Lecture 12: Classical Applications

Part IV: Applications — SCE Futures

Contents¶

Overview: SCE Applications Landscape
SQUIDs and Magnetometry
Josephson Voltage Standards
Single Photon Detectors (SNSPDs)
High-Speed ADCs and Digital RF
Medical Imaging
Scientific Instruments
Quantum Computing Control
Summary

In [1]:

Copied!





# Setup: Import libraries for visualizations
import matplotlib.pyplot as plt
import matplotlib.patches as patches
from matplotlib.patches import FancyBboxPatch, Rectangle, Circle, FancyArrowPatch
import numpy as np

# Color scheme
COLORS = {
    'primary': '#2196F3',
    'secondary': '#FF9800',
    'success': '#4CAF50',
    'danger': '#f44336',
    'dark': '#1a1a2e',
    'light': '#f5f5f5',
    'superconducting': '#00BCD4',
    'normal': '#9E9E9E',
    'purple': '#9C27B0',
    'teal': '#009688'
}

plt.rcParams['figure.facecolor'] = 'white'
plt.rcParams['axes.facecolor'] = 'white'
plt.rcParams['font.size'] = 11

print("Setup complete.")
# Setup: Import libraries for visualizations
import matplotlib.pyplot as plt
import matplotlib.patches as patches
from matplotlib.patches import FancyBboxPatch, Rectangle, Circle, FancyArrowPatch
import numpy as np

# Color scheme
COLORS = {
    'primary': '#2196F3',
    'secondary': '#FF9800',
    'success': '#4CAF50',
    'danger': '#f44336',
    'dark': '#1a1a2e',
    'light': '#f5f5f5',
    'superconducting': '#00BCD4',
    'normal': '#9E9E9E',
    'purple': '#9C27B0',
    'teal': '#009688'
}

plt.rcParams['figure.facecolor'] = 'white'
plt.rcParams['axes.facecolor'] = 'white'
plt.rcParams['font.size'] = 11

print("Setup complete.")

Setup complete.

1. Overview: SCE Applications Landscape¶

Superconducting electronics has a rich history of practical applications, many of which are in production today. This lecture covers the established and near-term applications of SCE - the technologies that work now and generate real value.

Application Maturity¶

Application	Maturity	Market	Key Technology
SQUIDs	Production	$100M+/yr	DC/RF SQUID
Voltage Standards	Production	National labs	Josephson junction arrays
SNSPDs	Production	$50M+/yr	Nanowire detectors
High-speed ADC	Demonstrated	Emerging	RSFQ comparators
Medical Imaging	Production	$500M+/yr	SQUID arrays
Quantum Control	R&D	Emerging	SFQ + AQFP

Why These Applications?¶

Each application leverages a unique property of superconductors:

Property	Enables	Applications
Zero resistance	Lossless current flow	SQUIDs, interconnects
Flux quantization	Φ₀ = 2.07 fWb exactly	Voltage standards, SQUIDs
Josephson effect	V = Φ₀ × f exactly	Voltage standards
Fast switching	ps-scale transitions	ADCs, digital logic
Low noise	Near quantum limit	SQUIDs, amplifiers
Single-photon sensitivity	Cooper pair breaking	SNSPDs

Comparison to AI Acceleration¶

Aspect	Classical Applications	AI Acceleration
Maturity	Production today	R&D phase
Scale	Small systems	Datacenter scale
Market	Specialized niches	Mass market potential
Competition	Often unique capability	Competes with GPUs
This lecture	✓ Focus	→ Lectures 13-14

In [2]:

Copied!





# Visualize: SCE applications by maturity and market size
fig, ax = plt.subplots(figsize=(12, 7))

applications = [
    ('SQUIDs/\nMagnetometry', 9, 7, 400, COLORS['success']),
    ('Voltage\nStandards', 10, 3, 200, COLORS['success']),
    ('SNSPDs', 7, 6, 300, COLORS['primary']),
    ('Medical\nImaging', 8, 8, 500, COLORS['success']),
    ('High-Speed\nADC', 5, 5, 250, COLORS['secondary']),
    ('Quantum\nControl', 3, 9, 350, COLORS['danger']),
    ('Digital\nLogic', 4, 6, 200, COLORS['secondary']),
]

for name, maturity, potential, size, color in applications:
    ax.scatter(maturity, potential, s=size, c=color, alpha=0.7, edgecolors='black', linewidth=2)
    ax.annotate(name, (maturity, potential), fontsize=10, ha='center', va='center', fontweight='bold')

ax.set_xlabel('Technology Maturity (1-10)', fontsize=12)
ax.set_ylabel('Market Potential (1-10)', fontsize=12)
ax.set_title('SCE Classical Applications: Maturity vs Potential', fontsize=14, fontweight='bold')
ax.set_xlim(0, 11)
ax.set_ylim(0, 11)
ax.grid(True, alpha=0.3)

# Quadrant labels
ax.axvline(x=5.5, color='gray', linestyle='--', alpha=0.3)
ax.axhline(y=5.5, color='gray', linestyle='--', alpha=0.3)
ax.text(2.5, 9.5, 'High Potential\nEmerging', fontsize=10, ha='center', style='italic', color='gray')
ax.text(8, 9.5, 'High Potential\nMature', fontsize=10, ha='center', style='italic', color='gray')

# Legend
ax.scatter([], [], s=100, c=COLORS['success'], label='Production', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['primary'], label='Growing', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['secondary'], label='Demonstrated', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['danger'], label='R&D', edgecolors='black')
ax.legend(loc='lower right', fontsize=10)

plt.tight_layout()
plt.show()

print("Green = Production today | Blue = Growing market | Orange = Demonstrated | Red = R&D")
# Visualize: SCE applications by maturity and market size
fig, ax = plt.subplots(figsize=(12, 7))

applications = [
    ('SQUIDs/\nMagnetometry', 9, 7, 400, COLORS['success']),
    ('Voltage\nStandards', 10, 3, 200, COLORS['success']),
    ('SNSPDs', 7, 6, 300, COLORS['primary']),
    ('Medical\nImaging', 8, 8, 500, COLORS['success']),
    ('High-Speed\nADC', 5, 5, 250, COLORS['secondary']),
    ('Quantum\nControl', 3, 9, 350, COLORS['danger']),
    ('Digital\nLogic', 4, 6, 200, COLORS['secondary']),
]

for name, maturity, potential, size, color in applications:
    ax.scatter(maturity, potential, s=size, c=color, alpha=0.7, edgecolors='black', linewidth=2)
    ax.annotate(name, (maturity, potential), fontsize=10, ha='center', va='center', fontweight='bold')

ax.set_xlabel('Technology Maturity (1-10)', fontsize=12)
ax.set_ylabel('Market Potential (1-10)', fontsize=12)
ax.set_title('SCE Classical Applications: Maturity vs Potential', fontsize=14, fontweight='bold')
ax.set_xlim(0, 11)
ax.set_ylim(0, 11)
ax.grid(True, alpha=0.3)

# Quadrant labels
ax.axvline(x=5.5, color='gray', linestyle='--', alpha=0.3)
ax.axhline(y=5.5, color='gray', linestyle='--', alpha=0.3)
ax.text(2.5, 9.5, 'High Potential\nEmerging', fontsize=10, ha='center', style='italic', color='gray')
ax.text(8, 9.5, 'High Potential\nMature', fontsize=10, ha='center', style='italic', color='gray')

# Legend
ax.scatter([], [], s=100, c=COLORS['success'], label='Production', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['primary'], label='Growing', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['secondary'], label='Demonstrated', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['danger'], label='R&D', edgecolors='black')
ax.legend(loc='lower right', fontsize=10)

plt.tight_layout()
plt.show()

print("Green = Production today | Blue = Growing market | Orange = Demonstrated | Red = R&D")

No description has been provided for this image

Green = Production today | Blue = Growing market | Orange = Demonstrated | Red = R&D

2. SQUIDs and Magnetometry¶

The SQUID (Superconducting Quantum Interference Device) is the most sensitive magnetometer ever built - and the most commercially successful SCE device.

Operating Principle¶

A SQUID consists of a superconducting loop interrupted by one (RF SQUID) or two (DC SQUID) Josephson junctions:

DC SQUID:
                    ┌───[JJ1]───┐
     Current ──────►│           │──────► Output
                    └───[JJ2]───┘
                         │
                      Φ (flux)

The critical current is modulated by magnetic flux through the loop:
I_c(Φ) = I_c0 |cos(π Φ/Φ₀)|

When biased just above its critical current, the SQUID voltage is exquisitely sensitive to flux:

$$V = R \cdot \frac{\partial I_c}{\partial \Phi} \cdot \Delta\Phi$$

SQUID Types¶

Type	Junctions	Bias	Sensitivity	Complexity
DC SQUID	2	DC current	~1 µΦ₀/√Hz	Higher
RF SQUID	1	RF tank circuit	~10 µΦ₀/√Hz	Lower

DC SQUIDs are more sensitive but require more complex readout. RF SQUIDs are simpler but noisier.

Sensitivity: The Numbers¶

Metric	Value	Comparison
Flux sensitivity	1 µΦ₀/√Hz	2 × 10⁻²¹ Wb/√Hz
Field sensitivity	1-10 fT/√Hz	Earth's field: 50 µT
Energy sensitivity	~ℏ (quantum limit)	Approaching fundamental limit

1 femtoTesla = 10⁻¹⁵ T. For comparison:

Earth's magnetic field: 50 µT (50 × 10⁻⁶ T)
MRI machine: 1.5-7 T
Human brain activity: 10-1000 fT
Human heart activity: 10-100 pT

SQUIDs can detect fields 10 billion times weaker than Earth's field.

In [3]:

Copied!





# Visualize: SQUID sensitivity compared to other magnetometers
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Magnetometer sensitivity comparison
sensors = ['Fluxgate', 'Hall Effect', 'GMR/TMR', 'OPM\n(Atomic)', 'SQUID\n(LTS)', 'SQUID\n(HTS)']
sensitivity = [1e-9, 1e-6, 1e-9, 1e-14, 1e-15, 1e-13]  # T/√Hz
colors = [COLORS['normal'], COLORS['normal'], COLORS['normal'], 
          COLORS['secondary'], COLORS['superconducting'], COLORS['primary']]

bars = ax1.barh(sensors, sensitivity, color=colors, edgecolor='black', linewidth=1.5)
ax1.set_xscale('log')
ax1.set_xlabel('Field Sensitivity (T/√Hz)', fontsize=11)
ax1.set_title('Magnetometer Sensitivity Comparison', fontsize=12, fontweight='bold')
ax1.set_xlim(1e-16, 1e-5)
ax1.grid(True, alpha=0.3, axis='x')

# Reference lines
ax1.axvline(x=1e-12, color='gray', linestyle='--', alpha=0.5)
ax1.text(1e-12, 5.5, 'pT', fontsize=9, ha='center')
ax1.axvline(x=1e-15, color='gray', linestyle='--', alpha=0.5)
ax1.text(1e-15, 5.5, 'fT', fontsize=9, ha='center')

# Right: Signal sources and required sensitivity
ax2.set_title('Magnetic Signal Sources', fontsize=12, fontweight='bold')

signals = [
    ("Earth's field", 5e-5, COLORS['normal']),
    ('MRI gradient', 1e-2, COLORS['normal']),
    ('Car at 10m', 1e-7, COLORS['secondary']),
    ('Human heart (MCG)', 5e-11, COLORS['primary']),
    ('Human brain (MEG)', 1e-13, COLORS['superconducting']),
    ('Single neuron', 1e-15, COLORS['danger']),
]

for i, (name, field, color) in enumerate(signals):
    ax2.barh(i, field, color=color, edgecolor='black', linewidth=1.5, height=0.6)

ax2.set_yticks(range(len(signals)))
ax2.set_yticklabels([s[0] for s in signals])
ax2.set_xscale('log')
ax2.set_xlabel('Magnetic Field (T)', fontsize=11)
ax2.set_xlim(1e-16, 1)
ax2.grid(True, alpha=0.3, axis='x')

# SQUID detection threshold
ax2.axvline(x=1e-15, color=COLORS['superconducting'], linestyle='-', linewidth=2)
ax2.text(3e-16, 5.5, 'SQUID\nlimit', fontsize=9, ha='center', color=COLORS['superconducting'])

plt.tight_layout()
plt.show()

print("SQUIDs are 1000-1,000,000× more sensitive than competing technologies.")
print("This enables unique applications like MEG and fundamental physics experiments.")
# Visualize: SQUID sensitivity compared to other magnetometers
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Magnetometer sensitivity comparison
sensors = ['Fluxgate', 'Hall Effect', 'GMR/TMR', 'OPM\n(Atomic)', 'SQUID\n(LTS)', 'SQUID\n(HTS)']
sensitivity = [1e-9, 1e-6, 1e-9, 1e-14, 1e-15, 1e-13]  # T/√Hz
colors = [COLORS['normal'], COLORS['normal'], COLORS['normal'], 
          COLORS['secondary'], COLORS['superconducting'], COLORS['primary']]

bars = ax1.barh(sensors, sensitivity, color=colors, edgecolor='black', linewidth=1.5)
ax1.set_xscale('log')
ax1.set_xlabel('Field Sensitivity (T/√Hz)', fontsize=11)
ax1.set_title('Magnetometer Sensitivity Comparison', fontsize=12, fontweight='bold')
ax1.set_xlim(1e-16, 1e-5)
ax1.grid(True, alpha=0.3, axis='x')

# Reference lines
ax1.axvline(x=1e-12, color='gray', linestyle='--', alpha=0.5)
ax1.text(1e-12, 5.5, 'pT', fontsize=9, ha='center')
ax1.axvline(x=1e-15, color='gray', linestyle='--', alpha=0.5)
ax1.text(1e-15, 5.5, 'fT', fontsize=9, ha='center')

# Right: Signal sources and required sensitivity
ax2.set_title('Magnetic Signal Sources', fontsize=12, fontweight='bold')

signals = [
    ("Earth's field", 5e-5, COLORS['normal']),
    ('MRI gradient', 1e-2, COLORS['normal']),
    ('Car at 10m', 1e-7, COLORS['secondary']),
    ('Human heart (MCG)', 5e-11, COLORS['primary']),
    ('Human brain (MEG)', 1e-13, COLORS['superconducting']),
    ('Single neuron', 1e-15, COLORS['danger']),
]

for i, (name, field, color) in enumerate(signals):
    ax2.barh(i, field, color=color, edgecolor='black', linewidth=1.5, height=0.6)

ax2.set_yticks(range(len(signals)))
ax2.set_yticklabels([s[0] for s in signals])
ax2.set_xscale('log')
ax2.set_xlabel('Magnetic Field (T)', fontsize=11)
ax2.set_xlim(1e-16, 1)
ax2.grid(True, alpha=0.3, axis='x')

# SQUID detection threshold
ax2.axvline(x=1e-15, color=COLORS['superconducting'], linestyle='-', linewidth=2)
ax2.text(3e-16, 5.5, 'SQUID\nlimit', fontsize=9, ha='center', color=COLORS['superconducting'])

plt.tight_layout()
plt.show()

print("SQUIDs are 1000-1,000,000× more sensitive than competing technologies.")
print("This enables unique applications like MEG and fundamental physics experiments.")

SQUIDs are 1000-1,000,000× more sensitive than competing technologies.
This enables unique applications like MEG and fundamental physics experiments.

SQUID Applications¶

Application	Field Range	Market	Key Players
Magnetoencephalography (MEG)	10-1000 fT	$200M/yr	Elekta, CTF, MEGIN
Magnetocardiography (MCG)	10-100 pT	$50M/yr	CardioMag, Biomagnetik
Geophysical exploration	pT-nT	$100M/yr	SQUID systems
Non-destructive testing	nT-µT	$50M/yr	Conductus, Tristan
Materials characterization	Variable	Research	Quantum Design, MPMS
Fundamental physics	fT	Research	Universities, national labs

Commercial SQUID Systems¶

Quantum Design MPMS (Magnetic Property Measurement System):

Standard tool in materials science labs worldwide
Measures magnetic susceptibility, hysteresis, moment vs. temperature
Sensitivity: 10⁻⁸ emu (10⁻¹¹ A·m²)
Used in: Superconductor characterization, nanoparticle studies, thin films

MEG Systems (306-channel helmet arrays):

Map brain activity with millisecond time resolution
Clinical use: epilepsy localization, pre-surgical planning
Research: cognitive neuroscience, brain-computer interfaces
Cost: $2-4M per system

SQUID Gradiometers¶

Most practical SQUID systems use gradiometers rather than magnetometers:

First-order gradiometer:        Second-order gradiometer:
                                
    ┌──┐  + coil                    ┌──┐  +1
    │  │                            │  │
    │  │  baseline                  ├──┤  -2
    │  │  (5-10 cm)                 │  │
    └──┘  - coil                    └──┘  +1
    
Measures: dB/dz              Measures: d²B/dz²
Rejects: uniform fields      Rejects: uniform + gradient

Gradiometers reject distant noise sources (power lines, Earth's field variations) while remaining sensitive to nearby sources (brain, heart, material under test).

3. Josephson Voltage Standards¶

The Josephson voltage standard (JVS) is arguably the most precise measurement device ever built, and it's based entirely on superconducting electronics.

The Josephson Effect: Voltage from Frequency¶

When a Josephson junction is irradiated with microwaves at frequency f, it develops quantized voltage steps:

$$V_n = n \cdot \frac{h}{2e} \cdot f = n \cdot \Phi_0 \cdot f$$

where:

n = step number (integer)
h = Planck's constant (exactly defined since 2019)
e = elementary charge (exactly defined since 2019)
Φ₀ = h/2e = 2.067833848... × 10⁻¹⁵ Wb (exact)

The voltage depends only on frequency and fundamental constants - no material properties!

Why This Matters¶

Before 1990, the "volt" was defined by electrochemical cells (Weston cells) that drifted over time. Now:

Era	Voltage Standard	Uncertainty
Pre-1972	Weston cells	~10⁻⁵
1972-1990	Josephson (advisory)	~10⁻⁸
1990-2019	Josephson (K_J-90)	~10⁻⁹
2019-present	Josephson (exact h, e)	Exact definition

Since the 2019 SI redefinition, the Josephson effect provides the exact realization of the volt.

Programmable Josephson Voltage Standards (PJVS)¶

Modern voltage standards use arrays of thousands of junctions:

System	Junctions	Voltage Range	Uncertainty
NIST PJVS	300,000	±10 V	<10⁻⁹
PTB (Germany)	70,000	±1 V	<10⁻⁹
Conventional JVS	20,000	1-10 V	~10⁻⁸

Each junction contributes ~70 µV per GHz of microwave frequency. To reach 10 V requires tens of thousands of junctions in series.

In [4]:

Copied!





# Visualize: Josephson voltage standard concept
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Shapiro steps (I-V curve with microwave)
I = np.linspace(-2, 2, 1000)
V_base = np.tanh(I * 3)  # Simplified I-V

# Add Shapiro steps
step_voltage = 0.3  # Arbitrary units for visualization
for n in range(-2, 3):
    step_I = np.linspace(-0.3, 0.3, 50) + n * 0.5
    step_V = np.ones_like(step_I) * n * step_voltage
    ax1.plot(step_I, step_V, 'b-', linewidth=3)

ax1.plot(I, V_base, 'gray', linewidth=1, alpha=0.5, linestyle='--', label='No microwave')
ax1.set_xlabel('Current (I/I_c)', fontsize=11)
ax1.set_ylabel('Voltage (normalized)', fontsize=11)
ax1.set_title('Shapiro Steps: Quantized Voltage Levels', fontsize=12, fontweight='bold')
ax1.axhline(0, color='gray', linewidth=0.5)
ax1.axvline(0, color='gray', linewidth=0.5)

# Annotate steps
for n in [-2, -1, 0, 1, 2]:
    ax1.annotate(f'n={n}', xy=(0.5, n*step_voltage), fontsize=10, 
                color=COLORS['primary'], fontweight='bold')

ax1.text(1.5, 0.7, 'V = n × (h/2e) × f\n= n × 483.6 MHz/µV × f', 
        fontsize=10, bbox=dict(boxstyle='round', facecolor='white', edgecolor='gray'))
ax1.set_xlim(-2, 2)
ax1.set_ylim(-0.8, 0.8)
ax1.grid(True, alpha=0.3)

# Right: Voltage standard uncertainty over time
years = [1960, 1972, 1990, 2000, 2010, 2019, 2024]
uncertainties = [1e-5, 1e-6, 1e-8, 1e-9, 1e-9, 1e-10, 1e-10]

ax2.semilogy(years, uncertainties, 'o-', color=COLORS['superconducting'], 
            linewidth=2.5, markersize=10)
ax2.set_xlabel('Year', fontsize=11)
ax2.set_ylabel('Voltage Uncertainty', fontsize=11)
ax2.set_title('Improvement in Voltage Standards', fontsize=12, fontweight='bold')
ax2.grid(True, alpha=0.3)

# Annotate key events
events = [
    (1972, 1e-6, 'Josephson\nadopted'),
    (1990, 1e-8, 'K_J-90\nstandard'),
    (2019, 1e-10, 'SI redefinition\n(exact h, e)'),
]
for year, unc, label in events:
    ax2.annotate(label, xy=(year, unc), xytext=(year-5, unc*0.1),
                fontsize=9, arrowprops=dict(arrowstyle='->', color='black'))

ax2.set_xlim(1955, 2030)
ax2.set_ylim(1e-11, 1e-4)

plt.tight_layout()
plt.show()

# Calculate actual numbers
f_GHz = 70  # GHz
Phi_0 = 2.067833848e-15  # Wb
V_per_junction = Phi_0 * f_GHz * 1e9  # Volts
print(f"At {f_GHz} GHz, each junction contributes: {V_per_junction*1e6:.1f} µV")
print(f"For 10V output, need: {10 / V_per_junction:,.0f} junctions")
# Visualize: Josephson voltage standard concept
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Shapiro steps (I-V curve with microwave)
I = np.linspace(-2, 2, 1000)
V_base = np.tanh(I * 3)  # Simplified I-V

# Add Shapiro steps
step_voltage = 0.3  # Arbitrary units for visualization
for n in range(-2, 3):
    step_I = np.linspace(-0.3, 0.3, 50) + n * 0.5
    step_V = np.ones_like(step_I) * n * step_voltage
    ax1.plot(step_I, step_V, 'b-', linewidth=3)

ax1.plot(I, V_base, 'gray', linewidth=1, alpha=0.5, linestyle='--', label='No microwave')
ax1.set_xlabel('Current (I/I_c)', fontsize=11)
ax1.set_ylabel('Voltage (normalized)', fontsize=11)
ax1.set_title('Shapiro Steps: Quantized Voltage Levels', fontsize=12, fontweight='bold')
ax1.axhline(0, color='gray', linewidth=0.5)
ax1.axvline(0, color='gray', linewidth=0.5)

# Annotate steps
for n in [-2, -1, 0, 1, 2]:
    ax1.annotate(f'n={n}', xy=(0.5, n*step_voltage), fontsize=10, 
                color=COLORS['primary'], fontweight='bold')

ax1.text(1.5, 0.7, 'V = n × (h/2e) × f\n= n × 483.6 MHz/µV × f', 
        fontsize=10, bbox=dict(boxstyle='round', facecolor='white', edgecolor='gray'))
ax1.set_xlim(-2, 2)
ax1.set_ylim(-0.8, 0.8)
ax1.grid(True, alpha=0.3)

# Right: Voltage standard uncertainty over time
years = [1960, 1972, 1990, 2000, 2010, 2019, 2024]
uncertainties = [1e-5, 1e-6, 1e-8, 1e-9, 1e-9, 1e-10, 1e-10]

ax2.semilogy(years, uncertainties, 'o-', color=COLORS['superconducting'], 
            linewidth=2.5, markersize=10)
ax2.set_xlabel('Year', fontsize=11)
ax2.set_ylabel('Voltage Uncertainty', fontsize=11)
ax2.set_title('Improvement in Voltage Standards', fontsize=12, fontweight='bold')
ax2.grid(True, alpha=0.3)

# Annotate key events
events = [
    (1972, 1e-6, 'Josephson\nadopted'),
    (1990, 1e-8, 'K_J-90\nstandard'),
    (2019, 1e-10, 'SI redefinition\n(exact h, e)'),
]
for year, unc, label in events:
    ax2.annotate(label, xy=(year, unc), xytext=(year-5, unc*0.1),
                fontsize=9, arrowprops=dict(arrowstyle='->', color='black'))

ax2.set_xlim(1955, 2030)
ax2.set_ylim(1e-11, 1e-4)

plt.tight_layout()
plt.show()

# Calculate actual numbers
f_GHz = 70  # GHz
Phi_0 = 2.067833848e-15  # Wb
V_per_junction = Phi_0 * f_GHz * 1e9  # Volts
print(f"At {f_GHz} GHz, each junction contributes: {V_per_junction*1e6:.1f} µV")
print(f"For 10V output, need: {10 / V_per_junction:,.0f} junctions")

At 70 GHz, each junction contributes: 144.7 µV
For 10V output, need: 69,085 junctions

4. Single Photon Detectors (SNSPDs)¶

Superconducting Nanowire Single-Photon Detectors (SNSPDs) are the highest-performance single-photon detectors available, revolutionizing fields from quantum communications to deep-space optical links.

Operating Principle¶

Nanowire (NbN, WSi, MoSi) ~100 nm wide, ~5 nm thick
Biased just below critical current

    Photon arrives
          ↓
    ┌─────●─────┐  ← Nanowire (superconducting)
    │           │
    
          ↓ photon absorbed, Cooper pairs broken
          
    ┌──▓▓▓▓▓──┐  ← Local hotspot (normal)
    │           │
    
          ↓ current diverted, voltage pulse
          
    Output: ~mV pulse, ~100 ps duration

Photon absorbed → breaks Cooper pairs → local "hotspot"
Current diverted around hotspot → superconductivity lost locally
Normal-state resistance develops → detectable voltage pulse
Heat dissipates → nanowire recovers → ready for next photon

Performance Metrics¶

Metric	SNSPD (best)	APD (InGaAs)	PMT
Detection efficiency	>95%	25%	40%
Dark count rate	<1 Hz	10-100 kHz	kHz
Timing jitter	<20 ps	50-100 ps	300 ps
Recovery time	~10 ns	~µs	~10 ns
Wavelength range	UV to mid-IR	NIR	UV-NIR
Operating temp	2-4 K	200-300 K	300 K

SNSPDs dominate in every performance metric except operating temperature.

SNSPD Applications¶

Application	Wavelength	Key Benefit	Status
Quantum key distribution (QKD)	1550 nm	Low dark counts, high efficiency	Deployed
Deep-space optical comm	1064/1550 nm	Single-photon sensitivity	NASA DSOC
LIDAR	1550 nm	Timing precision	R&D
Fluorescence lifetime imaging	Visible	ps timing jitter	Research
Quantum computing readout	Various	High fidelity	Growing

NASA DSOC: Deep Space Optical Communications¶

The Deep Space Optical Communications (DSOC) experiment on NASA's Psyche mission (launched 2023) uses SNSPDs for receiving laser signals from deep space:

Distance: Up to 2.5 AU (Mars distance)
Data rate: 10-100× better than RF
Detector: 64-pixel SNSPD array
Operating at 1 K

This demonstrates SNSPDs in a practical, high-stakes application.

Commercial SNSPD Systems¶

Vendor	Model	Efficiency	Dark Counts	Price
ID Quantique	ID281	>85%	<10 Hz	~$100K
Quantum Opus	Opus One	>90%	<1 Hz	~$150K
Photon Spot	Various	>80%	<100 Hz	~$80K
Single Quantum	Eos	>85%	<10 Hz	~$100K

The market is growing rapidly, driven by quantum communication and research applications.

In [5]:

Copied!





# Visualize: SNSPD performance comparison
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Detection efficiency vs wavelength
wavelengths = np.linspace(400, 2000, 100)
# Simplified efficiency curves
snspd_eff = 95 * np.exp(-((wavelengths - 1550)**2) / (2 * 400**2))
snspd_eff = np.clip(snspd_eff, 20, 95)
apd_eff = 25 * np.exp(-((wavelengths - 1000)**2) / (2 * 300**2))
apd_eff = np.clip(apd_eff, 5, 25)

ax1.plot(wavelengths, snspd_eff, 'b-', linewidth=2.5, label='SNSPD (NbN)')
ax1.plot(wavelengths, apd_eff, 'r--', linewidth=2, label='InGaAs APD')
ax1.fill_between(wavelengths, snspd_eff, alpha=0.2, color='blue')

ax1.axvline(x=1550, color='gray', linestyle=':', alpha=0.7)
ax1.text(1550, 85, 'Telecom\n1550 nm', fontsize=9, ha='center')

ax1.set_xlabel('Wavelength (nm)', fontsize=11)
ax1.set_ylabel('Detection Efficiency (%)', fontsize=11)
ax1.set_title('Single-Photon Detector Efficiency', fontsize=12, fontweight='bold')
ax1.legend(loc='upper right')
ax1.grid(True, alpha=0.3)
ax1.set_xlim(400, 2000)
ax1.set_ylim(0, 100)

# Right: Timing jitter comparison
detectors = ['PMT', 'APD\n(Si)', 'APD\n(InGaAs)', 'SNSPD\n(NbN)', 'SNSPD\n(WSi)']
jitter = [300, 100, 80, 30, 15]  # ps
colors = [COLORS['normal'], COLORS['secondary'], COLORS['secondary'], 
          COLORS['superconducting'], COLORS['primary']]

bars = ax2.bar(detectors, jitter, color=colors, edgecolor='black', linewidth=1.5)
ax2.set_ylabel('Timing Jitter (ps)', fontsize=11)
ax2.set_title('Detector Timing Precision', fontsize=12, fontweight='bold')
ax2.grid(True, alpha=0.3, axis='y')

# Annotate best
ax2.annotate('Best: <20 ps', xy=(4, 15), xytext=(3.5, 80),
            fontsize=10, arrowprops=dict(arrowstyle='->', color='black'))

plt.tight_layout()
plt.show()

print("SNSPDs dominate in both efficiency (>95%) and timing precision (<20 ps).")
print("The only drawback is the cryogenic operating temperature (1-4 K).")
# Visualize: SNSPD performance comparison
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Detection efficiency vs wavelength
wavelengths = np.linspace(400, 2000, 100)
# Simplified efficiency curves
snspd_eff = 95 * np.exp(-((wavelengths - 1550)**2) / (2 * 400**2))
snspd_eff = np.clip(snspd_eff, 20, 95)
apd_eff = 25 * np.exp(-((wavelengths - 1000)**2) / (2 * 300**2))
apd_eff = np.clip(apd_eff, 5, 25)

ax1.plot(wavelengths, snspd_eff, 'b-', linewidth=2.5, label='SNSPD (NbN)')
ax1.plot(wavelengths, apd_eff, 'r--', linewidth=2, label='InGaAs APD')
ax1.fill_between(wavelengths, snspd_eff, alpha=0.2, color='blue')

ax1.axvline(x=1550, color='gray', linestyle=':', alpha=0.7)
ax1.text(1550, 85, 'Telecom\n1550 nm', fontsize=9, ha='center')

ax1.set_xlabel('Wavelength (nm)', fontsize=11)
ax1.set_ylabel('Detection Efficiency (%)', fontsize=11)
ax1.set_title('Single-Photon Detector Efficiency', fontsize=12, fontweight='bold')
ax1.legend(loc='upper right')
ax1.grid(True, alpha=0.3)
ax1.set_xlim(400, 2000)
ax1.set_ylim(0, 100)

# Right: Timing jitter comparison
detectors = ['PMT', 'APD\n(Si)', 'APD\n(InGaAs)', 'SNSPD\n(NbN)', 'SNSPD\n(WSi)']
jitter = [300, 100, 80, 30, 15]  # ps
colors = [COLORS['normal'], COLORS['secondary'], COLORS['secondary'], 
          COLORS['superconducting'], COLORS['primary']]

bars = ax2.bar(detectors, jitter, color=colors, edgecolor='black', linewidth=1.5)
ax2.set_ylabel('Timing Jitter (ps)', fontsize=11)
ax2.set_title('Detector Timing Precision', fontsize=12, fontweight='bold')
ax2.grid(True, alpha=0.3, axis='y')

# Annotate best
ax2.annotate('Best: <20 ps', xy=(4, 15), xytext=(3.5, 80),
            fontsize=10, arrowprops=dict(arrowstyle='->', color='black'))

plt.tight_layout()
plt.show()

print("SNSPDs dominate in both efficiency (>95%) and timing precision (<20 ps).")
print("The only drawback is the cryogenic operating temperature (1-4 K).")

SNSPDs dominate in both efficiency (>95%) and timing precision (<20 ps).
The only drawback is the cryogenic operating temperature (1-4 K).

5. High-Speed ADCs and Digital RF¶

Superconducting ADCs leverage the ultra-fast switching of Josephson junctions to achieve sampling rates impossible with semiconductor technology.

Why Superconducting ADCs?¶

Advantage	Mechanism	Benefit
Ultra-fast comparators	ps JJ switching	>50 GHz sampling
Low aperture jitter	Quantized flux	High ENOB at high frequency
Low power	~µW per comparator	Dense integration
Low noise	Cryogenic operation	Better SNR

ADC Architectures¶

Architecture	Speed	Resolution	SCE Implementation
Flash	Fastest	Low (4-8 bit)	Parallel comparators
Sigma-Delta	Moderate	High (12-16 bit)	Oversampling + filtering
Pipelined	High	Medium (10-12 bit)	Staged conversion
SAR	Moderate	High	Sequential approximation

Demonstrated Performance¶

Project	Sample Rate	Resolution	Year	Notes
HYPRES/Northrop	20 GS/s	8-bit ENOB	2010s	RSFQ-based
NIST	10 GS/s	10-bit ENOB	2015	Sigma-delta
Yokohama	5 GS/s	6-bit	2020	AQFP comparator

Digital RF Receivers¶

The "holy grail" application: directly digitizing RF signals without analog downconversion.

Traditional RF receiver:
  Antenna → LNA → Mixer → IF Filter → ADC → DSP
                    ↑
               Local Oscillator

Superconducting digital RF:
  Antenna → LNA → SCE ADC → Digital Processing
                    ↑
            Direct digitization at GHz

Benefits:

Eliminate analog components (mixers, filters, LOs)
Software-defined radio becomes truly digital
Wideband operation without tuning
Ideal for: radar, communications, spectrum monitoring

In [6]:

Copied!





# Visualize: ADC performance comparison (Walden chart style)
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Sample rate vs resolution (simplified Walden chart)
# CMOS ADCs
cmos_rates = [1e6, 10e6, 100e6, 1e9, 10e9]  # Sample rate
cmos_enob = [16, 14, 12, 10, 6]  # Effective bits

# SCE ADCs (demonstrated/projected)
sce_rates = [5e9, 10e9, 20e9, 50e9]
sce_enob = [6, 8, 8, 6]

ax1.scatter(cmos_rates, cmos_enob, s=100, c=COLORS['secondary'], 
           label='CMOS ADCs', edgecolors='black', linewidth=1.5)
ax1.scatter(sce_rates, sce_enob, s=150, c=COLORS['superconducting'], 
           label='SCE ADCs', edgecolors='black', linewidth=2, marker='s')

ax1.set_xscale('log')
ax1.set_xlabel('Sample Rate (S/s)', fontsize=11)
ax1.set_ylabel('Effective Number of Bits (ENOB)', fontsize=11)
ax1.set_title('ADC Performance: CMOS vs SCE', fontsize=12, fontweight='bold')
ax1.legend(loc='upper right')
ax1.grid(True, alpha=0.3)
ax1.set_xlim(1e5, 1e11)
ax1.set_ylim(0, 18)

# Highlight SCE advantage region
ax1.axvspan(10e9, 100e9, alpha=0.2, color=COLORS['superconducting'])
ax1.text(30e9, 16, 'SCE\nAdvantage\nZone', fontsize=10, ha='center', 
        color=COLORS['superconducting'], fontweight='bold')

# Right: Digital RF concept
ax2.set_title('Digital RF Receiver Architecture', fontsize=12, fontweight='bold')
ax2.axis('off')

# Traditional
ax2.text(0.5, 0.95, 'Traditional (Analog)', fontsize=11, ha='center', 
        fontweight='bold', transform=ax2.transAxes)
trad_blocks = [('ANT', 0.05), ('LNA', 0.2), ('Mixer', 0.35), ('Filter', 0.5), 
               ('ADC', 0.65), ('DSP', 0.8)]
for name, x in trad_blocks:
    rect = FancyBboxPatch((x-0.05, 0.75), 0.1, 0.12, boxstyle="round,pad=0.01",
                          facecolor=COLORS['normal'], edgecolor='black', 
                          linewidth=1, transform=ax2.transAxes)
    ax2.add_patch(rect)
    ax2.text(x, 0.81, name, fontsize=8, ha='center', va='center', 
            transform=ax2.transAxes, color='white')
    if x < 0.8:
        ax2.annotate('', xy=(x+0.06, 0.81), xytext=(x+0.09, 0.81),
                    arrowprops=dict(arrowstyle='->', color='black', lw=1),
                    transform=ax2.transAxes)

# SCE Digital RF
ax2.text(0.5, 0.55, 'Superconducting (Digital)', fontsize=11, ha='center', 
        fontweight='bold', color=COLORS['superconducting'], transform=ax2.transAxes)
sce_blocks = [('ANT', 0.1, COLORS['normal']), ('LNA', 0.3, COLORS['normal']), 
              ('SCE\nADC', 0.5, COLORS['superconducting']), ('Digital\nProc', 0.7, COLORS['primary'])]
for name, x, color in sce_blocks:
    rect = FancyBboxPatch((x-0.07, 0.35), 0.14, 0.15, boxstyle="round,pad=0.01",
                          facecolor=color, edgecolor='black', 
                          linewidth=1.5, transform=ax2.transAxes)
    ax2.add_patch(rect)
    ax2.text(x, 0.425, name, fontsize=9, ha='center', va='center', 
            transform=ax2.transAxes, color='white', fontweight='bold')
    if x < 0.7:
        ax2.annotate('', xy=(x+0.08, 0.425), xytext=(x+0.12, 0.425),
                    arrowprops=dict(arrowstyle='->', color='black', lw=1.5),
                    transform=ax2.transAxes)

ax2.text(0.5, 0.15, 'Eliminates mixer, LO, IF filtering\n→ Software-defined, wideband operation', 
        fontsize=10, ha='center', transform=ax2.transAxes, style='italic')

plt.tight_layout()
plt.show()

print("SCE ADCs enable direct RF digitization at rates impossible with CMOS.")
print("Applications: radar, communications, spectrum monitoring, radio astronomy.")
# Visualize: ADC performance comparison (Walden chart style)
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Sample rate vs resolution (simplified Walden chart)
# CMOS ADCs
cmos_rates = [1e6, 10e6, 100e6, 1e9, 10e9]  # Sample rate
cmos_enob = [16, 14, 12, 10, 6]  # Effective bits

# SCE ADCs (demonstrated/projected)
sce_rates = [5e9, 10e9, 20e9, 50e9]
sce_enob = [6, 8, 8, 6]

ax1.scatter(cmos_rates, cmos_enob, s=100, c=COLORS['secondary'], 
           label='CMOS ADCs', edgecolors='black', linewidth=1.5)
ax1.scatter(sce_rates, sce_enob, s=150, c=COLORS['superconducting'], 
           label='SCE ADCs', edgecolors='black', linewidth=2, marker='s')

ax1.set_xscale('log')
ax1.set_xlabel('Sample Rate (S/s)', fontsize=11)
ax1.set_ylabel('Effective Number of Bits (ENOB)', fontsize=11)
ax1.set_title('ADC Performance: CMOS vs SCE', fontsize=12, fontweight='bold')
ax1.legend(loc='upper right')
ax1.grid(True, alpha=0.3)
ax1.set_xlim(1e5, 1e11)
ax1.set_ylim(0, 18)

# Highlight SCE advantage region
ax1.axvspan(10e9, 100e9, alpha=0.2, color=COLORS['superconducting'])
ax1.text(30e9, 16, 'SCE\nAdvantage\nZone', fontsize=10, ha='center', 
        color=COLORS['superconducting'], fontweight='bold')

# Right: Digital RF concept
ax2.set_title('Digital RF Receiver Architecture', fontsize=12, fontweight='bold')
ax2.axis('off')

# Traditional
ax2.text(0.5, 0.95, 'Traditional (Analog)', fontsize=11, ha='center', 
        fontweight='bold', transform=ax2.transAxes)
trad_blocks = [('ANT', 0.05), ('LNA', 0.2), ('Mixer', 0.35), ('Filter', 0.5), 
               ('ADC', 0.65), ('DSP', 0.8)]
for name, x in trad_blocks:
    rect = FancyBboxPatch((x-0.05, 0.75), 0.1, 0.12, boxstyle="round,pad=0.01",
                          facecolor=COLORS['normal'], edgecolor='black', 
                          linewidth=1, transform=ax2.transAxes)
    ax2.add_patch(rect)
    ax2.text(x, 0.81, name, fontsize=8, ha='center', va='center', 
            transform=ax2.transAxes, color='white')
    if x < 0.8:
        ax2.annotate('', xy=(x+0.06, 0.81), xytext=(x+0.09, 0.81),
                    arrowprops=dict(arrowstyle='->', color='black', lw=1),
                    transform=ax2.transAxes)

# SCE Digital RF
ax2.text(0.5, 0.55, 'Superconducting (Digital)', fontsize=11, ha='center', 
        fontweight='bold', color=COLORS['superconducting'], transform=ax2.transAxes)
sce_blocks = [('ANT', 0.1, COLORS['normal']), ('LNA', 0.3, COLORS['normal']), 
              ('SCE\nADC', 0.5, COLORS['superconducting']), ('Digital\nProc', 0.7, COLORS['primary'])]
for name, x, color in sce_blocks:
    rect = FancyBboxPatch((x-0.07, 0.35), 0.14, 0.15, boxstyle="round,pad=0.01",
                          facecolor=color, edgecolor='black', 
                          linewidth=1.5, transform=ax2.transAxes)
    ax2.add_patch(rect)
    ax2.text(x, 0.425, name, fontsize=9, ha='center', va='center', 
            transform=ax2.transAxes, color='white', fontweight='bold')
    if x < 0.7:
        ax2.annotate('', xy=(x+0.08, 0.425), xytext=(x+0.12, 0.425),
                    arrowprops=dict(arrowstyle='->', color='black', lw=1.5),
                    transform=ax2.transAxes)

ax2.text(0.5, 0.15, 'Eliminates mixer, LO, IF filtering\n→ Software-defined, wideband operation', 
        fontsize=10, ha='center', transform=ax2.transAxes, style='italic')

plt.tight_layout()
plt.show()

print("SCE ADCs enable direct RF digitization at rates impossible with CMOS.")
print("Applications: radar, communications, spectrum monitoring, radio astronomy.")

SCE ADCs enable direct RF digitization at rates impossible with CMOS.
Applications: radar, communications, spectrum monitoring, radio astronomy.

6. Medical Imaging¶

Superconducting electronics enables several medical imaging modalities that are impossible or impractical with conventional technology.

Magnetoencephalography (MEG)¶

MEG maps brain activity by detecting the tiny magnetic fields (10-1000 fT) produced by neuronal currents.

Aspect	Details
Signal source	Ionic currents in neurons
Field strength	10-1000 fT (10⁻¹⁴ to 10⁻¹² T)
Detector	300+ channel SQUID array
Time resolution	~1 ms
Spatial resolution	~5 mm

MEG System Configuration:

        ┌──────────────────────────────┐
        │     Helmet-shaped array      │
        │    ┌─────┐  ┌─────┐         │
        │    │SQUID│  │SQUID│  × 300+ │
        │    └──┬──┘  └──┬──┘         │
        │       │        │             │
        │    Gradiometer coils         │
        │       │        │             │
        │    ┌──┴────────┴──┐         │
        │    │   Patient    │         │
        │    │    Head      │         │
        │    └──────────────┘         │
        └──────────────────────────────┘
               Inside MSR
        (Magnetically Shielded Room)

Clinical Applications of MEG¶

Application	Use Case	Value
Epilepsy localization	Pre-surgical mapping	Identify seizure focus
Tumor mapping	Functional areas near tumors	Preserve critical function
Stroke recovery	Monitor rehabilitation	Track brain reorganization
Cognitive research	Brain function studies	Millisecond dynamics

Commercial MEG Systems¶

Vendor	System	Channels	Installed Base
MEGIN (Elekta)	TRIUX neo	306	~100 worldwide
CTF	MEG	275	~50 worldwide
Ricoh	MEG	160	~30 (Japan)

Market size: ~$200M/year, growing with clinical adoption.

In [7]:

Copied!





# Visualize: MEG and MCG signal ranges
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Brain signal sources and magnetic fields
sources = ['Alpha rhythm\n(occipital)', 'Evoked response\n(auditory)', 'Epileptic\nspike', 
           'Spontaneous\nactivity', 'Deep sources\n(hippocampus)']
fields = [300, 100, 1000, 50, 10]  # femtoTesla

colors = [COLORS['primary'] if f > 50 else COLORS['secondary'] for f in fields]
bars = ax1.barh(sources, fields, color=colors, edgecolor='black', linewidth=1.5)
ax1.set_xlabel('Magnetic Field at Scalp (fT)', fontsize=11)
ax1.set_title('MEG: Brain Signal Magnitudes', fontsize=12, fontweight='bold')
ax1.set_xscale('log')
ax1.set_xlim(1, 10000)
ax1.grid(True, alpha=0.3, axis='x')

# SQUID sensitivity line
ax1.axvline(x=5, color=COLORS['superconducting'], linestyle='--', linewidth=2)
ax1.text(3, 4.5, 'SQUID\nnoise floor', fontsize=9, ha='right', 
        color=COLORS['superconducting'])

# Right: MCG - heart signals
ax2.set_title('MCG: Cardiac Magnetic Mapping', fontsize=12, fontweight='bold')

# Heart magnetic field map (simplified)
x = np.linspace(-1, 1, 50)
y = np.linspace(-1, 1, 50)
X, Y = np.meshgrid(x, y)

# Dipole field pattern (simplified)
R = np.sqrt(X**2 + Y**2) + 0.1
theta = np.arctan2(Y, X)
Bz = np.sin(theta) / R**2
Bz = np.clip(Bz, -10, 10)

im = ax2.contourf(X, Y, Bz, levels=20, cmap='RdBu_r')
ax2.set_xlabel('Position (normalized)', fontsize=11)
ax2.set_ylabel('Position (normalized)', fontsize=11)
ax2.set_aspect('equal')

# Heart outline
heart_x = 0.3 * np.sin(np.linspace(0, 2*np.pi, 100))
heart_y = 0.3 * np.cos(np.linspace(0, 2*np.pi, 100))
ax2.plot(heart_x, heart_y - 0.1, 'k-', linewidth=2)
ax2.text(0, -0.1, '♥', fontsize=20, ha='center', va='center')

cbar = plt.colorbar(im, ax=ax2, label='Magnetic Field (pT)')

plt.tight_layout()
plt.show()

print("MEG: 10-1000 fT signals from brain, requires SQUID sensitivity")
print("MCG: 10-100 pT signals from heart, can detect arrhythmias non-invasively")
# Visualize: MEG and MCG signal ranges
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: Brain signal sources and magnetic fields
sources = ['Alpha rhythm\n(occipital)', 'Evoked response\n(auditory)', 'Epileptic\nspike', 
           'Spontaneous\nactivity', 'Deep sources\n(hippocampus)']
fields = [300, 100, 1000, 50, 10]  # femtoTesla

colors = [COLORS['primary'] if f > 50 else COLORS['secondary'] for f in fields]
bars = ax1.barh(sources, fields, color=colors, edgecolor='black', linewidth=1.5)
ax1.set_xlabel('Magnetic Field at Scalp (fT)', fontsize=11)
ax1.set_title('MEG: Brain Signal Magnitudes', fontsize=12, fontweight='bold')
ax1.set_xscale('log')
ax1.set_xlim(1, 10000)
ax1.grid(True, alpha=0.3, axis='x')

# SQUID sensitivity line
ax1.axvline(x=5, color=COLORS['superconducting'], linestyle='--', linewidth=2)
ax1.text(3, 4.5, 'SQUID\nnoise floor', fontsize=9, ha='right', 
        color=COLORS['superconducting'])

# Right: MCG - heart signals
ax2.set_title('MCG: Cardiac Magnetic Mapping', fontsize=12, fontweight='bold')

# Heart magnetic field map (simplified)
x = np.linspace(-1, 1, 50)
y = np.linspace(-1, 1, 50)
X, Y = np.meshgrid(x, y)

# Dipole field pattern (simplified)
R = np.sqrt(X**2 + Y**2) + 0.1
theta = np.arctan2(Y, X)
Bz = np.sin(theta) / R**2
Bz = np.clip(Bz, -10, 10)

im = ax2.contourf(X, Y, Bz, levels=20, cmap='RdBu_r')
ax2.set_xlabel('Position (normalized)', fontsize=11)
ax2.set_ylabel('Position (normalized)', fontsize=11)
ax2.set_aspect('equal')

# Heart outline
heart_x = 0.3 * np.sin(np.linspace(0, 2*np.pi, 100))
heart_y = 0.3 * np.cos(np.linspace(0, 2*np.pi, 100))
ax2.plot(heart_x, heart_y - 0.1, 'k-', linewidth=2)
ax2.text(0, -0.1, '♥', fontsize=20, ha='center', va='center')

cbar = plt.colorbar(im, ax=ax2, label='Magnetic Field (pT)')

plt.tight_layout()
plt.show()

print("MEG: 10-1000 fT signals from brain, requires SQUID sensitivity")
print("MCG: 10-100 pT signals from heart, can detect arrhythmias non-invasively")

MEG: 10-1000 fT signals from brain, requires SQUID sensitivity
MCG: 10-100 pT signals from heart, can detect arrhythmias non-invasively

Magnetocardiography (MCG)¶

MCG measures the magnetic field of the heart, complementing ECG:

Comparison	ECG	MCG
Measures	Electric potential	Magnetic field
Contact	Electrodes on skin	Non-contact
Affected by	Tissue conductivity	Not affected
Spatial resolution	Limited	Superior
Equipment	Portable, cheap	SQUID array, shielded room

MCG advantages:

Non-contact: No electrode placement issues
Volume conductor independent: Not distorted by tissue
Better localization: For arrhythmia source finding

Fetal MCG¶

A unique application: non-invasive fetal heart monitoring

Fetal ECG is contaminated by maternal signals
Fetal MCG can be spatially separated from maternal MCG
Enables detection of fetal arrhythmias before birth
Critical for high-risk pregnancies

Ultra-Low-Field MRI¶

Emerging application using SQUIDs instead of high-field superconducting magnets:

Conventional MRI	ULF-MRI
1.5-7 T field	~µT field
Large SC magnet	No magnet
SQUID not needed	SQUID detection
High SNR, fast	Lower SNR, slower
Metal artifacts	No artifacts
Shielded room	Shielded room

ULF-MRI can image near metal implants and may enable combined MEG/MRI.

7. Scientific Instruments¶

SCE enables scientific instruments with capabilities impossible to achieve otherwise.

Radio Astronomy¶

Superconducting mixers and amplifiers are essential for radio astronomy:

Component	Technology	Application
SIS mixers	Nb-AlOx-Nb	mm/sub-mm receivers
SQUID amplifiers	DC SQUID	Low-noise amplification
Digital spectrometers	RSFQ	High-speed signal processing

ALMA (Atacama Large Millimeter Array):

66 radio telescopes in Chile
SIS receivers for all high-frequency bands
Superconducting mixers enable detection of faint cosmic signals

Particle Physics¶

Experiment	SCE Component	Purpose
ADMX	SQUID amplifier	Axion dark matter search
Various	TES arrays	X-ray detection
CMB experiments	TES/KID arrays	Cosmic microwave background

Transition Edge Sensors (TES)¶

TES bolometers operate at the superconducting transition:

Resistance vs Temperature:

R ↑
  │         ┌─── Normal state
  │        /
  │       /  ← Operating point
  │      /     (steep transition)
  │─────/
  │    Superconducting state
  └────────────────────────► T
         T_c

Small temperature change → Large resistance change
→ Extremely sensitive power measurement

Sensitivity: ~10⁻¹⁸ W (attowatt level)

Applications:

X-ray spectroscopy (better than semiconductor detectors)
Cosmic microwave background measurements
Infrared astronomy

In [8]:

Copied!





# Visualize: TES operating principle and scientific applications
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: TES transition curve
T = np.linspace(0, 1.5, 200)
T_c = 1.0
width = 0.05

# Sharp transition at Tc
R = 1 / (1 + np.exp(-(T - T_c) / (width)))

ax1.plot(T, R, 'b-', linewidth=2.5)
ax1.fill_between(T[T < T_c - 2*width], R[T < T_c - 2*width], alpha=0.3, 
                color=COLORS['superconducting'], label='Superconducting')
ax1.fill_between(T[T > T_c + 2*width], R[T > T_c + 2*width], alpha=0.3, 
                color=COLORS['secondary'], label='Normal')

# Operating point
op_T = T_c
op_R = 0.5
ax1.scatter([op_T], [op_R], s=150, c=COLORS['danger'], zorder=5, 
           edgecolors='white', linewidth=2)
ax1.annotate('Operating\npoint', xy=(op_T, op_R), xytext=(op_T + 0.2, op_R + 0.2),
            fontsize=10, arrowprops=dict(arrowstyle='->', color='black'))

ax1.set_xlabel('Temperature (T/T_c)', fontsize=11)
ax1.set_ylabel('Resistance (R/R_n)', fontsize=11)
ax1.set_title('TES: Transition Edge Operation', fontsize=12, fontweight='bold')
ax1.legend(loc='lower right')
ax1.grid(True, alpha=0.3)
ax1.set_xlim(0, 1.5)
ax1.set_ylim(-0.05, 1.1)

# Annotation
ax1.text(0.3, 0.8, 'dR/dT → ∞\nat transition\n→ extreme\nsensitivity', 
        fontsize=9, style='italic', ha='center')

# Right: Scientific applications by wavelength
ax2.set_title('SCE Detectors Across the Spectrum', fontsize=12, fontweight='bold')

# Wavelength ranges
apps = [
    ('Radio\n(ALMA)', 1e-3, 1e-1, COLORS['normal']),
    ('mm-wave\n(CMB)', 1e-4, 1e-2, COLORS['secondary']),
    ('Infrared\n(astronomy)', 1e-6, 1e-4, COLORS['purple']),
    ('Optical\n(SNSPD)', 4e-7, 2e-6, COLORS['primary']),
    ('X-ray\n(TES)', 1e-11, 1e-8, COLORS['superconducting']),
    ('Gamma\n(TES)', 1e-13, 1e-10, COLORS['danger']),
]

for i, (name, wl_min, wl_max, color) in enumerate(apps):
    ax2.barh(i, wl_max - wl_min, left=wl_min, color=color, 
            edgecolor='black', linewidth=1.5, height=0.6)
    ax2.text(wl_min * 0.3, i, name, fontsize=10, va='center', ha='right')

ax2.set_xscale('log')
ax2.set_xlabel('Wavelength (m)', fontsize=11)
ax2.set_xlim(1e-14, 1)
ax2.set_yticks([])
ax2.grid(True, alpha=0.3, axis='x')

# Label key wavelengths
ax2.axvline(x=5e-7, color='green', linestyle=':', alpha=0.5)
ax2.text(5e-7, 5.7, 'Visible', fontsize=8, ha='center', rotation=90)

plt.tight_layout()
plt.show()

print("SCE detectors span from radio waves to gamma rays.")
print("Each wavelength regime uses optimized superconducting sensor technology.")
# Visualize: TES operating principle and scientific applications
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 5))

# Left: TES transition curve
T = np.linspace(0, 1.5, 200)
T_c = 1.0
width = 0.05

# Sharp transition at Tc
R = 1 / (1 + np.exp(-(T - T_c) / (width)))

ax1.plot(T, R, 'b-', linewidth=2.5)
ax1.fill_between(T[T < T_c - 2*width], R[T < T_c - 2*width], alpha=0.3, 
                color=COLORS['superconducting'], label='Superconducting')
ax1.fill_between(T[T > T_c + 2*width], R[T > T_c + 2*width], alpha=0.3, 
                color=COLORS['secondary'], label='Normal')

# Operating point
op_T = T_c
op_R = 0.5
ax1.scatter([op_T], [op_R], s=150, c=COLORS['danger'], zorder=5, 
           edgecolors='white', linewidth=2)
ax1.annotate('Operating\npoint', xy=(op_T, op_R), xytext=(op_T + 0.2, op_R + 0.2),
            fontsize=10, arrowprops=dict(arrowstyle='->', color='black'))

ax1.set_xlabel('Temperature (T/T_c)', fontsize=11)
ax1.set_ylabel('Resistance (R/R_n)', fontsize=11)
ax1.set_title('TES: Transition Edge Operation', fontsize=12, fontweight='bold')
ax1.legend(loc='lower right')
ax1.grid(True, alpha=0.3)
ax1.set_xlim(0, 1.5)
ax1.set_ylim(-0.05, 1.1)

# Annotation
ax1.text(0.3, 0.8, 'dR/dT → ∞\nat transition\n→ extreme\nsensitivity', 
        fontsize=9, style='italic', ha='center')

# Right: Scientific applications by wavelength
ax2.set_title('SCE Detectors Across the Spectrum', fontsize=12, fontweight='bold')

# Wavelength ranges
apps = [
    ('Radio\n(ALMA)', 1e-3, 1e-1, COLORS['normal']),
    ('mm-wave\n(CMB)', 1e-4, 1e-2, COLORS['secondary']),
    ('Infrared\n(astronomy)', 1e-6, 1e-4, COLORS['purple']),
    ('Optical\n(SNSPD)', 4e-7, 2e-6, COLORS['primary']),
    ('X-ray\n(TES)', 1e-11, 1e-8, COLORS['superconducting']),
    ('Gamma\n(TES)', 1e-13, 1e-10, COLORS['danger']),
]

for i, (name, wl_min, wl_max, color) in enumerate(apps):
    ax2.barh(i, wl_max - wl_min, left=wl_min, color=color, 
            edgecolor='black', linewidth=1.5, height=0.6)
    ax2.text(wl_min * 0.3, i, name, fontsize=10, va='center', ha='right')

ax2.set_xscale('log')
ax2.set_xlabel('Wavelength (m)', fontsize=11)
ax2.set_xlim(1e-14, 1)
ax2.set_yticks([])
ax2.grid(True, alpha=0.3, axis='x')

# Label key wavelengths
ax2.axvline(x=5e-7, color='green', linestyle=':', alpha=0.5)
ax2.text(5e-7, 5.7, 'Visible', fontsize=8, ha='center', rotation=90)

plt.tight_layout()
plt.show()

print("SCE detectors span from radio waves to gamma rays.")
print("Each wavelength regime uses optimized superconducting sensor technology.")

SCE detectors span from radio waves to gamma rays.
Each wavelength regime uses optimized superconducting sensor technology.

Kinetic Inductance Detectors (KIDs)¶

KIDs are an alternative to TES for large detector arrays:

Aspect	TES	KID
Readout	Individual SQUID per pixel	Frequency-multiplexed
Complexity	High (wiring)	Low (single feedline)
Pixels	~10,000 demonstrated	~100,000 possible
Sensitivity	Higher	Slightly lower
Operating T	~100 mK	~100 mK

KID principle: photon absorption changes kinetic inductance of superconductor, shifting resonator frequency.

KID Array Readout:
                                    
    Single microwave feedline
    ─────┬─────┬─────┬─────┬─────
         │     │     │     │
        ┌┴┐   ┌┴┐   ┌┴┐   ┌┴┐
        │1│   │2│   │3│   │4│   ← Resonators at
        └─┘   └─┘   └─┘   └─┘     different frequencies
        
    Each resonator = one pixel
    Frequency shift = absorbed power

Applications:

CMB experiments (BLAST-TNG, TolTEC)
Ground-based sub-mm astronomy
Future space missions

Josephson Parametric Amplifiers (JPAs)¶

JPAs provide near-quantum-limited amplification:

Parameter	JPA	HEMT
Noise temperature	~100 mK (near quantum limit)	~2 K
Bandwidth	~100 MHz	GHz
Gain	20 dB	40 dB
Operating T	~20 mK	4 K

Critical for:

Quantum computing readout
Axion dark matter searches
Ultra-sensitive measurements

8. Quantum Computing Control¶

Superconducting electronics may provide the scalability solution for quantum computers.

The Wiring Problem¶

Current quantum computers face a fundamental challenge:

Current approach:
                    Room Temperature (300K)
    ┌─────────────────────────────────────────┐
    │  Control electronics: AWGs, DACs, etc.  │
    │  (rack-mounted, ~kW per qubit)          │
    └───────────────────┬─────────────────────┘
                        │ Coax cables (many per qubit)
                        │ ~20 cables per qubit
    ┌───────────────────┴─────────────────────┐
    │              4K Stage                    │
    │        (just thermal anchoring)          │
    └───────────────────┬─────────────────────┘
                        │
    ┌───────────────────┴─────────────────────┐
    │             ~20 mK                       │
    │         Quantum processor                │
    │          (100s of qubits)                │
    └─────────────────────────────────────────┘
    
Problem: 1000 qubits × 20 cables = 20,000 cables!
Heat leak and physical space become impossible.

SCE Solution: Control at 4K¶

Approach	Temperature	Technology	Status
Room-temp control	300 K	CMOS	Current standard
Cryo-CMOS	4 K	Si CMOS	Active R&D
SCE control	4 K	SFQ/AQFP	Research

SCE advantages for quantum control:

Native 4K operation (no heating issues)
Ultra-low power (~pW per gate)
Fast switching (GHz rates for qubit control)
Reduced cable count: only power and data to 300K

In [9]:

Copied!





# Visualize: Quantum computer wiring challenge
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 6))

# Left: Wiring scaling problem
qubits = np.array([10, 50, 100, 500, 1000, 10000])
cables_per_qubit = 20
total_cables = qubits * cables_per_qubit

ax1.semilogy(qubits, total_cables, 'o-', color=COLORS['danger'], 
            linewidth=2.5, markersize=10, label='Current approach')

# With SCE at 4K
sce_cables = qubits * 0.1 + 100  # Much fewer cables to room temp
ax1.semilogy(qubits, sce_cables, 's--', color=COLORS['superconducting'], 
            linewidth=2.5, markersize=10, label='SCE at 4K')

ax1.set_xlabel('Number of Qubits', fontsize=11)
ax1.set_ylabel('Cables to Room Temperature', fontsize=11)
ax1.set_title('Quantum Computer Wiring Challenge', fontsize=12, fontweight='bold')
ax1.legend(loc='upper left')
ax1.grid(True, alpha=0.3)

# Practical limit line
ax1.axhline(y=10000, color='gray', linestyle=':', alpha=0.7)
ax1.text(5000, 12000, 'Practical limit', fontsize=9, color='gray')

# Annotations
ax1.annotate('1M qubits for\nerror correction', xy=(10000, 200000), 
            xytext=(3000, 500000), fontsize=9,
            arrowprops=dict(arrowstyle='->', color='black'))

# Right: SCE-enabled architecture
ax2.set_title('SCE-Enabled Quantum Architecture', fontsize=12, fontweight='bold')
ax2.axis('off')

# Temperature stages
stages = [
    ('300 K', 0.85, COLORS['danger'], 'Digital control\n(minimal)'),
    ('4 K', 0.55, COLORS['primary'], 'SFQ pulse generation\nAQFP readout processing'),
    ('20 mK', 0.2, COLORS['superconducting'], 'Qubit array\n+ local SCE demux'),
]

for label, y, color, desc in stages:
    rect = FancyBboxPatch((0.1, y-0.1), 0.8, 0.2, boxstyle="round,pad=0.02",
                          facecolor=color, edgecolor='black', 
                          linewidth=2, alpha=0.3, transform=ax2.transAxes)
    ax2.add_patch(rect)
    ax2.text(0.15, y, label, fontsize=12, fontweight='bold', 
            transform=ax2.transAxes, va='center')
    ax2.text(0.5, y, desc, fontsize=10, transform=ax2.transAxes, 
            va='center', ha='center')

# Arrows showing reduced wiring
ax2.annotate('', xy=(0.5, 0.65), xytext=(0.5, 0.75),
            arrowprops=dict(arrowstyle='<->', color='black', lw=2),
            transform=ax2.transAxes)
ax2.text(0.7, 0.7, 'Few cables\n(power, data)', fontsize=9, 
        transform=ax2.transAxes)

ax2.annotate('', xy=(0.5, 0.35), xytext=(0.5, 0.45),
            arrowprops=dict(arrowstyle='<->', color='black', lw=2),
            transform=ax2.transAxes)
ax2.text(0.7, 0.4, 'On-chip\nconnections', fontsize=9, 
        transform=ax2.transAxes)

plt.tight_layout()
plt.show()

print("SCE at 4K could reduce cables by 100× or more.")
print("This may be essential for scaling to 1M+ qubits.")
# Visualize: Quantum computer wiring challenge
fig, (ax1, ax2) = plt.subplots(1, 2, figsize=(14, 6))

# Left: Wiring scaling problem
qubits = np.array([10, 50, 100, 500, 1000, 10000])
cables_per_qubit = 20
total_cables = qubits * cables_per_qubit

ax1.semilogy(qubits, total_cables, 'o-', color=COLORS['danger'], 
            linewidth=2.5, markersize=10, label='Current approach')

# With SCE at 4K
sce_cables = qubits * 0.1 + 100  # Much fewer cables to room temp
ax1.semilogy(qubits, sce_cables, 's--', color=COLORS['superconducting'], 
            linewidth=2.5, markersize=10, label='SCE at 4K')

ax1.set_xlabel('Number of Qubits', fontsize=11)
ax1.set_ylabel('Cables to Room Temperature', fontsize=11)
ax1.set_title('Quantum Computer Wiring Challenge', fontsize=12, fontweight='bold')
ax1.legend(loc='upper left')
ax1.grid(True, alpha=0.3)

# Practical limit line
ax1.axhline(y=10000, color='gray', linestyle=':', alpha=0.7)
ax1.text(5000, 12000, 'Practical limit', fontsize=9, color='gray')

# Annotations
ax1.annotate('1M qubits for\nerror correction', xy=(10000, 200000), 
            xytext=(3000, 500000), fontsize=9,
            arrowprops=dict(arrowstyle='->', color='black'))

# Right: SCE-enabled architecture
ax2.set_title('SCE-Enabled Quantum Architecture', fontsize=12, fontweight='bold')
ax2.axis('off')

# Temperature stages
stages = [
    ('300 K', 0.85, COLORS['danger'], 'Digital control\n(minimal)'),
    ('4 K', 0.55, COLORS['primary'], 'SFQ pulse generation\nAQFP readout processing'),
    ('20 mK', 0.2, COLORS['superconducting'], 'Qubit array\n+ local SCE demux'),
]

for label, y, color, desc in stages:
    rect = FancyBboxPatch((0.1, y-0.1), 0.8, 0.2, boxstyle="round,pad=0.02",
                          facecolor=color, edgecolor='black', 
                          linewidth=2, alpha=0.3, transform=ax2.transAxes)
    ax2.add_patch(rect)
    ax2.text(0.15, y, label, fontsize=12, fontweight='bold', 
            transform=ax2.transAxes, va='center')
    ax2.text(0.5, y, desc, fontsize=10, transform=ax2.transAxes, 
            va='center', ha='center')

# Arrows showing reduced wiring
ax2.annotate('', xy=(0.5, 0.65), xytext=(0.5, 0.75),
            arrowprops=dict(arrowstyle='<->', color='black', lw=2),
            transform=ax2.transAxes)
ax2.text(0.7, 0.7, 'Few cables\n(power, data)', fontsize=9, 
        transform=ax2.transAxes)

ax2.annotate('', xy=(0.5, 0.35), xytext=(0.5, 0.45),
            arrowprops=dict(arrowstyle='<->', color='black', lw=2),
            transform=ax2.transAxes)
ax2.text(0.7, 0.4, 'On-chip\nconnections', fontsize=9, 
        transform=ax2.transAxes)

plt.tight_layout()
plt.show()

print("SCE at 4K could reduce cables by 100× or more.")
print("This may be essential for scaling to 1M+ qubits.")

SCE at 4K could reduce cables by 100× or more.
This may be essential for scaling to 1M+ qubits.

SCE Components for Quantum Control¶

Function	SCE Technology	Advantage
Pulse generation	SFQ	ps-precision timing
DAC	RSFQ/AQFP	Digital pulse shaping
Multiplexing	SFQ switches	Reduce cable count
Readout	SQUID + AQFP	Low-noise processing
Classical logic	AQFP	Error correction

Research Programs¶

Program	Focus	Participants
IARPA SuperCables	Cryogenic interconnects	Multiple
Google/UCSB	Cryo-CMOS + SCE	Industry
IBM	Modular cryo architecture	Industry
Academic	SFQ pulse generators	Universities

The Vision: Fault-Tolerant Quantum Computing¶

Fault-tolerant quantum computing requires millions of physical qubits for error correction.

Requirement progression:

NISQ era (now):      ~100 qubits    → Room-temp control works
Near-term:           ~1,000 qubits  → Challenging but possible
Fault-tolerant:      ~1M qubits     → Requires new approach

SCE provides a path to the million-qubit scale.

Key metrics (targets vary by architecture):

100s-1000s of qubits controlled per SCE chip (architecture dependent)
<1 µW power dissipation per qubit
<1 ns timing precision
99.9% pulse fidelity

9. Summary¶

Classical SCE Applications at a Glance¶

Application	Technology	Maturity	Key Achievement
SQUIDs	DC/RF SQUID	Production	fT sensitivity
Voltage Standards	JJ arrays	Production	Exact volt definition
SNSPDs	Nanowire	Production	>95% efficiency
High-speed ADC	RSFQ	Demonstrated	20+ GS/s
MEG/MCG	SQUID arrays	Production	Brain/heart imaging
Scientific	TES, KID, SIS	Production	Astrophysics, physics
Quantum Control	SFQ/AQFP	Research	Scalability solution

Why These Applications Succeed¶

Each successful SCE application leverages a unique capability that semiconductors cannot match:

Application	Unique SCE Capability
SQUIDs	Flux quantization → ultimate sensitivity
Voltage Standards	Josephson relation → exact metrology
SNSPDs	Cooper pair breaking → single photons
ADCs	ps switching → extreme speed
Medical Imaging	SQUID sensitivity → brain signals
Quantum Control	Native cryogenic + low power

Compared to AI Acceleration¶

Aspect	Classical Apps	AI Acceleration
Maturity	Production today	R&D phase
Market	$500M+/year	Future potential
Competition	Often unique	Competes with GPUs
Integration	Small scale OK	Needs >1M gates
Memory	Minimal	TB-scale challenge

Classical applications work because they:

Require small-scale integration
Don't need large memory
Provide capabilities impossible otherwise

In [10]:

Copied!





# Visualize: Summary - SCE applications landscape
fig, ax = plt.subplots(figsize=(12, 8))
ax.set_title('SCE Classical Applications: Maturity vs Market Impact', fontsize=14, fontweight='bold')

# Applications with maturity and market size
applications = [
    ('SQUIDs\n(Magnetometry)', 9, 7, 500, COLORS['success']),
    ('Voltage\nStandards', 10, 3, 200, COLORS['success']),
    ('SNSPDs', 8, 6, 400, COLORS['success']),
    ('High-Speed\nADC', 6, 5, 300, COLORS['secondary']),
    ('MEG/MCG\n(Medical)', 9, 8, 600, COLORS['success']),
    ('TES/KID\n(Science)', 8, 4, 250, COLORS['primary']),
    ('Quantum\nControl', 4, 9, 500, COLORS['danger']),
]

for name, maturity, impact, size, color in applications:
    ax.scatter(maturity, impact, s=size, c=color, alpha=0.7, 
              edgecolors='black', linewidth=2)
    ax.annotate(name, (maturity, impact), fontsize=10, ha='center', 
               va='center', fontweight='bold')

ax.set_xlabel('Technology Maturity (1-10)', fontsize=12)
ax.set_ylabel('Market Impact (1-10)', fontsize=12)
ax.set_xlim(0, 11)
ax.set_ylim(0, 11)
ax.grid(True, alpha=0.3)

# Legend
ax.scatter([], [], s=100, c=COLORS['success'], label='Production', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['primary'], label='Established R&D', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['secondary'], label='Demonstrated', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['danger'], label='Active R&D', edgecolors='black')
ax.legend(loc='lower right', fontsize=10)

# Quadrant annotations
ax.axvline(x=7, color='gray', linestyle='--', alpha=0.3)
ax.axhline(y=5.5, color='gray', linestyle='--', alpha=0.3)

plt.tight_layout()
plt.show()

print("\n" + "="*60)
print("KEY TAKEAWAY")
print("="*60)
print("Classical SCE applications are in production TODAY, generating")
print("$500M+/year in market value. These technologies leverage unique")
print("superconducting properties that semiconductors cannot replicate.")
print("="*60)
# Visualize: Summary - SCE applications landscape
fig, ax = plt.subplots(figsize=(12, 8))
ax.set_title('SCE Classical Applications: Maturity vs Market Impact', fontsize=14, fontweight='bold')

# Applications with maturity and market size
applications = [
    ('SQUIDs\n(Magnetometry)', 9, 7, 500, COLORS['success']),
    ('Voltage\nStandards', 10, 3, 200, COLORS['success']),
    ('SNSPDs', 8, 6, 400, COLORS['success']),
    ('High-Speed\nADC', 6, 5, 300, COLORS['secondary']),
    ('MEG/MCG\n(Medical)', 9, 8, 600, COLORS['success']),
    ('TES/KID\n(Science)', 8, 4, 250, COLORS['primary']),
    ('Quantum\nControl', 4, 9, 500, COLORS['danger']),
]

for name, maturity, impact, size, color in applications:
    ax.scatter(maturity, impact, s=size, c=color, alpha=0.7, 
              edgecolors='black', linewidth=2)
    ax.annotate(name, (maturity, impact), fontsize=10, ha='center', 
               va='center', fontweight='bold')

ax.set_xlabel('Technology Maturity (1-10)', fontsize=12)
ax.set_ylabel('Market Impact (1-10)', fontsize=12)
ax.set_xlim(0, 11)
ax.set_ylim(0, 11)
ax.grid(True, alpha=0.3)

# Legend
ax.scatter([], [], s=100, c=COLORS['success'], label='Production', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['primary'], label='Established R&D', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['secondary'], label='Demonstrated', edgecolors='black')
ax.scatter([], [], s=100, c=COLORS['danger'], label='Active R&D', edgecolors='black')
ax.legend(loc='lower right', fontsize=10)

# Quadrant annotations
ax.axvline(x=7, color='gray', linestyle='--', alpha=0.3)
ax.axhline(y=5.5, color='gray', linestyle='--', alpha=0.3)

plt.tight_layout()
plt.show()

print("\n" + "="*60)
print("KEY TAKEAWAY")
print("="*60)
print("Classical SCE applications are in production TODAY, generating")
print("$500M+/year in market value. These technologies leverage unique")
print("superconducting properties that semiconductors cannot replicate.")
print("="*60)

============================================================
KEY TAKEAWAY
============================================================
Classical SCE applications are in production TODAY, generating
$500M+/year in market value. These technologies leverage unique
superconducting properties that semiconductors cannot replicate.
============================================================

Key Numbers to Remember¶

Parameter	Value	Context
SQUID sensitivity	~1 fT/√Hz	10 billion × better than Earth's field
Josephson voltage	483.6 MHz/µV	Exact (h/2e)
SNSPD efficiency	>95%	Best single-photon detector
ADC sample rate	>20 GS/s	Direct RF digitization
MEG channels	300+	Whole-head brain imaging
Voltage standard uncertainty	<10⁻⁹	Exact SI definition

Connections to AI Acceleration¶

The classical applications covered here provide the technology foundation for AI acceleration:

Fabrication: Same Nb trilayer process
Testing: SQUID-based characterization
I/O concepts: Lessons from ADC, digital RF
Cryogenics: Shared infrastructure
Control electronics: Quantum control → AI control