670nm Red Light: Vision & Mitochondrial Research

670nm sits at the longer end of the visible red light spectrum, just before the transition to deep red and near-infrared. It is a wavelength that has attracted significant research attention for a specific and compelling reason: its potential to slow and reverse age-related decline in mitochondrial function, particularly in the retina.

The driving force behind this research is Glen Jeffery, Professor of Neuroscience at University College London (UCL), whose work has opened an entirely new application area for photobiomodulation. This article examines what makes 670nm distinctive, the UCL research programme, the eye health implications, and where this wavelength fits in the broader PBM landscape.

670nm and Cytochrome C Oxidase

Like all wavelengths in the PBM therapeutic window, 670nm acts primarily on cytochrome c oxidase (CCO), the terminal enzyme of the mitochondrial electron transport chain. Absorption measurements in cell monolayers show a 650-680nm band characteristic of the relatively oxidised form of CCO, with a peak near 665nm, so 670nm sits inside that band (Karu et al., 2005; PMID: 16125966).

Both 660nm and 670nm fall within that band, and either can drive the core PBM mechanism: photodissociation of nitric oxide from CCO, restoration of electron transport, and increased ATP production. The key question is not whether 670nm activates CCO (it does) but whether it offers any advantage over 660nm.

Penetration Depth

Tissue penetration at 670nm is comparable to 660nm, with estimates placing it at approximately 2–3mm through skin. The small wavelength increase from 660 to 670nm provides a marginal reduction in haemoglobin and melanin absorption, but the practical difference in penetration is negligible for most applications.

Where penetration becomes particularly relevant for 670nm is in the eye. The retina is accessible to light without requiring penetration through thick tissue — photons entering through the pupil reach the retinal photoreceptors and retinal pigment epithelium directly. This makes the retina an ideal target for PBM, as the light delivery challenge is minimal compared to treating, say, a knee joint or the brain.

Glen Jeffery and the UCL Research Programme

The Ageing Mitochondria Problem

Glen Jeffery’s research programme at UCL’s Institute of Ophthalmology is built on a well-established observation: mitochondrial function declines with age. The UCL group reports that the retina ages faster than other organs, with a 70% reduction in ATP over life (Shinhmar et al., 2020; PMID: 32596723). This decline is particularly acute in the retina because photoreceptor cells (rods and cones) have among the highest metabolic demands of any cell in the body. They contain densely packed mitochondria to support the enormous energy requirements of phototransduction.

As mitochondrial efficiency drops, photoreceptor function degrades. Colour contrast sensitivity (mediated by cones) declines measurably from around age 40. This age-related retinal decline is distinct from pathological conditions like age-related macular degeneration (AMD), though compromised mitochondrial function likely contributes to AMD susceptibility.

The Drosophila Studies

Jeffery’s team began with fruit fly (Drosophila melanogaster) models. Begum et al. (2015; PMID: 25788488) exposed fruit flies daily to 670nm light and found elevated ATP and reduced inflammation with age, a 100-175% increase in the number of flies surviving into old age, and significantly better mobility in those old flies.

The same group later tested bumblebees poisoned with the neonicotinoid pesticide imidacloprid. Bees also given daily 670nm light had corrected ATP levels, better mobility, and eye recordings showing the pesticide-induced deficits were corrected (Powner et al., 2016; PMID: 27846310).

The Mouse and Rat Studies

Moving to mammalian models, Kokkinopoulos et al. (2013; PMID: 22595370) gave aged mice only five 90-second exposures to 670nm light over 35 hours. This significantly increased mitochondrial membrane polarisation in the retina and reduced macrophage numbers, TNF-alpha and other inflammatory markers.

Begum et al. (2013; PMID: 23469078) exposed aged complement factor H knockout mice, a model of age-related macular degeneration, to 670nm for 6 minutes twice a day for 14 days. Cytochrome c oxidase increased and complement component C3, an inflammatory marker in the outer retina, fell significantly, suggesting the treatment also changes the immune environment of the retina.

Human Trials: The Breakthrough

The UCL team’s first human study was published in The Journals of Gerontology (Shinhmar et al., 2020; PMID: 32596723). Participants aged 28-72 used 670nm light, and their visual function was measured psychophysically. In those around 40 and older, colour contrast sensitivity on the blue (tritan) axis improved significantly, the red (protan) axis improved but not significantly, and rod thresholds improved significantly. The light had no impact in younger participants.

This study was notable for several reasons:

1. Healthy ageing, not disease: The participants had normal age-related decline rather than an eye disease
2. Age-dependent effect: Benefits appeared only in people around 40 and older
3. Early evidence: The findings still need larger, independent replication

The Circadian Factor

A follow-up study by Shinhmar et al. (2021; PMID: 34819619) gave single 3-minute 670nm exposures at about 8 mW/cm² and investigated the timing dependency. Participants were exposed either in the morning (8:00–9:00 AM) or in the afternoon (12:00–1:00 PM). Only the morning-exposure group showed significant improvement in colour contrast sensitivity. Afternoon exposure produced no measurable benefit.

The proposed explanation involves circadian fluctuations in mitochondrial membrane potential. Mitochondria exhibit rhythmic variations in activity, with a natural dip in efficiency during certain phases. Morning exposure at 670nm may coincide with a window when mitochondria are most responsive to photostimulation — when they are transitioning from reduced nighttime activity to daytime metabolic demands. If PBM provides a “boost” at the right phase of this cycle, it amplifies the effect. If applied when mitochondria are already at peak function (afternoon), the additional stimulation has less impact.

This finding has broader implications for all PBM research. If treatment timing affects outcomes, then many existing studies — which rarely controlled for time of day — may have underestimated the potential of PBM by averaging across optimal and suboptimal timing windows.

Applications Beyond the Eye

Neuroprotection

670nm has been investigated for broader neuroprotective effects. Eells et al. (2003; PMID: 12626762) showed that three brief 670nm LED treatments led to significant recovery of rod and cone function and protected the retina from histological damage in methanol-poisoned rats, and suggested photobiomodulation may help other eye diseases in which mitochondrial dysfunction plays a role.

Fitzgerald et al. (2010; PMID: 20822460) used a 670nm LED array in an animal model of partial optic nerve transection. Treatment reduced oxidative stress in the part of the nerve vulnerable to secondary degeneration and restored normal visual function on behavioural tests.

Wound Healing

Like other red wavelengths, 670nm has been studied for wound healing. Whelan et al. (2001; PMID: 11776448) reported NASA-funded LED work using a variety of wavelengths: LED increased growth of cultured cells including fibroblasts by 140-200%, wound size decreased by up to 36% when combined with hyperbaric oxygen in ischaemic rat models, and there were uncontrolled reports in Navy personnel. The abstract does not break these results down by wavelength.

Ageing and Systemic Mitochondrial Function

Jeffery’s work has implications beyond ophthalmology. If 670nm light can improve mitochondrial function in the retina — one of the most metabolically demanding tissues — it is plausible that similar effects could occur in other tissues with high mitochondrial density. Research is ongoing into whether 670nm exposure could benefit age-related declines in muscle function, cognition, and metabolic health.

However, the retina’s advantage is accessibility: light reaches it directly through the pupil. Other tissues require the light to penetrate through skin, fat, muscle, and bone, which attenuates the signal substantially. For deeper targets, longer wavelengths (810–850nm near-infrared) may be more practical due to their superior tissue penetration.

670nm vs 660nm: Is There a Meaningful Difference?

The honest answer is that 670nm and 660nm are very similar in their photobiological properties. Both activate CCO, both penetrate to approximately the same depth through skin, and both fall within the optimal red wavelength range for PBM.

The primary distinction is in the research literature:

• 660nm has the broadest evidence base across PBM applications — skin, pain, wound healing, inflammation
• 670nm has the strongest evidence specifically for retinal and mitochondrial ageing applications, driven primarily by Jeffery’s UCL programme

There is no evidence that 670nm is superior to 660nm for general PBM applications, nor that 660nm would be inferior to 670nm for eye health. The difference in CCO absorption between these two wavelengths is minimal. The UCL team chose 670nm based on prior animal research showing efficacy at this specific wavelength, and their human protocols validated it — but they did not test 660nm as a comparator.

For practical purposes, a 660nm device used at the eyes with a similar protocol might well produce comparable results, though this remains untested in a controlled trial.

Devices Offering 670nm

670nm is less commonly found in consumer PBM devices than 660nm. The devices that do include it tend to be:

• Multi-wavelength panels — Some premium panels include 670nm as part of a broader wavelength mix. This is less common than the standard 660nm + 850nm combination, but manufacturers like PlatinumLED offer models with 670nm diodes.
• Eye-specific devices — Following the UCL research, a small number of devices have been developed specifically for retinal PBM. These are typically simple, low-power LED devices designed for brief morning use. Some are available commercially in the UK.
• Clinical and research devices — Several research-grade devices used in published trials operate at 670nm, but these are not typically available to consumers.

If your primary interest is eye health and you want to follow the UCL work, the 2021 study used about 8 mW/cm² of 670nm light at the eye for 3 minutes between 8 and 9 AM, delivered from modified LED torches (Shinhmar et al., 2021). Consult an optometrist or ophthalmologist first if you have any eye condition.

Dosing for 670nm

Eye Health Protocol (Based on UCL Research)

• Wavelength: 670nm
• Irradiance: about 8 mW/cm² at the cornea in the 2021 study (the authors note earlier work commonly targeted about 40 mW/cm²)
• Duration: 3 minutes
• Timing: Morning (8:00–9:00 AM appears optimal)
• Frequency: In the 2021 study a single exposure improved colour contrast thresholds for about a week; the best repeat schedule has not been established
• Method: In the 2021 study, light was delivered to one eye down a white, internally reflecting tube that fitted over the eye.

Important safety note: Although the UCL studies used low irradiances, anyone considering eye-directed PBM should consult an ophthalmologist, particularly if they have existing eye conditions. The studies used carefully calibrated devices with known power outputs. Using an uncharacterised device at unknown power levels could pose a risk to retinal tissue.

General PBM Applications

For non-eye applications (skin, wound healing), standard PBM dosing applies: 10–50 mW/cm² irradiance, 3–10 J/cm² fluence, 3–7 sessions per week.

Limitations and Unknowns

• The UCL human studies involved small sample sizes (24 participants in the main trial). Larger replication studies are needed.
• The circadian timing effect has been demonstrated once. Independent replication is important before it can be considered established.
• Long-term safety and efficacy data (beyond two weeks of treatment) are not yet published for the eye health protocol.
• Whether 670nm is genuinely superior to 660nm for retinal applications, or whether 660nm would work equally well, has not been tested in a head-to-head trial.
• The mechanism for the circadian timing dependency is hypothesised but not conclusively demonstrated.

Summary

670nm occupies a unique position in the PBM wavelength landscape. Whilst it shares the core mechanism of action with neighbouring red wavelengths — CCO absorption, NO displacement, ATP increase — it has become specifically associated with a compelling research programme on ageing and eye health.

Glen Jeffery’s work at UCL has demonstrated that brief morning exposure to 670nm light can measurably improve visual function in adults over 40 by boosting mitochondrial function in retinal cells. This is a low-cost, low-risk, evidence-backed intervention — though it is still early-stage and awaiting larger replication trials.

For general PBM use, 670nm is functionally similar to 660nm. For eye health specifically, 670nm has the unique advantage of being the exact wavelength validated in human trials. If retinal ageing is your primary concern, this wavelength — applied in the morning for just three minutes — is the most directly evidence-supported option available.

References

1. 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
2. Begum R, et al. Treatment with 670 nm light up regulates cytochrome C oxidase expression and reduces inflammation in an age-related macular degeneration model. PLoS One. 2013;8(2):e57828. PMID: 23469078
3. Begum R, et al. Near-infrared light increases ATP, extends lifespan and improves mobility in aged Drosophila melanogaster. Biol Lett. 2015;11(3):20150073. PMID: 25788488
4. Powner MB, et al. Improving mitochondrial function protects bumblebees from neonicotinoid pesticides. PLoS One. 2016;11(11):e0166531. PMID: 27846310
5. Kokkinopoulos I, et al. Age-related retinal inflammation is reduced by 670 nm light via increased mitochondrial membrane potential. Neurobiol Aging. 2013;34(2):602-609. PMID: 22595370
6. 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
7. Shinhmar H, Hogg C, Neveu M, Jeffery G. Weeklong improved colour contrasts sensitivity after single 670 nm exposures associated with enhanced mitochondrial function. Sci Rep. 2021;11(1):22872. PMID: 34819619
8. Eells JT, et al. Therapeutic photobiomodulation for methanol-induced retinal toxicity. Proc Natl Acad Sci U S A. 2003;100(6):3439-3444. PMID: 12626762
9. Fitzgerald M, et al. Near infrared light reduces oxidative stress and preserves function in CNS tissue vulnerable to secondary degeneration following partial transection of the optic nerve. J Neurotrauma. 2010;27(11):2107-2119. PMID: 20822460
10. 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