1. This question is in regard to my Minecraft mod. To those who don't know, Minecraft is a first person survival game by Mojang, where the players' objective is to become self sufficient and survive in a hostile environment, having nothing but nature to rely upon for all of their resources. A mod is a modification made to the game (encouraged by Mojang) that adds something to the players' experience. Mods cannot be sold. A mod developer distributes his/her mod for free. My Minecraft mod focus on adding realistic chemistry to the game.
  2. My source for this question was http://geology.com/minerals/
  3. I'm aware these rock types are far from pure in nature, but I'm willing to disregard that, for the sake of simplicity, since mineralogy is already a very complex field, and realistic metal extraction from an ore goes far beyond the scope of the mod I have in mind.
  4. I will be doing similar questions about andesite and diorite.

For this question I'm focusing on a way to decompose granite into useful substances. In reality, thinking in industrial scale, this probably is never done, because of the costs I assume are involved and the return.

Most of these rocks contain Quartz ($\ce{SiO_2}$ crystalline) in abundance. Quartz, for whatever reason, happens to be a mineral found only in the nether in Minecraft, a hellish dimension the player can travel to after building a proper portal (can't be done early in the game). This means these rocks (in fact all rocks, but lets focus on granite, because it lacks any function in the game) can be a source of Quartz.

Granite is also rich in Orthoclase feldspar, which has the chemical formula $\ce{KAlSi_3O_8}$. Looks like a very poor source of aluminum and potassium, but a source nonetheless, and in a Minecraft game, the player doesn't need to worry about industrial scale production, since whatever he/she does, is only for him/her and friends. I don't know the structure of this species. It's probably an oxide, but unlike $\ce{SiO_2}$ it probably has some weird bonds going on.

Another major component of granite is Plagioclase feldspar, which is made of these two chemical compounds in equal concentrations: $\ce{NaAlSi_3O_8}$ and $\ce{CaAl_2Si_2O_8}$. Poor sources of aluminum, sodium and calcium.

The two other components of granite are amphiboles and micas. Amphiboles comprise a large variety of minerals, all of them contain a double hydroxyl group in their formulas. Similarly to orthoclase and plagioclase, they all contain silicon and oxygen. I doubt these are silicon oxides, or Arrhenius bases, they probably belong to some college level chemical group that we don't see in highschool. Here's a link to the chemical formulas of all Amphiboles: https://en.wikipedia.org/wiki/Amphibole#Chemical_formula They look like extremely poor sources of metals like aluminum, lithium, iron, sodium, calcium, magnesium and potassium.

Now micas. Micas are described by the formula


in which

X is K, Na, or Ca or less commonly Ba, Rb, or Cs;
Y is Al, Mg, or Fe or less commonly Mn, Cr, Ti, Li, etc.;
Z is chiefly Si or Al, but also may include Fe3+ or Ti.

Of interest in this formula are Rb and Cs, no matter how rare, because these are rare by definition. So, adding to the same elements as the above described amphiboles, micas have the advantage of also containing barium, rubidium (!), caesium (!!), manganese, chromium, Titanium and Lithium.

So my question is, what would be a lightly realistic approach to obtain usable compounds from granite? I'm sure real techniques do not exist, but what about science-fictional approaches? Chemists probably have unachievable dreams of tapping this or that potential source for obtaining huge amounts of a substance (until recently, one of those fancy dreams was a way to obtain drinkable water from sea water). I would love to hear any input from this (almost always) lovely community.

EDIT: I imagined a weird method: First, grind the stones to a thin powder, and then put them inside a "digestor" machine, that will use fictional enzymes and heat to break the compounds into quartz and oxides of each separate element, which are then reduced by electrolysis yielding the pure metals. Is it good enough?

  • $\begingroup$ Related: chemistry.stackexchange.com/questions/50668/… $\endgroup$
    – aventurin
    Aug 1, 2016 at 8:31
  • $\begingroup$ yep, kind of related, yes. Specially in the points where my question regards the extraction of aluminum from feldspar. Still, hopefully, I will get a full answer for this question, since it doesn't neet to be bound by regular chemistry rules, but it's open to Sci-Fi theories. $\endgroup$ Aug 1, 2016 at 9:17

2 Answers 2


First, some specific points:

For this question I'm focusing on a way to decompose granite into useful substances. In reality, thinking in industrial scale, this probably is never done, because of the costs I assume are involved and the return.

It's not only the cost. It's about having huge amounts of waste. You were mentioning Rb and Cs in your question. An average granite will have ppm (parts per million) amounts of Rb and Cs in the rock. Let's say you want to produce 1 kg of Rb, from a rock that has 50 ppm of Rb. The same rock will have, let's say 65% of SiO2. You will end up with 13,000 kg of SiO2 for each 1 kg of Rb!

Granite is also rich in Orthoclase feldspar...I don't know the structure of this species. It's probably an oxide, but unlike SiO2 it probably has some weird bonds going on.

It is a silicate. The metals (cations) in there are ionically bonded to silicate anionic groups (which themselves are ~covalent) such as [SiO4]4−, [Si2O7]6−, etc.

So my question is, what would be a lightly realistic approach to obtain usable compounds from granite?

No. See below.

I'm sure real techniques do not exist, but what about science-fictional approaches?

Real techniques do exist, it's just pointless do do them. You ask about science fiction, but in the beginning of your question you mentioned that your mod "focus on adding realistic chemistry to the game".

Second, a detailed answer:

It is not feasible to produce those elements just from a granite (or a diorite or andesite or any other "conventional" rock). Producing a commodity has several parts to it which determines whether we should do it or not. Since you do not actually care about costs, it may seem ok. But there are also two other important things: effort and waste (as mentioned above). Producing 1 kg of Rb from a rock that has 50 ppm (if you're lucky) requires mining 20000 kg. This is something that's extremely hard and time consuming unless you have the machines for that. Also, think about what you're going to do with the waste rock (13000 kg of SiO2 for example).

Even if somehow you managed to do it, and you have some magic processes that can extract the Rb with little effort out of the rock, you will probably want to do it for every other element in the periodic table. Then it becomes a huge effort. Think about the environmental impacts of digging out a mountain and then building pools of enzymes to digest the thing, and let's not forget the noise and pollution (and energy requirements) made by the machines that grind the rock to powder. Granite is a hard rock, you know?

This is why we extract mineral commodities from ore deposits. Luckily, mother nature provided us with certain types of rocks that are enriched in an element or a group of elements. Then, if we want to produce 1 kg of Rb, we don't have to mine 20000 kg of rock, we only need to mine 200 kg of rock or less.

Another question, why do you need Rb for, or any other metal? Do people mine stuff just for fun or do they actually need this for something, in-game? I would suggest that you come up with a list of elements that may be useful for someone in the game, and then think of an ore deposit that may hold this. For example, granites occasionally have certain "segregations" called pegmatites. These rocks are heavily enriched in certain elements, such as Li, Cs, the lanthanides, Be, Sn, etc. You can make up spots of pegmatite in your granite world so people can mine that. As a bonus, pegmatites usually host pretty crystals of stuff.

Once you come up with a list of elements, you might want to ask the next questions in the Earth Science SE, not here.

  • $\begingroup$ Thank you for the answer. I'm not sure you understand what I'm trying to acomplish. Granite, Diorite and Andesite are already part of the game, and they serve no purpose. Since I'm making a chemistry mod, I thought I would give those rocks some use. In Minecraft, the most common block is called "stone" the second most common is called dirt. Those are generic names and they are meant to be kind of useless. They are what separate you from the ores. But granite, diorite and andesite were introduced in the game with no purpose. I want to give them a purpose. $\endgroup$ Aug 24, 2016 at 7:06
  • $\begingroup$ What if the enzymes would use most of the $SiO_2$ as catalists or as part of their own molecules? If the Silica was part of the cycle somehow, being partially consumed so the digestor would be self suficient in terms of matter, and it would only require energy to work. When I mentioned I wanted realism, well as much as it is possible in science fiction... I thought this is what this tag was about. $\endgroup$ Aug 24, 2016 at 7:57
  • $\begingroup$ As for the need for the elements obtained from those rocks, I will introduce the need with my modification. $\endgroup$ Aug 24, 2016 at 8:02
  • $\begingroup$ @FinnTheHuman I'm not a biology expert, but it seems very unlikely that enzymes would be able to exploit SiO2. Usually, when you have bacterial stuff in rocks, it's redox. Oxidising or reducing some elements, but even then I find it very hard to think of a process where you can extract (trace) elements from a rock using non-industrial methods. Maybe it would be better to exploit the minerals without separating them to the constituent elements? For example, quartz (aka sand) can be used in glass-making and construction, feldspars (orthoclase and plagioclase) in ceramics, micas in cosmetics, etc $\endgroup$
    – Gimelist
    Aug 24, 2016 at 8:23
  • 1
    $\begingroup$ Ok then. Thank you very much. Since there's already magic in the game, I will do the digestor anyway even if it's less realistic than what I wanted. It's a shame. Though. I may revisit this at a later patch. Thanks again for your help. $\endgroup$ Aug 26, 2016 at 9:57

One option might be something like FFC Cambridge process - it can reduce most (possibly all) oxides directly, at the cost of potentially high amounts of energy. It's been tested with simulated lunar regolith and while that's not exactly the same as your granite, they're definitly related.

I don't know if the presence of the other elements would be an issue - I've actually asked a question myself today about the impact of impurities in the reaction. Will link if there's a useful answer.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.