Introduction

Atmospheric particulate matter reflects the presence of solid or liquid particles in the atmosphere. Its composition is intricate, with diverse sources. Due to their smaller volume and larger specific surface area, PM2.5 particles can absorb more harmful substances that may enter the human body through the respiratory system, posing risks to human health1. Some researchers have utilized electron SEM technology to capture microscopic images, studying the microscopic morphology and surface features of PM2.5 in detail, leading to a deeper understanding of the harm caused by PM2.52.

In recent years, single-particle analysis technology has provided micro-morphology and composition information of specific particles intuitively. It offers the advantages of short sampling time and small sample quantity. This technology is widely used in PM2.5 studies, including atomic force microscopy3, scanning electron microscopy, and transmission electron microscopy4,5. X-ray fluorescence spectrum (XRF)6: among various single-particle analysis techniques, electron microscopy (scanning electron microscope and transmission electron microscope) is the only method that can simultaneously observe the morphology and analyze the composition of PM2.57,8,9,10. It plays a significant role in studying the morphology, composition, particle size distribution, particle aggregation characteristics, and source analysis of particulate matter. Compared with transmission electron microscopy and scanning electron microscopy, the sample requirements are simple, and the analysis speed is fast, making it more suitable for PM2.5 studies. Scanning electron microscopy technology with an energy spectrum has been widely applied in mineral and material analysis11,12,13.

Given this, this paper adopts scanning electron microscopy (SEM/ED) analysis technology to conduct a comprehensive study on atmospheric particulate matter PM2.5 in Yining City during 2018–2019. The focus is on the mild pollution weather in winter to reveal its morphological characteristics, main chemical composition, chemical composition of single amorphous particles, and trace element characteristics. After 2018, heavy pollution occurred several times in Yining City. The research results from 2018 to 2019 provide certain data references for the comprehensive control of PM2.5 and the continuous improvement of air pollution.

The study provides new insights and understanding of PM2.5 concentration, regional tectonic and geomorphological dynamics, and potential climate and environmental impacts in Yining City. Fine particles from human activities and nature can enter the atmospheric circulation and influence regional and even global climate14. Xiaoyu Jiao using U isotopes, Results show that dust originating from the TP undergoes long-range transport and constitutes potentially significant component of the Asian and Northern Hemisphere atmospheric dust load15.

Materials and methods

Sample collection

The sampling point was selected on the roof of the Film Environmental Protection Bureau building (43° 92' N, 81° 23' E, 125 m), approximately 15 m above the ground and 300 m away from the South Ring Road, with no factories or enterprises nearby. This sampling point can be considered representative of the thorough mixing of atmosphere and traffic flow in the downtown area of Dining. The sampling period for four PM2.5 samples was from June 2018 to July 2019. The glass fiber filter membrane was preheated at 450 °C for 4 h and then placed in a drying dish after cooling. A TH-150C medium-flow PM2.5 cutting head was used for sampling; the sampling flow rate was 100 L/min, and meteorological parameters such as on-site temperature and air pressure were recorded. After sampling, the filter membrane must be equilibrated in a constant temperature and humidity test chamber for 24 h before analysis.

Sample analysis

Using the Hitachi JSM-7500FSEM scanning electron microscope (SEM), the maximum resolution is 1.0 nm (15 kV) and 1.4 nm (1 kV). The magnification is 1000×, the working voltage is 20 kV, the acceleration current is 15 MA, and the working distance is 15 nm. The micromorphology and composition of the collected PM2.5 particles were analyzed. The glass fiber filter membrane itself interferes with the energy spectrum analysis. The premise of the energy spectrum analysis assumes that the laser only falls on the particles. To prepare the samples, use pollution-free stainless steel scissors to gently cut about 1/10 of each sample filter film. Fix it on the sample tray and cut out an appropriate size of conductive adhesive with stainless steel scissors. Paste the clipped filter film sample on the metal post in order with conductive adhesive, press it flat, place the metal column with the sample in the automatic ion sputtering instrument, and plate 20 nm thick gold under vacuum conditions. To improve the electrical conductivity of the sample, it must meet the requirements of electron microscope observation. Fix the metal column with the sample in the sample table, level it (ensuring the surface of the metal column is at the same level as the sample table), and note the fixing order of these samples. The sample table of the prepared sample was placed in the SEM for observation and analysis, with magnifications ranging from 1000 to 10,000 times.

PM2.5 concentration

See Table 1.

Result and analysis

Evaluation of ambient air quality for PM2.5

Firstly, literature analysis was used in this study. Through literature collection, the development history of China's Air Quality Index and information resources were sorted out. The latest progress of China's AQI-related work was deeply analyzed, along with its evaluation indicators and classification concentration limits. Then, based on the Air Quality Index (IRAQI) and AQI calculation method given in the "Environmental Air Quality Index (AQI) Technical Regulations (Trial)" issued by the Ministry of Environmental Protection, the data on PM2.5 from June 2018 to July 2019 were statistically analyzed16. Secondly, one-fifth of the samples were weighed (see Figs. 1, 2), and then the concentration and AQI value of atmospheric particulate matter PM2.5 were calculated according to the formula; refer to Table 1.

Figure 1
Figure 1
Full size image

Trend of PM2.5 concentration in 2018.

Figure 2
Figure 2
Full size image

Trend of PM2.5 concentration in 2019.

Table 1 Sampling time, PM2.5 index, corresponding concentration value, and AQI value statistical table.
Full size table

Figure 1 illustrates the weight trend of the selected samples in 2018. It can be inferred from the figure that the weight of the samples decreased significantly in August, indicating a decrease in atmospheric particulate matter PM2.5 pollution and its impact. The weight of samples notably increased in November and December, suggesting high pollution levels of atmospheric particulate matter PM2.5 in December and February of the following year, with light pollution weather. Possible reasons for this include the lush trees in summer, which have an adsorption effect on PM2.5, and the stable summer climate that prevents PM2.5 accumulation. Additionally, winter meteorological conditions hinder PM2.5 diffusion, leading to continuous accumulation near the ground due to increased ground inversion frequency, elevating PM2.5 content. Figure 2 displays the sample weight trend in 2019, showing a significant increase in February. This could be attributed to the low local temperature in February, which hinders diffusion. Additionally, the formation of a low-pressure air mass in February may impede PM2.5 dispersion.

Morphological characteristics of PM2.5

Morphological characteristics of PM2.5 in a non-heating period

To observe the microscopic morphology of PM2.5, scanning electron microscopy (SEM) was used to analyze the collected samples. Figure 3 displays SEM images of PM2.5 particles captured at various magnifications (1000–10,000 times). The figure includes images labeled as follows: a: 2018.6.5; b: 2018.7.13; c: 2019.5.6; d: 2019.5.10.

Figure 3
Figure 3
Full size image

Microtomography of PM2.5 on non-polluted days.

The particle morphology of PM2.5 observed by SEM is shown in Fig. 3. The fibers and their attachment to PM2.5 particles can be seen, and the fibers are quartz fiber filter membranes. As can be seen from the PM2.5 figures at different magnifications, PM2.5 particles do not adhere uniformly to the membrane surface but loosely to it, and the membrane voids are visible. A closer look reveals that PM2.5 particles are mostly long chains, with chain-like aggregates of soot generally coming from motor vehicle exhaust or coal-burning. The coal-burning fly ash is spherical; regular spheres are less common in chains. In the non-heating season, the central and residential heating boilers in Yining have all stopped working. The source of particulate matter may be mainly motor vehicle exhaust. Some scholars have studied Yining. There are irregular clumps attached to the long chain. Zhang Qianghua and others17 think the irregular clumps are mainly smoke-dust complexes, biomass, and primary minerals. Some researchers18 have studied the micro-morphology of PM2.5 in many cities of Guizhou and found that the irregular aggregates are mostly composed of sulfate. According to the references, the clumps are not the same in different areas. Further analysis of irregular clumps requires the combination of microscopic Raman spectroscopy and Chemometrics methods19, or TEM EDX20 and aerosol time-of-flight mass spectrometry (ATO FMS) observations, which is my future research direction.

Morphological characteristics of PM2.5 in the heating season

Figure 4 shows the SEM images of PM2.5 particles collected at different magnifications (1000–10,000 times). Figures a: 2018.12.2; b: 2018.12.15; c: 2019.2.11; d: 2019.2.24.

Figure 4
Figure 4
Full size image

Microtomography of PM2.5 in light-pollution days.

Through SEM observation, the particle morphology characteristics of PM2.5 are depicted in Fig. 4. The glass fibers in the filter membrane adsorb particles with complex shapes, varying sizes, and different forms. The particles mainly exhibit a few large-volume long strips, small-volume irregular shapes, spherical particles (such as coal fly ash and tarball), oval shapes, cluster-like dust aggregates (chain, fluffy, and dense), flocculant, patellar and mineral particles (both regular and irregular), and other forms. Through SEM observation and comparison with literature data, ITIS concluded that the atmospheric particles in Yining City primarily consist of soot aggregates, coal-fired fly ash, motor vehicle exhaust dust, industrial dust, and other unidentified particles. Additionally16, temperature inversion is less likely to occur during non-heating periods, and more frequent rainfall and strong winds facilitate the dispersion and elimination of PM2.5.

Compared with the scanning results during the non-heating period, on mild PM2.5 pollution days in Yining City, there was an increase in the number of spherical and chain particles. More bar or irregular block particles were detected, and some surfaces showed soot aggregates. In the classification of PM2.5 particles, they can be categorized into three types: large-volume elongated strips, small-volume cubes, and small-volume cubes with an amorphous structure. The elongated strips are primarily sourced from coal dust, vehicle exhaust dust, construction cement dust, industrial dust, soil dust, and road dust. The amorphous structure mainly originates from the characterization of secondary pollutants. The main reasons for this situation are as follows: firstly, the high concentration and intensity of local pollutant emissions, notably the significant increase in coal burning during the heating period; secondly, the rise in exhaust emissions from motor vehicles due to low temperatures, leading to an increase in the release of PM2.5 and its precursors (sulfur dioxide, nitrogen oxides, etc.); and thirdly, unfavorable meteorological conditions for the dispersion of atmospheric pollutants. The increased occurrence of ground temperature inversion in winter causes pollutants to continuously accumulate near the ground, resulting in the progressive buildup of PM2.5 from primary emissions and secondary conversion in the lower atmosphere, reaching high concentration levels.

Natalie Mahowald's research focuses on evaluating the relative contributions of different dust sources to the dust sink using an end-member mixed analysis model and constrained least squares methods to derive the proportion of dust in each sediment or dust sample. The comparison of model data is designed to produce a consistent view across different time scales and wider spatial and environmental domains. So far, most efforts have focused on emissions and airborne dust. These smaller particles (< 50 μm) are entrained into the boundary layer, from where they can be transported over long distances21. The continuous accumulation of pollutants near the ground in winter may be the common result of increased emissions of local pollutants, dust, and long-distance transport, which needs further study.

The chemical composition of PM2.5 was analyzed using energy spectrum analysis

To determine the composition of each element in the sample PM2.5 (refer to Fig. 5), this study conducted an energy spectrum analysis of the sample. In Figure (a): December 2, 2018, (b): February 11, 2019, and (c): June 8, 2019.

Figure 5
Figure 5
Full size image

Energy Spectrum Analysis of Particulate Matter PM2.5

To understand the main chemical composition of PM2.5, the sample's chemical composition was analyzed through energy spectrum scanning. The results revealed that the primary chemical components of PM2.5 included O, S, P, C, N, Na, Ge, Rb, Fe, Mg, Al, and others.

Table 2 displays the atomic percentages in the energy spectrum analysis diagram. The analysis reveals that element C has a higher atomic ratio, reaching up to 16.48% and 12.14% in (a) and (b) respectively. The dominance of coal combustion in the energy structure leads to the oxidation of hydrocarbon emissions from combustion, resulting in the production of various organic acids, polycyclic aromatic hydrocarbons, and other organic particles formed through homogeneous and heterogeneous nucleation. Some samples may lack certain elements, with no fixed pattern, possibly linked to automobile exhaust emissions and other urban pollution sources. It is theorized that gaseous NOx and SO2 emitted from vehicle exhausts enter the atmosphere and form through nucleation under specific physical chemistry conditions. Elements Na and Mg may be present in the form of soluble inorganic ions like Na+ and Mg2+, potentially originating from soil dust. Chlorine (Cl) is absent in the energy spectrum analysis, with its presence typically associated with combustion and serving as an indicator of industrial soot pollution characteristics. Yining's industrial enterprises are actively promoting clean energy and have made significant progress in controlling industrial soot emissions22.

Table 2 Results of the Energy Spectrum Analysis Diagram.
Full size table

Conclusions

PM2.5 is a significant threat to public health, and China prioritizes the regulation of PM2.5. The aim of this research was to examine the morphological characteristics and primary components of PM2.5. The study utilized scanning electron microscopy to observe and analyze the microscopic structure of fine particles in the atmosphere from 2018 to 2019. The findings revealed that during mildly polluted conditions, glass fibers intertwined to create a tubular film. Particulate matter loosely adhered to the glass film's surface, forming larger three-dimensional shapes, as well as smaller long strips and irregular shapes. The composition included over ten trace elements. Moreover, the adsorption rate notably increased in winter, suggesting that mild pollution in the urban area was primarily due to winter heating. Automobile exhaust emissions emerged as a significant pollution source in the urban area, and there might also be long-distance input of PM2.5.