Sciencemadness Discussion Board - Making rubies using microwave

Making rubies using microwave

Bedlasky - 29-10-2021 at 19:51

Hi,

I came across this interesting method of making synthethic rubies.

https://www.youtube.com/watch?v=ybcdRQmQcHQ

elementcollector1 - 7-11-2021 at 10:18

Realistically, this is more of a 'ruby powder' or 'ruby bead' synthesis, given his setup, but there's no shame in that, and he definitely achieved ruby!

That said, I have to wonder how this would go if he could keep it molten for longer - with a silicon carbide block, maybe, or some other decent microwave receptor. Might at least get a ruby 'glass'.

I'm currently building a DIY at-home Czochralski furnace that reaches 2000 C to grow my own sapphires - it's complicated, but not out of reach for the ordinary chemist if they're willing to put in the work and understand the concepts.

[Edited on 11/7/2021 by elementcollector1]

Triflic Acid - 8-11-2021 at 06:11

Speaking about rubies, I've put in the last few weeks of work towards making a perfectly clear arc-melted ruby. What I've ended up doing is mixing potassium dichromate into the aluminum oxide and then arcing it. This reacts with the carbon produced and gives a transparent gem. I think that the potassium ions are screwing up my results, so CrO3 would be ideal. Does anyone have any tips on producing it from K2Cr2O7, preferably without destroying my dwindling stock of sulfuric acid. I have about ~500mL left and don't want to go distilling just yet.

woelen - 8-11-2021 at 06:53

Why use CrO3? Can't you make rubies with Cr2O3, which is much easier to obtain and also is much safer to use? At those very high temperatures, the CrO3 loses oxygen anyway and it does not oxidize anything, when it is mixed with Al2O3.

If you really want to use CrO3, you can buy that at a good price, both at Onyxmet and at Chemcraft. If you are in the EU, then I think that the net price, including shipping, will be comparable from both suppliers. If you are outside the EU, Chemcraft probably will be a little cheaper. I have purchased the chemical from both suppliers. Chemcraft's is a fine-grained solid, which most likely is easier to mix, but Onyxmet's is less humid, the latter is a perfectly dry sample of fairly large scales.

Ruby

MadHatter - 8-11-2021 at 10:45

Woelen, I'm with you on Cr₂O₃. I buy mine from a pottery shop.
BTW, if you want sapphire most likely you'll need titanium.

Triflic Acid, can you get or make ammonium dichromate ?
This is quite flammable and produces Cr₂O₃ as the only solid.

metalresearcher - 8-11-2021 at 11:04

I have tried it by mixing Al2O3 with 1-2% green Cr2O3 heating by a welder wit carbon electrodes and indeed a few small pinkish ruby beads appeared. Using a blackled lamp, they fluoresce nicely.

Quote:

I'm currently building a DIY at-home Czochralski furnace that reaches 2000 C to grow my own sapphires - it's complicated, but not out of reach for the ordinary chemist if they're willing to put in the work and understand the concepts.

elementcollector1: Making a Czochralski furnace ? That should be a tedious task, you need a very pure crucible and a strictly controlled device to pull the crystals slowly from the liquid. And how do you heat the furnace ? A temp of 2000+ C can only be obtained by special heating elements, even the 1700 C Kanthal Super don't suffice. And burning H2 or propane in oxygen costs a lot of high pressure O2 cylinders.
A (n abandoned) medical zeolite based O2 generator generated not sufficient O2.

Admagistr - 8-11-2021 at 15:37

Quote: Originally posted by Bedlasky

Hi,

I came across this interesting method of making synthethic rubies.

https://www.youtube.com/watch?v=ybcdRQmQcHQ

I tried to make rubies in the microwave, of aluminium wool, according to the YouTube instructions, but I failed, occasionally a microscopic ruby arose, but it was no success...That's why I invented my own method. I used a loop of aluminum wire, of high purity, diameter d = 1mm. Direct the first experiment ended in success... I'm sending some photos from the experiment...Adam

Admagistr - 8-11-2021 at 15:51

I have to add that I placed the Al wire in question on a powder mixture of Al2O3+Cr2O3. Composed of 98% Al2O3 and 2% Cr2O3. I'm sending a few more pictures that didn't fit in the previous posts. There's one ruby drop at the end of the overcooked Al wire

!Unfortunately, though, I had to stop the reaction after more than 20 seconds because something started burning in the microwave oven and it smelled very bad...

Admagistr - 8-11-2021 at 16:24

Quote: Originally posted by elementcollector1

Realistically, this is more of a 'ruby powder' or 'ruby bead' synthesis, given his setup, but there's no shame in that, and he definitely achieved ruby!

That said, I have to wonder how this would go if he could keep it molten for longer - with a silicon carbide block, maybe, or some other decent microwave receptor. Might at least get a ruby 'glass'.

I'm currently building a DIY at-home Czochralski furnace that reaches 2000 C to grow my own sapphires - it's complicated, but not out of reach for the ordinary chemist if they're willing to put in the work and understand the concepts.

[Edited on 11/7/2021 by elementcollector1]

Hi, do you have a video or a link to information about your homemade Czochralski oven!? With more enthusiasts involved in the construction, we can help each other and share our experiences! Thanks, Adam.

Rubies and unusual thermite mixture...

Admagistr - 8-11-2021 at 16:55

I'm sending a video I made about a year ago, capturing my idea and my efforts to create a ruby melt from Al2(MoO4)3+Cr2O3+Al, in a chamotte crucible. The ruby melt was indeed formed, even sprayed into the surrounding area in response, but it is heavily contaminated with molybdenum, which has not melted and separate from it, unfortunately. I have it confirmed by X-ray analysis. I mean, a lot of MoO3 was squeezing out, as you can see in the video...I recommend you watch the video in full size;-).But I really enjoyed the pyrotechnic effect...
I posted the video on a profile of a friend of mine who's an astrophysicist. His name is Stanislav Hledík. It was back when I didn't have my own YouTube account. The video is still unofficial, in a working version, and will only be translated into English later. But there are important equations in his description, describing the reaction that everyone here at the forum understands;-)...Note how long the mixture burns and stays hot...I invented this mixture myself:-)

https://www.youtube.com/…SRg

Admagistr - 8-11-2021 at 17:04

https://youtu.be/lEosDA7nSRg

Triflic Acid - 8-11-2021 at 21:22

Well, my issue is that the large rubies tend to be really cloudy for me. I want some nice clear gems for my collection. The K2Cr2O7 somewhat solved the problem, since it got decently less cloudy after adding it. I agree that most of it is likely losing oxygen and becoming worthless, but some of it does seem to be helping the issue. I'm not sure if this was unclear in my original post, but I'm adding a small amount of dichromate to the Cr2O3, not using just dichromate.

Admagistr - 9-11-2021 at 13:44

Quote: Originally posted by Triflic Acid

Well, my issue is that the large rubies tend to be really cloudy for me. I want some nice clear gems for my collection. The K2Cr2O7 somewhat solved the problem, since it got decently less cloudy after adding it. I agree that most of it is likely losing oxygen and becoming worthless, but some of it does seem to be helping the issue. I'm not sure if this was unclear in my original post, but I'm adding a small amount of dichromate to the Cr2O3, not using just dichromate.

There is no need to use Cr2O3 for rubies in synthesis at all. I am sending proof, with this photograph, that the rubies were made in a crucible of technical corundum, purity 96%, using only K2Cr2O7! The mixture in the crucible was melted at 1000 C with a mixture composition of Li2CO3+Al2O3+K2Cr2O7+MoO3 for approximately 7 hours.

DraconicAcid - 9-11-2021 at 13:59

Nice!

The synthesis of rubies has been on my mind since 1989. I may have to actually try it some day.

Admagistr - 9-11-2021 at 16:39

Quote: Originally posted by DraconicAcid

Nice!

The synthesis of rubies has been on my mind since 1989. I may have to actually try it some day.

Thanks for the compliment. Make sure you try and let me know how it goes, you might have more success than me;-). I am sending photos, from a similar experiment, but without Li2CO3 and K2Cr2O7. I only had a tiny platinum bowl that couldn't even fit a whole mixture of Al2O3+MoO3+Cr2O3, so I only filled it in part. I heated it at 1000 C for about six hours, in a crucible furnace lined with chammote. I used a CaO to capture the MoO3 vapor. MoO3 also works on beryls-emeralds, chrysoberyls, zircons, hafnons, bromellites synthesis and some spinels that do not contain Mg;-)! Adam

Edit(woelen): removed email address

[Edited on 11-11-21 by woelen]

Triflic Acid - 9-11-2021 at 17:07

You could likely have some success with a tantalum or tungsten crucible, if you could get your hands on one

Admagistr - 9-11-2021 at 17:20

Quote: Originally posted by Triflic Acid

You could likely have some success with a tantalum or tungsten crucible, if you could get your hands on one

Thanks for the advice. Tantalum might be used to go, but it would corrode...Tungsten would corrode even more. In fact, MoO3 is very corrosive to metals at high temperatures because it dissolves the protective layer of oxide on their surface:-(Metals seem to be the best material in platinum, but unfortunately it's terribly expensive. I thought of trying the CaWO4 crucible, but where to get it, or how to make it? Maybe that's a great idea, what do you think;-)?

Competition.

Admagistr - 10-11-2021 at 09:06

It occurred to me that the Supermoderators might run a competition for the prettiest photo/video of an amateur gem and the most original method. If there was a sponsor, some professional gem-maker, like Mr. Chatham, Mr. Oishi, ladies Osmer and Chase...The main prize could be Chatham emerald, Ramaura ruby, Alexandrite, bromellite, or chyrzoberyl, which contains gallium instead of aluminum;-) and crystallizes in a hexagonal system...The condition would be to publish the procedure and teach it to other enthusiasts on the forum;-)! It would be an incentive for others to publish their results and photos/videos;-). The competition could last one year....

unionised - 10-11-2021 at 10:44

Quote: Originally posted by Admagistr

Quote: Originally posted by DraconicAcid

Nice!

. You can also write privately on my : Adm agistr gmail om

It's a bad idea to put your email address on the web.
At best, it will get spammed to hell and back.

Admagistr - 10-11-2021 at 14:19

Quote: Originally posted by unionised

Quote: Originally posted by Admagistr

Quote: Originally posted by DraconicAcid

Nice!

. You can also write privately on my : Adm agistr gmail om

It's a bad idea to put your email address on the web.
At best, it will get spammed to hell and back.

Thank you for the warning. I'm not doing it anymore. Fortunately, this email address is not my main one, it is my second address and I only use it for similar cases...I am new to the forum, should I ask the moderator to remove it if possible?

woelen - 11-11-2021 at 00:10

Quote: Originally posted by Admagistr

[...] should I ask the moderator to remove it if possible?

I removed the email address. People can contact you through U2U.

Admagistr - 26-1-2022 at 18:26

I found this Japanese tutorial on the internet that I wanted to try. I translated it with a translator, but the added image didn't work for translate. Is there anyone here who speaks Japanese who would be willing to translate these instructions into English, please? Anyway, I gave it a try, bought 5 pencils of 9B hardness, which corresponds to natural graphite, and set the pencils on fire and let them burn. I obtained small graphite rod this way simply with point at the end. I took a glass cube for candle with a bowl in the middle and add Al2O3 mixed with 2% Cr2O3 into the depression. Then I stick into the mixture a piece of graphite rod, a few centimeters long, with the tip up, protruding about a centimeter above the mixture, placed it in the microwave and turned it on - nothing happened. Then I stick into the mixture second piece of graphite rod with the tip down near the first graphite rod, when I turned on the oven there was a blinding white glow and melting of the glass after a short time, eventually the glass exploded and the Al2O3+Cr2O3 mixture splattered all over the microwave owen. After the reaction was complete, one fused glass shard was covered with a few microscopic ruby crystals that fused into the glass. The next day I changed the experiment a bit, using a crucible from sintered corundum and stick 2 graphite rods close together in the oxide mixture, tips down. Under the corundum crucible I placed a disk of quartz glass which stood on a rotating glass plate. When I used only one of the graphite rod, again nothing happened and no plasma was produced. It just heated up the crucible a bit, with the oxide mixture. The reaction took a terrible course, dazzling plasma plumes appeared, white, purple and blue in colour, which repeatedly exploded, I thought the crucible had been shattered and nothing was left of it. When I opened the microwave owen, to my surprise, I found the cup was perfectly fine. A small amount of oxides had been scattered around the microwave, but it looked like nothing had happened and no rubies had formed. To find out what was wrong, I used a spoon to gradually remove the material from the crucible... Almost at the bottom of the crucible, something squeaked and I pulled out an opaque and irregular ruby pearl over 0.5 centimeters in size! The bottom of the crucible was blackened under the graphite tips. I will send photos later.
http://limejuice.jugem.jp/?eid=269

[Edited on 27-1-2022 by Admagistr]

oberkarteufel - 29-1-2023 at 02:40

Quote: Originally posted by Admagistr

I used a CaO to capture the MoO3 vapor.

Hi! Could you please elaborate on this? Was the CaO just put in a dish next to the crucible, or was the setup more complicated?

Also - could the molybdenum crucible be a good, cheaper ersatz for the platinum one?
Because, I dunno, maybe it gets eaten by the MoO3 melt, or is oxidixed at 1100*C too quickly and would require an inert atmosphere... Just asking the ones with more experience, need to check if this idea holds water.

I also have a question regarding the microwave synthesis.
In this link https://www.youtube.com/watch?v=ybcdRQmQcHQ&ab_channel=N... the guy has no success with a SiC kiln for glass melting. Could it be possible to combine the ignition by aluminium fiber with the SiC kiln to maintain higher temperature and provide slower cooling/better conditions for the melt to crystallize? Honestly, I'm not sure if microwaves would even be able to penetrate inside the through the carbide layer.

Zdravím mého jižního souseda.

Admagistr - 29-1-2023 at 09:35

Quote: Originally posted by oberkarteufel

Quote: Originally posted by Admagistr

I used a CaO to capture the MoO3 vapor.

Hi! The CaO was in a chamote bowl, which was standing on a chamote bowl, which contained a platinum crucible with reaction mixture.The first chamote bowl was covered by this second bowl.The two bowls were identical.I found, however, that contact of the chamote with platinum is not ideal, there is a little corrosion of platinum.It caused by SiO2 at high temperature, in contact with platinum.It is stated that it is necessary to use corundum ceramics, which does not have this defect.
The molybdenum crucible can only be used in the presence of argon, or in a vacuum, but MoO3 itself acts quite oxidatively and would certainly react-reduce MoO3 to lower molybdenum oxide.So much better would be a tantalum crucible, as a cheaper substitute for platinum, but it would also be corrosive, perhaps it would be worth using a large tantalum crucible, or a sealed pipe, for one or two uses.In this way we could eventually create really big and maybe even nice rubies!Ta2O5, which would be produced here could also act as a mineralizer for the synthesis of ruby crystals

.It would like to try, but tantalum is not very cheap either.
Your idea of microwave synthesis is interesting and it would like to try it, but you are right that SiC would absorb most, maybe all microwaves and aluminum would not get these microwaves, but it is just a theoretical assumption, it needs to be tested experimentally...I still want to go back to the aluminum wool, unlike the aluminum wire, I had almost no success with it, there was mostly no plasma generation.Moreover, it is not made of pure aluminum, but it is made of technical aluminum alloys, these other metals contaminate the resulting rubies and reduce their quality.But I probably used wool of inappropriate thickness.

Greetings to my northern neighbor.

Admagistr - 30-4-2023 at 13:52

I found this Japanese video with instructions on how to make really big rubies in the microwave!With the help of a burning charcoal, if Parakeet will be so kind, he will explain to us exactly how to do it...

https://www.youtube.com/watch?v=7KAytAsJHSY

Parakeet - 1-5-2023 at 05:25

Sure! Let me see…
Spread a mixture of 95% aluminum oxide and 5% chromium oxide in a microwave oven. (4:15)
Place a small piece of aluminum foil, and put a heated charcoal on it. (4:32)
Cover the charcoal with more oxide mixture. On top of this mound, put a mechanical pencil lead. (4:40)
Turn on the microwave oven to 700W.
If a fire breaks out, turn off the microwave and cover the place where the fire broke out with a layer of oxide mixture. (5:20)
After 7 hours of heating, collect the rubies.

For his second attempt, he put the mechanical pencil lead directly on the aluminum foil and placed a charcoal next to the lead. (Make sure that the lead touches the charcoal) (7:50)
Then after covering it with Al/Cr oxide powder, turn on the microwave and the rest was similar to the first attempt.

However, some details are not explained in the video. For example, I can see something like a sheet in the microwave oven (4:15), but it was not mentioned. Also, in the video they say safeguards are in place, but there is no specific explanation.
So just be careful if you actually do this.

If you have any questions, please feel free to ask!

Admagistr - 1-5-2023 at 07:37

Quote: Originally posted by Parakeet

Sure! Let me see…
Spread a mixture of 95% aluminum oxide and 5% chromium oxide in a microwave oven. (4:15)
Place a small piece of aluminum foil, and put a heated charcoal on it. (4:32)
Cover the charcoal with more oxide mixture. On top of this mound, put a mechanical pencil lead. (4:40)
Turn on the microwave oven to 700W.
If a fire breaks out, turn off the microwave and cover the place where the fire broke out with a layer of oxide mixture. (5:20)
After 7 hours of heating, collect the rubies.

For his second attempt, he put the mechanical pencil lead directly on the aluminum foil and placed a charcoal next to the lead. (Make sure that the lead touches the charcoal) (7:50)
Then after covering it with Al/Cr oxide powder, turn on the microwave and the rest was similar to the first attempt.

However, some details are not explained in the video. For example, I can see something like a sheet in the microwave oven (4:15), but it was not mentioned. Also, in the video they say safeguards are in place, but there is no specific explanation.
So just be careful if you actually do this.

If you have any questions, please feel free to ask!

Thank you so much, Parakeet! When I try it and something isn't quite clear, I'll ask you. I have experience with similar experiments in the microwave,I burned two artist pencils,which are made of pure graphite without any impurities and I stuck their carbon rods close together in the Al2O3+Cr2O3 mixture.The reaction was terrifying with explosions and ball lightning,so I couldn't even film it,I was tempted to end it prematurely,but I overcame my fear and let the reaction run in the set mode for 30 seconds until the end. I got the biggest ruby I have made so far,but of course it doesn't have gem quality,it is opaque,the size is between 0.5cm to 1cm.Do they do the reaction continuously for 7 hours,or do they always turn the microwave off and then keep repeating it until they reach a total time of 7 hours?Thanks!

Parakeet - 1-5-2023 at 14:40

Quote: Originally posted by Admagistr

Do they do the reaction continuously for 7 hours,or do they always turn the microwave off and then keep repeating it until they reach a total time of 7 hours?Thanks!

He does not clearly explains that. There is little explanation of how the microwave oven was operated.
However, the dial on his microwave oven only has a scale up to 15 minutes, so he probably repeated turning it on and off for 7 hours.

Microwave Corundum

davidfetter - 15-10-2023 at 19:32

Quote: Originally posted by Bedlasky

Hi,

I came across this interesting method of making synthethic rubies.

https://www.youtube.com/watch?v=ybcdRQmQcHQ

Admagistr - 16-10-2023 at 12:39

I just tried this yesterday with my younger daughter using a microwave (USD 60) I got for the purpose. Amazingly, the microwave oven isn't broken yet, and has produced things that'll glow. We used 1% (w/w) Cr₂O₃ in aluminum oxide. Because I was already at a pottery supply place getting the materials, I also got a couple of refractory plates, one for the bottom, a smaller one for the top, and a clay bit with a hole in it to use as a crucible.

We didn't have aluminum wool, so we tried copper, which didn't work right away, and parts cut from a steel scrubber (more flat coils than thin fibers). To everybody's delight, we got things that glowed under UV light, and are working up to making bigger ones.

I saw that program in Japanese on making way bigger bits of corundum in the microwave, but it was very light on details. Are there written ones somewhere else, ideally in English?

[Edited on 2023/10/16 by davidfetter][/rquote]
I haven't tried the Japanese process yet, but I'm about to.Do you have any pictures of the experiment?If you use copper or steel to start the reaction, I'm worried that you might contaminate the rubies.Have you observed anything like this?Higher concentrations of iron in the crystal lattice of ruby interfere with its luminescence. I have tried tungsten and molybdenum.Its oxides promote ruby crystallization,but the arc did not ignite,probably because the energy required to create a plasma from these materials exceeds that available in a microwave oven.The solution would probably be to coat Mo or W with a layer of graphite or aluminum.

davidfetter - 17-10-2023 at 12:35

Quote: Originally posted by Admagistr

I just tried this yesterday with my younger daughter using a microwave (USD 60) I got for the purpose. Amazingly, the microwave oven isn't broken yet, and has produced things that'll glow. We used 1% (w/w) Cr₂O₃ in aluminum oxide. Because I was already at a pottery supply place getting the materials, I also got a couple of refractory plates, one for the bottom, a smaller one for the top, and a clay bit with a hole in it to use as a crucible.

We didn't have aluminum wool, so we tried copper, which didn't work right away, and parts cut from a steel scrubber (more flat coils than thin fibers). To everybody's delight, we got things that glowed under UV light, and are working up to making bigger ones.

I saw that program in Japanese on making way bigger bits of corundum in the microwave, but it was very light on details. Are there written ones somewhere else, ideally in English?

[Edited on 2023/10/16 by davidfetter][/rquote]
I haven't tried the Japanese process yet, but I'm about to.Do you have any pictures of the experiment?If you use copper or steel to start the reaction, I'm worried that you might contaminate the rubies.Have you observed anything like this?Higher concentrations of iron in the crystal lattice of ruby interfere with its luminescence. I have tried tungsten and molybdenum.Its oxides promote ruby crystallization,but the arc did not ignite,probably because the energy required to create a plasma from these materials exceeds that available in a microwave oven.The solution would probably be to coat Mo or W with a layer of graphite or aluminum.

I tried to take a few with my phone camera, but the UV light I was using to see whether rubies had come out did terrible things to the images it captured. What sorts of cameras are people using for this kind of thing?

The steel scrubber bits definitely get wound into the ruby we've made so far. I'm looking for sources of Al wool to try as...microwave antennae? I guess that's kinda what they're doing.

Admagistr - 17-10-2023 at 13:58

The steel scrubber bits definitely get wound into the ruby we've made so far. I'm looking for sources of Al wool to try as...microwave antennae? I guess that's kinda what they're doing.[/rquote]

I have had success with thin aluminum wire, which I made into a ring, which had a gap of several millimeters between the opposite ends of the wire.I also have a problem with shooting in UV light, I have not been able to take a picture chromium doped Ga2O3, which in UV light had several beautiful colors depending on the angle at which the UV light fell. I wrote a thread about it,here on SM with photos,but only in natural light.In natural light it is green,in UV it is blue-green,blue,grey,purple and red,like the gemstone alexandrite!But alexandrite is more likely to be green in sunlight and purple-red in UV,it doesn't have as many colors in UV as Ga2O3:Cr3+.

[Edited on 17-10-2023 by Admagistr]

Mateo_swe - 17-4-2024 at 07:14

I would like to get a small ruby vee-bearing for a very low friction DIY pulsemotor.
I wonder if making some synthetic ruby pieces and somehow grind a vee hole in the middle can be done.
Grinding such a vee-dent in a ruby would require some grind powder of equal or harder substance, that means ruby or diamond dust.
Or a tool already coated with such powder.
But when thinking about it for a while i think there is a far easier source of such ruby bearings, old HQ watches.
I probaly can find an old HQ watch at some second hand store and i bet they have some of those bearings inside somewhere.
I looked online for such ruby bearings and havent found a good source so old watches is probably what i try next.

Admagistr - 17-4-2024 at 11:03

Quote: Originally posted by Mateo_swe

I would like to get a small ruby vee-bearing for a very low friction DIY pulsemotor.
I wonder if making some synthetic ruby pieces and somehow grind a vee hole in the middle can be done.
Grinding such a vee-dent in a ruby would require some grind powder of equal or harder substance, that means ruby or diamond dust.
Or a tool already coated with such powder.
But when thinking about it for a while i think there is a far easier source of such ruby bearings, old HQ watches.
I probaly can find an old HQ watch at some second hand store and i bet they have some of those bearings inside somewhere.
I looked online for such ruby bearings and havent found a good source so old watches is probably what i try next.

A long time ago, when I was a kid, I ground miniature screwdrivers out of sewing needles on an electric grinder and took apart a few old watches and got a lot of ruby bearings;-)! A medical plaster for minor injuries, which I wrapped around the top of the needles, served as a handle for the screwdrivers. I don't have those ruby bearings anymore, they got lost somewhere in the many years that have passed. Otherwise, rubies can be ground with a carborundum SiC wheel.

Twospoons - 17-4-2024 at 15:28

You do know you can just buy them, mounted and unmounted, along with matching bearing pins? Making your own seems like a very difficult route to go down.
https://www.swissjewel.com/products/jewel-bearings/