Accelerating Gallium Ions to 0.056% light speed
Вставка
- Опубліковано 9 лют 2025
- I made a liquid metal field emission electric thruster. It nearly broke me 😅
Watch this ad-free on Nebula: nebula.tv/vide...
🥰 CONSIDER SUBSCRIBING 🥰
🔬Patreon if that's your jam: / breakingtaps
📢Twitter: / breakingtaps
🦋Bluesky: bsky.app/profi...
🐘Mastodon: universeodon.c...
📷Instagram: / breakingtaps
💻Discord: / discord
Applied Ion Systems: appliedionsyst... ( / applied_ion )
So apparently Gallium/Indium is called "EGaln". I knew it wasn't technically galinstan since it doesn't have the tin, but couldn't find a proper name for it. Cheers to everyone for the note!
14:36 You made an Electric Harmonica
@@BreakingTaps I always thought galinstan was for both versions with and without tin. Thanks for the tip. You always learn sometimes useful here 👍
I have a design using mercury , no exhaust gases or molecules , your design spends the costly gallium indium , No need for fuel, just electricity and mercury . You can message me if you want to design it
@@Blue.star1 See I always thought that mercury would be best for such a thing but meh idk much myself
@@Jay-n2j5p vaporised mercury is extremely toxic. Galium and indium are not toxic. 👍
Plasmas can have negative resistance characteristics, which causes instability, which is why current limiting is needed for a stable discharge.
Aha! That makes a lot of sense, cheers for the info! Starting to think I should learn me some more EE, would have saved like a whole month of pain haha 😅 Really want to have another go at the slot emitters, think they will perform well once resistance is dialed in 🤞
Wtf does negative resistance even like
Upping the voltage and lowering the amps??
@@RENO_K energy from the plasma discharging into the circuit I think
As the voltage/current increases, the gas becomes more ionised so the resistance of the plasma actually decreases. Current increases more than it would have done with a stable resistance.
@@RENO_K(someone who understands this stuff better than me, feel free to correct anything I say here)
From the wikipedia article "Plasma channel": "Unlike a normal electrical conductor, the resistance (and voltage drop) across an unconfined plasma channel decreases with increasing current flow, a property called negative resistance. As a result, an electric spark that initially required a very high voltage to initiate avalanche breakdown within the insulating gas will rapidly evolve into a hot, low-voltage electric arc if the electrical power source can continue to deliver sufficient power to the arc."
The wikipedia page "Electric arc" has this to say about it: "An electric arc has a non-linear relationship between current and voltage. Once the arc is established (either by progression from a glow discharge[12] or by momentarily touching the electrodes then separating them), increased current results in a lower voltage between the arc terminals. This negative resistance effect requires that some positive form of impedance (as an electrical ballast) be placed in the circuit to maintain a stable arc."
The resistance of a purely ohmic resistor comes from how easy it is for electrons to move between atoms of a material, aka electron mobility. Initially before an arc is formed, the resistance of the circuit is very high. Pure vacuum technically is a perfect insulator, having infinite resistance, as there is no material through which electricity can conduct, but we can ignore that for now because neither space nor Breaking Taps' vacuum chamber are anywhere close to that. And even if he could reach perfect vacuum, field electron emission can cause free electrons to be generated allowing for the creation of an arc. Anyway, once the arc is formed, resistive heating rapidly heats the plasma, which effectively lowers its resistance. This increase in plasma temperature and lowering of resistance draws more and more current until it reaches the maximum current that the supply can provide. This is the negative resistance in effect.
And a week later all aluminum in shop crumbles
And every piece of steel is rusted from the HCL fumes 💀
I hadn't considered that part of vaporized gallium, sounds terrifying. I wonder how much Gallium you would need to vaporize to cause any serious harm.
@@lbgstzockt8493 I don't know but I believe it acts as a catalyst and not consumed, or at least very little.That means it could keep destroying anything it touches. I vapor form you could contaminate an entire room with a milligram.
@@JoeSmith-cy9wjthe ice-9 of metals.
This is why I refuse to try any projects that require it, I'm not careful to not accidentally get it somewhere it shouldn't be.
of course you did
Hahaha. Literally said that out loud when I saw the video title.
@@11macedonian same here ahahaha
XD
Makes sense... I mean, what else would you do if you had all this machining capability, no impulse control, stubbornness well north of the regular human scale, and an energy level that makes Red Bull look like de-ionized water.
Lord Helmet, fancy seeing you here.
I appreciated you writing that vacuum didn't work either. It's enjoyable to be taken serious as an audience 😉
And it was exactly what I thought too. lol
@@UberAlphaSirus I have 0 knowledge on any of this. But say you put a load of the liquid metal on top of the glass in a chamber. Could you then pressurize it and maybe something happens? Maybe does something maybe doesnt. I for sure have no idea
@@PwnsauceDzG the oxides are the really annoying part
Likely the sintered glass pores are mostly closed cells at the surface?
@@pseudolullus Ah okay. I completely missunderstood the issue. I thought the issue was more getting it to 'paint' on the liquid metal. But yeah i realise now you explained. Thanks!
BTW for high-voltages, old-school solid carbon resistors work well - not really made nowadays but can be found as old stock. I used them on my 1MV Marx generator and they are pretty much indesctructable
The problem with most film resistors is they are typically a coated cylinder with a spiral slot cut in it, so the creepage distance isn't very long. The thin film is also rather fragile and easily damaged by spark discharges.
Filing that tip away for future projects! Will set up some saved searches to see if I can scrounge up a supply. Was quite irritating to lose a resistor and have to cycle the chamber. Even the 10kv-rated resistors didn't seem to fare much better, a few of them popped. Should have been well under their power rating so I dunno. Maybe they overheated faster due to vacuum?
@@BreakingTaps even within their nominal power rating, high currents from very brief surges can damage the thin film as it has very little thermal mass.
@@mikeselectricstuff How about an inductor to reduce di/dt at the start of the arc? A high-voltage transformer might be a suitable inductor. If it has a flyback core the core would need to be disassembled and reassembled without it's airgap to increase the inductance.
@@maeanderdevan inductor would be ideal as it doesn't suffer quite the same losses as a purely resistive element. It also induces a varying voltage to try maintain a constant current. Finally, if you wound the inductor right, you could use it's magnetic field to focus the plasma like in a magnetron
Those kapton graphene circuits wouldn't be a horrible way to make these either
Awesome! that jet was fantastic
Video of one under a SEM when?
hell yeah. also nice to see your here :)
i love watching someone else suffer through a project for a change. makes me feel a bit more sane. great job on this one.
this channel is so great, sometimes we are looking close at metal, sometimes we are watching a cool milling process, sometimes we are making an ion engine. Love your work.
Audience: "So what are we getting next?"
Breaking Taps: "Hell if I know.. but it's gonna be COOL!"
I have to say I laughed out load and said “vacuum?” and then you flashed the black screen saying no vacuum did t work, lol 😂
It was almost too perfect
Zach is slowly becoming the Dude. :)
Another great vid! Thank you.
I do enjoy a good white russian 😅
I bet he's got a rug that really ties the lab together
What is the Dude?
@@nicolasolton Jeff Bridges' character from a movie called The Big Lebowski
@@nicolasolton " But sometimes, there's a man. And I'm talkin' about the Dude here. Sometimes, there's a man, well, he's the man for his time and place. He fits right in there. And that's the Dude, in Los Angeles. And even if he's a lazy man - and the Dude was most certainly that. Quite possibly the laziest in Los Angeles County, which would place him high in the runnin' for laziest worldwide. But sometimes there's a man, sometimes, there's a man. Aw. I lost my train of thought here..."
The secret to making Liquid metals like gallium or indium is ultrasonic energy. I had terrible results indium bonding silicon sputter targets. Using an ultrasonic horn to apply the liquid indium to the silicon gave amazing results. Ultrasonic energy nicely disrupts the oxide film.
Good advice, This technique was used to apply solder to glass by a youtuber some years back, one of the popular physics experiment channels.
I work at a company that makes gallium ion sources for ion microscopes, and I've got a couple of those sources you showed a picture of on my desk as desk ornaments! It's sooo cool to see someone being able to build something like this by themselves! Also, even in manufacturing, LMIS sources still arc and fail all the time and are often quite unstable.
I would imagine that the reason your 1M current limiting resistors were getting burnt up was that when the arcs occur, you're essentially putting the full 10kV across the 1M resistor, causing about 100W of of power going into the little 2-3W resistor. Increasing the resistance to 50-100M would solve this problem (and it sounds like that's what you ended up doing!)
Aha, yeah that makes sense! Seems I accidentally stumbled into the right solution haha 😄
Cool! I have accidentally made Taylor Cones with a van de graaff generator and hot wax. It makes a HUGE mess as the charged stream of wax is propelled at crazy high speed and is attracted to every surface in the room.
😂
This sounds interesting
How swift were those Taylor cones? I knew you were trouble when you turned on the generator.
@5:20 just a thought about the issue of breaking down the oxide film that immediately forms. This sounds like a very similar issue to what happens with aluminum. In welding we use AC voltage and adjust the percentage time balance between the electrode positive and negative. We want electronegative most of the time because it puts more heat into the part, but we use around 30% electropositive which creates a cleaning effect on the aluminum and breaks down the oxide. I’m wondering if you could do something similar here with the gallium. Obviously you would need to bond it to the negative side of the circuit initially, but maybe you could create an ignition circuit that is similar to a pilot arc that starts with cleaning the metal and gradually transitions entirely to DC as an active surface is achieved. If you’ve already thought of this to my apologies for saying the obvious. (I work in product research, design, and development and by consequence also end up doing a lot of fabrication work).
Get hold of a guy named Jerg Jergensen in Santa Barbara, CA. Ages ago he used to make ion sources for Hughes and he knows a lot about your predicament. :-)
Pronounced "Yerg".👍👋
Super cool space toy 😂. I know how long projects can break a person, especially vacuum systems. Super awesome and totally worth it. Stay awesome 👍
You never cease to impress and amaze me.
Back in the 80's as an undergraduate studying membrane biophysics, we had a lecturer who was a pioneer of patch clamping single ion channels using very fine micropipettes. It was (and still is) incredible to even imagine.
Oh that's awesome! Yeah patch pipettes just blow my brain whenever I think about it. Worked in a lab where one of the grad students was doing a lot of electrophysiology. Always amazed to see her clamp those tiny pipettes onto neurons and taking measurements. Just bonkers stuff!
Very cool video, it does sound like most of your issues are from the alloy though haha. I’m pretty glad I went the ionic liquid route for my masters and phd. Crazy coincidence though, I was randomly flipping through your older videos and saw this one just posted. You don’t see a lot of amateur electrospray work, nice job!
Congratulations on your success with the project.
I am incredibly jealous of you and others on YT that can work on all these interesting projects.
I am a mass spectrometrist and when doing nanoLC keeping the Taylor cone stable is a major problem and yeah the current is a huge part of this. If you go too low you will get it destabilized and form a droplet that is hard to get to reform a cone without intervention.
For our applications the cone will flitter if you go much over a microamp and it’s a careful balance of changing emitter position, voltage, and solvent mixture to get stable ionization.
Thanks!
I used to work with Caesium Ion sources a lot and they were twitchy but did squirt out quite a bit of current. They were made by having a tube blocked of at one end by a 'Frit' these are like a sintered metal but usually glass like anyhow, behind the little disc 'Frit' sat the caesium and by electrically heating the metal would diffuse through the 'Frit' and emit streams of Ion's ! juicy and ripe ready for extraction and used in various ways. I guess the Frit provided a clean reservoir of material that was far more stable than dipping a needle into a puddle ! just sayin LOL! great work and good luck !
Why don't rockets get sad? They always have a blast.
I like this one lol _😂😂😂😂😂_
I love all the incredible precision machining, microscopy, metrology ending with, basically just dipping the electrode into the metal like a Dunkaroos cookie into frosting. 😅
Thanks!
Thank you! ❤
man's gonna walk on the moon before he finishes that bird.
I wish i had a teacher like you when i was younger. You're easy enough to listen to. I've never seen this channel before. I love learning new things like this. Thanks.
scientists are bad at naming things ALL the times. very cool video so far, love it
One thing I like about your channel is that it shows how straightforward, easy, and simple experimental science is. ;)
Two thoughts: first, instead of just current-limiting resistors, how about pairing those with current-limiting inductors? Nothing prevents sudden jumps in current like inductors. Second, you could try pre-saturating (or supersaturating) gallium with hydrogen, so that instead of forming surface oxides, it will just produce surface water, which will evaporate in vacuum. One method of hydrogen supersaturation in other fields was using the metal as the cathode in a water electrolysis cell, where they could achieve hydrogen equivalent pressures in the mega-bar range; though those were with solid metals, so with liquid gallium you might just end up with gallium soda... but maybe just a bit of ambient-pressure hydrogen doping would be sufficient to reduce oxide skin formation.
This sounds fantastic. Imagine how clever one must feel to have invented that technique of saturation with hydrogen by splitting water!
@@lumotroph I think it was "discovered" rather than invented. Noble metal electrodes in common use for electrode materials, and also have a fairly high ability to incorporate hydrogen into their lattices. When some electrochemist (probably working on fuel cells) measured the amount of hydrogen that was in their platinum electrode, they noticed it was equivalent to compressing gaseous hydrogen into the metal at many thousand times of atmospheric pressure. Presumably before the hydrogen bubbles nucleate on the cathode, or when they're at ~nanometer-diameters, the surface tension produces those equivalent pressures.
Man, that was such a bear of a project. I admire your tenacity and resolve.
Always wanted to see something like this! Electric propulsion always intrigued me, so this is so awesome
To prevent the vibration of the slot design, try introducing a bend on either edge, parallel to the slot. Kind of like cutting a slit down the center of a piece of c channel.
you should call up the slow-mo guys to film some of this stuff!
Your videos consistently light a fire in my chest. One of curiosity; a burning desire to entrenche myself into the sciences.
Though my upbringing didn't allow for me to pursue a higher education, over the last 2 years I've got my diploma, (veledictorian) and will be going for mechanical engineering starting this next year, despite going on 30 years old.
Channels like yours and Veritasium consistently re-ignite and bring me back into alignment with my passions that can fade from the busy nature of day to day life. For that, I thank you.
Congrats, that's awesome! Goodluck with your studies! ❤️
About that wetting process, maybe try using ultra sounds.
hey some EE tips, the resistors arent dying because of voltage, the voltage rating on a resistor is to make sure it doesnt arc across the windings. Your resistor was burning up because it was too low wattage. Get a 10W sized resistor.
Hmm, that makes sense. Although I did have some die even when there wasn't any activity (no arcs, no emission, just sitting there ramping voltage). I don't think it was emitting an appreciable amount of electrons across the gap so shouldn't have been much current, right? Maybe it was still enough leakage current across the windings to exceed the wattage rating?
Really appreciate sharing your process and frustrations through this. Totally don't care about ion thrusters, but some of the material and morphological details about how to wrangle the surface interface properties of gallons tan is solid gold.
DIY FIB sounds like a cool project.
What about a gas plasma ion source rather than a LMIS? It would avoid all of the issues around gallium and you could just feed it with neon or xenon or argon or something. Plus PFIBs are cool.
I watch your videos thinking back to university, and wishing the profs back then were as entertaining and intriguing as your videos are. Good stuff.
Super cool project! Thank you for sharing!
A spectacular piece of home engineering as always! Love this stuff!
Totally cool, and also that pipette puller is awesome.
A bit of advice at around 22:15. I'm taking a couple of welding courses in college at the moment, one of which being Tungsten Inert Gas welding. A tungsten electrode is ground to a point (with a flat tip, but that's not important here) which the gas flows down to protect the arc generated from the oxygen in the atmosphere. The specific grinding method used is very important; if you grind the electrode so that the grooves are perpendicular to the electrode length, like the way you showed us, the gas will get sprayed around the plate and won't cover the arc. If you grind the tip so that the grooves are parallel, however, the gas gets directed straight into the arc.
Yep, however inthis case there is no gas in the vacuum and the surface roughness is in some way effecting the capillary forces. I am not sure which way would be better but testing with both would provide some information I expect. My gut tells me that the axial (parallel to tungsten would work better but that is nothing more than a hunch.
IT'S SO CUTE!
Sir, I worked a wafer fab for a brief bit as a repair & maintenance intern in thin films & you remain an inspiration.
Same reaction I had to seeing oxygen have a color for the first time. (:
I replicated the LIGraphene after I saw your video on it because you made it seem so approachable. Tried to make a finger capacitor on 6" wide roll of polyamide but couldn't get the power supply to power it appropriately. Never figured out the root cause & it drove me up the wall.
Just know your work is inspiring & I'm grateful to see your latest project.
Your projects are always cool, but this one is extra awesome. I love it. Super amazing job.
What an engineering tour de force (ahem, I will see myself out ;-) Thanks for sharing.
At the beginning i was thinking this sounded a lot like what Applied Ion Systems was working on. I didn't know you made those emitters for him! Been following AIS for years so super cool to se more DIY ion stuff.
His work is so great! I was so excited when he asked me to machine the emitters, was super cool to see the test fires shots and all that plasma 🤩
👍👍
Dude i can't imagine the work and effort you put into this video, i love your content! Congrats and thank you!
TIL building DIY satellite ion thruster is actually rocket science😁
Seriously, kudos that is amazing.
You're definitely one of my heroes. Thank you for doing what you do.
When i experimented with krytron tubes they would generate a very well collimated ion beam from the anode through the trigger cup electrode.. Im guessing that if you add a third electrode a lot of the issues would go away. The gallium at hv, the second electrode at a few hundred volts less through a high resistance, and the final electrode at ground or even negative HV. ❤
Allot of your stuff is way over my head but I can’t stop watching. So cool!
Dude, such a cool video. I was so engaged, when you finally got that plasma spike 20:15 I nearly jumped out my chair. Great work, sir.
Just decrease the current limiting resistor value by putting some in parallel. This will increase output to a point. Messing with aperture is much more work than swapping resistor values. Also look at HVR inc, they make really good hv resistors in small packages for relatively cheap. This is fantastic stuff doing it all yourself! There is a lot to keep track of in projects like this!
Thanks for the video. It's nice learning new stuff. Take care! 🙏💪
Chemical sharpening of tungsten is very easy, you just heat the tungsten and then stick it in oxidizer. It's sold as "Chem Sharp". The sander seems just as easy, though like in TIG welding I'd do it so that the scratches run down the cone, not around it.
Aha interesting, I'll look around for that. All I could find were academic paper references to really horrible acid combinations, didn't realize there was an easy OTC solution available! Noted on grinding direction, will keep that in mind
Thanks for doing this. One more item off my list.
Amazing that a ballast resister was all that was needed. I was thinking it would take super fast opamps to clamp the current after the oxide layer was broken. I just didn't know if any were that fast.
Very cool! Thanks for documenting your journey with this project. It does sound like it would have been absolutely maddening though
I imagine that using ultrasonic might knock the oxide layer around enough to improve wetting. Not sure if you tried it, I sure have no experience here, but it is an approach I would try. Ultrasonic cavitation might also be a good way to 'rough up' the surface of the cone.
Please don't push yourself too much. Thank you for your videos.
9:35 If the bridge is arcing because it's the closest part to the capillary ring then just 3d print a small press die to bend them outwards, sort of like the suspension on a speaker cone.
And if you machine the bridge in a slight twist, then reforming it will not stretch the metal but will just rotate the center section a little so there will be less chances to break when fabricating.
For getting the gallium to wet, you might try acid to break up the oxide and place it in a vacuum chamber prior to starting the wetting process, to avoid trapping air.
Although I feel like a total dumb@$$ watching this, you explain it so well that I feel like I am having beers talking about cars or football. Your channel really has no equal on YT. Freaking love this stuff!
I’ve done a lot of work with electrospray for biopharma applications. So, pressure-driven capillaries at atmospheric pressure. Totally different from liquid metal in a vacuum, but maybe some of the knowledge is transferable.
I notice that your ground plane is flat. I would recommend trying a shallow conical ground plane, with a 85-95 degree half angle. That helps intensify the electric field at the tip while reducing the tendency for the Taylor cone to shoot off horizontally.
The other possible issue I see with your glass capillaries is high voltage connection to the gallium. You can sputter coat the outside and tip of a pulled glass capillary to form a conductive layer, maintaining continuity from your high voltage connection to the gallium itself. Otherwise the glass insulates the gallium from the high voltage supply.
Really cool video! I love seeing this stuff tried in the home lab. Good luck with this one, I’ve confronted my own demons with electrospray and it isn’t fun.
I would love to see a FIB basics video. And I absolutely understand burnout from a project. Take as long as you need. I love your videos.
For negative resistivity behavior, the most suitable hv power supply is like tig welder, hf hv for cleaning and high current for the after clean running
One observation, one that I've long developed as a software developer: always start with the simplest form of what you're trying to do. Use lots of diagnostics, so you can observe what is happening. Then, once you have the basic version working, build on that. You didn't start with this ethos. Changing to it made all the difference.
BT contribution to science education is immense, thank you for all your 2024 output. A diversion from the Bird project. UA-cam video providers are people that finish projects, if other viewers are like me we have boxes of unfinished projects inspired by your projects.And we love it.
Nice fruster! Thank you for sharing!
god I love the lighting in your workshop recording spot
"I just wanted to do some advanced degree thesis, multi-discipline rocket science as a side project from my other mechatronic engineering hobby." Don't beat yourself up. We love your videos. These aren't solo projects. Burnout is real. I'm sure UA-cam will be fine with you releasing shorts instead of full long format videos. If your engagement drops because you are a human and have personal things to deal with, those of us that like your content are still going to come back to it.
Thank you Sir, You taught me some stuff. Two Thumbs Up!
yay! your videos are always my favorite these days
When you mentioned the oxide layer problem, I was expecting this video to involve some exotic and horrifying acids. Pleasantly surprised to see the solution wound up being a blow torch lol
please make the full thruster I love seeing this
the perseverance is inspiring. thanks
you can use tungsten electrode sharpners for get a finer tip. You can find those in any welding store or bought online. Also, if you aren't already using them they sell presharpened tungsten electrodes for welding, although idk how fine the tips would be it could be worth it to check on the electron microscope
I was thinking about this a couple weeks ago. Nice to see it’s being actually tested! 😊🎉😮
2:02 Forbidden toothbrush.
Really amazing work. I think if you made the non pointy end curved and smooth, I think you could get less arcing as well.
Absolutely amazing work! 🤗
Dude, of all my favourited UA-camrs, you’re my favourite. 👍👍
I liked the part with the sparks!
I found it neat about how it led to microscopes.
I was just wondering how to build one of these, youtube never disappoints
Gotchu fam
Also, with large enough resistor values, you might accidentally create an RC oscillator, where the electrospray current collapses the voltage on the conductors on that side of the high-megohm resistor ...causing the emission to temporarily halt, until the resistor charges up those conductors and the Taylor-cone turns on again. In other words, to limit the capacitance, the high-meg resistor needs to be located less than 1cm distance from the emitter-tip. Or, to produce slow pulsing on purpose, connect a grounded capacitor to the tip. (Measure the received droplets-current on a scope, to check if it starts pulsing. When pulsing, and also being swept using deflection-plates, it starts drawing dotted-lines on surfaces where the droplet-streams are impacting!
Here in our department at the UW, everything is 1KV electrospray devices, not liquid metals.
Oh wow, I hadn't even thought of that aspect! Was such a simple "circuit" it didn't even cross my mind, but I guess things get weird with such large voltages. Yeah the resistor was close, but not that close. A few centimeters away usually. Cheers for the info and tips, will do some more reading!
@@BreakingTaps I had it drawing dotted lines, but it depended strongly on the microamps. Try it all in air at 1ATM. The droplets concentrate into very thin laminar streams, called the "gas focusing" phenomonon seen in particle accelerators. They're moving at perhaps 30KPH in air, and will pop soap-bubbles. I never did try some faraday-cup measurements.
Build a beam-break burglar alarm, where the beam is microdroplets rather than light? Heh. Also, it occurred to me to try alcohol droplets in air, then aim it at a rainbow oil-film on water. Then scan the beam XY, to make a "cathode ray oscilloscope" at ambient pressure.
That, at the risk of understatement, is pretty neat.
Pretty sure galinstan has tin in it, along with the gallium and indium
came here for the stan
I really enjoyed this video! Fascinating physics!
I also have been suffering thrugh a project only half working but yours is so much cooler T_T
Magnetic field to guide the plasma and reduce arcing? 2) Haven't commercial plasma-cutters solved most of these feed/arc issues? At a whole lot more interesting power levels? 3) Replace plasma-cutter's gas with guiding magnetic field? 4) Heaviest/Densest safe-ish fuel might seem to be Bismuth? Perhaps not as convenient as InGa though?
For the resistors you can always have more in series to increase it's voltage rating and more in paralell to increase amps and less watt per resistor
Most "HV" resistors in production units is just 2 or 10 regular resistor's in series 😅👍🏼
I tried to patent a hydrogen thruster in 1976. It worked. It was basted on the capillary action with a trinary tube without the glass!
Galium focused ion beams come to mind 😊
Try electroplating the galeum onto the needle(add a metal coating or use a metal needle). Then heat it up a little and it will work better
Resistors with extremely high values are very difficult to make, especially for high voltages.
There's one surprisingly simple way to make them though:
A plastic tube with distilled water, plugged with electrodes on each side.
The resistance depends on the length and thickness of the tube, and the purity of the water.
I was making mine with dilute india ink, drawing long thin stripes on ppma acrylic. First put down dabs of silver ink, for the end connections to copper foil. But that was in air, and might behave differently in hard vacuum.