Not all wavelengths of red and near-infrared light behave the same. Red light is absorbed and scattered closer to the skin surface, while near-infrared light travels further into tissue before it is attenuated. This reference page lists the wavelengths used in photobiomodulation, says which ones have a verified wavelength-specific study behind them, and explains how to choose.
The short answer: red light around 630-670nm is used for skin, hair and surface wounds, and near-infrared around 810-850nm for muscle, tendon, joint and transcranial targets. Cell absorption measurements show cytochrome c oxidase bands in both regions, including near 760nm, so the 700-770nm range is not an absorption “dead zone”; it simply has little clinical research. Measured power density per wavelength matters as much as wavelength count.
Master Wavelength Reference Table
The table below lists each wavelength and, where we could verify one, a study that actually used it. Many rows have no wavelength-specific clinical source; we say so rather than attach a paper that did not test that wavelength. We have removed the penetration-depth and evidence-strength ratings that earlier versions of this table carried, because the cited sources did not establish them.
| Wavelength | Type | Verified wavelength-specific study | What that study tested |
|---|---|---|---|
| 620nm | Visible red | None cited | - |
| 630nm | Visible red | Wunsch & Matuschka, 2014 (PMID: 24286286) | Randomised controlled trial, 611-650nm or 570-850nm polychromatic light twice weekly for skin roughness, wrinkles and collagen density |
| 633nm | Visible red | Lee et al., 2007 (PMID: 17566756) | Randomised split-face trial of 633nm, 830nm, both or sham for facial wrinkles |
| 640nm | Visible red | None cited | - |
| 650-655nm | Visible red | Lanzafame et al., 2013 (PMID: 24078483) | Double-blind RCT, 655nm laser and LED helmet for male pattern hair loss |
| 660nm | Visible red | Papageorgiou et al., 2000 (PMID: 10809858) | Randomised trial, 415nm blue plus 660nm red light daily for acne |
| 670nm | Visible red | Shinhmar et al., 2020 (PMID: 32596723) | 670nm light and age-related decline in human colour and rod vision |
| 680-770nm | Visible red / NIR | None cited | - |
| 780nm | NIR | None cited | - |
| 808-810nm | NIR | Stergioulas, 2008 (PMID: 18341417) | RCT, 810nm laser vs placebo for frozen shoulder pain and disability |
| 830nm | NIR | Lee et al., 2007 (PMID: 17566756) | As above: 830nm alone and combined with 633nm for facial wrinkles |
| 850nm | NIR | Dogan et al., 2010 (PMID: 21120304) | RCT, 850nm laser vs placebo laser for shoulder impingement: no difference between groups |
| 880nm | NIR | None cited | - |
| 904nm | NIR (pulsed) | None cited | - |
| 940nm | NIR | None cited | - |
| 980nm | NIR | None cited | - |
| 1060nm | NIR | None cited | - |
| 1072nm | NIR | None cited | - |
| 1100nm | NIR | None cited | - |
The Visible Red Window: 620-700nm
The visible red spectrum is what most people picture when they think of red light therapy: the deep, warm glow that is visible to the naked eye. These wavelengths interact mainly with the skin and superficial tissues.
Why 660nm Dominates
660nm is used in a large share of consumer devices. The main reasons:
Cytochrome c oxidase absorption. Cytochrome c oxidase (CCO), the terminal enzyme of the mitochondrial respiratory chain, is considered the main photoacceptor for red and near-infrared light (Karu, 2010; PMID: 20681024). Absorption measurements in cell monolayers show a 650-680nm band characteristic of the relatively oxidised enzyme, with a peak near 665nm (Karu et al., 2005; PMID: 16125966). Downstream effects of this absorption include increases in ATP, a brief burst of reactive oxygen species and an increase in nitric oxide (Hamblin, 2017; PMID: 28748217).
Suitable for surface targets. Red light is absorbed closer to the surface than near-infrared, which suits skin and hair follicle targets.
Clinical use of nearby wavelengths. Trials cited on this page used red light close to 660nm: 611-650nm broadband light for skin rejuvenation (Wunsch & Matuschka, 2014; PMID: 24286286), 655nm for hair growth (Lanzafame et al., 2013; PMID: 24078483) and 660nm with blue light for acne (Papageorgiou et al., 2000; PMID: 10809858).
The Other Red Wavelengths
620-630nm wavelengths are used for superficial targets such as the epidermis and surface-level skin conditions, and some LED masks use 630nm. We have not found a head-to-head trial comparing 630nm with 660nm.
633nm is used in several LED face devices. Lee et al. (2007; PMID: 17566756) randomised 76 patients with facial wrinkles to 830nm, 633nm, both, or sham light twice a week for four weeks, and reported wrinkle reductions of up to 36% and elasticity increases of up to 19% versus baseline in the three active groups, with more collagen and elastic fibres on histology. For acne, the best-known trial used 415nm blue with 660nm red rather than 633nm: after 12 weeks of daily 15-minute treatment, inflammatory lesions improved by a mean of 76% (Papageorgiou et al., 2000; PMID: 10809858).
670nm is the wavelength used in Glen Jeffery’s research at University College London, which found that 670nm light improved colour contrast sensitivity (the blue, tritan axis) and rod thresholds in people around 40 and older, with no effect in younger participants (Shinhmar et al., 2020; PMID: 32596723). Earlier NASA-funded LED work by Whelan et al. (2001; PMID: 11776448) mixed cell culture, rat wound models and uncontrolled human reports across a variety of wavelengths; its abstract does not report results by wavelength, so it is not a 670nm-specific trial.
680-700nm wavelengths sit at the boundary between visible red and near-infrared. They have substantially less published research, and few commercial devices target them.
The Near-Infrared Window: 700-1100nm
Near-infrared (NIR) wavelengths are invisible to the human eye. They travel further into tissue than visible red light, which is why they are used for muscle, joint and transcranial targets. Light is still attenuated exponentially as it passes through tissue, and delivering enough dose to the brain through the skull is a recognised challenge (Salehpour et al., 2018; PMID: 29327206).
The Critical Wavelengths
810nm is a common NIR wavelength in photobiomodulation. Karu and colleagues recorded an absorption peak at 813nm in cell monolayers (PMID: 16125966). Applications include:
- Transcranial PBM (brain health). Naeser et al. (2014; PMID: 24568233) treated 11 people with chronic mild traumatic brain injury using red and near-infrared LED cluster heads on the scalp and reported improvements on executive-function and verbal-learning tests. It was an open-protocol study with no control group, and the abstract does not state 810nm. Reviews cover research in dementia, Parkinson’s disease, stroke, brain trauma and depression (Salehpour et al., 2018; PMID: 29327206).
- Shoulder pain. Stergioulas (2008; PMID: 18341417) found that 810nm laser reduced pain and disability scores in frozen shoulder compared with placebo, though range of motion did not improve significantly.
- Muscle performance. Vieira et al. (2012; PMID: 21870127) found that 808nm laser after endurance training reduced a fatigue index more than the same training without laser.
830nm was one of the two wavelengths in the Lee et al. (2007) skin trial above, and Karu’s absorption measurements include a peak at 830nm. For joint pain, the Bjordal et al. (2003; PMID: 12775206) systematic review found that low level laser therapy within a location-specific dose range reduced pain in chronic joint disorders; the abstract does not attribute this to a particular wavelength.
850nm is common in consumer panels, much as 660nm is for visible red. Clinical data specific to 850nm are thinner than marketing suggests. In Dogan et al. (2010; PMID: 21120304), 850nm laser plus exercise was no better than placebo laser plus exercise for shoulder impingement. Ferraresi et al. (2012; PMID: 23626925) is sometimes cited for 850nm muscle recovery, but it is a review of low level laser therapy for muscle across animal and human studies, not an 850nm experiment.
880nm is less commonly used in modern consumer devices than 850nm.
904nm is used mainly in pulsed laser devices rather than LED panels, particularly in physiotherapy. We do not cite a 904nm-specific trial on this page.
940nm and Beyond
940nm appears in some devices aimed at fat reduction or body contouring. We found no verified clinical trial to cite for it here.
980nm sits near a water absorption band, so more of its energy is absorbed by tissue water and converted to heat. It is used in some surgical and thermal laser applications.
1060nm appears in some multi-wavelength devices. We found no verified clinical trial to cite for it here.
Beyond 1100nm, water absorption increases and the interaction becomes increasingly thermal. This is generally treated as outside the photobiomodulation window.
The 700-770nm Range: Not an Absorption Dead Zone
Many red light therapy sites describe 700-770nm as a “dead zone” with poor absorption. The primary absorption data do not support that. Karu et al. (2005; PMID: 16125966) recorded absorption spectra of cell monolayers and found peaks at 616, 665, 760, 813 and 830nm. The 710-790nm region, including the 760nm peak, was characteristic of the relatively reduced form of cytochrome c oxidase, and the peaks matched the action spectra of long-term cellular responses.
What the 700-770nm range does lack is clinical research and consumer devices. Most trials and panels use 630-670nm or 800-850nm, so there is little outcome data to judge 700-770nm devices by. If a device relies on these wavelengths, ask the manufacturer what evidence supports its choice.
Multi-Wavelength vs Single-Wavelength Devices
Modern red light therapy devices range from single-wavelength panels to devices offering five or more wavelengths simultaneously. Understanding the trade-offs is important.
Two-Wavelength Devices (660nm + 850nm)
The most common configuration in consumer panels. This combination provides:
- One red wavelength in the oxidised-CCO absorption band and one near-infrared wavelength
- Superficial treatment (skin, hair) via 660nm
- Deeper treatment (muscle, joints) via 850nm
- The configuration most commonly sold for general use
For most users, a quality 660nm + 850nm device covers the common applications. This is a sensible starting point for general use.
Three-Wavelength Devices (e.g., 630nm + 660nm + 850nm)
Adding a third wavelength, typically a shorter red like 630nm, gives slightly broader red coverage. We found no trial showing a benefit over two wavelengths, though some users treating purely superficial skin conditions may prefer it.
Five-Wavelength Devices (e.g., 630nm + 660nm + 810nm + 830nm + 850nm)
Several premium devices (notably the BioMax series from PlatinumLED) offer five wavelengths spanning both visible red and NIR ranges. The rationale is:
- Multiple absorption peaks are targeted simultaneously
- Broader depth range, from shallow red to deeper near-infrared
- Condition versatility: a single device can be used for skin, muscle, joint and transcranial targets
The trade-off is cost and complexity. Total power is shared across the LED types, and the share each wavelength gets depends on how many LEDs of each type the panel carries and how efficient they are. Ask for measured irradiance per wavelength at your treatment distance rather than dividing the wall-plug wattage.
How to Decide
| Your primary goal | Recommended wavelengths | Why |
|---|---|---|
| Skin rejuvenation / anti-ageing | 630nm + 660nm | Red light used in skin trials |
| Acne | 415nm blue + 630-660nm red | Papageorgiou et al. (2000) used 415nm + 660nm |
| Hair growth | 650nm or 660nm | Lanzafame et al. (2013) used 655nm |
| Muscle recovery / sports | 810-850nm | Near-infrared reaches muscle |
| Joint pain / arthritis | 810-850nm | Near-infrared reaches joint structures |
| Brain health / TBI / cognition | Near-infrared (often 810nm) | Common in transcranial devices; research is early |
| General wellness (all-purpose) | 660nm + 850nm | Covers red and near-infrared |
| Deep tissue + superficial skin | 630nm + 660nm + 810nm + 850nm | Full-spectrum approach |
Common Wavelength Combinations in Popular Devices
| Device | Wavelengths | Configuration |
|---|---|---|
| Joovv Solo 3.0 | 660nm + 850nm | Dual-wavelength, switchable or combined |
| PlatinumLED BioMax 600 | 630nm + 660nm + 810nm + 830nm + 850nm | Five wavelengths, simultaneous |
| Mito Red MitoPRO 1500 | 630nm + 660nm + 830nm + 850nm | Four wavelengths |
| Rouge Ultimate | 660nm + 850nm | Dual-wavelength |
| Bestqool Pro300 | 660nm + 850nm | Dual-wavelength, budget option |
| Celluma Pro | 465nm + 640nm + 880nm | Three wavelengths (includes blue) |
| Omnilux Contour | 633nm + 830nm | Medical-grade, FDA-cleared |
| Kineon Move+ Pro | 808nm (laser) + 650nm (LED) | Laser/LED hybrid for joints |
| Flexbeam | 630nm + 660nm + 850nm | Three-wavelength, portable |
| Hooga HG1500 | 660nm + 850nm | Dual-wavelength, value |
Why Some Devices Use 5 Wavelengths vs 2
The five-wavelength approach rests on a plausible but unproven hypothesis: that targeting multiple absorption peaks simultaneously produces a synergistic effect greater than the sum of individual wavelengths.
The argument for multi-wavelength:
- Broader absorption coverage. CCO shows several absorption peaks across red and near-infrared. Hitting several simultaneously may activate a larger proportion of the enzyme population.
- Depth stratification. Different wavelengths are absorbed at different depths, so a multi-wavelength device treats shallow and deeper tissue at once.
- Diverse chromophore targeting. Beyond CCO, other cellular chromophores respond to different wavelengths. Multi-wavelength exposure may trigger multiple pathways.
The argument for simplicity (two wavelengths):
- Evidence concentration. The trials cited on this site mostly used one or two wavelengths, and we found very few trials directly comparing multi-wavelength with dual-wavelength outcomes. In Wunsch & Matuschka (2014), broadband 570-850nm light showed no advantage over red-only 611-650nm light.
- Power distribution. Adding wavelengths does not add power. Each wavelength’s share depends on the LED count and efficiency for that wavelength, so check measured irradiance per wavelength.
- Cost. Multi-wavelength devices cost more. If the clinical benefit is marginal, the additional investment may not be justified.
The honest answer: for most people, 660nm + 850nm is a sensible choice. Multi-wavelength devices are a reasonable choice if budget allows, but the incremental benefit over a quality dual-wavelength panel has not been demonstrated in clinical research.
Matching Wavelengths to Your Condition
When choosing a device, start with the condition you want to treat and work backwards to the wavelength:
| Condition Category | Target Tissue | Relative depth | Common Wavelengths |
|---|---|---|---|
| Surface skin (acne, redness, fine lines, texture) | Epidermis, upper dermis | Shallow | 630nm, 633nm, 660nm |
| Deep skin (collagen, scarring, pigmentation) | Dermis | Shallow | 660nm, 670nm |
| Hair growth | Hair follicle | Shallow | 650nm, 660nm |
| Superficial wounds | Epidermis, dermis | Shallow | 630nm, 660nm, 670nm |
| Muscle recovery | Skeletal muscle | Deep | 810nm, 850nm |
| Tendons / ligaments | Connective tissue | Moderate to deep | 810nm, 830nm |
| Joint pain / arthritis | Synovium, cartilage, capsule | Deep | 830nm, 850nm |
| Bone healing / density | Bone | Deep | Near-infrared; no verified wavelength-specific trial cited here |
| Brain / cognitive | Cerebral cortex | Deep (through skull) | Near-infrared |
| Nerve pain / neuropathy | Peripheral nerves | Moderate to deep | 810nm, 830nm |
| Fat reduction | Subcutaneous adipose | Moderate | No verified trial cited here |
Key Takeaways
- Red (630-670nm) and near-infrared (810-850nm) are the two workhorse ranges: red for skin and hair, NIR for deeper targets
- 700-770nm is not an absorption dead zone: cell measurements show a CCO absorption peak near 760nm, but there is little clinical research at these wavelengths
- Multi-wavelength devices offer theoretical advantages but no demonstrated superiority over well-made dual-wavelength panels
- Always match wavelength to your condition: surface skin issues need red, deeper tissue needs NIR
- Many wavelengths lack a wavelength-specific trial; treat confident claims about them with caution
- More wavelengths does not automatically mean better: measured power density per wavelength matters as much as wavelength count
Frequently Asked Questions
Which red light therapy wavelengths are the most important?
Most consumer devices and many of the trials cited here use red light around 630-670nm and near-infrared light around 810-850nm. Red light is absorbed closer to the surface, so it is used for skin and hair; near-infrared travels further into tissue, so it is used for joints, muscles and transcranial work.
Why do most red light therapy devices use 660nm?
Cytochrome c oxidase, the proposed photoacceptor, shows an absorption band at 650-680nm in its relatively oxidised form, with a peak near 665nm, so 660nm sits in that band. Red LEDs at 660nm are also widely available, and trials cited on this page used nearby red wavelengths, such as 655nm for hair growth and 660nm with blue light for acne.
Is 700-770nm a dead zone in red light therapy?
Not on absorption grounds. Karu and colleagues recorded an absorption peak near 760nm in cell monolayers, within a 710-790nm region characteristic of the relatively reduced form of cytochrome c oxidase. What the 700-770nm range lacks is clinical research and consumer devices, so there is little outcome data to judge it by.
Are five-wavelength devices better than two-wavelength devices?
Not proven. Multi-wavelength devices are a reasonable choice if budget allows, but we found no trial showing that five wavelengths beat a quality dual-wavelength panel. A panel’s total power is shared across its LED types, and the split depends on how many LEDs of each type it carries, so compare measured irradiance per wavelength rather than wall-plug wattage.
Which wavelength should I use for my condition?
Match the wavelength to the depth of the target tissue. Surface skin conditions such as acne, redness and fine lines are usually treated with red light (630-660nm). Hair growth trials have used 650-660nm. Muscle, joint and tendon targets usually call for near-infrared (810-850nm). Transcranial devices use near-infrared.
Which wavelength is best for deep tissue and brain health?
Near-infrared wavelengths in the 800-850nm region are the usual choice because they travel further into tissue than red light. Karu and colleagues recorded absorption peaks at 813nm and 830nm in cell monolayers. Transcranial research is still early and has not settled on a single best wavelength.
References
- Karu TI, Pyatibrat LV, Kolyakov SF, Afanasyeva NI. Absorption measurements of a cell monolayer relevant to phototherapy: reduction of cytochrome c oxidase under near IR radiation. J Photochem Photobiol B. 2005;81(2):98-106. PMID: 16125966
- Karu TI. Multiple roles of cytochrome c oxidase in mammalian cells under action of red and IR-A radiation. IUBMB Life. 2010;62(8):607-610. PMID: 20681024
- Hamblin MR. Mechanisms and applications of the anti-inflammatory effects of photobiomodulation. AIMS Biophys. 2017;4(3):337-361. PMID: 28748217
- Wunsch A, Matuschka K. A controlled trial to determine the efficacy of red and near-infrared light treatment in patient satisfaction, reduction of fine lines, wrinkles, skin roughness, and intradermal collagen density increase. Photomed Laser Surg. 2014;32(2):93-100. PMID: 24286286
- Lee SY, et al. A prospective, randomized, placebo-controlled, double-blinded, and split-face clinical study on LED phototherapy for skin rejuvenation. J Photochem Photobiol B. 2007;88(1):51-67. PMID: 17566756
- Lanzafame RJ, et al. The growth of human scalp hair mediated by visible red light laser and LED sources in males. Lasers Surg Med. 2013;45(8):487-495. PMID: 24078483
- Papageorgiou P, Katsambas A, Chu A. Phototherapy with blue (415 nm) and red (660 nm) light in the treatment of acne vulgaris. Br J Dermatol. 2000;142(5):973-978. PMID: 10809858
- Shinhmar H, et al. Optically improved mitochondrial function redeems aged human visual decline. J Gerontol A Biol Sci Med Sci. 2020;75(9):e49-e52. PMID: 32596723
- Whelan HT, et al. Effect of NASA light-emitting diode irradiation on wound healing. J Clin Laser Med Surg. 2001;19(6):305-314. PMID: 11776448
- Salehpour F, et al. Brain photobiomodulation therapy: a narrative review. Mol Neurobiol. 2018;55(8):6601-6636. PMID: 29327206
- Naeser MA, et al. Significant improvements in cognitive performance post-transcranial, red/near-infrared light-emitting diode treatments in chronic, mild traumatic brain injury: open-protocol study. J Neurotrauma. 2014;31(11):1008-1017. PMID: 24568233
- Stergioulas A. Low-power laser treatment in patients with frozen shoulder: preliminary results. Photomed Laser Surg. 2008;26(2):99-105. PMID: 18341417
- Vieira WH, et al. Effects of low-level laser therapy (808 nm) on isokinetic muscle performance of young women submitted to endurance training: a randomized controlled clinical trial. Lasers Med Sci. 2012;27(2):497-504. PMID: 21870127
- Bjordal JM, et al. A systematic review of low level laser therapy with location-specific doses for pain from chronic joint disorders. Aust J Physiother. 2003;49(2):107-116. PMID: 12775206
- Dogan SK, Ay S, Evcik D. The effectiveness of low laser therapy in subacromial impingement syndrome: a randomized placebo controlled double-blind prospective study. Clinics (Sao Paulo). 2010;65(10):1019-1022. PMID: 21120304
- Ferraresi C, Hamblin MR, Parizotto NA. Low-level laser (light) therapy (LLLT) on muscle tissue: performance, fatigue and repair benefited by the power of light. Photonics Lasers Med. 2012;1(4):267-286. PMID: 23626925
Find the right device
Compare 20+ red light therapy devices by wavelength, irradiance, and value.
1060nm & 1064nm: Beyond Standard RLT
In-depth analysis of 1060nm & 1064nm: beyond standard rlt with PubMed citations, absorption data, and clinical applications.
630nm Red Light: Benefits, Penetration & Studies
In-depth analysis of 630nm red light: benefits, penetration & studies with PubMed citations, absorption data, and clinical applications.
650nm Red Light: The Sweet Spot?
In-depth analysis of 650nm red light: the sweet spot? with PubMed citations, absorption data, and clinical applications.
Evidence-based RLT updates
No hype, just research. New studies, protocol updates, and device test results in your inbox.