The aerosol deposition efficiencies at the demarcated sections of the lungs under high-frequency breathing conditions are quantified to resolve the puzzle of the mechanistic link between human exposure to aerosol during exercise for planar and mutually orthogonal lung geometries. These are valuable results for developing dosimetry algorithms for future risk assessments involving PM-PAHs.
3.1. Airflow characteristics
Fluid flow in the G3–G6 is described using velocity distribution, recirculation and stagnation zones, and static pressure distribution on the walls (Mutuku et al., 2020a). The cross-sectional area at the inlet for G3 is 2.44×10− 5 m2 while that of the outlets at G6 are 4.88×10− 5 m2. The increasing effective cross-sectional area causes a decrease in velocity, as depicted in the 2-D velocity contours for 4 METs in Fig. 4a and 8 METs in Fig. 4b. Expectedly, the air velocity is generally higher in the airways for inhalation at 8 MTEs compared to 4 METs. Since the same boundary conditions are imposed at the inlets of the geometries, G3 and G4 of the two types of geometries have similar airflow patterns. The highest velocity regions are seen in the central airways, as the change in flow direction forms regions of low velocities at the peripheral walls of G4, according to Poiseuille's Law. This causes expansion, which, in turn, induces zones of low flow velocity and stagnation at the distal bifurcations of G5 and G6 (Kim et al., 2019). From Figs. 4a and b, the stagnation zones at the outer walls of each bifurcation decrease as the velocity at the inlet increases for breathing at 4 METs and 8 METs. Boundary layer separation along the curvatures causes recirculation and stagnation phenomena at the vicinities of the carina regions (Zhang and Kleinstreuer, 2002). Consequently, airflow appears more uniform in the planar geometry than in the mutually orthogonal one.
The average Reynolds number of flows for vigorous-intensity activity is 1946, implying laminar flow. However, the Reynolds number in a section of the sin-wave breathing curve exceeds 2100, as seen in Fig. 2a. Specifically, 16 out of 30 time-steps fit in the turbulent flow regime characterized by rapid changes in both the direction and magnitude of the two-phase flow. It is more chaotic than laminar flow and, therefore, accompanied by higher mixing levels, which leads to a higher momentum transfer between the gas and dispersed phases (Hvelplund et al., 2020).
The distribution of static pressures at bifurcation 4 (B4) and the area immediately preceding this section differed for the two geometries, as seen in Fig. 5a for 4 METs and Fig. 5b for 8 METs, due to the pressure variation caused by different orientations of the geometries. The magnitude of the negative pressure gradients is higher for inhalation at 8 METs than at 4 METs. Static pressure drops are usually caused by changes in kinetic energy and viscous dissipation (Zhang and Kleinstreuer, 2002). Since there is zero-gauge pressure at all outlets of the computational domains, pressure is calculated based on the static pressure at any location and that of the outlets. The first generations in both geometries have low static pressures due to the high velocity of flow regimes imposed at the inlets. Inner bifurcations have larger zones of higher static pressure compared to distal bifurcations. Consequently, higher static pressure zones are seen in the inner bifurcations of G5 in the planar geometry compared to the outer bifurcations, as seen in Fig. 5b. In comparison, zones of high static pressures in the mutually orthogonal bifurcation are uniform across all four branches of G5 as seen in Fig. 5b. The branching in the mutually orthogonal geometry introduces localized pressure variations and regions of higher pressures compared to the straight, simpler channel representation of the planar model.
3.2. Aerosols deposition
The deposition patterns and efficiencies displayed in Fig. 6 indicate an increase in DEs in the two geometries as the aerosol diameter increases for exercise at 4 METs. For breathing in an industrial area presented in Fig. 6a, DEs of 5.6 µm particles in the mutually orthogonal bifurcation exceeded that of the planar bifurcation by 4.3% with a relative difference of 8.2%. However, For 1 and 5.6 µm particles, the DEs in the two geometries are equal for moderate exercise in urban and rural areas, as seen in Fig. 6b and Fig. 6c. Overall, deposition is localized at the carina regions for both geometries, as seen in Fig. 6. It is noteworthy that while in the planar geometry, all bifurcations are hotspots for aerosol deposition, in the mutually orthogonal geometry, only B3 and B4 are the dominant hotspots. During moderate-intensity activity, the DEs of the 3.2 µm differed the most in the two geometries, as seen in Fig. 6, where the relative differences in the two geometries for the industrial, urban, and rural areas are 83.3%, 80.5%, and 84.4%, respectively, as summarized in Table 3. The variations in the DEs between the two geometries at a PM-PAHs concentration of 0.95 µg‧m-3 in a rural site are relatively low, as seen in Table 3.
Table 3
The mass (µg) of the deposited PM-PAHs, and their relative differences in DEs (%) at the two activity intensities and for the planar and mutually orthogonal bifurcations.
|
PM diameter (µm)
|
Mass of deposited PM-PAHs (µg)
|
Relative DE (%)
|
|
Intensity of activity
|
|
4 METs
|
8 METs
|
4 METs
|
8 METs
|
|
Planar
|
Orthogonal
|
Planar
|
Orthogonal
|
-
|
-
|
Industry
|
1.0
|
4.19×10− 6
|
6.81×10− 6
|
2.46×10− 4
|
7.46×10− 4
|
62.5
|
203.8
|
3.2
|
6.18×10− 4
|
1.03×10− 4
|
0.006
|
0.009
|
83.3
|
50.7
|
5.6
|
0.017
|
0.018
|
0.028
|
0.033
|
8.2
|
17.8
|
Total
|
|
0.018
|
0.018
|
0.034
|
0.042
|
-
|
-
|
Urban
|
1.0
|
9.95×10− 6
|
1.47×10− 5
|
2.23×10− 4
|
2.2×10− 4
|
47.4
|
1.2
|
3.2
|
6.18×10− 4
|
1.20×10− 4
|
0.006
|
6.32×10− 3
|
80.5
|
7.0
|
5.6
|
0.012
|
0.012
|
0.021
|
0.022
|
0
|
2.6
|
Total
|
0.012
|
0.012
|
0.027
|
0.028
|
-
|
-
|
Rural
|
1.0
|
1.10×10− 5
|
9.95×10− 6
|
2.06×10− 4
|
2.35×10− 4
|
9.5
|
13.7
|
3.2
|
7.72×10− 4
|
1.20×10− 4
|
0.004
|
3.48×10− 3
|
84.4
|
10.2
|
5.6
|
0.012
|
0.012
|
0.028
|
0.030
|
0
|
9.6
|
Total
|
0.012
|
0.012
|
0.032
|
0.033
|
-
|
-
|
Note: DE represents deposition efficiency
|
For vigorous-intensity activity, 1 µm and 3.2 µm particles depict lower DEs of 1.4–1.7% and 7.6–8.7%, respectively compared to the 5.6 µm aerosol whose range for the DEs in both geometries is 40.3–44.1% as seen in Fig. 7. This phenomenon can be attributed to turbulent dispersion at G4 causing the smaller and lighter aerosols to disperse throughout the computational domain. The relative differences between the DEs depicted in Fig. 7a of the microparticles in the planar and orthogonal geometries in the industrial area are 12.3%, 2.7%, and 4.6% for 1 µm, 3.2 µm, and, 5.6 µm aerosols, respectively. The relative differences between the DEs depicted in Fig. 7b of the microparticles in the planar and orthogonal geometries in the urban environment are 1.2%, 6.9%, and 2.8% for 1 µm, 3.2 µm, and, 5.6 µm aerosols, respectively. The relative differences between the DEs depicted in Fig. 7c of the microparticles in the planar and orthogonal geometries in the rural area are 13.7%, 10.2%, and 9.6% for 1 µm, 3.2 µm, and, 5.6 µm aerosols, respectively. Generally, there is an overestimation in the DEs obtained using the mutually orthogonal geometry for aerosols of all diameters. In incidences of turbulent flow, deposition occurs due to flow fluctuations (Darquenne, 2012). Irregular fluctuations continuously alter the particles' trajectories in terms of magnitude and direction until they deposit on the airway walls.
Previous literature shows that DEs for aerosols' are proportional to their Stokes numbers (Cheng et al., 1999; Mutuku et al., 2020a). Due to improved resolution and retrievability of data, results from this study indicate DEs are also affected by geometric factors, herein represented by bifurcation and rotator subroutines (Zhang and Kleinstreuer, 2001). Smaller-sized aerosols, including 1 µm and 3.2 µm positioned next to the airway walls, deposit easily, especially in the presence of complex secondary flow phenomena, as seen in Figs. 7a-c. This tendency is attributed to the particle's distribution at the inlet of the geometry and the interaction between the gaseous and dispersed phase, which improve their chances of touching the airway walls. Due to impaction, PM at the middle of the inlet cross-sectional area deposit at the carina region. This phenomenon is evident for the 5.6 µm aerosols, as seen in Figs. 7a-c. At higher Reynolds numbers of flows, expansion of the recirculation and stagnation zones pushes the particles toward the center of the carina region. In a study comparing the deposition patterns between nano and microparticles, the former deposited more uniformly in the computational volume under investigation (Zhang et al., 2005). This agreed well with the depositions of the size-differentiated particles, where the 1 µm and 3.2 µm particles deposited more uniformly in G3–G6 as compared to the 5.6 µm. Herein are deposition patterns, as seen in Figs. 6 and 7, particles where larger particles deposited in anterior sections of the computational domain, unlike in an investigation by Augusto et al. (2016) where smaller particles deposited in the first bifurcation while larger particles deposited in the third bifurcation. In a research by Longest and Holbrook (2012) on a human-specific airway tree model, the most important hotspot for particle flow is the first carina region characterized by a stagnation zone. Similar findings are obtained here where the carina region at the end of G3 is the most crucial hotspot, especially for the 5.6 µm particles.
The differences in DEs for the two geometries based on the Weibel model can be explained by the different extents in the x-, y-, and z-coordinates presented in Table 2. Importantly, the distance on the y-coordinate in the planar geometry exceeds that of the mutually orthogonal one by about 0.02 m. Additionally, the total surface area of the wall for the planar geometry is higher than that of the mutually orthogonal geometry. Consequently, it has a bigger area for the deposition of particles. Overall, the DEs of the aerosol in the mutually orthogonal bifurcation exceed that of the planar bifurcation, implying that the orientation of the bifurcation is a more significant parameter for PM deposition compared to the wall surface areas of the computational domains indicated in Table 2. Due to the considerable differences in the DEs of different size-resolved particulate PAHs in the two geometries under investigation, as seen in Table 3 and Table 4, a physiologically correct benchmark model is needed to establish the validity of the results.
Table 4
The total surface area (mm2) of the deposited PM-PAHs at the two different activity intensities and for the planar and mutually orthogonal bifurcations.
|
PM diameter (µm)
|
Surface area of deposited PM-PAHs (mm2)
|
|
Intensity of activity
|
|
4 METs
|
8 METs
|
|
Planar
|
Orthogonal
|
Planar
|
Orthogonal
|
Industry
|
1.0
|
2.5×10− 5
|
4.1×10− 5
|
1.5×10− 3
|
4.5×10− 3
|
3.2
|
1.2×10− 3
|
1.9×10− 4
|
1.1×10− 2
|
1.7×10− 2
|
5.6
|
1.8×10− 2
|
2.0×10− 2
|
3.0×10− 2
|
3.5×10− 2
|
Total surface area
|
1.9×10− 2
|
2.0×10− 2
|
4.3×10− 2
|
5.6×10− 2
|
Urban
|
1.0
|
6.0×10− 5
|
8.8×10− 5
|
1.3×10− 3
|
1.4×10− 3
|
3.2
|
1.2×10− 3
|
2.3×10− 4
|
1.1×10− 2
|
1.2×10− 2
|
5.6
|
1.3×10− 2
|
1.3×10− 2
|
2.3×10− 2
|
2.3×10− 2
|
Total surface area
|
1.4×10− 2
|
1.3×10− 2
|
3.5×10− 2
|
3.6×10− 2
|
Rural
|
1.0
|
6.6×10− 5
|
6.0×10− 5
|
1.2×10− 3
|
1.4×10− 3
|
3.2
|
1.4×10− 3
|
2.3×10− 4
|
7.3×10− 3
|
6.5×10− 3
|
5.6
|
1.3×10− 2
|
1.3×10− 2
|
3.0×10− 2
|
3.3×10− 2
|
Total surface area
|
1.4×10− 2
|
1.3×10− 2
|
3.8×10− 2
|
4.1×10− 2
|
Table 5
BaP-based toxicity equivalence factors for the PM-bound PAHs deposited in a 2s inhalation event.
|
|
Intensity of exercise
|
Location
|
TEFBaP at 4 METs
|
TEFBaP at 8 METs
|
Planar geometry
|
Industry
|
1.37×10− 10
|
2.31×10− 10
|
Urban
|
7.38×10− 11
|
1.58×10− 10
|
Rural
|
1.34×10− 9
|
3.20×10− 9
|
Mutually orthogonal geometry
|
Industry
|
1.48×10− 10
|
2.74×10− 10
|
Urban
|
7.39×10− 11
|
1.62×10− 10
|
Rural
|
1.34×10− 9
|
3.51×10− 9
|
3.3. PAH dose from the deposited aerosols
Establishing the dose of PAHs after inhalation is essential to characterize health risks accurately following the inhalation of PM-PAHs. Exposure to chemical compounds can be quantified in several ways, including frequency, duration, and magnitude. Herein, PAH bound on the surfaces of PM deposit on the epithelial exchange boundary layer after a single inhalation cycle of 2.1 s for moderate exercise and 1.2 s for vigorous intensity exercise. Herein, the frequency of exposure is one. Estimates of the exposure concentrations are usually applied to calculate the intake of the chemicals (Moorthy et al., 2015). The exchange surface area facilitates the transportation of chemicals from PM to the epithelial layer upon contact. Therefore, the deposited aerosols' surface areas (SAs) are important. The dose for aerosols with surface adsorbed PAHs depends on size because smaller particles have a higher specific surface area and therefore may contain more organic carbon, allowing more PAHS to adsorb (Sheu et al., 1997). Since the SAs indicate the doses of PAHs inhaled, cumulative surface areas for the deposited PM for the respective inhalation cycles are calculated and summarized in Table 4.
The 5.6 µm particles dominated the surface area of PM-adsorbed PAH deposited for inhalation at 4 METs presented in Table 4. The dominance was more exaggerated for inhalation at 4 METs, where they represented 94%, 92%, and 89% in the planar geometry and 99%, 98%, and 98% for the mutually orthogonal geometry, industry, urban, and rural areas, respectively. Their surface area at 8 METs was 70%, 65%, and 78% for the planar geometry and 63%, 64%, and 80% for the mutually orthogonal. At 4 MET, the larger particle aerosols had significantly higher DEs in the planar geometry than the mutually orthogonal geometry, as seen in Fig. 8a. This trend was inconsistent for exercise at 8 METs where the DEs differed insignificantly between the two geometries, as seen in Fig. 8b. For aerosols with diameters less than 6.7 µm, deposition in the orthogonal geometry exceeded that in the planar geometry.
During vigorous-intensity activity, the SA of the deposited particulate PAHs is the lowest for planar geometry in a rural area where it is\(3.8\times {10}^{-8}\)m2 while the maximum at\(8.7\times {10}^{-8}\)m2 in the mutually orthogonal bifurcations in an industrial area. For Moderate-intensity activity, the minimum and maximum are\(1.3\times {10}^{-8}\)m2 and\(2.0\times {10}^{-8}\)m2 for the mutually orthogonal bifurcations in rural and industrial areas, respectively. For inhalation at rest, the minimum SA of deposition is for planar geometry in an industrial area at\(3.0\times {10}^{-8}\)m2 while the maximum.\(6.0\times {10}^{-8}\)m2 for the mutually orthogonal bifurcations in a rural area. The standard deviations in SA of deposited PM three inhalation statuses are 1.9×10− 8, 3.2×10− 9, and 1.4×10− 9 for inhalation vigorous intensity, moderate intensity, and rest. This implies that the least variance from the mean is observed for rest inhalation status, while the highest variance is observed for vigorous-intensity activity. Maximum exposure is attained using mutually orthogonal geometry during vigorous-intensity activity in an industrial area, as seen in Table 3. In comparing toxic microparticles and nanoparticles, the latter had a higher surface area, reactivity, and levels of reactive oxygen species (Buzea et al., 2007).
Table 3 reveals the highest mass-based dose of deposited PAHs occurred in the industrial inhalation scenario, indicating a substantial deposition burden in potentially exposed populations. Notably, vigorous-intensity activity (8 METs) further amplified this effect, with peak depositions reaching 0.03 µg and 0.04 µg for planar and orthogonal geometries, respectively. This exceeds the average public exposure to PAHs reported by Moorthy et al. (2015) of 40–50 ng⋅kg− 1⋅day− 1. Interestingly, minimal differences were observed at moderate-intensity (4 METs). These findings suggest differential deposition patterns emerge at higher inhalation rates, potentially due to altered airflow dynamics within the complex bronchial geometries. The significant discrepancy between planar and orthogonal configurations at 8 METs highlights the importance of using anatomically realistic models for accurate assessment of PAH deposition during strenuous activity.
In Table. 5, the BaP-based toxicity equivalence factors (TEFBaP) for particulate matter-bound Polycyclic Aromatic Hydrocarbons (PAHs) deposited in the aftermath of a 2-second inhalation event reveal an elevation in exposure levels within rural locales, surpassing those observed in other regions despite the comparatively higher concentrations of PAHs in these areas. Specifically, in the context of a 4 METs exercise, the exposure magnitude in rural environments surpassed that of both urban and industrial counterparts by factors of 18 and 9, respectively, regardless of the geometric orientations. Furthermore, during an 8 METs exercise, the preeminence of exposure persisted in rural settings. In the case of mutually orthogonal geometry, the BaP-TEFs for deposited PM-bound PAHs in rural areas were 21 and 12 times higher than those in industrial and urban areas, respectively. Similarly, for planar geometry, the corresponding factors were 20 and 13, respectively. These findings emphasize the surprisingly elevated risk associated with exposure to PM-bound PAHs in rural settings, despite lower overall PAH concentrations. This underscores the indispensable importance of considering both mass concentrations and congener profiles in assessing the health risk posed by inhaled PM-bound PAHs. These results highlight the critical need for tailored interventions in rural areas to mitigate potential health risk.
It's crucial to consider that mass alone may not fully capture the potential health impacts of deposited aerosols. As demonstrated by (Kermani et al., 2021), the surface area of particles directly correlates with the influx of inflammatory immune cells in the lungs, indicating potential injury. Therefore, in cases like vigorous inhalation with increased deposition, surface area may be a more accurate determinant of dose and subsequent health effects compared to simply evaluating the deposited mass. The internal surface area of the lung is about 130 m2: this study's computational domain encompasses a smaller subset of 2.38×10− 2 m2 for the planar geometry and 2.27×10− 2 m2 for the mutually orthogonal geometry. This limited scale represents a key constrain. Furthermore, total concentrations of PM-PAHs overestimate the bioavailable PAHs concentration (Sánchez-Piñero et al., 2022), and therefore, information on the bioavailability of inhaled PAHs in Pulmonary, Oral, GI, and hepatic systems is helpful for a comprehensive understanding of inhaled PAH exposure and potential health risks.
3.4 The influence of particle size and flow velocity on the average penetration depth of micropollutants within the planar and orthogonal symmetrical bifurcations.
To understand the infiltration of deposited aerosols, the change of DEs of aerosols as the flow advances downstream (x-direction) for moderate and vigorous intensity activities are plotted in Figs. 9 and 10. During moderate and vigorous intensity exercises, deposition for the 5.6 µm is predominant in Figs. 9 and 10, where the carina region at the end of G3, which is 15 mm from the inlet, is the most significant hotspot. Deposition in this region is caused by inertial impaction due to high impaction forces and local turbulent vortices (Lambert et al., 2011). Unlike for vigorous intensity exercise, nearly all deposition of the 5.6 µm aerosol during moderate intensity exercise occurs at Carina 3 as seen in Figs. 9 and 10. For both geometries, the DEs for the 5.6 µm aerosol in C3 are almost equal due to similarities in airflow properties from the inlet of G3 to the end of G4 in the geometries.
Overall, a more significant proportion of the 1 µm and 3.2 µm aerosols proceed to the lower sections of the lungs for breathing at 4 METs than for 8 METs (Kim et al., 2019). For infiltration during moderate-intensity exercise, presented in Fig. 8, the 5.6 µm aerosols infiltrated the planar geometry until regions surrounding Carina 5, unlike in the mutually orthogonal geometry, where they infiltrated up to Carina 4. However, even though a great proportion of the 5.6 µm aerosol deposit on the carina region at the end of G3, a slight proportion also deposits in other sections of the geometry during vigorous-intensity exercise, as seen in Fig. 10. Similar results were also obtained in previous investigations where turbulent deposition was the dominant mechanism of deposition in G0–G5 (Kolanjiyil and Kleinstreuer, 2017). In another study by Longest and Vinchurkar (2007), turbulent mixing is shown to affect particle deposition in G3–G5. It is evident from the deposition patterns, especially for vigorous-intensity activity in Fig. 9, that the most critical surfaces for particle impaction in a human airway tree during inhalation are, carinas and the curved sections at the bifurcations. Different infiltration rates are caused by complicated flow fields upstream characterized by highly unsteady and non-linear flow fields of the third bifurcation, especially in the mutually orthogonal geometry.
Spherical shapes are adopted for the aerosol; therefore, their Stokes numbers are proportional to diameters. As shown in Tables 3 and 4, DEs are proportional to particle diameters and intensity of activities. The stokes number relates the PM-PAHs's stopping distance to the carina regions' size (Bui et al., 2020; Mutuku et al., 2020a). As such, aerosols with surface-adsorbed PAHs with higher Stokes numbers easily depart from the flow streamlines at obstructions such as carina regions and curved surfaces. In contrast, those with low Stokes numbers follow the streamlines of airflow. From Figs. 6, 7, 9, and 10, it is evident that distal carinas have lower DEs than the proximal ones due to a reduction in flow velocity as airflow progresses in the control volume.
DEs are applied to evaluate the risk of exposure for the respective diameters of aerosols with surface-adsorbed PAHs. Several factors influence particle deposition efficiencies and patterns, including the lung's geometry, breathing status, and physical and chemical properties of the PM (Choi et al., 2007; Zhang et al., 2005). The airflow phenomenon is critical to particle deposition patterns at the carina regions and along the curvatures (Chen et al., 2018b; Wessel and Righi, 1988). Several PM deposition mechanisms are investigated, including inertial impaction, turbulent mixing, and deposition caused by both complex secondary flows and centrifugal forces (Stapleton et al., 2000; Zhang and Kleinstreuer, 2011). The dominance of these mechanisms is determined by location in the lung's geometry, PM concentration, breathing intensity, and particle distribution at the inlet. Deposition by gravitational sedimentation is usually more pronounced beyond generation 9 of the Weibel geometry, and since G3–G6 are investigated here, this mechanism is ignored (Hofmann et al., 1995; Tsuda et al., 2011). In addition, diffusional forces exert more dominance in alveolar regions and, therefore, are neglected in this study.