850nm Near-Infrared: The Deep Penetration Wavelength

If 660nm is the gold standard for skin, 850nm is its counterpart for everything beneath the surface. This near-infrared wavelength penetrates 4–5cm into human tissue — deep enough to reach muscles, joints, bones, and even organs. It aligns with the second absorption peak of cytochrome c oxidase, carries a substantial body of clinical evidence, and has become the dominant near-infrared wavelength in consumer red light therapy panels worldwide.

850nm is invisible to the human eye. You cannot see it working. But the photons are there, passing through layers of skin, fat, and connective tissue to interact with mitochondria in cells that visible red light simply cannot reach.

The Second Absorption Peak of Cytochrome c Oxidase

Cytochrome c oxidase (CCO) has two major absorption regions relevant to photobiomodulation: one in the visible red range (peaking around 660nm) and one in the near-infrared range (peaking across the 810–850nm band). The near-infrared peak corresponds to the oxidised form of the enzyme — specifically, the electronic transitions within the CuA centre and the haem a moiety of CCO.

Karu (2008) mapped this absorption band and identified strong activity across the 810–860nm range, with the exact peak position varying slightly depending on the oxidation state of the enzyme and the measurement methodology. The breadth of this absorption band is actually advantageous — it means that wavelengths from 810nm through 860nm all interact meaningfully with CCO, giving manufacturers and researchers some flexibility in wavelength selection.

When 850nm photons are absorbed by the oxidised form of CCO, the same core mechanism operates as with 660nm: nitric oxide is photodissociated from the enzyme’s binding sites, electron transport is restored, oxygen consumption increases, and ATP production rises. The downstream signalling cascade — increased ROS signalling, NF-kB activation, gene expression changes — follows the same pathway (Hamblin, 2017; PMID: 28748217).

The key difference is not mechanism but location. Because 850nm penetrates centimetres rather than millimetres, these cellular effects occur in tissues that are inaccessible to visible red light.

Reference: Karu, T.I. (2008). “Mitochondrial Signaling in Mammalian Cells Activated by Red and Near-IR Radiation.” Photochemistry and Photobiology, 84(5), 1091–1099. PMID: 18651871

Penetration Depth: ~4–5cm

At 850nm, the primary absorbers that limit penetration in the visible range — melanin and haemoglobin — have dramatically reduced absorption coefficients. Melanin absorption drops roughly tenfold between 600nm and 850nm. Deoxyhaemoglobin absorption falls to near its minimum. Water absorption, which will eventually limit near-infrared penetration, remains negligible at 850nm (it does not become significant until approximately 970nm).

The result is that 850nm photons travel through tissue far more efficiently than visible red photons. Published tissue optics measurements (Bashkatov et al., 2005; Jacques, 2013) indicate an effective penetration depth of approximately 4–5cm at 850nm, meaning a therapeutically relevant fraction of photons reaches structures at that depth.

In practical anatomical terms, 4–5cm of penetration reaches:

Knee joint cartilage and synovial space — typically 1–2cm below the skin surface
Rotator cuff tendons — 2–4cm deep in the shoulder
Lumbar paraspinal muscles — 2–5cm from the skin surface
Hip joint — the superficial aspect, though the deepest part of the hip can be 7–10cm beneath the surface and remains challenging
Cerebral cortex — 1.5–3cm from the scalp surface (though 810nm is preferred for transcranial applications due to skull transmission properties)
Peripheral nerves — at various depths throughout the body

This depth advantage is why 850nm is essential for musculoskeletal and orthopaedic applications. A device emitting only 660nm cannot deliver meaningful energy to a knee joint or a deep muscle belly — the photons are absorbed and scattered before they reach the target.

Reference: Jacques, S.L. (2013). “Optical properties of biological tissues: a review.” Physics in Medicine and Biology, 58(11), R37. PMID: 23666068

Best Conditions for 850nm

Joint Pain and Arthritis

Osteoarthritis and inflammatory joint conditions are among the most compelling applications for 850nm near-infrared therapy. The target tissues — cartilage, synovial membrane, subchondral bone — sit beneath layers of skin, fat, and muscle that only NIR wavelengths can penetrate effectively.

Hegedus et al. (2009) studied 830nm laser therapy (closely comparable to 850nm) in knee osteoarthritis patients and found statistically significant reductions in pain and improvements in microcirculation compared to placebo. The treatment group showed a 50% reduction in pain scores after 10 sessions (PMID: 19735396).

Bjordal et al. (2003) conducted a systematic review and meta-analysis of low-level laser therapy for joint disorders, finding that wavelengths in the 820–830nm range produced significant pain reduction and functional improvement in osteoarthritis and other chronic joint conditions. The optimal dose was identified as 1–4 J per point, applied directly over the joint line (PMID: 14522484).

Alves et al. (2013) demonstrated in a rat model that 850nm irradiation at 4 J/cm2 reduced inflammation in experimentally induced arthritis, decreased TNF-alpha levels, and preserved cartilage integrity compared to untreated controls (PMID: 22926505).

Muscle Recovery and Exercise Performance

The evidence for 850nm in athletic recovery is particularly robust. Ferraresi et al. (2012) published a comprehensive review of PBM for exercise performance and found consistent benefits when 808–850nm light was applied before or after exercise: reduced creatine kinase (a biomarker of muscle damage), decreased delayed-onset muscle soreness (DOMS), faster return to baseline performance, and improved maximum voluntary contraction (PMID: 22817596).

Leal-Junior et al. (2015) built on this work with additional systematic reviews and meta-analyses, confirming that near-infrared PBM (predominantly 810–850nm) applied before exercise reduced muscle fatigue, and applied after exercise accelerated recovery. The effect sizes were clinically meaningful — approximately 20% reduction in CK levels and significant reduction in subjective soreness ratings (PMID: 25803069).

Baroni et al. (2010) specifically tested 850nm on the quadriceps muscles during an eccentric exercise protocol and found that pre-exercise irradiation at 850nm significantly reduced muscle damage and improved isometric force recovery at 24, 48, 72, and 96 hours post-exercise compared to placebo (PMID: 20662758).

The mechanism involves both metabolic and anti-inflammatory pathways: 850nm photons reach the mitochondria in muscle fibres, boost ATP availability for repair processes, and simultaneously downregulate inflammatory mediators (IL-1beta, IL-6) that contribute to DOMS.

Deep Tissue Inflammation

Chronic inflammatory conditions affecting deep structures — bursitis, tendinopathies in deep tendons, fascial inflammation — respond to 850nm precisely because the photons can reach the inflamed tissue. Surface-level anti-inflammatory effects from 660nm cannot substitute when the inflammation is centimetres below the skin.

Tumilty et al. (2010) reviewed the evidence for PBM in tendinopathy, finding that studies using near-infrared wavelengths (808–850nm) at adequate doses reported positive outcomes for Achilles tendinopathy, lateral epicondylitis, and rotator cuff tendinopathy (PMID: 19913903).

Naterstad et al. (2022) conducted a meta-analysis specifically examining PBM for musculoskeletal disorders and found that near-infrared wavelengths (808–904nm) produced significant pain reduction and functional improvement across multiple conditions, with optimal effects at doses of 2–8 J per treatment point (PMID: 34947893).

Wound Healing (Deep Structures)

While 660nm handles surface wound healing effectively, post-surgical healing involving deep tissue layers benefits from 850nm. After orthopaedic surgery, abdominal procedures, or deep tissue injuries, 850nm can penetrate to the internal wound site and promote healing through the same mechanisms — increased fibroblast activity, enhanced angiogenesis, modulated inflammation — but at greater depths.

Combination therapy with 660nm (for the skin incision) and 850nm (for the underlying tissue) is a logical approach for post-surgical recovery, though controlled clinical trials specifically testing this combination in surgical patients are still emerging.

Key Clinical Studies Citing 850nm (and Adjacent NIR Wavelengths)

Study	Year	Wavelength	Condition	Finding	PMID
Ferraresi et al.	2012	808–850nm	Muscle recovery	Reduced CK, DOMS; improved performance	22817596
Baroni et al.	2010	850nm	Exercise-induced damage	Reduced quadriceps damage, faster recovery	20662758
Leal-Junior et al.	2015	810–850nm	Athletic performance	Meta-analysis: reduced fatigue, accelerated recovery	25803069
Hegedus et al.	2009	830nm	Knee osteoarthritis	50% pain reduction, improved microcirculation	19735396
Bjordal et al.	2003	820–830nm	Chronic joint disorders	Significant pain reduction and functional improvement	14522484
Alves et al.	2013	850nm	Arthritis (animal)	Reduced TNF-alpha, preserved cartilage	22926505
Naterstad et al.	2022	808–904nm	Musculoskeletal disorders	Pain reduction and functional improvement	34947893

How 850nm Compares to 810nm and 830nm

All three wavelengths fall within the near-infrared CCO absorption band and have strong clinical evidence. The differences are subtle, but they matter for specific applications.

850nm vs 810nm

810nm is the preferred wavelength for transcranial photobiomodulation — delivering light through the skull to brain tissue. Tedord et al. (2015) measured photon transmission through human cadaver skulls and found that the 800–830nm range offered optimal transmission, with 810nm performing particularly well (PMID: 26098775). The Vielight Neuro, the most widely studied transcranial PBM device, uses 810nm pulsed at 10 Hz for its nasal applicator and 40 Hz for its transcranial LEDs.

For non-cranial deep tissue applications (muscles, joints), the difference between 810nm and 850nm is marginal. 850nm offers slightly deeper penetration in soft tissue because water absorption is still negligible at this wavelength, while at 810nm it is fractionally lower — but the practical difference amounts to millimetres at most. Both wavelengths interact strongly with CCO.

The reason consumer panels favour 850nm over 810nm is primarily LED manufacturing: gallium arsenide (GaAs) LEDs at 850nm are among the most efficient and cost-effective near-infrared emitters available, produced at massive scale for applications ranging from remote controls to security cameras. 810nm LEDs exist but are less common and sometimes less efficient at high power outputs.

Choose 810nm if your primary interest is brain health or neurological applications. Choose 850nm for general-purpose deep tissue therapy.

850nm vs 830nm

830nm has strong clinical heritage, particularly through the Thor Laser system (a medical-grade PBM device used in hundreds of studies) and the Celluma LED panel (FDA-cleared for pain, acne, and anti-ageing). Many of the foundational studies in orthopaedic PBM — the Bjordal meta-analyses, the Hegedus knee OA trial — used 830nm.

Biologically, 830nm sits closer to the estimated CCO absorption peak in the near-infrared (some researchers place it at 820–830nm rather than 850nm). This might suggest 830nm has a marginal advantage in CCO interaction. However, 850nm compensates with slightly deeper penetration, and the broad absorption band of CCO means both wavelengths are well within the active range.

In practice, the clinical outcomes for 830nm and 850nm are comparable. If you already own a device with 830nm, there is no compelling reason to switch to 850nm, and vice versa. The difference in therapeutic effect for most conditions is within the noise of individual variation.

Choose 830nm if you want to match the exact wavelength used in many clinical trials. Choose 850nm for the widest device selection, best LED efficiency, and marginally deeper penetration.

Why Most Panels Pair 660nm + 850nm

The pairing of 660nm and 850nm has become the default configuration for consumer red light therapy panels, and the logic is straightforward:

Complementary tissue targets. 660nm treats the surface — skin, dermis, superficial capillaries, hair follicles. 850nm treats the depths — muscles, tendons, joints, bones. Together, they cover virtually every tissue from the skin surface to several centimetres beneath it. A single treatment session with a dual-wavelength panel irradiates the full range of accessible tissue.

Both CCO peaks. The combination hits both the red (reduced-form) and near-infrared (oxidised-form) absorption peaks of CCO. In any given cell, CCO molecules exist in both oxidation states simultaneously. Delivering photons at both peaks maximises the chance of productive photon absorption regardless of the enzyme’s current state.

Additive anti-inflammatory effects. 660nm reduces surface inflammation (redness, swelling in skin) while 850nm reduces deep inflammation (joint swelling, muscle inflammation). For conditions with both surface and deep components — such as post-surgical recovery, sports injuries, or inflammatory arthritis — the combination addresses all layers.

LED efficiency. Both 660nm (AlGaInP) and 850nm (GaAs) LEDs are manufactured at scale, are highly efficient, and are available at high power outputs. The dual-wavelength panel does not sacrifice performance to achieve wavelength diversity.

Research validation. While many individual studies test a single wavelength, combination therapy with red + NIR has been validated in clinical contexts. Calderhead (2007) argued that multi-wavelength phototherapy is more likely to produce comprehensive tissue effects than single-wavelength therapy (PMID: 18049060). The widespread adoption of 660nm + 850nm panels effectively tests this hypothesis in millions of users worldwide.

The typical configuration

Most panels alternate 660nm and 850nm LEDs in a chequerboard or stripe pattern across the panel surface. Some devices allow users to select red only, NIR only, or combined modes. For general health and wellness, combined mode is recommended unless you have a specific reason to isolate one wavelength — for example, if treating only skin conditions and wanting to maximise 660nm irradiance without diluting power across 850nm LEDs.

Dosing Guidelines for 850nm

Published research and WALT (World Association for Photobiomodulation Therapy) recommendations suggest the following dose ranges for near-infrared PBM at 850nm:

Joint pain / arthritis: 4–8 J per treatment point, applied directly over the joint, 3–5 times per week
Muscle recovery (pre-exercise): 20–60 J total dose over the target muscle group, applied 3–6 hours before or immediately before exercise
Muscle recovery (post-exercise): 20–60 J total dose, applied within 4 hours of exercise
Deep tissue inflammation: 4–8 J per point, 3–5 times per week
General wellness: 3–6 J/cm2 per session, 3–5 times per week

At typical consumer panel irradiance levels of 60–100 mW/cm2 at 15cm distance (NIR component only), these doses correspond to treatment times of approximately 5–15 minutes per area depending on the target and panel specifications.

A critical consideration with deep tissue dosing: the dose that reaches the target tissue is substantially lower than the dose measured at the skin surface. If you deliver 6 J/cm2 at the skin, only a fraction — perhaps 10–30% depending on tissue type and depth — reaches a structure 3cm beneath the surface. This is why NIR dosing recommendations tend to be higher than red light doses: you need to compensate for attenuation through overlying tissue.

Safety Considerations

850nm is invisible, which creates a specific safety concern: you cannot see the light, so you may not realise you are exposing your eyes to high-intensity radiation. While the retina is less sensitive to near-infrared than to visible light, prolonged direct exposure to high-power 850nm LEDs can potentially cause thermal damage to the retina and lens.

Always follow the manufacturer’s eye safety guidelines. Many practitioners recommend closing your eyes during treatment (the eyelids provide some attenuation) or wearing protective goggles rated for near-infrared wavelengths if treating the face or head. For body treatments where you are not looking directly at the panel, the risk is lower but not zero due to reflections and scattered light.

Thermal effects are another consideration. At high irradiance levels, 850nm can produce a sensation of warmth — not from the visible light, but from absorption by water molecules in the skin and superficial tissue. This is generally mild and not harmful at typical consumer device power levels, but it serves as a useful indicator that energy is being delivered to tissue.

Summary

850nm near-infrared is the essential deep-penetration wavelength in red light therapy. It aligns with the near-infrared CCO absorption peak, penetrates 4–5cm into tissue, and has robust clinical evidence for joint pain, muscle recovery, deep tissue inflammation, and orthopaedic applications.

If your therapeutic targets are beneath the skin — joints, muscles, tendons, bones — 850nm is non-negotiable. Pair it with 660nm red light for comprehensive surface-to-depth coverage, or look at 810nm if transcranial brain therapy is your primary interest.

The 660nm + 850nm combination has become the industry standard not through marketing consensus but through photobiological logic: two wavelengths, two CCO peaks, two optical windows, and the broadest possible range of evidence-backed applications in a single device.

This article references peer-reviewed studies indexed on PubMed. It is for educational purposes only and does not constitute medical advice. Consult a healthcare professional before beginning any photobiomodulation protocol.