Toward Understanding the Influence of ... - ACS Publications

Article pubs.acs.org/Langmuir

Toward Understanding the Influence of Ethylbenzene in p-Xylene Selectivity of the Porous Titanium Amino Terephthalate MIL-125(Ti): Adsorption Equilibrium and Separation of Xylene Isomers Mariana A. Moreira,† Joaõ C. Santos,† Alexandre F. P. Ferreira,† José M. Loureiro,† Florence Ragon,‡ Patricia Horcajada,‡ Pascal G. Yot,§ Christian Serre,‡ and Alírio E. Rodrigues*,† †

Laboratory of Separation and Reaction Engineering (LSRE), Associate Laboratory LSRE/LCM, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias, 4200-465 Porto, Portugal ‡ Institut Lavoisier (UMR CNRS 8180), Université de Versailles Saint-Quentin-en-Yvelines, 45 Avenue des Etats-Unis, 78035 Versailles Cedex, France § Institut Charles Gerhardt Montpellier (UMR CNRS 5253), Université Montpellier 2, Place Eugene Bataillon, 34095 Montpellier Cedex 05, France S Supporting Information *

ABSTRACT: The potential of the porous crystalline titanium dicarboxylate MIL-125(Ti) in powder form was studied for the separation in liquid phase of xylene isomers and ethylbenzene (MIL stands for Materials from Institut Lavoisier). We report here a detailed experimental study consisting of binary and multi-component adsorption equilibrium of xylene isomers in MIL-125(Ti) powder at low (≤0.8 M) and bulk (≥0.8 M) concentrations. A series of multi-component breakthrough experiments was first performed using n-heptane as the eluent at 313 K, and the obtained selectivities were compared, followed by binary breakthrough experiments to determine the adsorption isotherms at 313 K, using n-heptane as the eluent. MIL-125(Ti) is a para-selective material suitable at low concentrations to separate p-xylene from the other xylene isomers. Pulse experiments indicate a separation factor of 1.3 for p-xylene over o-xylene and m-xylene, while breakthrough experiments using a diluted ternary mixture lead to selectivity values of 1.5 and 1.6 for p-xylene over m-xylene and o-xylene, respectively. Introduction of ethylbenzene in the mixture results however in a decrease of the selectivity.

■

materials,13 hydrotropes,14 hydroxylated silica gel,15,16 and more recently, metal−organic frameworks (MOFs).17−24 MOFs are emerging as adsorbents in adsorptive separations, because they combine highly tunable crystalline structures and organic−inorganic compositions with large surface areas and pore volumes.25,26 These materials have already shown promising potential applications in important fields, such as selective molecular sieving,25,27−29 heterogeneous catalysis,25,30−32 and gas storage.25,33,34 Up to now, only a few MOFs have shown interesting features in the selective adsorption of xylene isomers, including the vanadium terephthalate MIL-47,20,35,36 the aluminum terephthalate MIL-53(Al),18,21,24 the zirconium terephtalate UiO66,37 and the zinc terephthalates Zn(BDC)(Dabco),22 MOF-5, and MOF-monoclinic.17 The titanium 1,4-benzenedicarboxylate

INTRODUCTION Hydrocarbon mixtures or fractions containing C8 aromatics are often a byproduct of oil refinery processes. These hydrocarbon mixtures typically contain ethylbenzene and xylene isomers, which include m-xylene, o-xylene, and p-xylene. Between these isomers, p-xylene becomes the isomer with the broadest commercial importance, because it is used in the manufacture of terephthalic acid, which is the basis for polyethylene terephthalate (PET) production. The benchmark for xylene isomer separation is the Parex process based on the simulated moving-bed technology. Nowadays, the advances achieved in this technology pass through the use of new materials as adsorbents because this process has already been studied and optimized to increase its performance and efficiency.1−4 Currently, Faujasite-type zeolites are the adsorbents used in the Parex process;5−10 however, other types of materials proved to be suitable in xylene isomer separation, such as silicates,11 polymers,12 carbon © 2012 American Chemical Society

Received: December 16, 2011 Revised: January 18, 2012 Published: January 19, 2012 3494

dx.doi.org/10.1021/la204969t | Langmuir 2012, 28, 3494−3502

Langmuir

Article

Figure 1. View of octahedral and tetrahedral cages of the MIL-125(Ti) structure. Titanium polyhedra and carbon atoms are in gray and black, respectively. For a better understanding, free accessible diameters are represented through yellow spheres. found elsewhere.24,40 Previous to the experimental tests, the adsorbent was activated. This activation was carried out in a stainless-steel column by heating the material in vacuum, at 423 K, for 20 h. In this way, it was possible to remove the water present in the pores. After activation, the column was closed in vacuum and placed in the HPLC oven. Column and material properties are presented in Table 1.

(or terephthalate) MIL-125(Ti), considering its high thermal stability (633 K), its porous character (SBET = 1550 m2 g−1), its composition, and its photocatalytic properties,38 seems to be an attractive candidate for several applications, including separation. MIL-125(Ti) or Ti8O8(OH)4−(O2C−C6H4− CO2)6 is built up from cyclic octamers of corner or edge sharing titanium octahedra connected to 12 other octamers through the terephthalate linkers (or BDC), resulting in a porous three-dimensional (3D) quasi-cubic tetragonal structure delimiting two types of cages, one octahedral (12.5 Å) and one tetrahedral (6 Å), accessible through narrow triangular windows of 5−7 Å, i.e., of a size close to those of xylene isomers (Figure 1). 38,39 We report here the performance of MIL-125(Ti) in its powdered form on the separation by adsorption in liquid phase of xylene isomers. A detailed experimental study of binary and multi-component adsorption equilibrium of xylene isomers in MIL-125(Ti) powder at bulk concentrations is shown. In this way, a set of single- and multi-component pulse experiments (PEs) to assess selectivities was conducted, using n-heptane as the eluent at 313 K. To complete this study, multi-component breakthrough experiments (BEs) were also performed in the same conditions and the obtained selectivities were compared. Binary BEs were also performed to determine the adsorption isotherm at 313 K using n-heptane as the eluent.

■

Table 1. Geometry and Properties of the Column and Adsorbent Lcol Øcol Vcol Vbed mads particle size

properties

unit

10.0 0.46 1.64 1.49 0.51 powder, 3.2

cm cm cm3 cm3 g μm

Experimental Procedure for PEs. PEs were performed using MIL-125(Ti) as the adsorbent. Prior to any experiment, a blank experiment (without column) was conducted to determine the dead volumes and the experimental results were corrected accordingly. All runs followed the same experimental protocol. Initially, the column was fed with n-heptane at 0.20 cm3/min, and the oven was programmed to maintain the temperature at 313 K. A pulse of one of the xylene isomers or a mixture was injected in the column using a loop of 20 μL. At the same time, the fraction collector program was started. All the samples were analyzed by a Shimadzu gas chromatograph equipped with a fused silica capillary column WCOT-CP xylenes, with a diameter of 0.53 mm and 50 m length, and a flame ionization detector (FID). The analyses were performed using a column temperature of 308 K and an injector and detector temperatures of 423 K. Helium was used as the carrier gas with a flow rate of 238.5 cm3/min and with a split ratio of 30. Shimadzu LCSolution software was used for graphic visualization and data acquisition. Experimental Procedure for BEs. BEs were carried out in the laboratory scale unit for determining adsorption equilibrium data. All experiments were performed at 313 K, using a flow rate of 0.20 cm3/min. Initially, the column was fed with eluent until thermal equilibrium was reached. At this point, column feed was switched to the xylene solution and samples were collected. Reverse BEs were performed switching the column feed from the xylene solution to the eluent. The experimental protocol was repeated with different concentrations of xylene solutions in the eluent. All samples were analyzed through gas chromatography, according to the method described in the previous subsection.

EXPERIMENTAL SECTION

Chemicals and Equipment. p-Xylene (p-x), o-xylene (o-x), and m-xylene (m-x) of GC grade (purity > 99.0%), from Sigma-Aldrich, were used as adsorbates. Ethylbenzene (eb) of GC grade (purity > 99.0%) from Fluka was also used. n-Heptane (n-hep), also from SigmaAldrich, was used as the solvent with a 99.5% purity. The synthesis of the MIL-125(Ti) powder form was performed by following the synthesis conditions reported previously.38 To remove free acid entrapped in the pores, the synthesized product was dispersed at room temperature in N,N-dimethylformamide (DMF) under stirring overnight. Then, to remove the DMF out of the pores, the same procedure was repeated twice using MeOH instead of DMF. The product was dried at 373 K overnight. The resulting solid was characterized by X-ray diffraction (XRD), infrared spectroscopy (IR), thermogravimetric analysis (TGA), nitrogen adsorption porosimetry, helium density, and mercury porosimetry (see the Supporting Information). The experiments were performed using a laboratory-scale unit, which consists mainly of a high-performance liquid chromatography (HPLC) experimental setup. This unit has the advantage of allowing fixed-bed experiments using a small sample amount, being appropriate for powder adsorbents. More information about this equipment can be 3495


Langmuir

■

Article

RESULTS AND DISCUSSION

PEs. PEs were first performed in the column and, according to Table 2, using MIL-125(Ti) powder as the adsorbent and n-heptane as the eluent. Table 2. Summary of the PEs Using MIL-125(Ti) Powder as the Adsorbent runs

pulse composition (wt %)

1 2 3 4 5 6

100% p-x 100% m-x 100% o-x 100% eb 33.3 wt % p-x, 33.3% m-x, and 33.3% o-x 25% p-x, 25% m-x, 25% o-x, and 25% eb

From PEs, the retention times could be calculated, to determine the separation factors through eq 1

αi , j = tri/trj

(1)

where tri is the relative retention time of the xylene isomer i or j. The results of xylene isomers pulses in the MIL-125(Ti)-packed column are shown in Figure 2. As observed in Figure 2, it appears that MIL-125(Ti) is a para-selective material because p-xylene is the most retained isomer (Figure 2a). This confirms the selectivity toward this isomer even in the presence of the other isomers, which has been confirmed with the pulse of the ternary mixture (Figure 2b). However, the presence of ethylbenzene in the mixture influences the preference of MIL-125(Ti) for p-xylene (Figure 2c). This results in a shorter retention time for p-xylene and strongly impacts the selectivity. Furthermore, this material hardly discriminates m-xylene from o-xylene even in the presence of the other xylene isomers and ethylbenzene. The preference for ethylbenzene and p-xylene of MIL-125(Ti) can be explained taking into account the diameters of the different components. Between all of the compounds, ethylbenzene and p-xylene are the compounds with a smaller diameter (6.7 Å) compared to m-xylene (7.1 Å) and o-xylene (7.4 Å).19 As mentioned before,41 the favored adsorption of xylenes within the tetrahedral cages, whose dimensions are close to 6 Å38 (i.e., very close to those of the xylene isomer dimensions), is probably the origin of the material selectivity. Figure 3 represents the selectivity of p-xylene against each one of the other isomers [(a) m-xylene, (b) o-xylene, and (c) ethylbenzene], as a function of the concentration for all conducted experiments. In the case of the PEs, the largest concentration observed was used for the representation. One can clearly observe that, in the presence of ethylbenzene, selectivity values decrease compared to those obtained from single-component pulse tests, in agreement with the results previously shown in Figure 2. Binary Component BEs. Binary component BEs were performed to determine the adsorption equilibrium data of each xylene isomer and ethylbenzene in powder MIL-125(Ti). In this way, binary mixtures containing p-xylene (runs 7−12), m-xylene (runs 13−18), o-xylene (runs 19−24), or ethylbenzene (runs 25−30) in n-heptane were used to feed the column with the aim of determining the adsorbent capacity as a function of feed concentrations. The binary equilibrium data for n-heptane (runs 31−36) was also determined using similar experiments based on p-xylene as the eluent. This set of

Figure 2. Pulse curves performed at 313 K and 0.20 cm3/min, using nheptane as the eluent, on powder MIL-125(Ti) (full lines are just guiding lines for reader’s eyes): (a) single components of 100 wt % xylene isomers and 100 wt % ethylbenzene (runs 1, 2, 3, and 4), (b) ternary mixture of 33.3 wt % p-x, 33.3 wt % m-x, and 33.3 wt % o-x (run 5), and (c) quaternary mixture of 25 wt % p-x, 25 wt % m-x, 25 wt % o-x, and 25 wt % eb (run 6).

experiments is essential for later use in modeling and design of a separation process. Feed concentrations are shown in Table 3. Adsorption equilibrium data for the five components are included in Figure 4. The values were obtained using the adsorbed amount of each component, which was calculated by integration of the experimental breakthrough curves (qads) and reverse breakthrough curves (qdes). From the results presented in Figure 4, one confirms the para-selectivity of the material at low concentrations (see Figure S6 of the Supporting Information). However, the ideal 3496


Langmuir

Article

Figure 3. Selectivity versus concentration: (a) αp‑x/m‑x versus Cp‑x, (b) αp‑x/o‑x versus Cp‑x, and (c) αp‑x/eb versus Ceb.

Figures 5 and 6 show the breakthrough and reverse breakthrough curves of MIL-125(Ti) powder using ternary and quaternary xylene isomer mixtures at 313 K and 0.20 cm3/min, respectively. Selectivities between the different xylene isomers were calculated according to the adsorbed amounts, as shown in eq 2

Table 3. Feed Composition Used in Binary Component BEs

Cfeed (%, v/v)

runs 7, 13, 19, 25, and 31

runs 8, 14, 20, 26, and 32

runs 9, 15, 21, 27, and 33

runs 10, 16, 22, 28, and 34

runs 11, 17, 23, 29, and 35

runs 12, 18, 24, 30, and 36

0.5

1

5

10

20

30

αi , j = (qi /Ci)/(qj /Cj)

(2)

where qi and qj are the adsorbed phase concentrations of xylene isomers in MIL-125(Ti) and Ci and Cj are the equilibrium concentrations of the xylene isomers in the external liquid phase. In the present case, BEs were performed using an equimolar mixture of xylene isomers, and then eq 2 can be simplified, as follows:

selectivity (defined as the ratio of single-component adsorbed amounts at identical component concentrations) decreases with the concentration, leading to a more difficult separation. Nevertheless, the capacity of MIL-125(Ti) is significantly higher, for all isomers, than the Faujasite-type zeolites that have a reported saturation capacity of about 1−2 mol/kg.5−10 Multi-component BEs. To ascertain the influence of ethylbenzene in xylene isomer separation, BEs were performed using ternary (runs 37−39) and quaternary (runs 40−43) mixtures at different concentrations. The feed compositions used in all experimental runs are presented in Table 4. On the basis of the literature,1 a typical feed composition of the industrial PAREX unit was produced and diluted in n-heptane (run 44) to simulate a real industry column feed. Therefore, all tests were performed using powder MIL-125(Ti) as the adsorbent, at 313 K and a flow rate of 0.20 cm3/min.

αi , j = qi /qj

(3)

From the breakthrough and reverse breakthrough curves of ternary xylene isomer mixtures (Figure 5), one can confirm that MIL125(Ti) is a para-selective material, able to separate p-xylene from the other isomers, in particular at low concentrations (1 wt %; see panels e and f of Figure 5), with selectivity values reaching 1.5 and 1.6 for p-xylene over m-xylene and o-xylene, respectively (see Figure 3). However, xylene separation becomes more difficult when the concentration of xylene isomers in the mixture increases, resulting in a decrease in the selectivity values from 1.5 or 1.6 down to 1.0 when concentrations increase from ∼0.1 to ∼3 mol/dm3 3497


Langmuir

Article

Figure 4. Binary adsorption equilibrium data: (a) p-xylene (runs 7−12), (b) m-xylene (runs 13−18), (c) o-xylene (runs 19−24), (d) ethylbenzene (runs 25−30), and (e) n-heptane (runs 31−36), in MIL-125(Ti) powder at 313 K.

Table 4. Feed Concentration Used in Experimental Runs

breakthrough

reverse breakthrough

Cfeed,p‑x (mol/dm3) Cfeed,m‑x (mol/dm3) Cfeed,o‑x (mol/dm3) Cfeed,eb (mol/dm3) Cfeed,n‑hep (mol/dm3) Cfeed,xylene isomers and ethylbenzene (mol/dm3) Cfeed,n‑hep (mol/dm3)

run 37

run 38

run 39

run 40

run 41

run 42

run 43

run 44

2.82 2.82 2.82

0.57 0.56 0.56

0.06 0.07 0.06

2.20 2.17 2.15 2.15

5.31

6.66

0.79 0.80 0.78 0.78 4.46

0.18 0.18 0.18 0.18 6.28

0.07 0.08 0.08 0.07 6.61

0.80 2.04 0.60 0.59 3.65

6.83

6.83

6.83

6.83

6.83

6.83

6.83

6.83

keeps its selectivity for p-xylene (see Figures 3 and 6). These results are similar to those obtained in PEs, where selectivity values also decreased with the presence of ethylbenzene in the mixture. The selectivity toward ethylbenzene and p-xylene might be explained by sieving that occurs in the tetrahedral cages. These cages are mainly accessible to the p-xylene and ethylbenzene molecules and not the other isomers.41 However, when the concentration of xylene

(Figure 3). Noteworthy, the introduction of ethylbenzene in the mixture results in an inversion of the selectivity at higher concentrations when the fixed-bed column is fed with a solution containing ethylbenzene (see Figure 6). At higher concentrations of ethylbenzene, MIL-125(Ti) is selective for ethylbenzene (selectivity of 0.8 for p-xylene over m-xylene and o-xylene and 0.7 for p-xylene over ethylbenzene), whereas for lower concentrations, the material 3498


Langmuir

Article

Figure 5. Ternary xylene isomer mixture breakthrough and reverse breakthrough curves performed in MIL-125(Ti) powder at 313 K and 0.20 cm3/min: (a and b) 33.3 wt % p-x, o-x, and m-x, (c and d) 10 wt % p-x, o-x, and m-x in n-hep, and (e and f) 1 wt % p-x, o-x, and m-x in n-hep.

at 298 K, are αp‑x/m‑x = 3, αp‑x/o‑x = 2.2, and αm‑x/o‑x = 0.97.41 These selectivities are for diluted systems and are slightly higher than the selectivities reported in this work because of the fact that the cage windows of MIL-125(Ti)_NH2 are smaller than the cage windows of MIL-125(Ti). Other MOFs tested for xylene separation are usually ortho-selective, such as MIL-47,20,35,36 MIL-53,18,21,24 and UiO-66(Zr).37 Therefore, their reported selectivities are not comparable to the selectivities obtained for para-selective materials. In the state-of-the-art Faujasite-type zeolites, the reported selectivities are usually higher than the selectivities of this study for MIL-125(Ti) (αp‑x/m‑x = 5.25, αp‑x/o‑x = 4.63, αp‑x/eb = 1.94, and αm‑x/o‑x = 1, at 293 K).6

isomers increases, it is likely that the packing effects in the cages become more favorable for ethylbenzene than p-xylene. In addition, a BE using a simulation of the real PAREX feed was performed (see panels i and j of Figure 6). Despite the larger concentration of m-xylene present in the solution, the material still shows a preference for ethylbenzene, with selectivity values of 1.2 for p-xylene over m-xylene, 1.1 for p-xylene over o-xylene, and 0.8 for p-xylene over ethylbenzene. MIL-125(Ti) belongs to the group of three isostructural MOFs until now found to be para-selective. In this group, we can find MIL-125(Ti), MIL-125(Ti)_NH2, and CAU-1(Al).41 The reported selectivities for MIL-125(Ti)_NH2 materials, 3499


Langmuir

Article

Figure 6. Quaternary xylene isomer mixture breakthrough and reverse breakthrough curves performed in MIL-125(Ti) powder at 313 K and 0.20 cm3/min: (a and b) 25 wt % p-x, o-x, m-x, and eb, (c and d) 10 wt % p-x, o-x, m-x, and eb in n-hep, (e and f) 2.5 wt % p-x, o-x, m-x, and eb in n-hep, (g and h) 1 wt % p-x, o-x, m-x, and eb in n-hep, and (i and j) 10 wt % p-x, 7.5 wt % o-x, 7.5 wt % eb, and 25 wt % m-x in n-hep (synthetic PAREX industrial feed diluted in n-heptane).

■

CONCLUSION Remarkably, it can be confirmed that MIL-125(Ti) is a paraselective material, being suitable to separate p-xylene from the

other xylene isomers, particularly at low-feed xylene concentrations. PEs revealed a separation factor of 1.3 for p-xylene over o-xylene and m-xylene. BEs performed with a diluted 3500


Langmuir

Article

(10) Minceva, M. Separation/Isomerization of Xylenes by Simulated Moving Bed Technology. Ph.D. Dissertation, Faculdade de Engenharia da Unversidade do Porto, Porto, Portugal, 2004. (11) Gu, X.; Dong, J.; Nenoff, T. M.; Ozokwelu, D. E. J. Membr. Sci. 2006, 280 (1−2), 624−633. (12) Mohammadi, T.; Rezaeian, M. P. Sep. Sci. Technol. 2009, 44 (4), 817−840. (13) Wu, Z.; Yang, Y.; Tu, B.; Webley, P.; Zhao, D. Adsorption 2009, 15 (2), 123−132. (14) Ramesh, N.; Jayakumar, C.; Nagendra Gandhi, N. Chem. Eng. Technol. 2009, 32 (1), 129−133. (15) Kiselev, A. V.; Aratskova, A. A.; Gvozdovitch, T. N.; Yashin, Y. I. J. Chromatogr., A 1980, 195 (2), 205−210. (16) Ageev, A. N.; Kiselev, A. V.; Yashin, Y. I. Chromatographia 1980, 13 (11), 669−672. (17) Gu, Z.-Y.; Jiang, D.-Q.; Wang, H.-F.; Cui, X.-Y.; Yan, X.-P. J. Phys. Chem. C 2009, 114 (1), 311−316. (18) Finsy, V.; Kirschhock, C. E. A.; Vedts, G.; Maes, M.; Alaerts, L.; De Vos, D. E.; Baron, G. V.; Denayer, J. F. M. Chem.Eur. J. 2009, 15 (31), 7724−7731. (19) Loiseau, T.; Serre, C.; Huguenard, C.; Fink, G.; Taulelle, F.; Henry, M.; Bataille, T.; Férey, G. Chem.Eur. J. 2004, 10 (6), 1373− 1382. (20) Finsy, V.; Verelst, H.; Alaerts, L.; De Vos, D.; Jacobs, P. A.; Baron, G. V.; Denayer, J. F. M. J. Am. Chem. Soc. 2008, 130 (22), 7110−7118. (21) Alaerts, L.; Maes, M.; Giebeler, L.; Jacobs, P. A.; Martens, J. A.; Denayer, J. F. M.; Kirschhock, C. E. A.; De Vos, D. E. J. Am. Chem. Soc. 2008, 130 (43), 14170−14178. (22) Nicolau, M. P. M.; Bárcia, P. S.; Gallegos, J. M.; Silva, J. A. C.; Rodrigues, A. E.; Chen, B. J. Phys. Chem. C 2009, 113 (30), 13173− 13179. (23) Castillo, J. M.; Vlugt, T. J. H.; Calero, S. J. Phys. Chem. C 2009, 113 (49), 20869−20874. (24) Moreira, M. A.; Santos, J. C.; Ferreira, A. F. P.; Müller, U.; Trukhan, N.; Loureiro, J. M.; Rodrigues, A. E. Sep. Sci. Technol. 2011, 46 (13), 1995−2003. (25) Ferey, G. Chem. Soc. Rev. 2008, 37 (1), 191−214. (26) Long, J.; Yaghi, O. Themed Issue: Metal−Organic Frameworks; The Royal Society of Chemistry (RSC): London, U.K., 2009; Vol. 38, pp 1201−1202. (27) Kuppler, R. J.; Timmons, D. J.; Fang, Q. R.; Li, J. R.; Makal, T. A.; Young, M. D.; Yuan, D. Q.; Zhao, D.; Zhuang, W. J.; Zhou, H. C. Coord. Chem. Rev. 2009, 253 (23−24), 3042−3066. (28) Li, J. R.; Kuppler, R. J.; Zhou, H. C. Chem. Soc. Rev. 2009, 38 (5), 1477−1504. (29) Ma, S. Q. Pure Appl. Chem. 2009, 81 (12), 2235−2251. (30) Isaeva, V. I.; Kustov, L. M. Pet. Chem. 2010, 50 (3), 167−180. (31) Corma, A.; Garcia, H.; Xamena, F. X. L. Chem. Rev. 2010, 110 (8), 4606−4655. (32) Wang, Z.; Chen, G.; Ding, K. L. Chem. Rev. 2009, 109 (2), 322− 359. (33) Morris, R. E.; Wheatley, P. S. Angew. Chem., Int. Ed. 2008, 47 (27), 4966−4981. (34) Britt, D.; Tranchemontagne, D.; Yaghi, O. M. Proc. Natl. Acad. Sci. U.S.A. 2008, 105 (33), 11623−11627. (35) Alaerts, L.; Kirschhock, C. E. A.; Maes, M.; van der Veen, M. A.; Finsy, V.; Depla, A.; Martens, J. A.; Baron, G. V.; Jacobs, P. A.; Denayer, J. E. M.; De Vos, D. E. Angew. Chem., Int. Ed. 2007, 46 (23), 4293−4297. (36) Alaerts, L.; Maes, M.; Jacobs, P. A.; Denayer, J. F. M.; De Vos, D. E. Phys. Chem. Chem. Phys. 2008, 10 (20), 2979−2985. (37) Bárcia, P. S.; Guimarães, D.; Mendes, P. A. P.; Silva, J. A. C.; Guillerm, V.; Chevreau, H.; Serre, C.; Rodrigues, A. E. Microporous Mesoporous Mater. 2011, 139 (1−3), 67−73. (38) Dan-Hardi, M.; Serre, C.; Frot, T.; Rozes, L.; Maurin, G.; Sanchez, C.; Férey, G. J. Am. Chem. Soc. 2009, 131 (31), 10857− 10859.

ternary mixture led to selectivity values of 1.5 and 1.6 for p-xylene over m-xylene and o-xylene, respectively. This preference for p-xylene can be explained by the sieving effect that occurs in the tetrahedral cage of the material. The presence of ethylbenzene in the mixture leads to a decrease in the selectivity values, indicating that ethylbenzene influences the selectivity of the material toward the xylene isomers. Quaternary breakthrough results may indicate competitive adsorption between p-xylene and ethylbenzene. This could be concluded from the obtained selectivity values, which are 0.8 for p-xylene over m-xylene and o-xylene and 0.7 for p-xylene over ethylbenzene. On the other hand, selectivity strongly depends upon the concentration of the component mixture. Selectivity is higher when diluted quaternary mixtures are used in BEs, obtaining selectivity values of 1.5 for p-xylene over m-xylene and o-xylene and no selectivity of p-xylene relative to ethylbenzene (selectivity = 1.1).

■

ASSOCIATED CONTENT

S Supporting Information *

MIL-125(Ti) powder characterization by XRD, IR spectroscopy, TGA, nitrogen porosimetry, helium density, and mercury porosimetry; supplementary binary adsorption data; and supplementary multi-component adsorption data. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Author

*Telephone: +351-22-5081671. Fax: +351-22-5081674. E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The research leading to these results was partly funded by the European Community’s 7th Framework Programme (FP7/ 2007-2013) under Grant 228862. MACADEMIA is a largescale integrating project under the Nanosciences, Nanotechnologies, Materials, and New Production Technologies Theme in FP7. This work is partially supported by Project PEst-C/EQB/LA0020/2011, financed by FEDER through Programa Operacional Factores de Competitividade (COMPETE) and by Fundaçaõ para a Ciência e a Tecnologia (FCT).

■

REFERENCES

(1) Minceva, M.; Rodrigues, A. E. AIChE J. 2007, 53 (1), 138−149. (2) Minceva, M.; Rodrigues, A. E. UOP’S Parex: Modeling, Simulation and Optimization. Proceedings of ENPROMER 2005; Rio de Janeiro, Brazil, Aug 14−18, 2005. (3) Gomes, P. S.; Minceva, M.; Rodrigues, A. E. Sep. Sci. Technol. 2008, 43 (8), 1974−2002. (4) Minceva, M.; Rodrigues, A. E. Sep. Sci. Technol. 2003, 38 (7), 1463−1497. (5) Minceva, M.; Rodrigues, A. E. Chem. Eng. Res. Des. 2004, 82 (5), 667−681. (6) Santacesaria, E.; Morbidelli, M.; Danise, P.; Mercenari, M.; Carrà, S. Ind. Eng. Chem. Process Des. Dev. 1982, 21 (3), 440−445. (7) Santacesaria, E.; Morbidelli, M.; Servida, A.; Storti, G.; Carrà, S. Ind. Eng. Chem. Process Des. Dev. 1982, 21 (3), 446−451. (8) Carrà, S.; Santacesaria, E.; Morbidelli, M.; Storti, G.; Gelosa, D. Ind. Eng. Chem. Process Des. Dev. 1982, 21 (3), 451−457. (9) Hsiao, H. C.; Yih, S. M.; Li, M. H. Adsorpt. Sci. Technol. 1989, 6 (2), 64−82. 3501


Langmuir

Article

(39) Zlotea, C.; Phanon, D.; Mazaj, M.; Heurtaux, D.; Guillerm, V.; Serre, C.; Horcajada, P.; Devic, T.; Magnier, E.; Cuevas, F.; Ferey, G.; Llewellyn, P. L.; Latroche, M. Dalton Trans. 2011, 40 (18), 4879− 4881. (40) Moreira, M. A.; Santos, J. C.; Ferreira, A. F. P.; Loureiro, J. M.; Rodrigues, A. E. Ind. Eng. Chem. Res. 2011, 50 (12), 7688−7695. (41) Vermoortele, F.; Maes, M.; Moghadam, P. Z.; Lennox, M. J.; Ragon, F.; Boulhout, M.; Biswas, S.; Laurier, K. G. M.; Beurroies, I.; Denoyel, R.; Roeffaers, M. B. J.; Stock, N.; Duren, T.; Serre, C.; De Vos, D. E. J. Am. Chem. Soc. 2011, 133 (46), 18526−18529.

3502


Toward Understanding the Influence of ... - ACS Publications

Recommend Documents