Copper Pyrophosphate Plating: Orthophosphate & potassium pyrophosphate in bath

"Copper Electrodeposition for Nanofabrication of Electronics Devices"

on eBay or

AbeBooks

or Amazon

(affil links)

A. Hi Samer,

Harry last posted on this thread 13 years ago, so I'm confident he doesn't monitor it. Before we try to contact him this much later, please try to be more explicit about what you are looking for -- are you looking to retain his services as a consultant, or are you looking for free help?

If you are looking for free public help, hundreds of other people will read your posting and may be able to help you as well -- but that requires that you carefully detail your situation please because "yes", "no" or "maybe" as an answer won't help much :-)

Thanks & Regards,


Ted Mooney, P.E. RET
Striving to live Aloha
finishing.com - Pine Beach, New Jersey

Viscosity Tester

on eBay or

Amazon

(affil links)

A. Hi Dante,

Pyrophosphate decomposes to orthophosphate and there will always be some present. In small to moderate concentrations it has no effect on performance. However as it increases it begins to reduce the current density at which "burning" occurs. This is because the orthophosphate significantly increases the viscosity of the bath and reduces the transfer of copper and brightener to the plated surface.

There is no way of converting the ortho back to pyro in the bath. The only remedy is a part/total dump. The factors which affect the rate of decomposition are low pH (less than 8.6) insufficient anode area (very bright, electropolished, anodes) and high temperature.

In well maintained baths used for thin deposits, the drag-out will normally keep the orthophosphate at an acceptable level. But in electroforming baths (thick deposits), drag-out for a given plated surface area is much, much lower and the probability of too much ortho accumulating is greater.

I don't know what Atotech's figure of "7" means. I have some guidelines for acceptable ortho and if you give the complete analysis perhaps I can extract the information I need to make a useful comment.

Harry

"... perhaps I can extract the information I need to make a useful comment."

Yes, Harry, more info about this "7" would be helpful, but I for one have already found your posting very enlightening. Thanks!

Regards,

Ted Mooney, P.E.
Striving to live Aloha
finishing.com - Pine Beach, New Jersey

A. Yes Dante this is the information required.

Now that we know that the "7" has units of "oz/gal" it is possible to comment. 7 oz/gal (52 g/l) orthophosphate is very high and is most probably the cause of your problem. I regard 4 oz/gal (30 g/l) as being a reasonable maximum value. There is no hard and fast figure. Some users who plate simple shapes and use relatively low current densities might have no difficulties at 7 oz/gal. But for most users it is too high.

As I said previously, dumping is the only answer. There is much to be said for frequent small dumpings to keep the bath in a constant condition rather that doing total dumps and then allowing the orthophoshosphate to rise to critical level again. If it were my bath I would dump half.

The question of pH is a vexed one. A ph of 8 is probably best for plating. However the rate conversion of pyro to ortho increases at lower pH. Above 9 the stability is better but plating performance especially at lower current densities falls off. My preference is for 8.6 to 8.9.

Anode current density is important. With too high an anode CD (low area) the copper content and pH will fall and the anode electropolishes. Too high and the opposite happens and the anode becomes coated with a brown sludge: the brightener may precipitate on this sludge making control difficult. The ideal condition is a slowly rising pH and anodes having a clean pink appearance with slight brown film revealed when wiped with a finger (like light dust on furniture).

If this were a bath doing typical decorative plating I would blame bad control for the ortho being so high: most such baths never need to be dumped. But with electroforming and little drag out, a significant rise is likely. Whether your rate of rise is excessive is impossible to say.

Finally, you are probably thinking about cost of replacement and disposal of waste: I don't have any answer except "grin and bear it".

A. Hi Dante,

No, you will not have achieved a dump. The concentration of ortho (and everything else) as you evaporate down will rise and then fall when you top-up with water. There will be some break down of pyro to ortho, as already explained, causing the ortho to rise. The only way to bring down the ortho is to remove all or part of the bath and replace the lost water, copper and potassium pyrophosphate salts and other additives.

If the bath is not needed, allow it to cool. The break down of pyro is much, much lower at room temperature.

It is important to bring the bath volume to the working level and stir well before providing a sample for analysis.

The change in pH that apparently took place during the period when the bath was unused would certainly promote a rapid increase in ortho generation. But I don't believe that it happened. Much more probable is an error in measuring. It would take a lot of acid to cause such a fall. I'll make some comments on the measurement of the pH of pyro solutions if you think it would be helpful.

I can't comment on the swishing sound. I can only say that high ortho will reduce the quality of the deposit. Sounds (if you will excuse the pun) like a good research project.

I am curious as to the errors in the Atotech analysis. Will you let us know the new results when you get them? As a check on the analytical result for ortho I calculate the expected specific gravity and compare it with the actual value. In some ways this calculation is more useful because it includes any sulphate from pH adjustments and will also affect the viscosity. If you wish, I will crunch the numbers for you if you supply the copper (Cu), Pyrophosphate (P2O7) and SG measured at standard temperature. You are welcome to the calculation method or even to have it as a small executable file.

A. Hi again Dante,

Pitting is not normally a problem with copper pyro. Inadequate cleaning is usually the cause when it occurs.

Orthophosphate is not a contaminant as, for example, copper in nickel plating solutions. It increases the viscosity of the bath and reduces the rate of replenishment copper and brightener at the cathode. This limits maximum current density and brightening/grain refining. This could be your problem. You did not say in your posting whether you had done a partial dumping: I guess not.

You do not need to analyse for orthophosphate in order to an idea of the concentration and the likely effects. Here is a simple calculation:-
Divide the pyrophosphate in g/l, expressed as P2O7, by 687 and add 1. You will get the specific gravity of a new, ortho-free bath. For example, a bath containing 166 g/l P2O7:
166/687 = 0.24. Add 1 = 1.24 = calculated SG
If you prefer US oz/gal, the figure is 92 rather than 687.
Then compare the calculated value with a measured one.

Here is a an extract from a computer generated report (saves me retyping). It partly repeats comments given above above:

The difference between is measured and calculated SG is due to mainly orthophosphate. The effect of high ortho is to increase the viscosity of the solution. This limits the maximum current density before burning. There are no hard and fast rules. But, as a guide:-
A difference of less that 0.05 has little effect.
A difference of 0.05 to 0.10 will have growing effect.
A difference of 0.10 to 0.15 becomes increasingly severe and there are few circumstances where the bath will be useable without significant loss in performance.
The only way to remove orthophosphate is by dumping all or part of the bath and adding new chemicals.
Well maintained baths, especially with respect to pH, will run without the need to dump.

A. Hi Dante. Harry probably missed your followup (it was 2 years later), and I'm not expert on this ... but I think that although you can calculate the expected specific gravity of a new (pure) solution as Harry described, that you must measure the S.G. of your current (extant) solution, either with a hygrometer or by weighing a given volume of solution. The milligrams divided by the cubic centimeters is the specific gravity.

Regards,

"Copper & Copper Alloy Plating"
by R. Pinner

on eBay

or AbeBooks

or Amazon

(affil links)

A. Copper pyrophosphate is electrolytic, not electroless. It may (although I haven't heard of it) be possible to use hypophosphite as a reducer in a copper electroless bath (generally, HCHO is used as the reducer).

A. I don't know of an electroless copper bath based on pyrophosphate, but that isn't to say it doesn't exist. However, all pyrophosphate baths should be run at less than 60 °C otherwise the pyrophosphate hydrolyses and dissociates to orthophosphate. Hope this helps you.

A. Dear colleague, if you support pH 9-10 and the temperature 40-50 °C, you may use this bath for ages.

A. Hi Wade,

Are you trying to dissolve the potassium pyrophosphate at room temperature or are you bringing the temperature of the solution up at all? You should bring the make up water up to 50 °C before adding any potassium pyrophosphate. You will find it difficult to get what is in solution to dissolve, even with warming now, don't ask me why, but it is a devil to get to dissolve when wet.

A. Hi Wade,

I agree with Brian's comments and yet something does not feel right. If the potassium pyrophosphate if finely divided and does not clog the filter then it should dissolve. I suspect that the insoluble matter is not potassium pyrophosphate. I think that either you are not adding potassium pyrophosphate or you have insoluble matter which you are confusing with the potassium pyrophosphate that you have added.

Try a lab beaker scale test with hot copper pyrophosphate solution and adding potassium pyrophosphate slowly with good stirring and see if you still have undissolved "potassium pyrophosphate"

Best wishes

A. This response is three years late, but for the sake of completeness and to assist others who may search this, here we go: The ratio going low indicates copper concentration increasing due to anode dissolution. If your anode area is too high this will happen rapidly. Try reducing your anode area to slow the ratio decrease. Harry also discussed this earlier in the thread.