Elesorter (and Isosorter)

Synopsis:
This "machine" will have one input and 80 or so outputs. Throw anything except radioactive waste into the input, along with lots of power, and collect pure chemical elements from the outputs. Then you use those elements to make stuff. Plain ordinary rock is roughly 50% oxygen, 25% silicon, and 25% everything else: aluminum, iron, magnesium, titanium, etcetera. Think about how an ordinary "good grade of ore" might be 5% of what you want, while ordinary rock is at least 25% of what we want! Because we use EVERY stable element these days, for something or other. Sure, we don't need quite so much silicon or oxygen, but those can be combined back together (conveniently, two oxygens per silicon) to make quartz blocks, and we can use those...and remember, if we can process ordinary rock, then we can also recycle every trash heap on the planet.

For radwaste you need a fancier machine, an "Isosorter", because you want to sort out the stable from the radioactive isotopes. An Isosorter will be a lot fancier than an Elesorter, and maybe have 600 outputs, because, while there are around 1700 total known isotopes of the elements, most of them are both radioactive and too unstable to persist long enough to bother sorting them out. That is, just take your fresh radwaste and wait a month before sorting it, and only the longer-lived radionuclides will be left. One output will be needed for each, as well as the 272 outputs needed for the stable isotopes.

Details:
I first thought up this idea in the early 1970s. When I went to college I began taking courses with a goal of attempting to make the elesorter a reality, although I didn't brag about it at that time. Various things happened, and I ended up doing other stuff. Ah, well. Perhaps here I can share the dream, although I must admit that sometime in the 1980s I encountered a science-fictional description of something much like an elesorter. So, in one sense the idea is already "out there" -- but not as detailed as I will try to present here....

I'll start with the Isosorter first, because this is just a superlatively souped-up "mass spectrometer", which we already know how to make. First, in case you don't happen to know what an isotope is, then here is an explanation that those who do know can skip: Every atom is the smallest constituent particle of a pure chemical element, such as oxygen or gold. Any atom, as a whole, can be thought of as having two "obvious" components: a dense core called the nucleus (1/100,000 the size of the atom), and an orbiting "cloud" of electrons, the stuff of which electricity is made. Most of the physical and chemical properties of an element depend on the number of electrons that each of its atoms possess; for any one element, all its atoms have exactly the same number of electrons. Appropriate electrical equipment allows us to strip the electrons away, which lets us study atomic nuclei independently. They proved to be tough nuts to crack, but eventually we learned that almost all elements have nuclei that are made up of two obvious constituents, known as protons and neutrons. The protons have electrical properties that attract electrons on a one-for-one basis. Thus, the number of protons in an atomic nucleus (or "nuclide") determines the number of electrons that can make up its "cloud" -- that is, all nuclei of any one element have exactly the same number of protons, which attract that same number of electrons, which in turn are responsible for most physical and chemical properties of that element. Note that protons (and neutrons) are more than 1830 times as massive as electrons, which is why the phrase "a dense core called the nucleus" is relevant. Because atoms can often rather easily acquire or lose electrons, while it is much more difficult for them to acquire or lose protons, and because electrons orbit protons rather than vice-versa, the protons are considered to be the "controlling" part behind an element's properties, and their quantity is known as the Atomic Number. Meanwhile, neutrons have no significant electrical properties (they are named after the word "neutral"), but they do have the purpose of "gluing" the protons together in an atomic nucleus. Protons electrically repel each other, and will not stay next to each other without help. The details of that "glue" (usually called "The Strong Nuclear Force") do not concern us here, but what does matter is the fact that while every atom of a given element always has the same number of protons, it does not necessarily always have the same number of neutrons. The word "isotopes" is used to distinguish these varieties of a chemical element. Since neutrons have about the same mass as protons, they contribute significantly to the overall weight of an atom (a physical property), and can also affect the rate at which atoms react with each other (heavier isotopes react a little more slowly, and this is a chemical property). Atoms with too many or too few neutrons are almost always radioactive -- the chief exception being the element hydrogen, Atomic Number One (one proton), which makes up 90% of the observed Universe and has no trouble existing without any accompanying neutrons. (Makes sense: one proton by itself doesn't need "glue"! --but if one neutron is added anyway, the atom, called "deuterium", is still stable, while adding a second neutron yields the radioactive isotope called "tritium".)

In a mass spectrometer, matter is heated to become a plasma in a vacuum, the loose electrons are then removed, and the remaining ions are passed as a stream through a magnetic field. Since the path of motion of an ion through a magnetic field depends on its electric-charge-to-mass ratio, and every isotope is unique, an extensive enough magnetic field will allow a big enough "spread" of the ions, as they take different paths, so that each unique ion reaches its own unique destination. At those hundreds of sites, one for each different isotope, we collect the arriving ions, give them their electrons back, and condense the atoms back into piles of pure elemental isotopes. (I use "piles" in a generic sense only, since gaseous and liquid elements will need containers.) But sorting by mass spectromety is a very energy-expensive process, and so only should be used for appropriate stuff like radwaste (much of which is NOT actually radioactive; it is a mixture!). This is why I originally sought (and would rather continue to seek) some other process as the basis for an elesorter -- I might mention that mass spectromety was implied as being the basis of that science-ficitional elesorter I read about in the 1980s (but separation into isotopes wasn't mentioned, just separation into elements).

In the Synopsis I mentioned something about the composition of rocks: here is a more detailed breakdown, of the average composition of the the Earth's crust:
46.60% Oxygen
27.70% Silicon
08.13% Aluminum
05.00% Iron
03.63% Calcium
02.83% Sodium
02.59% Potassium
02.09% Magnesium
00.44% Titanium
00.14% Hydrogen (oceans are part of the Earth's crust)
00.04% Phosphorus
00.03% Sulfur
00.01% Nitrogen
99.23% (Total so far)
00.77% (about 70 other stable elements)

Let me state more clearly (and rather simplisticly) the way one tries to earn a profit from running an elesorter. Look at rocks; think "Oxygen mine! Extremely rich ore!"
So, extract the oxygen, and consider what's left: "Silicon mine! Oxygen extraction has enriched the "tailings" -- compute 27.70% divided by (100% - 46.60%) -- to become 51.87% Silicon ore!
So, extract the silicon, and consider what's left: "Aluminum mine! Prior extraction has enriched the tailings -- compute 8.13% divided by (100% - 46.60% - 27.70%) -- to become 31.63% Aluminum ore!
So, extract the aluminum, and consider what's left....

Every time you extract the most abundant remaining element, the percentage of the next element goes up, making it economically worth extracting, too (as soon as enough tailings are accumulated). In this manner you work your way down to the least abundant stable element (I think it's rhenium, Atomic Number 75, rarer even than the radioactive elements thorium and uranium), which has been enriched to 100% of "what's left". And that makes it profitable, too!

Now, how do we actually go about doing that? Especially since we aren't living on the Moon, where an oxygen mine is a Good Thing To Have? Well, as I implied in the Synopsis, we should work to process the rock to extract quartz (pure silicon dioxide), leaving everything else in the tailings. At this time I think, but am not certain, that by exploiting a natural property of crystal growth which tends to promote purity, and employing a process known as "zone refining", it may be possible to melt ordinary rocks and extract large pure quartz crystals, leaving (mostly oxides and sulfides of) all the other elements behind. Then, relatively ordinary already-existing techniques can be used for orderly extraction of those other elements, as already described.

An alternate approach (and one that might be preferable, when recycling trash heaps), is to take large quantities of fluorine gas, and let it chemically react with everything being processed. (Later, the fluorine is itself extracted for reuse here.) All carbon-rich and silicon-rich matter will yield quantites of carbon tetrafluoride and silicon tetrafluoride GASes -- and oxygen gas is released, too! -- which separate themselves from the tailings most cooperatively. Those gases can then be liquified, fractionated (separated), and the tetrafluorides would be processed to remove the fluorine for reuse. The tailings would be a mix of fluorides, and again we'd process them in sequence of most-abundant-first (a number of new processes, not currently in use anywhere, may be required, though).

In either case, an elesorter is going to be an energy-absorbing operation (but rather less so than an isosorter). Sooner or later we are actually going to need elesorters, because all the rich natural ores will have been used up, and the trash heaps will be the richest remaining places to find such strategically important materials as chromium.... We can also use elesorters to process polluted soil, and while this won't restore the soil, at least we recover those elements that poisoned it, and can perhaps plan to keep them out of the biosphere in the future.

One simple way to do that last thing involves all those quartz blocks I mentioned earlier. Besides building houses that don't have windows (but still need curtains; this stuff is VERY transparent), quartz blocks can be fused together to make containers; each container can be filled with some pure and reviled element like arsenic (there are a number of such elements), and then more quartz blocks can be fused to seal the containers. Then the containers can be stored away, separate from the biosphere, but available if needed --dumping them in the Marinas Trench may sound like a good idea, but as soon as they are subducted in-between tectonic plates, then we will find that those reviled elements are vitally needed. Better to have and not need, than to need and not have.