Research on the Mechanism and Source Changes of Urban O₃ Formation Under the Background of Increased Industrial Activity Levels

Abstract

The increase in industrial production can lead to more complex emissions of O₃ precursors, but the changes in the formation mechanism and source of O₃ are still unclear. Taking Jincheng as the typical industrial city, an observation-based model (OBM) is explored to analyze the changes in O₃ formation in 2022 and 2024. The results indicated that the concentration of NOx and VOCs in 2024 increased by 21.1% and 22.3%, respectively. And the concentrations of alkenes related to industrial processes increased significantly. RO₂+NO is the main pathway for O₃ formation (51.5~54.2%), while VOCs+OH· contributes most to the formation of RO₂. VOC and NOx both play important roles in O₃ formation, and the sensitivity of VOCs increased from 0.76 to 0.84 in 2022 and 2024, with alkenes increasing the most. Industrial processes and coal combustion are the important sources for O₃ and its precursors, and the contribution of the industrial process increased significantly during 2022 and 2024. In summary, the increase in the industrial activity level has led to the increase in alkenes, which has a key impact on the formation of O₃. Controlling the emission of alkene from the industrial process is the direction for the continuous control of O₃ pollution in industrial cities.

1. Introduction

O₃ is an important pollutant in the atmosphere, and the atmospheric O₃ pollution has been more severe in recent years [1,2,3]. O₃ is mainly generated by the reaction of precursors (VOCs and NOx) under light conditions [4,5], but there is a complex nonlinear relationship between the precursors and O₃ [6,7,8]. The precursor sources of O₃ generation in atmospheric environments are complex, among which VOCs come from vehicle emissions, industrial processes, fossil fuel volatilization, and solvent utilization, and NOx come from the emissions of vehicles, coal combustion, and heavy industries such as cement, iron and steel, and chemicals [9,10]. This study shows that industrial sources account for a relatively large proportion of VOCs emissions (30~80%) [11,12,13,14]. The complexity of industrial process emissions increases the difficulty of urban atmospheric O₃ pollution control. The industrial production in China has risen in recent years, and the increase in the level of industrial activities has led to a change in the situation of precursors emissions. As the pollution source that contributes greatly to O₃ formation, the influence of industrial processes on the mechanism and sensitivity of O₃ formation is still unclear within the background of the change in the precursor emission situation. Therefore, it is of great significance to study the changing trend of precursors and their influence on the O₃ formation process and O₃ sources under the background of the increase in industrial production. This is an important scientific basis for industrial cities to adjust control measures in a timely manner and reduce atmospheric O₃ pollution.

The current methods used to simulate atmospheric photochemical reaction mechanisms mainly include the air quality model system based on emission inventories and box models based on observed data. The air quality model system could be used to study the reaction mechanism and the source of pollutants. However, due to the uncertainty of the emission inventory, the simulation results in a lack of accuracy [15,16]. The box model focuses on simulating the local chemical formation process, including detailed chemical reactions, and can objectively reflect the actual atmospheric photochemical reaction characteristics in the local area [17,18]. There is a researcher who used the box model to analyze the important role of VOCs species and free radicals in the formation of O₃ [19]. Zhang used the box model to analyze the O₃ control scenarios, and the results showed that the ratio of VOCs/NOx should be no less than 0.72 to effectively control O₃ [20]. Box models are often coupled with ozone formation potential (OFP) or receptor models for O₃ source apportionment. The OFP is calculated based on ideal state parameters, and the results may deviate relatively from reality [21,22]. The box model coupled with the receptor model is based on the analysis of observed concentrations, which is more consistent with actual pollution source results. Lyu used this method to conclude that vehicle exhaust gas sources (27.8%) and coal combustion sources (21.8%) were the main contributors to O₃ generation in Wuhan [23]. Li’s research results show that industrial processes (29%), vehicle emissions (27%), and coal combustion (24%) are the main contributing sources of O₃ formation in Changzhi [24]. Therefore, the box model coupled with the receptor model approach was chosen in this study to explore the changes in the urban atmospheric O₃ formation and source contributions within the background of increasing industrial production.

Jincheng is a typical industrial city with coal resources. In June 2022, the raw coal production was 13.924 million tons, the steel production was 851,000 tons, the pig iron production was 418,000 tons, and the cement production was 321,000 tons. With the increase in the industrial activity level, the industrial products in 2024 have significantly increased. The production of raw coal, steel, pig iron, and cement in June 2024 were 14.35 million tons, 1.159 million tons, 470,000 tons, and 332,000 tons, respectively [25]. However, the mechanism of O₃ pollution caused by changes in industrial production is not clear. In this study, the observed concentrations of VOCs during the summer O₃ pollution episodes in 2022 and 2024 were collected. An OBM was employed to analyze the atmospheric O₃ formation, free radical reaction processes, and sensitive precursors in Jincheng. Additionally, the coupling of the OBM with positive matrix factorization (PMF) was used to identify the sources of O₃ and compare the mechanisms of atmospheric O₃ chemical formation before and after the increase in industrial production. The aim of this study is to provide a more accurate and effective basis for the prevention and control of air pollution in Jincheng.

2. Materials and Methods

2.1. Sample Collection and Quality Control

The sampling site (Figure 1), located on the rooftop of the Environmental Monitoring Station of Jincheng, is adjacent to major transportation arteries and is surrounded by densely populated residential and commercial areas. This location can better reflect the characteristics of the urban air quality. According to the industrial production data, the production amounts of steel, pig iron, and crude steel in June 2024 increased by 36.2%, 20.3%, and 12.7% compared with those of June 2022. In order to better compare the changes in O₃ formation mechanisms caused by changes in industrial production, the typical O₃ pollution episode in June 2022 (from 12 June 2022 to 30 June 2022, OE1) was selected as the research period before the increase in industrial production, and the typical O₃ pollution episode in June 2024 (from 8 June 2024 to 16 June 2024, OE2) was selected as the research period after the increase in industrial production. We collected the VOC concentrations during these two O₃ pollution episodes and synchronously recorded meteorological parameters, such as the temperature (T), relative humidity (RH), and atmospheric pressure (P).

VOCs data were analyzed using an online gas chromatograph–mass spectrometer/flame ionization detector (GC-MS/FID, EXPEC 2000), which can continuously monitor over 100 VOCs with a 1-h time resolution, including PAMS, OVOCs, Haloalkane, etc. The system is mainly composed of a low-temperature preconcentration device and a GC-MS instrument. Firstly, the collected air samples were enriched in the preconcentrator with a three-stage cold hydrazine concentrator, which is used to remove impurities such as H₂O, NO₂, and CO₂ from the samples. Subsequently, the samples were desorbed in the high-temperature environment of the thermal desorber [26,27]. After high-temperature thermal desorption, the samples entered the chromatographic separation section. C2–C5 VOCs were separated on the Plot (50 m × 0.32 mm × 8 μm) column and analyzed by the FID. Other VOCs species were separated on the DB-1 column (30 m × 0.25 mm × 1 μm) and analyzed by MS [28]. Finally, the concentration data of different species are obtained [29,30,31]. The sampling system adopts strict quality assurance and quality control procedures during the analysis process, meeting the requirements of the Specifications and Test Procedures for Ambient Air Quality Continuous Monitoring System with Gas Chromatography for Volatile Organic Compounds (HJ 1010-2018).

2.2. PMF

PMF is a multivariate statistical method for identifying and quantifying the sources of species based on the long time series of receptor site species component data [32], which has been widely applied to the source apportionment of VOCs in the atmosphere. The basic principle of the model is to use the weights to calculate the error of each volatile component in the sample, and to determine the main emission factors and their contributions by the least squares method [33]. The PMF model requires the input of pollutant concentration and uncertainty (U) data. If the species concentration is below the method detection limit (MDL), the 1/2 MDL is used as the input parameter for the species concentration, and the 5/6 MDL is the U [34]. If the species concentration is higher than the MDL, the corresponding U is calculated using the following equation:

$$\mathrm{U}=\sqrt{{(\mathrm{E}·\mathrm{c})}^2+{\mathrm{MDL}}^2},$$

(1)

where E is the error ratio; c is the concentration of the species involved in the PMF model.

2.3. OBM

The OBM is an observation-based chemical box model, which is widely used in atmospheric chemistry observation and simulation studies [35,36,37]. In this study, the OBM with a master chemical mechanism (MCM) was used to simulate the chemical processes in the atmosphere of Jincheng. Observed concentrations of pollutants (O₃, NO₂, CO, and VOCs), meteorological parameters (T, RH, and P), and photolysis rates were input into the model to calculate the production rates and loss rates of O₃ and to assess the sensitivity of the O₃ formation. The MCM (https://mcm.york.ac.uk/MCM; accessed on: 14 October 2024) provides a detailed chemical description of the atmospheric chemistry of 143 VOCs from emission to decomposition, containing 17,000 inorganic and organic reactions of about 6700 species, which can objectively reflect the actual atmospheric pollution conditions and the relationship between O₃ and its precursors [38].

2.3.1. O₃ Production Simulation

The O₃ production rate P(O₃) and the loss rate L(O₃) were calculated from the results of the OBM simulation, and P(O₃) can be quantified by the rate at which peroxyl radicals oxidize NO to NO₂ [39], which is calculated as follows:

$$\mathrm{P}\left({\mathrm{O}}_3\right)={\mathrm{K}}_{{\mathrm{HO}}_2+\mathrm{NO}}\left[{\mathrm{HO}}_2\right]\left[\mathrm{NO}\right]+{\mathrm{K}}_{{\mathrm{RO}}_2+\mathrm{NO}}\left[{\mathrm{RO}}_2\right]\left[\mathrm{NO}\right],$$

(2)

where [HO₂] and [RO₂] are the concentrations of HO₂ and RO₂ radicals, respectively. ${\mathrm{K}}_{{\mathrm{HO}}_2+\mathrm{NO}}$ and ${\mathrm{K}}_{{\mathrm{RO}}_2+\mathrm{NO}}$ are the reaction rates between the radicals and NO.

The L(O₃) was calculated from the reaction of O₃ photolysis reaction product O1D with H₂O, the reaction between OH and NO₂, and the reaction of O₃ with OH, HO₂, and alkenes, respectively.

$$\begin{gathered} \mathrm{L}\left({\mathrm{O}}_3\right)={\mathrm{K}}_{{\mathrm{O}}_3+\mathrm{OH}}\left[{\mathrm{O}}_3\right]\left[\mathrm{OH}\right]+{\mathrm{K}}_{{\mathrm{O}}_3+{\mathrm{HO}}_2}\left[{\mathrm{O}}_3\right]\left[{\mathrm{HO}}_2\right]+{\mathrm{K}}_{{\mathrm{O}}_3+\mathrm{alkenes}}\left[{\mathrm{O}}_3\right]\left[\mathrm{alkenes}\right] \\ +{\mathrm{K}}_{{\mathrm{NO}}_2+\mathrm{OH}}\left[{\mathrm{NO}}_2\right]\left[\mathrm{OH}\right]+{\mathrm{K}}_{\mathrm{O}1\mathrm{D}+{\mathrm{H}}_2\mathrm{O}}\left[\mathrm{O}1\mathrm{D}\right]\left[{\mathrm{H}}_2\mathrm{O}\right], \end{gathered}$$

(3)

Finally, Net(O₃) is the net O₃ production rate calculated from the difference between P(O₃) and L(O₃).

$$\mathrm{Net}\left({\mathrm{O}}_3\right)=\mathrm{P}\left({\mathrm{O}}_3\right) - \mathrm{L}\left({\mathrm{O}}_3\right)$$

(4)

2.3.2. RIR

Relative incremental reactivity (RIR) can quantitatively evaluate the impact of precursors on O₃ formation [18]. And in this study, the RIR was calculated based on the OBM to determine the sensitivity of different precursors to O₃ formation. The RIR values of different O₃ precursor fractions were calculated assuming 20% emission reduction, and the RIR was calculated as:

$$\mathrm{RIR}={\displaystyle \frac{\left[\mathrm{Net}\left(\mathrm{X}\right) - \mathrm{Net}\left(\mathrm{X} - \Delta \mathrm{X}\right)\right]/\mathrm{Net}\left(\mathrm{X}\right)}{\Delta \mathrm{S}\left(\mathrm{X}\right)/\mathrm{S}\left(\mathrm{X}\right)}},$$

(5)

where Net(X) is the net production rate of group X; Net(X − ΔX) is the net production rate after a change in emissions by ΔX; S(X) is the observed concentration of X; and ΔS(X) is the change in the observed concentration of X caused by the reduction. When the RIR is positive, it means that reducing X can lead to a decrease in O₃, and the larger the value, the more effective the cuts are in controlling O₃; when the RIR is negative, it means that reducing X will lead to an increase rather than a decrease in O₃.

The formula for calculating the average RIR value for each source [40] is as follows:

$$\stackrel{\_}{\mathrm{RIR}}={\displaystyle \frac{{\sum }_1^{\mathrm{n}}[\mathrm{RIR}\left(\mathrm{X}\right)\times \mathrm{Net}\left({\mathrm{O}}_3\right)]}{{\sum }_1^{\mathrm{n}}\mathrm{Net}\left({\mathrm{O}}_3\right)}},$$

(6)

$$\mathrm{contribution}\left(\mathrm{X}\right)={\displaystyle \frac{\stackrel{\_}{\mathrm{RIR}}\left(\mathrm{X}\right)\times \mathrm{con}\left(\mathrm{X}\right)}{{\sum }_1^{\mathrm{m}}[\stackrel{\_}{\mathrm{RIR}}\left(\mathrm{X}\right)\times \mathrm{con}\left(\mathrm{X}\right)]}}$$

(7)

where con(X) is the average concentration of source X originating from the PMF and m is the number of VOCs emission sources resolved by the PMF.

3. Results and Discussion

3.1. Characteristics of O₃ and Its Precursors

Through the analysis of the continuous sampling data in Jincheng during the summer, it was found that the data quality during the periods of OE1 and OE2 was relatively good. We sorted the daily maximum concentration of O₃-8H during the research period from small to large and took the value at the 90th percentile as the concentration of the 90th percentile of O₃ (O₃-8H_90%). The O₃-8H_90% concentrations were 111 ppb and 98 ppb during OE1 and OE2, which can reflect the highest O₃ pollution levels throughout the years 2022 and 2024 in Jincheng. Therefore, OE1 and OE2 were selected as representatives of the typical O₃ pollution episodes in Jincheng. Figure 2 shows the comparison of O₃ and its precursor concentrations during the observation period. Overall, the concentrations of NO₂ and VOCs components have increased from OE1 to OE2. The average concentration of NO₂ in OE2 (12.44 ± 1.48 ppb) during the period of increased industrial production is slightly higher than that in OE1 (10.28 ± 3.04 ppb). The average concentrations of TVOCs in Jincheng in OE1 and OE2 were 15.98 ± 6.52 ppb and 19.55 ± 7.54 ppb, which were lower than the concentrations of TVOCs in megacities, such as Nanjing [41], Zhengzhou [42], and Guangzhou [43], and slightly higher than those in the industrial areas of cities such as Linfen [44], Dalian [45], and Tianjin [46].

From the perspective of VOCs components, the concentrations of alkanes, alkenes, aromatics, and OVOCs in OE2 have all increased. The species with the largest concentration difference is propane (0.92 ppb). The emissions of aromatics and alkenes, such as m/p-xylene, acetaldehyde, cis-2-pentene, toluene, 1-hexene, and propylene, have increased in the context of the increased industrial production. Propane and propylene often come from combustion processes in industrial production [47], while m/p-xylene, acetaldehyde, cis-2-pentene, and 1-hexene are the main emission species in coal chemical, steel, and coking industries [48,49,50].

3.2. Simulation of Atmospheric Photochemical Reaction Process

3.2.1. O₃ Budget Analysis

Figure 3 demonstrates the diurnal variation in O₃ production rates and loss rates on polluted days. The Net (O₃) is determined by the difference between the production rates and loss rates. O1D+H₂O, O₃+OH, O₃+HO₂, NO₂+OH, and O₃+Alkenes were the main pathways of O₃ loss, in which O₃+HO₂·accounted for a larger share of the total loss pathways (39.74%, 40.55%), with the reaction rates of 1.95 ppb/h and 1.54 ppb/h, respectively.

O₃ is mainly produced through the reaction of NO with peroxyl radicals (HO₂·and RO₂·), and the O₃ production rate can be calculated as the sum of HO₂+NO and RO₂+NO, in which RO₂+NO dominated the production of O₃, accounting for over 50% of the total O₃ production rate, with peak rates of 14.56 ppb/h and 17.15 ppb/h. Higher reaction rates also mean that O₃ cannot preferentially react with NO to eliminate it, leading to the accumulation of O₃. The average rate of the daytime (6:00–18:00) Net (O₃) of OE2 is 9.24 ppb/h, which is 25.2% higher than that of OE1 (6.91 ppb/h). The higher Net (O₃) in Jincheng brought about severe O₃ pollution, which is related to the high concentrations of NO, HO₂·, and RO₂ in the atmosphere, and thus mitigating O₃ pollution requires focusing on the production of NO and peroxyl radicals.

3.2.2. Free Radical Cycle

In the OH-HO₂-RO-RO₂·radical cycling reaction pathway (Figure 4), the free radical budget and cycling reaction process during the OE1 and OE2 periods were comparatively demonstrated. The OH primary sources included H₂O₂+hv, O1D+H₂O, and HONO+hv, among which O1D+H₂O accounted for 2.15 ppb/h and 1.51 ppb/h. The reaction with NO and NO₂ to produce HONO and HNO₃ was the cycling reaction process to terminate OH. The primary source of HO₂· was mainly the photolysis of HCHO and OVOCs, with reaction rates of 0.28 ppb/h and 0.20 ppb/h and 0.28 ppb/h and 0.31 ppb/h, respectively, for different conditions. And the termination reaction of HO₂· was HO₂+HO₂ (1.74 ppb/h, 1.22 ppb/h) and HO₂+RO₂· (1.74 ppb/h, 1.59 ppb/h), respectively.

In the ROx cycle reaction process, the generation of HO₂· mainly occurs through the reaction of RO+O₂ and OH+HCHO pathways, and the termination process is mainly through the reaction with HO₂·and RO₂· to generate H₂O₂ and ROOH. The generation of OH· was mainly through the reaction of NO+HO₂, and the reaction rate of OE2 (7.50 ppb/h) was 10.5% higher than that of OE1 (6.71 ppb/h), which was one of the important pathways for O₃ formation. The loss process of OH· was mainly through the reaction with VOCs to generate RO₂. This reaction is the beginning of the participation of VOCs in the free radical cycling process, and it is also an important pathway for the generation of RO₂, a participant in the O₃ production reaction. Alkenes with higher reactivity contribute more to the reaction of OH+VOCs. Compared with OE1, cis-2-butene, cis-2-pentene, and propylene, which have increased reaction rates in OE2, are also species with significant concentration changes. RO₂+NO is an important reaction in free radical cycling reactions and one of the dominant reactions for O₃ formation. The daytime average reaction rate of this reaction showed a small difference between the two typical O₃ pollution episodes, but the peak reaction rate of OE2 (17.15 ppb/h) was significantly higher than that of OE1 (14.56 ppb/h). In general, the alkenes with larger reactivity in the reaction pathway of the free radical cycle are the focus of control. And these species not only play an important role in accelerating the free radical cycle but also promote the production of O₃ to cause pollution.

3.2.3. Sensitivity Analysis of O₃ Formation

Figure 5 demonstrates the sensitivity results of the O₃ generation to NOx, VOCs, and VOCs components (alkanes, alkenes, aromatics, alkynes, and OVOCs). The RIR values of each species component of VOCs and NOx are positive. Jincheng is in the O₃-VOCs-NOx transition area, and the reduction in both VOCs and NOx can alleviate O₃ pollution. Comparing OE1 to OE2 after the increase in industrial production, the RIR for VOCs increases from 0.76 to 0.84, and the RIR for alkenes increases from 0.42 to 0.63. NOx are important precursors that affect the O₃ formation in both OE1 and OE2, with RIR values of 0.68 and 0.51, indicating strong sensitivity. The sources of NOx in industrial cities are complex, and in addition to emissions from vehicles and coal combustion, heavy industries represented by coking, iron and steel, cement, and other industries also emit large amounts of NOx [10,51]. And the comparison with industrial cities, such as Changzhi (0.35) [52], Zibo (0.4) [53], and Dongying (0.37) [54], shows that NOx in Jincheng play a more prominent role in the O₃ formation process. Therefore, the control of O₃ precursors in Jincheng should be aimed at the effective reduction in alkenes and NOx, and the prevention and control of NOx cannot be ignored.

3.3. Source Apportionment of O₃ and Its Precursors

3.3.1. VOCs and NOx Source Apportionment

In this study, the PMF5.0 model was used to analyze the sources of O₃ precursors and their contributions. A total of 43 species were selected to participate in the PMF simulation, including 16 alkanes, 8 alkenes, 1 alkyne, 13 aromatics, 4 OVOCs, and NOx. The six factors were identified as the industrial process, biogenic sources, coal combustion, solvent utilization, diesel vehicle emission, and gasoline vehicle emission. Aromatics, such as m/p-xylene, p-ethyltoluene, and 1,2,3-trimethyltoluene, had high contributions in Factor 1, and these species are related to the solvent utilization in paints and coatings. Therefore, Factor 1 was recognized as solvent utilization [49]. In Factor 2, n-heptane, n-octane, ethylbenzene, other high-carbon chain alkanes, aromatics, and NOx are the main pollutants emitted from diesel vehicles [55]. Therefore, Factor 2 was recognized as diesel vehicle emissions. In addition to the high contribution of C2-C5 low carbon chain alkanes, cis-2-butene, trans-2-pentene, acetaldehyde, and propanal in Factor 3, we identified this factor as industrial processes since these substances are emitted from production processes in the coal chemical and steel industries [48,50], thus identifying this factor as industrial processes. Ethane and propane, which are large contributors to Factor 4, are often from natural gas and LPG volatilization and combustion sources [56], and acetylene and ethene are characteristic species of coal combustion processes [57], so Factor 4 was identified as coal combustion. Factor 5 was identified as gasoline vehicle emissions; C4-C6 alkanes, such as isopentane, toluene, and acetone, are species emitted in large quantities from gasoline vehicles [58,59], and MTBE is also a typical emission tracer from gasoline vehicles [60]. The contribution of isoprene emitted by plants in Factor 6 was 82.9%, so the factor was judged to be biogenic sources [61].

Figure 6 shows the results of the contribution of the six factors identified by the model to VOCs and NOx. During OE2, industrial processes contributed the largest proportion to VOCs (33%), which was significantly higher than that of OE1 (20%), followed by coal combustion (29%), gasoline vehicle emissions (18%), biogenic sources (7%), solvent utilization (7%), and diesel vehicle emissions (4%). Gasoline vehicle emissions and coal combustion were significant contributors of VOCs during OE1. Among the NOx sources, diesel vehicle emissions contributed the largest proportion (53%, 45%), followed by coal combustion (26%, 28%), industrial processes (12%, 22%), and gasoline vehicle emissions (8%, 5%). After the increase in industrial production, its contribution to atmospheric VOCs and NOx was also significantly elevated. In addition, the contribution of diesel vehicles in Jincheng is also more prominent. Coal is the dominant industry in this industrial city, which has a greater demand for road transportation. Combined with the characteristics of high-speed roads around the city, diesel vehicles have a greater impact on the atmospheric environment in Jincheng. Therefore, in addition to pollutant emissions from industrial companies, attention should also be paid to the impact of diesel vehicles on O₃ precursors in Jincheng.

3.3.2. O₃ Source Apportionment

Combined with the O₃ formation sensitivity results obtained in Section 3.2.3 for Jincheng, which is in the transition control area of VOCs and NOx, NOx are also quite important for O₃ formation. It is necessary for this study to carry out the O₃ source apportionment by considering VOCs and NOx together.

It has also been noted that for industrial cities in transition areas, NOx are very important in O₃ formation and source apportionment. Only considering VOCs and ignoring NOx in the analysis process will lead to an inaccurate O₃ source apportionment [62]. The results are shown in Figure 7, and diesel vehicle emissions are the first contributor to O₃ formation in both OE1 and OE2. The contribution of industrial processes in OE2 (30.8%) is significantly higher than that in OE1 (15.2%), which is mainly due to the increase in industrial production and the consequent increase in the emissions of highly reactive VOCs species, which have a greater impact on O₃ formation. The prominence of industrial processes and diesel vehicle emissions on VOCs and NOx in the comparison scenarios is also derived in Section 3.3.1. In general, diesel vehicle emissions and industrial processes were the focus of controlling O₃ and its precursors in Jincheng.

4. Conclusions

This study comparatively analyzed the changes in atmospheric O₃ precursors in a typical industrial city within the background of increasing industrial production in 2022 and 2024 and explored the changes in the atmospheric O₃ formation mechanism and source apportionment using the OBM. By comparison, it was found that in the context of increasing industrial production, the concentrations of NOx and VOCs in 2024 have increased by 21.1% and 22.3%, respectively. Among them, the concentrations of cis-2-pentene, 1-hexene, and propylene related to industrial processes increased greatly. RO+NO is the main free radical chemical reaction pathway for atmospheric O₃ formation, accounting for 51.5~54.2% of the total O₃ formation. The reaction of VOCs with OH· is an important pathway for precursors to participate in the radical cycle to generate RO₂·, with the peak reaction rate increasing from 7.58 ppb/h to 12.82 ppb/h, in which cis-2-pentene played an important role, and the daytime average reaction rate increased from 0.01 ppb/h to 0.18 ppb/h. VOCs and NOx both play important roles in the O₃ formation process, and the sensitivity of VOCs increased from 0.76 to 0.84, with alkenes increasing most significantly (0.42~0.63). These results indicate that alkenes play an important role in the formation of O₃ in the context of increased industrial activity. The comparative results of the precursors source apportionment indicated that the contribution of industrial processes to VOCs increased from 20% in 2022 to 33% in 2024. Diesel vehicles, the main transportation tool for industrial products, have also demonstrated an important emission contribution to NOx driven by industrial activities. Under the influence of industrial activities on precursors emissions, the contribution of industrial processes to O₃ formation has increased from 15.2% to 30.8%, and diesel vehicle emissions also contribute significantly to O₃ formation (30.6~34.8%). On the whole, the increase in the industrial activity level has led to a significant increase in alkenes emissions, which has a significant impact on the mechanism and sources of urban O₃ formation. Controlling the emission of alkenes from the industrial process is the direction for the continuous control of O₃ pollution in industrial cities, which is of great significance for reducing atmospheric O₃ pollution.