Ultraviolet Germicidal Irradiation: Future Directions for Air Disinfection and Building Applications†
- †This paper is part of the Symposium-in-Print on “Ultraviolet Germicidal Irradiation.”
Corresponding author e-mail: email@example.com (Shelly L. Miller)
Ultraviolet germicidal irradiation (UVGI) for air disinfection applications has relied on low-pressure mercury vapor lamps for decades. New design requirements have generated the need for alternatives in some uses. This study describes the current state of UVGI technology and describes future directions for technology development, including the use of lamps produced from nontoxic materials and light-emitting diode lamps. Important applications are discussed such as the use of ultraviolet germicidal lamps in developing countries, in heating, ventilating and air-conditioning systems to improve energy efficiency and indoor air quality, and for whole room disinfection.
Ultraviolet germicidal irradiation (UVGI) air disinfection is typically applied in upper room configurations in which louvered fixtures are mounted to the ceiling or walls to treat the upper areas of a room while keeping irradiation levels to a minimum in the occupied areas. UGVI is also used in ventilation systems to disinfect the air as it moves through the air-handling ductwork. Many studies of UVGI have been published including reviews that summarize the history of UVGI , describe its use for tuberculosis control [2, 3] and provide design guidelines for health-care settings .
Most if not all lamps currently sold for UVGI air disinfection applications are low-pressure mercury (Hg) vapor lamps. These lamps are typically about 30% efficient at converting input power into ultraviolet C (UVC) radiation. UVC refers to short wavelength electromagnetic radiation in the range 100–280 nm. Low-pressure mercury vapor lamps emit >90% of their total spectral power at 253.7 nm. They exist in a variety of shapes and power to fit every requirement.
Improvements in UVC lamps and fixtures to consider are better lifetime reliability, reflectors integrated in the lamp to efficiently redirect UVC radiation to one side, high-output lamps to make more compact systems, intelligent lamp drivers and controls to adapt performance depending on actual need (e.g. occupied/unoccupied room) or to warn in case of malfunction. New lamp technology development includes light-emitting diodes and the use of nontoxic materials. Applications in which UVGI could play a significant role include developing countries and countries with significant tuberculosis rates, cleaning of cooling coils in heating, ventilating and air-conditioning systems, and whole room/surface disinfection.
This study reviews the future direction of UVGI air disinfection applications and needs for different lamps and fixtures to meet design requirements.
Currently all lamps used in UVGI air disinfection applications are low-pressure Hg vapor lamps. These lamps are similar to an ordinary Hg fluorescent lamp, but the tube contains no fluorescent phosphor and the tube is made of fused quartz, instead of glass. These two changes combine to allow the 253.7 nm ultraviolet radiation produced by the mercury arc to pass out of the lamp unmodified (whereas, in fluorescent lamps, it causes the phosphor to fluoresce, producing visible light).
A wide range of lamps are commercially available, but can be generally categorized as either low output driven by conventional magnetic ballasts or high output driven by electronic ballasts . Lamp output energy and spectrum depends on lamp pressure, electrical current, voltage, excitation waveform, discharge ignition, and internal gas composition . Low-output lamps are driven at low power, whereas the high-output lamps are overdriven by increasing current input into the lamps to produce more output radiation. Ballasts that supply the electrical power to the UV lamps are built with varying technology including transformers and solid-state ballasts and must be chosen specifically for the lamp in use. Improvements in lamp ballasts have been reported by manufacturers and include improved energy-efficient designs, cold start capability and striation-reducing technology.
Developments in lamp hardware have led to new technologies such as low-pressure amalgam lamps, but these have not yet been utilized in air disinfection applications. Amalgam lamps can have slightly higher input conversion efficiencies, upwards of 38%, and operate at higher temperatures . Also, lamps produced from nontoxic materials are not presently available in germicidal fixtures, but are needed for many air disinfection applications (they are currently available for water disinfection applications). The potential release of mercury, for example, is a concern during disposal and for some environments and industries, such as aircraft manufacturers, due to the potential for hazardous material exposure if accidently broken . Improvements in low-pressure UVC lamps for upper air application to consider are better lifetime reliability, high-output lamps to make more compact systems and intelligent lamp drivers and controls to adapt performance depending on actual need (e.g. occupied/unoccupied room) or to warn in case of malfunction.
Nontoxic alternatives should be the future goal as green manufacturing expands. Ryan and colleagues  reported on two new lamp technologies that were investigated—a light-emitting diode (LED) and a xenon lamp. The advantage of an LED lies in its solid-state construction and that the radiation wavelength can be selected within the UVC range 200–280 nm for more application-specific wavelength output. An LED called the UVTop265 made by Sensor Electronic Technology Inc. (Columbia, SC), which emitted its peak wavelength at 265 nm, was tested in a single-pass flow-through control device . Although the LED functioned, it was very inefficient at producing UVC radiation (0.3%) and it required almost 10 W of active cooling to maintain operating temperatures. Research conducted at Harvard investigated the performance of a sample array of 9 LEDs with an average total UV output of 3.6 mW from Crystal IS, Inc. (Green Island, NY) (data not published). The array had an average irradiance of 7 μW/cm2 and an efficiency of 0.5% when measured with a planar UV meter (Gigahertz-Optik) at a distance of 10.2 cm away from the array at midheight of a benchtop flow chamber used previously in bacterial inactivation experiments . The average peak wavelength was 263 nm and the array was cooled by forced air. Unfortunately, the UV output of the array was not powerful enough to inactivate Bacillus atrophaeus spores irradiated in a single-pass chamber. Since single-pass flow-through device effectiveness depends on many factors including flow rate, it is possible that inactivation could have been achieved with this system with a lower flow rate, a more susceptible organism or more LEDs . According to a recent publication, Crystal IS, Inc. is producing new LEDs with more than triple the total UV output and 2% efficiency and is expecting to further increase this output in the coming years .
LED technology remains an attractive future alternative for design of distributed upper room UV over traditional mercury vapor lamps. However, to become a viable UVGI air disinfection solution, both the efficiency and cost of LEDs will need to continue to improve dramatically, whereas their operating voltage should be reduced . More studies are needed to demonstrate if LED efficiency offsets their hardware and operating costs compared with those from low-pressure mercury lamps. UVC LEDs are of great interest in other applications as well, such as water disinfection. In a comparison study examining the inactivation of E. coli in water, germicidal UV LEDs provided an equivalent level of treatment to that of conventional low-pressure UVC lamps . UV LEDs were used in combination with a TiO2 catalyst to disinfect airborne Staphylococcus aureus collected on a surface . Advancements in the field of UVC LEDs in other applications may prove beneficial, as they will translate to UVGI air disinfection applications as well.
The Xenon lamp tested by Ryan et al.  was a prototype and emitted its peak wavelength at 240 nm. This lamp was a 10 W unit that emitted approximately 1.4 W of UVC radiation. A disadvantage to this lamp was that it produced ozone. The obvious advantage to developing a lamp made from xenon is that it is a nontoxic alternative to mercury. However, more development work is needed to ensure that xenon lamps do not emit ozone into the indoor air, as ozone is a strong oxidant and toxic air pollutant. One approach that could be used is what is typically done with low-pressure mercury lamps, which are made “ozone free” by using doped quartz that blocks the UV wavelengths that produce ozone.
Pulsed UV lamps are available from the Xenon Corporation (Wilmington, MA). Review of the information on their website shows that they have lamps available at a peak wavelength of 240 nm that do not produce ozone. In pulsed UV lamps, intense radiation is pulsed several times per second and each pulse lasts between 100 ns and 2 ms. Systems can be made with adjustable pulse repetition frequency settings, although the ability to adjust the pulse duty cycle is limited by concerns with overheating the flash lamp. Major applications of these lamps are food decontamination, surface disinfection in medical facilities  and water purification. No tests have been conducted using these lamps for air disinfection to our knowledge.
Further innovations are needed to develop more efficient lamps, more versatile lamps and lamps that are made from nontoxic materials.
Current fixture designs implement louvers to direct radiation. For upper room applications, radiation is optimally kept in the upper 25–30% of the room volume. Louvers, however, absorb most of the UVC radiation generated by the lamps. New designs are needed that optimally direct and keep the radiation in the upper portion of the room, and that do not absorb radiation. Typical designs of upper room UVGI fixtures include parabolic reflectors placed behind the lamps, to greatly improve output from the fixture. The efficiency of UVC output from a fixture varies considerably. Rudnick et al.  determined that the efficiency of fixture-to-lamp UVC output ranged from 1.2 to 5.6%. New materials or coatings could be explored that would improve reflectivity. A proprietary high-reflectance surface coating (L2B Environmental Systems Inc., North Barrie, Ontario Canada) was tested and found to increase the average fluence rate in a single-pass flow-through reactor by a factor of 1.62 compared with uncoated aluminum . The spatial distribution of UV radiation was also affected  and resulted in higher levels throughout the length of the reactor as well as a higher peak radiation.
Development and use of economical UVGI refractive materials to produce an efficient lens are one alternative to direct UVC out of fixtures without the need for louvers. However, currently identified materials that are transparent to UVGI are either quite expensive (quartz, due to the high melting temperature required for shaping into a lens) or prevent collimation of the UVGI due to scattering (fluorinated ethylene propylene, due to the high surface roughness of these polymers at UV wavelengths).
Upper room systems should be designed to distribute the radiation evenly throughout the upper portion of the room. This can be difficult to do for large spaces as radiation decreases as the inverse of the square of the distance away from the point source (approximating a single UV fixture in a large room). Thus, the radiation peaks nearest the fixture and falls off rapidly. Methods to better distribute UVGI evenly at low cost are needed. Computer models have been developed to allow for better-designed and safer facilities with UVGI fixtures [16, 17].
One potential solution to these problems lies in utilizing LED sources as they become more efficient. Because of their tiny size, LEDs could be economically fitted with individual glass lenses to direct all of the UV emitted along a horizontal path. In addition, their very small size and potential to be configured in a wide range of sizes and geometries gives entirely new design options, such as spacing individual LEDs throughout the upper room to distribute UVGI more equally.
Safety continues to be a concern over the use of UVGI fixtures in the workplace, although adverse health effects are rare and occur typically as a result of improper maintenance procedures. A study of eye and skin exposure in a hospital, office and homeless shelter in which germicidal UV was being used showed that doses were minimal and well below the recommended level . Education regarding the use and operation of UVGI fixtures needs to be provided to all institutions using this technology. Professional associations could help to promote a better understanding of the safety of UV.
Other Applications and Technology
A low-cost, low-maintenance, effective solution is needed to support the use of UVGI in developing countries and countries with significant tuberculosis (TB) infection rates. Countries such as South Africa or Russia have high tuberculosis incidence and HIV coinfections are common as is drug-resistant TB [19, 20]. Engineering controls are a necessary part of a successful airborne infection control program and UV lamps are an obvious choice. However, fixtures that are commonly sold and used in the United States can be too expensive for developing countries. In 2011, the National Department of Health in South Africa implemented a moratorium on the purchase of UV germicidal irradiation fixtures for use in the departments of health due to incorrect installations and poor-quality UV lamps . More work is needed to support the appropriate use of UVGI technology in developing countries. Evidence-based professional design and installations must be used. In addition, appropriate procedures for maintenance and disposal of lamps need to be implemented, and health-care workers must be trained in the use of UVGI. Importantly, systems need to be continuously monitored and effectiveness should be independently validated, with the data readily available for equipment users. With these changes, UVGI could be successfully applied in developing countries and help to control TB. Harvard University is working to develop a guideline for effective, affordable upper air fixtures for developing countries and it is expected that the above-mentioned issues will be included in the guideline.
UVC coil cleaning systems installed in air-handling units in heating, ventilating and air-conditioning (HVAC) systems are increasingly being used in many climates across the United States. UVC coil cleaning technology uses specially designed fixtures that are installed just downstream or upstream of the cooling coils, irradiating the surfaces of the coil to disinfect system components. Few published studies to date have documented the impact of this application. There is anecdotal evidence that it saves energy by restoring and maintaining the air-handling unit's capacity to initial design levels [22, 23]. A study by the California Energy commission of UVC coil cleaning technology applied in schools showed that total fungal and gram-positive bacteria on cooling coil surfaces was reduced by 65–100%, and the system air flow did improve, but was not statistically significant due to sample size and conditions . A recent study also showed that UVC coil cleaning technology had a significant impact on the indoor air quality and improved health of patients in a hospital. UVC coil cleaning was implemented in a neonatal intensive care unit's air-handling unit. Results showed significantly decreased HVAC surface, environment (including air samples) and tracheal microbial colonization, as well as ventilator-associated pneumonia and use of antibiotics . In addition, a study by Menzies et al.  in office buildings found that use of UVC coil cleaning technology installed on the upstream side of the coil “was associated with significantly fewer work-related symptoms overall.” The study concluded that applying this technology in most North American offices could reduce work-related symptoms, in roughly 4 million employees, caused by microbial contamination of HVAC systems.
Another important application of germicidal lamps is UVC surface/air cleaning technology for whole room disinfection. UVC surface/air cleaning technology is implemented as stand-alone units with high-powered lamps that generate high fluence rate levels and are used in unoccupied spaces to disinfect the whole room. A recent study of a stand-alone unit showed significant reduction in aerobic colony counts and C. difficile spores on contaminated surfaces; no air sampling was performed . While the use of portable UVGI lamps to disinfect unoccupied rooms is not new, the tested unit has a unique feature in that it incorporates sensors that measure surface-reflected UVC from 360° to automatically calculate the pathogen-lethal UVC dose required for each room.
More research is needed on these applications of UVGI, and specifically on whether they will reduce energy costs in buildings due to poor performance of air-handling units, whether they will reduce nosocomial infections and whether they can improve indoor air quality.
The potential exists to greatly increase the use of UVGI air disinfection systems. Applications include upper room systems in health-care facilities and settings with high occupancy, air-handling units in office buildings and homes, and transportation. The number of facilities currently using these technologies is thought to be a very small fraction of the total buildings and hospitals in the United States that could benefit.
Innovations are needed in lamp design and materials used in lamps and fixture design. There is a real need for UVGI lamps produced from nontoxic materials, improvements in LED efficiency at reduced cost, and more versatile UVGI fixtures including fixtures that are affordable for developing countries. These improvements could lead to better performance and increased application. In addition, technologies such as UVC coil cleaning systems must be scientifically tested so that the energy savings and indoor air quality benefits that have been shown to date could be realized in more buildings. UVGI is an important technology to be included in a multitiered approach to improving indoor air quality and promoting healthy buildings.
We are grateful for the helpful review and comments by Jaak Goebers.
Shelly L. Miller is an Associate Professor at the University of Colorado Boulder in the Mechanical Engineering Department and in the Environmental Engineering Program. She investigates indoor air quality, assesses exposures to air pollutants and develops and evaluates air pollution control measures. Dr. Miller's current research projects include understanding the role of ventilation systems in the transmission of infectious agents in buildings and intermodal transportation, engineering controls for reducing exposures to infectious diseases, studying ultraviolet germicidal coil cleaning technology, source apportionment of particulate matter and associated health effects, characterization of indoor air quality and the microbial communities in homes and investigating urban air quality issues including industrial odor episodes. Dr. Miller has received funding for her research program from the US EPA, CDC, NIOSH, NSF, NIH, AHRAE, HUD, and various private foundations and industry sponsors.
Jacqueline Linnes earned her PhD from the Department of Bioengineering at the University of Washington. Her thesis work focused on hospital-acquired infection caused by bacterial adhesion to medical devices. During this time she obtained a certificate in global health and cofounded company that develops low-cost indicators to monitor ultraviolet irradiation reaching water during solar disinfection in resource-limited settings. Following her graduate work, she was a Fogarty funded postdoctoral fellow at the Melvin First Center for Innovative Air Disinfection Technologies through a collaboration between the Division of Global Health Equity at Brigham and Women's Hospital and the Innovations in International Health (IIH) program at the Massachusetts Institute of Technology. This research focused on developing novel ultraviolet germicidal irradiation technologies for the prevention of tuberculosis transmission in resource-limited settings. She has continued this work as a postdoctoral researcher at the Harvard School of Public Health.
Julia Luongo received her B.S. in Engineering from Swarthmore College in 2010 and her M.S. in Mechanical Engineering from the University of Colorado at Boulder in 2012. She is currently in her third year of the Ph.D. program in Mechanical Engineering at the University of Colorado at Boulder. Her thesis work investigates the effect of ultraviolet germicidal irradiation on energy efficiency and biofouling of heat exchangers in HVAC systems. Her interests include indoor environmental quality, building ventilation systems, disease transmission and the links between these topics.