How binders help gold lustre stick to the pot and how to choose the right vehicle

MODERN BINDERS

I didn’t start this journey with a clever theory about polymers. I didn’t even know where to start. All I really understood at the beginning was that the binder had to do two things: stick everything to the pot, and then burn away cleanly. Beyond that, I had no real way of knowing whether a polymer would behave or not. At the time, it felt like the only way to find out was to try it and see what came out of the kiln.

So I started with ethyl cellulose (CAS 9004-57-3), simply because I’d used it before.

On the bench it looked promising. It laid down a fairly level film and behaved nicely during application. Then it went into the kiln and everything fell apart. The gold ran, migrated, and pooled on the test tiles. I didn’t understand why at the time, only that the results were awful.

Looking back now, it’s obvious what was happening. The binder itself was softening and flowing as the temperature rose, dragging the gold along with it. The chemistry wasn’t failing in some subtle way. The whole film was moving.

That was my first real lesson: a binder that behaves beautifully at room temperature can still destroy everything once heat is involved.

CASTING AROUND BLIND

After that came a long stretch of Googling polymers, mostly in places I never expected to be looking. Cosmetics. Makeup. Nail products. Anywhere that dealt with thin films that had to dry level and stick.

That’s how I found the Paraloid range. These are a whole family of specially designed polymers and polymer blends, more at home stablising ancient dinasaur bones or old oil paintings than in a potters workshop.

There were plenty to choose from and no obvious way to narrow them down, so I picked one and hoped. Paraloid B67 came first.

It misbehaved almost immediately. The film laid down unevenly, dried poorly, and this was where I first learned phrases like coffee ring and coffee eye. I managed to partly tame these flaws by abandoning a single solvent and moving into blended solvent system. That was also the moment it finally clicked why historical lustres leaned so heavily on solvent blends rather than neat single solvents. That lesson turns out to matter later on.

Even with the drying issues largely under control, B67 still failed in the kiln. The fired gold showed a clear alligator-skin pattern and required burnishing to bring out any shine. As the film heated it shrank and cracked, causing the gold to deposit unevenly along those cracks and leave a rough surface behind.

So I moved on to Paraliod B44.

Different polymer, same disappointment. Patchy, dull gold. Nothing like the clean, bright surface I was aiming for. By this point it was obvious I was missing something fundamental. Swapping polymers blindly wasn’t going to get me there.

THE MISSING PIECE

The turning point came from a Johnson Matthey patent that mentioned poly(isobornyl methacrylate), giving good results in lustre systems.

At first, it felt like another dead end. I had almost no suppliers, and the one I did find quoted something in the region of six hundred dollars per gram. That wasn’t happening. So instead of buying it, I did the only thing left to do and started reading about it.

That’s where Tg and Td properly entered the picture. When I lined PMIB’s properties up against the polymers I’d already tried, the difference was immediate. Its glass transition temperature was far higher, and its decomposition temperature sat exactly where it needed to be relative to everything else happening in the firing.

Suddenly, there was a language to explain why all the earlier experiments had failed.

I hadn’t been choosing polymers at random. I’d been choosing them blind to the one thing that actually mattered: when they softened, and when they burned away. Every failure up to that point suddenly had a perfectly sensible explanation.

Not long after that, more by luck than planning, I stumbled across another supplier. Still expensive, but just affordable enough to justify pushing on. I bought some PMIB, convinced Nicola this really was another step forward, and quietly tried not to think about the steadily rising cost of this “small side project.”

WHEN IT FINALLY WORKED

PMIB gave good, even, bright results straight out of the firing, genuinely comparable with commercial gold lustres. For the first time, the gold looked like it was meant to be there. The chemistry in the kiln was finally behaving.

What I didn’t realise at the time was how much luck was involved. On that very first PMIB test, I’d landed almost exactly on a workable polymer percentage by accident. Later testing made it painfully clear how narrow that window actually is. Too much PMIB and the fired gold dulls, loses brightness with defects. Too little and the lustre starts to run and misbehave during application.

If that first test had been even slightly off, I’d probably have disappeared down yet another rabbit hole. Instead, it gave me something rare at that stage of the project: A major cinfidence boost in the project as a whole.

It still wasn’t perfect.

During drying, the films would crack and flake. PMIB is extremely brittle, almost glass-like, and that brittleness showed up long before the piece ever reached the kiln. I’d solved the high-temperature problem, only to uncover a low-temperature one. The gold film still worked but under magnification you could clearly see a crazing type pattern on the gold.

And that’s where the story naturally moves on. Once the binder itself was doing the right thing in the firing, the only way forward was to start gently modifying how it behaved as a film.

That’s where the micro-additions and solvents come in.