Rhenium tags along with molybdenite (MoS2) ore as ReS2. During roasting it oxidizes to Re2O7, which is unusually volatile for a metal oxide (sublimes near 360 °C, boils near 363 °C at 1 atm and is fully vapor by roasting temperatures) — so instead of staying in the MoO3 calcine, most of it rides the off-gas stream. This lab tracks that distribution with two coupled models:
v(T,O2) = 100 · Tfrac(T) · O2eff(O2) · sinter(T)
Tfrac = clamp((T−480)/170, 0, 1) ← saturates ~650°C
O2eff = under-oxidized below ~16% O2,
mild dilution penalty above it
sinter = calcine sintering above 680°C
traps volatilized Re before it escapes
captured = v · (1 − e^(−k·L/G)), k = 1.15
← classic gas-absorption NTU relation: capture
rises with liquid-to-gas contact, saturating fast
The remaining (100−v)% never left the calcine at all — it is not "lost", it heads to a separate acid-leach circuit downstream. Only the volatilized fraction that the scrubber fails to wet becomes an actual stack emission. SO2→SO3 conversion is shown alongside as a genuine side effect of excess O2 and temperature; it does not feed back into the Re balance here, but in practice the resulting acid mist is exactly why scrubber design and roast control are tuned together.
- Furnace → duct — particles spawn at the roaster; the fraction that turns orange (volatilized Re2O7) versus grey (stays in calcine, removed) is set live by v(T,O2).
- Scrubber tower — volatilized particles either fall as blue droplets (captured) or pass straight through to the stack (lost), split live by the L/G absorption curve.
- Sensitivity curve — volatilization vs. temperature at the current O2 setting, with a marker at the live operating point.