Next Article in Journal
Examining the Impact of Dimethyl Sulfide Emissions on Atmospheric Sulfate over the Continental U.S.
Previous Article in Journal
Physics-Informed Deep Learning for Reconstruction of Spatial Missing Climate Information in the Antarctic
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

Metal Composition and Source Identification of PM2.5 and PM10 at a Suburban Site in Pathum Thani, Thailand

1
Nuclear Technology Research and Development Center, Thailand Institute of Nuclear Technology (Public Organization), Nakhon Nayok 26120, Thailand
2
Department of Physics and Materials Science, Faculty of Science, Chiang Mai University, Chiang Mai 50200, Thailand
3
Nuclear Technology Service Center, Thailand Institute of Nuclear Technology (Public Organization), Nakhon Nayok 26120, Thailand
*
Author to whom correspondence should be addressed.
Atmosphere 2023, 14(4), 659; https://doi.org/10.3390/atmos14040659
Submission received: 13 February 2023 / Revised: 13 March 2023 / Accepted: 23 March 2023 / Published: 31 March 2023
(This article belongs to the Section Air Quality)

Abstract

:
This study reports the mass concentrations, elemental characterization and identification of the possible sources of PM2.5 and PM10 at a suburban site in Pathum Thani, Thailand. The sampling was done from 18 February 2021 to 14 September 2021. PM2.5 concentrations were between 0.39 μg/m3 and 174.26 μg/m3, while PM10 concentration ranged from 12.75 μg/m3 to 242.02 μg/m3. The average concentration of PM2.5 in the wet season (61.96 μg/m3) in the study area exceeded the national ambient air quality standards (NAAQS). Particle-induced X-ray emission (PIXE) was used to measure the element concentrations. The main elements, namely Fe, K, Cr and Ca, in both PM2.5 and PM10 showed significantly higher concentrations in the summer season. A qualitative inter-elemental correlation analysis, principal component analysis (PCA) and cluster analysis (CA) were applied for source identification of PM2.5 and PM10. The results for the three procedures were in good agreement. Four and three factors of sources were isolated by the PCA for PM2.5 and PM10, respectively. The main sources identified by PCA were, for PM2.5, soil dust and biomass burning (32%), road dust and industrial emission (25%), vehicle and industrial emission (10%) and soil dust (9%); for PM10, road dust and industrial emission (36%), crustal and biomass burning (30%) and industrial sources (10%).

Graphical Abstract

1. Introduction

Thailand is facing problems with air quality issues, resulting in adverse health consequences. It also has serious effects on climate forcing [1,2], global warming [2,3] and visibility [1,4]. The most polluted substances include particulate matter (PM2.5 and PM10) and gaseous species (sulfur dioxide; SO2, nitrogen dioxides; NO2, carbon monoxide; CO, hydrocarbon, etc.). Among air pollutants, air particulate matter (APM), PM2.5 (fine particles with a diameter <2.5 μm), and PM10 (coarse particles with a diameter between 2.5 and 10 μm) draw considerable attention because of their high contribution to air pollution and the detrimental health effects such as respiratory and cardiovascular disease [5,6], visibility effects [4,7] and climate change [1,2]. The levels of PM2.5 and PM10 are monitored by Thailand’s Pollution Control Department (PCD). It revealed that every year, a high concentration of PM2.5 is found from January to April and peaks around March, while the concentration of PM10 reaches from December to February and peaks around January. In 2019, the concentration of PM2.5 in Bangkok ranged from 4–104 μg/m3 and exceeded the national ambient air quality standards (NAAQS) for 34 days, while the concentration of PM2.5 in Chiang Rai, the northern area of Thailand, reached 353 μg/m3 which is seven times higher than the standard level. In addition, a high concentration of PM2.5 and PM10 can be found in central parts of Thailand, including Pathum Thani, Samut Prakan, Samut Sakhon, Nonthaburi and Saraburi.
The main sources of PM2.5 and PM10 in Thailand are automobiles, agriculture activities (crop residue burning; rice, cassava, sugarcane, soybean, potato, etc.; wildfires, and others related to agriculture industries) [8,9,10,11,12,13] and the industrial sector (petroleum refinery, natural gas separation, iron and steel, cement, production of rubber, thermal power plant, etc.) [9,12,13,14]. There have been a number of reports mentioning the possible sources of PM2.5 and PM10 in Thailand. Ahmad et al. (2022) [13] reported that the major sources of PM2.5 and PM10 in three areas of Bangkok (Ari, Din Daeng and Bangna) were local emission sources and sea salt, while vehicle emissions, coal combustion, biomass burning, secondary aerosol formation, industrial emissions and dust were found to be the most significant local sources of PM2.5 and PM10. Moreover, the health risk assessment of heavy metals revealed that the Cr in PM2.5 and PM10 had a carcinogenic risk to adults in Bangna and Din Daeng. The carcinogenic risk value of Cr exceeded the permissible limits, which is the most serious threat to the residents of Bangkok. Vongruang and Pimonsree (2020) [15] applied the WRF-CMAQ modeling system for source identification of PM10 in mainland Southeast Asia (MSEA), including Laos, Myanmar, Cambodia, Thailand, China, and Vietnam. It was found that the main source of PM10 was biomass burning which contributed 49% in MSEA, while it specifically contributed 45% in Thailand. Choochuay et al. (2020) [12] mentioned that the level of PM2.5 in Bangkok was affected by both local (vehicular exhausts) and regional (biomass-burning) sources. Hierarchical cluster analysis (HCA) results also revealed that the major sources of PM2.5 were a mixture of combustion activities (biomass burning, vehicular exhausts, fossil fuel, coal and industrial emissions). In addition, PCA extracted five important sources of PM2.5, including vehicular exhausts, biomass burning, sea salt aerosols, power plants, and industrial emissions. Kanjanasiranont et al. (2022) [8] investigated the concentrations and chemical compositions of PM10 in Bangkok and its vicinity areas, including Nonthaburi and Nakhon Pathom, Thailand. C, O and Si were found to be primary components with minor and trace elements, including Fe, Cu, Cr, Ca, Na and S. These elements indicated that their origin might be related to road dust, vehicle emission, and the erosion of building products. Kayee et al. (2020) [11] studied the metal concentrations and source apportionment of PM2.5 in Chiang Rai, the northern part of Thailand and Bangkok from January–April 2019. The results showed that the possible sources extracted by PCA for Chiang Rai were (1) crustal dust and biomass burning, (2) industrial sources, and (3) refuse incineration mixed with road dust; for Bangkok, (1) natural background, industrial emissions and coal combustion, (2) traffic emission, and (3) sea spray. In addition, high concentrations of crustal metals and an increase in the Pb isotope ratio (206Pb/207Pb) in Chiang Rai were noted during the haze. Kawichai et al. (2022) [16] studied the source identification of PM2.5 during a haze period (January–May 2017) in Chiang Mai, the northern area of Thailand, using stable carbon (δ13C) and nitrogen isotopes (δ15N). The results indicated that biomass burning was the main source of PM2.5, to which C3 and C4 plants contributed about 74% and 26%, respectively.
As shown in the aforementioned reports, most research focuses on the studies of source identification of PM2.5 and PM10 in urban sites, Bangkok and the northern part of Thailand, Chiang Mai and Chiang Rai, while only a few studies have been conducted on source identification of both PMs in the suburban area of Thailand, especially Pathum Thani. Pathum Thani is one of the central provinces located right next to Bangkok. It has many important industrial estates which produce steel wires, electronic parts, food processing, metal processing, etc. Pathum Thani also has many universities and academic and research institutes. Moreover, there are a large number of rice paddies and other crop-growing areas. There have been reports that Pathum Thani is facing an air pollution crisis [17]. The PCD recorded that the levels of PM2.5 in the Khlong Luang district, in Pathum Thani, reached 88, 74, and 79 μg/m3 in 2019, 2020, and 2021, respectively, which exceeded the state standard. A previous study has shown that the main contributing sources to PM2.5 in the Klongha district in Pathum Thani were traffic and biomass burning, while soil and construction were the major sources of PM10 [18].
This work measured the mass concentrations of PM2.5 and PM10 in summer and wet seasons from 18 February 2021 to 14 September 2021 at a suburban site in Pathum Thani, Thailand. The seasonal variations of mental concentration were determined by the particle-induced X-ray emission (PIXE) technique. In addition, the sources of both PM2.5 and PM10 were identified by inter-element correlation, cluster analysis (CA) and principal component analysis (PCA).

2. Materials and Methods

2.1. Study Area Description

Thailand is a tropical country located at 6°–20° latitude north and 98°–105° longitude East. As Thailand is situated so close to the equator, it experiences hot and humid temperatures all year round. The average temperature in Thailand ranges from 18 to 38 °C, and the mean annual rainfall varies from 1000–1800 mm.
Pathum Thani is part of the Bangkok Metropolitan Region (BMR) and is situated approximately 40 km north of Bangkok. It covers an area of 589 sq. km and has a population of 938,000. Most areas of the city are plains, with two rivers. The Chao Phraya River flows through the heart of the district, Muang district. The sampling site (14.24° N, 100.43° E) was located at a height of about 2 m above ground level in the Thailand Institute of Nuclear Technology, Khlong Luang district, Pathum Thani Province, Thailand (Figure 1). The site is influenced by a complex mixture of residential, commercial, agricultural and industrial emissions. It is surrounded by houses (wooden and brick), a canal, a paddy field and a field crop. The monitoring site also hosts many large and medium-scale industries such as semiconductors, steel wires, electronic power equipment, motor, machine tools and metal processing. These industries are located approximately 20 km northwest of the monitoring site.

2.2. Sampling Procedure and Instrumentation

PM2.5 and PM10 sampling were carried out using a Tisch TE-Wilbur sampler (Low volume air sampler) according to the Federal Reference Method (40 CFR, Part 53, 50 Appendix L). Samples were collected on 46.2 mm PTFE filters for 2-3 days a week for PM2.5 and PM10, respectively, at a flow rate of 16.67 L/min for 24 h. Sampling started at 9 a.m. and ended at 9 a.m. the next day. PM2.5 and PM10 sampling were done continuously from the 18 February 2021 to the 14 September 2021. Each filter was stored in a polyethylene bag and conditioned in an auto-desiccator at 25 °C and 40–50% relative humidity at least 24 h before and after sampling; then, they were weighed on a digital microbalance (Mettler-Toledo; sensitivity of 0.001 mg) to determine the gravimetric mass of PM2.5 and PM10. Due to the filter samples used in this study being PTFE, they are non-hygroscopic and possess low sensitivity to variations of temperature and humidity due to their permanent hydrophobicity, hence they have the great ability to limit water absorption [19]. In total, 77 samples of PM2.5 (31 during summer and 46 during the wet period) and 49 samples of PM10 (21 during summer and 28 during the wet period) were collected. The sampling parameters are summarized in Table 1.
Meteorological parameters, including temperature, relative humidity, atmospheric pressure, wind speed and rainfall, were collected from Khlong Nueng station, Pathum Thani, located approximately 18 km from the motoring site, as provided by the National Hydroinformatics Data Center available online (www.tiwrm.hii.or.th/v3/ (accessed on 15 September 2021)).

2.3. Elemental Analysis

The collected PM2.5 and PM10 samples were analyzed by the PIXE technique for the measurement of elemental composition. The PIXE measurements were performed using 2 MeV proton by 1.7-MV Tandetron tandem accelerator at the Department of Physics and Materials Science, Faculty of Science, Chiang Mai University, Thailand. The X-rays emitted were detected by means of a Si (Li) detector (Mirion Technologies (Canberra), Inc., Meriden, CT, USA). PIXE spectra were determined using the GUPIX software. The energy resolution of the detection system was determined by using a standard 304 stainless steel sample. The resolution was estimated from the full width at half maximum (FWHM) of the Fe Kα1 peak at 6.4 keV to be 180 eV ± 10 eV. Calibration of the PIXE system was performed by irradiating the suitable Micro Matter thin target standards [20]. The precision and accuracy of the PIXE technique are typically within ±15%. Detailed parameters are shown in Table 2.

2.4. Source Identification Methods

Three multivariate methods, principal component analysis (PCA), cluster analysis (CA) and Pearson correlation analysis, were used to identify potential sources of PM2.5 and PM10 at the sampling site. PCA was used to identify the number of significant factors or sources contributing to the PM mass, which are characterized by specific tracers or pollution sources, whereas CA analysis verifies the relationship and difference between these elements as markers of different sources [3]. Meanwhile, Pearson correlation analysis establishes the inter-elemental relationships and identifies the rotations among the sources of PM. PCA and CA were completed using the Minitab 17 statistical software, while Pearson correlation analysis was performed with the ORIGIN 8 software.
PCA with varimax rotation was performed for compression of the data by reducing the number of variables from a large set of variables to a small set of uncorrelated variables called principal components that still contain most of the information in the large set. The first component accounts for the largest possible variance (variability in the data), and each succeeding component accounts for as much of the remaining variability as possible. The number of significant factors was based on the cumulative variance (>70%) and eigenvalue (must be greater than 1), Kaiser’s criterion [21].
CA was performed to explore the set of homogeneous elemental clusters representing a particular receptor source. The process took place by dividing each metal into clusters. Subsequently, two or more metals with the highest similarity are combined until all elements are linked and are represented by a dendrogram source [3].

3. Results and Discussion

3.1. PM2.5 and PM10 Mass Concentrations and Meteorology

Thailand’s climate is influenced by two monsoons (the Southwest and Northeast monsoon) with three main seasons. The Southwest monsoon (from mid-May to September) usually brings warm and moist air from the Indian Ocean toward Thailand, resulting in abundant rain over the country. Meanwhile, the Northeast monsoon (from October to March) blows in from Indo-China, bringing relatively cool, dry conditions (less humidity and less chance of rainfall) to Thailand (except for the southern part of the country, where the rain blows in from the Pacific Ocean). In this dry period, haze events usually appear due to forest fires and the burning of agricultural residues. Three main seasons can be described as the following: the hot season (mid-February to mid-May), the wet season (mid-May to mid-October) and the cool season (mid-October to mid-February).
Temporal variations and summary statistics of meteorological parameters during the sampling period at Khlong Nueng station, Pathum Thani, located approximately 18 km from the motoring site, as recorded by the National Hydroinformatics Data Center, are shown in Figure 2 and Table 3, respectively. The maximum temperature ranged from 30.50 to 39.50 °C, and the highest maximum temperature was recorded during the pre-Southwest monsoon. Meanwhile, the minimum temperature was found to be from 20.5 to 29.0 °C and was highest during the Southwest monsoon. The average relative humidity was found to be 88.96%, while the maximum and minimum relative humidity were 70 and 98% in June and May, respectively (the Southwest monsoon). The maximum rainfall of 74.8 mm was noted during the pre-Southwest monsoon, which was above normal. However, during the Southwest monsoon, the rainfall was below normal. The average wind speed was found to be 0.35 m/s, while the maximum wind speed was 2.50 m/s. The air pressure ranged from 754 to 763 mmHg with an average of 757 mmHg.
Figure 2 and Table 3 show the temporary variations and summary statistics of PM2.5 and PM10 mass concentrations in suburban Pathum Thani, Thailand, during different seasons (summer and wet) from 18 February to 14 September 2021. Both 24-h monitoring PM2.5 and PM10 levels in Pathum Thani were compared with three guideline levels for allowable 24-h PM2.5 and PM10 concentrations, quoted from the World Health Organization (WHO), the United States Environmental Protection Agency (US EPA) and Thailand National Ambient Air Quality Standards (Thailand NAAQS). As can be seen in Figure 2, in summer days from mid-February to mid-May 2021, the maximum and minimum concentrations of PM2.5 were 133.73 and 1.65 μg/m3, respectively, with an average value of 27.53 μg/m3 (n = 33), which was slightly higher than the WHO air quality guideline (25 μg/m3). However, this mean value was lower than the US EPA (35 μg/m3) and the NAAQS (50 μg/m3). Meanwhile, the maximum and minimum concentrations of PM10 were 81.72 and 12.75 μg/m3, respectively, with an average value of 45.08 μg/m3 (n = 21). It can be noted that the mean concentration of PM10 was below the values defined by the WHO (50 μg/m3), the US EPA (120 μg/m3) and the NAAQS (150 μg/m3).
In wet seasons (mid-May to September), the concentration of PM2.5 varied from 0.39 to 174.26 μg/m3 with an average value of 61.96 μg/m3 (n = 46). The mean concentration of PM2.5 in this season exceeded the values defined by the WHO, the US EPA and the NAAQS, while the concentration of PM10 ranged from 16.06 to 242.02 μg/m3 with an average value of 81.51 μg/m3 (n = 28). The average concentration of PM10 exceeded the value defined by the WHO; however, it was still below the values defined by the US EPA and the NAAQS. Normally, during the wet period in Thailand, PM2.5 and PM10 concentrations are lower than in the summer and winter [22]. The mean concentration values of PM2.5 and PM10 over the same (wet) period at Khlong Nueng station, Pathum Thani, located approximately 18 km from the motoring site, as reported by the PCD (http://www.air4thai.com/webV2/history/ (accessed on 20 September 2021)) showed to be 11.81 and 30.62 μg/m3, which were lower than those obtained from our sampling site. The high mass concentrations of PM2.5 and PM10 during the wet season might be due to the road and building construction activities near the monitoring site combined with low rainfall during this period (see Figure 2).
Table 4 shows the average concentrations of PM2.5 and PM10 in several tropical countries. As shown in Table 4, the average concentration of PM2.5 in this study was higher than those in Chiang Mai [16], Jayapura, Indonesia [23] and Petaling Jaya, Malaysia [24]. However, it was much lower than those in Bangkok [12], which is the biggest urban site in Thailand. It was also significantly lower than those in Chiang Rai [10] and Pekanbaru, Indonesia [25]. Meanwhile, the average concentration of PM10 in this study was much higher than those in Pekanbaru [25] and Jayapura, Indonesia [23], as well as Belisario Quito, Los Chillos and Tababela, Ecuador [26]. It can also be noted that the average concentration of PM10 in this study was slightly higher than that of Barranquilla, Colombia [27].

3.2. Elemental Concentration

Table 5 presents the minimum, maximum and mean values, standard deviation (SD) and minimum detection limits (MDL) in ng/m3 for the elemental concentration in PM2.5 and PM10 samples collected from Pathum Thani Province. For the PIXE technique, the detection limit highly depends on the presence of the other elements in the samples due to its impact on spectral interference. As a result, the detection limits vary from one sample to another [28]. The results showed that Fe, K, Cr and Ca (30–158 ng/m3) were the most abundant in PM2.5. The second abundant was Zn, S, Ni, Si and Mn (15–21 ng/m3). Low concentrations were obtained for Cu, Ti and Cl (6–8 ng/m3). The relative contribution of Si, Fe, K, Ca, Mn and Ti was associated with crustal or natural origin. [29,30]. Meanwhile, Cr, Zn, S, Ni and Cu mostly originated from anthropogenic origin [7,29]. For PM10, the crustal elements Fe, K, and Ca, ranging from 71–292 ng/m3, were in high concentrations, and the besides crustal elements, anthropogenic Cr and Ni, also had high concentrations (72–114 ng/m3). Anthropogenic (Zn, S, and Cu and Cl) and crustal (Si, Mn and Ti) elements were found in medium concentrations (10–25 ng/m3).
Significant changes in concentrations of key elements for PM2.5 and PM10 during the whole period were observed, as shown in Figure 3. It can also be seen that Fe and Cr concentrations had similar temporal trends for both PM2.5 and PM10. For PM2.5, the seven concentration peaks of Fe and Cr occurred as follows: on 9 April, 30 April, 5 May, 19 May, 31 May, 30 June and 18 August at 3879, 872, 509, 1381, 521, 802 and 488 ng/m3 for Fe, and 948, 278, 156, 444, 179, 267 and 162 ng/m3 for Cr. For PM10, the five concentration peaks of Fe and Cr occurred as follows: on 22 April, 13 May, 20 May, 1 June and 20 July at 4145, 893, 1225, 1121 and 1753 ng/m3 for Fe, and 1033, 246, 341, 330 and 490 ng/m3 for Cr.
No significant pattern differences in element concentrations between summer and wet seasons for PM2.5 were observed, as shown in Figure 4. The mean concentrations of main elements for PM2.5, such as Fe, K, Cr and S, in the summer season were significantly higher than those in the wet season. The concentrations of element species followed the order of Fe > K > Cr > S > Ca > Zn > Ni > Mn > Si > Ti > Cu > Cl for the summer season, and Fe > K > Cr > Ca > Zn > Ni > Si > Mn > S > Cu > Cl > Ti for the wet season. For PM10, similar trends of element concentrations in the summer and wet seasons were noted (see Figure 4), and almost all element concentrations in the summer season were higher than those in the wet seasons, like in PM2.5. The orders of element concentrations were Fe > Ca > Cr > Ni > K > Zn > S > Mn > Si > Cl > Ti > Cu for the summer season and Fe > Ca > Cr > K > Ni > Zn > Cl > Si > Mn > S > Cu > Ti for the wet season. A significantly lower element concentration in the wet season (monsoon) than in the summer season for both PM2.5 and PM10 might be due to the metals being precipitated and washed out by frequent rain during the wet season [31].

3.3. Source Identification

3.3.1. Correlation Analysis

In order to identify qualitatively possible sources of PM2.5 and PM10 contributing to the elements, Pearson’s correlation analysis was performed. Pearson’s correlation coefficient matrix relates the mass of PM2.5 and the elements concentrations of PM2.5; the mass of PM10 and the elements concentrations of PM10 are given in Table 6 and Table 7, respectively. As shown in Table 6, a weak correlation coefficient (r: 0.230–0.257) was found between the mass of PM2.5 and elements from crustal and biomass-burning origin such as Ca, K, Cl and S. Meanwhile, the correlation coefficient between the mass of PM10 and elements was observed to be quite poor (r: 0.012–(−0.167)) (see Table 7). For the inter-elemental relationships of PM2.5, it can be seen that the highest correlations were between Cr-Mn (r = 0.967), Cr-Fe (r = 0.990), Cr-Ni (r = 0.966), Mn-Fe (r = 0.949), Mn-Ni (r = 0.947) and Fe-Ni (r = 0.968). These strong correlations revealed that all of these associated elements originated from road dust [32,33] and industrial emission sources; metallurgical, electromechanical, electronic industries and electroplating [34,35]. Fe, Mn and Cr are commonly used in the steel industry [7]. Cr and Ni are also often used in electroplating [7]. In addition, Zn releases from tire and brake abrasion as ZnO is used as a vulcanization agent to produce tires [33]. There were moderate positive correlations between Si-S (r = 0.601), Si-Ca (r = 0.632) and K-Ca (r = 0.629), Si-Cl (r = 0.578), Si-K (r = 0.547), which can be observed, suggesting they predominantly come from crustal sources [36]. Si classifies as clay, such as almandine or kaolinite, common in soils of tropical regions [19]. Ca and S are possibly from calcium sulfate. K and Ti are usually found in aluminosilicate clays. Cl can be absorbed from soil dust [19]. Subsequently, moderate positive correlations between S-K (r = 0.741) and Cl-K (r = 0.614), S-Cl (r = 0.511), S-Ca (r = 0.577) and Cl-Ca (r = 0.556), as well as a moderate negative correlation between K-Cr (−0.720), indicated that the source could be from biomass burning [37]. K and Cl are considered biomarkers for crop residue burning, such as wheat and rice [38]. K and Cl also probably relate to the use of fertilizers and herbicides in the field [39]. Meanwhile, S and Ca come from wood burning [40,41]. Moderate correlations between Zn-Cu (r = 0.524) and Mn-Cu (r = 0.597) involved vehicles exhaust [34,42] and industrial emission sources [43]. Another interesting correlation between Ti-Ni (r = 0.725) is probably attributed to industrial sources of different industrial types [44].
As can be seen in Table 7, for the inter-elemental relationships of PM10, Cr showed strong positive correlations with Mn (r = 0.971), Fe (r = 0.993), Ni (r = 0.969) and Cu (r = 0.746). Mn also displayed important correlations with Fe (r = 0.970) and Ni (r = 0.914). A strong correlation between Fe-Ni (r = 0.984) can be noted. These highest correlations were quite similar to PM2.5 and had similar source origin from road dust and industrial emission. Vaio et al. (2016) [45] discovered that Cr and Cu contained in PM10 can be attributed to industrial sources. Ahmad et al. (2022) [13] also reported that Mn and Ni may result from industrial emissions, while Cu and Cr originated from road dust. Truong et al. (2022) [46] found that Cr, Ni, Fe and Cu related to non-ferrous metallurgical industry, metal processing and refinery industry emissions. Meanwhile, Si was well correlated with Ca (r = 0.625) and had moderate correlations with S (r = 0.508). K also showed a strong correlation with Ca (r = 0.804) and a moderate correlation with Ti (r = 0.535). Ca displayed a moderate correlation with Ti (r = 0.520). This observation suggested that all of these associated elements were mainly from crustal sources [27]. Good correlations between S-K (r = 0.737) and S-Ca (r = 0.685) were noted, suggesting a possible contribution of biomass burning [47]. Another interesting correlation was Si-Ni (r = 0.696), which revealed that it possibly originated from industrial sources such as metallurgical processes [40] and steel industries [26].
As for the correlation analysis results of PM2.5 and PM10, the emission sources of PM2.5 and PM10 were quite similar. The main emission sources related to road dust and industrial, crustal and biomass burning. These sources are associated with the combination of several activities of the monitoring site, which comprises many large and medium-scale industries such as metallurgical, electromechanical, electronic industries and electroplating, as well as a significant number of commercial centers, agricultural land and residential areas.

3.3.2. PCA Analysis

PCA was employed to predict the potential sources of PM2.5 and PM10. The PCA results are summarized in Table 8, and the possible sources of PM2.5 and PM10, as identified using PCA, are shown in Figure 5. The obtained results for both PM2.5 and PM10 were in agreement with the inter-element relationships obtained from Pearson’s correlation matrix results. As can be seen in Table 8, PCA resolved four components explaining 76.7% cumulative variance for PM2.5. The strongest Factor 1 explained 32.4% of the total variance with an eigenvalue of 3.89. The influenced element consisted of Si, S, Cl, K and Ca. The elements such as Si, K and Ca may be from the local soil dust. Bangar et al. (2021) [48] and Onat et al. (2013) [49] reported that crustal dust (soil and surface dust) is the primary source of Si, K and Ca. Potassium (K), chlorine (Cl) and sulfur (S) were considered tracers of agricultural-residue (biomass) burning [34,47,48]. The area of Pathum Thani is mostly surrounded by rice paddy fields. Rice paddies are burned after the harvesting season from November to April. Therefore, this factor should be identified as a combined soil/surface dust and biomass burning.
Factor 2 accounted for 25.3% of the total variance with an eigenvalue of 3.04. It strongly correlated with Mn, Fe, Cr and Ni. This represented a mixture of road dust and industrial emission sources [50]. Mn, Fe and Cr are fingerprint elements for road dust sources [49]. Bie et al. (2021) [50] reported that Mn, Fe and Cr are the main chemical components in profiles of industrial emission. Wang et al. (2019) [7] reported that Cr and Ni are emitted from industrial sources such as chemical, metallurgical, electromechanical, and electronic industries. There is a large industrial zone composed of various industries, such as metal processing, steel wires and electronic power equipment, located approximately 20 km northwest of the monitoring site.
Factor 3 explained 10.2% of the total variance with Zn and Cu as the main elements. Zn and Cu are commonly recognized as a source of the mechanical wear of motor vehicles and tire wear [7]. Ahmad et al. (2020) [51] reported that Zn and Cu might have originated from the oil-based combustion industry. Therefore, a plausible source for Factor 3 would be a combined vehicle exhaust and industrial emission [50].
Factor 4 accounted for 8.7% of the total variance, and it was mainly made up of Ti. Ti most likely originated from soil dust [19]. This factor was likely a soil dust source.
In the PCA analysis of PM10, three factors explained 75.8% of the total variance. Factor 1 showed high loading of Mn, Fe, Cr and Ni, and it accounted for 33.6% of the total variance. It was assigned to a mixed resuspended road dust and industrial emission very similar to PM2.5 Factor 2. The elements such as Mn, Fe and Cr are the main elements in road dust [52,53]. Cr and Ni may be associated with the various industrial activities [26,54] located approximately 20 km northwest of the monitoring site, such as metal processing, steel wires and electronic power equipment. Cr and Ni are emitted from the combustion process and metallurgical industries [52,55].
Factor 2 explained 29.6% of the total variance and was dominated by S, K, Ca, Ti and Zn. Elements such as K, Ca, Ti and Zn could be attributed to crustal dust [6]. The presence of S and K indicated the contribution of biomass burning [44]. Factor 2 can therefore be identified as a mixture of crustal dust and a biomass-burning source similar to PM2.5 Factor 1.
Factor 3 explained 10.0% of the total variance, and it was characterized by Si, Cl and Cu. Si and Cl might be from the industrial coal combustion process [54]. Panda and Nagendra (2018) [31] reported that Cu can be emitted from industrial combustion processes and metal industries. Therefore, this Factor may originate from industrial sources.

3.3.3. CA Analysis

The CA was also calculated to support the PCA results, as shown in Figure 6. Figure 6 shows that the grouping obtained with the CA analysis was in good agreement with the PCA results for both PM2.5 and PM10. For PM2.5, the first group was characterized by a group of elements with Si, S, Ca, Cl and K, which was very similar to Factor 1 of PCA, representing soil dust with a significant contribution to biomass burning. The second cluster included Cr, Fe, Mn and Ni and matched well with Factor 2 of PCA, indicating road dust combined with industrial emission. The third group showed a link between Zn and Cu analogous to Factor 3 of PCA. The possible sources were vehicle exhaust and some contributions from industrial emissions. The last group characterized by CA was Ti, which appeared on its own and was not linked to any elements. Ti is a marker element for crustal dust. This result was consistent with Factor 4 of PCA.
The dendrogram of the CA analysis for PM10 is shown in Figure 7. The first cluster with elements such as Si, Cl and Cu reparented the industrial emission. This was in good agreement with Factor 3 of PCA. The second cluster showed a group of elements such as Cr, Ni, Mn and Fe. This was absolutely in agreement with Factor 1 of PCA and represented a mixed source related to anthropogenic activities, including road dust and industrial emissions. The third cluster was a link between S, K, Ca, Ti and Zn and similar to Factor 2 of PCA. The presence of these elements can be attributed to crustal dust (K, Ca, Ti and Zn) together with the source of biomass burning (S and K).
It can be concluded from the CA analysis results that the major pollution sources of both PM2.5 and PM10 were quite similar. There were four main sources of biomass burning, a mix of road dust and industrial emission, vehicle exhaust and industrial emission, and crustal dust contained in PM2.5. Meanwhile, for PM10, there were three main contributory pollution sources, and these were industrial emission, a mix of road dust and industrial emission and a mix of crustal dust and biomass burning.

4. Conclusions

In this work, the elemental characterization and source identification of PM2.5 and PM10 at a suburban site in Thailand during summer and wet seasons from 18 February 2021 to 14 September 2021 was performed. The results showed that the annual mean concentrations of PM2.5 and PM10 were 48.10 ± 46.97 μg/m3 and 65.90 ± 47.51 μg/m3, respectively. It can be observed that the concentration of PM2.5 in the wet season, with a mean value of 61.96 μg/m3 was higher than the NAAQS. The elements concentration in PM2.5 were in descending order as follows: Fe > K > Cr > Ca > Zn > S > Ni > Si > Mn > Cu > Ti > Cl, while the order of elements concentration for PM10 was Fe > Ca > Cr > Ni > K > Zn > S > Cl > Si > Mn > Ti > Cu. The high concentrations of Fe, K, Cr, Ni and Ca indicated that they originated from industrial sources, except for K and Ca, which could be soil dust but could also be emitted from biomass burning. A significant seasonal variation in elemental concentrations of PM2.5 and PM10 was observed, with their highest values during the summer season followed by the wet season. The source identification using PCA was influenced by four and three possible factors for PM2.5 and PM10, respectively. The identified sources for PM2.5 were soil dust and biomass burning, road dust and industrial emission, vehicle and industrial emission and soil dust, respectively. In the case of PM10, the possible sources were road dust and industrial emission, crustal and biomass burning and industrial sources, respectively. These results were also confirmed by the CA and correlation analysis data.

Author Contributions

S.N.: Investigation, methodology, data curation; W.K.: Formal analysis, data curation, visualization; U.T.: validation, resources, investigation; W.S.: Validation, resources, investigation; R.F.: Investigation, resources; D.R.: Conceptualization, funding acquisition, writing—original draft, writing—review and editing. All authors have read and agreed to the published version of the manuscript.

Funding

This research is funded by the Regional Office of the Regional Cooperative Agreement for Research, Development and Training related to Nuclear Science and Technology for Asia and the Pacific (RCARO) (Contract No.: RCARP02/RC08) and the National Research Council of Thailand (NRCT). This work was also supported in part by the International Atomic Energy Agency (IAEA) 2022–2025 Technical Cooperation Programme, THA0016.

Data Availability Statement

Data will be available by request to the corresponding authors.

Conflicts of Interest

The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Diapouli, E.; Fetfatzis, P.; Panteliadis, P.; Spitieri, C.; Gini, M.I.; Papagiannis, S.; Vasilatou, V.; Eleftheriadis, K. PM2.5 source apportionment and implications for particle hygroscopicity at an urban background site in Athens, Greece. Atmosphere 2022, 13, 1685. [Google Scholar] [CrossRef]
  2. Zhang, G.; Ding, C.; Jiang, K.; Pan, G.; Wei, X.; Sun, Y. Chemical compositions and sources contribution of atmospheric particles at a typical steel industrial urban site. Sci. Rep. 2020, 10, 7654. [Google Scholar] [CrossRef] [PubMed]
  3. Waheed, S.; Jaafar, M.Z.; Siddique, N.; Markwitz, A.; Brereton, R.G. PIXE analysis of PM2.5 and PM2.5–10 for air quality assessment of Islamabad, Pakistan: Application of chemometrics for source identification. J. Environ. Sci. Health Part A 2012, 47, 2016–2027. [Google Scholar] [CrossRef] [PubMed]
  4. Dong, D.; Qiu, T.; Du, S.; Gu, Y.; Li, A.; Hua, X.; Ning, Y.; Liang, D. The chemical characterization and source apportionment of PM2.5 and PM10 in a typical city of Northeast China. Urban Clim. 2023, 47, 101373. [Google Scholar] [CrossRef]
  5. Usman, F.; Zeb, B.; Alam, K.; Valipour, M.; Ditta, A.; Sorooshian, A.; Roy, R.; Ahmad, I.; Iqbal, R. Exploring the mass concentration of particulate matter and its relationship with meteorological parameters in the Hindu-Kush Range. Atmosphere 2022, 13, 1628. [Google Scholar] [CrossRef]
  6. Ramírez, O.; Campa, A.M.S.; Amato, F.; Ruth, A.; Catacolí, R.A.; Rojas, N.Y.; Rosa, J. Chemical composition and source apportionment of PM at an urban background site in a high-altitude Latin American megacity (Bogota, Colombia). Environ. Pollut. 2018, 233, 142–155. [Google Scholar] [CrossRef]
  7. Wang, S.; Yu, R.; Shen, H.; Wang, S.; Hu, Q.; Cui, J.; Yan, Y.; Huang, H.; Hu, G. Chemical characteristics, sources, and formation mechanisms of PM2.5 before and during the Spring Festival in a coastal city in Southeast China. Environ. Pollut. 2019, 251, 442–452. [Google Scholar] [CrossRef]
  8. Kanjanasiranont, N.; Butburee, T.; Peerakiatkhajohn, P. Characteristics of PM10 levels monitored in Bangkok and its vicinity areas, Thailand. Atmosphere 2022, 13, 239. [Google Scholar] [CrossRef]
  9. Kliengchuay, W.; Srimanus, W.; Srimanus, R.; Kiangkoo, N.; Moonsri, K.; Niampradit, S.; Suwanmanee, S.; Tantrakarnapa, K. The association of meteorological parameters and AirQ+ health risk assessment of PM2.5 in Ratchaburi province, Thailand. Sci. Rep. 2022, 12, 12971. [Google Scholar] [CrossRef]
  10. Niampradit, S.; Kliengchuay, W.; Mingkhwan, R.; Worakhunpiset, S.; Kiangkoo, N.; Suntorn, S.; Hongthong, A.; Siriratruengsuk, W.; Muangsuwan, T.; Tantrakarnapa, K. The elemental characteristics and human health risk of PM2.5 during haze episode and non-haze episode in Chiang Rai Province, Thailand. Int. J. Environ. Res. Public Health 2022, 19, 6127. [Google Scholar] [CrossRef]
  11. Kayee, J.; Sompongchaiyakul, P.; Sanwlani, N.; Bureekul, S.; Wang, X.; Das, R. Metal concentrations and source apportionment of PM2.5 in Chiang Rai and Bangkok, Thailand during a biomass burning season. ACS Earth Space Chem. 2020, 4, 1213–1226. [Google Scholar] [CrossRef]
  12. ChooChuay, C.; Pongpiachan, S.; Tipmanee, D.; Suttinun, O.; Deelaman, W.; Wang, Q.; Xing, L.; Li, G.; Han, Y.; Palakun, J.; et al. Impacts of PM2.5 sources on variations in particulate chemical compounds in ambient air of Bangkok, Thailand. Atmos. Pollut. Res. 2020, 11, 1657–1667. [Google Scholar] [CrossRef]
  13. Ahmad, M.; Manjantrarat, T.; Rattanawongsa, W.; Muensri, P.; Saenmuangchin, R.; Klamchuen, A.; Aueviriyavit, S.; Sukrak, K.; Kangwansupamonkon, W.; Panyametheekul, S. Chemical composition, sources, and health risk assessment of PM2.5 and PM10 in urban sites of Bangkok, Thailand. Int. J. Environ. Res. Public Health 2022, 19, 14281. [Google Scholar] [CrossRef] [PubMed]
  14. Johnston, H.J.; Mueller, W.; Steinle, S.; Vardoulakis, S.; Tantrakarnapa, K.; Loh, M.; Cherrie, J.W. How harmful is particulate matter emitted from biomass burning? A Thailand perspective. Curr. Pollut. Rep. 2019, 5, 353–377. [Google Scholar] [CrossRef] [Green Version]
  15. Vongruang, P.; Pimonsree, S. Biomass burning sources and their contributions to PM10 concentrations over countries in mainland Southeast Asia during a smog episode. Atmos. Environ. 2020, 228, 117414. [Google Scholar] [CrossRef]
  16. Kawichai, S.; Prapamontol, T.; Cao, F.; Song, W.; Zhang, Y. Source identification of PM2.5 during a smoke haze period in Chiang Mai, Thailand, using stable carbon and nitrogen isotopes. Atmosphere 2022, 14, 1149. [Google Scholar] [CrossRef]
  17. Peng-in, B.; Sanitluea, P.; Monjatturat, P.; Boonkerd, P.; Phosri, A. Estimating ground-level PM2.5 over Bangkok Metropolitan Region in Thailand using aerosol optical depth retrieved by MODIS. Air Qual. Atmos. Health 2022, 15, 2091–2102. [Google Scholar] [CrossRef] [PubMed]
  18. Wimolwattanapu, W.; Hopke, P.K.; Pongkiatkul, P. Source apportionment and potential source locations of PM2.5 and PM2.5–10 at residential sites in metropolitan Bangkok. Atmos. Pollut. Res. 2011, 2, 172–181. [Google Scholar] [CrossRef] [Green Version]
  19. Rincon, G.; Quintana, G.M.; Gonzalez, A.; Buitrago, Y.; Gonzalez, J.C.; Molina, C.; Jones, B. PM2.5 exceedances and source appointment as inputs for an early warning system. Environ. Geochem. Health 2022, 44, 4569–4593. [Google Scholar] [CrossRef]
  20. Cohen, D.D.; Bailey, G.M.; Kondepudi, R. Elemental analysis by PIXE and other IBA techniques and their application to source fingerprinting of atmospheric fine particle pollution. Nucl. Instrum. Methods Phys. Res. Sect. B 1996, 109–110, 218–226. [Google Scholar] [CrossRef]
  21. De La Cruz, A.H.; Roca, Y.B.; Suarez-Salas, L.; Pomalaya, J.; Tolentino, D.A.; Gioda, A. Chemical characterization of PM2.5 at rural and urban sites around the Metropolitan area of Huancayo (Central Andes of Peru). Atmosphere 2019, 10, 21. [Google Scholar] [CrossRef] [Green Version]
  22. Kanchanasuta, S.; Sooktawee, S.; Patpai, A.; Vatanasomboon, P. Temporal Variations and Potential Source Areas of Fine Particulate Matter in Bangkok, Thailand. Air Soil Water Res. 2020, 13, 1–10. [Google Scholar] [CrossRef]
  23. Santoso, M.; Lestiani, D.D.; Kurniawati, S.; Damastuti, E.; Kusmartini, I.; Atmodjo, D.P.D.; Sari, D.K.; Hopke, P.K.; Mukhtar, R.; Muhtarom, T.; et al. Assessment of urban air quality in Indonesia. Aerosol Air Qual. Res. 2020, 20, 2142–2158. [Google Scholar] [CrossRef]
  24. Hassan, H.; Latif, M.T.; Juneng, L.; Amil, N.; Khan, M.F.; Fujii, Y.; Jamhari, A.A.; Hamid, H.H.A.; Banerjee, T. Chemical characterization and sources identification of PM2.5 in a tropical urban city during non-hazy conditions. Urban Clim. 2021, 39, 100953. [Google Scholar] [CrossRef]
  25. Siregar, S.; Idiawati, N.; Lestari, P.; Berekute, A.K.; Pan, W.; Yu, K. Chemical composition, source appointment and health risk of PM2.5 and PM2.5-10 during forest and peatland fires in Riau, Indonesia. Aerosol Air Qual. Res. 2022, 22, 220015. [Google Scholar] [CrossRef]
  26. Zalakeviciute, R.; Rybarczyk, Y.; Granda-Albuja, M.G.; Suarez, M.V.D.; Alexandrino, K. Chemical characterization of urban PM10 in the Tropical Andes. Atmos. Pollut. Res. 2020, 11, 343–356. [Google Scholar] [CrossRef]
  27. Gómez-Plata, L.; Agudelo-Castañeda, D.; Castill, M.; Teixeira, E.C. PM10 source identification: A case of a coastal city in Colombia. Aerosol Air Qual. Res. 2022, 22, 210293. [Google Scholar] [CrossRef]
  28. Lestiani, D.D.; Santoso, M. Analytical methods INAA and PIXE applied to characterization of airborne particulate matter in Bandung, Indonesia. At. Indones. 2011, 37, 52–56. [Google Scholar] [CrossRef]
  29. Zhang, T.; Su, Y.; Debosz, J.; Noble, M.; Munoz, A.; Xu, X. Continuous measurements, and source apportionment of ambient PM2.5-bound elements in Windsor, Canada. Atmosphere 2023, 14, 374. [Google Scholar] [CrossRef]
  30. Shaltout, A.A.; Boman, J.; Hassan, S.K.; Abozied, A.M.; Al-Ashkar, E.A.; Abd-Elkader, O.H.; Yassin, M.A.; Al-Tamimi, J.H. Elemental composition of PM2.5 aerosol in a residential–industrial area of a Mediterranean Megacity. Arch. Environ. Contam. Toxicol. 2020, 78, 68–78. [Google Scholar] [CrossRef]
  31. Panda, S.; Nagendra, S.M.S. Chemical and morphological characterization of respirable suspended particulate matter (PM10) and associated heath risk at a critically polluted industrial cluster. Atmos. Pollut. Res. 2018, 9, 791–803. [Google Scholar] [CrossRef]
  32. Wiseman, C.L.S.; Levesque, C.; Rasmussen, P.E. Characterizing the sources, concentrations and resuspension potential of metals and metalloids in the thoracic fraction of urban road dust. Sci. Total Environ. 2021, 786, 147467. [Google Scholar] [CrossRef] [PubMed]
  33. Samek, L.; Stegowski, Z.; Styszko, K.; Furman, L.; Zimnoch, M.; Skiba, A.; Kistler, M.; Kasper-Giebl, A.; Rozanski, K.; Konduracka, E. Seasonal variations of chemical composition of PM2.5 fraction in the urban area of Krakow, Poland: PMF source attribution. Air Qual. Atmos. Health 2020, 13, 89–96. [Google Scholar] [CrossRef]
  34. Gope, M.; Masto, R.E.; George, J.; Balachandran, S. Tracing source, distribution and health risk of potentially harmful elements (PHEs) in street dust of Durgapur, India. Ecotoxicol. Environ. Saf. 2018, 154, 280–293. [Google Scholar] [CrossRef] [PubMed]
  35. Wang, Y.; Zhang, Y.; Li, X.; Cao, J. Refined source apportionment of atmospheric PM2.5 in a Typical City in Northwest China. Aerosol Air Qual. Res. 2021, 21, 200146. [Google Scholar] [CrossRef]
  36. Evagelopoulos, V.; Begou, P.; Zoras, S. In-depth study of PM2.5 and PM10 concentrations over a 12-year period and their elemental composition in the Lignite Center of Western Macedonia, Greece. Atmosphere 2022, 13, 1900. [Google Scholar] [CrossRef]
  37. Rahman, M.S.; Bhuiyan, S.S.; Ahmed, Z.; Saha, N.; Begum, B.A. Characterization and source apportionment of elemental species in PM2.5 with especial emphasis on seasonal variation in the capital city “Dhaka”, Bangladesh. Urban Clim. 2021, 36, 100804. [Google Scholar] [CrossRef]
  38. Saggu, G.S.; Mittal, S.K. Source apportionment of PM10 by positive matrix factorization model at a source region of biomass burning. J. Environ. Manag. 2020, 266, 110545. [Google Scholar] [CrossRef] [PubMed]
  39. Sillapapiromsuk, S.; Chantara, S.; Tengjaroenkul, U.; Prasitwattanaseree, S.; Prapamontol, T. Determination of PM10 and its ion composition emitted from biomass burning in the chamber for estimation of open burning emissions. Chemosphere 2013, 93, 1912–1919. [Google Scholar] [CrossRef]
  40. Popouen, A.J.; Benchrif, A.; Kezo, P.C.; Agbo, D.D.A.; Koua, A.A.; Bounakhla, M.; Monnehan, A.G. Elemental Composition of PM2.5 and PM10 in the Industrial Area of Yopougon, Abidjan, Côted’Ivoire. J. Environ. Prot. 2022, 13, 385–397. [Google Scholar] [CrossRef]
  41. Molnár, P.; Gustafson, P.; Johannesson, S.; Boman, J.; Barregård, L.; Sällstena, G. Domestic wood burning and PM2.5 trace elements: Personal exposures, indoor and outdoor levels. Atmos. Environ. 2005, 39, 2643–2653. [Google Scholar] [CrossRef]
  42. Xia, L.; Gao, Y. Characterization of trace elements in PM2.5 aerosols in the vicinity of highways in northeast New Jersey in the U.S. east coast. Atmos. Pollut. Res. 2011, 2, 34–44. [Google Scholar] [CrossRef] [Green Version]
  43. Juda-Rezler, K.; Reizer, M.; Maciejewska, K.; Błaszczak, B.; Klejnowski, K. Characterization of atmospheric PM2.5 sources at a Central European urban background site. Sci. Total Environ. 2020, 713, 136729. [Google Scholar] [CrossRef] [PubMed]
  44. Khan, J.Z.; Sun, L.; Tian, Y.; Shi, G.; Feng, Y. Chemical characterization and source apportionment of PM1 and PM2.5 in Tianjin, China: Impacts of biomass burning and primary biogenic sources. J. Environ. Sci. 2021, 99, 196–209. [Google Scholar] [CrossRef] [PubMed]
  45. Vaio, P.D.; Magli, E.; Barbato, F.; Caliendo, G.; Cocozziello, B.; Corvino, A.; Marco, A.D.; Fiorino, F.; Frecentese, F.; Onorati, G.; et al. Chemical composition of PM10 at urban sites in Naples (Italy). Atmosphere 2016, 7, 163. [Google Scholar] [CrossRef] [Green Version]
  46. Truong, M.T.; Nguyen, L.S.P.; Hien, T.T.; Pham, T.D.H.; Do, T.T.L. Source apportionment and risk estimation of heavy metals in PM10 at a Southern Vietnam megacity. Aerosol Air Qual. Res. 2022, 22, 220094. [Google Scholar] [CrossRef]
  47. Deng, J.; Jiang, L.; Miao, W.; Zhang, J.; Dong, G.; Liu, K.; Chen, J.; Peng, T.; Fu, Y.; Ahou, Y.; et al. Characteristics of fine particulate matter (PM2.5) at Jinsha Site Museum, Chengdu, China. Environ. Sci. Pollut. Res. 2022, 29, 1173–1183. [Google Scholar] [CrossRef]
  48. Bangar, V.; Mishra, A.K.; Jangid, M.; Rajput, P. Elemental Characteristics and source-apportionment of PM2.5 during the post-monsoon season in Delhi, India. Front. Sustain. Cities 2021, 3, 648551. [Google Scholar] [CrossRef]
  49. Onat, B.; Sahin, U.A.; Akyuz, T. Elemental characterization of PM2.5 and PM1 in dense traffic area in Istanbul, Turkey. Atmos. Pollut. Res. 2013, 4, 101–105. [Google Scholar] [CrossRef] [Green Version]
  50. Bie, S.; Yang, L.; Zhang, Y.; Huang, Q.; Li, J.; Zhao, T.; Zhang, X.; Wanga, P.; Wang, W. Source appointment of PM2.5 in Qingdao Port, East of China. Sci. Total Environ. 2021, 755, 142456. [Google Scholar] [CrossRef]
  51. Ahmad, M.; Rihawy, M.S.; Haydr, R.; Tlass, M.; Roumie, M.; Srour, A. PIXE and statistical analysis of fine airborne particulate matter (PM2.5) in Damascus. Nucl. Instrum. Methods Phys. Res. Sect. B 2020, 462, 75–81. [Google Scholar] [CrossRef]
  52. Li, J.; Chen, B.; Campa, A.M.; Alastuey, A.; Querol, X.; Rosa, J.D. 2005–2014 trends of PM10 source contributions in an industrialized area of southern Spain. Environ. Pollut. 2018, 236, 570–579. [Google Scholar] [CrossRef] [PubMed]
  53. Bencharif-Madani, F.; Ali-Khodja, H.; Kemmouche, A.; Terrouche, A.; Lokorai, K.; Naidja, L.; Bouziane, M. Mass concentrations, seasonal variations, chemical compositions and element sources of PM10 at an urban site in Constantine, northeast Algeria. J. Geochem. Explor. 2019, 206, 106356. [Google Scholar] [CrossRef]
  54. Jaafari, J.; Naddafi, K.; Yunesian, M.; Nabizadeh, R.; Hassanvand, M.S.; Ghozikali, M.G.; Shamsollahi, H.R.; Nazmara, S.; Yaghmaeian, K. Characterization, risk assessment and potential source identification of PM10 in Tehran. Microchem. J. 2020, 154, 104533. [Google Scholar] [CrossRef]
  55. Millán-Martínez, M.; Sánchez-Rodas, D.; Campa, A.M.; Rosa, J.D. Contribution of anthropogenic and natural sources in PM10 during North African dust events in Southern Europe. Environ. Pollut. 2021, 290, 118065. [Google Scholar] [CrossRef] [PubMed]
Figure 1. Location of sampling site.
Figure 1. Location of sampling site.
Atmosphere 14 00659 g001
Figure 2. Temporal variations of PM2.5 and PM10 mass concentrations and meteorological parameters (rainfall, maximum temperature, minimum temperature, air pressure, relative humidity and wind speed) at the Pathum Thani sampling site in the suburban area of Thailand from 18 February to 14 September 2021.
Figure 2. Temporal variations of PM2.5 and PM10 mass concentrations and meteorological parameters (rainfall, maximum temperature, minimum temperature, air pressure, relative humidity and wind speed) at the Pathum Thani sampling site in the suburban area of Thailand from 18 February to 14 September 2021.
Atmosphere 14 00659 g002
Figure 3. Variation of elements concentrations of PM2.5 and PM10 samples over Pathum Thani Province from 18 February to 14 September 2021.
Figure 3. Variation of elements concentrations of PM2.5 and PM10 samples over Pathum Thani Province from 18 February to 14 September 2021.
Atmosphere 14 00659 g003
Figure 4. Metals concentration variations of PM2.5 and PM10 in summer and wet seasons. The line in the box represents the median, the cross shows the mean values, the bottom and top of the vertical lines represent the values of the 5th and 95th percentiles, respectively, the dot denote the data beyond the mean values).
Figure 4. Metals concentration variations of PM2.5 and PM10 in summer and wet seasons. The line in the box represents the median, the cross shows the mean values, the bottom and top of the vertical lines represent the values of the 5th and 95th percentiles, respectively, the dot denote the data beyond the mean values).
Atmosphere 14 00659 g004
Figure 5. Source identification of PM2.5 and PM10 derived from PCA analysis over Pathum Thani area during February–September 2021.
Figure 5. Source identification of PM2.5 and PM10 derived from PCA analysis over Pathum Thani area during February–September 2021.
Atmosphere 14 00659 g005
Figure 6. Dendrogram of the cluster analysis for PM2.5.
Figure 6. Dendrogram of the cluster analysis for PM2.5.
Atmosphere 14 00659 g006
Figure 7. Dendrogram of the cluster analysis for PM10.
Figure 7. Dendrogram of the cluster analysis for PM10.
Atmosphere 14 00659 g007
Table 1. Sampling parameters.
Table 1. Sampling parameters.
ParametersData
Sampling site14.24° S, 100.43° E
Sampling period18 February 2021–14 September 2021
Summer season (18 February 2021–15 May 2021)
Wet season (16 May 2021–14 September 2021)
Total number of samples collected
PM2.577 (31 for summer; 46 for wet period)
PM1049 (21 for summer; 28 for wet period)
Average sampling time (h)24
Average flow rate (L min–1)16.67
Average volume (m–3)24
Table 2. Parameters for PIXE analysis used in this study.
Table 2. Parameters for PIXE analysis used in this study.
Analysis EquipmentTandetron Tandem Accelerator at Department of Physics and Materials Science, Faculty of Science, Chiang Mai University, Thailand
Proton energy2 MeV
DetectorSi(Li)
Beam intensity2 nA
Beam spot area1 mm2
Analysis time1000 s
Analysis softwareGUPIX
Table 3. Summary statistics of PM2.5 and PM10 mass concentration and meteorological parameters in Pathum Thani during the sampling period from 18 February to 14 September 2021.
Table 3. Summary statistics of PM2.5 and PM10 mass concentration and meteorological parameters in Pathum Thani during the sampling period from 18 February to 14 September 2021.
VariableMeanMinMaxSD
PM2.5 (μg/m3)48.100.39174.2646.97
PM10 (μg/m3)65.9012.75242.0247.51
Rainfall (mm)4.150.0074.8011.32
Min Temperature (°C)25.8820.5029.001.24
Max Temperature (°C)35.7030.5039.502.10
Air pressure (mmHg)756.60753.51763.031.70
Relative humidity (%)88.9670.0098.008.17
Wind speed (m/s)0.350.002.500.52
Table 4. PM2.5 and PM10 mass concentrations of this study and several tropical countries.
Table 4. PM2.5 and PM10 mass concentrations of this study and several tropical countries.
Study AreaLocation TypeSampling PeriodPM Size FractionAverage Conc. (μg/m3)References
Pathum Thani, ThailandSuburbanFeb 2021–Sep 2021PM2.548 ± 47This study
PM1066 ± 48
Bangkok, ThailandUrbanAug 2017–Mar 2018PM2.577 ± 21[12]
Chiang Mai, ThailandUrbanJan 2017–Mar 2017PM2.536 ± 16[16]
Chiang Rai, ThailandPeri-urbanJul 2021–Aug 2021PM2.563 ± 26[10]
Pekanbaru, IndonesiaUrbanApr 2014–Jul 2014PM2.563 ± 3[25]
PM1028 ± 2
Jayapura, IndonesiaSuburbanJan 2017–Dec 2017PM2.56 ± 3[23]
PM1012 ± 3
Barranquilla, ColombiaUrbanMar 2016–Apr 2016PM1052 ± 11[27]
Belisario Quito, EcuadorUrbanJan 2017–Oct 2017PM1025 ± 9[26]
Los Chillos, EcuadorIndustrial 26 ± 11
Tababela, EcuadorPeri-urban 26 ± 5
Petaling Jaya, MalaysiaSuburbanJan 2017–Feb 2018PM2.528 ± 10[24]
Table 5. Elemental concentrations of PM2.5 and PM10 in Pathum Thani Province in ng/m3.
Table 5. Elemental concentrations of PM2.5 and PM10 in Pathum Thani Province in ng/m3.
ElementMDLPM2.5PM10
MeanMinMaxSDMeanMinMaxSD
Si0.3117.901.0872.6717.5015.181.6757.7112.94
S0.2620.860.71107.9221.1317.701.4450.0514.42
Cl0.246.630.4830.256.2615.441.2086.2017.27
K0.6674.791.23355.0173.8170.7910.07220.6751.69
Ca1.1730.711.51137.6428.78139.1329.18485.9490.23
Ti0.227.040.30125.0417.8810.510.9237.608.42
Cr0.5765.431.44948.16163.74113.590.751033.32243.05
Mn0.3915.340.54125.8124.0014.121.33124.4024.34
Fe1.42157.741.763879.18491.03291.8222.424145.35661.64
Zn1.7821.172.3076.8219.7924.722.9582.0818.75
Ni1.3118.801.8494.0625.1571.783.78595.56152.15
Cu1.227.771.6935.085.9510.242.3153.3410.74
Table 6. Pearson’s correlation coefficients relating the mass of PM2.5 and the elements concentrations of PM2.5.
Table 6. Pearson’s correlation coefficients relating the mass of PM2.5 and the elements concentrations of PM2.5.
MassSiSClKCaTiCrMnFeZnNiCu
Mass1
Si−0.2031
S−0.2560.6011
Cl−0.2570.5780.5111
K−0.2300.5470.7410.6141
Ca−0.2520.6320.5770.5560.6291
Ti−0.1520.0290.0640.2100.4920.1891
Cr−0.1570.037−0.009−0.146−0.720−0.0180.3011
Mn−0.2070.0330.063−0.0590.0390.0730.4440.9671
Fe−0.1400.1230.088−0.0420.0030.0240.0410.9900.9491
Zn−0.011−0.0070.2620.2090.2300.384−0.117−0.153−0.062−0.0851
Ni−0.081−0.239−0.117−0.286−0.112−0.0080.7250.9660.9470.968−0.1171
Cu−0.0250.0940.2120.1750.1940.2020.0940.0990.5970.1330.524−0.0211
Shaded correlation coefficients are statistically significant at p < 0.05.
Table 7. Pearson’s correlation coefficients relating the mass of PM10 and the elements concentrations of PM10.
Table 7. Pearson’s correlation coefficients relating the mass of PM10 and the elements concentrations of PM10.
MassSiSClKCaTiCrMnFeZnNiCu
Mass1
Si0.0711
S−0.1670.5081
Cl0.0120.4800.2521
K−0.1070.3930.7370.2591
Ca−0.0650.6250.6850.3860.8041
Ti−0.1400.2650.3670.2130.5350.5201
Cr0.1510.3540.187−0.022−0.182−0.149−0.1811
Mn−0.0800.2970.189−0.031−0.024−0.031−0.1680.9711
Fe−0.0590.3670.2610.0150.0210.030−0.0520.9930.9701
Zn0.084−0.0370.2960.0050.3830.4840.090−0.0500.028−0.0081
Ni0.0410.6960.3710.021−0.146−0.209−0.2060.9690.9140.984−0.3061
Cu−0.1670.2790.1760.1210.0770.4410.2740.7460.1250.1450.2040.0491
Shaded correlation coefficients are statistically significant at p < 0.05.
Table 8. Summary of principal component analysis (PCA) of elements in PM2.5 and PM10 in Pathum Thani Province.
Table 8. Summary of principal component analysis (PCA) of elements in PM2.5 and PM10 in Pathum Thani Province.
ElementPM2.5PM10
Factor 1Factor 2Factor 3Factor 4Factor 1Factor 2Factor 3
Si0.8310.123−0.0520.1090.3050.370−0.719
S0.8430.0810.140.0470.2190.790−0.222
Cl0.724−0.0490.033−0.181−0.0870.171−0.743
K0.794−0.0060.181−0.413−0.0400.914−0.098
Ca0.8070.0400.186−0.106−0.0540.858−0.380
Ti0.1720.0850.058−0.952−0.1490.646−0.223
Cr0.0020.9850.0450.0240.990−0.052−0.085
Mn0.1340.9720.057−0.0050.9690.059−0.051
Fe0.0120.9540.0970.0530.9890.047−0.099
Zn0.242−0.0490.7920.0830.0380.6950.178
Ni0.0250.637−0.098−0.1690.973−0.057−0.117
Cu0.0280.0940.817−0.1550.162−0.006−0.625
EV3.8863.0391.2271.0504.3513.5511.199
V (%)32.425.310.28.736.329.610.0
CV (%)32.457.767.976.736.365.875.8
EV = eigenvalue; V = variance contribution rate (%); CV = cumulative variance (%).
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Nuchdang, S.; Kingkam, W.; Tippawan, U.; Sriwiang, W.; Fungklin, R.; Rattanaphra, D. Metal Composition and Source Identification of PM2.5 and PM10 at a Suburban Site in Pathum Thani, Thailand. Atmosphere 2023, 14, 659. https://doi.org/10.3390/atmos14040659

AMA Style

Nuchdang S, Kingkam W, Tippawan U, Sriwiang W, Fungklin R, Rattanaphra D. Metal Composition and Source Identification of PM2.5 and PM10 at a Suburban Site in Pathum Thani, Thailand. Atmosphere. 2023; 14(4):659. https://doi.org/10.3390/atmos14040659

Chicago/Turabian Style

Nuchdang, Sasikarn, Wilasinee Kingkam, Udomrat Tippawan, Wiranee Sriwiang, Ratchai Fungklin, and Dussadee Rattanaphra. 2023. "Metal Composition and Source Identification of PM2.5 and PM10 at a Suburban Site in Pathum Thani, Thailand" Atmosphere 14, no. 4: 659. https://doi.org/10.3390/atmos14040659

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop