UV lamps and UVC LEDs compared for water disinfection

UV Lamps in 2026: Mercury UV Lamps vs UVC LEDs, Wavelengths, Efficiency, and Water

Ultraviolet disinfection has quietly become one of the most important tools in water treatment, air purification, and surface sanitation. For decades, that job belonged almost entirely to mercury vapor lamps. In 2026, that picture is changing fast. UVC LEDs have closed much of the efficiency gap, international mercury regulations are tightening, and buyers now have a real choice to make instead of a default. This article breaks down how each technology actually works, what the wavelengths mean in practice, where the efficiency numbers stand today, and which option makes more sense for water disinfection projects right now.

UV Wavelengths and Why They Matter

"UV" covers a wide band of the electromagnetic spectrum, but disinfection depends on a narrow slice of it. UVC, roughly 200 to 280 nanometers, is the range that damages the DNA and RNA of bacteria, viruses, and protozoa enough to stop them from reproducing. Within that band, absorption by nucleic acids peaks around 260 to 265 nanometers, which is why manufacturers describe this zone as the germicidal sweet spot.

This matters because not every UV source hits that sweet spot the same way. Low-pressure mercury lamps emit an almost single wavelength at 253.7 nanometers, commonly rounded to 254 nm, which sits close enough to the DNA absorption peak to be highly effective. UVC LEDs, by contrast, can be manufactured to emit at a chosen wavelength, and most disinfection-grade LEDs today are built around 265 to 280 nm because that range balances germicidal effectiveness with manufacturing yield. A separate category, far-UVC LEDs and excimer lamps around 222 nm, sits outside the traditional germicidal zone and behaves quite differently, which we cover further down.

How Mercury UV Lamps Work

A mercury vapor lamp passes an electric current through mercury gas sealed inside a quartz tube, exciting mercury atoms so they release UV photons. Low-pressure lamps run at a low internal gas pressure and produce that near-monochromatic 254 nm output efficiently. Medium-pressure lamps run hotter and at higher pressure, producing a broader, polychromatic spectrum with more total power per lamp, which makes them useful for high-flow municipal water plants and large air-handling systems.

A common variant, the amalgam lamp, replaces liquid mercury with a mercury-indium amalgam. It runs hotter than a standard low-pressure lamp and delivers more UV output per fixture, letting operators use fewer lamps for the same disinfection dose, though it still relies on mercury.

Mercury lamps have real operational drawbacks. They need a warm-up period before reaching full output, their UV yield drops in cold water or cold ambient air, the quartz envelope is fragile, and they require a ballast and relatively high voltage to strike an arc. Because they contain mercury, spent lamps require hazardous waste handling rather than ordinary disposal. Typical service life runs about 9,000 to 12,000 hours, which in continuous operation often translates to an annual replacement cycle.

How UVC LEDs Work

UVC LEDs are semiconductor devices, generally built from aluminum gallium nitride or aluminum nitride materials, that emit UV light directly when current passes through them, with no gas, filament, or warm-up period involved. They reach full output instantly, can be dimmed or pulsed, tolerate cold temperatures without a drop in UV yield, and contain no mercury at all.

The tradeoff has historically been efficiency and cost. Because UVC LEDs are point sources rather than the natural line source a mercury tube provides, disinfection systems built around them need carefully engineered arrays and reflectors to spread UV dose evenly across a chamber, which adds design complexity, particularly for higher water flow rates.

Efficiency in 2026: Where the Numbers Actually Stand

Wall-plug efficiency, the percentage of electrical input converted into usable UV output, is the clearest way to compare these technologies, and it is where mercury has held its advantage the longest. Mercury lamps at 254 nm typically reach wall-plug efficiencies in the range of 25 percent, a figure researchers at the University of California, Santa Barbara cited directly when benchmarking new UV LED devices against conventional lamps.

UVC LEDs have improved substantially but generally remain behind that mark. Most commercially available disinfection LEDs today operate between 3 and 10 percent wall-plug efficiency, according to semiconductor manufacturer Crystal IS. Recent lab and near-commercial breakthroughs are narrowing the gap, though. In 2025, ams OSRAM announced a 265 nm UVC LED validated at 10.2 percent wall-plug efficiency by Germany's Physikalisch-Technische Bundesanstalt, roughly double the efficiency of its previous generation, with commercial availability targeted for late 2026 and a rated lifespan beyond 20,000 hours. Around the same period, Stanley Electric reported a 265 nm UVC LED reaching 7.5 percent wall-plug efficiency, about three times its prior result, with a company target of 10 percent by the end of fiscal year 2026 and sample shipments beginning in March 2026.

Efficiency also depends heavily on wavelength. Research on far-UVC LEDs published in 2024 found wall-plug efficiency climbing steadily with wavelength within that band, from roughly 0.5 percent at 227 nm up to about 2.4 percent at 235 nm, well below the efficiency achievable at longer UVC wavelengths. At the opposite end, researchers pushing toward less germicidally optimal but longer wavelengths near 280 to 298 nm have posted much higher lab results, including a UC Santa Barbara device reaching 19 percent wall-plug efficiency, and earlier work from Chinese research teams reporting over 20 percent using a novel monolithic device structure at 280 nm. In short, the closer an LED's wavelength gets to the ideal germicidal range or into far-UVC territory, the harder efficiency becomes to achieve, while efficiency improves as wavelength shifts longer, away from the disinfection sweet spot.

The Mercury Phase-Out and What It Means for Buyers

The Minamata Convention on Mercury, the global treaty governing mercury-containing products, has been reshaping the lighting industry for several years. At its fifth Conference of the Parties in Geneva in late 2023, delegates from roughly 147 countries agreed to phase out the manufacture, import, and export of general-purpose fluorescent lighting by 2027. It is important to note that this obligation applies to manufacturing, import, and export, not to the continued use of lamps already installed.

Specialty UV lamps used for water and air disinfection, curing, and similar industrial purposes have historically been treated as a separate category under the Convention, with countries able to register exemptions that can be extended over time. Several of those exemptions were reviewed and extended at the Convention's sixth Conference of the Parties in 2025. Even so, the direction of travel is clear: mercury-containing UV sources face shrinking regulatory runway worldwide, and equipment manufacturers are actively developing mercury-free alternatives well ahead of any hard deadline.

Which Technology Wins for Water Treatment

In practice, the answer in 2026 depends on scale. Large municipal water treatment plants still lean heavily on medium-pressure and amalgam mercury lamps because a single fixture delivers substantial UV output through a well-understood line-source reactor design, and the cost per unit of delivered UV dose remains lower than an LED array of equivalent output. Retrofitting a plant validated for mercury-based reactors is also a significant engineering and regulatory undertaking, which slows adoption even where LED technology would otherwise be attractive.

UVC LEDs have found their strongest footing in point-of-use and low-flow applications: under-sink drinking water units, portable purifiers, small-scale rural or off-grid systems, and equipment where instant on-off operation, ruggedness, or the complete absence of mercury outweighs the efficiency gap. As LED wall-plug efficiency climbs toward and past the 10 percent mark and lifespans stretch beyond 20,000 hours, that footprint is expanding into mid-sized commercial systems as well.

Far-UVC (222 nm): A Different Use Case Entirely

Far-UVC, centered around 222 nm and typically produced by krypton-chloride excimer lamps rather than standard LEDs, has generated significant research interest because of its safety profile in occupied spaces. Multiple studies, including human ocular safety trials and long-term skin exposure research, have found that 222 nm light is absorbed almost entirely in the outermost dead-cell layers of skin and the eye's tear film, meaning it does not appear to reach living cells the way longer UVC or UVB wavelengths do. That has made far-UVC a candidate for continuous air disinfection in occupied rooms, something impossible with conventional germicidal UV. It is not, however, a serious contender for water treatment. Current far-UVC efficiency is far lower than either mercury lamps or mainstream UVC LEDs, and its primary value lies in air and surface applications rather than liquid disinfection.

Choosing Between Mercury and UVC LEDs in 2026

For high-volume municipal or industrial water treatment, mercury lamps, particularly medium-pressure and amalgam types, generally remain the more cost-effective choice today, though that advantage is narrowing. For point-of-use systems, portable equipment, mercury-free product lines, or applications where instant-on operation and cold-temperature performance matter, UVC LEDs are increasingly the practical default. Regulatory pressure under the Minamata Convention, combined with steady efficiency gains from manufacturers like ams OSRAM and Stanley Electric, means the gap will keep closing. Buyers making long-term equipment decisions in 2026 should weigh not just today's efficiency numbers, but the direction both technologies are heading.

Frequently Asked Questions

What wavelength is most effective for UV water disinfection? Wavelengths between about 260 and 265 nanometers align most closely with the peak DNA and RNA absorption that damages pathogens, which is why low-pressure mercury lamps at 254 nm and UVC LEDs tuned near 265 nm are both considered highly effective for disinfection.

Are UVC LEDs as efficient as mercury lamps yet? Not quite, as of 2026. Mercury lamps typically reach around 25 percent wall-plug efficiency at 254 nm, while most commercial UVC LEDs operate between 3 and 10 percent, though recent lab results near 265 nm have reached the 7 to 10 percent range and continue to improve.

Is far-UVC light at 222 nm safe for people? Multiple published safety studies, including human ocular trials and extended skin exposure research, have found no significant skin or eye damage from 222 nm exposure within tested dose limits, because the light is absorbed in outer, non-living tissue layers before it can reach living cells.

Are mercury UV lamps being banned? The Minamata Convention has set a 2027 target for phasing out the manufacture, import, and export of general-purpose mercury fluorescent lighting, but specialty germicidal and industrial UV lamps fall under a separate exemption category that several countries have renewed, so mercury UV lamps remain legally available in many markets while regulatory pressure continues to build.

Can UVC LEDs handle high water flow rates like municipal treatment plants? Not as easily as mercury systems today. Because LEDs are point sources rather than line sources, reactors built around them need arrays and careful hydraulic design to deliver an even dose, which has generally limited them to lower-flow applications, though improving LED efficiency and array design are gradually pushing that ceiling higher.

Retour au blog