Assessment of the risk of exposure to Air pollutants and identifying the affecting factors on making pollution by PCA, CFA

ABSTRACT Today, one of the most important problems for humans is disease and mortality due to air pollutants, especially in large and industrial cities. This research is to evaluate the health risk of children and adults’ exposure to heavy metals and PAHs in the air of cities of Arak, Isfahan, Ahvaz, Tabriz, Shiraz, Karaj and Mashhad and identifying the factors affecting pollution, based on sampling data from 120 stations, was done during 2019–2020. ICP OES and GC/MS were used to determine the amount of heavy metals and PAHs, respectively. US-EPA models were used for carcinogenic and non-carcinogenic risk assessment and Factor analysis models, PCA, CFA with SPSS and LISREL software were used to identify factors affecting contamination. The risk index for children and adults, except in the cities of Arak and Isfahan, was obtained within the allowable range and the carcinogenicity and non-carcinogenicity risk of all metals, except Mn in Arak and Isfahan for children and adults was seen below the threshold. The carcinogenic risk of all PAHs was also obtained to an acceptable extent. During the statistical analysis, combustion of fuels, vehicle exhaust, tyre wear and industrial activities were identified as the most important factors in causing pollution.


Introduction
Today, with the increase in industrial activities and overuse of fossil fuels, human exposure to dangerous air pollutants has increased [1][2][3][4].Air pollutants contain very complex mixtures, such as gases, particles, liquid droplets or mixtures of them [5][6][7][8][9][10][11].These pollutants have noticeable negative effects on public health and the environment, which is one of the global problems [12][13][14][15][16][17][18][19][20][21].Heavy metals are one of the most important air pollutants that have been bound to air particles and would enter the body easily by inhalation.Since these mixtures have non-degradable properties, high toxicity and adverse health effects, they have been considered by many researchers in recent years [22][23][24][25][26][27][28][29][30][31][32].According to the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA) list, rare elements were introduced as the most harmful priority pollutants in 2007 [33].Studies show that the accumulation of these mixtures in adipose tissue, bones and muscles would affect the normal functioning of body organs.Therefore, exposure to them, even to a very small extent, can have detrimental effects [34][35][36][37][38][39][40][41][42][43][44][45][46][47][48][49][50][51][52].For example, zinc (Zn) can affect the nervous system and cause reducing the consciousness [53].Another air particulate metal that causes an increase in lung cancer, according to the World Health Organization (WHO), is arsenic (As) [54].In addition, exposure to it can lead to skin problems, especially skin cancer [55].Lead (Pb) also can have negative effects on internal organs including the reproductive system, nervous system, kidneys and blood enzymes [23,54,[56][57][58].Research has shown that people who were exposed to nickel (Ni) are more likely to suffer from cardiovascular disease [59].Based on reports, Mercury (Hg) also causes dysfunction to children's intelligence and causes heart disease in them [60].Iron (Fe) is one of the rare elements that is needed daily for the body, but is harmful to it in large amounts [61,62].Also, a large amount of copper (Cu) can have problems such as; neurological disorders, mutation and, most importantly, increase the resistance of cancer cells to medicines [63,64].Cadmium (Cd), in addition to carcinogenicity, can also cause serious damage to the nervous system, cardiovascular system, kidneys and bones [25,65].
Studies have been done about assessment of the health risk of air pollutants in cities of Iran, but despite the announcement of the World Health Organization due to high levels of air pollutants, Iran's metropolises have been named as the most polluted cities in the world [76] and identifying factors affecting air pollution is essential for the protection of human health and the environment [77].So far, no comprehensive study has evaluated the health risk of human exposure to heavy metals and PAHs and identifying the factors affecting air pollution in seven large industrial cities of Iran.Therefore, assessing the health risk of exposure to heavy metals and PAHs and identifying the factors that cause pollution for the population of these somewhat polluted areas is very important.Preparing controlling programmes and assessing the health risks of air pollutants is one of the most logical ways to improve air quality, especially in industrial and polluted metropolises.Such research encourages policymakers and officials to codify and enforce appropriate laws.Therefore, considering the sensitivity of the issue, the purposes of this study are 1) to assess the health risk of exposure to heavy metals and PAHs and 2) to identify the factors affecting air pollution in seven large cities of Iran.

Data collection
In this study, the amounts of 11 heavy metals (Pb, Mn, Al, Hg, Cu, Fe, Ni, As, Cd, Zn and Cr) and 16 samples (PAH, Nap, Acy, Ace, Flu, Pyr, Anth, Phen, Chr, Flrt, BaA, Bbf, Bkf, BaP IND, DBahA and BghiP) were analysed from 120 stations in the air of 7 industrial cities of Iran during 2019-2020.The study areas and sampling points are shown in Figure 1.Sampling was performed during 24 hours using an Aircheck pump (skc, model 44xr) containing Personal Modular Impactor (PMI2.5-model225,351, SKC) and 5 L/m constant flow rate.Air samples were collected with a filter (mm in diameter and 1 μ in pore size 37) Poly Tetra Fluoro Ethylene (PTFE) (USEPA 2016).The filters were placed in a desiccator at 23 ± 1°C and 40 ± 5% relative humidity for 48 hours and measured by electronic microbalance with a sensitivity level of 0.0001 mg (Mettler Toledo AB204 N).To make the variance of the measurements less than 10 µg, all filters were weighed 3 times before and after sampling.The weight difference between the primary filter and the PM-loaded filter divided by the total passed air volume was obtained during the sampling of the air pollutant density.The filters were divided into two parts and were used for the analysis of metals and PAHs.The Inductively Coupled Plasma Optical Emission Spectrometer (ICP -OES) was used in order to determine heavy metals.In such a way that in the quartz vial, these filters were placed with 4cc HNO3 (65-68%) and 1cc HCl (36-38%).Then, the vials were then placed on hot water at 90°C for 2 hours for complete digestion [78].After lowering the temperature, the solution was passed through a 0.45 μm syringe filter and made up to 25 cc with distiled water.Eventually, the analysis was performed with the ICP-OES device (PerkinElmer instrument model optimum 8000 (Sheltin, USA) [79].To analyse the PAHs, PM-loaded filters were placed in a 25 ml beaker and 4 ml of acetonitrile was added to them, and then the caps were closed with parafilm and immersed in ultrasonic for half an hour by NIOSH 5515method.Eventually, the solution was poured into a suitable vial using a syringe and was analysed by GC-MS (Agilent model 7890B, Agilent-MS 5975B, Model [H1] EI).Chromatographic columns HP5-MS (30 m, 0.25 mm, 0.25 um) and helium gas were used as carriers with a constant flow rate of 2.5 mL/min.The methods of analysis of the samples and devices used in this research are appropriate and are used in many cases [78].

Health risk assessment model
Health risk assessment is a method to estimate the effect of exposure to a specific pollutant through three paths: 1) swallowing; 2) inhalation and 3) skin absorption [69].Because particles smaller than 2.5 microns are very small, therefore, swallowing pathways and skin can be ignored [79].Therefore, during this research, the health risk of carcinogenic and non-carcinogenicity due to exposure to heavy metals and PAHs through inhalation (mouth and nose) was estimated by the U.S. (EPA) Environmental Protection Agency's assessment model [69,80].
The process of assessing the health risk of heavy metals consists of the two following steps: (1) Carcinogenic risk assessment The carcinogenic risk was obtained according to equation (1)(2)(3).
The International Agency for Research on Cancer (IARC) has classified As, Ni, Cd and Cr as human carcinogenesis by inhalation in Group 1 [80].Pb has also been suggested as a potential carcinogen in Group 2A.
(2) Non-carcinogenic risk assessment Non-carcinogenic risk was obtained by calculating hazard quotient (HQ) according to Equation (4) [81].Here, D is the dose contacted of HM (mg/kg-day), and RfD is the reference dose of heavy metals for Cr ( [54,83]. If it is (HQ>1) it means that there is a risk of non-carcinogenicity and if (HQ < 1) the risk of non-carcinogenic risk is unexpected [69,80].
To estimate the overall non-carcinogenic risk, the HI risk index was calculated according to Equation ( 5) [8].
The health risk assessment process of PAH compounds consists of the following two steps: (1) Carcinogenic Potency of PAHs (BaPequi) The carcinogenic potential of air PAHs as benzo (a) pyrene equivalence (BaPequi) according to their TEFs was estimated according to Equation (6).
(2) Incremental Lifetime Cancer Risk (ILCR) The USEPA standard model was used to calculate the health risk for both groups of children and adults exposed to air PAHs [86].ILCR was estimated by inhalation using Equation ( 7) [86] Here ILCR incremental lifetime cancer risk; CS concentration of the PAH compounds in air based on toxic equivalent of BaP using the TEF; CSF inhalation carcinogenic slope factor (3.85 mg/kg/day) [86]; BW body weight (15 kg for children and 70 kg for adults) (USEPA, 1989); AT Averaging time (70 years or 25,550 days); EF exposure frequency (365 days/year for adults and children; ED exposure duration (years) (24 years for adults and 6 years for children; IR Inhalation inhalation rate (20 m 3 /day for adults and 10 m 3 /day for children); PEF particle emission factor 1.36 × 10 9 m 3 /kg [86] .

Statistical and software analysis
Principal component analysis (PCA) and confirmatory factor analysis (CFA) were used to model air pollutants.PCA and CFA are one of the multivariate analysis methods [85], which are used to reduce the amount of data and identify hidden factors in causing pollution [84,85,87].In PCA model, Bartlett's Test and KMO index were used to determine the appropriateness of the analysis factor for the data.The amount of KMO index varies between 0 and 1, if the amount of KMO is more than 0.8, it is appropriate to use factor analysis [88].The RMSEA index was used to fit the CFA model.According to the structural equation model, if the RMSEA's amount is less than 0.05, the model has a good fit, if it is between 0.05 and 0.08, the model is acceptable, if it is between 0.08 and 0.1, the fit is moderate and if it is greater than 0.1, the fit is poor.GFI, CFI and IFI indices were also used in this model.The amount of these indicators is between 0 and 1.An amount greater than 0.9 indicates a well-fitting model.Another indicator used was RMR.An amount of less than 0.5 of this index indicates an excellent fit [89].All of the data were analysed using the SPSS 26.0 and LISREL 8.8 software and a significance level of p < 0.05.
Spatial distribution of heavy metals and PAHs was performed using ArcGIS10.8software.The levels of pollutants in the studied cities were significantly different from each other.Amplitude of changes of heavy metals and PAHs in all studied cities as described as Cr (0.03-160.25),Zn (0.01-874.64),Cd (0.01-339.69),As (ND-111.96),Ni (ND −874.40),Fe (0.01-1256.64),Cu (0.01-691.97),Hg (ND −118.58),Al (56.94-4044.52),Mn (0-354.6),Pb (12.36-885.89)and Nap (0-6.49),Acy (0-5.49),Ace (0.01-6.58),Flu (0-5.69),phen (0.01-5.49),Anth (0-4.59),Pyr (0-6.58),Flrt (0.01-5.64),Chr (0-5.97),BaA (0-6.47),Bbf (0-6.49),Bkf (0.01-5.82),BaP (0.07-7.53),DBahA (0.01-5.67),IND (0.01-4.34),BghiP (0.01-4.61) ng/ m 3 .Al and BaP They had the highest levels among heavy metals and PAHs, respectively, therefore it can be said that these pollutants were observed in most stations.The main reasons for the high level of some heavy metals and PAHs are the existence of steel industries, petrochemical, oil and gas, dust and vehicles industries [69,81,90].The lowest amount of heavy metals and PAHs was related to Shiraz.Decreased industrial activity and traffic congestion may be involved herein.Pb levels in Mashhad were lower than the other cities, and according to Al-Masri et al.'s (2006) and Kemp's (2002) studies, it could be related to fewer vehicle in this city.However, Pb concentration was high in Shiraz; It seems similar to the research of Najafi et al, Latif et al.And Kamani et al.The use of leaded fuels is one of the main reasons.The results of this study showed that Pb and Cd had the highest amount in Shiraz and Arak, respectively.Zhang et al also reported the highest metal levels in Pb, Cd, in their study in Hubei, China [91].As metal was another pollutant in the air of the studied cities.We suspect emissions from coal mills are due to the presence of this metal [78].Pb, Cd and As were observed in Mashhad and Shiraz, respectively, much lower than the National Ambient Air Quality Standards (NAAQS), (500 ng m -3 , 5 ng m -3 and 6 ng m -3 ), respectively.While the amount of As in the cities of Tabriz, Ahvaz and Karaj was higher than the standard due to the high use of agricultural pesticides, metal smelting and fuel burning, which was consistent with the report of Wang et al [83].During this study, the amount of Cd and Zn in Arak city was significantly high, which can be similar to the research conducted by Hou et al, due to metal extraction and smelting in Daye city [91].Also, the concentrations of Cu, Cd and Pb in Arak are similar to the research of Hou et al in Daye city; was high due to the presence of ferrous and non ferrous minerals [77].What was considered in this research was the presence of metals such as Fe and Al as air pollutants; re-suspension of soil and local road dust may be the main reasons for their presence, which was matched according to the results of research Hasheminassab et al.The existence of energy conversion industries, refineries and power plants can be considered as the reasons for the presence of three metals Fe, Cu and Pb in urban air, which was similar to the Naimabadi et al and Sorooshian et al' study.In the present study, the amount of Mn in Isfahan and Arak is much higher than the threshold provided by WHO (0.15 μg m -3 ), Wuhan (0.048-0.15 μg m -3 ), Nanjing (0.015-0.032 μg m -3 ), Spain (0.004-0.012 μg m -3 ), Shanghai (0.07-0.19 μg m -3 ) was observed [91].We believe that the existence of a mining plant in these areas could be one of the main reasons, according to a 2019 study by Zhang et al.Ni level in Arak was very high due to the use of gas and fuel oil by the steel plant, which was similar to the results of the Filby Branthaver study [8].While Neisi et al expressed that the iron is the cause of Ni high level in S1 region, during their research in Ahvaz [8].Another result of this study was that the concentration of PAHs in all cities except Ahvaz and Mashhad was lower than the research conducted by Ma et al, in Hong Kong (0.4 −16.3 ng/m 3 (and Wang et al in Shanghai (6.5-7.2 ng/m 3 ) [78].The highest and lowest levels of PAH were related to BaP and Nap, Acy, Flu, Anth, Pyr, Chr, BaA, Bbf in Mashhad and Shiraz, respectively.The high level of PAHs in Ahvaz compared to other cities can be related to the increase in vehicle emissions and fossil fuels, the activity of industries and power plants, population density and coal burning in it, that was matched by the study result of Mon et a, Qishlaqi & Beiramali, Kumar et al, Nguyen et a, Wang et al and Shabbaj et al [85,86,92] Reduction of traffic density, decrease of human activities and suitable weather conditions in Shiraz can be attributed to the low levels of PAHs, that was similar according to the study results of Gope et al in India, Wang et al in China, Guangzhou and Shabbaj et al in Saudi Coastal [85,86,93].The amount of Nap in Ahvaz city was high due to vehicle exhaust, diesel exhaust which was consistent with the results of the report of Pengchai et al [85].In Ahvaz, the amount of Flrt and Pyr was higher than other cities studied, which seems to be the main reasons for the increase in industrial activities and combustion of fuels, as in the Qishlaqi & Beiramali study [85].BghiP level in Karaj was higher than other places, that it seems the abundance of vehicles is probably one of the reasons as stated by Khairy et al [94].The levels of carcinogenic PAHs BaA, BaP, Chr, B (b) F, B (k) F, IND and DBahA were higher in Ahvaz than the rest.We think that the increase in the activity of industries and refineries is one of the possible causes.Other results of this study was higher levels of PAHs with 4-6 rings (Pyr, Chr, Flrt, BaA, Bbf, Bkf, BaP IND, DBahA BghiP) compared to 2-3 rings (Nap, Acy, Ace, Flu, phen, Anth) in Shiraz, which was consistent with the results of a study by Kumar et al in Jamshedpur, India, which is mostly due to their presence in the particulate phase and is usually emitted from automobile fuels [94].While 2-3 ring PAHs are present in the gas phase and are mostly released from coal combustion [95].While Bai et al in their study in China reported high concentrations of PAHs with lower rings [94].Overall, the average PAHs were 1.90 ng/m 3 , that was lower than Beijing, China 28.5, Tehran, Iran 44.2, Lucknow, India 54.23, Semiurban Jamshedpur, India 87.34, Rural Jamshedpur, India 73.33 and was below Shenzhen, China 128, Urban Jamshedpur, India 190.2 ng/m 3 [95].

Concentrations and spatial distributions of heavy metals and PAHs in air
The spatial distribution of heavy metals and PAHs in the air of the mentioned cities using ArcGIS10.8software is shown in Figure 2. Due to the fact that most of these pollutants are emitted from the industrial activities, refineries and fuel combustion, their concentration is expected to be higher in cities with more petrochemical industries and complexes and more vehicle traffic.The highest average of total heavy metals, and PAH was distributed in Ahvaz and the lowest in Mashhad.In general, it was observed that the level of pollutants decreases from northeast to southwest.Therefore, southwestern Iran should be given more attention to control air pollution.The results of this study reflect the high level of air pollution in some intended cities.These seven cities are among the industrial cities of Iran.It seems that increasing fuel and energy consumption, increasing industrial activities, greenhouse gas emissions, large population and hydrological characteristics of the region are involved in different levels of pollutants in the studied cities [82,83].

Human health risk assessment of heavy metals and PAHs
One of the best options for determining the potential risk of exposure to air pollutants is a health risk assessment that was conducted in this research for heavy metals and PAHs.The results of carcinogenic and non-carcinogenic risk assessment of rare elements and PAH compounds in the air of 7 important industrial cities of Iran for children and adults are presented in Tables S1 and Tables S2.Among the studied metals, Cr, Cd, As, Ni and Pb are carcinogens and the rest are non-carcinogenic [78].Although some elements such as Fe, Zn and Cu are beneficial for the body; their presence in polluted air can cause serious health problems [62,96,97] PAHs are other air pollutants that can easily enter the lungs through inhalation of polluted air and dust, so they are dangerous to the health of residents of polluted cities.In addition, research has shown that PAHs can accumulate in the liver, kidneys, ovaries, and spleen and cause problems once they enter the body [74].If 0.04 mg/kg of fluoranthene, 0.3 mg/kg of anthracene, 0.03 mg/kg of pyrene, 0.06 mg/kg of acenaphthene and 0.04 mg/kg of fluorene of human body weight enter the body daily, according to the EPA, there's no harm to the health [90,92].In the present study, the HQ of all metals except Mn in Arak and Isfahan was less than 1 for children and adults and it indicates that there are no non-carcinogenic health effects.The highest and lowest HQ belonged to Mn and Zn in Arak and Shiraz, respectively.Chabukdhara and Nema stated in their study that Cr and Cd have the maximum and minimum noncarcinogenic risk for children and adults, respectively [92].Traczyk and Gruszecka-Kosowska showed in their study that Mn had the highest and Sn the lowest noncarcinogenic risk for exposed individuals [98].In Arak city HQ Cd (1.17 × 10 −2 ) was higher than all elements except Cr and Mn.As a result of our research, during the study of Gholizadeh et al, HQ Cd (1.55 × 10 −1 ) was obtained more than other elements in Yazd [92].Kidney disease is one of the most common complications of exposure to Cd metal [82].Another result of this study was the high HQ Cr in Arak compared to other cities.We think that similar to the claim of Chen et al. in Shanghai, the increase in the activity of steel and iron industries is its reason [55].Maleki et al. also stated in their study in Tehran that mechanical wear, natural resources, friction processes of vehicles and mechanical wear were the reasons for the high non-carcinogenic risk of Cr [99].Prolonged exposure to Cr can have adverse health effects such as lung cancer and hepatocellular carcinoma [100] The risk of non-carcinogenic Mn in Arak and Isfahan was higher than the standard.According to Zhang et al. in Hubei, the continuation of manganese ore extraction in these cities seems to be one of the main reasons [91].In this study, it was discovered that the level of non-carcinogenic risk of Ni in all cities was within acceptable limits.Contrary to this conclusion, Chen et al, during their study in China, cited increased industrial activity as one of the main reasons for the high levels of non-carcinogenic Ni [55].Another result of this study was the high level of HQ Pb in Shiraz due to the presence of ferrous and non-ferrous minerals compared to Cu, Zn and Cd, which was consistent with the research of Hou et al, in China [77].Goudarzi et al and Zheng et al also reported a higher non-carcinogenic risk of Pb than other metals in their research [81].Neurological problems and developmental disorders in children can be considered as health problems of exposure to Pb [82].The non-carcinogenic risk of Cr and Mn in children in Arak was higher than 0.1, so according to Chen et al, exposure to them can lead to neurological problems and developmental disorders in children [55].HQ Hg had the highest amount in Ahvaz.Inhalation of mercury vapours causes serious health problems such as damage to the nervous system, gastrointestinal tract and kidneys [82].The HI amount of Cu, Zn and Pb metals were obtained within the allowable range.The reduction in the number of motor vehicles seems to be the main reason for this, which can be confirmed by the results of the study of Chen et al; they observed that the metals were lower at the BH site than at the BTH, YRD and PRD sites due to the smaller number of HI vehicles [55].In general, except for the cities of Arak, Isfahan and Ahvaz, hazard indexes (HI) of metals for children and adults were obtained within acceptable limits and indicate that their inhalation has no side effects on human health.In a similar study, Zhou et al in Hefei, Chin, reported HI levels of metals in the safe range for children and adults.Chen et al's study in Shanghai, China, also discovered low levels of heavy metals HI for children and adults are below safe levels [55].In this study, the total non-carcinogenic risk of metal inhalation for children and adults was 2.26 and 6.63, respectively.Similar to the results of our study, Traczyk and Gruszecka Kosowska in Kraków, Poland reported a total noncarcinogenic risk of 26.4 and 15.0 in children and adults, respectively, which was higher than safe [98].
During this study, the risk of Cd carcinogenesis in all cities except Shiraz was estimated within the allowable range, which contradicted the results of research by Gholizadeh et al. [84].Cr carcinogenic risk was also reported to be acceptable in all cities.Contrary to the results of this study, Zhou et al. estimated the level of Cr carcinogenicity to be higher than 10 −6 during their study in Hefei [101].Gastrointestinal problems, disorders of the body's immune system and respiratory system, genetic changes and skin disorders are the most obvious problems of Cr on human health [83].During this study, the risk of Pb carcinogenesis in all cities except Arak was estimated to be lower than the allowable limit.Contrary to the results of the present study, a study conducted in Yazd reported a safe risk of Pb carcinogenesis [84].The risk of carcinogenesis in children and adults was observed to be safe for all cities.The total carcinogenic risk of metal inhalation for children and adults was in the range of 1.0 × 10 −4 to 1.0 × 10 −6 and indicates the acceptable risk of carcinogenesis.As in the present study, Wu et al., reported their overall carcinogenic risk for children and adults during their research in Ningbo, China, 6.24 × 10 -6 , 2.50 × 10 -5 , respectively, which was within acceptable limits [80].Also, in another research, Traczyk and Gruszecka Kosowska in Kraków, Poland, stated that the total carcinogenic risk for children and adults was 1.77 × 10 −4 and 1.51 × 10 −4 , respectively [98].In general, the study found that risk levels were higher in adults than in children.According to our findings, Wu et al., in Ningbo, China, reported that the risk of cancer was higher in adults than in children [80] that the risk in children was higher due to higher sensitivity than adults, which contradicted the results of this study [55,101].
BaP is known as the most toxic and mutagenic PAHs [69,86].The risk calculation of PAHs was performed by BaP (BaPeq) method.To determine BaPeq (TEQ), toxic equivalence factors (TEFs) were multiplied in each PAH component [69].The ILCR assesses the carcinogenic risk of exposure to PAH compounds in the air [69,93].The highest and lowest ILCRs for children and adults were 7.95E-07 and 9.21E-12, respectively, which were related to pah Bap and Flrt in Shiraz.According to EPA ILCR reports, between 0.0001 and 0.000001 are in the acceptable range [69], above 0.0001 is high risk and below 0.000001 is carcinogenic risk [102].In this study, ILCR PAHs in all cities were lower than the allowable limit.As in the present study, Liu et al. in China reported that ILCR is below the EPA limit for people confronted with PAHs [69].Contrary to the research of Goudarzi et al., in Ahvaz, who reported a higher risk of carcinogenicity of PAHs than the EPA standard [69].Also, in another study by Tsai et al., ILCR expressed PAH compounds for carbon black manufacturing workers in the range (0.00435) above the EPA limit [69].We suspect that PAH compounds do not pose a threat to human health, and it contradicted to the results of the research by Goudarzi et al. and Balcıoğlu et al. [69].
In general, the results showed that the total level of carcinogenic risk of PAHs in the present study for children and adults except in Arak, Ahvaz and Mashhad was lower than the standard and with the results of research conducted in Xuzhou, China (4.12E −06 and 3.98E-06 for children and adults, respectively) [90], at Ghaziabad, India (1E-06 and 1E-04 for children and adults, respectively) [90], at Tianjin, China (2.55E-06 and 9.33E-06 for children and adults, respectively) [90], at Ghana (2.6 E-07 and 3.6E-0406 for children and adults, respectively) [90], at Jamshedpur, India (3.69E-06 for children and 19.40 E-06 for adult) [95].Also, in a study conducted in 67 Chinese cities, the risk of carcinogenicity caused by PAHs was within the acceptable range (0.06 × 10 −6 to 7.56 × 10 −6 ) [103].In a study in 11 Chinese cities, Ma et al. reported a total carcinogenic risk of PAHs of 99.7% higher than 1 × 10 −4 and 9.50% higher than 1 × 10 −6 for the three groups of children, adolescents and adults, which indicate that potential risks are carcinogens of these air pollutants [103].Another result of this study was the higher level of carcinogenic risk due to inhalation of atmospheric PAHs in adults than in children, which was similar to the studies of Yan et al., Xia et al., Ma et al. and Ali [74,103].Nematollahi et al. also reported a higher level of carcinogenic risk of PAHs in Yazd than children [104].This may be due to the longer exposure time of adults than children, which was similar to the results of Ghanavati et al. in Abadan [105].Contrary to the results of this study, Gope et al. and Škrbić et al. during their study in the cities of Durgapur, India, and Novi Sad, Serbia, respectively, expressed a higher risk of carcinogenesis due to inhalation of PAHs in children than adults due to their low weight [85].Wang et al. also reported that the ILCR due to exposure to atmospheric PAHs was significantly higher in children than in adults because the internal organs, nerves, and immune system were growing in them, so they were more affected [93,105].

Trace element Principal component analysis & Confirmatory factor analysis
Investigating the relationships between the data is a proper way to identify the factors affecting air pollution, which was done by PCA in this study [84,105].The results of Bartlett's Test (χ 2 = 1273.74,df = 28, P < 0.001) and the KMO index = 0.868 showed the adequacy of the data to perform the analysis factor.The results of Anti-image-Matrices also showed that none of the metals had a diameter correlation less than 0.5.The metals Cd, Fe and Al were excluded from the analysis factor due to the complexity of the factor loading.
Based on the results of Table 1, factor analysis of data identified four factors in air pollution in the studied cities.These four factors determine 78.8% of the variance of air pollution.The first factor, with a rotating eigenvalue of 2.16, predicts an average of 27% variance in air pollution in these cities.This factor is composed of As, Hg and Ni, respectively, based on the degree of correlation of the metal.According to studies of coal combustion, gasoline vehicles, steel and plating industries, tyre wear is associated with the release of these pollutants into the air [54,55,80,83].The second factor, consisting of Cu and Mn with a rotating Eigenvalue of 1.62, determines about 20.18% of the variance of air pollution in large cities and indicates their release from tyre erosion, fossil fuel combustion and the steel industry [55,77,84].The third factor, with a rotating Eigenvalue of 1.47, determines an average of 18.42% of the variance of air pollution changes.This factor is composed of Cr and Zn metals.Vehicles, tyre and marble wear and industrial activities, especially the steel industry are the main causes of the presence of these pollutants in the air [54,55,77,80,83,84].The fourth factor is Pb alone, which averages 13.21% of the variance of air pollution changes with a rotating eigenvalue of 1.07.Vehicles and human activities seem to be one of the reasons for the presence of this pollutant in the air [54,55,77,80,83,84].
In order to confirm the air pollution factors of 7 big cities based on heavy metals obtained from EFA analysis, CFA confirmatory factor analysis was used.In fitting the CFA model in this study, the RMSEA index of 0.065 was obtained and indicates the acceptable fit.The GFI, CFI and IFI indices were 0.97, 0.99 and 0.99, respectively, indicating that the fit model is good.An RMR index of 0.3 was observed, which indicates an excellent fit.The CN index was 294.79, indicating that the minimum sample required for CFA was 294.Therefore, the number of samples in this study was enough to perform CFA (n = 434).According to the R 2 index, which is the root of the standardised coefficients, all heavy metals in factors F1, F2 and F3 and Pb have a high predictive level.
As shown in Figure 3, all heavy metals of the EFA model were determined in four factors, a statistical amount of t above 2 in the four determined factors (P < 0.05).Therefore, all these factors played a significant role in determining air pollution.In the first factor As with a coefficient of 0.85, 72% predicts the hidden variable of the first factor (R 2 = 0.72).Ni with a coefficient of 0.74 predicts 55% of the latent variable of this factor (R 2 = 0.55).Hg is between these two values.In the second factor, based on the CFA results based on standardised coefficients, Cu with a coefficient of 0.70 predicts 49% of the hidden variable of factor 2. In this factor, Mn had the least effect with a coefficient of 0.65 and a predictability level of 43% (R 2 = 0.43).In the third factor Zn with a standardised coefficient of 0.77, had the highest level of predictability in this factor (R 2 = 0.60).In this factor, Cr with a coefficient of 0.73 had the lowest level of predictability (R 2 = 0.53).The heavy metal Pb in this model, as a factor, had a direct impact on air pollution based on the CFA model.The direct impact coefficient was equal to 0.27, and the predictive level of contamination was equal to 7.1% (R 2 = 0.071).

PAH Principal component analysis & Confirmatory factor analysis
PCA was used to determine the relationships between PAHs and to identify the causes of air pollution in seven major cities of Iran.The results of Bartlett's Test (χ 2 = 4583.2,df = 160, P < 0.001) and the KMO index = 0.958 showed the adequacy of the data to perform the analysis factor.The results of Anti-image-Matrices also showed that none of the PAHs had a diameter correlation less than 0.5.Based on the results of Table S3, a factor analysis of data identified four factors in air pollution in the studied cities.These four factors predict 71.46% of the variance of air pollution.The first factor, with the eigenvalue rotating 3.74, accounts for an average of 23.4% of the variance in air pollution in these cities.Based on the degree of correlation, this factor is composed of Chr, Nap, Flrt and Anth, respectively, which are mainly emitted from vehicles, industrial activities, combustion of fuels (oil and diesel) and coal [105,106].The second factor, consisting of DBahA, BghiP, Bbf, Bkf and BaA with a rotated eigenvalue of 3.44, represents an average of 21.49% of the variance of air pollution changes in these 7 cities.Emissions of fossil fuels, diesel, gasoline and coal combustion were determined to be among the reasons for the presence of these pollutants in the air [105,107].The third factor with a rotated Eigenvalue of 3.16% consisting of 6 PAH IND, Ace, Pyr, Phen, Flu and Acy represents an average of 19.77% of the variance of pollution changes in 7 major cities.Vehicles, fuel combustion, coal and biomass seem to be their main sources [95,105,106].The fourth influential factor in the pollution of big cities is the only BaP, which with eigenvalue 1.09 represents an average of 6.79% of air pollution in these cities.Gasoline and diesel vehicles are sources of this pollutant [105,107].
Confirmatory factor analysis of CFA was used to confirm the air pollution factors of large cities based on PAHs obtained from EFA analysis.In the CFA model fit, the RMSEA index of 0.074 was obtained, indicating that the fit was acceptable.GFI, CFI and IFI were 0.91, 0.98 and 0.98, respectively, indicating that the fitted model is good.RMR index of 0.3 was observed, which represents an excellent fit.The CN index was gained equal to 171.4,indicating that it was the minimum sample required for CFA 171.Therefore, the number of samples in this study was enough to perform CFA (n = 434).According to the R 2 index, which is the root of the standardised coefficients, all PAHs in factors F1, F2 and F3 and BaP have a high level of predictability.
According to Figure 4, all PAHs of EFA model in four determined factors had a statistical amount of t above 2 (P < 0.05).Therefore, all these factors played a significant role in determining air pollution.According to Figure 4 in the first factor Chr with a coefficient of 0.78, 61% predicts the hidden variable of the first factor (R 2 = 0.61).Anth with a coefficient of 0.73, 53% predicts the latent variable of this factor (R 2 = 0.53).Nap and Flrt are located between these two values.In the second factor, based on the CFA results based on standardised coefficients, Bbf with a coefficient of 0.84 predicts 71% of the hidden variable of factor 2. In this factor, BghiP had the least effect with a coefficient of 0.72 and a predictability level of 52% (R 2 = 0.52).The PAHs of DBahA, Bkf and BaA had a predictive level between these two values.In the third factor, Phen with a standardised coefficient of 0.84, had the highest level of predictability in this factor (R 2 = 0.71).In this factor, IND with a coefficient of 0.72 had the lowest level of predictability (R 2 = 0.52).PAHs Ace, Pyr, Flu and Acy had predictive levels between these two pollutants.PAH BaP in this model as a factor directly affected air pollution based on the CFA model.The direct impact factor was 0.31 and the level of predictability of the contamination rate was 9.6% (R 2 = 0.096).

Conclusion
This was a comprehensive study in which the possible risks of exposure to 11 heavy metals and 16 PAHs of air from 120 sampling stations in 7 large industrial cities of Iran were analysed and the effective factors in causing pollution during 2019-2020 were analysed.Al and BaP had the highest amount among 11 heavy metals and 16 PAH, respectively, which show the severe contamination of some sites with these pollutants.Ahvaz city had the highest amount of pollutants compared to other cities, and this was due to increase of industrial activities, consumption of fossil fuels and dust storms.Assessment of carcinogenic and non-carcinogenic risk of heavy metals for children and adults except Mn in Arak and Isfahan was below the threshold, indicating that they have no adverse non-carcinogenic effects on population health.Also, the level of carcinogenic risk of all PAHs states their acceptable amount.It was also discovered that combustion of fuels and coal, vehicle exhaust gases, tyre wear and industrial activities were probably the most effective factors in the occurrence of pollutants.The results of this study can provide policy makers with appropriate scientific solutions such as reducing the use of dangerous fuel, proper dust management, proper control of pollutants from emission sources to reduce and manage air pollutants in these cities.Whereas the issue of health risk assessment and identification of factors affecting pollution is a vital and important issue.More research is needed in this area.

Figure 1 .
Figure 1.Study areas and sampling points.
Heavy metal concentrations and PAHs of sampling points are given in Fig S1 and Fig S2.

Figure 2 .
Figure 2. Spatial distributions of trace element (right) and PAHs(left) in air in seven cities.

Figure 3 .
Figure 3. Confirmatory factor analysis loadings for trace element concentrations in air in seven cities.

Figure 4 .
Figure 4. Confirmatory factor analysis loadings for PAHs concentrations in air in seven cities.
. While Zhou et al. in Hefei, China, Chen et al. in Shanghai, China, Cui et al. in Beijing, Yang et al. in China, Hou et al. in China, Zheng et al. in Northeast of China, stated

Table 1 .
Principal component analysis loadings for trace element concentrations in air in seven cities.