Separation of Monobranched and Dibranched Isomers of n-Hexane

J. Phys. Chem. B 2010, 114, 4465–4470

4465

Separation of Monobranched and Dibranched Isomers of n-Hexane on Zeolitic Molecular Sieves: A Thermodynamic Study Arse`ne H. Yonli,†,‡ Nicolas Bouillault,† Isabelle Gener-Batonneau,† and Samuel Mignard† Laboratoire de Catalyse en Chimie Organique, UniVersite´ de Poitiers, UMR CNRS 6503, 40 AVenue du Recteur Pineau 86022 Poitiers Cedex-France, and Laboratoire de Physique et de Chimie de l’EnVironnement, UniVersite´ de Ouagadougou, UFR SEA, 03 BP 7021 Ouagadougou 03 Burkina Faso ReceiVed: September 3, 2009; ReVised Manuscript ReceiVed: February 19, 2010

A thermodynamic study was realized by competitive adsorption over zeolitic adsorbents to determine the efficiency of these solids for the separation of monobranched and dibranched isomers of n-hexane. The effect of the zeolite structure was studied. The medium-pores ZSM-5-type zeolites were better than the large-pores BEA and MOR zeolites. The size and number of the extraframework cations had an important influence on the efficiency of the separation over ZSM-5 zeolites. The sodic Na6ZSM-5 sample was found to be the better adsorbent for the separation of the studied mixture because of steric hindrance induced by the presence of Na+ cations in the zeolite structure. The initial composition of the mixture also had an important influence on the separation. In fact, when the initial mixture was equimolar the monobranched isomer was preferentially adsorbed, whereas when the molar percentages of the isomers were different in the initial mixture the adsorption of the majority isomer was favored. The temperature of the adsorption was another important parameter influencing the separation. Indeed, when the temperature of adsorption was low the separation was more effective. At an adsorption temperature of 333 K the Na6ZSM-5 sample was the most efficient by adsorbing 65% of the monobranched isomer and only 35% of the dibranched isomer. 1. Introduction The reglementation for the composition of gasoline is going to be more and more severe in many countries. The aromatic, nitrogenous, sulfured, and oxygenated compounds which are compounds with a high octane number have to be drastically reduced in the composition of gasoline.1-4 One of the solutions for maintening the octane number is to increase the gasoline content in highly branched paraffins and to decrease linear and monobranched ones. Yet, the initial treatment of crude oil produces gasoline with low octane number. The octane number of gasoline can be improved by isomerization of C5-C10 linear paraffins and by removal of nonconverted linear paraffins by adsorption on 5A zeolite beds. The separation of dibranched and monobranched paraffins significantly enhances gasoline octane number, which increases with the degree of branching. However, this kind of separation is difficult using classic methods because the physical and chemical properties of the isomers are very close. Due to their shape-selective adsorption properties zeolites are well-designed adsorbants for this separation.5-7 Several zeolites with a high Si/Al ratio, such as ZSM-5,7-15 BEA,10,16-18 MOR,10,16 and TON,19,20 are described in the literature to achieve separation of n-alkanes/isoalkanes mixtures. It was found that both adsorption and diffusion in zeolites depend on the size and shape of sorbates,8,9,21 the framework Si/Al ratio of the zeolite, and the nature and number of extraframework cations.22-24 ZSM-5, MOR, and BEA zeolites preferentially adsorb n-alkanes. The adsorption phenomenon is governed by the steric hindrance and not by the existence of preferential sorption * To whom correspondence should be addressed. E-mail: yarsene@ hotmail.com. † Universite´ de Poitiers. ‡ Universite´ de Ouagadougou.

sites.15 According to Monte Carlo simulations,25-31 the selfdiffusion of alkanes in ZSM-5 zeolite is inversely proportional to their branching degree. Nevertheless, the diffusion of nalkanes decreases in the presence of branched alkanes in ZSM-5 zeolite.25 Moreover, Krishna et al.31 pointed out an inflection point in the sorption isotherm of hexane isomers in silicalite, indicating the presence of two sorption sites (channel intersections and straight channels). According to Krishna et al., this effect can be used in the separation of hexane isomers. These experimental or theoretical approaches are in majority devoted to single-component sorption studies in zeolites with a high framework Si/Al ratio. The aim of the present work is to study the separation of hexane monobranched and dibranched isomers in binary mixtures over different zeolites. The effect of the zeolite structure will be studied using ZSM-5, BEA, and MOR zeolites. The effect of the extraframework cations, initial composition of the mixture, and temperature of adsorption will be also studied for mixtures of 3-methylpentane and 2,3dimethylbutane. 2. Materials and Methods 2.1. Adsorbents and Reagents. ZSM-5-type zeolites were provided by Zeolyst International. BEA and MOR zeolites were supplied by Su¨d-Chemie. 3-Methylpentane (3MP) and 2,3dimethylbutane (23DMB) with a purity of 99.9% were provided by Fluka Chemie AG. The global framework Si/Al ratio was determined by elemental analysis. The BET surfaces and porous volumes were determined by nitrogen adsorption at 77 K with a TRISTAR Micromeritics instrument. 2.2. Binary Mixture Adsorption Procedures. The adsorption experiments were performed by volumetry. The experimental volumetric setup is represented in Figure 1.

10.1021/jp908513z  2010 American Chemical Society Published on Web 03/12/2010

4466

J. Phys. Chem. B, Vol. 114, No. 13, 2010

Yonli et al.

Figure 1. Experimental volumetric setup.

We have two flaskes containing pure and degassed monobranched and dibranched paraffins in the liquid state, a mixing room, two pressure sensors, and two He calibrated volumes (V1 and Vr). V1 volume is connected to a gas chromatograph (GC) equipped with a DB1 column and a FID detector. All tubings and valves are heated at 333 K to avoid liquid condensation. A defined pressure for the two sorbates was sent to the mixing room. The initial composition of the mixture was defined according to the partial pressure of each sorbate in the mixture, by assuming the sorbates followed the equation of perfect gases. The mixture was homogenized for 4 h. The adsorption experiments were performed at 298 or 333 K. The zeolite samples (ca. 150 mg) were placed in the adsorber and outgassed under secondary vacuum at 623 K for 12 h prior to the sorption measurements and cooled down to 298 or 333 K. A first increment of pressure of the homogenized mixture was introduced in the volume V1. A small part of this gaseous mixture was analyzed by GC to determine its composition before adsorption. The pressure (P1) in the volume V1 was noted. The mixture was then slackened in the volume Vr. A progressive decrease of the pressure occurred due to the adsorption on the zeolite. The pressure decrease was recorded versus time until stabilization. When the pressure was stabilized in the volume Vt (with Vt ) V1 + Vr), the volume Vr was isolated, the value of residual pressure (Pt) in the volume V1 was noted, and the composition of the residual gaseous phase was analyzed by GC. From the values V1, Vr, P1, and Pt and the gas-phase composition before and after adsorption, the adsorbed amount for each constituent was calculated. Relation 1 gives the adsorbed amounts of each constituent i in the mixture for the kth equilibrium (k ) 1 for the first equilibrium)

nia(k + 1) )

[

V1 (y (k + 1)P1′ - yi(k + 1)Pt′) + RT 1,i Vr (y (k)Pt - yi(k + 1)Pt′) + nia(k) (1) RT i

]

Figure 2. Adsorption isotherms for 3MP/2DMB mixture (50/50) over H0.2ZSM-5 at 333 K. Opened and filled symbols correspond to two different experiments.

where P1 is the gas partial pressure in the initial volume V1, Pt is the gas partial pressure in the total volume Vt, Vr is the volume of the adsorber, P′1 is the gas partial pressure in the volume V1 for the (k + 1)th point, P′t is the gas partial pressure in the volume Vt for the (k + 1)th point, yi(k) and nia(k) are, respectively, the molar fraction of constituent i at equilibrium in the gas phase and the amount of constituent i for the kth adsorption. Thus, the entire adsorption isotherm for each constituent in the mixture is obtained step by step by incrementing the pressure in the volume V1. For each zeolite, experiments have been carried out 2 or 3 times and the resulting points on the adsorption isotherm are a superposition of each trial. In Figure 2, two mixture sorption experiments on H0.2ZSM-5 are reported, showing the good repeatability of the obtained results. This Figure presents the adsorption isotherms for 3MP/2DMB mixture (50/50) over H0.2ZSM-5 at 333 K are shown.

Hexane Isomers Separation over Zeolites

J. Phys. Chem. B, Vol. 114, No. 13, 2010 4467

TABLE 1: Physico-Chemical Characteristics of the Studied Zeolites designation

formula of unit cell

molar mass (g/mol)

global Si/Al ratio

BET surface (m2/g)

total porous volume (cm3/g)

microporous volume (cm3/g)

H0.2ZSM-5 H6ZSM-5 Na6ZSM-5 Na4ZSM-5 Na7BEA Na8MOR

H0.19Al0.19Si95.48O192 H6.62Al6.62Si89.38O192 Na6.62Al6.62Si89.38O192 Na3.37Al3.37Si92.63O192 Na7.11Al7.11Si56.89O128 Na8.42Al8.42Si39.58O96

5750 5760 5905 5834 3996 3065

504 13.5 13.5 27.5 12.5 4.7

368 377 354 257 744 490

0.203 0.170 0.146 0.113 0.675 0.207

0.149 0.144 0.132 0.099 0.235 0.182

3. Results and Discussion 3.1. Effect of the Structure of the Zeolite on the Adsorption of Mixture of Monobranched and Dibranched Isomers of n-Hexane. The samples nomenclature and physicochemical characteristics are reported in Table 1. The unit cell formulas have been obtained using the global Si/Al ratio and assuming that all Al atoms are located in the zeolite framework. From the nitrogen adsorption experiments it appears that for all the samples microporous volumes are predominant except for the BEA sample, which exhibits a non-negligible mesoporous volume. In the case of ZSM-5 samples, the microporous volume is reduced by the presence of extraframework Na+ cations. The binary mixture adsorption experiments were performed from initial mixtures containing 50% of both 3MP and 23DMB. The sieving effect was studied over ZSM-5, BEA, and MOR zeolites at 333 K. The resulting adsorption isotherms on

Na6ZSM-5, Na7BEA, and Na8MOR samples are reported on Figure 3a, 3b, and 3c. The isotherms are well fitted by the Langmuir model

na Kp ) a 1 + Kp nsat

(2)

where na and nasat are the adsorbed amount at pressure p and the adsorbed amount at the saturation of the porous volume. K is the Langmuir constant. The Langmuir parameters obtained for the different zeolites are reported in Table 2. The sorption capacities of the three zeolites expressed in mmol · g-1 for 3MP and 23DMB decreases in the following order: Na7BEA > Na8MOR > Na6ZSM-5. This order is in

Figure 3. (a) Adsorption isotherms for 3MP/2DMB mixture (50/50) over Na6ZSM-5 at 333 K: 3MP partial amount (4), 23DMB partial amount (2), and total amount (*). (b) Adsorption isotherms for 3MP/2DMB mixture (50/50) over Na7BEA at 333 K: 3MP partial amount (4), 23DMB partial amount (2), and total amount (*). (c) Adsorption isotherms for 3MP/2DMB mixture (50/50) over Na8MOR at 333 K: 3MP partial amount (4), 23DMB partial amount (2), and total amount (*).

4468

J. Phys. Chem. B, Vol. 114, No. 13, 2010

Yonli et al.

TABLE 2: Langmuir Thermodynamic Parameters for 50/50 Mixture Adsorption of 3MP and 23DMB on Na6ZSM-5, Na7BEA, and Na8MOR at 333 K Na6ZSM-5

Na7BEA

a

Na8MOR

K n sat (molecules/ K n sat (molecules/ K nasat (molecules/ (bar-1) UC; mmol · g-1) (bar-1) UC; mmol · g-1) (bar-1) UC; mmol · g-1) 3MP 1387 23DMB 13 151

a

1.8 (0.305) 1.0 (0.169)

3029 3003

2.6 (0.651) 2.4 (0.600)

1826 1877

0.9 (0.293) 0.9 (0.293)

coherence with the available microporous volume (Table 2) and data from the literature.16 For Na6ZSM-5 sample, the total adsorbed amount is 2.8 molecules by unit cell. This value is slightly lower than the value found in the literature for silicalite or for highly siliceous ZSM-5. Indeed, according to the literature,14,28,31 in ZSM-5type zeolites there are two different sorption sites located at the channel intersections and in the straight channels. 23DMB as well as 3MP were adsorbed preferentially at the channel intersections. Furthermore, the maximum capacity for monobranched and dibranched hexane isomers determined by singlecomponent adsorption13,15,28,32 is 4 molecules by unit cell. The observed difference is certainly due to the presence in the channel intersection of extraframework sodium cations.33 Furthermore, the adsorbed amount of monobranched isomer is higher than that of the dibranched isomer for the Na6ZSM-5 sample; this can be ascribed to the steric hindrance due to a 23DMB critical diameter higher than that of 3MP and very close to ZSM-5 pore aperture. The total adsorbed amount in Na8MOR is higher than that in Na6ZSM-5, but the partial adsorbed amounts of 3MP and 23DMB in Na8MOR are the same: thus, no thermodynamic separation effect is observed for the Na8MOR sample. 3MP and 23DMB molecules occupy a porous volume of 0.076 cm3, which represents only 42% of the available Na8MOR porous volume. Moreover, this value is lower than the available volume in the 12-member straight channels (0.65 × 0.70 nm), which is 0.117 cm3 · g-1.16 3MP and 23DMB molecules seem to be only adsorbed in the 12-ring straight channels, the small side pockets (0.26 × 0.57 nm) being inaccessible. This inaccessibility of the mordenite side pockets has also been observed by Huddersman et al.16 and is probably due to the presence of sodium ions in the vicinity of the 8-member windows.34 Thus, as 12-member straight channels are too large compared to the 23DMB and 3MP kinetic diameter, no separation is observed. The Na7BEA sample has a higher porous volume than MOR and ZSM-5 samples with two 12-member channel systems with larger aperture (0.55 × 0.55 and 0.76 × 0.64). By comparison with those of the MOR sample, the smaller channels of the BEA sample seem to be accessible to both 3MP and 23DMB molecules, which occupy around 69% of the available porous volume. However, the partial amounts of 3MP and 23DMB are very close, indicating that the majority of the adsorption phenomena occur in large channels which are less sensible to the steric hindrance of the two molecules. The study of the Langmuir constants reported in the Table 2 shows that the two isomers have close K values when they are adsorbed in Na8MOR (1826 and 1877 bar-1, respectively, for 3MP and 23DMB) and Na7BEA zeolite (3029 and 3003 bar-1 for 3MP and 23DMB, respectively). On the contrary, the Langmuir constants for 3MP and 23DMB are very different in the case of Na6ZSM-5 (1387 and 13151 bar-1, respectively, for 3MP and 23DMB). This probably indicates that in the case of MOR and BEA zeolites where molecules are adsorbed in channels with a diameter higher than in ZSM-5, molecules are

sufficiently far from the channel walls so they do not specifically interact with them. In Na6ZSM-5 zeolite, the channels diameters are very close to the kinetic diameter of the two molecules and this induces a confinement effect. This effect is enhanced in the case of 23DMB, which possesses a kinetic diameter higher than 3MP. According to Corma et al.,35 the confinement effect is responsible at low loading for an increase of the sorption energy. At low loading, molecules try to achieve the best fit between their size and the framework environment in order to enhance their van der Waals attraction energy. This achievement is easier for 23DMB, which fit better channel intersection than 3MP. Thus, a better confinement of 23DMB than 3MP in Na6ZSM5 could explain its higher Langmuir K constant. 3.2. Effect of Extraframework Cations in ZSM-5 Type Zeolites. Since ZSM 5-type zeolites seem to be more promising for the separation of monobranched and dibranched hexane isomers, the effect of extraframework cations on this separation was studied over H0.2ZSM-5, H6ZSM-5, Na4ZSM-5, and Na6ZSM-5 samples. The adsorption isotherms for H6ZSM-5, Na4ZSM-5, and H0.2ZSM-5 samples are shown in Figure 4a, 4b and 4c, respectively. All isotherms have been fitted by the Langmuir model, and the resulting Langmuir parameters are reported in Table 3. For all the samples, the total adsorbed amount is in agreement with the microporous available volume. The two protonic ZSM-5 zeolites adsorbed a quite similar amount of 3MP and 23DMB. Nevertheless, Langmuir constants are higher on H0.2ZSM-5 for the two isomers. Acidity seems to induce a negative electrostatic effect upon the host-guest interaction. The high hydrophobicity of the H0.2ZSM-5 sample could certainly explain the better affinity of 3MP and 23DMB with the zeolite compared to H6ZSM-5 sample. Surprisingly, the results obtained for Na4ZSM-5 are not those expected. Nevertheless, the shapes of the sorption isotherms (Figure 4b) are unusual and the plateau slope is higher by comparison with the other zeolites, indicating a multilayer adsorption on the external surface. However, according to the saturation amount calculated from the Langmuir model, the 3MP/23DMB separation seems to be slightly better on Na4ZSM-5 compared to protonic ZSM-5. This effect is enhanced when the extraframework cations number increases in Na6ZSM5. Extraframework sodium cations located preferentially at the channel intersections33 reduce the available volume for sorbates and increase the confinement effect in favor of dibranched isomer, which is less adsorbed in capacity but better retained (higher K value). The adsorption capacity of the monobranched isomer increases clearly, inducing a promising separation between 3MP and 23DMB. To conclude on this point we can say that the nature and number of extraframework cations are important for the separation of monobranched isomers from dibranched isomers over ZSM-5 type zeolites. 3.3. Effect of the Initial Composition of the Mixture. The effect of the initial composition of the mixture on the separation of monobranched and dibranched isomers of hexane was studied over Na6ZSM-5. The adsorption experiments were performed under three conditions: a first mixture composed by 50% of both 3MP and 23DMB, a second mixture with a composition of 80% and 20%, respectively, for 3MP and 23DMB, and a third mixture composed by 20% and 80% of 3MP and 23DMB, respectively. The adsorption isotherms were well fitted by the Langmuir model, and the adsorption data are reported in Table 4. For the first mixture with 50% of the two sorbates we have a preferential adsorption of the monobranched isomer (65% of

Hexane Isomers Separation over Zeolites

J. Phys. Chem. B, Vol. 114, No. 13, 2010 4469

Figure 4. (a) Adsorption isotherms for 3MP/2DMB mixture (50/50) over H6ZSM-5 at 333 K: 3MP partial amount (4), 23DMB partial amount (2), and total amount (*). (b) Adsorption isotherms for 3MP/2DMB mixture (50/50) over Na4ZSM-5 at 333 K: 3MP partial amount (4), 23DMB partial amount (2), and total amount (*). (c) Adsorption isotherms for 3MP/2DMB mixture (50/50) over H0.2ZSM-5 at 333 K: 3MP partial amount (4), 23DMB partial amount (2), and total amount (*).

TABLE 3: Langmuir Thermodynamic Parameters for 50/50 Mixture Adsorption of 3MP and 23DMB on H0.2ZSM-5, H6ZSM-5, Na4ZSM-5, and Na6ZSM-5 at 333 K H0.2ZSM-5

3MP 23DMB

H6ZSM-5

Na4ZSM-5

Na6ZSM-5

K (bar-1)

nasat (molecules/UC)

K (bar-1)

nasat (molecules/UC)

K (bar-1)

nasat (molecules/UC)

K (bar-1)

nasat (molecules/UC)

2637 4029

1.7 1.3

1525 2919

1.8 1.6

1028 1384

1.4 1.0

1387 13151

1.8 1.0

TABLE 4: Adsorption Percentages at Adsorption Equilibrium for 3MP and 23DMB with Varying Compositions of Mixture over Na6ZSM-5 at 333 K mixture 50/50 mixture 80/20 mixture 20/80 (3MP/23DMB) (3MP/23DMB) (3MP/23DMB) 3MP adsorbed at equilibrium 23DMB adsorbed at equilibrium

65%

84%

23%

35%

16%

77%

3MP). However, when the ratios of the isomers in the mixture are very different (0.8 and 0.2, for example), the majority isomer is the most adsorbed by the zeolite. Thus, the selectivity of the separation depends not only on the nature of the adsorbent but also on the mixture proportions. This fact has also been observed by Schuring et al.25 It can be explained by the fact that statistically the compound “in excess” (for example, 23DMB molecules) has a higher probability to be adsorbed at the channel

intersections, inducing a more difficult access to the straight channel for the other compound (3MP in the example). Further studies should be realized for gas mixtures with less different isomers ratios in order to assess in depth the influence of the initial composition of the mixture on the efficiency of the separation. We can assume from the present data that for an efficient separation of monobranched isomers from dibranched isomers the percentage of the dibranched isomers has to be equal to or lower than the percentage of the monobranched isomers in the initial mixture. 3.4. Effect of the Temperature of Adsorption. The effect of the adsorption temperature was studied over H0.2ZSM-5 for a 50/50 mixture of 3MP and 23DMB. The experiments were performed at 298 and 333 K. As it should be expected, the adsorption capacities determined by Langmuir modelization are higher when the adsorption temperature is low. Indeed, at 298 K, we have, respectively, 3.99 and 2.45

4470

J. Phys. Chem. B, Vol. 114, No. 13, 2010

molecules adsorbed per unit cell for 3MP and 23DMB while at 333 K only 1.7 molecules of 3MP and 1.3 molecules of 23DMB are adsorbed per unit cell. On the contrary, the performances in separation for H0.2ZSM-5 are different when the adsorption is performed at 298 or 333 K. Indeed, at 298 K 62% of the monobranched isomer and 38% of the dibranched isomer are adsorbed by the zeolite. At 333 K the separation is less effective: 57% of the monobranched isomer and 43% of the dibranched isomer are adsorbed. Thus, it seems that separation of a mixture of monobranched and dibranched isomers of hexane is more efficient when the experiments are performed at low temperature. 4. Conclusions This study performed over zeolitic adsorbents had to determine the better solids and better conditions for an efficient separation of dibranched isomers of n-hexane from monobranched isomers in order to improve the octane number of gasoline without adding polluting compounds. The performances of zeolites were compared according to their structure. The medium-pores ZSM-5 zeolites were more efficient for separation than the large-pores BEA and MOR zeolites because of the presence of two different sorption sites and a better confinement effect due to the channel diameters in the case of ZSM-5 type zeolites. The preferential adsorption of the monobranched isomer was ascribed to the steric hindrance due to the dibranched 23DMB isomer. The effect of the extraframework cations was also studied over ZSM-5 samples. The efficiency of protonic ZSM-5 samples was lower than that of sodic ZSM-5 samples. The presence of the extraframework cations plays an important role for separation of isomers of n-hexane. The separation of isomers is better when the extraframework cations are big enough to enhance the shape selectivity for the adsorption of the isomers. The initial composition of the mixture of isomers was also very important for the separation process. When we had an equimolar composition of isomers in the initial mixture the monobranched isomer was preferentially adsorbed over Na6ZSM-5. However, when a compound is in majority in the initial mixture, it is always preferentially adsorbed. The effect of the temperature of adsorption on the separation of isomers was also studied, and the separation was found to be more efficient for low temperatures by comparison with higher temperatures. Thus, the Na6ZSM-5 zeolite sample seems to be the most efficient adsorbent. For an initial equimolar mixture it shows a promising separation performance: 65% of 3MP adsorbed against 35% of 23DMB.

Yonli et al. References and Notes (1) Rossini, S. Catal. Today 2003, 77, 467. (2) Marcilly, C. In Zeolites and Mesoporous Materials at the dawn of the 21th Century; Galarneau, A., Di Renzo, F., Fajula, F., Vedrine, J. , Eds.; Studies in Surface Science and Catalysis; 2001; Vol. 135, p 37. (3) Ma, X.; Sun, L.; Song, C. Catal. Today 2002, 77, 107. (4) Hernandez-Maldonado, A. J.; Yang, R. T. Ind. Eng. Chem. Res. 2003, 42, 3103. (5) Dessau, R. M.; Chen, N. Y.; Garwood, W. E. J. Catal. 1978, 52, 453. (6) Chen, N. Y.; Garwood, W. E.; Dwyer, F. G. Shape selectiVity catalysis industrial applications; Marcel Dekker: New York, 1989. (7) Choudhary, V. R.; Akolekar, D. B. J. Catal. 1989, 117, 542. (8) Haaj, W.; Lago, R. M.; Weiz, P. B. Faraday Discuss. Chem. Soc. 1981, 72, 317. (9) Wu, P.; Debebe, A.; Ma, Y. H. Zeolite 1983, 3, 118. (10) Denayer, J. F.; Souverijns, W.; Jacobs, P. A.; Martens, J. A.; Baron, G. V. J. Phys. Chem. B 1998, 102, 4588. (11) Millot, B.; Methivier, A.; Jobic, H.; Clemenc¸on, I.; Rebours, B. Langmuir 1999, 15, 2534. (12) Jolimaitre, E.; Tayakout,-Fayolle, M.; Jallut, C.; Ragil, K. Ind. Eng. Chem. Res. 2001, 40, 914. (13) Bellat, J. P.; Lemaire, E.; Simon, J. M.; Weber, G. Adsorption 2005, 11, 109. (14) Ferreira, A. F. P.; Mittelmeijer-Hazeleger, M. C.; Bliek, A. Adsorption 2007, 13, 10. (15) Gener, I.; Rigoreau, J.; Joly, G.; Renaud, A.; Mignard, S. Stud. Surf. Sci. Catal. 2002, 142, 1679. (16) Huddersman, K.; Klimczyk, M. J. Chem. Soc., Faraday Trans 1996, 92, 143. (17) Ba´rcia, P. S.; Silva, J. A. C.; Rodrigues, A. E. Microporous Mesoporous Mater. 2005, 79, 145. (18) Ba´rcia, P. S.; Silva, J. A. C.; Rodrigues, A. E. Adsorpt. Sci. Technol. 2007, 25, 169. (19) Eder, F.; Lercher, J. A. J. Phys. Chem. B 1997, 101, 1273. (20) Pieterse, J. A. Z.; Veefkind-Reyes, S.; Seshan, K.; Lercher, J. A. J. Phys. Chem. B 2000, 104, 5715. (21) Denayer, J. F.; Souverijns, W.; Jacobs, P. A.; Martens, J. A.; Baron, G. V. J. Phys. Chem. B 1998, 102, 4588. (22) Olson, D. H.; Bisio, A. Butterwarths: England, 1984; p 251. (23) Choudhary, V. R.; Srinivansan, V. R. J. Catal. 1986, 102, 32. (24) Funke, H. H.; Kovalchick, M. G.; Falconer, J. L.; Woble, R. D. Ind. Eng. Chem. Res. 1996, 35, 1575. (25) Schuring, D.; Jansen, A. P. J.; van Santen, R. A. J. Phys. Chem. B 2000, 104, 941. (26) Beerdsen, E.; Smit, B.; Calero, S. J. Phys. Chem. B 2002, 106, 10659. (27) Vlugt, T. J. H.; Schenk, M. J. Phys. Chem. B 2002, 106, 12757. (28) Schenk, M.; Vidal, S. L.; Vlugt, T. J. H.; Smit, B.; Krishna, R. Langmuir 2001, 17, 1558. (29) Vlugt, T. J. H.; Krishna, R.; Smit, B. J. Phys. Chem. B 1999, 103, 1102. (30) Krishna, R.; van Baten, J. M. Sep. Purif. Technol. 2007, 55, 246. (31) Smit, B.; Krishna, R. Chem. Eng. Sci. 2003, 58, 557. (32) Zhu, W.; Kapteinjn, F.; Linden, B.; Moulijn, J. A. Phys. Chem. Chem. Phys. 2001, 3, 1755. (33) Kuc˘era, J.; Nachtigall, P. Phys. Chem. Chem. Phys. 2003, 5, 3311. (34) Devautour, S.; Abdoulaye, A.; Giuntini, J. C.; Henn, F. J. Phys. Chem. B 2001, 105, 9297. (35) Sastre, G.; Corma, A. J. Mol. Catal. A 2009, 305, 3.

JP908513Z