I'm only YouTube-level informed on how silicon manufacturing works, but something that is, perhaps intentionally, not made clear to someone unfamiliar with the field is that this is not manufacturing chips in space. This is to grow the crystals only, the very first step in silicon chip manufacturing. This is how you get the ingot, then you slice it to get the wafers upon which the chips are built. The reason you would even consider doing it in space in the first place is because, on Earth, gravity and other forces are stronger and result in lower-purity crystals. Basically, what I'm getting at, is that I believe this is pretty much a glorified oven. Moving the entire manufacturing process in space wouldn't make sense, as I don't think the benefits to other steps of the process like CVD would outweigh the insane costs of sending things into orbit.
I'm curious what the thermal management system on this looks like. On one hand, vacuum being in essence a perfect insulator works in favor of keeping the silicon hot for the very long time it takes to pull a boule while requiring very little energy. On the other hand, you have to make sure the control electronics don't also heat up to 1000C. I'm also curious how you keep the molten silion separate from the crystal without gravity keeping it in the crucible. I bet a lot of interesting engineering going on here.
> "The work that we're doing now is allowing us to create semiconductors up to 4,000 times purer in space than we can currently make here today," says Josh Western, CEO of Space Forge.
> "This sort of semiconductor would go on to be in the 5G tower in which you get your mobile phone signal, it's going to be in the car charger you plug an EV into, it's going to be in the latest planes."
Okay, but, we have 5G towers, car chargers, and planes right now?
I understand that purer material is better, but to what extent are the impurities of current wafer production methods limiting us? Why is shooting the furnace into space the best option? Why is making wafers 4+ orders of magnitude more expensive the solution we should go for?
4000 times purer seems a questionable claim to me. 4000 times purer then what? Current Earthbound state of the art? A guy in shed with a vacuum pump and a heater? On what axis: crystal defects or contamination?
High vacuums aren't at all impossible on Earth and silicon boules are already single crystals.
What exactly about their process permits such a huge quality improvement?
Is the idea that it will manufacture all of these chips and then both the 'factory' and the resulting materials will return from space, or that the factory would stay in orbit and send materials back?
The idea is that the furnace will melt the silicon into a 'boule' (cylinder). <https://en.wikipedia.org/wiki/Boule_(crystal)> The solid boule will be returned to Earth for lab analysis.
I think that this is far from the first experiment in growing crystals in space, so they must have good evidence for what properties to expect from single crystal silicon grown in space, my guess is that the purity is not so much which elements are or are not included, but more the perfection of the physical crystaline structure, which may introduce the posibility of the crystal having different characteristics in different planes that can be exploited for various purposes.
The crystal planes indeed do have different properties which are exploited already. Some planes have better flatness and defect rates then others, and etching rates and undercutting varies according to the surface orientation and the orientation of long thin details to the bulk lattice.
Silicon boules are grown with defined orientations. There's a system of flat edges ground into the wafers which indicate the orientation.
I'm only YouTube-level informed on how silicon manufacturing works, but something that is, perhaps intentionally, not made clear to someone unfamiliar with the field is that this is not manufacturing chips in space. This is to grow the crystals only, the very first step in silicon chip manufacturing. This is how you get the ingot, then you slice it to get the wafers upon which the chips are built. The reason you would even consider doing it in space in the first place is because, on Earth, gravity and other forces are stronger and result in lower-purity crystals. Basically, what I'm getting at, is that I believe this is pretty much a glorified oven. Moving the entire manufacturing process in space wouldn't make sense, as I don't think the benefits to other steps of the process like CVD would outweigh the insane costs of sending things into orbit.
I'm curious what the thermal management system on this looks like. On one hand, vacuum being in essence a perfect insulator works in favor of keeping the silicon hot for the very long time it takes to pull a boule while requiring very little energy. On the other hand, you have to make sure the control electronics don't also heat up to 1000C. I'm also curious how you keep the molten silion separate from the crystal without gravity keeping it in the crucible. I bet a lot of interesting engineering going on here.
“I've noticed the pattern of big technical ideas. They’re obvious in hindsight. They cost money early.” ~
“What’s worked for me is a rough three-question filter,” Moxley continues,
“What assumption would be most easily disproven if it’s false?” “What
Is it something that can be cheaply verified in weeks, not months?
“Who would notice if this quietly failed?”
When I don’t skip this, what ends up happening is that I am endorsing the wrong thing. When I do, good ideas also die prematurely.
What the others do, curious to see.
Do you write out assumptions or is it an informal process?
How early do you bring outsiders to poke holes?
Any heuristics for distinguishing between "hard but right" and "just hard"?
Examples always appreciated. Failures too.
> "The work that we're doing now is allowing us to create semiconductors up to 4,000 times purer in space than we can currently make here today," says Josh Western, CEO of Space Forge.
> "This sort of semiconductor would go on to be in the 5G tower in which you get your mobile phone signal, it's going to be in the car charger you plug an EV into, it's going to be in the latest planes."
Okay, but, we have 5G towers, car chargers, and planes right now?
I understand that purer material is better, but to what extent are the impurities of current wafer production methods limiting us? Why is shooting the furnace into space the best option? Why is making wafers 4+ orders of magnitude more expensive the solution we should go for?
4000 times purer seems a questionable claim to me. 4000 times purer then what? Current Earthbound state of the art? A guy in shed with a vacuum pump and a heater? On what axis: crystal defects or contamination?
High vacuums aren't at all impossible on Earth and silicon boules are already single crystals.
What exactly about their process permits such a huge quality improvement?
One thing that is unclear to me from the article:
Is the idea that it will manufacture all of these chips and then both the 'factory' and the resulting materials will return from space, or that the factory would stay in orbit and send materials back?
The idea is that the furnace will melt the silicon into a 'boule' (cylinder). <https://en.wikipedia.org/wiki/Boule_(crystal)> The solid boule will be returned to Earth for lab analysis.
Just a tiny part of the operation will happen in space. The result of that will return, then will be made into chips on earth.
Return.
I think that this is far from the first experiment in growing crystals in space, so they must have good evidence for what properties to expect from single crystal silicon grown in space, my guess is that the purity is not so much which elements are or are not included, but more the perfection of the physical crystaline structure, which may introduce the posibility of the crystal having different characteristics in different planes that can be exploited for various purposes.
The crystal planes indeed do have different properties which are exploited already. Some planes have better flatness and defect rates then others, and etching rates and undercutting varies according to the surface orientation and the orientation of long thin details to the bulk lattice.
Silicon boules are grown with defined orientations. There's a system of flat edges ground into the wafers which indicate the orientation.