A Deep Dive Into The Chemistry Of Retrobright

Considerable effort is often required to rejuvenate the yellowed and grungy plastic cases of retrocomputing gear. One generally does well to know their enemy in order to fight it, though, which is where this guide to the chemistry of plastic yellowing and whitening (PDF) comes in handy.

“The Retrobright Mystery” was written and sent in to us by [Caden Xu], a high school student who also goes by the alias [Saltypretzel]. The paper begins with an excellent description of the chemistry of plastic yellowing. We had always heard that the yellowing in ABS, or acrylonitrile-butadiene-styrene, the plastic most commonly used for cases back in the day, was primarily caused by brominated compounds added to the plastic as flame retardants. It turns out that’s only a minor contributor, with the bulk of yellowing occurring thanks to a complex chain of reactions starting with free radicals liberated from the butadiene copolymer through a reaction requiring oxygen and energy.

Reactive radicals from the decomposing synthetic rubber, added to ABS to increase its flexibility, unroll the benzene ring in styrene copolymers to form a conjugated compound called 2-hydroxymuconic acid. The alternating double and single bonds in this compound tend to absorb light towards the blue end of the spectrum strongly, so the accumulation of 2-HMA in the plastic over time thus makes it reflect more and more yellow and red wavelengths, giving aged ABS its unhealthy bronze glow.

Luckily, just as ketchup smears and grass stains, both rich in conjugated compounds like lycopene and chlorophyll, can be bleached out of existence, so too can yellowed plastics. [Caden] notes that Retrobright, which contains a powerful dose of hydrogen peroxide, does its whitening trick by breaking the UV-absorbing double bonds in 2-HMA. There’s little that can be done about the embrittlement of the ABS caused by the breakdown of butadiene copolymers, but at least it’ll look good.

We found this guide quite comprehensive and instructive, and it should only help retrocomputing fans in their restoration efforts. For those less interested in the chemistry, [Bob Baddeley] published an overview of the yellowing of plastic and manufacturing steps to avoid it, and we covered the more practical considerations of Retrobright treatment too.

48 thoughts on “A Deep Dive Into The Chemistry Of Retrobright

  1. “There’s little that can be done about the embrittlement of the ABS caused by the breakdown of butadiene copolymers”

    I’m interested in knowing what little CAN be done. Got a couple of items where the plastic seems to now have the strength of chocolate, hard chocolate maybe at reasonable temps, but yah, really weak.

    Also need ways to stop that stupid rubberised coating turning to tar on various stuff.

    1. Definitely. That is something that we need to figure out or else our computers might turn to dust in a few years. Storing computer equipment in gases like nitrogen and carbon dioxide will probably prevent yellowing as there is no free reactive oxygen there to oxidize the butadiene, but that is quite impractical and expensive.

      1. Vacuum storage bags are getting quite popular now, for blankets and such, which you can find at the dollar and discount stores at about $2 a bag. I would presume those are airtight. So maybe storing stuff in those and displacing the air out of them with a CO2 cylinder or nitrogen cylinder would do the trick. Though if you wanna take them out when you feel like it, it’s kind of a PITA each time.

        1. While air-tight packaging might prevent oxidation, it could be harmful to other components. Leaking plasticizers can damage rubber and other materials, turning them into goo or dust. This is known to kill belts and rollers in tape recorders and to severely shorten the life of electrolytic caps (most of them have a rubber seal). Storing old electronics in a cool place (and not in the attic) is the better option, this will slow down all chemical reactions.

          1. Thanks for your comment. I’m thinking about storing modern electronic devices for long time (10+years) data retention purposes in a vacuumed bag. The devices I want to put are a HDD, a plastic SD card, an aluminum USB key and a plastic LTO tape. Would you recommend not using vacuum in this case ? FWIW I’m adding a 49% humidity regulator pouch in it too to stabilize humidity.

        2. I wouldn’t vacuum bag the item directly. The atmospheric pressure on the bag will crush or deform the thing you want to protect, better off getting a can of freeze spray or canned air, put your item in a sealed box with two blockable holes at the top, and fill it up generously with freeze spray or canned air, as it is usually nitrogen (check your brand label). I use this to purge air out of paint tins as the cold gas from freeze spray does a good job of staying heavy and push the rest of the natural air out before sealing the lid.

    2. Some of my black and Decker hand drills got sticky. I used baby powder to infuse the sticky bits and, even after the baby powder rubbed away, the handles are not sticky /tarry anymore.

        1. I think it’s just adhering to the surface of the sticky coating. I’ve done the same with corn starch while doing roofing jobs with tar, glue, and rubberized coatings. If it gets on your clothes/tools/hands, rub in corn starch (or talc, chalk) until there is no more possible sticky area to adhere to.

          1. Corn starch is also fantastic for soaking up engine/gearbox oil spills on garage floors. Sprinkle it on, leave it to soak up the oil for a day or two then use a blade scraper to lift it up. It is still a (lumpy) powder at that point, so you can spread it back and let it soak up some more. Eventually it becomes fully saturated so just scrape and dispose.

    3. I have been using clear epoxies for years and more recently clear Gorrila Glue painted on in thin layers with various size brushes to reinforce brittle plastics. I try to put the layers on the hidden areas when possible because I usually end up with a glossy finish which is not always desirable. Clear nail polish too sometimes If the solvents won’t harm the material. Works great much more often than not. Good Luck.

      1. Titebond III might be a candidate for that too. Now in vintage popup camper communities, there’s a couple of schools of thought about preserving the ABS “road covers” or what are the actual roofs really, one says to paint it if it’s not yet deteriorated, they see more sun but are a lot thicker than many casings, but don’t get contact wear, per se. The other school says to paint it all over with LABS, liquid ABS, which is ABS in solvent, this material also being used to help patch up any extant cracks, preferably in thicker form than would be appropriate as a coating. However, for such an application, there’s not much worry about losing fine detail, you just want the same general shape and it to stay dry underneath. Whereas with computer and other casings you’d tend to lose design details like lines, moulded text, grille shapes etc if you put much of that on. Possibly it would work to reinforce from the back, where there is room, maybe shore up screw posts etc so they don’t rip out.

    4. Also need ways to stop that stupid rubberised coating turning to tar on various stuff.

      I don’t know how to stop it but I know how to clean the tar (and thus stopping the reaction since it removes some material). I use isopropyl alcohol and a magic eraser and that works like a charm. Used successfully on old Thinkpads and various plastic enclosure with a 100% success rate.

    1. Ah yes, the “new information comes in, but I’m gonna parrot the old wives tales anyway”. It’s especially hilarious because the two points you state are mutually exclusive. The yellowing returns because retrobriting is a surface treatment, but the yellow compounds are diffused deep below the surface. Over time they diffuse back out to the surface. And *since* it’s the case that retrobriting can only affect the surface, how exactly is the plastic becoming more brittle? It was *already* brittle. You just don’t realise until you break something, which usually happens when you’re in the middle of a restoration project.

      1. Both yellowing and retrobright should all be superficial processes. Plastics contain oxidation stabilizers inside of them to make sure that the yellowing stays at the surface. Some plastics are not stabilized properly. In that case, the yellowing can spread throughout the entire plastic. Retrobright should also be superficial, unless the plastic has cracks inside of it. The small superficial fissures from retrobright could lead to growing cracks, but they aren’t immediately bad to the plastic. Speaking about yellowing coming back, yellowing can come back in two forms. It can come back from the isomerization of the muconic acid molecules, or they can come back from further oxidation of the plastic, where the latter is the more detrimental.

  2. I’ve read somewhere that peroxide in retrobright process can be replaced by ozone. Which would be great because we can make retrobright oven that does not need any consumable refills, since both ozone and uv light can be made using electricity. And it would also be useful for sterilizing during pandemic. Also buying ozone generator is easier than getting concentrated peroxide and will not get you listed as a terrorist. Only downside is that you need to make sure not to breathe ozone.

    1. I had varied results with my ozone attempts. It made a bit of a difference with my A1200 case but did little to the keycaps. I have an ozone generator with a fan that I used in a large cardboard box with the plastic in. I dunno if having it better sealed would have helped, or possibly purging it with CO2 first. I’m planning another try with some UV light as I read that combining UV and peroxide or UV and ozone improved the process.

  3. I wonder if spraying a UV blocking clearcoat would help? Createx has recently released a range of UVLS clears that spray really well with an airbrush if thinned about 10-30% with 4011 flash reducer. I’m about to replace the palmrest and screen bezel of my Thinkpad T420 (they used a very soft plastic for these) and plan to spray them before installing in the hopes that they’ll last longer. Might spray a rubberized coating too.

    1. I’ve heard positive results from clearcoating things, unfortunately the paint means you generally can’t use the device without worrying about wearing the paint off and ending up with uneven yellowing

      1. Another thing I’m experimenting with on 3D printed parts is Solarez “I can’t believe it’s not lacquer” UV-cured resin. It goes on very thin and is so hard I doubt it would ever rub off. That would also address the embrittlement problem. What I haven’t figured out yet is what sort of pre-treatment the plastic needs in order for the resin to coat nicely. So far I’ve been getting a lot of issues where the resin retracts into pools after being applied rather than settling out.

  4. Given the fact that with time all old plastics will wind up extremely yellow, and retrobrighting will just serve to weaken the plastic to the consistency of chocolate over time, having found a set of dies for the Commodore 64C case was a god-send. Now all we need are a set of dies for the Key Caps and we’re gold.

    Given the fact that there are people out there with some pretty decent CNC machines in their backyard shop, or have access to some really Pro grade CNC machines, having someone create fresh CAD models of old cases and using those to design new injection moulding dies for old systems would be awesome.

  5. Some years ago I read about the rare Dornier Do-17 bomber recovered from the English Channel.
    After removing the encrustation of 60-odd years in salt water (a story in itself), they mentioned treating the parts with a material called ‘Renaissance Wax’.

    I looked into that, and it’s a microcrystalline synthetic wax that preservationists at places like the Smithsonian use for artifacts, as it leaves a virtually invisible and inert coating.

    I wonder if a coat of something like this would help keep the plastic surfaces sealed from oxygen? I’ve not tried it myself as this stuff seems very expensive for the amount in the container, but could be worth a try?

    1. Fickle lot we English, join the EU, leave the EU, put all the Dorniers into the Channel, take all the Dorniers out of the Channel. Yes, it’s sounding like some kind of surface sealing is the way to go. With that in mind, I’m starting to wonder about Armor All, which numbers of ppl say is “no good because it doesn’t have significant UV blockers” well maybe that’s not the point then.

      1. For the love of all that is holy, PLEASE do not use Armor-All on a restoration project. In fact, don’t even use it in your car.

        Armor-All embrittles the plastic, possibly not as much as Retr0Brite, but it still weakens the structural integrity of the plastic assembly.

        Case in point: Fleet of 25 Ford trucks, each issued to an individual driver. Every driver that Armor-All’d the dashboard had to have at minimum one replacement dashboard during that vehicle’s fleet life. Non treated dashboards survived the fleet-life of the truck in blazing Florida sun. This happened without fail over 25 years’ worth of trucks and drivers.

        1. Ah, that sounds like a fair comparison. I had somewhat positive thoughts about it because I got 5 new tires, the ones mounted on car got shined up with armor all tire stuff, one I got fed up of being in the way in the trunk, and went back to carrying the space saver which fit in the hole, so it went in the garage. 3 years later, about halfway down the tread, I think derp, I should be using the spare in the rotation, so get it out and it’s starting to check, while the others weren’t. It hadn’t had the armor all. BUT yes later, I hear that running tires on a vehicle, with all the constant deformation, actually helps the plasticisers spread around inside the material, kinda pumping in and out, so stops surface drying and cracking. So I then thought maybe was the AA, maybe it wasn’t.

      2. Armor All can stay tacky. I’ve also seen what it looks like when it does dry and multiple applications have been done. It’ll crack and where the cracks are, what it’s put on deteriorates.

        Seems similar to the glossy crap often sprayed in the engine compartments of used cars. Looks all shiny, until it goes bad, cracks, and where the coating cracks are, the rubbers and soft plastics crack.

        1. On a hunch, I took to spraying my rubber parts with silicone lubricant, which has seemed to make them last longer, but I haven’t done any A-B-A type tests to be certain of it.

  6. FWIW, I had some luck removing surface yellowing on an “ivory” colored, early ’80’s computer, using one of those white melamine sponges (damp with detergent in water) that are sold under various brands such as “Mr. Clean Magic Eraser,” “Scotch Bright Soap Scum Erasers,” and any number of other brands, including one available through Dollar Tree. I’d bought these for use against soap scum in the shower stall, and was impressed. They worked very well removing surface yellowing and various surface stains and blemishes.

  7. How about an airbrushed on coating of cyanoacrylate? What it’s sprayed on would have to be totally moisture free and kept in a moisture free environment until it dries or the resin might turn chalky white.

    1. You would have to keep the CA’d surfaces well away from any CD’s, CDROMs, DVDs and the like for quite some time.
      Even the styrene CD jewel boxes fog white with CA vapours. Apart from that, it certainly could seal a surface and I’ve found it works beautifully for woods such as balsa and cork.

  8. Is the “clear” epoxy yellowing coming from the same kind of reaction?
    I’ve been thinking of a slightly blue transparent coating to reverse the process on artworks I made with that kind of stuff…

  9. Some years ago I stored a Dreamcast on a plastic box, his case was white, to my surprise the last year I opened the box only to see that my Dreamcast turned yellow, inside a box that was inside a room with it’s windows blocked with aluminum foil, and happened the same with a keyboard on another box, that room develop high temperatures on summer (45°C), I also think that yellowing is caused by heat.

  10. Over 25 Years Ago, i was attempting to remove the stickyness of an extremely Yellow tandy 1000EX case.
    At first, i tried scrubbing the topcase with a sponge and a mild household cleaner, and failed, No effect on the plastic. As a last resort, i boiled a kettle of water, mixed 2 or 3 household cleaning products, placed the topcase in a plastic tub, dumped the boiling water and cleanig mix over the case.
    After About a minute, a massive yellow and Brown “mushroom cloud” came Loose from the case and drifted to the surface. After Quickly rinsing the case with cold water, i had a factory new case, Used it for many years after That. Wish i could rember the recipe of my “proto Retrobright” mixture :)

    1. Maybe two of the cleaning products were ammonia and bleach, generating chloramine vapors that would take the surface layer off of pretty much anything (including your lungs!!).

      But then, you’re still alive to write comments here, so perhaps not.

    2. Yes, about 25 years ago I also found that the yellow Mr Clean style household cleaners would remove yellowing, however, what they had in them then, does not seem to be what they have in them now. I believe it was some kind of ammonium chloride or other ammonium salt.

    1. Because it’s black?

      If you get nothing, try replacing the voltage converter, 7805 I think it is. If it’s a scrambled screen likely the ULA fried. Two of the more common faults.

  11. years ago i cleaned typewriters and cases with an ammonia solution, 10 0r 20 to 1, cant remember exactly. with a soft long bristled brush. it takes off yellow from plastic and all the dirt. water wash off, dry then a light wd40 spray and plastics were like new.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.