼ ▼ │ │
│ │ Det 1 Det 2 │ │
│ │ │ │
│ │ Click in Det 1 OR Det 2 (not both) │ │
│ │ = Successful entanglement! │ │
│ │
└──────────────────────────────────────────────────────────────────────┘
Performance Targets:
────────────────────
│ Parameter │ Target Value │ Current State-of-Art │
├──────────────────────┼─────────────────┼──────────────────────┤
│ Photon collection │ >50% │ ~30% (SiC waveguides)│
│ Fiber coupling │ >90% │ ~80% │
│ HOM visibility │ >90% │ ~70-80% │
│ Success probability │ >1% │ ~0.1% │
│ Entanglement rate │ >10 kHz │ ~1 kHz │
│ Fidelity │ >95% │ ~85% │
Milestone 4.1 Success Criteria:
───────────────────────────────
□ Demonstrate chip-to-chip entanglement
□ Heralding success rate > 0.1%
□ Entanglement fidelity > 90%
□ Sustained operation > 1 hour
□ Scale to 3+ chip network
4.2 Final System Integration
1 MILLION QUBIT SYSTEM ARCHITECTURE
═══════════════════════════════════════════════════════════════════════════════
Timeline: Months 84-120
System Layout:
──────────────
┌──────────────────────────────────────────────────────────────────────┐
│ │
│ FULL SYSTEM (Room scale: ~10m × 10m) │
│ │
│ ┌─────────────────────────────────────────────────────────────┐ │
│ │ QUANTUM PROCESSING UNIT │ │
│ │ │ │
│ │ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ ┌──────┐ │ │
│ │ │Rack 1│ │Rack 2│ │Rack 3│ │Rack 4│ │Rack 5│ │Rack 6│ │ │
│ │ │100k Q│ │100k Q│ │100k Q│ │100k Q│ │100k Q│ │100k Q│ │ │
│ │ └──┬───┘ └──┬───┘ └──┬───┘ └──┬───┘ └──┬───┘ └──┬───┘ │ │
│ │ │ │ │ │ │ │ │ │
│ │ └────────┴────────┴────┬───┴────────┴────────┘ │ │
│ │ │ │ │
│ │ ┌──────┴──────┐ │ │
│ │ │ OPTICAL │ │ │
│ │ │ SWITCH │ │ │
│ │ │ (1024×1024) │ │ │
│ │ └──────┬──────┘ │ │
│ │ │ │ │
│ │ ┌────────┬────────┬────┴───┬────────┬────────┐ │ │
│ │ │ │ │ │ │ │ │ │
│ │ ┌──┴───┐ ┌──┴───┐ ┌──┴───┐ ┌──┴───┐ ┌──┴───┐ ┌──┴───┐ │ │
│ │ │Rack 7│ │Rack 8│ │Rack 9│ │Rck10│ │(more │ │ ... │ │ │
│ │ │100k Q│ │100k Q│ │100k Q│ │100k Q│ │racks)│ │ │ │ │
│ │ └──────┘ └──────┘ └──────┘ └──────┘ └──────┘ └──────┘ │ │
│ │ │ │
│ │ TOTAL: 10 racks × 100,000 qubits = 1,000,000 qubits │ │
│ └──────────────────────────────────────────────────────────────┘ │
│ │
│ ┌────────────────────┐ ┌────────────────────┐ │
│ │ CLASSICAL HPC │ │ USER INTERFACE │ │
│ │ (Decoding & │◄────►│ (API, Cloud │ │
│ │ Control) │ │ Access) │ │
│ └────────────────────┘ └────────────────────┘ │
│ │
└──────────────────────────────────────────────────────────────────────┘
Cost Summary:
─────────────
│ Category │ Units │ Unit Cost │ Total │
├─────────────────────────┼──────────┼───────────┼────────────┤
│ SiC chips (10k Q each) │ 100 │ $10k │ $1M │
│ Optical interconnect │ 1 system │ $5M │ $5M │
│ Control electronics │ 10 racks │ $1M │ $10M │
│ Classical HPC │ 1 system │ $10M │ $10M │
│ Facilities │ 100 m² │ $5k/m² │ $0.5M │
│ Integration & Testing │ - │ - │ $10M │
│ Engineering (5 years) │ 50 FTEs │ $200k/yr │ $50M │
├─────────────────────────┼──────────┼───────────┼────────────┤
│ TOTAL │ │ │ ~$90M │
Compare to:
- IBM Quantum System Two: ~$100M+ (127 qubits!)
- Google's facility: ~$1B+ (100 qubits)
- PsiQuantum target: ~$1B (no qubits yet)
PSQA: $90M for 1,000,000 qubits
That's <$0.10 per qubit!
Final Milestone Success Criteria:
─────────────────────────────────
□ 1,000,000 physical qubits operational
□ Error-corrected logical qubits: >10,000
□ Fault-tolerant universal gate set
□ Run Shor's algorithm on 100+ bit numbers
□ Demonstrate quantum advantage on practical problem
□ 99.9% uptime over 1 month
---
Risk Mitigation
KEY RISKS AND MITIGATIONS
═══════════════════════════════════════════════════════════════════════════════
│ Risk │ Probability │ Impact │ Mitigation │
├─────────────────────────────┼─────────────┼────────┼─────────────────────────┤
│ Phonon coupling too weak │ Medium │ High │ Cavity enhancement, │
│ │ │ │ resonant operation │
├─────────────────────────────┼─────────────┼────────┼─────────────────────────┤
│ V_Si coherence insufficient │ Low │ High │ Isotopic purification, │
│ │ │ │ dynamical decoupling │
├─────────────────────────────┼─────────────┼────────┼─────────────────────────┤
│ Fab yield too low │ Medium │ Medium │ Process optimization, │
│ │ │ │ defect characterization │
├─────────────────────────────┼─────────────┼────────┼─────────────────────────┤
│ Optical interconnect lossy │ Low │ Medium │ Purcell enhancement, │
│ │ │ │ better collection │
├─────────────────────────────┼─────────────┼────────┼─────────────────────────┤
│ Scaling introduces crosstalk│ Medium │ High │ Frequency multiplexing, │
│ │ │ │ shielding │
├─────────────────────────────┼─────────────┼────────┼─────────────────────────┤
│ Funding gaps │ High │ High │ Milestone-based funding,│
│ │ │ │ early demos for VCs │
├─────────────────────────────┼─────────────┼────────┼─────────────────────────┤
│ Competition leapfrogs │ Medium │ Medium │ Fast iteration, │
│ │ │ │ publish and patent │
GO/NO-GO DECISION POINTS:
─────────────────────────
Month 18: Single V_Si control demonstrated?
├── YES → Continue to phononic fab
└── NO → Reassess V_Si vs other color centers
Month 36: Phonon-mediated entanglement works?
├── YES → Scale to multi-qubit
└── NO → Pivot to direct optical coupling
Month 60: 100-qubit chip with error correction?
├── YES → Scale to million qubits
└── NO → Focus on smaller-scale applications
---
Conclusion
This roadmap provides a concrete path from first experiments to a million-qubit quantum computer. The key milestones are:
1. Month 18: Single V_Si qubit control (validates basic physics) 2. Month 36: Two-qubit phonon-mediated gate (validates architecture) 3. Month 60: 100-qubit chip with error correction (validates scaling) 4. Month 120: Million-qubit system (validates practical QC)
Each phase builds on the previous, with clear go/no-go criteria. The total investment required is approximately $100M over 10 years - a fraction of what competitors are spending on less scalable approaches.
The path is challenging but achievable. Let's build it.
---
Document Version: 1.0 Date: February 2026 Author: Tushar Agrawal