This study intends to explore the effective and flexible solutions to cope with airborne transmission in hospital outpatient rooms. Such solutions might be used as an additional measure during pandemics and as an independent measure in regions with incomplete health facilities and limited resources. It first investigates the dispersion characteristics of exhaled pollutants during typical expiratory activities and then evaluates the effectiveness of a low-volume air cleaner and local exhaust in reducing the risk of cross infection using a newly proposed index, i.e., personal exposure reduction effectiveness (PERE). The results show that, though wearing a face mask largely obstructs the horizontal dispersion of exhaled particles and thus avoids short-range direct transmission, the influence of particles leaked from the edges of a face mask on the doctor and the next patient cannot be ignored. Under the conditions without wearing a face mask, a background ventilation rate of 60 m3/h plus a 50 m3/h desk-mounted air cleaner is effective to prevent the direct exposure of the doctor from the patient's exhaled particles, with the PERE reaching 90.1%. Under the conditions with wearing a mask, a background ventilation rate of 60 m3/h plus a 30–50 m3/h local exhaust above the patient's head removes 85.7%–88.5% of leaked particles, achieving a PERE of 96.6%–100%. The aforementioned PERE value during the two types of conditions is 137% (or 70.4%–71.4%) higher than that under only a background ventilation of the stipulated 120 m3/h by standard. These findings should provide ideas and information for improving the mitigating system of airborne transmission in hospital outpatient rooms.

Hospital outpatient rooms are at the forefront of providing medical consultations and treatments. However, these rooms are often enclosed indoor places with limited uniformly mixing ventilation, where the medical staff suffer from high exposure risks to infectious viruses exhaled from infected patients, particularly during pandemics and seasonal flus.1 Such a situation becomes more severe in regions with incomplete health facilities and limited resources, where insufficient ventilation is a common problem of outpatient rooms and improving the ventilation level is restricted by its energy consumption. The exposure risk also exists for patients who visit the same outpatient room after an infected patient. In order to mitigate risks and minimize the socio-economic impact of hospital outpatient infections, it is imperative to conduct research on the airborne transmission of pollutants during the medical consultation process and propose effective, flexible, and low-cost control measures.

The airborne transmission of infectious diseases between occupants indoors involves three main aspects:2 the generation of virus-laden aerosols from an infected person, the spread of virus-laden aerosols, and the inhalation of virus-laden aerosols by an exposed person. Understanding the characteristics of aerosol generation and dispersion is crucial for comprehending the process of aerosol spread and formulating precise prevention and control measures.3 Virus-laden exhaled flow during various expiratory activities, such as breathing, coughing, and talking, is generally considered to be a potential source of infectious aerosols.4–8 Previous studies have reported abundant information on the characteristics of exhaled flow,9–13 including flow velocity, flow shape, flow direction, and propagation distance through manikin/subject experiments. This information provided critical boundary conditions for computational fluid dynamics (CFD) simulations, which were a widely used research method to evaluate the effectiveness of control measures. Additionally, a few studies have revealed the dispersion characteristics of exhaled pollutants while wearing various types of face masks,3,14–18 such as surgical masks, N95 respirator (both tightly and loosely fit), and homemade sewn masks. However, there is still a lack of comparative analysis of the dispersion characteristics during various expiratory activities under the conditions without and with face masks.

Indoor air quality (IAQ) control strategies, such as ventilation, air cleaning, and source control, have been widely recommended by various international and national organizations.19,20 Among these strategies, ventilation is one of the most influential engineering methods to remove indoor pollutants and mitigate the risk of cross-infection.21,22 The room air distributions are shaped by two key factors: room airflow pattern and ventilation flow rate.2,23–26 The location of inlets and exhausts is critical for the development of room airflow pattern.24–26 Studies have shown that, in rooms with wall jet inlets, placing the exhausts near the ceiling led to lower pollutant concentrations compared to placing them near the floor. In contrast, the placement of exhausts was less crucial in rooms with ceiling inlets, as the room air was well mixed.26 On the other hand, the minimum air change rate (ACH) is typically recommended by ventilation standards and guidelines19 to maintain an acceptable IAQ in one specific area. Increasing ACH is beneficial but may not always be sufficient for the effective infection control.27 Additionally, advanced air distribution methods have shown a better performance in reducing the risk of cross-infection compared to the total volume air distribution.28 In hospital outpatient rooms, on one hand, doctors and patients suffer from a high risk of direct cross-infection due to their close proximity and fixed positions.29 On the other hand, as mentioned before, outpatient rooms often have inadequate ventilation in countries and regions with limited resources, and complete-mixing ventilation in these rooms is limited in achieving the localized control of pollutants.30,31 Therefore, the portable air cleaner is an alternative method for providing equivalent clean air and offering flexible placement options. Some past studies have evaluated the effectiveness of air cleaners in various settings for removing indoor particles,32–37 such as offices, classrooms, and hospital wards. They confirmed that air cleaners were indeed an effective measure for infection control. The effectiveness of air cleaners was considerably influenced by various factors, such as its location, supply air volume, and supply air direction. Previous studies have reported that increasing the supply air volume of air cleaners would increase their removal efficiency of indoor particles. However, it would also increase indoor noise levels, potentially causing complaints from nearby individuals. It is therefore crucial to investigate the method and the effectiveness of using air cleaners, particularly those with low supply air volume (namely, low noise and energy consumption), in hospital outpatient rooms where doctors and patients are in close proximity.

Ventilation (total volume air distribution) and air cleaning are regarded as secondary solutions for dealing with the exhaled aerosols after their release into the surrounding air. Source control, however, is widely recognized as the most effective measure.20 Although source control may be challenging to be applied in practice, they remain a topic of active exploration by researchers. Source control measures based on ventilation with advanced air distribution method mainly include personal ventilation (PV) and personal exhaust (PE), which have shown excellent performance in reducing the risk of cross-infection while minimizing energy consumption.38 PV and PE are typically used in places where occupants (seated or standing) spend a long time,39 such as offices, theaters, and vehicles. Examples of PV and PE configurations31,39–41 include bed-mounted, desk-mounted, chair-mounted, headsets, partition-integrated, and so on. In hospital outpatient rooms, it is important to prevent the dispersion of exhaled aerosols and quickly remove them, particularly during pandemics and seasonal flus when patients are mandatory to wear face masks. However, there is currently no report on the application of local ventilation systems for individuals wearing face masks in a seated position. Therefore, further study is necessary to investigate the design methods and the effectiveness of local ventilation to cope with emergency pandemics.

The current practice of using total volume ventilation systems with mixing air distribution in hospital outpatient rooms is generally ineffective in terms of aerosol removal and airborne transmission mitigation, while localized measures should be more cost-effective. This study aims to address the doctor-to-patient and patient-to-patient cross-infection in hospital outpatient rooms and to examine the performance of two localized measures, namely, low-volume air cleaner and local exhaust. First, the distribution of exhaled pollutants is characterized to build up the boundary condition for evaluating cross-infection. Second, the effectiveness of low-volume air cleaners with various locations and supply air directions in reducing doctors' inhalation exposure to particles exhaled from infected patients is examined. Finally, the effectiveness and optimal design method of the low-volume local exhaust system with different configurations and ventilation flow rates in localized infection control are explored. This study would provide new effective, flexible, and energy-saving solutions for mitigating the risk of cross-infection in hospital outpatient rooms, which might potentially be used as an additional measure during pandemics and as an independent measure in regions with incomplete health facilities and limited resources.

As illustrated in Fig. 1, two physical models with a dimension of 5 × 4 ×3 m3 (L × W × H) were numerically constructed. Based on our field observations, the mixing ventilation was widely used in outpatient rooms, which was thus used in the two numerical outpatient rooms. The air was supplied from the upper inlet (0.2 × 0.2 m2) and exhausted through the upper outlet (0.2 × 0.2 m2). As shown in Fig. 1(a), three positions were chosen to place an air cleaner, namely, one on the desk, the second on the wall, and the third on the floor. Clean air from the air cleaner was supplied either from the top vent or the side vent, and room air was removed through the exhaust vent. In addition, the local ventilation system was installed on the chair for patients [see Fig. 1(b)], with four configurations of air distribution provided with two side vents and one top vent [see Figs. 1(c)–1(f)]. To accurately reproduce the airflow interactions indoors, two human-shaped models were employed to represent an infected patient (IP, without/with a face mask) and a susceptible doctor (SD, with a face mask). These two sitting thermal manikins had the same geometry, with a height of 1.2 m and a surface area of 1.6 m2. The mouth opening area was 100 mm2, matching the mouth area of an average person.14 It was reported that the thickness of the face mask was roughly in a range of 0.5–2.3 mm, and the gap size between the face mask and the user's face was roughly in a range of 4–14 mm.42,43 Hence, in this study, the thickness of the face mask was set to be 2.0 mm, and the largest gap size less than 10.0 mm was located at the top of the face mask.

FIG. 1.

The computational domain representing the outpatient room and the local ventilation system with four configurations (IP: infected patient and SD: susceptible doctor): (a) room with air cleaners, (b) room with the local ventilation system, (c) inlet (side vent II) and outlet (side vent I and top vent), (d) inlet (side vent I and II) and outlet (top vent), (e) outlet (top vent), and (f) inlet (side vent II) and outlet (side vent I).

FIG. 1.

The computational domain representing the outpatient room and the local ventilation system with four configurations (IP: infected patient and SD: susceptible doctor): (a) room with air cleaners, (b) room with the local ventilation system, (c) inlet (side vent II) and outlet (side vent I and top vent), (d) inlet (side vent I and II) and outlet (top vent), (e) outlet (top vent), and (f) inlet (side vent II) and outlet (side vent I).

Close modal

The room walls were well-insulated and considered to be adiabatic.44 According to Hospital Facilities Design Guidelines (Air conditioning volume, HEAS-02-2022), established by the Healthcare Engineering Association of Japan (HEAJ),45 the ACH of outpatient rooms should be no less than 2 h−1. However, in practice, outpatient rooms in regions with limited resources often have insufficient ventilation, making it challenging to achieve the recommended ventilation flow rate of 120 m3/h that is 2 h−1 for the present configuration. To create an indoor space resembling poorly ventilated outpatient rooms, the background ventilation flow rate was set to be 60 m3/h for most conditions. Meanwhile, the maximum supply air volume for both air cleaners and local ventilation system was set to be 50 m3/h to represent the low-volume ones. Additionally, when the air cleaner and the local ventilation system were not presented, the conditions with different background ventilation flow rates (namely, 60 and 120 m3/h) were compared. The air was assumed to be incompressible ideal gas to simulate the buoyancy effect.46 The detailed boundary conditions are shown in Table I.

TABLE I.

Boundary conditions for the CFD simulation.

Models Boundaries Supply/exhaust air volume (m3/h) Temperature (°C) Boundary type
Room  Inlet  60  22 (Ref. 44 Velocity-inlet 
Outlet  30 (Ref. 44 Pressure-outlet 
Air cleaner  Inlet  50  26  Velocity-inlet 
Outlet  50  26  Mass-flow-outlet 
Local ventilation system  Inlet  50  26  Velocity-inlet 
Outlet  10, 20, 30,40, 50  26  Velocity-inlet 
Mouth  Mouth opening  see Fig. 2   34  Velocity-inlet 
Models Boundaries Supply/exhaust air volume (m3/h) Temperature (°C) Boundary type
Room  Inlet  60  22 (Ref. 44 Velocity-inlet 
Outlet  30 (Ref. 44 Pressure-outlet 
Air cleaner  Inlet  50  26  Velocity-inlet 
Outlet  50  26  Mass-flow-outlet 
Local ventilation system  Inlet  50  26  Velocity-inlet 
Outlet  10, 20, 30,40, 50  26  Velocity-inlet 
Mouth  Mouth opening  see Fig. 2   34  Velocity-inlet 

The two thermal manikins were heated at a constant temperature of 32 °C,2 close to the skin temperature of an average person when seated in a state of thermal comfort. Three typical expiratory activities including breathing, coughing, and talking were examined, and their airflow patterns are shown in Fig. 2. For breathing [see Fig. 2(a)], the pulmonary ventilation rate and the breathing frequency were defined to be 6.0 L/min and 12 times/min,44 respectively. The coughing process was represented by a combination of gamma-probability-distribution functions, each lasting for 2 s (Refs. 3 and 47) [see Fig. 2(b)]. Previous studies have reported that the exhaled flow rate over time can be represented as a constant for talking, with an average (or peak) velocity in a range of 1.6–3.9 m/s.12,14,48 In this study, the transient process of talking was thus simplified to a constant inhalation or exhalation process at a velocity of 2 m/s [see Fig. 2(c)].

FIG. 2.

Airflow patterns during different expiratory activities: (a) mouth breathing,44 (b) coughing,3 and (c) talking.14 

FIG. 2.

Airflow patterns during different expiratory activities: (a) mouth breathing,44 (b) coughing,3 and (c) talking.14 

Close modal

The present simulations included two main parts. In the first part (Case I, see Table II), the species transport model was applied to characterize the dispersion of exhaled pollutants and the distribution of exhaled flow during three expiratory activities, without and with wearing a face mask. In the second part (Cases II and III, see Tables III and IV), the discrete phase model (DPM) was used to examine the effectiveness of air cleaners and local ventilation system in reducing risks of cross-infections during 5-min (300 s) medical consultations in the outpatient room. For the first part, since the movement of droplet nuclei less than 3–5 μm can be well represented with the tracer gas simulation,49 carbon dioxide (CO2) with a mass fraction of 4.3% was chosen as the tracer gas and released from the mouth of an infected patient (IP) in model A [see Fig. 1(a)]. It should be noted that, in this part, the SD and air cleaners were set as non-heated walls. For the second part, the consultation scenarios with an infected patient (IP) and a susceptible doctor (SD) were simulated (see Fig. 1). The SD wore a face mask and talked with the IP, which was simplified as a constant inhalation process.14 Meanwhile, the IP, without or with wearing a face mask, talked to the SD, which was simplified as a constant exhalation process.14 Although droplets released by a human talking are distributed in a wide size range (0.1–1000 μm), over 80% of them are fine droplets smaller than 36 μm.12,50 The vast majority of the droplets then quickly evaporate and reach their equilibrium diameters less than 0.3 s (Refs. 51–53) in indoor spaces. Therefore, the instant evaporation process was not considered in this study. The equilibrium diameter of a droplet is roughly 26% of its initial size.54 This means over 80% of the droplet nuclei generated by a human talking are smaller than 9.36 μm. In addition, these droplet nuclei may contain viruses, and studies have shown that viruses are enriched in small aerosols with particle sizes smaller than 5 μm.55 Moreover, specific size ranges with most viruses have been identified, including peaks around 0.2–0.8 μm and 3.5–5.0 μm, each representing a different generation site and production process.56 Therefore, aerosols with diameters of 0.5, 2.5 (particulate matter 2.5, a commonly used air quality parameter),57 and 5 μm were examined in this study. These particles were released from the mouth of the IP with a mass flow rate of 5.24 × 10−11 kg/s.52 For boundary conditions of the discrete phase, the escape condition was applied for all outlets and the mouth of SD; the reflection condition was applied for all supply air inlets, the mouth of IP, and the ceiling; and the trap condition was defined for the remaining parts of the models.

TABLE II.

A list of simulated scenarios (model A) regarding three expiratory activities.

Case I Face mask (IP) Pollutant type Expiratory activity
Yes  Tracer gas  Breathing 
No  Tracer gas  Breathing 
Yes  Tracer gas  Coughing 
No  Tracer gas  Coughing 
Yes  Tracer gas  Talking 
No  Tracer gas  Talking 
Yes  Aerosol particle  Talking 
No  Aerosol particle  Talking 
Case I Face mask (IP) Pollutant type Expiratory activity
Yes  Tracer gas  Breathing 
No  Tracer gas  Breathing 
Yes  Tracer gas  Coughing 
No  Tracer gas  Coughing 
Yes  Tracer gas  Talking 
No  Tracer gas  Talking 
Yes  Aerosol particle  Talking 
No  Aerosol particle  Talking 
TABLE III.

A list of simulated scenarios (model A) regarding air cleaners.

Case II Location Air supply direction Supply air volume (m3/h) Face mask (IP) Pollutant type Expiratory activity
1 (D-H)  Desk  Horizontal  50  No  Aerosol particle  Talking 
2 (D-V)  Desk  Vertical upward  50  No  Aerosol particle  Talking 
3 (W-H)  Wall  Horizontal  50  No  Aerosol particle  Talking 
4 (W-V)  Wall  Vertical upward  50  No  Aerosol particle  Talking 
5 (F-H)  Floor  Horizontal  50  No  Aerosol particle  Talking 
6 (F-V)  Floor  Vertical upward  50  No  Aerosol particle  Talking 
7 (D-H)  Desk  Horizontal  50  Yes  Aerosol particle  Talking 
8 (D-V)  Desk  Vertical upward  50  Yes  Aerosol particle  Talking 
9 (W-H)  Wall  Horizontal  50  Yes  Aerosol particle  Talking 
10 (W-V)  Wall  Vertical upward  50  Yes  Aerosol particle  Talking 
11 (F-H)  Floor  Horizontal  50  Yes  Aerosol particle  Talking 
12 (F-V)  Floor  Vertical upward  50  Yes  Aerosol particle  Talking 
13  No air cleaner  Background ventilation flow rate: 60 m3/h 
14  No air cleaner  Background ventilation flow rate: 120 m3/h 
Case II Location Air supply direction Supply air volume (m3/h) Face mask (IP) Pollutant type Expiratory activity
1 (D-H)  Desk  Horizontal  50  No  Aerosol particle  Talking 
2 (D-V)  Desk  Vertical upward  50  No  Aerosol particle  Talking 
3 (W-H)  Wall  Horizontal  50  No  Aerosol particle  Talking 
4 (W-V)  Wall  Vertical upward  50  No  Aerosol particle  Talking 
5 (F-H)  Floor  Horizontal  50  No  Aerosol particle  Talking 
6 (F-V)  Floor  Vertical upward  50  No  Aerosol particle  Talking 
7 (D-H)  Desk  Horizontal  50  Yes  Aerosol particle  Talking 
8 (D-V)  Desk  Vertical upward  50  Yes  Aerosol particle  Talking 
9 (W-H)  Wall  Horizontal  50  Yes  Aerosol particle  Talking 
10 (W-V)  Wall  Vertical upward  50  Yes  Aerosol particle  Talking 
11 (F-H)  Floor  Horizontal  50  Yes  Aerosol particle  Talking 
12 (F-V)  Floor  Vertical upward  50  Yes  Aerosol particle  Talking 
13  No air cleaner  Background ventilation flow rate: 60 m3/h 
14  No air cleaner  Background ventilation flow rate: 120 m3/h 
TABLE IV.

A list of simulated scenarios (model B) regarding the local ventilation system.

Case III Ventilation configuration Supply air volume (m3/h) Pollutant type Expiratory activity
Configuration A  50  Aerosol particle  Talking 
Configuration B  50  Aerosol particle  Talking 
Configuration C  50  Aerosol particle  Talking 
Configuration D  50  Aerosol particle  Talking 
Configuration C  40  Aerosol particle  Talking 
Configuration C  30  Aerosol particle  Talking 
Configuration C  20  Aerosol particle  Talking 
Configuration C  10  Aerosol particle  Talking 
Case III Ventilation configuration Supply air volume (m3/h) Pollutant type Expiratory activity
Configuration A  50  Aerosol particle  Talking 
Configuration B  50  Aerosol particle  Talking 
Configuration C  50  Aerosol particle  Talking 
Configuration D  50  Aerosol particle  Talking 
Configuration C  40  Aerosol particle  Talking 
Configuration C  30  Aerosol particle  Talking 
Configuration C  20  Aerosol particle  Talking 
Configuration C  10  Aerosol particle  Talking 

1. Continuous phase

The continuous phase flow is described by the Reynolds-averaged Navier–Stokes (RANS) equations of mass, momentum transport, and energy. The steady, incompressible, and turbulent forms of the equations are given as follows.

The mass equation is
u i x i = 0 .
(1)
The momentum equation is
u i t + u j u i x j = 1 ρ p x i + 1 ρ τ i j x j + g i .
(2)
The energy equation is
x i ( u i ( ρ e + ρ ) ) = x i ( λ eff ( T x i + u j τ i j ) ) + S h .
(3)
The shear stress transport (SST) kω model58 was found to perform best in predicting the dispersion of the exhaled airflow through the mask among RANS models.46 In Eq. (2), the viscous stress tensor τij is defined by
τ ij = μ [ ( u i x j + u j x i ) 2 3 δ i j u k x k ] .
(4)
For turbulence simulation predicted by the RANS model, the fluctuating velocity component u i is defined by7 
u i = f i ξ i 2 3 k ,
(5)
f u = 1 + 0.285 ( y + + 6 ) exp ( 0.455 ( y + + 6 ) 0.53 ) ,
(6)
f v = 1 exp ( 0.02 y + ) ,
(7)
f w = 3 f u 2 f v 2 ,
(8)
y + = y v τ w ρ .
(9)

The definitions of the parameters in the above formula are listed in Table V.

TABLE V.

The definition of parameters.

Parameters Meanings
ui  The velocity in the xi direction 
gi  The gravitation acceleration in the xi direction 
p  Pressure 
ρ  Air density 
T  Temperature 
μ  Dynamic viscosity 
τij  Viscous stress tensor 
e  Specific energy 
λeff  Effective thermal conductivity 
Sh  Heat source 
u i   Fluctuating velocity component 
ξi  Random numbers from the standard normal distribution 
fi  Damping factors 
y+  Dimensionless wall distance 
ν  Kinematic viscosity 
τw  Wall shear stress 
Parameters Meanings
ui  The velocity in the xi direction 
gi  The gravitation acceleration in the xi direction 
p  Pressure 
ρ  Air density 
T  Temperature 
μ  Dynamic viscosity 
τij  Viscous stress tensor 
e  Specific energy 
λeff  Effective thermal conductivity 
Sh  Heat source 
u i   Fluctuating velocity component 
ξi  Random numbers from the standard normal distribution 
fi  Damping factors 
y+  Dimensionless wall distance 
ν  Kinematic viscosity 
τw  Wall shear stress 

2. Discrete phase

We used the Euler–Lagrange method to establish a discrete phase model for simulating the behavior of aerosol particles. Given that the aerosol volume fraction in this study is much smaller than 10−6, resulting in a dilute flow, we applied one-way coupling59 based on the Euler–Lagrange approach to establish the coupling between the continuous and particle phases.60–62 The location of the single particle xp, can be calculated as follows:63,
d x p d t = u p ,
(10)
where up is the particle velocity derived from the balance between inertial forces and external forces applied to the particle
m p d u p d t = F p ,
(11)
where mp is the particle mass and ∑Fp is the sum of all forces that are applied on the particle, as calculated below
F p = F drag + F gravity + F a ,
(12)
where Fa is other additional forces applied on the particle, including the Saffman lift force, the virtual mass force, the Brownian force, the thermophoresis force, and the Magnus. According to the basic theory of aerosol dynamics, additional forces depend on the properties of airflow and particles.64 Since the particles are small enough, we only consider the Brownian force, the thermophoresis force, and the Saffman lift force.63,65 The Saffman lift force may be relatively large near room walls for fine particles.64 Thus, the total forces on the particles are as follows:
m p d u p d t = F drag + F gravity + F thermo + F saff + F brownian .
(13)
For the above equation, the terms on the right-hand side represent the drag force, the gravity, the thermophoresis force, the Saffman lift force, and the Brownian force, respectively. The drag force (Fdrag) can be calculated as follows:
F drag = π ρ a d p 2 C D | u a u p | 2 8 C c ,
(14)
where dp is the diameter of the particle, ρa is the air density, ua is the air velocity, up is the particle velocity, Cc is the Cunningham correction factor, and CD is the drag coefficient, which is calculated based on the spherical drag law, as follows:
C D = a 1 + a 2 Re p + a 3 Re p 2 ,
(15)
Re p = d p ρ a | u a u p | μ ,
(16)
where a1, a2, and a3 are constants that apply over several ranges of the Reynolds number given by Abkarian et al.,11  Rep is the particle Reynolds number, and μ is the dynamic viscosity.
The Saffman force (Fsaff) used is from Li and Ahmadi66 and is a generation of the expression provided by Saffman,67 as defined as follows:
F saff = m p 2 K ν 1 / 2 ρ d i j ρ p d p ( d l k d k l ) 1 / 4 ( u a u p ) ,
(17)
where K is the constant coefficient of Saffman's lift force (K = 2.594) and dij is the deformation tensor.

3. Solver settings

The second-order upwind discrete scheme was adopted to discretize the momentum, energy, and turbulence, while the “PRESTO” scheme was used to discretize the pressure equation considering the pressure gradient near the boundaries.68 Pressure-implicit with splitting of operators (PISO) algorithm was applied for solving the pressure–velocity coupling of transient flow.68 The initial conditions of the transient results were provided by the steady-state airflow solution, which was solved by the semi-implicit method for the pressure-linked equation (SIMPLE) algorithm. In the transient simulation, the fully implicit scheme was used for time integration, because it was unconditionally stable for time step size. The time step size was set to 0.001 (case I3-4) and 0.1 s (remaining cases) according to previous studies.46,68 Convergence was achieved when the scaled residuals reached 10−4 for all quantities except energy, which was less than 10−6.68 

As the porous media model69 is capable of solving problems regarding pressure drop/loss without modeling microscopic flow details, it was adopted in this simulation to examine the fluid flow regimes and the penetration through the face mask. The source term of the momentum equation [Eq. (18)], the pressure drop through the porous media model [Eq. (19)], and the pressure drop through the face mask [Eq. (20)] can be calculated as follows:
S i = ( μ α ν i + C 2 1 2 ρ | ν | ν i ) ,
(18)
Δ P 1 = S i Δ n ,
(19)
Δ P 2 = A ν 2 + B ν ,
(20)
where Si is the source term due to the effect of the porous media model; ΔP1 is the pressure drop through the porous media model; ΔP2 is the pressure drop through the face mask; μ is the hydrodynamic viscosity coefficient; α is the permeability coefficient; 1/α is the viscous resistance coefficient; C2 is the inertial resistance coefficient; ρ is the density of the fluid; ν is the air speed; Δn is the thickness of the porous media; and A and B are the fitting polynomial coefficients, which were obtained from the experiments conducted by Jia et al.46 on surgical masks.

The computational domains were filled with a hybrid unstructured grid composed of polyhedrons and hexahedrons [see Figs. 3(a) and 3(b)]. Due to manikins' geometrical complexity and the expected high velocity and temperature gradients, a bounding box (1.5 -length × 1-width × 1.5 m3-height) covering both manikins was used for grid refinement to accurately simulate the airflow interactions around the manikins. In the model A, the grid independence test was conducted in terms of the velocity distribution above the patient head (without the face mask), and three polyhedron-based grids, namely, coarse: 1.87 × 106, medium: 3.53 × 106, and fine: 5.05 × 106 were assessed. As shown in Figs. 3(c) and 3(d), the average relative deviation (MRE)44 of the velocity magnitudes between the medium and the coarse was 16.3%, while that between the fine and the medium was 3.9%. Since the difference in the velocity distribution above the patient head between the fine and the medium grids was small, the medium grid of 3.53 × 106 was selected for further simulations, and its maximum skewness of the cells and y+ at the human body surface were less than 0.85 and 1.0, respectively. Similarly, a grid with 3.85 × 106 cells was applied for the room model B.

FIG. 3.

Mesh domain and the grid independence test of the model A by comparing the velocity distribution above the patient head (without the face mask): (a) mesh in the plane of X = 1.945 m passing over the IP, (b) mesh in the plane of Y = 3.06 m passing over the SD, (c) sampling line (Y, Z1, Z2 and Z3) with ten sampling points, and (d) velocity distribution along the sampling line Z3; MRE represents the average of relative differences between two groups of data.

FIG. 3.

Mesh domain and the grid independence test of the model A by comparing the velocity distribution above the patient head (without the face mask): (a) mesh in the plane of X = 1.945 m passing over the IP, (b) mesh in the plane of Y = 3.06 m passing over the SD, (c) sampling line (Y, Z1, Z2 and Z3) with ten sampling points, and (d) velocity distribution along the sampling line Z3; MRE represents the average of relative differences between two groups of data.

Close modal
For the first part, the distribution of the exhaled pollutant concentration along different sampling lines around the human breathing zone, as shown in Fig. 3(c), was presented and analyzed. The dimensionless exhaled pollutant concentration (C′) was calculated by Eq. (21). For the second part, the inhalation fraction (IF)44 was introduced to evaluate the exposure risk of doctors [see Eq. (22)]; the personal exposure reduction effectiveness (PERE)70 was used to assess the effectiveness of air cleaners and local ventilation system in reducing the exposure risk for doctors [see Eq. (23)]; and the removal efficiency (RE) was used to evaluate the effectiveness of air cleaners and local ventilation system in removing indoor particles [see Eq. (24)]
C = C i C 0 ,
(21)
where Ci represents the maximum value of the pollutant concentration recorded during mouth breathing, coughing, and talking at the sampling point i, which is located at distances of 0.02, 0.04, 0.06, 0.08, 0.10, 0.12, 0.14, 0.16, 0.18, and 0.2 m away from the mouth along either the horizontal or the vertical line [see Fig. 3(c)]; and C0 is the initial exhaled CO2 concentration
I F ( t ) = 0 t in Q i n ( t ) C i n ( t ) d t 0 t ex Q e x ( t ) C e x ( t ) d t ,
(22)
where Cin is the mass of aerosol particles inhaled by the SD, kg; Cex is the mass of aerosol particles exhaled by the IP, kg; Qin is the inhaled flow rate during the inhalation process, m3/h; and Qex is the exhaled flow rate during the exhalation process, m3/h
PERE = C I , 0 C I C I , 0 C A ,
(23)
where CI,0 is the mass of aerosol particles inhaled by the SD when without air cleaners and the local ventilation system, kg; CI is the mass of aerosol particles inhaled by the SD with air cleaners and the local ventilation system, kg; and CA is the mass of aerosol particles released from the supply air of air cleaners and the local ventilation system. In general, the air provided by air cleaners and local ventilation system was totally clean air, meaning that, in this study, CA = 0. When PERE < 0, the air cleaners and the local ventilation system cannot protect the doctor and may even present a harmful effect; when 0 < PERE < 1, the PERE value indicates the probability of exposure risk reduction for doctors
R E = C A e x C e x ,
(24)
where CA-ex is the mass of aerosol particles exhausted from the outlet of air cleaners and the local ventilation system, kg; and Cex is the mass of aerosol particles exhaled by the IP, kg.

Model validations were performed by reproducing several sets of experiments. The first was the distribution of the exhaled pollutant concentration around the human breathing zone. We measured the exhaled CO2 concentration along the centerline of the human mouth in an empty chamber room with a dimension of 7.2-length × 5.0-width × 3.0 m3-height, as shown in Fig. 4(a). The indoor air temperature was 27 °C. Six subjects were employed and instructed to perform three expiratory activities: breathing, coughing, and talking. For the breathing process, they breathe through their mouth for 1 min. For the coughing process, the subjects were asked to hold their breath after coughing. For the talking process, the subjects continuously enunciate the numbers from 1 to 20 for 1 min. Detailed information on the six subjects is presented in Table VI. All measurements were repeated three times to improve the reliability and accuracy of the results. The CO2 concentration was measured by RS-I20 CO2 transmitter, which was calibrated to ensure accuracy before performing the experiments. Its measuring range was 0–10 000 ppm with a measuring accuracy of ±(50 ppm ± 3% read value). Figure 5 shows the comparison of experimental and simulated results of CO2 concentration. Although the subjects were asked to maintain the same respiratory rate during the experiment, variations in individual metabolic rates, influenced such as age and gender, led to differences in exhaled CO2 concentrations in front of the mouth. There were thus some obvious differences among individuals in the experimental results, but the overall trend remained consistent across all subjects. The simulated results had the same trend with the experimental data, indicating the rationality of the boundary conditions employed for expiratory activities in present simulations.

FIG. 4.

A sketch of the environmental chamber and arrangements: (a) our environmental chamber for the measurement of the exhaled CO2 concentration and (b) experimental arrangements for the measurement of the aerosol particle concentration by Zhang and Chen.71 

FIG. 4.

A sketch of the environmental chamber and arrangements: (a) our environmental chamber for the measurement of the exhaled CO2 concentration and (b) experimental arrangements for the measurement of the aerosol particle concentration by Zhang and Chen.71 

Close modal
TABLE VI.

Detailed information of human subjects.

NO. Age Gender Height (m) Weight (kg)
F1  26  Female  1.62  56 
F2  25  Female  1.60  58 
F3  24  Female  1.66  50 
F4  23  Female  1.68  60 
M1 (smoker)  25  Male  1.72  62 
M2  29  Male  1.73  70 
NO. Age Gender Height (m) Weight (kg)
F1  26  Female  1.62  56 
F2  25  Female  1.60  58 
F3  24  Female  1.66  50 
F4  23  Female  1.68  60 
M1 (smoker)  25  Male  1.72  62 
M2  29  Male  1.73  70 
FIG. 5.

A comparison of the CO2 concentration distribution along the centerline of human/manikin mouth (without the face mask) between simulated (case I2, 4, 6, and 8) and experimental results. The CO2 concentration (C) was normalized by the initial exhaled CO2 concentration (C0).

FIG. 5.

A comparison of the CO2 concentration distribution along the centerline of human/manikin mouth (without the face mask) between simulated (case I2, 4, 6, and 8) and experimental results. The CO2 concentration (C) was normalized by the initial exhaled CO2 concentration (C0).

Close modal

The second validation was regarding the distribution of the airflow velocity and the dispersion of aerosol particles indoors. The experiment was conducted by Zhang and Chen71 in a full-scale environmental chamber with under-floor air distribution (UFAD). The experimental arrangements are depicted in Fig. 4(b). Two air supply openings were located on the floor with a total ventilation rate of 0.0944 m3/s, while the exhaust opening was mounted in the ceiling where particle concentration measurements were taken. The particle source was located at 0.3 m above the floor, with the mean particle size of 0.7 μm. The airflow velocity and particle concentration were monitored at five height levels (0.4, 0.6, 1.2, 1.6, and 1.8 m) of six vertical lines from P1 to P6 [see Fig. 4(b)]. We numerically reproduced their experiments using the turbulence model and solver settings described in the present study and compared the simulated and experimental results, as shown in Fig. 6. It should be noted that, although there were some differences resulting from the fluctuation of airflow near the particle source, as explained by Zhang and Chen,71 the simulated results of the distribution of airflow velocity and particle concentration along sampling lines highly agreed with experimental results. Hence, the present numerical model can well predict the distribution of airflow velocity and dispersion of aerosol particles.

FIG. 6.

A comparison of the distribution of airflow velocity and particle concentration along sampling lines between simulated and experimental results:71 (a) velocity distribution and (b) particle distribution; the particle concentration (C) was normalized by the steady-state nominal particle concentration (C0) at the exhaust outlet.

FIG. 6.

A comparison of the distribution of airflow velocity and particle concentration along sampling lines between simulated and experimental results:71 (a) velocity distribution and (b) particle distribution; the particle concentration (C) was normalized by the steady-state nominal particle concentration (C0) at the exhaust outlet.

Close modal

1. Qualitative analysis of exhaled pollutant distribution

Figure 7 shows the systematical comparisons of concentration evolutions of exhaled CO2 under the conditions without and with a face mask during breathing, coughing, and talking. It can be observed from Figs. 7(a), 7(c), and 7(e) that, when not wearing a face mask, compared to breathing, coughing and speaking with high initial momentum can result in higher pollutant concentrations in front of the mouth. Additionally, as shown in Figs. 7(b), 7(d), and 7(f), when wearing a face mask, for all three expiratory activities, exhaled pollutants mainly concentrated inside the face mask, but some of them would leak from the top and bottom gaps, especially during talking. The concentration of leaked pollutants at different locations depended on the gap sizes between the face mask and user's face (mainly influenced by mask-wearing habits and mask types) and the pulmonary ventilation rates during expiratory activities (mainly influenced by age, gender, and activity level). These results indicated that wearing a face mask can effectively block the forward throughflow and the horizontal dispersion of exhaled pollutants generated from various expiratory activities. However, leaks that cannot be ignored may still occur on the outermost surface of the mask and through gaps between the mask and the user's face.

FIG. 7.

A comparison of evolutions of the CO2 dimensionless concentration during breathing, coughing, and talking: breathing: (a) without a face mask and (b) with a face mask; coughing: (c) without a face mask and (d) with a face mask; talking: (e) without a face mask and (f) with a face mask.

FIG. 7.

A comparison of evolutions of the CO2 dimensionless concentration during breathing, coughing, and talking: breathing: (a) without a face mask and (b) with a face mask; coughing: (c) without a face mask and (d) with a face mask; talking: (e) without a face mask and (f) with a face mask.

Close modal

2. Quantitative analysis of exhaled pollutant and flow distribution

Figures 8 and 9 show the distribution of the exhaled pollutant concentration during breathing, coughing, and talking, under the conditions without and with wearing a face mask. The concentration data were collected at sampling points [see Fig. 3(c)] along the horizontal and vertical lines. The maximum values recorded within the duration of mouth breathing, coughing, and talking were used to indicate the highest potential level of pollutants at each sampling point. First, it can be seen from Figs. 8(a), 8(c), and 9(a) that, under the condition without a face mask, the highest concentration data were obtained from the point closet to the mouth, with a dimensionless concentration of 63.3%, 92.1%, and 83.78% for breathing, coughing, and talking, respectively. Under the condition with a face mask, the highest concentration at the same point was 0.9%, 16.0%, and 1.01% (not shown in this study) for breathing, coughing, and talking, respectively. Second, as shown in Figs. 8(b), 8(d), and 9(b)–9(d), the data of concentration along vertical lines were significantly lower compared to those along horizontal lines, and the pollutants were almost concentrated around the outermost surface and the top gap of the face mask. Overall, under specific boundary conditions in this study, wearing a face mask can reduce the dimensionless concentration of exhaled pollutants by 98.6%, 82.6%, and 98.8% in front of the mouth, during breathing, coughing, and talking, respectively.

FIG. 8.

Distribution of the exhaled pollutant concentration without/with a face mask: mouth breathing (30 s): (a) along the horizontal centerline of the mouth and (b) along the vertical lines Z1 and Z2; coughing (2 s): (c) along the horizontal centerline of the mouth and (d) along the vertical lines Z1 and Z2; The exhaled pollutant concentration (C′) represented the concentration after normalization by the initial exhaled CO2 concentration (C0).

FIG. 8.

Distribution of the exhaled pollutant concentration without/with a face mask: mouth breathing (30 s): (a) along the horizontal centerline of the mouth and (b) along the vertical lines Z1 and Z2; coughing (2 s): (c) along the horizontal centerline of the mouth and (d) along the vertical lines Z1 and Z2; The exhaled pollutant concentration (C′) represented the concentration after normalization by the initial exhaled CO2 concentration (C0).

Close modal
FIG. 9.

Concentration distribution of exhaled pollutants during talking (30 s) without/with a face mask (a) along the horizontal centerline of the mouth, (b) along the vertical line Z1, (c) along the vertical line Z1, and (d) along the vertical line Z2 with a face mask.

FIG. 9.

Concentration distribution of exhaled pollutants during talking (30 s) without/with a face mask (a) along the horizontal centerline of the mouth, (b) along the vertical line Z1, (c) along the vertical line Z1, and (d) along the vertical line Z2 with a face mask.

Close modal

1. Effectiveness in reducing exposure risk of the doctor

Figure 10 shows the airflow patterns in the room (XZ plane, Y = 3.06 m) under different conditions without air cleaners. The velocity in the plane (Y = 3.06 m) is relatively small (less than 0.3 m/s), with the dominant flow being the human thermal plume around the doctor. Figure 11 shows airflow patterns (XZ plane, Y = 3.06 m) and particle distributions at the moment of 300 s under the conditions without and with air cleaners (with different locations and supply air directions). First, as shown in Fig. 11(a), when the patient (pink manikin) was not wearing a face mask, the airflow distributions around the doctor's breathing zone were largely affected by the supply air from air cleaner on the desk (D–H and D–V conditions), compared to the condition with no air cleaner used. The supply air created a vortex in front of the doctor's chest, altering the airflow near her breathing zone and directing exhaled particles upward through the thermal plume, thereby reducing the risk of inhalation exposure. Additionally, in the F–H condition, the horizontal supply air from the air cleaner on the floor can directly remove particles away from the breathing zones of both the patient and the doctor. However, in the conditions of W–H, W–V, and F–V, particle distributions were similar to those observed when no air cleaner was present, with particles mainly accumulating in the breathing zone and above the doctor's head. Second, in Fig. 11(b), it can be seen that when the patient was wearing a face mask, particles released from the patient's mouth mostly concentrated around her face and above her head. This can be attributed to particle leakage through gaps near the nose and chin. The leaked particles dispersed to the upper zone and then spread throughout the outpatient room, while only a few particles being present in the doctor's breathing zone. It should be noted that the particle distribution in the doctor's breathing zone was influenced by various factors, including room airflow, expiratory flow, human thermal flume, and airflows generated by air cleaners. The airflow velocities around the breathing zone were all below 0.3 m/s;72 thus, the supply air from air cleaners would not cause any drafts for the doctor. These results indicated that the airflow distribution around the breathing zones of doctors and patients was crucial to the particle distribution, and the impact of air cleaner on particle distribution depended on its location and the direction of air supply.

FIG. 10.

Airflow patterns in the room (XZ plane, Y = 3.06 m) under different conditions without air cleaners: (a) patient without a face mask and (b) patient with a face mask.

FIG. 10.

Airflow patterns in the room (XZ plane, Y = 3.06 m) under different conditions without air cleaners: (a) patient without a face mask and (b) patient with a face mask.

Close modal
FIG. 11.

Airflow patterns around the doctor's breathing zone (XZ plane, Y = 3.06 m) and particle distributions around the patient and the doctor at 300 s without and with air cleaners (with different locations and supply air directions): (a) patient (pink manikin) without a face mask and (b) patient (pink manikin) with a face mask; the term “particle birth time” refers to the moment when particles release from the patient's mouth.

FIG. 11.

Airflow patterns around the doctor's breathing zone (XZ plane, Y = 3.06 m) and particle distributions around the patient and the doctor at 300 s without and with air cleaners (with different locations and supply air directions): (a) patient (pink manikin) without a face mask and (b) patient (pink manikin) with a face mask; the term “particle birth time” refers to the moment when particles release from the patient's mouth.

Close modal

Figure 12 shows the temporal evolutions of the doctor's inhalation fraction (IF) under the conditions without and with air cleaners (with different locations and supply air directions). First, when no air cleaner present [see Figs. 12(a) and 12(b)], it can be seen that, when the patient without (with) wearing a face mask, increasing the background ventilation flow rate from 60 to 120 m3/h would increase (decrease) the average IF (over 300 s) for the doctor from 0.6% (0.007%) to 0.8% (0.005%). Additionally, the IF values under these conditions were higher than those under some properly designed conditions with 50 m3/h air cleaners used. These results indicated that the proper use of low-volume air cleaner can be one feasible method to compensate the insufficient background ventilation. Second, when the patient was not wearing a face mask [as shown in Fig. 12(a)], the IF initially increased and then stabilized over time. The average IF (over 300 s) under different conditions ranged from 0.06% (D–H) to 1.21% (W–H). Conversely, when the patient was wearing a face mask [as shown in Fig. 12(b)], the IF value started to sharply increase after the moment of 100 s for all conditions except for the condition of W–V. It was possible that particles from a patient wearing face mask moved upwards to the upper zone and required some time to be inhaled by the doctor. The average IF (over 300 s) under different conditions ranged from 0.0007% (W–V) to 0.0110% (F–H). Overall, when the patient was not wearing a face mask, a desk-mounted air cleaner with horizontal supply air (D–H) can effectively reduce the doctor's IF during consultations. Additionally, when the patient was wearing a face mask, regardless of the use of air cleaners, the doctor's IF was considerably lower than that when the patient was not wearing a face mask.

FIG. 12.

Temporal evolutions of the inhalation fraction (IF) of the doctor without and with air cleaners (with different locations and supply air directions): (a) patient (IP) without a face mask and (b) patient (IP) with a face mask.

FIG. 12.

Temporal evolutions of the inhalation fraction (IF) of the doctor without and with air cleaners (with different locations and supply air directions): (a) patient (IP) without a face mask and (b) patient (IP) with a face mask.

Close modal

Figure 13 shows the personal exposure reduction effectiveness (PERE) under different conditions of air cleaners. In Fig. 13(a), when the patient was not wearing a face mask, the horizontal supply air from a wall-mounted air cleaner (W–H) had a negative effect (PERE < 0), increasing inhalation exposure to particles for the doctor due to its inhibition of particle dispersion toward areas further from the breathing zone. In contrast, floor and desk-mounted air cleaners had positive effects on reducing inhalation exposure for the doctor. Among these conditions, the desk-mounted air cleaner with the horizontal supply air (D–H) provided the best performance, with an average PERE of 90.1%, followed by the condition of F–H (74.8%). In Fig. 13(b), when the patient was wearing a face mask, the floor-mounted air cleaner with the horizontal supply air (F–H) significantly increased the inhalation exposure to particles for the doctor (at the moment of 120 s). This can be explained by that the horizontal supply air accelerated the flow of particles toward the breathing zone of the doctor. In contrast, for other conditions, it took more time for particles to move toward the breathing zone of the doctor. Among these conditions, the wall-mounted air cleaner with the vertical upward supply air (W–V) was most effective, with an average PERE of 91.2%, followed by the condition of D–H (64.7%). Overall, when the patient without (with) wearing a face mask, the low-volume air cleaner under the condition of D–H (W–V) provided the best performance in reducing doctor's exposure risk, with an average PERE of 90.1% (91.2%).

FIG. 13.

Personal exposure reduction effectiveness (PERE) under different conditions of air cleaners: (a) patient (IP) without a face mask and (b) patient (IP) with a face mask. Note: at the moment of 60 s, in (b), PERE was not available due to the CI,0 with the value of zero [see Eq. (23)] for the condition without the air cleaner.

FIG. 13.

Personal exposure reduction effectiveness (PERE) under different conditions of air cleaners: (a) patient (IP) without a face mask and (b) patient (IP) with a face mask. Note: at the moment of 60 s, in (b), PERE was not available due to the CI,0 with the value of zero [see Eq. (23)] for the condition without the air cleaner.

Close modal

2. Removal efficiency of aerosol particles

Figure 14 shows the temporal evolutions of removal efficiency (RE) of air cleaners under different conditions. In Fig. 14(a), when the patient was not wearing a face mask, the RE of wall-mounted (W–H and W–V) and the desk-mounted with vertical upward supply air (D–V) air cleaners initially increased and then gradually stabilized over time. Among all conditions, the highest RE was observed with the condition of W–H (50.2%), followed by the condition of W–V (40.2%). Additionally, in Fig. 14(b), when the patient was wearing a face mask, the RE of air cleaners gradually increased over time. Among all conditions, the highest RE was observed with the condition of W–V (1.3%). The large differences in RE of air cleaners can be explained by that, when the patient was wearing a face mask, the dispersion direction of exhaled particles was redirected upward, more time was thus needed for the air cleaner to effectively remove them. Overall, when the patient was not wearing a face mask, low-volume air cleaner under the condition of W–H provided the best performance in removing indoor particles. However, when the patient was wearing a face mask, the efficiency of air cleaners in removing indoor particles was limited.

FIG. 14.

Temporal evolutions of the removal efficiency (RF) under different conditions of air cleaners: (a) patient (IP) without a face mask and (b) patient (IP) with a face mask.

FIG. 14.

Temporal evolutions of the removal efficiency (RF) under different conditions of air cleaners: (a) patient (IP) without a face mask and (b) patient (IP) with a face mask.

Close modal

Figure 15 depicts the distribution of particles released from the patient's mouth under the condition of W–H, which exhibited the highest RE among the conditions examined. It can be observed that the number of particles increased over time, and the indoor distribution of particles varied due to different dispersion characteristics of particle sources. In Fig. 15(a), when the patient was not wearing a face mask, particles were mainly distributed in the zone between the doctor and the patient, as well as in the upper space of the room. However, when the patient was wearing a face mask [see Fig. 15(b)], particles were observed to be distributed above the patient's head and in the zone behind the patient, due to the effect of room ventilation. These results indicated that the implementation of air cleaners could not quickly reduce indoor particle concentration. Therefore, to prevent cross-infection among patients visiting later due to the accumulation of virus-laden particles in outpatient rooms, particularly during pandemics when it was mandatory for patients to wear face masks, it was critical to implement advanced measures to rapidly remove particles generated by infected patients.

FIG. 15.

Distribution of particles released from the patient's mouth with the wall-mounted air cleaner with horizontal supply air (W–H): (a) patient (IP) without a face mask and (b) patient (IP) with a face mask.

FIG. 15.

Distribution of particles released from the patient's mouth with the wall-mounted air cleaner with horizontal supply air (W–H): (a) patient (IP) without a face mask and (b) patient (IP) with a face mask.

Close modal

1. Influence of ventilation configurations

While the use of air cleaner cannot rapidly remove the exhaled particles from an infected patient, the accumulation of particles indoors posted a risk of cross-infection to patients visiting later. To rapidly remove virus-laden particles, a local ventilation system with four different ventilation configurations was designed for situations where it was mandatory for patients to wear face masks. The distribution of indoor particles at the moment of 300 s was illustrated in Fig. 16. It can be seen that the configuration of the local ventilation system significantly affected the distribution of indoor particles. The local ventilation system with the top outlet [configuration C, see Fig. 16(c)] resulted in the least particle accumulation indoors, as it effectively removed particles leaking from the top gap of the face mask and moving upward due to thermal plumes. Despite the local ventilation system with ventilation configurations A and B also being equipped with top outlets, more severe particle accumulation was observed compared to that with only the top outlet. Additionally, the application of the local ventilation system with side air supply and exhaust vent [configuration D, see Fig. 16(d)] led to the most significant indoor particle accumulation. These results emphasized the importance of the design configuration of the local ventilation system on the removal efficiency of particles generated by a user.

FIG. 16.

Distribution of particles released from the patient's mouth under different ventilation configurations of the local ventilation system: (a) configuration A: inlet (side vent II) and outlet (side vent I and top vent), (b) configuration B: inlet (side vent I and II) and outlet (top vent), (c) configuration C: outlet (top vent), and (d) configuration D: inlet (side vent II) and outlet (side vent I).

FIG. 16.

Distribution of particles released from the patient's mouth under different ventilation configurations of the local ventilation system: (a) configuration A: inlet (side vent II) and outlet (side vent I and top vent), (b) configuration B: inlet (side vent I and II) and outlet (top vent), (c) configuration C: outlet (top vent), and (d) configuration D: inlet (side vent II) and outlet (side vent I).

Close modal

Figure 17 shows the average PERE (over 300 s) and the temporal evolutions of RE for the local ventilation system with four configurations. In Fig. 17(a), using the local ventilation system with configuration A had a negative effect, which increased the doctor's risk of inhalation exposure to particles by 253.3% (PERE) compared to that under the condition without a local ventilation system. This was due to that the simultaneous use of side exhaust and top exhaust caused particles to disperse into the air from the baffle closest to the doctor and then increased particle accumulation in the doctor's breathing zone. However, the local ventilation system with the ventilation configuration C had the best performance in reducing exposure risk of the doctor with a PERE of 100%, followed by the configuration D (92.2%) and then the configuration B (87.7%). In Fig. 17(b), the local ventilation systems with configurations A, B, and C all exhibited high particle removal efficiency (>80%), with the configuration C giving the highest average RE of 88.5%. Conversely, the lowest RE was observed for the configuration D due to the inefficiency of side exhaust in removing a large number of particles [see Fig. 16(d)]. Overall, the local ventilation system with configuration C performed best in both reducing doctor's exposure risk and removing particles in a rapid way.

FIG. 17.

The average PERE (over 300 s) and temporal evolutions of RE for the local ventilation system with four different ventilation configurations: (a) PERE and (b) RE.

FIG. 17.

The average PERE (over 300 s) and temporal evolutions of RE for the local ventilation system with four different ventilation configurations: (a) PERE and (b) RE.

Close modal

2. Influence of ventilation rates

To reduce energy consumption and alleviate potential noise problems resulting from excessive ventilation rate, we further investigated the influence of the exhaust ventilation rate on the effectiveness of the local ventilation system with configuration C (local exhaust), as shown in Fig. 18. The results indicated that the local exhaust displayed excellent performance in reducing the doctor's exposure risk and rapidly removing particles, with an average PERE of 96.6%–100% and an RE of 85.7%–88.5% within the airflow rate range of 30–50 m3/h. Hence, we recommend adjusting the airflow rate within this range based on personnel requirements while guaranteeing the effectiveness of the local exhaust.

FIG. 18.

The PERE and RE of the local ventilation system with configuration C under different exhaust ventilation rates: (a) PERE and (b) RE.

FIG. 18.

The PERE and RE of the local ventilation system with configuration C under different exhaust ventilation rates: (a) PERE and (b) RE.

Close modal

The present study showed that wearing a face mask had a significant impact on pollutant distribution in the human breathing zone under various expiratory activities. Wearing a face mask can effectively obstruct the horizontal dispersion of pollutants compared to the condition without a face mask. Exhaled pollutants were mainly concentrated within the face mask during breathing and coughing, while obvious leakage of pollutants at the top and bottom gaps between the face mask and the user's face was observed during talking. Previous experimental studies by Kolewe et al.73 and Viola et al.74 reported that pollutants also leaked from the side gap during the steady breathing flow at 17.7 L/min and coughing with a peak velocity of 8 m/s. The characteristics of leakage were definitely dependent on the dynamics of expiratory activities, facile features, mask wearing habits, and the type of pollutants. Additionally, wearing a face mask greatly alters the trajectory of exhaled pollutants, particularly those generated from expiratory activities with high momentum, resulting in high concentrations of pollutants on the outermost surface of the face mask and in the region above the head. Overall, wearing face masks can serve as an effective measure to reduce the risk of cross-infection between people in short distances by inhibiting the horizontal dispersion of exhaled pollutants.75 However, considering the leakages, the use of a face mask alone may not be sufficient, especially in high-risk scenarios.

The results of the present study suggested that the proper use of air cleaners with low flow rates can effectively reduce the exposure risk of doctors in outpatient rooms. Air cleaners have been widely used in various places such as hospital wards,76 dental clinics,77 and office rooms,34 where its proper use, including appropriate locations and supply air directions, can effectively remove indoor particles. Past studies mainly focused on the ones with high flow rates, which, however, could cause high noise and energy consumption. The findings of the present study indicated that, first, when the patient without (with) wearing a face mask, increasing the background ventilation flow rate from 60 to 120 m3/h would increase (decrease) the average IF (over 300 s) for the doctor from 0.6% (0.007%) to 0.8% (0.005%), with an average PERE of −33.3% (28.6%). Second, during normal periods when patients are not mandatory to wear face masks, a desk-mounted air cleaner with a horizontal supply air (D–H) was suggested to be applied in outpatient rooms. This air cleaner can be placed on the desk in front of the doctor, slightly closer to the patient, making the direction of the supply air opposite to the doctor's face [see Fig. 1(a) for the layout]. A background ventilation flow rate of 60 m3/h plus the desk-mounted air cleaner with a horizontal supply air can effectively reduce the doctor's inhalation exposure risk, with an average PERE of 90.1%. This was achieved mainly by the fact that the horizontal supply air changed the airflow pattern near the doctor's breathing zone, causing the airflow carrying particles to move away from the doctor's breathing zone. During pandemic periods, the use of a face mask by patients should be the primary method to reduce the risk of inhalation exposure for doctors. Although a background ventilation flow rate of 60 m3/h plus the wall-mounted air cleaner with vertical upward supply air (W–V) was most effective, with an average PERE of 91.2%, virus-laden particles released from infected patients can accumulate indoors, increasing the risk of cross-infection to the doctor and the patients visiting later. In this circumstance, low-volume air cleaner can provide equivalent clean air to dilute the particles indoors.

The results in Sec. III D indicated that when the patient was wearing a face mask, using a local ventilation system, equipped with a physical board and employing various ventilation configurations (except for configuration A), can considerably reduce the doctor's exposure risk and indoor particle concentration. Previous studies reported that physical boards, such as those used in canteens,78 aircraft cabins,60 and classrooms,79 can effectively reduce aerosol transmission in various settings. However, the use of physical boards may not effectively remove indoor particles. Park et al.68 numerically investigated various ventilation configurations for the fast removal of pathogen-laden aerosols in a negative-pressure unit, indicating that ceiling air exhausts mounted above patients can remove 99% of aerosols within 4.5 min. They emphasized the significance of preventing the wide dispersion of aerosols released during breathing and their rapid evacuation through air exhausts. In this study, during pandemic periods when patients are mandatory to wear face masks, as aforementioned in Sec. IV B, increasing the background ventilation flow rate can reduce the exposure risk of doctors, but it was more effective to use a background ventilation flow rate of 60 m3/h plus the local exhaust system with flow rates of 30–50 m3/h above the patient's head. This system can remove 85.7%–88.5% of particles within 60 s and significantly reduce the risk of inhalation exposure to infectious particles for doctors, achieving PERE values ranging from 96.6% to 100%. Additionally, compared to the local ventilation system with other ventilation configurations, the local exhaust with top outlet performs the best in terms of efficiency, simplification, and energy saving. In practice, a detachable local exhaust with the top outlet in combination with chairs may be one feasible approach, but alternative solutions may also exist.

The present study was intended to explore low-cost solutions for effectively mitigating the risk of cross-infection in outpatient rooms with inadequate ventilation that widely exist in countries and regions with limited resources. It therefore focused on only low background ventilation flow rates and low flow rates of air cleaner and local exhaust system. In addition, this study was limited with fixed positions of a doctor and a patient in seated posture.

This study first characterized the dispersion of exhaled pollutants during various expiratory activities (namely, breathing, coughing, and talking) and then evaluated the effectiveness of a low-volume air cleaner and local ventilation system in mitigating the risk of cross-infection in hospital outpatient rooms with inadequate ventilation. The following conclusions are made.

  1. Wearing a face mask can effectively reduce the exhaled pollutant concentration in front of the face mask by 98.6%, 82.6%, and 98.8% during breathing, coughing, and talking, respectively. However, the non-negligible leakages through the gaps between the mask and the user's face lead to the accumulation of exhaled particles indoors, suggesting the necessity of combining a face mask with effective and fast removal methods to mitigate the exposure risk of doctor and follow-up patients.

  2. Under the conditions when patients are not mandatory to wear face masks, a background ventilation flow rate of 60 m3/h plus a 50 m3/h desk-mounted air cleaner with horizontal supply air (D–H) is effective in reducing the risk of inhalation exposure of doctor with an average PERE of 90.1%, which is 137% higher than that under only the background ventilation with the stipulated flow rate of 120 m3/h. This is achieved because the horizontal supply air of the air cleaner changes the airflow pattern near the doctor's breathing zone, redirecting the particles carried by the airflow away from the doctor's breathing zone. The air cleaner can be placed on the desk in front of the doctor, slightly closer to the patient, making the direction of the supply air opposite to the doctor's face.

  3. Under the conditions when patients are mandatory to wear face masks, the leaked particles from patients' face masks and their accumulation indoors cannot be ignored. A background ventilation flow rate of 60 m3/h plus a 30–50 m3/h local exhaust above patient's head removes 85.7%–88.5% of exhaled particles and considerably reduce the exposure risk of the doctor, achieving an average PERE value of 96.6%–100%, which is approximately 70% higher than under only the background ventilation with the stipulated flow rate of 120 m3/h. In practice, a detachable local exhaust with the top outlet in combination with a chair may be one feasible solution.

This study was supported by the Fundamental Research Funds for the Central Universities (No. 531118010378). The authors wish to thank Professor A. Melikov from the Technical University of Denmark for his valuable discussions and suggestions.

The authors have no conflicts to disclose.

Jie Zong: Conceptualization (equal); Formal analysis (equal); Investigation (equal); Methodology (equal); Writing – original draft (equal); Writing – review & editing (equal). Chen Lin: Methodology (equal); Resources (equal); Writing – review & editing (equal). Zhengtao Ai: Conceptualization (equal); Methodology (equal); Project administration (equal); Resources (equal); Supervision (equal); Writing – review & editing (equal).

The data that support the findings of this study are available from the corresponding author upon reasonable request.

1.
J.
Yang
,
C.
Sekhar
,
D. K. W.
Cheong
, and
B.
Raphael
, “
A time-based analysis of the personalized exhaust system for airborne infection control in healthcare settings
,”
Sci. Technol. Built Environ.
21
,
172
178
(
2015
).
2.
Z. T.
Ai
and
A. K.
Melikov
, “
Airborne spread of expiratory droplet nuclei between the occupants of indoor environments: A review
,”
Indoor Air
28
,
500
524
(
2018
).
3.
J. K.
Gupta
,
C. H.
Lin
, and
Q.
Chen
, “
Flow dynamics and characterization of a cough
,”
Indoor Air
19
,
517
525
(
2009
).
4.
F. A.
Berlanga
,
I.
Olmedo
, and
M.
Ruiz de Adana
, “
Experimental analysis of the air velocity and contaminant dispersion of human exhalation flows
,”
Indoor Air
27
,
803
815
(
2017
).
5.
C.
Darquenne
,
A. A. T.
Borojeni
,
M. J.
Colebank
,
M. G.
Forest
,
B. G.
Madas
,
M.
Tawhai
, and
Y.
Jiang
, “
Aerosol transport modeling: The key link between lung infections of individuals and populations
,”
Front. Physiol.
13
,
923945
(
2022
).
6.
Z.
Li
,
H.
Wang
,
X.
Zhang
,
T.
Wu
, and
X.
Yang
, “
Effects of space sizes on the dispersion of cough-generated droplets from a walking person
,”
Phys. Fluids
32
,
121705
(
2020
).
7.
J.
Wang
and
G.
Du
, “
COVID-19 may transmit through aerosol
,”
Ir. J. Med. Sci.
189
,
1143
1144
(
2020
).
8.
S. K.
Das
,
J. E.
Alam
,
S.
Plumari
, and
V.
Greco
, “
Transmission of airborne virus through sneezed and coughed droplets
,”
Phys Fluids
32
,
097102
(
2020
).
9.
L.
Bourouiba
,
E.
Dehandschoewercker
, and
J. W. M.
Bush
, “
Violent expiratory events: On coughing and sneezing
,”
J. Fluid Mech.
745
,
537
563
(
2014
).
10.
C.
Xu
,
P. V.
Nielsen
,
G.
Gong
,
L.
Liu
, and
R. L.
Jensen
, “
Measuring the exhaled breath of a manikin and human subjects
,”
Indoor Air
25
,
188
197
(
2015
).
11.
M.
Abkarian
,
S.
Mendez
,
N.
Xue
,
F.
Yang
, and
H. A.
Stone
, “
Speech can produce jet-like transport relevant to asymptomatic spreading of virus
,”
Proc. Natl. Acad. Sci. U.S.A.
117
,
25237
25245
(
2020
).
12.
C. Y. H.
Chao
,
M. P.
Wan
,
L.
Morawska
,
G. R.
Johnson
,
Z. D.
Ristovski
,
M.
Hargreaves
,
K.
Mengersen
,
S.
Corbett
,
Y.
Li
,
X.
Xie
, and
D.
Katoshevski
, “
Characterization of expiration air jets and droplet size distributions immediately at the mouth opening
,“ J. Aerosol Sci.
40
,
122
133
(
2009
).
13.
B.
Wang
,
H.
Wu
, and
X. F.
Wan
, “
Transport and fate of human expiratory droplets—A modeling approach
,”
Phys. Fluids
32
,
083307
(
2020
).
14.
J. K.
Gupta
,
C. H.
Lin
, and
Q.
Chen
, “
Characterizing exhaled airflow from breathing and talking
,”
Indoor Air
20
,
31
39
(
2010
).
15.
J. W.
Tang
,
A. D.
Nicolle
,
J.
Pantelic
,
M.
Jiang
,
C.
Sekhr
,
D. K.
Cheong
, and
K. W.
Tham
, “
Qualitative real-time schlieren and shadowgraph imaging of human exhaled airflows: An aid to aerosol infection control
,”
PloS One
6
,
e21392
(
2011
).
16.
Z.
Lei
,
J.
Yang
,
Z.
Zhuang
, and
R.
Roberge
, “
Simulation and evaluation of respirator faceseal leaks using computational fluid dynamics and infrared imaging
,”
Ann. Occup. Hyg.
57
,
493
506
(
2013
).
17.
R.
Perić
and
M.
Perić
, “
Analytical and numerical investigation of the airflow in face masks used for protection against COVID-19 virus—Implications for mask design and usage
,”
J. Appl. Fluid Mech.
13
,
1911
1923
(
2020
).
18.
C.
Chen
,
C. H.
Lin
,
Z.
Jiang
, and
Q.
Chen
, “
Simplified models for exhaled airflow from a cough with the mouth covered
,”
Indoor Air
24
,
580
591
(
2014
).
19.
X.
Zhang
and
J.
Wang
, “
Dose-response relation deduced for coronaviruses from coronavirus disease 2019, severe acute respiratory syndrome, and middle east respiratory syndrome: meta-analysis results and its application for infection risk assessment of aerosol transmission
,”
Clin. Infect. Dis.
73
,
e241
e245
(
2021
).
20.
J.
Zhang
, “
Integrating IAQ control strategies to reduce the risk of asymptomatic SARS CoV-2 infections in classrooms and open plan offices
,”
Sci. Technol. Built Environ.
26
,
1013
1018
(
2020
).
21.
C.
Graetz
,
P.
Düffert
,
R.
Heidenreich
,
M.
Seidel
, and
C. E.
Dörfer
, “
The efficacy of an extraoral scavenging device on reducing aerosol particles ≤ 5 μm during dental aerosol-generating procedures: An exploratory pilot study in a university setting
,”
BDJ Open
7
,
19
(
2021
).
22.
M. H.
Weir
and
C. N.
Haas
, “
Quantification of the effects of age on the dose response of variola major in suckling mice
,”
Hum. Ecol. Risk Assess.: Int. J.
15
,
1245
1256
(
2009
).
23.
Z.
Liu
,
J.
Ma
,
J.
Lv
,
Y.
Wang
,
J.
He
,
G.
Yao
, and
G.
Cao
, “
Transmission characteristics of infectious pathogen-laden aerosols in a negative-pressure operating room
,”
J. Hazard. Mater.
446
,
130650
(
2023
).
24.
S.
Kumar
and
M. D.
King
, “
Numerical investigation on indoor environment decontamination after sneezing
,”
Environ. Res.
213
,
113665
(
2022
).
25.
G.
Xu
,
X.
Xing
,
V. T.
Nguyen
,
H. J.
Poh
, and
J.
Lou
, “
CFD-driven optimization of air supplies deployment in an air-conditioned office
,”
IOP Conf. Ser.: Earth Environ. Sci.
238
,
012054
(
2019
).
26.
J. A.
Khan
,
C. E.
Feigley
,
E.
Lee
,
M. R.
Ahmed
, and
S.
Tamanna
, “
Effects of inlet and exhaust locations and emitted gas density on indoor air contaminant concentrations
,”
Build. Environ.
41
,
851
863
(
2006
).
27.
K.
Grosskopf
, “
Balancing ventilation and energy use in hospitals: A case study of bioaerosol transport in healthcare environments
,”
Antimicrob. Resist. Infect. Control
2
,
P396
(
2013
).
28.
R.
Cermak
,
A. K.
Melikov
,
L.
Forejt
, and
O.
Kovář
, “
Performance of personalized ventilation in conjunction with mixing and displacement ventilation
,”
HVAC&R Res.
12
,
295
311
(
2006
).
29.
X.
Li
,
Z.
Ai
,
J.
Ye
,
C. M.
Mak
, and
H. M.
Wong
, “
Airborne transmission during short-term events: Direct route over indirect route
,”
Build. Simul.
15
,
2097
2110
(
2022
).
30.
A. K.
Melikov
, “
Advanced air distribution: Improving health and comfort while reducing energy use
,”
Indoor Air
26
,
112
124
(
2016
).
31.
Z. D.
Bolashikov
,
M.
Barova
, and
A. K.
Melikov
, “
Wearable personal exhaust ventilation: Improved indoor air quality and reduced exposure to air exhaled from a sick doctor
,”
Sci. Technol. Built Environ.
21
,
1117
1125
(
2015
).
32.
H.
Qian
,
Y.
Li
,
H.
Sun
,
P. V.
Nielsen
,
X.
Huang
, and
X.
Zheng
, “
Particle removal efficiency of the portable HEPA air cleaner in a simulated hospital ward
,”
Build. Simul.
3
,
215
224
(
2010
).
33.
W. G.
Lindsley
,
R. C.
Derk
,
J. P.
Coyle
,
S. B.
Martin
, Jr.
,
K. R.
Mead
,
F. M.
Blachere
,
D. H.
Beezhold
,
J. T.
Brooks
,
T.
Boots
, and
J. D.
Noti
, “
Efficacy of portable air cleaners and asking for reducing indoor exposure to simulated exhaled SARS-CoV-2 aerosols—United States, 2021
,”
Morb. Mortal. Wkly. Rep.
70
,
972
976
(
2021
).
34.
H.
Ma
,
H.
Shen
,
T.
Shui
,
Q.
Li
, and
L.
Zhou
, “
Experimental study on ultrafine particle removal performance of portable air cleaners with different filters in an office room
,”
Int. J. Environ. Res. Public Health
13
,
102
(
2016
).
35.
J. H.
Lee
,
M.
Rounds
,
F.
McGain
,
R.
Schofield
,
G.
Skidmore
,
I.
Wadlow
,
K.
Kevin
,
A.
Stevens
,
C.
Marshall
,
L.
Irving
,
M.
Kainer
,
K.
Buising
, and
J.
Monty
, “
Effectiveness of portable air filtration on reducing indoor aerosol transmission: Preclinical observational trials
,”
J. Hosp. Infect.
119
,
163
169
(
2022
).
36.
J.
Cai
,
W.
Yu
,
B.
Li
,
R.
Yao
,
T.
Zhang
,
M.
Guo
,
H.
Wang
,
Z.
Cheng
,
J.
Xiong
,
Q.
Meng
, and
H.
Kipen
, “
Particle removal efficiency of a household portable air cleaner in real-world residences: A single-blind cross-over field study
,”
Energy Build.
203
,
109464
(
2019
).
37.
J.
Li
,
G.
Fan
,
Y.
Ou
, and
Q.
Deng
, “
Characteristics and control strategies of indoor particles: An updated review
,”
Energy Build.
294
,
113232
(
2023
).
38.
P.
Ole Fanger
, “
Human requirements in future air-conditioned environments
,”
Int. J. Refrig.
24
,
148
153
(
2001
).
39.
Z.
Zhai
and
I. D.
Metzger
, “
Insights on critical parameters and conditions for personalized ventilation
,”
Sustainable Cities Soc.
48
,
101584
(
2019
).
40.
Z.
Bolashikov
,
A.
Melikov
, and
M.
Brand
, “
Reduced exposure to coughed air by a novel ventilation method for hospital patient rooms
,” in
Proceedings of the International Symposium on Contamination Control
(
2012
).
41.
R. K.
Dygert
and
T. Q.
Dang
, “
Mitigation of cross-contamination in an aircraft cabin via localized exhaust
,”
Build. Environ.
45
,
2015
2026
(
2010
).
42.
A.
Khosronejad
,
C.
Santoni
,
K.
Flora
,
Z.
Zhang
,
S.
Kang
,
S.
Payabvash
, and
F.
Sotiropoulos
, “
Fluid dynamics simulations show that facial masks can suppress the spread of COVID-19 in indoor environments
,”
AIP Adv.
10
,
125109
(
2020
).
43.
T.
Dbouk
and
D.
Drikakis
, “
On respiratory droplets and face masks
,”
Phys. Fluids
32
,
063303
(
2020
).
44.
J.
Zong
,
Z.
Ai
, and
G.
Ma
, “
Accurate evaluation of inhalation exposure based on CFD predicted concentration in the breathing zone towards personalized and smart control
,”
J. Build. Eng.
71
,
106404
(
2023
).
45.
J.
Watanabe
,
Y.
Iwamatsu-Kobayashi
,
K.
Kikuchi
,
T.
Kajita
,
H.
Morishima
,
K.
Yamauchi
,
W.
Yashiro
,
H.
Nishimura
,
H.
Kanetaka
, and
H.
Egusa
, “
Visualization of droplets and aerosols in simulated dental treatments to clarify the effectiveness of oral suction devices
,”
J. Prosthodontic Res
.
68
,
85
91
(
2023
).
46.
Z.
Jia
,
Z.
Ai
,
X.
Yang
,
C. M.
Mak
, and
H. M.
Wong
, “
Towards an accurate CFD prediction of airflow and dispersion through face mask
,”
Build. Environ.
229
,
109932
(
2023
).
47.
T.
Dbouk
and
D.
Drikakis
, “
On coughing and airborne droplet transmission to humans
,”
Phys. Fluids
32
,
053310
(
2020
).
48.
S.-B.
Kwon
,
J.
Park
,
J.
Jang
,
Y.
Cho
,
D.-S.
Park
,
C.
Kim
,
G.-N.
Bae
, and
A.
Jang
, “
Study on the initial velocity distribution of exhaled air from coughing and speaking
,”
Chemosphere
87
,
1260
1264
(
2012
).
49.
Z. T.
Ai
,
C. M.
Mak
,
N.
Gao
, and
J.
Niu
, “
Tracer gas is a suitable surrogate of exhaled droplet nuclei for studying airborne transmission in the built environment
,”
Build. Simul.
13
,
489
496
(
2020
).
50.
D. K.
Milton
,
M. P.
Fabian
,
B. J.
Cowling
,
M. L.
Grantham
, and
J. J.
McDevitt
, “
Influenza virus aerosols in human exhaled breath: Particle size, culturability, and effect of surgical masks
,”
PLoS Pathogens
9
,
e1003205
(
2013
).
51.
J. K.
Gupta
,
C.-H.
Lin
, and
Q.
Chen
, “
Transport of expiratory droplets in an aircraft cabin
,”
Indoor Air
21
,
3
11
(
2011
).
52.
J.
Redrow
,
S.
Mao
,
I.
Celik
,
J. A.
Posada
, and
Z.-G.
Feng
, “
Modeling the evaporation and dispersion of airborne sputum droplets expelled from a human cough
,”
Build. Environ.
46
,
2042
2051
(
2011
).
53.
Y.
Yan
,
X.
Li
,
Y.
Shang
, and
J.
Tu
, “
Evaluation of airborne disease infection risks in an airliner cabin using the Lagrangian-based Wells-Riley approach
,”
Build. Environ.
121
,
79
92
(
2017
).
54.
L.
Liu
,
J.
Wei
,
Y.
Li
, and
A.
Ooi
, “
Evaporation and dispersion of respiratory droplets from coughing
,”
Indoor Air
27
,
179
190
(
2017
).
55.
J.
Gralton
,
E. R.
Tovey
,
M.-L.
McLaws
, and
W. D.
Rawlinson
, “
Respiratory virus RNA is detectable in airborne and droplet particles
,”
J. Med. Virol.
85
,
2151
2159
(
2013
).
56.
C. C.
Wang
,
K. A.
Prather
,
J.
Sznitman
,
J. L.
Jimenez
,
S. S.
Lakdawala
,
Z.
Tufekci
, and
L. C.
Marr
, “
Airborne transmission of respiratory viruses
,”
Science
373
,
eabd9149 (2021)
.
57.
C.
Krittanawong
,
Y. K.
Qadeer
,
R. B.
Hayes
,
Z.
Wang
,
G. D.
Thurston
,
S.
Virani
, and
C. J.
Lavie
, “
PM2.5 and cardiovascular diseases: State-of-the-art review
,”
Int. J. Cardiol. Cardiovasc. Risk Prev.
19
,
200217
(
2023
).
58.
K.
An
and
J. C. H.
Fung
, “
An improved SST k−ω model for pollutant dispersion simulations within an isothermal boundary layer
,”
J. Wind Eng. Ind. Aerodyn.
179
,
369
384
(
2018
).
59.
S.
Elghobashi
, “
On predicting particle-laden turbulent flows
,”
Appl. Sci. Res.
52
,
309
329
(
1994
).
60.
K.
Talaat
,
M.
Abuhegazy
,
O. A.
Mahfoze
,
O.
Anderoglu
, and
S. V.
Poroseva
, “
Simulation of aerosol transmission on a Boeing 737 airplane with intervention measures for COVID-19 mitigation
,”
Phys. Fluids
33
,
033312
(
2021
).
61.
L.
Huang
,
W.
Hetharia
,
A.
Grech La Rosa
,
S.
Tavakoli
,
D.
Khojasteh
,
M.
Li
,
S.
Riyadi
,
D.
Setyawan
,
I. K. A. P.
Utama
, and
G.
Thomas
, “
Offshore COVID-19 risk assessment based on a fishing vessel
,”
Ocean Eng.
285
,
115408
(
2023
).
62.
G.
Busco
,
S. R.
Yang
,
J.
Seo
, and
Y. A.
Hassan
, “
Sneezing and asymptomatic virus transmission
,”
Phys. Fluids
32
,
073309
(
2020
).
63.
M.
Ahmadzadeh
and
M.
Shams
, “
Passenger exposure to respiratory aerosols in a train cabin: Effects of window, injection source, output flow location
,”
Sustainable Cities Soc.
75
,
103280
(
2021
).
64.
B.
Zhao
,
Y.
Zhang
,
X.
Li
,
X.
Yang
, and
D.
Huang
, “
Comparison of indoor aerosol particle concentration and deposition in different ventilated rooms by numerical method
,”
Build. Environ.
39
,
1
8
(
2004
).
65.
M. D.
Allen
and
O. G.
Raabe
, “
Slip correction measurements of spherical solid aerosol particles in an improved Millikan apparatus
,”
Aerosol Sci. Technol.
4
,
269
286
(
1985
).
66.
A.
Li
and
G.
Ahmadi
, “
Dispersion and deposition of spherical particles from point sources in a turbulent channel flow
,”
Aerosol Sci. Technol.
16
,
209
226
(
1992
).
67.
P. G.
Saffman
, “
The lift on a small sphere in a slow shear flow
,”
J. Fluid Mech.
22
,
385
400
(
1965
).
68.
J.
Park
,
K. S.
Lee
, and
H.
Park
, “
Optimized mechanism for fast removal of infectious pathogen-laden aerosols in the negative-pressure unit
,”
J. Hazard. Mater.
435
,
128978
(
2022
).
69.
A. O.
Mahgoub
,
S.
Ghani
,
M. M.
Rashwan
,
S. M.
Ismail
, and
E. A.
ElBialy
, “
Simulation of spectators' aerodynamic drag using porous models approximation
,”
Build. Environ.
184
,
107248
(
2020
).
70.
A. K.
Melikov
,
R.
Cermak
, and
M.
Majer
, “
Personalized ventilation: Evaluation of different air terminal devices
,”
Energy Build.
34
,
829
836
(
2002
).
71.
Z.
Zhang
and
Q.
Chen
, “
Experimental measurements and numerical simulations of particle transport and distribution in ventilated rooms
,”
Atmos. Environ.
40
,
3396
3408
(
2006
).
72.
C. J.
Kähler
,
T.
Fuchs
, and
R.
Hain
, “
Can mobile indoor air cleaners effectively reduce an indirect risk of SARS-CoV-2 infection by aerosols?
” Morbidity and Mortality Weekly Report (MMWR) (
2020
).
73.
E. L.
Kolewe
,
Z.
Stillman
,
I. R.
Woodward
, and
C. A.
Fromen
, “
Check the gap: Facemask performance and exhaled aerosol distributions around the wearer
,”
PLoS One
15
,
e0243885
(
2020
).
74.
I. M.
Viola
,
B.
Peterson
,
G.
Pisetta
,
G.
Pavar
,
H.
Akhtar
,
F.
Menoloascina
,
E.
Mangano
,
K. E.
Dunn
,
R.
Gabl
,
A.
Nila
,
E.
Molinari
,
C.
Cummins
,
G.
Thompson
,
T. M.
Lo
,
F. C.
Denison
,
P.
Digard
,
O.
Malik
,
M. J. G.
Dunn
,
C. M.
McDougall
, and
F. V.
Mehendale
, “
Face coverings, aerosol dispersion and mitigation of virus transmission risk
,”
IEEE Open J. Eng. Med. Biol.
2
,
26
35
(
2021
).
75.
Z.
Zhang
,
T.
Han
,
K. H.
Yoo
,
J.
Capecelatro
,
A. L.
Boehman
, and
K.
Maki
, “
Disease transmission through expiratory aerosols on an urban bus
,”
Phys. Fluids
33
,
015116
(
2021
).
76.
H.
Qian
,
Y.
Li
,
P. V.
Nielsen
, and
X.
Huang
, “
Spatial distribution of infection risk of SARS transmission in a hospital ward
,”
Build. Environ.
44
,
1651
1658
(
2009
).
77.
C.
Chen
,
B.
Zhao
,
W.
Cui
,
L.
Dong
,
N.
An
, and
X.
Ouyang
, “
The effectiveness of an air cleaner in controlling droplet/aerosol particle dispersion emitted from a patient's mouth in the indoor environment of dental clinics
,”
J. R. Soc. Interface
7
,
1105
1118
(
2010
).
78.
J.
Ye
,
Z.
Ai
, and
Y.
Chang
, “
Aerosol transmission in queuing and dining scenarios in canteens and the effectiveness of control measures
,”
J. Build. Perform. Simul.
16
,
660
679
(
2023
).
79.
M.
Abuhegazy
,
K.
Talaat
,
O.
Anderoglu
, and
S. V.
Poroseva
, “
Numerical investigation of aerosol transport in a classroom with relevance to COVID-19
,”
Phys. Fluids
32
,
103311
(
2020
).