Make plastic printed circuits with a standard laser cutter

today on Applied Science we're going to talk about plastic printed circuits so these have been available commercially for a while but in this video I'm going to show you a new process tour we can take a 3d printed part put it in the laser cutter dip it in some chemicals and end up with this this shows a 1/2 millimeter pitch a leadless part soldered down with o4o to components and oh you also get plated through holes for free of course because we just make the holes in the laser cutter and this process will plate those holes so I wanted to show this was more than just a curiosity this is actually a pretty functional process and I'm gonna give you all the details in this video and then in the second half we're going to talk about the commercial process how this is done at scale and you can actually do a similar thing in the laser cutter by mixing some copper oxide which actually makes copper traces without any plating baths I mentioned copper plating bats and so let's start talking about ways that we can get copper to stick to another object and so one of the most common ways is to start with a copper electrode and something that you want to cover in copper and put both of these into an acid bath and then run an electrical current between the two plates and the copper ions will go through the solution and coat the other object with some additives in the solution the problem is I mean this requires the object that you want to plate to be conductive Rank's they're running an electric current through there and that's not gonna work for us because we're starting with a plastic substrate so there's a whole other kind of copper plating technology called electroless copper plating and in fact it's used quite a bit in the circuit board industry so this is actually not a solid piece of copper this is a piece of circuit board and in the standard process what you want to do is drill holes in this thing but then the like the the through-hole is not plated the hole with nut will not be conductive and if we want to make a connection from one side of the board to the other the fabrication house that makes these circuit boards needs to plate the inside of that hole with copper so this electroless copper process is used actually everyday all around the world to make circuit boards you would think with such a common process used by you know probably thousands of businesses around the world that the details would be well worked out unfortunately when it comes to electroless copper trade secrets rule everything there are a few patents but believe it or not there's actually not even enough information and patents to really figure out what's going on it takes a lot of trial and error and I'm going to give you the condensed version in this video and give you the exact measurements that you can hopefully replicate this the whole idea behind an electrolysis copper system is to make a solution full of copper that's just ready to fall out of solution right so we've got this beaker here filled up with the magic liquid and it's right on the verge of dropping all that copper out and making metal but the trick is it's not doing it right like the beaker is not coated with copper it's just shy of being to that point where the copper would fall out so the trick is if we put an object in here that we want a plate with copper how do we make the object different than the beaker and the tanks and the pumps and everything else you might have connected to this and the answer is that we first dip the object we want to coat in copper in a catalyst and so that's over here and so the basic process is to really wash this really well dip it in the catalyst wash it again and then dip it into the magic solution and take it out and you'll have a copper part you probably guessed that since this magic solution is right on the edge of dropping the copper out all by itself it is pretty unstable so this this solution here will probably not last more than a day or two I've experimented with different formulations and some are more stable than others but generally if you want it to run quickly like in other words you you want your part to be coated with copper quickly that means that the bath has to be very aggressive and in that case it's also unstable and just won't last very long on the other hand if you're okay with going slow you can mix up the bath to be less aggressive and so it'll be more stable but it will also plate copper more slowly if you think about it this process also requires that the freshly laid down copper continues the process right like if we dip this on did part in the catalyst then dip it in the bath and it starts to plate because we have that nice catalyst on there as soon as that first atomic layer of copper is laid down we don't want the process to stop we want this to keep going ideally forever and so the other trick is that this bath has to be Auto catalytic so that all the freshly deposited copper will continue that reaction so all these requirements you know the thing has to be kind of teetering right on a knife edge it has to be quick but also stable it has to be Auto catalytic mean that there's quite a few requirements placed on the chemical composition let's talk about what's in this magic plating bath first we need a source of copper ions and so in that case we're going to use copper sulfate and already if you start searching for recipes for this you'll run into a slight problem sometimes the recipe will call for copper sulfate without saying whether they're talking about copper sulfate pentahydrate now it's really the same chemical it's just that the pentahydrate means there's some water in there so if they're calling for two grams well 2 grams of copper sulfate anhydrous is different than two grams of copper sulfate pentahydrate and so at first it might seem like a you know a pedants sort of point that's just annoying but but actually it's critically important and if you don't specify it's kind of hard to tell what's going on so I'm gonna show you the labels of all the stuff that I use so that it's completely unambiguous what I've what I've done here so copper sulfate pentahydrate 5 grams per liter potassium sodium tartrate tetrahydrate I just call this tartrate on my car charts this is 20 grams per liter sodium hydroxide 5 grams per liter it doesn't have to be particularly pure I'm just using this drain opener there's no aluminum chunks in here it's just sodium hydroxide sodium carbonate 5 grams per liter and finally a solution of formaldehyde 37% 10 milliliters per liter another little gotchu is you might hear the word formal in well formalin is basically just 37% from aldehyde in water with maybe a stabilizer in there and then even more confusingly you might get a like a 10% formalin solution so a 10% formalin is just 3.7 percent from l dehyde here's a quick rundown of what's happening in here chemically obviously the copper sulfate is the source of the copper ions that we want to eventually plate the object the tartrate is a chelating agent that allows the copper to dissolve in the water more easily without the tartrate the copper sulfate wouldn't dissolve and a high enough concentration for this thing to be useful we want the the liquid to hold a lot of copper the formaldehyde is a reducing agent and so the formaldehyde if there were nothing else in here would cause the copper to just start crashing out a solution if there was enough of it basically the copper is going to be stored in an oxidized state in solution and then adding the formaldehyde is going to reduce it and cause it to go back into copper metal the sodium hydroxide is just a pH adjuster this whole process only works at high pH very high pH in fact about 12 and a half or 13 so we need sodium hydroxide to do that and the sodium carbonate is a buffer so the as this process continues it actually creates its own acid like if for every molecule of a copper or every atom of copper deposited there'll be some amount of acid created and if we don't do anything about that it's going to shift the pH of our solution and then cause the thing not to work well anymore so the trick with adding sodium carbonate is that it can absorb some acid being created by the process and maintain the pH where it should be so I've tried a lot of different recipes and found this one by far to be the best it's pretty stable it will last for at least a day or two and if you stick exactly to these values you shouldn't have any problems just for the records I've been doing everything in 300 ml batches in beakers on stir plates and typically the way that I mix this up is to dissolve the tar straight first then the copper sulfate then the NaOH the carbonate and finally the formaldehyde and wait until each one of these things is dissolved before adding the next ingredient okay let's talk about this catalyst solution so here it is and I've also been mixing this up in this is actually a 250 ml batch and the idea here is that we start with water add some hydrochloric acid this is going to be 15 ml of hydrochloric acid - 250 ml of water and then we're going to add a tiny amount of palladium chloride point-o 6 to 5 grams in this case and it dissolves very slowly so you want to take your time and heat it up a little bit 40 or 50 degrees C to help the stuff dissolve don't move on to the next step until it's completely dissolved and then after the palladium chloride is dissolved you add 3 grams of stannous chloride and it's you can see the bottle here this solution goes through some interesting color changes after adding the stannous chloride it's sort of greenish dark green and then if you wait about an hour or two which is what you have to do in order for the solution to be ready it turns this interesting brown color an alternative approach to this palladium chloride colloidal solution is to use the two ingredients separately so you can have a stannous chloride solution here and a palladium chloride solution here and you can do it sequentially so you dip your part first in the stannous chloride rinse it and then dip it into the palladium chloride rinse it and then go into the bath and the benefits of this idea is that since it's not a colloid these solutions are much more stable so that you can just use them for a long long time you don't have to worry about it falling out however I haven't had much luck using this process I pretty much always go with the colloidal and we'll explain why in the next few steps okay so the basic process of completely plating an object like this piece of circuit board material is to wash it very carefully in fact washing it is another critical step that is sometimes overlooked you got to use something pretty powerful like this Alka Knox cleaner and you want to heat it up to 40 or 50 degrees C and then dip the part in there with agitation for a good 20 minutes to make sure the part is completely squeaky clean if there is even a trace of oil and I mean when I say a trace I mean there's like oil droplets in the air normally and if you just leave a part out on the counter it will eventually accumulate a very very thin layer of oil on it and so you got to get rid of that so you take it out of the Alka Knox rinse it quickly into the activator rinse it into the plating bath rinse it and then we're done getting all that working took quite a while but unfortunately with plastic it's even more difficult as we'll see so after getting this electroless copper process working pretty reliably on circuit board material I decided to start trying it out on 3d printed parts and so I'm using a formlabs form 2 printer and I'm using the clear resin and right away I realized I had a big problem so this is part number one the first of many samples and I used the laser cutter to mark these stripes here at different power levels with the laser so this is about 12% and this is maybe 20% of my 60 watt laser cutter in sort of a raster mode and as you can see we got a big problem one the part plates where I don't want copper and then where the laser hit it the plating is actually not as good so we have two problems one we have to figure out a way to prevent the copper from sticking where we don't want it and then we also have to sort of invert what's happening with the laser here so you can kind of separate this whole thing into two problems I spent probably a month or two researching you know off and on this whole process and how there's existing technology to make selective copper deposition like this and it seems like surfactants and controlling sort of surface affinity is a pretty big deal so I started experimenting with sodium lauryl sulfate which came up in one of the papers that I found and that researcher was using this surfactant which is short for surface active agent it basically changes how hydrophilic the surface of a material is right so if you've heard of detergent herbs or soap even it's basically a way of bridging this surface that doesn't want to interact with water the surfactant allows the water to start interacting with the surface and what this comes down to is basically dipping the part into the catalyst means that that catalyst has to come into really intimate contact with the surface in order to stick there and then if it sticks there then when we put it in the plating bath we get copper there so this whole thing comes down to controlling whether the catalyst sticks or not to the part so as it turns out sodium lauryl sulfate is an anionic surfactant meaning that the molecule is negatively charged and it's other part of the molecule is hydrophilic so it will attach to positively charged surfaces because it's negatively charged and allow that surface to be hydrophilic and when I started experimenting with this I actually found that it had the opposite effect so when I was using SLS it would cause the laser Daria's to be even worse like you know taking the copper even worse than without it and so this this process took a very long time but I eventually figured out that maybe I should try a surfactant of the opposite charge a cationic surfactant and here it is the secret ingredient that makes all of this possible is wet & Forget that's right this is actually 10 percent alcohol dimethyl benzyl ammonium chloride and doesn't sound very common as it turns out this surfactant kills moss mold and mildew it's also great for making circuit boards so my best guess to what's happening here is that the laser cutter when it touches an area of the plastic will charge the surface and then when we put it into the go ahead and forget the charge is appropriately set up such that that surfactant will cause the area to become super hydrophilic and then when we dip it in the catalyst the Palladium particles stick there then we put it in the plating bath we get copper there and so I figured out that by going through this process I could make the laser Daria's take copper really well but then we had the remaining problem of getting rid of the copper where we don't want it to solve that problem I also tried many different things and found that coating the piece of plastic before lasering it with this lubricant called tri flow actually worked really well and really what this is is just a an oil a very light oil but I've tried a lot of different kinds and found that tri flow works by far the best and so the idea is that we coat the part in tri flow now it's not hydrophilic at all because you know oil repels water of course right so the oil kind of soaks into the top layer of the part and if we were to go through the whole process with that we would get absolutely nothing in fact this is what you'd get right so this went through the whole process with the oil and it takes no copper at all and that's because the oil is really effective at keeping the palladium away there's no palladium we put it in the copper bath there's no activation and there's no plating so now we have kind of all the ingredients coming together and when I was getting up to about you know test number 66 or whatever here it started to fall into place and I realized that I could coat the sample with oil then laser it then put it in the surfactant then activate it and then I played it with copper and right around sample number 83 I really dialed it in and realized that this was going to work a lot of process variables to control here so let's just step through it and you can see at all okay to start off we start with a 3d printed part and I'm just making flat little test coupons to make life easy and as an optional step I dip it in sodium lauryl sulfate a very dilute solution and I don't think this is really necessary but it's just kind of extra insurance it makes sure the surface is absolutely unresponsive to any plating process next I blow-dry it and then apply the Tri flow and you really don't have to soak it I mean just putting a few drops of Tri flow on there and then wiping it off completely is really all that's necessary next I put it in the laser cutter and let it do its thing using a raster pattern to make the traces and then using a cut pattern to make the little vias and the process window on this is pretty big like I've tried higher power and lower power and it doesn't make a huge difference you don't want to do a high power thing though because it removes too much material and then the copper areas are just physically lower than the surface and that makes soldering more difficult so basically you want to use as little laser power as possible then we dip it into the secret secret ingredient the wet & Forget and i diluted this down to 1% so a 10 to 1 dilution from the commercial product gives a 1% surfactant solution and I leave it in for 1 minute next I rinse the part in distilled water and then put it into the palladium colloidal bath for about 2 and 1/2 minutes next rinse it in distilled water again and then put it into the copper plating bath at about 40 degrees C with stirring for about an hour and a half now the air bubbler here is controversial remember how I was saying that this bath works by having the formaldehyde reduce the copper causing the copper oxide to become copper metal and if you put an air bubbler in there you're introducing more oxygen into the bath and that will actually slow down what's happening that's good because it makes the bath more stable you can kind of control it a little better but I actually found a pretty big downside at first I thought oh I know it's a great idea I'll use the air bubbler and it will push away all the little hydrogen bubbles that form on the as part of this process right because you know you get a hydrogen bubble forming on your part then it's not gonna plate there anymore and so it makes the results inconsistent so at first I was thinking the air bubbler is great one because it stabilizes the bath and two it you know sweeps away all this hydrogen but as it turns out it actually completely destroys some parts there's if you put it right in the Airstream eventually a bubble an air bubble will touch the copper and very slightly oxidize it and then it's not active anymore and it actually halts the entire process from working so I eventually gave up on the air bubbler and found that this bath for which I gave the recipe in this video is stable enough where it doesn't really need the oxygen bubbler after the plating is done it's pretty much it just cleaning it up with a little bit of naphtha and just rubbing it pretty firmly with a paper towel will take off any traces of copper that they're sort of sticking to that oil layer that you don't really want and I have to admit later on I found out that sandblasting the 3d printed part helps a lot and I'm trying to figure out how to make the process work without a sandblaster because that's an additional piece of equipment that you might need the goal in doing this whole thing was to make a process that anyone with a 3d printer and a laser cutter could do in in a hackerspace the process still works without sandblasting of course it just makes the results more consistent especially for fine pitch work if you're doing larger traces like 20 or 30 mil wide traces none of this matters at all it's it's totally fine if you're going that big so that's basically where we are the remaining problems are the adhesion if the copper to the plastic is really good actually until the flux hits it and I know it's the flux because I've plated some things continuously in copper and you can solder to here in the adhesion is pretty good I add a wire on here there's the surface mount component and you can push on it pretty hard and it's it's not coming off I eventually yanked the wire off but it took a lot of force to to rip this open the trouble is if you're doing you know a relatively thin trace like this and soldering it with paste or whatever the flux gets under the copper and then that destroys the adhesion it's really terrible and that demo piece that I showed off at the beginning of the video is not strong like I admit it's it's a very fragile piece and so figuring out how to get the copper to stick to the plastic and withstand the flux is really the trick it's not the temperature because I put my iron right on here and it's still holding just fine even the trick is that since the copper was continuous it kept the flux out so you can see here what the air bubbler did notice in the upper one this second p in the word applied looks all corroded it actually came out of the bath looking just like that and this one I had the air bubbler going and realized what was happening and funny enough it kind of spreads you know that that second P is completely corroded but the letters next to it are totally fine it's weird how the corrosion kind of spreads to all the areas that the copper is touching but not further and then the second one I realized what was happening and quit with the air bubbler and ended up with better results there let's finish up by talking about how these are actually used commercially already so all these test coupons are flat and rectangular of course but the real excitement with making a plastic part with a printed circuit on it is that it doesn't have to be flat it can have all kinds of weird geometry and if you combine this with through holes and everything else you can do all kinds of cool stuff and the chances are you're looking at one of these parts right now or it's in your pocket this is actually a piece of a cell phone and these are the conductive traces that have been applied to this 3-dimensional plastic part this is actually the antenna for this cell phone that this came out of and so that the conductive trace reps around the outside and makes contact with the phone circuit port here and then provides the antenna designer this 3d surface to design an antenna on which is pretty nice I picked this one up at a trade show just and they were showing off their high resolution capability I'd say this is about maybe 8 mil maybe even 6 mil trade space probably more like 8 though and they've gold coated it so that it doesn't corrode of course but I think it's just copper under there the way these are made is pretty interesting these guys are also laser defined but they use a very expensive pulse laser they don't use a co2 laser like you'd find in your average hackerspace and part of my goal in doing this whole project was to avoid needing this very expensive pulsed laser so the way this one works is they add a special powder to the plastic when this is injection molded and the powder is basically a catalyst it's not palladium but it's a similar catalyst so if the injection molded this whole part you would end up with something non conductive because the catalyst by itself is not conductive but then they put this in a machine that hits it with a laser selectively and where the laser hits this catalyst it becomes active then they dip the whole thing into an electrode list copper bath just like I showed in the video and those areas that are now activated become coated with copper when I started this entire journey months ago my original goal was to mix the powdered catalyst with the 3d printed resin and then take it out of there and then hit it with the laser and try to go through the whole process and then once I realized that we could do this on the surface without bothering with you know powdered catalysts and whatever it's you know it's way easier it's way better it could work with potentially any resin that comes out of that 3d printer also one other note probably a lot of you are asking does it work on PLA no and the unfortunately it doesn't and the reason is that the plating bath is very high pH and PLA disintegrates in high pH solutions so unfortunately the chemistry of depositing copper like that makes it such that it just simply won't work with PLA of course I tried it just to be sure and not only does the part get ruined it actually ruins the entire plating bath what happens is the the PLA starts to disintegrate and then all of the catalysts and little bits of copper that have started to deposit you know move out into the solution and the whole thing crashes in a matter of thirty minutes basically all the copper drops out and you have to start over again anyway I did go a little bit down the path of trying out these embedded catalysts and so to test this out and stuff some epoxy with copper oxide which is one of these latent catalysts that you can put into a resin and interestingly enough if you put copper oxide and epoxy together and then put that in the laser cutter and D focus the laser beam it will actually make copper traces all by itself you don't have to plate anything there's no path or anything like that and at first I might think we'll wait a minute but how does amine heating up copper oxide doesn't turn it into copper the trick is that the hot laser beam is vaporizing the epoxy resin and it's creating a really strong reducing environment because it's very hot in there and the epoxy is trying to burn but it's not able to get enough oxygen going and so the the hot burning epoxy actually reduces the copper oxide down the copper metal and it makes a conductive trace I mean it's not very good and as soon as you flex it it's gonna pull apart so it's it's not really usable but it's kind of an interesting trick okay well you know put your questions in the comment section I'll do my best to answer them I'm gonna continue working on this and if anyone else is really interested in it let me know and I'll give you as many details as I can okay see you next time bye

Channel: Applied Science

Views: 375,537

Keywords: electroless copper, pcbs, electronics, copper plating, laser cutter, laser, tartrate, DIY, copper circuit, circuitry, applied science, krasnow, ben krasnow, prototype, circuit board, PCB, making, plated through hole, plated, copper plated, laser defined, LPKF

Id: Z228xymQYho

Length: 27min 49sec (1669 seconds)

Published: Mon Sep 03 2018