[HN Gopher] UK company sends factory with 1,000C furnace into space
___________________________________________________________________
UK company sends factory with 1,000C furnace into space
Author : vekerdyb
Score : 131 points
Date : 2025-12-31 08:04 UTC (3 days ago)
(HTM) web link (www.bbc.co.uk)
(TXT) w3m dump (www.bbc.co.uk)
| VoidWhisperer wrote:
| 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?
| dogma1138 wrote:
| Return.
| coldtea wrote:
| 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.
| ahazred8ta wrote:
| 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.
| skippyboxedhero wrote:
| The idea is that they get government funding from credulous
| civil servants. There is no actual idea here, there is no
| business. The idea that a country that is unable to supply
| basic infrastructure is suddenly going to build CPUs in space
| is obvious bullshit.
| Squeeze2664 wrote:
| 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.
| HPsquared wrote:
| In a way that's good because they don't need logistics for the
| entire supply and production chain up there, they can just drop
| the (small and presumably valuable) silicon crystals back to
| Earth.
| guepe wrote:
| Note that growing ingots is an incredible feat at that purity
| and size that they achieve on earth. It's already a very very
| hard step in a crazy process for entire chip manufacturing.
| RobotToaster wrote:
| Leaving aside the small issue of getting there, wouldn't it
| be easier to achieve in space given there is fewer impurities
| like air floating around?
| stevenwoo wrote:
| Even given that the sending finished products back to earth
| in same clean room conditions for next step seems
| challenging to make profitable. If it's a proof of concept,
| okay, but to take it further the lithography step takes RV
| sized machines.
| meindnoch wrote:
| >this is not manufacturing chips in space
|
| *crisps
|
| It's from the UK.
| mkl wrote:
| Chips are definitely a thing in the UK. Like French fries but
| usually chunkier:
| https://en.wikipedia.org/wiki/Fish_and_chips
| whynotmaybe wrote:
| Which is called Fish and Chips in Canada, even though it's
| served with fries.
| NooneAtAll3 wrote:
| I heard some restaurant getting sued for selling "fish
| and chips" without fish, but I don't remember how it
| ended
| thayne wrote:
| Same in the US
| teekert wrote:
| In the Netherlands we have two words for fries and you
| know if someone is from the north or the south based on
| their use: Patat, north en Friet, south, particularly in
| the South people are sensitive to using the wrong,
| northern word. (And chips are just crisps here.)
| adventured wrote:
| The chunkier fries are akin to what the US calls steak
| fries and are very common in the US as well.
| OJFord wrote:
| That's gastro pub chips though, not chippy chips.
| shermantanktop wrote:
| If you are lucky, you get triple-fried chips. Which are
| just as good/bad as they sound.
| omnicognate wrote:
| > just as good/bad as they sound.
|
| I.e. extremely good and not at all bad.
| zdragnar wrote:
| What Americans call chips (potato or corn) the UK typically
| calls crisps.
| sillyfluke wrote:
| There is a Douglas Adams inspired The Great British Bake Off
| subplot out there somewhere.
| willis936 wrote:
| Do gravity-based defects outweigh displacement defects from
| fast particles?
|
| Also, there are large headwinds from having to ship up a large
| quantity of raw material and have to deorbit a payload so
| fragile that any amount of shock is unacceptable. Maybe a high
| purity silicon boule pays for these headwinds with room for
| profit on top. I'm skeptical, but time will tell.
| mattmaroon wrote:
| It would be pretty easy for them to have done the math in
| advance and I'm sure they're aware that space travel is
| expensive.
| dmurray wrote:
| I'm sure they've done the maths and determined that it's
| not economically feasible at all.
|
| But if it works as a proof of concept, in three or four
| generations time perhaps they'll have a scalable process
| which pays for itself.
| jdranczewski wrote:
| If you've managed to find more details about what process
| exactly they're implementing I'd be glad to see it - I assumed
| plasma-based growth, since the BBC article mentions that it's a
| plasma that is at 1000C here (making heat dissipation less of a
| problem too), but if they're growing ingots that would usually
| be done from liquid silicon, which sounds like a mess in space.
| So are they doing plasma-growth of ingots (which I haven't
| heard of, but I haven't heard of many things), or are they
| bringing wafers up and growing ultra-pure layers on top... The
| website is not super clear on this from what I've seen.
| MORPHOICES wrote:
| "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.
| kruffalon wrote:
| This looks like an interesting comment, to me
|
| But I'm having a hard time parsing it.
|
| Is it a quote? Who is Moxley?
|
| Where do the different statements begin and end?
| miningape wrote:
| Hey chatgpt, play out a conversation with "Moxley"
| tomatotomato37 wrote:
| 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.
| HPsquared wrote:
| Something like a vacuum flask, I imagine. Vacuum is a very good
| insulator already and you minimise radiative heat transfer
| (infrared glow) by making a surface shiny and metallic usually
| (low emissivity)
| jstanley wrote:
| If you have to bring your own vacuum flask, don't you lose
| half the benefit of doing it in space?
| rokkamokka wrote:
| Isn't the primary benefit the lack of gravity?
| imtringued wrote:
| You can reduce metals through vacuum pyrolysis at much
| lower temperatures without a reducing agent if you have a
| vacuum. This could make industrial scale processing of
| steel relatively easy on the moon.
| pfdietz wrote:
| Reducing ferric oxide to magnetite, perhaps, but I think
| if you tried that with ferrous oxide you'd get iron vapor
| coming off along with the oxygen.
|
| An issue with any high temperature process is things
| start evaporating. This is part of why carbothermal
| reduction of aluminum oxide doesn't work: at the required
| temperature aluminum oxide is volatile.
|
| (There are thermochemical water splitting technologies
| that exploit partially reducing transition or rare earth
| oxides at high temperature, then reacting them with steam
| at a bit lower temperature to make hydrogen. I believe
| cerium oxides are the current best approach there,
| although still not competitive.)
| RobotToaster wrote:
| I imagine you just need the flask part, the vacuum is
| rather easy in space.
| willis936 wrote:
| Good electrical conductors are also good thermal conductors.
| It's a fun system challenge to minimize what needs to be hot,
| but some things will have to get hot. It could be reduced to
| a photodiode, transistor, and a relay.
|
| But how do you get the power to the heater in a compact way?
| aeonik wrote:
| Diamond is my favorite exception to this, one of the best
| thermal conductor and insulators.
| HPsquared wrote:
| Boron nitride too. I guess the thermal vibrations
| transmit well through a stiff microstructure.
| pfdietz wrote:
| One notable exception to this is superconductors. One might
| naively think that because superconductors have zero
| electrical resistance, they also have zero thermal
| resistance. But this is wrong (sorry, Larry Niven)! The
| superconducting charge carriers (Cooper Pairs) have zero
| entropy, so they can't carry heat. Thermal conductivity of
| a superconducting material drops when it becomes
| superconductive.
|
| I believe high Tc superconductors have been used (or at
| least proposed to be used) as current leads for carrying
| current into low Tc superconductors from somewhat higher
| temperature normal conductors.
| marcosdumay wrote:
| Vacuum is only a nearly perfect insulator until a few hundred
| degC. After that, radiation dominates over every other form of
| heat exchange, and it stops making any difference.
|
| A couple of meters long steel rod with a dissipator on the end
| can easily keep electronics at Earth surface temperatures even
| if you heat the other end to 1000degC.
| crote wrote:
| > "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?
| PunchyHamster wrote:
| There is no reason to do it with current processes. It could
| possibly reduce defect rate/increase yields but I'm not sure
| impurities are even leading cause now.
|
| But we might discover other uses, that's what science is.
| threethirtytwo wrote:
| No way. This is being done because there's a predictable path
| to profitability. It's not just random shot in the dark
| science you can sometimes see in academia. It's just this
| path isn't clear to us laymen... I know because launching
| into space isn't something that will be done just for science
| wizzwizz4 wrote:
| Hubble Space Telescope. James Webb Space Telescope.
| threethirtytwo wrote:
| I should've specified corporate vs. academia. Those
| telescopes clearly aren't corporate. More akin to
| academia.
|
| Corporations won't go to space just for "science" and
| that is the case here.
| wizzwizz4 wrote:
| That doesn't mean they're going to space for traditional
| profitability, though. The advertising budget is a
| classic example: none of those activities are profitable,
| but they alter market behaviour such that other parts of
| the company can make money.
| threethirtytwo wrote:
| Yeah going to space is not an "advertising budget".
| wizzwizz4 wrote:
| Which doesn't mean it's not another example of the same
| phenomenon. _Many_ companies do things that are not, and
| will never be, profitable (in the traditional sense),
| because they have ulterior motives.
| threethirtytwo wrote:
| False. The objective a company is to make profit,
| anything else is a side quest and is less relevant.
|
| There isn't an "ulterior" motive here. The motive is
| clear: profit. We just can't pinpoint the exact pathway
| to it in terms of details.
| wizzwizz4 wrote:
| Not all companies have "make profit" as their primary
| goal. But what I meant by "ulterior motive" was that
| their space factory may never provide enough output for
| that alone to justify the cost. Their plan may be for the
| factory in space to bring money in other ways (e.g.
| making them eligible for grants they would otherwise not
| have received).
| intended wrote:
| I came across space forge due to some algorithmic discovery
| on YT a few months ago.
|
| If I am recalling it correctly - They're focusing on goods
| that can be made, that have a high $ value vs volume and
| density. High purity silicon is what they identified.
| MomsAVoxell wrote:
| Come on, it's the BBC - which has a much wider spectrum of
| audience than you and me.
|
| Think with it a little - the statement "in the 5G tower" is
| intended to bring the context of this event closer to those who
| are not knowledgeable about this technology, but would
| nevertheless read the article. You and I may understand that
| the economies of scale don't make sense yet - but they _could_
| , some day, if this technology succeeds, be relevant to the
| local neighborhood.
|
| To many, the 5G tower is the most mystical, mysterious
| technology in their neighborhood - and indeed, the silicon
| ingots being manufactured this way would, eventually, find
| their way to the local neighborhood if this technology is
| successful.
|
| It would probably have been more appropriate to say "some day
| these ingots will power the supercomputers in your pocket",
| which would be an accurate statement - but that is a whole
| order of magnitude of different economic scale than in the
| industrialization of cell networks. Maybe it'd be more
| appropriate for the BBC writer to have said that satellites
| might one day benefit from space-grown silicon wafers - but
| that is still to distant to the Mom and Pop readership they're
| targeting in these articles...
| georgefrowny wrote:
| 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?
| fsh wrote:
| The best silicon single crystals still contain impurities at
| the 1E-11 level. This project is about doing crystal growing in
| low gravity (the ultra-high purity is only achieved due to the
| growth process). The vacuum of space is a lot worse than what
| can be achieved in the lab, especially in low orbits.
| boothby wrote:
| What I'm struggling with is the hard radiation, which causes
| defects and even impurities even if you start with
| isotopically pure feedstock
| twic wrote:
| There have been experiments to create higher vacuums in
| space, which have been used to grow semiconductor crystals:
|
| https://scfh.ru/en/papers/vacuum-in-the-wake/
|
| https://en.wikipedia.org/wiki/Wake_Shield_Facility
|
| I'm not saying it's practical, but it's pretty cool.
| 3eb7988a1663 wrote:
| What is the current defect rate preventing from happening? If
| you can lower it three orders of magnitude, what then becomes
| possible?
| notahacker wrote:
| Earth's surface isn't a microgravity environment. The principle
| that you can get purer crystalline structures with less
| interference from the earth's gravitational pull has explored
| on space stations since the 1970s; that aspect and potential
| applications in higher performing semiconductors and drugs are
| fairly well understood
|
| What hasn't been solved yet and Space Forge (and Varda et al
| for biotech) are hoping to solve are the unit economics of
| reentry vehicles to actually make it viable for manufacturing.
| metalman wrote:
| 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.
| georgefrowny wrote:
| 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.
| actionfromafar wrote:
| UK - the only country to achieve to-orbit capability only to
| shelve it after the first flight.
| fsh wrote:
| As an ESA member state, the UK has had launch capabilities via
| Arianespace for a couple of decades.
| rsynnott wrote:
| France essentially did that too, in favour of the ESA (I think
| the French launcher may have had a couple launches, but
| certainly no more than ten).
| sandworm101 wrote:
| Um... i see some red flags. This test to heat gasses in a furnace
| looks a little sus. It is basically a micowave oven ... in
| space!! My kitchen microwave can hit 1000c. Let styropyro at it
| and it could probably do 10,000c. Then i came across this gem on
| thier website.
|
| >> Radiators facing cold space can freely produce temperatures
| near absolute zero for ultra-fast curing without the need for
| cryogenics.
|
| https://www.spaceforge.com/
|
| I dont see a path to a product. I see a company farming
| investment and government programs with overhyped "experiments"
| that have been already done many times. They even talk about
| testing a heat shield for reentry as if that tech is somehow new.
| Want to bring samples back? Send your microwave to the space
| station and bring them back like everyone else.
| MomsAVoxell wrote:
| This is really, really exciting. The moment this spark ignites,
| it will herald a new reality for industrialization of space, and
| I for one cannot wait to see it succeed.
|
| One of the things I truly believe will elevate our species is
| space industrialization. If, in 20 years or so, we send a fleet
| of space furnaces to 16 Psyche [1], there is a very real
| possibility that we will be able to move a lot of Earths heavy
| industrial processes to space. Can you imagine - 3D printed
| Starship hulls being made from the immense resources of 16
| Psyche?
|
| Literal pallets of iPhones being landed from near earth orbit.
|
| It sounds like whacky science fiction now, and for now it really
| is just that, but the launch and successful mission of Space
| Forge and other companies like it bring us all a single step
| closer to seeing that reality play out.
|
| I truly hope we can survive long enough to move heavy metal
| industry to space, and use that event to return the Earth to a
| garden state. It's a long shot, but oh what a beautiful world it
| would be in 100 years time if this dream can be kept alive, and
| actually achieved.
|
| [1] - https://en.wikipedia.org/wiki/16_Psyche
| marcosdumay wrote:
| Research has gone this direction several times. Every time it
| did, ground factories discover some way to improve their
| quality so that space manufacturing became irrelevant again.
|
| That's not to say it will happen again. But it's not a certain
| thing.
| MomsAVoxell wrote:
| Wins at both ends, whats to not like?
| thayne wrote:
| > a heat shield named Pridwen after the legendary shield of King
| Arthur will be deployed to protect the spacecraft from the
| intense temperatures it will experience as it re-enters the
| Earth's atmosphere.
|
| Why bring back the entire spacecraft and not just the finished
| product?
|
| I'm also curious how they handle cooling the silicon, since
| dissipating heat in space is kind of difficult.
| 3eb7988a1663 wrote:
| Harder engineering to separate the components? Making the oven
| + crystal formation work is probably maxing out their novelty
| budget without trying to make the craft partially disassemble.
|
| Plus, they can study the oven after the process which is likely
| to be helpful if the entire experiment poops the bed.
| rdiddly wrote:
| _" 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."_
|
| LOL! Talk about an anticlimax. Either this is a lack of
| imagination on the CEO's part, or he's dumbing it down for us, or
| that level of silicon purity/regularity, is one of those nice-
| sounding but impractical Platonic ideals that ends up being kind
| of a waste. But it, and/or solving the attendant problems, might
| be an important precursor for some future innovation.
| golem14 wrote:
| This is a microwave oven. I wonder how feasible is to have a
| solar concentrator, if you can build it out of foil, mostly, like
| a light sail.
|
| You would have to 1) keep turning it toward the sun and b) reduce
| time in earth's shadow, which means a polar orbit?
___________________________________________________________________
(page generated 2026-01-03 23:01 UTC)