The other interesting thing about "iron fire" (fast oxidation of iron) is that it creates a "vacuum" of sorts that "sucks" oxygen to itself. Ordinary carbonaceous "fire" creates carbon monoxide (CO) or carbon dioxide (CO2), which are gases that can take the place of consumed oxygen (02) gas. Carbon monoxide production releases two molecules of CO gas per one O2 molecule consumed. Thus, such a carbon fire requires a "convection" current to remove the hot carbon mon/dioxide (out the top) to make room for more cold oxygen to be brought in (at the bottom).
By contrast, an "iron fire" converts the oxygen gas (and possibly also nitrogen gas, but that is tangent) into a solid (rust). Thus, the burning iron metal effectively sucks atmospheric oxygen INTO the pile of burning metal, regardless of convection currents. Convection currents are a strong mechanism for REMOVING heat from a fire. Of course convection currents will also be present even in a huge iron pile furnace, but a result of direct conversion of oxygen gas into a solid (rust) is that there are weaker convection currents and that means that the heat of combustion escapes more slowly from the metal fire furnace than from a carbonaceous fire furnace. Thus, since the heat of combustion does not leave with the combustion products, a metal-air furnace could become much "hotter" faster than a carbon-air furnace of the same scale (e.g., at the same oxygen demand level).
Theoretically, there is no limit upon the temperature that such a large air-metal-fire could attain. It could, in theory, attain a temperature high enough to not only melt iron, but also to boil (vaporize) iron, but not at the same location at the same time. (You cannot maintain solid, liquid, and gaseous iron at the same location, because "melting" and "vaporization" occur at greatly different temperatures). The difficulty with that however is that the molten (burning) iron would tend to settle into a pool, having a smaller surface area (on its top surface only), thus reducing its rate of oxidation.
