Future Disinfection of Touch Screens by Far-UVC-LEDs?—A Feasibility Study

Touch Screens are commonly contaminated with staphylococci and other pathogens and can therefore lead to the spread of infections both in the medical environment and in everyday life. Therefore, the feasibility study presented here aims to investigate whether touch screens could be disinfected automatically and quickly in the future by applying Far-UVC radiation, which is known for its strong antimicrobial impact despite its low hazard to humans. The study was carried out on staphylococci on a small quartz pane as a simplified touch screen model. In one setup, the 236 nm Far-UVC-LEDs radiation was coupled laterally into the quartz plate. For comparison, the staphylococci were also irradiated from the top by these Far-UVC-LEDs in another setup. In both approaches, the bacteria were reduced by about 90% within approximately 5 seconds. Hence, the approach with side-mounted Far-UVC-LEDs in particular appears to be a realistic possibility for the future automated disinfection of touch screens if the performance and lifetime of Far-UVC-LEDs continue to improve.


Future Disinfection of Touch Screens by Far-UVC-LEDs?-A Feasibility Study
Ben Sicks , Oksana Gurow, and Martin Hessling Abstract-Touch Screens are commonly contaminated with staphylococci and other pathogens and can therefore lead to the spread of infections both in the medical environment and in everyday life.Therefore, the feasibility study presented here aims to investigate whether touch screens could be disinfected automatically and quickly in the future by applying Far-UVC radiation, which is known for its strong antimicrobial impact despite its low hazard to humans.The study was carried out on staphylococci on a small quartz pane as a simplified touch screen model.In one setup, the 236 nm Far-UVC-LEDs radiation was coupled laterally into the quartz plate.For comparison, the staphylococci were also irradiated from the top by these Far-UVC-LEDs in another setup.In both approaches, the bacteria were reduced by about 90% within approximately 5 seconds.Hence, the approach with sidemounted Far-UVC-LEDs in particular appears to be a realistic possibility for the future automated disinfection of touch screens if the performance and lifetime of Far-UVC-LEDs continue to improve.

I. INTRODUCTION
W ITH the global trend towards digitalization, the appli- cation of touch screens is also increasing in all areas of life [1].This includes everyday situations such as public ticket machines, but also in medical facilities like hospitals [2] and is of particular importance here, as microorganisms can be transmitted through contact between screen and fingers, potentially leading to the spread of nosocomial infections.In fact, in previous studies, no touch screen was found to be free of bacteria, even in the medical environment.At least potentially pathogenic staphylococci were always detected [3].
In principle, it would be possible to treat each screen with chemical disinfectants after every single use, but this would involve a great deal of effort, so this practice is not common.
An earlier study investigated the possibility of automatic touch screen disinfection applying UVA radiation or visible violet light [4].These wavelengths were selected so that the radiation could not endanger the human operator.The experiments were carried out on glass plates contaminated with staphylococci.The radiation from the LEDs was coupled into the pane from the side to minimize exposure to the user.Actually, a bacterial reduction by more than two orders of magnitude was achieved.However, the time required for a 90% disinfection was at least one hour, which would pose practical limitations in potential future applications.
Another, recently much-discussed UV spectral range for disinfection applications is the so-called Far-UVC (200-240 nm).On the one hand, this radiation can very effectively destroy the DNA of bacteria, fungi and viruses and thus provides a strong antimicrobial effect.On the other hand, human cells are usually well protected because this radiation is strongly absorbed by proteins, and so, for example, the outer protein-rich dead skin cells in the stratum corneum protect deeper-lying vital cells [5], [6], [7], [8], [9], [10], [11], [12], [13].
Therefore, the study presented here aims to investigate the extent to which Far-UVC would be suitable for automatic touch screen disinfection.Unfortunately, the choice of Far-UVC sources is currently limited.In addition to krypton chloride excimer lamps with a peak emission at 222 nm, which are too bulky to be installed in small devices, there are only a few commercially available Far-UVC-LEDs in the spectral range 230-240 nm with limited radiation powers in the order of only one or a few milliwatts and lifetimes in the order of a few hundred hours.
It is expected that the properties of Far-UVC-LEDs will improve in terms of performance, lifetime and peak wavelength in the future.However, due to the currently low LED outputs, the application of Far-UVC-LEDs for automatic touch screens is only being investigated for a small quartz plate here, with the quartz plate acting as a highly simplified touch screen front window / touch screen model.The irradiation tests are carried out with a staphylococcus strain, as staphylococci are the most common form of touch screen contaminations [3].
Different irradiation approaches are carried out to investigate the impact of Far-UVC radiation on bacterially contaminated quartz plates.Firstly, Far-UVC-LEDs are attached to the side of the quartz plate, which then acts as a light guide.Due to the evanescent field or lack of total reflection in the concept of geometrical optics, bacteria are irradiated over the whole quartz surface without emitting radiation in the direction of the operator.For comparison, the staphylococci are also irradiated conventionally from the top by these LEDs to assess the efficiency of the light guide approach.

A. Far-UVC-LEDs and Experimental Setups
The first irradiation setup was based on four 236 nm LEDs type SF1 (flat lens) from Silanna UV (Pinkenba, Australia).Their emission spectrum is illustrated in Fig. 1.The total radiant power at 40 mA was about 1.6 mW per LED, determined with a calibrated photo spectrometer type CAS140D of Instrument Systems (Munich, Germany).The emission angle of the LEDs was 128 • .
To implement the first irradiation approach, four of these LEDs were installed at the sides of a quartz pane measuring 50 × 50 × 4 mm 3 (Fig. 2A), the edges of which were not polished.The LEDs were not positioned in the middle of the edges, but slightly shifted to achieve a greater homogeneity and avoid a hotspot in the center of the quartz plate.This quartz plate acted as a light guide with a surface from which the light was reflected, unless the reflection was prevented by microbial contamination, such as bacteria, for example.In the perception of geometrical optics, the light then passed through the bacteria as indicated in Fig 2A .The second setup is schematically presented in Fig. 2B, where the quartz plate was conventionally irradiated from the top with the microorganisms on its surface.The irradiation was performed by four of the above-described 236 nm Far-UVC-LEDs at a distance of 2 cm.The irradiance in the center of the quartz plane, as determined with the CAS140D spectrometer, was 216 µW/cm 2 .

B. Test Bacteria and Irradiation Analysis
The selected test organism was Staphylococcus carnosus (S. carnosus), a non-pathogenic staphylococcus strain that exhibits a similar Far-UVC sensitivity as the pathogenic Staphylococcus aureus (S. aureus) [14].S. carnosus was first propagated in tryptic soy extract medium [15] to a concentration of about 10 8 bacteria/ml, then washed several times in phosphate buffered saline (PBS) and concentrated to 10 8 bacteria per ml once again.
A home-built spraying device homogeneously contaminated the surface of the quartz plate with the bacterial suspension to be irradiated and an another quartz plate that was not irradiated and served as control.The number of living/surviving bacteria on the irradiated quartz plate was determined after 5, 10, 15, 20, 30 and 60 seconds for the different irradiation setups.Additionally, the bacterial concentrations on the unirradiated control quartz plate were determined at the start and end of irradiation.
This microbial assessment was achieved with the aid of eSwabs from Copan (Brescia, Italy).According to the sampling procedure reported by Madsen et al. [16], an area of approx.25 × 25 mm 2 was sampled on the quartz plates using the swab provided, transferred into a destined isotonic solution.The bacteria were then spread on agar plates at various dilution levels.After one day in a 37 • C incubator, the surviving bacteria could be determined on the basis of the visible grown colonies.Before irradiation, the quartz plate was divided into 4 sections (as hinted in Fig. 2A) and two opposite quadrants were sampled at each point in time.To compensate for the limited number of sample areas, the experiments were repeated many times with different irradiation periods.At least 3 independent experimental runs were carried out for each irradiation duration.

III. RESULTS
The observed average bacterial reduction is given in a semi-logarithmic representation in Table I and Fig. 3.In all setups, the Far-UVC radiation led to a strong bacterial reduction of about 90% in about 5 seconds and almost three orders of magnitude in 15 seconds, while the unirradiated control revealed no significant decrease in the number of staphylococci.
However, after the initial phase, where the reductions appear to follow exponential curves -straight lines in the half-logarithmic representation in Fig. 3 -the bacterial reduction slows down with higher irradiation times and doses, resulting in only marginal further reductions.
Based on the observed reductions after 10 s for 236 nm top surface irradiation and the determined irradiance values, an average log reduction dose of 0.9 mJ/cm 2 was calculated for Staphylococcus carnosus.

IV. DISCUSSION AND CONCLUSION
The most important result is that rapid automatic disinfection of small touch screens using Far-UVC by lateral coupling of the radiation is technically already conceivable even with the low power of today's Far-UVC-LEDs, as this approach is surprisingly efficient.The total output power of the four 236 nm LEDs of only 6.4 mW was distributed over an area of 25 cm 2 of the quartz plate, which formally corresponded to 256 µW/cm 2 .The disinfection effect achieved was approximately the same or even better as with the 236 nm top irradiation with real irradiance of 216 µW/cm 2 , respectively.This observed high efficacy is consistent with the results of our previous investigation on laterally coupled UVA radiation [4].However, it is worth noting that the current peak wavelength of the Far-UVC-LEDs at 236 nm is not optimal.The ACGIH 236 nm threshold limit values (TLVs) for skin irradiation are approximately 71.5 mJ/cm 2 compared to 479 mJ/cm 2 at 222 nm [17].For the aforementioned 236 nm irradiance of 256 µW/cm 2 , the user should currently only touch the screen for a maximum of about 280 seconds if it does not switch off on contact.At 222 nm this period would extend to over 30 minutes.
The TLVs for eyes are even lower, but as the eye does not touch the screen and there is practically no irradiation without touching -the total irradiance measured at a distance of only 2 cm is already below 9 µW/cm 2 -therefore, the eye is not at risk.
Although the present experiments were carried out only with non-pathogenic staphylococci, S. carnosus has a similar Far-UVC sensitivity to pathogenic S. aureus, including methicillin-resistant S. aureus (MRSA) and other important pathogens such as Pseudomonas aeruginosa.Therefore, this disinfection performance is not limited to (non-pathogenic) staphylococci.
The fast disinfection times, which are almost a factor of 1000 shorter than in the previously conducted UVA investigation, suggest that in the future, fast, automatic touch screen disinfection using laterally coupled Far-UVC-LED radiation is also realistic for large screens if the performance and lifetime of these LEDs continues to increase.

Fig. 1 .
Fig.1.Far-UVC-LED emission spectrum together with absorption spectra of DNA and protein (the absorption spectra were taken from previous measurements[8]).

Fig. 2 .
Fig. 2. Irradiation of bacterially contaminated quartz plates from the side via the light guide effect (A) and from the top (B), by four 236 nm LEDs.The dotted gray lines in (A) divide the quartz plate in four sample areas.

Fig. 3 .
Fig. 3. Change in bacterial concentrations for irradiated and non-irradiated quartz plates over time for both irradiation approaches in a semi-logarithmic representation; the upper X-axis indicates the average dose for irradiation from the top; the error bars indicate the standard deviation of the individual experiments.

TABLE I OBSERVED
BACTERIAL REDUCTIONS FOR IRRADIATED SAMPLES AND UNIRRADIATED CONTROLS GIVEN AS LOG-REDUCTION FOR DIFFERENT IRRADIATION DURATIONS