Copper plated vias

[Drirr]

Hi,
Long time ago I succesfully made my first 2 layer copper plated PCBs using mixture of graphite and some other stuff.
But.. I decided to step further and found https://hackaday.io/project/180422-process-01-an-open-source-2-layer-pcb-process

I'm not a chemist. I'll be glad for any advice.
Problem is that final solution is yellow but it should be dark green/dark brown?

Process as described in guide:
----------------
Main Solutions
1. Cleaning solution. Around 10% by weight sulfuric acid. Concentration not critical. Deionized water preferred. You can prepare this solution
by dissolving 13mL of concentrated sulfuric in 225mL of deionized water. Stable, can be stored indefinitely and reused many times.
2. Palladium chloride activator.
Instructions for preparing 50mL
40mL deionized water
0.02g PdCl2
2.25g SnCl2·2H2O
7.75g NaCl
2.5mL 34% HCl
adjust the volume by adding deionized water till 50mL
Stir until every added reagent dissolves before adding the next one.
Should last one year or more in a capped container.
---------

I believe that problem may be using kitchen salt as "NaCl" or maybe PdCl2 bought on ebay is messy? SnCl2 is 97% purity, destilled water was used.

PS I know that making PCBs at home does not worth. Main reason are 50GHz microwave filter prototypes on special substrate.

Thanks! David

[coppercone2]

don't know but I will get off my ass and try it (ordered reagents)

BTW you don't explain the pictures. I don't know what I am looking at.

And also, just talking to people that have been around for a while, some people seem to think that the quality of copper pipes sold at home depot has decreased over the years.

[Drirr]

@coppercone2 I am interested in your results. Today I failed again still not able to make it. I also tried some chemistry forum on reddit. I will let you know.

There are some photos:
    I mixed destilled water with 20mg PdCl2

https://photos.app.goo.gl/JgviyNDDSzt2MHp56

https://photos.app.goo.gl/hDrZPWXGahwFdFt59

https://photos.app.goo.gl/vDFHE7mAC59jSxty5

2) I added SnCl2

https://photos.app.goo.gl/3YW55gwYP8CwFenw7

https://photos.app.goo.gl/VFVfEddPDA3qGKxK8

And now some foggy unsoluble sediment appears:

https://photos.app.goo.gl/YKraQbfikpjnTc3Z8

and now I do not now what to do, I even tried to filter sediment but my final solution (after adding NaCl and 34% HCl) is pale yellow all the time. And.. not working at all.

[coppercone2]

i got palladium and some very high purity sncl2 . want to try this still


alright I got everything but the albino goat head needed for this god damn odyssey

at least you can make as much as you need, my experience with plating kits is you end up with a bunch of expensive bottles with muck on the bottom.

And I suspect its nothing but salt and a bit of some proprietory crap like triton or some BS like EDTA with a 1000% markup

Like, maybe its some finely tuned mixture at first, but after like a few uses, it is probobly severely degraded. Unless you pour out that commercial plating stuff after a single use, I suspect you get little benefit from whatever the manufacturer did. If they are using it alot you need basically some kinda plating technican to maintain that mix. since you gotta do it anyway, making it yourself seems just cheaper. The nickel bright I had after a few tiny uses ended up having some kinda sediment develop in it.. .super funky. like after plating 3 square inches of copper. I am super suspicious its not worth the mark up so long you follow best practices like getting whatever additives (there is a few of those)

[coppercone2]

lol. mettler. 20 miligrams of pdcl2 almost resulted in my banging my head through the ceiling of all places. those scales test you. I need to straiten out one of the rings now because it got a bit deformed (mechanical dac).

how in the fucking fuck do they fucking expect you to work on that god damn machine. It needs a cast frame and removable access panels.  I ended up having to modify tweezers to work through the lid. insane swiss

[coppercone2]

lol, my solution turned milky white. Its not black or orange. I wonder if the palladium I got was fake.

it bubbles and turns steel black if you drip it on steel.

I wonder if he means 0.2 grams of palladium. because 0.02 aint look like shit. The chinese solutions are 5g / liter.

this would be .4 grams a liter. I have a feeling that PDF is fake with fake pictures etc. I don't see how it will turn 50ml of water a dark black color. Like how the HELL could it do that. I wonder if its modified for hackaday because the cheap articles are the only ones that get popular. maybe its too expensive. I am really suspicious.

[coppercone2]

its 90% dissovled. Its clear with white powder on the bottom. The PdCl2 seeming disappeared. I wonder if I should add more.

My purity for the salt and tin was 99.9+ % and nothing from the store. The PdCl2 is chinese. 

[jpanhalt]

1)  This is an old thread, if it actually worked, I might expect more responses before now.
2)  Electroless copper as describe will be extremely thin.  I just drill a hole (#72 drill or smaller), insert a wire, solder both sides, and cut off the excess with flush cutters.  That's what I did.
3)  I haven't etched a DIY board in years.
4)  If your heart is set on plated through holes, why not start with a double sided board, pre-etched or not, drill, and add a solderable "silver ink" to each hole.  (https://pubs.acs.org/doi/abs/10.1021/ja209267c )  Unfortunately, the ACS no longer allows a meaningful link to the procedure. It is widely available and is a modification of the very old Tollens reagent that allows plating to almost any arbitrary thickness.  It used to be used for making telescope mirrors, and maybe still is for hobbyists.  The original paper noted that the nano-deposit of pure silver was solderable.  The reagent is relatively stable in the dark, as you might suspect.

[coppercone2]

I think it might do something its just maybe not whats in the picture. I found like 7 different formulas but all of them have more palladium. I added some more and left it on a warm stir plate.

I mean more then likey the Pd is off, because its ebay china.


But with this, you can plate 3d prints, without the paint. Paint is obnoxious. I.e. my waveguide horns made me cry when I wanted to plate the aperture nicely. Painting a small rectangle hole is hard.

getting a plating solution with a good parameters (level) might be the key to utilizing 3d printers for high microwave frequencies. you have a fair bit of precision and resolution but the paint messes it all up.

of course, 1 or two more fails makes buying the made sensitizer it the economical option. I thought it might be fun and cheaper to make but it looks like its not so easy

the high resolution of the dlp printer can be harnessed

[jpanhalt]

Please define the distinction you make between "paint" and "ink."  Did you bother to look up either Tollens or the link I gave?  There is no added matrix.  It is pure silver. 

Actually, no real difference between that and the copper plating solution you are trying that doesn't work, except the former does not require a questionable catalyst.  In the questionable solution you are trying to duplicate, Pd  acts as a catalyst to for the reduction of a copper salt by formaldehyde. In the method I presented, which works, the silver salt is simply reduced during evaporation and mild warming.

Is this related to any real project or just an exercise in posting?  Or, is your purpose to get something copper colored rather than simply a good electrical and solderable conductor?

[coppercone2]

I don't wanna buy your crap. Its just for the poster that was asking to show that someone else is interested in his experiment. I went through the effort of buying the stuff like he did and we both ran into a problem. I dunno what the hell you want

i will actually instead buy some palladium from a reputable supplier and try it again. this thread is about catalyst electroless plating

this is seriously like two people trying to get a transistor to work and your pushing relays. maybe your method works. it does not mean we are not interested in the traditional chemistry way, even if its harder.

if you just wanna buy something, you can get colloidal palladium which is a replacement for the acid palladium catalyst method. or use a vacuum deposition machine. or or or we started something already

[coppercone2]

well I left it on a stir plate heated for a day. it looks like there was some residue after it settled.

But the weird thing is this

if I dissolve the red pdcl2 in HCl (just a few ml) without water, it kinda dissolves slowly. Clumps a bit.  Well I assume that means its not powdered brick. I add salt it retains color.  If I add salt maybe it dissolves more. As soon as I add the tin chloride hydrate, all the color goes away.

I wonder what they sold me.

But I think I will wait for fresh reagents before trying again.

Iron (rust) I think would react strongly with acid and be very soluble in most forms. It must be something bizarre.

At this point I am more curious as to what it is then the actual plating. Maybe they crush up red ceramic and sell that.

[coppercone2]

Ok part of the problem might be diluted HCl and then improper tin chloride handling.

So unless you read about it, no one tells you that tin chloride, like tin lead free solder, is a quirky mf. You need to dissolve it in an acid of sufficient concentration to mess with the reaction balance to prevent tin oxychloride or something from forming. I noticed that for the palladium decomposer.

there is that old story of someone being jailed in old russia because a train full of tin disintegrated because of siberian weather. and those antarctic guys that might have got iced because their tin storage containers disintegrated because of the cold. And tin whiskers. Its that kind of element.

That HCl is mad old, maybe it lost it potency. So i will try this in a few weeks with fresh acid and americanized palladium. It looks like for tin chemistry, you need to be really certain about your HCl. I need to actually work out the numbers in this poorly documented PDF to see if it even makes sense.

But I guess I should expect about as much from hack a day, I am sure it was gutted so the bloomberg editors don't delete it for being over complicated for the mass market. Maybe they accused him of being a corporate entity when they saw reaction mathematics.  They don't accept complicated stuff you worked out usually if they think pop sci audiance might be intimidated (even though their not gonna try making anything there)


probobly what you wanna do is dissolve the palladium in acid water, then dissolve the tin in separate acid water, make sure it dissolved properly, then mix those (and how do you deal with the salt? not sure).

[T3sl4co1l]

Well, SnCl2 is simple enough: it's a mild reducing agent, and amphoteric.  Don't let it oxidize, or you get some Sn(OH,Cl)4 contamination, and a lower reductive potential, or even an oxidizing one instead (if enough Sn(IV) dissolves).  Amphoteric is... not quite the key aspect here, but the fact that Sn(II) is kind of equally comfortable in high, medium or low pH (as the stannate(II)(2-) ion, Sn(OH)2 precipitate, or Sn(2+) ion, respectively) means you need to maintain the correct pH, and it can easily lose Cl- and precipitate at neutral pH.  Especially avoid hard tap water (includes Ca and HCO3 ions), use distilled or deionized, etc.  Maybe boil it not just to remove hardness (if applicable) but sparge O2 out of it as well (again, another source of oxidation).

You may be able to substitute another reducing agent for Sn(II), sulfite or thionite for example, including adding it after mixing, but likely that changes the whole mix.  The Sn ions may have some effect themselves, I don't know offhand.

Pure HCl probably pays, too.  I've seen some pretty green (Fe contamination) stuff at the hardware store, myself.  Whether Fe(II) or (III), that probably screws further with the redox balance.  You can always titrate the acid to assess its concentration.

Starting with known pure and freshly prepared reagents will give the best chance of success.

Tim

[coppercone2]

well maybe its bad reagents because the stuff I got looks like milk in distilled water in mere seconds.

I am gonna try to make the tin activator step 2 first with the science HCl first . If that don't work maybe its bad tin. I spent way more money then I wanted to on this slay ride


Why do I have the feeling everyone that does this makes their own tin chloride and has no idea what happens if you actually get powder tin chloride and put it in water. This reagent manufacturer looks fairly legitimate, I would be surprised if their stock pile is that fucked up. I mean this merits a size 25 font disclaimer about tin.

I got a roll of pure tin solder from a reputable manufacturer for my gallistan experiments. I can try to make it too but I thought buying the powder would be purer. 

[coppercone2]

ok I bought a glass gas wash bottle with the glass diffuser I think it will be good for putting nitrogen into distilled water. lord knows from what I am seeing this might as well be the manhattan project.


This is so god damn bizzare, a salt that oxidizes lol

I was thinking to use an aquarium stone but I bet ill find out its contaminating it with something or another


Maybe its good for degassing high voltage oil too. I wonder if you can get silicone oil out of the frit, or if its no bueno. If I do that maybe I can rig it to a perestalic pump to cycle xylene through it or something like that if it ever treats silicone oil. But it would make me feel better about mineral oil too


I am waiting for the next one, chinese tin contaminated with ____ that inhibit reactions. hello xrf spectrometer

[coppercone2]

so others don't waste their time, I found this

http://www.dawnmist.org/chemicals/sncl2.htm

Basically the tin chloride I got is bad. i see going through some amazon reviews, there is plenty of bad tin chloride going around from various vendors.

I think because its so simple people are getting tripped up on electroplating, because sodium chloride is so stable, does not mean tin chloride is