Study on the Effects of Wet Flue Gas Desulfurization on Particulate Matter Emission from Industrial Coal-Fired Power Plants

Abstract

This study aimed to investigate the effects of wet flue gas desulfurization (WFGD) on particulate matter (PM) emissions in coal-fired power plants (CFPPs) using an electrical low-pressure impactor (ELPI). The investigation was conducted on five industrial CFPPs of various loads in China to clarify the influence factors of WFGD on PM₁₀ emissions. After WFGD, the proportion of PM_2.5 to PM₁₀ in the outlet flue gas increases, which showed that the WFGD system is selective in the capture of PM, with a significant effect on the capture of large particle sizes. The investigation found that four spray layers have a better effect on the capture of particles than two spray layers. Additionally, the investigation also found that unit load is not the main factor affecting the efficiency of PM₁₀ capture by WFGD. Instead, the factors affecting the capture efficiency of PM₁₀ by WFGD are the inlet flue gas temperature and the dust concentration. Relatively higher inlet flue gas temperature and lower inlet dust concentration will both result in higher emission of PM_0.1~1 from the WFGD outlet. These findings suggest that a matched integration of WFGD and CFPP is essential for ultra-low PM emission control and green industry development.

1. Introduction

China has abundant coal reserves, and the country is one of the largest consumers of coal resources in the world. Coal-fired power plants (CFPPs) are one of the main sources of pollutants, including gaseous pollutant NO_x, SO₂, and particulate matter (PM), which are harmful for both the environment quality and human health [1,2]. PM, in particular, can accumulate toxic substances through respiratory exposure due to its high specific surface area and formation mechanism [3,4,5,6]. Given the significant impact of CFPPs on air quality, it is crucial to control their emissions of pollutants, especially PM smaller than 2.5 μm, from which we can reap significant economic and environmental benefits [7,8].

Hence, the Chinese government has been making CFPPs a key regulatory source for controlling air pollution [9]. In 2014, China issued the “Reformation and Upgrading Action Plan for Coal Energy Conservation and Emission Reduction (2014–2020)” [10], which requires ultra-low emission standards of less than 10, 35, and 50 mg/Nm³ (standard temperature and pressure (273.15 K, 1 atm) with 6% O₂) for PM, SO₂, and NO_x, respectively. More stringently, dust emissions were required to be less than 5 mg/Nm³.

In China, the control of three primary air pollutants—NO_x, PM, and SO₂—is a priority for coal-fired boilers. To achieve this, selective catalytic reduction (SCR), electrostatic precipitator (ESP), and wet flue gas desulfurization (WFGD) technologies are widely installed in sequence [11,12,13,14]. Although ESP can remove up to 99% of fly ash, it is less effective for fine particles (smaller than 2.5 µm), and achieving the 5 mg/Nm³ emission standard is challenging [15,16,17]. Since the first successful ultra-low emission CFPPs demonstration in 2014 [18], research on PM emissions has been increasing. Studies of CFPP emissions suggest that the comparison of fine and coarse particle ratios may vary depending on the treatment or source of the emissions [19], but the dominance of fine particles is dangerous to human health [20] because fine particles can penetrate more deeply into the lungs than coarse particles [21].

In general, there is a strong association between sulfur dioxide emissions and particulate matter: fuels containing high levels of sulfur are associated with high levels of sulfate and ash, which increase the organic particulates produced during combustion, and higher sulfur levels in fuels are also associated with larger particle sizes [22]. There are dry desulfurization, semi-dry desulfurization, and wet desulfurization technologies for flue gas desulfurization. In China, WFGD technology is a more mature and widely used desulfurization technology, which uses a liquid desulfurizer, so it uses alkaline liquid desulfurizer to absorb SO₂ in the flue gas [23], such as limestone-gypsum desulfurization [24], double alkali desulfurization [25], magnesium oxide desulfurization [26], etc. Wet electrostatic scrubbing (WES) and WFGD technology can remove dust to some extent while desulfurizing [27]. Previous studies have investigated the removal efficiency of WFGD for PM in different units. For instance, Liu et al. [28] reported WFGD removal efficiencies of 28.7% and 39.6% for PM_0.1 and PM_2.5, respectively, in a 1000 MW unit. Li et al. [29] investigated the concentration and composition of PM in ultra-low emission power plants and the overall PM elimination efficiency was extremely good for the combination of ESP, WFGD, and WESP. Sui et al. [30] investigated the distribution and composition of PM₁₀ imported and exported from ESP, WFGD, and WESP after the ultra-low emission retrofit and it was found that the combined application of low-temperature economizer, retrofitted ESP, FGD, and WESP can reduce fine particle emissions from coal-fired power plants to less than 1 mg/m³. Wu et al. [31] proposed a non-homogeneous condensation process of limestone-gypsum desulfurization combined with waste heat recovery, and the results showed that the inlet flue gas with high relative humidity can remove fine particulate matter, and the high liquid-gas ratio of WFGD system is also beneficial to the effective removal of fine particulate matter. Yang et al. [32] investigated the control of pollutants in coal-fired power plants by WFGD and WESP. The removal efficiencies of PM₁₀ by WFGD and WESP were 64.0–70.3% and 57.5–79.2%, respectively, with WFGD having a greater effect on the particle size distribution. The studies mainly focused on one or two plants, and little attention was paid to the factors influencing WFGD efficiency for dust removal.

The main goal of this research was to investigate the effects of WFGD systems on PM emission, including PM₁₀ and PM_2.5 concentration, diameter distribution, and their ratio, in WFGD systems in five industrial CFPPs of different loads in China, using an electrical low-pressure impactor (ELPI) system to measure particle size distribution at the inlet and outlet of the WFGD. The five CFPPs were selected to represent coal-fired boilers for ultra-low emission retrofit in China, ranging from a 75 t/h industrial boiler to 600 MW units. The study analyzed the grade efficiency and emission ratio of particles with different diameters. Additionally, the effect of spray layer amount, inlet flue gas temperature, and dust concentration on PM₁₀ emissions from the wet desulfurization outlet was investigated in four 330 MW units at Power Plant C. This study provides practical data covering plants of multiple types and investigates various factors affecting the efficiency of WFGD for dust removal. This study provided a large amount of data based on actual industrial boiler applications, and the effect factors of the combination of ESP and FGD have been evaluated; more importantly, the contribution of different operating parameter has been tested, which can be used to improve the design and operation of the integration of WFGD and ESP systems in CFPPs, ultimately reducing the emission of harmful pollutants and protecting the environment and human health.

2. Experimental Section

2.1. APCDs in the CFPPs

This study conducted tests on the PM₁₀ mass concentrations at the inlet and outlet of wet desulfurization towers in five coal-fired power plants in China, with unit loads ranging from 75 t/h to 600 MW. These plants represent the ultra-low emission retrofit of coal-fired boilers in China. This study included detailed information on the characteristics of each plant, as shown in

Table 1. Description of the studied plants.

Description	Plant A	Plant B	Plant C	Plant D	Plant E
boiler type	CFB boiler	PC boiler	PC boiler	PC boiler	CFB boiler
installed capacity (MW)/BMCR (t/h)	3 × 15/75	4 × 220/670	4 × 330/1100	2 × 600/1900	2 × 135/440
flue gas flow (×104 m3/h)	10	45	110	145	88
flue gas temperature (°C)	140	95	140 (110)	-	120
mist eliminator type	Tubular type	2nd roof ridge type	3rd roof ridge type	3rd roof ridge type	3rd roof ridge type+ Mechanical grille type
APCDs	SCR + ESP + WFGD	SCR + LTE + ESP + WFGD	SCR + (LTE) + ESP + WFGD	SCR + ESP + WFGD	SCR + LTE + ESP + WFGD

, including the air pollution control devices (APCDs) installed in each, such as SCR for NO_x control, low-temperature economizer (LTE) for heat utilization, ESP for particle removal, and WFGD for the further reduction of PM and SO₃ emissions. Figure 1a provides a schematic diagram of the APCDs and sampling points in the CFPPs. The ELPI system was used to sample particles and measure the particle size distribution. This study also investigated the effect of different inlet flue gas temperatures and concentrations and the number of spray layers on PM₁₀ emissions from the wet desulfurization outlet in four 330 MW units at Power Plant C. Overall, the study aimed to analyze the emission and removal characteristics of PM₁₀ in WFGD and provide information on factors affecting the efficiency of WFGD for dust removal.

2.2. Total Particle Concentration and Flue Gas Temperature Test

The test method for total PM concentration in this paper mainly refers to China’s “Sampling Method for Particulate Matter and Gaseous Pollutants in Exhaust from Stationary Sources” GB/T16157-1996 [33], using the Pitot tube dynamic pressure balance isometric sampling method. The instrument used in this study for the testing of total particulate matter concentration was a Pitot tube parallel automatic smoke sampler, model WJ-60B, produced by Qingdao Laoshan Electronic Instrument General Factory Co. (Qingdao, China), and the main technical specifications of the sampler are shown in

Table 2. Main technical specifications of WJ-60B sampler.

Description	Range	Resolution	Accuracy
Sampling flow rate	5~60 L/min	0.1 L/min	±2.5%
Isometric sampling flow rate	1~45 m/s	0.1 m/s	<±4%
Flue gas dynamic pressure	0 Pa~2000 Pa	1 Pa	<±1.5%
Flue gas static pressure	−20 kPa~20 kPa	0.01 kPa	<±4%
Flue gas full pressure	−20 kPa~20 kPa	0.01 kPa	<±4%
Pressure before flow meter	−30 kPa~0 kPa	0.01 kPa	<±2.5%
Temperature before flow meter	−20 °C~100 °C	0.1 °C	<±1.5%
Flue gas temperature	0 °C~500 °C	1 °C	≤±3 °C
Dry bulb temperature	0~100 °C	0.1 °C
Wet bulb temperature	0~100 °C	0.1 °C
Humidity content	0~60%	0.1%	<±1.5%

. The sampler is also equipped with a temperature sensor for testing the flue gas temperature while testing the PM concentration.

2.3. PM₁₀ Graded Concentration Test

In this study, the particle size distribution of PM₁₀ was measured using an ELPI from DEKATI, Finland, which combines electrical detection with a conventional cascade impactor to acquire real-time aerosol measurement [34]. The ELPI is capable of the real-time measurement of both number and mass particle size distribution [35]; the main specifications of ELPI are shown in

Table 3. Main technical specifications of ELPI.

Description	Range
Rated flow rate	10 L/min
Range of particle size	0.03~10 μm
Number of sampling trays	12
Operating temperature	5~40 °C
Operating humidity	0~60% RH
Flue gas temperature	−30 kPa–0 kPa
First stage sampling tray pressure	100 mBar
Response time	<5 s

. Figure 1b,c show the schematic diagram of the ELPI, which consists of a charger and a sampler. The particles are charged in the charger and then enter the sampler, which has 12 independent sampling trays with microcurrent probes. The microcurrent probes capture the current signal and convert it into particle size distribution.

In addition, the ELPI was compared to the results of simultaneous sampling tests with conventional membrane and cartridge sampling guns under the same flue gas conditions in order to show the stability and reliability of ELPI. In Plant E, a filter membrane sampler and a filter cartridge sampler were used alongside ELPI. Each test was conducted simultaneously under the same flue gas conditions for a duration of 50 min. The conventional samplers pumped approximately 1000 L of gas while the ELPI recorded 3000 data points at a rate of one test result per second. The distribution of the flue gas flow field at the sampling point is not entirely uniform, causing fluctuations in the dust concentration. Therefore, the average of continuous 60 data points was used to determine the average mass concentration of PM per minute.

2.4. Particle Matter Sampling Description

The particle size distributions (PSDs) are presented with the expression of D_p—dM/dlog(D_p) [36]:

$$\mathrm{d}M/\mathrm{dlog}\left(D_{\mathrm{p}}\right)=\frac{\Delta M}{\log D_{\mathrm{p},\mathrm{upper}}-\log D_{\mathrm{p},\mathrm{lower}}}$$

(1)

where $D_{\mathrm{p}}$ is the aerodynamic diameter, μm. $\Delta M$ (mg·m⁻³) is the mass concentration in each stage. $\log D_{\mathrm{p},\mathrm{upper}}$ and $\log D_{\mathrm{p},\mathrm{lower}}$ are the upper and lower cut-off diameters of each stage, respectively.

Based on the mass concentration of PM, the grade efficiency $\eta$ can be obtained using the following expression [37]:

$$\eta =\frac{c_{inlet}-c_{outlet}}{c_{inlet}}\times 100\%$$

(2)

where $c_{inlet}$ and $c_{outlet}$ are the mass concentrations of PM at the inlet and outlet of WFGD in mg/m³, respectively.

3. Results and Discussion

3.1. Comparison of ELPI and Conventional Sampler

The results of the comparison indicated that the PM₁₀ mass concentration ranged from 1 mg/m³ to 7 mg/m³, with an average value of 2.84 mg/m³ and a mean square deviation of 77%, demonstrating the reliability of the ELPI measurement.

The real-time and every minute test results of ELPI are shown in Figure 2a. The PM concentration varied smoothly between 2 mg/m³ to 4 mg/m³. The ELPI instrument can be considered a reliable method of expressing dust concentration, as the average concentration in any 60-s period only varies by 2 mg/m³.

To compare the data obtained from ELPI with traditional sampling methods, the study divided the 3000-s ELPI data into six parts and calculated their average values separately. As shown in Figure 2b, three out of five tests using the filter cartridge sampler resulted in negative concentrations, with a mean value of −0.44 mg/m³ and a mean square deviation of 151%. Two out of five tests using the filter membrane sampler also resulted in negative concentrations, with a mean value of 2.02 mg/m³ and a mean square deviation of 518%. In contrast, all six groups of data sampled using ELPI were positive, with a mean value of 2.84 mg/m³ and a mean square deviation of 56.7%. Therefore, the test results of the traditional filter membrane sampler and filter cartridge sampler were highly deviated and lacked credibility. On the other hand, the test results obtained from ELPI were evenly distributed, showed a smooth change trend, and had little deviation. As a result, ELPI was used for all soot concentration tests in this study.

Using ELPI, Figure 2c,d display the mass concentration and ratio of PM_2.5 and PM₁₀ at the inlet of the WFGD at different operating capacities, as well as the mass concentration of PM_2.5 and PM₁₀ at the inlet of the WFGD in different plants.

3.2. Characteristics of PM Emission at the Inlet and Outlet of the Industrial Boiler

Plant A has an industrial boiler with a rated evaporation capacity of 75 t/h. To examine the changes in PM concentration after wet desulfurization under different concentration conditions at the desulfurization inlet, two boiler loads (high and low) of 75 t/h and 48 t/h, respectively, were used during the experiment. Figure 3a,b illustrate the particle size distribution and grade efficiency in the desulfurization process for Plant A at 48 t/h and 75 t/h loads.

3.2.1. For Plant A at 48 t/h Load

At 48 t/h load, the WFGD tower had a PM₁₀ mass concentration of 11.69 mg/m³ at the inlet and 3.76 mg/m³ at the outlet, with a capture efficiency of 67.84% for PM₁₀. The PM_2.5 mass concentration at the inlet was 1.90 mg/m³, which reduced to 1.03 mg/m³ at the outlet. The capture efficiency of the WFGD tower for PM_2.5 was 45.79%.

Figure 3c illustrates the emission ratio of particles with different diameters at the inlet and outlet of WFGD for Plant A, at 48 t/h load. The mass concentration of PM_1~10 decreased the most, by approximately 39.7%, followed by PM_0.5~1, which decreased by about 21.1% after desulfurization. However, both PM_0.1 and PM_0.1~0.5 showed a slight increasing trend after the process.

Figure 3e depicts the real-time distribution of mass concentration of PM₁₀ and PM_2.5 at the inlet and outlet of WFGD for Plant A. At the lower load, the average mass concentrations of PM_2.5 were 1.90 mg/m³ and 1.03 mg/m³ at the inlet and outlet, respectively, whereas the average mass concentrations of PM₁₀ were 11.69 mg/m³ and 3.76 mg/m³ at the inlet and outlet, respectively. The mass concentration ratio of PM_2.5 to PM₁₀ was 16.25% at the inlet and 27.39% at the outlet.

3.2.2. For Plant A at 75 t/h Load

At 75 t/h load, the PM₁₀ mass concentration of the WFGD tower at the inlet and outlet were 13.29 mg/m³ and 3.87 mg/m³, respectively. The WFGD tower had a capture efficiency of 70.88% for PM₁₀. The PM_2.5 mass concentration at the inlet was 2.08 mg/m³, which reduced to 1.82 mg/m³ at the outlet. However, the capture efficiency of the WFGD tower for PM_2.5 was only 12.50%.

Figure 3d illustrates the emission ratio of particles with different diameters at the inlet and outlet of WFGD for Plant A, at 75 t/h load. All PM_0.1~10 mass concentrations decreased, with PM_0.1 mass concentration increasing after desulfurization. After the desulfurization process, the mass concentration of PM_1~10 changed significantly, whereas the change in PM₁ mass concentration was relatively small. However, PM_0.5 mass concentration showed an increasing trend, primarily due to the uneven distribution of local airflow in the desulfurization tower, resulting in an increased secondary carryover of desulfurization slurry, and leading to increased particle emissions in PM_0.5.

Figure 3f depicts the real-time distribution of mass concentration of PM₁₀ and PM_2.5 at the inlet and outlet of WFGD for Plant A. At the higher load, the average mass concentrations of PM_2.5 were 2.08 mg/m³ and 1.82 mg/m³ at the inlet and outlet, respectively, whereas the average mass concentrations of PM₁₀ were 13.29 mg/m³ and 3.87 mg/m³ at the inlet and outlet, respectively. The mass concentration ratio of PM_2.5 to PM₁₀ was 15.65% at the inlet and 47.03% at the outlet. The ratio increased in the outlet due to the higher capture efficiency of PM_1~10 by WFGD and the large proportion of PM_1~10 in PM₁₀.

The WFGD process effectively captures PM₁₀, resulting in a more concentrated real-time distribution of PM_2.5 and PM₁₀ mass concentrations. Notably, the capture efficiency of PM₁₀ is closely related to the boiler load. As the load increases, the mass concentration of PM₁₀ at the inlet also increases, but so does the capture efficiency. The higher the operating load, the higher the particulate capture efficiency for the same APCD case, which is consistent with the results of previous studies [29]. However, changes in boiler load do not appear to affect the particle size distribution characteristics of PM₁₀ before and after desulfurization.

3.3. Characteristics of PM Emission at the Inlet and Outlet of Coal-Fired Boilers

3.3.1. For Plant B with 220 MW

Plant B has a rated generating capacity of 220 MW and was operated at full load during the experiment. Figure 4a illustrates the particle size distribution and grade efficiency of PM₁₀ at the inlet and outlet of WFGD. The results show that the mass concentration of PM_0.1~5 increased to varying degrees. The PM₁₀ mass concentration at the WFGD tower was 7.78 mg/m³ at the inlet and 7.45 mg/m³ at the outlet, with a capture efficiency of 4.24% for PM₁₀. The PM_2.5 mass concentration was 1.29 mg/m³ at the inlet and 1.36 mg/m³ at the outlet, representing a 5.14% increase. Figure 4b presents the emission ratio of particles with different diameters at the inlet and outlet of WFGD for Plant B. The results show that after desulfurization, the mass concentration of PM_0.1 significantly reduced by 93.7%, with an absolute value of 0.0364 mg/m³. However, the mass concentrations of PM_0.1~10 all increased to varying degrees, with the mass concentration of PM_0.1~1 increasing by 149.21%, PM_1~2.5 increasing by 29.63%, and PM_2.5~10 increasing by 5.08%.

The real-time distribution of mass concentration of PM_2.5 and PM₁₀ at the inlet and outlet of the WFGD were analyzed, as shown in Figure 4c, indicating that the distribution range of PM₁₀ mass concentration before and after desulfurization is very similar. The average mass concentrations of PM_2.5 were 1.29 mg/m³ at the inlet and 1.36 mg/m³ at the outlet, whereas the average mass concentrations of PM₁₀ were 7.78 mg/m³ at the inlet and 7.45 mg/m³ at the outlet. The mass concentration ratio of PM_2.5 to PM₁₀ was 16.58% at the inlet and 18.26% at the outlet. These results indicate that PM_2.5 constitutes a larger proportion of PM₁₀ after wet desulfurization, which is consistent with the conclusion of previous studies that ESP and WFGD are more effective in removing coarse particles than fine particles [32,37,38].

3.3.2. For Plant E with 135 MW

Plant E comprises two 135 MW coal-fired units that are structurally identical with the same size of WFGD towers and similar flue gas operating conditions, with the only difference being the mist eliminator. Specifically, Unit 1 uses a three-stage roof mist eliminator, whereas Unit 2 still employs the traditional mechanical grille mist eliminator. This study focuses on comparing the performance of the two mist eliminators under the same operating conditions.

According to Figure 5a,c, the PM₁₀ mass concentration distribution at the WFGD inlet is similar for both Unit 1 and Unit 2 under the same operating conditions, with PM₁₀ mass concentrations of 14.87 mg/m³ and 15.26 mg/m³, respectively. The PM_2.5 mass concentrations for Unit 1 and Unit 2 are 1.02 mg/m³ and 1.23 mg/m³, respectively, and the PM_2.5 mass concentrations as a percentage of PM₁₀ are 6.86% and 8.06%, respectively. Figure 5b shows that after wet desulfurization, the PM_2.5 outlet mass concentration increased in both Unit 1 and Unit 2 by 9.80% and 69.10%, respectively. The PM₁₀ outlet mass concentration of Unit 1 decreased by 51.71%, whereas that of Unit 2 increased by 53.93%.

Additionally, Figure 5c shows the real-time mass concentration distribution of PM_2.5 and PM₁₀ at the WFGD inlet and outlet of Unit 1 and Unit 2. The PM₁₀ mass concentration at the WFGD tower outlet of Unit 2 with the traditional mechanical grille mist eliminator is highly dispersed, with the instantaneous mass concentration of PM₁₀ reaching up to 150 mg/m³, the instantaneous mass concentration of PM_2.5 up to 10 mg/m³, and the average mass concentration of PM_2.5 accounting for 8.85% of PM₁₀. In contrast, the PM₁₀ mass concentration at the outlet of the WFGD tower of Unit 1 with the three-stage roof-type mist eliminator is more concentrated, and the average mass concentration of PM_2.5 is 15.60% of PM₁₀.

Moreover, the movement of different particle sizes is influenced by different flow fields. The flow field inside the traditional mechanical grille mist eliminator is relatively smooth and uniform. PM_0.5~10 particles are subject to significant flow field traction, making it easier for them to escape from the mist eliminator by following the flow field. PM_0.5 particles are subject to less flow field traction, but due to their Brownian motion and electrostatic effects, the probability of collision with the surface of the mechanical grille mist eliminator increases.

In conclusion, the traditional mechanical grille has a higher capture efficiency for PM_0.5 particles. On the other hand, the three-stage ridge type mist eliminator has a higher capture efficiency for PM_0.5~10 particles.

3.3.3. For Plant D with 600 MW

Plant D comprises two 600 MW coal-fired units. The real-time mass concentration curves of PM₁₀ and PM_2.5 at the inlet and outlet of WFGD, using an ELPI instrument, are presented in Figure 6a. The smoothness of the curves indicates the accuracy and stability of the ELPI-based thermostatic and dilution sampling method for testing PM₁₀ of high temperature and high humidity flue gas. Figure 6b presents the particle size distributions and grade efficiency of PM₁₀. At the WFGD inlet, the PM₁₀ mass concentration was 14.48 mg/m³, with a PM_2.5 mass concentration of 0.94 mg/m³, accounting for 6.49% of PM₁₀. At the WFGD outlet, the PM₁₀ mass concentration was 3.09 mg/m³, with a PM_2.5 mass concentration of 0.62 mg/m³, accounting for 20.06% of PM₁₀, which was 13.57% higher than that at the WFGD inlet. The capture efficiency of WFGD for PM₁₀ was 78.66% and for PM_2.5 was 34.04%. Figure 6c shows the emission ratio of particles with different diameters, highlighting that the capture efficiency for PM_0.1~1 is relatively low, whereas that for PM_0.1 and PM_1~10 is relatively high. Figure 6d presents the real-time distribution of the mass concentration of PM₁₀ and PM_2.5 at the inlet and outlet of WFGD of Plant D, revealing a more concentrated and credible range of PM_2.5 and PM₁₀ mass concentration distribution at the WFGD inlet and outlet.

3.4. Influencing Factors of PM₁₀ Emissions from Wet Desulfurization

Plant C has four 330 MW coal-fired units, all of which have undergone ultra-low emission retrofit. This study examines the PM₁₀ capture characteristics of wet desulfurization, focusing on the following factors: (1) number of spray layers, (2) flue gas temperature at the desulfurization inlet, and (3) PM concentrations at the desulfurization inlet. The emission concentrations of particulate matter for each unit were monitored after the completion of the retrofit.

3.4.1. The Influence of the Number of Spraying Layers

Experiments were conducted in the Unit 1 WFGD tower of Plant C to compare the capture efficiency of different numbers of spraying layers. The WFGD inlet mass concentration of PM₁₀ was kept constant, and the distribution of PM₁₀ mass concentration at the WFGD outlet was compared under two and four spraying layers by shutting down the corresponding slurry circulation pumps. Figure 7a shows the particle size distribution and grade efficiency at different numbers of spraying layers. The PM₁₀ mass concentration was 14.36 mg/m³ at the WFGD inlet, with PM_2.5 mass concentration accounting for 12.81% of PM₁₀, and the PM_2.5 mass concentration was 1.84 mg/m³. Figure 7a,b indicate that the capture efficiency of WFGD is relatively poor for PM_1~2.5 and more efficient for PM_0.1 and PM_2.5~10. The overall PM₁₀ grade efficiency improves when the number of spray layers is increased from two to four. Figure 7c shows the real-time distribution of mass concentration of PM₁₀ and PM_2.5 at the inlet and outlet of the four spraying layers of Plant C. The PM₁₀ mass concentration was 4.05 mg/m³, with PM_2.5 mass concentration accounting for 29.38% of PM₁₀, and PM_2.5 mass concentration was 1.19 mg/m³ at the outlet of WFGD of four spraying layers. The capture efficiency of PM₁₀ in the four-layer spraying WFGD tower is 71.80%, and the capture efficiency of PM_2.5 is 35.33%. Additionally, the mass concentration distribution of PM₁₀ in the WFGD outlet is extremely concentrated, indicating that the capture efficiency of PM₁₀ in the four-layer spraying is extremely high and stable. However, compared to two sprays, in the case of four sprays, the submicron PM concentration at the outlet is higher and the PM_0.1 capture efficiency decreases because more small droplets are entrained by the flue gas, leading to an increase in fine particulate matter [32]. Small droplets cannot be fully captured by the mist eliminator, and some of them penetrate, leading to an increase in submicron particles [39].

3.4.2. The Influence of the Temperature of Inlet Flue Gas

The WFGD of Unit 2 in Power Plant C had an inlet flue gas temperature of 140 °C, and a low-temperature economizer was later added to the inlet of the electric precipitator. The inlet flue gas temperature dropped to 100 °C after the addition, and the flue gas temperature at the outlet of the desulfurization tower remained relatively stable. Figure 8a presents the particle size distributions and grade efficiency of PM₁₀. At the flue gas temperature of 140 °C, the total mass concentration of PM₁₀ at the inlet of desulfurization was 17.37 mg/m³, whereas the total mass concentration of PM_2.5 was 2.06 mg/m³, and the ratio of PM_2.5 mass concentration to PM₁₀ was 11.86%. In comparison, the total mass concentration of PM₁₀ at the outlet of desulfurization was 3.79 mg/m³, the total mass concentration of PM_2.5 was 0.80 mg/m³, and the ratio of PM_2.5 mass concentration to PM₁₀ was 21.22%. The ratio of PM_2.5 mass concentration to PM₁₀ increased by 9.36% before and after WFGD. The capture efficiency of WFGD was 78.20% for PM₁₀ and 61.00% for PM_2.5.

At the flue gas temperature of 100 °C, the total mass concentration of PM₁₀ at the inlet of WFGD was 14.31 mg/m³, whereas the total mass concentration of PM_2.5 was 4.18 mg/m³, and the ratio of PM_2.5 mass concentration to PM₁₀ was 29.21%. The total mass concentration of PM₁₀ at the outlet of WFGD was 4.52 mg/m³, the total mass concentration of PM_2.5 was 1.61 mg/m³, and the ratio of PM_2.5 mass concentration to PM₁₀ was 35.54%. The ratio of PM_2.5 mass concentration to PM₁₀ increased by 6.33% before and after WFGD. The capture efficiency of WFGD was 68.39% for PM₁₀ and 61.54% for PM_2.5.

Figure 8b shows that the mass concentration of PM_0.5~10 is reduced after desulfurization at both 140 °C and 100 °C flue gas temperatures. However, for PM_0.5, the concentration increases after desulfurization at 140 °C flue gas temperature and decreases after desulfurization at 100 °C flue gas temperature. The higher inlet flue gas temperature leads to an increase in the fine particle concentration at the WFGD outlet, which is consistent with the findings of his study [31]. This phenomenon is mainly attributed to the large temperature difference between the flue gas and the sprayed desulfurization slurry under high flue gas temperature conditions. When the flue gas comes into contact with the desulfurization slurry in the desulfurization tower, the evaporation of the desulfurization slurry increases the water content of the gas and reduces the gas temperature, which depends, to a large extent, on the temperature difference between the flue gas and the desulfurization slurry [40]. During the heat exchange between the flue gas and the desulfurization slurry, the thermal energy of the slurry particles formed by spraying increases, making them more easily carried out by the airflow. In addition, the roof-type mist eliminator is highly efficient for capturing PM_0.5~10, but has low efficiency for capturing PM_0.5.

3.4.3. The Influence of the Different PM Concentrations at the Desulfurization Inlet

In Power Plant C, Unit 4, the flue gas temperature at the inlet of the electric precipitator and wet desulfurization system is around 110 °C. By adjusting the electric field operation parameters of the electric precipitator, two PM₁₀ mass concentration levels are formed at the wet desulfurization inlet: high and low, as shown in Figure 9a.

Under high concentration conditions, the total PM₁₀ mass concentration at the desulfurization inlet is 19.32 mg/m³, with a total PM_2.5 mass concentration of 4.68 mg/m³, and the proportion of PM_2.5 mass concentration to PM₁₀ is 24.21%. Under low concentration conditions, the total PM₁₀ mass concentration at the WFGD inlet is 8.40 mg/m³, with a total PM_2.5 mass concentration of 1.77 mg/m³, and the proportion of PM_2.5 mass concentration to PM₁₀ is 21.09%.

Figure 9b shows the PM_2.5 and PM₁₀ ratio at the WFGD inlet and outlet under high and low concentration conditions. After WFGD treatment, the PM₁₀ mass concentration is reduced by 79.72% and PM_2.5 mass concentration is reduced by 63.80% under a high-concentration WFGD inlet PM₁₀ mass concentration of about 20 mg/m³. Under a low-concentration WFGD inlet PM₁₀ mass concentration < 10 mg/m³, the PM₁₀ mass concentration is reduced by 32.03%, but the PM_2.5 mass concentration increases by 14.01%.

The efficiency of PM₁₀ capture by WFGD is dependent on the mass concentration of PM₁₀ at the WFGD inlet. The higher the mass concentration of PM₁₀ at the inlet, the greater the capture efficiency. This is because a higher mass concentration results in a greater number concentration, leading to an increased likelihood of collision and combination between particulate matter and slurry particles. However, for PM_2.5, a low concentration at the WFGD inlet causes a significant decrease in number concentration, making it difficult to capture PM_2.5 through WFGD. In addition, the secondary carrying effect of airflow increases PM_2.5 emissions at the WFGD outlet.

3.5. Summary

In this study, data from the experimental power plant were collected and analyzed. Figure 10a displays the statistical results of the ratio of PM_2.5 mass concentration to PM₁₀ at the inlet and outlet of wet desulfurization. The average ratio of PM_2.5 mass concentration to PM₁₀ at the WFGD inlet is 16.25%, ranging from 6.5% to 28.74%, whereas the average ratio at the WFGD outlet is 28.11%, ranging from 17.38% to 64%. The ratio of PM_2.5 mass concentration to PM₁₀ at the WFGD outlet is significantly higher than that at the inlet, indicating particle size selectivity of the WFGD for the capture of particulate matter.

Figure 10b shows the statistical results of the capture efficiency of PM₁₀ and PM_2.5 by WFGD. The average capture efficiency of PM₁₀ by wet desulfurization is 51.39%, whereas the capture efficiency of PM_2.5 is about 16.88%. WFGD has a significant effect on capturing large particle size (PM₁₀) compared to small particle size (PM_2.5).

Figure 10c displays the inlet and outlet concentrations of PM_2.5 and PM₁₀ in the WFGD tower at different plant operating capacities and the percentage of PM_2.5 relative to PM₁₀. The results indicate that the higher the operating capacity, the higher the inlet PM₁₀ concentration, the lower the PM_2.5 concentration, and the smaller the PM_2.5 to PM₁₀ ratio. After wet desulfurization, the PM₁₀ export concentration decreases significantly at different operating capacities, whereas the PM_2.5 export concentration does not change significantly. Furthermore, the proportion of PM_2.5 to PM₁₀ is generally higher than that at the inlet.

4. Conclusions

In this study, we investigated the effects of WFGD on the particle size distribution and grade efficiency of PM₁₀ in five different plants. Our main findings can be summarized as follows:

(1)

The ELPI is an effective test method with more concentrated distribution, lower bias, and higher confidence in the test data of particle matter concentration than filter membrane sampler and filter cartridge sampler.

(2)

WFGD showed significant capture efficiency for PM₁₀ compared to PM_2.5, indicating that WFGD has particle size selectivity for the capture of particulate matter, with a better capture effect on larger particles and poorer capture effect on smaller particles.

(3)

The three-stage roof-type mist eliminator in the tower of WFGD had higher PM₁₀ capture efficiency than the conventional mechanical grid mist eliminator and two-stage ridge-type mist eliminator.

(4)

Compared to the two-layer spray, the four-layer spray in the WFGD tower is more beneficial in reducing PM₁₀ and PM_2.5 emissions, but its capture efficiency for submicron particles is lower because more small droplets are entrained.

(5)

The PM₁₀ capture efficiency of WFGD is affected by WFGD inlet flue gas temperature and particle matter concentration. Higher inlet flue gas temperature and lower inlet particle matter concentration can both increase WFGD outlet PM_0.1~1 emissions.