The science of red light masks

Red light is everywhere – you’ll see it used by celebrities, skincare influencers, and biohackers (not sure why male wellness influencers get their own special name). What does it do, and which products actually work? Let’s dig into the science behind red light therapy, or photobiomodulation.

The video version of this post was part of a 3-way collaboration with Becky Stern who looked at how red light masks are constructed, and Ruth Amos who hacked together her own red light helmet and lounge suit.

Light and skin

Light can do stuff to skin. I know this sounds like weird pseudoscience, and frankly a lot of stuff around red light gives off kinda cult vibes – it’s hard to take it seriously when it’s promoted by people like this:

But there are lots of other examples of skin absorbing light, and that energy making something happen.

In laser hair removal, very intense pulses of light are absorbed by pigment in hair, which turns into heat and kills the hair root.

UVB from the sun is absorbed by 7-dehydrocholesterol, a bond breaks and it turns into pre-vitamin D.

UVB can also be absorbed by DNA, which reacts and can lead to skin cancer, melanocytes (pigment-producing cells) pumping out more pigment, and melanocytes multiplying.

Red light and skin

In the 1960s, lasers started being used for medical treatments, and there were questions about whether this really intense light might trigger cancer. So a scientist performed safety tests which involved zapping shaved mice with red light every day (I know, this doesn’t help with the weird cult vibes). He didn’t find any cancer, but the zapped mice regrew their hair faster.

He used red light because the first lasers were red, and this “mice with nice hair” finding led to more investigation of light treatments.

There’s now been thousands of studies published over the years on red light therapy (more generally called photobiomodulation or low-level laser therapy (LLLT)), including clinical trials. These found benefits for a whole host of things including osteoarthiritis, wound healing, pain relief, and of course, skin improvements.

The evidence really isn’t as solid for skin as for the main topical skin treatments like sunscreen and retinoids (which are also usually a lot cheaper), but I think it’s pretty clear red light can do something. I was pretty surprised – before I looked into it properly, the vibes were so, so bad!

Related post: Answering (Almost) Every Sunscreen Question

(Full disclosure: I haven’t done any sponsorships with masks or light devices, but I’ve been sent a whole bunch of free masks over the years that I haven’t managed to use consistently. I’m a bit more motivated to try them now, but I’m really bad at forming habits… we’ll see.)

Mechanism of action

The science is a bit murky for reasons we’ll get into, but scientists largely agree that red light does something called photobiomodulation. Photo means light, it’s modulating (speeding up or slowing down) biological processes and reactions that were naturally happening in your skin.

Red light seems to be absorbed by mitochrondria – yes, the powerhouse of the cell. Our bodies produce energy through this overall respiration reaction:

glucose + oxygen → water + carbon dioxide + energy

This energy powers all the chemical reactions inside our cells. There are lots of little reactions that add up to make this overall process, and a lot of them happen inside the mitochondria.

Inside the mitochrondria there’s an enzyme called cytochrome c oxidase, which chops up oxygen. To control how much oxygen chopping happens, there’s a bunch of nitric oxide sitting in the way (it’s not choppable). It’s thought that red light breaks the glue sticking the nitric oxide onto the enzyme, so more oxygen can come in and get chopped. This produces more energy and generally speeds up natural processes.

The effects in skin:

• Faster healing and repair
• Reduced inflammation
• Faster cell proliferation (including fibroblasts in the dermis – these cells produce collagen and elastin)
• Improved volume, elasticity and texture (e.g. from the increased collagen and elastin)

The released nitric oxide also widens blood vessels and increase blood flow, which could also contribute to the skin benefits.

To make things more complicated (a recurring theme), some cells that don’t have the cytochrome c oxidase enzyme still respond to red light. So there are probably other things in mitochondria that absorb light and increase respiration. There might also be completely different mechanisms that haven’t been found yet.

But what lights will actually make all of this happen? That’s where things get even messier.

Why is wavelength important?

There are two important things when it comes to light and skin.

First, wavelength: this is basically the colour of the light, and we’re using “colour” very loosely here. Things in your skin that absorb light (chromophores) will absorb some colours of light, and not others.

For example, red cellophane doesn’t absorb much red light – the red laser doesn’t get much dimmer, and most of the light goes through. But blue cellophane will make it get a lot dimmer. This means energy is absorbing into the blue cellophane and creating the potential for a reaction (the energy gets stuck in the cellophane), but there’s much less potential for the red cellophane. 

The blue laser doesn’t match as nicely, but it still clearly has the opposite effect – it’s dimmer through the red cellophane but brighter through the blue. So this time, there’s more potential for a reaction in the red cellophane, since the energy is absorbing there.

In other words, to get something to happen, you need to match the light wavelength (or colour) with the chromophores you want to target in the skin.

Red light

For skin treatments, the most studied wavelengths are red, partly because that’s what they researched first (the rats with the good hair).

As well as matching the right chromophores, red light also gets deeper into your skin, since it isn’t absorbed by as much stuff near the very top. That’s why you see red when you shine a white light through skin.

In general, it seems like red wavelengths roughly between 600 and 700 nm have similar effects on skin. The main two wavelengths you’ll see in LED skin products are 633 and 660 nm. LEDs aren’t lasers, so they put out wavelengths on either side as well, and there’s wiggle room with the central wavelength (usually +/- 10 nm).

Near infrared (NIR)

Some skin products also produce near-infrared wavelengths. These are slightly longer wavelengths than red (sort of like an invisible redder version of red), which can penetrate even deeper.

770 to 1200 nm has similar overall effects on skin as red wavelengths, but they seem to be absorbed by a different part of the cell: photoreceptors in the cell membrane, and on the membranes inside the cell.

Some studies have found that near-infrared might be more beneficial than red light, possibly because it penetrates deeper. The most popular wavelength for skin is 830 nm.

Other wavelengths

In between the beneficial red and NIR wavelengths, there’s a gap at 700-770 nm. It seems like these wavelengths don’t have the usual effects, possibly because oxygen absorbs them which could mess with respiration.

This is another complexity with photobiomodulation – some wavelengths will undo the effects of other wavelengths. Sunlight is a good example – it contains all of the wavelengths, including the beneficial red and infrared ones, but excess sunlight damages skin.

Blue light (most popular wavelength = 415 nm) is mostly used for acne. It’s absorbed by substances called porphyrins in acne bacteria. This generates free radicals which kill the bacteria, much like how benzoyl peroxide works. However, blue light can also darken brown post-acne hyperpigmentation marks and cause them to fade more slowly (the light both giveth and taketh away).

Yellow might do something anti-aging-ish, but there’s a lot less evidence. Other wavelengths such as green have even less evidence.

“Dose”

Aside from wavelength, the other big thing we care about is “dose” – how much light are we getting?

Again, it’s kind of messy and those quotation marks around “dose” are there for a reason. “How much light” isn’t a straightforward question, and there are lots of different metrics that need to be considered.

Light is kind of like balls being thrown at your skin. The official definition of “dose” is the total energy of light delivered to the skin, measured in joules (J) – in this analogy, it’s the total weight of the balls.

Wavelength (nm) is the weight of each ball, and aside from determining the targeted chromophores in skin, a shorter wavelength means less light balls in the same total dose (e.g. 10 balls of 700 nm red light and 8.6 balls of 600 nm red light are the same dose in joules). 

But even with the same wavelength, the same total dose can have different effects on skin, depending on many other factors (this list is not exhaustive). Let’s say you have 10 red light balls of 630 nm:

• Fluence (J/cm²) is how spread out the balls are over the skin
• Power (W) is how quickly the balls are shot towards the skin (1 watt = 1 joule per second)
• Irradiance (W/cm²) is probably the most useful number. It’s a mix of fluence and power – it’s essentially how bright the light is, but doesn’t tell you the total dose unless you consider treatment time
• Treatment time (sec or min) is how long the light is shooting into your skin
• Frequency of sessions is how often you perform the light treatments
• Application method is how the light is applied: touching the skin, from a distance, or moving across skin

Application method is very important, but not often discussed. If the light is pressing into your skin it mostly goes in, but if it’s further away, a lot of the light can be reflected. There’s estimates where 60% (even 90%!) of the light bounces right off.

Direct pressure on your skin also makes skin absorb light better, for example by pushing blood away which means the light can penetrate deeper. This is why the World Association for Photobiomodulation Therapy has on or off skin as one of the essential parameters you need to report in clinical trials.

This complexity is really annoying – studies usually don’t use the same parameters, so it’s difficult to compare protocols because all these factors will make a difference.

More wavelength complexity

This is a bit beyond the scope of this post, but some wavelengths within a single “colour” are more powerful than others.

This information is usually expressed as an action spectrum (you might remember these from this post about UV wavelengths from nail lamps). Different action spectra match different biological effects and there’s a lot of debate around which ones are most relevant.

Related post: Are gel nails bad for you? UV, skin cancer and allergies

Here’s one example for DNA synthesis – a higher line means a greater effect, so you need a lot less 620 nm light to get the same effects as 650 nm. Mess! This is a good source for more information on photobiomodulation action spectra.

Masks are too weak to work?

This leads us onto a common myth:

“Red light masks are useless, because the total light energy they put out is relatively low.”

I’ve been seeing this around a lot, such as in this video from a medical student (Eviba Carter) that went viral last year. 

He makes the argument that since fluence = irradiance x time, 10 minutes with a mask (average irradiance 2 mW/cm²) is the same as 6 seconds with a brighter panel (~200 mW/cm² at 6 inches) – you get the same total “dose” in terms of fluence (the same number of light balls going into your skin).

For example, his video states:

• “If I take a general average of the whole surface, I get a pathetic 2 milliwatts per centimeter squared. For perspective, to get the same amount of total red light shown to reduce wrinkles in this Omnilux panel study, it would require wearing the mask for 17.5 hours.”
• “I can get the same amount of total light with 6 seconds on my panel rig as a 10 minute mask session.”

He also summarises “papers with dose and outcome information as a public resource”, simply stating the fluence that each paper uses (in J/cm²).

There are several problems with this line of reasoning:

Compressing “dose” parameters into a single number doesn’t work

The idea that the total dose is all that matters is called the Bunsen-Roscoe law of reciprocity. It turns out that this “law” is more of a gentle suggestion – it breaks all the time with light treatments.

Again, biology is complicated. Eating three meals in the morning doesn’t give the same result as spacing them out. Your body won’t process the food, or the light, the same way.

Biphasic dose-response relationship

With light treatments, there’s a biphasic dose-response relationship. 

As we go across, we’re increasing the amount of light:

• If the light is too low, you don’t get any effects
• With more light, eventually you pass a threshold, and the more light you use, the more benefits you get
• Once you have too much light, you actually get less benefits
• At some point, you can even get negative effects!

These negative effects might be because too many free radicals are generated from the respiration reaction, or perhaps too much heat building up. Anecdotally, the main obvious issues in skin seem to be irritation and hyperpigmentation (again, the light giveth and taketh away).

This means there’s a “therapeutic window” in the middle, with the best results achieved with a medium irradiance. There are many in vitro, animal and human studies showing this relationship – biphasic dose response in photobiomodulation has been reviewed by Barolet et al., Huang et al. (and their update) and Sommer et al.. A few examples relevant to skin:

• van Breugel et al: The same dose (180 mJ) was applied to each well with different irradiances (0.55-5.98 mW) and exposure times, but cell proliferation was only markedly improved for cells treated at 1.24 mW for 154 seconds
• Lanzafame et al.: Wound closure rate was better with 8 mW than 0.7, 2 or 40 mW, with same total dose
• Gál et al.: With 5 J/cm² total fluence, 4 mW/cm² achieved significantly increased wound tensile strength but not 15 mW/cm²

To add more complexity, this biphasic relationship isn’t just for the intensity (irradiance) of the light, but also total light (energy/dose). Where you get the best results and when you start getting negative effects is also going to be different depending on many factors, including the wavelength, cells and results involved, how deep the cells are…

Application differences

There’s additional complexity as well with masks versus panels, in terms of application.

First there’s application method: even if a mask and a panel were putting out the same total dose, a lot more of the light ends up in your skin from a mask. 

• In a mask, many of the bulbs will like be touching your skin (depending on mask design and how well it fits your face), so more of the light will actually get in. None will be touching for a panel. This can result in a huge difference, since it’s estimated that 60% of the light can just reflect off skin as discussed above.
• Again, pressure will blanch the skin, removing superficial blood that might absorb – this allows the light to penetrate deeper.
• Additionally, the white inside of the mask can bounce any reflected light back towards the skin, which won’t happen with a panel.

There’s also the fact that a mask is essentially a bunch of bright spots spread over the skin, which isn’t going to have the same effect as if you averaged it out and had a continuous sheet of lower brightness – again, this is the Bunsen-Roscoe law breaking. 

There are gaps between the lights, but since we’re stimulating processes deep in the skin, the surrounding areas of skin are also affected, not just the bits directly under the LED bulbs. For example, there’s a cool experiment where two wounds were created 2 inches apart. Only one was treated with light, but the other wound healed faster too.

So it’s misleading to say masks are a scam and everyone should just use a panel to get the same dose – comparing one single number for “dose” just doesn’t make sense scientifically. It would be nice to see a proper clinical trial comparing them for different skin benefits. For now, anecdotally, it seems to be quite varied – some people see more benefits from panels while others prefer masks, some see side effects from one but not the other, plus there’s variation with different devices in the one category too.

Mask or panel?

If you want to start using a mask or panel, the most important thing seems to be finding a product you’re likely to use consistently – for all of these devices, the effects seem to be cumulative.

Advantages of panels

• Small panels tend to be cheaper
• Treatment times tend to be shorter
• More options for devices
• Easier to use on different parts of the body
• Can be uncomfortable if you don’t like staring into a bright light

Advantages of masks

• More portable and don’t need to be attached to a wall socket (I’ve found that I can cook dinner while wearing one)
• Take up less space
• Easier to fit into luggage if you travel a lot
• Can be uncomfortable if you don’t like the feeling of them on your face

Again, whether red/infrared light treatments are “worth it” for you depends on many factors – you need to consider what benefits you want out of it (skin and otherwise), what your budget is, and whether you can stick to using it. But I do think they offer benefits that standard cosmetic skincare products don’t.

(The situation with not having many studies, and studies being sponsored by brands because there’s not much public funding for skincare research is a whole different mess I’m not going into here. It’s pretty much the stuff in my retinol post but worse, and with a whole bunch of scammy stuff and weird vibes layered on top. Most studies are on in-clinic panels, but there are more studies coming out on masks, including this one on the Omnilux Men’s mask.)

Related post: Is Retinol a Scam? The Science

How to choose products

In my opinion, the best way to deal with all this complexity is to pick a product type (mask or panel) based on what you think you can benefit more from (I highly recommend reading reviews). Then for the products you’re considering, check that they have parameters that can do what you want (wavelength, irradiance, treatment schedule (time and frequency) and delivery mode (on skin or off skin)). Also consider practical aspects, like warranty and whether reviews report that they break easily.

For masks, these parameters tend to be what people agree works – it’s what the leading mask companies (Omnilux, CurrentBody) use:

Wavelengths (nm)	General skin benefits (anti-aging, reduced inflammation)	Red: 600-700 nm (most tested = 633, 660 nm) Infrared: 770-1200 nm (most tested = 830 nm)
	Acne	Blue: 400-500 nm (most tested = 415 nm)
Irradiance (mW/cm²)	Red + infrared	30-35 mW/cm²
	Blue	30-44 mW/cm²
Treatment time	10 min, 3-5 times a week
Number of LEDs	60-66 combined dual LEDs	These are sometimes listed as double the number e.g. 120 or 132

I’d recommend getting something close to these parameters – an exact match is probably not necessary, but there’s less certainty around their effects and effectiveness. If a mask has slightly more LEDs but each with a slightly lower irradiance, for example, it’ll probably also be effective.

Higher energy, more LEDs, more time, more sessions etc. can lead to greater benefits but also greater side effects, likely depending on individual skin tolerance.

If you’re looking for a mask to buy, I highly recommend this amazingly comprehensive review post from Vanessa of Goals to Get Glowing.

References

Stausholm MB, Naterstad IF, Joensen J, et al. Efficacy of low-level laser therapy on pain and disability in knee osteoarthritis: Systematic review and meta-analysis of randomised placebo-controlled trials. BMJ Open. 2019;9(10):e031142. doi:10.1136/bmjopen-2019-031142

Chow RT, Johnson MI, Lopes-Martins RAB, Bjordal JM. Efficacy of low-level laser therapy in the management of neck pain: A systematic review and meta-analysis of randomised placebo or active-treatment controlled trials. Lancet. 2009;374(9705):1897-1908. doi:10.1016/S0140-6736(09)61522-1

Jagdeo J, Austin E, Mamalis A, Wong C, Ho D, Siegel DM. Light‐emitting diodes in dermatology: A systematic review of randomized controlled trials. Lasers Surg Med. 2018;50(6):613-628. doi:10.1002/lsm.22791

Ngoc LTN, Moon JY, Lee YC. Utilization of light-emitting diodes for skin therapy: Systematic review and meta-analysis. Photodermatol Photoimmunol Photomed. 2023;39(4):303-317. doi:10.1111/phpp.12841

Cheng K, Martin LF, Slepian MJ, Patwardhan AM, Ibrahim MM. Mechanisms and pathways of pain photobiomodulation: A narrative review. J Pain. 2021;22(7):763-777. doi:10.1016/j.jpain.2021.02.005

Yadav A, Gupta A. Noninvasive red and near-infrared wavelength-induced photobiomodulation: Promoting impaired cutaneous wound healing. Photodermatol Photoimmunol Photomed. 2017;33(1):4-13. doi:10.1111/phpp.12282

GembaRed. Skin Contact Method for Clinical Grade Red Light Therapy (vide0). February 22, 2024. 

GembaRed. Proper Red Light Therapy Dosing – Contact Method and Skin Reflectance. October 17, 2019.

Karu TI, Kolyakov SF. Exact action spectra for cellular responses relevant to phototherapy. Photomed Laser Surg. 2005;23(4):355-361. doi:10.1089/pho.2005.23.355

Karu T. Action Spectra: Their Importance for Low Level Light Therapy. October 14, 2008.

Barolet D, Christiaens F, Hamblin MR. Infrared and skin: Friend or foe. J Photochem Photobiol B. 2016;155:78-85. doi:10.1016/j.jphotobiol.2015.12.014

Huang YY, Chen ACH, Carroll JD, Hamblin MR. Biphasic dose response in low level light therapy. Dose Response. 2009;7(4):358-383. doi:10.2203/dose-response.09-027.Hamblin

Huang YY, Sharma SK, Carroll J, Hamblin MR. Biphasic dose response in low level light therapy – an update. Dose Response. 2011;9(4):602-618. doi:10.2203/dose-response.11-009.Hamblin

Sommer AP, Pinheiro AL, Mester AR, Franke RP, Whelan HT. Biostimulatory windows in low-intensity laser activation: lasers, scanners, and NASA’s light-emitting diode array system. J Clin Laser Med Surg. 2001;19(1):29-33. doi:10.1089/104454701750066910

van Breugel HH, Bär PR. Power density and exposure time of He-Ne laser irradiation are more important than total energy dose in photo-biomodulation of human fibroblasts in vitro. Lasers Surg Med. 1992;12(5):528-537. doi:10.1002/lsm.1900120512

Lanzafame RJ, Stadler I, Kurtz AF, et al. Reciprocity of exposure time and irradiance on energy density during photoradiation on wound healing in a murine pressure ulcer model. Lasers Surg Med. 2007;39(6):534-542. doi:10.1002/lsm.20519

Gál P, Mokrý M, Vidinský B, et al. Effect of equal daily doses achieved by different power densities of low-level laser therapy at 635 nm on open skin wound healing in normal and corticosteroid-treated rats. Lasers Med Sci. 2009;24(4):539-547. doi:10.1007/s10103-008-0604-9

Hopkins JT, McLoda TA, Seegmiller JG, David Baxter G. Low-level laser therapy facilitates superficial wound healing in humans: A triple-blind, sham-controlled study. J Athl Train. 2004;39(3):223-229.

Kim DS, Song KU, Lee HK, et al. Synergistic effects of using novel home-use 660- and 850-nm light-emitting diode mask in combination with hyaluronic acid ampoule on photoaged Asian skin: A prospective, controlled study. J Cosmet Dermatol. 2020;19(10):2606-2615. doi:10.1111/jocd.13599

Kim MS, An J, Lee JH, et al. Clinical validation of face‐fit surface‐lighting micro light‐emitting diode mask for skin anti‐aging treatment. Adv Mater. 2024;36(50):2411651. doi:10.1002/adma.202411651

Couturaud V, Le Fur M, Pelletier M, Granotier F. Reverse skin aging signs by red light photobiomodulation. Skin Res Technol. 2023;29(7):e13391. doi:10.1111/srt.13391

Hong JH, Kim DY, Kye YC, Seo SH, Kim DH, Ahn HH. Lightening effect of a light-emitting diode mask on facial skin: A colorimetric assessment. Ann Dermatol. 2019;31(2):247-248. doi:10.5021/ad.2019.31.2.247

Nestor MS, Swenson N, Macri A, Manway M, Paparone P. Efficacy and Tolerability of a Combined 445nm and 630nm Over-the-counter Light Therapy Mask with and without Topical Salicylic Acid versus Topical Benzoyl Peroxide for the Treatment of Mild-to-moderate Acne Vulgaris. J Clin Aesthet Dermatol. 2016;9(3):25-35.

Mineroff J, Austin E, Feit E, et al. Male facial rejuvenation using a combination 633, 830, and 1072 nm LED face mask. Arch Dermatol Res. 2023;315(9):2605-2611. doi:10.1007/s00403-023-02663-w

Goals to Get Glowing. LED Masks: Deep Dive into Low Level Light Therapy & Comparison of 55+ LED Devices. January 17, 2021. Last updated November 5, 2024.

Other resources I found helpful:

Carroll J. PBM & The Brain: Why I do not believe any of you. Presented at: MGH Brain PBM Clinic Rounds. May 3, 2024. 

Hamblin MR. Mechanisms and applications of the anti-inflammatory effects of photobiomodulation. AIMS Biophys. 2017;4(3):337-361. doi:10.3934/biophy.2017.3.337

de Freitas LF, Hamblin MR. Proposed mechanisms of photobiomodulation or low-level light therapy. IEEE J Sel Top Quantum Electron. 2016;22(3):7000417. doi:10.1109/JSTQE.2016.2561201