Density Functional Theory study of the structural and electronic properties of H₃PO₄/ZSM-5

Abstract

In this work, a Density Functional Theory (DFT) study has been carried out to investigate the structural and electronic properties of H₃PO₄/ZSM-5 (extra-framework and framework modification). Cluster models at different T sites (T6, T9, and T12) are suggested. The local structure of the extra-framework modified cluster for H₃PO₄/ZSM-5 shows that there is a hydrogen bond interaction between the hydrogen atoms in H₃PO₄ and the oxygen atoms in the zeolite framework, involving O₄ and H_a. Additionally, the O_zeo always tends to keep in a straight line distribution. Mulliken charge analysis of the cluster models concerned shows that the charge indeed transfers from the oxygen atoms toward P, Al and H atoms. The charge transfer from ZSM-5 to H₃PO₄ could enhance the interaction between the H₃PO₄ and the ZSM-5. Based on the DFT study above, the mutual relationship between the acidic sites on ZSM-5 and phosphoric acid might be more clearly confirmed.

---

1. Introduction

It has long been known that bio-ethanol reacts with HZSM-5 zeolite (MFI) to give a mixture of hydrocarbons which are rich in ethylene, called bio-ethanol to ethylene (B.E.T.E.).^1–5 Zeolites play an important role in catalytic reactions in the chemical industry. Studies involving zeolites are attracting the attention of global researchers.^6–8 A number of works have reported that the introduction of phosphorus acid into HZSM-5 by wet impregnation could provide the catalyst with a much lower selectivity for aromatic byproducts, and a lower reaction temperature.^9–15

Many experiments have confirmed the existence of interactions between H₃PO₄ and the solid acid of HZSM-5. Abubakar et al.⁹ analyzed the structure and mechanism of phosphate-modified HZSM-5 for methanol conversion. It was found that (1) XRD revealed no bulk phases other than the zeolite; (2) ²⁷Al, ²⁹Si, and ¹H MAS (magic-angle spinning)-NMR all showed extensive dealumination of the zeolite framework from an initial SiO₂/Al₂O₃ ratio of 80 to a final value that was typically near 240; and (3) ³¹P MAS-NMR showed that, with the presence of water and methanol, phosphorus was present in the modified zeolite as phosphoric acid. However, through dehydration or calcination, it was present as P₄O₁₀. Ramesh et al.^11,15 studied a P-modified HZSM-5 catalyst in selective ethanol dehydration. In their results, ²⁷Al MAS NMR spectra suggested that the addition of P facilitates the cleavage of the Si–O–Al bond, which leads to partial dealumination. The NH₃-TPD results indicated that the total acidity and the amount of strong acid sites decreases with P loading.

At present, five framework modification models have been proposed through experiments (as shown in Fig. 1). As early as 1986, Lercher et al.¹⁸ proposed a model to describe the interaction between the bridging hydroxyls of ZSM-5 and orthophosphoric acid. Corma et al.¹⁹ summarized the models present in the literature for describing the interaction between HZSM-5 and phosphorous species, for elucidation of the status of the phosphorus species and the hydroxyl groups. Xue et al.²⁰ proposed the mechanism of the interaction between phosphorus and HZSM-5. Lü et al.²¹ used DFT to investigate the five models, and found that among the five proposed models, model B (where the terminal oxygen of the phosphorus species interacts with aluminum) is the most plausible model for describing the phosphorus species grafting on a zeolite framework. However, Lü et al.²¹ didn't fix any atoms in the five models, and the optimized structures could not maintain the ZSM-5 channels. So, the terminal H atoms of the models in this study were fixed in order to maintain the ZSM-5 crystal.

Fig. 1 Models proposed for the interaction of phosphorus with the Brönsted acid sites of HZSM-5.²⁰ (A) proposed by Kaeding et al.¹⁶ and Védrine et al.;¹⁷ (B) proposed by Lercher et al.,¹⁸ (C) proposed by Corma et al.;¹⁹ (E) proposed by Xue et al.²⁰

In this work, a DFT study was carried out in order to understand the structural and electronic properties of H₃PO₄/ZSM-5. The main aims of this work are: (1) to investigate the local structural configuration of the extra-framework and framework modified HZSM-5; (2) to elucidate the electronic properties of the extra-framework and framework modified HZSM-5; (3) to understand the inherent relationship between H₃PO₄ and the acidic sites on the HZSM-5.

2. Computational models and methods

2.1 Cluster selection

ZSM-5 has an MFI framework topology, characterized by a three-dimensional pore system with straight and sinusoidal channels. The pore vacancies are defined by 10 member oxygen-rings that are wide enough (about 5.5 Å) to allow molecules as large as benzene to pass through. As shown in Fig. 2, among the twelve crystallographic distinct T sites, the T4 and T10 sites (marked in green) only appear in the 10 member O-ring which forms the straight channel. Similarly, the T8 and T11 sites (marked in blue) only appear in the 10 member O-ring which forms the sinusoidal channel. Thus, there are eight T sites which are common to both straight and sinusoidal channels.^22,23 Many studies have reported that the T6, T9 and T12 sites (marked in purple) are the most stable positions for the Al atom,^24–26 but some other studies have suggested that the T12 site is the most stable position for the Al atom.^27–30

Fig. 2 The unit structure of ZSM-5 zeolite.

In this work, 8T (H₃SiO)₃Si–O(H)–T(OSiH₃)₃(T=Si, Al) cluster models are applied to investigate the modifications at the T6, T9, and T12 sites, respectively. The terminal Si atoms are saturated with H atoms, and all terminal Si–H bond lengths r(Si–H) are fixed at 1.470 Å along the direction of the Si–O bond, as determined from crystallographic data.³¹ To allow for site relaxation upon Al (P) substitution, only the terminal H atoms are fixed at crystallographic locations. The initial configuration of ZSM-5 is taken from the siliceous ZSM-5 crystal.

2.2 Computational methods

All calculations are performed using the DMol³ program from MS (Materials Studio).^32,33 The double numerical plus polarization (DNP) basis set is used in the calculation to describe the valence orbital of all of the atoms concerned. The numerical basis sets in DMol3 minimize or even eliminate basis set superposition error (BSSE), in contrast to Gaussian basis sets, where BSSE can be a serious problem.^34,35 The DFT study with the DMol³ module is performed with the generalized gradient corrected approximation (GGA) using the BLYP functional.^36,37 The advantage of the BLYP method employed in this study is that it has a correction term for dispersion forces.³⁸ In this study, no symmetry constraints are used for any cluster models. The atom charges are calculated using the approach proposed by Mulliken.³⁹

The relative stability of the H₃PO₄ in each extra-framework site is evaluated by calculating the binding energy (BE):

E_bind = −{E(H₃PO₄/HZSM-5) − E(H₃PO₄) − E(HZSM-5)}

Where E(H₃PO₄/HZSM-5), E(H₃PO₄) and E(HZSM-5) represent the electronic energy for the optimized H₃PO₄/HZSM-5 cluster, the isolated H₃PO₄ and the HZSM-5 cluster, respectively.

The deprotonation energy (DE) is calculated to estimate the acidity of the bridging hydroxyl groups, which is expressed as the total energy difference between the initial (neutral) and the deprotonated (anionic) forms of the HZSM-5 or H₃PO₄/HZSM-5 clusters. The DE value cannot be directly measured experimentally. The difference between the DE values of two different OH groups can be determined from the difference in the binding energies of a probe basic molecule.⁴⁰

3. Results and discussion

3.1 Structural configurations of H₃PO₄/HZSM-5

3.1.1 Extra-framework modifications of H₃PO₄/HZSM-5. Fig. 3 shows the optimized structures for the T6, T9, and T12 sites of extra-framework modified H₃PO₄/HZSM-5 clusters. The main structural features of the optimized clusters are listed in Table 1.

Fig. 3 Optimized structural configurations of extra-framework modified H₃PO₄/HZSM-5 cluster models. (a) T6, (b) T9 and (c) T12.

Table 1 Properties of optimized extra-framework modified H₃PO₄/HZSM-5 cluster models^a

Sites		Bond name and length, Å				Bond angle, deg.
		P–O	O–H	Ozeo–Ha	Al–O	Al–O–Si
a O1–O4 are denoted as the individual oxygen atoms in H3PO4. Oa, Ob, Ozeo are denoted as the individual framework oxygen atoms in HZSM-5.
T6		P–O1 1.597	O1–H1 1.013 O2–H2 0.975	1.132	Al–Oa 1.743	Init.134.569
		P–O2 1.598	O3–H3 0.999 O4–Ha 1.325		Al–Ob 1.747	Opt.133.488
		P–O3 1.590	Oa–H3 1.909 Ob–H1 1.751		Al–Ozeo 1.807
		P–O4 1.521			Al–Od 1.693
T9		P–O1 1.625	O1–H1 0.977 O2–H2 0.975	1.059	Al–Oa 1.759	Ini.133.404
		P–O2 1.607	O3–H3 1.040 O4–Ha 1.471		Al–Ob 1.707	Opt.131.276
		P–O3 1.569	Oa–H3 1.540 Ob–H1 5.712		Al–Ozeo 1.845
		P–O4 1.514			Al–Od 1.709
T12		P–O1 1.599	O1–H1 1.009 O2–H2 0.975	1.104	Al–Oa 1.733	Init.128.060
		P–O2 1.602	O3–H3 1.011 O4–Ha 1.379		Al–Ob 1.738	Opt.126.010
		P–O3 1.587	Oa–H3 1.716 Ob–H1 1.748		Al–Ozeo 1.814
		P–O4 1.521			Al–Od 1.691

---

It can be seen that the proton H_a in the bridging hydroxyl group migrates toward the O₄ atom in the H₃PO₄ for all extra-framework modified cluster models. Before modification, the O_zeo–H_a bond length in the HZSM-5 is about 0.977 Å. After the introduction of H₃PO₄ into ZSM-5, the O_zeo–H_a bond length is extended to 1.059 Å, 1.104 Å and 1.132 Å for different models, while the distance between O₄ atom in H₃PO₄ and H_a atom is 1.325 Å, 1.379 Å and 1.471 Å for different models. This indicates the existence of a hydrogen bond interaction between the O₄ and H_a atoms. This interaction plays a vital role in the thorough dispersion of the H₃PO₄, resulting in a better catalytic performance. Additionally, O₄, H_a and O_zeo are always situated on the same straight line.

The distances between the hydrogen atoms in the H₃PO₄ and the framework oxygen atoms are shown in Table 1. It can be seen that the O_a–H₃ and O_b–H₁ bond distances range from 1.716 to 1.909 Å. This suggests the existence of hydrogen bond interactions between the hydrogen atoms in H₃PO₄ and the framework oxygen atoms in the zeolite. For the T9 site, only the O_a and H₃ atoms show a hydrogen bond interaction. Such results cannot be experimentally obtained since the EXAFS technique can only determine the average bond distance.⁴¹ This information might be very important for exploring the reaction mechanism.

The interaction between the H₃PO₄ and the HZSM-5 will cause some changes in the Al–O bond length and the Al–O–Si bond angle. The Al–O bond lengths are listed in Table 1. For all cluster models, the bond lengths between Al and the O_a, O_b and O_zeo atoms are longer than that of Al–O_d. This can be attributed to the strong hydrogen bond interactions between the hydrogen atoms in H₃PO₄ and the oxygen atoms in the zeolite framework. The Al–O–Si bond angles are also shown in Table 1. Redondo et al.²⁴ reported that larger T–O–T bond angles could represent a stronger acidity (lower proton affinity). The optimized Al–O–Si bond angles for all cluster models are smaller than those for the initial HZSM-5, which might indicate that the acidity of the zeolite decreases after the H₃PO₄ extra-framework is modified. This is consistent with the NH₃-TPD results.^11,15 The sequence of acid strength (from weak to strong) at different T sites is: T12 < T9 < T6.

3.1.2 Framework modification of H₃PO₄/HZSM-5. The initial framework modification structures are from model B. Fig. 4 shows the optimized structures for the T6, T9, and T12 sites of the H₃PO₄/HZSM-5 framework modified clusters. The main structural features of the optimized clusters are listed in Table 2.

Fig. 4 Optimized structural configurations of the framework modified H₃PO₄/HZSM-5 cluster models. (a) T6, (b) T9, (c) T12.

Table 2 Properties of the framework modified H₃PO₄/HZSM-5 cluster models^a

Sites		Bond name and length, Å			Bond angle, deg.
		P–O	O–H	Al–O	P–O–Si
a O1–O4 are denoted as the individual oxygen atoms in H3PO4. Oa, Ob, Ozeo aredenoted as the individual framework oxygen atoms in HZSM-5.
T6		P–O1 1.594 P–O2 1.575	Ob–H1 1.875	Al–Oa 1.550	Init. 134.569
		P–O3 1.471 P–Ozeo 1.851	Oa–H2 1.546	Al–Ob 1.598	Opt 122.097
				Al–Oc 1.602
				Al–Ozeo 1.588
T9		P–O1 1.608 P–O2 1.564	Oa–H1 1.490	Al–Oa 1.606	Init. 134.404
		P–O3 1.475 P–Ozeo 1.822	Od–H2 3.325	Al–Ob 1.608	Opt. 119.292
				Al–Oc 1.612
				Al–Ozeo 1.709
T12		P–O1 1.585 P–O2 1.575	Ob–H1 1.645	Al–Oa 1.727	Init. 128.060
		P–O3 1.470 P–Ozeo 1.846	Oa–H2 1.553	Al–Ob 1.729	Opt.116.809
				Al–Oc 1.683
				Al–Ozeo 1.884

---

As shown in Fig. 4, the optimized structures turn out to be the same as the structure of model A in Fig. 1. The distance between the hydrogen atoms in the H₃PO₄ and the framework oxygen atoms are shown in Table 1. It can be seen that the O_b–H₁ and O_a–H₂ bond distances are from 1.490, 1.645 and 1.875 Å, respectively. This suggests the existence of a hydrogen bond interaction between the hydrogen atoms in the H₃PO₄ and the framework oxygen atoms in the zeolite. For the T9 site, only the O_a and H₁ atoms show a hydrogen bond interaction.

3.2 Electronic properties of H₃PO₄/HZSM-5

The Mulliken population analyses of the P, Al, O atoms (the zeolite framework oxygen atoms and the oxygen atoms in H₃PO₄), and H atoms (the proton hydrogen atom and the hydrogen atoms in H₃PO₄) are listed in Table 3.

Table 3 Mulliken population analyses of the extra-framework modified H₃PO₄/HZSM-5 cluster models

Sites		Atoms and Mulliken charge, |e|
		P	O	H	Al
T6		1.519	O1 - 0.604 O2 - 0.527 O3 - 0.570	H1 0.354H2 0.288	1.350
			O4 - 0.725 Oa - 0.919 Ob - 0.914	H3 0.356 Ha 0.490
T9		1.494	O1 - 0.553 O2 - 0.528 O3 - 0.604	H1 0.287H2 0.286	1.374
			O4 - 0.755 Oa - 0.942 Ob - 0.846	H3 0.403 Ha 0.465
T12		1.508	O1 - 0.598 O2 - 0.529 O3 - 0.590	H1 0.357H2 0.285	1.362
			O4 - 0.744 Oa - 0.926 Ob - 0.919	H3 0.369 Ha 0.470

---

The Mulliken net charges of the framework oxygen atoms and the oxygen atoms in H₃PO₄ are also listed in Table 3. It can be seen that there is a charge transfer from the O atoms to the P, Al and H atoms. The charge transfer from the ZSM-5 to the P and H atoms in the H₃PO₄ could enhance the interaction between the H₃PO₄ and the ZSM-5.

The Mulliken population analyses of the framework modified cluster models are listed in Table 4. A similar conclusion could be obtained from the Mulliken net charges analysis, i.e. that a charge transfer from the O atoms to the P, Al and H atoms occurs.

Table 4 Mulliken population analyses of the framework modified H₃PO₄/HZSM-5 cluster models

Sites		Atoms and Mulliken charge, |e|
		P	O	H	Al
T6		1.547	O1 - 0.562 O2 - 0.610 O3 - 0.600	H1 0.331H2 0.387	1.380
			Ozeo - 0.984 Oa - 0.928 Ob - 0.907
T9		1.495	O1 - 0.548 O2 - 0.586 O3 - 0.586	H1 0.299H2 0.385	1.411
			Ozeo - 0.965 Oa - 0.893 Ob - 0.849
T12		1.523	O1 - 0.580 O2 - 0.597 O3 - 0.603	H1 0.359H2 0.375	1.388
			Ozeo - 0.968 Oa - 0.924 Ob - 0.921

---

3.3 Deprotonation energy (DE) and binding energy (BE) of H₃PO₄/HZSM-5

The deprotonation energies and binding energies for each T site are shown in Table 5, where ΔE_D = E_Al–O–Si − E_Al–OH–Si.

Table 5 Deprotonation energies and binding energies for each T site^a

Sites		Deprotonation energies, kcal mol^−1			BEext, kcal mol^−1
		ΔED-un	ΔED-ext	ΔED-fra
a ΔED-un, ΔED-ext and ΔED-fra represent the deprotonation energies of unmodified, extra-framework modified, and framework modified zeolite, respectively. BEext represents the binding energy of extra-framework modified zeolites.
T6		293.5662	295.6677	302.4867	21.0946
T9		294.7945	295.7426	302.6203	21.6091
T12		298.1843	298.4150	304.1262	25.6921

---

A higher BE value indicates a higher stability of the configuration. By comparing the BE value at different T sites, it can be found that the most stable extra-framework modified cluster model is that at the T12 site, for which the BE value is about 4 kcal mol⁻¹, higher than for the other models.

It is well known that the acidity of the catalyst is important for ethanol dehydration.⁴² The catalytic performance of the H₃PO₄ is highly dependent on the amount of acid in the ZSM-5. The deprotonation energies (DE) could be used as an effective method to evaluate the acidity of the zeolite.^43,44,45 The lower the DE value of an OH group is, the stronger its Brönsted acidity is. As shown in Table 5, the ΔE_D-ext and ΔE_D-fra values are higher than the ΔE_D-un value, which suggests that these modifications could decrease the Brönsted acid strength of ZSM-5. The obtained results for DE are rather consistent with the T–O–T bond angles discussed above. For the extra-framework cluster models, an H₃PO₄H⁺ group is formed when the proton migrates toward the O₄ atom in the H₃PO₄. Therefore, these cluster models can be considered as a proton H substituted by H₃PO₄H⁺.

3.4 Probe molecules

The acidity information for a zeolite can be obtained through the interaction of a basic probe molecule with the acid sites of the zeolite. In this study, DFT was used to calculate the interaction of the probe molecules and ZSM-5, to obtain the acidity information for ZSM-5. The stable adsorption configuration of the probe molecule NH₃ adsorbed on the T₁₂ site of ZSM-5 is shown in Fig. 5. In Fig. 5, blue represents N atoms, yellow represents Si atoms, red represents O atoms, deep purple represents Al atoms, light purple represents P atoms and white represents H atoms.

Fig. 5 The structures of NH₃ adsorbed on the T₁₂ site of ZSM-5 zeolite. (a) Un-before optimization, (b) Un-after optimization (c), Ex-before optimization (d), Ex-after optimization (e), Fr-before optimization and (f) Fr-after optimization.

As can be seen from Table 6, the ΔE_Int of the unmodified ZSM-5 is the largest. The ΔE_Int of the extra-framework modified ZSM-5 is smaller, and the ΔE_Int of the framework modified ZSM-5 is the smallest. Thus, the Bronsted acidity of ZSM-5 in descending order is Un > Ex > Fr (that is, unmodified > extra-framework modified > framework modified), which is in agreement with our previously discussed results for deprotonation energies, and the experimental results of Tynjälä et al.⁴⁶

Table 6 The interaction energies of NH₃ with the zeolite

Model	ET (Ha)	ENH3 (Ha)	ET-NH3 (Ha)	ΔEInt (kcal mol^−1)
Un	−2807.7385	−56.5578	−2864.3314	21.98
Ex	−3452.0915	−56.5578	−3508.6835	21.48
Fr	−3375.5873	−56.5578	−3432.1783	20.81

---

3.5 Ethanol adsorption

The stable configuration models of ethanol adsorption on the zeolite before and after P modification are shown in Fig. 6. The structural geometry parameters and the adsorption energy of the zeolite model when ethanol molecules are adsorbed on the zeolite before and after P modification are shown in Table 7. The adsorption energy of ethanol can be found with the following formula:

E_a = E_T-C2H5OH − E_T – E_C2H5OH

Where E_a represents the adsorption energy of ethanol in kcal mol⁻¹; E_T-C2H5OH represents the total energy of the system after ethanol is adsorbed on the catalyst surface in Ha; E_T represents the total energy before adsorption in Ha and E_C2H5OH represents the total energy of ethanol before adsorption in Ha;

1Ha = 627.51 kcal mol⁻¹.

Fig. 6 The structures of ethanol adsorbed on the P modified ZSM-5.

Table 7 The adsorption energies of ethanol adsorbed on the P modified ZSM-5

Model	Ozeo–Ha (Å)	O–Ha (Å)	ET (Ha)	EC2H5OH (Ha)	ET-C2H5OH (Ha)	Ea (kcal mol^−1)
Un	1.051	1.510	−2807.7426	−155.0447	−2962.8152	−17.50
Ex	1.027	1.593	−3452.0915	−155.0447	−3607.1607	−15.44
Fr	1.044	1.519	−3375.5873	−155.0447	−3530.6532	−12.33

---

From Table 7, the adsorption energy of ethanol molecules before and after modification are −17.50 kcal mol⁻¹, −15.44 kcal mol⁻¹ and −12.33 kcal mol⁻¹, respectively.

It can be seen that the ethanol always tends to form a hydrogen bond with the H₂ atom in the H₃PO₄, but not with the original H_a atom in the ZSM-5. Therefore it could be speculated that after the introduction of phosphorus acid into the HZSM-5, the Brønsted acid proton could be considered as the H₂ atom in H₃PO₄. This could explain the high stability of H₃PO₄/HZSM-5.

4. Conclusions

In this work, a DFT study confirms the hydrogen bond interactions between hydrogen atoms in H₃PO₄ and the framework oxygen atoms in ZSM-5, with the O₄, H_a and O_zeo atoms always distributed in the same straight line. This configuration is the most stable structure for extra-framework modified HZSM-5. For framework modified models, the optimized structure is consistent with model A. The DFT study also confirms that a close relationship exists between the acid sites of the ZSM-5 and the H₃PO₄. It is suggested that the acid strength of the ZSM-5 is weakened after H₃PO₄ modification, which is consistent with the NH₃-TPD results. The hydrogen bond interaction between the H₃PO₄ and the framework oxygen atoms in the ZSM-5 could result in the formation of an H₃PO₄H⁺ group, and thus the acidic sites on the ZSM-5 could keep the highly dispersed state of the H₃PO₄. It is speculated that after the introduction of the H₃PO₄ into the HZSM-5, the Brønsted acid proton could be considered as the H₂ atom in the H₃PO₄.

References

1. R. Le Van Mao, T. M. Nguyen and G. P. McLaughlin, Appl. Catal., 1989, 48, 265–277.
2. W. R. Moser, R. W. Thompson, C.-C. Chiang and H. Tong, J. Catal., 1989, 117, 19–32.
3. T. M. Nguyen and R. Le Van Mao, Appl. Catal., 1990, 58, 119–129.
4. I. Takahara, M. Saito, M. Inaba and K. Murata, Catal. Lett., 2005, 105, 249–252.
5. I. Takahara, M. Saito, H. Matsuhashi, M. Inaba and K. Murata, Catal. Lett., 2007, 113, 82–85.
6. N. Wang, M. Zhang and Y. Yu, RSC Adv., 2014, 4, 4324–4329.
7. X. Q. Zhang, R. A. van Santen and A. P. J. Jansen, Phys. Chem. Chem. Phys., 2012, 14, 11969–11973.
8. X. Q. Zhang, T. T. trinh, R. A. van Santen and A. P. J. Jansen, J. Am. Chem. Soc., 2011, 133, 6613–6625.
9. S. M. Abubakar, D. M. Marcus, J. C. Lee, J. O. Ehresmann, C. Y. Chen, P. W. Kletnieks, D. R. Guenther, M. J. Hayman, M. Pavlova, J. B. Nicholas and J. F. Haw, Langmuir, 2006, 22, 4846–4852.
10. D. S. Zhang, R. Wang and X. X. Yang, Catal. Lett., 2008, 124, 384–391.
11. K. Ramesh, L. M. Hui, Y. F. Han and A. Borgna, Catal. Commun., 2009, 10, 567–571.
12. N. N. Zhan, Y. Hu, H. Li, D. H. Yu, Y. W. Han and H. Huang, Catal. Commun., 2010, 11, 633–637.
13. Z. X. Song, A. Takahashi, I. Nakamura and T. Fujitani, Appl. Catal., A, 2010, 384, 201–205.
14. N. H. Xue, R. Olindo and J. A. Lercher, J. Phys. Chem. C, 2010, 114, 15763–15770.
15. K. Ramesh, C. Jie, Y. F. Han and A. Borgna, Ind. Eng. Chem. Res., 2010, 49, 4080–4090.
16. W. W. Kaeding and S. A. Butter, J. Catal., 1980, 61, 155–164.
17. J. C. Védrine, A. Auroux, P. Dejaifve, V. Ducarme, H. Hoser and S. Zhou, J. Catal., 1982, 73, 147–160.
18. J. A. Lercher and G. Rumplmayr, Appl. Catal., 1986, 25, 215–222.
19. T. Blasco, A. Corma and J. Martínez-Triguero, J. Catal., 2006, 237, 267–277.
20. N. Xue, X. Chen, L. Nie, X. Guo, W. Ding, Y. Chen, M. Gu and Z. Xie, J. Catal., 2007, 248, 20–28.
21. R. Lü, Z. Cao and S. Wang, J. Mol. Struct.: THEOCHEM, 2008, 865, 1–7.
22. A. Chatterjee and R. Vetrivel, J. Chem. Soc., Faraday Trans., 1995, 91, 4313–4319.
23. A. Chatterjee and R. Vetrivel, J. Mol. Catal. A: Chem., 1996, 106, 75–81.
24. A. Redondo and P. J. Hay, J. Phys. Chem., 1993, 97, 11754–11761.
25. A. Chatterjee and A. K. Chandra, J. Mol. Catal. A: Chem., 1997, 119, 51–56.
26. M. S. Stave and J. B. Nicholas, J. Phys. Chem., 1995, 99, 15046–15061.
27. J. G. Fripiat, F. Berger-André, J.-M. André and E. G. Derouanc, Zeolites, 1983, 3, 306–310.
28. E. G. Derouane and J. G. Fripiat, Zeolites, 1985, 5, 165–172.
29. S. R. Lonsinger, A. K. Chakraborty, D. N. Theodorou and A. T. Bell, Catal. Lett., 1991, 11, 209–217.
30. A. Chatterjee and R. Vetrivel, Microporous Mater., 1994, 3, 211–218.
31. H. van Koningsveld, H. van Bekkum and J. C. Jansen, Acta Crystallogr., Sect. B: Struct. Sci., 1987, 43, 127–132.
32. B. Delley, J. Chem. Phys., 1990, 92, 508–517.
33. B. Delley, J. Chem. Phys., 2000, 113, 7756–7764.
34. M. Elanany, M. Koyama, M. Kubo, P. Selvam and A. Miyamoto, Microporous Mesoporous Mater., 2004, 71, 51–56.
35. B. Kalita and R. C. Deka, Eur. Phys. J. D, 2009, 53, 51–58.
36. A. D. Becke, Phys. Rev. A, 1988, 38, 3098–3100.
37. C. Lee, W. Yang and R. G. Parr, Phys. Rev. B: Condens. Matter Mater. Phys., 1988, 37, 785–789.
38. A. F. Jalbout, A. J. Hameed, F. I. Jimenez and M. Ibrahim, J. Organomet. Chem., 2008, 693, 216–220.
39. R. S. Mulliken, J. Chem. Phys., 1955, 23, 1833–1840.
40. G. N. Vayssilov and N. Rösch, J. Phys. Chem. B, 2001, 105, 4277–4284.
41. B. Wichtelová, Z. Sobalík and J. Dedecek, Appl. Catal., B, 2003, 41, 97–114.
42. K. M. Abd El-Salaam and E. A. Hassan, Surf. Technol., 1982, 16, 121–128.
43. N. Gonzales, A. Chakraborty and A. Bell, Catal. Lett., 1998, 50, 135–139.
44. M. Sierka, U. Eichler, J. Datka and J. Sauer, J. Phys. Chem. B, 1998, 102, 6397–6404.
45. N. O. Gonzales, A. T. Bell and A. K. Chakraborty, J. Phys. Chem. B, 1997, 101, 10058–10064.
46. P. Tynjala and T. T. Pakkanen, Microporous Mesoporous Mater., 1998, 20, 363–369.

---