Is blue fluorescence bad, and does it matter on a lab-grown diamond?

Blue fluorescence is not damage, not a treatment and not a warning sign, and on most stones it is invisible unless someone points a UV lamp at it. The evidence from the laboratory that invented the colour scale runs against the trade's instinct: viewed face-up, strongly blue fluorescent diamonds were judged, on average, to look slightly better rather than worse. On a lab-grown diamond the question mostly evaporates, because most such stones show no reaction to the long-wave lamp that grading reports use.

That is the short answer. Fluorescence is the one line on a grading report where fear and evidence have pulled apart for half a century, and the gap is measured in money.

What is diamond fluorescence, and what makes it blue?

Fluorescence is the visible glow a diamond emits while it is sitting under ultraviolet light, produced by atomic-scale defects in the carbon lattice that absorb UV energy and release it again as visible colour. Switch the lamp off and the glow stops. Nothing has changed in the stone; it holds a defect that answers to a wavelength the eye cannot see.

In natural diamonds the usual culprit is the N3 centre: a gap in the lattice surrounded by three nitrogen atoms, with its emission line at 415 nm and a broad band that reads as blue. Yun Luo and Christopher Breeding set this out in Gems and Gemology in 2013. Laboratories describe the reaction to long-wave ultraviolet, nominally 365 nm; short-wave at 254 nm is used for identification and does not appear on a report.

Boron is often blamed for blue fluorescence, which is a muddle. Boron-bearing type IIb diamonds are better known for phosphorescence, the blue-green afterglow that continues once the lamp is off, which colourless HPHT-grown stones frequently show.

An inconvenient detail from that 2013 paper: real UV lamps are not pure. They commonly emit extra lines at 404 and 435 nm, and filters age, so two lamps can disagree about one stone. A shop-counter torch is not evidence.

How common is fluorescence, and what do the grades mean?

The most solid count remains GIA's own, published in 1997 from a random sample of 26,010 of its grading reports for colourless to faint yellow stones. Roughly 65 per cent had no reported fluorescence. Of the 35 per cent that did, 3,465 were faint and 5,710 ran from medium to very strong. Within that group, 5,533 fluoresced blue and 162 fluoresced something else, chiefly yellow, white or orange. In the D-to-F range, 4,250 of 11,901 stones reacted.

Two caveats. Those figures cover stones sent to one laboratory nearly three decades ago, and GIA calls fluorescence an identifying characteristic rather than a grading factor; the report describes its intensity, it does not grade it.

Those price columns come from Rapaport lists reproduced by GIA researchers in 2021. They apply to natural diamonds, and are dated.

Did GIA's own study really find that fluorescence does no harm?

Broadly, yes, and the study deserves reading rather than quoting. Thomas Moses, Ilene Reinitz, Mary Johnson, John King and James Shigley published it in Gems and Gemology, Winter 1997, pages 244 to 259. The team screened over 1,000 polished diamonds to assemble four sets of six round brilliants at E, G, I and K colour, matched for size, clarity, polish and proportion. Forty-six observers took part: 25 laboratory graders, five trade graders, five trade observers and 11 members of the public, across five lighting environments.

Face-up, stones described as strongly or very strongly fluorescent were on average reported as having a better colour appearance. Table-down, the grading position, fluorescence had no widely perceptible effect. The public group could not make the distinctions at all: for them, fluorescence had no overall effect on colour or transparency.

Now the parts that marketing copy leaves out. The sample was 24 stones. Researchers excluded so-called overblues, the extremely fluorescent stones with an oily or hazy look, because they could not find enough, and yellow-fluorescing stones for the same reason. Only round brilliants were tested. Experienced observers disagreed from one stone to the next. On transparency, half the table-down observations showed no difference; the rest showed a clear trend for weakly fluorescent stones to be judged more transparent.

GIA returned to the problem in 2021 with instruments instead of eyes. Yun Luo and colleagues found that strong fluorescence produces a minor, measurable contrast loss face-up in some stones, but that the milky look reported by the trade comes mainly from light-scattering structural defects and nano-inclusions. Fluorescence made those stones look worse; it did not create haze in clean ones. UV content of the viewing light matters enormously: sunlight carries roughly 3 to 5 per cent, most indoor lighting under 1 per cent.

Nor is the literature unanimous. A 2018 study by Bouman and co-authors found improved colour grades table-down outdoors, with no clear correlation face-up, close to the reverse of the 1997 pattern. Anyone calling the science settled has read one paper.

If the evidence is neutral, why does fluorescence still cost money?

Habit explains most of it, and a price list does the rest. Rapaport has published estimated fluorescence discounts since 1993, and the direction of travel has been unkind. In its list of 7 November 1997, stones of I to N colour with very strong fluorescence carried a premium of up to 4 per cent, on the reasoning that blue offsets a yellow tint. By 2020 the same list had turned those premiums into discounts for I to K colour, minus 6 to 12 per cent at Strong in the IF to VVS band, while no figures at all were published for L and below, or for Very Strong at any of those colours.

The picture is starting to move. Rapaport News reported on 14 June 2026 that GIA will add fluorescence comments to its standard reports for natural D-to-Z diamonds from the fourth quarter. One will note that fluorescence may improve appearance in UV-rich light such as daylight. The other, which GIA said would apply to fewer than 0.2 per cent of natural D-to-Z diamonds, will note that fluorescence can make existing characteristics such as reduced transparency or milkiness more noticeable. Around 10 per cent of natural D-to-Z diamonds would be eligible for a comment at all. At the time of the announcement the wording was not final, and the change covers natural stones only.

Does fluorescence matter on a lab-grown diamond?

It matters less than most people expect. GIA's review of several hundred CVD-grown diamonds examined up to mid-2016 found only 7 per cent of near-colourless samples had measurable long-wave fluorescence, all of it faint to very faint. Its companion review of several thousand HPHT-grown stones found 2 per cent of colourless samples reacting to long-wave UV, against 88 per cent under short-wave; across all HPHT colours, 74 per cent showed nothing to long-wave. Both papers set this against the 35 per cent for natural stones of the same colour.

Colour of the glow differs too. Where natural stones give blue from the nitrogen-based N3 centre, lab-grown stones that react tend towards orange or red from NV centres, or green from H3. Colourless HPHT material often shows that long-lived blue-green phosphorescence linked to trace boron. Under the DiamondView, which uses deep ultraviolet below 225 nm, HPHT stones classically show a cross-shaped pattern and CVD stones growth striations. Our note on how lab-grown diamonds are made explains what produces them.

Two cautions, because this is where confident writing goes wrong. GIA warned in 2024 that relying on phosphorescence, or any single observation, is not advisable for colourless HPHT stones, since low-dose irradiation can remove the afterglow. It also reported CVD stones whose deep-UV images resemble natural type II diamonds closely enough to require cathodoluminescence imaging. Fluorescence pattern is a signal used alongside spectroscopy, never a verdict on its own.

For a buyer the picture is simpler. An IGI laboratory-grown report carries a Fluorescence line like any other, and on the reports we handle it generally reads None. We could not find a published figure for what share of IGI-graded lab-grown stones read above None, and will not invent one. GIA's forthcoming comments do not apply here either, since its lab-grown reports stopped carrying 4Cs grades on 1 October 2025, a change covered in why GIA stopped grading lab-grown 4Cs.

What should a careful buyer do with any of this?

The fluorescence line is a prompt to look, not a grade to optimise. On a natural colourless stone with strong or very strong blue, the discount is real and the risk it prices applies to a small minority of stones, so the saving goes to whoever inspects. Unedited daylight video tells a buyer more than a light box does. Haziness, where it exists, shows as a softening of the facet pattern, and usually originates elsewhere. Colour grade matters more in almost every case, as our guide to the D-to-Z colour scale sets out.

Unflatteringly for us, this article will save most of our customers nothing. Every stone we sell is lab-grown and IGI-certified, and lab-grown material rarely fluoresces, so there is no discount to pass on and no hazard to disclaim. The question earns its keep when someone is weighing a natural stone elsewhere, told without evidence that blue is bad. On our engagement rings the fluorescence reading appears on every report we send, and it is almost always the least interesting line on it.