Risk Evaluation of Pollutants Emission from Coal and Coal Waste Combustion Plants and Environmental Impact of Fly Ash Landfilling

Emission factors (EFs) of gaseous pollutants, particulate matter, certain harmful trace elements, and polycyclic aromatic hydrocarbons (PAHs) from three thermal power plants (TPPs) and semi-industrial fluidized bed boiler (FBB) were compared. EFs of particulate matter, trace elements (except Cd and Pb), benzo[a]pyrene, and benzo[b]fluoranthene exceed the upper limits specified in the EMEP inventory guidebook for all combustion facilities. The comparison of trace elements and PAHs content in fly ashes (FAs) from lignite and coal waste combustion in TPPs and FBB, respectively, as well as the potential environmental impact of FAs disposal, was performed by employing a set of ecological indicators such as crustal enrichment factor, risk assessment code, risk indices for trace elements, and benzo[a]pyrene equivalent concentration for PAHs. Sequential analysis shows that the trace elements portion is the lowest for water-soluble and exchangeable fractions. The highest enrichment levels in FAs are noticed for As and Hg. Based on toxic trace elements content, FAs from TPPs represent a very high ecological risk, whereas fly ash from FBB poses a moderate ecological risk but has the highest benzo[a]pyrene equivalent concentration, indicating its increased carcinogenic potential. Lead isotope ratios for Serbian coals and FAs can contribute to a lead pollution global database.


Introduction
Coal combustion is both the primary energy source in many countries and one of the major anthropogenic sources of atmospheric, water, and soil pollution. It induces a variety of environmental and health issues due to the release of gaseous pollutants such as sulfur dioxide (SO 2 ), nitrogen oxide (NOx), carbon monoxide (CO), and solid particulate matter (PM) [1]. Millions of tons of coal are burned in different power plants worldwide [2], resulting in enormous quantities of combustion residues, such as fly and bottom ashes. Branch TPP Nikola Tesla (TPPs Nikola Tesla A and B, Kolubara A and Morava) generates more than 50% of electricity, while TPPs Kostolac A and B produce around 17% of overall electricity in Serbia [3]. Approximately 40 million tons of lignite from the Kolubara and Drmno basins are burned in thermal power plants, producing 6 million tons of FAs [4]. Since coal resources are limited [5], using alternative fuels such as coal waste in various combustion technologies is increasingly important. Fluidized bed combustion has proven to be an efficient and environmentally acceptable technique for producing energy from low as FAs from TPPs and FBB (8 samples) were analyzed. Sampling and sample preparation followed a previously described methodology [34]. Table S1 in Section S1 (Supplementary) contains the details about sampling locations.

Emission of NOx, CO, SO 2 and Total PM from TPPs and FBB
The content of flue gas pollutants was determined according to the standards [35][36][37][38][39][40]. The flue gases sampling was carried out at the stack or at the flue gas line in front of the stack. Multicomponent gas analyzer Horiba PG 350E (O 2 , CO 2 , CO, NOx, and SO 2 ), gas conditioning unit PSS5 M&C Tech group and heated hose JCT Analysentechnik GmbH were used. Utilized gas sampling equipment was in accordance with standards [35][36][37]39,40]. Flue gas flows for all TPPs and FBB are shown in Table 1. Particulate matter sampling was performed by the isokinetic sampling system Heated Paul Gothe gas sample probe with a fine filter and pre-filter [38]. Total PM included all particulate matter with diameters higher than 0.3 µm. Sampling flow range: 0.5-4.0 m 3 /h.

Determination of Trace Elements and Lead Isotopic Ratios 2.3.1. Sequential Extraction of FAs
The sequential extraction of trace elements (As, Be, Cd, Co, Cr, Cs, Cu, Ga, Ge, Hg, Mn, Mo, Ni, Pb, Sb, Sr, U, and V) was performed as described in the literature [41] with a modified last step. Details are explained in Section S2 (Supplementary).

ICP-MS Analysis
Trace elements concentrations were determined by the inductively coupled plasma mass spectrometry (ICP-MS) using Agilent 7500ce instrument equipped with Octopole Reaction System in FullQuant mode. ICP-MS calibration was performed by Agilent Multi-Element Calibration Standards. Standard solutions and blanks were prepared in 2% HNO 3 . The analyses of each fraction (F1-F6) were performed in 3 replicates. The isotope analysis mode was used for lead isotopic measurements. The accuracy of the isotope ratio measurements was evaluated by analyzing a certified isotopic standard NIST SRM 981. External corrections due to a mass bias of the ICP-MS were performed by measuring a 5 g/L solution of NIST SRM 981. Mass bias correction factors were automatically calculated for the ratios 207 Pb/ 206 Pb (0.9925) and 208 Pb/ 207 Pb (0.9968). Other details about the ICP-MS analysis are described in Section S3 (Supplementary).

PAHs Extraction
The extraction of PAHs from FAs was performed by a solid-liquid extraction. More details about PAHs extraction are provided in [34].

HPLC Analysis
A Thermo Fisher Scientific Dionex UltiMate 3000 HPLC system with a diode array detector was applied to determine 16 priority PAHs along with two naphthalene substituted derivatives (1-methyl naphthalene and 2-methyl naphthalene). The details of the modified HPLC analysis are provided in [34].  [10,11,18,[42][43][44] and data acquired from the field measurements (flue gas velocity, cross-sectional area, coal consumption, and ash production) were employed to estimate As, Cd, Cr, Cu, Hg, Ni, and Pb emission factors, as well as EFs for benzo [ CEF specifies the degree to which each trace element in fly ash is enriched relative to its content crustal core. To reduce the alterations among diverse samples, CEF values are estimated in relation to a reference element with low occurrence variability. In this paper, Mn is used as a reference element. CEF n/Mn is calculated using the following formula: CEF n/Mn = C n C Mn f ly ash C n C Mn crustal core (1) where C n is the concentration of each investigated trace element and C Mn is the concentration of manganese, both in fly ash samples and crustal core, respectively.

Risk Assessment Code (RAC)
Usually, RAC defines the potential environmental risk of trace elements leaching from complex matrices, such as ashes [24,46]. It is calculated as the ratio of the sum of the watersoluble fraction (F1), the exchangeable fraction (F2), and the carbonate bound fraction (F3) to the total element concentration in a representative sample, expressed in percentage.

Pollution Index (PI)
PI identifies the pollution level of all investigated trace elements that can pose harm to the soil [47]. PI is calculated by the following equation: where PI i is the single pollution index; and C n and C b are the elemental concentration in fly ash and crustal core, respectively.

Risk Index (RI)
RI estimates the individual ecological risk level for the most hazardous trace elements (As, Cd, Co, Cr, Cu, Hg, Ni, and Pb), while RI sum evaluates the overall ecological risk of the investigated sample. It is calculated by the following equations: where T r represents the toxic response for each potentially toxic trace element [23]. BaP eq is used to estimate the PAHs overall toxic potency in investigated samples. BaP eq is determined according to the following equation: where c i is the individual PAH concentration, and TEF i represents the toxic equivalency factor of each PAH [48]. BaPE is calculated according to the following equation [20]:

Results and Discussion
3.1. NOx, CO, SO 2 and Total PM Emission from TPPs and FBB The highest NOx concentration was noticed for TPP Nikola Tesla A, while total PM concentration was the highest for TPP Kolubara A ( Figure 1). The increased content of SO 2 in TPP Kostolac B surpasses its corresponding limit value (listed in Table S2) [49]. Higher sulfur content in lignite utilized in TPP Kostolac B most likely leads to elevated SO 2 content in the flue gas. CO is higher in FBB than in other combustion facilities, possibly due to incomplete combustion and lower combustion temperature in FBB ( Figure 1). The electrostatic precipitators installed in TPPs have high efficiency (more than 99%) in collecting particulate matter, but they are usually inefficient for capturing fine and/or ultrafine particles [50,51].
where Tr represents the toxic response for each potentially toxic trace element [23].

BaPeq and BaPE Values for FAs
BaPeq is used to estimate the PAHs overall toxic potency in investigated samples. BaPeq is determined according to the following equation: where ci is the individual PAH concentration, and TEFi represents the toxic equivalency factor of each PAH [48]. BaPE is calculated according to the following equation [20]:

NOx, CO, SO2 and Total PM Emission from TPPs and FBB
The highest NOx concentration was noticed for TPP Nikola Tesla A, while total PM concentration was the highest for TPP Kolubara A ( Figure 1). The increased content of SO2 in TPP Kostolac B surpasses its corresponding limit value (listed in Table S2) [49]. Higher sulfur content in lignite utilized in TPP Kostolac B most likely leads to elevated SO2 content in the flue gas. CO is higher in FBB than in other combustion facilities, possibly due to incomplete combustion and lower combustion temperature in FBB ( Figure 1). The electrostatic precipitators installed in TPPs have high efficiency (more than 99%) in collecting particulate matter, but they are usually inefficient for capturing fine and/or ultrafine particles [50,51]. The EF values of all investigated pollutants are listed in Table 2, along with their corresponding lower and upper limits and the average value obtained from the EMEP inventory guidebook. EFs for total particulate matter (Table 2A)   The EF values of all investigated pollutants are listed in Table 2, along with their corresponding lower and upper limits and the average value obtained from the EMEP inventory guidebook. EFs for total particulate matter (   [52,53] or for total PM (this paper).

Sequential Extraction of Trace Elements from FAs
The chemical speciation of 18 trace elements in FAs between the water-soluble fraction (F1), the exchangeable fraction (F2), the carbonate bound fraction (F3), the metal oxide bound fraction (F4), the organic bound fraction (F5), and the residual fraction (F6) was performed. Overall trace element concentrations within six fractions ( Figure 2) range from 709.81 mg/kg in CFB to 1360.90 mg/kg in TPPKb. The lowest concentrations are within F1 fractions for all FAs, going from 9.35 mg/kg in CFB to 22.45 mg/kg in TPPKb. The highest concentrations are noticed for F4 fractions in TPPKb and CFB and F6 fractions in TPPKs and TPPNT.

Sequential Extraction of Trace Elements from FAs
The chemical speciation of 18 trace elements in FAs between the water-solub tion (F1), the exchangeable fraction (F2), the carbonate bound fraction (F3), the met bound fraction (F4), the organic bound fraction (F5), and the residual fraction ( performed. Overall trace element concentrations within six fractions ( Figure 2  The distribution of all investigated elements among fractions (F1-F6) is shown ure 3. The water-soluble fraction is of major environmental concern since anions, chlorides and sulfates, are easily available for plant and soil uptake [54]. Accor Figure 3, most compounds of the studied trace elements are not water-soluble, so p of their F1 fractions are low. The exception is molybdenum, which can easily form soluble species [15]. The distribution of all investigated elements among fractions (F1-F6) is shown in Figure 3. The water-soluble fraction is of major environmental concern since anions, such as chlorides and sulfates, are easily available for plant and soil uptake [54]. According to Figure 3, most compounds of the studied trace elements are not water-soluble, so portions of their F1 fractions are low. The exception is molybdenum, which can easily form water-soluble species [15].
In addition, some of the investigated elements ( Figure 3) in all FAs (Cs, Ga, Ge, Hg, Ni, Sb, and U) have low levels in F1 and F2 fractions as shown in the literature [46]. Among previously mentioned elements, germanium prevails in the water-soluble phase up to 3.33% in TPPKs, probably due to the presence of GeO 2 and GeS 2 [55]. In F2 fractions, the proportions of Mo and Sr are the highest (from 3.0% for Sr in TPPNT to 15.42% for Mo in TPPKs). Sr solubility is likely due to its cationic leaching pattern and removal of the exchangeable cation sorbed on the ash particles surface [56]. F3 fraction mainly consists of elements as carbonates or oxides/hydroxides (Cd, Cu, Mn, Mo, V, and Sr) [57,58]. pH reduction increases the mobility of the carbonate form of these trace elements [59]. Trace elements in the F4 extraction step are moderately mobile and sensitive to redox potential changes [21]. The investigated elements proportion in F4 fractions ranges from 0.74% for Hg (TPPKb) to 73.62% for As (TPPNT). Elements of the F5 fraction can occur as oxidizable minerals, e.g., sulfides [22], and their mobility is relatively low unless they undergo the oxidation process. Co has a maximal portion of 26.45% in CFB among all elements within F5 fractions. The majority of elements have the highest distributions in residual fractions in all FAs (Cs, Ga, Ge, Hg, Ni, Pb, Sb, Sr, and U) or are evenly distributed between F4 and F6 fractions (As, Cr, Cu, and V). Since elements from the residual fractions have strong bonds with the mineral crystal lattice, their mobility is low, and they are not considered to be of environmental concerns.  In addition, some of the investigated elements ( Figure 3) in all FAs (Cs, Ga, Ge, Hg, Ni, Sb, and U) have low levels in F1 and F2 fractions as shown in the literature [46]. Among previously mentioned elements, germanium prevails in the water-soluble phase up to 3.33% in TPPKs, probably due to the presence of GeO2 and GeS2 [55]. In F2 fractions, the proportions of Mo and Sr are the highest (from 3.0% for Sr in TPPNT to 15.42% for Mo in TPPKs). Sr solubility is likely due to its cationic leaching pattern and removal of the exchangeable cation sorbed on the ash particles surface [56]. F3 fraction mainly consists of elements as carbonates or oxides/hydroxides (Cd, Cu, Mn, Mo, V, and Sr) [57,58]. pH reduction increases the mobility of the carbonate form of these trace elements [59]. Trace elements in the F4 extraction step are moderately mobile and sensitive to redox potential changes [21]. The investigated elements proportion in F4 fractions ranges from 0.74% for Hg (TPPKb) to 73.62% for As (TPPNT). Elements of the F5 fraction can occur as oxidizable minerals, e.g., sulfides [22], and their mobility is relatively low unless they undergo the oxidation process. Co has a maximal portion of 26.45% in CFB among all elements within F5 fractions. The majority of elements have the highest distributions in residual fractions in all FAs (Cs, Ga, Ge, Hg, Ni, Pb, Sb, Sr, and U) or are evenly distributed between F4 and F6 fractions (As, Cr, Cu, and V). Since elements from the residual fractions have strong bonds with the mineral crystal lattice, their mobility is low, and they are not considered to be of environmental concerns.
Among investigated elements, As, Cd, Cr, Hg, Ni, and Pb are carcinogenic (C), while Be, Co, Cs, Cu, Ga, Ge, Mn, Mo, Sb, Sr, U, and V are considered to be non-carcinogenic (NC). The carcinogenic element prevails in F6 fractions and goes from 39.52% in TPPNT to 42.92% in CFB, while the non-carcinogenic portion is up to 91.31% in the F2 fraction of CFB (Figure 4a). Figure 4b displays the distributions of the C and NC elements among the six fractions with the overall content set to be 100%. Carcinogenic element portions in the Among investigated elements, As, Cd, Cr, Hg, Ni, and Pb are carcinogenic (C), while Be, Co, Cs, Cu, Ga, Ge, Mn, Mo, Sb, Sr, U, and V are considered to be non-carcinogenic (NC). The carcinogenic element prevails in F6 fractions and goes from 39.52% in TPPNT to 42.92% in CFB, while the non-carcinogenic portion is up to 91.31% in the F2 fraction of CFB (Figure 4a). Figure 4b displays the distributions of the C and NC elements among the six fractions with the overall content set to be 100%. Carcinogenic element portions in the environmentally significant fractions (F1-F3) range from 5.07% in TPPKs to 7.76% in TPPKb, while non-carcinogenic element distributions vary from 12.23% in TPPNT to 13.50% in TPPKs (Figure 4b).

Lead Isotope Ratios
Coal combustion is a predominant source of lead pollution, and because of that lead isotopic composition in FAs and coals has a major influence on total Pb isotopic fingerprint [29]. Figure S1a demonstrates the ranges of lead isotope ratios among investigated fly ashes and coals for 206 Pb/ 207 Pb (from 1.162 to 1.206), 208 Pb/ 207 Pb (from 2.434 to 2.533), and 208 Pb/ 206 Pb (2.045 to 2.167). In addition, Figure S1b shows the values of 208 Pb/ 206 Pb and 206 Pb/ 207 Pb from available literature data [29,30,32,33]. The 3D diagram of all determined Pb isotope ratios for Serbian coals and fly ashes are shown in Figure 5a. Figure 5b compares 206 Pb/ 207 Pb literature data for coal combustion worldwide [29,33] with data obtained in this study. The mean value of 206 Pb/ 207 Pb ratio for Serbian coals and FAs is 1.186 ± 0.012, and it highly correlates with the United Kingdom (1.187), Switzerland (1.181), and the USA (1.189). Therefore, identifying the lead pollution fingerprint for Serbian coals and FAs contributes to lead pollution studies. environmentally significant fractions (F1-F3) range from 5.07% in TPPKs to 7.76% in TPPKb, while non-carcinogenic element distributions vary from 12.23% in TPPNT to 13.50% in TPPKs (Figure 4b).

Lead Isotope Ratios
Coal combustion is a predominant source of lead pollution, and because of that lead isotopic composition in FAs and coals has a major influence on total Pb isotopic fingerprint [29]. Figure S1a demonstrates the ranges of lead isotope ratios among investigated fly ashes and coals for 206 Pb/ 207 Pb (from 1.162 to 1.206), 208 Pb/ 207 Pb (from 2.434 to 2.533), and 208 Pb/ 206 Pb (2.045 to 2.167). In addition, Figure S1b shows the values of 208 Pb/ 206 Pb and 206 Pb/ 207 Pb from available literature data [29,30,32,33]. The 3D diagram of all determined Pb isotope ratios for Serbian coals and fly ashes are shown in Figure 5a. Figure 5b compares 206 Pb/ 207 Pb literature data for coal combustion worldwide [29,33] with data obtained in this study. The mean value of 206 Pb/ 207 Pb ratio for Serbian coals and FAs is 1.186 ± 0.012, and it highly correlates with the United Kingdom (1.187), Switzerland (1.181), and the USA (1.189). Therefore, identifying the lead pollution fingerprint for Serbian coals and FAs contributes to lead pollution studies.     [52,53] can be explained by the poorer quality of lignite utilized in all TPPs compared with coal provided in EMEP.

Trace Elements EFs
Despite having different powers, the EFs of Cr, Cu, Hg, and Pb in TPPs Kolubara A and Nikola Tesla A are comparable, at the same time TPP Kolubara A exhibits EFs higher from around 40% for As and Ni to 106% for Cd. Higher temperatures, better process optimization, and better air pollution management in TPPs, combined with lower feed fuel quality in FBB, result in much higher EFs of all trace elements (aside from Hg) for FBB than for TPPs (Table 2B). Table S3 compares As, Cr, Hg, and Pb emission concentrations, ranging from 4.02 µg/m 3 for Hg in FBB to 42.68 µg/m 3 for Pb in TPP Kostolac B, to their respective emission limits for coal-fired units [60,61]. Arsenic emission concentrations from all combustion facilities exceed limit values compared with both standards (Table S3), while only Hg emission from FBB meets US EPA criteria.

Environmental Concerns of Investigated Trace Elements from FAs
3.5.1. Crustal Enrichment Factor Normalized to Mn (CEF n/Mn ) CEF n/Mn values for all investigated elements were determined and presented in Figure 6. To comprehensively evaluate the pollution level of elements for examined FAs, five classes for CEF [62] were displayed in Table S4. Among investigated trace elements, As and Hg show the highest enrichment, which is more pronounced for all FAs from TPPs compared with CFB. CFB, on the other hand, has the greatest CEF n/Mn values among all FAs for Be, Cr, Ga, Sb, Sr, and U. No enrichment is observed for Be, Cs, Ga, and Sr.   [60,61]. Arsenic emission concentrations from all combustion facilities exceed limit values compared with both standards (Table S3), while only Hg emission from FBB meets US EPA criteria.

Crustal Enrichment Factor Normalized to Mn (CEFn/Mn)
CEFn/Mn values for all investigated elements were determined and presented in Figure  6. To comprehensively evaluate the pollution level of elements for examined FAs, five classes for CEF [62] were displayed in Table S4. Among investigated trace elements, As and Hg show the highest enrichment, which is more pronounced for all FAs from TPPs compared with CFB. CFB, on the other hand, has the greatest CEFn/Mn values among all FAs for Be, Cr, Ga, Sb, Sr, and U. No enrichment is observed for Be, Cs, Ga, and Sr.

Risk Assessment Code
The toxic trace elements content in various matrices is typically expressed as total or water leaching concentrations. The content and leaching patterns of potentially toxic substances in coal combustion residues can provide valuable information for landfilling or be a limiting factor for application [63]. Table S5 compares the content of the most toxic elements in the investigated fly ashes with the European countries' legislation. According to the Canadian Environmental Protection Act, six elements (As, Cd, Hg, Ni, Cr, and Pb) are defined as toxic substances [64]. As and Hg for all FAs from TPPs, as well as Pb for all FAs, are above limit values [65], while other potentially hazardous elements are below these limits. The trace elements from F1 fractions can be defined as non-hazardous [66], as

Risk Assessment Code
The toxic trace elements content in various matrices is typically expressed as total or water leaching concentrations. The content and leaching patterns of potentially toxic substances in coal combustion residues can provide valuable information for landfilling or be a limiting factor for application [63]. Table S5 compares the content of the most toxic elements in the investigated fly ashes with the European countries' legislation. According to the Canadian Environmental Protection Act, six elements (As, Cd, Hg, Ni, Cr, and Pb) are defined as toxic substances [64]. As and Hg for all FAs from TPPs, as well as Pb for all FAs, are above limit values [65], while other potentially hazardous elements are below these limits. The trace elements from F1 fractions can be defined as non-hazardous [66], as shown in Table S5.
RAC classification (Table S4) is important for estimating the potential leachability of elements from the sample matrix to the environment. The results for RAC (Figure 7a) suggest that Mo presents the highest risk. Sr displays medium risk of TPPNT and CFB, while values for TPPKb and TPPKs are above 30%. Be (up to 16.47%) and V (up to 18.93%) in all FAs, as well as Cd (up to 28.10%) for FAs from TPPs, pose a medium risk to the environment. Among investigated trace elements, only Co, Ge, and Mo have higher RAC values for CFB compared with the other FAs from TPPs. Although RAC values for some trace elements (Figure 7a) indicate a high environmental risk, some of them, such as Mo, are not considered environmentally relevant [14,65].   Table S6 summarizes PI values for all investigated trace elements along with their pollution levels [67]. PIs indicate very high pollution levels for As (from 11.25 in CFB to 42.17 in TPPKb) and Hg (from 6.71 in CFB to 29.35 in TPPNT), while Cr, Ni, and Sb show a high level of pollution (Table S6). Among investigated FAs, Be, Ga, and Sr do not pose any pollution (class 1), while only Sb shows the highest pollution in CFB compared with other FAs.

Pollution Indices and Risk Indices
The potential ecological risk index (RI) categorizes different matrices based on the degree of contamination (Table S4 shows ecological risk limits). Figure 7b depicts RIs for the most toxic elements (As, Cd, Co, Cr, Cu, Hg, Ni, and Pb) in investigated fly ashes. Since the RI values for all FAs from TPPs are higher than 600, they represent a very high ecological risk, while CFB shows a moderate ecological risk [68,69].

PAHs Content in FAs
Individual and total PAH concentrations are presented in Table 3. Overall PAH contents are from 286.69 ng/g (TPPNT) to 33,378.53 ng/g (CFB), which is in accordance with the literature [19,28,70,71]. Figure S2 shows PAHs distribution by ring number (a) and the  Table S6 summarizes PI values for all investigated trace elements along with their pollution levels [67]. PIs indicate very high pollution levels for As (from 11.25 in CFB to 42.17 in TPPKb) and Hg (from 6.71 in CFB to 29.35 in TPPNT), while Cr, Ni, and Sb show a high level of pollution (Table S6). Among investigated FAs, Be, Ga, and Sr do not pose any pollution (class 1), while only Sb shows the highest pollution in CFB compared with other FAs.

Pollution Indices and Risk Indices
The potential ecological risk index (RI) categorizes different matrices based on the degree of contamination (Table S4 shows ecological risk limits). Figure 7b depicts RIs for the most toxic elements (As, Cd, Co, Cr, Cu, Hg, Ni, and Pb) in investigated fly ashes.
Since the RI values for all FAs from TPPs are higher than 600, they represent a very high ecological risk, while CFB shows a moderate ecological risk [68,69].

PAHs Content in FAs
Individual and total PAH concentrations are presented in Table 3. Overall PAH contents are from 286.69 ng/g (TPPNT) to 33,378.53 ng/g (CFB), which is in accordance with the literature [19,28,70,71]. Figure S2 shows PAHs distribution by ring number (a) and the total and carcinogenic PAH contents of the investigated FAs (b). The four ring PAHs have the highest yield in CFB (68.29%) and TPPKb (66.07%), while the sum of two and three ring PAHs predominates in TPPKs (75.44%) and TPPNT (68.16%). Flu and Fla are the most abundant among examined PAHs (Table 3), as expected, since they can be commonly found in products of incomplete combustion of fossil fuels [70].

PAH Emission Factors
PAHs belong to the most hazardous and persistent organic compounds, particularly BbF, BkF, BaP, and IP, and since they can easily reach the atmosphere, estimation of their EFs is important [72]. Additionally, Table 2C shows PAHs emission for the combustion of lignite in TPPs and coal waste in FBB, as well as their EMEP emission limits [45].
PAHs portion in the finest ash particles emitted along with flue gases ranges from 39.96% (TPPs Kostolac B and Nikola Tesla A) to 94.63% (FBB). TPPs Kostolac B and Nikola Tesla A have BaP emission that exceeds the upper limit specified by EMEP, most likely due to lower lignite quality than coal listed in the EMEP (Table 2C). BbF, BkF, BaP, and IP emissions from TPP Kolubara A are within permissible limits. The estimated EF values for FBB (Table 2C) are elevated probably due to reduced combustion temperature, poor coal quality, and low cyclone efficiency.

Potential Environmental Effects of PAHs from FAs
To assess the carcinogenic potential of studied FAs, calculated overall BaP eq and BaPE are shown in Table 3. The total BaP eq for PAHs ranges from 5.21 ng/g (TPPNT) to 876.73 ng/g (CFB), while BaPE varies from 3.66 ng/g to 632.15 ng/g for TPPKs and CFB, respectively. The calculated BaP eq and BaPE values are in accordance with the literature findings for different FAs [28,73]. Figure S3 shows BaP eq ratios for each PAH expressed relative to BaP (set as 100%). BaA proportions are the highest and range from 50.25% (TPPNT) to 150.83% (CFB).

Conclusions
The environmental impact of byproducts generated during coal combustion in various TPPs (Kolubara A, Kostolac B, and Nikola Tesla A) and coal waste burning in an experimental semi-industrial fluidized bed boiler was investigated in this study. The aim was to determine the emission of gaseous pollutants (NOx, CO, and SO 2 ) and total PM, estimate trace elements and PAHs emission factors, assess the environmental risk of FAs landfilling, and establish lead isotope ratios for Serbian coals and FAs.
The general conclusions are: • The determined lead isotope fingerprint for investigated coals and FAs is within ranges of other countries and can be particularly useful in the source apportionment of lead pollution.
This paper provides comprehensive systematic research of the overall environmental impact of different combustion facilities from the emission and fly ash landfilling perspective. To reduce the environmental risk of ash disposal, it is important to monitor harmful trace elements, and persistent organic pollutants such as PAHs, as well as to enhance environmental pollution control in the Serbian energy sector.
Supplementary Materials: The following supporting information can be downloaded at https:// www.mdpi.com/article/10.3390/toxics11040396/s1:, Figure S1: (a) Lead isotope ratios ( 206 Pb/ 207 Pb and 208 Pb/ 207 Pb vs. 208 Pb/ 206 Pb) for all investigated samples from Serbia (8 fly ashes and 4 coals); (b) Lead isotope ratios ( 206 Pb/ 207 Pb vs. 208 Pb/ 206 Pb) depending on originating country; Figure S2: (a) PAHs proportions by their ring number (R2-R6); (b) The total and carcinogenic PAHs content for fly ashes (TPPKb, TPPKs, TPPNT and CFB); Figure S3: The individual BaP eq ratios relative to BaP (%); Table S1. Locations of combustion facilities and lignite mining basins; Table S2: Limit values of NOx, CO, SO 2 and total PM concentrations (mg/Nm 3 ) in flue gases of combustion facilities with different capacities; Table S3: Flue gas emissions of trace elements (µg/m 3 ); Table S4: Adjusted indices of element enrichment degree, environmental risk and ecological risk; Table S5: Comparative view of total heavy metal concentrations and their water leachates with literature data; Table S6

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.
Acknowledgments: Special thanks to Slavica Dramićanin and Jelena Erić for their support in laboratory experiments.

Conflicts of Interest:
The authors declare no conflict of interest.