Waste Disposal & Sustainable Energy

, Volume 1, Issue 2, pp 143–150 | Cite as

Formation and growth mechanisms of ultrafine particles in sludge-incineration flue gas

  • Yanlong Li
  • Jiaqi Man
  • Zhengquan Fang
  • Yunbin Zhao
  • Feng Wang
  • Rundong LiEmail author


Atmospheric particulate matter with diameter < 2.5 μm now makes up much of the air pollution in China, but it is the ultrafine particles (UFPs) with diameter < 90 nm that are of particular interest. This is because UFPs are strongly linked with human health for two reasons: they contain a variety of hazardous substances and they can deeply penetrate human respiratory systems. Therefore, scanning electron microscopy combined with X-ray dispersive energy spectrometry was used to characterize the morphology and surface texture, as well as the elemental composition of 60 UFPs. The UFPs was generated in a sewage sludge-incineration power plant in Zhejiang Province. This was done to determine the microstructure of the ultrafine particles and to follow the evolution of particle surface elemental composition with increasing particle size. Then, a comparison of the characteristic time for nucleation, condensation and coagulation was done to estimate the dominant mechanism. The results show that the UFPs have generally irregular shapes (cotton-like, irregular balls, sheets, etc.) and that they usually aggregate to form a mass. With increase in the size of a UFP, the mass fraction of the elements presents clearly changed: Na, K and Fe gradually decreased; while Ca, Si and Al, as well as the heavy metals Cu, Zn and Ni increased. Characteristic time estimation is a convenient and effective tool for identifying the predominant mechanisms during combustion. In this study, calculations of characteristic time were used to reveal a mechanism of vaporization, nucleation, condensation and coagulation, which drives the formation and growth of ultrafine particles.


Ultrafine particle Characteristic time Nucleation Condensation Coagulation 


Sewage sludge causes serious environmental pollution because it contains some harmful and toxic substances, including viruses, bacteria, dioxins, non-biodegradable organic compounds and heavy metals [1, 2]. Incineration has been considered the most thorough, quick and economical way of sludge disposal due to its advantages on stabilization, volume reduction and resource recovery [3, 4]. Nevertheless, even though incineration has advantages of the reduction of sewage sludge volume and lower disposal costs. On account of almost 30% of the solids occur as incineration residues in the forms of fly ash and air pollution control (APC) residues [5], it is still not a perfect method to deal with sludge In recent decades, fly ash, which contains higher concentrations of toxic heavy metals [6, 7] and dioxin [8], has become an increasingly important issue [9]. Especially revealed of the ultrafine particles (UFPs, diameter < 90 nm). Typically, UFPs account for about 90% of the total number–concentration of particles in the atmosphere [10, 11] and exhibited even more severe toxicity relative to larger particles of the same composition [12]. The ultrafine particles are strongly linked with adverse health effects, including cardiovascular and pulmonary ailments and premature deaths in people with heart or lung disease [13]. For these reasons, UFP air pollution has become a topic of great interest in terms of air quality harm, public health and global climate. Of particular interest, are the processes of driving formation of ultrafine particles.

Many scholars have carried out research on the physical and chemical properties of ultrafine particles. Abramesko and Tartakovsky [14] investigated the UFP air pollution inside diesel-propelled passenger trains in order to reduce train passenger exposure to these harmful particles. Buonanno [15] characterized the UFPs emitted at the stack of a waste-to-energy plant from a dimensional, chemical and morphological point of view. He found that the particle size number distribution at the stack presented a mode at about 90 nm, whereas the one measured before the fabric filter showed a higher value (about 150 nm). However, there is little information about the formation of ultrafine particles in the process of sludge incineration. Typically, particulates are characterized using X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and scanning electron microscopy (SEM). These technologies are widely employed for studying the microstructure, morphology and surfaces of solids, respectively. Through XRD it is possible to identify the microstructural state of particles (if the particle is crystalline or amorphous) and to identify the different crystalline phases. SEM offers the possibility to relate visually the morphology and size. XPS provides qualitative and semi-quantitative information regarding the chemical composition, oxidation state and bonding energy of the elements composing the particles [16].

Although conventional analytical techniques such as XRD, XPS, SEM, XRF, Raman spectroscopy, ICP-AES and ICP-MS allowed to evaluation the morphology and average concentration of trace elements in bulk samples, they do not provide information on the microstructure or chemical composition of a single ultrafine particle [17]. Single-particle characterization can provide information on the evolution of size distribution and chemical composition of UFPs. Previous studies on UFPs have been focused on their classification based on size, concentration and chemical composition of whole particle masses, with relatively few details on the size, shape and chemical composition of individual particles.

For the reasons given above, field emission scanning electron microscopy (FE-SEM) combined with X-ray dispersive energy spectrometry (EDS) was used to characterize UFPs. At high magnification, FE-SEM offers a quick and easy way of analyzing particular matter (PM) down to 10 nm in diameter, and can provide brilliant images of the particle morphology, shape, size and roughness. Chemical information as well, can be acquired for individual particles using energy dispersive spectroscopy (EDS) [18]. In this work, the UFPs were not present individually, but were aggregated to form larger particles. For this reason, particles of < 2.5 μm were obtained by sieving, and more than 50 images of PM2.5 were obtained. From these, the morphology and chemical composition of 60 UFPs adsorbed to larger particles were analyzed using SEM–EDS to characterize the evolution of UFPs.

In addition, quantitative simulation of UFP formation was provided using mathematical models. Kulmala et al. [19] used measurements during nucleation episodes of evolving size distributions down to 3 nm to calculate the apparent source rate of 3 nm particles and the particle growth rate. Wu and Biswas [20] compared the characteristic times of nucleation and condensation, to determine the fate of metallic species in a combustion environment. In such cases, there are many processes of particle formation and growth occurring simultaneously due to the high temperatures of the combustion environment (e.g., nucleation, condensation, coagulation and so on). For better understanding of them, it was necessary to identify the relative importance of these mechanisms in the combustion environment by comparing their characteristic times.

Materials and methods

Sample preparation

The particles used in these experiments were collected in a sewage sludge-incineration power plant in Zhejiang Province, China. The particles were collected from bag filters, which are very common at sludge-incineration plants. First, the samples were randomly collected over 3 days. The mass of each sample was 5 kg to guarantee a representative particle sample. Then according to the centrifugal principle, an air classifier was used to separate particles of different size. The rotation speed of the induced-draft fan was adjusted to regulate the size of the internal negative pressure traction of the equipment. This guaranteed the fineness and precision of the grading treatment, and we obtained a fine powder of approximately 2.5 μm. Compared with the use of sizing screens, high-precision air classification equipment causes minimal damage to the original particles; thus, the accuracy of the final classification was improved significantly.

Composition and phase analysis of the particles

The chemical composition of the sample was analyzed using an X-ray fluorescence spectrum analyzer (ZSX100e), as shown in Table 1. X-ray diffraction (PRO/MPD) was used to analyze the mineral phase of the particle, as shown in Fig. 1.
Table 1

Main components of a ultrafine particles (wt%)



















Fig. 1

Mineral phase of the ultrafine particles

Microstructure and surface composition analysis of the particle

In this investigation, the morphology, crystal structure, surface topography and chemical composition of 60 UFPs were investigated using a Nova NanoSEM450 FE-SEM. The electron microscope resolution was 2 nm and the magnification was ~ 7 × to ~ 500,000 ×. The FE-SEM was fitted with a backscatter electron detector that not only gave visual imaging magnification up to 500,000 ×, but could also distinguish electron density in the sample. That is to say, heavier elements appeared brighter and the lighter elements appeared darker. Having the FE-SEM fitted with EDS-detectors provides the option of scanning an area or a spot for elemental composition. The atomic content of particles fixed on a substrate can in this way be analyzed by focusing the electron beam on an isolated particle and detecting the resulting X-ray emissions [21]. Thus, the images of the particles and the specific sizes of 60 UFPs were obtained. The surface chemical composition of individual particles was obtained using EDS, and the elements detected included oxygen, sodium, potassium, iron, calcium, silicon, aluminium, sulphur and trace elements.

Results and discussion

Ultrafine particles characteristics

The main chemical components of the PM2.5 were determined by X-ray fluorescence spectrometer (XRF) and normalization methods. The results from Table 1 show that the PM2.5 collected was mainly composed of the elements C, O, Fe, Al, Ca, Si, S and P. The particles contained more sulphur because, in the process of sludge incineration, the sludge was mixed with coal to improve its calorific value. The high calcium content in the particles may be because, to control the acid gas generated during burning, a certain amount of lime was added. The XRD results regarding the PM2.5 fraction showed that the main components were silicon dioxide and iron oxide; followed by calcium sulphate. Aluminosilicates were present in lower proportion, and occasionally, aluminium dioxide was also found.

Next, the SEM–EDS was used to provide detailed image information on the morphology and surface texture of individual particles, as well as the surface elemental composition of the UFPs. As noted by Vassilev and Vassileva [22], SEM is one of the best and most widely used techniques for the chemical and physical characterization of fly ash.

Each individual UFP was characterized by randomly selecting 3–4 fields of view, from which the surface elemental composition and morphology were obtained and compiled. The morphology of the particles is shown in Fig. 2. It can be seen that the particles are generally irregular in shape (cotton-like, irregular balls, sheets, etc.) and are usually aggregated to form a mass. Moreover, the particles are mostly amorphous and polycrystalline polymers, neat crystals are rare.
Fig. 2

The morphology of the ultrafine particles

The EDS detected O, K, Na, Ca, Si, S, Al, Fe, Cu, Zn and other heavy metal elements in the 60 UFPs. The diameters of some UFPs are indicated in Fig. 3.
Fig. 3

The diameter of the ultrafine particles

Apparent composition analysis of ultrafine particles

The EDS results characterizing the surface elemental composition of 60 UFPs are shown in Fig. 4. All the mass fractions of the elements detected were based on inorganic minerals contained in the UFP samples, and the amount of carbon (including organic carbon and inorganic carbon) was not considered. The elemental O was also not considered. It can be seen that the surface elemental composition of the UFP changed significantly with increase of the particle size from 10.17 to 155.9 nm.
Fig. 4

The surface elemental composition of 60 ultrafine particles

Figure 4a shows the mass fractions of alkali metals: Na and K gradually decreased with increase in the particle size from the initial fractions (1.5%) to the last (0.2%). Elemental iron showed the same obvious trend, decreasing from the initial (37%) to the final (18%) condition (Fig. 4b). However, the mass fractions of the main elements Ca, Si and Al gradually increased with increase in the particle size. There were some calcium-rich particles of size 71.69 nm. The calcium-rich particles usually showed more roundness than other particles [23]. There were also Si-rich particles (95.47 nm and 139.8 nm), which were characterized by a prismatic, regular morphology typically observed in crystalline materials [16]. Figure 4c, d shows the mass fractions of heavy metals. The fractions of Cu, Zn and Ni gradually increased with increase in particle size, while the mass fraction of Pb, Cd, Cr and As tended to be constant. The heavy metal mass fraction was very low: < 1.5%.

Many scholars have carried out researches on the formation mechanism of fly ash. Seames [24] proposed a tri-modal particle size distribution that included a submicron fume region to describe the particle formation. Wang et al. [25] considered that PM1 is formed from vaporization and then subsequent nucleation and accumulation of material. In more detail, Li et al. [26] found that UFPs mostly formed by vaporization and nucleation of mineral metals associated with the volatiles. These gas-phase minerals came mostly from two sources: the volatilization of mineral components present in water-soluble or ion-exchangeable form (Na, Cl, etc.), and vaporized gas from mineral inclusions through a reduction reaction as MOn (solid) + CO = MOn-1 (vapor) + CO2. In this case, MOn and MOn-1 refer to the refractory oxides (SiO2, CaO, etc.) and the corresponding volatile sub-oxides (SiO) or metal vapous (Ca), respectively [27, 28, 29]. It can be seen from Fig. 4 that, with increase in particle size, the mass fractions of the volatile alkali metal elements Na and K gradually decreased. This indicates that the initially occurring Na and K nucleation was homogeneous. This was followed by Fe, which provided heterogeneous nucleation of secondary nanoscale particles. Then, secondary nanoscale particles were formed by coagulation, heterogeneous condensation and collision. During this process, elemental Si, Ca and Al condensed to form nanoscale particles, so the mass fractions of Si, Ca and Al gradually increased.

With regard to heavy metal concentrations, the elements with higher boiling temperature presented higher concentrations at lower diameters. It disclosed that incomplete evaporation during combustion followed by consequent condensation of semi-volatile compounds on solid nuclei [15]. From Fig. 4a, c, d, the concentration of the mass fractions of trace heavy metal elements is significantly higher in these UFPs, than it was for the alkali metals. The specific process by which trace heavy metal elements transform from the vapor phase to the condensed phase is important in assessing the toxicity of waste products.

Characteristic times for UFP formation and growth

Nucleation is the particle-formation process that involves transformation of a vapor or liquid to clusters of the vapor “monomer” by a series of reversible steps [30]. When the vapor temperature decreases or when vapor is rapidly produced by a chemical reaction, the vapor concentration exceeds the critical concentration required for formation of stable particles; then nucleation occurs [31]. Consequently, the nucleation characteristic time \( \tau_{N} \) can be defined as follows [31, 32]:
$$ \tau_{N} = \frac{{\left( {N^{*} } \right)_{t = 0} }}{{I_{t = 0} }}, $$
where \( \tau_{N} \) is the nucleation characteristic time (s), \( N^{*} \) is the concentration of particles corresponding to saturated vapor (cm−3), \( I_{t = 0} \) is the initial nucleation rate at a certain temperature (cm−3 s−1), which can be obtained by Girshick and Chiu [33] as follows:
$$ I = v_{1} \left( {\frac{2\sigma }{{\pi m_{1} }}} \right)^{{\frac{1}{2}}} n_{s}^{2} Sexp\left[ {\theta - \frac{{4\theta^{3} }}{{27\left( {\ln S} \right)^{2} }}} \right], $$
$$ \theta = \frac{{\sigma s_{1} }}{{K_{B} T}}, $$
where \( v_{1} \) is the molecular volume of the condensing species (cm3), \( \sigma \) is surface tension (N/m), \( m_{1} \) is the molecular mass of the condensing species (g), \( n_{s} \) is the concentration of monomer in saturated state(cm−3), \( S \) is a saturation coefficient, \( \theta \) represents the dimensionless surface energy, \( s_{1} \) is the surface area of the condensing species (cm2), \( K_{B} \) is the Boltzmann constant and \( T \) is the temperature.
The critical diameter of nucleation can be obtained as follows:
$$ d_{p} = \frac{{4\sigma v_{1} }}{{K_{B} T\ln S}}, $$
where \( d_{p} \) is the critical diameter of nucleation (nm).
Condensation is the particle-growth process caused by condensation of supersaturated vapor onto the surfaces of existing particles. The condition for condensation to occur is that the vapor concentration of the species has to be greater than that of the particle surface concentration. Condensation involves consumption of vapor by impingement or diffusion of molecules to the particle surface [31]. The characteristic time for condensation in the free molecular regime is given as follows:
$$ \tau_{\text{cond}} = \frac{1}{{\sqrt {2K_{B} \frac{T}{{m_{1} }}} \left( {\frac{{6N_{\text{av}} v_{1} C_{\emptyset } }}{\pi N}} \right)^{{\frac{2}{3}}} N\sqrt \pi }}, $$
where \( N_{\text{av}} \) is the Avogadro number, \( C_{\emptyset } \) is an equivalent initial vapor concentration (\( = C_{o} + {\raise0.7ex\hbox{${N\pi d_{p,o}^{3} }$} \!\mathord{\left/ {\vphantom {{N\pi d_{p,o}^{3} } {\left( {6N_{\text{av}} v_{1} } \right)}}}\right.\kern-0pt} \!\lower0.7ex\hbox{${\left( {6N_{\text{av}} v_{1} } \right)}$}} - C_{s} \)), where the subscript “o” indicates initial values and \( C_{s} \) is the surface concentration, and \( N \) is the concentration of particles corresponding to gaseous molecules (cm−3).
In this study, it is assumed that the vapor concentration is constant, which is valid at low initial particle loading. The physical meaning of the characteristic time for constant vapor concentration is the representative time to grow particles from the initial size [34]. So, the characteristic time is as follows:
$$ \tau_{\text{cond}} = \frac{{d_{p,o} }}{{\sqrt {\frac{{2K_{B} T}}{{\pi m_{1} }}} N_{\text{av}} v_{1} \left( {C_{o} - C_{s} } \right)}}. $$
Coagulation is a particle–particle collision and coalescence process that occurs under condition of Brownian movement or convection, and leads to a reduction in the particle concentration and increase in the particle size. It was assumed that after the two particles collided, they formed a stable larger particle and maintain a monodispersity aerosol system [31]. Therefore, the characteristic time for coagulation is given by [32]:
$$ \tau_{\text{coag}} = \frac{2}{{\beta N_{{{\text{tot}},o}} }}, $$
where \( N_{{{\text{tot}},o}} \) is the initial total particle number concentration and \( \beta \) is the coagulation coefficient which can be obtained as follows [35]:
$$ \beta \left( {v_{i} ,v_{j} } \right) = \left( {\frac{3}{4\pi }} \right)^{1/6} \left( {\frac{{6K_{B} T}}{{\rho_{p} }}} \right)^{1/2} \left( {\frac{1}{{v_{i} }} + \frac{1}{{v_{j} }}} \right)^{1/2} \left( {v_{i}^{1/3} + v_{j}^{1/3} } \right)^{2} , $$
where \( \beta \left( {v_{i} ,v_{j} } \right) \) is the collision rate function of the two particles \( i \) and \( j \), \( v_{i} \) and \( v_{j} \) is the volume of the particles, respectively, \( \rho_{p} \) is the particle density, and \( N_{{{\text{tot}},o}} \) is the concentration of a single particle size.
The expressions derived were for a spherical, monodisperse aerosol, and the Kelvin effect was neglected for simplicity [34]. Assuming that all of the sodium elements were converted into gaseous sodium chloride, the particle surface pressure would be zero. The coagulation process was assumed to involve two particles of identical size. The calculation results from these mechanisms are presented in Fig. 5.
Fig. 5

The nucleation, condensation and coagulation’s characteristic time of ultrafine particles

The calculations indicated that the critical diameter for homogeneous nucleation is dp = 0.419 nm, and that the nucleation characteristic time is 0.1268 ms. This presented that the nucleation process occurs first and that the process is very quick. Once the UFPs form, the vapor caused rapid growth of the submicron particles by condensation, and simultaneously, the particles collided with each other to form particle aggregated of about 1 nm. As Fig. 5 shows, the characteristic time of condensation is shorter than the time of coagulation when the particle size is < 1 nm. Then, as the size of the particles increases, coagulation becomes the dominant mechanism of particle growth. The characteristic time of condensation gradually increased because the increase of the specific surface area leads to difficulty for condensation of vapor onto the particles. However, the coagulation coefficient increases with the square of the particle number concentration and increases non-linearly with decreasing particle size [30]. This is because the characteristic time of coagulation gradually decreases with increasing particle size.


The morphology and surface chemical composition of 60 UFPs were characterized using SEM–EDS. The particles were usually irregular spheres, and the submicron particles (PM1) resulted from coalescence of UFPs. The mass fractions of heavy metals (Cu, Zn and Ni) were higher than those of alkali metals (Na and K), indicating that the UFPs have high heavy metal toxicity and must be properly handled.

With increase of the particle size from 10.17 to 155.9 nm, the mass fractions of Na, K and Fe gradually decreased, while the main elements Ca, Si and Al, as well as the heavy metals Cu, Zn and Ni, gradually increased. This indicates that, during the process of particle formation, the initial part of the sequence involves Na and K nucleation, followed by Fe nucleation. Then, the elements Si, Ca and Al, and parts of the heavy metal elements, condense onto the nanoscale particles.

According to the calculation of characteristic times, the nucleation process occurs initially, then rapid growth of the submicron particles to about 1 nm occurs by condensation. When the particle size is > 1 nm, coagulation becomes the dominant mechanism of particle growth.



The authors acknowledge support by the National Natural Science Foundation of China in the form of a research grant (No. 51576134) and the National Key R&D Program of China (2017YFC0703100).


  1. 1.
    Nowak B, Aschenbrenner P, Winter F. Heavy metal removal from sewage sludge ash and municipal solid waste fly ash—a comparison. Fuel Process Technol. 2013;105:195–201.CrossRefGoogle Scholar
  2. 2.
    Syed-Hassan SS, Wang Y, Hu S, et al. Thermochemical processing of sewage sludge to energy and fuel: fundamentals, challenges and considerations. Renew Sust Energ Rev. 2017;80:888–913.CrossRefGoogle Scholar
  3. 3.
    Murakami T, Suzuki Y, Nagasawa H, et al. Combustion characteristics of sewage sludge in an incineration plant for energy recovery. Fuel Process Technol. 2009;90:778–83.CrossRefGoogle Scholar
  4. 4.
    Deng W, Yan J, Li X, et al. Emission characteristics of dioxins, furans and polycyclic aromatic hydrocarbons during fluidized-bed combustion of sewage sludge. J Environ Sci. 2009;21:1747–52.CrossRefGoogle Scholar
  5. 5.
    Kasina M, Kowalski PR, Michalik M. Metals accumulation during thermal processing of sewage sludge characterization of fly ash and air pollution control (APC) residues. Energy Procedia. 2016;97:23–30.CrossRefGoogle Scholar
  6. 6.
    Li R, Li Y, Yang T, et al. A new integrated evaluation method of heavy metals pollution control during melting and sintering of MSWI fly ash. J Hazard Mater. 2015;289:197–203.CrossRefGoogle Scholar
  7. 7.
    Wang D, Guo H, Cheung K, et al. Observation of nucleation mode particle burst and new particle formation events at an urban site in Hong Kong. Atmos Environ. 2014;99:196–205.CrossRefGoogle Scholar
  8. 8.
    Liu J, Fu J, Ning X, et al. An experimental and thermodynamic equilibrium investigation of the Pb, Zn, Cr, Cu, Mn and Ni partitioning during sewage sludge incineration. J Environ Sci. 2015;35:43–54.CrossRefGoogle Scholar
  9. 9.
    Cheng H, Hu Y. Curbing dioxin emissions from municipal solid waste incineration in China: re thinking about management policies and practices. Environ Pollut. 2010;158:2809–14.CrossRefGoogle Scholar
  10. 10.
    Chuang MT, Chen YC, Lee CT, et al. Apportionment of the sources of high fine particulate matter concentration events in a developing aerotropolis in Taoyuan, Taiwan. Environ Pollut. 2016;214:273–81.CrossRefGoogle Scholar
  11. 11.
    González Y, Rodríguez S, García JCG, et al. Ultrafine particles pollution in urban coastal air due to ship emissions. Atmos Environ. 2011;45:4907–14.CrossRefGoogle Scholar
  12. 12.
    Jones AM, Harrison RM. Emission of ultrafine particles from the incineration of municipal solid waste: a review. Atmos Environ. 2016;140:519–28.CrossRefGoogle Scholar
  13. 13.
    Lee SW. Fine particulate matter measurement and international standardization for air quality and emissions from stationary sources. Fuel. 2010;89:874–82.CrossRefGoogle Scholar
  14. 14.
    Abramesko V, Tartakovsky L. Ultrafine particle air pollution inside diesel-propelled passenger trains. Environ Pollut. 2017;226:288–96.CrossRefGoogle Scholar
  15. 15.
    Buonanno G, Stabile L, Avino P, et al. Chemical, dimensional and morphological ultrafine particle characterization from a waste-to-energy plant. Waste Manag. 2011;31:2253–62.CrossRefGoogle Scholar
  16. 16.
    Gonzalez LT, Rodríguez FEL, Sanchez-Domínguez M, et al. Chemical and morphological characterization of TSP and PM2.5 by SEM–EDS, XPS and XRD collected in the metropolitan area of Monterrey, Mexico. Atmos Environ. 2016;143:249–60.CrossRefGoogle Scholar
  17. 17.
    Ribeiro J, DaBoit K, Flores D, et al. Extensive FE-SEM/EDS, HR-TEM/EDS and ToF-SIMS studies of micron- to nano-particles in anthracite fly ash. Sci Total Environ. 2013;452–453:98–107.CrossRefGoogle Scholar
  18. 18.
    Ramirez-Leal R, Valle-Martinez M, Cruz-Campas M. Chemical and morphological study of PM10 analysed by SEM–EDS. Open J Air Pollut. 2014;3:121–9.CrossRefGoogle Scholar
  19. 19.
    Kulmala M, Vehkamaki H, Petaja T, et al. Formation and growth rates of ultrafine atmospheric particles: a review of observations. Aerosol Sci. 2004;35:143–76.CrossRefGoogle Scholar
  20. 20.
    Wu CY, Biswas P. Competing effects of chemical reaction, nucleation and condensation on metallic aerosol formation in waste incineration systems. Proceedings of 1994 Technical Meeting of Central States, The Combustion Institute, Madison. 1994;342–347.Google Scholar
  21. 21.
    Wilkinson K, Lundkvist J, Seisenbaeva G, et al. A cost-effective method for monitoring airborne particulate matter using tabletop SEM–EDS. Air Pollut. 2010;136:407–18.Google Scholar
  22. 22.
    Vassilev SV, Vassileva CG. Methods for characterization of composition of fly ashes from coal fired power stations: a critical overview. Energy Fuels. 2005;19:1084–98.CrossRefGoogle Scholar
  23. 23.
    Genga A, Baglivi F, Siciliano M, et al. SEM–EDS investigation on PM10 data collected in central Italy: principal component analysis and hierarchical cluster analysis. Chem Cent J. 2012;6:S3.CrossRefGoogle Scholar
  24. 24.
    Seames WS. An initial study of the fine fragmentation fly ash particle mode generated during pulverized coal combustion. Fuel Process Technol. 2003;81:109–25.CrossRefGoogle Scholar
  25. 25.
    Wang Q, Zhang L, Sato A, et al. Effects of coal blending on the reduction of PM10 during high-temperature combustion 1. Mineral transformations. Fuel. 2008;87:2997–3005.CrossRefGoogle Scholar
  26. 26.
    Li G, Li S, Dong M, et al. Comparison of particulate formation and ash deposition under oxy-fuel and conventional pulverized coal combustions. Fuel. 2013;106:544–51.CrossRefGoogle Scholar
  27. 27.
    Li G, Li S, Huang Q, et al. Fine particulate formation and ash deposition during pulverized coal combustion of high-sodium lignite in a down-fired furnace. Fuel. 2015;143:430–7.CrossRefGoogle Scholar
  28. 28.
    Niu Y, Liu X, Wang S, et al. A numerical investigation of the effect of flue gas recirculation on the evolution of ultra-fine ash particles during pulverized coal char combustion. Combust Flame. 2017;184:1–10.CrossRefGoogle Scholar
  29. 29.
    Zhang L, Ninomiya Y, Yamashita T. Formation of submicron particulate matter (PM1) during coal combustion and influence of reaction temperature. Fuel. 2006;85:1446–57.CrossRefGoogle Scholar
  30. 30.
    Lighty JS, Veranth JM, Sarofim AF. Combustion aerosols: factors governing their size and composition and implications to human health. J Air Waste Manag Assoc. 2000;50:1619–22.CrossRefGoogle Scholar
  31. 31.
    Biswas P, Wu CY. Control of toxic metal emissions from combustors using sorbents: a review. J Air Waste Manag Assoc. 1998;48:113–27.CrossRefGoogle Scholar
  32. 32.
    Zhuang Y, Biswas P. Submicrometer particle formation and control in a bench-scale pulverized coal combustor. Energy Fuels. 2001;15:510–6.CrossRefGoogle Scholar
  33. 33.
    Girshick SL, Chiu CP. Kinetic nucleation theory: a new expression for the rate of homogeneous nucleation from an ideal supersaturated vapor. J Chem Phys. 1990;93:1273–7.CrossRefGoogle Scholar
  34. 34.
    Wu CY, Biswas P. Particle growth by condensation in a system with limited vapor. Aerosol Sci Technol. 1998;28:1–20.CrossRefGoogle Scholar
  35. 35.
    Friedlander SK. Smoke, dust, and haze. Topics in chemical engineering. New York: Oxford University Press; 2000.Google Scholar

Copyright information

© Zhejiang University Press 2019

Authors and Affiliations

  • Yanlong Li
    • 1
  • Jiaqi Man
    • 1
  • Zhengquan Fang
    • 1
  • Yunbin Zhao
    • 1
  • Feng Wang
    • 1
  • Rundong Li
    • 1
    Email author
  1. 1.Key Laboratory of Clean Energy, College of Energy and EnvironmentShenyang Aerospace UniversityShenyangPeople’s Republic of China

Personalised recommendations