AI summary by TwinMind
Lecture 13: CECE and Bolometry
MIT OpenCourseWare
1h 19m 43s
Electron Cyclotron Emission (ECE) Recap
- Multiple peaks in emissivity spectrum at frequencies ωₙ = n × (eB/γmₑ) × relativistic factors
- Peaks broadened asymmetrically by particle distribution
- Frequencies depend only on magnetic field (EB/mₑ)
- Toroidal geometry: B ∝ 1/R creates spatial mapping
- Different radial positions (R₁, R₂, R₃) have different B-fields (B₁, B₂, B₃)
- Each location emits at different frequencies
- Low harmonics (m=1,2) in optically thick regime → black body spectrum I(ν) = Tν²/c²
- Temperature profile measurement: frequency → B-field → spatial coordinate → temperature
- Accessibility considerations: O-mode and X-mode propagation
- CMA diagram reference posted by Sean
Correlation ECE (CECE)
- Problem: Standard ECE cannot measure small temperature fluctuations (ΔTₑ/T ≈ 1%)
- Noise >> temperature fluctuation signal
- Fluctuations related to turbulence (key for fusion reactor economics)
- ASDEX Upgrade Setup:
- Bandpass filter: 110 GHz ± 5 GHz (10 GHz bandwidth)
- Heterodyne mixing with 100 GHz → 0-10 GHz beat frequency
- Multiple narrow channels: ~100 MHz bandwidth each, 250 MHz spacing
- Non-overlapping frequency channels → non-overlapping spatial regions
- Sampling ~100 microns spatial regions (1000× smaller than plasma radius)
- Digitization at 4 MS/s (affordable compared to hundreds of GS/s)
- Correlation Technique:
- Adjacent channels S₁, S₂ measure nearby plasma volumes
- Turbulent eddy size Rₜᵤᵣᵦ ≈ 100 μm > ΔR₁₂
- Same temperature fluctuation in both channels
- Signal: S = temperature fluctuation + noise
- Cross-correlation: ⟨S₁S₂⟩ ∝ temperature² (noise terms average to zero)
- Performance: 1% temperature fluctuations at 13 radial locations, 100 kHz time resolution
Bremsstrahlung (Free-Free) Radiation
- Heavy ions + electrons → electron deflection → photon emission
- Isotropic emission (unlike ECE anisotropy)
- Multiple calculation approaches (classical, semi-classical, quantum) give same scaling
- Maxwellian-averaged emissivity:
- 4πJ(ν) = nₑnᵢZ²T⁻¹/²exp(-hν/T) × constants × gaunt factor
- Units: watts/Hz/m³
- Gaunt factor G̅ ≈ 1-2 (weakly varying with hν/T from 0.1 to 10)
- Spectrum: Simple exponential decay with photon energy
- Always present as "irreducible minimum" emission
- Constant over small frequency ranges → can subtract as background
- Can couple into O/X modes depending on observation geometry
- Optical thickness considerations at low frequencies
Recombination (Free-Bound) Radiation
- Electron + ion → bound electron + photon
- Energy conservation: hν = ½mv² + Z²Ry/n²
- Discrete energy levels: En = -Z²Ry/n² (n = 1,2,3,...)
- Ridberg energy Ry = 13.6 eV
- Emissivity per level n:
- Similar form to bremsstrahlung with additional terms
- Strength ∝ Z²/n³ (strongest for n=1)
- Additional exponential factor: exp(-Z²Ry/n²T)
- Spectrum: Bremsstrahlung continuum + recombination edges
- Sharp edges at photon energies > kinetic energy thresholds
- Series of edges for n=1,2,3,4...
- Total emission: J = Jᵦᵣₑₘ + Σₙ Jₙ
- Ion charge Z affects energy level positions → different recombination lines
Bolometry
- Purpose: Measure total radiated power Pᵣₐd = ∫J(ν)dν
- Motivation: Power balance in fusion plasma
- Pₐₗₚₕₐ + Pₑₓₜ = Pcond + Sᵣₐd
- Need to quantify/minimize radiative losses
- Basic Principle:
- Radiation heats resistor → resistance change R(T)
- Voltage divider: VR = V₀R/(R+M)
- Measure resistance → temperature → incident power
- Heat equation: P(t) = C(dT/dt) + (ΔT/τ)
- C = heat capacity, τ = thermal transport time constant
- Wheatstone Bridge Design:
- Four resistors: M₁,M₂ (measurement), R₁,R₂ (reference)
- Measurement resistors see plasma, reference resistors shielded
- Bridge voltage Vᵦᵣᵢdgₑ ∝ (R-M) → differential measurement
- Isolates small temperature change ΔT from background heating
- Phase-locked loop: V₀ oscillating at ~50 kHz for noise reduction
- Absorber Design:
- ~100 μm thick, 1 mm gold absorber (broad spectral response)
- Resistor zigzag pattern on substrate back
- Heat capacity C and thermal diffusion τ determine time response
- Limitations:
- Measures all energy (radiation + neutrons + particles)
- Neutron damage requires thicker substrates → longer τ → worse time resolution
- ITER bolometers will be significantly worse than current devices
- Standard design from ASDEX Upgrade (1980s) used worldwide
