Shift from 630nm to 660nm and tissue response changes — here's what light absorption science says about why

Last updated: August 13, 2026 | 13-minute read

Red light therapy and photobiomodulation (PBM) are often summarized with a simple rule: visible red light is used for more superficial targets, while near-infrared light generally retains more relative intensity at greater tissue depths. That rule is useful as a starting point, but it is not a complete explanation of how light behaves in living tissue.

Wavelength matters because tissue absorption and scattering are wavelength-dependent. However, wavelength alone does not determine treatment depth, cellular response, safety, or clinical benefit. The result also depends on tissue composition, pigmentation, blood content, beam geometry, contact, irradiance, exposure time, pulsing, illuminated area, and the biological context in which the light is used.

This guide explains how tissue absorbers differ from proposed cellular photoacceptors, why fixed penetration-depth claims are unreliable, how to evaluate an LED spectrum, and why optical testing must be kept separate from photobiological safety, quality-system controls, and regulatory status.

Core principle: An emission wavelength can support a plausible optical and biological rationale, but wavelength alone does not prove target engagement or clinical efficacy.

What the light absorption spectrum actually shows

Light absorption spectra in red light therapy

An absorption spectrum describes how strongly a material absorbs light at different wavelengths. In tissue, no single curve determines the outcome. Light encounters multiple absorbers as well as structures that reflect and scatter photons.

The most important distinction is between major tissue absorbers, which largely influence where optical energy travels and is deposited, and proposed cellular photoacceptors, which may participate in downstream photobiological signaling.

Many PBM studies use wavelengths from approximately 600 to 1100 nm, but this should not be presented as a hard border between effective and ineffective light. The term tissue optical window describes a region in which the combined effects of absorption and scattering can permit greater relative transmission than at many shorter or longer wavelengths. Its exact boundaries vary with the tissue, the optical model, and the definition of penetration being used.

Below this broad region, absorption by melanin and blood and stronger scattering generally reduce transmission. At longer wavelengths, water and lipid absorption bands become increasingly important. These trends are gradual and structured, not two vertical "absorption walls." Wavelengths just outside a simplified range do not automatically become biologically inactive.

Absorption, scattering, reflection, and transmission all matter

Wavelength-dependent absorption, reflection, scattering, and transmission

When light reaches skin, several processes occur simultaneously:

• A fraction is reflected at the surface.
• Some photons are scattered and change direction within tissue.
• Some are absorbed by melanin, blood, water, lipids, proteins, or other constituents.
• A decreasing fraction may continue to greater depths.
• Absorbed energy may be converted into heat or participate in photochemical and photobiological processes, depending on the absorber, dose, and biological context.

This is why a simple power reading cannot fully describe tissue exposure. Two devices can show the same irradiance at the treatment plane while differing in spectrum, beam angle, illuminated area, spatial uniformity, pulse waveform, or thermal stability.

The Beer–Lambert relationship is often written as:

[
A = \varepsilon c l
]

or, in an exponential attenuation form:

[
I(z) = I_0 e^{-\mu z}
]

These equations are useful simplified descriptions of absorption in suitable media. Living tissue, however, is a heterogeneous and strongly scattering medium. A realistic estimate of light distribution may require a modified Beer–Lambert approach, diffusion theory, radiative-transfer modeling, Monte Carlo simulation, or direct measurements made under conditions that resemble intended use.

Therefore, a higher molecular absorptivity at one wavelength does not automatically mean that more useful energy reaches a cellular target. Competing absorption, scattering, tissue concentration, optical path length, and incident photon distribution must also be considered.

Cytochrome c oxidase: a proposed photoacceptor, not a complete mechanism

Cytochrome c oxidase (CCO) is complex IV of the mitochondrial electron transport chain. It contains heme and copper centers with wavelength-dependent spectral features. Research has frequently proposed CCO as an important red and near-infrared photoacceptor in PBM.

That proposal should not be rewritten as a settled sequence in which every absorbed photon activates CCO, increases ATP, and produces a therapeutic response. Several limitations matter:

1. CCO spectra depend on oxidation state, ligands, species, preparation, and measurement method.
2. An absorption spectrum is not the same as an action spectrum showing biological response versus wavelength.
3. Other proposed mechanisms include nitric-oxide-related signaling, transient changes in reactive oxygen species and calcium, light-sensitive ion channels, and possible effects involving interfacial or mitochondrial water.
4. Cellular responses can be biphasic: a dose that produces one effect under one condition may produce a smaller, absent, or different effect at another dose.
5. In vitro target engagement does not automatically establish a clinical benefit in humans.

A scientifically defensible description is that red and NIR photons may interact with cellular photoacceptors, including CCO, and may influence mitochondrial and signaling pathways under some conditions. The magnitude and direction of the biological response depend on wavelength, delivered dose, cell state, tissue, protocol, and outcome being measured.

Why 630, 660, 810, 830, 850, and 1060–1064 nm are all used

Qualitative wavelength comparison for red and near-infrared light attenuation in tissue

Commercial and research wavelengths are shaped by several factors at once: tissue optics, proposed action spectra, historical research protocols, LED and laser availability, optical efficiency, heat management, and manufacturing cost. It is therefore inaccurate to say that molecular physics alone selected one exact set of wavelengths.

The correct conclusion is qualitative: under comparable conditions, NIR wavelengths often retain more relative intensity at depth than visible red wavelengths. The exact amount reaching a target cannot be inferred from wavelength alone.

Why fixed penetration-depth claims are unreliable

Claims such as "660 nm penetrates 3 mm" or "850 nm penetrates 5 cm" appear precise but usually omit the definition of penetration.

Depending on the source, "penetration depth" may mean:

• the depth at which intensity falls to (1/e) of a reference value;
• the depth at which 10%, 1%, or another fraction remains;
• the deepest location where a detector can measure light;
• the greatest depth visited by a small fraction of scattered photons;
• or a proposed biological treatment depth.

These definitions are not interchangeable. Experimental work in ex vivo tissue has reported optical penetration depths on the order of millimeters for several red and NIR wavelengths, while other studies may detect a small fraction of incident light at greater depths. Detectable photons at a location do not automatically mean that the dose is sufficient to produce a biological or clinical effect there.

Actual transmission depends on:

• tissue type and thickness;
• melanin concentration and skin pigmentation;
• blood volume and oxygenation;
• fat, water, and connective-tissue content;
• wavelength and spectral bandwidth;
• beam diameter, angle, divergence, and spatial profile;
• contact, compression, and coupling conditions;
• incident irradiance and exposure duration;
• and the depth metric used by the study.

For product evaluation, qualitative attenuation diagrams and measured multi-distance data are usually more honest than fixed anatomical boundaries.

Spectral accuracy: what should actually be measured

Quality-control technician measuring the stabilized spectrum of a generic LED therapy panel

Nominal wavelength labels are insufficient for technical evaluation. An LED specified as "660 nm" emits across a band rather than at one infinitely narrow wavelength. Its spectrum can also vary with LED bin, drive current, junction temperature, optics, and operating time.

A useful model-specific optical report should state:

• measured peak wavelength for every enabled channel;
• wavelength tolerance;
• full width at half maximum (FWHM);
• spectral irradiance or band-integrated irradiance;
• whether channels were measured separately and together;
• drive mode, pulse frequency, pulse width, duty cycle, and off-state output where relevant;
• treatment distance and sensor position;
• stabilized operating temperature and warm-up time;
• spatial measurement grid and uniformity;
• instrument model, calibration date, and measurement uncertainty;
• and the exact model and configuration tested.

Thermal drift is real, but its significance should be measured rather than exaggerated. A 10–15 nm difference does not automatically move a broadband LED completely outside a biologically relevant band. The correct question is whether the stabilized spectrum stays within the documented design tolerance and whether that tolerance is appropriate for the intended protocol.

Independent third-party measurement can improve confidence, but "third party" is not a substitute for a complete test method. A calibrated in-house laboratory can also generate useful engineering data when the procedure, instrument, conditions, and uncertainty are disclosed.

Irradiance is not the same as dose

Irradiance describes optical power per unit area, commonly reported in mW/cm². Radiant exposure describes energy per unit area, commonly reported in J/cm².

For continuous output, or for a valid time-averaged irradiance measurement:

[
\text{Radiant exposure (J/cm²)} = \frac{\text{average irradiance (mW/cm²)} \times \text{time (s)}}{1000}
]

For an ideal rectangular pulse with zero output during the off period:

[
\text{average irradiance} \approx \text{peak irradiance} \times \text{duty cycle}
]

The duty cycle must not be applied twice if the meter or report already provides time-averaged irradiance.

A dose calculation at the treatment plane still does not equal dose at a deep tissue target. Tissue attenuation and geometry must be considered. Conversely, a higher surface irradiance is not automatically better because PBM responses can be biphasic and application-specific.

A practical framework for selecting and evaluating wavelengths

Decision matrix for evaluating red light therapy wavelengths and complete device protocols

Instead of asking for one universally "best" wavelength, evaluate the complete protocol.

1. Define the intended use and target

Specify the anatomical site, target depth, target population, desired outcome, and whether the use is general wellness, research, cosmetic, or medical. Avoid assigning a medical purpose based only on the presence of red or NIR LEDs.

2. Compare the complete emitted spectrum

Review measured peak wavelength, tolerance, FWHM, channel-specific output, stabilized temperature, and model-specific test conditions. Do not infer the delivered spectrum from LED part numbers alone.

3. Evaluate treatment-plane output

Request irradiance at the stated treatment distance, an illuminated-area definition, and a spatial grid showing average, minimum, maximum, and uniformity. A center-point peak is not representative of a full panel or large treatment area.

4. Calculate the actual protocol dose

Record exposure time, average irradiance, pulse parameters, radiant exposure, frequency of use, and whether multiple wavelength channels operate simultaneously or sequentially.

5. Match claims to evidence

Compare the proposed device protocol with the wavelength, dose, placement, population, comparator, and outcome used in the cited study. Evidence for one device or protocol should not be transferred automatically to another.

6. Separate four different questions

• Optical performance: What light does the device actually emit at the treatment plane?
• Photobiological safety: What eye and skin hazards were assessed under the intended conditions?
• Quality-system control: Are design, production, inspection, complaints, and changes documented and audited?
• Regulatory status: What claims and intended uses are permitted in the specific market for the specific model?

Passing one checkpoint does not prove the others. For example, an electrical-safety mark does not validate wavelength accuracy, and regulatory registration does not prove ATP elevation or clinical benefit.

FDA status depends on intended use and classification—not wavelength alone

In the United States, a red or near-infrared light-emitting product may be treated as a general-wellness product, a medical device that requires a 510(k), or a medical device that is exempt from 510(k) premarket notification. The route depends primarily on the intended use, claims, technology, user population, and applicable FDA classification—not simply on whether the device emits 660 or 850 nm light.

One relevant example is the FDA classification for an infrared therapeutic heating lamp. Under 21 CFR 890.5500, an infrared lamp is described as a medical-purpose device that emits infrared energy, approximately 700 to 50,000 nm, to provide topical heating. When the product fits the infrared therapeutic heating-lamp category, it is classified as Class II and may be exempt from 510(k) premarket notification, subject to the limitations in 21 CFR 890.9. FDA product code ILY is currently listed as "510(k) Exempt."

FDA's current ILY classification entry separately states "GMP Exempt? No." In other words, the listed exemption concerns 510(k) premarket notification; it is not a blanket exemption from applicable manufacturing or quality-system controls.

This exemption must be described carefully:

• It is a 510(k) premarket-notification exemption, not an exemption from all FDA requirements.
• It does not mean that FDA has approved, cleared, certified, or endorsed the product.
• Establishment registration, device listing, labeling, quality-system requirements, medical-device reporting, corrections and removals, and other controls may still apply, depending on the manufacturer and device.
• The exemption applies only when the device remains within the generic type and limitations of the classification.
• A different intended use, a different fundamental technology, lay use where the generic type was limited to professional use, or claims extending beyond topical heating can change the regulatory pathway and may require a 510(k) or another submission.
• A visible-red-only lamp or a device promoted primarily for nonthermal PBM effects should not automatically be placed under the infrared therapeutic heating-lamp classification.

Some low-risk light-emitting products intended only for general wellness may fall outside the scope of FDA's PBM medical-device guidance, but that is a separate regulatory analysis—not a 510(k) exemption. Manufacturers should confirm the product code, classification, intended use, exemption limitations, and current database status for each model before making an FDA claim.

The most accurate public wording is therefore:

"Certain infrared therapeutic heating lamps may be exempt from 510(k) premarket notification under their applicable FDA classification. Exemption status is device- and intended-use-specific and does not denote FDA approval or exemption from all regulatory controls."

What safety and quality documents can—and cannot—support

Different documents answer different questions:

Key takeaways

The light absorption spectrum is one part of a larger light–tissue interaction problem. Major tissue absorbers such as melanin, hemoglobin, water, and lipids influence reflection, absorption, scattering, and transmission. Cytochrome c oxidase is a widely proposed cellular photoacceptor, but CCO absorption should not be treated as automatic proof of ATP elevation or clinical benefit.

Visible red wavelengths are generally used for more superficial exposure, while near-infrared wavelengths often retain more relative intensity at depth under comparable conditions. Exact penetration depths cannot be assigned from wavelength alone, and detectable light at depth does not establish a therapeutic dose.

A technically credible device evaluation should review the measured spectrum, FWHM, tolerance, stabilized temperature, spatial irradiance, pulse waveform, radiant exposure, treatment geometry, safety assessment, quality controls, and market-specific regulatory status. No single wavelength, power number, certificate, registration, or exemption can replace the complete protocol.

FAQ

What are the best wavelengths for red light therapy?

There is no universal best wavelength. Common research and commercial choices include 630, 633, 660, 810, 830, 850, and 1060–1064 nm. Visible red is generally used when more superficial exposure is desired, while NIR often retains more relative intensity at depth. The appropriate choice depends on the target, dose, geometry, evidence, safety assessment, and intended use.

Does 850 nm penetrate several centimeters into tissue?

Some photons may be detected at centimeter-scale depths under particular conditions, but that is not the same as an optical penetration depth or a therapeutically effective dose. Published values vary because tissue, beam size, measurement method, and the definition of penetration vary. Avoid assigning a fixed depth to 850 nm without defining the metric and conditions.

Is cytochrome c oxidase proven to be the only target of red and NIR light?

No. CCO is a widely proposed and extensively discussed photoacceptor, but the complete PBM mechanism remains under investigation. Other mitochondrial, signaling, ion-channel, nitric-oxide-related, and water-associated mechanisms have also been proposed.

Why do medical recommendations for PBM vary?

Recommendations are indication-specific. Some uses, such as selected protocols for preventing oral mucositis in certain cancer-treatment settings, appear in clinical guidelines. For many other uses, studies remain heterogeneous in wavelength, dose, placement, device design, population, and outcome. Evidence for one indication should not be generalized to every red light therapy claim.

Can red light therapy regrow gums?

Regrowth of lost gingival tissue has not been established as a general clinical outcome of red light therapy. PBM has been studied as an adjunct in periodontal and oral-care contexts, but it should not replace diagnosis and treatment by a dental professional.

What irradiance is needed?

There is no universal irradiance requirement. Published protocols use a wide range, and the relevant value depends on wavelength, target, illuminated area, time, pulse parameters, and clinical context. Always report the distance, measurement method, average rather than only peak irradiance, and calculated radiant exposure.

Are some red light therapy lamps FDA-exempt?

Some devices may be exempt from 510(k) premarket notification, but the exemption is classification- and intended-use-specific. For example, an infrared therapeutic heating lamp that fits 21 CFR 890.5500 and its limitations may fall under FDA product code ILY, which is listed as 510(k)-exempt. This does not mean "FDA approved," and it does not remove other applicable FDA obligations. Red-light, cosmetic, PBM, or disease-treatment claims outside that classification require a separate regulatory assessment.

References

1. Jacques SL. Optical properties of biological tissues: a review. Physics in Medicine & Biology. 2013;58:R37–R61. https://doi.org/10.1088/0031-9155/58/11/R37
2. Karu TI. Primary and secondary mechanisms of action of visible to near-IR radiation on cells. Journal of Photochemistry and Photobiology B. 1999;49:1–17. https://pubmed.ncbi.nlm.nih.gov/10365442/
3. Mason MG, Nicholls P, Cooper CE. Re-evaluation of the near infrared spectra of mitochondrial cytochrome c oxidase. Biochimica et Biophysica Acta. 2014;1837:1882–1891. https://doi.org/10.1016/j.bbabio.2014.08.005
4. de Freitas LF, Hamblin MR. Proposed mechanisms of photobiomodulation or low-level light therapy. IEEE Journal of Selected Topics in Quantum Electronics. 2016;22:7000417. https://doi.org/10.1109/JSTQE.2016.2561201
5. Stolik S, Delgado JA, Pérez A, Anasagasti L. Measurement of the penetration depths of red and near infrared light in human ex vivo tissues. Journal of Photochemistry and Photobiology B. 2000;57:90–93. https://doi.org/10.1016/S1011-1344(00)00082-8
6. Ash C, Dubec M, Donne K, Bashford T. Effect of wavelength and beam width on penetration in light–tissue interaction using computational methods. Lasers in Medical Science. 2017;32:1909–1918. https://doi.org/10.1007/s10103-017-2317-4
7. International Electrotechnical Commission. IEC 62471:2006—Photobiological safety of lamps and lamp systems. https://webstore.iec.ch/en/publication/7076
8. U.S. Food and Drug Administration. Photobiomodulation (PBM) Devices—Premarket Notification [510(k)] Submissions. Draft Guidance, January 2023; not for implementation. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/photobiomodulation-pbm-devices-premarket-notification-510k-submissions
9. U.S. Food and Drug Administration. Product Classification: ILY—Lamp, Infrared, Therapeutic Heating. https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfPCD/classification.cfm?ID=ILY
10. Electronic Code of Federal Regulations. 21 CFR 890.5500—Infrared lamp. https://www.ecfr.gov/current/title-21/chapter-I/subchapter-H/part-890/subpart-F/section-890.5500
11. Electronic Code of Federal Regulations. 21 CFR 890.9—Limitations of exemptions from section 510(k). https://www.ecfr.gov/current/title-21/chapter-I/subchapter-H/part-890/subpart-A/section-890.9
12. U.S. Food and Drug Administration. Class I and Class II Device Exemptions. https://www.fda.gov/medical-devices/classify-your-medical-device/class-i-and-class-ii-device-exemptions
13. U.S. Food and Drug Administration. Important Reminders about Registration and Listing. https://www.fda.gov/medical-devices/device-registration-and-listing/important-reminders-about-registration-and-listing
14. MASCC/ISOO. Clinical practice guidelines for the management of mucositis. https://mascc.org/resources/mascc-guidelines/