By Sophie Harris

These are four copper alloy classical table casters. They are cast and chased with decorative scrolls on each side, lion's feet and acanthus leaves.

They display overall general wear and minor bright scuffs from being situated in a public area. The second caster was missing and had a wooden (lignum vitae) replacement. A thick natural patchy oxide patina covers all surface areas.

The client requested a mellow antique two thirds clean and the replacement of the caster.

The wheel was turned on a lathe, aged and patinated to match the originals. It has also been stamped to show it's a replacement.

The process of cleaning involved multiple tests. The first was a spot test in an inconspicuous area with Triton X and white spirit. White spirit seemed to have more of an effect but wasn't removing the oxide. The result was indicative of a cross-linked protective coating as well as signs of multiple layers of wax, possibly microcrystalline, which would have been used to polish the surface. A waxy green copper corrosion product (due to moisture) was visible around screw holes. Random patches of white paint possible deposited by the local painter and decorator were also visible on top surfaces.

Non-abrasive steam cleaning was the next step with Vulpex, which is an alkaline conservation soap, and a brass brush. It was hoped this would soften the oxides, which could easily be removed by steam cleaning, however this didn't work either.

As soaps did not seem to be lifting the heavy oxides, chemical cleaning was the only option. Ethylene diamine tetra-acetic acid (EDTA) is a chelating agent that grabs corrosion oxides and makes them water-soluble so they can be removed. I tested a pre-made EDTA gel that was supposed to be pH 7 but it was in fact far too acidic (anything too acidic or basic will etch the surface of a metal) and within seconds it de-zinced the surface, which is why making your own recipes is advisable. I made my own EDTA disodium hydroxide solution with a 7.5 pH. I could see it was effective by a change on the surface colouration, and so to keep the solution in place on the surface I put it into a 5% sodium carboxymethyl cellulose gel. The gel keeps the solution from evaporating & from running off the surface of the object too quickly. I carried out a series of time tests to see how long it would take the solution to chelate, this had the perfect result at half an hour.

Once I had achieved an overall two-thirds clean appearance, it was a case of making a protective tinted coating that would complement the colours of the table that they came from. A final bristle brush clean ensured the removal of all cleaning agents, as leaving these on the surface will do damage over a long period of time.