Novel non intrusive continuous use ZeBox technology to trap and kill airborne microbes

Preventing nosocomial infection is a major unmet need of our times. Existing air decontamination technologies suffer from demerits such as toxicity of exposure, species specificity, noxious gas emission, environment-dependent performance and high power consumption. Here, we present a novel technology called “ZeBox” that transcends the conventional limitations and achieves high microbicidal efficiency. In ZeBox, a non-ionizing electric field extracts naturally charged microbes from flowing air and deposits them on engineered microbicidal surfaces. The surface’s three dimensional topography traps the microbes long enough for them to be inactivated. The electric field and chemical surfaces synergistically achieve rapid inactivation of a broad spectrum of microbes. ZeBox achieved near complete kill of airborne microbes in challenge tests (5–9 log reduction) and \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$>90\%$$\end{document}>90% efficiency in a fully functional stem cell research facility in the presence of humans. Thus, ZeBox fulfills the dire need for a real-time, continuous, safe, trap-and-kill air decontamination technology.

Electric field extracts charged microbes from the flow. Microbes are naturally charged 37,38 ; therefore, in an electric field, they are impelled towards the electrode of opposite polarity. Figure 1 depicts this process schematically. Here, X-axis points along the flow and Z-axis points away from the attracting electrode. A microbe initially at distance z 0 from the attracting electrode travels a distance R in the streamwise direction, called its "range", as it descends to z = 0 . Whether or not the microbe hits the electrode depends on its length, the microbe's initial distance z 0 , strength of the electric field, charge on the microbe and the type of flow (laminar or turbulent). The Reynolds number for the flow between electrodes in ZeBox is ∼ 10 3 and a rectangular duct flow (or even plane Poiseuille flow) undergoes transition at this Reynolds number and could be turbulent 39,40 . Analyzing microbe's motion in a turbulent flow is difficult because of its complicated, stochastic nature. Supplementary information S1 analyzes microbe's motion and its maximum range in a laminar flow instead. The settling speed is obtained by equating electrostatic and drag forces on the microbe, while also accounting for its changing streamwise speed as it settles (the steady laminar velocity profile of the background flow being known); the result of the analysis is a universal dimensionless curve for microbe's range, refer Supplementary Fig. 1, from which the efficiency of ZeBox may also be computed given its operating parameters.
Earlier studies on resuspension of dust from flat surfaces due to a flow show that, whenever the hydrodynamic force and torque exerted by the flow exceed those that keep the particles attached to the surface (for example, Van der Waals force), the particles can either detach by lifting off or slide and roll on the surface 41,42 . In our case, lifting off of microbes from the electrode is unlikely due to the strong electric field, but they can nevertheless slide and roll and thus escape away due to the electrode's finite length (refer Fig. 2). Since the microbicidal surface requires a minimum duration of contact to inactivate microbes depending on how sensitive or hardy it is, a fraction of the deposited microbes could escape while still viable. Therefore, the ability of the microbicidal surface to trap and hold microbes until they are inactivated becomes important.   www.nature.com/scientificreports/ Three dimensional topography of the microbicidal surface traps the microbe. The microbicidal surface employed in ZeBox has a highly uneven topography at the microbial scale, populated with well-like depressions to trap and hold microbes. Figure 3a, b show the scanning electron microscope (SEM) images of the surface at different magnifications appropriate to the microbial scale. Figure 3c shows streamlines in a numerically simulated two dimensional flow (using OpenFOAM-7) over a surface with square shaped wells, to qualitatively illustrate the kind of flow obtained over an uneven topography. Any bulk flow may be approximated as simple shear flow sufficiently close to a solid surface. A simple shear flow is characterized entirely by its shear rate, estimated as U/H for our case; U ≈ 1 cm/s is the average flow speed between electrode plates and H = 1 cm is the gap between them. The flow Reynolds number based on shear rate and characteristic dimension of the square well, d, is Re ≡ (U/H)d 2 /ν , where ν = 1.5 × 10 −5 m 2 /s is the kinematic viscosity of air. From Fig. 3b, d ∼ 10 µ m, which yields Re ∼ 10 −5 . A simple shear flow was imposed on the flow domain (refer Fig. 3c) by moving its uppermost boundary horizontally at constant speed to achieve the aforementioned Reynolds number. The important feature of the flow for our purpose is the recirculating region set up within the wells, in which the streamlines of the flow form closed loops. This feature is quite general for a flow over an uneven topography and which presumably enhances the efficacy of the microbicidal surface further in regard to trapping microbes. Once the microbe falls into one of the wells, brought there either in the course of its rolling over the surface or directly by the electric field, the recirculating flow can confine it to the well for a sufficiently long duration. Table 1 shows the efficacy of microbicidal surfaces (in terms of log 10 reduction, where n-log 10 reduction implies reduction in the initial microbial load by a factor of 10 n ) with different topographies, which we call 2-D and 3-D surfaces, in flow experiments. A 2-D surface is a single layer of cotton fabric while a 3-D surface is a multilayered 90:10 polyethylene : cotton fabric. In the presence of electric field, 3-D microbicidal surface performs better than the 2-D surface. When the electric field is absent, the microbes are not extracted from the flow and hence both surfaces perform similarly.
Electric field and chemical microbicidal-surfaces synergistically achieve rapid inactivation of microbes. In contrast to electrostatic precipitators, the applied electric field in ZeBox plays two roles: it pulls microbes from the flow on to the microbicidal surface and then accelerates their subsequent inactivation. Table 2 shows log 10 -reduction in the microbial load in spot experiments, with 3 kV/cm electric field applied between electrodes. The microbicidal surface achieves the highest reduction in microbial load in the presence of the electric field. Quaternary ammonium compounds (QAC) are membrane-active agents which inactivate microbes by targeting their cytoplasmic membrane [43][44][45][46] , but first, they must breach the outer cell wall. In the present design, QAC is tethered to the 3-D surface by long flexible chains, which presumably helps the QAC to orient itself to puncture holes in the microbe. The external electric field increases the trans-membrane voltage of the cell above its resting value, leading to an electric current that presumably flows through these pores as they form the path of least resistance. This current flow may be analogous to the electroporation of bacteria in which the pores formed in the cell wall are stabilized 47 . The intracellular components then leak from the pores, as is seen in the SEM pictures. This process leads to the irreversible killing of the cells. Therefore, the chemical surface in tandem with the electric field displays an enhanced electro-chemical microbicidal action compared to what they would have achieved separately.   49,50 . Figure 4 shows the collated data on the variation in log 10 microbial load (n-log 10 microbial load equals 10 n microbes) over time after ZeBox was turned on. ZeBox proves to be extremely effective in rapidly decreasing the viable microbial load in a closed space. It achieved 9.9 log 10 -reduction (i.e. 99.999999999% reduction) of E. coli in 10 min (n log 10 -reduction equals reduction by a factor of 10 n ). For other microbes ZeBox brought about 5 to 9 log 10 -reduction (i.e. 99.999-99.9999999% reduction) of the viable microbial load.
SEM images of microbicidal action. Scanning electron microscopy (SEM) studies were done to see how microbes trapped on the microbicidal surface are killed. E. coli and A. fumigatus spores were chosen because they form two extremes on the scale of sensitivity, with spores being hardy. Figure 5a,e show the microbes in control conditions. Due to electro-chemical action at the three dimensional microbicidal surface, their cell membrane undergoes morphological changes followed by complete degradation. Figure 5b,c, obtained after 5 min of contact, reveals puncturing and blebbing of the E. coli cell membrane. Ultimately, the cells burst and their intracellular contents spill out (Fig. 5d,f) signaling a complete degradation of the microbes.

ZeBox reduces microbial load in open room.
ZeBox's performance was also tested in a real life setting, i.e. in a room with constant influx of microbes from outside or due to internal sources. A working tissue culture laboratory in a building with central air-conditioning, but without High Efficiency Particulate Air (HEPA) filters, was chosen for the purpose. Figure 6a shows the schematic plan-view of the lab and the measurement locations. The working people in the lab were the primary source of microbial contamination. Figure 6b shows that the microbial load at location-03 where tissue culture work was carried out was >3000 CFU/m 3 initially. ZeBox  www.nature.com/scientificreports/ reduced the microbial load in the lab to ∼ 10 CFU/m 3 within about 3 h after it was turned on. This low level was consistently maintained so long as ZeBox was operational. When it was turned off at day 10, the microbial load rebounded to its original level. During its operation, ZeBox effectively decontaminated a zone of dimensions ∼ 10 feet × 10 feet (refer Fig. 6a), which demonstrates its potential to decontaminate a smaller region of interest in a relatively large open room, with uncontrolled movement of personnel and without needing physical partitions.
ZeBox does not produce ozone. Since ZeBox employs non-ionizing electric field, it does not produce ozone (verified in standardized laboratory tests, data not shown here). This is an immense advantage over conventional microbicidal technologies such as plasma and PCO. Also, it consumes <20 Watt-hour of electric energy during its operation.

Discussion and conclusions
ZeBox technology exploits the fact that microbes (bacteria, viruses, spores and fungi) are naturally charged and therefore can be readily manipulated by an electric field. Using a non-ionizing electric field, microbicidal surfaces with three dimensional topography and electro-chemical kill mechanism, ZeBox achieves significantly higher microbicidal rate compared to other technologies.  www.nature.com/scientificreports/ Knowing the total reduction in microbial load, as shown in Fig. 4, is inadequate to gauge ZeBox's efficacy because any level of decontamination may be achieved given sufficient time. Therefore, an overall microbicidal efficiency must be determined while factoring in the time of operation as well as the volume of the room being decontaminated. Towards this end, we may think in terms of the number of nominal air changes in a room achieved in a given duration and the consequent reduction in microbial load for each air change. In time t, Qt/V number of nominal air changes is achieved, where Q is the air flow rate through ZeBox and V is the volume of the room. If η is the corresponding microbicidal efficiency, then N 0 initial number of viable microbes in the room decreases to N = N 0 (1 − η) Qt/V after time t. Using this formula and the latest-time data from Fig. 4 whose ordinate is log 10 N , we may back-calculate η for a specified time duration. For experiments with viruses, Q/V ≈ 1.5 air changes per minute inside the test chamber, which implies 7.5 air changes during the duration of the experiment, refer Fig. 4b. If, for example, we consider PhiX virus then log 10 N = 0, log 10 N 0 ≈ 6 and Qt/V = 7.5 , which gives η = 1 − 10 −6/7.5 = 84 % . For the microbes in Fig. 4a, the test chamber was ∼ 5 times larger (refer "Materials and methods" section), hence the air change rate was lower by the same factor. The microbicidal efficiency of ZeBox lies in the range of 83-99 % for all the tests. Considering the variety of sensitive and hardy microbes employed, ZeBox is about equally effective against all of them. Supplementary information S2 provides a theoretical estimation of the microbicidal efficiency of ZeBox. To estimate the efficiency using the theory provided in Supplementary information S1, the charge on the microbes must be deduced; towards this end, we measured their zeta potential and used the Debye-Hückel theory which governs the distribution of electric potential around a charged particle 54 , in order to relate the microbe's zeta potential to its charge (refer Supplementary equation (8)). The resulting theoretical estimate of Zebox's efficiency aligns reasonably well with that deduced from experimental data.
Airborne microbes of size < 2 µ m can remain suspended in air for several hours before settling down and therefore must be inactivated to reduce the transmission of infections. ZeBox technology presents a universal solution because: • Freely floating microbes are trapped and killed with high efficiency, eliminating the possibility of future growth. • The airflow is parallel to antimicrobial surfaces with almost no resistance; therefore, unlike HEPA filters, it has low energy utilization. • There are no chemical emissions or production of free radicals or ozone; the technology is safe for continuous use in the presence of humans and animals. • It is equally effective for different varieties of sensitive and hardy microbes.

Materials and methods
Challenge tests. Test setup. An air-sealed test chamber of dimensions 3 ft × 4 ft × 3 ft (approximately 1000 liters in volume) was built with multiple sampling and nebulization ports. The environmental parameters such as relative humidity and temperature could be monitored using a probe located inside the chamber. During experiments, various microorganisms were aerosolized using a 6-jet collision nebulizer (MESA LABS, BGI) into the chamber, and the device efficiency was monitored by collecting and measuring microbial concentration at different time intervals. A second test chamber of dimensions 3ft × 2.5 ft × 1 ft (approximately 220 liters in volume) placed inside a biosafety cabinet, with similar aerosolization and sampling port configuration, was used for tests with viruses. Aerosolization of test microbes. A 6-jet Collison nebulizer (MESA LABS, BGI) was used to aerosolize the test microbes into the test chamber. Dry air from a compressed air cylinder at a pressure of 10 psi was used to operate the nebulizer. The nebulizer produces bioaerosols of a 2-5 µ m diameter that allows them to float in the air present in the test chamber for a definite period. The length of the nebulization period varied depending on the type of experiment and microorganism, but typically ranged between 30-40 min.
Sampling of air for viable microbes. The airborne survival of the test microbe and the activity of the air decontamination devices were determined by collecting the air from the chamber at the rate of 12.5 liter/min using SKC biosampler 51 , filled with sterile buffer ( 1× Phosphate buffer saline, pH 7.2). Collected samples were analyzed to understand the quantity of viable microorganism present by diluting and plating them onto suitable growth media. The plated samples were incubated at 37 ± 2 • C for bacteria and 25 ± 2 • C for fungal species and allowed Spot experiments. E. coli cells were grown in the standard medium. A known titre of cells were spotted onto a 25 mm 2 surface and incubated for various time duration, both with and without electric field. Surfaces were resuspended in 500 µ l of sterile 1X PBS, which was then plated on standard agar plates to enumerate the microbes.
Limit of detection. Microbial enumeration is guided by two parameters, Limit of Detection (LOD) and Limit of Quantification (LOQ). For the present assays used to quantify the microbial load inside the test chamber, the LOD was 10 CFU for bacterial and fungal load and 5 PFU for viral load. However, LOD is always less than LOQ 53 . In many of our experimental analysis, post operating ZeBox device, the microbial detected numbers were in around LOD and hence, the exact LOQ was often indeterminant. Field tests. Air sample collection. A working tissue culture laboratory in a national stem cell research facility was chosen for study. This laboratory was situated in a building which had central airconditioning but the absence of a HEPA-enabled air handling unit resulted in frequent contamination of tissue culture samples. A handheld air sampler (SAS Super 100) was used, which could sample 100 liters of air per minute. Tryptic Soy Agar and Sabouraud dextrose agar plates were used to sample bacteria and fungi, respectively from the air. A fixed volume of air was sampled using the bio-sampler. Plates were placed in and removed from the bio-sampler in an aseptic manner. Plates were incubated at 25 ± 2 • C (for fungal cultivation) and 37 ± 2 • C (for bacterial cultivation) for 48 h. Post-incubation, the number of colonies appeared were enumerated and converted to CFU/ m 3 using statistical conversion provided by the manufacturer. Control plates were used to ensure the sterility of the entire process.