Let There Be Germicidal Light: This $500 Fixture Could Stop the Next Pandemic, from Complex Systems

Summary

• AeroLamp is currently offering roughly $500 222-nm far-UVC lamps. Vivian Belenky compares their germicidal effect to roughly 30–50 equivalent air changes, while Misha Gurevich says the emitters produce around 100 milliwatts. One lamp covers about 250 square feet, so a classroom generally needs two to three lamps, perhaps four for a larger room. Professional installation is roughly as expensive as the lamps; the bulbs are rated for at least 10,000 hours while maintaining 70% output, or about five to six years in an 8-hour workweek.
• The market is still extremely small: Vivian estimates worldwide sales may be only a couple hundred lamps per year. Comparable lamps sell for roughly $2,000–$3,500, while AeroLamp is offering $500 units. Vivian thinks prices could reach roughly $100 within the next couple of years, even without technological innovation, although current prices can still matter for large installations.
• The safety mechanism is primarily mechanical. Far-UVC is absorbed by pathogen DNA, RNA and proteins; the 20-micron layer of dead, protein-rich skin cells absorbs nearly all 200–235-nm light, principally 222 nm. Eyes are more vulnerable: the tear layer absorbs only about 15%, the safe eye dose is lower than the safe skin dose, and long-term studies remain limited. Vivian says there are one-year and three-year eye-safety studies underway in Japan, with some uncertainty about the exact industry dose limits.
• The strongest preliminary field-relevant evidence discussed is a South African TB study reporting 90% transmission suppression in an animal setup where guinea pigs were exposed to humans only through the air. Vivian notes that TB is relatively resistant—perhaps ten times more resistant than typical respiratory viruses—so the result is encouraging but not a direct test of flu or coronavirus transmission.
• The likely early sites are long-term-care and senior centers, TB hotspots, hospital waiting rooms, universities, transport hubs, boarding schools and eventually schools. Vivian thinks elementary and middle schools could have the greatest benefit because of poor ventilation and children’s susceptibility, but expects parent and school conservatism to delay those deployments. Randomized trials are difficult, and early installations may see sublinear benefits until coverage becomes broad.
• The pandemic-prevention case is stronger than the ordinary-cold case. Vivian is uncertain how much far-UVC can prevent close-range colds, but is much more confident it could suppress a future highly contagious respiratory pandemic. She points to traditional 254-nm UVC’s historical use against measles, which she estimates has a reproduction number around 20, versus “one point something” for COVID-19 at its worst.
• The main bear case is transmission dynamics, not basic feasibility. Speaker 2 warns that if common airborne illnesses are mostly transmitted through short-range, high-dose interactions, far-UVC may have little everyday business case even if it remains important pandemic infrastructure. Vivian estimates a typical installation provides one equivalent air change every two minutes—about 90% reduction in coronavirus or influenza virus in eight minutes and 99% in roughly fifteen—but says close two-foot conversational exposures are harder to interrupt. She considers no benefit at all unlikely.
• Adoption is constrained by awareness, evidence, price and supply-chain scale rather than a single decisive regulatory obstacle. Gurevich emphasizes information and broad awareness; Vivian favors trial deployments as a more robust way to build confidence. The best current krypton-chloride emitters come from one Japanese company and require hydrogen fluoride gas, while solid-state far-UV emitters remain a future possibility. Far-UVC is additive to ventilation and filtration, and employers may have a direct business case because they bear sickness and caregiver-absence costs. One speaker views $500 home deployment as uneconomic for most households; Vivian argues that special cases such as babies or immunocompromised people can justify it.

Deep dive

1. The mechanism: protein absorption is both what makes it work and what makes it safer

Vivian Belenky explains that far-UVC is absorbed by pathogen DNA and RNA, as well as by essentially all proteins. That protein absorption is what makes it safer than longer germicidal UV wavelengths. Pathogens remain in the air after exposure, but they can no longer replicate, and the effect is rapid.

Her performance analogy is an extremely strong air purifier: instead of adding perhaps one or two air changes in a space, far-UVC can provide the equivalent of roughly 30–50.

Patrick McKenzie initially assumes that the wavelength is present in sunlight. Belenky corrects him: sunlight is primarily UVA and UVB, while UVC is completely blocked by the ozone layer. Sunlight is still mildly germicidal, but it compensates for its weaker per-photon effect with much greater total light, prompting McKenzie’s aside that Oliver Wendell Holmes was empirically disproven by the science.

2. The safety story is mechanical, with the eye as the harder case

Belenky calls the safety mechanism “much more of a mechanical story than a chemical or biological story.” Humans have a roughly 20-micron layer of dead skin cells, the stratum corneum, that is full of proteins and absorbs nearly all far-UVC. This is specific to the shorter 200–235-nanometer wavelengths, principally 222 nm.

Longer UVC wavelengths such as 254 and 265 nm, used in water disinfection, do not have the same degree of protein absorption. Belenky will not categorically say they cause cancer, but says they are not pleasant to be exposed to and are thought to be less carcinogenic than UVB.

The eye is more vulnerable because it has no equivalent dead-cell layer. The tear layer absorbs only about 15% of the incident far-UVC, with the remainder absorbed in the epithelium. A relatively low dose can therefore cause eye pain or discomfort. Eyelids, eyelashes, eyebrows and the brow ridge reduce normal exposure, but the safe effective eye dose is lower than the safe skin dose.

Belenky says the technology has not existed long enough for extremely long-running studies. She mentions one one-year and one three-year eye-safety study underway in Japan, with the qualification “I think.” She also says the likely failure mode is more like staring into a bright light than looking into an infrared laser: pain should cause a person to stop looking, rather than allowing a silent blinding exposure. The remaining uncertainty is how to translate the safety evidence into industry practice.

3. Corner-mounted units are an existence proof, not the end state

AeroLamp’s current units are usually mounted in room corners because the lamps have a narrow beam angle. Pointing one corner toward the opposite corner maximizes beam path length and therefore the average dose across the room.

Belenky ultimately expects ordinary overhead fixtures—“just a boring ceiling fixture,” comparable to other routine building infrastructure—in offices, hospitals and schools. The current design is easier to install and demonstrates that a customer can buy a unit and use it immediately.

Gurevich describes the current lamps as an “existence proof” that the technology is ready to deploy. He says there are no insurmountable technical or logistical barriers, while acknowledging the ordinary difficulties of operating a high-tech business. The product is not limited to secret government laboratories or $10,000 installations.

The installation advantage over older UVC systems is safety. Upper-room systems using older wavelengths must be installed carefully because a bad installation can cause rapid eye damage. Gurevich says a 222-nm system is intrinsically safer and can be installed by an ordinary electrician. For a basic deployment, McKenzie describes the task as installing a somewhat unusual light fixture rather than solving an unsolved materials-science problem.

4. Unit economics and deployment scale

Gurevich estimates roughly 250 square feet of coverage per lamp. A typical classroom needs two to three lamps, perhaps four for a larger room. A small building might require four to five figures of lamp spending; a large institution such as a university could need hundreds of lamps and six figures in lamp costs alone. Professional installation may cost about as much as the lamps, while a do-it-yourself installation can be nearly free: a unit can be attached to a wall and plugged into an outlet in about ten minutes.

Belenky calls budgeting roughly the same amount for installation as for the lamps a conservative mass-scale rule of thumb. The exact figure depends on the electrical system and ceiling type.

The manufacturer rates the current bulbs for at least 10,000 hours while maintaining 70% output. Belenky has seen data suggesting 13,000–14,000 hours may be possible. At eight hours per day during a workweek, that implies roughly five to six years; at continuous 24/7 operation, about a year and a half.

McKenzie’s structural argument is that the product fits on a shipping container and should therefore become cheaper as production scales. He contrasts that with medical interventions whose costs remain high because they require substantial labor, arguing that lighting for a hospital is rarely unaffordable in the same way.

5. Where the first evidence and deployments may appear

Belenky says elementary and middle schools could produce the greatest benefit because they are often poorly ventilated and children are immunologically naive. However, she is not sure there is enough safety evidence for a large fraction of parents to welcome a relatively new technology in schools. She expects private or specialty schools to move before public schools.

She also identifies long-term-care centers and hospital waiting rooms as high-value sites. In a waiting room, the aggregate occupancy is high even though individual visitors are relatively transient, reducing the chance that one person receives a potentially concerning cumulative dose while still providing substantial population-level exposure.

Gurevich is more optimistic about early evidence in specialized settings with limited social mixing or unusually susceptible pathogens. He highlights tuberculosis hotspots and long-term-care or senior centers, where he expects noticeable transmission reductions relatively quickly. Universities may also be early adopters because they involve older students rather than children.

The South African TB work is presented as preliminary evidence. Belenky says it reports 90% transmission suppression in TB wards, but clarifies that it is an animal study: guinea pigs are exposed to humans only through the air, and researchers monitor how many develop tuberculosis.

She also cautions that TB is relatively resistant to far-UVC—perhaps ten times more resistant than a typical respiratory virus such as flu or coronavirus—and that TB may not transmit through the air in exactly the same way. The result is encouraging but does not settle the broader question.

6. The pandemic case and the role of building codes

Belenky calls pandemic prevention “by far the most exciting element” of far-UVC, even though she avoids emphasizing pandemics when speaking with ordinary customers because people remain traumatized by COVID-19.

She is less certain about ordinary colds, which may require extended close interaction, but more confident about suppressing a future respiratory pandemic. Her reasoning is that highly contagious pathogens provide more opportunities for environmental interventions to interrupt transmission. She points to traditional 254-nm UVC’s use in controlling measles, which she estimates has a reproduction number of roughly 20, compared with COVID-19’s “one point something” at its worst.

McKenzie emphasizes the coordination advantage. Vaccination and masking require many individual decisions, while a building owner can install an environmental intervention unilaterally. Gurevich adds that infection-prevention requirements could eventually be incorporated into building codes and then deployed through normal commercial renovation cycles.

He identifies ASHRAE 241 as the current infection-prevention standard, while noting that it is still under construction and not broadly adopted by authorities having jurisdiction. If a building code requires a particular amount of infection prevention and UVC is the cheapest and easiest way to provide it, he expects buildings to use it.

7. Far-UVC complements ventilation and filtration

Belenky does not present far-UVC as a replacement for other clean-air interventions. Buildings should receive more outdoor air, more filtered recirculated air and in-room air cleaners. She says a MERV 13 portable air filter can be as effective as a higher-rated filter while being quieter.

Filtration is still useful alongside UVC because pathogens are not the only airborne pollutants. Dust, allergens, particulate matter and chemical pollutants require other interventions. Far-UVC has a modest effect on allergens through protein absorption, but much less than conventional filtration.

For pathogens, however, Belenky says simply moving air can be insufficient in large, densely occupied spaces such as auditoriums, lecture halls and gyms. Meeting a clean-air standard without UVC could be cost-prohibitive for many buildings.

McKenzie argues that far-UVC can stack with vaccines, filters, ventilation and other interventions. By lowering circulating pathogen levels, it may make existing measures more effective while requiring less individual coordination than vaccination or masking.

8. Adoption is a social-diffusion problem, but current prices and supply also matter

Vivian estimates that worldwide sales may be only a couple hundred lamps per year, and she is not sure whether that figure is increasing. If adoption becomes socially normalized, she suggests LED lighting as a reference case, but expects at least a decade of broad deployment after the technology truly takes off because commercial buildings renovate on roughly a ten-year cycle.

She says the absence of a decisive barrier is frustrating. There is still substantial safety and other research to do, but she does not see one critical missing study that must be completed before deployment can begin, nor does she see a major regulatory barrier. Her description is a “social-diffusion question”: how does the idea become something people know they can and should do?

Gurevich sees broad information and awareness as the main bottleneck. Belenky would allocate substantial resources to trial deployments because she thinks formal evidence may be a more robust way to build awareness. She also warns that too much attention too early could have an “IFSR effect,” so the company wants to scale responsibly.

9. Supply chain and the price ladder

Gurevich says the best current emitters are produced by a single Japanese company whose business is focused on high-end, high-margin products. Even after removing that margin, the emitter cost is roughly $15–$20 per unit. The manufacturing process also requires hydrogen fluoride gas and is less straightforward than LED production, though he does not view it as fundamentally unscalable.

Belenky pushes back on the simple comparison with LEDs. LED chips are themselves highly complex and capital-intensive, but once the capital investment is made, production can scale very effectively. Krypton-chloride excimer lamps are currently on a less aggressive cost curve and probably will not reach white-LED economics. Solid-state, chip-based far-UV emitters could eventually improve the cost structure, but those technologies are still far out.

Current lamp prices are more than the emitter cost alone. Gurevich says many products sell for around $2,000, with one reported sale at $3,500. He distinguishes price from cost: the current price reflects a tiny industry in which companies need high margins to survive, rather than a fundamental cost floor.

Vivian says this is why AeroLamp is offering a $500 lamp and believes the price could fall to roughly $100 within the next couple of years, or sooner, without any major technological innovation. At the same time, large installations requiring 100 lamps can still face a meaningful current price barrier.

10. The bear case is transmission dynamics

When McKenzie asks what could cause the thesis to fail, Speaker 2 identifies the key uncertainty as the actual structure of airborne transmission. If a large share of ordinary colds and flu are transmitted at short range, with one person delivering a large dose directly to another, there may be little long-range transmission for far-UVC to intercept.

That would produce an uncomfortable outcome: far-UVC could remain crucial pandemic infrastructure while lacking a near-term commercial case. Prevention is difficult to sell when the benefit is an uncertain pandemic a decade or two in the future, rather than illnesses or absences prevented within the next year.

Vivian quantifies the environmental effect as roughly one equivalent air change every two minutes. Under a shared-air transmission model, that would reduce coronavirus or influenza virus by about 90% in eight minutes and 99% in roughly fifteen minutes. If infection requires sharing air with an ill person for 30 minutes or an hour, that could substantially reduce risk. If infection instead comes from a large dose delivered during a close two-foot conversation, environmental treatment has less opportunity to work.

She adds that lower viral dose might still reduce disease severity, even when it does not fully prevent infection. Her conclusion is that the benefit could range from marginal to socially transformational, but she would be very surprised by no benefit at all.

11. Homes, immune development and the indoor microbiome

The home market produces a disagreement. One speaker argues that a $500 lamp is not cost-effective for most households with limited disease transmission, though the calculation changes for wealthy people, people who strongly value avoiding illness and immunocompromised people. Vivian pushes back using her own experience after having a baby: she kept two lamps in her living room and used them when hosting people.

She argues that illness in a tiny baby can lead to an emergency hospital visit, hospitalization and potentially a spinal tap, making prevention more valuable in some households even if the broader societal benefit of home deployment is modest. She still sees schools, transport hubs and other shared environments as higher priorities for broad social suppression.

On the hygiene-hypothesis objection, Vivian strongly rejects the idea that children need clinical viral infections to train their immune systems. She says current thinking places more emphasis on environmental and commensal bacteria and other microorganisms. She cites a study in which childhood RSV exposure had a neutral-to-negative effect on future illness, and says measles can damage immunological memory.

Her categorical claim is that viral illness is purely harmful and that immune training does not require clinical infection. If a child must get sick, she would rather it happen at ten than five, and at five than one, but would prefer that it not happen at all. McKenzie jokes that people could intentionally infect themselves later; Belenky responds that this technology is called vaccines.

The indoor microbiome remains less studied. Belenky says far-UVC is much less effective on surfaces than in air and that surface bacteria are extremely hardy, so her bet is that any cost would be minor. She does not claim the question is settled. McKenzie also notes that turning off a lamp is a relatively easy way to stop exposure compared with reversing a biological intervention.

12. The commercial wedge: clean air, employers and awareness

Gurevich’s closing appeal is for broad awareness of clean-air technologies, not merely for customers to buy AeroLamps. He says very few people have heard of UV disinfection, and even fewer have heard of 222 nm. Once people understand the concept, he thinks the pitch is straightforward: a science-fiction-like technology that reduces the risk of getting sick.

The discussion of a hypothetical first $1 million produces another disagreement. Gurevich favors awareness, while acknowledging that trial deployments could be the better choice. Belenky favors trials as a more robust path to credibility and says celebrity promotion could be very good, very bad or have no effect. Gurevich’s hypothetical celebrity pick is Paris Hilton, because associating clean air with high-end hotels could be valuable.

Belenky says formal studies are unusually vulnerable to a heckler’s veto: one person’s discomfort can prevent an institutional review board from approving an installation. Voluntary deployment by a building owner is easier, and once the technology is installed, its effects can be studied later. In offices, she says employees are usually more supportive than building managers, who must decide whether to spend the money.

The nonpolitical business case is the employer’s own balance sheet. A company with highly paid employees bears the cost of employee sickness and of parents missing work to care for sick children. Belenky says her economic analysis found that caregiver absenteeism is a significant part of the cost of colds.

The founders point to financial firms that adopted ventilation and filtration measures early in COVID-19 after seeing the business case. Their broader argument is that clean air can become a normal part of building infrastructure, supported by employers, building owners, standards and public awareness rather than relying only on individual public-health choices.