Environ. Sci. Technol. 2002, 36, 1086-1092
Adsorption Characteristics of SO2 on Activated Carbon Prepared from Coconut Shell with Potassium Hydroxide Activation YOUNG-WHAN LEE* AND JIN-WON PARK Department of Chemical Engineering, Yonsei University, Sinchon-dong, Seodaemun-gu, Seoul 120-749, Korea JAE-HOON CHOUNG AND DAE-KI CHOI Environment & Process Technology Division, Korea Institute of Science & Technology, P.O. Box 131, Cheongryang, Seoul 130-650, Korea
The adsorption characteristics of SO2 were studied with KOH-impregnated granular activated carbon (K-IAC). To confirm selective SO2 adsorptivity of K-IAC using a fixed bed adsorption column, experiments were conducted on the effects of KOH and of linear velocity, temperature, and concentration. In addition, changes in features before and after adsorption were observed by utilizing FTIR, XRD, ToF-SIMS, and AES/SAM, examining the surface chemistry. K-IAC adsorbed 13.2 times more SO2 than did general activated carbon (GAC). The amount of SO2 adsorbed increased as linear velocity and concentration increased and as temperature decreased. At lower temperature, the dominant reaction between KOH and SO2 produces K2SO3 and H2O. Any H2O remaining on the surface is converted into H2SO4 as SO2 and O2 are introduced. Then, the KOH and SO2 reaction produces K2SO4 and H2O. The surface characterization results proved that adsorption occurred through chemical reaction between KOH and SO2. The SO2 adsorbed K-IAC exists in the form of stable oxide crystal, K2SO3 and K2SO4, due to potassium. The basic feature given to the surface of activated carbon by KOH impregnation was confirmed to be acting as the main factor in enhancing SO2 adsorptivity.
Introduction
the bypass line of the existing process, consumes less energy, and produces no secondary pollutants. Activated carbon used mainly as an adsorbent to be packed on adsorber is widely used because of its high affinity for adsorption of many lowconcentration harmful gases (below 0.3-0.25%) due to its large surface area, intricate pore structure, and hydrophobic features (7-10). The raw materials for activated carbon include woods such as wood chip and coconut shell and coals such as lignite and bituminous (11, 12). Wooden materials are known to make activated carbon with excellent adsorptivity. The activated carbon used in this study was from coconut shell. The ability of activated carbon to remove sulfur oxides has been known from some time, and a great deal of research on it continues. Summarizing previous study results, activated carbon is adsorbed in the form of SO3 in an SO2-O2 atmosphere and in the form of oxidized H2SO4 in an SO2-O2-H2O atmosphere. It is reported that at that time that if O2 and H2O are added to the reaction, they serve to increase the SO2 adsorption rate of activated carbon (13, 14). Davini reports (15) that basic surface groups present in activated carbon greatly enhance SO2 adsorption. It was suggested that the amount of SO2 adsorbed is affected by the chemical features of the basic group. As a result, studies on removing acid gases, such as SO2, by making the surface of activated carbon into basic feature through impregnation of alkaline hydroxide have recently gained much attention. In this study, KOH was selected as the alkaline hydroxide impregnated on the activated carbon. As an impregnating substance, KOH is known to have excellent selective adsorption capacity of acid gases (16-18). However, there is very little literature on activated carbon produced by such chemical activation, especially with regard to SO2 behavior on impregnated activated carbon and surface chemistry. Guo and Lua (19) emphasized the surface chemistry of adsorbents in determining adsorption capacity using KOH impregnated activated carbon. Other researchers also stressed the importance of surface chemistry in dealing with adsorptivity of impregnated activated carbon (20, 21). The objective of this study on the adsorption characteristics of SO2 to K-IAC was to confirm the selective adsorption of K-IAC for SO2 in a fixed-bed adsorption column. Toward this end, the study investigated the SO2 adsorption behaviors according to the function of impregnation, linear velocity, temperature, and concentration and examined the features of surface chemistry before and after adsorption.
Experimental Section
The major air pollutant emitted in the combustion of fossil fuels including petroleum and coal, both major energy sources, is SO2, a sulfur dioxide. SO2 has been proven to be detrimental to human health (1, 2), and this has encouraged the development of SO2 emission control technologies. Most SO2 is produced when oxygen in the air combines with sulfur species gas emissions. Oxidized sulfur content first becomes SO2 and then further oxidizes into SO3. Both chemicals are called sulfur oxides, symbolized as SOx (SO2 and SO3) (3, 4). The most common means of reducing SO2 emissions is the wet scrubber. The wet scrubber has a high SO2 removal efficiency, but it is costly to install and operate, and the disposal of the pollutant rich wastewater is also problematic (5, 6). On the other hand, the fixed bed adsorber used in this study is simple in design, easily installed by connection to
Adsorbent Preparation. The adsorbent was prepared by impregnating a KOH (Junsei Chemical Co.) solution into GAC obtained from coconut shell (Dongyang Carbon Co.). The GAC was sieved through an 8/16 mesh and treated with N2 flowing for 4 h at 413 K. The incipient wet impregnation process was used for the actual impregnation. The KOH solution was prepared and mixed after GAC was added. All H2O was removed by means of a rotary vacuum evaporator at 368 K for 3 h. The K-IAC was stored in a desiccator to protect it from common airborne moisture and contaminants, which reduce adsorption capacity. Great care was taken in the processes of impregnation, drying, and storage of the adsorbent because they greatly influence adsorption capacity. Through atomic absorption spectroscopy (AAS) analysis, potassium loading of the K-IAC was found to be 9.96 wt %.
* Corresponding author phone: (822)958-5885; fax: (822)958-5809; e-mail:
[email protected].
Apparatus and Method. The fixed-bed adsorption system used in this experiment is shown in Figure 1. The fixed bed
1086
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 5, 2002
10.1021/es010916l CCC: $22.00
2002 American Chemical Society Published on Web 01/22/2002
TABLE 1. Experimental Conditions and the Amounts Adsorbed runb no. 1 2 3 4 5 6 7 8 9
adsorbenta
SO2 concn (ppm)
temp (K)
contact time (s)
linear velocity (cm/s)
adsorbed amt (SO2 mmol/ g-adsorbent)
GAC K-IAC K-IAC K-IAC K-IAC K-IAC K-IAC K-IAC K-IAC
163 163 413 1007 164 162 165 163 163
403 403 403 403 303 353 473 403 403
0.068 0.068 0.068 0.068 0.068 0.068 0.068 0.200 0.100
29.563 29.563 29.563 29.563 29.563 29.563 29.563 10.014 20.003
0.0401 0.5298 0.5964 0.7442 0.7281 0.5627 0.3855 0.2221 0.3618
a 1.067 g of K-IAC consisting of 9.96 wt % potassium was introduced into the column. b The adsorption tests were performed in the presence of 21% O2 for 3 h.
FIGURE 1. Schematic diagram of experimental system.
TABLE 2. Characteristics of GAC and K-IAC adsorption column was a 316 stainless steel tube, 400 mm in length with an inside diameter of 10.9 mm. Inside the columns, steel mesh was placed in the upper and lower extremities of adsorbent to support the samples and minimize channeling. The temperature of the column was maintained with an electric furnace located at its outer wall. Along the system line, temperature was maintained by using a heat band and heat insulating material and was regulated with a PID temperature controller. The temperature was measured by connecting K-type thermocouple (Omega Engineering Inc.) located inside the line and connected to a recorder. The 2% SO2/N2 was shown to be diluted to the desired concentration by a mass flow controller (Brooks Co., Model 5280E). In the fore end of the adsorption column, an inline static mixer was installed. The concentration of SO2 emitted from the bypass line and adsorption column was analyzed by means of a flue gas analyzer (Eurotron Instruments S.p.A., GreenLine Mk II). Daily SO2 analyzer calibrations were performed with N2 (zero value) and SO2 certified by manufacturer analysis as having a concentration near 80% of the full-scale range (span value) of the analyzer. During the adsorption experiment, analysis of the column outlet concentration of SO2 was controlled by a computer software package (Eurotron Instruments S.p.A.). Analytical data were captured every minute and read into a computer, generating 180 analytical values from every of 3 h. For all experiments, 1.067 g of K-IAC (equivalent to 2 cm of bed depth) were used at concentrations of 163-1014 ppm, linear velocity of 10.01429.563 cm/s, and temperatures of 303-473 K. The column was packed with adsorbent and purged with He at 403 K for 10 min. Then, the zero value of the SO2 analyzer was checked before proceeding with the experiment. After each experiment was completed, all SO2 remaining in line was removed by purging with He at 473 K for 30 min. Sample Characterization. The specific surface area and pore volume of GAC and K-IAC were determined by an automatic volumetric sorption analyzer (Quantachrome, Autosorb 1) using N2 adsorption at 77 K. Prior to the measurements, the sample was outgassed at 473 K under an N2 flow for at least 3 h. The BET surface area of the samples was calculated from the N2 adsorption isotherm by assuming the area of a N2 molecule to be 0.162 nm2. The t-plot method was applied to calculate the micropore volume. The micropore volume was estimated to be the liquid volume of N2 at a relative pressure of 0.993. Before and after adsorption, a surface chemical analysis was performed for samples by utilizing an FTIR (Brucker IFS 45), XRD (PW-1830, Philips), static type ToF-SIMS (PerkinElmer, PHI Model 700 ToF-SIMS/SALI), and an AES/SAM (Perkin-Elmer, PHI Model 670).
property m-3
bulk density, kg BET surface area, m2 g-1 mesopore area, m2 g-1 micropore volume, cm3 kg-1 average pore diameter, Å
K-IAC [GAC, loss (%)] 570 719 [1160, 38] 15 [42, 64] 232 [347, 33] 11.5
Time-of-Flight Secondary Ion Mass Spectrometry (ToFSIMS) analysis was carried out using a system equipped with a two-stage reflectron-type analyzer. A low dose of pulsed Cs+ primary ion beam, with an impact energy of 10 keV, was employed. The spectrometer was run at an operating pressure of 10-9 mbar. The primary ion beam was directed on a square area of 50 µm × 50 µm. The system was operated in high sensitivity mode with a pulse width of 50 ns and with a beam current of 0.5 nA, resulting in a primary ion dose of approximately 4 × 1011 ions cm-2 analysis-1. SIMS spectra were acquired over a mass range of m/z ) 1-100 in negative modes. The Auger Electron Spectroscopy (AES) survey scan and sputter depth profile were recorded using a primary beam energy Ep ) 10 keV, primary beam current Ip ) 0.0099 µA, and beam diameter ∼0.4 µm. The resolution of the cylindrical mirror analyzer was set to 0.6%. The argon ion beam, with an ion energy of 1.5 keV and a current density of 0.6 µAm2, was produced by a differentially pumped ion gun. The sputter profiles were analyzed using the PC PHI-MATLAB software package. In this analysis, AES was used quantitatively to help determine the chemical state of the potassium (K), oxygen (O), carbon (C), and sulfur (S) on the K-IAC. Morphology analysis of the K-IAC was done by Scanning Auger-electron Microscopy (SAM).
Results and Discussion SO2 Adsorption on K-IAC. The experiment conditions on SO2 adsorption in runs 1-9 are reported in Table 1. The physical properties of the K-IAC and GAC used are shown in Table 2. Because of KOH impregnation, the surface area of K-IAC decreased. The BET surface area of K-IAC was found to be 719 m2/g, 38% less than that of GAC. The micropore volume decreased by 33%. The reason for this was blockage of pores in K-IAC by KOH impregnation. To confirm the effect of K-IAC, the breakthrough curve of SO2 comparing K-IAC and GAC in the presence of oxygen is shown in Figure 2. In a breakthrough experiment on two adsorbents at concentration of 163 ppm, temperature of 403 K, and linear velocity of 29.563 cm/s, GAC progressed to breakthrough at the moment SO2 was introduced, reaching saturation point (C/C0 ) 1) within 30 min. However, K-IAC did not reach a saturation point until after 180 min, recording C/C0 ) 0.91. VOL. 36, NO. 5, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1087
TABLE 3. Integrated Values of ToF-SIMS Peak Intensity (I.P.I.) of Main Ion Species main peak OHKOSOSO2SO3SO4CCH-
m/z
nonadsorbed K-IAC
SO2 adsorbed K-IAC (run 2)
external surface (times)
nonadsorbed K-IAC
SO2 adsorbed K-IAC (run 2)
after 20 Å sputtering (times)
17.0027 55.9588 47.9969 63.9618 79.9567 95.9516 12.0000 13.0078
293703 532 15022 22179 24762 747 26578 94495
97692 297 27595 47342 80545 34545 9856 10242
3.01V 1.79V 1.84v 2.13v 3.25v 46.24v 2.70V 9.23V
187763 1330 16523 14871 8457 1755 41565 34184
95270 1143 82222 87709 63815 16721 8140 5452
1.97V 1.16V 4.98v 5.90v 7.55v 9.53v 5.11V 6.27V
FIGURE 2. Breakthrough curves of SO2 over GAC and K-IAC: (a) GAC (run 1) and (b) K-IAC (run 2). Gas composition: 163 ppm SO2, balance air. T ) 403 K. L.V. ) 29.563 cm/s. Comparing the adsorbed amount, GAC recorded only 0.04 mmol SO2/gGAC, whereas K-IAC showed 0.5298 mmol SO2/ gK-IAC or 13.2 times that of GAC. Such results confirm that K-IAC shows good SO2 selectivity, and it can be determined that KOH provided a strong basic functional group to the GAC, playing an important role in providing adsorption site to allow selective adsorption of SO2. It can, therefore, be concluded that the changes in surface chemistry caused by KOH impregnation are a major factor in reducing SO2 emissions. In their previous reports of investigation results, Davini (15) and Carrasco-Marı´n et al. (22) also suggested that the number of basic suface sites with good SO2 selectivity is important because the amount of SO2 being chemically adsorbed depends on the chemical characteristics of surface basic groups. In addition, they reported consistent results showing the value of pore volume or surface area as having insignificant effect on the actual adsorption capacity of adsorbent. Therefore, it can be suggested that the degree of adsorptivity can largely be affected by the basic hydroxide ions and catalysis of potassium that induce selective SO2 adsorption in KOH, which is an impregnant. Figure 3 shows the results of breakthrough experiments with different amounts of SO2 introduced under isothermal conditions of 403 K, changing linear velocity to 10.014 cm/s (run 8), 20.003 cm/s (run 9), and 29.563 cm/s (run 2). At a linear velocity of 29.563 cm/s, the slope of breakthrough curve is steeper. This occurs even if the SO2 concentration is the same because the higher the linear velocity introduced in the column, the larger the absolute amount of SO2 becomes, reducing the time to breakthrough and reaching the saturation point (C/C0 ) 1) faster. The amounts adsorbed in run 8, run 9, and run 2 were 0.2221, 0.3618, and 0.5298 mmol SO2/gK-IAC, respectively. Figure 4 shows the result of breakthrough curve at temperatures of 303 K (run 5), 353 K (run 6), 403 K (run 2), 1088
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 5, 2002
FIGURE 3. Effect of linear velocity on SO2 breakthrough time of K-IAC: (a) 10.014 cm/s, (b) 20.003 cm/s, and (c) 29.563 cm/s. Gas composition: 163 ppm SO2, balance air. T ) 403 K.
FIGURE 4. Effect of temperature on SO2 breakthrough time of K-IAC: (a) 303 K, (b) 353 K, (c) 403 K, and (d) 473 K. Gas composition: 162165 ppm SO2, balance air. L.V. ) 29.563 cm/s. and 473 K (run 7). Adsorptivity increased as the temperature decreased. The amount adsorbed in run 5, run 6, run 2, and run 7 were 0.7281, 0.5627, 0.5298, 0.3855 mmol of SO2/gKIAC, respectively. Muller (23) reported that physical adsorption of SO2 using GAC is greater at low temperatures through eqs 1 and 2; SO2 produced H2SO3 and H2SO4 on the surface reacted with H2O or H2O and O2. (* refers to the state adsorbed on the surface of activated carbon.)
SO2 + H2O f H2SO3*
(1)
SO2 + H2O + 1/2O2 f H2SO4*
(2)
In addition, Siedlewski (24) concluded that with the presence of O2, SO2 is adsorbed on GAC and oxidizes to SO3 (eqs 3-5).
SO2 f SO2*
(3)
O2 f O2 *
(4)
SO2 + O f SO3 *
(5)
Many reviews containing the works of other researchers state that if O2 and H2O exist in GAC, reactions indeed occur according to eqs 1-5. However, there has been no clear examination on this besides the very few pieces of literature on the reaction mechanism on the surface during adsorption of SO2 using impregnated activated carbon with modified surfaces to provide chemical activation. Since chemisorption is dominant rather than physisorption in K-IAC, higher temperature can be considered conducive to good adsorptivity as it allows SO2 to overcome the activation barrier and raises the collision fraction of sufficient collision energy to react with potassium. However, in this experiment, the adsorptivity decreased as the temperature increased. With regard to the variation in adsorptivity according to temperature, the following reactions can be expected to take place between K-IAC and SO2. At lower temperature the following occurs: 1. H2O produced by the reaction between KOH and SO2 on the surface of K-IAC becomes adsorbed on the surface, reacting with SO2 and O2 to produce H2SO4, and H2SO4 reacts with KOH to form K2SO4, a stable oxide crystal. Because decomposition takes place once the temperature reaches 603 K (24), most of the H2SO4 remains on the surface at a low temperature, and because SO2 is more soluble than H2SO4 at low temperature, and the reaction between KOH and H2SO4 is considered to be superior to the direct reaction between KOH and SO2. 2. The equilibrium yield of SO3 by reaction of SO2 and O2 increases with decreasing temperature (26). Oxidation from SO2 to SO3 will, therefore, be facilitated. 3. Since research reports show (13, 21) physisorption occurs readily on the activated carbon at low temperature while K2SO4 by K-IAC are formed simultaneously through chemisorption forming, physisorption is expected to occur at the same time as chemisorption. Thus, at low temperature, the following reaction mechanism is expected to be dominant. The reaction between KOH and SO2 induces production of K2SO3 and H2O. The H2O that remains on the surface is easily converted into H2SO4 by SO2 and O2 introduced. Then, with the reaction between 2 mol of KOH, K2SO4 is formed, simultaneously producing 2 mol of H2O, circulating to eq 8, creating a repetitive reaction.
2KOH + SO2 f K2SO3* + H2O*
(6)
K2SO3*+ 1/2O2 f K2SO4*
(7)
H2O*+ SO2 + 1/2O2 f H2SO4*
(8)
2KOH + H2SO4* f K2SO4* + 2H2O*
(9)
In comparison, at high temperatures of 373 K to 473 K the following occurs: 1. H2O produced by reaction between KOH and SO2 does not stay on the surface and vaporizes; it is expected that oxidation into K2SO4 by adsorption of SO2 following reaction between KOH and SO2 will take place, which is contrary to low-temperature cases.
FIGURE 5. Effect of inlet gas concentration on SO2 breakthrough time of K-IAC: (a) 163 ppm, (b) 413 ppm, and (c) 1007 ppm. T ) 403 K. L.V. ) 29.563 cm/s. 2. Even if H2O is present, SO2 does not readily become H2SO4 since its solubility for H2O at high temperature is low (25). 3. However, since chemisorption of oxygen readily takes place on the surface of activated carbon at high temperature, oxidation into K2SO4 is facilitated, and the O2 byproduct greatly facilitates SO oxidation (26). As a result, it can be considered that adsorptivity will not show large variences from that at low temperature. The reaction mechanism superior at higher temperature is as follows: At high temperature, there is no source for H2SO4 production. Instead, K2SO3 forms with 2 mol of KOH and 1 mol of SO2 and then again through 2 mol of SO2, taking up the selective adsorption site of K-IAC in the stable K2SO4, emitting SO and O2 at the same time. Equations 6 and 10 and eqs 7 and 11 are reactions that take place both at high and low temperature.
2KOH + SO2 f K2SO3* + H2O
(10)
K2SO3*+ 1/2O2 f K2SO4*
(11)
K2SO3*+ 2SO2 f K2SO4* + 2SO + 1/2O2
(12)
In conclusion, it can be explained that depending on the temperature, two reaction mechanisms occur competitively, bringing about differences in adsorptivity. Figure 5 shows the comparison of SO2 at different concentration levels of 163 ppm (run 2), 413 ppm (run 3), and 1007 ppm (run 4). The breakthrough time decreased as the concentration of SO2 increased. As the concentration of SO2 increased, the driving force of gas-solid mass transfer increased, increasing mass transfer rate of SO2. The amounts adsorbed in run 2, run 3, and run 4 were 0.5298, 0.5964, 0.7442 mmol SO2/gK-IAC, respectively. At high SO2 concentration, the increase in adsorption rate led to increase in the adsorbed amount of SO2 for 180 min. Surface Chemistry Characterization. Fourier Transform Infrared Spectroscopy (FTIR) spectra that measured nonadsorbed K-IAC (a) and K-IAC adsorbed for 3 h with 1007 ppm of inlet SO2 concentration (b, run 4) is shown in Figure 6. During the adsorption of SO2 by K-IAC, a strong band emerged by the OH group that is part of -SO-OH group at 3435 cm-1, 1159 cm-1 of an -SO3- group was created, and 1113 cm-1 of an -O-SO3- group was very strongly formed. In addition, at 984 cm-1, a weak band of an -SO2- group emerged (28). From the FTIR spectrum of nonadsorbed K-IAC, the following were observed: near 3600-2600 cm-1, basic -OH groups caused by KOH; -OH groups formed by VOL. 36, NO. 5, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1089
FIGURE 6. FTIR spectra of the K-IAC in the 4000-400 cm-1 region: (a) nonadsorbed and (b) SO2 adsorbed (C0 ) 1007 ppm, run 4).
FIGURE 7. X-ray diffraction patterns of K-IAC: (a) nonadsorbed, (b) SO2 adsorbed (C0 ) 163 ppm, run 2), (c) SO2 adsorbed (C0 ) 413 ppm, run 3), and (d) SO2 adsorbed (C0 ) 1007 ppm, run 4). moisture adsorption due to exposure in the air; aromatic CdC, and aromatic C-H, which originally existed as the support for activated carbon; 1820-1690 cm-1 of ketone Cd O group; 1630 cm-1 of aromatic CdC group; fingerprint region of CdO, C-O, -OH groups that exist as functional groups of activated carbon under 1600 cm-1; out-of-plane bending of C-H group in benzene derivatives, and other peaks. However, these had low FTIR transmittance on K-IAC itself, making it difficult to distinguish. The weak peaks that make microscopic changes by functional groups produced by SO2 adsorption are believed to have been hidden or overlapped. Figure 7 represents comparison of peaks via X-ray Diffractometer (XRD) between nonadsorbed K-IAC and K-IAC that adsorbed SO2 for 3 h at different levels of concentration of 163 ppm (run 2), 413 ppm (run 3), and 1007 ppm (run 4). As for the nonadsorbed K-IAC of (a), a high-intensity peak of 2θ ) 26.580 was observed, which was confirmed to be a peak created by GAC. This peak was not found on any of SO2 adsorbed K-IAC. This can be explained by formation of new crystals on the external surface by potassium and basic OHion as adsorption of SO2 took place, covering the surface of 1090
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 5, 2002
the original K-IAC. As the concentration of SO2 increased, peak intensities increased significantly at 2θ ) 21.240, 29.700, 30.740, and 43.360, and slightly at 2θ ) 35.800, and 37.040. Since the 2θ of major peaks increased as the concentration of SO2 adsorption became higher, it was possible to identify substances formed on the surface. First, based on the information on the surface functional group obtained from FTIR analysis, possible substances were investigated using software from an international center for diffraction data (JCPDS, PCPDFWIN v.200), and with the reagents of suspected substances, XRD was measured. As a result, it was possible to predict that the substances newly formed on the surface are consistent with the 2θ of the main peak of K2SO4. However, the activated carbon itself showed very low sensitivity on FTIR and XRD. There was also the need to collect firm data for the predicted product and specific details regarding diffusion onto the external surface and into the interior and the role of selective adsorption sites. For these reasons, ToF-SIMS and AES/SEM were also used. ToF-SIMS is capable of analyzing remaining substances on the surface in ppb. It is extremely sensitive and can detect molecules. In analyzing SO2 in ppm adsorbed on K-IAC, ToFSIMS can provide a large amount of information on samples due to its extreme sensitivity and ability to simultaneously analyze organic and inorganic substances (29-31). During the analyses of SO2 adsorbed K-IAC, tendencies were interpreted by comparison with nonadsorbed K-IAC, strengthening the credibility of the data. The result gained by the polarity of negative ions detected in the analysis is represented within 1-100 m/z where information can be obtained on major peaks of interaction between KOH and SO2. Table 2 show the results of comparisons among integrated values of ToF-SIMS ion species peak intensity (I.P.I.) of nonadsorbed K-IAC and K-IAC adsorbed for 3 h with an inlet SO2 concentration of 163 ppm (run 2). Comparison for major ion species was made with nonadsorbed K-IAC, and SO2 adsorbed K-IAC; it was compared with K-IAC that is sputtered 20 Å based on SiO2 (32). Differences were indeed found as a result of comparing major peaks from quantitative analysis of the SIMS spectrum with the I.P.I. The ions of SO2 adsorbed K-IAC with notable increase of counts compared to nonadsorbed K-IAC were observed to be SO4-, SO3-, SO2-, and SO-, which increased 46.4, 3.25, 2.13, and 1.84 times, respectively, after SO2 adsorption. The increase in the order of SO4- . SO3- f SO2- f SO- occurred because SO2 makes the fastest contact with the exterior surface of K-IAC, producing predominantly oxide crystals of SO4- species. It can be considered that as a consequence, SO4- showed dominant distribution. In other words, it can be seen, before and after adsorption, that SO4- which was nearly nonexistent before adsorption was formed in high intensity after adsorption. After 20 Å sputtering, SO2 adsorbed K-IAC showed 9.53, 7.55, 5.90, and 4.98 times of SO4-, SO3-, SO2-, and SO-, respectively, compared to nonadsorbed K-IAC. The I.P.I. of SO4- distributed on K-IAC after 20 Å sputtering showed decrease by 5 times compared to SO2 adsorbed K-IAC before sputtering. On the other hand, the amounts of SO3-, SO2-, and SO- showed rather increasing trends, which indicates that most had not reached the termination reaction inside the external surface; i.e., with K2SO4 blocking the pores of K-IAC, surface reaction no longer progressed inside the pores. Therefore, SO4- showed a decreasing trend, while nonreacted SO, SO2 molecules, and K2SO3 in pores were present, resulting in a continuously increasing intensity of SO-, SO2-, and SO3-. OH- and KO- decreased after SO2 adsorption. This proves that KOH provides selective adsorption sites and elicits chemical reaction. OH- provided selective adsorption sites and K+ acted as a catalyst that adsorbed SO2 on the surface as a stable oxide crystal. For nonadsorbed K-IAC, C- and CH- were observed by carbon layer. They are major groups
FIGURE 9. AES montage display obtained from (a) nonadsorbed K-IAC and (b) SO2 adsorbed K-IAC (C0 ) 1007 ppm SO2/air, run 4).
FIGURE 8. AES survey scan obtained from K-IAC adsorbed for 3 h with an inlet SO2 concentration of 1007 ppm: (a) before sputtering and (b) after 60 min sputtering (sputter rate based on SiO2 ) 160 Å/min). in activated carbon (33) and showed a relatively large decrease after SO2 adsorption. This can be interpreted as a phenomenon occurring due to oxide crystals covering the carbon layer. This was investigated in the result of analysis by AES/ SAM (Figure 8). The conclusion of ToF-SIMS showed clearly that SO4- is dominantly distributed due to formation of K2SO4 on the external surface, supporting findings by FTIR and XRD. Figure 8 represents the AES survey scan of K-IAC adsorbed for 3 h with an inlet SO2 concentration of 1007 ppm of run 4. Figure 8(a) is the result of immediate analysis on SO2 adsorbed K-IAC from run 4, and Figure 8(b) is the result from measuring K-IAC that had sputtered 60 min on the same position of the surface as Figure 8(a). Observing the surface enlarged 2500 times using SAM, parts (position 1, position 2) with two major differences were analyzed. The areas where the oxide crystals are created are seen as white in the SAM image because they are charged by the instrument itself. The white part represents formation of oxide crystals following SO2 adsorption. In the SAM image, position 1 of Figure 8(a) clearly shows oxide crystals, where charging takes place strongly, and position 2 of Figure 8(a) is the part where charging is not seen with the naked eye, assuming that oxide crystals of white color are covered very thinly on K-IAC. As a result, it was found that there is a close relationship among the intensities of K, S, and O by analyzing AES survey scan. It was confirmed that the chemical reaction among K, S, and O is taking place on the surface, and accordingly, it further indicated the formation of oxide crystals such as KxSOy (x ) 2, y ) 3 or 4), as predicted by results from surface analysis. The intensities of K, S, and O were all shown to be higher in
position 1 of Figure 8(a) where there seems to be more oxide crystals than those in position 2 of Figure 8(a). In addition, after sputtering of 60 min (Figure 8(b)), the oxide crystals became very pale in the SAM image, showing notable decrease in peak intensity of K, S, and O to Figure 8 (a). On the other hand, C showed no decrease, presenting a larger position 2 than position 1 in Figure 8(a), but showing similar intensity of position 2 and position 1 in Figure 8(b). It can be stated as the phenomenon appeared from sputtering of the oxide crystal that was covered on the surface of activated carbon before sputtering. Such facts indicate that C did not take part in the reaction during the oxide crystal formation process. These results confirm that adsorption took place through chemical reaction between KOH and SO2. These findings also suggest that, due to SO2 adsorption, the intensity for each element of the survey scan contributed to the degree of formation of oxide crystal made of KxSOy forms. Figure 9(b) shows the result illustrated in Auger montage display with the results from observation of the depth profile from 60 min sputtering with K-IAC from Figure 8(a). Sputtering is a result obtained from 80 cycles for 60 min on SO2 adsorbed K-IAC (run 4). As the inlet SO2 concentration increased, there was little change in intensity of C against sputter time, which is the major component of activated carbon, while the intensity of K, S, and O showed gradual decreases. The decreasing of peak intensity in K, S, and O can be considered the result of the blockage of adsorption sites in the internal layer due to the formation of KxSOy crystal and also as a function of the depth of SO2 diffused. On the other hand, Figure 9(a) shows that the nonadsorbed K-IAC, C, K, and O did not decrease as they did after adsorption; 20 cycles were gained for 10 min with 0.5 min intervals. It could be understood that, due to SO2 adsorption, K, S, and O interact with each other and form chemical bonds on the surface of K-IAC. This, furthermore, suggests a strong interaction between K and SO2. Here, KxSOy crystal can be confirmed; because of potassium, the chemicals created on the surface during adsorption are ionic compounds such as K2SO3 and K2SO4. This is because potassium is highly reactive and cannot exist as a single element and, instead, exists in chemical compounds. Figure 10 shows the samples that reached saturation by 3 h of adsorption with 1007 ppm SO2/air in atomic % using PHI matlab program through AES montage display and depth profile analysis. In the initial sputter time, C, K, S, and O each VOL. 36, NO. 5, 2002 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
1091
From the study on surface characteristics of K-IAC following adsorption of SO2, it was confirmed that the surface chemistry of K-IAC was greatly affected by strong chemical interaction between KOH and SO2.
Literature Cited
FIGURE 10. AES depth profile after Matlab application for C, K, O, and S obtained from K-IAC adsorbed for 3 h with an inlet SO2 concentration of 1007 ppm considering sensitivity factor. showed the atomic percentages of 69.70, 18.66, 3.65, 7.99, respectively, but after 6 min (sputter depth of 960 Å based on SiO2 sputter rate) sputtering, each showed drastic changes to 84.91, 9.83, 1.89, 3.37, respectively. As explained in the survey scan analysis result, this illustrates that K2SO4 forms a layer of certain thickness on the exterior surface of K-IAC. Up to 18 min (sputter depth of 2880 Å based on SiO2 sputter rate), C showed a continuous increase, while K, S, and O showed decreases. The atomic percentages for C, K, S, and O were 85.90, 9.63, 1.46, 2.63, respectively. The atomic percentages of S and O showed continuous decreases for even longer periods of time, each falling to 1.05 and 2.79, respectively, at 60 min, but C showed only a small change, falling only to 84.69 at 60 min. K by KOH impregnation, regardless of product formation, showed gradual increase after 18 min, as S and O showed relative decreases. Up to 2880 Å, SO2 is relatively well diffused and adsorbed, but, at deeper layers, the degree of blocking caused by production of oxide crystals on the exterior surface increases as reaction time elapses, reducing the amount of SO2 diffusing into the pores, resulting in small amounts of distributed S and O. The results of AES/SAM and AES depth profile analysis can be interpreted in relation with the results from ToF-SIMS; that KOH directly provides selective adsorption sites for SO2 inducing adsorption. Through surface characterization, it was confirmed that the adsorbed SO2 exists as stable ionic crystals, K2SO3 and K2SO4, due to potassium. However, blocking of pores due to clusters of K2SO4 was observed. Eventually, when adsorbing SO2, K-IAC acts as the selective adsorption site with OH- groups of KOH providing basic atmosphere and K+ facilitating the stable adsorption of SO2 by K-IAC.
1092
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 36, NO. 5, 2002
(1) Srinivasan, A.; Grutzeck, M. W. Environ. Sci. Technol. 1999, 33, 1464. (2) Alvarez, E.; Blanco, J.; Knapp, C.; Oivares, J.; Salvador, L. Catal. Today 2000, 59, 417. (3) Forzatti, P.; Nova, I.; Beretta, A. Catal. Today 2000, 56, 431. (4) Lapina, O. B.; Bal’zhinimaev, B. S.; Boghosian, S.; Eriksen, K. M.; Fehrmann, R. Catal. Today 1999, 51, 469. (5) Vladea, R. V.; Hinrichs, N.; Hudgins, R. R.; Suppiah, S., Silveston, P. L. Energy Fuels 1997, 11, 277. (6) Shi, Y.; Littlejohn, D.; Chang, S.-G. Environ. Sci. Technol. 1996, 30, 3371. (7) Aarna, I.; Suuberg, E. Energy Fuels 1999, 13, 1145. (8) Tsuji, K.; Shiraishi, K. I. Fuel 1997, 76, 549. (9) Yun, J. H.; Choi, D, K.; Kim, S. H. Ind. Eng. Chem. Res. 1998a, 37, 1422. (10) Yun, J. H.; Choi, D, K.; Kim, S. H. AICHE J. 1999, 45, 751. (11) Edwards, I. A. S. InIntroduction to Carbon Science, 1st ed.; Marsh, H., Ed.; Butterworth: London, 1989; Chapter 1. (12) Guo, J. A.; Lua, C. Microporous Mesoporous Materials 1999, 32, 111. (13) Lisovskii, A.; Semiat, R.; Aharoni, C. Carbon 1997, 35, 1639. (14) Lizzio, A. A.; DeBarr, J. A. Energy Fuels 1997, 11, 284. (15) Davini, P. Carbon 1990, 28, 565. (16) Lee, Y. W.; Choi, D. K.; Park, J. W. Ind. Eng. Chem. Res. 2001, 40, 3337. (17) Chiang, H. L.; Tsai, J. H.; Tsai, C. L.; Hsu, Y. C. Sep. Sci. Technol. 2000, 35, 903. (18) Barton, S. S.; Dacey, J. R.; Evans, M. J. B. Colloid Polymer Sci. 1982, 260, 762. (19) Guo, J.; Lua, A. C. Sep. Purif. Technol. 2000, 18, 47. (20) Biniak, S.; Szyman ˜ ski, G.; Sieldwski, J.; SÄ wiatkowski, A. Carbon 1997, 35, 1799. (21) Mochida, I.; Korai, Y.; Shirahama, M.; Kawano, S.; Hada, T.; Seo, Y. Yoshikawa, M.; Yasutake, A. Carbon 2000, 38, 227. (22) Carrasco-Marı´n, F.; Utrera-Hidalgo, E.; Rivera-Utrilla, J.; MorenoCatstilla, C. Fuel 1992, 71, 575. (23) Muller, H. Ullmann’s Encyclopedia of Industrial Chemistry; New York, 1994; Vol. A25, p 635. (24) Siedlewski, J. Int. Chem. Eng. 1965, 5, 608. (25) Perry, R. H., Green, D. W., Maloney, J. O. Perry’s Chemical Engineers’ Handbook, 7th ed.; McGraw-Hill: New York, 1997; pp 2-25. (26) Flagan, R. C., Semfeld, J. H. Fundamentals of air pollution engineering, 1st ed.; Prentice Hall: NJ, 1988; p 217. (27) Raymundo-Pin ´ ero, E.; Cazorla-Amoro´s, D.; Salinas-Martinez de Lecea, C.; Linares-Solano, A. Carbon 2000, 38, 335. (28) Shin, S.; Jang, J.; Yoon, S.-H.; Mochida, I. Carbon 1997, 35, 1739. (29) Spevack, P.; Deslandes, Y. Appl. Sur. Sci. 1996, 99, 41. (30) Vickers, P. E.; Castle, J. E.; Watts, J. F. Appl. Sur. Sci. 1990, 150, 244. (31) Lamontagne, B.; Semond, F.; Adnot, A.; Roy, D. Appl. Sur. Sci. 1995, 90, 447. (32) Briggs, D., Brown, A., Vickerman, J. C. Handbook of Static Secondary Ion Mass Spectrometry (SIMS); Wiley: Chichester, 1989. (33) Davies, M. C.; Lynn, R. A.; Hearn, P. J.; Paul, A. J.; Vickerman, J. C.; Watts, J. F. Langmuir 1996, 12, 3866.
Received for review May 1, 2001. Revised manuscript received August 13, 2001. Accepted October 24, 2001. ES010916L
