Whiter Than White

I wrote this article for the April 2015 issue of tED Magazine. Reprinted with permission.

From walls to paper to plastics to clothing, white abounds in the built environment. Lighting research, however, has focused on the rendering of colors. The color rendering index (CRI), in fact, is based on the rendering of colors by a white-light source and does not characterize how well the source renders white.

To manufacturers of white materials and goods, white matters. Many manufacturers use fluorescent whitening agents (FWAs) to make objects appear “whiter than white.” These agents (which incidentally occur naturally in some natural materials, such as human teeth) absorb ultraviolet and violet light and re-emit blue light, causing fluorescence and a perception of enhanced whiteness. The degree of enhancement depends on the composition and quantity of the FWA.

An object is not visible unless it reflects light, however, and the color of objects is entirely dependent on the color composition of the light source. The ability of light sources to render objects accurately is expressed by the source’s CRI rating and color composition expressed using correlated color temperature, measured in kelvins (K). As CRI again does not capture rendering of whites, what’s relevant here is color temperature. This is why a white wall appears different under an incandescent source compared to noon daylight. It is most noticeable when viewing objects illuminated separately by light sources with different color emissions.

Another factor is whether the light emission is capable of exciting the FWA and producing fluorescence and the resulting whiter-than-white effect. Again, for an FWA to fluoresce, the output of the light source must contain a UV or violet component. Incandescent and halogen lamps produce some violet emission. The spectral emission of LED sources, however, is precisely engineered, often with no output below 430 nanometers (violet).

A 2014 study led by Kevin W. Houser at Pennsylvania State University’s Department of Architectural Engineering and funded by LED product manufacturer Soraa, Inc., “Whiteness Perception Under LED Illumination,” explored the issue. The study attempted to determine whiteness perception using three types of psychological experiments (forced choice, selection, sorting) with various amounts of FWAs and under five light sources. The light sources included a filtered halogen lamp, a typical blue-pumped LED and violet-pumped LEDs with three violet emission levels (2.5, 5 and 6.5 percent violent composition in the spectral output). Thirty-nine people with normal color vision participated in the experiments.

The results focused on two effects related to whiteness rendering and perception—chromaticity and the presence of FWAs. Regarding chromaticity, as expected, a blue shift generally produced higher whiteness perception. Regarding FWAs, however, all lamps except the blue-pumped LEDs produced higher whiteness perception values as the amount of FWAs increased.

LED sources with some violet emission excite fluorescent whitening agents used in many white materials and goods, making them appear whiter than blue-pumped LEDs, according to a new study conducted at Penn State University. Image courtesy of Soraa.

In an article presenting the research findings, published in Leukos, the Journal of the Illuminating Engineering Society, the researchers stated: “The [blue-pumped LED] exhibited serious problems in accurately rendering the calibrated whiteness standards. We would expect similar problems to occur with the many manmade objects that contain FWAs… When the FWAs contained in objects cannot be excited, as is the case with blue-pumped LEDs, observers will be unable to differentiate white objects. This has implications for the illumination of white textiles, plastic, makeup, paints and papers.”

Since the large majority of LED products use blue-pumped phosphors to produce white light, the results are concerning.

Houser et al concluded, “We believe that whiteness rendering should be considered alongside color rendition when evaluating a source’s overall color quality.”

Craig DiLouie

See Full Bio