Adsorption of CO2 on Sodium-Exchanged Ferrierites: The Bridged

Jan 27, 2009 - Isosteric heats of adsorption of CO2 on sodium-exchanged ferrierites (Na-FER) with different Si/Al ratios were obtained from the adsorp...
1 downloads 0 Views 963KB Size
Download PDF
2928

J. Phys. Chem. C 2009, 113, 2928–2935

Adsorption of CO2 on Sodium-Exchanged Ferrierites: The Bridged CO2 Complexes Formed between Two Extraframework Cations ˇ ejka*,† Angeles Pulido,‡ Petr Nachtigall,*,‡,§ Arnosˇt Zukal,† Irene Domínguez,† and Jirˇ´ı C J. HeyroVsky´ Institute of Physical Chemistry, AS CR, V.V.i., DolejsˇkoVa 3, 18223 Prague 8, Czech Republic, Institute of Organic Chemistry and Biochemistry, AS CR, V.V.i., FlemingoVo na´m. 2, 16610, Prague 6, Czech Republic, and Department of Physical and Macromolecular Chemistry, Faculty of Science, Charles UniVersity in Prague, HlaVoVa 2030, 12840, Prague 2, Czech Republic ReceiVed: NoVember 14, 2008; ReVised Manuscript ReceiVed: December 23, 2008

Isosteric heats of adsorption of CO2 on sodium-exchanged ferrierites (Na-FER) with different Si/Al ratios were obtained from the adsorption isotherms recorded in the temperature range between 273 and 333 K. The isosteric heats of adsorption significantly depend on the content of the Na+ cation in the zeolite. Large isosteric heats of adsorption were obtained for Na-FER when Si/Al ) 8.7. On the basis of calculations employing the periodic density functional theory (DFT) model, these large isosteric heats are attributed to the formation of the linearly bridged CO2 adsorption complexes formed between a pair of Na+ cations located in cationic positions. The CO2 adsorption complexes formed on such dual-cation sites are ca. 10 kJ/mol more stable than the CO2 adsorption complexes where CO2 interacts only with a single Na cation. 1. Introduction The economically feasible and environmentally friendly capture, storage, and separation of carbon dioxide is currently being investigated because of its technological importance.1 Microporous and mesoporous molecular sieves, namely, metal-organic frameworks, as well as functionalized silicas, activated carbon, and zeolites, are materials focused on for CO2 capture and separation. With regard to that, it is necessary to optimize a specific material capable of capturing CO2 just enough to allow for a subsequent CO2 release without unavoidable energetic demands. From this point of view, metalexchanged zeolites are of particular interest as the interaction strength between the CO2 and the solid adsorbent can be tuned by changing the metal cation,2 cation concentration,3 and/or the size and topology of the zeolite micropores.4 It is therefore important to increase our understanding of the details of the CO2-zeolite interaction. Several studies have dealt with CO2 adsorption on alkali-metal-exchanged zeolites (e.g., refs 2-8); however, the possibility of the formation of CO2 adsorption complexes with a pair of extraframework cations has not been considered thus far. In zeolites with a sufficiently large concentration of alkalimetal cations, the simultaneous interaction of the adsorbed molecule with more than one cation has been proposed and discussed in the past (e.g., refs 9-11). The details of CO adsorption complexes formed on a pair of Na+ or K+ cations have been reported based on the combination of density functional theory (DFT) calculations and variable-temperature IR spectroscopy, identifying the spectroscopic fingerprint of such “linearly bridged CO adsorption complexes” on “dual-cation sites”.12-15 Considering the fact that the carbon monoxide manifests bridged adsorption complexes, the existence of similar * Corresponding authors. E-mail: [email protected] (P.N.); [email protected] (J.C.). † J. Heyrovsky´ Institute of Physical Chemistry. ‡ Institute of Organic Chemistry and Biochemistry. § Charles University.

bridged adsorption complexes of carbon dioxide with a pair of cations should also be investigated. It is the goal of this study to show that such complexes of CO2 can indeed exist and that they play an important role in CO2 adsorption in metal-exchanged zeolites. For that purpose, the results of combined theoretical (periodic DFT modeling) and experimental (isosteric heats of adsorptions) investigations are presented for sodium-exchanged ferrierites (Na-FER). FER was chosen for this study as it is possible to prepare it with different Si/Al ratios, thus, different concentrations of Na+ ions, which is not possible in the case of the most frequently studied zeolites X and Y. The differences in the isosteric heats of adsorption are interpreted at the atomic level. 2. Materials and Methods 2.1. Experimental Details. 2.1.1. Materials. The ferrierite with a Si/Al ratio of 26.6 was obtained from Zeolyst Int. in ammonium form. Na,K-ferrierite with Si/Al ratio of 8.7 was provided by TOSOH Co., Japan. This sample was converted into ammonium form using 1 M NH4NO3 solution four times for 8 h at room temperature (100 mL of solution per 1 g of zeolite), washed several times with distilled water, filtered, and dried at 353 K overnight. Afterward, from the ammonium forms of ferrierites the sodium-exchanged samples (Na-FER) were obtained by ion exchange repeated with 1 M aqueous solution of NaNO3 at 333 K. In this case, the procedure had to be repeated seven times in order to obtain the complete ion exchange. The ferrierites modified by sodium cation exchange are denoted as Na-FER/A (Si/Al ) 26.6) and Na-FER/B (Si/Al ) 8.7). The chemical composition obtained using X-ray fluorescence spectroscopy (Philips PW 1404) is presented in Table 1. The Na-FER samples were characterized by X-ray powder diffraction (Bruker D8 with Cu KR radiation), scanning electron microscopy (JEOL JSM-5500LV), and infrared spectroscopy (Nicolet Prote´ge´ 460) using KBr pellets and self-supported wafers activated at 723 K for 3 h inside an IR cell.

10.1021/jp810038b CCC: $40.75  2009 American Chemical Society Published on Web 01/27/2009

Adsorption of CO2 on Na-FER

J. Phys. Chem. C, Vol. 113, No. 7, 2009 2929

TABLE 1: Chemical Analysis and Micropore Volume of the Na-FER sample

Si/Al

Na/Al

Vmicroa

Na/UC

A B

26.6 8.7

0.92 0.97

0.107 0.118

1.30 3.71

a

In cm3/g.

2.1.2. Adsorption Measurements. The adsorption isotherms of carbon dioxide in the temperature range of 273-333 K were determined using an ASAP 2020 (Micromeritics) static volumetric apparatus. In order to attain a sufficient accuracy in the accumulation of the adsorption data, the ASAP 2020 was equipped with pressure transducers covering the 133 Pa and 1.33 and 133 kPa ranges. Before the adsorption experiment began, the sample was outgassed under the turbomolecular pump vacuum using a special heating program allowing the slow removal of the majority of the preadsorbed water at low temperatures so as to avoid any potential structural damage to the sample as a result of hydrothermal alteration. Starting at ambient temperature, the sample was outgassed at 383 K until a residual pressure of 0.5 Pa was obtained. After further heating at 383 K for 1 h, the temperature was increased to 623 K. This temperature was maintained for 8 h. A homemade thermostat, maintaining the temperature of the sample with an accuracy of (0.01 K, was used for the measurement of the carbon dioxide adsorption at 273, 293, 313, and 333 K (the exact temperature of each measurement was determined using a platinum resistance thermometer.). Prior to each measurement of the carbon dioxide isotherm, the sample was evacuated at 423 K for 12 h. Nitrogen and argon isotherms were measured at 77 K. The adsorption isosteres were calculated from the experimental adsorption isotherms; to this end, the isotherms were transformed into coordinates of log p versus a. The values of log p corresponding to a series of adsorbed amounts a ) 5, 10, 15,..., 55 cm3/g STP were calculated using a polynomial interpolation procedure. The isosteric heats of adsorption qst were calculated from the slope of the adsorption isosteres using the equation

d(log p)/d(1/T) ) -qst/2.303R

(1)

where R is the gas constant. 2.2. Calculations. The ferrierite structure, FER, has an orthorhombic unit cell (UC), T36O72, where T is either a Si or Al atom, and a space group Immm with four tetrahedral sites (T1:T2:T3:T4 ) 1:2:2:4) and eight framework oxygen atoms that are symmetrically independent.16 It should be pointed out that this numbering scheme differs from that used in Database of Zeolite Structures, where T1 and T4 positions are switched. The eight-ring (8R) entrance window to the FER cage is formed at the intersection of the main (M) and perpendicular (P) channels. The equilibrium volume of the all-silica FER unit cell (with cell parameters of a ) 19.1468, b ) 14.3040, and c ) 7.5763 Å, and a volume of 2076.70 Å3) fitted previously17 was used for all the calculations. The calculations were performed for CO2 interacting with the most stable Na+ cation sites in the vicinity of a particular framework Al atom found previously.13,18 There are three stable Na+ sites in the vicinity of Al in the T4 position that are within 3 kJ/mol; therefore, CO2 adsorption complexes on each of these cation sites were considered. The notation introduced for the Cu-FER zeolite17 has been adopted for Na-FER: the Na+ cation site is denoted as Xn/Tm, where X refers to the main channel, perpendicular channel, or intersection (M, P, or I,

Figure 1. Dependence of the stabilization energy arising from the second Na+ ion on the RNa-Na distance calculated with the 2x(1-TNa) cluster model (depicted above the plot); energy is plotted with respect to 1-TNa-CO2 and 1-TNa at infinite distance.

respectively), n specifies a particular site within the channel system, and m designates the framework T-site (using the numbering scheme from ref 16). The details of the Na+ cation sites (coordination, localization, etc.) can be found in ref 13 along with the notation used for the cation sites. The high-silica sample was modeled with the UC having a composition of NaAlSi35O72 (corresponding to Si/Al ) 35) and the sample with a larger Na content was modeled with UC having a composition of Na4Al4Si32O72 (corresponding to Si/Al ) 8). The calculations were performed using a periodic DFT model implemented in the VASP program19-21 while employing the Perdew-Burke-Ernzerhof (PBE) exchange-correlation functional,22 the projector augmented wave approximation (PAW) of Blo¨chl,23,24 and the plane wave basis set with a kinetic energy cutoff of 400 eV, with the Brillouin-zone sampling restricted to the Γ-point. Zero-point energy (ZPE) corrections were calculated within the harmonic approximation using nine degrees of freedom for every CO2 molecule in the model system, keeping the framework atoms fixed. Adsorption enthalpy at 300 K, ∆Hads(300 K), was calculated from the internal energy of adsorption, ∆Uads(0 K), using the ideal gas equation. To investigate the stability of the linear CO2 adsorption complex formed between a pair of Na+ cations and to find the optimum distance between two Na+ cations in a “dual-cation site”, calculations on simple cluster models were performed. A 1-TNa and a double 2x(1-TNa) clusters, Al(OH)4Na and Al(OH)4Na-NaAl(OH)4, respectively, were used (see Figure 1). All calculations on cluster models were performed under the C2V symmetry constraint, using the G03 program suite,25 employing the PBE exchange-correlation functional and valencetriple-ζ basis sets with polarization functions;26 interaction energies were corrected for the basis set superposition error.27 Calculations shown in Figure 1 indicate that at the optimum distance of two Na+ cations (RNa-Na ) 7.3 Å) the linear CO2 adsorption complex formed on this simple model of dual-cation site is 13 kJ/mol more stable than the CO2 adsorption complex on the 1-TNa cluster model. 3. Results 3.1. Experimental Results. 3.1.1. Sample Characterization. X-ray powder diffraction patterns of the parent ferrierites (Figure 2) evidence a high crystallinity of both samples under study. Diffraction patterns of the sodium forms of ferrierites reveal that their crystallinity remains unchanged after repeated ion-exchange treatment. This conclusion is in a good agreement with the data of scanning electron microscopy (SEM). SEM micrographs of samples A and B show typical ferrierite

2930 J. Phys. Chem. C, Vol. 113, No. 7, 2009

Pulido et al.

Figure 2. X-ray diffraction patterns of parent ferrierites and Na-FER/A and Na-FER/B.

Figure 3. Scanning electron micrographs of sodium-exchanged ferrierites.

Figure 4. FTIR spectra of the parent ferrierites and Na-FER/A and Na-FER/B.

morphology, mostly forming sheets like discs or agglomerates of several cross-linked discs (Figure 3). It is seen that the resulting crystal size of ∼1 µm and crystal morphology are not influenced by the Si/Al molar ratio or alkali cation exchange.

The completeness of the ion exchange in the Na-FER is confirmed by the chemical analysis and infrared spectroscopy. The ratio Na+/Al of both samples is close to 1 (cf., Table 1). This corresponds to 1.30 and 3.71 Na+ cations per UC for FER/A and FER/B, respectively. The IR band at 3610 cm-1

Adsorption of CO2 on Na-FER

J. Phys. Chem. C, Vol. 113, No. 7, 2009 2931

Figure 5. Adsorption isotherms of nitrogen on Na-FER/A and Na-FER/B at 77 K. (The adsorption isotherm for Na-FER/B is offset vertically by 10 cm3/g STP.) Figure 7. Adsorption isotherms of CO2 on Na-FER/A and Na-FER/B at 273 and 333 K.

Figure 6. Adsorption isotherms of argon on Na-FER/A and Na-FER/B at 77 K. Inset: the low-pressure region.

corresponding to the Brønsted acid Si(OH)Al group is generated during the thermal activation of the zeolite if total exchange did not take place. The absence of this band in both samples evidence the completeness of the sodium ion exchange (Figure 4). Nitrogen adsorption isotherms obtained for Na-FER/A and Na-FER/B are presented in Figure 5. Both isotherms are typical for microporous materials showing a marked increase in the amount of adsorbed nitrogen at high relative pressures (p/p0 > 0.8), which can be related to slit-shaped pores among platelet particles. The argon adsorption isotherms are shown in Figure 6. It can be seen that the isotherm on the Na-FER/B with lower Si/Al ratio and higher concentration of Na+ cations is slightly steeper. Micropore volumes of both samples determined from the argon isotherms using the Horvath-Kawazoe method are given in Table 1. Unfortunately, due to a slow kinetics of argon adsorption it was not possible to measure isotherms in the region of very low equilibrium pressures. Therefore, the micropore size

distribution was calculated only in the limited range of micropore diameters and the median pore diameter could not have been determined. 3.1.2. CO2 Adsorption Isotherms. Carbon dioxide isotherms recorded on the Na-FER/A and Na-FER/B are presented in Figure 7 in linear coordinates a versus p. For clarity, the isotherms are shown only for the lowest and highest temperatures 273 and 333 K, respectively. The adsorption isotherms at both temperatures clearly illustrate the decisive role of the mode of interaction among sodium cations and adsorbed CO2 molecules. The Na-FER/B is characterized by the higher concentration of Na+ cations in comparison with the Na-FER/A. As a result, the amount of CO2 adsorbed on the Na-FER/B increases with increasing equilibrium pressure much faster than the amount of CO2 adsorbed on the Na-FER/A. Since both ferrierites are characterized practically by the same micropore volume, the limiting amount of CO2 corresponding to the filled micropores should be also practically identical. Therefore, the differences between CO2 isotherms on samples A and B arise only from the different concentrations of sodium cations. All carbon dioxide isotherms on the Na-FER/A and Na-FER/B are shown in Figure 8 in semilogarithmic coordinates a versus log p because these coordinates enable us to illustrate the region of low pressures in sufficient resolution. On the basis of these data, adsorption isosteres were calculated. To this end, values of log p corresponding to a series of adsorbed amounts were calculated using a polynomial interpolation procedure. Isosteres obtained in this way are shown in coordinates log p versus 1/T in Figure 9. The linear fit shows that the isosteric adsorption heat does not depend on the temperature and it depends only on the amount adsorbed. The experimental isosteric heats of adsorption of CO2 on Na-FER/A and Na-FER/B are reported in Figure 10. Isosteric heat of adsorption of CO2 on the Na-FER/B is higher than that on the Na-FER/A in the whole range of amount adsorbed. In the region of a < 15 cm3/g STP the isosteric heat of adsorption attains ca. 52 kJ/mol (Na-FER/B). Isosteric heat of adsorption of CO2 on the Na-FER/A quickly decreases with increasing amount adsorbed; it attains maximum value of ca. 45 kJ/mol for a ) 10 cm3/g STP.

2932 J. Phys. Chem. C, Vol. 113, No. 7, 2009

Pulido et al.

Figure 8. Adsorption isotherms of CO2 on Na-FER/A and Na-FER/B at temperatures 273, 283, 293, 303, 313, 323, and 333 K (in the direction from left to right).

Figure 9. Adsorption isosteres of CO2 on Na-FER/A and Na-FER/B. All the isosteres are marked with the corresponding amount adsorbed in cm3/g STP. (Straight lines represent a linear fit.)

3.2. Calculations. The calculated interaction energies and adsorption enthalpies are reported in Table 2 for the CO2 adsorption complexes formed on the single Na cations and for the bridged CO2 adsorption complexes formed between a pair of Na+ ions (dual-cation sites); selected adsorption complexes are depicted on Figure 11. The adsorption enthalpies, ∆Hads(300 K), calculated for CO2 on isolated cation sites are between -33 and -36 kJ/mol. The smaller ∆Hads(300 K) ) -30 kJ/mol calculated for the P6/T1 site is caused by the fact that the Na+ cation is located inside the six-ring (in coordination with four framework O atoms) and the adsorbed CO2 molecule cannot be bound as efficiently as in the case of more open Na+ sites (see Figure 1 in ref 18). Adsorption complexes with two CO2 molecules can form on all Na+ sites in Na-FER (Table 2); the strongest ability to bind more than one CO2 molecule was found for the Na+ sites on the channel intersection (I2/T2 and I2/T4) and for the M7/T3 site in the main channel, where the adsorption enthalpy of the second CO2 is only 5-7 kJ/mol smaller than for the first CO2 molecule. Owing to the size of the Na+ cation and its tight coordination with the framework as well as the topology of the FER, the third CO2 molecule cannot enter the

first coordination sphere of the Na+ directly and its interaction with the Na+ site is significantly weaker. The exception is the M7/T3 site, where two CO2 molecules can approach the Na+ in the direction of the perpendicular channel and another pair of CO2 molecules can be aligned along the main channel direction (Figure 12). Even in this case, however, the two CO2 molecules located along the perpendicular channel are farther from the Na+ ion (about 3 Å) as compared to 2.4 Å distance found for two CO2 molecules located along the main channel; thus, the interaction of the third and fourth CO2 molecules with the M7/T3 site is significantly weaker (Table 2). The stability of the CO2 adsorption complexes on the dualcation sites is always improved with respect to the stability of the corresponding complex on the isolated Na+ site. In the most stable bridged CO2 complexes on the dual site in Na-FER, the I2/T2-I2/T2 dual site, the stability of the CO2 bridged adsorption complex is 11 kJ/mol larger when compared to the corresponding isolated I2/T2 site (see I2/T2-I2/T2 and I2/T2 adsorption complexes in Figure 11). The CO2 molecule is almost linearly bridged between two Na+ cations separated by 7.3 Å. This is an ideal separation of Na+ cations for the formation of

Adsorption of CO2 on Na-FER

J. Phys. Chem. C, Vol. 113, No. 7, 2009 2933

Figure 10. Isosteric adsorption heats of CO2 on Na-FER/A and Na-FER/B.

Figure 11. CO2 adsorption complexes formed on isolated cation sites in Na-FER, P6/T1 (a), M7/T3 (b), I2/T2 (c), and on dual-cation site I2/T2-I2/T2 (d). The CO2 molecule and Na atoms are depicted as spheres; the O, Na, Si, and Al atoms are depicted in red, violet, gray, and black, respectively.

TABLE 2: Calculated Characteristics of the CO2 Adsorption Complexes in Na-FER Na+ sitinga P6/T1 I2/T2 M7/T3

I2/T4 P8/T4 M5/T4

P6/T1-P6T1 (7.6) I2/T2-I2/T2 (7.3) M7/T3-P8/T2 (7.4) I2/T4-M5/T4 (6.2) M5/T4-I2/T2 (6.8)

D(Na+-O)b

CO2/ ∆Hads UC ∆Eintc (300 K)d

2.33 2.44,2.51 2.34 2.37,2.38 2.36 2.40,2.42 2.41,2.42, 2.78 2.39,2.42, 2.95,3.18 2.34 2.37,2.37 2.31 2.37 2.33,3.95

1 2 1 2 1 2 3 4 1 2 1 1 2

-23 -16 -29 -24 -29 -21 -11 -4 -28 -25 -26 -28 -11

-30 -21 -36 -29 -35 -28 -17 -11 -35 -30 -33 -34 -16

Dual-Cation Sites 2.33/3.06 2.47/2.50 2.58/3.01 2.34/2.48 2.37/2.50

1 1 1 1 1

-25 -40 -32 -35 -37

-32 -47 -38 -42 -43

a Na+ site/Al position notation is defined in the text. The distances between the two Na+ cations forming a dual-cation site are given in parentheses (in angstroms). b The distance between Na+ and the oxygen atoms of CO2; the distances for both CO2 oxygen atoms to the closest Na+ are given in the case of dual-cation sites. c In kJ/mol. d ∆Hads from ∆Uads(0 K) and the ideal gas equation; ∆Uads(0 K) ) ∆Eint + ∆ZPVE.

a bridged complex with CO2. This is apparent from Figure 1, where the stabilization of the linearly bridged CO2 complex on the dual-cation site is plotted as a function of the distance of the two Na+ cations. Although other dual-cation sites reported in Table 2 also show some enhanced stability (with respect to the isolated sites), the stabilization due to the presence of the secondary Na+ cation is in the range of 2-6 kJ/mol as a consequence of the distance between the cations not being ideal or the localization of the cation and the framework topology not allowing the formation of linearly bridged adsorption complexes. Only a modest stabilization of the CO2 adsorption

Figure 12. Adsorption complex of four CO2 molecules formed on the M7/T3 site (view along the perpendicular channel). See the caption of Figure 11 for the coloring scheme.

complex due to the secondary Na+ cation was found for the P6/T1-P6/T1 dual-cation site; a weaker interaction of the adsorbent with the Na+ cation in P6/T1 site is due to the fact that Na+ cation is coordinated inside the six-ring and the interaction with adsorbent is shielded by the nearby framework oxygen atom.13,18 4. Discussion The adsorption enthalpies calculated at the DFT level and reported in Table 2 can be compared with the experimental isosteric adsorption heats reported in Figure 10. Adsorption heats found experimentally for Na-FER/B (Si/Al ) 8.6) are, for low coverage, about 52 kJ/mol. The largest ∆Hads(300 K) was calculated for CO2 adsorption on the I2/T2-I2/T2 dual-cation site (-47 kJ/mol) and smaller adsorption enthalpies (in the range from -38 to -43 kJ/mol) were found for the adsorption on other dual-cation sites. Isosteric adsorption heats found for highsilica sample A (Si/Al ) 26.6) are 45 kJ/mol at the lowest coverage, and they decrease to 31 kJ/mol at higher coverages. Calculated adsorption enthalpies for the high-silica FER model are in the range from -36 to -30 kJ/mol for mono(carbon dioxide) complexes, and they decrease for poly(carbon dioxide)

2934 J. Phys. Chem. C, Vol. 113, No. 7, 2009 complexes. Thus, calculated adsorption enthalpies are about 10 kJ/mol underestimated with respect to the corresponding experimental values. It is now widely accepted that the DFT method is capable to describe covalent and electrostatic interactions with a good accuracy, whereas the weak intermolecular interactions dominated by the dispersion interaction cannot be described.28 It is therefore assumed that the calculated adsorption enthalpies are about 10 kJ/mol underestimated due to the lack of dispersion interaction in the DFT description of the system. In addition, the dispersion interactions between the adsorbate and the zeolite are not site-specific; thus, it is assumed that the contribution due to the dispersion interaction is constant (∼10 kJ/mol) for all the adsorption enthalpies calculated. The experimentally determined adsorption heats of CO2 on Na-FER samples clearly show that the qst is significantly higher for the Na-FER sample with higher concentration of Na+ ions. Isosteric adsorption heats decreases with increasing CO2 dosage, and they level off at about 31 and 44 kJ/mol for Na-FER/A and Na-FER/B, respectively, at about 45 cm3 of CO2/1 g of sample. This dosage corresponds to approximately 4.5 CO2 in the unit cell. Both samples show that the qst starts to decrease rapidly for higher amounts of CO2 in the unit cell. This is consistent with the estimate of 4 and 4.5 CO2 molecules in FER/ UC for Na-FER/A and Na-FER/B, respectively, based on the pore-volume measurements of FER (0.107 and 0.118 mL/g for samples A and B, respectively, using N2 adsorption isotherms, Table 1) and a CO2 molar volume of 56.9 mL/g at 293 K. In the case of high-silica FER (Si/Al ) 26.6, corresponding to 1.3 Na/UC), qst decreases from 45 kJ/mol obtained for a low CO2 coverage (10 cm3/g, corresponding to 0.76 CO2 molecule per Na+) to ca. 34 kJ/mol at approximately 2 CO2/Na+ (a ) 25 mL/g). At this CO2 coverage range, the observed qst can be attributed to the formation of complexes of Na+ with a single CO2 molecule, first on the more open Na+ sites (I2/T2, I2/T4, M7/T3), followed by the formation of complexes of Na+ with 2 CO2 molecules on these sites and complexes on the sites where Na+ is less accessible for CO2 (P6/T1 site). For the CO2 coverages exceeding 2 CO2/Na+, additional CO2 molecules either do not interact directly with the Na+ cations or they are weakly bonded to the cation sites as indicated in Figure 12; qst is then determined by the nonspecific interactions with the framework (dispersion and electrostatic interactions with contributions due to the partial loss of the translational and rotational degrees of freedom of CO2 molecules upon entering the zeolite pores). This interpretation is in agreement with the recent investigation of the CO2 adsorption on Na-ZSM-5 performed by a step change response method.29 A rather different dependence of qst on the CO2 loading is observed for Na-FER/B (Si/Al ) 8.7, which corresponds to 3.7 Na+/UC). For the CO2 coverages of up to 2 CO2/UC (i.e., ∼0.5 CO2/Na+), the maximum isosteric heat was observed (52 kJ/mol). For higher CO2 loading, the qst monotonically decreases and levels off at approximately 45 kJ/mol at 4.5 CO2/UC (about 1.2 CO2/Na+). Apparently, there are enough Na+ cations in this sample to allow each CO2 molecule in the system to be involved in the specific interaction with at least one extraframework Na+. On the basis of the differences in qst obtained for both Na-FER samples and the results of DFT calculations, it has been concluded that the large qst (∼50 kJ/mol) observed for the sample with Si/Al ) 8.7 is due to the formation of bridged CO2 complexes on dual-cation sites (Figure 11d). Sample B (Si/Al ) 8.7) contains almost 4 Na+ ions in the UC. Although the distribution of Na+ ions is not known for such a sample (and depends on the distribution of the framework

Pulido et al. Al atoms, which is also unknown), statistically there are 22% and 44% of Al atoms in T2 and T4 sites, respectively; therefore, the population of the Na+ sites in the eight-ring windows at the intersection of the M and P channels (the I2/T2 and I2/T4 sites) is significant. These sites are particularly favorable for the formation of a dual-cation site, as is clear from Table 2 and Figure 11. This situation resembles the one observed previously for the CO interaction with Na-FER and K-FER as reported in refs 12 and 13, where more details about the dual-cation sites can be found. Due to the character of the CO2 molecule, the formation of bridged adsorption complexes, where each oxygen atom of CO2 interacts with a different cation site, can be expected also in zeolites other than FER and with different alkali-metal cations (currently under investigation). It should be also mentioned that the interaction strength of the CO2 with the isolated extraframework cations follows the same trends as observed for the CO adsorption on alkali-metal-exchanged zeolites:12,13,30,31 a stronger interaction between the cation and the zeolite framework results in a weaker interaction with the adsorbed molecule. The isosteric heat reported here for Na-FER/B (Si/Al ) 8.7) is larger than reported earlier for other sodium zeolites, e.g., Na-CHA,32 Na-Y, or Na-X.3 A strong dependence of the differential heats of adsorption on the Na+ density in the FAU supercage was reported based on the combined microcalorimetric measurements and GCMC simulations3 but with a different interpretation proposed. Instead of the formation of bridged CO2 complexes on dual-cation sites, the strong interaction of the CO2 with the NaX zeolite (Si/Al ) 1) was attributed to the population of the SIII sites, where the Na+ cations are more exposed to the open space of the supercage. A similar interpretation could be applied also for CO2 interaction with Na-FER; however, on the basis of the following arguments the interpretation of the differences in qst for Na-FER with different Na+ concentrations based on the existence of dual-cation sites is more appropriate: (i) The differences in calculated adsorption enthalpies for individual Na+ site are rather small in Na-FER (in the range from -33 to -36 kJ/mol, Table 2). Only the adsorption on P6/T1 site is calculated to be smaller (-30 kJ/mol); however, assuming the statistical distribution of framework Al, there is only 11% of Al in T1 site. (ii) The existence of dual-cation sites has been clearly demonstrated by the combination of variable-temperature IR spectroscopy and DFT modeling for CO adsorption on Na-FER where, in addition to the changes in the adsorption enthalpy, a distinct spectroscopic fingerprint of CO adsorption complexes on dual-cation sites at 2158 cm-1 has been described.13 5. Conclusions The isosteric heats of adsorption of CO2 in Na-FER significantly depend on the content of Na+ cation in the sample. In addition to the zeolite-framework topology and the size of the extraframework cation, the concentration of the cation in the sample can also be used to tune the interaction strength between the CO2 and the solid adsorbent. Na-FER with a high content of Na+ cation (Si/Al ) 8.7) shows a very large isosteric heat of adsorption of CO2 (up to 52 kJ/mol); this strong interaction is explained by the formation of linearly bridged CO2 complexes on the dual-cation sites. Except for the low CO2 loading, the difference between the qst obtained for Na-FER samples with Si/Al ratios of 26.6 and 8.7 is in the range of 10-15 kJ/mol. On the contrary, the maximum CO2 loading of the Na-FER samples does not substantially depend on the Na+ concentration.

Adsorption of CO2 on Na-FER Acknowledgment. The access to the METACentrum computing facilities provided under the research intent MSM6383917201 is acknowledged. P.N. acknowledges the support of the GACR (Grant No. 203/06/0324) and the ME CR (Grant Nos. LC512 and MSM0021620857); A.P. thanks Research Project No. Z4 055 0506. The work of I.D. has been supported by the “INDENS” project (MRTN-CT-2004-005503) while the work of J.C. and A.Z. has been supported by the GACR (Grant No. 203/08/0604). References and Notes (1) Idem, R.; Tontiwachwuthikul, P. Ind. Eng. Chem. Res. 2006, 45 (8), 2413–2413. (2) Pawlesa, J.; Zukal, A.; Cejka, J. Adsorption 2007, 13 (3-4), 257– 265. (3) Maurin, G.; Llewellyn, P. L.; Bell, R. G. J. Phys. Chem. B 2005, 109 (33), 16084–16091. (4) Goj, A.; Sholl, D. S.; Akten, E. D.; Kohen, D. J. Phys. Chem. B 2002, 106 (33), 8367–8375. (5) Plant, D. F.; Maurin, G.; Deroche, I.; Gaberova, L.; Llewellyn, P. L. Chem. Phys. Lett. 2006, 426 (4-6), 387–392. (6) Garrone, E.; Bonelli, B.; Lamberti, C.; Civalleri, B.; Rocchia, M.; Roy, P.; Arean, C. O. J. Chem. Phys. 2002, 117 (22), 10274–10282. (7) Montanari, T.; Busca, G. Vib. Spectrosc. 2008, 46 (1), 45–51. (8) Bonelli, B.; Civalleri, B.; Fubini, B.; Ugliengo, P.; Arean, C. O.; Garrone, E. J. Phys. Chem. B 2000, 104 (47), 10978–10988. (9) Salla, I.; Montanari, T.; Salagre, P.; Cesteros, Y.; Busca, G. Phys. Chem. Chem. Phys. 2005, 7 (12), 2526–2532. (10) Bordiga, S.; Palomino, G. T.; Paze, C.; Zecchina, A. Microporous Mesoporous Mater. 2000, 34 (1), 67–80. (11) Nachtigallova, D.; Bludsky, O.; Arean, C. O.; Bulanek, R.; Nachtigall, P. Phys. Chem. Chem. Phys. 2006, 8 (42), 4849–4852. (12) Garrone, E.; Bulanek, R.; Frolich, K.; Arean, C. O.; Delgado, M. R.; Palomino, G. T.; Nachtigallova, D.; Nachtigall, P. J. Phys. Chem. B 2006, 110 (45), 22542–22550. (13) Nachtigall, P.; Rodriguez Delgado, M.; Frolich, K.; Bulanek, R.; Turnes Palomino, G.; Lopez Bauca, C.; Otero Arean, C. Microporous Mesoporous Mater. 2007, 106, 162. (14) Arean, C. O.; Delgado, M. R.; Frolich, K.; Bulanek, R.; Pulido, A.; Bibiloni, G. F.; Nachtigall, P. J. Phys. Chem. C 2008, 112 (12), 4658– 4666. (15) Arean, C. O.; Delgado, M. R.; Bauca, C. L.; Vrbka, L.; Nachtigall, P. Phys. Chem. Chem. Phys. 2007, 9 (33), 4657–4661. (16) Vaughan, P. A. Acta Crystallogr. 1966, 21, 983.

J. Phys. Chem. C, Vol. 113, No. 7, 2009 2935 (17) Bludsky, O.; Silhan, M.; Nachtigall, P.; Bucko, T.; Benco, L.; Hafner, J. J. Phys. Chem. B 2005, 109 (19), 9631–9638. (18) Arean, C. O.; Nachtigallova, D.; Nachtigall, P.; Garrone, E.; Delgado, M. R. Phys. Chem. Chem. Phys. 2007, 9 (12), 1421–1436. (19) Kresse, G.; Hafner, J. Phys. ReV. B: Condens. Matter 1993, 48 (17), 13115–13118. (20) Kresse, G.; Hafner, J. Phys. ReV. B: Condens. Matter 1994, 49 (20), 14251–14269. (21) Kresse, G.; Furthmuller, J. Phys. ReV. B: Condens. Matter 1996, 54 (16), 11169–11186. (22) Perdew, J. P.; Burke, K.; Ernzerhof, M. Phys. ReV. Lett. 1996, 77 (18), 3865–3868. (23) Blochl, P. E. Phys. ReV. B: Condens. Matter 1994, 50 (24), 17953– 17979. (24) Kresse, G.; Joubert, D. Phys. ReV. B: Condens. Matter 1999, 59 (3), 1758–1775. (25) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J. J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03; Gaussian, Inc.: Pittsburgh, PA, 2003. (26) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys. 1992, 97 (4), 2571–2577. (27) Boys, S. F.; Bernardi, F. Mol. Phys. 1970, 19, 553. (28) Koch, W.; Holthausen, M. C. A Chemist’s Guide to Density Functional Theory; Wiley-VCH: Weinheim, Germany, 2000. (29) Wirawan, S. K.; Creaser, D. Microporous Mesoporous Mater. 2006, 91 (1-3), 196–205. (30) Nachtigall, P.; Bulanek, R. Appl. Catal., A 2006, 307 (1), 118– 127. (31) Nachtigallova, D.; Nachtigall, P.; Bludsky, O. Phys. Chem. Chem. Phys. 2004, 6 (24), 5580–5587. (32) Zhang, J.; Singh, R.; Webley, P. A. Microporous Mesoporous Mater. 2008, 111 (1-3), 478–487.

JP810038B