View Full Version : LED lighting and Chromium Yellow = Browning?

03-04-2013, 06:05 AM
Someone mentioned this news article on my museum photographers listserv and I thought I would post it here:


I'll post it to the PACCIN listerv too.

Of course it doesn't say what kind or brand of LEDs did the damage, but I'll try to find out.


03-04-2013, 06:22 AM
Well the listserv came through again, here are some responses:

My colleague and I were just at an LED seminar at the Smithsonian this past Friday, and interestingly enough, one of the presentations was on how this was essentially a blown-out-of proportion story based on misunderstandings on LEDs. I don’t have the information in front of me, but a highly reputable lighting scientist, Jim Druzak, from the Getty (I think?) and a colleague presented a paper on these fallacies picked up by the media. The problem was an inherent one of instabilities in non-organic yellow, early-modern paints, not in the lighting itself. I can look for more information, if anyone’s interested.

Tim Gierschick, Preparator
Barnes Foundation

This was just discussed at the LED Lighting conference Friday. It is bunk. There is no UV in the output of the LEDs used.

The conservator who presented had a great deal of data, so I'll post a link as soon as I can locate it.

Jim Williams
Exhibition Designer/Preparator
Kent State University Museum

03-04-2013, 07:54 AM
Here is the response from The Smithsonian American Art Museum Lunder Conservation Center's program "Gallery Illumination: LED Lighting in Today's Museums on March 1st, 2013. (Thanks to Kevin Marshall from the listserv)

How safe is Van Gogh’s Sunflowers to LED Lighting? Quite safe, actually!

In 2012 a team of European scientists and conservators published the results of their work on chrome yellow pigments used by Vincent Van Gogh in a series of articles in the journal “Analytical Chemistry”. The team had uncovered the chemistry of why certain of these pigments in some paintings by Van Gogh and others tend to darken over the years from the effects of light and other environmental factors.

Unfortunately the research team also made the very strong, but substantially untested, assertion in a press release that the pigments in question are likely to be selectively sensitive to the white LEDs increasingly used in museums (http://www.vangogh.ua.ac.be/). Significantly, they did not draw this conclusion in their peer-reviewed publications, possibly because it was based on highly unrealistic accelerated light exposure test using a test illuminant completely unlike an LED. In fact the accelerated aging light source is unlike any other museum lighting short of displaying the paintings outdoors, without glazing, in the sun. They also inferred fading properties of an LED with a spectrum that is unlikely to be recommended for a museum because of its color properties. Astonishingly, given the importance of the issue they have raised, they did not actually test or even model the damage potential of that or any other LED, nor did they compare its putative effect with conventional tungsten lighting, which they neglected to explain must in principle also affect the pigments to a similar degree.

In an attempt to determine which wavelength ranges contributed most to the darkening effect, model paints were exposed to a xenon lamp variously filtered to (1) a mixture of ultraviolet radiation (UV) and visible light from 300-800 nm, (2) purely UV from 300-400 nm, (3) a band wrongly described as “blue” light from 335 to 525 nm, and (4) red light greater than 570 nm. Not surprisingly, the darkening of their model paints followed the expected trend: a mixture of UV and visible produced the largest change; followed by the same range excluding visible light; then their so-called “blue”; and lastly, red which contributed little or nothing to the undesirable colour change.

Visible light refers to radiation between 400 and 800 nm, however their “blue” light included UV at levels and energies far in excess of what would be found in a museum. It is simply not possible to infer from their test what the relative contribution to damage the real blue portion of an LED’s spectrum (400-500 nm) might be, compared to that caused by the high levels of UV included in their “blue” source. What is well known is that once into the UV, the photochemical damage potential of light normally increases exponentially with decreasing wavelength. On this basis it is likely that the radiation most responsible for the damage observed by the researchers was below 400nm, which no white light LED (however poorly chosen) contains.

The research team’s December 17, 2012 press release was picked up and quoted by a large number of reputable news sources, including Spiegel Online and the Huffington Post. Conservators around the world reacted quickly to headlines such as “Are Van Gogh's Sunflowers Wilting? Priceless Paintings May Be Turning 'Coffee Brown' Due To LED Lights” with a concern that would be justifiable if the research team’s alarmist conclusion about LEDs could be sustained. The headlines were no media beat-up; the first paragraph of the press release stated, “[t]he scientists recommend that museums identify all paintings with this type of chrome yellow, and protect them in particular from the increasingly popular LED lights as these emit a large amount of blue”, with one of the lead Italian authors even remarking, “[w]e were surprised to note that already under conditions of illumination currently considered safe, some of our test samples started changing color quickly.”

Far from being “safe”, they did not explain that because xenon lamp accelerated-aging test chambers typically employ light levels 500-1000 times greater than that used in museum gallery lighting, the actual cumulative light dose responsible for the dramatic darkening depicted in the press release was probably equivalent to 50-100 years of museum exposure. No wonder it occurred “quickly”. Finally, to compound the misinformation about the energy and intensity of their “blue” light, the press release included an LED spectrum with the label “[e]mission spectrum of a typical "white" LED, containing a substantial portion of harmful blue light”. The spectrum shown in fact is not typical of - nor even recommended for - any light-sensitive artifacts. They also omitted the important qualifier that in order to adequately render the colors of these paintings satisfactorily any illuminant must contain a proportion of “harmful blue light”, not just LEDs. For works of art on paper, textiles, decorative arts, and paintings with known sensitivity issues, the types of LED recommended by conservation scientists are those that most closely resemble the visible portion of the spectrum of tungsten lighting that, for the most part, have been used for the last century to illuminate museums. These LEDs have small blue peaks and represent no greater risk than the lighting they replace; indeed there is some research indicating that they are marginally safer. LEDs with larger amounts of blue, like the one illustrated in the paper, have a useful role as an appropriate illuminant for artifacts with demonstrably little or no light sensitivity, for which natural daylight is considered necessary to present the works as the artist intended, precisely because they do not need to be filtered to remove UV.

While we agree with the authors of this press release to the extent that art museums have a responsibility to protect their collections, and because light damage is usually irreversible they must carefully weigh the damage versus display issues. However we believe it is wrong both in principle and on the available evidence to characterize contemporary LED lighting in general as less safe than the traditional alternatives; and it is irresponsible to publicly imply that museums have begun to phase them in without considering their relative damage potential. In view of the energy and many other advantages that LED lighting offers, and the availability of a wide range of white light LEDs with virtually identical spectral properties to tungsten lighting, it would be better argued that museums would be foolish not to consider their adoption.

James R. Druzik
Senior Scientist
The Getty Conservation Institute

Naomi J. Miller
Senior Lighting Engineer
Technology Planning and Deployment, Pacific Northwest National Laboratory

Scott Rosenfeld
Lighting Designer
Smithsonian American Art Museum

Bruce Ford
Conservation Scientist
Melbourne, Australia

Joseph Padfield
Conservation Scientist
The National Gallery, London, UK