Hg⁰ Removal by V₂O₅ Modified Palygorskite in Simulated Flue Gas at Low Temperature

Abstract

The V₂O₅-modified palygorskite (V₂O₅/PG catalysts) were prepared and used for Hg⁰ removal in simulated flue gas at low temperature. It was found that the V₂O₅/PG catalyst had excellent performance for Hg⁰ removal at 150 °C. O₂ exhibited a positive effect on Hg⁰ removal over V₂O₅/PG, while SO₂ and H₂O showed an inhibiting effect. However, Hg⁰ removal efficiency showed a promotion trend in the presence of H₂O, SO₂, and O₂. The Brunauer–Emmett–Teller (BET) method, scanning electron microscopy (SEM), and X-ray photoelectron spectroscopy (XPS) were applied to characterize the physicochemical properties of the V₂O₅/PG catalyst. Mercury temperature-programmed desorption (Hg-TPD) experiments were also conducted to identify the mercury species adsorbed on the V₂O₅/PG catalyst, and the pathway of Hg⁰ removal over V₂O₅/PG was also discussed. The used V₂O₅/PG catalyst after Hg⁰ removal was regenerated, and its capability for Hg⁰ removal can be completely recovered. The V₂O₅/PG-Re-300 °C catalyst showed excellent performance and good stability for Hg⁰ removal after regeneration.

1. Introduction

Mercury, as a trace toxic chemical, is harmful and threatening to people’s health and the environment [1,2,3]. In China, about 38% of annual mercury emissions are from coal-fired power plants, which are regarded as the main anthropogenic source of mercury emissions [4,5,6]. Therefore, more and more attention has been paid to mercury emission control from coal-fired flue gas. To control mercury release, a series of environmental regulations and laws have been promulgated. The Minamata Convention on Mercury has been in effect since 16 August 2017, which means that reducing atmospheric mercury emissions has been a consensus-driven and compulsive goal of many countries around the world [7]. Moreover, the new Emission Standard of Air Pollutants for Thermal Power Plants in China has been executed, and explicitly stipulates that the emission limitation of mercury and its compounds in coal-fired flue gas is 30 μg/m³ [8]. Therefore, the control situation of mercury in coal-fired flue gas is very serious. The urgent task is to develop and research effective and economical technology for mercury emissions in coal-fired power plants.

Generally, coal-fired flue gas mainly contains several components, including O₂ (3–7%) (v/v), moisture (8–10%) (v/v), SO₂ (100–3000 ppm), etc. [9,10]. Mercury in coal-fired flue gas mainly exists in three forms: oxidized mercury (Hg²⁺), particulate mercury (Hg^p), and elemental mercury (Hg⁰) [11,12]. Among them, wet flue gas desulfurization (WFGD) can effectively remove Hg²⁺ [13,14], and electrostatic precipitators (ESPs) or fabric filters (FFs) can easily capture Hg^p [15]. However, due to its high volatility, stability, and insolubility, Hg⁰ is difficult to remove by existing air pollution control devices (APCDs) [16,17]. Therefore, researchers focus their efforts on controlling the emission of Hg⁰ from flue gas.

Various methods have been studied and used to control Hg⁰ emission, such as activated carbon injection (ACI) [18,19], catalytic oxidation, photochemical oxidation, and adsorption [20,21,22]. Among them, ACI technology is considered as the most effective Hg⁰ control technology [23]. However, the large-scale industrial application of ACI technology is limited due to its high operation cost [24,25,26].

In recent years, the natural mineral has attracted the attention of researchers since the raw materials are abundant in reserves and are cost-effective [27,28]. Palygorskite (noted as PG), as a kind of natural magnesium aluminum silicate clay mineral, has a unique crystal structure, well-developed pores, high specific surface area, high adsorption ability, good cohesiveness and mechanical properties, low price, and can be used as a good catalyst carrier. However, few studies have been reported on Hg⁰ removal by PG from flue gas at low temperature.

It is well known that V₂O₅-based selective catalytic reduction (SCR) catalysts are widely used for NO_x removal in coal-fired flue gas [29,30]. Meanwhile, it was also found that V₂O₅ exhibited high catalytic oxidation activity for Hg⁰ and can effectively oxidize Hg⁰ to Hg²⁺ [31,32,33,34]. However, V₂O₅-based SCR catalysts themselves have limited ability to adsorb the formed Hg²⁺ at high operating temperatures (300–450 °C), and the formed Hg²⁺ must be removed by the downstream WFGD unit [13]. Therefore, it is necessary to develop cost-effective and environmentally friendly technologies for Hg⁰ control in coal-fired flue gas at low temperature.

In this work, palygorskite-supported V₂O₅ catalysts (V₂O₅/PG) were prepared and used to remove Hg⁰ in simulated flue gas at low temperature (120–210 °C). The V₂O₅/PG catalysts were characterized by BET, SEM, XPS, and Hg-TPD. Moreover, the effects of V₂O₅ loading, temperature, flue gas components (SO₂, H₂O, and O₂), space velocity (GHSV), and the reaction pathways were also studied in the presence of SO₂. The regeneration of the used V₂O₅/PG catalyst after Hg⁰ removal and reuse for Hg⁰ removal were also studied.

2. Results and Discussion

2.1. Effect of Reaction Temperature

Figure 1 shows the effect of reaction temperature (120 °C, 150 °C, 180 °C, and 210 °C) on Hg⁰ removal over 5V₂O₅/PG catalyst. It can be seen that the initial E_Hg of 5V₂O₅/PG reached 98% at 120 °C, and that it decreased gradually to 73.8% at 420 min. As the temperature increased from 150 °C to 180 °C and 210 °C, E_Hg decreased from 83.1%, 78.4%, and 52.5% at 420 min, respectively. Generally, adsorption becomes weaker at high temperature while oxidation will be enhanced. The difference of E_Hg at different temperatures may be due to the different effect extent of adsorption and oxidation at different temperatures [35]. The 5V₂O₅/PG catalyst had the highest Hg⁰ removal capability at 150 °C, which was much lower than the operating temperature of the V₂O₅-based SCR catalyst. Meanwhile, the V₂O₅/PG catalyst had much higher adsorption ability than that of V₂O₅-based SCR catalyst. Furthermore, the optimum reaction temperature (150 °C) was very close to the actual exhaust gas temperature of flue gas in a power plant, indicating that the V₂O₅/PG catalyst can be used to remove Hg⁰ from flue gas without an extra heat supply.

2.2. Effect of V₂O₅ Loading

Figure 2 shows the effect of V₂O₅ loading on Hg⁰ removal over V₂O₅/PG at 150 °C. Since it was found that a V₂O₅ loading range of 1% to 5% was favorable in our previous research, V₂O₅ loading (1% to 5%) was selected in this paper [36]. It can be seen that E_Hg of the PG carrier was very low, and that it decreased rapidly from 50% to 8.3% in 300 min. The E_Hg of the V₂O₅/PG was obviously higher than that of the PG, which was mainly due to the contribution of V₂O₅. V₂O₅, as the active site for Hg⁰ oxidation, was of great importance in Hg⁰ removal by V₂O₅/PG. It was found that lattice oxygen of V₂O₅ played a critical role in Hg⁰ oxidation in our previous research, which was consumed in Hg⁰ oxidation and can be subsequently replenished by gas-phase O₂ [35]. With the increase of V₂O₅ loading (1% to 5%), the E_Hg increased from 38.6% to 86.0% at 420 min. Since the 5V₂O₅/PG catalyst showed the best Hg⁰ removal performance, 5V₂O₅/PG was used in the following experiments.

The specific surface areas and other pore parameters of the PG and V₂O₅/PG catalysts are summarized in

Table 1. Properties of the PG and V₂O₅/PG catalysts.

Sample	BET Surface Area (m2/g)	Average Pore Size (nm)	Total Pore Volume (cm3/g)
PG	151.43	10.51	0.32
1V2O5/PG	139.91	10.67	0.30
3V2O5/PG	130.24	10.85	0.26
5V2O5/PG	109.31	11.04	0.21

. It can be seen that the BET surface area and total pore volume of the V₂O₅/PG catalyst are lower than those of the pure PG carrier. With the increase of the V₂O₅ loading, the BET area of the catalyst decreased obviously (from 151.4 m²/g to 109.3 m²/g). This may be due to the pores of the PG carrier being blocked by the active component of V₂O₅ and/or the sloughing of the PG skeleton during the calcination process, which led to the decrease of the BET area of the V₂O₅/PG catalyst. It is well known that a larger BET area can provide more reaction sites for the adsorption of Hg⁰. However, the specific surface area of the V₂O₅/PG catalyst showed little effect on Hg⁰ removal, i.e., the BET area was not the decisive factor affecting Hg⁰ removal, while the oxidation activity of V₂O₅ played a critical role in Hg⁰ removal. This was similar to the results reported in the literature [35].

Figure 3A,B shows the SEM morphology of 5V₂O₅/PG. It can be seen that the V₂O₅/PG catalyst had a porous structure, which was conducive to the distribution of V₂O₅ on the surface of the PG carrier and the adsorption of Hg⁰ on the surface of the V₂O₅/PG catalyst. The Lewis acid–base properties of the PG and V₂O₅/PG catalysts were analyzed and the detailed results are shown in the Supporting Information. The results showed that there were several different acid sites on the V₂O₅/PG catalyst surface, which were beneficial for Hg⁰ adsorption on V₂O₅/PG.

2.3. Effect of Flue Gas Components on Hg⁰ Removal

Figure 4A shows the effects of flue gas components on Hg⁰ removal over 5V₂O₅/PG. It can be seen that the 5V₂O₅/PG catalyst exhibited excellent Hg⁰ removal capability (E_Hg = 90.9%) within 400 min in a N₂ atmosphere. However, after 0.15% SO₂ was introduced, an obviously inhibitive effect was observed, and the E_Hg decreased to 43.5%. A similar negative effect (E_Hg = 53.48%) was also observed by adding 5% H₂O into N₂. These may be due to the competitive adsorption of SO₂ (or H₂O) with Hg⁰ on the surface of 5V₂O₅/PG in a N₂ atmosphere, as well as the reaction of SO₂, H₂O, and V₂O₅ [18,35]. Adding 8% O₂ into the N₂ atmosphere promoted Hg⁰ removal over V₂O₅/PG, and the E_Hg increased to 96.8%, indicating that O₂ played a positive role in Hg⁰ removal. Since no Hg⁰ oxidation by O₂ was measured in the gas phase, the effect of O₂ should be on V₂O₅, i.e., V₂O₅ was reduced to V⁴⁺ in Hg⁰ oxidation and lost its oxidation activity, while O₂ can replenish O to the used V₂O₅ sites and resume its oxidation activity. This was consistent with SO₂ removal and Hg⁰ oxidation over V₂O₅/AC catalysts in our previous research [37], and is similar to Hg⁰ oxidation on metal oxide catalysts [38,39].

Figure 4B shows Hg⁰ removal over 5V₂O₅/PG in a complex atmosphere containing more components. As mentioned above (Figure 4A), E_Hg was 43.5% in N₂ + SO₂, and it decreased to 34.2% after adding 5% H₂O into N₂ + SO₂. However, after 8% O₂ was added into N₂ + SO₂ or N₂ + H₂O, the E_Hg increased to 72.4% and 77.4%, respectively. This suggested that O₂ played a critical role and could offset the negative effect of H₂O (or SO₂) on Hg⁰ removal to a certain extent [18,40,41,42]. The E_Hg increased to 86.9% after 5% H₂O was added into N₂ + O₂ + SO₂, which may be due to the formation of ${\mathrm{SO}}_4^{2-}$, which could react with Hg⁰ [43,44,45]. HgSO₄ was formed, as shown in the following TPD experiment results.

2.4. Effects of GHSV on Hg⁰ Removal

The GHSV, as an important industrial parameter, can directly affect the contact time between the flue gas components and catalyst bed, and it has an important influence on the Hg⁰ removal process. Figure 5 shows the results of Hg⁰ removal by 5V₂O₅/PG at different GHSV (6000 h⁻¹, 10,000 h⁻¹, and 15,000 h⁻¹) at 150 °C. It can be seen that GHSV had an obvious effect on Hg⁰ removal over 5V₂O₅/PG. As the GHSV increased from 6000 h⁻¹ to 15,000 h⁻¹, the E_Hg decreased from 86.9% to 57.9% in 400 min. The lower GHSV (6000 h⁻¹) is more beneficial for Hg⁰ removal over 5V₂O₅/PG. This may be due to the lower GHSV increasing the contact time of Hg⁰ and 5V₂O₅/PG [35]. The 5V₂O₅/PG catalyst exhibited good Hg⁰ removal activity at 6000 h⁻¹, which can match the actual GHSV of flue gas in a power plant and was suitable for mercury removal in power plants.

2.5. The Pathway of Hg⁰ Removal over V₂O₅/PG

The speciation of Hg adsorbed on the surface of V₂O₅/PG was characterized by XPS, and the results are shown in Figure 6. It can be seen that there was only one peak, at 102.90 eV, for the fresh 5V₂O₅/PG, which could be attributed to Si 2p in the PG carrier. As for the spent 5V₂O₅/PG after Hg⁰ removal, two peaks were observed at 103.1 eV and 107.15 eV, respectively. The peak at 103.1 eV can be assigned to Si 2p of the PG carrier, while the peak at 107.15 eV was attributed to Hg²⁺ [24,28].

To further investigate the mercury species adsorbed over V₂O₅/PG, the experiments of Hg temperature-programmed desorption (Hg-TPD) were performed for the fresh and spent 5V₂O₅/PG. As shown in Figure 7, it can be seen that there was no Hg release peak during the whole Hg-TPD process for the fresh 5V₂O₅/PG sample. However, as for the spent 5V₂O₅/PG sample, two Hg release peaks appeared at 240 °C and 495 °C, respectively. This indicated that there were two forms of mercury on the surface of the spent 5V₂O₅/PG, which can be ascribed to HgO and HgSO₄ [46,47]; the relevant reactions may be described as follows, and the enthalpy change of adsorption and reactions were calculated in the Vienna ab initio simulation package (VASP 5.4.4) [48]:

Hg⁰ (gas) → Hg⁰ (ad)      (△H = −18.337 kJ/mol)

O₂ (gas) → O₂ (ad)       (△H = −12.589 kJ/mol)

SO₂ (gas) → SO₂ (ad)      (△H = −33.545 kJ/mol)

2Hg⁰ (ad) + O₂ (ad) → 2HgO (ad)    (△H = 147.448 kJ/mol)

SO₂ (ad) + O₂ (ad) + Hg⁰ (ad) → HgSO₄ (ad) (△H = −161.557 kJ/mol)

Additionally, it can be seen that the Hg adsorbed on 5V₂O₅/PG began to release at 150 °C, which could well explain the result of Figure 2 that the E_Hg decreased as the temperature was higher than 150 °C.

2.6. Regeneration of V₂O₅/PG Catalyst after Hg⁰ Removal

The above results showed that the 5V₂O₅/PG catalyst had an excellent Hg⁰ removal capability at low temperature. To investigate the reusability of the 5V₂O₅/PG catalyst, the 5V₂O₅/PG after Hg⁰ removal was regenerated and reused for Hg⁰ removal again. Since the V₂O₅/PG catalyst was prepared by calcining at 300 °C in air, a regeneration temperature was chosen from 300 °C to 500 °C to resume its catalytic activity. The results of the PG, 5V₂O₅/PG, 5V₂O₅/PG-Re-300 °C, 5V₂O₅/PG-Re-400 °C, and 5V₂O₅/PG-Re-500 °C for Hg⁰ removal are shown in Figure 8. It can be seen that, with the rise of regeneration temperature (from 300 °C to 500 °C), the E_Hg of 5V₂O₅/PG-Re-x decreased obviously. 5V₂O₅/PG-Re-300 °C showed the highest activity (E_Hg = 90.3%) at 420 min, even higher than that of the fresh 5V₂O₅/PG catalyst (E_Hg = 83.1%). The E_Hg of 5V₂O₅/PG-Re-400 °C and 5V₂O₅/PG-Re-500 °C were 66.7% and 46.8%, respectively. This may be due to the change of structure and chemical properties of 5V₂O₅/PG during regeneration, which caused partial active sites loss on the surface of the 5V₂O₅/PG catalyst at higher regeneration temperatures (400 °C and 500 °C) [36].

3. Materials and Methods

3.1. Catalysts Preparation

The PG was mixed with distilled water in a certain proportion (1 g: 4–8 mL), dried at 110 °C for 6 h, then calcined at 300 °C for 6 h in air. The obtained PG samples were crushed and screened into a 40–60 mesh.

The xV₂O₅/PG catalysts (x is the mass fraction of V₂O₅ in the V₂O₅/PG catalyst, and x = 0–5 wt%) were prepared by impregnating PG particles with different V₂O₅ loadings, which was similar to the V₂O₅/AC catalysts in our previous research [36]., Briefly, the PG particles are impregnated in ammonium metavanadate solution with the required concentration for 1 h, dried at 60 °C for 5 h, then 110 °C for 8 h. Finally, the samples were calcined at 300 °C for 4 h in air.

3.2. Catalytic Activity Evaluation of V₂O₅/PG

The catalytic activity of the V₂O₅/PG catalyst for Hg⁰ removal was tested in a bench-scale fixed-bed reactor; a detailed description of the experimental schematic is in Figure 9. The Hg⁰ removal activity evaluation device system mainly includes simulated flue gas, a programmed temperature controller, a fixed-bed reactor, an online mercury analyzer and a tail gas treatment cleaner. A total of 0.5 g V₂O₅/PG was loaded into the quartz tube. The simulated flue gas includes 8% O₂, 0.15% SO₂, 5% H₂O and balance N₂, and was controlled by a mass flowmeter. The GHSV was approximately 6000 h⁻¹. The temperature (120–210 °C) was controlled by a digital temperature controller. The Hg⁰ vapor was produced from a mercury permeation tube (VICI Metronics) and carried out by N₂ with a constant flow rate. Hg⁰ concentration was detected by an online RA-915M Mercury Analyzer (Lumex Co, Ltd., St. Petersburg, Russia). Finally, the tail gas was treated by a device equipped with activated carbon.

The Hg⁰ removal efficiency (E_Hg) was defined as follows:

$${\mathrm{E}}_{\mathrm{Hg}}(\%)=\frac{{\mathrm{C}}_0-{ \mathrm{C}}_1}{{\mathrm{C}}_0}\times 100 \%$$

(1)

where C₀ and C₁ represent the Hg⁰ concentration (μg/m³) at the inlet and outlet of the reactor, respectively.

3.3. Characterization

The N₂ adsorption–desorption tests were carried out by an Autosorb-iQ analyzer (Quantachrome, Boynton Beach, FL, USA). The specific surface areas were calculated by the Brunauer–Emmett–Teller (BET) method, the pore structure parameters were analyzed by the Barrett–Joyner–Halenda (BJH) method.

The morphologies of the V₂O₅/PG samples were observed, which was performed on a scanning electron microscope (SEM) (JSM-7001F, JEOL, Akishima City, Japan).

The X-ray photoelectron spectroscopy (XPS) analysis was performed on an ESCALAB 250Xi spectrometer (Thermo Fisher, Waltham, MA, USA) using an Al Kα X-ray source at room temperature. All binding energies (BE) were adjusted with the C 1s binding energy value of 284.6 eV.

The Hg temperature-programmed desorption (Hg-TPD) experiments were conducted in a quartz tube reactor using a 0.1 g sample in N₂ (200 mL/min). The sample was firstly used to remove Hg⁰ at 150 °C for 6 h, then swept with N₂ at 150 °C for 2 h, and finally, heated from 30 °C to 600 °C with a heating rate of 5 °C/min. The outlet gas from the reactor was introduced into a KBH₄ solution to reduce the possibly existing Hg²⁺ to Hg⁰. The Hg⁰ concentration in the effluent gas after the KBH₄ solution was continuously measured by an on-line mercury analyzer (RA-915M, Lumex, St. Petersburg, Russia).

4. Conclusions

The V₂O₅/PG catalyst had excellent Hg⁰ removal capability, which was mainly due to the V₂O₅/PG catalyst combined with the adsorption ability of PG and the catalytic oxidation activity of V₂O₅. Hg⁰ was oxidized to form HgO and HgSO₄, and then adsorbed on the V₂O₅/PG catalyst. The E_Hg of the V₂O₅/PG catalyst increased with the increase of V₂O₅ loading, and the 5V₂O₅/PG catalyst showed the highest E_Hg at 150 °C. O₂ exhibited a promoting effect on Hg⁰ removal, while SO₂ and H₂O showed an obvious inhibitory effect. However, when O₂, H₂O, and SO₂ were added together, the E_Hg showed a promoting trend. The used V₂O₅/PG catalyst after Hg⁰ removal can be regenerated and its capability for Hg⁰ removal can be completely recovered, and the V₂O₅/PG-Re-300 °C catalyst showed excellent performance and good stability for Hg⁰ removal after regeneration.