If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Establishment and application of test methodology demonstrating the functionality of air purification systems in reducing virus-loaded aerosol in indoor air
Corresponding author. Address: Netherlands Organization for Applied Scientific Research, Department of Microbiology and Systems Biology, Sylviusweg 71, 2333 BE, Leiden, the Netherlands. Tel.: +31 6211 345 87.
The global COVID-19 pandemic has resulted in a greater interest in improving the ventilation of indoor environments in order to remove aerosolized virus and thus reduce transmission. Air purification systems have been proposed as a solution to improve aerosol removal.
Aim
The aim was to determine the efficacy of air purification systems in reducing the viral load in the environmental air of a room.
Methods
A containment room equipped with HEPA filter on air intake and exhaust was constructed. It was connected via an inlet with the BSL-2 facility. From the BSL-2, Feline Coronavirus (FCoV)-loaded aerosols were released into the containment room. After nebulization, air sampling was performed to determine the viral load in air prior to assessing the clean air delivery rate of the air purification systems. The infectivity of the captured viruses was also examined.
Findings
The air purification systems realized a 97–99% reduction in viral load in air in 1 h.
Captured infectious FCoV was reduced by 99.9%–99.99% by use of an ESP technology.
Conclusions
The air purification systems, using ESP technology or HEPA filter, reduce the viral load in air. The ESP purifiers inactivate captured FCoV viruses. Therefore, air purification systems can be used as an adjunctive infection control measure.
The global COVID-19 pandemic caused a dramatic change in our socio-economic life. Social distancing, self-isolation and travel restrictions caused a reduced workforce across all economic sectors resulting in a loss of jobs [
According to current knowledge about transmission, SARS-CoV-2 transmission primarily occurs between people when an infected person is in close contact with another person. The extent to which the virus will transmit between individuals depends on the amount of infectious virus being shed and expelled by a person, the type of contact that person has with other individuals, the setting where exposure occurs, and what preventative measures are in place [
Anonymous. Preventing and mitigating COVID-19 at work: policy brief, 19 May 2021. WHO/2019-nCoV/Workplace_actions/Policy_brief/2021.1. Available at: Preventing and mitigating COVID-19 at work: policy brief, 19 May 2021 (who.int) [last accessed 11 2022].
]. In normal circumstances, particles will also be lost by gravitational deposition. Micro-organisms including viruses migrate as droplet nuclei in indoor air. These can be removed by processes such as ventilation, filtration, electrostatic precipitation, and combinations thereof [
Combined use of an electrostatic precipitator and a high-efficiency particulate air filter in building ventilation systems: effects on cardiorespiratory health indicators in healthy adults.
Electrostatic filtration by electrostatic precipitator (ESP) uses an electrostatic charge to generate air ions. This technology is commonly used in industry to charge particulate matter such as dust, smoke and fungal spores and filter them by deposition on oppositely charged plates [
]. HEPA filters consist of fibrous media made with multiple layers of randomly arranged fibres that trap airborne particles via three mechanisms, inertial impaction, interception and diffusion. Their combined action dictates the efficiency of the filter. Although there are no recommendations by the CDC for application of HEPA purifiers for air decontamination of airborne SARS-CoV-2, HEPA purifiers should be considered an adjunctive infection control measure [
This study aimed to establish and apply a method for determining the efficacy of air purification systems in reducing virus-loaded aerosols in a room. Four air purification systems were included in the study representing ESP and HEPA filtration. These were tested for their efficacy in removing viral aerosol, bacteriophage MS2 [
] from air. FCoV was used as a surrogate for SARS-CoV-2 because it represents an enveloped virus highly similar to SARS-CoV-2. It has a similar envelope structure with spike-protein, virus size of about 120 nm and ssRNA type of genome [
A containment room was constructed for performing the air purification experiments. The room had a volume of approximately 50 m3 (width 4.5 m × depth 4.3 m × height 2.6 m) and was physically closed from the external environment by a door with rubber seals and door-lock for controlled entrance. For sampling, the containment room was connected to a BSL-2 laboratory via a double door lock equipped with a UVC and chemical disinfection system that enabled the workers to manage the processes in the containment room and to collect samples in a controlled way without risk of spreading viral particles outside the containment room. The intake and exhaust air of the containment room were both HEPA filtered. The room was cleared of biological material after every experiment by UVC disinfection (Signify, Eindhoven, The Netherlands) and by air refreshment of 300 m3 air per hour via HEPA filtering. Shaded places are treated by chemical disinfection. Air sampling was performed by using a Coriolis air sampler (Bertin Technologies, Montigny-le-Bretonneux, France).
Air purification systems
The air purification systems used in the study were supplied by the partners of the PPP Room disinfection project. The air purification systems applied ESP or HEPA technology with an airflow of 400–500 m³/h.
Virus nebulization and sampling
The viruses, MS2 bacteriophages and FCoV viruses, were aerosolized using an eFlow® rapid nebulizer (PARI GmbH, Germany). This nebulizer applies a vibrating plate and appears to be more gentle on viral particles compared with jet nebulizers and is used for nebulization of high counts of bacteriophages against tuberculosis [
]. These particles were transported and released into the containment room via a tube with an airflow of 1 bar pre-pressure. To prevent accidental release of virus from the nebulizer system to the environmental air, it was placed in a closed plastic box in the BSL-2 laboratory from which the nebulized virus containing mist was transported via a tube of 50 cm directly into the containment room.
Air sampling was performed by the Coriolis sampler (Bertin Instruments, France) that samples 300 L of air per minute. This sampler demonstrated good biological sampling efficiencies and physical sampling efficiencies for viruses in a comparative test [
Comparative testing and evaluation of nine different air samplers: end-to-end sampling efficiencies as specific performance measurements for bioaerosol applications.
]. The samples were collected in a small beaker containing 15 mL of sampling liquid, which was placed in a UVC-closed transport box after sampling to prevent UVC illumination. Sampling liquid was sterile saline solution for MS2 bacteriophage and DMEM cell culture medium including 0.45% glucose, 2 mM L-glutamine and 100 U/mL penicillin and 100 U/mL streptomycin without 2% fetal calf serum for FCoV. Air samples taken by the Coriolis sampler in the containment room were transported into the BSL-2 laboratory via the inlet equipped with a double door. Application of UVC and chemical disinfection with VirkonS (Biosecurity BV, The Netherlands) prevents environmental release of virus upon transportation of the contained material in the transport box via the inlet into the BSL-2 laboratory. The samples were processed in the biosafety cabinet of the BSL-2 laboratory.
Sample analysis
Samples were analysed using two different approaches. Firstly, by detection of biological activity, implying infection of target cells and subsequent lysis of the cells. For MS2 bacteriophage samples, these samples were serially diluted and spread on Luria-Bertani Agar (LBA) (Oxoid Limited, USA) plates in overlay with confluently seeded Escherichia coli in LBA 0.5% w/v top agar. The plates were incubated at 37 °C for 24 h, after which the plaques of lysed E. coli cells were counted. For FCoV-containing samples, an infectivity assay was performed [
] whereby serial dilutions were seeded on cell cultures of CRFK cells on DMEM including 0.45% glucose, 2 mM L-glutamine and 100 U/mL penicillin and 100 U/mL streptomycin supplemented with 2% fetal calf serum in 96-well microtitre plates. After two days of incubation at 37 °C in a CO2 incubator, the wells of the plate were inspected for the cytopathic effect of the virus on the monolayer of CRFK cells and scored for TCID50.
Secondly, samples were analysed by detection and quantification of the viral genomes using reverse transcriptase quantitative PCR (RT-qPCR). Prior to PCR, RNA was extracted and purified from the FCoV samples by using the QIAamp Viral RNA Mini Kit (Qiagen, Netherlands) according to the supplier's protocol. A one-step RT-qPCR was performed on a mixture of 5 μL purified viral RNA, 0.4 μM Fel-CoV-F forward primer, 5′-TGCCACAGGATGGGCTTAC-3′, 0.5 μM Fel-CoV-R reverse primer, 5′-TTGTCAGTACGTGCTTCT GTTGAG-3′, 0.25 μM Fel-CoV-P-FAM/MGB probe, FAM-5′- ACGTAAAATCTAAAGCTGGTG-3′-MGB, and TaqMan Fast virus 1-step Master mix (Applied Biosystems, USA) in a total reaction volume of 25 μL. The thermocycler programme was 5 min at 50 °C, 20 s at 95 °C, 40 cycles alternating 15 s at 95 °C and 60 s at 60 °C. The RT-qPCR assay yields information on the quantity of the amount of target FCoV RNA based on a standard curve. It reflects the number of viruses both infectious and non-infectious that were sampled and thereby the basis to calculate the number of FCoV genome equivalents; the size of the FCoV RNA genome is about 30 kb [
Decay of MS2 loaded aerosol in the containment room
MS2 phage suspension of about 1 × 1010 pfu/mL was prepared by culturing the phage on E. coli on LBA plate, scraping the top agar layer from the plate and resuspending in 5 mL phage buffer according to DSMZ protocol (DSMZ, Germany). After subsequent centrifugation at 4000 g and filtration of the supernatant through a 0.45-μm filter, the filtrate was nebulized at 0.28 mL/min for 5 min into the containment room of 50 m3 using the e-Flow vibrating mesh nebulizer (Pari GmbH, Germany) while the air mixing was constantly achieved by four fan units. One minute after nebulization, the t = 0 sample was collected by the Coriolis sampler at 300 L/min for 3 min, which resulted in 900 L of sampled air. Subsequent samples were taken at t = 15 min, t = 30 min, t = 45 min and t = 60 min. These phage recovery data indicate the natural decay of MS2 phage in the containment room. Following the same nebulization and sampling conditions, the air purification systems' effect was explored on additional removal of MS2 phage loaded aerosol in the containment room. The decay realized by the air purification system was calculated by total decay with purifier action minus natural room decay without purifier action based on triplicate tests.
Decay of FCoV loaded aerosol in the containment room
The FCoV suspension was prepared by virus culturing on CRFK cells on DMEM including 0.45% glucose, 2 mM L-glutamine and 100 U/mL penicillin and 100 U/mL streptomycin supplemented with 2% foetal calf serum. After centrifugation at 4000 g, the FCoV suspension was nebulized. The decay of FCoV loaded aerosol was measured upon the action of the air purification systems. In the experiments with FCoV, a TCID50 load of FCoV in the suspension was about 2.5 × 105/mL. Because infectious virus could hardly be retrieved from the containment room at t = 0, the virus was detected using the FCoV specific RT-qPCR. All tests were performed in triplicate.
Calculation of clean air delivery rate, single pass efficiency and log and percentage reduction
The effectiveness of an air purification system in reducing the viral load in air was assessed while the aerosol load was constantly homogenized in the room. Based on the ANSI/AHAM AC-1-2006 test method [
Anonymous. Association of Home Appliance Manufacturers Method for Measuring Performance of Portable Household Electric Room Air Cleaners. ANSI/AHAM AC-1-2006. Approved American National Standard. Available at: ANSI/AHAM AC-1-2006 - Method for Measuring Performance of Portable Household Electric Room Air Cleaners [last accessed 11 2022].
], air cleaning for smoke, fine dust and pollen can be expressed as Clean Air Delivery Rate (CADR). In line with this, the CADR was calculated for virus-loaded aerosols. The following two formulae were used:
Ctx = Ct0 ℮−κtx, where C is concentration at time point tx or t0 and κ is decay constant.
CADR (m3/h) = V(κt – κn), where V is volume test chamber, κt is test decay constant and κn is natural decay constant.
Percentage reduction was calculated based on log reduction by the formula:
P = (1-10−L) × 100, where P is percentage reduction and L is log10 reduction.
Determination of virus inactivation upon capturing in purifier system
The air purification system using ESP technology contains a collector tube for capturing the particles. The captured virus particles were analysed for remaining infectivity. This was performed by examining the ratio of the TCID50 per quantity of virus RNA determined by RT-qPCR on the FCoV virus particles collected on the collector inside the purifier system. For this, nebulized FCoV was directly released from 50 cm distance into the inlet of the active air purification system for 5 min. After virus nebulization, the fan units in the containment room were switched on and the virus containing air in the room was exchanged with fresh air by 1 h ventilation of 300 m3 of air through the HEPA filter in the exhaust of the room and additional UVC treatment for virus inactivation of environmental virus in the containment room. Then, three areas of 10 × 10 cm of the internal collector tube in the air purification system were sampled individually by wet swab to collect virus material. The swabs were each directly submerged into 3 mL of DMEM medium with 50% v/v glass beads to resuspend the viruses into the medium by vortexing for 1 min. These virus suspensions were examined for TCID50 and viral RNA content by RT-qPCR after RNA isolation. For reference, the natural stability of the virus on the collector surface was explored by nebulizing the virus on top of stainless-steel surfaces. The plates were dried in the biosafety cabinet for 15 min and then swab samples were taken after 1 h. The recovered viruses from the surface were analysed for TCID50 and total viral RNA by RT-qPCR. The difference between TCID50/fg of viral RNA ratio of the control and the collector surface of the purifier system allowed calculation of the reduction of viral infectivity.
Results
Decay of MS2 loaded aerosol in the containment room
MS2 phage decay in the containment room was measured with and without an active air purification system. Natural decay of phage presence and decay by the action of the air purification systems are shown in Table II. The natural phage MS2 decay in the room without an active air purification system appeared to be about 0.3 log units in 1 h. In the presence of each of the four types of air purification systems, A, B, C and D, we observed greater phage MS2 decay compared with the natural decay, ranging from about 1.7 to 2.3 log units or 98%–99.5% reduction in a period of 1 h.
Table IIMS2 decay averaged data and data corrected for natural decay in the room in the absence of a purifier system for 60 min
System
MS2 decay in log units
Corrected MS2 decay in log units
Corrected MS2 decay percentage
Natural
0.3 ± 0.5
0
0.0
System A
2.5 ± 0.4
2.2
99.4
System B
2.6 ± 0.3
2.3
99.5
System C
2.0 ± 0.3
1.7
98.0
System D
2.6 ± 0.2
2.3
99.5
Decay data is provided as Log and percentage reduction in 60 min.
Decay of FCoV loaded aerosol in the containment room
FCoV loaded aerosol decay in the containment room was measured with and without an active air purification system. In the absence of an active air purification system, infectious viruses were hardly detected. Only 0.8 log TCID50 per mL sample was obtained at t = 0. A decline of infectious properties of the airborne viruses at low relative humidity was previously also observed by Oswin et al. [
]. Based on quantity determination of FCoV virus in sampled air using the FCoV specific RT-qPCR, the number of genome equivalents of the virus could be detected in the room. At t = 0, the number of FCoV genome equivalents is about 9.1 log/m3 and in the absence of an active air purification system, the number reduced to about 8.7 log/m3 of air in 1 h (Figure 1). In the presence of an active air purification system, the FCoV reduction in air proceeded faster compared with the natural reduction of FCoV particles in the containment room (Figure 1, Table III). The natural reduction in the containment room is about 0.4 log units per hour. The decay attributable to the air purification systems ranges from about 1.5 to 2.3 log units or 96.8%–99.5% reduction in 1 h (Table III).
Figure 1Feline Coronavirus (FCoV) genome equivalent decay in containment room per cubic metre. Natural decay is indicated in the upper blue line with diamonds and the decay with additional action of four air purification systems A, B, C and D are given in green line, dark blue line, light blue line and dark green line, respectively.
Table IIIFeline Coronavirus (FCoV) genome equivalent decay based on RT-qPCR averaged data and data corrected for natural decay in the room in absence of a purifier system for 60 min expressed as log and percentage reduction in 60 min
Calculation of clean air delivery rate and virus inactivation
The CADR of the different air purification systems was calculated for both MS2 bacteriophage clearance (based on plaque forming units) and for FCoV particle clearance (based on qPCR quantified viral RNA). The CADR values obtained with the air purification systems for both types of particles, e.g., aerosol containing MS2 and aerosol containing FCoV, were in a range of about 169–290 m3/h (Table IV).
Table IVClean Air Delivery Rate (CADR) of purifier systems based on exposure of nebulized MS2 bacteriophage and nebulized Feline Coronavirus (FCoV) particles in separate experiments in a 50-m3 containment room
Virus could not be retrieved from HEPA matrix; no TCID50 nor qPCR amplifiable RNA was detected.
CADR is calculated based on experimental data shown in previous paragraphs. Infectivity reduction of captured FCoV virus is shown in the latter two columns.
a Virus could not be retrieved from HEPA matrix; no TCID50 nor qPCR amplifiable RNA was detected.
Apart from the CADR, it was also relevant to conclude on the fate of the captured FCoV particles. Because the aerosolized FCoV loses its infectivity at low relative humidity, the FCoV-loaded aerosol was directly inhaled by the active air purifier systems. The subsequently sampled viral material from the inner metal collector surfaces of the ESP systems showed at least a factor 1000 to 10,000 less TCID50 per fg viral RNA compared with FCoV containing aerosol sedimented on steel plates. This experiment indicated a reduction of viral infectivity of at least 3 log or 99.9% of captured viral particles by the ESP-based air purification systems (Table IV). Attempts to retrieve viral material from HEPA by grinding and vigorous shaking in DMEM appeared unsuccessful.
Discussion
The configuration of the containment room with four ventilators in the corners realizes a constant homogeneous distribution of virus loaded aerosol in the room. This was visually confirmed by releasing smoke from the point of virus release at one side of the room. By the action of the air purification systems, a constant intake of aerosol is realized and a concomitant output of purified air. The collected data shows therefore a decay of the viral load in the air as captured by the sampler over time. By correcting the natural decay of viral load in the room, the contribution of the air purification system to the decay could be determined as shown in Table II, Table III, which show the reduction of virus in 1 h. Irrespective of the type of virus, MS2 phage or FCoV, the purifiers realize a decay of about 96.8–99.5% in 1 h. Beswick et al. [
] showed similar MS2 decay data for air cleaning systems in a comparable room setting.
The decay of bacteriophage MS2 could be analysed by quantifying the number of infectious particles in the samples of air. The numbers of this virus in the air were in the range of about 108 plaque forming units per cubic metre; thus, there was enough virus to show plaque formation when seeding the particles collected from 900 L of air in 15 mL of sampling liquid with the Coriolis sampler. This results in about 108 infectious viruses in 15 mL if all viruses are collected. Despite some sampling efficiency loss, MS2 virus always showed a plaque count on plates with E. coli target bacterium. Unfortunately, infectious particles of FCoV could only be found in the samples at low TCID50 counts; the reason for this was that the number of FCoV that can be captured in 15 mL sampling liquid is about TCID50 of 100. This is close to the limit of detection with additional inactivation of FCoV because of low relative humidity of the air. Oswin et al. [
] showed that the low humidity does severely affect infectivity to about 10% of the starting value for SARS-CoV-2 over 20 min. Because FCoV has an identical structure and size [
], the fate of the virus is anticipated to be similar. An alternative method to detect and quantify the virus particles was by targeting the RNA genome of the virus by RT-qPCR. This seemed successful and yielded insight into the numbers of viral genomes sampled regardless of the virus' infectiousness. The decay of FCoV in the containment room as a result of the action of the air purification systems show a similar trend of about 2 log reduction in 1 h as roughly also observed for the decay of MS2. This trend is likely the result of the similar type of aerosols released by the MESH nebulizer that releases similar volumes of aerosols, which may be influenced by the different nature of the virus and the medium. Importantly, it provides confidence in the applied methodology and demonstrates a similar result of particle capturing of the purifier systems. However, typical differences in virus capturing were observed when having a more detailed look at the log reduction data and calculated CADR. The trend based on the CADR data in Table IV suggests a slightly better CADR for MS2 compared with FCoV-loaded particles. This may suggest a slightly different type of aerosol for the two different viruses. None the less, the ESP- and HEPA-based purifiers show a CADR range that has been previously described for medium-sized particles in the range of 0.5–3.0 μm [
]. Despite the minor differences in CADR observed when challenging the systems with two distinct types of virus-loaded aerosol, these differences show a similar trend depending on the air purification system. This is in line with the statement that the CADR permits an intercomparison of performance among various air cleaners [
] expressed the fear of re-aerosolization of infectious SARS-CoV-2 from HEPA and ESP air purification systems. They proposed in their simulated filtration performance study that UVC treatment should be added as technology to air purification systems to eradicate virus infectivity. Moreover, they stated that the biological disinfection effect on SARS-CoV-2 aerosols exposure to ions and electric field should be investigated. The latter has been carried out in this study using the surrogate virus FCoV that has identical structural and physical characteristics to SARS-CoV-2 but only has a different host range. The outcome of our study demonstrates that ESP inactivates the captured FCoV by 3–4 log units that equals a reduction of infectiousness of about 99.9% to 99.99%. Similar inactivating effects on captured bacteriophages by an ESP-based system has been reported by Kettleson et al. [
Our findings indicate that the HEPA- and ESP-based air purification systems evaluated in this study reduced the viral load in air. The captured viruses were inactivated by a percentage of 99.9–99.99% in the ESP systems that apply a strong electric field for capturing airborne particles. From this study it can be concluded that the evaluated air purification systems can be used as adjunctive infection control measure.
Conflict of interest statement
The authors declare the research was conducted to find out whether air purification systems could be used as additional infection control measure. Although the project was also financially supported by the partners Genano, Euromate, Formula Air and VFA Solutions, their support did not pose any conflict of interest as TNO executed the work as an independent research organization. The protocol and work were designed and carried out without any input from the manufacturers.
Funding sources
This work was financially supported by Health Holland, Genano, Euromate, Formula Air and VFA Solutions.
References
Nicola M.
Alsafi Z.
Sohrabi C.
Kerwan A.
Al-Jabir A.
Iosifidis C.
The socio-economic implications of the coronavirus pandemic (COVID-19): a review.
Anonymous. Preventing and mitigating COVID-19 at work: policy brief, 19 May 2021. WHO/2019-nCoV/Workplace_actions/Policy_brief/2021.1. Available at: Preventing and mitigating COVID-19 at work: policy brief, 19 May 2021 (who.int) [last accessed 11 2022].
Combined use of an electrostatic precipitator and a high-efficiency particulate air filter in building ventilation systems: effects on cardiorespiratory health indicators in healthy adults.
Comparative testing and evaluation of nine different air samplers: end-to-end sampling efficiencies as specific performance measurements for bioaerosol applications.
Anonymous. Association of Home Appliance Manufacturers Method for Measuring Performance of Portable Household Electric Room Air Cleaners. ANSI/AHAM AC-1-2006. Approved American National Standard. Available at: ANSI/AHAM AC-1-2006 - Method for Measuring Performance of Portable Household Electric Room Air Cleaners [last accessed 11 2022].
00078-6&id=gr1.jpg){kind=link}