How does a "first surface" mirror work? (2^15 sub special!)

oh my god holy crap that worked there's a mirror that's phenomenal it's not a very high quality mirror but it's a mirror [Music] hopefully this is going to be the first video on a longer project because there are other things that i want to make that are going to require 3d printed mirrors that said today i'm just going to be showing you the technique these are some first tries from like a month ago i'm going to be talking about what a first surface mirror is how i've been able to make one out of plastic and as much of the chemistry as i can understand i am a material scientist not a chemist when light hits an object it can either go straight through that object it can be absorbed and end up heating that object or it can reflect and bounce away if you shined light on a surface and then zoomed way in you would see that the oscillating electric field that is light is actually causing electrons to accelerate back and forth near the surface of the material in a metal that's electrically conductive so that this motion of electrons can happen pretty easily these shaking electrons do a great job of re-radiating that energy back out into the air as reflected light they only hold on to that energy for a split second in order to reflect light specularly like a mirror you also need a really really flat surface how flat well that depends on the light arecibo's giant radar dish was able to very accurately reflect 12.6 centimeter light about the wavelength of old wifi signals with a mirror made from perforated aluminum sheets incidentally this is almost exactly the same perforated metal design that you have right now covering your microwave's window it doesn't look like a mirror but it's more than good enough when you're dealing with really really massive light waves and i don't mean massive as in actual mass i mean massive as in very long optical light on the other hand has wavelengths in the hundreds of nanometers so we need to work quite a bit harder to get a smooth enough surface for specular reflection of optical wavelengths unfortunately for me wanting to make 3d printed mirrors a raw 3d printed shape has layer lines so this is a benchy it was printed at 200 micron layer height and i want to smooth out the hull and turn that into a mirror all according to plan this is what you saw in the intro you could remove material until you had a smooth surface for example with sandpaper but that still leaves a pretty rough interface especially in plastic instead i opted to add material that would conform to the layer lines on one side and form an extremely smooth surface on the other all by itself it may sound like magic but it's surface tension so let's get started so after some preliminary uh mechanical smoothing i need to clean the parts up so that they have a nice surface that will accept the resin so now i've got a whole bunch of parts that have had water on them what is the fastest most overkill way to dry things in a hurry can't have any water in effect this stuff that i'm pouring out is a pretty common resin for smoothing 3d printed parts it's a two-part epoxy that cures at room temperature so it's pretty easy to work with a friend of mine in town actually owns an sla printer and i considered using some of his uv cure resin to smooth parts but i skipped that out of fear for oxygen contamination and sticky rough surfaces although for reasons that you'll see in a few minutes maybe that'll be worth trying in the future after the epoxy is mixed you basically just paint the stuff onto your parts and although i don't think it's how you're supposed to use it i've been using really thick coats to let surface tension do as much work as possible as this stuff gets in between all the sanded down layer lines anywhere that it touches air is an energetically expensive boundary nature doesn't like different materials touching each other like this that's why falling water droplets make spheres spheres have the lowest surface area to volume ratio oh shoot that one was a bit much and in this case the liquid resin naturally smooths out to minimize its own surface area the upshot of this is that we can use it to take jagged edges and make them round however these thick coats mean that i do get some issues with flowing resin so for large parts i actually started slowly spinning them while they cured to avoid forming big drips anywhere turns out that works pretty well for ornamental parts but doesn't even come close to giving me the precision that i wanted for optical mirrors there will be more on that in the future now that the parts that i want to make into mirrors are all epoxy coated and they have a smooth surface the next step is the really fun part the chemical silvering these are going to be first surface mirrors whereas your bathroom mirror is actually a second surface mirror like i said before for a mirror to work at all you need a surface a boundary between two different materials and if that critical interface is between the metal film and air then you've got yourself a first surface mirror the light bounces off of the first surface in general though if you take a flat surface and coat it with metal that film is really really easy to damage or dent or chemically degrade so outside of really weird applications like telescope mirrors or optics for physics experiments that's really not done instead the metal is applied to the back of a sheet of extremely flat glass which means that incoming light passes through the air glass boundary bounces off of the glass metal boundary and then goes through the glass air boundary again before reaching your eye but if you have a second surface mirror you can like breathe on it or poke it with your finger without completely destroying the metal coating so they've got that going for them which is nice now the whole point here is that you're taking these nasty chemicals and spraying them into the air so i will be wearing a respirator the whole time that i'm doing this this silvering process is actually a set of sprays that evenly distribute the precursors across the surface you want to coat and then they react on the surface deposit elemental metallic silver and the excess chemicals wash away because this is actually a commercial kit i can't say with certainty what's inside but one very likely candidate is silver nitrate i'm assuming that whatever's happening in these sprays is a variant of the tollens reaction through a convoluted set of reactions the silver ions in silver nitrate are converted into diamine silver which can be reduced into elemental silver which can deposit as a solid metal film on a smooth glass or in this case plastic surface one thing that weirds me out about the spray silvering technique is the necessity of mixing these chemicals at the last possible second before reaction i don't know which of these precursors is embedded in which spray presumably just preventing them from reacting until mixed but i can say that if the silver fails to deposit an additional reaction takes place that ruins the solution very quickly a couple months ago when i tried this process for the first time notice that i was working on the floor because i didn't even have tables in the garage yet i didn't prepare the surface properly when i sprayed the silver containing solution and the reducing agent at the plastic surface the solution just balled up on the hydrophobic resin and the reaction to make silver happened within the liquid droplets instead of on the plastic surface after the reaction was done there was this gross brown slurry everywhere there's still residue of it inside this plastic basin i'm using as a spray booth after some later wikipedia it turns out that this brown slurry is actually a precipitate of silver oxide all the silver atoms that don't adhere to the surface as a metal end up reacting with oxygen in the liquid and seeing as silver oxide isn't soluble it makes all these tiny little specks you might look at this and go oh my god look at all the leftover silver but this is actually just a sheet of aluminum foil all of the really gross bit the part that's like brown that's actually the leftover silver oxide first step is to clean it really well and i'm just going to spray this down with water and make sure there's no dust on it you may notice as i spray water on this that the water like shrinks and forms little droplets it doesn't actually want to stick to the surface this is a problem because we want to coat the surface with reactive chemicals so the first thing we need to do is apply the wetting agent which is basically a surfactant that's going to stick all of the liquid to the presumably non-polar inert surface of the resin it even makes bubbles just like soap a surfactant you're probably used to now we add a little bit of the sensitizer i need to wait 30 seconds i just get to sound like darth vader breathing through this thing drink some water oh i don't like that at all i've done something to the back surface we'll see okay let's go [Music] oh my god what happened to the surface so yeah i love watching the beautiful smooth silvering reaction when it works but what you just saw when i tried to repeat the process so that i could explain the steps on camera this happened darn well it's silver but man a mirror it ain't it worked the other day what a mess look at that and when i tried to silver another smooth part this play button it still didn't make a mirror so to determine whether it was the silvering process or whether it was the parts i decided i needed a more scientific control and i found an old smooth sample from my original batch of resin coated prints that i hadn't yet silvered and drum roll please so that's the mirror finish that i wanted it worked somehow i had managed to mix a batch of resin that reacted with the chemicals during the silvering process and actually roughened to the point where it wouldn't be a mirror anymore sometimes it was so bad that you could see the roughness even before the silvering agent went on so something in the hard resin was probably reacting with something in the wetting agent or even the water so what did i do well i tried another batch and then i tried another batch and then i tried another batch unfortunately i kept doing something wrong or at least different than the first time that i tried it when everything worked perfectly all i wanted was to be able to recreate that based on the fact that some of my coded parts seemed like they remained kind of tacky and sticky for a few days my hypothesis was that i had screwed up the ratio of the resin a and b and prevented it from curing in the way that it wanted to i had also mixed all subsequent batches outside in the garage for ventilation after realizing with the first batch that the smell was pretty terrible for me to be mixing this in my office and then just because i like adding complexity i was also trying to get better optical surfaces and wanted a less viscous resin which the instructions say can be achieved by watering it down with acetone so now i had three variables and a deeply wired instinct to collect as much data as scientifically possible so i ran a grid search with every combination of these parameters for a total of 12 different resin mixtures after seeing these the next day i added another variable mixing time so i actually now have a four dimensional data set and i duplicated a few of these original conditions again with a much longer mixing time before spreading the resin on the parts it's worth pointing out that if you ever see a scientist do a big grid search like this it generally means that they don't know what's actually going on and they're just sort of throwing variables at the wall and seeing what'll stick and in this case that is exactly what i was doing okay so some of these are clearly not cured because they're still stuck together i'm not a chemist i don't really know how these polymerization reactions work so massive grid search land through four dimensional parameter space was really the only avenue that i could walk down and hope for a useful result after a few days waiting for all these samples to cure i attempted to silver them all along with another previously unsilvered control sample from that magical first batch and thankfully most of them didn't work i say thankfully most as in thankfully not all because a few of them did produce mirror finishes turns out that curing them inside is better for reasons that i really don't yet understand acetone doesn't seem to do much and if you screw up the ratio of part a and part b it's better to air on the side of an a-rich mixture if you don't want the surface to react with the silvering solution chemicals so that was an awful lot of work to solve not really that many questions but there you have it it was time to mix one more batch and make a play button also as a psa about this resin on fdm parts if you try to put it in a vacuum chamber to eliminate bubbles it has the exact opposite effect i thought i was going to be so clever [Music] [Music] well this looks to me like a mirrored silver play button which means it's time for the laser now this is the time where i really need to thank all of you viewers of the channel alpha phoenix went from about 18 000 subscribers to over 100 000 subscribers in the span of literally a couple weeks and i was just watching all the charts sort of along for the ride this button is technically for the 2 to the 15 milestone but i did decide to make this one the actual silver play button because it's the first video i'm publishing since the channel hit 100k i'm thrilled that so many people are interested in all of this stuff and although this happened while i was moving the worst possible time for me to keep up a production schedule i want to thank all of you for sticking around and i hope that you find all the videos to come both entertaining and informative thanks for watching ah also while i'm in here key lime it's a kind of green i swear [Music] you

Channel: AlphaPhoenix

Views: 680,667

Keywords:

Id: SCuUQOTehqw

Length: 17min 4sec (1024 seconds)

Published: Fri Nov 12 2021