Restore dirty/corroded IC pins

[tony359]

Hello all,

I have a 1980 PCB which I am going to wash into a PCB cleaner water-based fluid from Electrolube (SWAS05L) which I have tried before and I found excellent. Some of the IC's are not shiny anymore and most show some corrosion on the legs. I tried with a fiber pen and while they get better, they don't come back shiny.

Someone on a forum mentioned using TARN-X but I believe it is forbidden in Europe because of the chemical it uses (Thiourea). It is also made of Sulfamic acid so I purchased a small bag of it, made a solution and tested with some spare chips I had, along with a copper coin. Unfortunately the Sulfamic Acid doesn't seem to be very effective - or maybe the solution is just not strong enough.

Can anybody recommend a way to clean those chips?

Thank you!
Tony

[helius]

The first port of call for cleaning corrosion is acetic acid, for a very simple reason: it is volatile and does not leave an acidic residue.
You can use other organic acids like citric or oxalic, but they need to be thoroughly rinsed to remove all residues. A popular rust remover is based on tannic acid, another choice.
Tarnish removal products are designed to remove tarnish on silver, which can be very tough and resists most chemicals. Many silver specialists, however, caution against Tarn-X because its ingredients (thiourea and sulfamic acid) contain sulfur, which accelerates silver corrosion. I believe similar concerns may disqualify it for use on electronics.

[TimFox]

Any suggestions for liquid cleaner for tin-plated surfaces with mild corrosion from long storage?

[tony359]

Thanks Helius

I see Vinegar has 8% of acetic acid - do you think that would work? What would be your choice between acetic, citric, oxalic and tannic?

Thank you!

[DrG]

It really does not sound that bad - can you post some clear pics? Are they dull or corroded?

This is corrosion:
https://www.eevblog.com/forum/index.php?action=dlattach;topic=235362.0;attach=959522;image

and BTW I did try EZest (something like tarnX) and it did not do much, but I was able to get the chip to work

What I am getting at is that if you have already used a fiber pen (an abrasive I am thinking), then some acetic acid like in vinegar or citric acid in a lemon might help a bit - as has already been suggested.

They are 40 years old and I don't know how you can get them shiny new, or why they need to be.

I might dab some 90+% isopropyl alcohol on them to remove anything on them that might enhance a reaction.

[tony359]

They are 40 years old and I don't know how you can get them shiny new, or why they need to be.

Same reason why I am going to wash the board: I am restoring this machine and I'd like it to look as good as possible. I'll retrobrite the plastics too. I appreciate I may not be able to remove 40 years of time of course - I'm just searching for the best method. I'll share a picture tomorrow but no, they are not so corroded as in the link. Maybe "stained" is a better word

[bitter_mike]

If this is an IC from the 1980s, the leads are most likely plated with tin/lead (90/10 was very common back then). Liquid solder flux would probably do the job. Afterwards, you would have to clean that up with a bit of solvent like acetone.

[tony359]

Here are a couple of pictures.
The one with black leads was easy to clean and the legs come back brand new. The one in the socket won't come up as nice even if I scrub it with the fiber pen. A couple of chips have been bathing in Vinegar for a while now - unfortunately I do not see any changes. I have some citric acid in granules I could try.

Bitter_mike
Do you mean just applying the flux or applying the flux and then warming the leg with the solder tip?

[bitter_mike]

I meant to just apply it and scrub it off, but now that I'm seeing the pictures I don't think that would work. That white stuff is a mixture of tin(IV) oxide and lead(Iv) oxide. As a general matter, acids will not dissolve them. You need a solution containing chelating agents, like EDTA or glucconate. In terms of commercial products that might do it, there's a product in the US called "Calcium Lime Rust" or CLR. I don't know if its available where you are but I suspect there are similar products where you are. They are meant for getting scale off faucets and such. Those have the same problem as you: insoluble salts of various metals which are rendered soluble by the chelating agents in the product.

[DrG]

I don't exactly know why (maybe because I have some experience with coin collecting where "dipping" is widely practiced with silver but almost all other kinds of cleaning is strictly a no-no), but I am really enjoying this thread. I do hope that you will keeps us apprised of what works and what does not.

[tony359]

There is a similar product in the UK but it seems to be based on phosphoric acid.

http://zepcommercial.co.uk/products/kitchen-bathroom/calcium-lime-rust-remover/

This one seems different though
https://www.cleaningproducts.net/chemspec-one-earth-calcium-lime-amp-rust-remover-1-litre?gclid=Cj0KCQjw3Nv3BRC8ARIsAPh8hgJyQtzZCwK6naej9g37VEXkRdSCRxbR1mF4d31BnrhUC5UuSPPbkjIaAi2jEALw_wcB

Finally this one which is a rust remover though - non acidic
https://lilleytileandstone.co.uk/ecoprotec-rust-stain-remover-1ltr.html?gclid=Cj0KCQjw3Nv3BRC8ARIsAPh8hgIuDQYnZnVgJvHCE76vTj89OlQDzcMU1aRlKdK7-KvCIxCylBlbL1QaAkFQEALw_wcB

What would be a key component I should look for?

[bitter_mike]

The two most common chelating agents in products like these are EDTA (ethylenediaminetetracetic acid) and gluconic acid (or more commonly, salts like sodium gluconate). The trouble is that in most jurisdictions, they don't have to disclose them on the MSDS so its hard to know if they're in there. CLR is just such an old product that it's formulation is known (it also used to contain phosphoric acid).

I would give the first one a try. If it can dissolve calcium salts, it has some kind of respectable chelating agent in it. Calcium salts are insoluble without one.

[tony359]

I've got the Zep cleaner. While it's very effective on a copper coin in just 60 seconds, it's not really doing anything to my test chips. On one, it wiped the little identification text though
The picture of the chip was taken after 3-4 minutes submerged in the fluid.

[bitter_mike]

Ok, so my next instinct here is to mix something up when I go back into the office on Tuesday and run a few tests on some straight stannic oxide. Are you committed enough to doing this that you'd be willing to buy some stuff off Amazon and mix it up or is that too far for you?

[tony359]

If it's not too expensive I can definitely give it a go!

[Ian.M]

First you need to know what you are dealing with.  Its a hermetic ceramic package so the leads are probably some sort of nickel iron alloy (a close coefficient of thermal expansion match between the leadframe and the ceramic is essential) e.g. see https://www.astm.org/Standards/F30.htm

They were originally tinned, probably a tin lead alloy as 'bitter_mike' pointed out, as even back at the beginning of the silicon era, the problems with whisker growth from pure tin coatings were well known to metallurgists with experience in semiconductor packaging.  The dark spots you are seeing are probably where the original tinning is completely gone, probably due to oxidisation creeping under it from a scratch at an edge, then it simply flaking off.

To restore them, you'd need to get the pins clean and oxide free, then re-plate them.  Electrolytic tin brush plating one pin at a time would be one approach, taking care *NOT* to pass current through the IC , or electroless copper then electroless tin over the copper.   I'd be very wary of re-tinning them with solder paste and hot air due to the risk of heat damage.  Flash plating with copper to make them wettable and re-tinning each pin individually with a temperature controlled soldering iron and lots of flux might be an option.

N.B.  Restoring the plating on the pins is relatively high risk - you can make it look good but may well kill the IC in the process.   If you use a pure tin plating even if the IC initially survives without damage, you are setting yourself up for tin whisker problems in the future that could short out and kill the IC.

[tony359]

You know, I was thinking just today that those stains may not be just stains or oxidation but areas where the original coating has now gone. And indeed I was thinking that I can brush and clean as much as I want but they won't come back shiny.

Don't get me wrong here: my purpose was to see if an easy and riskless method was available to quickly clean old chips. I'm not going to attempt to to re-coat corroded ones. Happy to do more tests bitter_mike but I just wanted to make this clear

[bitter_mike]

Ian might be right about the plating being gone. It's hard to tell without throwing it under an XRF or in an SEM and checking. Either way, I don't think it would hurt to try what I was thinking. Even if the original plating is gone in those areas, it may clean up the darker oxides. They will just grow back over time, but it won't make it any worse.

[helius]

I'd be very wary of re-tinning them with solder paste and hot air due to the risk of heat damage.  Flash plating with copper to make them wettable and re-tinning each pin individually with a temperature controlled soldering iron and lots of flux might be an option.

I had a similar idea, although mine was to spray a thin flux coat over the chip and then dip into a solder pot. (Is there anything on the underside that could bond with solder? That might be unwise if so). Note that iron may require more active fluxes to wet properly, which could be a future corrosion risk if any residue remained.

[CatalinaWOW]

Since the goal here is apparently purely cosmetic have you considered coating the whole board with a insulating coating and then just painting the pins to the appearance you want.

[tony359]

Catalina,
I suspect that would make future service a total nightmare I'm sure that Apple //e will require more TLC over time

bitter_mike: by all means. I'm ok to proceed! We may as well find a method that works well for other chips in better shape.

[bitter_mike]

So I have bad news. I tried a few things and while the thermodynamics are on our side, the kinetics are not. I made up some mixtures using both EDTA and gluconate and the only way I could get any stannic oxide to dissolve was by heating them quite hot (50 degC) for a long time (several hours). Even then, I didn't get total dissolution. The only thing that dissolved it quickly was using a zinc powder / hydrochloric acid mixture I found in the literature and you don't want to mess with that. Lots of hydrogen comes off, the zinc powder is a fire hazard in its own right and HCl is nasty business too. This is to say nothing of the damage which might occur to the chip during this process.

Needless to say, this is why we try to prevent the formation of stannic oxide in tin baths as much as possible and use filtration to remove it rather than trying to dissolve it.

Sorry I don't have better news. The only way left I see is via some mechanical polishing or something.

[tony359]

Hey Mike,

I do appreciate the time you spent on this, thank you! Bad news is still good news: your very scientific approach just confirms that I do not have to spend further time and money on various chemicals and that to me is a "good news". As you say I plan to polish the chips manually with the fiber pen and then wash the whole chips and PCB in Electrolube SWAS05L  which is a water based PCB cleaner that I really like and works wonder.

Again, thank you for lending your skills to this thread!

[fi11222]

Hi, even though I am quite late to the party, I have decided to add my 2 cents here bc there does not seem to be much resources on the Internet on this topic apart from here.

So, after reading this thread I tried scotch-brite and it worked pretty well so I decided to go full industrial and bought a Dremel-type grinder with tips like these (first img below)
" alt="" class="bbc_img" />
(On the left, new. On the right, after polishing 12 ICs)
On Amazon: https://www.amazon.in/gp/product/B086LFXM9N/ref=ppx_yo_dt_b_asin_title_o03_s01?ie=UTF8&psc=1
They are basically Scotch-Brite on a wheel. Quite nice, but they wear out fast.

I have found a batch of old Soviet TTL 74xx clones and many of them have badly corroded pins so I needed something that would do the trick in bulk. The main concern was electrical contact. Before polishing, the ICs failed on the tester. After, they passed.

In order to hold the chips and be able to treat several at a time, I built this little rig (img 3 below)
" alt="" class="bbc_img" />

Here is a before/after comparison (img 2 below)
" alt="" class="bbc_img" />