Effect of ceiling fan in mitigating exposure to airborne pathogens and COVID-19

Abstract

Ceiling fans are the ubiquitously used electrical appliance in indoor spaces that affect the local airflow pattern and, consequently, transmission of airborne pathogens and respiratory droplets. This study numerically investigated the effect of airflow induced by the ceiling fan and ventilation rate on aerosol distribution to mitigate exposure to airborne pathogens and COVID-19. A full-scale room with a ceiling fan, natural ventilation and an occupant was modelled through transient computational fluid-particle dynamics (CFPD). To analyze the relationship between the ceiling fan rotation speed and the aerosol distribution, a ceiling fan was operated with 160, 265 and 365 revolutions per minute (RPM). The effect of the ceiling fan on particles was analyzed for particles of different sizes. The increasing ceiling fan rotation speed, the percentage deposition of the aerosol particles with diameters >40 μm was increased. The effect of different ventilation rates on aerosol distribution was evaluated. The increased ventilation rate, the percentage of the total aerosol particles flushed out was increased. The effectiveness of the mask in mitigating the exposure risk of airborne pathogens was also investigated. In combination with the natural ventilation and mask, the ceiling fan was demonstrated to have the potential to reduce airborne pathogen transmission in indoor spaces.

Introduction

The fatalities due to COVID-19 and the number of infected people from COVID-19 are increasing day by day. The ongoing pandemic of COVID-19 has infected over 339.5 million and caused ∼5.5 million deaths worldwide as of 20 January 2022.¹ The COVID-19 disease is transmitted through direct inhalation of the virus-laden aerosols/respiratory droplets or contact with contaminated surfaces.^2–4 Respiratory droplets and virus-laden aerosols are dominated mainly by short-range airborne transmission.^5,6 In contrast, the smaller droplets and aerosol particles diffuse faster and travel long distances, and may cause wider transportation of the viruses.⁷

Respiratory droplets/aerosols are generated in indoor spaces by talking, sneezing and coughing.⁸ The pieces of evidence show the probable transmission of virus-laden aerosol in indoor spaces of hospital wards,⁹ restaurants¹⁰ and offices.¹¹ In indoor spaces, the distribution of aerosol and droplets is mainly governed by the ventilation provisions and local indoor airflow patterns.^12–16 Hence, adequate provisions of outdoor ventilation for mitigation of exposure to virus-laden aerosols are recommended. Other measures to mitigate COVID-19 infection, such as social distancing and change in the operation pattern in the built environment, have also been suggested. Table 1 presents the exhaustive summary of different mitigation measures, such as social distancing, ventilation provisions and change in operation pattern in the built environment, that had been studied in the literature and the recommendations from those studies to maintain the COVID-19 appropriate indoor and outdoor environment.

Table 1.

Literature summary of different mitigation measures of COVID-19.

Mitigation measures	Authors	Description	Recommendations
Social distancing	Chu et al.100	Observational studies across 16 countries and 6 continents	- Physical distancing of 1 m or more
		Comparative studies in healthcare and non-healthcare settings
	Vos et al.101	Studied the effect of social distancing on travel pattern	- Make public transport a safer way of travelling in time of social distancing
	Liu et al.102	Predicted the effect of social distancing on COVID-19 spread using a deep neural network	- Aggressive and extensive social distancing interventions to flatten the COVID-19 pandemic
	Thu et al.103	Analyzed the relationship between social distancing measures, COVID-19 confirmed cases and deaths	- Different levels of stringent social distancing measures based on the severity of COVID-19
	Xiao et al.104	Examined the kind of spatial layouts needed and the proportion of physical isolation required on a micro scale	- Correctly planning the layout of a space - Increase the numbers and widths of entries into public spaces - Barriers installation between flows of pedestrians moving in opposite directions
	Daniel R et al.105	Measured events of physical distancing of less than 2 m, its duration and causes in a supermarket	- Increase the number of checkout points - Decrease the clients' checkout time to maintain physical distancing
	Koh et al.106	Assessed the impact of social distancing measures in containing the virus transmission empirically	- Stringent social distancing measures to reduce the reproduction number (R0) quickly
		Three categories (international travel restrictions, lockdown type measures and cancellation of mass gatherings) were analyzed
	Cowling et al.107	Examined the influence of non-pharmaceutical interventions on COVID-19 transmission in Hong Kong	- Social distancing to reduce the virus transmission rate
	Kamga et al.108	Reviewed the social distancing interventions deployed by public transit	- Barrier between driver and passengers - Physical distance of 2 m in the waiting area - Physical distance of 1m or more in the passengers' vehicle
	Chen et al.109	Assessed the effect of non-lockdown social distancing (population-wide control measures) and/or testing-contact tracing (individual-specific control measures), alone or combined, in terms of the basic reproductive number (R0) and the trajectory of the epidemic in South Korea	- Combination of social distancing and testing-contact tracing interventions for successful control of COVID-19 outbreak
	Ugail et al.110	Studied the reconfiguration of common physical environments with appropriate physical distancing measures using a well-known circle packing problem	- Use of the circle packing problem algorithm to reconfigure the common physical spaces
	Fu et al.6	Estimated the interpersonal viral exposure using inhalation fraction and breathing thermal manikins	- Increase in the interpersonal distance due to the findings of reducing inhalation fraction with an increment of interpersonal distance
Ventilation provisions	Dao et al.111	Analyzed ventilation system in hospital isolation rooms with infectious patients	- Exhaust outlet is placed above the patient’s head to enhance the performance of ventilation with higher droplet removal efficiency
		Analyzed the effects of various air outlet positions on the removal efficiency of infectious droplets in an isolation room
	Srivastva et al.112	Studied the effect of air disinfection systems on removing the COVID-19 infection risk in office buildings	- 100% outside air combined with air disinfection system with the existing HVAC system
	Bazant and Bush113	Provided guidelines to limit indoor airborne transmission of COVID-19 in the indoor environment	- Increase in the fresh air supply delivered by an HVAC system - Use of fans to promote airflow - Installation of clean filters and up-gradation to more effective filters
	Berry et al.114	Reviewed the methods to reduce the probability of the airborne spread of COVID-19 in ventilation systems and enclosed spaces	- Ventilation of an enclosed space either through opening a window or increasing the fresh air supply for an HVAC system - Ventilation in combination with air filtration, UVGI, air ionization, non-thermal plasma inactivation and chemical disinfectants
	Li et al.115	Studied experimentally the reduction in infection risks with ventilation rate in hospital	- Increase in air change rate and application of air cleaner to dilute/filter the viral aerosols in the waiting rooms
	Motamedi et al.116	Used CFD to model airborne pathogen transmission of COVID-19 in confined spaces under different ventilation strategies	- Cross ventilation and mechanical ventilation were more effective than the single sided and no ventilation
		Ventilation strategies included: Single sided ventilation, cross ventilation, mechanical ventilation and no ventilation
	Barrio et al.117	Studied the effect of natural ventilation in maintaining the indoor environment of school buildings before and after COVID-19 during the heating season	- Mechanical ventilation with heat recovery should complement natural ventilation - All schools should prepare a plan to progressively improve the indoor environment through an upgrade of windows openings and allowing natural ventilation
	Elsaid et al.118	Studied indoor air quality (IAQ) by improving the ventilation process, air-conditioning systems and their components to confront the emerging situation of COVID-19	- Avoid recirculation of air in the HVAC system - Feed 100% of total fresh air to HVAC - Negative pressure maintenance in ventilation system for COVID-19 infected patients' wards
	Beaussier et al.119	Aerodynamically analyzed hospital ventilation according to seasonal variations to mitigate the effect of COVID-19	- Temperature difference between both sides of the hospital is a more important factor of airflow contamination than ventilation systems and should be taken into account before deciding the location of infected patients on the ward
	Park et al.120	Reported the effect of natural ventilation on COVID-19 exposure risk and energy consumption in a school building	- Cross ventilation to minimize the infection possibility in high density public buildings - If cross ventilation is not possible, it is advised to use auxiliary fan to achieve the same effect as cross ventilation
	Lipinski et al.121	Investigated the role of different ventilation strategies such as displacement ventilation, natural ventilation and naturally assisted ventilation in mitigating the infection risk of COVID-19	- Displacement ventilation strategies, natural or naturally assisted ventilation strategies to provide an effective starting point for reclaiming the building safe for use
	Lee et al.122	Investigated the effect of air cleaner on reducing concentrations of indoor-generated viruses with or without natural ventilation	- Operation of the air cleaner along with open windows and doors
	Burridge et al.123	Reviewed the different ventilation strategies in buildings to mitigate COVID-19	- Ventilation as the primary measure to control airborne transmission in indoors
	Zhou et al.124	Numerically investigated the airborne infection in naturally ventilated hospital wards with central-corridor type	- Natural ventilation is not recommended in central-corridor type hospital ward
	Zeong et al.125	A vertical laminar airflow system has been investigated through CFD and an experimental approach to prevent aerosol transmission of SARS-CoV-2 in building space	- With optimized parameters vertical laminar airflow system is recommended to remove aerosol from the room
	Zheng et al.126	Used statistical methods to evaluate the interventions for respiratory disease transmission on cruise ships	- Increased air change rate in enclosed spaces - High efficiency particulate air filters and ultraviolet germicidal irradiation devices in the ventilation system
	Wan et al.127	Investigated four ventilation strategies Unidirectional–downward Unidirectional–upward, single-side–ceiling and single-side–floor, using a multiphase numerical model to mitigate the effect of COVID-19	- Unidirectional system over single sided ventilation, which is more effective in containing the droplets - Floor supply ventilation over ceiling supply due to expiratory droplets and droplet nuclei’s longer vertical settling time
	Mirikar et al.128	Numerically investigated the transmission of virus-laden droplets inside a conference room	- Higher air change rate due to the direct relationship between high ventilation rate and the number of droplets extracted from the outlet vent
	Burridge et al.129	Estimated the airborne infection risk using monitored CO2	- Well ventilated spaces due to its unlikely significant contribution to airborne infection
	Cravero et al.130	Numerically investigated the COVID-19 carrying droplets exhaled by coughing and breathing	- Masks with different ventilation strategies
	Ye et al.131	Reviewed the guidelines for the operation and management of HVAC systems issued by Chinese governmental departments and professional institutions during the pandemic of COVID-19	- Sufficient fresh air through mechanical ventilation and/or natural ventilation for non-medical buildings - Avoid the use of return air as much as possible when a building is occupied - Reasonable natural ventilation scheme, even if the room is equipped with a mechanical ventilation system
	Harmon et al.132	A web based application tool was developed and used to compare the different indoor risk mitigation strategies by estimating airborne transmission risk	- Increased ventilation rate and high relative humidity (30–40%) due to its potential of reducing the additional people being infected
	Xu et al.14	Studied the performance of airflow distance from personalized ventilation on personal exposure to airborne droplets from different orientations	- Optimization of parameters such as the relative position of the personalized ventilation terminal, the infected and healthy person and the distance of personalized ventilation from the healthy person
Operational patterns	Guo et al.133	Compares HVAC related guidelines during the pandemic from various countries and regions, including those issued by ASHRAE, REHVA, SHASE, Architectural Society of China and the Chinese Institute of Refrigeration	- Operation of HVAC system 2 h before and after occupancies - Negative pressure in toilets - Lower speed operations of HVAC to ensure an effective circulation of air that removes the virus from the building with limited energy penalties
	Awada et al.134	Reviewed the occupant health in buildings during normal operations and COVID-19	- New HBI strategies, such as voice activated elevators, doors and water fountains, hands-free light switches and thermostats, surfaces with antibacterial fabrics and finishes AI enabled real-time sensing, learning, decision making and prediction
	Zheng et al.135	Analyzed the ability of HVAC systems to control the transmission of SARS-CoV-2 and corresponding energy impacts of HVAC system	- Auxiliary equipment such as HEPA filters, UVGI systems and heat recovery devices
	Krarti et al.136	Reviewed the electricity demand in residential buildings due to COVID-19-induced lockdown and suggested operation changes for the built environment	- A wide range of energy efficiency and renewable energy solutions are recommended to reduce the energy consumption of residential buildings - Energy efficiency solutions: LEDs, smart power strips, smart thermostats, intelligent shades and blinds and high star rated electrical equipment - Renewable energy solutions: BIPV, solar thermal cooling, solar electrical cooling, solar thermal collector and photovoltaic thermal collector (PV/T)
	Ding et al.137	Qualitatively analyzed the operation of the HVAC system for environmental control to minimize the COVID-19 infection	- Recommendation of the use of HEPA filter, UVGI system, intelligent control of the humidity and temperature in the indoor environment
	Wang et al.138	Studied the surface distribution of SARS-CoV-2 in Leishenshan hospital in China	- Air disinfection, object surface cleaning and disinfection, instrument equipment surface disinfection and hand hygiene have been recommended as reliable and useful

Investigating the characteristics of the viral-laden respiratory droplets under different ventilation strategies and airflow conditions in confined and crowded spaces such as lifts, restaurants, classrooms, hospitals and toilets is of most importance for preventing the super spreading events, for these places are mostly over-occupied and inadequately ventilated. The following paragraphs explain the different ventilation strategies used in the practical settings of public spaces and their effect on respiratory droplets.

The characteristics of the respiratory droplets in the confined space of the elevators have been studied, and vertical evaporations (from top to bottom in decreasing order) were reported because of the droplets of smaller size at the top due to buoyancy.¹⁷

The face-to-face seating was found to increase the risk of being infected by the sneeze of the person sitting across the table in a numerical investigation of the transmission of respiratory droplets caused by sneezing in a small cafeteria with six tables and one counter.¹⁸ Furthermore, in the same study, the effect of airflow induced by air conditioning on the transmission of respiratory droplets was investigated. The reported study found that the horizontal air conditioner could cause vortices inside the cafeteria and could spread the droplets transversely. Lastly, proper air distribution, such as vertical airflow supply for indoor activities, was suggested to restrain the spread of respiratory droplets.

The fate of airborne bacteria and viruses in negatively pressurized indoor spaces was investigated,^19,20 the larger number of air change rates in negatively pressurized indoor spaces was reported could lower the concentration of bacteria and viruses.

Transmission and deposition of aerosol in a realistic air-conditioned classroom with nine students and an instructor was investigated using the Eulerian-Lagrangian approach²¹ and Re-Normalization Group (RNG) k-ε turbulence model.²² The reported results showed that the distribution of aerosol in the room is non-uniform and depends on the layout of the air-conditioning system.

The virus transmission in a crucial public space, that is, public toilets, was investigated.^23,24 These studies found that the virus-laden particles could reach up to 1 m from the ground during flushing. This could cause a potential massive infection through public toilets.

The airborne transmission of the virus in indoor space is more than the virus spreading outdoors due to the low-level turbulence in the room.²⁵ Inferring from the above studies, effective mitigation measures require an increase in the amount of clean outdoor air to be supplied into indoor spaces.²⁶ This can be achieved by supplying the outdoor air in the ventilation system of the heating ventilation and air-conditioning (HVAC) system or through natural ventilation.²⁷ In all the above-mentioned studies, the characteristics of the respiratory aerosols that were studied under the conventional air-conditioning systems suggested that the vertical airflow for indoor activities could restrain the spread of respiratory droplets. The vertical supply of airflow can be achieved efficiently through a ceiling fan.

In tropical and subtropical countries, occupants mainly deploy natural ventilation with a ceiling fan to achieve thermal comfort.²⁸ However, the effect of airflow induced by the ceiling fan on the airborne transmission of pathogens and the tradeoffs between the different ceiling fan rotation speeds combined with the different ventilation provisions with regard to respiratory aerosol distribution and virus transmission has not been extensively studied. As per the authors’ knowledge, only one study²⁹ in the literature reported the effect of the airflow induced by the ceiling fan on virus-laden aerosol transmission. The limitation of the reported study is that it was performed under the steady-state of ceiling fan induced airflow conditions, whereas, in real life, for the effective mitigation strategy of COVID-19 or airborne pathogens, there is a need to know the spatial as well as temporal behaviour of the virus-laden aerosols and droplets affected due to the rotation of the ceiling fan. Therefore, this study aimed to elucidate the behaviour of the aerosol distribution in the indoor space under the influence of the ceiling fan and natural ventilation numerically through transient computational fluid and particle dynamics (CFPD).

The effect of the ceiling fan on aerosol distribution in a naturally ventilated indoor space was investigated, and the role of fan rotation speed and natural ventilation rate in mitigating airborne transmission was analyzed. For this purpose, a real-size room of a residential building with a ceiling fan, natural ventilation provision and a cuboid occupant’s body was modelled. The ceiling fan rotation speed was varied by keeping the ventilation rate constant to analyze the relationship between the fan rotation speed and aerosol distribution pattern. Similarly, the natural ventilation rate was varied by keeping the ceiling fan rotation speed constant to analyze the flushed out concentration of the aerosol from the indoor space to the outdoors with the ventilation provisions. Lastly, the effect of wearing a mask on aerosol distribution under the ceiling fan and natural ventilation was also investigated. This study provides a better understanding of the role of the ceiling fan and natural ventilation in mitigating the airborne pathogen transmission in the built environment and contributes to the future design of the ceiling fan for an integrated built environment.

Methods

Room model, mask schematic and spatial mesh

A three-dimensional model of a realistic room comprising an occupant, a ceiling fan and two windows (one for the inlet and the other for the outlet) was selected for this study, similar to that reported in the literature.³⁰ The computational model of the room is shown in Figure 1(a). The selected room has dimensions of 3.6 × 3.6 × 3.2 m³. There are two windows in the room with dimensions of 1.5 × 0.5 m² and 2 × 1.5 m², respectively. The ceiling fan chosen has a diameter of 1.3 m with a hub height of 0.047 m, hub diameter of 0.24 m and rake angle of 8°. The ceiling fan was placed at a height of 2.7 m from the floor of the room. This model considered an occupant in a sitting position (0.65 m) near one side of the wall (the usual sitting place in a room). The occupant was modelled as a cuboid body (0.5 × 0.25 × 1 m³) and a cuboid head (0.15 × 0.15 × 0.2 m³) through which particles and air were generated into the room. The simplified cuboid body was selected due to its practice available in the literature.³¹ No other obstacles were considered in the model for simplicity. To model the airflow induced by the ceiling fan, the computational domain was divided into two sub-domains: the rotating domain and the stationary domain. The criteria of rotating domain selection and the methodology of the ceiling fan modelling technique are described in our previous paper.³² The unstructured tetrahedral mesh element was used for both the stationary and rotating domains (Figure 1(b)). Since there is an interface between the stationary and rotating domain, a sharp velocity change would occur at the interface. To capture the flow induced by the ceiling fan, the body sizing element was used for the rotating domain to refine the mesh element further. Face sizing refinement was employed for meshing of the occupant.

Figure 1.

Solid model of the computational domain and the computational mesh used in the CFD simulation.

Furthermore, the effect of a mask to mitigate the airborne pathogens in the presence of a ceiling fan in the room was studied by constructing a rectangular-shaped virtual mask at a distance of 5 mm from the mouth of the occupant. Figure 2 shows the schematic diagram of the chosen mask. The mask has an area of 160 × 20 mm², and the depth of the mask was kept as 2 mm. The placing of the mask, in this case, emulated the scenario where the mask was usually sealed at the bottom of the chin but remained open near the cheeks and nose.

Figure 2.

Schematic of the mask arrangement in the computational domain.

Numerical models

Numerical model for ceiling fan and aerosol particles dynamics

Ansys CFX³³ was used to simulate the transient particle behaviour exhaled from occupants under the airflow induced by the ceiling fan. Airflow induced by the ceiling fan was predicted first by solving the continuity and momentum equation independent of the discrete phase using the steady-state Navier-Stokes equations in conjunction with the shear stress transport (SST) turbulence model.³⁴ The SST model was selected due to its property of combining the strengths of the k-ε and k-ω models. This can deal with higher and lower Reynolds numbers and accurately model the boundary layers and the flow separation under adverse pressure gradient conditions, which is likely to arise due to airflow induced by the ceiling fan. The governing equations for the incompressible airflow is given as equations (1) and (2)

$$\frac{\partial u_j}{\partial x_j}=0$$ (1)

$$\frac{\partial u_i}{\partial t}+\frac{\partial \left(u_i{u}_j\right)}{\partial x_j}=-\frac{1}{⍴}\frac{\partial p}{\partial x_j}+\frac{1}{⍴}\frac{\partial {\tau }_{ij}}{\partial x_j}+g_i$$ (2)

where $u_j$ is the fluid velocity, $p$ is the pressure, $⍴$ is the density and $g_i$ is the gravitational acceleration. The viscous term ${\tau }_{ij}$ is defined by equation (3), as

$${\tau }_{ij}=\mu \left(\frac{\partial u_i}{\partial x_j}+\frac{\partial u_j}{\partial x_i}\right)$$ (3)

where $\mu$ is the dynamic viscosity. Once the continuous phase of airflow solution converges, the flow field is then frozen, and is used to model the discrete phase particle model (aerosol model). The trajectory of the particles can be predicted by solving equation (4), the translation of the discrete phase equation given as

$$\frac{d{m}_d{u}_{d,i}}{dt}=F_i^D+F_i^A+F_i^G$$ (4)

where $F_i^D$ is the drag force between the air and the particle. $F_i^A$ represents the other additional forces, including the pressure force, virtual mass force, Basset force, Brownian force and Saffman’s lift force. $F_i^G$ denotes the gravity force. As the droplets due to human sneezing are sufficiently small in the range of 1–300 μm,³⁵ the virtual mass and pressure forces are neglected. The turbulent dispersion of particles and the random effects of turbulence on particle dispersion are taken into account using the particle dispersion model available in CFX. Drag force on the particles is modelled due to the smaller size of the particles than the size of the mesh. The drag force is determined by equation (5), as

$$F_i^D=\frac{1}{8}\pi ⍴d_d^2{C}_D\left(\overrightarrow{u}-\overrightarrow{u_d}\right)\left|\overrightarrow{u}-\overrightarrow{u_d}\right|/C_c$$ (5)

where $⍴$ is the particle’s density. $d_d$ is the particle’s diameter, $C_c$ is the Cunningham correction factor.³⁶$u$ and $u_d$ are fluid and particle’s velocity, respectively. The drag coefficient ($C_D$) is defined by equation (6), as

$$C_D=a_1+\frac{a_2}{R_{ed}}+\frac{a_3}{R_{ed}^2}$$ (6)

where the constants $a_1$, $a_2$ and $a_3$ are determined by the particles' Reynolds number ($R_{ed}$) that depends on the particles' diameter.

The particles' size distribution exhaled from the occupant was modelled through the Rosin-Rammler³⁷ approach. In this approach, the inject point, that is, the occupant’s mouth, was modelled by seeding a different range of particle radii by invoking a presumed probability density function. The Rosin-Rammler distribution was determined by equation (7)

$$f\left(r\right)=\frac{{qr}^{q-1}}{{\overline{r}}^q}\mathit{exp}\left[-{\left(\frac{r}{\overline{r}}\right)}^q\right]$$ (7)

where $q$ and $\overline{r}$ are the exponential factors and average radius, respectively, which are based on the aerosol injection flow rate as an input parameter for the considered seeding particles. In this study, the droplet’s material was sodium chloride (NaCl) in a liquid thermodynamic state. The molar mass and density of the sodium chloride were taken as 58.44 kg/kmol and 2160 kg/m³, respectively. The particles of diameter 1–50 $\mu$ m were selected, for 1–50 $\mu$ m particles accounted for most of the particles from a cough.³⁸ With an injection rate of ∼1200 particles per second, a total of 44,536 particles were injected into the room.

Numerical model for the mask

The mask was modelled as a permeable material. Permeability of the mask is the property through which aerosol particles escape from the mask. The transmissivity of the particles from the mask depends on its permeability and pressure gradient across the mask, as given by equation (8)

$$\frac{dp}{{dx}_i}=-\frac{\mu }{K_{perm}}{U}_i-K_{loss}\frac{\rho }{2}\left|U\right|U_i$$ (8)

where $K_{perm}$ is the permeability coefficient and $K_{loss}$ is the resistance loss coefficient. $\mu$ is the viscosity of the fluid. $\rho$ is the density of the fluid. $U$ is the fluid velocity. $K_{perm}$ is defined by equation (9)

$$K_{perm}=\frac{{D_p}^2{\epsilon }^3}{150{\left(1-\epsilon \right)}^2}$$ (9)

where $D_p$ is the equivalent spherical diameter of the particle. $\epsilon$ is the volume porosity of the material. In this study, $\epsilon$ was selected as 0.86⁸⁰ for the N95 mask. The resistance loss term ($K_{loss}$) which governs the inertia is represented by equation (10)

$$K_{loss}=\frac{3.5\left(1-\epsilon \right)}{D_p{\epsilon }^3}$$ (10)

Using the spherical diameter ($D_p$) of the particle, that is, 30 μm (Rosin-Rammler size considered in this study) and volume porosity ($\epsilon$) in equations (9) and (10), the permeability and loss coefficient were calculated.³⁹

Boundary conditions

In order to define the boundary conditions at the inlet and outlet, the considered room was simulated using the EnergyPlus⁴⁰ for the site of Mumbai climatic conditions. The simulation shows that the window with a higher dimension act as an inlet to the room, and the window with a lower dimension act as an outlet. The inlet boundary condition was set as velocity with a normal speed of air as 0.11 m/s which corresponded to the ∼29 air change per hour (ACH) of the room. The outlet boundary condition was taken as a pressure outlet with the relative pressure equal to 0 atm. The ceiling fan rotation speed was set as 160 RPM, 265 RPM and 365 RPM emulating as low, medium and high fan rotation speed of the typical residential ceiling fan. The mouth of the cuboid body was considered as an injector of the aerosol particles following the Rosin-Rammler distribution function with a Rosin-Rammler size of 30 μm and Rosin-Rammler power of 2. The particles were injected into the room from the mouth of the cuboid body with a normal speed of 5 m/s, which was comparable to the velocity of air expelled by coughing or sneezing.⁴¹ The boundary condition for all the wall surfaces was set as no-slip wall boundary conditions.

Computational grid-independence study

A grid-independence study was conducted to check the credibility of the numerical results. To perform the grid independence test, the velocity profile of the airflow induced by the ceiling fan at the height of 1.1 m from the plane of the ceiling fan was plotted with respect to different numbers of mesh elements. Figure 3 shows the velocity profile for the mesh elements of 1.4 million, 0.7 million and 0.4 million. The figure shows that for the mesh elements of 0.7 million, there is not a very significant difference in velocity profile obtained with 1.4 million mesh elements. Compared to the finest grid of 1.4 million, the differences between velocity profiles obtained on the grid size of 0.7 million and 0.4 million were 5.49% and 11.78%, respectively. Therefore, the present results on the grid size of 0.7 million were considered grid-independent.

Figure 3.

Grid-independence study: spatial distribution of the ceiling fan induced velocity profile at the height of 1.2 m from the plane of the ceiling fan. On the x-axis, 0 represents the centre of the ceiling fan. In the figure, M represents a million.

Study design

The present study investigates the effect of airflow induced by the ceiling fan and outdoor air ventilation on the distribution of exhaled aerosol particles (emulating the droplets containing the viral load). The distribution pattern of the aerosol particles was also investigated in the presence of a mask. The aerosol source considered here was an occupant (assumed in a sitting position near a wall (0.65 m from the wall). The effect of airflow induced by the ceiling fan on aerosol was studied by changing the fan rotation speed from low to high (in this study, low = 160 RPM, medium = 265 RPM and high = 365 RPM) and observing the aerosol deposition and settling time. The effect of ventilation through windows was also investigated by changing the ventilation rate from low to high. For low, medium and high ventilation rates, the normal air velocity from the inlet was set as 1, 2 and 3 m/s, respectively.

Particles’ size is the most important determinant of aerosol behaviour. According to aerodynamics of the aerosol particles’ diameter, the aerosol particles were divided into PM₁₀ (diameter <10 μm), PM_2.5 (diameter <2.5 μm), PM₁ (diameter <1 μm) and nanoparticles.^42,43 According to WHO, respiratory particles were divided into aerosols and droplets according to the particles’ sizes. In this study, respiratory particles with a diameter <5 μm were considered as aerosols, and those with a diameter >5 μm as droplets.⁴⁴ However, the dichotomy of 5 μm between the aerosols and droplets is not supported by the novel aerosol science. This definite criteria between aerosols and droplets undermine the notion of transmission of COVID-19.^45,46 The variability of transmission among respiratory pathogens appears to be less dependent on the physical particle size. The transmission is affected more by biological factors such as the size of the emitted inoculum, the ability of the pathogen to survive desiccation and other stresses of aerosolization and airborne transport. It also depends on environmental factors such as air movement, temperature and humidity.⁴⁷ During coughing and sneezing, droplets of different sizes (100–1000 μm) are formed,⁴⁸ and they degenerate into the range of 1–100 μm particles in a short span of time.⁴⁹ Therefore, in this study, the effect of airflow induced by the ceiling fan on the behaviour of the particles was analyzed by segregating the particles into bins of 10 μm intervals.

The behaviour of the aerosol particles in the presence of the mask was also studied. To emulate the effect of the mask, a thin porous layer was created near the occupant’s mouth.

Results and discussion

Model validation

Firstly, to validate the numerical model of the ceiling fan, an experimental case study⁵⁰ was modelled. The ceiling fan induced airflow predicted by the CFD model was compared with the experimental data. The detailed air velocity distribution around a ceiling fan (20 cm below and above the ceiling fan plane) in a large experimental room for the downward and upward ceiling fan induced airflow was measured. Figures 4(a) and (b) depict the comparison of the experimental and numerically predicted results at 20 cm above and below the ceiling fan for the downward and upward ceiling fan induced airflow, respectively. The figure shows that the CFD simulations reproduced the ceiling fan induced airflow well and characterized the flow field around the ceiling fan regardless of the flow direction. The differences of 85% between the velocity components obtained through the numerical model and experimental results were less than 0.5 m/s. The maximum discrepancies were shown in the w-component of the velocity profile. As a whole, considering the differences between the experimental and CFD results, the numerically predicted values of the velocity components are reasonably good and acceptable. To model the ceiling fan setup, a rotating reference frame method was used, which shows sufficient agreement with the experimental data in predicting the experimentally measured velocity profile. The numerical modelling of the ceiling fan has been discussed in detail in our previous work.³⁴

Figure 4.

Comparison of different components of ceiling fan induced air velocity: (a) at 20 cm below the fan plane for the downward flow; (b) at 20 cm above the fan plane for the upward flow. In the figure, filled markers represent the experimental data, and unfilled markers show numerical data.

To validate the CFPD model for predicting the concentration of the aerosol particles, the trend of saliva droplets’ size distribution⁵¹ was considered. The trend of distribution was selected due to the unavailability of the experimental data of particles' dynamics under the influence of the ceiling fan. Figures 5(a) and (b) exhibit the particles' size distribution obtained through the CFPD model and reported in literature,⁵¹ respectively. The figure shows that the CFPD model considered in this study can predict well the distribution of the particles in a similar fashion as reported in the literature.⁵¹

Figure 5.

Particles’ size distribution: (a) predicted by the CFPD model; (b) reported in the literature.⁵¹

Airflow characteristics of the ceiling fan

The velocity distribution under the influence of a ceiling fan is essential information to understand the behaviour of aerosol particles' transport. Figure 6 shows the airflow distribution pattern at the central plane of the room at three fan rotations, which are 160 RPM (Figure 6(a)), 265 RPM (Figure 6(b)) and 365 RPM (Figure 6(c)). As the fan rotation speed was increased, the airflow induced by it was shown to converge towards the centre of the ceiling fan and the strong downward airflow from the ceiling fan would form a cone on the floor. The regions of different air velocity intensities would be created due to the ceiling fan. At the cylindrical region below the ceiling fan, the airflow speed is the highest, and as the distance from the tip of the fan’s blade towards the walls was increased, the air velocity would be reduced. From the flow pattern, if the occupant sneezes in the cylindrical region of the ceiling fan, there is a high chance that the aerosol droplets would settle down to the surfaces quickly. In contrast, the settling time may increase when the aerosol is exhaled in other regions. Also, there is a chance of the droplets being transported to other regions within the indoor environment. The effect of airflow on particle dynamics is discussed in the following sections.

Figure 6.

Airflow pattern at the central plane of the room induced by the ceiling fan rotating at speeds: (a) 160 RPM, (b) 265 RPM and (c) 365 RPM.

Particle dynamics under ceiling fan induced airflow

Figure 7 shows the three-dimensional distribution of the aerosols at the time elapsed of 4 s, 30 s and 60 s under different airflows induced by the ceiling fan rotating at 160 RPM, 265 RPM and 365 RPM. After 4 s from the aerosol release, the particles acquired a parabolic trajectory. The particles slowly disperse with time. Some of the particles settled down on the floor under the airflow induced by the ceiling fan. Some of the particles were raised up with the circulation of the airflow. In the figure, the particles' tracking is shown through the trajectory lines. The colour of the trajectory lines is proportional to the particles' diameter. The balls' size that is shown in the figure is in proportion to the particles' diameter, whereas the colour of the balls corresponds to the particles' travelling time. The travel time of particles with a larger diameter was less compared to the particles with a smaller diameter. As the rotation speed of the ceiling fan was increased, the parabolic path of the particles was reduced. This shows that when the rotation speed of the ceiling fan was increased, the particles with larger diameters settled down quickly by flowing towards the floor and dispersing along the floor. In contrast, the particles with less diameter were transported to different places under the airflow induced by the ceiling fan. This was due to the negligible effect of gravity and inertia forces on small aerosol particles (Dia. ≈ ≤40 μm) in comparison to the influence of the indoor airflow, whereas the large-sized aerosol particles (50 ≈ ≤ Dia. ≈ ≤200 μm) were more affected by the gravity and inertia forces.^52–54 The time and rotation-wise observations show that when the rotation speed of the ceiling fan remained low, most of the particles settled down on the floor with the time, whereas a few particles of diameter less than 10–13 μm remained in the air even after 60 s. Similarly, when the rotation speed was 265 RPM and 365 RPM, the particles of ≤37 μm diameters remained in the air even after 60 s. These findings are consistent with the data reported in the literature.^55–57 The literature also reported that airflow would prevent small airborne droplets from settling, and this would become stronger with higher airflows. The figure also shows vortices were created due to the airflow induced by the ceiling fan near walls. Particles were rising with the air near the vortices. Some particles were trapped in the air recirculation zones, and a few particles were transported to the ceiling fan’s rotating domain were redistributed within the room with the ceiling fan’s descending airflow.

Figure 7.

Particle distribution under different airflow induced by the ceiling fan. The first, second and third row corresponds to the ceiling fan rotation speed of 160 RPM, 265 RPM and 365 RPM, respectively. The first, second and third column corresponds to the time elapsed of 4 s, 30 s and 60 s, respectively.

The evaporation of the droplets was not considered in this study; however, in reality, the droplets of small diameters would evaporate up to the time of 60 s.⁵⁸ At 25°C, the evaporation time for small droplets (diameter ∼140 μm) would be about 6 s, which was increased to 27 s for larger size droplets (diameter ∼1.4 mm).⁵⁹

Effect of ceiling fan induced airflow on particles deposition

Figure 8 shows the effect of airflow induced by the ceiling fan on particle deposition. The figure shows the percentage of deposition for total particles and particles of diameter >35 μm. The droplets with diameters <30 μm experience negligible gravity and inertia.⁶⁰ Our qualitative analysis of particles (explained in previous section) also shows that particles with diameters <30 μm remain in the air even after 60 s.

Figure 8.

Effect of different fan rotation speeds on particle deposition.

The percentage of the total particle deposition shows that as the airflow induced by the ceiling fan was increased (through increment in ceiling fan rotation speed), the percentage of particle deposition would be reduced. This is due to the drag force created by airflow recirculation in the room. Due to the very low gravitational force acting on particles, particles of small diameter move with the airflow recirculation. As the airflow recirculation was increased due to the high rotation speed of the ceiling fan, a more significant number of particles that become unstable would start moving with the airflow recirculation, whereas for particles with a higher diameter (i.e. >35 μm), the percentage of particle deposition would be increased as the airflow was increased. This is because the downward airflow induced by the ceiling fan would force these particles towards the floor. Due to their higher weight, particles with higher diameters would be unable to move with the air recirculation.

Figure 9 shows variation in the deposition of particles of different sizes with the different ceiling fan speeds (low, medium and high). The deposition of particles of relatively smaller sizes, that is, <40 μm, was inversely related to the ceiling fan rotation speeds. The figure also shows that the deposition of particles of relatively bigger sizes, that is, >40 μm, was directly related to the ceiling fan rotation speeds. Further, the figure shows that the percentage of the particles' deposition was increased with the increasing diameter of the particles at all different fan rotation speeds. Table 2 shows the percentage increase or decrease in the deposition of particles as the ceiling fan rotation speed was varied from low to high. The negative and positive signs represent the decrease and increase in the particles' deposition percentage, respectively. The analysis of the table showed that the deposition of particles of smaller sizes was decreased with the increasing of ceiling fan rotation speed, whereas the deposition of particles of bigger sizes was increased with the increasing of ceiling fan rotation speeds.

Figure 9.

Effect of different ceiling fan rotations (low, medium and high) on the deposition of particles of different sizes.

Table 2.

Percentage increase and decrease of the deposition of the particles at different ceiling fan rotation speeds.

Particles size	Percentage increase or decrease of the deposition of the particles
	Low $\to$ Medium	Low $\to$ High	Medium $\to$ High
<10 μm	−50.36	−67.12	−33.72
10 to <20 μm	−60.62	−78.67	−45.83
20 to <30 μm	−14.68	−54.8	−47.02
30 to <40 μm	−2.34	−22.27	−20.40
40 to <50 μm	5.24	14.07	8.39
50 to <60 μm	5.32	14.57	8.78
60 to <70 μm	8.49	12.79	3.96

Figure 10 exhibits the temporal variation in the deposition of particles >35 μm under the effect of different fan rotation speeds. In the beginning, there was less percentage deposition on the floor; however, the percentage of deposition was increased with the advancement of time. The percentage deposition was increased with the fan rotation speed throughout the simulation period for the low and medium fan rotation speeds. In contrast, at higher fan rotation speeds, the percentage deposition of particles remained low at the beginning. This was due to strong airflow recirculation, which carried the particles . After 40 s, the percentage deposition under high fan rotation speed started increasing. This was because the downward flow of the ceiling fan, which diverged near the floor, ascended near the wall and again converged to the ceiling fan’s rotating domain (Figure 7). At a higher fan rotation speed, this airflow pattern became stronger. The particles with less diameter were transported with this air current, reached the ceiling fan rotating domain and descended on the ground with the downward flow of the ceiling fan. Therefore, the percentage deposition under high fan rotation speed was increased with time compared to the medium and low fan rotation speeds.

Figure 10.

Temporal variation in particles' deposition under the influence of airflow induced by the ceiling fan rotating at different speeds.

Figure 11 shows the effect of airflow induced by the ceiling fan on the average travel time of total particles. As the airflow was increased due to the increment in ceiling fan rotation speed, the overall time travel by the particles was reduced. After the period of ∼20 s, the travel time of particles was saturated. This means that particles started settling down on the floor after the time elapsed of ∼20 s. At the high fan rotation speed, the average travel time of the particles was the lowest (approximately ∼4 s). The average travel time was observed with medium and low fan rotation speeds as 5.1 s and 5.7 s, respectively. This indicates that the higher fan rotation speed forced the particles to settle down quicker than the medium and low fan rotation speeds.

Figure 11.

Effect of fan rotation speed on the particle travel time before deposition.

Figure 12 shows the average travel time of particles of different sizes with different ceiling fan rotation speeds (low, medium and high). The figure shows that the travelling time of particles of size <40 μm was increased as the rotation of the ceiling fan was increased, and the travelling time of the particles of size >40 μm was decreased as the ceiling fan’s rotation speed was increased. Table 3 shows the percentage increase or decrease of the travelling time of particles of different sizes with different ceiling fan rotation speeds. The negative and positive signs represent the decrease and increase in the particles' deposition percentage, respectively. From the analysis of the table, the higher fan rotation speeds forced the particles of larger sizes to settle down quickly, whereas the settling time of particles of smaller sizes was increased with the increasing fan rotation speeds.

Figure 12.

Effect of different ceiling fan rotations (low, medium and high) on travelling time of particles of different sizes.

Table 3.

Percentage increase or decrease in the settling time of the particles of different sizes with different ceiling fan rotation speeds.

Particles size	Percentage increase or decrease of the settling of the particles
	Low $\to$ Medium	Low $\to$ High	Medium $\to$ High
<10 μm	13.9	21.29	6.49
10 to <20 μm	15.36	25.74	8.99
20 to <30 μm	15.20	22.99	6.76
30 to <40 μm	3.44	9.64	5.99
40 to <50 μm	−4.88	−66.09	−64.37
50 to <60 μm	−2.88	−73.67	−72.89
60 to <70 μm	−13.64	−68.66	−63.71

Effect of the ventilation rate on particles’ concentrations

Figure 13 shows the temporal variation of the percentage of particles flushed out from the room with different ventilation rates. To understand the effect of ventilation rate on dispersion of exhaled aerosol particles, the ceiling fan rotation speed was kept constant at 365 RPM, and the inlet velocities with respect to low, medium and high ventilation rates were set as 1, 2 and 3 m/s, respectively. To determine normal velocities to inlet as boundary conditions, the Wells-Riley equation was used, as given by equation (11)

$$P=1-\exp \left(\frac{-Iqpt}{Q}\right)$$ (11)

where I is the number of infectors, q is the quanta generation rate (quanta/h), p is the pulmonary ventilation rate for a person (m³/h), t is the exposure time interval (h) and Q is the room outdoor ventilation rate (m³/h). To calculate the normal velocity, first room outdoor ventilation rate Q was calculated by making the infection probability zero for 1 h exposure time interval, and with the values of I = 1, q = 142 (quanta/h) and p = 0.92 (m³/h). After calculating Q, it was divided by the area of the inlet windows to obtain the inlet velocity for the boundary condition of the CFD simulation. The infection probabilities corresponding to 1, 2 and 3 m/s were observed as 0.02, 0.008 and 0.005, respectively.

Figure 13.

Effect of ventilation on particle flushed out behaviour.

As the ventilation rate was increased from 1 m/s to 3 m/s, the percentage of particles flushed out from the room was increased. The percentages of particles flushed out from the indoor space corresponding to the low, medium and high ventilation rates are 1.3%, 6.3% and 14.4%, respectively. The low concentration reduction with respect to the ventilation rate may be attributed to aerosol dispersion throughout the room and the dominance of the local airflow field induced by the ceiling fan in determining the concentration distribution of the aerosol. The figure shows that after the time horizon of 40 s, the percentage of particles flushing out from the room was saturated. This signifies that after 40 s, most of the particles were settled down, and the remaining particles were flushed out of the room.

Figure 14 shows the percentage of flushed out particles of different sizes with different ventilation rates (low, medium and high). The figure shows that the percentage deposition of particles up to the size <10 μm is low. This means that the motion of particles of smaller sizes was mainly governed by the ceiling fan induced airflow. As particles' sizes were increased up to the particles of size <30 μm, the percentage of flushed out particles were increased. This shows that particles of sizes between 10 and 30 μm were not much affected by the gravity and downward force of the ceiling fan induced airflow and were thus governed by the ventilation flow rates. As particles' sizes were increased from 30 μm, the motion of particles was governed by the gravitational force, and hence the particle flushed out percentage was decreased with the increasing size of particles.

Figure 14.

Effect of ventilation on flushed-out behaviour of particles of different sizes.

Comparative analysis of the effect of airflow induced by the ceiling fan and ventilation

For the effect of ceiling fan only on the particles behaviour, the results of section ‘Effect of ceiling fan induced airflow in particles deposition’ was used. Because for this case ventilation rate was low at 0.11 m/s, whereas for the case of the effect of ventilation only on the behaviour of particles, CFD simulations were performed by setting the ceiling fan rotation speed to zero and ventilation rates as 1, 2 and 3 m/s for low, medium and high ventilation rates. Figure 15 shows the effect of different ventilation rates on the deposition of particles of different diameters. The effect of different ceiling fan rotation speeds on the deposition of particles of different diameters is shown in Figure 9. Through the comparative analysis of Figure 15 and 9, the percentages of deposited particles were more due to the presence of the ceiling fan compared to the ventilation only. The comparative analysis also shows that the deposition of particles was increased with the increasing ceiling fan rotation speeds, whereas the deposition of the particles was decreased with the increased ventilation rate. Through the analysis of the effect of ventilation only on particle deposition, the deposition of the particles was decreased with the increased ventilation rate. This was because, at high ventilation rate, particles would become unstable and governed by the airflow profile. The figure shows that the percentage of the deposition of particles was increased with the increasing diameter of particles for all ventilation rates. This was due to the dominance of gravity force on the larger particles.

Figure 15.

Effect of ventilation rate on the deposition of the particles of different diameters.

Table 4 shows quantitively the percentage increase and decrease in the deposition of the particles. The table shows that the deposition of particles was decreased by 73.3 and 8.9% as settings were changed from high and medium ceiling fan rotation speeds to high and medium ventilation rates, whereas as the setting was changed from low ceiling fan rotation speed to low ventilation rate, the deposition of particles was increased by 12.3%. Through the combined analysis, the built environment should be operated with the combination of the ceiling fan and ventilation provision, as the ceiling fan would facilitate the deposition of particles and the ventilation would help to flushing out the remaining particles from the indoor environment to outdoor.

Table 4.

Percentage increase and decrease of the deposited particles when the setting was changed from ceiling fan to ventilation rate.

Level of ventilation and ceiling fan rotation speed	% increase or decrease of particles deposited
	Ceiling fan $\to$ Ventilation
High	73.3 (decrease)
Medium	8.9 (decrease)
Low	12.3 (increase)

Effect of mask in mitigating exposure risk

Masks are primarily intended to reduce the exposure of virus-laden droplets for the wearer and by the wearer, which is especially relevant for asymptomatic and presymptomatic person.^61,62 The recent progress in investigating the efficiency of facemasks to prevent COVID-19 spread and the importance of wearing masks has been reviewed in literature.⁶³ Various facemasks ranging from simple homemade cloth masks to ventilated respirators have played their role in the current COVID-19 pandemic. The efficiency of the homemade mask has been studied through CFD,⁶⁴ and 90% effectiveness of the mask in blocking droplets during coughing and sneezing has been reported. However, although masks would reduce droplet transmission, several droplets could be transmitted away from the mask^65–67 due to leakage between the mask and face. The use of a mask would not provide complete prevention from airborne droplet transmission.

The fit factor of the mask is an important feature in evaluating its efficiency. The face seal of a modified strapless form-fitting sealed version of a surgical mask using quantitative fit testing (QNFT) was tested and compared with the performance of N95 and unmodified loose-fitting surgical masks.⁶⁸ The fit factor for the sealed surgical mask was reported significantly higher than that of the loose-fitting surgical mask but lower than that of the N95 mask. This indicates significantly lower inward leakage of ambient aerosols with the sealed surgical mask than that of the loose-fitting surgical mask but still higher than the N95 mask. Furthermore, amongst different activities to evaluate the fit factor of different masks; talking has a greater effect on reducing the overall fit factor for the sealed surgical masks than for the N95 mask. To address the fitting issue of masks, López‐Rebollar⁶⁹ designed a mask using valves and deflectors inside the mask. Through CFD analysis, they found that the exhaled air flow had a long time and recirculation path within the mask before exiting. Further, they reported that filters of the valves operated as a barrier, preventing the flow through them efficiently.

The benefit of masks as a protective intervention for the community has been reviewed, and masks have been reported as an effective source control measure in mitigating COVID-19.⁷⁰ In a study,⁷¹ the investigation of the infection rate of COVID-19 with and without a mask was carried out. 820 people were interviewed and 53.3% of them wore masks. Non-mask users were found to be infected at a rate of 16.4%, while mask users were infected at a rate of 7.1%. The importance of masks in preventing COVID-19 compared to without masks in a community setting can be understood with the following examples:

• • The study of 124 Beijing households with confirmed COVID-19 cases >1 showed that mask usage reduced the secondary transmission within the households by 79%.⁷²

• • In a study conducted in 70 schools in Massachusetts during the 2020-21 school year, 11.7% secondary attack rate was reported for unmasked cases versus 1.7% for masked interactions.⁷³

• • A case–control study from Thailand, which involved more than 1000 persons for the interview as a part of a contact tracing exercise, reported that those who wore a mask during high-risk exposures experienced a greater than 70% reduced risk of acquiring infection compared with persons who did not wear masks under these circumstances.⁷⁴

• • In a nationwide analysis of U.S. data collected during 1 July to 4 Sept. 2021, counties without school mask requirements experienced a larger increase in COVID-19 cases after the start of school compared with counties with school mask requirements.⁷⁵

An experimental study was conducted to evaluate the efficiency of masks according to the type of materials and their fitting with the face.⁷⁶ N95 has been designed with many occupational health features to protect workers from inhaling hazardous aerosols in the workplace. The N95 mask is characterized by its tight-fitting design (face seal, nose clip and stretchable straps), and the material of the N95 is dense with high particle filtration efficiency. Therefore, N95 showed a generally high mask efficiency compared to the loose-fitting masks. Therefore, to investigate the effect of the mask in reducing the exposure risk of airborne pathogens under a ceiling fan, a mask with properties of N95 was modelled as a porous medium near the mouth’s surface. The computational domain of the mask is shown in Figure 2. The porosity of the mask was considered as 0.86. The permeability of the mask was calculated using equation (9). To capture the mask filtration efficiency, the particles’ concentration was calculated for the upstream and downstream. For the upstream, the volume between the mouth’s surface and the inner surface of the mask was considered. For the downstream, the volume between the exterior surface of the mask and the distance travelled by the particles after transmitting from the mask was chosen. The filtration efficiency ($FE$) of the mask was calculated by equation (12), as

$$FE=\frac{\left(C_u-C_d\right)}{C_u}$$ (12)

where $C_u$ and $C_d$ are the upstream and downstream particles’ concentration, respectively. Figure 16 shows the particles’ sizewise upstream and downstream concentrations and mask filtration efficiency. The figure shows that for particles of different size bins, the mask has different filtration efficiency. For particles <30 μm, the filtration efficiency was 77%, whereas the filtration efficiency was 79% for particles of diameter >30 μm. This shows that particles with higher diameters have less transmissibility for the selected mask than those with smaller diameters. Therefore, the mask with a finer pore size would be beneficial in mitigating the exposure risk of airborne pathogens.

Figure 16.

Effect of particle sizes on the observed filtration efficiency of a mask. In the figure, upstream data represents the particle concentration between the face surface and the inner surface of the mask. Downstream data shows particle concentration from the exterior surface of the mask to the room.

The particles' distribution in the presence of a mask and without a mask is shown in Figure 17. Particles travel horizontally larger distances without the mask in the room compared to the case when the mask is placed in front of the mouth. The maximum distance travelled by the particles horizontally was 1.4 m without a mask compared to 1.2 m when the mask was placed. The figure also shows that without the mask, the particles would bend towards the ground quickly due to the presence of a higher concentration of particles with larger diameters in the total number of particles compared to the case when the mask was placed, whereas in the presence of the mask, particles with larger diameters were filtered by the mask, and particles with smaller diameters were transmitted by the mask in the room, which remained suspended due to the buoyancy forces. The figure shows a difference between the distribution pattern of particles with and without the mask. In the case of without a mask, the distribution followed the parabolic path, whereas in the presence of a mask, the distribution is of the balloon shape. Without the mask, the particles were directly injected into the room without any hindrance. In the presence of the mask, particles of less diameter travel horizontally after being transmitted through the mask, whereas the particles of larger diameter travel after escaping from the space between the mouth surface and mask, which caused the distribution of particles to form a balloon shape. This study shows that the mask should be used with minimized gaps between the mask and face to maximize its effectiveness.

Figure 17.

Particles’ distribution in the room at 1s for (a) without a mask and (b) with a mask. The droplets’ symbol size is larger than the other plots for visualization.

Due to constraints on computational resources, the numerical model for mask was simulated for 1s only on (Intel® Core(TM)i7-7700 CPU @3.60 GHz). Though the results were for a short time, it served to analyze the mask’s effectiveness in mitigating airborne pathogens in indoor spaces.

Relationship between aerosol size, COVID-19 and other airborne pathogens

The aerosol produced by sneezing and coughing can travel up to 7–8 m. Studies have shown that when people sneeze or cough, the droplets larger than 10 μm sediments nearby, form fomites and risk direct and indirect transmission of the airborne pathogens and COVID-19 virus, whereas, when the droplets smaller than 10 μm leave the airway, becomes droplet nuclei or aerosols.⁷⁷ These aerosols can stay airborne in the atmosphere for much longer, and aerosols particles less than 2.5 μm can enter the alveoli directly⁷⁸ and particles sized between 6 and12 μm can enter the upper airways of the head and neck.⁷⁹ As compared to the nasal cavity and trachea, when the virus accumulates in the alveoli, small doses can cause infection.⁸⁰ In contrast to sneezing and coughing, finer aerosol particles <1.5 μm are produced during breathing and talking⁸¹ and can travel larger distance and create the possibility of spreading airborne pathogens and COVID-19 virus as there is evidence of the presence of COVID-19 virus in submicron and ultra-micron aerosols.⁸² Literature shows that aerosols of different sizes have a positive correlation with the prevalence of COVID-19. Timeseries analysis between aerosol particles of PM_2.5 and PM₁₀ sizes, and COVID-19 infection shows a positive correlation due to large surface area and strong adsorption capability.^83–88

Like COVID-19 other airborne pathogens such as influenza virus are contained in the aerosol particles within the respiratory size range.⁸⁹ Sizewise particles’ analysis for the influenza-like viruses shows that most of the viral content of the virus are found in particles of size ranging <1 μm, 1–4 μm and >4 μm.^90,91 Another study showed that influenza transmission was mediated by the aerosols >1.5 μm, even if the majority of the aerosols exhaled were <1.5 μm.^92,93 Airborne pathogens such as M tuberculosis were mostly found in aerosol particles <4.7 μm, and some were found in large particles of size >7 μm. In a study of drug-resistance-tuberculosis highest count of bacilli was found in the cough aerosols of size ranging between 2.1 and 4.7 μm.⁹⁴Pseudomonas aeruginosa was found in the cough aerosols of patients with cystic fibrosis of particles size slightly greater than that of patients of M tuberculosis.⁹⁵

From the above description of aerosols’ sizes and their relationship with airborne pathogens and COVID-19, the virus that can cause COVID-19 is transmitted by both small and large particle aerosols. Further, other airborne pathogens such as Influenza, M tuberculosis, Pseudomonas aeruginosa and Measles are mostly caused by the smaller aerosol particles (1–4.7 μm). Since a large number of patients and healthcare workers are likely to have frequent exposure to highly infectious aerosols in healthcare settings, and there is a possibility that they have more cumulative inhaled doses, clear guidance of control measures, for example, masks keeping the size of the aerosols in mind would reduce the probability of infection and also a reduction in the inhaled inoculum.

International implications of the results

As compared with air-conditioning and mechanical ventilation (ACMV), a ceiling fan is a more sustainable cooling modality and is more widely accessible in low–income and middle–income countries. According to Lancet’s historical day-time analysis,⁹⁶ ceiling fan use can be universally recommended to both young and older adults across large areas of northern Europe, northeastern regions of the USA, Canada, all of South America and much of southeast Asia.

Recently there has been increasing interest in using ceiling fan integrated air-conditioning systems to achieve energy savings while maintaining occupant comfort. Recognizing the importance of ceiling fans in mixing the indoor air, ASHRAE and Center for Disease Control and Prevention (CDC)⁹⁷ have recommended using ceiling fans in indoor environments to reduce the likelihood of stagnant air pockets where viral concentrations can accumulate. The ceiling fan could also enhance the window opening efficiency.⁹⁸ In spite of being ubiquitous in every household, the influence of the ceiling fan on the airborne transmission and the tradeoffs between ventilation rate and ceiling fan speeds with regard to airborne transmission is not clear.

The results presented in this study provide the impact of ceiling fan induced airflow on airborne pathogens and also the relationship between the ventilation rate, ceiling fan rotation speed and airborne pathogens concentration in the indoor environment. The results presented in this study could be applicable to all 12 countries where the Lancet study⁹⁹ has recommended the use of ceiling fans. The results of this study could be an initial start to performing a more comprehensive region-specific study regarding the role of ceiling fans in mitigating airborne pathogens from indoor environments.

Application possibilities in the built environment

Increasingly, ceiling fans are found in applications varying from industrial and warehouse applications to offices and high-end hospitality settings and everything in between. ASHRAE standard-55,¹⁴⁰ provides for increased airspeed in proportion to the increased operative air temperature while maintaining thermal comfort. Despite the spatial variation of airspeed in the built environment, ceiling fan integrated air conditioning and mechanical ventilation provide uniform thermal perceptions. This shows that there are plenty of recommendations available in the literature to use ceiling fans in the built environment to achieve thermal comfort while minimizing the cooling energy consumption.

The numerical simulation presented in this study was performed by considering the full-scale room and reasonably real conditions. The three rotation speeds of the ceiling fan have been considered according to the presently available rotation ranges for the ceiling fan. The occupant position in the simulation model presents the typical sitting place in a typical room of the built environment. Therefore, the results presented in this study can be used to determine the sitting position of the occupant in the room by observing the concentration of the airborne pathogen in the room. Also, by observing the relationship between the ceiling fan rotation speed and particles' deposition concentration, the optimum ceiling fan rotation speed can be determined for actual scenarios. Furthermore, through the analysis of the ventilation rate, ceiling fan induced airflow and particle concentration, the optimum natural ventilation rate can be determined for the built environment to mitigate the effect of the airborne pathogens. The presented results show that when the occupant uses a mask, the concentration of particles could be reduced in the built environment. This implies that when there is a likelihood of an infected person in the built environment, the results presented in this study suggest using a mask in addition to the ceiling fan operation and natural ventilation.

Conclusion

Ceiling fans are ubiquitously used in the built environment to achieve thermal comfort. Understanding the aerosol distribution pattern under the influence of airflow induced by the ceiling fan is critical to airborne pathogens and COVID-19 mitigation measures. The present study investigated the aerosol distribution under airflow induced by the ceiling fan operated at different ceiling fan rotation speeds, under different ventilation rates, and lastly, with the mask. A computational domain of a real-size room containing two windows, (one on the east façade (as inlet) and the other on the south façade (as outlet)), a ceiling fan and one occupant was constructed. Computational fluid and particle dynamics (CFPD) was used to model the aerosol exhaled by the occupant. Three case scenarios were considered here to investigate the aerosol behaviour in indoor spaces: the different fan rotation speeds, different ventilation rates through windows and the wearing of mask. The conclusions drawn from this study are summarized as follows:

• • Ceiling fans can play an important role in COVID-19 mitigation measures. Airflow induced by the ceiling fan would increase the percentage deposition of aerosol particles (i.e. particles with a diameter >35 μm) and would reduce the particles settling time.

• • The percentages of deposition of the aerosol particles of diameter >35 μm with fan rotation speeds of 160 RPM, 265 RPM and 365 RPM were found as 20.7%, 22.5% and 24.7%, respectively. In contrast, the percentages of deposition of total aerosol particles at corresponding fan rotation speeds are 58.9%, 55.6% and 31.7%.

• • The reason for the anomaly, that is, a decreasing trend in total particle deposition with increasing ceiling fan speed mentioned in the above point, was identified as the higher fan rotation speed. Through its strong air recirculation, the ceiling fan could lift the aerosol particles with small diameters against the gravitational force, leading to the particles' recirculation.

• • The average travelling time of the aerosol particles was investigated under the influence of the ceiling fan rotating at different speeds. The average aerosol travelling time was decreased with increasing fan rotation speed. The average aerosol travelling times with the fan rotation speed of 160 RPM, 265 RPM and 365 RPM are 5.6 s, 5.2 s and 4.4 s, respectively.

• • Ventilation could help to reduce the aerosol concentration in the indoor space. In this study, three ventilation rates, that is, low, medium and high, were considered that corresponds to the inlet airflow rate of 1m/s, 2m/s and 3m/s and facilitate 1.3%, 6.3% and 14.4% of total aerosol flushing out from the indoor space to outdoor. The results infer that ventilation could be an essential measure for COVID-19 mitigation.

• • The wearing of mask covering could help to reduce the concentration of aerosol particles of larger diameter in the room, which could be a potential viral load bearer in real-case scenarios. The filtration efficiency of the mask for the particles of diameter <30 µm was found to be 77%. In contrast, for particles of diameter >30 µm, the filtration efficiency was 79%. Thus, the mask with a smaller pore size and a well-covering face can be an effective mitigation measure for COVID-19.

Finally, this study investigated the importance of the most widely used electrical appliance in almost every household in tropical countries, that is, a ceiling fan, in mitigating airborne pathogens such as COVID-19, TB, influenza and measles. Since increased ventilation provisions and a downward supply of airflow have been recommended to mitigate airborne diseases, its efficiency can be enhanced by optimizing the ceiling fan induced airflow.

There are some limitations of the present study, which include non-consideration of the energy equation. The current model does not consider the dynamic process of droplets exhaled by occupants, such as evaporation, condensation, coagulation, resuspension and phase change. However, this study shows qualitatively and quantitatively the distribution pattern of aerosols under the effect of ceiling fan induced airflow, ventilation that may occur in real life, and the wearing of mask covering. This study would be helpful for further investigation of the impact of the ventilation provisions for COVID-19 appropriate indoor environment on thermal comfort and energy consumption.

ORCID iD

Brijesh Pandey https://orcid.org/0000-0003-0999-8180