LIGHTCELL / REDUCED RESEARCH MODEL
Can trapped light hold sodium out of thermal equilibrium?
A hydrogen–oxygen flame burns inside a sapphire cell seeded with sodium salt. Excited sodium atoms emit the yellow D lines at 589 nm — but the vapor is so opaque at that exact wavelength that each photon is reabsorbed thousands of times before escaping. Every reabsorption re-excites an atom, so the excited population can sit far above what temperature alone would allow. That departure from equilibrium is what this page solves — live, on your GPU — and it is the physical basis of a combustion-driven sodium lamp feeding a matched photovoltaic cell.
Watch the cell. The bright full-chamber flash at t=0 is the seeded starting condition. The solver then relaxes to the true solution: a small flame anchored at the burner lip. Dimmer is not dead — D-line glow falls steeply with temperature.
Read the departure coefficient. The big yellow number is the solved 3p population divided by the detached Boltzmann prediction. Above 1 means radiation trapping and chemistry are beating thermal equilibrium.
Audit the books. Every watt is tracked from metered fuel to light, heat, exhaust and storage. The rate audit and energy ledger below expose the full balance — including the closure residual.
Display-only camera response: solved sodium emission and trapped D-line radiation viewed through the complete cell with sensor exposure and saturation.
b₃ₚ > 1 means the solved excited population exceeds the thermal-equilibrium prediction at the same local temperature — the non-LTE claim, quantified.
Boltzmann is not fed back into the simulation.
The dip at each line center is self-reversal: light from the hot core is reabsorbed by the cooler sodium shell it must cross to escape. A model without real radiation trapping cannot produce it.
inner-wall heat flux and hotspot solving
Nozzle and reaction regime pending…
Cantera and public-observable checks pending…
Measured Na + OH + N₂ and NaOH + H rates evaluated against the unperturbed Cantera radical pool. This is a burden screen, not a coupled inhibited-flame prediction.
141 frames from 10–150 s: the white-clipped column spans every analyzed axial row, so the video supports persistent axial confinement and a steady centerline. Saturated RGB cannot establish radiance, temperature, D-line power, or sodium concentration. The post identifies hydrogen confinement but does not identify the salt.
Choose the stabilized branch for a persistent burner-lip flame, or the ignition transient to test whether the one-time seed survives. The 15 Jul protocol holds metered H₂ fixed while an area-averaged chamber source ramps from bypassed to captured; its accelerated timing and geometry are not reconstructed from the video. Set physical inputs, then press Start / rerun from t=0 for a comparable run.
Waiting for GPU state…
Metered fuel power in; light, heat, exhaust, and stored energy out. The residual is what the integration fails to account for — it should settle near zero, and it is displayed rather than hidden.
A 40% wire-to-wire target with an 80% electrolyzer requires a 50% H₂ → electric generator. Required fuel → PV-absorbed D-line power:
The public ~100 W optical, 10 W integrated electrical, and cell-efficiency peaks came from different operating points. This panel never multiplies them together.
Physical-time axisymmetric transport of coaxial H₂ and oxidizer jets, differential species diffusion, finite-rate heat release bounded by a Cantera thermochemical ceiling, inner/outer sapphire conduction, conserved sodium, D1/D2 statistical equilibrium, and six photon groups.
Populations come from chemical pumping, radiation, species-resolved collisions and D1↔D2 mixing. Detailed balance lets LTE emerge only in the thermal limit. The Boltzmann population remains a detached diagnostic.
NaCl/NaOH neutralization, the fixed 1% chemical excitation yield, local mixture, species-dependent pressure broadening, frequency redistribution and measured optical boundary spectra remain uncertainty-bearing inputs—not operator knobs. H₂/O₂/N₂/H₂O quenching now uses measured 1500–2500 K flame cross sections.
Prescribed axisymmetric coaxial-jet/turn/return velocity field, a face-aligned burner inlet, one-step H₂/O₂ heat release, mixture-averaged scalar diffusion, three groups per D line and P1 transport with a partial-current optical boundary. Stabilized mode replaces unresolved burner-lip radicals and boundary-layer recirculation with a localized flame-holder activity; transient mode removes it. Cantera HP equilibrium, premixed flame speed, resolved Voigt spectra, and sapphire stress are comparison diagnostics—not replacement closures.