A Manikin-Based Study of Particle Dispersion in a Vehicle Cabin

Abstract

Recently, there has been a growing interest in understanding how respiratory particles spread within passenger cars, especially in light of ongoing challenges posed by infectious diseases. This study experimentally investigates dispersion patterns of respiratory airborne particles (<1 µm) within these confined spaces. The main objective is to introduce a manikin-based method for studying particle dispersion and assessing in-cabin air quality. To achieve this, a respiratory manikin as a particle source has been developed and tested under various use-cases, including variations in source emission (breathing vs. speaking), the HVAC ventilation mode (fresh and recirculation), and the blower level of the HVAC system (low and high). The findings reveal that for an infection source on the first row of the vehicle when cabin airflow originates from the front panel, the seat directly behind the particle source is associated with the highest particle exposure, while the seat adjacent to the source offers the lowest exposure. Among the tested configurations, the recirculation mode with an active HEPA filter and high blower level shows the lowest particle concentration at recipients’ breath levels during both breathing and speaking. These findings can be used to enhance the design of passenger cars to reduce the transmission of potentially pathogen-laden particles.

1. Introduction

According to the National Human Activity Pattern Survey (NHAPS), an American resource for assessing exposure to environmental pollutants, individuals spend an average of about 6% of their time in enclosed vehicles [1]. Traveling in passenger cars, especially ride-share vehicles, has elevated the risk of respiratory disease transmission in the wake of the COVID-19 pandemic [2,3,4,5]. Respiratory infections primarily spread through virus-laden fluid particles, including airborne particles (≤5 μm) and droplets (>5 μm) [6] originating from the respiratory tract of an infected individual and released from the mouth and nose [7,8,9]. The significance of virus transmission through airborne particles (≤5 μm) has been widely acknowledged [10,11,12]. Due to their small size and low mass, airborne particles can remain suspended in the air for several hours, increasing the likelihood of penetrating deeper into the respiratory tract of susceptible individuals [12]. Predominant particle sizes released during human breathing and speaking are around 1 μm [13,14,15,16,17]. The airborne particles below 1 µm can carry various respiratory pathogens, including the measles virus (50–500 nm) [18], the influenza virus (100 nm–1 μm) [19], and SARS-CoV-2 (0.07–0.09 μm) [20,21]. To properly mitigate the risk of airborne transmission in vehicle cabins and develop strategies for enhancing air quality within the breathing zone of passengers, it is crucial to understand how aerosols are transmitted within these confined spaces.

Previous studies, employing computational fluid dynamics (CFD) and/or experimental approaches, have significantly contributed to understanding in-cabin air quality. Mathai et al. [22] conducted CFD simulations to assess airborne transmission risks within passenger cars, specifically between the driver and the rear-right passenger, with their focus only on window ventilation. Sarhan et al. [3] used CFD simulations to predict when and who will become infected by a coronavirus while sharing a passenger car with a patient of COVID-19 or similar viruses. Their model predicts the number of aerosol droplets inhaled by every individual inside the car cabin through breathing and speaking. Similarly, Arpino et al. [23] employed CFD simulations to analyze the spatial distribution of virus-laden respiratory particles under diverse conditions, considering variations in ventilation modes; airflow rates of a heating, ventilation, and air-conditioning (HVAC) system; emission use-cases (breathing vs. speaking); and the position of the infected individual within the car cabin. Despite these valuable contributions, their studies lack empirical validation through experimental tests.

Vehicle road tests are expensive and time-consuming; despite that, there are some studies focused on experimental investigations. Kumar et al. [24] conducted experimental studies to examine the trade-offs between in-car aerosol concentrations (from traffic exhaust emissions), ventilation settings, and respiratory infection transmission. In their experiments, occupant exposure to CO₂ levels served as a proxy for assessing COVID-19 transmission risks. The use of CO₂ as a surrogate for investigating airborne transmission is supported by Ai et al. [25]. However, it is acknowledged that particle simulation is a more effective approach than tracer gas if in experimental studies the number and size distribution of the generated particles can be accurately controlled and the influence of background concentration on measurement accuracy and repeatability can be eliminated. Lednicky et al. [26] used a real case of a COVID-19 patient in their experiments to assess the presence of viable virus particles within a car. The inclusion of an actual infected person adds a level of realism to the experiments. However, conducting comprehensive experiments including various use-cases involving human subjects in risky situations remains a challenge and may raise ethical dilemmas or questions about the well-being and safety of participants. In the study by Kim et al. [27], experiments were conducted to assess the risk of COVID-19 infection within a car under two air circulation methods, air conditioning and opening windows. However, a limitation of their research is that they assumed the rate of infectious aerosol production is the same throughout the entire vehicle, without accounting for specific locations of the infected individuals.

Despite advancements in understanding in-cabin air quality and infectious particle dispersion within vehicles, there remains still a necessity to conduct vehicle tests under various experimental use-cases while addressing the limitations present in the existing studies. The primary objective of this study is first to introduce a methodology involving a respiratory manikin as a particle emitter source for assessing cabin air quality. Second, this study aims to provide practical insights into the dispersion patterns of respiratory airborne particles (<1 µm) within passenger cars.

Experimentation in car cabins using a particle-emitting manikin is practically nonexistent in previous field tests. Manikin-based studies provide several advantages; there is no patient recruitment and no risk of severe adverse effects. Additionally, manikins allow for the creation of stable and replicable experimental conditions, which can be valuable for conducting comparative tests [28].

This study can improve the understanding of how respiratory particles disperse within vehicle cabins. While the concentration of particles alone is not a direct indicator for assessing health risks, it can offer insights into potential risks if the virus is airborne and remains stable. Accurate knowledge of how exhaled particles disperse will facilitate the prevention of disease propagation.

In this study, experiments were conducted using the manikin as a particle-emitting passenger positioned in the front-right seat of a car cabin. Three different scenarios were tested: (i) source emission variations (breathing vs. speaking), (ii) variations in the HVAC ventilation mode (fresh and recirculation), and (iii) variations in blower level of the HVAC system (low and high).

2. Materials and Methods

2.1. Structure of the Respiratory Manikin

The respiratory pattern of the manikin involved cycles of breathing that were repetitive and imitated the pattern of a real human. It included inhalation, exhalation, and a short pause between each inhalation and exhalation cycle. The manikin system was composed of four primary sections: a mixing section, an artificial lung, valve interconnections, and an artificial head; see Figure 1.

The main components of the mixing section included the aerosol generator ATM 222, a source of pressurized air, and a mixing chamber. The aerosol generator with 300 hPa nozzle pressure released Di-Ethyl-Hexyl Sebacate (DEHS) into the buffer tank. DEHS particles from the buffer tank were mixed with filtered pressurized air in the 1.0 L mixing chamber. The mixing section was connected to the artificial lung section through several valves. Two pumps, each with a volume of 3.0 L, acted as artificial lungs. A stepper motor and linear drive controlled the pumps during the breathing and speaking actions. Particles were released through the nose and mouth of the artificial head. The artificial head of the manikin closely resembled the human anatomy. The realistic modeling of the throat and oral cavity created a lifelike flow field when exhaling from the mouth and nose, as shown in Figure 2. The manikin system was controlled by a programmable logic controller (PLC) from Beckhoff and the windows control and automation technology (TwinCAT).

Settings of Respiratory Functions

Experiments were conducted involving two respiratory functions of the manikin, including breathing and speaking. The exhaled air volume per breath was 0.5 L, corresponding to a typical adult male at complete rest. Certainly, there are individual variations due to size, age, gender, and health state [29]. In the scope of this study, the differentiation between breathing and speaking modes in terms of particle emission rates relied exclusively on the respiration intensity, particularly concerning volume flow rate and frequency. The assumption made was that when speaking, the respiration intensity was higher compared to breathing [30,31].

Table 1. Settings of the respiratory manikin during breathing and speaking modes.

Respiratory Mode	Frequency in Cycles/min	Exhaled Volume in L/min	One Cycle Duration in Seconds	Delay Time Between Inhalation and Exhalation in Seconds
Breathing	10.0	5.0	6.0	0.5
Speaking	15.0	15.0	4.0	0.5

reports the main settings of the manikin.

The literature offers a range of insights into particle emission rates and particle sizes during breathing and speaking. For example, Gregson et al. [32] found that the number of particles produced during speaking “Happy Birthday” for 20 s is influenced by the loudness. Notably, particles released during breathing covered a wide range of speaking with different loudness. However, Asadi et al. [10] highlighted a significant increase in particle numbers during speaking compared to breathing. Bagheri et al. [33] found that the highest particle number concentrations produced by all the activities, including breathing and speaking, is found within the 0.1-1.0 µm diameter range.

After reviewing the literature, the focus of this study was placed on particles less than 1.0 µm due to their involvement in carrying various respiratory pathogens. Figure 3 compares particle concentrations across different size ranges for both breathing and speaking scenarios.

The calculation of the interquartile range (IQR = Q_0.75 − Q_0.25) for the boxplots, which shows the spread of the middle 50% of the data for particle concentrations, is presented in

Table 2. Interquartile range (IQR = Q_0.75 − Q_0.25) of particle concentrations across different particle sizes. Comparison between breathing and speaking cases.

Particle Size in µm	IQRbreathing in pcs./cm3	IQRspeaking in pcs./cm3	Relative Difference in IQR (Breathing vs. Speaking)
0.3	75.0	254.5	−71%
0.374	63.4	211.6	−70%
0.465	42.6	154.2	−72%
0.579	17.0	60.2	−72%
0.721	9.1	33.3	−73%
0.897	5.0	17.7	−72%
1.117	1.3	4.1	−68%
Average			−71%

. The relative difference in particle concentration between breathing and speaking cases is calculated according to Equation (1), and the results are shown in Table 2:

$${\displaystyle \frac{{\mathrm{I}\mathrm{Q}\mathrm{R}}_{breathing}-{\mathrm{I}\mathrm{Q}\mathrm{R}}_{speaking}}{{\mathrm{I}\mathrm{Q}\mathrm{R}}_{speaking}}}\times 100$$

(1)

The results of the IQR indicated that on average, the concentrations during the breathing case are 70% lower than during the speaking case.

The depicted data in Figure 3 were the result of experiments conducted inside a duct with a setup that is shown in Figure 4. The duct section was configured with a constant volume flow rate of 350 m³/h, wherein the air was mixed with the aerosols emitted by the manikin. The experiments were conducted for both breathing and speaking modes, each lasting 15 min. The resulting mixture of particles and air was measured at the end of the duct using TSI Optical Particle Sizer spectrometer model 3330 (OPS). Section 2.2.2 provides detailed information about the OPS device and its sampling setup.

2.2. Experimental Setup

2.2.1. Case Study: Car Cabin

The experiments were conducted within a car cabin (4-door mid-sized gasoline station wagon) situated in a stationary position in the climate-controlled test hall of the Institute for Energy Efficient Buildings and Indoor Climate at RWTH Aachen University. The airflow within the car originated only from the front panel through four vents (left, center-left, center-right, and right), and the windows were closed. The study comprised experiments conducted in both fresh and recirculation ventilation modes. In the fresh mode, external air was drawn into the vehicle cabin, whereas the recirculation mode involved the circulation of cabin air, as illustrated in Figure 5. Notably, the air emitted from the vents in both use-cases (fresh and recirculation) was free of particles due to the presence and activation of high-efficiency particulate air (HEPA) filters. Furthermore, experiments were conducted under different blower levels. In the fresh mode, tests were performed at blower levels 4 (low) and 6 (high). To keep the total air volume flow comparable in both fresh and recirculation modes, blower levels 3 (low) and 5 (high) in recirculation mode were used. In this study, when referring to blower levels, “low” designates level 3 in recirculation mode and level 4 in fresh mode, while “high” means blower level 5 in recirculation mode and blower level 6 in fresh mode.

Throughout the experiments, the cabin temperature was maintained at its lowest setting to ensure that temperature fluctuations did not influence particle measurements within the cabin.

Across the study, the manikin representing a particle-emitting passenger was positioned at the front-right seat. The manikin did not emit any heat, neither from its skin nor from exhaled air. During the breathing use-case, the manikin’s head was oriented toward the front of the vehicle, while during the speaking use-case, the head was rotated 60 degrees toward the driver’s side, as shown in Figure 6.

2.2.2. Instrumentation and Measurement Locations

In this study, particle concentrations were measured using TSI optical particle sizer spectrometer (OPS) model 3330. The OPS device has a size resolution of <5% at 0.5 μm and a concentration range of 0–3000 particles/cm³ [34]. For precise measurements, the OPS instrument was configured with 16 channels, covering a particle size range from 0.3 to 10 µm (0.3, 0.374, 0.465, 0.579, 0.721, 0.897, 1.117, 1.391, 1.732, 2.156, 2.685, 3.343, 4.162, 5.182, 6.451, 8.032, and 10 µm). Before commencing the experiments, as multiple OPS devices were to be operated simultaneously, a factor referred to in this study as “scaling factor” was computed and applied to concentrations of particles below 1 µm. The concentrations of particles exceeding 1 µm accounted for only about 1% of the entire range and were considered negligible.

The scaling factor was determined by placing five OPS devices under consideration at a designated sampling point inside the cabin. The vehicle cabin served as calibration chamber, where boundary conditions, which are detailed in the caption of Figure 7, were under control. Additionally, to guarantee consistent particle losses, identical conductive tubes were used throughout both calibration and experimental phases.

Equation (2) outlines the formula used to determine the scaling factor for each OPS device, where $\mathrm{“}{SF}_{i, {OPS}_j}\mathrm{”}$ is the scaling factor for a specific particle size ($i$) of a specific OPS device ($j$), $“{mean concentration}_{{i,OPS}_{1-5}}”$ is the mean concentration of the particle size ($i$) across all five OPS devices, and $“{concentration}_{{i,OPS}_j}”$ is the concentrations of particle size ($i$) measured by a specific OPS device ($j$). Given that the calibration experiment was repeated three times, the mean of three repetitions was taken to establish the definitive scaling factor.

$${SF}_{i,{OPS}_j}={\displaystyle \frac{{mean concentration}_{i,{OPS}_{1-5}}}{{concentraion}_{i,{OPS}_j}}}$$

(2)

The sampling locations within the cabin were strategically chosen. To identify the optimal measurement point for the emitter positioned at the front-right seat, five measurement tubes were installed around the manikin (right, left, and front sides of the mouth and right and left sides of the headrest); see Figure 8b,c. Three experiments, each lasting 10 min, with breathing settings were carried out, and the integral value for each breathing cycle was calculated and is illustrated in Figure 8a. The integral value corresponds to the area under a curve representing the fluctuation of total particle numbers over time.

Based on Figure 8a, particle concentrations measured at the mouth area were higher compared to those at the headrest area. The data indicated greater variability and fluctuations in particle numbers at the mouth area, likely due to higher turbulence in this region. In contrast, the particle concentrations at the headrest area were relatively stable, showing fewer fluctuations. Between the two measured points within the headrest area, the right point displayed higher concentrations compared to the left point, primarily due to the lower airflow in the right vent compared to the center register. Therefore, the sampling location of the emitter was determined to be the right side of the headrest, as it reflects the highest concentrations with the highest stability among the measurement points surrounding the manikin, as shown in Figure 9a.

The other measurement locations within the vehicle were situated at the breath levels of the front-left, rear-left, and rear-right seats, as shown in Figure 9. Throughout the tests, continuous monitoring of panel concentrations was performed to validate the correct setup of the respective ventilation settings and ensure the introduction of particle-free air into the cabin.

2.2.3. Particle Source Consistency

Prior to conducting the experiments, the consistency of the emitted particles from the manikin was ensured by subjecting it to 15 min of controlled breathing and speaking sessions within the duct setup, as explained in Figure 4. The total particle numbers generated in each session were compared to those recorded on other test days, as seen in Figure 10.

Table 3. The IQR of total particle number concentrations across various experiment days. The percentages show the relative differences of total particle numbers on the day (i) compared to the mean concentrations over the 6 days of experiments.

Experiment Days	IQR in pcs./cm3	Relative Difference from the Mean Concentrations
Day 1	133.1	−2%
Day 2	134.5	−1%
Day 3	145.4	7%
Day 4	123.8	−9%
Day 5	137.4	1%
Day 6	140.4	3%

presents the IQR of total particle numbers across various experiment days corresponding to the sample graph of Figure 10. It also reveals that the relative differences from the mean concentrations that are calculated according to Equation (3) are less than 10%.

$${\displaystyle \frac{{\mathrm{I}\mathrm{Q}\mathrm{R}}_{day i}-{\mathrm{I}\mathrm{Q}\mathrm{R}}_{mean, day 1-6}}{{\mathrm{I}\mathrm{Q}\mathrm{R}}_{mean, day1-6}}}\times 100$$

(3)

This comparative analysis enabled us to evaluate the consistency of particle emissions, ensuring the reliability and comparability of results across various experimental sessions.

2.3. Data Analysis

A total of 24 experiments (including repetitions) were conducted. The boundary conditions of eight experimental use-cases, including the manikin mode, ventilation mode, blower level, and blower intensity, are reported in

Table 4. The experimental use-cases within the car cabin.

Use-Case	Manikin Mode	Ventilation Mode	Blower Level	Blower Intensity
1	Breathing	Fresh	4	Low
2	Breathing	Fresh	6	High
3	Breathing	Recirculation	3	Low
4	Breathing	Recirculation	5	High
5	Speaking	Fresh	4	Low
6	Speaking	Fresh	6	High
7	Speaking	Recirculation	3	Low
8	Speaking	Recirculation	5	High
Number of repetitions				3
Total number of experiments				24

. Each experiment was repeated three times to ensure the reliability and consistency of the results. The mean concentrations from the three repetitions were calculated and subjected to analysis. After each individual test, a designated interval was allocated to allow for flushing the cabin of particles and the return of concentrations to the levels indicated by the OPS device located at the panel (particle-free air due to activation of the HEPA filter system). This standardized procedure ensured the comparability of the three repetitions and facilitated accurate analysis and comparison of the results.

During the monitoring phase, data were collected from five OPS devices at 1 s intervals for 15 min. The initial four minutes and final one minute of the data were excluded to focus on the steady-state range. Scaling was applied by multiplying the data with the scaling factor, as explained in Equation (2). Subsequently, for each measurement location, the integral value was computed based on the curve representing the variation in total particle numbers (<1 µm) over 10 min (steady-state data). The integral calculation provides a detailed understanding of how particle concentrations changed over time, capturing variations and trends within the data. Finally, the averaged particle concentration was determined by dividing the integral value by the length of the interval and then averaging the results obtained from three repetitions.

3. Results

3.1. Influence of the Source Emission: Breathing and Speaking

Figure 11 shows the concentration of particles at various measurement locations under different use-cases. In the breathing use-case, across all tested ventilation settings, the front-right measurement location consistently displays the highest particle concentrations when compared to other seats. The rear-right measurement location, corresponding to the seat directly behind the emitter location, shows the second-highest particle concentrations. Conversely, measured concentrations at the front-left and rear-left locations are closely matched, indicating that both the left-side sampling locations of the cabin experience similar and comparatively lower levels of exposure to particles emitted during breathing. In this study, the particle concentrations at the front-left seat are 55% lower compared to the rear-right seat (the average of all breathing experiments).

The results of the speaking use-case are depicted in Figure 11. In this use-case, different patterns emerge depending on the ventilation mode. Under fresh mode ventilation (Figure 11a,b), the rear-right location receives the highest concentrations, while in the recirculation mode, the highest concentrations are detected at the front-right location (Figure 11c,d). It is worth noting that for all cases in the speaking mode, where the emitter’s head is turned toward the driver, contrary to expectation, the driver (front-left) experiences significantly lower particle concentrations compared to the seat behind the emitter (rear-right). On average, across all speaking experiments, it is found that particle numbers at the front-left seat are 58% less than those at the rear-right seat.

A comparison of particle concentrations at measurement locations between the breathing and speaking use-cases clearly shows that particle concentrations are lower during the breathing case than during the speaking case. The difference between breathing and speaking is represented by percentage values on the graph, showing reduction in breathing compared to speaking. On average (front-right excluded), a 58% decrease in particle levels is observed at sampling locations during the breathing mode relative to the speaking mode.

In both breathing and speaking use cases, higher particle concentrations are observed when fresh air was activated, and the blower level was low (554 pcs./cm³ in the breathing mode and 1282 pcs./cm³ in the speaking mode). Conversely, the lowest particle concentrations occur in the recirculation mode with a high blower level (256 pcs./cm³ in the breathing mode and 524 pcs./cm³ in the speaking mode). These values represent the averages measured across the front-left, rear-left, and rear-right positions.

3.2. Influence of the HVAC Ventilation Mode and Blower Level

Figure 12 illustrates a comparison of particle concentrations at recipients’ measurement locations. The comparison is made between two use-cases: one with the fresh air mode and another with the recirculation mode. The difference in particle concentrations between fresh and recirculation cases is represented by percentage values on the graphs. Across all cases, a consistent reduction in particles received by recipients is observed in the recirculation case compared to the fresh case, with the HEPA filter active in both ventilation modes. Also, the percentage of reduction indicates that the recirculation mode is more effective at reducing particle concentrations at the left side of the cabin (front-left and rear-left positions), while the rear-right position consistently shows the smallest reduction in particle concentrations.

The effect of the blower level of the HVAC system on the particle numbers at the recipients’ locations is shown in Figure 13. Clearly, fewer particles are observed at the higher blower levels compared to the lower blower levels. This trend is consistent across all use-cases. In the fresh air mode during breathing, the high blower level (blower level 6) decreased particle concentrations by approximately 42–45% compared to the low blower level (blower level 4), with the largest reduction (44.63%) observed at the front-left position. Similarly, during speaking with the fresh air mode, the high blower level achieved particle reductions between 28 and 37%, with the maximum reduction (37.27%) at the front-left location.

In the recirculation mode, increasing the blower level from 3 to 5 reduced particle concentrations by 28–31% during breathing and 42–46% during speaking, with speaking activities showing more substantial reductions.

The obtained results from Figure 12 and Figure 13 suggest that the HVAC ventilation mode and the blower level influence particle numbers that reach the breath level of recipients. Among the experimented use-cases, the recirculation mode with an active HEPA filter and a high blower level (referred to as blower level 5 in this study) shows the least particle numbers at the recipients’ breath levels during both breathing and speaking activities.

4. Discussion

4.1. Influence of the Source Emission: Breathing and Speaking

According to Figure 11, the higher front-right concentrations observed during breathing case align with the placement of the respiratory manikin (emitter) at the front-right seat, logically resulting in increased concentrations at this point. While the elevated rear-right concentrations can be attributed to the direction of airflow from the panel, which transports the emitted particles from the front row to the back row, making the seat behind the emitter the most susceptible to particle exposure. The front-left and rear-left positions showed similar particle concentrations, both experiencing relatively low exposure levels to particles from breathing emissions. Findings are in agreement with results from CFD simulations conducted by Sarhan et al. [3]. In their study involving an infected driver, it was observed that the passenger sitting directly behind the driver is at a higher risk of inhaling more contaminated aerosol particles. Conversely, the passenger seated beside the driver is exposed to fewer particles in comparison to other passengers within the vehicle.

In Figure 11, in the speaking use-case, a discrepancy in the seat with the highest particle concentrations across different ventilation modes was observed. In the recirculation mode, the front-right seat showed the highest concentrations, while in the fresh mode, the highest concentration was observed at the rear-right seat. This discrepancy is likely linked to the positioning of the measurement tube at the front-right (emitter) seat. In the speaking use-case, the manikin’s head was turned 60 degrees towards the driver, while in the breathing use-case, the manikin’s head faced the front of the vehicle, as shown Figure 6. In both use-cases, the concentrations were measured at the headrest-right location, as seen in Figure 9a. With the ventilation setting in the fresh mode, the airflow from the panel guides the emitted particles during speaking towards the back row, passing through the space between the right and left seats. Consequently, most of the particles exit the cabin through the air extractor in the trunk, bypassing the measurement tube at the front right. In contrast, in the recirculation mode, where cabin air is recirculated and the inlet is positioned beneath the glove box, the suction created by recirculation counteracts the airflow from the panel. This phenomenon likely contributes to the higher concentrations measured at the headrest of the emitter in the recirculation case compared to the fresh case. This implies that attempting to compare particle concentrations in different use-cases at the front-right location (emitter) in this study may not lead to accurate comparisons. Consequently, this measurement location is excluded from the graphs in Figure 12 and Figure 13.

In the speaking mode, despite the emitter’s head being oriented toward the driver, the driver is exposed to significantly lower particle concentrations compared to the seat behind the emitter. This observation indicates that the impact of airflow direction from the vents plays a more significant role than the change in the orientation of emitted particles.

A reduction in particle concentrations within the cabin was observed when the emitter was breathing rather than speaking, aligning with respiratory settings of the manikin. This is consistent with the results of the study conducted by Arpino et al. [23]. In their study, the risk of infection within the car cabin was significantly lower when the infected passenger was breathing compared to when speaking. A discrepancy in particle concentration differences between the breathing and speaking use-cases was noticed in this study. Inside the cabin, particle levels during breathing were 58% lower compared to the speaking mode. However, according to Table 2, which reported the concentrations within the duct test section, particle number concentrations during breathing are about 70% lower than those during the speaking case. This variation may be attributed to the position of the measurement tube within the duct test setup. Positioned perpendicular to the particle flow, the tube might not capture all particles passing through the duct. Additionally, a discrepancy between the velocity of air entering the sampling tube and the velocity of the air within the duct may also contribute to this variation.

4.2. Influence of the HVAC Ventilation Mode and Blower Level

In Figure 12, a reduction in particle concentrations at recipients’ seats in the recirculation mode compared to the fresh mode was observed. One plausible explanation for this phenomenon might be the air suction and the changed flow field created by the recirculation door, which is located under the glove box. Specifically, this suction draws in particles emitted from the manikin, reducing the particle count that ultimately reaches the recipient points. On the other hand, in the fresh mode, the front-to-back airflow direction leads to increased particle numbers received by recipients. This finding appears to contrast with the results of the study conducted by Kim et al. [27], which reported a higher probability of COVID-19 infection in the recirculation mode compared to the fresh air mode. However, in their study, the emitter was located in the back seat, whereas the analysis in the current study focuses on a different emitter position. When the emitter is positioned in the back seat, the fresh air mode is a more effective solution. In this configuration, the front-to-back airflow originating from the panel directs particles toward the rear and ultimately into the trunk of the car, resulting in lower particle concentrations at the front seats. In this study, where the emitter is located in the front seat, the flow field generated by recirculation reduces particle concentrations at other seats within the cabin.

The impact of the HVAC blower level on particle concentrations within the cabin was investigated in Figure 13, showing an expected trend. By increasing the blower level, the measured particle concentrations were reduced. Researchers Arpino et al. [23] and Sarhan et al. [3] have conducted CFD simulations to explore the impact of the HVAC flow rate on the risk of infection from SARS-CoV-2. Arpino et al. [23] discovered that reduced ventilation rates elevated the risk of infection, and Sarhan et al. [3] noted an increase in the time required to inhale a specific quantity of particles with an increased air velocity in the HVAC system.

4.3. Limitations and Future Research

It is important to note that this study has some limitations. First, the use of DEHS by the respiratory manikin to generate particles might result in different behaviors compared to particles produced by humans, particularly in terms of deposition and evaporation processes. Moreover, the investigation does not account for the warm and moist exhaled air cloud that typically surrounds exhaled particles in real-world situations. Furthermore, the study does not incorporate realistic body geometry and skin temperature considerations for both the manikin and recipients. The complex human body shape and the buoyancy flows generated around it influence pollutant transport, exposure, and air distribution in confined spaces, aspects that merit attention in future research. Additionally, the experimental scope could benefit from the inclusion of further use-cases, such as examining mixed ventilation modes (windshield defrosting inlet and front inlets), varying the emitter’s location, and conducting road tests under driving conditions.

5. Conclusions

A manikin-based method was introduced in this study to capture the spatial distribution of aerosol particles within a car cabin. The focus was on the infection pathway via sub-micrometer particles (<1 µm). The infection route via respiratory droplets was not in the scope of this study. Particle source stability, crucial for measurement accuracy and repeatability, was ensured through pretests. The methodology was investigated with different conditions, such as changes in ventilation settings (the ventilation mode and blower level) and emission use-cases (breathing and speaking). Throughout the experiments, the respiratory manikin representing a particle-emitting passenger during both breathing and speaking use-cases was placed at the front-right seat, and the airflow from panel registers was consistent for all the experiments. The main findings of this study are as follows:

• The introduced manikin-based method could capture variations in particle concentrations at different locations and due to variations in the ventilation mode, blower level, and emission source.
• Identifying the optimal measurement location around the emitter to accurately capture the highest concentrations with a high degree of reliability is a challenge. Key factors such as the airflow pattern (e.g., direction, volume flow rate, uniform/non-uniform spreading) and the direction of the emitter’s head (particle source) play important roles in this process.
• Within a car cabin, the seat directly behind the emitter has the highest concentrations of encountered respiratory particles. The average of all experimental use-cases indicated that the front-left and rear-left seats demonstrate 56% and 49% lower particle concentrations, respectively, compared to the rear-right seat.
• The passenger seat beside the emitter has the lowest particle concentrations. On average, concentrations in this seat are lower than the rear-right seat by 56% and rear-left seat by 14%.
• Increasing the airflow from blower level 3 to 5 in the recirculation mode or from blower level 4 to 6 in the fresh mode reduces particle levels within the cabin, with measurements indicating an average reduction of 38% across all experimental conditions.
• In this study, under comparable air volume flow conditions and similar exhalation activities, the recirculation mode with an efficient particle filter was more effective in reducing particle concentrations compared to the fresh air mode. On average, it achieved a 33% reduction in particle concentrations across all experimental conditions.

This study contributes to a better understanding of the dispersion of exhaled particles within vehicle cabins. The results may serve as essential inputs for future engineering investigations, especially in the design of public transport and passenger cars, to reduce the transmission of aerosols.