Red Light Therapy for Mitochondria: What the Science Shows
Red light therapy for mitochondria rests on one main finding: an enzyme inside mitochondria, cytochrome c oxidase, absorbs red and near-infrared light, and in cell and animal studies this is followed by more ATP, the cell's energy molecule, and a short burst of signals. In people, small studies have measured that enzyme signal rising in the tissue under the light within minutes. Below: the mechanisms, the evidence in cells, animals and people, wavelengths, dose, study settings and how to use a Kivorla mask or panel.
Key takeaways
- In a placebo-controlled study of 11 healthy adults, 1064 nm light on the forearm raised the oxidised cytochrome c oxidase signal in the tissue below, and the change was significant after about 1 minute. Wang, 2016
- In a sham-controlled study, an 810 nm LED on the forearm also raised the cytochrome c oxidase signal, with a similar time course to lasers and a smaller amplitude. Pruitt, 2022
- In strength-trained mice, LED light at 630 and 850 nm over the legs and lower back raised ATP more than tenfold. Ferraresi, 2015
- In mouse wounds, 635 nm light worked best at 2 J/cm², while 50 J/cm² gave a worse result than no light at all. Huang, 2009
- Cytochrome c oxidase absorbs light at roughly 600 to 700 nm and 760 to 940 nm, ranges that match the wavelengths often reported to work. Peeples, 2026
What does red light therapy for mitochondria do?
It gives cells a small, short-lived nudge: light absorbed by cytochrome c oxidase is linked to more ATP and to signals that change how the cell behaves for a while afterwards. Most of this detail comes from cell and animal work, with a few small human studies measuring the enzyme directly.
Mitochondria make ATP, and cytochrome c oxidase is the last enzyme in that chain. A 2018 review names it as the primary site of light absorption in mammalian cells. Hamblin, 2018 Cleveland Clinic describes red light as "thought to work by acting on the 'power plant'" of the cell. Cleveland Clinic
| Level | What has been shown | Study |
|---|---|---|
| Cells in a dish | 660 nm light raised cytochrome c oxidase activity and ATP in human skin cells | Houreld, 2012 |
| Animals | ATP up more than tenfold and muscle performance more than doubled in trained mice | Ferraresi, 2015 |
| People | Oxidised cytochrome c oxidase and blood oxygenation rose in forearm tissue under 1064 nm light | Wang, 2016 |
| People | The same signals rose under 800, 850 and 1064 nm lasers and an 810 nm LED | Pruitt, 2022 |
The evidence is strong in the lab and early in people. Our photobiomodulation therapy guide covers the wider evidence for pain and skin.
How does cytochrome c oxidase absorb red and near-infrared light?
The enzyme contains copper centres that absorb light, and its absorption peaks line up with the wavelengths that change cell behaviour. Researchers call it the leading photoacceptor, though the debate is still open.
In a classic 2004 study, researchers shone each wavelength from 600 to 860 nm on cells and found response peaks at 620, 680, 760 and 820 nm. Karu, 2004 A chemical that blocks the enzyme changed that pattern, pointing to cytochrome c oxidase as the absorber.
Human skin cells showed the same link. In a 2012 study, 660 nm light at 5 or 15 J/cm² raised the enzyme's activity, and ATP rose in normal cells and in cells modelling diabetes. Houreld, 2012
What do nitric oxide, reactive oxygen and calcium do after a session?
They carry the message on: light may free nitric oxide from the enzyme, produce a brief burst of reactive oxygen species and shift calcium, which together switch on genes for repair. The nitric oxide step is still described as a hypothesis.
- Nitric oxide: the leading idea is that light releases nitric oxide that was blocking the enzyme, so electron transport restarts. Hamblin, 2018
- Reactive oxygen: in normal cells, light causes "a brief burst of reactive oxygen species" that acts as a signal. Hamblin, 2017
- Calcium: light- and heat-sensitive ion channels in the cell membrane are a second route. Wang, 2017
- Water: one opinion paper proposes that longer near-infrared, such as 1064 nm, makes the water around mitochondria less viscous so the ATP-making motor turns faster. Sommer, 2019
The 2017 review adds: "One of the most reproducible effects of PBM is an overall reduction in inflammation." Hamblin, 2017 One researcher has also reported responses with mitochondria blocked, so they may not be the whole story. Peeples, 2026
Does red light raise ATP in people, or only in cells?
ATP rises are a frequent finding in cells and animals, but ATP is hard to measure in living people. The closest human evidence is a measured rise in the cytochrome c oxidase signal in the tissue under the light.
Most red light therapy mitochondria ATP data come from the lab. In trained mice, LED light at 80 mW/cm² and 7.2 J/cm² raised ATP more than tenfold. Ferraresi, 2015
In people, researchers used near-infrared spectroscopy, which reads the enzyme's state through the skin. In 11 healthy adults, 1064 nm light on the forearm raised oxidised cytochrome c oxidase and oxygenated haemoglobin against placebo, with the enzyme change significant after about 1 minute. Wang, 2016 A 2022 follow-up found the same rise with 800, 850 and 1064 nm lasers and an 810 nm LED. Pruitt, 2022
ATP works on a slower clock. In cultured mouse muscle cells, it peaked about 6 hours after the light, as our article on red light therapy for energy describes.
One 2026 biopsy study found a single high-power laser session did not change mitochondrial respiration in healthy young adults' thigh muscle or skin, so effects in healthy tissue may be small or need repeated sessions. Aussieker, 2026
Which wavelength suits mitochondria best: 660, 850 or 1064 nm?
No single best wavelength is established, because cytochrome c oxidase absorbs across red and much of the near-infrared. Red suits skin, near-infrared reaches deeper, and in one forearm study 1064 nm sustained its increases longer than 800 nm during the session.
| Wavelength | Rough depth | Mitochondria evidence | Kivorla devices |
|---|---|---|---|
| 630 to 670 nm red | About 2 to 3 mm | Inside the 600 to 700 nm absorption band; ATP up in skin cells at 660 nm | Mask 665 nm; panel 630, 660, 670 nm |
| 800 to 850 nm near-infrared | About 5 to 10 mm | Inside the 760 to 940 nm band; forearm signal up with lasers and an LED | Mask 850 nm; panel 810, 830, 850 nm |
| 1060 to 1064 nm near-infrared | Deeper near-infrared | Larger forearm changes than 800 nm, sustained longer during the session, in one study | Mask 1064 nm; panel 1060 nm |
The depths are rough guides; our 660 nm vs 850 nm guide explains them. In the 2022 study, 1064 nm "sustained the increases longer" than 800 nm. Pruitt, 2022 Above about 900 nm, light may also act through calcium channels. Heiskanen, 2018
In human stem cells, 660 nm red and 810 nm near-infrared raised ATP, while shorter blue and green wavelengths acted differently. Wang, 2017 For mitochondria-focused sessions, the mask's red mode is the relevant setting; see our 1064 nm face mask guide.
Do LEDs reach mitochondria, or do you need a laser?
LEDs do: in a sham-controlled study, an 810 nm LED raised the forearm enzyme signal with a similar time course to the lasers. Its change was smaller, and the lasers ran at higher irradiance.
A review of about 40 laser-versus-LED papers concluded that the effect "is not dependent on lasers or coherence". Heiskanen, 2018 It found LEDs about 100 times cheaper per mW.
Lasers warmed skin by up to about 4°C in the 2022 study, but a small thermal check in 2016 suggested heat did not explain the enzyme change. The panel's surface stays at or below 40°C.
Can too much red light work against your mitochondria?
In cells and animals, yes: past a certain dose, ATP and membrane potential fall again and a second, larger rise in reactive oxygen appears. That is why a longer session is not a stronger one.
In brain cells given 810 nm light from 0.03 to 30 J/cm², ATP rose at lower doses and fell at higher ones. Huang, 2011 Reactive oxygen had a small first peak at 3 J/cm² and a larger rise at 30 J/cm².
- Rat arthritis: 3 J/cm² helped at 5 mW/cm² but not at 50 mW/cm². Huang, 2009
- Mouse brain injury: 36 J/cm² at 50 mW/cm² helped, while 360 J/cm² at 500 mW/cm² lost the benefit. Huang, 2011
The same authors found no convincing report of this curve in patients. In people, no skin trial has found daily use did worse than two or three sessions a week. The rule from our guide to how much red light therapy is too much still applies: follow the timer, one session a day at most. Stop and let the skin settle if redness lasts for hours, or with stinging, a headache or new dark patches.
Why do some people notice little after mitochondria-focused sessions?
Healthy cells may respond less than stressed ones, and dose matters in both directions. Feeling more energetic is also different from a change in cell energy.
Light raised reactive oxygen in normal cells but lowered it in cells under oxidative stress. Hamblin, 2017 A 2026 Nature feature reports that effects may be stronger in stressed or unhealthy cells, and quotes one researcher: "We still don't fully understand what we're doing." Peeples, 2026
Cleveland Clinic notes "there's not enough evidence to support most uses" so far. Cleveland Clinic Anyone with lasting tiredness, a diagnosed mitochondrial condition or diabetes should speak to their doctor or specialist first.
What settings did the mitochondria studies use?
Human studies used single sessions of 8 to 15 minutes with the light on the skin. Cell and animal work showed benefits at low doses and losses at high ones.
| Study | Who | Light | Strength or dose | Time | Sessions | Where |
|---|---|---|---|---|---|---|
| Wang, 2016 | 11 healthy adults | 1064 nm laser | 250 mW/cm² | About 7.3 minutes | Single | Forearm, about 2 cm away |
| Pruitt, 2022 | 2 groups of 10 adults | 800, 850, 1064 nm lasers; 810 nm LED | 250 to 330 mW/cm² lasers; about 135 mW/cm² LED | 8 minutes | Single per wavelength | Forearm |
| Powner, 2024 | 30 healthy adults | 670 nm LED array | 40 mW/cm² | 15 minutes | Single | Upper back, about 16 inches away |
| Ferraresi, 2015 | Trained mice | 630 + 850 nm LEDs | 80 mW/cm², 7.2 J/cm² | Not stated | 6 training sessions | Legs and lower back |
In the back study, one 15-minute session was followed by a 27.7% smaller rise in blood glucose after a sugar drink than with placebo, in healthy adults only. Powner, 2024 Our devices are not intended for blood sugar or diabetes.
Study doses do not convert into a home session. The pattern that carries over: red and near-infrared, on bare skin, for modest times.
How do you use a Kivorla mask or panel for mitochondria-focused sessions?
Use the mask's red mode for the face and the panel's red and near-infrared channels for the body, on clean, intact skin. There is no proven mitochondria setting, so stay near the studied ranges and inside the manual's limits.
The 4-Wavelength Red Light Therapy Mask covers the face:
- Mode: the red mode combines 665 nm red with 850 and 1064 nm near-infrared, from 72, 72 and 20 VCSEL laser diodes.
- Start: 50% brightness on the 10-minute timer; step up to 70 or 90% only if your skin stays calm.
- Limits: 10-minute sessions up to 4 times a week at high brightness, or 15-minute sessions up to 3 times a week after a week of three 10-minute sessions. Never more than 20 minutes a day.
- Practical: cordless, run by voice or a handheld controller with a colour screen; about 3.5 hours to charge for around 40 minutes of use, and it cannot be used while charging.
For the neck, back or legs, the 9-Spectrum Red Light Therapy Panel carries 480, 590, 630, 660, 670, 810, 830, 850 and 1060 nm, each dimmable from 1 to 100%:
- Channels: red and near-infrared on, 480 and 590 nm turned down, continuous light.
- Time and distance: 10 to 15 minutes per area, about 6 inches for a joint and 12 inches for a broad area such as the back.
- Gentler setting: to stay near the modest strengths in the human studies, dim each wavelength, set a shorter timer (it starts at 1 minute) or step back to 12 inches.
- Spacing: at least 6 hours between 30-minute sessions.
- Position: hang it on a door or wall with the included door hook, pulley, hanging cables and screws, or stand it upright on the floor. Near the neck, place it side-on or behind you, never aimed straight at your face.
Who should not use red light, and how do you protect your eyes?
Do not use either device if you are pregnant, have epilepsy, a thyroid condition or take thyroid medication, take photosensitising medication, or have broken skin on the area. Keep your eyes closed every session, with the supplied goggles on when you use the panel.
The panel manual adds: not while breastfeeding or preparing for pregnancy; not after sunburn on the body in the past 2 weeks; not with seizures triggered by light; not by sleeping or unconscious persons; not with Fitzpatrick skin type V or VI. Anyone taking a medicine should check with their pharmacist or doctor before using the device.
Some research has looked at red light and the eye under laboratory conditions. Neither Kivorla device is ever looked into: the mask's eye opening has no diodes over the eyes, and the AAD says "Don't substitute sunglasses for protective goggles." AAD Our guide to red light therapy goggles explains why.
Standard shipping to the US, UK and Germany is free. Read the precautions before ordering. Under our return and refund policy, returns are accepted within 30 days: an unopened device with its seals intact is refunded less a 10% restocking fee, and an opened device that is undamaged and complete is refunded less 20%.
Kivorla devices are general wellness devices, not medical devices, and are not intended to treat fatigue, diabetes or any mitochondrial condition. Speak to your doctor about lasting tiredness or a diagnosed condition, and keep any treatment your specialist has advised.