UV Light and Porcelain: How to Spot Restoration and Repair

Use UV light to find restoration on Chinese porcelain: which lamp, how to read fluorescence, where to look first, and what ultraviolet cannot prove.

Take a blue-and-white bowl into a windowless room, switch off the light, and play a long-wave ultraviolet torch across the surface at a shallow angle. For the first few seconds you see nothing but a dull violet wash. Then the eye adjusts, and a hairline that was invisible in daylight comes up as a thin pale thread running from the rim towards the foot, with a cloudy halo around it where somebody has filled and sprayed. The bowl has not changed. What you know about it has.

This guide walks you through ultraviolet examination as a buying tool: which lamp to use and how to set up a room so the results mean anything, how to read the colours that appear on a glazed surface, how to sequence an examination so you do not miss the obvious places, what UV genuinely cannot tell you, and how to turn what you find into a price and a straight conversation with the seller.

Why porcelain fluoresces at all

Ultraviolet light is simply light at a wavelength shorter than the eye can see. Some materials absorb those photons and re-emit the energy as visible light, which is why a surface that looks uniform in daylight can look patchy in the dark under a UV lamp.

A high-fired porcelain glaze is a glass. It is chemically stable, largely inert, and usually reads as a flat, dull purple or a very faint bluish grey. It is not that porcelain does not fluoresce; it is that it fluoresces evenly and weakly.

Restoration materials are a different order of substance altogether. Adhesives, fillers, sprayed coatings and retouching paints are organic resins or modern pigments applied cold, and they respond to ultraviolet in their own way. The diagnostic signal is never the colour by itself. It is the contrast: a patch that behaves differently from the glaze around it.

That distinction matters more than any colour chart. Two Qianlong dishes can fluoresce slightly differently and both be untouched. A single dish that fluoresces two ways within one rim has almost certainly been worked on.

Choosing a lamp and setting up the room

Wavelength. Use a long-wave lamp at 365 nanometres. This is the standard used in conservation studios and auction house viewing rooms, and it is the one restoration materials respond to most usefully.
Filter. A proper lamp carries a filter, often called Wood's glass, that blocks the visible violet the diode also emits. The cheap torches sold at 395 nanometres flood the object with visible purple light, which washes out precisely the faint differences you are trying to see.
Darkness. Complete darkness, not dimness. A windowless room, a cupboard, a bathroom at night. Any ambient light raises the floor and kills weak signals.
Eye adaptation. Give yourself two or three minutes in the dark before you judge anything. The first look is always the least informative.
Safety. Wear UV-blocking glasses, never look into the beam, and do not play the lamp across your own skin. Keep sessions short.
Distance and angle. Hold the lamp twenty to thirty centimetres away and work at a raking angle, sweeping slowly. Straight-on light hides the very edges you want.

Reading what you see

Materials vary and the literature is full of exceptions, so treat the following as tendencies rather than a key.

Flat, even, dull purple. Almost always the original glaze, behaving as glass behaves.
Chalky, milky or greenish patches. The classic appearance of a filler and the retouching over it. Fills sit in chips, over rim losses and along breaks, and they rarely stop exactly where the damage stopped.
Amber or orange bloom. Older organic adhesives and shellac-based coatings tend this way. On an object with a long repair history you may see orange from an early repair and something quite different from a later one over the top.
Dead, absorbing dark patches. Retouching paints made with titanium white absorb ultraviolet rather than re-emitting it, so a sprayed area can appear darker than the glaze around it. Restorers have known this for decades, and a dark patch is as suspicious as a bright one.
A bright line following a contour. A break edge that has been glued and dressed. Follow it with your eye: breaks run, chips do not.
An even haze over one whole region. A sprayed coating applied across a wider area than the damage, intended to blend the join. On a vase this often covers an entire shoulder or a whole neck.

Two false positives are worth knowing before you accuse anyone of anything. Old paper labels, and the glue that held them, fluoresce strongly on the base. So do detergent residues, waxes and fingerprints, all of which wipe away.

A working sequence

Examining at random means you will find what is obvious and miss what matters. Use the same order every time.

Look in daylight first. Note the shape, the glaze, the wear and any damage you can already see. UV is a second opinion, not a first one.
Rim and mouth. The most frequently damaged part of almost every form, and the first place a restorer works.
Foot rim and base. Look for fills at the foot, and for the halo of a reground or replaced foot. Base examination also pairs naturally with the questions asked in dating porcelain by its foot rim.
The interior. Turn the light into the bowl or the jar mouth. Restorers finish the outside to a much higher standard than the inside.
Extremities. Handles, spouts, finials, ears, flanges, dragon heads. Anything that projects has been knocked off at some point in four hundred years, and the join is usually at the narrowest point.
The mark. A repainted or strengthened mark fluoresces against the base glaze. What the mark says and whether it is original are separate questions, addressed in the guide to Chinese porcelain marks.
Return to daylight. Confirm every suspicion with touch, with a loupe and with a gentle tap. Fills feel warmer and slightly softer than glaze under a fingernail, and a filled or cracked piece rings dead rather than clear.

What ultraviolet cannot tell you

It is not an age test. Nothing about fluorescence dates a piece, and a lamp will not separate an eighteenth-century dish from a careful twentieth-century copy of one. That work belongs to the body, the foot, the glaze surface and the potting, and it is set out step by step in the porcelain authentication checklist.

It does not find every restoration. The most expensive restorations are made with ceramic materials and fired, so they behave under UV much as glaze does. Very recent work using modern conservation-grade resins can also read quietly. A clean result raises confidence; it does not close the question.

It says nothing useful about a piece with no old surface at all. A recently made object is uniformly new, so it fluoresces uniformly. Evenness is only reassuring on something whose age you have established by other means.

It does not distinguish natural glaze variation from intervention. Some later glazes, particularly certain nineteenth and twentieth-century wares, fluoresce more brightly than earlier ones across the whole surface, and kiln grit, crawled glaze and firing flaws all read oddly. The wear patterns and bubble structure covered in the note on glaze bubbles and surface wear are a better guide to whether a surface has lived.

Finally, it does not grade the work. A lamp tells you an area has been treated. Only removal under a conservator's hands tells you how much of the original survives beneath.

What restoration does to value

Condition works as a multiplier on Chinese porcelain, and the size of that multiplier depends on rarity far more than on the size of the damage. On a rare early piece, collectors accept substantial restoration because there is no alternative example to buy. On a common late Qing vase, a restored neck can take most of the value away, because a sound one will be along shortly.

The distinction the market draws is between honest original damage and concealed intervention. A visible chip, a stable firing crack or an old riveted repair are facts a buyer can price. A sprayed shoulder that was not disclosed is a different matter, and it is the reason UV examination has become routine at every level of the trade. The vocabulary for describing what you find, and the line between a fault and a flaw, are set out in the guide to porcelain condition and damage.

There is also a simple economic rule worth remembering. Professional restoration is charged by the hour, and on a modest piece the work will often cost more than the object can ever fetch. Buy the sound example if a sound example exists.

Asking the seller

Once you have looked, ask plainly. Has this piece been restored, and what was done to it? Reputable dealers answer that question directly and most volunteer it first, because a returned vase costs them more than a discount would have.

At auction, request the condition report in writing and ask specifically for ultraviolet images rather than daylight ones. Ask when the piece was last examined, and by whom. Keep the reply: a written condition statement becomes part of the object's file, alongside the invoices and export papers discussed in the guide to provenance documents.

If you are buying remotely and cannot examine the piece yourself, make the sale conditional on your own inspection on arrival, and agree the return terms before you pay rather than after.

Common questions

Does ultraviolet light damage porcelain?

Not in the way it damages paper, silk or pigment. A fired glaze and a porcelain body are stable under the brief exposures involved in examination. The materials at risk are the organic ones: old adhesives can yellow and some retouching paints shift colour with prolonged exposure, which is an argument for short examinations rather than leaving a piece under a lamp. The real hazard is to you, not to the object, so wear protective glasses and keep the beam off your skin and eyes.

Will a hairline crack always show under UV?

No. A tight, clean hairline that has never been glued or filled may show only as a faint dark line, or not at all, because there is no foreign material in it to fluoresce. Cracks reveal themselves more reliably by transmitted light, by the tap test and under a loupe at the rim. UV is at its best on what has been added to a piece, not on what has simply parted.

Is fluorescence a test of age?

It is not. Glaze chemistry varies by kiln, by period and by ware, and it varies within a single period, so no reading maps onto a date. Treat ultraviolet as a condition tool. Age is settled by potting, foot, glaze surface, decoration and wear, considered together.

Can I use a cheap 395 nanometre torch?

You can see gross restoration with one, and it is better than nothing in a saleroom. The problem is that it emits a great deal of visible violet light, which raises the background and hides subtle differences, so you will find the crude fills and miss the careful ones. A filtered 365 nanometre lamp is an inexpensive upgrade and it changes what you are able to see.

I have found restoration on a piece I already own. What now?

Nothing urgent. Record it with photographs in daylight and under UV, note it in your own file for the object, and leave it alone unless the restoration is failing, which usually shows as lifting edges, discoloured fills or a join that has become visible in daylight. If it is failing, go to a ceramics conservator rather than attempting removal: the solvents that release old fills also attack overglaze enamels and gilding.

Browse the collection

Porcelain condition is a subject we write about rather than one we stock. Kiln and Ink publishes limited-edition archival giclee reproductions of the twentieth-century Chinese ink masters, each hand-numbered and issued with a Certificate of Authenticity, where condition questions are about paper and light rather than fills and sprayed glaze. You can see the current editions in the collection.