Since microscopists frequently spend hours looking at specimens, the question of the safety of LED illumination is significant. Concerns regarding the possible damaging effects of blue and ultraviolet light after conversion of scopes to LED light sources have been raised by several authors describing conversion techniques. Traditional incandescent light bulbs produce most of their illumination in the red or far-red end of the spectrum, with modest output in the visible range and minimal ultraviolet production. Consequently, incandescent light that is comfortable to observe is unlikely to cause damage. This is not necessarily the case with an LED light source, where shorter-wavelength blue light excites a phosphor to produce a second peak of mid to longer visible wavelengths. The latter is less likely to be injurious to the retina unless the light intensity is high, but the UV component and the amount of primary-emission blue light are often poorly characterized.
The retina is a complex organ with ten layers, five of which form a complex, interconnected nerve network performing some basic image enhancement. The arrangement of the retinal layers is exactly upside down from what logic would suggest: the nerve cell layers are on top, and the photoreceptors are mounted upside down at the back of the retina. Light must therefore traverse the nerve layers before activating the light-sensitive photoreceptors (images from Ted Montgomery’s site):
There are two types of photoreceptive cells in the retina: rods and cones. Rods are the more sensitive and are responsible for low-light vision, but cannot distinguish colors, having only one light-sensitive pigment, or chromophore: rhodopsin. This pigment, most sensitive to green light at 498nm, consists of a complexly-coiled protein that crosses and recrosses the cell membrane of the rod cell. It forms a pocket for the pigment retinal, a Vitamin A derivative. An opsin protein acts like a carefully-balanced tangle of springs; when a photon strikes the retinal molecule, it undergoes a subtle structural shift (conformational change) that twitches the opsin into a slightly different shape. This change also affects the segment of the protein inside the cell membrane, where an active enzymatic site is exposed, resulting in a chemical cascade that causes the rod cell to depolarize and create a nerve impulse. This process is an important mechanism in cells where transmembrane signaling occurs:
Cones, the color-sensitive photoreceptor cells, are less sensitive, are the primary cells responding to daylight light levels, and have three chromophore pigments, L, M, and S, responding respectively at 564nm (Long wave, red), 534nm (Medium wave, green) and 420nm (Short wave, blue). The normal human eye is therefore populated by L, M, and S cones.
The mechanisms of retinal damage by light are complex and incompletely understood, and the literature on this subject is both enormous and confusing (see Organisciak and Vaughan as well as Youssef et al). Two broad categories of damage are recognized: thermal damage and photochemical damage. Thermal damage reflects heat-induced retinal damage and can occur across a fairly broad range of visible wavelengths, but requires significant intensity. At very high light intensities with focused spot retinal heating, shearing and mechanical effects from heating of the aqueous components can also become a factor. Both effects are of concern with powerful sources such as infrared YAG lasers, but should not be a significant risk to the average microscopist.
Photochemical damage is of more concern; the risk for this form of retinal injury is concentrated in the shorter wavelengths (i.e., blue and violet) of the visible spectrum and peaks in the ultraviolet. There also seems to be some photochemical damage that can occur in the midrange of the visible spectrum as well; this is likely to reflect energy absorption and subsequent chemical reactions by the chromophore pigments.
The human eye is protected from much of the effect of high levels of visible light by the aversion response – pupillary constriction, head turning, and blinking. In other words, bright light hurts, and we respond reasonably quickly, reacting protectively in about one-quarter second. This response offers considerable protection from light that we can see. Unfortunately, the effectiveness of this reaction decreases significantly on both ends of the visible spectrum. Infrared light passes readily through the corneal and vitreous; on exposure to a powerful Nd:YAG infrared laser, the patient may feel only a “pop” as focal explosive boiling creates a permanent blind spot. Similarly, the aversion response to blue or violet light is weak, and the eye has no significant protective response to ultraviolet wavelengths.
Research results and published opinions are contradictory. Much work has been done in this area, as light-induced damage mimics many of the features of chronic degenerative processes such as macular degeneration, and chronic light exposure may be a factor in these processes. Some studies have used very intense exposures and may not be good models for the effects of long-term, lower-level blue or UV light exposure levels. In 2013, the US Department of Energy published a fact sheet on architectural light sources that stated “…LED products are no more hazardous than other lighting technologies that have the same CCT (color temperature)…” Then in 2015, a French research team published a highly sophisticated study showing biochemical and structural damage in the retinas of rats exposed to overhead LED illumination. Sources used included white and blue LEDs from Cree Industries and Nichia Corporation.
Ultraviolet light is almost entirely absorbed by the cornea, lens, and vitreous, producing lens cataracts, denatured protein swirls in the vitreous, and and the equivalent of a sunburn on the cornea. These injuries are clearly more related to high-energy sources like welding arcs than an afternoon spent watching the lazy movements of a hydra.. However, blue light can reach the retina, and it is the effect of the powerful blue peak pumping the phosphor that needs to be evaluated.. A recent paper from Taiwan showed changes in the mouse retina after exposure to “…Low-powered Family LED Lighting…”
Cree, Inc. has done safety testing with its LED products and has concluded that its blue and royal blue LEDs can pose a hazard
To be continued…..
Note on learning new things: I should really admit that, like much of my writing, I started off in complete and utter ignorance of all three of these topics. Furthermore, I have never actually DONE an LED conversion of a classic microscope. Like much of what I write, this series is a notebook of my learning process – a reflection of how I organize my thinking so that new information falls into order, like shiny ornaments on a well-designed Christmas tree. Since I’m doing all this work, I might as well write it down so others can share it. I’m also driven by my own curiosity and sense of fun at learning new things that I can use on my own projects. It should also be noted that complete ignorance of a topic has never stopped anyone from loudly expressing an opinion or writing voluminously about it. If you doubt this, listen to any politician.
Building Technology Office, US Department of Energy, Solid State Lighting technology fact Sheet. “Optical Safety of LEDs.” Published June 2013. http://apps1.eere.energy.gov/buildings/publications/pdfs/ssl/opticalsafety_fact-sheet.pdf
Wikipedia. “Rhodopsin.” https://en.wikipedia.org/wiki/Rhodopsin
Zhou, X. E. et al. “Structure and Activation of Rhodopsin.” Acta Pharmacol Sin. 2012 Mar; 33(3): 291–299.