Simultaneous removal of sulfur dioxide and polycyclic aromatic hydrocarbons from incineration flue gas using activated carbon fibers

Abstract

Incineration flue gas contains polycyclic aromatic hydrocarbons (PAHs) and sulfur dioxide (SO₂). The effects of SO₂ concentration (0, 350, 750, and 1000 ppm), reaction temperature (160, 200, and 280 °C), and the type of activated carbon fibers (ACFs) on the removal of SO₂ and PAHs by ACFs were examined in this study. A fluidized bed incinerator was used to simulate practical incineration flue gas. It was found that the presence of SO₂ in the incineration flue gas could drastically decrease removal of PAHs because of competitive adsorption. The effect of rise in the reaction temperature from 160 to 280 °C on removal of PAHs was greater than that on SO₂ removal at an SO₂ concentration of 750 ppm. Among the three ACFs studied, ACF-B, with the highest microporous volume, highest O content, and the tightest structure, was the best adsorbent for removing SO₂ and PAHs when these gases coexisted in the incineration flue gas.

Implications

Simultaneous adsorption of sulfur dioxide (SO₂) and polycyclic aromatic hydrocarbons (PAHs) emitted from incineration flue gas onto activated carbon fibers (ACFs) meant to devise a new technique showed that the presence of SO₂ in the incineration flue gas leads to a drastic decrease in removal of PAHs because of competitive adsorption. Reaction temperature had a greater influence on PAHs removal than on SO₂ removal. ACF-B, with the highest microporous volume, highest O content, and tightest structure among the three studied ACFs, was found to be the best adsorbent for removing SO₂ and PAHs.

Introduction

Incineration is the main technique for treating municipal solid waste in Taiwan. Some pollutants, including organic compounds, sulfur dioxide (SO₂), heavy metals, and fly ash, may be emitted during incineration. To meet the stringent emission standards for various contaminants in Taiwan, a spray dryer integrated with a bag filter is used conventionally as an air pollution control device (APCD) for incinerators, and powdered activated carbon is injected into flue gas before it enters the bag filter in order to remove the organic compounds. However, the spray dryer suffers from some disadvantages, such as blockage of the spray nozzle and production of sludge. Therefore, it is important to devise a new technique to replace the conventional APCD for incinerators.

Adsorption techniques are widely used to remove organic compounds from industrial waste gas streams, and powdered activated carbon, activated carbon fibers (ACFs), aluminum oxide, silica gel, and zeolite are commonly used adsorbents. Among these adsorbents, ACFs offer many advantages, including higher adsorption and desorption rates, much higher surface areas, a uniform micropore structure, and a lower pressure drop (Fuertes et al., 2003; Huang et al., 2003; Lillo-Ródenas et al., 2005; Muñiz et al., 2000; Park and Kim, 2001; Yang et al., 2011). The use of ACFs for removing organic compounds from waste gas streams has been extensively studied, indicating that the removal efficiencies of organic compounds are affected by the specific surface area, pore size distribution, pore volume, and surface functional groups of the ACFs and by vapor pressure, molecular size, and molecular weight of organic compounds (Das et al., 2004; Gaur et al., 2006; Lin et al., 2013; Liu, 2006; Liu et al., 2012; Mangun et al., 2001; Yi et al., 2009).

However, most species of the organic compounds explored in previous studies are volatile organic compounds (VOCs), such as benzene, toluene, and xylene (Das et al., 2004; Gaur et al., 2006; Huang et al., 2003; Lin et al., 2013; Liu et al., 2012; Yi et al., 2009). The 16 species of polycyclic aromatic hydrocarbons (PAHs) listed by the U.S. Environmental Protection Agency (EPA) as priority pollutants are rarely discussed in research of ACFs (Liu, 2006). In this study, a fluidized bed incinerator was used to simulate practical incineration flue gas.

Incineration flue gas contains PAHs and SO₂. The presence of SO₂ in the incineration flue gas could affect the removal efficiency and adsorption capacity of PAHs. Liu (2006) previously determined that the removal efficiency of PAHs from incineration flue gas using ACFs exceeded 90%. Previous research has shown that SO₂ could be effectively removed by ACFs (Daley et al., 1997; Ling et al., 1999; Liu, 2008). For example, Liu et al. (2012) investigated the simultaneous removal of toluene and SO₂, and reported that SO₂ inhibited the adsorption of toluene on ACFs and the removal of toluene decreased with an increase in the concentration of SO₂. In this study, the simultaneous removal of PAHs and SO₂ from incineration flue gas and the effect of SO₂ on the removal of PAHs were examined at different SO₂ concentrations (0, 350, 750, and 1000 ppm), reaction temperatures (160, 200, and 280 °C), and for three types of ACFs. Three adsorption temperatures, 160, 200, and 280 °C, were investigated because the temperature of flue gas in the APCD is 200 °C (Chi et al., 2006).

Experimental

Experimental procedure

The three commercial ACFs used in this experiment were provided by Taiwan Carbon Technology Co., Ltd., in Taiwan. Polyacrylonitrile (PAN) was used as a precursor in the production of these three ACFs. The thicknesses of ACF-A, ACF-B, and ACF-C were 0.4, 1.2, and 0.4 mm, respectively. The densities of ACF-A, ACF-B, and ACF-C were 0.22, 0.07, and 0.29 g·cm⁻³, respectively. The ACFs were cleaned using deionized water and then dried in an oven at 105 °C for 24 hr before the experiment was conducted. Thereafter, they were loaded into the ACF adsorber. The height and inner diameter of the ACF adsorber were 150 and 80 mm, respectively.

The laboratory-scale fluidized bed incinerator integrated with an ACF adsorber used in this study is shown in Figure 1. The combustion chamber was heated to the desired temperature, 800 °C, by electrical heaters, and air entered the incinerator at a flow rate of 60 L·min⁻¹. When the temperature reached a steady state, polypropylene (PP) granules and sulfur enclosed in a small polyethylene (PE) bag was thrown into the incinerator at the rate of one sample per 15 sec to simulate the flue gas. Subsequently, the ACF adsorber was turned on and sampling was initiated as described in the next paragraph. Detailed compositions of feedstock and operating parameters can be seen in Table 1.

Figure 1. Experimental system of fluidized-bed incinerator and activated carbon fibers (ACFs) adsorber: (1) air compressor; (2) flowmeter; (3) combustion chamber; (4) electrical heater; (5) thermal feedback controller; (6) thermocouple; (7) feeder; (8) ACF adsorber; (9) sampling; (10) induced fan.

Table 1. Composition of feedstocks and operating conditions

Test	Composition of Artificial Feedstocks (g/sample)			ACF Type	SO2 Concentration (ppm)	Adsorption Temperature (°C)
	Polypropylene	Sulfur	PE Bag
Run 1		—		ACF-A	0	200
Run 2		0.02		ACF-A	350	200
Run 3		0.03		ACF-A	750	200
Run 4		0.04		ACF-A	1000	200
Run 5	0.60	0.03	0.22	ACF-A	750	280
Run 6		0.03		ACF-A	750	160
Run 7		0.03		ACF-B	750	200
Run 8		0.03		ACF-C	750	200

Notes: PE = polyethylene; ACFs = activated carbon fibers.

Sampling and analytical methods

To estimate the removal efficiency of PAHs and SO₂, the flue gas was sampled before entering and after exiting the ACF adsorber. The sampling time periods for SO₂ after and before the passage of the flue gas through the ACF adsorber were 2–5 and 6–9 min, respectively. A flue gas analyzer (model 2800P; IMR Environmental Equipment Inc., Florida, USA) was used to determine the concentration of SO₂. Before the experiments, the analyzer was calibrated with standard gases. The accuracy and resolution of the SO₂ analyzer were 5% and 1 ppm, respectively. The concentration range for the SO₂ monitor was 0–4000 ppm (Liu, 2008).

PAHs were sampled using the EPA Modified Method 5. The sampling time periods for PAHs after and before the passage of the flue gas through the ACF adsorber were 10–15 and 15–20 min, respectively. The sampling flow rate for PAHs was 10 L·min⁻¹. The removal efficiency of PAHs was defined as follows:

1

PAHs_inlet and PAHs_outlet represent the total concentration of 16 PAHs before entering and after exiting the ACF adsorber, respectively. Figure 2 shows the sampling train for PAHs. The gas passed through a heated filter packed with fiberglass to collect particles, and then through a cooling tube to capture the gaseous PAHs onto a XAD-4 adsorbent (Sigma-Aldrich Co., Missouri, USA). Next, the particles and XAD-4 containing PAHs were extracted using dichloromethane over 20 hr through the Soxhlet extraction process. The extraction solution was concentrated to 1 mL using a K-D evaporator concentrator (Chenghong Co., Kaohsiung, Taiwan), and these samples were analyzed using a gas chromatograph with a flame ionization detector (Shimadzu GC-2014; Kyoto, Japan) (Liu, 2006). The R² of all calibration curves are above 0.99.

Figure 2. Sampling train for PAHs: (1) sampling probe; (2) heated filter and heating hose; (3) thermometer; (4) cooling tube and XAD-4 adsorption tube; (5) 200 mL distilled water; (6) silica gel; (7) flow meter; (8) connect to vacuum pump.

Characterization of ACFs

The Brunauer-Emmett-Teller (BET) surface area and porosities of various ACFs were determined using an ASAP 2020 vacuum volumetric sorption instrument (Micromeritics Instrument Co., Georgia, USA) from the N₂ adsorption-desorption isotherms at 77 K. Before N₂ sorption analysis, the samples were preheated at 473 K for degassing and cooled at room temperature under vacuum conditions (Lin et al., 2013). The volume of the micropores (pores of size <2 nm) was calculated using the t-plot method. The mesopore volumes (pores of size 2–50 nm) were calculated by subtracting the micropore volumes from the total pore volume at P/P₀ = 0.98. The surface functional groups of the ACFs were identified using a Fourier-transform infrared (FTIR) spectrometer (Digilab FTX350; American Digilab, Massachusetts, USA). The detailed FTIR analysis procedure is described in Liu (2008). Additionally, the contents of C, N, O, and H on the ACFs were determined using an elemental analyzer (EA) (Elementar vario EL III; Hanau, Germany). A Zeiss Axioplan 2 microscope (Jena, Germany) equipped with an AxioCam HR color charge-coupled device camera (Jena, Germany) was used to analyze the photomicrographs of the ACFs.

Results and Discussion

Characterization of ACFs

The detailed specific surface area and porosity of the three ACFs, determined through N₂ sorption analysis, can be viewed in Table 2. The three ACFs have similar specific surface area and total pore volume. Whereas the micropore and mesopore volumes of ACF-A and ACF-C are similar, the micropore and mesopore volumes of ACF-B are higher and lower, respectively, than the similar corresponding volumes of ACF-A and ACF-C.

Table 2. Specific surface area and porosity of activated carbon fibers (ACFs) by N₂ isotherms

			t-Plot Method
Sample	SBET^a (m^2/g)	VT^b (cm^3/g)	Smicro^c (<2 nm) (m^2/g)	Vmicro^d (<2 nm) (cm^3/g)	Vmeso^e (2–50 nm) (cm^3/g)
ACF-A	1175	0.57	534	0.23	0.34
ACF-B	1142	0.51	914	0.41	0.10
ACF-C	1204	0.58	579	0.25	0.34

Notes: ^aSpecific surface area by the Brunauer-Emmett-Teller (BET) method. ^bTotal pore volume at P/P₀ = 0.98. ^cMicropore surface area. ^dMicropore volume. ^eMesopore volume.

The FTIR spectra of the three ACFs can be viewed in Figure 3. The spectra show peaks in the range 900–1300 cm⁻¹, which correspond to both –C–O stretching and –O–H bending modes, and at ˜1600 cm⁻¹, which corresponds to the –C=O groups (Fanning and Vannice, 1993; Gaur et al., 2006; Lin et al., 2013). Bands at ˜2380 cm⁻¹ and at ˜3800 cm⁻¹ in the FTIR spectrum of ACF-A are assigned to the ketene group (=C=O) and the nitrogen species, respectively (Gaur et al., 2006; Tsuji and Shiraishi, 1997). Table 3 lists the elemental analysis of the three ACFs and indicates that the O content of the three ACFs follows the order of ACF-B > ACF-C > ACF-A.

Figure 3. FTIR spectrum for different activated carbon fibers (ACFs).

Table 3. Element content of the three activated carbon fibers (ACFs) by an elemental analyzer

	Element Content (wt. %)
Sample	C	N	O	H	O/C
ACF-A	65.19	2.62	27.66	3.82	0.42
ACF-B	52.15	2.58	40.02	4.49	0.77
ACF-C	59.14	2.22	30.42	4.02	0.51

It is clear from Figure 4 that the knitted structure of ACF-B is tighter than that of ACF-A and ACF-C, which might affect its efficiency in removing SO₂ and PAHs. SO₂ and PAHs can be absorbed onto fly ash, and ACFs with tighter structures are very efficient at removing fly ash. In addition to the tightness of ACF knitted structures, the pore volume and surface functional groups of ACFs may affect the adsorption capacities and mechanisms of SO₂ and PAHs on ACFs. Other than type of ACFs, the presence of SO₂ in the incineration flue gas can decrease PAH removal drastically because of competitive adsorption. Additionally, reaction temperature may also affect the adsorption capacities and mechanisms of SO₂ and PAHs on ACFs. Additional detail regarding the effects of SO₂ concentration, reaction temperature, and the type of ACFs on the simultaneous removal of SO₂ and PAHs will be provided below.

Figure 4. The knitted structure of a piece of activated carbon fibers (ACFs): (a) ACF-A; (b) ACF-B; (c) ACF-C (25×).

Effect of SO₂ concentration on PAH removal

The effects of SO₂ concentration on the removal efficiencies of SO₂ and PAHs are apparent in Figure 5. The results indicate that the removal efficiency of PAHs is 90% in the absence of SO₂. When the concentration of SO₂ is 350 ppm, the removal efficiencies of SO₂ and PAHs are 92% and 33%, respectively. The presence of SO₂ in the incineration flue gas leads to a drastic decrease in removal of PAHs because of competitive adsorption. These results are in agreement with those of Liu et al. (2012), which reported that the presence of SO₂ decreased the removal of toluene because the polarity and the dipole moment of SO₂ were stronger than those of toluene. Hence, SO₂ first adsorbs onto a polar or basic adsorption site and then oxidizes to SO₃ in the presence of O₂, which prevents the PAH removal efficiency from increasing.

Figure 5. Effects of SO₂ concentration on the removal efficiencies of SO₂ and PAHs (reaction condition: reaction temperature = 200 °C, ACF type: ACF-A).

Figure 5 shows that the removal efficiency of SO₂ decreases markedly when the concentration of SO₂ increases to 750 ppm, and beyond this concentration no changes are apparent in removal of PAHs with increasing SO₂ concentration. This may occur because SO₂ was first adsorbed by ACFs when SO₂ and PAHs coexisted in the incineration flue gas and the effect of SO₂ concentration on SO₂ removal was greater than that on PAH removal. Further, when the concentration of SO₂ became very high, the molecules of SO₂ collided with each other, thereby reducing the number of opportunities for SO₂ not only to diffuse from the surface of the ACFs into the pores in the ACFs, but also to be adsorbed onto the ACFs (Liu, 2008). As a result, the removal efficiency of SO₂ decreased with increasing SO₂ concentration.

Effect of reaction temperature on SO₂ and PAH removal

The effects of reaction temperature on the removal efficiencies of SO₂ and PAHs are shown in Figure 6. SO₂ removal is unaffected by an increase in temperature from 160 to 280 °C at an SO₂ concentration of 750 ppm. Liu (2008) previously reported that an increase in reaction temperature from 150 to 260 °C had a considerable effect on SO₂ removal using ACFs modified with HNO₃ at an SO₂ concentration of 200 ppm, with the highest SO₂ removal efficiency observed of 80% at 200 °C. In this study, changes in the reaction temperature had less of an influence on SO₂ removal at a high SO₂ concentration (750 ppm) than at a low SO₂ concentration (200 ppm). This may be so because the removal efficiency of SO₂ at a high SO₂ concentration of 750 ppm (<30%) was too low to be affected by the reaction temperature.

Figure 6. Effects of reaction temperature on the removal efficiencies of SO₂ and PAHs (reaction condition: SO₂ concentration = 750 ppm, ACF type: ACF-A).

The results also show that the removal efficiency of PAHs follows the order of 280 °C > 160 °C > 200 °C. The concentration of two-ring PAHs (naphthalene) is much higher than the concentrations of other PAH species in incineration flue gas. The boiling point of naphthalene is 218 °C, which is close to 200 °C (Liu, 2006). Thus, the naphthalene is removed by ACFs by condensation at 160 °C. However, the condensation of naphthalene on the ACF surface would block the pores on the ACF surface, thus decreasing the PAH removal. Further, because of the blocking of the pores on the ACF surface, the micropores inside the ACFs cannot be completely utilized during the adsorption process. Therefore, 280 °C was the optimum temperature for the adsorption of PAHs on ACF-A in this study. Reaction temperature had a greater influence on PAH removal than on SO₂ removal when SO₂ and PAHs coexisted in the incineration flue gas. Liu (2006) previously confirmed that the adsorption of PAHs on ACFs is attributed to condensation and physical adsorption, whereas the adsorption process between PAHs and fly ash is attributed to chemical adsorption.

Effect of type of ACFs on SO₂ and PAH removal

Figure 7 shows the effects of different types of ACFs on the removal efficiencies of SO₂ and PAHs at 200 °C, because in practice the temperature of flue gas entered into the APCD of incinerators is 200 °C. Removal efficiencies of SO₂ and PAHs when ACF-A was used as the adsorbent were similar to the corresponding values when ACF-C was used. When ACF-B was used as the adsorbent, the removal efficiencies of SO₂ and PAHs increased to 54% and 89%, respectively. Removal efficiencies of SO₂ and PAHs were the highest when ACF-B was used, which is in accordance with the order of micropore volume of the ACFs (see Table 2). Table 2 confirms that the three ACFs have similar specific surface area and total pore volume. It can therefore be concluded that the removal of SO₂ and PAHs depends mainly on the micropore volume of the ACFs reported in this study.

Figure 7. Effects of different ACF types on the removal efficiencies of SO₂ and PAHs (reaction condition: SO₂ concentration = 750 ppm, reaction temperature = 200 °C).

Liu (2006) previously confirmed that micropore volume plays an important role in determining the removal of gaseous PAHs when the concentration of PAHs is low and the BET surface area is sufficient for PAH removal. Kisamori and Mochida (1994) previously demonstrated that a certain amount of CO₂ was released by the decomposition of O functional groups at 200 °C, thus reducing the amount of active sites but increasing the mesopore volume and, in turn, increasing the storage capacity of H₂SO₄ on the ACFs. As can be seen in Table 3, the O content of ACF-B was the highest among the three ACFs.

It is possible to simultaneously remove fly ash and PAHs from incinerator flue gas by using ACFs because PAHs can be condensed or adsorbed on fly ash and ACFs are very efficient in removing fly ash. Further, since ACF-B has the tightest knitted structure among the three ACFs, fly ash can be separated effectively by ACF-B. Thus, the removal of the PAHs, which are adsorbed on fly ash, will improve because of the tight structure of ACF-B. Thus, ACF-B, with the highest microporous volume, highest O content, and tightest structure among the three ACFs, was the best adsorbent for removing SO₂ and PAHs when they coexisted in incineration flue gas.

Conclusions

The simultaneous removal of PAHs and SO₂ from incineration flue gas was examined in this study. The effect of SO₂ on the removal of PAHs and the effects of reaction temperature and type of ACFs on removal of SO₂ and PAHs were determined. The results indicate that the removal efficiency of PAHs by using ACF-A was 90% in the absence of SO₂. When the concentration of SO₂ was 350 ppm, the removal efficiencies of SO₂ and PAHs by using ACF-A were 92% and 33%, respectively. The presence of SO₂ in the incineration flue gas can decrease the removal of PAHs drastically because of competitive adsorption onto the ACFs. However, removal of PAHs was insignificant when the concentration of SO₂ increased from 350 ppm to greater than 750 ppm (SO₂ concentration up to 1000 ppm were examined). Moreover, the variation in SO₂ removal was insignificant at an SO₂ concentration of 750 ppm when the temperature increased from 160 to 280 °C, and the removal efficiency of PAHs followed the order 280 °C > 160 °C > 200 °C. ACF-B, with the highest microporous volume, highest O content, and tightest structure among the three studied ACFs, was found to be the best adsorbent for removing SO₂ and PAHs when these gases coexisted in incineration flue gas.

Funding

The authors thank the National Science Council of the Republic of China (Taiwan) for financially supporting this research under contract no. NSC 96-2211-E-131-019-MY3.

Additional information

Notes on contributors

Zhen-Shu Liu

Zhen-Shu Liu is an associate professor, Wen-Kai Li was a postgraduate student, and Ming-Jui Hung is an assistant professor in the Department of Safety, Health and Environmental Engineering at the Ming Chi University of Technology, New Taipei City, Taiwan.

Wen-Kai Li

Zhen-Shu Liu is an associate professor, Wen-Kai Li was a postgraduate student, and Ming-Jui Hung is an assistant professor in the Department of Safety, Health and Environmental Engineering at the Ming Chi University of Technology, New Taipei City, Taiwan.

Ming-Jui Hung

Zhen-Shu Liu is an associate professor, Wen-Kai Li was a postgraduate student, and Ming-Jui Hung is an assistant professor in the Department of Safety, Health and Environmental Engineering at the Ming Chi University of Technology, New Taipei City, Taiwan.