Red Light Therapy and the Two Numbers That Decide Whether It Works 660 versus 850. The wavelength is the whole story.
A panel of glowing diodes looks the same to the eye. To your cells, the difference between 660 nanometers and 850 is the difference between treating skin and reaching a joint. Here is the mechanism, the physics, and the evidence, without the hype.
Red and near-infrared diodes look alike, but they stop at very different depths in the body. The wavelength, not the brightness, sets the target.
- Red light therapy feeds light to cytochrome c oxidase inside your mitochondria, which then makes more cellular energy. It is photochemical, not heat.
- 660nm red reaches skin and hair. 850nm near-infrared reaches centimeters into muscle and joint tissue.
- More is not better. The benefit follows a biphasic curve, so too little does nothing and too much flattens or reverses the effect.
Stand in front of a red light therapy panel and the experience is almost anticlimactic. What you cannot see is the actual event, several layers beneath your skin, inside the energy factories of your cells. Almost every limit on the treatment comes down to two numbers printed on the box.
What red light therapy actually does inside a cell
Certain wavelengths of red and near-infrared light are not simply absorbed and turned into warmth. They are picked up by cytochrome c oxidase, an enzyme deep inside your mitochondria and the final stop in the chain that produces ATP.
Under stress from inflammation, fatigue, aging, or low oxygen, nitric oxide lodges onto that enzyme and clogs the spot where oxygen should bind. The light knocks it loose and the cell makes more ATP.
Red light therapy is not adding energy to your tissue, it is removing a brake. The brief rise in reactive oxygen species that follows switches on genes involved in repair, collagen production, and inflammation control.
Light absorbed by cytochrome c oxidase photodissociates inhibitory nitric oxide, restoring electron transport and increasing ATP, the single event from which most of photobiomodulation’s effects flow.
de Freitas & Hamblin, IEEE Journal of Selected Topics in Quantum Electronics, 2016
Why 660nm and 850nm are not interchangeable
Tissue is not transparent. Light travelling into the body is absorbed by melanin and hemoglobin and scattered by the cells it passes through. There is a narrow band, roughly 650 to 1,100 nanometers, where both dip and light travels deepest.
- 660nm red light is absorbed near the surface and penetrates only a few millimeters, reaching skin, superficial vessels, and hair follicles.
- 850nm near-infrared scatters less and slips deeper, on the order of centimeters, reaching muscle, tendon, and joint.
- The gap between them matters too. Wavelengths around 700 to 770nm sit in a trough of the enzyme's absorption and are comparatively inactive.
The practical takeaway is blunt. A 660nm-only mask is reasonable for fine lines or scalp follicles and useless for an arthritic knee, because the light cannot get there. Ask which wavelengths a device emits and how deep the target sits.
The biphasic dose: why more light is not more benefit
Red light therapy does not obey the rule that if some is good, more is better. It follows a biphasic dose-response, where below a threshold nothing happens and within a sweet spot you get more ATP and more repair signaling.
Push past the optimal dose and the curve bends back down. Huang and colleagues documented that ATP production and cell proliferation rise then fall as dose climbs, with the useful window landing around a few joules per square centimeter.
The goal is not maximum exposure but the right exposure, repeated consistently. A few minutes per area a few times a week beats marathon sessions.
What the evidence actually supports
The strongest evidence sits where the physics cooperates, at the surface. Randomized work has shown improved skin complexion and greater collagen density, and near-infrared light around exercise reduced markers of muscle damage.
The mechanism is real, the wavelength determines the target, and the dose has a ceiling. Red light therapy is a legitimate tool for skin, hair, recovery, and joint pain, not a cure-all. Full-body red light therapy is available at Citrin Longevity in West Hollywood, Los Angeles.
How to read a red light therapy claim
Three questions separate signal from noise. Which wavelengths does it emit, 660nm for skin and hair or 850nm for anything deeper than a few millimeters, and what is the dose? Does the claimed benefit match a tissue the light can actually reach?
None of this makes red light therapy magic, and that is rather the point. It is a specific intervention with a specific mechanism, real evidence in defined areas, and physical limits you can reason about.
- Hamblin, M.R. Photobiomodulation, Photomedicine, and Laser Surgery, 2018. Mechanisms and mitochondrial redox signaling in photobiomodulation. Reviews how light absorbed by cytochrome c oxidase raises ATP and modest reactive oxygen species, driving repair and pre-conditioning effects. PubMed
- Huang, Y.-Y., Chen, A.C.-H., Carroll, J.D. & Hamblin, M.R. Dose-Response, 2009;7(4):358–383. Establishes the biphasic (Arndt-Schulz) dose-response: low doses stimulate while high doses lose or reverse the benefit. PubMed
- Avci, P., Gupta, A., et al. Seminars in Cutaneous Medicine and Surgery, 2013;32(1):41–52. Low-level laser (light) therapy in skin. Reviews evidence for wrinkles, scars, and collagen via fibroblast activation. PubMed
- Wunsch, A. & Matuschka, K. Photomedicine and Laser Surgery, 2014;32(2):93–100. Randomized controlled trial: 660nm red-light treatment significantly reduced fine lines and roughness and increased intradermal collagen density in 136 volunteers. PubMed
- de Freitas, L.F. & Hamblin, M.R. IEEE Journal of Selected Topics in Quantum Electronics, 2016;22(3):348–364. Proposed mechanisms of photobiomodulation. Details nitric oxide photodissociation from cytochrome c oxidase and the downstream signaling cascade. PubMed
