ADDED VALUE PRODUCTS OF ACID LEACH ALUMINUM PROCESS
                                                         by David A. Dietzler  2007
Highland regolith will have all iron bearing minerals including ilmenite and chromite extracted magnetically leaving mostly calcic plagiolcase, some sodic plagioclase, a bit of potassium feldspar and some calcium and magnesium bearing pyroxenes and olivine along with some trace minerals.  Although ilmenite and chromite are more abundant in the mare, there has been some mixing of the two major kinds of regolith due to meteoric splashing. Subsequently, the mare is about 35% plagioclase and the highlands 60% or more plagioclase on the average.  If a rock or soil is 90% or more plagioclase it is called anorthositic or just anorthosite.  The iron bearing minerals, ilmenite and chromite are removed to simplify the next process described herein and to be processed by other methods to get iron, titanium and ferrochrome.

This "de-ironed" regolith will then be leached in sulfuric acid made from scavenged hydrogen and sulfur implanted by the solar wind and oxygen from molten silicate electrolysis. A solution of aluminum sulfate, magnesium sulfate, sodium sulfate and potassium sulfate will be obtained. This will be boiled down to obtain dry salts, the water condensed and used to reconstitute acid stocks. Unfortunately, the water will not also contain a little HCL, HF and H3PO4 formed by reaction of acid with apatite.  From NASA's Advanced Automation for Space Missions, Appendix 5E we find that this net reaction occurs:

                          3Ca3(PO4)2*Ca(F,Cl)2 + H2SO4 ==> H3PO4 + HF + HCl + CaSO4

These acids will react with the minerals and be lost. Only traces exist so the purity of products will not be harmed.  If there was a way to extract apatite from the bulk regolith we could leach it with H2SO4 and obtain these acids. These acids could be separated by distillation and would have uses as is and as sources of chlorine, flourine and phosphorus. The phosphoric acid would also be a soil pH conditioner and when reacted with a cation it would serve as phosphate fertilizer. Phosphorus is also used as an n-type solar cell dopant. More research into apatite extraction is called for.  Perhaps the flourine and chlorine will be released when regolith is roasted during volatile harvesting. 

Insoluble silica and barely soluble calcium sulfate will also form in the leaching tanks and can be kept separate from the solution of highly soluble sulfate salts. These two substances can be separated electostatically at negative 13,416 volts and positive 7800 volts respectively. A trace of barium can also be extracted. Insoluble barium sulfate is repelled at a lower voltage than CaSO4 so it will be separated from the silica with the CaSO4.  These can be roasted to oxides and barium oxide can be leached out of CaO with pure ethanol. Barium monoxide is used as a detergent in lubricating oils.  Barium chloride can be used in steel heat treating molten salt baths.

The aluminum, magnesium, sodium and potassium salts can also be separated electrostatically.  I cannot find details about the voltages and methods of separating these salts, but I am confident it can be done. Since they are water soluble I doubt that froth floatation, electrostatic separation's biggest competitor, is possible or desirable on the Moon. On the dry, air and humidity free Moon without even air currents to disturb the process of electrostatic separation this method seems best. From "Electrostatic Separation of Mixed Granular Solids" by Oliver C. Ralston (Elsevier Publishing Co. 1961) we read that "electrostatic separation [is] the most versatile method for treating mixtures of dry solids while floatation is the most versatile method of treating mixtures of wet solids." The salts from the sulfuric acid leach will definitely be dried as well a ground up and screened to the right mesh size.

Also from Ralston's book we find that there are many kinds of electrostatic separation processes and that, "what cannot be effected by one process, may be easy work for another type." Ralston lists these processes:

1) charging by conduction
2) charging by induction
3) charging by dielectric hysteresis
4) contact potential
5) charging by spray, corona or electric leakage
6) charging by thermionic emission
7) photoelectric charging
8) pyroelectric polarization, piezoelectric polarization
9) the dielectric medium sep. process

So there is bound to be a way to separate these and many other solids on the Moon with electrostatic separation. The regolith is about as dry as anything can be and when screened to the right mesh size it may be possible to concentrate trace element bearing mineral particles with electrostatic separation. Who knows? We might get copper and zinc out of the regolith at last. More research is called for.

Getting back to the acid leach process, some sulfates will be used as is, some roasted to oxides and some of the oxides reduced to pure metallic Al and Mg. Sodium can also be obtained. Oxides of sulfur from roasting will be reacted to SO3 over an upported vanadium pentoxide catalyst and the SO3 will spontaneously react with water to reform sulfuric acid. The leaching vats and plumbing will be made of corrosion resistant high silicon alloy iron from molten silicate electrolysis.

Calcium hydroxide, CaOH, can be made simply by mixing lime, CaO, with water.  Magnesium oxide MgO, reacts slowly with water to make magnesium hydroxide, MgOH.  A suspension of MgOH is milk of magnesia.  MgOH can be reacted with HCl to make magnesium chloride, MgCl.  Also, MgO can be carbochlorinated to make MgCl. Sodium oxide, Na2O can be reacted with HCl or carbochlorinated to get salt, NaCl.  Sodium hydroxide can be obtained by electrolyzing a solution of NaCl in a Castner-Kellner cell that uses mercury as the cathode.  Since the Moon lacks mercury this would have to be upported. Fortunately the mercury is not used up in the process.  Potassium hydroxide, KOH, requires a little more chemical magic.  Most of the potassium sulfate from the acid leach will be used as is for fertilizer.  Aluminum hydroxide requires more chemical magic but there's no great need for it as there is CaOH or NaOH. Carbon dioxide generated by carbochlorination, that is mixing the oxide with carbon dust in a stream of hot chlorine gas, will be reacted to methane, CH4, in Sabatier reactors (basically just pipes containing a nickel or ruthenium catalyst) and the CH4 pyrolized to pure carbon for reuse.  We cannot waste carbon on the Moon  

Here is a list of useful products from all these substances obtained by this process:

SiO2: pure silica, heat resistant glass for lab ware and cook ware, windows exposed to temperature extremes, making sodium silicate (water glass) and optical glass fibers, optical insturments and radio frequency control crystals
CaSO4: plaster, wallboard, molds for casting Al and Mg, medical uses (IE casts), cement modifier (5% in Portland cement to retard setting time), polishing powder, paints, reduction of zinc minerals, drying industrial gases, food additive, soil conditioner, dessicant
CaO: lime, refractory, steel making flux, cement, mortar, white washes, soda-lime glass, soil pH conditioner, making milk paint, poultry feed, removal of phosphate from sewage and pH control, neutralization of acid wastes, producing sodium carbonate by Solvay process, CO2 absorbent, food additive                             
CaOH: slaked lime; mortar, cement, white wash, caustic, making calcium salts, body hair removal (active ingredient in Nair and Magic), dehairing animal hides, soil conditioner, disinfectant, water softening, food additive, making low grade rubber and petrochemicals (not many petrochemicals on the Moon but we may get hydrocarbons from asteroids someday.) 
Aluminum Oxide: alundum, excellent abrasive for grinding and polishing (emery), hi temp. ceramic (mp 2000 C.), synthetic gems for jewelry, insturment bearings and wire drawing dies; dehydrating agent and catalyst 
Aluminum metal: structural uses, mirrors and solar furnace reflectors, rocket fuel, alloying magnesium, electrical wires and cables, coils for generators, electric motors and mass drivers,
Aluminum Sulfate: paper making (sizing, pH control), dyeing mordant, sewage treatment, foaming agent in fire foams, cloth fireproofing, tanning leather, catalyst in ethane manufacture, water proofing concrete, clarifier for fats and oils, food additive
Aluminum Chloride: strong antiperspirant, making pharmaceuticals and organics (Friedel-Crafts catalyst), making butyl rubber and hydrocarbon resins
MgSO4: epsom salts, medical uses, laxative
MgO: magnesia, firebrick hi temp ceramic
Magnesium: obtained by reducing MgO with silicon, explosives and structural uses, alloying Al
MgOH: magnesium hydroxide, stomach antacid, soil pH modifier
Na2SO4: will be roasted to oxide of sodium
Na2O: soda-lime glass, making sodium silicate
NaOH: lye, making sodium silicate, soap, laboratory reagent, 
NaCl, table salt, made by reacting lye or sodium oxide with HCl, steel heat treating baths, cooking, nutrient, when added to lye soap it separates into cake soap and glycerin. Glycerin is used in lotions and skin moisturizers and might serve as a body lubricant for slipping in and out of elastic counter pressure space suits.
Sodium metal can be obtained by electrolysis of NaCl in a Downs cell. Sodium is used in sodium vapor lamps.                                                                            
Potassium sulfate: Fertilizer. Plant fluids contain large amounts of potassium and need this element to grow.                                                                                        
Potassium hydroxide: KOH; water electrolysis cell electrolyte

Last but not least, molten sodium, salt and magnesium chloride can each be used for thermal energy storage and heat transfer in titanium vessels and pipes since titanium resists corrosion by chlorides.  Also, a sodium-potassium eutectic can be used for nuclear reactor coolant and may find uses in solar thermal systems too.
Since there is plenty of iron and some carbon on the Moon we could make Downs cells on the Moon.  It might also be possible to boil sodium out of molten silicates during electrolysis.  Chlorine  will be recycled to react with sodium oxide to get NaCl. Table salt consumed by humans will be excreted and recovered from life support recycling system.  Possibly, the sewage after conversion to a sterile solution of nitrates and salts by the SWCO (Super Critical Water Oxidizer) will be processed chemically, or the solution will be passed through a plastic filter to remove NaCl or there will even be plants that accumulate salt, so salt demands and therefore chlorine demands will not be high on the Moon. If we keep mining sodium and chlorine and making NaCl instead of recycling it from the LSS we could encounter salt pollution in our biosphere system.
See: Life Support
                                                     Reasoning
Iron bearing minerals are removed magnetically to "clean up" the feedstock so there are no sulfates of iron to separate too.  This simplifies the process.  One might ask, why not just take Ca or Al from fluxed electrolysis and react it with H2SO4 to get CaSO4 or Al2(SO4)3?  This would require a sealed vat and plumbing to pipe off the hydrogen:  Ca + H2SO4 ==> CaSO4 + H2 gas.  The hydrogen would then have to be reacted with oxygen to get water that would be reacted with SO3 to reconstitute acid stocks so this doesn't simplify things much.  One might also ask, why not take Si from fluxed electrolysis and react it with oxygen to get silica?  If this was done in a fluidized bed a layer of SiO2 would form around an inner core of silicon that prevents further oxidation and we would not get pure silica for glass.  If we tried to burn ultrafine particles of silicon in some kind of device that resembles a rocket engine and condense the silicon dioxide (silica) we'd have a really "hairy" gizmo at work that had a limited lifetime and the condensed silica would all have to be chipped up and melted and cast.  That would be a lot of work.  Acid leaching "de-ironed" highland regolith for silica, CaSO4 and other products is simple and direct.