Effect of Pressure on NOx Adsorption by Activated Carbons - American

Effect of Pressure on NOx Adsorption by Activated. Carbons. Aurora M. Rubel* and John M. Stencel. Center of Applied Energy Research, University of Ken...
2 downloads 0 Views 194KB Size
704

Energy & Fuels 1996, 10, 704-708

Effect of Pressure on NOx Adsorption by Activated Carbons Aurora M. Rubel* and John M. Stencel Center of Applied Energy Research, University of Kentucky, 3572 Iron Works Pike, Lexington, Kentucky 40511-8433 Received September 26, 1995X

Exposing activated carbons to nitric oxide and oxygen at temperatures between 295 and 400 K leads to the conversion of the NO to NO2 at the carbon surface and the adsorption of NO2. In the current study, the conversion and adsorption kinetics of the NO to NO2 reaction were examined at total pressures between 1 and 28 bar using a commercially available activated carbon. Up to 144 mg of NO2/g of carbon was adsorbed at 28 bar and 343 K, of which 115 mg of NO2/g of carbon was reversibly adsorbed-desorbed during repeated pressurization-depressurization cycles. The amount of NO2 adsorbed at 17 bar and 373 K was similar to the amount adsorbed at 1 bar and 343 K, whereas it was 3-4 times greater than the amount adsorbed at 1 bar and 373 K. Analysis of the adsorption isotherm suggested the mechanism of NO2 uptake was associated with micropore filling with a monolayer of adsorbed NO2 formed at a total gas pressure near 10 bar. Time profiles for the desorption of NO2, CO2, and O2 during pressure release and temperature-induced desorption suggested the influence of critical temperatures and pressures of the adsorbed gases and that van der Waals adsorption forces are important during adsorption and condensation within the pores of the carbon.

Introduction Interest in the abatement of NOx emissions from energy production and transportation sources has increased in recent years as a consequence of knowledge showing its potential acidifying, greenhouse and ozone formation effects.1 Although a tremendous amount of research and development has been and continues to be performed in catalytic NOx reduction,2 combustion NOx concentrations are very dilute, generally less than 0.1%, and catalysts are often poisoned or inhibited by coexisting gases.3 The possibility of using adsorptiondesorption processes for NOx sequestering or decomposition is beginning to be investigated. Such processes could serve as a stand alone method of NOx emission control or as a means to sequester and then concentrate NOx for increasing the efficiency of catalytic reduction processes which are not especially suited to dilute gas streams. One requirement for an efficient adsorption-desorption process is a material that has a high, reversible adsorption capacity. A number of potential sorbents have been examined, including metal oxides,4,5 ionexchanged zeolites,6-8 FeOOOH and Fe2O3 dispersed on and in activated carbon fibers,9-13 and pitch-based Abstract published in Advance ACS Abstracts, April 1, 1996. (1) Boer, F. P.; Hegedus, L. L.; Gouker, T. R.; Zak, K. P. Chemtech 1990, 20, 312. (2) Bosch. H.; Jansen, F. Catal. Today 1988, 2, 369. (3) Ritter, J. A.; Yang, R. T. Ind. Eng. Chem. Res. 1990, 29, 1023. (4) Arai, H.; Machida, M. Catal. Today 1994, 22, 97. (5) Stiles, A. B.; Klein, M. T.; Gauthier, P.; Schwarz, S.; Wang, J. Ind. Eng. Chem. Res. 1994, 33, 2260. (6) Zhang, W.; Yahiro, H.; Mizuno, N.; Izumi, J.; Iwamoto, M. Langmuir 1993, 9, 2337. (7) Zhang, W.; Yahiro, H.; Iwamoto, M. J. Chem. Soc., Faraday Trans. 1 1995, 91, 767. (8) Li, Y.; Armor, J. N. Appl. Catal. 1991, 76, L1. (9) Kaneko, K.; Fukuzaki, N.; Ozeki, S. J. Chem. Phys. 1987, 87, 776. X

0887-0624/96/2510-0704$12.00/0

activated carbon fibers.14 While activated carbons have been investigated as possible NOx reduction catalysts,15-18 studies indicated that they exhibited low NO adsorption capacities and slow NO adsorption rates.19,20 However, recent work has shown that NOx capacities and adsorption kinetics are very good for commercially available activated carbons when exposing them to simulated combustion gas mixtures containing O2, CO2, SO2, and H2O at temperatures between 293 and 393 K.21-25 During adsorption, the carbon catalytically converted NO in the combustion gas to NO2, which was then adsorbed, and subsequently desorbed during temperature-induced desorption.21-25 At atmospheric pressure (10) Kaneko, K.; Kobayashi, A.; Suzuki, T.; Ozeki, S.; Kekei, K.; Kosugi, S.; Kuroda, H. J. Chem. Soc., Faraday Trans. 1 1988, 84, 1975. (11) Kaneko, K.; Ozeki, S.; Inouye, K. Atmos. Environ. 1987, 21, 2053. (12) Kaneko, K.; Funamoto, T.; Ozeki, S. Ind. J. Environ. Protect. 1988, 8, 681. (13) Kaneko, K.; Shindo, N. Carbon 1989, 27(6), 815. (14) Kaneko, K.; Nakahigashi, Y.; Nagata, K. Carbon 1988, 26(3), 327. (15) Illan-Gomez, M. J.; Linares-Solano, A.; Radovic, L. R.; SalinasMartinez de Lecea, C. Energy Fuels 1995, 9, 97. (16) Illan-Gomez, M. J.; Linares-Solano, A.; Radovic, L. R.; SalinasMartinez de Lecea, C. Energy Fuels 1995, 9, 104. (17) Illan-Gomez, M. J.; Linares-Solano, A.; Radovic, L. R.; SalinasMartinez de Lecea, C. Energy Fuels 1995, 9, 112. (18) Ahmed, S. N.; Baldwin, R.; Derbyshire, F.; McEnaney, B.; Stencel, J. Fuel 1993, 72, 287. (19) Teng, H.; Suuberg, E. M. J. Phys. Chem. 1993, 97, 478. (20) Teng, H.; Suuberg, E. M.; Calo, J. M. Proc. 19th Conf. Carbon 1989, 574. (21) Rubel, A. M.; Stencel, J. M.; Ahmed, S. N. Proc. AIChE 1993 Summer Natl. Meet., Seattle, WA 1993, paper no. 77b. (22) Rubel, A. M.; Stewart, M. L.; Stencel, J. M. Prepr.sAm. Chem. Soc., Div. Pet. Chem. 1994, 39(1), 137. (23) Rubel, A. M.; Stewart, M. L.; Stencel, J. M. Prepr. Pap.sDiv. Fuel Chem. 1994, 39(3), 879. (24) Rubel, A. M.; Stewart, M. L.; Stencel, J. M. JMR 1994, 10(3), 562. (25) Stencel, J. M.; Rubel, A. M. In Coal Science and Technology 24; Coal Science, Proceeding of the Eighth International Conference on Coal Science; Elsevier: Amsterdam, 1995; p 1791.

© 1996 American Chemical Society

NOx Adsorption by Activated Carbons

and 343 K, the NO adsorption capacities were determined to be as high as 150 mg of NO2/g of carbon if, and only if, O2 was present in the reactant mixture. Other coreactants such as CO2 and H2O did not inhibit nor were significantly adsorbed while, under the conditions studied, SO2 did not inhibit NO2 adsorption but was coadsorbed at a significantly lower level (20-50 mg/g of carbon).25 Hence, at atmospheric pressure it would be possible to extract NO from combustion flue gas at relatively low temperatures even though it is at near-trace levels in comparison to the other gases which are present. Very little work has been reported for pressurized adsorption-desorption of NO over activated carbons, although we have reported that at atmospheric pressure, NO2 condensed within the micropores of the carbon and, in that state, was not associated with the active site(s) causing the NO + 1/2O2 f NO2 reaction.24 The extent of micropore filling4 in activated carbon fiber has been related to the van der Waals attractive force contstant, a, for different gases.14 The van der Waals a constant for NO2 is nearly 4 times greater than that for NO,26 making NO2 the favored adsorbate. However, increasing temperature has a disordering effect which decreases van der Waals attraction27 and negatively impacts adsorption capacity. This study was initiated to examine the effects of pressure on the adsorption capacity and kinetics of NOx adsorption over an activated carbon with specific attention to the effect of pressure to enhance the amount of NO2 adsorbed and abate the detrimental effect of temperature. Experimental Section A Cahn C1100 high-pressure microbalance and accompanying reactor was used for this study. The balance controls were interfaced to a dedicated personal computer for data acquisition. A cylindrical porous (100 micron) stainless steel bucket, 10 mm o.d., 8 mm i.d., and 20 mm in length, was used as the sample holder. This bucket when loaded was suspended by a platinum wire and hook from the balance beam to a position within the heated portion of the microbalance’s stainless steel high-pressure cell. No corrosion was observed on the sample holder or the high-pressure cell during this study. The highpressure cell was enclosed within a vertical, split tube, furnace which was regulated by a programmable temperature controller. Pressure in the reactor was maintained by a backpressure regulator. Gases flowing into the reactor were controlled by high-pressure mass flow controllers. A VG Micromass quadrapole MS was used to monitor the gases leaving the reactor immediately down stream from the back-pressure regulator. A heated (443 K) fused silica capillary was used to transfer a small aliquot of the gases leaving the pressure cell to an inert Metrasil molecular leak which interfaced the capillary with the enclosed ion source of the MS. The MS has a Nier type enclosed ion source, a triple mass filter, and two detectors (a Faraday cup and a secondary emissions multiplier). The MS was controlled by a dedicated personal computer which was also used to acquire and review scans. The identification of desorbed gases was done by using the major mass ions, 44, 32, and 18, for CO2, O2, and H2O, respectively. The major mass ion for both NO and NO2 is 30. The relative abundance of mass 46 for NO2 gas is approximately 40% but in mixtures of gases this value can (26) CRC Handbook of Chemistry & Physics; CRC Press: Boca Raton, FL, 1974; Vol. 55, p D157. (27) Sienko, M. J.; Plane, R. A. Chemistry; McGraw Hill: New York, 1976; p 138.

Energy & Fuels, Vol. 10, No. 3, 1996 705 change. Therefore, NO and NO2 were identified by comparing peak intensities of the mass ion ratio, 30/46, during desorption and relating these intensities to those obtained using known mixtures of NO or NO2 and all combinations of gases used during our study. This procedure assured proper identification of NO vs NO2. Although cracking of NO2 by MS filaments was a potential source of error in defining the presence of NO versus NO2, our procedure and the data obtained have eliminated this possibility. Additionally, higher 30/46 ratios as a consequence of cracking favor the erroneous identification of NO2 as NO and not the opposite as found in this study. Approximately 0.6-0.7 g of carbon was loaded in the zeroed sample bucket. The weight of the carbon was monitored continuously. Each batch of carbon was subjected to several adsorption and desorption cycles. The flow rate through the reactor was measured at room temperature and pressure and was maintained at 100 mL/m during the entire experiment which involved gas switching between He and the combustion gas mixture. The sample was first preconditioned by heating to 453-473 K in a flow of He to remove any preadsorbed NOx. The carbon was then cooled to the desired adsorption temperature before the first exposure to the simulated combustion flue gas. As the carbon approached equilibrium with adsorbate (g98%) as determined by a weight gain of less than 0.1% in 600 s, the gases flowing through the reactor were again switched to He and reactor pressure was lowered to atmospheric beginning a pressure release desorption step. Weight loss was monitored until no further changes were observed. The weight loss during this period was attributed to reversibly adsorbed species. System pressure was then reestablished with flowing combustion gas mixture and another adsorption cycle was begun. The adsorption variables studied were temperature (343 and 373 K) and pressure (1, 10, 17, and 28 bar). A commercially produced activated carbon was used in this study. The carbon was physically activated using steam and had N2 BET total, mesopore, and micropore surface areas of 460, 20, and 440 m2/g, and volumes of 0.69, 0.45, and 0.24 mL/ g, respectively. The NOx adsorption capacity of this carbon determined by thermal analysis-mass spectrometry was 120 mg of NO2/g of carbon at atmospheric pressure and 343 K when a simulated flue gas, containing 2.0% NO, 5% O2, 15% CO2, 0.4% H2O, and He as the balance, was used. The reactant gas used for most of this work had the same composition. The gases were added simultaneously except during experiments where the carbon was presaturated with CO2/O2 before NO was added to the reactant mixture. Control adsorption experiments were performed to determine the contributions of He and CO2/O2 to the total weight gained by the carbon. The gases and gas mixtures used during these experiments were He alone, NO in He, O2/CO2 in He, and O2/NO in He.

Results Adsorption-desorption profiles were generated for each experiment from the weight loss-gain and temperature data. Figure 1 shows a single adsorptionreversible desorption step at a pressure up to 17 bar (span between points a and b) which followed a preconditioning step (up to point a). The mass uptake observed during adsorption was 200 mg/g of carbon at a temperature of 343 K. After the carbon was nearly saturated, the gas flowing through the reactor was switched to ultrapure He and the pressure was released at point b. The weight loss following pressure release was attributed to a reversibly adsorbed species. Some adsorbed material, approximately 40 mg/g of carbon or 20% of the weight gain, remained on the carbon even after 104 s of purging with He at 343 K. Since the focus

706

Energy & Fuels, Vol. 10, No. 3, 1996

Rubel and Stencel

Figure 1. A single adsorption-reversible desorption profile.

Figure 2. Adsorption isotherm, VNO2/Vµ vs total pressure, P, for 343 K adsorption; VNO2 ) condensed gas volume.

of the present study was reversible adsorption of NOx, this more strongly bound adsorbate, considered as irreversibly bound, was not studied in detail. However, occasional temperature-induced desorption to 473 K performed to regenerate the carbon to its initial state indicated that the temperature of maximum evolution of this irreversibly adsorbed material was 411 K and regeneration resulted in a 1-2% loss of irreversible adsorption capacity. The adsorption capacity associated with the reversible adsorbed species was not affected by repeated adsorption-desorption cycles; as many as 10 cycles were performed using the same sample without loss of capacity. The effect of pressure on the adsorption of combustion flue gas components was studied using the adsorptiondesorption profiles just described. Figure 2 is a plot of the ratio of the volume of condensed NO2 adsorbed/ carbon microporous volume (VNO2/Vµ) versus the total pressure, P, at a temperature of 343 K. The adsorption data is typical of a Type I isotherm,28 suggestive of micropore volume filling, the extent to which is nearly complete at pressures near 10 bar. Figure 3a-c shows the weight loss and intensities of mass spectral peaks as a function of time during pressure release to 1 bar from either 17 or 28 bar and during temperature-induced desorption. For simplicity, only trends for masses 30, 32, and 44 are displayed. The mass ion ratio, 30/46, for the data in Figure 3 was compared to the mass ion ratio obtained from standard mixtures and was consistent with NO2 being the only detectable nitrogen oxide species evolved during pressure release and thermal desorption. A pressure release at 343 K caused the evolution of about 60-70% of the total mass uptake on and within the carbon. (28) Brunauer, S.; Deming, L. S.; Deming, W. E.; Teller, E. J. Am. Chem. Soc. 1940, 62, 1723.

Figure 3. Integrated weight loss mass spectral data for pressure release and thermal-induced desorptions; (a) pressure release from 17 bar; (b) pressure release from 28 bar; (c) thermal-induced desorption.

Increasing the adsorption pressure from 17 to 28 bar caused increased intensity for ion mass 30 (NO2) and decreased or held constant the intensity for ion mass 32 (O2) and ion mass 44 (CO2) during a pressure release desorption. The intensities for ion masses 32 and 44 were greater for the 17 bar experiments than for any other pressure used during adsorption. The data in Figure 3a,b, when compared to the data in Figure 2, indicate that O2 and CO2 may be important components of the reversibly adsorbed species (see below). Subsequent to eliminating the reversibly adsorbed species, the temperature-induced desorption data in Figure 3c show that very little O2 evolved from the carbon, whereas significant amounts of CO2 and NO2 were present. The order in which NO2, O2, and CO2 were desorbed from the carbon was different during pressure release versus temperature-induced desorption. During pressure release, the evolution of O2 and CO2 began immediately and maximized at approximately 115 s before the peak in the NO2 desorption. Subsequent to pressure release and during a thermal desorption cycle, the peak for NO2 evolution occurred 150 s before the CO2 desorption peak. The temperatures of these maxima were 413 and 428 K for NO2 and CO2, respectively. It is likely that, whatever the source of the CO2 in the mass spectral data, the origin of the CO2 detected during pressure release was different than the origin of CO2 during temperature-induced desorption. Because only a trace amount of O2 was detected during temperatureinduced desorption, its evolution time was not quantified. A comparison of desorption spectra from 10, 17, and 28 bar experiments clearly demonstrated that the amount of NO2 released from the carbon increased with

NOx Adsorption by Activated Carbons

Energy & Fuels, Vol. 10, No. 3, 1996 707

Figure 4. Contributions of He, NO, and CO2/O2 to total adsorption.

Figure 6. Effect of temperature and pressure on the relative percentage of reversibly and irreversibly adsorbed NO2 removed from the carbon.

increasing adsorption pressure. In contrast, the amounts of stored O2 and CO2 did not appear to be uniformly affected by pressure. To quantify the relative contribution of NO2, CO2, and O2 to the total mass uptake, control experiments were used to provide the relative amount of weight gain which could be related to each of these adsorbates. For example, data in Figure 4, for experiments performed at 17 bar, confirmed our previous findings at atmospheric pressure;21-25 less than 10 mg of NO/g of carbon was adsorbed if O2 was not present with NO in the reactant gas. This value is approximately 2% of the mass uptake for the case of simultaneous exposure to NO + O2. During CO2/O2 adsorption, the mass uptake can be as great as 78 mg/g of carbon (Figure 4), or about 36% of the total uptake when NO is also present in the reactant stream. In addition, Figure 5 shows that the relative contributions of the reactants to the total mass uptake were sensitive to temperature with less than 10% of the total uptake at 373 K resulting from CO2/O2. The remainder of the mass uptake for the carbons was the result of NO + 1/ O f (NO ) 2 2 2 adsorded. Figure 6 shows the relative percent of reversibly and irreversibly adsorbed NO2 removed from the sample. At 343K and 1 bar, approximately 65% of the NO2 in the carbon was reversibly desorbed by pressure release; this percentage increased to 80% at 17 bar. At 373 K and 1 bar, approximately 70% of the NO2 in the carbon was reversibly desorbed; this percentage increased to greater than 95% at 17 bar. Hence, increased presssure decreased the relative amount of NO2 which was irreversibly adsorbed at each temperature.

Their results established pressure and temperature dependencies of this reaction, pointed to possible surface adsorbed species such as (NO)2, thought to be important in the NO to NO2 reaction, and quantified the merits of using the heterogeneous reactions over activated carbon and silica gel in comparison to homogeneous, gas phase reactions. Recent work21-24,30 has confirmed the importance of gaseous oxygen as a coreactant for converting NO to NO2 over activated carbon within a temperature range of 300-420 K. The upper value of this range is low enough to limit carbon gasification (e1%21) but yet enables rapid adsorption rates and high adsorption capacities. In agreement with the data presented by Teng and Suuberg,19,20 the fitting of the NO + O2 adsorption data displayed in Figures 2-6 using a Elovich type analysis would lead to a negative apparent activation energy, the significance of which is only related to the reversible nature of the adsorption process. The Type I isotherm in Figure 2 suggests micropore filling. Even though the data in Figure 2 is not ideally suited for the use of Langmuir theory which is strictly valid only for monolayer adsorption, plotting P/V versus P, typical of Langmuir analysis, produces two linear regions having different slopes, one which extends from 0 to 10 bar and the other from 10 to 28 bar. The point common to these two linear fits is at a total pressure of 10.4 bar (partial pressure of NO2 ) 0.21 bar), a value at which the volume uptake data in Figure 2 shows VNO2/Vµ ) 1. Hence, it is possible that at total pressures near 10 bar, there is monolayer of NO2 formed within the micropores. The BET adsorption theory is applicable to multilayer adsorption. Using the BET equation in the form of P/[n(P0 - P)] ) 1/(nmC) + (C - 1)/(nmC)(P/P0), where n is the ratio VNO2/Vµ at pressure P, nm is the monolayer capacity ratio VNO2/Vµ, C is a free energy term, and P0 is taken as the critical pressure for NO2 (101 bar), a plot of P/[n(P0 - P)] versus P/P0 can be constructed in which the slope of the line is represented by (C - 1)/ nmC with an intercept of 1/(nmC). This plot, presented in Figure 7, produces the following intercept and slope:

Discussion

(C - 1)/nmC ) 0.95; and 1/(nmC) ) 6.4 × 10-3

The use of activated carbon to catalytically oxidize NO was examined by Rao and Hougen as early as 1952.29

These values imply that when nm ) VNO2/Vµ ) 1.04, a monolayer of NO2 has been formed within the carbon.

Figure 5. Effect of temperature on the relative contribution of CO2/O2 to total uptake.

(29) Rao, M. N.; Hougen, O. A. Chem. Eng. Prog. 1952, 48, 110.

(30) Grzybek, T.; Papp, H. Appl. Catal. B: Environ. 1992, 1, 271.

708

Energy & Fuels, Vol. 10, No. 3, 1996

Rubel and Stencel

Figure 7. Application of the BET adsorption equation to data from Figure 2. Table 1. Critical Temperature (TC), Critical Pressure (PC), and Kinetic Diameter for Gases Used during Adsorption Studies gas

TC (K)

PC (bar)

kinetic dia (nm)

CO2 O2 NO NO2

304 155 180 431

74 51 65 101

0.330 0.346 0.317 0.340

Therefore, viewing the isotherm in Figure 2 by either a Langmuir or BET approach suggests identical conclusions; i.e., the NO2 adsorbed within the carbon is primarily located within the micropores to a monolayer capacity at total pressures of less than 10 bar. As noted in previous work,4 micropore filling is not effective for gases which have critical temperatures less than the temperature used during adsorption. The critical temperatures (TC) and pressures (PC) of CO2, O2, NO, and NO2, presented in Table 1, show that only the critical temperature of NO2 is greater than the temperatures used during the current study. Hence, at atmospheric pressure the amounts of NO, O2, and CO2 adsorbed in the micropores were expected to be small in comparison to the amount of NO2. This behavior was observed in the adsorption data at 1 bar when the temperature was either 343 or 373 K (see Figure 5). However, at 17 bar and 343 K, the amount of CO2 adsorbed was 78 mg/g of carbon, a value representing 36% of the total uptake of the carbon. This CO2 uptake was possibly a consequence of the increased pressure because TC(CO2) is close to the 343 K adsorption temperature. The low value of the critical temperature for NO also indicates that it will not be the species adsorbed in the carbon micropores, in contrast to previous discussion.4,31 Rather, the quantitative mass spectral identification of NO2 during desorption points to the importance of adsorbed NO2 instead of adsorbed NO, and to the probable occurrence of the labile reactions, NO + 1/2O2 f NO2, and/or 4NO f 2(NO)2 f 2NO2 + N2, which have taken place over the carbon. The absence of amu 28 from the mass spectra suggested that the first reaction was the major contributor to the production of NO2 and the second reaction, if it occurred, played a minor role. Because the kinetic diameter of CO2 (see Table 1) is only slightly smaller than the kinetic diameter of NO2, (31) Wang, Z. M.; Shindo, N.; Otake, Y.; Kaneko, K. Carbon 1994, 32, 515.

any enhanced filling of the pores by either of these gases as a result of steric limitations is expected to be minimal. However, the van der Waals force constant a for NO2 is 1.5 times greater than the a constant for CO2. Hence, the attractive force between NO2 molecules upon entering the pores would be greater than for CO2. This difference may account for the desorption of CO2 before NO2 during pressure release. During temperatureinduced desorption, the evolution of CO2 subsequent to NO2 desorption may be related to a small amount of carbon gasification. The total amount of gaseous NO2 (at STP) adsorbed into the activated carbon at a pressure of 28 bar and temperature of 343 K is approximately 70 cm3/g of carbon. This value is significantly larger than the amount of NO reported to occupy the microporosity of cation-exchanged zeolites.2 In addition, the amount of irreversibly adsorbed NO2 in the activated carbon, at 17 bar and 373 K, is less than 5% of the total amount of adsorbed NO2. Such a value is as good as or less than values reported for the cation-exchanged zeolites. In addition, as stated in the Results section, increasing the pressure increased the relative amount of reversibly versus irreversibly adsorbed NO2. This behavior would be beneficial to pressure swing processing and is possibly a consequence of the different sites at which the reversible and irreversible specie are located. For example, reversibly adsorbed NO2 is associated with a condensed phase within the micropores whereas the irreversibly adsorbed NO2 is probably associated with sites that convert NO f NO2. The number of these irreversible NO2 sites is not expected to change significantly with pressure whereas the amount of condensed NO2 within the micropores does increase with increasing pressure. Summary and Conclusions The amount of NO2 reversibly adsorbed on activated carbons increased with pressure, a fact that could potentially be used to mitigate the detrimental effect of temperature on NO2 adsorption. At 373 K and 17 bar pressure, the amount of NO2 adsorbed was nearly the same as at 343 K and atmospheric pressure and 3-4 times greater than at 373 K and atmospheric pressure. Analysis of the NO2 adsorption isotherm by both Langmuir analysis and the BET adsorption theory suggested that the reversible NO2 adsorption mechanism involved micropore filling with a monolayer adsorbed near 10 bar. The time profiles during pressure release and temperature-programmed desorptions were consistent with important roles for critical temperature and pressure and van der Waals forces in the adsorption and condensation of NO2, CO2, and O2 within the micropores of the carbon. As with our previous work, the carbon acted as a catalyst to convert the NO in the presence of O2 to NO2 which was the adsorbed and desorbed species. This is potentially important for byproduct considerations since NO2 is an important feedstock for HNO3 and agricultural nitrates. This study suggested that activated carbon may be a candidate adsorption material for a process that captures and sequesters NOx. EF9501861