Journal of Hazardous Materials 351 (2018) 29–37 Contents lists available at ScienceDirect Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat Washable antimicrobial polyester/aluminum air filter with a high capture efficiency and low pressure drop T Dong Yun Choia,1, Ki Joon Heob,c,1, Juhee Kangb, Eun Jeong Ana, Soo-Ho Junga, Byung Uk Leec, ⁎⁎ ⁎ Hye Moon Leea,d, , Jae Hee Jungb,e,f, a Powder and Ceramics Division, Korea Institute of Materials and Science, Changwondaero 797, Seongsan-gu, Changwon, 51508, Republic of Korea Center for Environment, Health, and Welfare Research, Korea Institute of Science and Technology (KIST), Hwarang-ro 14-gil 5, Seongbuk-gu, Seoul, 02792, Republic of Korea c Aerosol and Bioengineering Laboratory, Department of Engineering, Konkuk University, Neungdong-ro 120, Gwangjin-gu, Seoul, 05029, Republic of Korea d Alink Co. Ltd., Chanwondaero 797, Seongsan-gu, Changwon, 51508, Republic of Korea e Green School, Korea University, Anam-ro 145, Seongbuk-gu, Seoul, 02841, Republic of Korea f Division of Energy & Environment Technology, KIST School, Korea University of Science and Technology, Hwarang-ro 14-gil 5, Seongbuk-gu, Seoul, 02792, Republic of Korea b G RA P H I C A L AB S T R A C T A R T I C L E I N F O A B S T R A C T Keywords: Air filter Antimicrobial filter Bioaerosol Conductive fiber Particulate matter Here, we introduce a reusable bifunctional polyester/aluminum (PET/Al) air filter for the high efficiency simultaneous capture and inactivation of airborne microorganisms. Both bacteria of Escherichia coli and Staphylococcus epidermidis were collected on the PET/Al filter with a high efficiency rate (∼99.99%) via the electrostatic interactions between the charged bacteria and fibers without sacrificing pressure drop. The PET/Al filter experienced a pressure drop approximately 10 times lower per thickness compared with a commercial high-efficiency particulate air filter. As the Al nanograins grew on the fibers, the antimicrobial activity against airborne E. coli and S. epidermidis improved to ∼94.8% and ∼96.9%, respectively, due to the reinforced hydrophobicity and surface roughness of the filter. Moreover, the capture and antimicrobial performances were ⁎ Corresponding author at: Center for Environment, Health, and Welfare Research, Korea Institute of Science and Technology (KIST), Hwarang-ro 14-gil 5, Seongbuk-gu, Seoul, 02792, Republic of Korea. ⁎⁎ Corresponding author at: Powder and Ceramics Division, Korea Institute of Materials and Science, Changwondaero 797, Seongsan-gu, Changwon, 51508, Republic of Korea. E-mail addresses: hyelee@kims.re.kr (H.M. Lee), jaehee@kist.re.kr (J.H. Jung). 1 These authors contributed equally to this work. https://doi.org/10.1016/j.jhazmat.2018.02.043 Received 13 October 2017; Received in revised form 8 February 2018; Accepted 22 February 2018 Available online 24 February 2018 0304-3894/ © 2018 Elsevier B.V. All rights reserved. Journal of Hazardous Materials 351 (2018) 29–37 D.Y. Choi et al. stably maintained during a cyclic washing test of the PET/Al filter, indicative of its reusability. The PET/Al filter shows great potential for use in energy-efficient bioaerosol control systems suitable for indoor environments. 1. Introduction antimicrobial activity, and adsorption of gaseous pollutants. However, multiple layers of filters result in high pressure drops along with long airflow pathways, and require more energy consumption and frequent filter replacement [23]. Therefore, considerable attention is being devoted to the development of multifunctional air filters that integrate at least two layers with different functions into a single layer [24,25]. To enable high-efficiency PM capture, fibrous filters should be thicker or be composed of densely packed nanofibers. Such filter structures inevitably increase airflow resistance, consequently leading to greater energy consumption [26,27]. However, if electrostatic forces between the particles and fibers are established, the capture efficiency can be improved substantially without an increase in filter pressure drop [28]. Recently, our group reported a conductive polyester/aluminum (PET/Al) fibrous filter showing high-efficiency electrostatic PM removal (∼99.99% for 30–400 nm particles) with a low pressure drop [29]. Because the PET/Al filter is electrically conductive, strong electric fields can be created around the filter by directly introducing electric potential, and the charged PM is effectively deposited onto the fibers via Coulomb forces. There have been several reports of the mild antimicrobial activity of alumina (Al2O3) particles against microorganisms [30–32]. In addition, thin oxide layers (3–10 nm) form on the Al nanostructures of the PET/Al filter, which could confer antimicrobial activity. Improving the antimicrobial properties of the PET/Al filter via the structural control of the Al layers would be of great significance to support its potential as a bifunctional filter for total air quality treatment. In this study, we demonstrate the bifunctionality of the PET/Al filter regarding its high effective capture and inactivation of airborne bacteria. This filter showed a pressure drop 10 times lower per thickness compared with a commercial high-efficiency particulate air (HEPA) filter. Both Escherichia coli and Staphylococcus epidermidis were captured on the PET/Al filter with a high efficiency of ∼99.99% via electrostatic forces. The antimicrobial activity of the filter itself for each species was increased to ∼94.8% and ∼96.9%, respectively, due to the enhanced Bioaerosols are classified as airborne biological particulate matter (PM), including microorganisms (e.g., viruses, bacteria, and fungi), biological particulate fragments, and toxins. Individual bioaerosols range in size from submicroscopic particles (< 0.01 μm) to particles larger than 100 μm, which are readily transmitted by wind and can float for a long time in the atmosphere [1–3]. Because they can be inhaled or attach to humans in their airborne state, they are an etiological agent for respiratory and infectious human diseases [4]. Thus, the control of microorganisms suspended in air is currently an active research field driven by the increasing demand for occupational and public health safety [5–7]. Fibrous filters are used widely as a method for removing hazardous bioaerosols due to their fascinating features, such as a light weight, cost-effectiveness, easy of fabrication, and universal applicability [1,8–10]. However, microorganisms that are captured in such filter media can remain viable, and some can even grow and propagate by absorbing air moisture and nutrients in dust. These organisms can become resuspended in the air upon the deterioration of the filters or accidental physical impact during maintenance [11]. In addition, volatile organic compounds, an indoor air carcinogen produced by microbial metabolism, can be emitted [12]. Hence, many efforts have been made to inactivate biological PM by depositing antimicrobial components onto filter fibers, such as inorganic materials (e.g., Ag, Cu, and TiO2 nanoparticles (NPs)) and organic materials (e.g., Sophora flavescens, Euscaphis japonica, and tea tree oil NPs) [13–20]. However, because antimicrobial NPs physically adhere to the fiber surfaces, these filters show poor durability against washing treatments for reuse. Moreover, the antimicrobial activity of the filters progressively diminishes due to the accumulation of dust covering the functional particles [21,22]. Modern air-cleaning devices are composed of multiple filters with different functions that support high-efficiency PM capture, Fig. 1. (a) Diagram of the chemical solution process for conductive polyester/aluminum (PET/Al) filter fabrication. (b) Configuration of the electrostatic filtration device composed of a carbon fiber ionizer and two PET/Al filters. Electric fields are formed between the front filter and the ionizer as well as the back filter and the front filter. Inflowing particles are negatively charged by the ionizer, and are captured by Coulomb forces towards the front PET/Al filter. (c) Schematic diagram of the experimental setup used in the filtration tests. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). 30 Journal of Hazardous Materials 351 (2018) 29–37 D.Y. Choi et al. (particles/cm3) of the bacterial bioaerosol at the inlet and outlet of the filter, respectively. The size and number concentration of bacterial bioaerosols were measured with an aerodynamic particle sizer (Model 3321, TSI Inc.) at both the inlet and outlet of the electrostatic filtration device marked with a dashed blue box in Fig. 1c. The airflow face velocity was fixed to 3.4 cm/s for all filtration tests unless noted otherwise. To eliminate any effects on time-dependent particle generation, efficiency measurements were conducted more than four times for each experimental condition. surface roughness and wettability of the chemically grown Al layer. More importantly, the capture and antimicrobial performances remained high during repeated reuse of the PET/Al filter after washing with water. This work presents a new approach for the simultaneous removal and inactivation of biological PM, which is attractive for lowcost, energy-efficient air quality applications. 2. Materials and methods 2.1. Al-coated conductive fibrous filter fabrication 2.5. Antimicrobial test A PET nonwoven filter (fiber diameter = 30 μm; porosity = 78.3%) was used as a backbone membrane to fabricate the PET/Al filter. Fig. 1a illustrates the chemical solution (CS) process, which included two steps: (i) catalytic treatment of the PET filter by titanium isopropoxide (Ti(Oi-Pr)4) and (ii) dip-coating of the catalytically treated filter into an Al precursor ink, AlH3{O(C4H9)2}, which was synthesized via an ethereal reaction of aluminum chloride (AlCl3) with lithium aluminum hydride (LiAlH4) in dibutyl ether (O(C4H9)2). More detailed explanations of the Al precursor ink preparation and CS process are described in our previous work [29,33–36]. After the bacteria were deposited onto the PET/Al filter for about 10 min, they were left for an additional contact time of 5 min. Then, the filters were placed in 15 mL (Vextraction) of phosphate-buffered saline (PBS) containing 0.01% Tween 80 and vortexed for 5 min to transfer the bacteria from the filters to the PBS solution. The resulting bacterial suspension was serially diluted onto plates containing nutrient agar (catalog no. 213000, Becton Dickinson) and incubated at 37 °C for 24 h. The colonies that grew on the plates were counted. The bacterial inactivation efficiency (ε) was calculated as follows: 2.2. Electrostatic filtration device design λPET = Fig. 1b presents a sketch of the electrostatic filtration device used in this study. The device was composed of a carbon fiber ionizer to charge bioaerosols and two conductive PET/Al filters to capture them. The ionizer was positioned 5 cm ahead of the front of the PET/Al filter. A constant voltage of −10 kV was applied to the ionizer when negative ion generation was required. Thus, the bacterial aerosols gained negative charges before passing through the PET/Al filters. PET/Al filters were installed on both sides of a 5-mm-thick plastic separator, and each filter was connected to a metal electrode to introduce electric potential. The front PET/Al filter was connected to a positive high-voltage source as required; the back PET/Al filter was grounded for all filter tests. λPET/Al = ⎜ (3) Cinlet⋅Qsampling⋅η⋅ζ extraction Vextraction λPET/Al ⎞ , ε = ⎛1 − λPET ⎠ ⎝ , (4) ⎟ (5) where λPET and λPET/Al are the active proportions of bacteria from the PET filter (control) and PET/Al filter, respectively. N is the total concentration of bacteria (particles/mL) in the extraction suspension plated onto the agar. Cinlet is the total concentration of airborne bacteria from the nebulizer. Qsampling is the total airflow sampling volume. ζextraction is the physical extraction of bacteria at each filter, defined as the ratio of the number of particles transferred from the filter to the extraction liquid to the number of particles removed from airflow using the filter. The values of ζextraction for all filter samples were as high as ∼96%, comparable to those obtained using the method proposed by Wang et al. [37]. To evaluate the capture and inactivation efficiency of the PET/Al filter, we selected two species of bacteria with different cell wall structures, S. epidermidis (Gram-positive bacteria) and E. coli (Gramnegative bacteria). These bacteria are commonly found in indoor environments and on human skin, and are used widely in bioaerosol research. The S. epidermidis and E. coli cultures were incubated in a nutrient broth medium (catalog no. 234000, Becton Dickinson) at 37 °C for 24 h and 18 h, respectively. When the bacterial suspension reached an optical density at 600 nm of ∼0.8, the bacteria were harvested by centrifugation (5000 × g, 10 min) and washed three times with distilled water. Once the concentration of the resulting suspension reached ∼108 colony forming units (CFU)/mL, 1 mL of bacterial suspension was mixed with 19 mL of distilled water and loaded into a one-jet Collison Nebulizer (CN 241, BGI Inc.). 2.6. Characterization Field-emission scanning electron microscopy (FE-SEM) images and energy dispersive spectroscopy (EDS) mapping results were acquired using a scanning electron microscope (SU8230, Hitachi). Sheet resistance measurements of the PET/Al filters were carried out using the four-probe van der Pauw method (FPP-HS8, DASOLENG). The pressure drop between the upstream and downstream sides of the PET/Al filter was measured with an electronic manometer (FCO332, Furness Controls) with a detection range of 100 Pa. The porosity, average pore diameter, and pore size distribution of the filters were determined using the mercury intrusion porosimetry (MIP) analysis (Micromeritics AutoPore V 9600, Micromeritics Corp.). The air permeability of the filters was evaluated with an air permeability tester (FX 3300, TEXTEST) at a constant pressure drop of 125 Pa. Ozone emissions at the outlet of the electrostatic filtration device were analyzed for various electric field conditions using an ultraviolet (UV) photometric ozone monitor (Model 49C, Thermo Environment Instruments Inc.) under a sampling flow rate of 1.0 L/min. The static water contact angle (WCA) of the filters was determined following the sessile drop method using a contact angle analyzer (Phoenix 300 Plus, SEO Co., Ltd.). 2010). 2.4. Filtration test Fig. 1c presents a schematic diagram of the apparatus used to measure the capture efficiency of PET/Al filters. Droplets containing test bacteria were atomized using the nebulizer supplied with 1.0 L/min airflow under 1.0 psig. Before the bacterial bioaerosols were introduced to the filter medium, moisture was thoroughly removed by passing through a diffusion dryer. The capture efficiencies (η) of the PET/Al filters were calculated using the following equation: Coutlet , Cinlet (2) CFUPET/Al , NPET/Al NPET orNPET/Al = 2.3. Bacterial aerosol preparation η=1− CFUPET , NPET (1) where Cinlet and Coutlet represent the particle concentrations 31 Journal of Hazardous Materials 351 (2018) 29–37 D.Y. Choi et al. 3. Results and discussion that affect filter performance. We evaluated the characteristics of the pore structure and the air permeability of the PET/Al filter, and the corresponding results were compared with those of the raw PET filter. Fig. S1 shows pore size distribution from MIP analysis for the PET filter and the PET/Al filter. Both filters showed a pore size distribution consisting of air pores without mesopores and capillary pores. The pores in each filter are mainly distributed between 30 and 300 μm. The PET filter has a porosity of 75.9% and an average pore size of 105.1 μm, quite similar to those of the PET/Al filter; the porosity and average pore size of the PET/Al filter are 75.0% and 119.8 μm, respectively. The air permeability of PET filter was 303 cm3/cm2/s, and that of PET/Al filter was 296 cm3/cm2/s. Above results imply that our CS process for the creation of Al thin layers can retain the integrity of the original porous structure of the PET filters. 3.1. Conductive PET/Al fibrous filter preparation The PET/Al filters were produced following the CS process developed in our previous work [29]. Fig. 2a describes the formation mechanism of the Al nanostructures on each unit fiber during the CS process. The PET filter treated catalytically with Ti(O-i-Pr)4 was immersed in the Al precursor ink composed of AlH3{O(C4H9)2}. AlH3 solvated in dibutyl ether begins to decompose directly into Al and 1.5 H2 at the fiber surface with the assistance of the catalyst, which enables the creation of conductive Al layers at room temperature. Al nuclei are produced ubiquitously over the fiber surface and grow gradually until they completely cover the fiber surface. The existence of Al features coating the fiber surface can be easily distinguished by the naked eye, as the color of the raw PET filter changes from white to metallic gray after the chemical reaction of the Al precursor ink (left images, Fig. 2b and c). No severe damage to the fibers was observed and the network structure of the fibers remained (Fig. 2b-ii and c-ii). The magnified SEM images in Fig. 2b-iii and c-iii show the fiber surfaces of the PET and PET/Al filters, respectively. The bumpy surface of the PET/Al fibers was clearly differentiated from the smooth surface of the PET fibers. Fig. 2d presents the EDS mapping images of the PET/Al filter. Carbon (cyan dots) and aluminum (yellow dots) were distributed evenly over the fiber surface, confirming the successful formation of Al nanostructures on the fibers via the CS process. Porosity and distribution of pore sizes are important characteristics 3.2. Electrical and airflow resistance properties To demonstrate the tunable electrical properties of the PET/Al filter, we investigated the variation in electrical resistance according to the immersion time in the Al precursor ink. Fig. 3a displays the sheet resistance values of the PET/Al filters and the corresponding areal loading mass of Al as a function of coating time. The quantity of Al deposited on the fibers increased linearly with increasing time, exhibiting a constant formation rate of Al features on the fiber surface. The growth rate of the Al thin film decelerated as the surface increased. Therefore, the sheet resistance decreased rapidly during a reaction time of < 45 min, but the resistance change slowed thereafter. The PET/Al filter had a somewhat high resistance of ∼770 Ω/sq at an immersion time of 15 min (PET/ Fig. 2. (a) Schematic illustration of the formation mechanism of Al thin films on the filter fibers. (b) Photograph of a 15 cm × 15 cm raw PET filter (left); scanning electron microscopy (SEM) image showing its fiber structure (middle) and magnified SEM image of the fiber surface (right). (c) Photograph of a 15 cm × 15 cm PET/Al filter with size of (left); SEM image showing its fiber structure (middle) and magnified SEM image of the fiber surface (right). (d) SEM image of the PET/Al fibers (top), and the corresponding energy dispersive spectroscopy mapping images of carbon (middle) and aluminum (bottom). Scale bars represent 20 μm. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). 32 Journal of Hazardous Materials 351 (2018) 29–37 D.Y. Choi et al. Fig. 3. (a) Variations in the electrical sheet resistance (red circles) and Al loading mass (blue squares) of the PET/Al filters according to the immersion time in the Al precursor ink. (b) Pressure drop versus flow rate characteristics of the PET/Al filters prepared at different coating times (0, 15, 30, and 45 min; Raw PET, PET/Al–15, PET/AL–30, and PET/Al–45, respectively). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). 3.3. Bacterial capture efficiency Al–15); however, its resistance dropped to 1.1 Ω/sq after 180 min (PET/ Al–180). Zhang et al. [38] reported that 50% of energy costs are related to heating, ventilation, and air conditioning systems and that 30% of these costs are associated with air filtration in commercial buildings. To save the energy used for protecting air quality, it is important to minimize the pressure drop across a filter while ensuring high capture efficiency, given that the flow resistance is related to the amount of energy consumed by a fan. Fig. 3b shows the pressure drop curves of the raw PET, PET/Al, and commercial HEPA filters as a function of the airflow rate. The face velocity was set in the range of 1.5–7.5 cm/s to include the standard airflow velocity of 5.33 cm/s suggested by the US Department of Energy (DOE) [39,40]. The PET/Al filters were prepared at coating times of 15, 30, and 45 min. Because the pressure drop of a filter is directly proportional to its thickness, the measured pressure drop data were normalized to the thickness of each sample. The PET and PET/Al filters were about 0.25 mm thick, while the HEPA filter was about 0.7 mm thick. The airflow resistance characteristics of the PET/Al filters did not differ greatly from that of the PET filter even after the formation of the Al thin layer. Moreover, the pressure drop per thickness of the PET/Al filters was roughly 10 times lower than that of the HEPA filter, showing superior air permeability. The capture performance of the PET/Al filters was evaluated using E. coli and S. epidermidis bioaerosols. Fig. S2 depicts the size distribution curves of the two species, which showed similar trends. Fig. 4a shows the capture efficiency of E. coli as a function of size. When uncharged E. coli was introduced to the electrically grounded PET/Al–180 filter, E. coli was removed with an efficiency of ∼31%, which was mainly driven by the mechanical filtration mechanism. Because the PET/Al filter was electrically conductive, the filter gained a high electric potential via the external high-voltage device, and strong electric fields were generated around the filter fibers. Thus, charged bioaerosols were effectively captured by the fibers via by electrostatic attraction. To examine the electrostatic capture performance, we applied voltages of −10 kV and + 10 kV to the ionizer and the front PET/Al filter, respectively. The capture efficiency of the charged E. coli was significantly augmented by Coulomb forces, and the average efficiency increased to about 99.99%. The overall capture performance for S. epidermidis did not differ greatly from that of E. coli since they had similar size properties (Fig. S3). The application of a high electric field was beneficial for augmenting the electrostatic capture efficiency, but could cause a harmful level of ozone generation from the ionizer. The ozone concentration was about 1.8 ppb when voltages of −10 kV and +10 kV were applied to the ionizer and the PET/Al filter, respectively (Fig. S4), far below 50 ppb, which is the standard for electrostatic air cleaners (UL 867) [41]. Fig. 4. (a) Capture efficiency of the PET/Al–180 filter depending on particle size of E. coli cells. (b) Bar graph showing a comparison of the E. coli capture efficiencies for the PET/Al filters fabricated at various coating times. SEM images of the deposition morphology of E. coli collected by (c) mechanical filtration and (d) electrostatic filtration. Each inset shows a magnified image of the region in the orange rectangular box. 33 Journal of Hazardous Materials 351 (2018) 29–37 D.Y. Choi et al. As demonstrated above, the quantity of Al on the fibers increased proportionally with coating time, thereby improving electrical conductivity. Therefore, we investigated the filtration performance of the conductive PET/Al filters depending on the degree of Al loading. Fig. 4b shows the E. coli capture efficiency of the PET/Al filters prepared at various immersion times in Al precursor ink. The efficiencies of the PET/Al filters driven by the mechanical filtration mechanism were similar regardless of filter type, indicating that an increase in Al layer thickness had a minimal effect on the filtration characteristics. However, the PET/Al filters conferred excellent electrostatic capture performance, independent of the electrical properties, once the electrical networks were formed due to the formation of Al nanostructures on all fiber surfaces. Regardless, no relationship between the capture performance for the two bacteria species and immersion time was identified, and the difference was not significant (p-value > 0.05). Fig. 4c and d show the difference in the deposition density of E. coli bacteria captured on the PET/Al filters by mechanical and electrostatic forces during the same filtration time, respectively. Bacteria were deposited sparsely onto the fibers when driven predominantly by mechanical forces. However, substantially more bacteria accumulated on the fiber surfaces when driven by electrostatic attraction, and dendrites of bacteria formed (inset, Fig. 4d). SEM analysis results visually convey the outstanding electrostatic capture ability of the PET/Al filter. from 89.3% to 94% when the Al coating time increased from 5 min to 15 min. Beyond 15 min, the performance showed no significant correlation (p-value > 0.05) with the degree of Al deposition, since it was already saturated (> 94%). However, the efficiency of S. epidermidis gradually improved from 79.7% to 96.9% with increasing Al coating time. The images in Fig. 5b correspond to the cultures of both bacteria sampled from each of the raw PET (control), PET/Al–15, and PET/ Al–180 filters; the results showed that the PET/Al filter has strong antimicrobial activity. Because Gram-positive S. epidermidis has a greater resistance to air exposure than Gram-negative E. coli, the natural decay of E. coli is more dominant than S. epidermidis [42,43]. Thus, the number of S. epidermidis colonies cultured from the control filter was always higher than that of E. coli colonies, resulting in a smaller λPET value for E. coli. Although the inactivation performance of E. coli was somewhat overestimated when considering the above points, the PET/ Al–180 filter showed better antimicrobial properties against S. epidermidis (96.9 ± 0.59%) than E. coli (94.8 ± 1.18%). The difference in the antimicrobial resistance to E. coli and S. epidermidis may be due to their dissimilar cell structures and physiologies; Gram-negative bacteria have a cell membrane structure capable of resisting antimicrobial agents [44,45], while the cell walls of Gram-positive bacteria can bind larger quantities of several metals than the cell envelopes of Gram-negative bacteria [46]. 3.4. Antimicrobial performance of the PET/Al filter 3.4.2. Antimicrobial mechanism of the PET/Al filter One of the most important mechanisms by which cells adapt to surrounding environments is their adhesive interaction with solid surfaces. The initial bacterial adhesion process is regulated by electrostatic interactions, which is enhanced when the cell wall of a bacterium lays flush against a filter surface. A filter surface with nanoscale roughness 3.4.1. Evaluation of the inactivation efficiency of the PET/Al filter We investigated the antimicrobial activities of the PET/Al filter based on the quantity of Al generated on the PET fibers, as shown in Fig. 5a. The inactivation efficiency against E. coli increased slightly Fig. 5. (a) Bacterial inactivation efficiencies for E. coli (red bars) and S. epidermidis (green bars) of the PET/Al filters prepared at different Al coating times. (b) Digital images of recultivated E. coli colonies and S. epidermidis colonies on agar culture plates. Changes in the surface morphology of the PET/Al filter with Al deposition times of (c) 15 min, (d) 45 min, (e) 90 min, and (f) 180 min. Each inset indicates the magnified SEM image of a fiber surface and the scale bars represents 2 μm. (g) Water contact angle (WCA) of the PET/Al filters according to Al coating time. (h) Images of the static WCA measurements performed on the raw PET filter (left) and PET/Al–180 filter (right). (i) Images describing the dynamic water adhesion behavior on the surface of the PET/Al–180 filter. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article). 34 Journal of Hazardous Materials 351 (2018) 29–37 D.Y. Choi et al. antimicrobial efficiencies for E. coli and S. epidermidis were improved to ∼75% and ∼85%, respectively, for an ion exposure time of 10 min. In our study, however, the inactivation performance of the PET/Al filter had already been saturated (> 98%) by the antimicrobial activity of the Al nanostructures and electric filed before introducing a negative ion treatment. Thus, these factors seem to conceal the effect of negative ions on the antimicrobial performance. Also, it is possible that antimicrobial activity by the negative ions could be worked during transportation of bacteria to the filter medium. However, the short exposure (less than 2 s) of bacteria to negative ions would be insignificant to antimicrobial performance, as supported by the results of Lee et al. [57]. could prevent close contact with the cell wall due to the relative rigidity of the cell wall [47]. Thus, the nanorough surface could hinder the preliminary adhesion step of bacteria, and consequently lead to apoptosis [48]. Fig. 5c–f show SEM images in sequence for the PET/Al filters prepared at coating times of 15, 45, 90, and 180 min. The filter surface became increasingly bumpy with longer coating times, and the size of the Al nanograins increased. Based on these results, the enhanced roughness of the PET/Al filter could augment its antimicrobial performance. In addition, the morphological changes in the fiber surface could influence the hydrophobicity of the filter, resulting in the interference of cell-substratum interactions during proliferation and differentiation processes [49]. To examine the wettability of the PET/Al filters depending on their surface morphology, the WCA was measured (Fig. 5g). The contact angle of water droplets on the PET/Al filter monotonically increased with Al loading content. For example, the WCA of the raw PET filter was 108.4°, whereas the PET/Al–180 filter had a higher WCA of 135.6°, indicative of enhancement of the hydrophobicity (Fig. 5h). Fig. 5i displays an image of a 3–μL water droplet during dynamic contact with the surface of the PET/Al–180 filter. When detaching from the filter, the droplet was not transferred to the filter and maintained its original shape, revealing extremely low water adhesion of the PET/ Al–180 filter surface [50]. This reinforced hydrophobicity may be attributed to the increased grain size and surface roughness of the Al nanostructures on the PET fibers. Together with physical antimicrobial activities, another important mechanism driving cell death is the disruption of cell walls by reactive oxygen species (ROS) generated from metal oxides, such as ZnO2 and TiO2 [30,51,52]. Several studies have demonstrated that Al2O3 NPs show mild antimicrobial properties. For example, Jiang et al. [31] observed antimicrobial activities of Al2O3 NPs against both Gram-positive and Gram-negative bacteria under dark conditions, indicating the possible production of free radicals under dark conditions [53]. Sadiq et al. [32] found that Al2O3 NPs had a minor antimicrobial effect on E. coli at high concentrations of Al2O3 NP up to 1000 μg/mL. These results support the hypothesis that the high antimicrobial activity of our PET/ Al filter may be ascribed to both the physical (surface roughness and hydrophobicity) and chemical (ROS and free radicals) characteristics of the Al nanostructures formed via the CS process. 3.6. Filter reusability To support economical maintenance, filters should be washable to enable the recovery of their functions. We examined the performance changes of the PET/Al filters under a cyclic washing test. Each cycle consisted of the filtration and antimicrobial experiments, cleaning of the test filter in an ultrasonic water bath for 10 min, UV irradiation for 10 min, and complete drying in an electric oven at 50 °C for 2 h. Fig. 7 shows the variations in the capture and inactivation performances for both E. coli and S. epidermidis during five cycles of the washing test. The capture efficiency was stably maintained regardless of the bacteria species species (E. coli, 99.4 ± 0.80%; S. epidermidis, 99.6 ± 0.77%) (Fig. 7a). Though the results are not shown here, the pressure drop characteristic did not change during a cyclic washing test. Furthermore, during the reusability test, the reused filter still showed high antimicrobial activities against both bacteria with no apparent degradation of function; E. coli and S. epidermidis were inactivated with efficiencies of 99.2 ± 0.28% and 98.8 ± 1.3%, respectively (Fig. 7b). The PET/Al filter retained its bifunctionality even after the cleaning process owing to the robust durability of the chemically grown Al structures on the PET fibers. These results demonstrate the reusability of our PET/Al filter after washing with water. 4. Conclusions We present a bifunctional PET/Al fibrous filter showing good performance regarding the electrostatic capture and inactivation of airborne microorganisms. The PET/Al filter captured bacterial bioaerosols of E. coli and S. epidermidis with an extremely high efficiency due to the electrostatic attraction between the filter and bacteria, with a pressure drop much lower than that of a commercial HEPA filter. The capture performance of the PET/Al filter was independent of the physical properties of the Al thin layers, such as its electrical resistance, surface roughness, and wettability. However, the antimicrobial activity increased with increasing surface roughness and hydrophobicity. The effects of the application of an electric field and negative ions on the 3.5. Effects of electric field and charge on antimicrobial performance We examined the changes in the bacteria inactivation performance of the PET/Al filter when electric fields and charges were introduced. Fig. 6 shows the inactivation efficiencies for E. coli and S. epidermidis according to the onset of an electric field around the filter and the electrical charging of bacteria. The PET/Al–180 filter was used as the test sample. For comparison, its intrinsic inactivation performance is included in Fig. 6 for the case that electric field and charges were not applied (left two bars). When a high electric potential of 10 kV was applied to the filter (middle two bars), the inactivation efficiency for E. coli increased from 94.8% to 98.7%, while that for S. epidermidis increased slightly from 96.9% to 98.2% (p-value < 0.001). The application of the electric field was more effective in E. coli because Grampositive bacteria are less sensitive to electric field effects than Gramnegative bacteria [54]. The survival rate of exposed bacteria depends on a combination of the electric field strength and treatment time [55]; however, it is currently unknown whether the antimicrobial mechanism on electric field effects is derived from structural damage or metabolic dysfunction of bacteria. When negative charges were additionally applied to bacteria (right two bars), the inactivation performance was very slightly increased. However, the results were not statistically significant (t-test pvalue > 0.05). This outcome is somewhat different with previous studies of the antimicrobial effect of negative ions against bacteria deposited on a normal air filter. Kim et al. [56] reported that the Fig. 6. Comparison of the inactivation efficiency of the PET/Al–180 filter according to the filtration method. 35 Journal of Hazardous Materials 351 (2018) 29–37 D.Y. Choi et al. [3] G.J. Smith, D. Vijaykrishna, J. Bahl, S.J. Lycett, M. Worobey, O.G. Pybus, S.K. Ma, C.L. Cheung, J. Raghwani, S. Bhatt, Origins and evolutionary genomics of the 2009 swine-origin H1N1 influenza A epidemic, Nature 459 (2009) 1122–1125. [4] C.E. Main, Aerobiological, ecological, and health linkages, Environ. Sci. Technol. 29 (2003) 347–349. [5] W.W. Nazaroff, Indoor bioaerosol dynamics, Indoor Air 26 (2016) 61–78. [6] J. Douwes, P. Thorne, N. Pearce, D. Heederik, Bioaerosol health effects and exposure assessment: progress and prospects, Ann. Occup. Hyg. 47 (2003) 187–200. [7] J.E. Lee, B.U. Lee, G.N. Bae, J.H. Jung, Evaluation of aerosolization characteristics of biocontaminated particles from flood-damaged housing materials, J. Aerosol Sci. 106 (2017) 93–99. [8] W.J. Fisk, Health benefits of particle filtration, Indoor Air 23 (2013) 357–368. [9] B. Stephens, J. Siegel, Ultrafine particle removal by residential heating, ventilating, and air‐conditioning filters, Indoor Air 23 (2013) 488–497. [10] J.S. Kang, H. Kim, J. Choi, H. Yi, S.C. Seo, G.N. Bae, J.H. Jung, Antimicrobial air filter fabrication using a continuous high-throughput aerosol-based process, Aerosol Air Qual. Res. 16 (2016) 2059–2066. [11] B.U. Lee, Life comes from the air: a short review on bioaerosol control, Aerosol Air Qual. Res. 11 (2011) 921–927. [12] R. Simmons, S. Crow, Fungal colonization of air filters for use in heating, ventilating, and air conditioning (HVAC) systems, J. Ind. Microbiol. Biotechnol. 14 (1995) 41–45. [13] J.P. Ruparelia, A.K. Chatterjee, S.P. Duttagupta, S. Mukherji, Strain specificity in antimicrobial activity of silver and copper nanoparticles, Acta Biomater. 4 (2008) 707–716. [14] J.H. Jung, G.B. Hwang, J.E. Lee, G.N. Bae, Preparation of airborne Ag/CNT hybrid nanoparticles using an aerosol process and their application to antimicrobial air filtration, Langmuir 27 (2011) 10256–10264. [15] K. Donaldson, R. Aitken, L. Tran, V. Stone, R. Duffin, G. Forrest, A. Alexander, Carbon nanotubes: a review of their properties in relation to pulmonary toxicology and workplace safety, Toxicol. Sci. 92 (2006) 5–22. [16] G.N. Bae, J.H. Jung, Aerosol-processed nanomaterials for antimicrobial air filtration, J. Nanosci. Nanotechnol. 5 (2016) 4487–4492. [17] J.H. Jung, G.B. Hwang, S.Y. Park, J.E. Lee, C.W. Nho, B.U. Lee, G.-N. Bae, Antimicrobial air filtration using airborne Sophora flavescens natural-product nanoparticles, Aerosol Sci. Technol. 45 (2011) 1510–1518. [18] G.B. Hwang, K.J. Heo, J.H. Yun, J.E. Lee, H.J. Lee, C.W. Nho, G.N. Bae, J.H. Jung, Antimicrobial air filters using natural euscaphis japonica nanoparticles, Plos One 10 (2015) e0126481. [19] G.B. Hwang, K.M. Sim, G.N. Bae, J.H. Jung, Synthesis of hybrid carbon nanotube structures coated with Sophora flavescens nanoparticles and their application to antimicrobial air filtration, J. Aerosol Sci. 86 (2015) 44–54. [20] J. Choi, B.J. Yang, G.N. Bae, J.H. Jung, Herbal extract incorporated nanofiber fabricated by an electrospinning technique and its application to antimicrobial air filtration, ACS Appl. Mater. Interfaces 45 (2015) 25313–25320. [21] Y.H. Joe, W. Ju, J.H. Park, Y.H. Yoon, J. Hwang, Correlation between the antibacterial ability of silver nanoparticle coated air filters and the dust loading, Aerosol Air Qual. Res. 13 (2013) 1009–1018. [22] K.K. Foarde, J.T. Hanley, Determine the efficacy of antimicrobial treatments of fibrous air filters, ASHRAE Trans. 107 (2001) 156. [23] A. Luengas, A. Barona, C. Hort, G. Gallastegui, V. Platel, A. Elias, A review of indoor air treatment technologies, Rev. Environ. Sci. Biotechnol. 14 (2015) 499–522. [24] Y. Zhao, Z.-X. Low, S. Feng, Z. Zhong, Y. Wang, Z. Yao, Multifunctional hybrid porous filters with hierarchical structures for simultaneous removal of indoor vocs, dusts and microorganisms, Nanoscale 9 (2017) 5433–5444. [25] H. Souzandeh, K.S. Johnson, Y. Wang, K. Bhamidipaty, W.-H. Zhong, Soy-proteinbased nanofabrics for highly efficient and multifunctional air filtration, ACS App. Mater. Interfaces 8 (2016) 20023–20031. [26] D. Thomas, P. Contal, V. Renaudin, P. Penicot, D. Leclerc, J. Vendel, Modelling pressure drop in HEPA filters during dynamic filtration, J. Aerosol Sci. 30 (1999) 235–246. [27] W.J. Fisk, D. Faulkner, J. Palonen, O. Seppanen, Performance and costs of particle air filtration technologies, Indoor Air 12 (2002) 223–234. [28] K.M. Sim, H.S. Park, G.N. Bae, J.H. Jumg, Antimicrobial nanoparticle-coated electrostatic air filter with high filtration efficiency and low pressure drop, Sci. Total. Environ. 533 (2015) 266–274. [29] D.Y. Choi, S.-H. Jung, D.-K. Song, E.J. An, D. Park, T.-O. Kim, J.H. Jung, H.M. Lee, Al-coated conductive fibrous filters with low pressure drop for efficient electrostatic capture of ultrafine particulate pollutants, ACS Appl. Mater. Interfaces 9 (2017) 16495–16504. [30] X. Zhu, L. Zhu, Z. Duan, R. Qi, Y. Li, Y. Lang, Comparative toxicity of several metal oxide nanoparticle aqueous suspensions to zebrafish (Danio rerio) early developmental stage, J. Environ. Sci. Health A. Tox. Hazard. Subst. Envrion. Eng. 43 (2008) 278–284. [31] W. Jiang, H. Mashayekhi, B. Xing, Bacterial toxicity comparison between nano-and micro-scaled oxide particles, Environ. Pollut. 157 (2009) 1619–1625. [32] I.M. Sadiq, B. Chowdhury, N. Chandrasekaran, A. Mukherjee, Antimicrobial sensitivity of Escherichia coli to alumina nanoparticles, Nanomedicine 5 (2009) 282–286. [33] H.M. Lee, S.Y. Choi, K.T. Kim, J.Y. Yun, D.S. Jung, S.B. Park, J. Park, A novel solution‐stamping process for preparation of a highly conductive aluminum thin film, Adv. Mater. 23 (2011) 5524–5528. [34] H.M. Lee, H.B. Lee, D.S. Jung, J.-Y. Yun, S.H. Ko, S.B. Park, Solution processed aluminum paper for flexible electronics, Langmuir 28 (2012) 13127–13135. [35] H.M. Lee, S.Y. Choi, A. Jung, S.H. Ko, Highly conductive aluminum textile and paper for flexible and wearable electronics, Angew. Chem. 125 (2013) 7872–7877. Fig. 7. Reusability of the conductive PET/Al filter. Monitoring of the (a) capture efficiency and (b) inactivation efficiency over five cleaning cycles. antimicrobial performance were not significant, as most of the bacteria for both bioaerosols were inactivated by the PET/Al filter alone. The bifunctionality of the PET/Al filter was stably maintained after multiple washings due to the robust durability of the chemically grown Al layers, indicative of its reusability. In biological settings, a variety of environmental factors must be considered, such as humidity and temperature. Thus, field experiments are required to confirm the applicability of the PET/Al filter in real environments. This study provides valuable information on the development of an energy-efficient bioaerosol control system suitable for use in indoor environments. Notes The authors declare no competing financial interest. Acknowledgments This work was supported by a grant from the Railway Technology Research Project of the Ministry of Land, Infrastructure and Transport (18RTRP-B082486-05) and was partially supported by the Korea Institute of Science & Technology (KIST) Institutional Program, by the Ministry of Environment (2016000160008, Public Technology Program based on Environmental Policy), and by the Basic Science Research Program through the National Research Foundation of Korea (NRF) funded by the Ministry of Science (2015R1D1A1A09056879). We thank Alink Co., Ltd. for their help in fabricating the filters and Al precursor ink. Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jhazmat.2018.02.043. References [1] W.C. Hinds, Aerosol Technology: Properties, Behavior, and Measurement of Airborne Particles, John Wiley & Sons, 2012. [2] C. Beggs, The airborne transmission of infection in hospital buildings: fact or fiction? Indoor Built Environ. 12 (2003) 9–18. 36 Journal of Hazardous Materials 351 (2018) 29–37 D.Y. Choi et al. (1985) 1893–1898. [47] J.T. Seil, T.J. Webster, Antimicrobial applications of nanotechnology: methods and literature, Int. J. Nanomed. 7 (2012) 2767. [48] K. Shellenberger, B.E. Logan, Effect of molecular scale roughness of glass beads on colloidal and bacterial deposition, Environ. Sci. Technol. 36 (2002) 184–189. [49] S. Kim, U.T. Jung, S.-K. Kim, J.-H. Lee, H.S. Choi, C.-S. Kim, M.Y. Jeong, Nanostructured multifunctional surface with antireflective and antimicrobial characteristics, ACS Appl. Mater. Interfaces 7 (2015) 326–331. [50] S. Zhang, N. Tang, L. Cao, X. Yin, J. Yu, B. Ding, Highly integrated polysulfone/ polyacrylonitrile/polyamide-6 air filter for multilevel physical sieving airborne particles, ACS Appl. Mater. Interfaces 8 (2016) 29062–29072. [51] R. Brayner, R. Ferrari-Iliou, N. Brivois, S. Djediat, M.F. Benedetti, F. Fiévet, Toxicological impact studies based on Escherichia coli bacteria in ultrafine ZnO nanoparticles colloidal medium, Nano Lett. 6 (2006) 866–870. [52] S.M. Dizaj, F. Lotfipour, M. Barzegar-Jalali, M.H. Zarrintan, K. Adibkia, Antimicrobial activity of the metals and metal oxide nanoparticles, Mater. Sci. Eng. C Mater. Biol. Appl. 44 (2014) 278–284. [53] M. Green, E. Howman, Semiconductor quantum dots and free radical induced DNA nicking, Chem. Commun. (Camb.) (2005) 121–123. [54] H. Hülsheger, J. Potel, E.-G. Niemann, Electric field effects on bacteria and yeast cells, Radiat. Environ. Biophys. 22 (1983) 149–162. [55] M. Yao, G. Mainelis, H.R. An, Inactivation of microorganisms using electrostatic fields, Environ. Sci. Technol. 39 (2005) 3338–3344. [56] Y.S. Kim, K.Y. Yoon, J.H. Park, J. Hwang, Application of air ions for bacterial decolonization in air filters contaminated by aerosolized bacteria, Sci. Total Environ. 409 (2011) 748–755. [57] S.G. Lee, J. Hyun, S.H. Lee, J. Hwang, One-pass antibacterial efficacy of bipolar air ions against aerosolized Staphylococcus epidermidis in a duct flow, J. Aerosol Sci. 69 (2014) 71–81. [36] H.M. Lee, S.-Y. Choi, A. Jung, Direct deposition of highly conductive aluminum thin film on substrate by solution-dipping process, ACS Appl. Mater. Interfaces 5 (2013) 4581–4585. [37] Z. Wang, T. Reponen, S.A. Grinshpun, R.L. Górny, K. Willeke, Effect of sampling time and air humidity on the bioefficiency of filter samplers for bioaerosol collection, J. Aerosol Sci. 32 (2001) 661–674. [38] R. Zhang, C. Liu, P.-C. Hsu, C. Zhang, N. Liu, J. Zhang, H.R. Lee, Y. Lu, Y. Qiu, S. Chu, Nanofiber air filters with high temperature stability for efficient PM2. 5 removal from the pollution sources, Nano Lett. 16 (2016) 3642–3649. [39] D.O.E. Standard, Specification for HEPA Filters Used by DOE Contractors, in, DOESTD-3020-97, (1997). [40] S. Zhang, H. Liu, F. Zuo, X. Yin, J. Yu, B. Ding, A controlled design of ripple‐like polyamide‐6 nanofiber/nets membrane for high‐efficiency air filter, Small 13 (2017). [41] Q. Zhang, P.L. Jenkins, Evaluation of ozone emissions and exposures from consumer products and home appliances, Indoor Air 27 (2017) 386–397. [42] B.U. Lee, S.H. Yun, J.H. Jung, G.-N. Bae, Effect of relative humidity and variation of particle number size distribution on the inactivation effectiveness of airborne silver nanoparticles against bacteria bioaerosols deposited on a filter, J. Aerosol Sci. 41 (2010) 447–456. [43] G.B. Hwang, B.M. Kwon, S.J. Lee, B.U. Lee, K.M. Sim, G.N. Bae, J.H. Jung, Effects of antimicrobial air filters on the viability and culturability of airborne bacteria, Clean (Weinh) 10 (2016) 1264–1427. [44] X. Liang, M. Sun, L. Li, R. Qiao, K. Chen, Q. Xiao, F. Xu, Preparation and antibacterial activities of polyaniline/Cu 0.05 Zn 0.95 O nanocomposites, Dalton Trans. 41 (2012) 2804–2811. [45] A. Azam, A.S. Ahmed, M. Oves, M. Khan, A. Memic, Size-dependent antimicrobial properties of CuO nanoparticles against Gram-positive and-negative bacterial strains, Int. J. Nanomed. 7 (2012) 3527–3535. [46] T. Beveridge, W. Fyfe, Metal fixation by bacterial cell walls, Can. J. Earth Sci. 22 37