Hand-held UV lights sold online to kill microbes are effective only in specific conditions and can be unsafe when used improperly

A video promoting the supposed germicidal properties of a hand-held UV lamp recently circulated on social media. It claims that the lamp will “kill 99.99% of all the bacteria, viruses, and mold […] in 10 seconds and that it “delivers 2,350 µW/cm²”, which is strong enough to kill “99.99% of viruses and bacteria, but is 100% safe for humans and pets”. The germicidal properties of ultraviolet (UV) light are widely accepted among the scientific community^[1,2]. However, the claim that the particular product shown in the video is efficient and safe is inaccurate and unsupported by available evidence.

UV light is electromagnetic radiation in a range of wavelengths lower than the visible light spectrum. UV light is broadly divided into 3 categories: UV-A, ranging from 315 nanometers (nm) to 399 nm, UV-B (280 nm – 314 nm) and UV-C (100 nm – 279 nm). While UV-A and UV-B emitted by the sun reach the surface of the Earth, UV-C light is completely absorbed by the ozone layer in the atmosphere. Therefore, UV-C is typically only encountered in artificial light sources such as decontamination systems.

Indeed, UV light can inactivate microbes such as bacteria and viruses by damaging their DNA or RNA^[1]. And research has suggested that hand-held devices similar to the one shown in the video can function under carefully calibrated experimental conditions^[3,4] unlikely to be encountered during daily use in a household setting.

The germicidal potential of UV light peaks within the UV-C range and most germicidal UV devices use wavelengths around 254 nm^[2]. The effectiveness of a UV source depends on its fluence: the amount of energy received by the exposed surface, expressed in J/m². The fluence threshold that must be reached in order to inactive a pathogen varies by species or strain. Fluence itself is a product of irradiance (in W/m²) and time of exposure (in seconds). Therefore, a lamp with a weaker irradiance would require more time to decontaminate a surface.

Neither the video nor the manufacturer’s website provide any experimental data to support their claims regarding germicidal efficiency or safety of the device. Although scientists have collected abundant evidence demonstrating the germicidal properties of UV light, no data is presented in the video to support the claims regarding the particular device. In 2015, the U.S. Federal Trade Commission charged companies that were selling other UV disinfectant devices with “engag[ing] in false and unsubstantiated advertising” by claiming that their devices “ha[d] been proven to eliminate 99.99% of targeted germs and viruses in as little as 10 seconds” without providing scientific evidence. Without such evidence, such claims remain unsupported.

Here, we review the two main claims in the video regarding the germicidal efficiency and safety of the lamp used as advertised in a home environment.

1. Germicidal efficiency

When asked to review this claim, scientists acknowledged that the nominal irradiance of the device shown in the video could certainly kill many bacteria and certain viruses, but only under ideal conditions (read their comments below). However, they found the statement that the device could “kill 99.99% of all the bacteria, viruses, and mold […] in 10 seconds” to be inaccurate. William Anderson, a professor of chemical engineering at the University of Waterloo in Ontario, says that the UV fluence of the device could theoretically be “sufficient to kill 99.99% of many bacteria and some (but not all) viruses”.

Fariborz Taghipour, a professor of chemical and biological engineering at the University of British Columbia in Vancouver, explained that “Each microorganism has a particular sensitivity to UV, so you can never claim it kills everything. This [fluence] is relatively good (you can likely inactivate 99.9% of some common pathogens, including SARS-CoV-2), but the claim of 99.99% for everything is false.” In fact, the standards for UV decontamination of water require fluences much higher than what the lamp in the video can deliver in 10 seconds, which is about 23 mJ/cm². Taghipour says, for example, that New York City water treatment plants require 40 mJ/cm².

The reviewers also state that the claim of high germicidal efficiency is misleading because it relies on ideal use conditions. Potential caveats are numerous but neither the promotional video nor the manufacturer’s website convey these. As mentioned above, that germicidal efficiency is determined by the fluence received by the surface that is being disinfected. The claims suggest that 99.99% decontamination can be reached in 10 seconds. According to the specifications of the lamp, a 10 second exposure would deliver a fluence of 23,500µJ/m², which is indeed enough to kill certain microorganisms.

However, this would only work if the lamp were held stationary for 10 seconds at every spot to be decontaminated. This is contrary to what the video illustrates, says Anderson: “In the video, the device is waved around for 10 seconds over a larger surface. In this case, no part of the surface will receive a sufficient dose to achieve 99.99% disinfection. To adequately disinfect a cell phone, for example, one would need to move the device very slowly from one end to the other such that all parts of the surface are exposed for 10 seconds. Just based on a rough estimate of dimensions, this could take over 1 minute of slow and steady motion“. This omission of the conditions under which the lamp would have to be used in order to be effective is therefore misleading. Anderson adds: ”Users may think they have done something by waving it around, but physics tells us that the effect will be minimal, resulting in a false sense of security.”

Another caveat is distance of use. Even though the irradiance at the stated output is 2,350µW/cm²), the final fluence that reaches a surface strongly depends on the distance between the lamp and the object being decontaminated. Anderson explains that irradiance decreases by the square of the distance. If, for example, the UV light has an irradiance of 2,350µW/cm² on the surface of an object held 5 cm away, this would be reduced to 588µW/cm² if the distance were doubled to 10 cm. “To achieve the same [claimed] 99.99% germicidal dose would now require 40 seconds of exposure on the single one spot,” says Anderson.

Finally, as with visible light, UV light can be blocked from reaching a surface.Surfaces that are not smooth and instead contain corners and crevices, such as keyboards or clothes, would be much more difficult to decontaminate. The video shows the device being used on surgical masks, but the folds of tissue would likely block at least some of the UV light emitted by the lamp, preventing complete decontamination.

2. Safety

The video further states that the device is “100% safe for humans and pets”. However, our reviewers stated unanimously that UV-C exposure is not safe for humans. The depth of penetration of UV-C light through the skin and into the body is shorter than that of UV-A or UV-B light^[5], but UV-C can still damage the eyes and skin. In answer to the question of whether the hand-held device is safe for domestic use, Taghipour says: “It depends on the exposure time (like microbes), but I would say not a chance. […] If the exposure time is just one second per day, it is safe…but not more”.

Anderson concurs with this statement, explaining that: “UV-C light is in fact mutagenic (causes DNA and RNA mutations and damage) […] In addition to radiation burns, like sunburn, it can initiate cancer.”

Ronald Hofmann, a professor of engineering at the University of Toronto and President of the International Ultraviolet Association, further notes that it would be challenging in terms of both engineering and calibration to maintain the device’s germicidal potential while keeping users safe from exposure: “while there might be a “sweet spot” where there is enough UV-C to kill the organisms without harming a person, engineering to that sweet spot on a consumer device will be difficult”.

Of note, shorter UV-C wavelengths, sometimes called far-UV-C (200 -220 nm), are thought to potentially be safe for humans while also maintaining germicidal activity^[6,7]. The manufacturer of the lamp claims that it uses far-UV-C, but contradicts this statement by listing the wavelength of the emitted light as 253 nm, which is regular UV-C. Furthermore, far UV-C for disinfection is still being developed: “There is evidence in the lab that it might be safe for human skin and eyes and yet still disinfect, but this is very preliminary and I could not personally recommend bringing such devices out into the public without more testing for possible unintended consequences,” warns Hofmann.

In summary, the video does not present proof that the device it advertises achieves the claimed germicidal efficiency. Even though the technical specifications of the lamp (wavelength and irradiance) would theoretically indicate efficient germicidal activity, it is unlikely that home users will use it under the exact conditions needed to achieve full efficiency (i.e., duration and distance of exposure, unobstructed smooth surfaces). In addition, even under ideal conditions, the bold claim that it will “kill 99.99% of all the bacteria, viruses, and mold […] in 10 seconds” is inaccurate due to the diverse UV resistance of different pathogens. The claim that it is 100% safe to humans is also inaccurate. Our reviewers unanimously warned of the mutagenic effects of UV-C light as well as the risks of skin burns and corneal damage.

Altogether, the many caveats and risks associated with using UV-C light as a decontaminant make it much better suited for use in industrial and research settings. “UV-C is an excellent disinfectant and is currently used for many applications. However, such applications are usually under very controlled conditions with well-engineered devices. Releasing UV-C into general consumer products will require very careful engineering and standard-setting“, says Hoffmann. Anderson agrees, saying: “In general, for practical purposes, these hand-held devices are largely useless and potentially dangerous […] using a chemical disinfectant wipe is much faster and likely more effective”.

The video makes a number of statements that could be true, but it depends on the circumstances. The demonstration of the UV-C device in the video is unlikely to be effective, however, for reasons outlined below (as well as some other general comments).

1. The UV spectrum is generally broken up into 3 parts, UV-A, UV-B, and UV-C.
2. At the surface of the earth, sunlight only contains UV-A and UV-B. The UV-C is filtered out by ozone in the stratosphere. The video seems to imply some sort of equivalence between sunlight and the UV-C device, which is not true. UV-C is not “natural” in the sense that humans, plants, and animals are never exposed to it by sunlight, only by artificial sources.
3. UV-C light is not 100% safe for people. The ICNIRP (International Commission on Non-Ionizing Radiation Protection) has set an exposure limit of 6 mJ/cm² over an 8-hour day for occupational safety. A device that emits 2,350 µW/cm² will exceed this exposure limit in about 2.5 seconds if placed near the unprotected skin or eyes.
4. UV-C light is in fact mutagenic (causes DNA and RNA mutations and damage), which is how it kills bacteria and viruses. In addition to radiation burns, like sunburn, it can initiate cancer.
5. The stated output of 2,350 µW/cm² (technically called “irradiance”) has no direct relationship with being 99.99% effective in killing bacteria and viruses. What matters for germicidal efficiency is the dose [fluence], which is the product of irradiance times exposure time.
6. If this device was used for 10 seconds (stationary on one spot), held very close to the target surface, the dose would be 23.5 mJ/cm² (2,350 times 10 divided by 1,000 to convert from µJ to mJ). According to data compiled by the IUVA (International UV Association), this dose is sufficient to kill 99.99% of many bacteria and some (but not all) viruses. Therefore, this claim in the video is somewhat true. However, it requires that the device be held very close to the surface and in one stationary spot for the full 10 seconds.
7. The stated irradiance of 2,350 µW/cm² depends strongly on the distance from the device to the surface, through what is often called the “inverse square law”. It is not clear in the video at what distance this irradiance is valid, but it is likely quite close to the lamp. For illustration, let’s assume that it is at a distance of 1 cm. If the user happens to hold the device 2 cm away from the target surface, the irradiance will then drop to about 588 µW/cm². To achieve the same 99.99% germicidal dose would now require 40 seconds of exposure on the single one spot. Therefore, the germicidal effectiveness is very sensitive to how the device is held and moved.
8. Likewise, if the surface being treated is not flat (like a keyboard), the deeper parts will not be very well irradiated because of the distance increase.
9. In the video, the device is waved around for 10 seconds over a larger surface. In this case, no part of the surface will receive a sufficient dose to achieve 99.99% disinfection. To adequately disinfect a cell phone for example, one would need to move the device very slowly from one end to the other such that all parts of the surface are exposed for 10 seconds. Just based on a rough estimate of dimensions, this could take over 1 minute of slow and steady motion for a typical cell phone on one side. In general, for practical purposes these hand-held devices are largely useless and potentially dangerous. Few, if any, users are patient enough to hold the device very close to a surface and move it very slowly such that a sufficient UV-C dose will be delivered for disinfection purposes. Users may think they have done something by waving it around, but physics tells us that the effect will be minimal, resulting in a false sense of security. Using a chemical disinfectant wipe is much faster and likely more effective.

Inactivation of microorganisms is a function of UV dose (or UV fluence), with the unit of mJ/cm². UV dose is a product of fluence rate […] with the unit of mW/cm² * irradiation time (s). […]

[Regarding the claim that] “it delivers 2,350 µW/cm²”: well, it really depends on the distance (unless the beam is collimated). So at what distance is this irradiance delivered?

Assuming that at a particular distance it delivers 2,350 µW/cm², which is equivalent to 2.3 mW/cm² multiplied by 10 seconds…it will give a dose of 23 mJ/cm². (Please note that you have to keep the device for 10 seconds at each location where it delivers 2,350 µW/cm², so to cover the entire length of your keyboard going from side to side, it may take a much longer time.)

Is that enough to kill all the germs by 99.99%? No. Each microorganism has a particular sensitivity to UV, so you can never claim it kills everything. This dose is relatively good (you can likely inactivate 99.9% of some common pathogens, including SARS-CoV-2), but the claim of 99.99% for everything is false. For UV water disinfection, for example, regulations ask for 40 mJ/cm² (that is the UV dose that the New York UV water treatment plant delivers).

Is it safe for humans: again it depends on the exposure time [..] . But I would say not a chance. […] If the exposure time is just 1 second/day, it is safe, but not more.

In a nutshell, UV at the correct wavelength can kill viruses, but one can’t issue a blanket statement like “UV will kill 99.99% of viruses”. It is all about the UV dose applied and whether any of the viruses are completely blocked by the UV light due to being hidden in a microscopic crevice on the material surface, etc. So it all comes down to whether the device is properly designed, and whether it is used properly. This is complex and there are no standards right now to guide this since it is a relatively new technology.

[The statement that the device “delivers 2,350 µW/cm²”, which is “strong enough to kill 99.99% of viruses and bacteria, but is 100% safe for humans and pets”] is technically false. First of all, 2,350 µW/cm² is not a “dose”, but rather a rate of application of UV radiation. A Watt (W) is a unit of energy per second, and disinfection is a function of total energy received by the organism. You therefore need to define for how many seconds the UV is applied to come up with the total energy. Also, while there might be a “sweet spot” where there is enough UV-C to kill the organisms without harming a person, engineering that sweet spot on a consumer device will be difficult. Reputable manufacturers of UV-C devices therefore recommend against all exposure to the skin or eyes, to be safe.

The one theoretical exception is a specific subset of UV called “far UV”. There is evidence in the lab that it might be safe for human skin and eyes and yet still disinfect, but this is very preliminary and I could not personally recommend bringing such devices out into the public without more testing for possible unintended consequences.

I heard a statement [in the video] that “sunlight can kill diseases”. This is extremely misleading. Technically, given a long enough exposure to the sun, some disease-causing organisms will die. But life evolved on Earth in the presence of sunlight, so the majority of microorganisms contain some defense mechanisms against the sun. So, the sun is an extremely poor disinfectant overall.

In general, UV-C is an excellent disinfectant and is currently used for many applications. However, such applications are usually under very controlled conditions with well-engineered devices. Releasing UV-C into general consumer products will require very careful engineering and standard-setting. The engineering exists, but so far the standard-setting is still trying to catch up. As a consequence, I would urge consumers to be very careful about using UV-C disinfection devices, since there is a risk involved.

REFERENCES

• 1 – Cutler and Zimmerman. (2011) Ultraviolet irradiation and the mechanisms underlying its inactivation of infectious agents. Animal Health Research Reviews.
• 2 – Reed. (2010) The History of Ultraviolet Germicidal Irradiation for Air Disinfection. Public Health Report.
• 3 – Allen et al.(2020) The effectiveness of germicidal wipes and ultraviolet irradiation in reducing bacterial loads on electronic tablet devices used to obtain patient information in orthopaedic clinics: evaluation of tablet cleaning methods. The journal of Hospital Infections.
• 4 – Byrns et al. (2017) The uses and limitations of a hand-held germicidal ultraviolet wand for surface disinfection. Journal of Occupational and Environmental Hygiene.
• 5 – Meinhardt et al (2008) Wavelength-dependent penetration depths of ultraviolet radiation in human skin. Journal of Biomedical Optics.
• 6 – Welch et al. (2018) Far-UVC light: A new tool to control the spread of airborne-mediated microbial diseases. Scientific Reports.
• 7 – Buonanno et al. (2017) Germicidal Efficacy and Mammalian Skin Safety of 222-nm UV Light. Radiation Research.