/sci/ - Science & Math

>>15193023
solar cells on earth are made using toxic chemicals which are hard to obtain on the Moon. Worse yet they require water and there ain't fucking much of that on the Moon. They use a process which doesn't need these chemicals. They take dirt, melt it to lava and pass electricity through it to electrolyze it, turning molten oxides in dirt into metals. They get a mixture of molten iron and silicon along with some other crap from this. Somehow, they're able to refine out the silicon from the molten iron, get aluminum out, and make glass. All of these are no small feat. Oh and did I mention that lava is fucking corrosive? Not only does it melt crap, molten silicates are very good at dissolving stuff. Getting silicon out has been tried before, but it didn't work great. No one got to the point where they could make solar cells. I don't think anyone's extracted aluminum from fucking lava like this either. Yeah, Hall–Héroult process does this, but that requires cryolite. Can't get cryolite on the goddamn moon.
>>15193033
If they really worked out all the steps to make solar cells and the solar cells efficiency isn't utter shit it's a breakthrough.
>>15193043
>>15193041
as I said, it's a first step. If you can't power a von neumann machine, you can't build one. Now, depending on how good the silicon they produced is they might be able to make shitty transistors and associated electronics. Huge, not very fast, but better than vacuum tubes. The process for making solar cells ain't too different from making transistors.

>>11788895
>>electrolysis of metal oxide
did somebody just mention MAGMA ELECTROLYSIS? You can make oxygen, iron, silicon, and other materials by electrolyzing melted dirt(MAGMA). It's really nice because it doesn't require precious volatile elements like hydrogen.

>>10864815
Lunar bungee jumping!
>>10864740
There's molten oxide electrolysis, heat regolith till it's molten, stick electrodes in, apply electricity. It's being investigated here on earth to make steel without coal

>>10645440
read the fucking links asshole.
this details mining robots for mining moondust:
https://en.wikisource.org/wiki/Advanced_Automation_for_Space_Missions/Appendix_5D
This details some of the chemical processing involved for making robots from mined moondust
https://en.wikisource.org/wiki/Advanced_Automation_for_Space_Missions/Appendix_5E
>>octahedron manufacturing robot?
The most difficult part of making a self replicating robot is assembling things. Robots aren't that good at assembling things. You need precise fixtures for every part you have, doing this for a self replicating machine would be difficult. Modular blocks get around this problem. The GIF I posted is just one an example of this.
>> aren't going to make metal parts out of moon dust.
bullshit. Moon dust contains 0.5% free iron. You can get straight up metallic iron by just passing magnets over moon dust. For more complicated processing we can do HF acid leach like that mentioned above or we can use a new up and coming process magma electrolysis. In magma electrolysis you melt moondust to magma and electrolyse it. This yields a melt of iron and silicon. With some processing you can separate out the silicon from the iron.

>>10097106
>>solar is not enough
bullshit. Just build a bigger solar array. Also the day on the moon is 14 earth days long and there aren't cloudy days.
>>nuclear
>>building them in space
no. You could ship up an entire ready to go kilopower reactor.
>>send materials to earth
use the materials to build a solar power satellite that ships power to earth.
>>without fuel or orbital bombardment
build a mass driver, and have your projectiles use a heatshield, made from say lunar spun basalt glass, to slow down.
>>10097095
>>scanning potentially interesting sites to prospect
or just use a process that can extract useful materials from regular lunar soil.
>>what do you do when the drones break?
advance robotics to the point where a human on earth can diagnose the problem and control the drones so as to repair the broken ones. We need robots that can autonomously carry out high level commands like 'remove that screw,' 'plug that wire in,' 'pick up that rock.' This way the human can have the robot do everything we need to repair another robot with the time lag.

>>10028283
Extracting resources on Mars for use anywhere except Mars is questionable. There is potentially a case for exporting propellant though.
>>10028373
THIS IS NOT AN ANTARCTICA COLONIZATION THREAD. Yeah, yeah you're making a point about how it'd be easier and that we have no reason to colonize Mars, but that's not what I made this thread for. This thread focuses on the how it could be accomplished if it can be accomplished at all, rather than the why. Also, I'd really like to show with numbers that Musk's Mars plans are full of shit with hard numbers that address 'how' it cannot be done or 'how' impractical it is.
>> coal
You don't need coal to make steel, you can use molten oxide electrolysis:
https://www.technologyreview.com/s/611961/this-mit-spinout-could-finally-clean-up-steel-one-of-the-globes-biggest-climate-polluters/
The process in the above article above was invented to work on the Moon. You can electrolyze molten rock to get ferrosilicon alloy and oxygen. You can further process the ferrosilicon to seperate the iron and silicon. There are other ways to reduce metal oxides than just using coal. Hydrogen reduction is another means of doing so.
>>ore
Ore is rock that contains certaim elements at slightly higher concentrations. Near term it probably makes more sense to use what is available around you through processes like molten oxide electrolysis.
>>10028390
Ok cool, now how much propellant do you actually have to produce? Where are these calculations? What assumptions did they make? Is this the ideal case? CO2 is easy, you can extract that from air, water is difficult because you have to dig for it.

Now I have a hunch that one of the big limiting factors for how fast we can set up a Mars colony will be water. You need hydrogen for propellant, plastics, refining regolith. There are likely a number of processes that will consume hydrogen.