Setup
  (layers from the top down: the bottom one, the substrate, is the first laid down)
  the doped thermo-optic phase shifter, at its defaults:
    layers:
      cladding       half-space   SiO2, n 1.444024
      Si             220 nm       Si, n 3.475700 in its shapes; SiO2, n 1.444024 elsewhere
      BOX            2 µm         SiO2, n 1.444024
      substrate      half-space   Si, n 3.475700
    shapes:
      on Si: rectangle centred at (160, 0) µm, 320 µm along x by 500 nm along y
    doped thermo-optic phase shifter: width 0.5 µm, thickness 0.22 µm, slab 0.09 µm, heater_width 1 µm, buffer 0.8 µm, cladding 3.32 µm, length 320 µm, pitch 5 µm, power 0 mW, loss 0 dB/cm, wavelength 1.55 µm, span 0.1 µm
  the thermo-optic phase shifter, at its defaults:
    layers:
      cladding       half-space   SiO2, n 1.444024
      cover          1 µm         SiO2, n 1.444024
      TiN            100 nm       SiO2, n 1.444024
      gap            2 µm         SiO2, n 1.444024
      Si             220 nm       Si, n 3.475700 in its shapes; SiO2, n 1.444024 elsewhere
      BOX            2 µm         SiO2, n 1.444024
      substrate      half-space   Si, n 3.475700
    shapes:
      on TiN: rectangle centred at (160, 0) µm, 320 µm along x by 7.5 µm along y
      on Si: rectangle centred at (160, 0) µm, 320 µm along x by 500 nm along y
    thermo-optic phase shifter: width 0.5 µm, thickness 0.22 µm, heater_width 7.5 µm, heater_thickness 0.1 µm, height 2 µm, cover 1 µm, length 320 µm, pitch 5 µm, power 0 mW, loss 0 dB/cm, wavelength 1.55 µm, span 0.1 µm
  both:
    wavelength: 1.55 µm
    grid: the 2D model's own grids, and grids twice as fine (the gate); the analytic model's closed forms
    solver: the static (heat) solver's rise and the full-vector mode's perturbation (2D); closed forms (analytic); the paper's measurements (measured)
    tolerance: P_π within 10 % of the paper's simulation and 15 % of its measurement; the analytic within 15 % of the 2D; the 2D grid within 1 %
  materials:
    SiO2: Sellmeier, n² = A + Σ Bλ²/(λ² − C²), A = 1, B = 0.6961663, 0.4079426, 0.8974794, C = 0.0684043, 0.1162414, 9.896161 µm; valid 0.21 to 3.71 µm
      source: I. H. Malitson, J. Opt. Soc. Am. 55, 1205 (1965), doi:10.1364/JOSA.55.001205, at 293.15 K
      fused silica, Eq. (1), checked against the paper; fit residual 10.5e-6 on average; thermal and deposited oxides differ slightly
      n 1.444024 at 1.55 µm
    Si: a table of 35 wavelengths, 1.2 to 14 µm, between them natural cubic splines; valid 1.2 to 14 µm
      source: H. H. Li, J. Phys. Chem. Ref. Data 9, 561 (1980), doi:10.1063/1.555624, at 293 K
      crystalline silicon; Table 1, 293 K, checked against the paper; uncertainty ±2e-3 (Li)
      n 3.475700 at 1.55 µm

Jacques et al. 2019's heaters through photonoxide's thermo-optic phase shifter devices, at 1550 nm, 320 um long

doped thermo-optic phase shifter:
  analytic  P_pi  27.607 mW   tau   6.06 us   f_3dB  19.99 kHz
            (exact: no grid)
  2D        P_pi  26.078 mW   tau   4.24 us   f_3dB  23.55 kHz
            (heat on 192 × 75 cells graded from 20.000 nm to 0.500 um (growth 1.200) over 40 × 18 um; mode on 10.000 nm cells)
  measured  P_pi  22.800 mW   tau   2.20 us   f_3dB  85.00 kHz
            (measured: Section 3.1 and Fig. 8, the MZI's arms 200 um apart, the widest pitch measured (23.3 mW at 50 um; Fig. 7's fits 23.2 +- 0.7, 22.7 +- 1.1 and 22.1 +- 1.4 mW))
  2D P_pi on grids 1 and 1/2, mW        26.089246   on 2 grids, error 1.1e-2 ≤ 2.61e-1 ok
  2D P_pi, mW (simulated)               26.089246   paper 25.5       ± 2.5500000000000003e0 ok
  2D P_pi, mW (measured)                26.089246   paper 22.8       ± 3.42e0    ok
  analytic P_pi, mW                     27.606579   2D 26.078324  ± 3.912e0      ok

thermo-optic phase shifter:
  analytic  P_pi  29.851 mW   tau  11.15 us   f_3dB  15.22 kHz
            (exact: no grid)
  2D        P_pi  33.360 mW   tau  11.34 us   f_3dB  14.86 kHz
            (heat on 174 × 99 cells graded from 20.000 nm to 0.500 um (growth 1.200) over 40 × 18 um; mode on 10.000 nm cells)
  measured  P_pi  29.200 mW   tau   7.20 us   f_3dB  33.00 kHz
            (measured: Section 3.1 and Fig. 8, the MZI's arms 200 um apart, the widest pitch measured (31.5 mW at 50 um))
  2D P_pi on grids 1 and 1/2, mW        33.358791   on 2 grids, error 1.7e-3 ≤ 3.34e-1 ok
  2D P_pi, mW (simulated)               33.358791   paper 32.2       ± 3.2200000000000006e0 ok
  2D P_pi, mW (measured)                33.358791   paper 29.2       ± 4.38e0    ok
  analytic P_pi, mW                     29.850921   2D 33.360489  ± 5.004e0      ok
all 8 within tolerance of the paper
