Roles of Sulfuric Acid in Elemental Mercury Removal by Activated

Jun 8, 2012 - M. Rumayor , N. Fernandez-Miranda , M.A. Lopez-Anton , M. Diaz-Somoano , M.R. Martinez-Tarazona. Fuel Processing Technology 2015 132, ...
0 downloads 0 Views 1MB Size
Download PDF
Article pubs.acs.org/est

Roles of Sulfuric Acid in Elemental Mercury Removal by Activated Carbon and Sulfur-Impregnated Activated Carbon Eric A. Morris, Donald W. Kirk, and Charles Q. Jia* Department of Chemical Engineering & Applied Chemistry, University of Toronto, 200 College St., Toronto, Ontario, Canada M5S 3E5

Kazuki Morita Institute of Industrial Science, University of Tokyo, 4-6-1, Komaba, Meguro-ku, Tokyo, 153-8505, Japan S Supporting Information *

ABSTRACT: This work addresses the discrepancy in the literature regarding the effects of sulfuric acid (H2SO4) on elemental Hg uptake by activated carbon (AC). H2SO4 in AC substantially increased Hg uptake by absorption particularly in the presence of oxygen. Hg uptake increased with acid amount and temperature exceeding 500 mg-Hg/g-AC after 3 days at 200 °C with AC treated with 20% H2SO4. In the absence of other strong oxidizers, oxygen was able to oxidize Hg. Upon oxidation, Hg was more readily soluble in the acid, greatly enhancing its uptake by acid-treated AC. Without O2, S(VI) in H2SO4 was able to oxidize Hg, thus making it soluble in H2SO4. Consequently, the presence of a bulk H2SO4 phase within AC pores resulted in an orders of magnitude increase in Hg uptake capacity. However, the bulk H2SO4 phase lowered the AC pore volume and could block the access to the active surface sites and potentially hinder Hg uptake kinetics. AC treated with SO2 at 700 °C exhibited a much faster rate of Hg uptake attributed to sulfur functional groups enhancing adsorption kinetics. SO2-treated carbon maintained its fast uptake kinetics even after impregnation by 20% H2SO4.

■

INTRODUCTION Coal combustion continues to be the dominant source of anthropogenic mercury (Hg) worldwide,1 and, once in the atmosphere, this Hg is capable of being transported across intercontinental distances.2,3 For reasons such as these, stringent regulations and emissions targets have been implemented by several governments such as the United States,4 Canada,5 and the European Union.6 One of the simplest and most extensively researched Hg control techniques is injection of activated carbon (AC) downstream of the boiler.7 In recent years, however, the consensus in the industry has been that sulfur dioxide (SO2) and sulfur trioxide (SO3), formed as products of sulfur oxidation reactions during coal combustion, seriously hinder attempts to remove Hg through AC injection.7−9 As possible mechanisms for Hg uptake inhibition by SO2 and SO3, Presto et al. have suggested competitive adsorption, scavenging of oxygen and/or halogen surface groups, and formation of sulfuric acid (H2SO4) leading to pore blockage.10,11 Despite these recent findings, an industrial process (Outokumpu Process) has existed since the 1970s for capturing Hg from nonferrous smelter flue gases using concentrated sulfuric acid (H2SO4) at elevated temperatures.12 In addition, a patented technique exists for removing Hg from liquid © 2012 American Chemical Society

hydrocarbons using AC impregnated with H2SO4, hydrochloric acid (HCl), or phosphoric acid.13 Some recent studies have even concluded that direct impregnation of H2SO4 on AC surfaces has a positive impact on Hg capture.14,15 Morris et al.16 compiled a review of research regarding the impact of sulfur on Hg capture with ACs, and found the discrepancy in the available literature could be explained by variability between studies with regards to gas−solid contact time (indicating kinetic limitations to Hg−H2SO4 reaction), or to the amount of H2SO4 on the surface (indicating pore blockage and swamping of Hg-binding sites). This study is intended to improve the current understanding of the role played by H2SO4 in the complex process of Hg adsorption on AC surfaces. This is carried out by investigating the feasibility of Hg reacting with H2SO4 and the impacts of the carbon surface, oxygen (O2), and sulfur functional groups deposited by reaction of carbon with SO2 at high temperature. Conditions used in Hg uptake are highly idealized and intended to provide fundamental insights as opposed to simulating Received: Revised: Accepted: Published: 7905

March 28, 2012 May 31, 2012 June 8, 2012 June 8, 2012 dx.doi.org/10.1021/es301209t | Environ. Sci. Technol. 2012, 46, 7905−7912

Environmental Science & Technology

Article

diluted with deionized water, centrifuged, and the liquid portion decanted. The dilution, centrifuging, and decanting steps were repeated twice. The liquid portions of the liquidphase samples were then added to the appropriate flasks. The solutions were then diluted to 250 mL with deionized water. Aqueous-phase samples not involving carbon were simply diluted to 250 mL. Total Hg in the samples was measured with inductively coupled plasma-atomic emission spectroscopy (ICP-AES) using a PerkinElmer Optima 7300. Pore size distributions were determined using nitrogen adsorption at 77 K with a Quantachrome Autosorb-1-C and a combination of quenched solid density functional theory (QSDFT) and the BJH technique.

practical situations in which several competing factors are at work.

■

EXPERIMENTAL SECTION Experimental procedures are provided in greater detail. Sample Preparation. All samples used in this study were derived from Calgon BPL, a bituminous coal-based activated carbon. BPL was first crushed and sieved to a particle size range of 150−300 μm. To produce SO2-treated samples (SO2−BPL), approximately 10 g of BPL was contacted with 50% SO2 in nitrogen flowing at 200 cm3/min for 20 min at 700 °C. Impregnation of samples with H2SO4 was done in a fashion similar to that of Presto and Granite.10,11 Solutions of H2SO4 in deionized water were prepared at concentrations of 5.4 and 20.1 wt %. Approximately 3 g of BPL or SO2−BPL was placed in a beaker and the desired solution was added to the solid at a ratio of about 1 mL liquid for every 1 g of solid. The resulting slurry was quickly stirred with a glass rod to ensure complete wetting of the particles, and the mixture was placed in an oven at 110 °C overnight for drying. In this way, two samples series were produced: BPL series (BPL, BPL+5%, and BPL+20%) and SO2−BPL series (SO2−BPL, SO2−BPL+5%, and SO2− BPL+20%). Vapor-Phase Experiments. For each of the two sample series, approximately 0.1 g of each sample was weighed into three ceramic crucibles; one for the raw and two for the H2SO4impregnated samples. All three crucibles were placed in a glass desiccator along with another crucible containing a drop of liquid Hg. The desiccator was placed within a muffle furnace set to 100, 150, or 200 °C. These temperatures were chosen to emulate thermal conditions typically observed downstream of coal-fired utility boilers. All experiments were performed in air with the Hg vapor concentration assumed to be at or near to saturation. For the BPL series, samples were removed after 7, 17, 24, 48, and 72 h of mercury contact. For the SO2−BPL series, samples were extracted after 6, 24, 48, and 72 h. Vapor-phase adsorption experiments to be done in the absence of O2 were done by suspending two crucibles (one containing a carbon sample, the other containing a small drop of liquid Hg) within a vertical tube furnace under N2 at 200 °C. The sample was removed after 3 days, and no intermediate lengths of time were studied. Liquid-Phase Experiments. Approximately 0.1 g of liquid Hg was placed in four 20 mL vials. Each vial was filled to capacity with either deionized water, 5.4 wt % H2SO4, 20.1 wt % H2SO4, or reagent grade (∼96 wt %) H2SO4. For oxygen-free experiments, pure N2 was blown across the surface of each vial for 30 s before it was tightly capped. Otherwise, the caps were either left off completely or pure O2 was bubbled into each vial at approximately 10 cm3/min. Samples were held in their respective conditions for 3 days, after which the H2SO4 solution was separated from the Hg by decanting. For experiments investigating the effect of a carbon surface, approximately 0.1 g of BPL was placed in vial prior to the solution. To facilitate its removal, liquid Hg was placed within an open 2 mL vial completely submerged within the solution. For all work using carbons, O2 was bubbled into the solution. Sample Analysis. Carbon samples from the vapor-phase experiments were weighed to approximately 10 mg in 50 mL centrifuge tubes. For the liquid-phase experiments involving carbon, liquid and solid portions were separated by centrifuging. All solid samples were left to digest overnight in aqua regia, then agitated at 45 °C for 3 h. Each sample was then

■

RESULTS AND DISCUSSION Aqueous Phase Hg−H2SO4 Reaction at Room Temperature. Effect of O2 and H2SO4. Figure 1 shows the results

Figure 1. Dissolved Hg concentrations in water and H2SO4 solutions after 3 days for the cases of O2 limited (black), open top (gray), and O2 bubbled (white). Error bars represent analytical standard deviation of the ICP-AES total Hg measurement.

of ICP-AES analysis regarding the effect of O2 on the reaction between Hg and H2SO4. It is immediately apparent that bubbling O2 into the solution greatly increased the amount of mercury detected under all conditions. For the case of pure water, the only conceivable mechanism by which this could occur is through direct oxidation of Hg by O2: Hg +

1 O2 → HgO(s) ΔG°(25°C) = −58.6kJ 2

(1)

At room temperature, reaction 1 is known to occur slowly in the presence of moist air.17 It is unknown if the surrounding aqueous phase accelerated this reaction. It is likely that HgO only existed on the surface of the Hg droplet, at the site of formation; Figure 2 illustrates how this Hg−HgO complex could thereupon be attacked by hydronium ions to form Hg2+ in solution. The solubility of HgO in water at 25 °C is known to be on the order of 50 mg/L,18 therefore the absorption of Hg2+ by the aqueous phase appears to not have been at equilibrium. It can be seen that Hg dissolution in the solution increased with increasing H2SO4 concentration above 5% when O2 was bubbled. The formation of mercuric sulfate (HgSO4) via reaction 2 was considered as a possible explanation for this positive correlation:17 7906

dx.doi.org/10.1021/es301209t | Environ. Sci. Technol. 2012, 46, 7905−7912

Environmental Science & Technology

Article

Figure 2. Diagram illustrating mechanism of Hg absorption by water and H2SO4 (aq) and adsorption on AC surfaces. Oxygen functional groups on AC are shown as O*.

HgO(s) + H 2SO4 ΔG°(25°C) = −37.5kJ → HgSO4 (s) + H 2O

(2)

Reactions 1 and 2 are used to describe the chemistry of the Outokumpu process for the removal of Hg from metallurgical smelter flue gas as a solid precipitate of HgSO4. In practice, however, the process is carried out at 150 to 200 °C.12 It should be pointed out that both solid HgO and HgSO4 may be dissolved in H2SO4 where mercury exists as Hg2+. The dissolution step will likely help push the overall reaction forward by removing the products. Thermodynamic simulations using the HSC Chemistry software package indicate that both reactions are feasible at room temperature, but the fact that no precipitate was observed indicates that the concentration of Hg2+ was below its solubility limit. Thus, the positive dependence on acid concentration seen in Figure 1 correlation observed was most likely due to the enhanced solubility of HgO and HgSO4 in acidic solutions. These results verify that O2 is a key player in the capture of Hg by H2SO4 at low temperatures. The sample series for which the vials were left open to ambient air without bubbling O2 showed insignificant change in mercury uptake from the O2 limited (sealed vial) series. Evidently, passive dissolution of O2 into the solutions was insufficient to promote oxidation of Hg. In other words, Hg dissolution was hindered by slow kinetics. It should be pointed out that bubbling O2 may enhance Hg dissolution by accelerating two processes: oxidation of Hg to HgO through increased O2 concentration and induced mixing from the action of the bubbles and dissolution of HgO through enhanced mixing. Measurements of dissolved oxygen content would help to determine the relative contributions of enhanced mass transfer and O2 supply. Effect of AC Surfaces. The results of adding small amounts of BPL to the H2SO4 solutions are shown in Figure 3 alongside those of the same system with no carbon. In Figure 3, the amount of Hg dissolved includes both Hg in the solution as well as Hg adsorbed on the carbon and is therefore presented as total Hg uptake in mass units. For the case of pure water, it is interesting that adding AC actually decreased the amount of Hg taken up by the system. ACs are well-known for their ability to remove Hg2+ from aqueous solution.19 The reduced Hg dissolution after adding AC was not expected and suggests that the presence of the carbon interfered with reaction 1. One possible mechanism is that the carbon scavenged dissolved O2, as shown in Figure 2, thereby limiting its availability for

Figure 3. Total dissolved Hg (in solution and/or associated with AC) after 3 days in contact with Hg while O2 was bubbled for the cases of no AC added (black) and BPL added (gray). Error bars represent analytical standard deviation of the ICP-AES total Hg measurement.

reaction with Hg. Previous studies have shown that ACs may adsorb up to 60 mg of dissolved O2 per gram of AC (6 mg for a 0.1 g sample),20,21 which is significant considering that the solubility of pure O2 in water at room temperature is approximately 40 mg/L (0.8 mg for a 20 mL sample).18 Future work will need to directly measure dissolved oxygen concentration to verify this conjecture. However, the same experiment done in 5% H2SO4 yielded the opposite result in that BPL had a positive effect on Hg uptake. Assuming the above hypothesis regarding oxygenscavenging by AC is true, then this effect could be due to the acid preventing the adsorption of oxygen to AC, thus making O2 available for reaction 1. However, the lack of data regarding dissolved O2 adsorption to AC under acidic conditions means that this is only speculation at this point. Another possibility is that the presence of acid promoted the dissolution of HgO enough to negate the inhibiting effect observed with AC in pure water. The net increase observed (approximately 25%) due to BPL may be attributed to Hg2+ adsorption by the carbon as illustrated in Figure 2. Because Hg2+ is a product of reaction 2, its removal from the solution should push the reaction forward. An enhancement in Hg uptake was also observed with BPL for the 20% and 96% H2SO4 systems. In 20% H2SO4, the increase in Hg uptake was much greater than that in 5% H2SO4 (approximately 57%) presumably due to the enhanced solubility of Hg2+ in the stronger acid. The greater dissolved concentration of Hg2+ led to more Hg being adsorbed to the 7907

dx.doi.org/10.1021/es301209t | Environ. Sci. Technol. 2012, 46, 7905−7912

Environmental Science & Technology

Article

Figure 4. Hg loading on BPL (A), BPL+5% (B), and BPL+20% (C) at temperatures of 100 °C (◊), 150 °C (■), and 200 °C (Δ). Error bars for 100 and 150 °C represent analytical standard deviation of the ICP-AES total Hg measurement, whereas error bars for 200 °C represent standard deviation of duplicate experiments.

Figure 5. Hg loading at 100 °C (A), 150 °C (B), and 200 °C (C) on BPL (◊), BPL+5% (■), and BPL+20% (Δ). Error bars for 100 and 150 °C represent analytical standard deviation of the ICP-AES total Hg measurement, whereas error bars for 200 °C represent standard deviation of duplicate experiments.

increased from 100 to 200 °C.27 This translates to an increase in equilibrium Hg vapor concentration from 1.3 to 102 mg/L, which is certain to have had a substantial impact on the amount adsorbed. In addition, the diffusivity always increases with temperature. Despite the increased uptake at 200 °C, however, the Hg loading measured for BPL is still quite low compared to that of H2SO4-treated samples. The effects of temperature on Hg loading for BPL+5% and BPL+20% are shown in parts B and C of Figure 4, respectively. An enhancement due to temperature was observed for both series at 150 and 200 °C, which is consistent with an increased rate of reaction. The Outokumpu process (reactions 1 and 2) is typically carried out in this temperature range,12 so Hg, O2, and H2SO4 will certainly react to form HgSO4 under these conditions. However, as previously mentioned, the greatly increased Hg vapor concentration at higher temperatures complicates any attempt to explain the enhanced Hg uptake. The maximum amount of Hg loaded was found for BPL+20% after 72 h at 200 °C. The Hg loading was determined to be over 500 mg/g, or half the total sample weight, which is one of the highest yet reported. Effect of H2SO4 Treatment. A better comparison can be made between samples treated with different concentrations of H2SO4 at a given temperature because Hg vapor concentration and diffusivity are not variables. In part A of Figure 5, the effect of increased loading of H2SO4 is shown for 100 °C. It is readily apparent that acid treatment improved the ability of BPL to adsorb Hg vapor under these conditions, which is attributed to the formation of HgSO4 within the pores of the carbon. When BPL was treated with H2SO4 of higher concentrations, more

AC surface, driving the subsequent dissolution of additional Hg2+. It is interesting to note that the enhancement observed due to BPL was not as great when 96% H2SO4 was used. This might be a result of the high acid concentration, since it is generally accepted that binding of oxidized Hg to AC surfaces occurs at Lewis basic sites.22,23 The overabundance of acidic ions in solution could have resulted in competitive adsorption at these sites, thus reducing adsorbed Hg. Vapor-Phase Hg−H2SO4 Reaction on AC and SO2Treated AC Surfaces at High Temperature. Effect of Temperature. For untreated BPL, it is apparent from part A of Figure 4 that very little Hg was adsorbed at 100 and 150 °C. This is in good agreement with previous fixed-bed reactor results in which virgin activated carbons were found to have poor Hg capacity at temperatures exceeding 100 °C.24−26 Increasing the temperature to 200 °C, however, increased Hg capture by BPL from 2 to 14 mg/g after 72 h. This may be explained by the presence of O2 during the experiment. Oxidation of Hg0 is believed to be a necessary step in its capture,22,23 and O2 should provide this role given enough thermal energy. This observed positive temperature dependence is attributed to a greater rate of Hg supply to the activated carbon due to a higher vapor pressure and a faster mass transport of Hg, although it may also suggest a chemical step that is accelerated at higher temperatures. This chemical step could be the oxidation of Hg in the air and/or on the carbon surface. However, it is not possible to explicitly determine if this effect isdue to enhanced rate of reaction or higher Hg concentration in the gas phase. The vapor pressure of Hg is known to increase from roughly 0.02 to 2 kPa as temperature is 7908

dx.doi.org/10.1021/es301209t | Environ. Sci. Technol. 2012, 46, 7905−7912

Environmental Science & Technology

Article

needed to clarify whether Hg uptake to H2SO4-impregnated sorbents under real flue gas conditions is limited by mass transfer or reaction kinetics. Effect of Oxygen. The liquid-phase experiments shown in Figure 1 confirmed that O2 is a necessary participant in the capture of Hg by H2SO4 at room temperature. In Figure 7, the

H2SO4 should remain in the pores of BPL after drying. This is confirmed in Figure 6, which shows the pore volume and pore

Figure 6. QSDFT+BJH pore size distribution with respect to pore volume. Micropore volume is shown in black, mesopore volume is gray, and macropore volume is white. Percent distributions are given as numerical values. Figure 7. Effect of O2 on Hg loading of BPL (black) and BPL+20% (white) after 72 h at 200 °C. Error bars represent standard deviation of duplicate experiments.

size distributions of BPL, BPL+5%, and BPL+20%. Treatment with H2SO4 reduced the total available pore volume, and this effect increased with increasing acid concentration: soaking in 20% H2SO4 resulted in an 80% decrease in pore volume. This was accompanied by an overall decrease in pore volume for all pore size ranges, particularly the microporous range, which is consistent with the filling of pores with liquid, H2SO4 in this case. A greater Hg loading is therefore anticipated with the BPL treated with H2SO4 of higher concentrations. As shown in part A of Figure 5, however, there was no enhancement seen for BPL+20% relative to BPL+5%, which suggests that the amount of Hg taken up by H2SO4-treated BPL at 100 °C was limited by Hg supply, intraparticle diffusion of Hg vapor, or both. In part B of Figure 5, the same comparison is made at 150 °C. As with 100 °C, Hg uptake was greatly enhanced by treatment with 5% H2SO4, but no further improvement was detected for BPL+20%. Like at 100 °C, the uptake of Hg to H2SO4-treated BPL at 150 °C was limited by Hg availability and/or mass transfer as opposed to the availability of H2SO4 on the carbon. At 200 °C (part C of Figure 5), the greater availability of H2SO4 on BPL+20% becomes apparent. This is attributed to a combination of increased Hg vapor concentration and enhanced Hg diffusivity. The relative contributions of either factor cannot be determined from the available data, but, in any case, the results confirm the role of H2SO4 in determining Hg loading capacity of H2SO4-treated AC. Considering the substantial loss of available pore space seen in Figure 6, the remarkable increase in Hg uptake seen in part C of Figure 5 after acid treatment could only be due to absorption of Hg in H2SO4. These results confirm that, under certain conditions, Hg can react with H2SO4 on a carbon surface and that this can greatly improve Hg uptake capacity. Results obtained by Presto et al.10,11 indicated that Norit Darco Hg, which had been soaked in 95% H2SO4 showed extremely poor Hg removal at 149 °C in a simulated flue gas. That study, however, used a fixed bed reactor with a gas stream flowing at 8 L/min, thereby greatly reducing the gas−solid contact time. The high concentration of H2SO4 used likely led to nearcomplete pore blockage meaning that the vast majority of the acid was inaccessible to Hg vapor. Fixed bed studies using AC treated with different concentrations of H2SO4 are therefore

increase observed for untreated BPL from less than 1 mg/g in the absence of O2 to 14 mg/g in air shows that this is also true at 200 °C for carbon surfaces devoid of H2SO4. This evidence further confirms that, in the absence of other oxidizing species, O2 plays a critical role in mercury sorption to carbon surfaces. As shown in Figure 7, the enhancing effect of O2 on Hg uptake was also observed with H2SO4-treated carbon at elevated temperature. The amount of Hg captured by BPL +20% was increased to over 500 mg/g with O2 from 150 mg/g in the absence of O2. However, 150 mg/g was still quite substantial. One possible explanation is that the system was not truly free of oxygen: activated carbons are known to contain a large amount of oxygen in the form of both surface functional groups and adsorbed O2.28 It is conceivable that this inherent oxygen content was adequate to support reaction 1, however the lack of Hg capture by BPL without O2 does not support this. The other possibility is that the temperature was sufficient to allow H2SO4 to act as an oxidizer in the absence of O2: Hg + 2H 2SO4 ΔG°(200°C) = 15.1kJ → HgSO4 + SO2 + 2H 2O (3)

In reaction 3, S6+ oxidizes Hg to Hg2+ while being reduced to S4+. Thermodynamic simulation of reaction 3 at 200 °C using HSC Chemistry yielded a positive Gibbs free energy change. At temperatures exceeding 260 °C, ΔG° becomes negative. With a large supply of Hg and H2SO4 along with little SO2 and H2O in the system; however, a negative Gibbs free energy change should be feasible making H2SO4 the oxidizing species for Hg at 200 °C. Without further study, this remains a hypothesis. Figure 8 provides an illustration of the proposed mechanism for capture of vapor Hg on an AC surface with or without H2SO4 present. After being physically sorbed to the surface, Hg is oxidized by either physisorbed O2 or acidic carbon−oxygen surface functional groups to give adsorbed Hg2+. The oxidized Hg then binds to Lewis base surface sites shown in Figure 8 as being either carbon−oxygen or carbon−sulfur functional groups. The Hg may remain in this state or be absorbed by 7909

dx.doi.org/10.1021/es301209t | Environ. Sci. Technol. 2012, 46, 7905−7912

Environmental Science & Technology

Article

adsorption kinetics resulting from SO2 treatment. The presence of sulfur groups on the surface may facilitate the binding of Hg to the surface, whereupon it is converted to HgSO4 through the action of the acid, which acts as a reservoir. Inspection of the early stages of reaction for both samples clearly shows a much faster initial rate of uptake for SO2−BPL+5%. The same discussion also applies to the comparison between BPL+20% and SO2−BPL+20% (part C of Figure 9). This is interesting because it indicates that the sulfur functional groups imparted by SO2 treatment are not masked by H2SO4 despite the large amount of acid present. These sulfur groups are thus likely concentrated near the outer surface of the carbon particles rather than within the pores, which become filled with acid. The SO2-treated sample appears to have reached saturation at just over 500 mg/g after only 24 h, much like with SO2−BPL +5%. BPL+20% achieved a similar Hg loading after 72 h, but it cannot be determined if that sample had reached saturation. Further study is clearly needed to determine the exact cause of the enhancement by high-temperature SO2 treatment. This study concludes that the reaction of Hg with H2SO4 on an AC surface is feasible and dramatically improves overall Hg uptake capacity, particularly with the presence of O2. However, in a practical setting such as powder AC injection in a coal-fired utility flue gas where the AC-gas contact time is on the order of seconds, the relatively slow kinetics of Hg absorption may result in H2SO4 being an impediment to effective Hg capture. The reported hindering effect of SO3 and/or H2SO4 on Hg capture during powder AC injection was likely due to slow kinetics. This may be remedied through high-temperature reaction of activated carbon with SO2, which greatly improves the rate of Hg uptake without compromising the exceptional capacity afforded by the presence of H2SO4. Further studies under fixed bed conditions are needed to experimentally validate the improved kinetics brought about by SO2 treatment.

Figure 8. Diagram illustrating mechanism of Hg adsorption on AC and SO2-treated AC with H2SO4. Oxygen and sulfur functional groups on AC are shown as O* and S*, respectively.

H2SO4 present on the surface. In the absence of O2 acting as an oxidizer, Hg is oxidized directly by H2SO4 (this is shown as a dotted line to indicate a less thermodynamically favorable route). Effect of Sulfur Impregnation with SO2. Inspection of part A of Figure 9 shows that SO2 treatment of BPL brought about an enormous improvement in Hg uptake at 200 °C. The literature contains an abundance of studies, which concluded that impregnation of sulfur onto a virgin activated carbon greatly enhances its overall performance in Hg capture,14,29−33 but this is the first time that it has been demonstrated for sulfur impregnation through high-temperature reaction with SO2. The sulfur groups on the surface are not likely to be in the form of elemental sulfur but rather as thermally stable surface complexes such as sulfide, disulfide, sulfoxide, and sulfone.34−37 Using X-ray absorption near-edge structure spectroscopy, Feng et al.38 found no correlation between Hg uptake to different sulfur-impregnated ACs and their contents of sulfoxide and sulfone. However, a reasonable correlation was determined for thiophene, which can be considered an aromatic sulfide. This study seems to confirm the ability of organic sulfur groups to bind Hg, however further work needs to be done to confirm the absence of elemental sulfur. When SO2−BPL is treated with 5% H2SO4, there is still a substantial improvement in Hg uptake over BPL+5%, as indicated by part B of Figure 9. SO2−BPL+5% appeared to have reached is saturation capacity of approximately 425 mg/g much earlier than BPL+5%, which suggests improved

■

ASSOCIATED CONTENT

S Supporting Information *

Apparatus used for SO2 treatment of Calgon BPL, SO2 treatment procedure, vapor-phase Hg adsorption procedure, apparatus used for liquid-phase Hg uptake work, liquid-phase Hg adsorption procedure, and sample analysis. This material is available free of charge via the Internet at http://pubs.acs.org.

Figure 9. Hg loading on BPL (◊) and SO2−BPL (■) with no H2SO4 treatment (A), treatment with 5% H2SO4 (B), and treatment with 20% H2SO4 (C) at 200 °C. Error bars for SO2−BPL, SO2−BPL+5%, and SO2−BPL+20% represent analytical standard deviation of the ICP-AES total Hg measurement, whereas error bars for BPL, BPL+5%, and BPL+20% represent standard deviation of duplicate experiments. 7910

dx.doi.org/10.1021/es301209t | Environ. Sci. Technol. 2012, 46, 7905−7912

Environmental Science & Technology

■

Article

(15) Uddin, Md.A.; Yamada, T.; Ochiai, R.; Sasaoka, E.; Wu, S. Role of SO2 for elemental mercury removal from coal combustion flue gas by activated carbon. Energy Fuels 2008, 22 (4), 2284−2289, http://dx. doi.org/10.1021/ef800134t. (16) Morris, E. A.; Morita, K.; Jia, C. Q. Understanding the effects of sulfur on mercury capture from coal-fired utility flue gases. J. Sulfur Chem. 2010, 31 (5), 457−475, http://dx.doi.org/10.1080/17415993. 2010.493199. (17) Patnaik, P., Ed. Handbook of Inorganic Chemicals; McGraw-Hill: New York, 2003; 559−581. (18) Green, D. W.; Perry, R. H., Eds. Perry’s Chemical Engineer’s Handbook, 8th ed.; McGraw-Hill: Blacklick, OH, 2007; 2−18 and 2− 130. (19) Cai, J. H.; Jia, C. Q. Mercury removal from aqueous solution using coke-derived sulfur-impregnated activated carbons. Ind. Eng. Chem. Res. 2010, 49 (6), 2716−2721, http://dx.doi.org/10.1021/ ie901194r. (20) Prober, R.; Pyeha, J. J.; Kidon, W. E. Interaction of activated carbon with dissolved oxygen. AIChE J. 1975, 21 (6), 1200−1204, http://dx.doi.org/ 10.1002/aic.690210621. (21) Matsis, V. M.; Grigoropoulou, H. P. Kinetics and equilibrium of dissolved oxygen adsorption on activated carbon. Chem. Eng. Sci. 2008, 63 (3), 609−621, http://dx.doi.org/doi:10.1016/j.ces.2007.10.005. (22) Olson, E. S.; Laumb, J. D.; Benson, S. A.; Dunham, G. E.; Sharma, R. K.; Miller, S. J.; Pavlish, J. H. The multiple site model for flue gas-mercury interactions on activated carbons: the basic site. Prep. Pap. Am. Chem. Soc., Div. Fuel Chem. 2003, 48 (1), 30−31, http:// www.anl.gov/PCS/acsfuel/preprint%20archive/Files/48_1_ New%20Orleans__03-03_0368.pdf. (23) Olson, E. S.; Mibeck, B. A.; Benson, S. A.; Laumb, J. D.; Crocker, C. R.; Dunham, G. E.; Sharma, R. K.; Miller, S. J.; Pavlish, J. H. The mechanistic model for flue gas−mercury interactions on activated carbons: The oxidation site. Prep. Pap. Am. Chem. Soc., Div. Fuel. Chem. 2004, 49 (1), 279−280, http://www.anl.gov/PCS/ acsfuel/preprint%20archive/Files/49_1_Anaheim_03-04_0822.pdf. (24) Sinha, R. K.; Walker, P. L., Jr. Removal of mercury by sulfurized carbons. Carbon 1972, 10 (6), 754−756, http://dx.doi.org/10.1016/ 0008-6223(72)90085-1. (25) Krishnan, S. V.; Gullett, B. K.; Jozewicz, W. Sorption of elemental mercury by activated carbons. Environ. Sci. Technol. 1994, 28 (8), 1506−1512, http://dx.doi.org/10.1021/es00057a020. (26) Miller, S. J.; Dunham, G. E.; Olson, E. S.; Brown, T. D. Fuel Process. Technol. 2000, 65−66, 343−363, http://dx.doi.org/10.1016/ S0378-3820(99)00103-4. (27) Huber, M. L.; Laesecke, A.; Friend, D. G. The Vapor Pressure of Mercury (NISTIR 6643). National Institute of Standards and Technology: Boulder, CO, 2006; http://www.nist.gov/mml/ properties/upload/NISTIR-6643.pdf. (28) Boehm, H.-P.; Diehl, E.; Heck, W.; Sappok, R. Surface oxides of carbon. Angew. Chem. Int. Ed. 1964, 3 (10), 669−677, http://dx.doi. org/10.1002/anie.196406691. (29) Otani, Y.; Emi, H.; Kanaoka, C.; Uchijima, I.; Nishino, H. Removal of mercury vapor from air with sulfur-impregnated adsorbents. Environ. Sci. Technol. 1988, 22 (6), 708−711, http://dx. doi.org/10.1021/es00171a015. (30) Korpiel, J. A.; Vidic, R. D. Effect of sulfur impregnation method on activated carbon uptake of gas-phase mercury. Environ. Sci. Technol. 1997, 31 (8), 2319−2325, http://dx.doi.org/10.1021/es9609260. (31) Liu, W.; Vidic, R. D.; Brown, T. D. Optimization of sulfur impregnation protocol for fixed-bed application of activated carbonbased sorbents for gas-phase mercury removal. Environ. Sci. Technol. 1998, 32 (4), 531−538, http://dx.doi.org/10.1021/es970630+. (32) Hsi, H.-C.; Rood, M. J.; Rostam-Abadi, M.; Chen, S.; Chang, R. Effects of sulfur impregnation temperature on the properties and mercury adsorption capacities of activated carbon fibers (ACFs). Environ. Sci. Technol. 2001, 35 (13), 2785−2791, http://dx.doi.org/10. 1021/es001794k. (33) Feng, W.; Borguet, E.; Vidic, R. D. Sulfurization of carbon surface for vapor phase mercury removal − I: Effect of temperature

AUTHOR INFORMATION

Corresponding Author

*Phone: +1 416 946 3097, fax: +1 416 978 8605, e-mail: cq. [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS For financial support, the authors wish to thank NSERC and the Consortium of Sustainable Materials (COSM), a working partnership between the University of Tokyo and the University of Toronto.

■

REFERENCES

(1) Pacyna, E. G.; Pacyna, J. M.; Sundseth, K.; Munthe, J.; Kindbom, K.; Wilson, S.; Steenhuisen, F.; Maxson, P. Global emission of mercury to the atmosphere from anthropogenic sources in 2005 and projections to 2020. Atmos. Environ. 2010, 44 (20), 2487−2499, http://dx.doi.org/10.1016/j.atmosenv.2009.06.009. (2) Seigneur, C.; Vijayaraghavan, K.; Lohman, K.; Karamchandani, P.; Scott, C. Global Source Attribution for Mercury Deposition in the United States. Environ. Sci. Technol. 2004, 38 (2), 555−569, http://dx. doi.org/10.1021/es034109t. (3) Travnikov, O. Contribution of the intercontinental atmospheric transport to mercury pollution in the Northern Hemisphere. Atmos. Environ. 2005, 39 (39), 7541−7548, http://dx.doi.org/10.1016/j. atmosenv.2005.07.066. (4) Cooney, C. M. Mercury rule lets hazardous air pollutants off the hook. Environ. Sci. Technol. 2005, 39 (11), 232A−233A, http://dx.doi. org/10.1021/es053283. (5) Valupadas, P. Alberta mercury regulation for coal-fired power plants. Fuel Process. Technol. 2009, 90 (11), 1339−1342, http://dx.doi. org/10.1016/j.fuproc.2009.07.006. (6) Díaz-Somoano, M.; Unterberger, S.; Hein, K. R. G. Mercury emission control in coal-fired plants: The role of wet scrubbers. Fuel Process. Technol. 2007, 88 (3), 259−263, http://dx.doi.org/10.1016/j. fuproc.2006.10.003. (7) Sjostrom, S.; Dillon, M.; Donnelly, B.; Bustard, J.; Filippelli, G.; Glesmann, R.; Orscheln, T.; Wahlert, S.; Chang, R.; O’Palko, A. Influence of SO3 on mercury removal with activated carbon: Full-scale results. Fuel Process. Technol. 2009, 90 (11), 1419−1423, http://dx.doi. org/10.1016/j.fuproc.2009.08.019. (8) Jarvis, J.; Meserole, F. SO3 effect on mercury control. Power Eng. 2008, 112 (1), 54−60. (9) Feeley, III, T.J.; Jones, A. P.; Brickett, L. A.; O’Palko, B. A.; Miller, C. E.; Murphy, J. T. An update on DOE’s Phase II and Phase III mercury control technology R&D program. Fuel Process. Technol. 2009, 90 (11), 1388−1391, http://dx.doi.org/10.1016/j.fuproc.2009. 05.012. (10) Presto, A. A. Granite, E. J. Impact of sulfur oxides on mercury capture by activated carbon. Environ. Sci. Technol. 2007, 41 (18), 6579−6584, http://dx.doi.org/10.1021/es0708316. (11) Presto, A. A.; Granite, E. J.; Karash, A. Further investigation of the impact of sulfur oxides on mercury capture by activated carbon. Ind. Eng. Chem. Res. 2007, 46 (24), 8273−8276, http://dx.doi.org/10. 1021/ie071045c. (12) Habashi, F. Metallurgical plants: How mercury pollution is abated. Environ. Sci. Technol. 1978, 12 (13), 1372−1376, http://dx.doi. org/10.1021/es60148a011. (13) Ohtsuka, K. Acid-containing activated carbon for adsorbing mercury from liquid hydrocarbons. U.S. Patent 5,891,324, 1999. (14) Werner, M.; Heschel, W.; Wirling, J. In Enhanced adsorption of elemental mercury by sulfurized activated lignite HOK©. Paper presented at: Twenty-Third Annual International Pittsuburgh Coal Conference. In: Proceedings of the Twenty-Third Annual International Pittsburgh Coal Conference; September 25−28, 2006; Pittsburgh, PA, USA [CD-ROM]. 7911

dx.doi.org/10.1021/es301209t | Environ. Sci. Technol. 2012, 46, 7905−7912

Environmental Science & Technology

Article

and sulfurization protocol. Carbon 2006, 44 (14), 2990−2997, http:// dx.doi.org/10.1016/j.carbon.2006.05.019. (34) Puri, B. R.; Hazra, R. S. Carbon-sulphur surface complexes on charcoal. Carbon 1971, 9 (2), 123−134, http://dx.doi.org/10.1016/ 0008-6223(71)90125-4. (35) Chang, C. H. Preparation and characterization of carbon-sulfur surface compounds. Carbon 1981, 19 (3), 175−186, http://dx.doi. org/10.1016/0008-6223(81)90040-3. (36) Humeres, E.; Peruch, M. G. B.; Moreira, R. F. P. M.; Schreiner, W. Reactive intermediates of the reduction of SO2 on activated carbon. J. Phys. Org. Chem. 2003, 16 (10), 824−830, http://dx.doi.org/10. 1002/poc.673. (37) Morris, E. A.; Jia, C. Q.; Morita, K. Effects of O2 on characteristics of sulfur added to petroleum coke through reaction with SO2. Ind. Eng. Chem. Res. 2010, 49 (24), 12709−12717, http://dx.doi. org/10.1021/ie101388q. (38) Feng, W.; Borguet, E.; Vidic, R. D. Sulfurization of carbon surface for vapor phase mercury removal − II: Sulfur forms and mercury uptake. Carbon 2006, 44 (14), 2998−3004, http://dx.doi. org/10.1016/j.carbon.2006.05.053.

7912

dx.doi.org/10.1021/es301209t | Environ. Sci. Technol. 2012, 46, 7905−7912