| Comparing DRI to Crucible Steel Production |
| Dave Dietzler, 2007 DRI Direct reduction of iron bearing minerals with CO gas involves many complications. Irony mnerals like ferrosilite have to be extracted from mare regolith and sintered into briquettes. Large qtys. of CO gas are needed for the reduction and the CO2 that forms must be cleaned, cooled and reacted with hydrogen to form CO and water that must be electrolyzed to recover hydrogen. Large furnaces heated by electricity made of stone, cast basalt and ceramics from magma electrolysis are needed; a maze of alloy pipes that can resist heat and corrosion are also needed. Heat exchangers, gas cleaners, RWGS reactors must all be built. Compressors, pumps, valves and gas storage tanks must be constructed on the Moon. Iron from the DRI furnace must be sent to a steel making furnace, probably an electrical open hearth furnace. Large amounts of flux are also needed and this is a challenge. There is no limestone or any other sedimentary rock on the Moon. Thermal decomposition of anorthite to CaAl2O4 and even thermal decomposition of CaAl2O4 to CaO and Al2O3 might be the only ways to make flux. Even then huge solar furnaces will be needed to produce the flux, and this adds to the complexity of the whole process. Irony minerals can also be heated in the vacuum to about 1200 C. and FeO will boil off. This could be condensed, formed into briquetts and reduced without lots of flux because it is free from silica; however, this adds to the complexity of the whole process. The complexity of all this is so great that the only justification is the production of huge qtys. of iron and steel. Decades of lunar development would be needed before such things could exist. Small systems might be built after a number of years, but do we need small systems for producing limited amounts of iron and steel except for experimental purposes before scaling things up? Crucible or "Blister" This system can use iron obtained from iron fines and/or from magma electrolysis that is rolled into thin sheets and reacted with carbon slowly in electrically heated stone, basalt and/or ceramic boxes. The steel can then be cleaned with small amounts of flux. Not much flux is needed because the iron is mostly free from silica to begin with unlike the iron minerals reduced directly with CO gas in the "blast furnace" system. The equipment is simple. No alloy pipes, compressors, pumps, gas storage tanks, heat exchangers, RWGS reactors, briquetting, or production of vast amounts of flux is necessary. The carburizing boxes don't have to endure high pressures as well as high temperatures. Unbeneficiated regolith or better yet, regolith that has had the ilmenite extracted electrostatically for separate processing to TiO2 and titanium and iron bearing minerals that are then extracted magnetically can be dropped into molten silicate electrolysis units. Iron and silicon will result and higher yeilds of iron will be had with irony minerals rather than plain regolith. This does not have to be briquetted, but it has to be cast into slabs as it leaves the magma unit then rolled into sheets or thin plates once it has cooled enough but is still somewhat soft. Even so, this involves less complication than briquetting and certainly less energy too. Can this system produce as much steel as DRI systems, should they ever be built? It takes 7 to 10 days to carburize the iron and convert it to steel. Seems like big direct reduction furnaces could produce more iron and steel. Perhpas large carburizing boxes or larger numbers of them could compete. A 5m x 5m x 1m carburizing box could produce roughly 200 tons of steel per 7-10 day run. If the run can be done in 7 days then 400 tons of steel could be produced every lunar dayspan. Perhaps ten of these could produce 2000 to 4000 tons of steel every dayspan or 24,000 to 48,000 tons of steel per year. That's quite a bit. If a "blast furnace" produced 300 tons every 24 hours that would be 4200 tons/dayspan and 50,400 tons/yr. . Are we making a false assumption about the superior productivity of a lunar DRI furnace system versus a crucible steel system just because the crucible steel system is slower???? How much steel? Finally, how much steel do we need? To build 1000 SPS rated at 10GWe apiece we would need over 100 million tons of steel. Some may object to the use of tubular steel for the SPS frame structure, but plain carbon mild steel has a higher strength to weight ratio than unalloyed aluminum. Aluminum must be alloyed with lithium or copper to give it high strength, but these are present on the Moon in mere parts per million and they can't simply be roasted out of regolith like volatiles can. Launching millions of tons of copper to alloy 100 million+ tons of aluminum for SPS is out of the question. Steel is the only reasonable metal for SPS construction. Glass-glass composites are certainly the runner up, if not the outstanding material for SPS construction, assuming they are flexbile enough not to crack when the SPS is rotated and oriented towards the Sun or shipped from L5 to GEO. Are SPSs the way to go? A thousand SPSs would mean 1000 rectennas about 10 to 15 km in diameter. Is this plausible? Perhaps fewer SPSs are more realistic for supplying 24/7 baseload power to industrial districts. Residences would rely on battery packs in the cellar by night in a world powered more and more by roof top solar panels and large rural solar farms connected to cities by superconducting cables. One hundred SPS would still require 10 million+ tons of steel. With 100 5mx5mx1m carburizing boxes we could make 2.4 to 4.8 million tons of steel per year! The job would take 2 to 4 years, and to make 100 million tons 20 to 40 years. Without all the complexities of "blast furnaces." What if SPS doesn't pan out at all? What if helium 3 fusion is the lunar money maker? Thousands of mining machines would be necessary but thier total mass might be 20,000 tons! Do we need to produce millions of tons of steel, or just tens of thousands of tons? Either way, it looks like we could do the job by adapting the old blister steel or crucible steel or cementation process to work on the Moon. We must not forget that to produce millions of tons of steel we would have to mine enormous areas for iron fines (possibly at the same time we mine helium 3?) or build large numbers of magma electrolysis units to churn out iron and more silicon and oxygen than we know what to do with. So there is a limitation there. This matter demands deeper investigation. |
 |