A Study on the Deterioration of Atmospheric Conditions in Road Areas Based on the Equal-Pollution Model and Fluid Dynamics Simulations

Abstract

This study investigates the impact of roadside building development and vehicle exhaust emissions on atmospheric deterioration in urban highway areas. By integrating satellite-based building coverage data with an equal-pollution vehicle conversion method (based on human toxicity potential), we establish a computational fluid dynamics framework to simulate pollutant dispersion. Key results reveal the following: (1) Street canyon morphology, particularly its geometric symmetry, dominates diffusion patterns. Wide canyons (aspect ratio = 3.3) reduce CO accumulation by over 30% compared to deep canyons (aspect ratio = 0.3), highlighting the role of built form in regulating pollution distribution. (2) Under idealized conditions, photocatalytic pavement mitigates pollutant concentrations at human breathing height by 28.7–56.7%, demonstrating the potential of uniformly applied material solutions. These findings provide a validated theoretical basis for optimizing urban road design and evaluating environmental policies, with considerations for spatial layout and material treatment.

1. Introduction

The dense road network has played a crucial role in enhancing transportation service levels, improving economic efficiency, and elevating residents’ travel quality. However, the motor vehicle exhaust and traffic noise generated during operation have caused irreversible damage to the ecological environment of road areas. Motor vehicle exhaust emissions represent a major source of pollution with significant impact on road operations, primarily consisting of harmful gases such as hydrocarbons (HCs), carbon monoxide (CO), and nitrogen oxides (NO_x). These gases pose a substantial threat to human respiratory and immune systems and contribute to ecological imbalances in the surrounding atmospheric and soil environments. Motor vehicle exhaust contains more than 200 chemical compounds. Among them, NO_x can impair respiratory function and contribute to bronchitis, emphysema, and other pulmonary diseases, while also exacerbating cardiovascular conditions. CO readily induces hypoxia, dizziness, and fatigue in humans. Hydrocarbons, classified as carcinogens by most health agencies, can trigger declines in lung function, asthma, and other related ailments.

The emphasis on road area atmospheric pollution originated from environmental assessment systems, and conducting environmental evaluations for engineering projects has become a legal requirement in most countries [1,2]. The United States enacted the National Environmental Policy Act (NEPA) in 1969 [3], which introduced the mandatory requirement for environmental assessments: “All federal agencies must provide a detailed description of the environmental impact of any proposed federal action that significantly affects the quality of the human environment”. This legislation sparked a surge of attention toward environmental issues.

“Environmental protection” is one of China’s fundamental national policies, and the principle of “prevention first, combined prevention and control” is enshrined in the country’s basic environmental and resource protection laws. To implement this principle, Article 6 of the Environmental Protection Law of the People’s Republic of China (Trial Implementation) [4], promulgated in 1979, stipulates: “All units should pay attention to preventing pollution and destruction of the environment during site selection, design, construction, and production”. However, for a long period, environmental assessments for newly constructed and expanded road projects were largely overlooked. It was not until 1990 and 1996, when the transportation sector introduced the Environmental Protection Management Measures for Transportation Construction Projects [5] and the Technical Specifications for Environmental Impact Assessment of Highway Construction Projects [6], that legal and technical foundations for environmental evaluations in road construction projects were established. With the ongoing advancement of the “dual carbon” system, road industry researchers and relevant authorities are increasingly focusing on the environmental issues arising from road construction.

Regarding the atmospheric environment evaluation in road areas, the Environmental Impact Assessment Specification for Highway Construction Projects (JTG B03-2006) [6] requires the assessment of the current atmospheric conditions along the road, including baseline monitoring, evaluation, and environmental air quality forecasting prior to construction. For the environmental impacts caused by motor vehicle exhaust emissions during the operational phase of highways, it mandates the use of model prediction methods or analogy analysis to estimate the diffusion concentrations. It also provides analogy prediction Formulas (1) and (2), emission source strength calculation methods (3), and model prediction methods.

$$C_{\mathrm{PR}}=C_{\mathrm{mR}}\cdot {\displaystyle \frac{Q_{\mathrm{p}}{U}_{\mathrm{m}}\sin {\theta }_{\mathrm{m}}}{Q_{\mathrm{m}}{U}_{\mathrm{p}}\sin {\theta }_{\mathrm{p}}}}$$

(1)

$$C_{\mathrm{p}}=C_{\mathrm{PR}}+C_{\mathrm{PO}}$$

(2)

$$Q_{\mathrm{j}}={\displaystyle \sum _{i=1}^{4}3}{600}^{-1}{A}_{\mathrm{i}}{E}_{\mathrm{ij}}$$

(3)

Estimating the motor vehicle exhaust emissions under various types and operating conditions is a crucial component in evaluating the atmospheric environmental impact along roadways. Numerous factors influence motor vehicle exhaust emissions, including traffic volume, vehicle type distribution, operational conditions, and the energy forms used by vehicles. In road atmospheric environmental impact assessment models, motor vehicle emissions are typically represented by emission factors [7]. Emission factors for most vehicle types are derived from laboratory vehicle emission tests [8], while traffic emissions are statistically calculated based on the mileage and operating conditions of different vehicle categories.

In addition, numerous studies have been conducted by scholars and institutions worldwide on the evaluation of atmospheric environmental impacts along roadways. In particular, significant progress has been made in the development of dispersion models, which have become increasingly refined and better aligned with real-world conditions. Factors such as vehicle type, fuel type, pollutant species, and emission patterns have been progressively incorporated into these models [9]. According to literature statistics, five models—MOBILE, COPERT, HBEFA, EMFAC, and ARTEMIS—are most frequently applied in current research, as illustrated in Figure 1 [8].

During the 1970s, researchers developed a variety of road exhaust dispersion models based on Gaussian theory [10], capable of predicting exhaust diffusion patterns using meteorological parameters, road geometry, and fluid diffusion principles. However, these models exhibited significant discrepancies between simulated and observed field data due to the simplification of vehicle-induced wake turbulence [11,12]. Subsequently, research institutions conducted controlled field experiments and tracer gas diffusion studies to analyze the influence of vehicle wake turbulence on exhaust dispersion, leading to the refinement and development of more advanced models such as HIWAY 2, CALINE 3 and 4, and GM. These updated models improved prediction accuracy under varying traffic, meteorological, and atmospheric conditions. Moreover, to address the limitations of earlier models—such as the inability to account for wind inclination and infinite roadway lengths—further iterations were introduced in models like GFLSM and CALINE 4.

The analysis methods for studying the diffusion patterns of vehicle pollutants primarily include field measurements, wind tunnel experiments, and numerical simulations [13]. To date, in China, the calculation of motor vehicle exhaust diffusion in highway environmental impact assessments is typically based on the provisions outlined in the appendix of the Environmental Impact Assessment Specification for Highway Construction Projects [6].

Among these, studies using fluid dynamics to analyze motor vehicle exhaust pollutant diffusion often focus on relatively enclosed spaces such as shafts and tunnels, or on the influence of specific facilities and their parameters on pollutant diffusion. In underground tunnels and shafts, where ventilation conditions are poor, the diffusion of motor vehicle exhaust pollutants typically requires focused attention. Wang et al. [14] and Nie et al. [15], respectively, conducted fluid dynamics modeling and calculations on pollutant diffusion patterns for special vehicle types operating in confined spaces. Researchers in related fields have used fluid dynamics simulation techniques to examine the effects of factors such as rooftop greening systems [16], building height, facade details [17], lateral entrainment settings [18], barrier height [19], and road landscaping parameters [20,21] on motor vehicle exhaust diffusion patterns. At urban road intersections, where vehicle speeds are low and frequent starts and stops occur, pollutant concentrations are generally higher. Hassan et al. [22] and Sun et al. [23] also explored these conditions using various models and found that ANSYS Fluent 2023R1 simulation results closely matched field measurements.

In summary, current research on motor vehicle exhaust diffusion patterns and atmospheric environment prediction along roadways primarily focuses on three areas: (1) conducting environmental impact assessments (EIA) for road construction projects in accordance with relevant laws and regulations; (2) improving and refining Gaussian-based exhaust pollutant diffusion calculation models by government departments and researchers; and (3) using fluid dynamics simulations for modeling and analyzing the diffusion patterns of exhaust pollutants in road area environments. However, the improvement of EIA reports and calculation models is largely policy-driven, with exhaust pollutant diffusion predictions being operational but lacking precision. Studies based on fluid dynamics simulations tend to focus on individual vehicles or street-scale models, often failing to integrate the interactive effects between these two dimensions.

Therefore, this study explores three aspects: the frequency of human activities within road areas, motor vehicle emission intensity, and fluid dynamics simulations of motor vehicle exhaust diffusion in road areas. The study analyzes the impact of motor vehicle exhaust emissions on human activity areas, establishes an equal-pollution motor vehicle exhaust emission intensity model, and develops a fluid dynamics model for atmospheric environment deterioration in road areas. The diffusion patterns of motor vehicle exhaust are analyzed from both the vehicle and street dimensions. The technical approach is illustrated in Figure 2.

This study seeks to address this gap by examining atmospheric deterioration in road environments through an integrated framework that accounts for the adverse health implications of exhaust constituents, the multiscale dynamics of pollutant dispersion, and the functional efficacy of photocatalytic pavement materials. The primary contributions are threefold: (1) proposing an equivalent-pollution vehicle conversion method based on human toxicity potential; (2) analyzing exhaust diffusion patterns from both individual vehicles and street canyons using CFD; and (3) evaluating the effectiveness of photocatalytic pavement for exhaust mitigation”.

2. Materials and Methods

2.1. Building Coverage Rate in Highway Areas near Urban Regions Photocatalytic Material

2.1.1. Acquisition of Satellite Imagery

With the continuous expansion of urban areas in recent years, some of the early constructed ring highways and airport expressways have gradually been surrounded by urban built-up areas. This has led to not only issues of exhaust pollution affecting the health of residents in surrounding urban roads but also the diffusion of motor vehicle exhaust emissions from highways to nearby residential and commercial areas, as well as other densely populated locations. According to the Environmental Impact Assessment Specification for Highway Construction Projects (JTG B03) [6], it is required to assess the atmospheric environmental impact within a 200–300 m range on both sides of the highway centerline when necessary. To quantify the changes in the range of residential clusters on both sides of the road, this section takes the ring expressways of six cities—Beijing, Shanghai, Xi’an, Nanjing, Wuhan, and Jinan—as examples, and satellite imagery from the past decade was acquired.

Using ArcGIS software (version 10.7), geographic geometry files were created based on the WGS84 latitude and longitude projection coordinate system, encompassing a 200 m range on both sides of the highway centerline for the six cities—Beijing, Shanghai, Xi’an, Nanjing, Wuhan, and Jinan—within their near-urban areas. These files were then imported into the “Shuijing Micro Map” V4.1 software, which provides access to open-source historical satellite imagery for identifying the building coverage areas along the road.

As the clarity of satellite images is closely related to factors such as cloud cover, atmospheric conditions, and vegetation at the time of capture, in order to enhance the accuracy of building area identification, satellite images from clear dates in the years 2010, 2012, 2014, 2016, 2018, and 2020 were manually selected for use. These images were employed as the source for identifying building areas. The selection of road segments for each city is shown in

Table 1. Information on Selected Road Segments for Satellite Imagery Acquisition.

City	Selected Road Segment	Opening Year	Total Length (km)
Beijing	G4501	2009	187.66
Shanghai	G1503	2009	189
Xi’an	G3002	2003	80.35
Nanjing	G2503	2012	166
Wuhan	G4201	2004	104
Jinan	G2001	2002	100

, with the scope of the selected areas depicted in Figure 3.

The levels of satellite imagery range from Level 1 to Level 21, with corresponding scales and spatial resolutions as shown in

Table 2. Satellite Imagery Levels and Specifications.

Level	Scale	Spatial Resolution	File Size (Beijing)
Level 1	1:591.66 million	156.54 km	192 K
Level 2	1:295.83 million	78.27 km	192 K
Level 3	1:147.91 million	39.14 km	192 K
Level 4	1:73.96 million	19.57 km	192 K
Level 5	1:36.98 million	9.78 km	192 K
Level 6	1:18.49 million	4.89 km	192 K
Level 7	1:9.24 million	2.45 km	192 K
Level 8	1:4.62 million	1.22 km	192 K
Level 9	1:2.31 million	611.50 m	192 K
Level 10	1:1.16 million	305.75 m	192 K
Level 11	1:578,000	152.87 m	1.12 MB
Level 12	1:289,000	76.44 m	3.00 MB
Level 13	1:144,000	38.22 m	10.50 MB
Level 14	1:72,200	19.11 m	33.75 MB
Level 15	1:36,100	9.55 m	125.06 MB
Level 16	1:18,100	4.78 m	500.25 MB
Level 17	1:9030	2.39 m	1.92 GB
Level 18	1:4510	1.19 m	7.51 GB
Level 19	1:2260	0.60 m	29.83 GB
Level 20	1:1130	0.30 m	118.97 GB
Level 21	1:564	0.15 m	475.29 GB

. For building area identification, the image’s spatial resolution must meet or exceed 0.60 m [24,25]. However, high-resolution imagery is typically difficult to obtain, and as image clarity increases, the file size grows exponentially, resulting in an exponentially higher computational load for processing (Table 2). Based on this, the satellite imagery used in this study is at Level 20, with a spatial resolution of 0.6 m. After testing, this resolution was found to meet the requirements for building area identification.

2.1.2. Building Area Identification and Coverage Area Calculation

This study, implemented in Python 3.8.3, proposes an automated method for identifying building areas within road corridors from satellite imagery and calculating their areal coverage. The raw images were first processed through grayscale conversion, Gaussian filtering for noise reduction, and Canny edge detection to extract structural contours. Hough-transform line detection, combined with road centerline information obtained from KML files, was then employed to accurately delineate roadside zones. Building footprints were subsequently extracted via image segmentation, and misidentified regions were removed. The refined results were finally used to compute the areal coverage of roadside buildings.

The calculation method for the area coverage ratio is given in Equation (4). The identification results are shown in Figure 4, where the white areas represent building coverage and the black areas represent other regions.

$$Ar={\displaystyle \sum _1^i\frac{A_{ai}}{A_{ti}}}$$

(4)

where Ar is the area coverage ratio of building regions within a 200 m range on both sides of a road, in percentage (%); A_ai is the pixel count of building areas in the $i$-th image; A_ti is the total pixel count of the $i$-th image.

2.2. Motor Vehicle Exhaust Emission Factors

According to the 2022 China Mobile Source Management Environmental Yearbook [26] published by the Ministry of Ecology and Environment, the nationwide emissions of CO, HC, NO_x, and PM from automobiles in 2021 were 6.935 million tons, 1.820 million tons, 5.685 million tons, and 64,000 tons, respectively. The emission contribution rate by vehicle type is shown in Figure 5.

The annual average driving mileage of motor vehicles and the total number of vehicles are the main factors influencing the total motor vehicle exhaust emissions. The annual average driving mileage refers to the average distance traveled by each vehicle type per year. It is an important indicator for evaluating vehicle usage intensity and traffic volume, as well as a key metric for assessing energy consumption and environmental impacts in the transportation sector. This section uses the vehicle kilometers of travel (VKT) data recommended by the Technical Guidelines for the Compilation of Road Motor Vehicle Air Pollutant Emission Inventories [27], as shown in

Table 3. Emissions by Vehicle Type (tons).

Vehicle Type/Pollutant	CO	HC	NOx	PM
Micro Passenger Car	34,675	9100	56.85	0.64
Small Passenger Car	4,597,905	1,263,080	238,770	1216
Medium Passenger Car	34,675	3640	34,110	448
Large Passenger Car	208,050	27,300	602,610	4032
Micro Truck	69.35	18.2	56.85	0.64
Light Truck	1,276,040	183,820	261,510	21,824
Medium Truck	48,545	10,920	221,715	3520
Heavy Truck	735,110	322,140	4,326,285	32,960

.

Motor vehicle emission factors refer to the amount of pollutants emitted per unit distance traveled by a vehicle. According to the Technical Guidelines for the Compilation of Road Motor Vehicle Emission Inventories [27], the calculation formula for vehicle emission factors is shown in Equation (5). Based on the cumulative average driving mileage of various vehicle types obtained from public security traffic management departments, the Technical Guidelines for the Compilation of Road Motor Vehicle Emission Inventories [27] determined the comprehensive baseline emission factors for each vehicle type under standard meteorological conditions of 15 °C temperature, 50% humidity, and a baseline driving condition of 30 km/h, using gasoline and diesel with sulfur contents of 50 ppm and 350 ppm, respectively.

In this study, heavy-duty diesel trucks and small/micro passenger cars are selected as the research objects for individual vehicle exhaust diffusion patterns. The comprehensive baseline emission factors for these vehicles are shown in

Table 4. Recommended Comprehensive Baseline Emission Factors for Heavy-Duty Diesel Trucks.

Emission Standard	CO (g/km)	HC (g/km)	NOx (g/km)	PM2.5 (g/km)	PM10 (g/km)
Pre-National I	13.6	4.083	13.823	1.3222	1.45
Stage I	5.79	0.897	9.589	0.623	0.692
Stage II	3.08	0.52	7.934	0.502	0.558
Stage III	2.79	0.255	7.934	0.243	0.27
Stage IV	2.2	0.129	5.554	0.138	0.153
Stage V	2.2	0.129	4.721	0.027	0.03

and

Table 5. Recommended Comprehensive Baseline Emission Factors for Small/Micro Passenger Cars.

Emission Standard	CO (g/km)	HC (g/km)	NOx (g/km)	PM2.5 (g/km)	PM10 (g/km)
Pre-National I	25.72	2.685	1.971	0.0228	0.031
Stage I	6.71	0.663	0.409	0.026	0.029
Stage II	2.52	0.314	0.324	0.011	0.012
Stage III	1.18	0.191	0.1	0.007	0.008
Stage IV	0.68	0.075	0.032	0.003	0.003
Stage V	0.046	0.056	0.017	0.003	0.003

. Based on the motor vehicle ownership data by vehicle type and emission standard published in the China Mobile Source Environmental Management Yearbook [26] and the China Statistical Yearbook [8] (

Table 6. Vehicle Ownership by Type.

Vehicle Type	Ownership (10,000 Vehicles)
Passenger Cars
Large	153
Medium	65.45
Small	25,651.99
Micro	145.4
Freight Vehicles
Heavy	907.09
Medium	95.88
Light	2253.26
Micro	2.25

and Figure 6), the representative comprehensive baseline emission factors are calculated using Equation (6), with the results shown in

Table 7. Comprehensive Speed Correction Factors.

Vehicle Type	Speed (km/h)	CO	HC	NOx	PM10
Heavy-duty Truck	<20	1.33	1.39	1.37	1.32
	20~30	1.11	1.12	1.11	1.11
	30~40	0.92	0.91	0.92	0.92
	40~80	0.66	0.63	0.64	0.67
	>80	0.61	0.46	0.38	0.48
Passenger Car	<20	1.69	1.68	1.38	1.68
	20~30	1.26	1.25	1.13	1.25
	30~40	0.79	0.78	0.9	0.78
	40~80	0.39	0.32	0.86	0.32
	>80	0.62	0.59	0.96	0.59

.

$${\mathrm{EF}}_{i,j}={\mathrm{BEF}}_i\times {\phi }_j\times {\chi }_j\times {\lambda }_j\times {\theta }_j$$

(5)

where EF_i,j is the emission factor for the $i$-type vehicle in the $j$-region; BEF_i is the comprehensive baseline emission factor for the $i$-type vehicle; φ_i is the meteorological correction factor for the $j$-region; γ_j is the average speed correction factor for the $j$-region; λ_i is the deterioration correction factor for the $i$-type vehicle; θ_i is the correction factor for other operating conditions of the $i$-type vehicle (such as load factor, fuel quality, etc.).

$${\mathrm{EF}}_i^p={\displaystyle \sum }^j{\mathrm{BEF}}_{i,j}\times {\displaystyle \frac{PAR{C}_j}{PAR{C}_t}}$$

(6)

where EFⁱ_p is the representative comprehensive baseline emission factor for the $i$-type vehicle; BEF_ij is the comprehensive baseline emission factor for the $i$-type vehicle under the $j$-emission standard; PARC_j is the ownership of the $j$-emission standard $i$-type vehicle, in vehicles; PARC_t is the total vehicle ownership, in vehicles.

When determining the motor vehicle exhaust emission inventory, regional factors, meteorological factors, vehicle deterioration factors, and other influencing factors need to be corrected based on Equation (5). In this study, the simulation of motor vehicle exhaust diffusion only involves driving speed and wind speed conditions, with other factors set according to standard conditions and not corrected. The driving speed is directly related to the internal combustion engine operation conditions of the motor vehicle. Under different operating conditions, the exhaust emissions will also vary. Based on experimental and survey data from a relevant research team at Tsinghua University, the Technical Guidelines for the Compilation of Road Motor Vehicle Emission Inventories [27] propose driving speed correction factors for different vehicle types at varying speeds. For the simulation of motor vehicle exhaust diffusion, representative emission characteristics are required as input parameters. Ownership data for different vehicle types and emission phases were obtained from the China Mobile Source Environmental Management Yearbook [26] and the China Statistical Yearbook [8], as shown in Table 6 and Figure 6. The weighted driving speed correction factor is obtained by applying the ownership ratio of vehicles with different emission standards, resulting in the comprehensive speed correction factor shown in Table 7. Using the representative comprehensive baseline emission factor and the comprehensive speed correction factor from

Table 8. Representative Comprehensive Baseline Emission Factors.

Vehicle Type	CO (g/km)	HC (g/km)	NOx (g/km)	PM10 (g/km)
Heavy-duty Truck	2.47	0.2	5.99	0.17
Small Passenger Car	0.86	0.12	0.07	0.01

, the corrected emission factors are calculated using Equation (7), with the results shown in

Table 9. Speed-Corrected Emission Factors.

Vehicle Type	Speed (km/h)	CO	HC	NOx	PM10
Heavy-duty Truck	<20	3.28	0.28	8.2	0.23
	20~30	2.74	0.22	6.65	0.19
	30~40	2.27	0.18	5.48	0.16
	40~80	1.62	0.13	3.82	0.11
	>80	1.51	0.09	2.28	0.08
Passenger Car	<20	1.45	0.2	0.1	0.01
	20~30	1.08	0.15	0.08	0.01
	30~40	0.68	0.09	0.06	0
	40~80	0.34	0.04	0.06	0
	>80	0.53	0.07	0.07	0

.

$${\mathrm{EF}}_i^{p,v}={\mathrm{EF}}_i^p\times {\displaystyle \frac{PAR{C}_j}{PAR{C}_t}}\times {\gamma }_{i,j}$$

(7)

where EF_i^p,v is the speed-corrected representative comprehensive baseline emission factor for the $i$-type vehicle; EF_i^p is the representative comprehensive baseline emission factor for the $i$-type vehicle; PARC_j is the ownership of the $j$-emission standard $i$-type vehicle; PARC_t is the total vehicle ownership; γ_j is the speed correction factor for the $i$-type vehicle under the $j$-emission standard.

In the simulation of single-vehicle exhaust diffusion, the input parameter is the time-based emission rate, converted according to Equation (8).

$$E_{i,j}^{\prime }={\displaystyle \frac{60\times E_{i,j}^v}{V}}$$

(8)

where E^t_i,j is the emission rate of the $j$-type pollutant from the $i$-type vehicle, in g/min; E^v_i,j is the speed-corrected baseline comprehensive emission factor for the $j$-type pollutant from the $i$-type vehicle, in g/km; V is the vehicle speed, in km/h.

2.3. Establishment of the Fluid Dynamics Model

2.3.1. Single-Vehicle Exhaust Diffusion Model

This study employs Fluent for fluid dynamics simulation. The modeling process involves several key steps: defining the study area, establishing the fundamental assumptions underlying the simulation, creating grid divisions for the model, setting boundary conditions, and configuring the calculation methods. The specific modeling procedure is outlined as follows.

• Basic Model Parameters

Through a review of relevant standards and literature, the body dimension data of typical passenger cars and heavy-duty trucks were compiled. Representative data were selected, and the final simulation dimensions for the passenger car and heavy-duty truck were determined as follows:

Passenger car: Length (L) = 5 m, Width (W) = 2 m, Height (H) = 1.4 m

Heavy-duty truck: Length (L) = 21 m, Width (W) = 2.6 m, Height (H) = 2.6 m

Additionally, the determination of the computational domain is essential for defining the spatial scale of the simulated fluid. Based on relevant literature [28], the computational domain for the model is set to 10 L × 4 W × 5 H.

2.

Basic Assumptions

In order to solve for pollutant concentrations, wind speed, wind direction, and other parameters within the computational domain, certain simplifications and assumptions must be made regarding the fluid properties. The fundamental equations used in the fluid dynamics simulation in this section include the mass conservation Equation (9), the momentum conservation Equation (10), and the energy conservation Equation (11).

$$-{\displaystyle \frac{\partial \rho }{\partial t}}dxdydzdt=\left[{\displaystyle \frac{\partial \left(\rho {\mathrm{u}}_x\right)}{\partial x}}+{\displaystyle \frac{\partial \left(\rho {\mathrm{u}}_y\right)}{\partial y}}+{\displaystyle \frac{\partial \left(\rho {\mathrm{u}}_z\right)}{\partial z}}\right]dxdydzdt$$

(9)

$$u_j{\displaystyle \frac{\partial u_i}{\partial x_j}}=-{\displaystyle \frac{1}{\rho }}{\displaystyle \frac{\partial \overline{p}}{\partial x_i}}-{\displaystyle \frac{\partial }{\partial x_j}}\left(\overline{u_i^{\prime }{u}_j^{\prime }}\right)+v{\nabla }^2{u}_i$$

(10)

$${\displaystyle \frac{D}{Dt}}{\displaystyle {\iiint }_V\rho }\left(e+{\displaystyle \frac{1}{2}}u\cdot u\right)dV={\displaystyle {\iiint }_V\rho }f\cdot udV+{\displaystyle {\iint }_A{\tau }_n}\cdot udA+{\displaystyle {\iint }_{A}k}{\displaystyle \frac{\partial T}{\partial n}}dA+{\displaystyle {\iiint }_V\rho }qdV$$

(11)

3.

Mesh Generation

The mesh generation feature in Fluent was used to create surface and volume meshes for the two models. The generated mesh images are shown in Figure 7.

Turbulence Model: The simulations utilize the Realizable k-ε turbulence model with standard wall functions. This model was selected for its robustness and accuracy in simulating flows involving recirculation and separation, which are characteristic of vehicle wakes and street canyon flows.

Solver Type: The calculations were performed using a steady-state (stationary) solver. This is appropriate for simulating time-averaged flow fields and pollutant distributions under constant boundary conditions, which is the focus of our study.

Grid Quality: A detailed description of the mesh has been added. The computational domains were discretized using unstructured tetrahedral and polyhedral cells. Grid independence was achieved by refining the mesh until key solution variables (e.g., velocity magnitude at specific points, overall concentration field) showed changes of less than 2%. The following quality metrics were ensured:

Skewness: Maximum skewness was maintained below 0.85.

Orthogonal Quality: Minimum orthogonal quality was greater than 0.2.

Wall Resolution: The grid was refined near walls and the vehicle surface to resolve boundary layers, with the dimensionless wall distance y+ values predominantly in the range of 30–300, suitable for standard wall functions.

4.

Boundary Conditions

The fluid domain is defined with one outlet boundary, two inlet boundaries, and the remaining boundaries as walls. Inlet 1 is located at the rear of the vehicle to simulate the emission of exhaust pollutants, while Inlet 2 simulates the wind speed during vehicle motion. Detailed parameters are provided in

Table 10. Boundary Condition Setup.

Boundary Condition	Passenger Car				Heavy Truck
	Location	Dimension	Type	Parameter	Location	Dimension	Type	Parameter
inlet1	Rear of the vehicle model	Φ5 cm	Mass Flow Inlet		Vehicle Underside		Mass Flow Inlet
inlet2	Left side of the fluid domain	4 W × 5 H	Velocity Inlet	Simulated Vehicle Speed	Left Side of Fluid Domain		Velocity Inlet	Simulated Vehicle Speed
outlet	Right side of the fluid domain	4 W × 5 H	Outlet		Right Side of Fluid Domain		Outlet
wall1	Other locations in the fluid domain		Stationary Wall–No-Slip		Other Locations in Fluid Domain		Stationary Wall–No-Slip
wall2	Other locations on the vehicle model		Stationary Wall–No-Slip		Other Parts of Vehicle Model		Stationary Wall–No-Slip

.

The pollutant concentration parameter for the mass flow inlet is input according to Fluent 2023R1 software requirements, based on the mass flow rate and the mass fraction of pollutants. After conversion using the emission factors for passenger cars and heavy trucks at different speeds, as given in Equation (8), the results are determined. The conversion results are presented in

Table 11. Mass Flow Inlet Input Parameters.

Vehicle Type	Speed (km/h)	Mass Flow Rate	CO	HC	NOx
Heavy Truck	<20	0.000196	0.278912	0.02381	0.697279
	20~30	0.00016	0.28512	0.022893	0.691988
	30~40	0.000132	0.286255	0.022699	0.691047
	40~80	9.28 × 10−5	0.290844	0.023339	0.685817
	>80	6.47 × 10−5	0.389175	0.023196	0.587629
Passenger Car	<20	2.92 × 10−5	0.828571	0.114286	0.057143
	20~30	2.18 × 10−5	0.824427	0.114504	0.061069
	30~40	1.38 × 10−5	0.819277	0.108434	0.072289
	40~80	7.33 × 10−6	0.772727	0.090909	0.136364
	>80	1.12 × 10−5	0.791045	0.104478	0.104478

. “The mass fractions at inlet1 (Table 11) are set as idealized values, representing a pure pollutant mixture without air. This simplification may alter fluid properties like density but ensures consistency in comparing diffusion trends under varying vehicle speeds and types”.

5.

Calculation Method

In the Fluent solver, the gravitational acceleration is set to −9.8 m/s² along the y-axis. The energy equation is activated, and component transport is enabled (including inlet diffusion, diffusion energy terms, and complete multicomponent diffusion). The required CO, HC, and NOx components are imported from the Fluent database under the “Materials—Mixture” option. The solver employs the SIMPLE method, with 3000 iterations specified [29].

2.3.2. Street Canyon Modeling

The street canyon modeling, including the determination of the computational domain, mesh generation, and boundary condition definition, is carried out following the fluid dynamics simulation process. The specific steps are as follows.

• Determination of Canyon Geometry

Nicholson [30] first proposed the street canyon model in 1975, using it as a simplified model for assessing atmospheric pollutants. With the development of fluid dynamics simulation software, the application of the street canyon model has expanded, gradually becoming a simplified model for studying road-related hydrology, atmospheric conditions, urban planning, and other related fields [31]. This section adopts seven different street canyon model configurations, including ideal street canyons, wide street canyons, deep street canyons, asymmetric street canyons, short street canyons, and long street canyons, to simulate varying building configurations around roads. The parameters are shown in

Table 12. Street Canyon Models.

No.	Type	Width-to-Height Ratio (W/H)	Height-to-Length Ratio (H/L)	Model
Canyon-1	Ideal Street Canyon	1	4
Canyon-2	Wide Street Canyon	3.3	4
Canyon-3	Deep Street Canyon	0.3	4
Canyon-4	Asymmetric Street Canyon	1/0.6	4
Canyon-5	Short Street Canyon	1	2.7
Canyon-6	Long Street Canyon	1	8.3

.

2.

Determination of Computational Domain and Mesh Generation

The first step in conducting a fluid dynamics simulation is to define the computational domain. The computational domain is set as a rectangular prism with dimensions of 24 H in length, 24 H in width, and 20 H in height, with the street canyon positioned at the geometric center of the bottom of the prism, as shown in Figure 8. Additionally, the grid generation function provided by Fluent is used to mesh the established model, as illustrated in Figure 9.

3.

Boundary Conditions

One side of the fluid domain is designated as Inlet 1, a velocity inlet, with the inlet material set as air to simulate the direction and intensity of natural wind. The opposing wall is set as an Outflow boundary to allow free exit of the flow. The surfaces of the buildings and other boundaries of the fluid domain are all set as Wall, with smooth wall conditions. A horizontal plane, 0.5 m in height, is positioned at the center of the road, designated as Inlet 2, a mass flow outlet, to simulate exhaust gas emissions from the road’s line source.

Table 13. Equivalent Traffic Volume Conversion Examples.

Traffic Condition	Passenger Cars (Vehicles)	Average Speed (km/h)	Heavy Trucks (Vehicles)	Average Speed (km/h)	Equivalent Value
Free Flow	843	53	815	46	94569
Slow-moving	1264	31	925	27	223,069

provides conversion examples for two typical four-lane road conditions: smooth traffic and slow traffic.

$$E_j={\displaystyle \frac{EVT\times E_j^s}{v_a}}$$

(12)

where E_j is the linear emission source intensity of the $j$-th pollutant, in g/(min·km); EVT is the equivalent of the passenger car standard condition derived from the equivalent traffic volume; E^s_j is the emission factor for the $j$-th pollutant under the passenger car standard condition, in g/min; v_a is the weighted average vehicle speed for different vehicle types, in km/h.

4.

Basic Assumptions and Calculation Method

The same basic assumptions and calculation methods as those in Section 2.3.1 are applied.

2.4. Experimental Scheme

2.4.1. Single-Vehicle Exhaust Diffusion

The diffusion rate of exhaust pollutants from vehicles varies under different driving conditions, influenced by vehicle type, speed, and pollutant emission rate. To investigate the diffusion patterns of vehicular exhaust under various operating conditions, this section presents simulation schemes for different vehicle types and conditions, as shown in

Table 14. Single-Vehicle Exhaust Diffusion Simulation Scheme.

No.	Vehicle Type	Speed (km/h)	Simulated Wind Speed (m/s)	Simulation Condition
VED-M-I	Passenger Car	0	0	Idle
VED-M-L	Passenger Car	10	2.78	Low Speed
VED-M-H	Passenger Car	50	13.89	Free Flow
VED-H-I	Heavy Truck	0	0	Idle
VED-H-L	Heavy Truck	10	2.78	Low Speed
VED-H-H	Heavy Truck	50	13.89	Free Flow

. The vehicle speed is simulated by adjusting the inlet wind speed at Inlet 1.

2.4.2. Street Canyon Exhaust Diffusion

Using street canyon morphology, traffic operation conditions, wind speed, and wind direction as variables, Fluent is employed for fluid dynamics simulation to explore the diffusion characteristics of vehicular exhaust pollutants in street canyons under various conditions. The street canyon morphology settings are outlined in Table 12 while wind speeds are set at 0.2 m/s, 4.4 m/s, and 10.8 m/s to simulate still air, light breeze, and strong wind conditions, respectively. Wind direction is adjusted to 0°, 30°, 45°, and 90° relative to the street axis. The simulation schemes are presented in

Table 15. Street Canyon Fluid Dynamics Simulation Scheme.

No.	Street Canyon Type	Wind Speed (m/s)	Wind Direction (°)	Linear Source Pollution Intensity
S1V1D30	Canyon-1	4.4	30	Free Flow
S2V1D30	Canyon-2	4.4	30	Free Flow
S3V1D30	Canyon-3	4.4	30	Free Flow
S4V1D30	Canyon-4	4.4	30	Free Flow
S5V1D30	Canyon-5	4.4	30	Free Flow
S6V1D30	Canyon-6	4.4	30	Free Flow
S1V2D30	Canyon-1	0.2	30	Free Flow
S1V3D30	Canyon-1	10.8	30	Free Flow
S1V1D0	Canyon-1	4.4	0	Free Flow
S1V1D45	Canyon-1	4.4	45	Free Flow
S1V1D90	Canyon-1	4.4	90	Free Flow
S1V1D90S	Canyon-1	4.4	30	Slow-moving

.

2.4.3. Experimental Design of Estimation of the Application Effect of Photocatalytic Pavement Materials for Exhaust Gas Degradation

The street canyon model used in this section of the fluid dynamics simulation is the Canyon-1 ideal street canyon model from Table 12 in Section 2.3.2, with a width-to-height ratio (W/H) of 1 and a height-to-length ratio (H/L) of 4. The computational domain, grid division, boundary conditions, basic assumptions, and calculation methods are all the same as those described in Section 2.3.2.

By reducing the road pollutant line-source intensity, the photocatalytic exhaust gas degradation function of eco-pavement materials is simulated, aiming to investigate the changes in pollutant concentrations in the surrounding atmospheric environment after the pavement is endowed with exhaust gas degradation functionality. The pollutant emissions under congested traffic conditions are reduced by 30% and 50%, with the reduction results shown in

Table 16. Line-Source Emission Intensities at Different Reduction Rates.

Condition	Reduction Rate (%)	Line-Source Intensity (kg/(min·km))
		Total Emission	CO	HC	NOx
Congestion	0	3.08	2.52	0.33	0.22
Congestion	30	2.16	1.77	0.23	0.16
Congestion	50	1.54	1.26	0.17	0.11

. The exhaust pollutant diffusion scenarios for a regular street canyon are then simulated at a wind speed of 4.4 m/s and four wind direction angles (0°, 30°, 45°, 90°) according to the emission intensities in Table 16. The specific simulation parameter settings are provided in

Table 17. Simulation Parameters for Estimating the Effectiveness of Photocatalytic Pavement Materials for Exhaust Gas Degradation.

No.	Wind Direction Angle (°)	Line-Source Emission Intensity Reduction Rate (%)
C0D0	0	0
C0D30	30	0
C0D45	45	0
C50D45	45	50
C50D90	90	50
C0D90	90	0
C30D0	0	30
C30D30	30	30
C30D45	45	30
C30D90	90	30
C50D0	0	50
C50D30	30	50

.

3. Results and Discussions

3.1. Equivalent Pollutant Vehicle Conversion Method Based on Human Toxicity Potential

In the study of atmospheric environment deterioration along roadways, it is necessary to estimate the emission rate of linear source pollutants. To assess the pollution impact of different vehicle types on the surrounding environment, the “equivalent pollution” principle is proposed, as illustrated in Figure 10. The direct impact of exhaust emissions is on human health, and the Human Toxic Potential (HTP) is an indicator used in life cycle assessment to quantify the toxicity of a pollutant to human health.

This section, based on the fundamental principles of industrial ecology and relevant literature [32], derives the equivalent factors of the major exhaust pollutants in terms of 1,4-C6H4Cl₂-eq, as shown in

Table 18. Pollutant Equivalent Factors Based on Human Toxic Potential.

Pollutant	Equivalent Factor (1,4-C6H4Cl2-eq)
CO	0.012
NOx	1.2
PM10	0.82

. Using representative composite baseline emission factors (Table 8), the equivalent values of pollutants for heavy-duty trucks and passenger cars at different speeds are obtained (

Table 19. Pollutant Equivalent Emission Values for Different Vehicle Types Based on Human Toxic Potential (Unit: 1,4-C₆H₄Cl₂-eq Equivalent).

Vehicle Type	Speed (km/h)	CO	NOx	PM10	Total
Heavy Truck	<20	0.03936	9.84	0.1886	10.06796
	20~30	0.03288	7.98	0.1558	8.16868
	30~40	0.02724	6.576	0.1312	6.73444
	40~80	0.01944	4.584	0.0902	4.69364
	>80	0.01812	2.736	0.0656	2.81972
Passenger Car	<20	0.0174	0.12	0.0082	0.1456
	20~30	0.01296	0.096	0.0082	0.11716
	30~40	0.00816	0.072	0	0.08016
	40~80	0.00408	0.072	0	0.07608
	>80	0.00636	0.084	0	0.09036

). Additionally, based on the 30–40 km/h operating condition for passenger cars, the equivalent values at different speeds are calculated using Equation (13), as shown in

Table 20. Equivalent Values for Standard Condition Passenger Cars.

Vehicle Type	Speed (km/h)	Equivalent Value (Standard Condition Passenger Car)
Heavy Truck	<20	125.5983
	20~30	101.9047
	30~40	84.01248
	40~80	58.55339
	>80	35.17615
Passenger Car	<20	1.816367
	20~30	1.461577
	30~40	1
	40~80	0.949102
	>80	1.127246

.

$$VE{E}_{i,v}=\sum _j{E}_{i,j}^{p,v}\times PD{V}_x$$

(13)

where VEE_i,v is the 1,4-C₆H₄Cl₂-eq equivalent value of pollutant emissions for the $i$-th vehicle type at speed $v$; E^p,v_i,j is the emission factor for the $j$-th pollutant from the $i$-th vehicle type at speed $v$; PDV is the pollutant’s 1,4-C₆H₄Cl₂-eq equivalent factor based on human toxicity potential.

3.2. Single-Vehicle Pollutant Concentration Field Analysis

To investigate the diffusion patterns of vehicular exhaust pollutants in the XZ, YZ, and XY dimensions, CO concentration distribution maps were output for two vehicle types at the following locations: 10 cm beneath the vehicle (XZ-1), directly beneath the vehicle (XZ-2), 10 cm below the vehicle (XZ-3), at the vehicle’s rear (YZ-1), 10 cm behind the vehicle (YZ-2), and at the exhaust pipe center (XY-1). These concentration distribution maps are shown in Figure 11, Figure 12, Figure 13, Figure 14, Figure 15 and Figure 16.

3.2.1. Passenger Car

By analyzing the fluid dynamics simulation results, it is evident that the diffusion patterns of different types of pollutants are generally similar. Therefore, this section takes the CO concentration field as an example to analyze the diffusion behavior of exhaust pollutants. From Figure 11, it can be observed that on the same plane, as the simulated vehicle speed increases, the exhaust diffusion range becomes wider, with high-concentration areas concentrated directly behind the vehicle’s rear at the Inlet 2 location. Additionally, from the CO concentration cloud maps at different heights on the XZ plane, it is apparent that the diffusion range at the exhaust pipe height is significantly smaller than that above and below the exhaust pipe. This suggests that once the exhaust is expelled into the fluid space, under the influence of airflow, it exhibits a dual diffusion trend, both upward and downward.

Furthermore, by comparing the cloud maps 10 cm above and below the exhaust pipe, it can be observed that due to the airflow beneath the vehicle, the exhaust diffusion range and concentration below the exhaust pipe at a height of 10 cm are greater than those above it.

Figure 12 illustrates the CO concentration distribution along the vehicle’s longitudinal center plane. The cloud map from the XY plane indicates that, under different operating conditions, the exhaust from the passenger car consistently diffuses downward and backward. This behavior aligns with the analysis from the XZ plane at different heights. There are two reasons for this phenomenon: first, in both congested and free-flow conditions, due to the vehicle’s forward motion, a low-pressure zone forms behind it, inhibiting the upward diffusion of exhaust pollutants; second, the exhaust pipe of most passenger cars is set at a downward angle, and consequently, the initial direction of the Inlet 2 boundary is also set at a 45° downward angle. Therefore, when pollutants enter the simulation domain, they initially have a downward velocity.

Figure 13 displays the CO concentration distribution on the YZ plane. It can be seen that, under idle conditions, due to the absence of the low-pressure zone behind the vehicle, diffusion in the vertical direction occurs more rapidly. Moreover, the pollutants are primarily distributed within a 1.3 m range above the ground. The cloud maps at different distances from the vehicle’s rear exhibit similar distribution patterns.

3.2.2. Heavy Truck

Figure 14, Figure 15 and Figure 16 illustrate the exhaust pollutant diffusion patterns of heavy-duty trucks in three different dimensions. Due to differences in vehicle type, exhaust pollutant emission ratios, and exhaust pipe configurations, the diffusion patterns of heavy-duty truck exhaust pollutants differ significantly from those of passenger cars.

Since the exhaust pipe of heavy-duty trucks is typically positioned at the vehicle’s underside, exhaust pollutants in the YZ horizontal plane are largely blocked by the vehicle’s body and do not diffuse upward. As a result, in Figure 14a, there is virtually no high-concentration CO region, with only low-concentration areas appearing along the vehicle’s sides. From the concentration field at the exhaust pipe and 10 cm below, a similar diffusion pattern to that of passenger cars is observed. As the height of the horizontal plane decreases, the high-concentration exhaust pollutant region expands.

In Figure 15, the CO concentration cloud map at the vehicle’s center plane reveals that, compared to passenger cars, the exhaust pollutant concentration at the rear of the heavy-duty truck is relatively lower. This is because the exhaust pipe is not positioned at the rear of the vehicle. The Inlet 2 location is positioned near the rear center of the vehicle’s underside, causing most of the exhaust pollutants to accumulate beneath the vehicle.

As shown in Figure 13, on the YZ plane, the faster the vehicle travels, the higher the concentration of pollutants at the vehicle’s underside. The ground clearance of most heavy-duty trucks is typically below 0.4 m. Therefore, for heavy-duty trucks, the high-concentration pollutant region is concentrated within a 0.4 m range from the ground.

Furthermore, given that NO_x is the predominant pollutant emitted from heavy-duty trucks, the areas identified with elevated pollutant concentrations in our simulations would correspond to regions with potentially significant NOx exposure, which carries distinct health and environmental risks compared to CO.

3.3. Street Canyon Pollutant Concentration Field Analysis

3.3.1. Impact of Street Canyon Morphology

In investigating the impact of street canyon morphology on the diffusion patterns of exhaust pollutants, the traffic operating condition is set to free-flow, with a wind speed of 4.4 m/s and a wind direction angle of 30° relative to the street axis. The building models are based on the six scenarios outlined in Table 12. The simulation results are then analyzed by capturing the CO concentration field on the building and street surfaces, representing the diffusion behavior of exhaust pollutants.

From Figure 17 it can be observed that the CO concentration field varies significantly under the same environmental wind and exhaust emission intensity across different street canyon morphologies.

Firstly, from the simulation results of the ideal street canyon shown in Figure 17a, where the width-to-height ratio of the model is constant, pollutants primarily diffuse from the downwind side, with a small portion of pollutants crossing over the tops of buildings and spreading outward. Additionally, the pollutant concentration on the external facades of the buildings directly reflects the impact of exhaust pollutants on the surrounding environment. High-concentration areas of exhaust pollutants on the building walls indicate that the pollutant concentration near the buildings on both sides of the road reaches high levels, making it easier for exhaust pollutants to pose a threat to residents within the buildings. Due to the wind direction, a high-concentration CO area appears on the windward side of the building’s rear facade.

Figure 17b shows the simulation results for a wide street canyon. Because the distance between the buildings is larger, more natural wind can pass through the street canyon, facilitating the diffusion of exhaust pollutants. There is virtually no distribution of exhaust pollutants on the external facades of the buildings, indicating that the building morphology in a wide street canyon can alleviate the impact of vehicle emissions on the human activity zones on both sides of the buildings.

Figure 17c illustrates the simulation results for a deep street canyon. Compared to other models, the street canyon is too narrow, and the natural wind passage is small, which hinders the diffusion of exhaust pollutants. This results in significant concentration buildup on the external facades of the buildings on both sides.

In the concentration map of the asymmetric street canyon shown in Figure 17d, due to the higher side blocking the natural wind, the airflow above the lower side of the building is slower. However, as the height increases, the obstruction from the taller building is reduced, and the wind speed increases. This wind speed difference creates a low-pressure zone above the shorter building’s side, allowing exhaust pollutants to diffuse not only from the rear of the street but also to spread over the lower building. Consequently, CO concentrations are higher at the top and facade of the lower building compared to other models.

Figure 17e,f show the simulation results with changes in street canyon length. The diffusion patterns of exhaust pollutants are similar to those in the ideal street canyon. However, in the long street canyon, CO concentration accumulation areas appear on the side and top of the buildings at the windward end of the canyon.

In conclusion, buildings in wide and long street canyons are less affected by exhaust pollutants, while buildings in narrow and asymmetric street canyons experience greater exposure to pollution.

3.3.2. Impact of Wind Speed

By varying the wind speeds of 0.2 m/s, 4.4 m/s, and 10.8 m/s at a wind direction angle of 30° in the ideal street canyon model (Canyon-1), the impact of wind speed on the diffusion patterns of vehicle pollutants is explored. The CO concentration fields on the building and street surfaces are output, as shown in Figure 18.

From Figure 18, at a wind speed of 0.2 m/s, simulating a light breeze, CO does not diffuse from the rear of the street canyon but instead accumulates between the buildings. The CO concentrations on the street and building facades are significantly higher compared to the other two wind speed scenarios. As the wind speed increases to 4.4 m/s, simulating moderate wind conditions, the wind speed in the canyon passage between the buildings increases. CO continuously diffuses to the rear of the street canyon and spreads outward towards the sides of the canyon. When the wind speed reaches 10 m/s, high concentrations of exhaust pollutants appear only on the windward side of the buildings. At other locations, the concentration of exhaust pollutants is much lower than in the two lower wind speed scenarios.

3.3.3. The Impact of Wind Speed

By varying the wind direction angles of 0°, 30°, 45°, and 90° at a wind speed of 4.4 m/s in the ideal street canyon model (Canyon-1), the influence of wind direction on the diffusion patterns of vehicular pollutants was investigated. The CO concentration fields on the building and street surfaces are presented in Figure 19.

From Figure 19a, it can be observed that when the wind direction is parallel to the street canyon axis, the wind speed within the canyon is maximized, allowing pollutants to diffuse rapidly from the rear of the street canyon and reducing the concentration of exhaust gases on the facades of the buildings. As the wind direction deviates from the street canyon axis, with the change in angle, the windward side of the buildings begins to be influenced by crosswinds, gradually leading to the accumulation of CO at certain concentrations. Additionally, due to the difference in wind speeds, pollutants begin to diffuse over the tops of the buildings.

Comparing Figure 19b and Figure 19c, both the accumulation of pollutants on the building’s side facades and the diffusion over the tops of the buildings increase as the angle between the wind direction and the street canyon axis increases. This is because the larger the wind direction angle, the slower the airflow within the street canyon, which in turn increases the airspeed difference between the building top and the surrounding air. The greater the difference in wind speeds, the larger the pressure difference, resulting in more pollutants rising and diffusing over the tops of the buildings.

When the wind direction is perpendicular to the street canyon axis, natural wind does not directly enter the canyon. Instead, pollutants spread through the ends and tops of the buildings, relying on the pressure differences. This leads to a significant reduction in the airspeed within the street canyon compared to other scenarios, causing a slowdown in the diffusion rate of exhaust pollutants and resulting in an accumulation of pollutants within the canyon.

3.3.4. The Impact of Traffic Operating Conditions (Line-Source Intensity)

By varying the mass flow rate at the inlet 2 (1.31 kg/(min·km) and 3.08 kg/(min·km)) in the ideal street canyon model (Canyon-1) with a wind speed of 4.4 m/s and a wind direction angle of 30°, the emission scenarios for both free-flow and congested traffic conditions were simulated. The impact of different traffic volumes and operating conditions on the pollutant diffusion patterns in the street canyon was investigated. CO was used as the indicative pollutant, and concentration fields on the building and street surfaces were plotted, as shown in Figure 20.

According to Figure 20, the contour shape and direction of the concentration fields remain largely unchanged across different emission intensities, indicating that the exhaust pollutant emission strength has little effect on the direction and pattern of pollutant diffusion. However, as shown in Figure 20b compared to Figure 20a, the cloud diagram’s color intensity increases, and the diffusion range slightly expands. This suggests that during traffic congestion, as vehicle speeds decrease and the number of vehicles per unit distance increases, the line-source intensity of vehicle emissions rises. This increase is reflected in the fluid dynamics simulation, where the CO concentration field indicates a more severe accumulation of pollutants within the street canyon, thereby exacerbating the environmental impact of traffic emissions on the surrounding area.

3.3.5. Impact Patterns

Through the calculation and analysis of models with different street canyon shapes, wind speeds, wind direction angles, and vehicular pollutant emission intensities, the following patterns can be summarized:

Street Canyon Shape: The shape of the street canyon significantly influences the diffusion pattern of exhaust pollutants. Wider street canyons facilitate faster pollutant dispersion, while in asymmetric street canyons, taller buildings can block the natural wind, causing pollutants to diffuse over the tops of the buildings. In deep street canyons, the accumulation of exhaust pollutants is most severe.

Wind Speed: Wind speed does not alter the diffusion pattern of pollutants within the street canyon, but higher wind speeds significantly accelerate the pollutant dispersion rate.

Wind Direction: The diffusion path and pattern of pollutants vary significantly with different wind directions. In general, smaller angles between the wind direction and the street canyon axis are more favorable for the diffusion of exhaust pollutants.

Traffic Operating Conditions: Traffic operating conditions are closely related to the emission intensity of the road’s line-source pollutants. The emission intensity has little effect on the diffusion pattern of pollutants but results in an increase in pollutant concentrations within the affected areas.

3.4. Estimation of the Application Effect of Photocatalytic Pavement Materials for Exhaust Gas Degradation

Fluid dynamics simulations are conducted according to the emission concentration reduction rates and wind direction angle parameters listed in Table 18. The CO concentration field at the representative breathing height (XZ-1.5 m horizontal plane) is output to analyze the impact of different emission concentration reduction rates on the CO concentration field under varying wind direction angles.

In Figure 21, the contour of the CO concentration field remains unchanged under the same wind direction; thus, it is inferred that the application of photocatalytic pavement materials for exhaust gas degradation does not alter the diffusion pattern and distribution of exhaust gases. However, by comparing the CO concentration field under the same wind direction, it can be observed that the concentration in the same area gradually fades with the increase in the reduction rate. It is evident from the figure that the areas most affected are primarily the diffusion zones at both ends of the street and those near the buildings. When no reduction in the emission line source is applied, the high-concentration exhaust pollution area almost covers the entire street canyon. After a 30% reduction in the exhaust pollution source, a significant decrease is observed in the regions immediately adjacent to the buildings. Following a 50% reduction, the concentration in the areas close to both sides of the buildings approaches zero. The areas with the highest human activity, such as shops, buildings, and sidewalks along the roadside, will benefit greatly from the application of photocatalytic pavement, contributing significantly to the health of residents and the improvement of the atmospheric environment.

To quantitatively characterize the extent of pollution reduction in the road domain under different pollutant reduction rates, a representative point (Point R) was selected from the midpoint of the street length at a distance of 0.5 m downwind from the building, with an elevation of 1.5 m. The CO concentration values at this point are shown in

Table 21. R-point CO Concentration and Reduction Rate.

No.	CO Concentration (kmol/m3)	CO Concentration (ppm)	Reduction Rate (%)
C0D0	2.2153	79.12	0
C0D30	3.3265	118.8	0
C0D45	3.5427	126.53	0
C0D90	3.9642	141.58	0
C30D0	1.5795	56.41	28.7
C30D30	2.2167	79.17	33.36
C30D45	2.2763	81.3	35.75
C30D90	2.4795	88.55	37.45
C50D0	1.1215	40.05	49.38
C50D30	1.5963	57.01	52.01
C50D45	1.6153	57.69	54.41
C50D90	1.7156	61.27	56.72

and Figure 22.

Based on the CO concentration reduction rates calculated in Table 21, R-point CO Concentration and Reduction Rate, the CO concentration reduction rate at the representative point R under different pollutant emission concentration reduction rates ranges from 28.70% to 56.72%. Overall, the CO reduction rate exhibits a high correlation with the reduction rate of pollutant emissions. Furthermore, the CO concentration reduction rate at the representative point also shows a systematic variation with different wind angles. Generally, the larger the wind angle, the greater the CO concentration reduction rate. This is because, at smaller wind angles, the wind speed within the street is higher, leading to a faster air circulation rate, resulting in a relatively lower reduction effect, which is also lower than the emission source reduction rate. As the wind angle increases, the internal air circulation speed within the street decreases, and the CO concentration reduction effect becomes more pronounced. Therefore, the application of photocatalytic ecological pavement materials can significantly reduce the concentration of exhaust pollutants in the human breathing zone, with the effectiveness strongly correlated with the exhaust degradation efficiency.

4. Conclusions

This chapter investigates the development of buildings along urban expressways and the current state of motor vehicle exhaust emissions, based on surveys of the building coverage rate and vehicle emission data. An “equivalent pollution vehicle conversion method” is proposed, and fluid mechanics simulations are employed to study the exhaust diffusion patterns of individual vehicles and the atmospheric degradation along roadways. The application effect of photocatalytic pavement for exhaust mitigation is also analyzed. The findings underscore the significance of developing photocatalytic pavement for exhaust mitigation and provide data support for subsequent chapters on the evaluation of exhaust degradation performance. The main conclusions are as follows.

Building Coverage Rate: By utilizing geographic information and image recognition technology, satellite maps of six cities (Beijing, Shanghai, Xi’an, Nanjing, Wuhan, Jinan) over the past decade were analyzed for building coverage along their urban expressways. It was found that, during the early stages of construction, building coverage around the expressways was approximately 1%. As urban development progressed, human activity within the roadway areas increased, resulting in a rise in building coverage, reaching as high as 5.7% depending on the city’s development characteristics and speed.

Vehicle Emission Factors: Statistical analysis of relevant data provided representative composite baseline emission factors for different vehicle types, which were then corrected for speed. By applying human toxic potential factors for pollutants, the equivalent emission values for small passenger vehicles were established for different vehicle types and speeds, resulting in the creation of an equivalent pollution vehicle conversion method.

Fluid Mechanics Simulation: Based on statistical data of vehicle dimensions, fluid mechanics simulations were conducted using Fluent software to model small passenger cars and heavy-duty trucks. Relevant boundary conditions and emission parameters were determined. The results showed that under the influence of both driving speed and vehicle type, the distribution range of exhaust pollutants at lower levels was broader, with high-concentration areas concentrated below the vehicle’s height.

Exhaust Emission Intensity in Different Traffic Conditions: Using the equivalent pollution conversion model and current exhaust emission data, the line-source exhaust emission intensity was calculated for two traffic conditions: free-flowing and congested. The emission intensities were 1.31 kg/(min·km) and 3.08 kg/(min·km), respectively. By altering factors such as wind speed and canyon model parameters, the simulations demonstrated that street canyon morphology significantly affects exhaust pollutant diffusion patterns. Wide and short street canyon models facilitated faster diffusion, while high wind speeds increased the diffusion rate without altering the pollutant diffusion path, thereby reducing exhaust pollutant concentrations. The wind direction angle also influenced the pollutant diffusion path, with wind directions closer to the street canyon axis being more conducive to pollutant dispersion.

Evaluation of Photocatalytic pavement for exhaust mitigation: Based on reductions in line-source emission strength, the application effects of exhaust-degrading ecological pavement materials were assessed. Simulations were conducted for three scenarios with emission reduction rates of 0%, 30%, and 50%, and the pollutant concentration fields were calculated for the XZ-1.5 m plane. The results showed that reducing the emission intensity did not affect the diffusion pattern of pollutants, but concentrations of pollutants at the street’s end and near buildings significantly decreased. The CO concentration reduction at representative point R ranged from 28.70% to 56.72%. This indicates a reduction in exhaust pollutant concentrations at the human breathing height within activity zones, demonstrating that the application of exhaust-degrading ecological pavement materials can significantly improve air quality in the roadway environment.

In summary, this study established an integrated framework to assess air quality deterioration in road areas, linking macro-scale traffic activity to micro-scale pollutant dispersion.

The practical implications of these findings are twofold. Firstly, the equivalent-pollution method provides a valuable tool for environmental impact assessments, enabling a more health-centric evaluation of traffic policies. Secondly, the CFD results offer direct guidance for urban planning, suggesting that optimizing street canyon geometry and selectively deploying photocatalytic pavements in critical areas can be effective strategies for mitigating exposure risks.

Despite the insights gained, this work opens up several avenues for future research. The immediate priority is the on-site validation of the simulated dispersion patterns. Furthermore, extending the model to account for chemical transformations between pollutants and conducting a thorough life-cycle cost–benefit analysis of mitigation technologies like photocatalytic pavement would significantly enhance the practical applicability and depth of future studies.