Background
Indoor air quality (IAQ) has a significant influence on health, comfort and well-being of building occupants. It has been demonstrated that poor IAQ could jeopardize health and well-being, which in turn will affect the quality of work and ultimately lower the productivity of workers [
1].
One major source of indoor air pollution is the presence of micro-organisms, which could cause even more serious problems than some organic and inorganic air contaminants. This is particularly more phenomenal in cases of inadequate ventilation, as the condensation in ventilation system can act as a breeding ground for harmful bacteria which are dispensed through the ventilation ducts. Environmental airborne bacteria such as
Pseudomonas aeruginosa,
Streptomyces albus,
Bacillus subtilis and complex populations of micro-organisms within normal flora were all etiological agents to hypersensitivity pulmonary diseases. Several additional infectious agents such as
Legionella pneumophila and
Mycobacterium tuberculosis (MTB) pose even more grave concerns to the IAQ, as these airborne pathogenic bacteria are known to cause severe illness in humans. Meanwhile, viruses such as influenza virus were originally thought to be only transmitted from person to person via aerosols of body fluids. However, in a recent study conducted by Weistein et al. [
2], the production of infectious droplet nuclei of diameter < 5 μm could remain suspended and disseminated by air current to infect a susceptible host. A good and reliable disinfection system, therefore, is required to disinfect the airborne microorganisms in order to maintain good IAQ.
Adopting vacuum-UV (VUV) lamps, for instance, the ozone producing low-pressure Hg vapor lamps, can be an effective mean of disinfecting the airborne microorganisms. Many existing infection control products use low pressure mercury vapor lamps as light source. This is a source of high energy photons with low cost. Recently, pulsed xenon light source technology emitting a broad spectrum (200-300 nm) of UV light is an emerging alternative to low pressure mercury vapor lamps that allows much faster surface disinfection because of the high peak power [
3]. Nevertheless, the pulsed nature of this technology would limit its use in continuous air disinfection system. Electrical discharge of low pressure mercury vapor mainly emits 254 nm ultraviolet light C (UVC) and 185 nm VUV light. However, existing products mainly use the lamps with doped quartz envelope that absorbs 185 nm photons to prevent the formation of potentially dangerous ozone. Nevertheless, ozone is also a powerful disinfectant and the valuable disinfection opportunity of the 185 nm VUV light becomes waste heat.
Ozone is an issue that bothers on safety if it remains in the output of an air treatment system. However, ozone can be easily destroyed before leaving the air treatment system if proper catalyst is adopted [
4,
5]. Also, some photocatalysts can utilize and destroy ozone in addition to its photocatalytic activity [
6].
The 254 nm UVC light adopted in conventional infection control products can disinfect the illuminated objects since the 254 nm radiation can disrupt the genetic materials of airborne pathogens and render them inviable [
7].VUV has an even stronger ionizing power than UVC light and can generate high concentration reactive species such as ozone and OH radicals [
7]. In other words, apart from direct illumination, VUV can inactivate bacterial growth by the radicals generated during VUV irradiation. Therefore, adopting VUV lamps can enhance the air disinfection capability of air cleaning systems. A previous study [
4] conducted by Huang et al. demonstrated that 64% toluene removal with VUV irradiation alone and the use of photocatalyst enhanced the toluene removal from 64 to 82%. The experiment adopting UVC lamps and the use of photocatalyst removed only 14% of toluene. The result demonstrated that VUV light could be an effective measure for chemical degradation in ventilation systems. When it comes to disinfection, extensive research has been carried out on UVC light and effective destruction of both airborne [
8‐
20] and other human pathogens [
21‐
29] has been shown. Nevertheless, disinfection using VUV light has attracted very little attention. This would be caused by the relative low prevalence of VUV light sources. Kim et al. [
30] found that the disinfection time required to attain the same extent of inactivation of aerosolized MS2 bacteriophage, using low pressure mercury vapor lamps with both 254 nm UVC and 185 nm VUV output was much shorter than the lamps with 254 nm UVC only. The disinfection time of ozone only (without UV) process at ozone concentrations equivalent to the ozone level generated by the mercury vapor lamps was also significantly faster than using lamps with 254 nm emission only. Besides, Huang et al. [
4] reported the inactivation of
E coli by low pressure mercury vapor lamps. Additionally, some researchers tested the disinfection of water with VUV light and it was reported that the efficiency was quite low compared to disinfection with UVC light [
31,
32]. The reason is due to the low penetration power of VUV light in water [
33]. Moreover, the disinfection of human pathogens by VUV light was rarely reported. In our opinion, only Christofi et al. [
34] reported the disinfection of the microbial films of 3 types of pathogenic bacteria using ozone producing low-pressure Hg vapor lamps. Therefore, the effect of VUV light against human pathogens is yet to be elucidated. In this study, we evaluated the germicidal effect of VUV light on common bacteria including
Escherichia coli ATCC25922 (
E. coli), Extended Spectrum Beta-Lactamase-producing
E. coli (ESBL), Methicillin-resistant
Staphylococcus aureus (MRSA) and
Mycobacterium tuberculosis (MTB), and that on influenza viruses H1N1 and H3N2. Influenza viruses and MTB are inherent airborne pathogens while
E. coli ATCC25922 is always the first indicator organism to monitor disinfection efficacy. The more drug resistant ESBL and MRSA were chosen as examples to monitor disinfection efficacy on human pathogens. Some suspensions of these bacteria and viruses were absorbed into nitrocellulose filter papers during the experiments and the disinfection under the environment with a moderate barrier to light was evaluated.
Discussion
High-energy vacuum-UV light is efficient in disinfection. Similar to other UV disinfection mechanisms, direct illumination of VUV could result in the formation of new bonds between adjacent nucleotides, causing photochemical damage on DNA strands and eventually inactivating the replication of microorganisms.
In addition, the high-energy VUV could also lead to the formation of both OH radicals and O3, which diffuse into anywhere that is shielded from direct UV irradiation and inhibit the growth of microorganism. This explained the excellent bactericidal efficiency of VUV light disinfection even in the presence of the opaque nitrocellulose filter. Our result has further revealed the potential of VUV light to provide a thorough disinfection, even for dust particles and large aerosols contaminated with pathogens where direct UV illumination cannot penetrate.
In this study, we demonstrated that VUV light disinfection is effective against Escherichia coli, Extended Spectrum Beta-Lactamase-producing E. coli and Methicillin-resistant Staphylococcus aureus. For the best tested situation, with the criterion of 3-log10 inactivation of bacteria, valid germicidal result can be achieved with ≤10 min of VUV treatment. Additionally, more than 5-log10 reduction in viable plate count can be attained below 15 min of disinfection.
In the disinfection tests against seasonal influenza viruses H1N1 and H3N2, we also demonstrated that viral load could be effectively reduced by 4-log10 folds after 20 min VUV illumination and this also satisfied the criterion of valid germicidal result. Additionally, more than 3-log10 reduction in viral load can be attained with < 10 min of treatment.
Mycobacterium tuberculosis, on the other hand, required a more intense disinfection.
At 20 min disinfection, VUV light disinfection could only result in a 3-log10 reduction in viable plate count. This is insufficient according to our 5-log10 reduction criterion for mycobacterial disinfection. It was only after 30 min of disinfection that the required 5-log10 reduction of
Mycobacterium tuberculosis viable bacterial load could be achieved regardless of the bacterial concentration. This is concordant to previous studies [
19,
35,
36] where mycobacterial species were generally more resistant to UV disinfection. This is probably accounted by the thicker lipid cell wall in
Mycobacterium species.
The tested variations in concentrations of bacteria did not manifest a trend in the rate of inactivation. For E. coli and ESBL, higher bacterial concentration resulted in lower rates of inactivation. Experiments with MTB showed a different trend. Nevertheless, no obvious trend was showed in the experiments with MRSA.
From literature, various research teams reported the UV dosages required attaining 99.9% (3-log) inactivation of various bacteria or viruses under light from low pressure mercury vapour lamps. For example, the UV dosages in mJ/cm
2 for 3-log inactivation of T7 phage,
E coli.,
Staphylococcus aureus,
Mycobacterium avium and
Mycobacterium phlei are 10 [
37], 5 [
37], 9 [
34], 18 [
20] and 158 [
34], respectively. Most of their experiments were conducted with bacteria and viruses virtually unprotected. In our experiment, attaining 3-log inactivation typically required 10 min. Considering that our equipment provided 21 and 2.3 mW/cm
2 light power at 254 nm and 185 nm, and the total UV power is ~ 23 mW/cm
2. The UV dosage of 10 min illumination is ~ 14,000 mJ/cm
2, far higher than the usual values. This could be the consequence of our testing condition created by loading the suspended bacteria or viruses onto nitrocellulose filter paper. Some bacteria were actually protected from direct UV light by the shading effect of filter paper which is different from the testing setup in the literature.
In order to provide sufficient disinfection against all the microorganisms we included in this study, we suggested the use of Mycobacterium reduction as a benchmark test for future disinfection instrument designs that incorporates the VUV light system.
Although, the disinfection under the environment with a moderate barrier to light was successful, there are limitations in the present study. The current pilot study on the disinfection efficacy of VUV light disinfection was conducted in laboratory-controlled conditions. For example, due to safety consideration, device type testing on aerosolized bacteria and viruses is not possible. All bacterial and viral inoculums were prepared in liquid suspension and illuminated by VUV on a Petri dish, which differed from actual environmental settings.
Conclusion
Airborne pathogens are important indoor air quality concerns. A good and reliable disinfection system is a must to maintain good indoor air quality. Vacuum-UV lamps with ozone production were found to be effective for inactivating various human pathogens. With the best tested situation, 3-log10 inactivation of Escherichia coli, Extended Spectrum Beta-Lactamase-producing E. coli, Methicillin-resistant Staphylococcus aureus and seasonal influenza viruses can be achieved with ≤10 min of VUV treatment except Mycobacterium tuberculosis which needed about 20 min. This demonstrated the high resistance against UV disinfection of MTB. Valid germicidal results, reflected with 3-log10 inactivation for bacteria, 4-log10 inactivation for viruses and 5-log10 inactivation for MTB, can be obtained with all tested pathogens. The duration of VUV treatment required for valid germicidal result of most of the bacteria was ≤10 min while MTB needed about 30 min. 20 min was adequate for the influenza viruses. This indicated that VUV light is an effective approach against different environmental and pathogenic microorganisms, and can potentially be used for air-purifying units in future ventilation systems.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.