Black dots on PCB - what am I doing wrong?

[ChildOfVision]

Hi to everyone!
Long story short: In last dozen of attempts (I'm using toner-transfer method) everything seems fine until I remove toner after apparently successful etching - I got black stuborn dots and marks all over the copper (regardless whether it is a larger copper filling or trace/pad) - please see picture. It is, as I know, black copper-oxide - I can't remove it by any means! It is not "standard" pitting, as someone could conclude from picture - copper is not etched totally in "dot-areas" - it looks like somebody imprinted black particles into healthy copper surface! I tried different plates (from different batch), every sort of papers (including TRF-Green!) but with no avail. Note: after laminating (toner sticks to PCB very well), if I remove toner (without etching) I just get clean copper surface!
View attachment 62790
I'm using muriatic acid (20%) and hydrogen-peroxide (12%) without adding water.
Any opinion appreciated! Especially if somebody had the same problem!

TIA!

 

[duffy]

Why is this blue? The black specks must be toner - did you try soaking the board in acetone to remove them?

 

[Joe G]

it looks like the same thing i get, under a microscope it looks like toner residue, i've also had small letters showing very faintly after i tranfer the layout to the board.
i just clean it very lightly ( i use bar keepers cleaner before i etch it, to remove any toner residue and to double check traces are correct before etching). just go lightly
w/ the cleaning.



and as duffy stated. clean it w/ acitone

 

[ChildOfVision]

It is blue because of: 1) white-balance on my camera was not properly set, and 2) I try different angles to make spots more visible.
Black specks aren't toner (at least I think so), and, yes, I did soak the whole board in acetone - but with no results. And, if it is toner why it isn't in level of copper (better to say: on copper), but rather deeper in copper foil, like somebody pressed them in (if you understand my bad English).

 

[PCBWING]

Chemical reaction is always very complicated. Side reaction or toner stripping is possible.I saw the factory management practices and felt the content of the various components and the temperature demanding requirements are very strict.

 

[duffy]

So it IS pitting. Pitting is often caused by leaving it in the etchant too long. Are you heating the solution and agitating?

 

[ChildOfVision]

@Joe G: Yes under microscope (35-40x) it looks like toner residue, but in your case WAS it toner or something else?

i just clean it very lightly ( i use bar keepers cleaner before i etch it, to remove any toner residue and to double check traces are correct before etching). just go lightly
w/ the cleaning.

Sorry, I didn't understand you well. Google-translated word "lightly" as "gentle", so I don't know where to be gentle? You said that you clean PCB BEFORE etching to remove toner residue - there is no any toner residue on my board(s) before etching, and about traces, sorry I didn't understand you... (thanks anyway!)

 

[ChildOfVision]

So it IS pitting. Pitting is often caused by leaving it in the etchant too long. Are you heating the solution and agitating?

It depends what do you mean under "pitting"; for me where is pitting there is no copper and you can see small hole through copper, especially if you're working with FR4 and inspect PCB toward source of light. BUT, in my case, I can "ferret out" that dot with a needle and then orange/yellow copper appear! So I concluded it is CuO, not a toner. But OK, we can call that "pitting".
I do not heat solution (room temp.) and just swing container slowly.
But it must be something wrong with etching solution, probably it is too strong (too much peroxide?). When I put PCB in the solution it immediately turns bright-green, but etching took too long, over 45 min.!

 

[PCBWING]

Hi to everyone!
Long story short: In last dozen of attempts (I'm using toner-transfer method) everything seems fine until I remove toner after apparently successful etching - I got black stuborn dots and marks all over the copper (regardless whether it is a larger copper filling or trace/pad) - please see picture. It is, as I know, black copper-oxide - I can't remove it by any means! It is not "standard" pitting, as someone could conclude from picture - copper is not etched totally in "dot-areas" - it looks like somebody imprinted black particles into healthy copper surface! I tried different plates (from different batch), every sort of papers (including TRF-Green!) but with no avail. Note: after laminating (toner sticks to PCB very well), if I remove toner (without etching) I just get clean copper surface!
View attachment 62790
I'm using muriatic acid (20%) and hydrogen-peroxide (12%) without adding water.
Any opinion appreciated! Especially if somebody had the same problem!

TIA!

Try Ferric chloride [FeCl3]

 

[ChildOfVision]

Chemical reaction is always very complicated. Side reaction or toner stripping is possible.

I totally agree with you!

Try Ferric chloride [FeCl3]

I will, although it is messy. I can't buy it in my town, and I must have certain PCB "yesterday"...

However, I will try to tin all copper with a solder (to save as much can be saved).

 

[jpanhalt]

It looks like pitting to me, even in the photograph you posted. Some of the spots seem clearly to be indented. "Pitting" does not mean the etch went through the entire copper layer. In this image you can see a similar pitting from a defective mask:

View attachment 62796

I doubt it is a copper oxide on the surface, because both copper oxides are soluble in HCl, which you had an excess of. Of course, in the crevice of a pit, there could be some copper oxide.

To prove it is pitting, make your regular toner transfer. Then use a paint marker over some areas. After etching, I will bet the areas with paint to not have the black spots.

Peroxide/HCl etchant is quite aggressive in my experience. I use the photoresist method. You could actually see it attacking that resist. So, I would second the suggestion to try ferric chloride. You could also try one of those papers that is supposed to be ironed one over the resist to fill defects.

Finally, I found that going over a photomask with a dry erase marker filled defects and gave better images/etches. Details are given in my blog. I have never tried that with toner transfer and can think of some reasons it might not work, but if you are doing some experiments, it might be worth a try on a small portion of the board. I would apply the dry erase to the board after you transfer the image and before you etch.

John

 

[ChildOfVision]

"Pitting" does not mean the etch went through the entire copper layer. In this image you can see a similar pitting from a defective mask...

As first, thank you for clarification about term "pitting". I said before that is not pitting, but after your "definition" I can say -YES it IS pitting (also as Duffy suggested)! And, yes, it is clearly the same thing as in your picture (although it doesn't look "so black" like on my PCB)!

I doubt it is a copper oxide on the surface, because both copper oxides are soluble in HCl, which you had an excess of. Of course, in the crevice of a pit, there could be some copper oxide.

It is not copper oxide on the surface of copper before laminating! I cleaned the copper thoroughly with: fine sand-paper, some dishwasher-detergent and "Scotch-brite" sponge, and finally with acetone - it is very clean and shiny (actually it was pretty clean even before cleaning).
But what are these black specks? If it is toner: why I can't remove it with acetone-soaking? Btw. I also tried much stronger ketone - methyl-ethyl-ketone - soaking over night - no avail! If it is any of copper-oxide it should be soluble in HCl (as you point) - but that spots won't go out even after long soaking in HCl ?! Is it any way to get rid of that stuff - H2SO4?

To prove it is pitting, make your regular toner transfer. Then use a paint marker over some areas. After etching, I will bet the areas with paint to not have the black spots.
You could also try one of those papers that is supposed to be ironed one over the resist to fill defects.

I will try! Did you mean real paint or "permanent-marker" pen (like "Sharpie" or "Edding") because in my experience "Sharpie" or similar has no chance in HCl+H2O2 ?
This board, I took picture of, is actually made with TRF-Green foil! I don't have any pic. of it, but it looked really great! Perfect transfer + "embossed" layer of TRF! But it seems that even TRF-Green is too sensitive to HCl solution! After etching - no any signs of pitting or any fault! So, I drilled it, and then remove toner+TRF and only then I saw that black dots!
Some time ago, I tried to make PCB (about 150x100mm) with shiny leaflet paper I took from some store - that was total disaster; even during etching I realised something is wrong! I can't find this board right now, so I can't show you how it looks, maybe later! What I got, was not just black specks, but more like black "sinus"-wave-lines, more or less parallel along the whole board! That "lines" are in the same direction as the board entered laminator (if you understand me) - obviously, the bad paper wrinkled during lamination and made things even worse!

Peroxide/HCl etchant is quite aggressive in my experience. I use the photoresist method. You could actually see it attacking that resist. So, I would second the suggestion to try ferric chloride.

I agree! I think peroxide is more problematic! I know for many "recipes" how to mix HCl and H2O2, and I always wondered why, for example, for 1 liter of solution you have to add about 700ml of (preferably distilled water) - not counting the water contained in, say, 20% HCl solution! Why so much water? It seems that "pure water" plays bigger role than bare concentrations of HCl&H2O2! Currently, I'm using HCl(20%) and pharmacy grade 12% peroxide WITHOUT adding any water! One more thing: there are not any little bubbles (as I seen (long time) before, when I used to add extra water)! The solution turns light green immediatelly after putting the board in, but etching took over an hour!

About dry erase marker: (did you mean whiteboard marker?) - yes it can improve photomask, but I doubt it could survive, even in pure water. (But I can try!)

Thank you for your suggestions!

 

[jpanhalt]

TRF-green is one brand of what I was referring to as an iron-on after toner transfer.

Dry-erase marker = white-board marker

I did mean "real" paint, not a standard permanent marker. DyKem makes a product called Brite-Mark. Marvy (Uchida of America, made in Japan) makes a marker called DecoColor. The DecoColor is available at craft stores in the US. I get Brite-Mark at welder and industrial supply places. It may also be at craft stores. What you find may be labeled for writing on greasy surfaces. They have a little ball inside for mixing, so they rattle when shaken.

As for the dry-erase, I don't know its chemistry. The carrier solvent smells a little like an acetone/ethanol mixture. Some glyme type of solvent may also be present by smell. I just tested the dried material. The black does not appear to be soluble in water. But it is, of course, easily rubbed off and water helps in that process. I don't think it will work with peroxide, but it may with ferric chloride. It is not for large areas. It only fills microscopic pits, so it does not need to be too rugged. Again, I can think of may reasons it won't work. But there is only one way to know for sure.

John

 

[ChildOfVision]

John, I can "smear" visible pores on the toner with Edding permanent marker instead of whiteboard marker. Anyway, I think I got some clue! I just found on the (plastic) bottle with my HCl that it is produced 1997 (!) and shelf life is 3 years! If so, my HCl is expired for more than 11 years!? Did you know that muriatic acid has shelf-life? I didn't! I will now try to measure pH of it, and tell you (all) result. So, I still think that black spots are CuO, a mineral also known as tenorite (Wiki Tenorite). If I really have too weak acid, I can't get CuO dissolved! So I must repeat soaking PCB in fresh HCl, to see if it is really tenorite!
The question is: what causes forming of tenorite? Some time ago I put blank piece of PCB into pure peroxide, and copper just turns brown (as it should). So, peroxide alone is not strong enough oxidiser to make "higher grade" oxide of copper, i.e. CuO.
Another thing (maybe you can figure something out): during one of my attempts to make PCB (with this weak (as I think) HCl) I put some HCl into container and added very little peroxide (no additional water); I put PCB in that solution and color of a liquid almost instantly turned to light green BUT copper didn't change color to brown (as usual)! I waited some 10 min. without any reaction. So I concluded there must be too little peroxide; I added a bigger "splash" of peroxide (and a little HCl) - the whole PCB changed color to BLACK!? I realised that PCB is already ruined enough so I remove half etched board from solution and saw that all remained copper is BLACK!? Under running water almost complete toner together with protective layer, simply detached from board in large "patches". Maybe deteriorated HCl "enriched" solution with additional oxygen that, in combination with peroxide, caused such a strong oxidation? I don't know, I am not a chemist.

 

[PCBWING]

John, I can "smear" visible pores on the toner with Edding permanent marker instead of whiteboard marker. Anyway, I think I got some clue! I just found on the (plastic) bottle with my HCl that it is produced 1997 (!) and shelf life is 3 years! If so, my HCl is expired for more than 11 years!? Did you know that muriatic acid has shelf-life? I didn't! I will now try to measure pH of it, and tell you (all) result. So, I still think that black spots are CuO, a mineral also known as tenorite (Wiki Tenorite). If I really have too weak acid, I can't get CuO dissolved! So I must repeat soaking PCB in fresh HCl, to see if it is really tenorite!
The question is: what causes forming of tenorite? Some time ago I put blank piece of PCB into pure peroxide, and copper just turns brown (as it should). So, peroxide alone is not strong enough oxidiser to make "higher grade" oxide of copper, i.e. CuO.
Another thing (maybe you can figure something out): during one of my attempts to make PCB (with this weak (as I think) HCl) I put some HCl into container and added very little peroxide (no additional water); I put PCB in that solution and color of a liquid almost instantly turned to light green BUT copper didn't change color to brown (as usual)! I waited some 10 min. without any reaction. So I concluded there must be too little peroxide; I added a bigger "splash" of peroxide (and a little HCl) - the whole PCB changed color to BLACK!? I realised that PCB is already ruined enough so I remove half etched board from solution and saw that all remained copper is BLACK!? Under running water almost complete toner together with protective layer, simply detached from board in large "patches". Maybe deteriorated HCl "enriched" solution with additional oxygen that, in combination with peroxide, caused such a strong oxidation? I don't know, I am not a chemist.

I think the "splash" was a bad thing. You should try to know the proportion of the various components, at least use the graduated cylinder to add the appropriate ingredients and uniform mixing.

 

[ChildOfVision]

@PCBWING:
Can you please help me about proportions of HCl, H2O2 and water? I searched alot but I get very different "recipes". Furthermore, toner is IMO much sensitive to the etching solution than, say photoresist, and John (jpanhalt) said he has problems even with photoresist! So, can you PLEASE tell me, for my combination: HCl-19%, H2O2-12% what volumes of each to measure? You don't need to be precise. I hope you are working with toner transfer method - in that case your proportions would be "very proven"!
And, is it advisable to use foam-etching (for both ferrichloride and HCl&H2O2)? I can easily build vertical (or horizontal) cuvette, I have all material (except heater).

TIA!

 

[PCBWING]

Chemical equation

Hi, ChildOfVision.
I found this chemical equation, and I hope it would be useful to you.
View attachment 62815

 

[ChildOfVision]

Thank you very much for your efforts! I will study now that equations!

 

[jpanhalt]

I found this chemical equation, and I hope it would be useful to you.
View attachment 62815

Are you sure the principal gas evolved is chlorine and not oxygen?

Why isn't the reaction just:

(1) H2O2 + Cu + 2HCl ==> CuCl2 +2H20
(2) H2O2 by a variety of reactions ==> H20 + O2(gas) (See: more examples below)

Lots of equations can be written for the reactions of copper, HCl, and H2O2. For example, one can easily find examples in which hydrogen peroxide in aqueous solutions will oxidize HCl to give chlorine gas. Chemists are accustomed to thinking that electronegativity increases as one moves to the right in the periodic table and decreases as one moves down. Chlorine is both to the right and down one row from oxygen. So, is it more or less electronegative than oxygen? By experiment, chlorine is slightly less electronegative than oxygen. Thus, from that consideration alone, one can understand how oxygen or peroxides can oxidize chlorides to give chlorine (gas), which is consistent with the equation you show. However, we know from everyday experience that the oxidation of ordinary table salt (NaCl) in water to give NaOH plus chlorine(gas) is a very slow reaction.

Thus, the question is not whether the reaction you suggest can occur, but rather, which of the several other known reactions is the predominant reaction under the conditions of light, copper ions, and acidity present in the etching process. One needs also to consider whether the reaction is purely ionic or whether other species, such as free radicals or excited oxygen are involved. (For more on that, read DuPont's material on its potassium persulfate etchant.)

Here are some additional reactions to consider:

(3) 2Cl2 +2H20 ==> 4HCl + O2(gas)
(4) Cl2 + H20 ==> HOCl + HCl
(5) HCl + H2O2 ==> H20 + HOCl
(6) HOCl + H2O2 ==> O2 +HCl + H2O

Chlorine is quite soluble in water, and the reactions in equations 3 and 6 are catalyzed by copper ions.

From what I have read, while the actual chemistry of the etching process may be complex, the empirical finding is that the predominant gas evolved is oxygen, not chlorine.

John



BTW, Those animations are cute and can be potentially useful. In this instance, it made reading and analyzing the equation you propose difficult. Is there a way to freeze them?

Note: HOCl is also written as HClO; I prefer the former.

 

[duffy]

which of the several other known reactions is the predominant reaction under the conditions of light

Any way you could guess what wavelength of photon would stimulate this? If you could stick a blacklight, or a germicidal lamp or something in there and speed it up, that would be pretty sweet.