Metal Coating using PLASMA (Part 1 - How it works)

this video was made possible by curiosity stream what does this playing card this flower and these microscope slides have in common well the answer is they were made using magic and by magic of course i mean engineering specifically they were made using this thing which does this and also this this was by far the most mechanically complex project i have ever attempted and required machining dozens of parts to as tight a tolerance as i could physically maintain we used literally every single tool in the new workshop and it really shows this thing needed machined metal 3d printed pieces laser-cut panels cold rolled metal all kinds of electronics a custom vacuum system and a litany of knobs and gauges so what is this thing and why am i so excited in short this device takes a sample of a material you're interested in and can coat almost anything in that material the layer is quite thin but even thin films of many materials can have amazing properties for example here's some glass i coded in different metals look at that mirror the clarity is amazing and the conductivity of these is fantastic but it can do more than just metals most things you can get in disk form can be thrown on here theoretically it could be ceramics like titanium nitride or even a piece of superconductor like ybco or something really weird like the special ceramic used for making glow powder and you can easily pattern how the material is laid down on the sample by simply masking off areas you don't want to be coated strictly speaking there's actually two different but very similar devices contained in this vacuum chamber the first simply uses heat to vaporize a material and spray it at the thing you're coating the other uses plasma to vaporize the surface of the material and then again spray it at whatever you're coding these two processes are called thermal evaporation deposition and magnetron sputtering and both are examples of thin film coding techniques without you necessarily knowing it you've interacted with thousands upon thousands of items which were coded in various materials using these devices almost every piece of electronics you've ever interacted with had components made with this machine but also every piece of metallized plastic many drill bits a huge number of optical components some of which you're wearing on your face are also made using these devices and that barely scratches the surface so let's get into it and see how this machine works and how we build it this project took almost two months to build and get working and that was with two people working full time i have almost 90 hours of footage of machining and assembling basically every piece of this thing and then tons of the testing and of course using it to make some fun stuff i can't hope to cram all of that into one video or it would be way too long so what i'm going to do is break it up into three pieces this one the first will focus on how the thing works going through all the systems and using it to make some fun stuff some of which i showed earlier the other two will either focus on all of the machining and how we physically made all the pieces and built the thing or diving into all the weird things we can make with this like possibly growing some diamond as those videos are released i'll add cards to them here and when i release the build video i also plan on releasing all the stl and cad drawings of all the pieces of the system so that anyone who wants to build their own can more easily do so we're just finishing up the markup and adjusting them to the dimensions that we actually ended up using rather than what was originally called for since we made some adjustments on the fly to make sure you don't miss either of those videos or any of the other ones coming up you'll definitely want to subscribe and ring that bell so you get notifications as soon as they come out okay enough sidetracks let's dive in first off the whole system can be broken down into subsystems the main ones being vacuum cooling gas and power but there are others like the sputtering system the thermal evaporation system and the motion controls to adjust the position of things in the chamber starting with probably the most important piece we have the vacuum system itself this removes as much gas from the chamber as possible which is crucial for both the magnetron and the thermal systems to work it's broken into two stages the first is a mechanical pump also called the roughing pump this brings the chamber down to a decent vacuum but not perfect it's more than enough to make some nice plasma but if you try and flow any gas into the chamber the pump is easily overwhelmed and the pressure will spike way too high so to assist it we have a second stage pump called a diffusion pump this one only works once the system is already at a pretty decent vacuum and actually attempting to turn it on at atmospheric pressure would make it just catch fire that's because diffusion pumps are super super weird they have no moving parts yet my little pump here claims it can pump about 150 liters per minute and it does that simply by boiling and condensing oil creatively basically the oil is heated at the bottom and the vapor is sent up through a column of inverted cone-shaped pipes the vapor is forced through gaps between the pipes in a downward direction and then hits the cooled wall of the pump this makes it condense and fall back into the boiler as this happens the refluxing oil drags any molecules of gas that happen to be floating by down to the bottom part of the pump where it's redirected out to the mechanical pump to be removed it's a bit of a nerve wracking process starting this thing the system is first brought down to around 100 to 200 millitorr before the diffusion pump can be turned on at this point there's this lag where nothing happens i've got the fan blowing at the cooling fins which is cooling the wall of the pump and the boiler is dumping lots of power into the oil but it takes a while for things to heat up and the oil to begin to vaporize as it heats up any water or contaminants in the oil evaporate or burn away this causes the pressure to spike a bit depending on how much junk has gotten into the pump if i haven't used it in a few days the spike will be larger than if it's say my second run of the day because more water vapor from the air will have been absorbed into the oil once the oil itself boils though the vapor rises up through the column and the downward jets form and suddenly the system will start pumping when this happens the pressure will rapidly drop to a nice low level for the weebs in the crowd the observant viewer may notice that this is the same sort of vacuum pump used in dr stone but i digress one side effect of the whole boiling oil thing is that some can end up in the chamber if it were to go too high so we made this water cooled baffle as a barrier between the two any oil vapor is forced by the geometry to touch the baffle which makes it instantly condensed before it ever makes it into the chamber speaking of which the chamber is this beautiful pyrex jar that i got super lucky and scored on ebay for a mere 300 new these things cost like two to four grand easily the vacuum removes gas but for at least the magnetron to work you do actually still need a small amount of unreactive gas in the chamber so that plasma will actually form this is controlled by two mass flow controllers on the control panel these each allow a small controlled amount of gas into the chamber and before you say so no they aren't upside down we took them apart and flipped the internal cartridge so that the control knob is on the vacuum side this gives a more accurate reading and makes the indicator ball less likely to bob around though it's definitely not perfect a digital mass flow controller would have been better but these certainly work there's two of them so that we can do what's called reactive sputtering normal sputtering just uses argon but if you mix in a tiny bit of reactive gas as the metal is flying at the thing you're coating it'll react mid-flight and turn into the reaction product of the metal and gas so if you add oxygen you get oxides nitrogen you get nitrides etc super useful as ceramics like that have a ton of amazing and useful properties even for the thermal evaporation which doesn't use plasma and just wants as low of a vacuum pressure as possible being able to flood the chamber with a bit of argon to flush out any junk is really useful most things are cleaned with acetone before they're coated but acetone is a little bit sort of sticky in this context so a flush of argon helps make it let go and get sucked out of the system moving on let's talk power the magnetron needs very high voltages while the evaporation needs very high current to handle both of these problems i'm using microwave oven transformers these are super super dangerous and definitely need to be treated with respect one of them is unmodified and can supply up to 2000 volts while the other has the secondary coil replaced with two turns of really thick gauge wire to supply several hundred amps both are controlled using a variac this is a special type of transformer which outputs zero to 120 volts based on where you turn the dial this way i can throttle the amount of power going into either system i can also control which one is on from the control panel thanks to a few switches what's great is that there's a lot of redundancy to actually send any power into the system you need to flip four switches in a variety of places so you really need to be sure that you want to do it one last thing to note is that the magnetron needs dc power to work so the transformer is connected to a string of diodes arranged in a special diamond pattern this is called a full bridge rectifier and takes two ac inputs and outputs a nice dc voltage all right one more system to go and we can actually fire this up and that is the coolant this one is nice and simple and basically just consists of a pump a reservoir tank and a peltier cooler silicon tubing sends water from the pump through the cooler to the various things that need cooling like the baffle and the magnetron but silicon tubing isn't actually good in a vacuum system so for the tubes that run from the base plate and into the sputter head we use teflon in fact any tubing which is exposed to the high vacuum side of the system is teflon we only use vinyl before the diffusion pump and to connect the gas tanks to the flow controllers with vacuum power gas and cooling sorted we can finally fire this baby up let's start simple with a little microscope slide if there's any dirt or oil on the slide it'll prevent the materials from sticking properly so the slides are cleaned well with acetone but this alone would be insufficient to clean these properly there's a whole protocol that uses really nasty chemicals to truly scrub any contaminants from the surface but for this demo i'm not going to bother one thing i have been experimenting with though is passing the slides over the flame of a little mini tesla coil the plasma is full of hot reactive oxygen and actually does a pretty good job of burning off contaminants and it also charges the slide which makes it a little more receptive to the coatings i realize this seems like a lot of work but it's only really necessary for glass other things like say flowers or 3d printed parts don't need it glass is just so unreactive that you need to treat it to make the layer stick really well don't get me wrong it'll coat just fine without all this effort but the metal adhesion layer is much stronger with this extra treatment with the bell jar off i first need to choose what materials i want to use i have all these different disks of different elements and materials for the magnetron and for the thermal system i have these molybdenum or tungsten boats and little pellets or pieces of material or filament wires since there's two ways to coat things there's two ways to load the machine for the thermal version i can either put a little piece of something in a boat and then suspend it between these electrodes or for things like titanium i can make a filament out of them and just heat that directly by wrapping a thin wire of them between the two electrodes for the magnetron i can take a disk from my collection and load it into the sputter head i do this by removing the ground shield which is this outer metal piece and exposing the core here there are four bolts and a retainer ring that hold the discs onto the sputter head i just remove one and loosen the others and i can slide one disc out and replace it with a new one since i expect to do a lot of bits of glass and things about the size of a standard microscope slide i made the sample holder have a slot that will fit that while leaving as much of the glass or slide exposed as possible microscope slides just fit right in and once they're in place the system can be closed up and pressurized once everything is running and ready to go i'll either give it a puff of argon if i'm just going to be using the thermal system or a flow of argon when i go to use the magnetron starting the magnetron is a little bit like starting a really finicky engine it usually takes a minute to balance the voltage gas flow and the total amount of power going into the system to get a nice stable plasma too little gas or too little voltage and the plasma can't form too much of either and a ton of current will flow and the system will spray too much material too quickly or you'll just roast everything that you're trying to coat because the plasma gets too hot this is why i always start with the shutter closed once i've got it balanced with a nice stable plasma then i'll open the shutter also this is really helpful since the disc will develop a layer of oxides and other junk on them while they're in storage so this is a way to shed all of that junk before exposing the thing you're actually coating so that only clean good material will actually be used using the thermal system is much simpler if you go too hot you'll melt the boat and too low and nothing will happen so you just balance where you start to see vapor flying and just kind of stop there or in my case i know that about 80 on the variac is enough for at least aluminum so i just start there and then fine tune it as needed for a more even coating i can use the motion controls to move the slide back and forth over the plasma plume eventually i plan on sticking a stepper motor here so that it can just do this with the push of a button i'm not going to get too much into how the magnetron itself works i'll get more into that in the doing weird stuff with the magnetron video but here's the quick version there are two electrodes the central one and the outer shell when you put voltage across these two plasma will form between them inside the central electrode is a very strong ring magnet which is sitting on a turned steel pole piece if you look at a cross section of the magnetic field this forms you see it makes a sort of m shape this will concentrate the plasma into two spots at the bottom of the m translated back to 3d this forms a super hot ring of plasma on the surface of the electrode the reason for the cooling lines that i showed earlier is to remove any excess heat from the core of the magnetron partly so that it doesn't melt but also so that the magnets don't get too hot since that would ruin them it also makes it so only the very surface of the metal is hot and the rest is much colder as argon atoms in that super hot plasma hit the inner electrode sort of like billiard balls it knocks atoms free from the surface of the electrode those atoms go flying up at the thing you're coating and just like steam condensing on a cold window the atoms condense and form a layer on the thing you're coating you can actually see the ring in some shots but better than that you can see this pattern left on the discs there's a thin line where the most metal was being ablated off this is actually called a racetrack pattern alright now we know how this works how to load it and how to run it let's make some stuff in a project i did about two years ago which i'll talk about a little bit later i made these microelectrode arrays meant for growing human neurons on them as i was filming this i found the vinyl cutter files and figured that i should probably make a new one using the new system just like last time i cut out some vinyl masked off the areas i didn't want to be coated and then threw it into the machine this time i'm using pure silver instead of sterling since i don't have to use a bread plate anymore same as last time i did a layer of silver and then a layer of gold i absolutely love that you can literally see the layer go down it feels like magic every time i watch it happen to finish it off once it's all coated the vinyl is removed to reveal the final array this is probably about as small as the electrodes can be made using vinyl like this so for the next neuron attempt i'll be using proper photoresist another thing to note is that switching materials is a little bit slow before the machine can be opened the diffusion pump needs to cool down enough so that it doesn't just ignite when you let air back in so each layer took about an hour from first starting the system to being able to open it again the custom playing card was made in basically the same way i just did gold on both sides and then had the fun task of peeling off all the little vinyl stickers to slowly reveal the pattern but when it was done the final card looks absolutely stunning if the machine didn't take so long to heat up and cool down i would do a whole deck like this but as it stands this one card took about three hours so that's a negatory on that one moving on let's coat something a little bit more fun this is a real crow's skull that i got on ebay i love the contrast of the black beak and the rest of the bone so i definitely want to keep that i masked off the beak using some aluminum tape and then loaded the skull onto the holder with a bit of soft aluminum wire it's hard to convey how delicate this thing is so i took a lot of care not to break it i went with silver as the coating material since silver is one of the easiest materials to sweater and frankly i just thought it would look the best in contrast to the beak once things were pressurized the coating process went super smoothly but can we just stop for a second and appreciate how freaking amazing this looks this is what i mean by this machine is basically magic if i were to show this to someone and say the 1500s i would 100 be burned at the stake for being a witch after it was coated i removed the mask and it worked perfectly so that the whole thing came out looking amazing this is going to be a great addition to my collection all right skulls and glass are both pretty solid let's try a real challenge you've been seeing these for the whole video so i can finally tell you how i made them i started with fresh flowers but they need to be dry to go into the machine so to do that quickly and preserve their shape i stuck a tray full of them into my new freeze dryer this worked great though it did have the little side effect of making some of the petals turn into mini balloons also loading them into the machine was really tricky both because they're super delicate but also i can't have any of the pedals falling off and falling into the diffusion pump this is definitely not something your pi would ever let you do but after that the rest of the process was super easy i just set the power nice and low so that i didn't accidentally carbonize the petals and just let them slowly coat and metal i tried a few different ones but silver and gold worked the best by far something interesting to note is that the metal is mostly deposited in line of sight with the plasma if i turn the flower around you can see the back is basically totally uncoated it's almost like shining a light made of metal at it and the uncoated bit is the shadow it makes for a really weird and interesting effect and finally just to really drive the point home that this can coat basically anything here's two quick examples the first is actually the vacuum grease around the bottom of the bell jar as i was testing the thermal system i noticed that the grease itself actually got coated so for those asking if a liquid could be coated technically yes last but not least some 3d printed parts these are some cortical stacks from altered carbon that we printed on our new resin printers and they coated really well i did one side in titanium to get this cool blue effect and one side and aluminum for a simple metallic look i think this is going to be awesome for any model making or cosplay type projects where i want plastic to look like real metal i think this pretty conclusively demonstrates that almost anything can be coated with this and now that you've seen what this can do let's talk about why i built this first is that the key to science is to hold a good grudge this is not the first time i've built either of these devices far from it i've built several iterations of these over the years and there's videos on both already on this channel which i've linked to below the first one i made that was actually functional was built about two years ago it was made out of a dollar store jar stuff you could get at a good hardware store one or two pieces i turned on a metal lathe and a few pieces i got off ebay so the complexity of this new machine wasn't technically necessary for this to work but it had a lot of advantages which we covered earlier that last system was actually pretty great for what it was it worked really well coded a large variety of objects and all sorts of different materials and didn't cost very much at all and could largely be made out of stuff you can find at a good hardware store it even managed to coat this dead hornet in silver which looked awesome but also shows that even the tiniest details will get coded using this technique but that aside the machine couldn't do a few things first highly reactive metals like titanium and aluminum that system was a little bit leaky so oxygen would get into the chamber and react with the metals before they vaporized properly a layer of oxide would keep forming on the surface and prevent any metal from actually flying also it was made out of little jars so there was very little room to put stuff in there and no way to manipulate the objects as they're being coated without opening the thing up and it needed large pieces of metal you wanted to use because i was using that as the bottom seal of the machine if you wanted to do say a copper coating you need a big plate of copper for the whole jar to sit on as i've demonstrated the new system addressed all of that so my grudge is finally settled and i can now move on to thinking about all the amazing things this can be used for instead of ways to finally make it work the next reason i wanted this is because i want to make some very specific things and this is the only way to do it specifically i want to revisit the project i mentioned earlier where i grow human neurons on a microelectrode array and connect it up to a computer there's obviously a lot you can do with that and i plan on doing a lot with that but if i can make it work the first thing i'm going to do is make it pass me some butter that project is pretty wild last time i made the same sort of electrode array that i showed earlier out of silver and gold on microscope slide then took it to my friend's lab where we tried growing actual human neurons on it but the arrays didn't hold up very well and the electrodes were too large to be really useful now with this new system i can make proper arrays out of better materials and use larger base plates so they're much easier to work with so think of this as step 1 in giving that project a proper attempt i've linked the original video below if you're interested finally the last reason i wanted this machine is because this is the groundwork for two other projects that both require a vacuum system like this the first one being an ion engine and the second being a nuclear fusion reactor a lot of pieces of this system are going to be reused in those projects or at the very least act as a prototype for various pieces and building this project got us used to our new machine tools so we have a better idea of what materials we can actually use and how big or small different features can be based on our abilities with the machines in the next video in this series we'll get into the more exotic materials as well as reactive sputtering so i can really show off what this thing can do and hopefully by then i'll have some photoresist ready and we can make some prototypes of the new arrays this project was a lot of fun and super hard but it was only possible thanks to a few people the first is ben from applied science we based this system off of his and he was super gracious with his time and advice and helped us troubleshoot our system so huge thanks to him i've got links to his amazing videos below the next was my fantastic friend and crew member frank he is a super talented engineer and not only was he responsible for counting almost every piece of this thing he was beside me for the whole project like literally i did all the lathe work and he did all the mill work so we could just pass parts back and forth quickly or hash out tolerances and dimensions on the fly if needed you'll be seeing him and his work a lot more in the future so big thanks to him and keep an eye out for those videos soon and of course we need to thank the fantastic sponsor of this video curiositystream curiositystream is an amazing streaming service that is hosted thousands of fantastic non-fiction documentaries and tv shows on all sorts of different topics from history and science to travel food technology and more and they have 35 curated collections of award-winning titles hand-picked by experts one i really enjoyed was called cities of the sea i am a massive bionerd obviously and i'm also a scuba diver so learning more about marine life is always my jam but they really have something for everyone and if you use the coupon code thought emporium you get a whole year's subscription for only 14.99 so head to curiositystream.com thought emporium to enjoy thousands of fantastic titles finally i of course need to thank my amazing patrons channel members and supporters on kofi this project was really expensive but your amazing support is what lets me build crazy things like this so thank you so much you are all amazing and if you'd like to help me keep making crazy projects like this there's some links below and as always you know what to do hit that like button subscribe and of course ring the bell to see when i post new videos and be sure to follow me on twitter and instagram to see updates on these projects long before they end up in videos that's all for now and i'll see you next time

Channel: The Thought Emporium

Views: 423,871

Keywords: physics, vacuum, plasma, high voltage, metal, art, coating, gold, silver, flowers, craft, educational, science, engineer, engineering, applied science, chamber, magnetron, sputtering, evaporation, thin film, thin film deposition

Id: XHQxghdj2ks

Length: 23min 45sec (1425 seconds)

Published: Tue Sep 21 2021