Hierarchical ZSM-5 zeolite catalysts for the selective oxidation of benzene

(1)

Hierarchical ZSM-5 zeolite catalysts for the selective oxidation

of benzene

Citation for published version (APA):

Koekkoek, A. J. J. (2011). Hierarchical ZSM-5 zeolite catalysts for the selective oxidation of benzene. Technische Universiteit Eindhoven. https://doi.org/10.6100/IR717554

DOI:

10.6100/IR717554

Document status and date: Published: 01/01/2011 Document Version:

Publisher’s PDF, also known as Version of Record (includes final page, issue and volume numbers) Please check the document version of this publication:

• A submitted manuscript is the version of the article upon submission and before peer-review. There can be important differences between the submitted version and the official published version of record. People interested in the research are advised to contact the author for the final version of the publication, or visit the DOI to the publisher's website.

• The final author version and the galley proof are versions of the publication after peer review.

• The final published version features the final layout of the paper including the volume, issue and page numbers.

Link to publication

General rights

Copyright and moral rights for the publications made accessible in the public portal are retained by the authors and/or other copyright owners and it is a condition of accessing publications that users recognise and abide by the legal requirements associated with these rights. • Users may download and print one copy of any publication from the public portal for the purpose of private study or research. • You may not further distribute the material or use it for any profit-making activity or commercial gain

• You may freely distribute the URL identifying the publication in the public portal.

If the publication is distributed under the terms of Article 25fa of the Dutch Copyright Act, indicated by the “Taverne” license above, please follow below link for the End User Agreement:

www.tue.nl/taverne

Take down policy

If you believe that this document breaches copyright please contact us at:

openaccess@tue.nl

providing details and we will investigate your claim.

(2)

(3)

(4)

Hierarchical ZSM-5 zeolite catalysts for the selective oxidation of benzene

PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Technische Universiteit Eindhoven, op gezag van de rector magnificus, prof.dr.ir. C.J. van Duijn, voor een

commissie aangewezen door het College voor Promoties in het openbaar te verdedigen op dinsdag 1 november 2011 om 16.00 uur

door

Arie Johannes Job Koekkoek

(5)

Dit proefschrift is goedgekeurd door de promotoren:

prof.dr.ir. E.J.M. Hensen en

prof.dr. R.A. van Santen

A catalogue record is available from the Eindhoven University of Technology Library

ISBN: 978-90-386-2779-3

This work was supported by the Netherlands Organisation for Scientific Research (NWO).

(6)

(7)

(8)

Chapter 1

Introduction

1.1 Zeolites

!

Zeolites are crystalline aluminosilicate minerals consisting of regular frameworks of aluminiumoxide and siliconoxide tetrahedra combined with group IA and IIA elements (e.g. sodium, potassium, calcium and magnesium). A conventional way of representing zeolites is by using M2/nO . Al2O3 . γSiO2 . wH2O

with γ an integer between 2 and 200, w the water contained in the framework and M the balancing cation with valence n (e.g. Na+_{, K}+_{, Mg}2+_{) [1]. Zeolites occur in}

nature as crystalline minerals. The history of zeolite synthesis starts with a laboratory preparation report of levynite (LEV) by St. Claire Deville [2]. Although this report dates as early as 1862, the synthesis of zeolites as we know it today starts with the work of Barrer and Milton in the late 1940’s and early 1950’s [3,4]. In the following decade several new zeolite structures were discovered, but the major breakthrough came with the use of organic ammonium cations in the synthesis. The traditional synthesis mixtures consisted of purely inorganic components. By use of organic cations the first high-silica zeolite, i.e. Beta (BEA) [5], was discovered in 1967 followed in 1972 by that of the by now very important ZSM-5 (MFI) zeolite [6].

Figure 1.1: Structural image of the framework of Levynite, Beta and MFI type zeolites with the oxygen atoms (red) on the corners and the silicon atoms (yellow) as the bridges.

(11)

From there on numerous types of new zeolites and framework type families have been discovered. Currently, 179 different zeolite structures are known with the number rising each year. Nowadays the worldwide consumption of synthetic zeolites is estimated about 1.7-2 million metric tons, where the consumption of natural zeolites is around 2.5 million metric tons [7-9]. The major applications of zeolites comprise detergents, adsorbents, desiccants and catalysis.

1.2 Acidity of zeolites

!

The presence of [AlO4] tetrahedra in the [SiO4] framework causes a net

negative charge. This charge is usually compensated by loosely fixed cations (e.g. Na+_{, K}+_{, Ca}2+_{and Mg}2+_{). When a proton is used instead, a strong Brønsted acidic}

material is created. The high acidity of these sites in zeolites can be related to the rigid structure of the crystalline framework. This prohibits yielding of the local structure upon interaction of molecules with the acidic proton. Next to these strong Brønsted acidic sites, zeolites can contain large amounts of weakly acidic silanols on defect sites and on the edges of the crystalline framework.

Figure 1.2: Schematic representation of a strong Brønsted acidic site (left) and a weakly acidic silanol site (right) in a silicate.

The total acidity of a zeolite therefore has to be regarded as the product of the total number of acidic sites and their intrinsic acidity. The intrinsic acidity of the sites is determined by the local structure around the acidic site. A tetrahedral aluminium in the next nearest neighbor tetrahedron decreases the intrinsic acidity of the acidic site [10]. Accordingly, the intrinsic acidity of the acid site increases by dilution of the framework Al atoms. This is very well demonstrated in the case of zeolite Y, commonly used as cracking catalyst. In general a conventional zeolite synthesis can be used to obtain a sodium Y zeolite with a Si/Al ratio of 2.5. Ion exchange

1 Introduction

(12)

with ammonium cations, followed by calcination yields the zeolite in its acidic form HY. As the amount of alumina in this material is very high, the Brønsted acidity is limited due to next nearest neighbor interactions. By steam treatment aluminum can be extracted from the framework effectively increasing the Si/Al ratio, and it was shown that an increase of the Si/Al ratio to 4.8 yielded a zeolite that is more active than the parent due to the decrease of the amount next nearest neighbor alumina [11]. When more aluminum is extracted the activity of the catalyst decreases again as the effect of the loss of acidic sites is larger than the increase of acidity of the sites.

1.3 Zeolites in catalysis

!

With an estimated 55% on value base of the total yearly world consumption [1], catalysis is one of the major applications for zeolites. Further noteworthy is the fact that of the 179 known zeolite structures only 18 are used in commercial processes. By far the largest application of zeolites as catalysts is in the field of petrochemical industry. The major application of zeolites in refineries is fluid catalytic cracking (FCC), where generally silica-enriched forms of zeolite Y (FAU) are used [12]. Other catalytic reactions involving zeolites include reforming, hydrocracking and chemical synthesis. One of the major challenges in the FCC process covers the cracking of heavy molecules. Typical bottom end fraction molecules have an average carbon number of 25-35 giving them a dynamic molecular size of 12-20 Å. Using the simple model proposed by Spry and Sawer [13], it can be estimated that the pore size should be 10-20 times larger than the molecules, meaning that a material with pores in the mesopore range (2-50 nm) is needed for optimal diffusion. This implies that larger molecules cannot enter the zeolite pores, although they might be cracked on the external surface of the zeolite. Nowadays the FCC catalysts consist of a mixture of components including Y and ZSM-5 zeolites combined in an amorphous silica-alumina matrix. The advantage of the use of amorphous silica-alumina is the presence of meso- and macroporosity. Although this improves the cracking efficiency of the catalyst, inevitably the formation of coke increases due to the lack of shape selectivity.

Introduction 1

(13)

Figure 1.3 Schematic drawing of the pore architecture of a modern composite FCC catalyst [14].

1.4 Ordered mesoporous silicas

!

Due to the need to process large molecules there is an increasing demand for novel porous materials that can overcome the diffusional limitations and pore size constraints of conventional zeolites. This has spurred the synthesis of large pore zeolites and related large pore silicas. The first report of the synthesis of an ordered mesoporous material dates from 1969, but due to the lack of decent characterization the remarkable features of this material were not recognized at the time [15,16]. In 1992 researchers of Mobil Oil Company reported the synthesis of a silica with a long range ordered pore system in the mesopore range [17-19]. The material, which was named Mobil Composition of Matter No. 41 (MCM-41), is an amorphous silicate having a hexagonal P6mm arrangement of the mesopores. Slight modifications to the synthesis procedure quickly led to the discovery of the cubic MCM-48 and lamellar MCM-50 mesoporous silicates. However, despite their favorable pore diameter in the mesopore region, these silicas lack the acidity and hydrothermal stability of crystalline zeolites because the pore walls are amorphous.

Yanagisawa and coworkers reported a less versatile approach to the synthesis of hexagonally ordered mesoporous silicates and aluminosilicates around the same time [20,21]. They used intercalation of surfactants in the layered silicate Kanemite. Via wrapping of the sheets and formation of a hexagonal phase, ordered

1 Introduction

(14)

mesoporous silicates and aluminosilicates were obtained. They named their materials Folded Sheet Materials-n (FSM-n) were n denotes the number of carbon atoms in the surfactant used.

Figure 1.3: Transmission Electron micrographs of the MCM-41 (a,b) and SBA-15 (c,d) ordered mesoporous silicas.

In 1998 Stucky and coworkers developed a new synthesis route to form a hexagonally ordered silicate with the P6mm structure [22]. Their material named Santa Barbara No. 15 (SBA-15) has a tunable pore size in the 4-30 nm range and a thicker pore wall when compared to its MCM-n counterparts. The synthesis of SBA-15 occurs under acidic conditions, whereas MCM-type silicas materials are made under alkaline conditions. As a result of the thicker pore walls, the material possesses increased hydrothermal stability, and due to the use of a block copolymer template containing moieties with different hydrophobicity the walls of the SBA-15 are pinched with micropores interconnecting the larger mesopores.

Introduction 1

(15)

1.5 Synthesis of ordered porous silicates

1.5.1 Synthesis of zeolites

! In the past decades extensive studies have been made on the formation mechanism of zeolites. Despite considerable progress, the exact details underlying the formation of zeolites on the atomic scale remain elusive due to the complexity involved in the formation of the zeolite nuclei and their growth [23] As known from general crystal growth theory the formation of a zeolite depends on parameters as temperature, aging and seeding the process. Besides, the formation depends on the exact composition of the synthesis gel including pH, water content, Si/Al ratio, silica and alumina source, type of template, template concentration and ionic strength. As a result of this complexity there is still much debate about the exact mechanism of the crystallization of zeolites. Nevertheless, the synthesis of all zeolites has several steps in common [24]: a mixture is made of the structural components of the zeolite in an appropriate form (Si, Al, P etc.) in a aqueous medium, typically under basic conditions containing a structure directing agent (SDA). This heterogeneous partially reacted mixture is commonly referred to as the primary amorphous phase and contains all the building elements of the zeolite [24,25]. Note that this primary phase can be colloidal and thus invisible to the naked eye as is the case in the so-called clear solution synthesis [26]. Heating of the mixture under autogenic pressure in an autoclave and aging induces the formation of a secondary amorphous phase. This phase is in pseudo equilibrium with the solution and possesses some short-range order [27,28]. After a certain induction period the formation of the nuclei takes place, from which the zeolite will grow at the expense of the amorphous phase.

1.5.2 Synthesis of ordered mesoporous silicates

! The synthesis of the MCM-41 type silicas was a considerable breakthrough in the field of porous silica materials. By variation of the synthesis conditions and the surfactant types numerous different ordered mesoporous silicas have been made via the cooperative assembly mechanism. Table 1.1 provides an overview of the most common pathways proposed for the synthesis of ordered mesoporous silicates. The synthesis mechanism of most ordered mesoporous silicates proceeds via the so-called cooperative pathway mechanism [29]. During the synthesis there

1 Introduction

(16)

occurs electrostatic interaction between inorganic silicate species in solution (I-₎

and the surfactant used as the mesoporogen (S+_^^^^_{). For the MCM-41 and}

MCM-48 this is a simple electrostatic interaction between the positively charged ammonium head group of the surfactant. For other mesoporous silicate this interaction can be slightly more complicated, especially when the synthesis proceeds under acidic conditions. As the silicates are positively charged at lower pH (I+_{), there can be no direct interaction between the silicate species in solution}

and a positively charged surfactant. Stucky and coworkers proposed the involvement of the anion in the counter ion mediated mechanism: a negatively charged ion (X-_{) acts as a bridge between the positively charged surfactant and}

silicate species. A further extension of this proposal led to a whole series of cooperative assembly mechanisms including the use of negatively charged surfactants in a I+_S-_{or I}-_M+_S+_{type mechanism [30-32]. Typical for all these}

mechanisms is the formation of the surfactant micelle early during the synthesis. Formation of the ordered mesostructure then follows assisted by the interaction of the micelle with the silicate species in solution.

Table 1.1: Formation pathways for the synthesis of mesoporous materials1_.

Template Interaction Synthesis

conditions Examples Ionic surfacant I-_S+_^^^^^ I+_X-_S+_^^^^^ I-_X+_S-_^^^^^ Basic Acidic Basic MCM-41, MCM-48 SBA-1, SBA-3 [31] Metal oxides [37] Non-ionic surfactant I0_S0_^^^^^ _Acidic _{SBA-15, MSU [38]}

Covalent bonding I-S^^^^^ - Nb-TMS [39]

Nano casting - - CMK-n [40]

1_{Table adapted from reference [41]}! 1_{Table adapted from reference [41]}! 1_{Table adapted from reference [41]}! 1_{Table adapted from reference [41]}!

The easiest way to control the size of the mesopores is by varying the length of the alkyl chain of the surfactant. As the silicate wall form around the micelle increasing (or decreasing) the size of the micelle directly leads to an increase of the pore size.

Introduction 1

(17)

Better control over the pore size and fine-tuning can be achieved by using a mixture of surfactants with short and long alkyl chains [33]. As the surfactant controls the structure of the mesopores as well, it is not always possible to change the size of the alkyl chain to the desirable length. In this case the use of swelling agents as n- alkanes [34], 1,3,5-trimethylbenzene (mesitylene) [35] or fatty acids [36] can provide a solution.

1.5.3 Synthesis of hierarchical porous silicates

! Ordered mesoporous silicates may have their pore size in the desirable region for several catalytic applications. However, due to their amorphous silicate network they lack the stability and acidity usually encountered in zeolites. Therefore materials, which combine the chemistry and stability of zeolites with the large pores of mesoporous silicas, are of current interest. Synthesis of zeolites having both micro- and mesopores is not easy[42]. Usually zeolites are synthesized at much higher temperatures than their amorphous mesoporous counterparts. As a consequence, the simple approach of addition of mesoporous surfactant template to a zeolite synthesis gel does not work, as the interaction and stability of the growing zeolite with the mesopore template is too weak to form a mesophase. If the temperature of such a mixture is lowered on the other hand, only an amorphous silica will form. Even after fine-tuning such synthesis one at best obtains intimate mixtures of mesoporous silica and zeolite [43].

To overcome these challenges to synthesize zeolite combining micro- and mesoporosity (hierarchical zeolites) several methods have been adopted in the last decade [44,45]. The best known and very widely employed method is steam calcination of faujasite zeolite, which forms the basis of modern refinery cracking catalysts. The principle reason to steam calcine these materials is to increase the framework Si/Al ratio so as to bring about high intrinsic acidity of the protons. An additional benefit is that mesopores are created, which improve accessibility and limit mass transport of reactants and products. Despite the wide experience in this field, it is clear that the zeolite domain size of these steam calcined zeolites is relatively large [46]. Another related and relatively simply and cheap top-down approach is to etch pores in a zeolite crystal by desilication [47,48]. Part of the silicate framework is removed by treatment with hydroxide as the etchant. Although the method works remarkably well for several zeolite structures

1 Introduction

(18)

including MFI, BEA and MTW, its effectiveness strongly depends on the Si/Al of the framework. When this ratio is too high (highly siliceous zeolite) the material tends to dissolve. When the ratio is too low, on the other hand, the framework is stabilized by the alumina preventing the desilication. This limits the possibility to desilicate the important faujasite zeolites. On the other hand, Van Donk et al. [49] have shown that a properly dealuminated faujasite can be modified in such way by desilication that trimodal porosity is obtained.

The easiest bottom up approach to synthesize hierarchical zeolites is the use of hard templates. Here a carbon material, very often in the form of beads or rods with a diameter similar to the desired mesopore size, is used as mesoporogen [50]. The carbon black material is impregnated with a zeolite precursor sol, followed by steaming of the sol under autogenic pressure in an autoclave thus inducing crystallization. Removal of the carbon template by calcination now yields a highly mesoporous zeolite. This method is versatile, gives a good control over the mesopore size and is not limited to any Si/Al ratio. However it can be difficult to fully remove the carbon template without damaging the zeolite framework.

Several years ago, several novel routes utilizing a direct soft templating mechanism were reported based on the use of a mesoporogen that covalently binds to the zeolite framework. The groups of Ryoo and Pinnavia solved this problem independently in a quite similar way [51,52]. The increased interaction of the organic mesoporogen was achieved by the introduction of a covalent linkage between the mesoporogen and the growing silicate framework. To this end, the mesoporogen was functionalized by orthosilicate functionalities. While Xiao used polymers with numerous orthosilicate groups, Ryoo adapted the conventional surfactant used for the synthesis of the MCM type mesoporous silicates with an trimethoxysilyl functionality. Both approaches resulted in the formation of zeolites having substantial mesoporosity, although the actual templating mechanism of these organosilane templates remains unclear.

Introduction 1

(19)

1.6 Scope of the thesis

Zeolites are widely used in as catalysts, especially in oil refining and the petrochemical industries. Nowadays the cracking of heavy oil feeds as well as the processing of larger (bio)molecules demands for improved catalysts, which can overcome the small pore size and imminent diffusion limitations of conventional zeolites. Ordered mesoporous materials often cannot replace zeolites as they are amorphous and do not exhibit the required Brønsted acidity. Besides, zeolites can be functionalized with transition metals to endow them with sites of broader applications. One such example is the use of iron, whose combination with ZSM-5 makes promising catalysts for the selective oxidation of benzene to phenol [53]. A challenge in this reaction is the very fast deactivation of the catalyst due to coking. The aim of this thesis is to investigate in detail the catalytic properties of hierarchical zeolites, especially the HZSM-5 and Fe/ZSM-5. To this end, a number of methods are explored and the texture and surface properties of the resulting zeolites are characterized in detail.

Chapter 2 of this thesis validates that the Brønsted acidity of ordered mesoporous Al/SBA-15 silicas is much lower than that of zeolites in contrast to many literature claims that Al/SBA-15 is highly acidic and can be used to replace zeolites. Instead, these materials share large resemblance in their catalytic behavior and aluminium speciation with amorphous silica-aluminas. This stresses the need for crystalline materials for acid catalysis. Therefore, the remainder of the thesis focuses on the synthesis, characterization and catalytic reactivity of hierarchical zeolites. Chapter 3 compares several routes to obtain hierarchical Fe/ZSM-5 zeolites, viz. carbon templating, desilication and organosilane templating, to be applied in the decomposition of nitrous oxide and the selective oxidation of benzene to phenol. Compared to a Fe/Al-SBA-15 catalyst [54], these catalysts are much more active because of their crystallinity and higher active site density. The smaller the zeolite domain size the higher catalyst stability. The most active catalysts are obtained by organosilane templating. In Chapter 4 and 5 the synthesis mechanism of octadecyl-dimethyl-(3-trimethoxysilyl-propyl)-ammonium chloride templated HZSM-5 and Fe/ZSM-5 is investigated in great detail. Chapter 4 reports the physicochemical characterization of these samples as a function of the hydrothermal crystallization time, while Chapter 5 discusses the catalytic activities for n-heptane hydroconversion and benzene oxidation. Chapter 6 explores the possibility for dry gel conversion of an amorphous precursor material prepared by

octadecyl-1 Introduction

(20)

dimethyl-(3-trimethoxysilyl-propyl)-ammonium chloride templating to which tetrapropylammonium is added as the structure directing agent. Finally, Chapter 7 of this thesis explores the possibilities to decrease the zeolite domain size further in one or in three dimensions.

Introduction 1

(21)

References

[1] E.M. Flanigen, R.W. Broach, S.T. Wilson, Zeolites in industrial separation and catalysis, 2010, 1-26.

[2] H. de St Claire Deville, Comp. Rend. Seances Acad. Sci., 1862, 54, 324. [3] R.M. Barrer, J. Chem. Soc., 1948, 127-132.

[4] R.M. Barrer, L. Hinds, E.A. White, J. Chem. Soc., 1953, 1466-1475. [5] R.L. Wadlinger, G.T. Kerr, E.J. Rosinski, US Patent No. 3.308.069, 1967. [6] R.J. Agauer, G.R. Landolt, US Patent No. 3.702.886, 1972.

[7] B. Yilmaz, U. M Müller, Top. Catal., 2009, 52, 888-895.

[8] J. Cejka, H. van Bekkum, A. Corma, F. Schüth, Introduction to zeolite science and practice, 3rd revised edn. Elsevier, 2007, 1-12.

[9] G. Bellussi, Stud. Surf. Sci. Catal., 2004, 154 A, 53-56.

[10] L.A. Pine, P.J. Maher, W.K. Wachter, J. Catal., 1984, 85, 466-476.

[11] Q.L. Wang, G. Gianetto, M. Torrealba, G. Perot, C. Kappenstein, M. Guisnet, J. Catal., 1991, 130, 459-470.

[12] W. Vermeiren, J.P. Gilson, Top. Catal., 2009, 52, 1131-1161.

[13] J.C. Spry, W.H. Sawyer, AIChE 68th annual meeting, Paper, (1975), 16-20. [14] P. O’Connor, A.P. Humphries, Accessibility of functional sites in FCC, 1993,

Prepr. – Am. Chem. Soc. Div. Pet. Chem., 38, 598-603.

[15] V. Chiola, J.E. Ritsko, C.D. Vanderpool, US Patent No. 3.556.725, 1971. [16] F. Di Renzo, H. Cambon, R. Dutartre, Microp. Mater., 1997, 10, 283-286.

[17] J.S. Beck, C.T.W. Chu, I.D. Johnson, C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.W. Vartuli, WO/Patent 91/11390, 1991.

[18] C.T. Kresge, M.E. Leonowicz, W.J. Roth, J.C. Vertuli, J.S. Beck, Nature, 1992, 359, 710-712.

[19] J.S. Beck, J.C. Vertuli, W.J. Roth, M.E. Leonowicz, C.T. Kresge, K.D. Smitt, C.T.W. Chu, D.H. Olson, E.W. Sheppard, S.B. McCullen, J.B. Higgins, J.L. Schlenker, J. Am. Chem. Soc., 1992, 114, 10834.

[20] T. Yanagisawa, T. Shimizu, K. Kuroda, C. Kato, Bull. Chem. Soc. Jpn., 1990, 63, 988-992.

[21] S. Inagaki, Y. Fukoshima, K. Kuroda, J. Chem. Soc. Chem. Commun., 1993, 680-682.

[22] D. Zhao, J. Feng, Q. Huo, N. Melosh, G.H. Fredrickson, B.F. Chmelka, G.D. Stucky, Science, 1998, 279, 548-552.

[23] R.M. Barrer, Hydrothermal Chemistry of Zeolites, Academic Press, 1982, 360.

1 Introduction

(22)

[24] C.S. Cundy, P.A. Cox, Microp. Mesop. Mater., 2005, 82, 1-78.

[25] M.A. Nicolle, F. Di Renzo, F. Fajula, P. Espiau, T. Des Courieres, Proceedings of the 9th international zeolite conference, 1993, 313-320.

[26] S. Mintova, N.H. Olson, V. Valtchev, T. Bein, Science, 1999, 283, 958-960.

[27] C. Kosanowic, S. Bosnar, B. Subotic, V. Svetlicic, T. Misic, G. Drazic, K. Havancsak, Microp. Mesop. Mater., 2008, 110, 177-185.

[28] T. Wakihara, S. Kohara, S. Sankar, S. Saito, M. Sanchez-Sanchez, A.R. Overweg, W. Fan, M. Ogura, T. Okubo, Phys. Chem. Chem. Phys., 2006, 8, 224-227. [29] A. Monnier, F. Schüth, Q. Huo, D. Kumar, D. Margolese, R. S. Maxwell, G. D.

Stucky, M. Krishnamurty, P. Petroff, A. Firouzi, M. Janicke, B. F. Chmelka, Science, 1993, 261, 1299-1303.

[30] A. Firouzi, D. Kumar, L.M. Bull, T. Besier, P. Sieger, Q. Huo, S.A. Walker, J.A. Zasadzinski, C. Glinka, J. Nicol, D.I. Margolese, G.D. Stucky, B.F. Chmelka, Science, 1995, 267, 1138-1143.

[31] Q. Huo, R. Leon, P.M. Petroff, G.D. Stucky, Science, 1995, 268, 1324-1327. [32] Q. Huo, D. Margolese, G.D. Stucky, Chem. Mater., 1996, 8, 1147-1160. [33] S. Namba, A. Mochizuki, M. Kito, Chem. Lett., 1998, 569-570.

[34] M. Lindén, P. Ågren, S. Karlsson, P. Bussian, H. Amenitsch, Langmuir, 2000, 16, 5831-5836.

[35] N. Ulagappan, C.N.R. Rao, Chemm. Commun., 1996, 2759-2761.

[36] A. Lind, J. Andersson, S. Karlsson, P. Ågren, P. Bussian, H. Amenitsch, M. Lindén, Langmuir, 2002, 18, 1380-1385.

[37] Q. Huo, D. Margolese, P. Feng, T. Gier, P. Sieger, R. Leon, R. Petrov, U. Ciesla, F. Schüth, G.D. Stucky, Nature, 1994, 368, 317-321.

[38] S.A. Bagshaw, E. Prouzet, T.J. Pinnavaia, S.J.L. Billinge, T.P. Rieker, J. Am. Chem. Soc., 1995, 269, 1242-1244.

[39] D.M. Antonelli, J.Y. Ying, Angew. Chem. Int. Ed., 1996, 35, 426-430. [40] R. Ryoo, S.H. Joo, S. Sun, J. Phys. Chem. B, 1999, 103, 7743-7746. [41] A. Taguchi, F. Schüth, Microp. Mesop. Mater., 2005, 77, 1-45.

[42] K. Egeblad, C.H. Christensen, M. Kustova, C.H. Christensen, Chem. Mater., 2008, 20, 946-960.

[43] S.A. Bagshaw, N.I. Baxter , D.R.M. Brew, C.F. Hosie, Y.T. Nie, S. Jaenicke, C.G, Khuan, J. Mater. Chem., 2006, 16, 2235-2244.

[44] J. Perez-Ramirez, C.H. Christensen, K. Egeblad, C.H. Christensen, J.C. Groen, Chem. Soc. Rev., 2008, 37, 2530-2542.

[45] R. Chal, C. Gerardin, M. Bulut, S. van Donk, Chemcatchem, 2011, 3, 67-81.

Introduction 1

(23)

[46] S. van Donk, A.H. Janssen, J.H. Bitter, K.P. de Jong, Cat. Rev. Sci. Eng., 2003, 45, 297-319.

[47] J.C. Groen, L.A.A. Peffer, J.A. Moulijn, J. Perez-Ramirez, Chem. Eur. J., 2005, 11, 4983-4994.

1 Introduction

(24)

Chapter 2

Brønsted acidity of aluminium-rich SBA-15 type

mesoporous aluminosilicas

Summary

A series of aluminum-rich SBA-15 amorphous aluminosilicas was prepared by different synthesis methods. Decreasing the silicon to aluminium ratio below 14 results in materials with a less ordered structure than that of well-formed SBA-15 despite the use of a single source molecular precursor or the use of acetone as the solvent. It was not possible to obtain SBA-15 with a silicon to aluminium ratio of 1. The Brønsted acidity of the mesoporous aluminosilicates, as probed by selective H-D exchange FTIR, was found to be very similar to conventional amorphous silica-aluminas. The slightly higher activity in the hydroisomerization of n-heptane for the SBA-15 type silicates is ascribed to the presence of many weakly acidic hydroxyl groups present on the surface.

(25)

2.1 Introduction

Zeolites are widely used as solid acid catalysts, especially in the petrochemical industry and oil refining [1]. A general problem with the use of zeolites is the smallness of their pores (< 2 nm). This not only leads to diffusion limitations in reactions of hydrocarbon molecules and incomplete utilization of the internal zeolite voids, but also implies that part of the feed cannot be converted because large molecules cannot enter the micropores. An example of the latter is in the hydrocracking of heavy oil feeds.

To resolve these problems several strategies can be applied such as the modification of the zeolite texture to enhance their mesoporosity [2,4], the synthesis of zeolites with larger pores [5,6] and the use of (ordered) mesoporous silicas [7,8]. Ordered mesoporous silicas such as M41S and SBA-n were discovered in the 1990’s [9-11] as a promising class of materials for catalysis. Mesoporous silicas have attracted widespread attention in the field of catalysis and separation [12,13], because they can overcome pore size constraints and diffusion limitations of conventional zeolites. However, due to the lack of ordering of the atoms in their walls they do not offer the same acidity as usually found in crystalline zeolites. Despite this, many authors have claimed that Al-functionalized ordered mesoporous silicas obtained in a variety of ways exhibit high Brønsted acidity and are useful alternatives for zeolites in acid catalysis [14-19].

SBA-15, which is synthesized with the pluronic P123 triblock copolymer, has attracted widespread attention because of its thicker pore walls compared to the M41S family of mesoporous silicas, resulting in improved (hydro)thermal stability [20,21]. Due to the acidic synthesis conditions of SBA-15 the incorporation of aluminium is different than for other mesoporous aluminosilicates synthesized at higher pH. In the latter case often alumina phases are formed [22]. As the major application in catalysis of aluminosilicates is in the field of acid catalyzed reactions, accurate determination of the Brønsted acidity of ordered mesoporous silicas is essential. In recent years several attempts have been made to quantify the acidic properties of SBA-15 [23.24], although this remains challenging due to its amorphous nature. Deng et al. showed the absence of bridging hydroxyl groups in NMR spectra of Al-SBA-15, although adsorption of basic molecules as pyridine indicated that Brønsted acidic sites are present [16]. Therefore, they concluded that the acid sites in Al-SBA-15 originate from terminal silanols in the vicinity of aluminium atoms, which upon exposure to basic molecules, convert to bridging

2 Brønsted acidity of Aluminum rich SBA-15...

(26)

hydroxyls. Such models have been suggested before in connection with the recently resolved nature of Brønsted acidity of amorphous silica-aluminas [25-27]. These amorphous silica-aluminas resemble Al/SBA-15 in the absence of an IR spectrum that is commonly found for acid zeolites. FTIR studies on the adsorption of ammonia and pyridine on Al-SBA-15, on the other hand, suggested the presence of both Brønsted acid sites quite similar to those in zeolites next to several different sites of medium Brønsted acidic strength [28,29]. The upset of these studies seems to be that Al/SBA-15 contains Brønsted acid sites (BAS) but their strength as compared to zeolites and their density remain unclear.

Recently, it has been found that the acidity of the broad range of aluminosilicates including zeolites, clays and amorphous silica-aluminas is predominantly determined by sites of zeolitic strength [30,31]. The much lower acid activity of amorphous silica-alumina relates to a very low density of bridging hydroxyl groups. In this picture the number of Al substitutions in amorphous silica is very limited. This situation is very different from the ease of substitution of aluminium for silicon in tetrahedral positions of the crystalline zeolite framework. As discussed above, many groups have claimed high Bronsted acidity of Al-substituted SBA-15. Therefore, in this work, efforts were made to establish the nature of the Brønsted acidity in Al/SBA-15 by the H/D exchange FTIR technique [30,31]. In order to have a set of Al/SBA-15 materials in a wide compositional range for comparison in acidity characterization by IR spectroscopy and acid catalytic testing, we have employed a procedure for synthesis of high-quality Al/ SBA-15 [32]. Al/SBA-15 with a relatively high Al content was prepared by use of a single-source molecular precursor already containing the desired Si-O-Al bond [32,33]. This can be combined with the use of acetone as the solvent [34] to avoid hydrolysis of the Si-O-Al bond in this molecular precursor. The texture and morphology of these materials was extensively characterized. The catalytic activity of these Al/SBA-15 supports was then benchmarked in an acid catalyzed reaction, viz. the hydroconversion of n-heptane, against an amorphous silica-alumina and a zeolite. A subset of these materials was characterized for strong Brønsted acidity by H/D exchange FTIR.

Brønsted acidity of Aluminum rich SBA-15... 2

(27)

2.2 Experimental

2.2.1 Synthesis of Al-SBA-15

SBA-15 with Si/Al ratios of 14, 30 and 60 were synthesized under acidic conditions as described by Li et Al. [35]. In a typical procedure 2 g of pluronic P123 was dissolved in 75 ml hydrochloric acid solution of pH 1.5. The solution was stirred at 313 K for 6 hours. A second solution was made by adding 3.2 ml of TMOS and 5 ml of hydrochloric acid solution of pH 1.5 to a calculated amount of aluminium triisopropoxide. This suspension was vigorously stirred in a closed flask for 1.5 h during which it became clear. The silicon solution was quickly added to the surfactant solution and stirred for 20 hours at 313 K. The resulting white suspension was transferred to a Teflon lined stainless steel autoclave and heated to 373 K for 20 h. The resulting white solid was recovered by filtration and was washed three times with demineralized water and two times with ethanol. The sample was dried overnight at 373 K. Templates were removed by calcination for 10 h at 823 K in a flow of O2/N2 (1:1, 100 ml/min) with a heating ramp of 2 K/min.

The samples are denoted as SBA-w-x, where x is the Si/Al ratio.

Reference Y zeolite and amorphous silica alumina (ASA) materials were similar to those described in Ref. [31].

2.2.2 Synthesis of Al-SBA-15 with a single source molecular precursor

! Aluminosilicates with low Si/Al ratios of 10, 4 and 1.3 were synthesized using di-sec-butoxyaluminoxytriethoxysilane as a single source molecular precursor (SSMP) as described by Li et al. [32,33]. In a typical procedure 2 g of pluronic P123 and 23 mg NH4F were dissolved in 75 ml diluted hydrochloric acid

of pH 1.5. The solution was stirred at 313 K for 6 h. An amount of SSMP and TMOS were quickly added to the surfactant solution and the resulting mixture was stirred at 313 K for 20 h. The aged suspension was transferred to a Teflon lined stainless steel autoclave and heated to 373 K for 20 h. The product was recovered by filtration and washed three times with demineralized water and two times with ethanol and dried over night at 373 K. Templates were removed by calcination for 10 h at 823 K in a flow of O2/N2 (1:1, 100 ml/min) with a heating ramp of 2 K/min.

The samples are denoted as SBA-ssmp-x, where x is the Si/Al ratio.

(28)

2.2.3 Hydrolysis controlled synthesis

A second set of Al-rich SBA-15 with Si/Al ratios of 2, 9 and 16 was synthesized using the method proposed by Su et al. [34]. In a typical procedure 4 g of Pluronic P123 was dissolved in 6.3 ml acetone followed by addition of 5.4 ml hydrochloric acid solution of pH 1.5. The resulting solution was stirred at 313 K for 6 h. In a second flask an amount of TMOS and SSMP were added to 50 ml acetone, and the solution was added to the surfactant solution. The resulting mixture was stirred at 313 K for 24 h and subsequently transferred to a Teflon lined stainless steel autoclave, which was heated at 373 for 24 h. The product was recovered by filtration and washed three times with demineralized water and two times by ethanol and dried over night at 313 K. Templates were removed by calcination at 823 K in a flow of O2/N2 (1:1, 100 ml/min) with a heating ramp of 2 K/min. The

samples are denoted as SBA-a-x, where x is the Si/Al ratio.

2.2.4 Catalyst characterization

The Al content of the catalysts was determined by inductively coupled plasma optical emission spectroscopy (ICP-AES). Measurements were performed on a Spectro CIROS CCD spectrometer. Prior to the measurement the samples were dissolved in a mixture of HF:HNO3:H2O (1:1:1).

XRD patterns were recorded on a Bruker D4 Endeavor powder diffraction system using Cu Kα radiation. With a scanning speed of 0.0057 º min-1_{in the range of 0.5º ≤}

2Θ ≤5º and 0.01 º min-1_{in the range of 5º ≤ 2Θ ≤ 60º. XRD-crystallinities were}

determined using the TOPAS 3.0 software.

Nitrogen sorption isotherms were measured at -196 °C on a Micromeritics Tristar 3000 system in static measurement mode. The samples were outgassed at 150 °C for 3 h prior to the sorption measurements. The Brunauer-Emmett-Teller (BET) equation was used to calculate the specific surface area (SBET) from the adsorption

data obtained (p/p0= 0.05-0.25). The mesopore volume (Vmeso) and mesopore size

distribution were calculated using the Barrett-Joyner-Halenda (BJH) method on the adsorption branch of the isotherm.

Transmission electron micrographs were obtained with a FEI Tecnai 20 at an electron acceleration voltage of 200 kV. Typically, a small amount of catalyst was suspended in ethanol, sonicated and dispersed over a Cu grid with a holy carbon film.

(29)

Scanning electron microscopy (SEM) was performed using a Philips environmental scanning electron microscope FEIXL-30 ESEM FEG in high-vacuum mode at low voltage

Magic angle spinning (MAS) 27_{Al single pulse NMR spectra were recorded on a}

Bruker DMX-500 NMR spectrometer equipped with a 2.5 MAS probe head operating at a magnetic field of 11.7 T (the Al resonance frequency at this field is 130.3 MHz). The 27_{Al chemical shift is referred to a saturated Al(NO}₃₎₃_{solution. In}

a typical experiment 10 mg of well-hydrated sample was packed in a 2.5 mm zirconia rotor. The MAS sample rotation speed was 25 kHz. The relaxation time was 1 s and the pulse length was 1 µs. Ammonia treated samples were first heated to 100 oC with a heating ramp of 5 o_{C/min in a flow of 50 ml/min N2,}

subsequently the flow was switched to 1% of NH3 in N2 and the sample was kept at 100 o_{C for 1 h. The sample was allowed to cool to RT and 10 mg was directly}

packed in a 2.5 mm zirconia rotor. The MAS sample rotation speed was 25 kHz. The relaxation time was 1 s and the pulse length was 1 µs.

2.2.5 Acidity characterization

The concentration of strong BAS in the aluminosilicates was determined by catalytic activity measurements of the hydroconversion of n-heptane of Pd-loaded aluminosilicates. To this end, a sieve fraction (250–500 μm) of SBA-15 catalyst was loaded with 0.4 wt% Pd via incipient wetness impregnation with a solution of appropriate concentration of Pd(NH3)4(NO3)2. The resulting material was calcined

at 300 °C. Prior to testing, the catalyst was reduced at 450 °C at 30 bar in flowing hydrogen (100 ml/min). Hydroconversion of n-heptane was carried out at 30 bar at a H2/hydrocarbon ratio of 24 mol/mol. The reaction temperature was lowered to

200 °C. Reaction products were evaluated by online gas chromatography (Hewlett-Packard GC-5890 equipped with an HP-5 column and a flame ionization detector). Hydroisomerization/hydrocracking of n-heptane is a first order reaction. From the temperature required to obtain 40% conversion, relative values for the rate constant k for two catalysts at a reference temperature Tref can be determined using

the expression

ln (k1/k2) = Eact,1/R(1/T40,1 – 1/Tref) – Eact,2/R(1/T40,2 – 1/Tref) (1a)

or, if we choose the second catalyst as the reference Tref = T40,2

(30)

ln (k/kref) = Eact/R(1/T40 – 1/T40,ref) (1b)

in which Eact is the activation energy (kJ/mol-) and R (J/mol.K). If we assume the

intrinsic acidity of each site to be constant, Eact/R(1/T40 – 1/T40,ref) scales with the

ratio of the acid site densities between the two catalysts.

H/D exchange of hydroxyl groups with perdeuterated benzene was followed in situ by infrared spectroscopy as described elsewhere [30,31]. Infrared spectra were recorded in transmission mode in a Bruker IFS-113v FTIR spectrometer with a mid-infrared DTGS detector. Typically, a powdered sample was pressed into a self-supporting wafer with a density ρ = 10 mg/cm2_{and placed in an in situ cell. After}

calcining the catalyst wafer at 550 °C , the catalyst was evacuated to a pressure better than 210-6_{mbar and temperature was lowered to 30 °C. A background}

spectrum was recorded. Perdeuterobenzene (C6D6, Merck, purity 99.96 %) was

introduced into the cell from a glass ampoule. The total volume of C6D6

administered to the cell was 0.33 mmol ± 1 %, resulting in a pressure of 10 mbar. IR spectra were recorded for different exposure times and different temperatures. For each spectrum, 125 scans were accumulated at a resolution of 2 cm-1_{. Difference}

spectra were obtained by subtracting the initial spectrum of the dehydrated sample from the spectra after exposure to C6D6.

!

2.3 Results

2.3.1 Structural characterization

The elemental compositions of the calcined materials as determined by ICP are collected in table 2.1. For all SBA-15 the Si/Al in the final product is higher than that in the synthesis gel. As for the synthesis of a-2 and SBA-15-SSMP-1.6 initially silicon and aluminium were present in one complex (no additional silicon added), it immediately follows that the Si-O-Al bonds in this double alkoxide were decomposed.

(31)

Table 2.1: Physical properties of the calcined mesoporous aluminosilicates. Table 2.1: Physical properties of the calcined mesoporous aluminosilicates. Table 2.1: Physical properties of the calcined mesoporous aluminosilicates. Table 2.1: Physical properties of the calcined mesoporous aluminosilicates. Table 2.1: Physical properties of the calcined mesoporous aluminosilicates. Table 2.1: Physical properties of the calcined mesoporous aluminosilicates. Table 2.1: Physical properties of the calcined mesoporous aluminosilicates.

Catalyst Si/Al gel Si/Al final SBET (m2/ g) Vpore (cm3/ g) dpore (nm) a0 2 (nm) SBA-15-w-60 28 60 1005 1.48 6.2 13.313.3 SBA-15-w-30 20 30 911 1.39 6.1 10.710.7 SBA-15-w-14 10 14 842 1.47 6.7 10.910.9 SBA-15-SSMP-10 5 10 550 2.44 9.1 12.512.5 SBA-15-SSMP-4.0 3 4 378 1.52 9.2 13.513.5 SBA-15-SSMP-1.3 1 1.3 287 0.43 br.1 _n.d._n.d.33 SBA-15-a-16 10 16 552 0.62 br.1 _n.d._n.d.33 SBA-15-a-9.0 5 9 268 0.73 br.1 _n.d._n.d.33 SBA-15-a-2.0 1 2 341 0.53 3.6 n.d._n.d.33

1_{Broad mesopore size distribution, with no sharp maximum;} 2_{Calculated from}

XRD as a0 = 2.d(100)/√3; 3Not determined due to absence of low angle reflections. 1_{Broad mesopore size distribution, with no sharp maximum;} 2_{Calculated from}

XRD as a0 = 2.d(100)/√3; 3Not determined due to absence of low angle reflections.

The structure and texture of calcined materials was investigated by XRD and N2

adsorption. Fig. 2.1 contains the physisorption isotherms, BJH pore size distributions and low angle XRD diffractograms of the SBA-w-x catalysts. These samples have a type IV isotherm with a type I hysteresis loop between p/p0 =

0.7-0.9 indicative of ordered mesoporosity. The corresponding BJH pore size distributions taken from the adsorption branch of the isotherms show a narrow pore size distribution centred around 6 nm. An increase of the Si/Al ratio typically results in a small increase in the mesopore diameter. In the corresponding low angle XRD diffractograms, all three samples show the reflections characteristic of the hexagonal P6mm structure. The patterns of SBA-w-60 and SBA-w-30 are very similar. SBA-w-14, on the other hand, has a slightly larger unit cell as a result of larger pores and/or thicker pore walls. Representative TEM micrographs of the SBA-w-x series are shown in figure 2.2. All three aluminosilicas show long ordered mesopores with the hexagonal arrangement of the pores typical for SBA-15.

(32)

Figure 2.1: N2 physisorption (left), BJH pore size distribution (middle) and small angle

XRD diffractograms (right.) Samples denote: () SBA-15-w-60, () SBA-15-w-30 and () SBA-15-w-14.

Figure 2.2: Representative TEM images of (a) SBA-15-w-14 (b) SBA-15-w-30 and (c) SBA-15-w-60.

27_{Al NMR MAS spectra of the calcined SBA-w-x aluminosilicas are depicted in Fig.}

2.3. The spectra show that all three samples contain tetrahedral (~53 ppm) and octahedral (~0 ppm) Al. The latter signal consists of a broad

signal centered around -5 ppm and a sharper signal around 0 ppm. The NMR spectra were deconvoluted to investigate the ratio of these two types of Al coordination. The tetrahedral to octahedral ratio is close 60:40 irrespective the

(33)

initial Al content (Table 2.2). To gain more detailed insight into the nature of the octahedral Al species, an NMR spectrum of SBA-w-14 was recorded after dehydration at 120 °C and exposure to ammonia. Under such conditions octahedral aluminum that is part of a silica-alumina phase will revert to tetrahedral coordination [38,39]. On the contrary, octahedral Al part of an alumina-type phase will remain unaffected. The NMR spectrum obtained after exposure to ammonia does not contain a signal at 0 ppm anymore. This implies that in principle all Al in this sample is part of a silica-alumina phase and is highly dispersed on or in the silica network. No Al has segregated into alumina-type of domains. As this sample has the highest Al content, it is reasonable to conclude that SBA-w-30 and SBA-w-60 also do not contain a separate alumina phase.

Table 2.2:Relative amount of Al species in the mesoporous aluminosilicates as determined from NMR.

Catalyst Altetra (%) Aldist. (%) AlAloctaocta (%) (%)

SBA-15-w-60 61 0 3939 SBA-15-w-30 60 0 4040 SBA-15-w-14 63 0 3737 SBA-15-SSMP-10 27 (74)1 _{41 (0)}1 _{32 (26)}_{32 (26)}11 SBA-15-SSMP-4.0 39 (55)1 _{36 (0)}1 _{25 (61)}_{25 (61)}11 SBA-15-SSMP-1.3 18 (21)1 _{40 (0)}1 _{42 (79)}_{42 (79)}11 SBA-15-a-16 49 (59)1 _{32 (0)}1 _{19 (41)}_{19 (41)}11 SBA-15-a-9.0 40 (70)1 _{39 (0)}1 _{21 (30)}_{21 (30)}11 SBA-15-a-2.0 23 (33)1 _{38 (0)}1 _{39 (41)}_{39 (41)}11

1_{Numbers between brackets denote values of the uncalcined samples.} 1_{Numbers between brackets denote values of the uncalcined samples.} 1_{Numbers between brackets denote values of the uncalcined samples.} 1_{Numbers between brackets denote values of the uncalcined samples.}

(34)

Figure 2.3: 27_{Al NMR spectra of the calcined SBA-15-w series catalysts.}

Figure 2.4: N2 physisorption isotherms (left), BJH pore size distributions (middle) and low

angle XRD diffractograms (right) of the SBA-SSMP series catalysts. Samples represent: () SBA-SSMP-1.3 (), SBA-SSMP-4.0 and () SBA-SSMP-10.

The physisorption isotherms, BJH pore size distributions and low angle XRD diffractograms of the SBA-ssmp catalysts can be seen in figure 2.4. The isotherm of SBA-ssmp-1.3 has a very low uptake over the whole pressure range, indicating a very low porosity. SBA-ssmp-4.0 and SBA-ssmp-10, on the other hand, have

(35)

isotherms with a strong uptake around P/P0 = 0.8 followed by a second step at P/

P0 = 0.95. This type of behavior is points to a structure of meso- and macroporosity.

This is also reflected by the BJH pore size distributions, which show a mesopore size centered around 9 nm in combination with a relatively broad macropore size distribution. The XRD pattern of SBA-ssmp-1.3 does not exhibit a reflection in the low angle region in contrast to the characteristic P6mm reflections in the corresponding diffractograms of SBA-ssmp-4.0 and SBA-ssmp-10. The peaks are shifted to larger d-spacings as compared to the SBA-w series and they are also much broader. The latter observation generally implies that the long range order is quite weak, i.e. these two SBA-ssmp samples likely consists of SBA-15 type particles with only a few repeating units perpendicular to the pore axis. Representative TEM images of the SBA-15-ssmp series are depicted in figure 2.5. As already suggested by the XRD and N2 physisorption results, SBA-15-ssmp-1.3 is

a non-porous amorphous aluminosilica. Both ssmp-4.0 and SBA-15-ssmp-10 consist of small fragments of SBA-15 with a size in the range of 50-200 nm. The small size of these particles is causing the broadening of the low-angle XRD reflections. The macropores in these samples, which can clearly be observed in the TEM images for SBA-15-ssmp-10, derive from the void space between the SBA-15 fragments agglomerated into larger particles.

Figure 2.5: Representative TEM images of (a) SBA-ssmp-1.3, (b) SBA-SSMP-4.0 and (c) SBA-SSMP-10.

Figure 2.6 shows the 27_{Al MAS NMR spectra of the SBA-15-SSMP series catalysts}

before and after calcination. Those of the as-synthesized samples evidence the presence of tetrahedral and octahedral Al. The relative contributions of the different forms of Al are listed in Table 2.2. It is seen that the amount of octahedral Al increases with increasing Al content. When these samples are calcined, an

(36)

additional signal around 28 ppm appears, which is due to aluminium in a distorted tetrahedral coordination or to five-coordinated Al. Similar to the SBA-15-w-14 the calcined SBA-15-ssmp catalysts have been exposed to ammonia to gather insight in the structure of the octahedral part of the aluminium. The SBA-15-SSMP-10 shows, similar to the SBA-15-w-14, a shift of the octahedral part of the aluminium to tetrahedral coordination. Apparently all the aluminium in this sample is in a silica-alumina like phase [38]. In the case of SBA-15-SSMP-4.0 on the other hand, only a part of the octahedral aluminium species revert to a tetrahedral coordination. This suggests that for this material having a high load of aluminium, part of the aluminium is in a pure alumina like phase. However as the high angle XRD diffractogram does not show the presence of any crystalline aluminate the alumina phase is most likely purely amorphous.

Figure 2.6: 27_{Al MAS NMR spectra of the as synthesized (left) and calcined (right)}

SBA-SSMP series catalysts. Dotted lines indicate ammonia treated samples.

The textural data and diffractograms of the third set of samples prepared using acetone as a solvent (SBA-15-a) are shown in figure 2.7The isotherms of the three catalysts synthesized in acetone do not show any trend as a function of the Al content. All the catalysts show limited uptake of nitrogen in the mesopore region, which implies a very small mesopore volume as compared with conventional SBA-15. The nitrogen uptake for SBA-15-a-16 occurs over the whole relative pressure range, indicating that this material does not have any well-defined pore size as is also evident from the corresponding BJH pore size distribution. For

(37)

SBA-15-a-9 the isotherm has no additional nitrogen uptake below p/p0 = 0.9

showing that this catalyst has no mesoporosity at all. The steep increase above this value is typical for an amorphous aluminosilica with macroporous voids between the particles. SBA-15-a-2 has an isotherm with a small uptake step at p/p0 = 0.5 and

a second small step can be observed at a p/p0 > 0.9. The small overall uptake of

nitrogen implies that the pore volume of this material is limited as compared to conventional SBA-15. Also the pore size is smaller compared to the SBA-15-w and SBA-15-SSMP samples. Nevertheless, the sample does not contain a fraction of mesopores with a narrow PSD around 3.5 nm

Figure 2.7: N2 physisorption (left), BJH pore size distributions (middle) and low angle XRD diffractograms (right) of the SBA-a series catalysts. Samples denote () SBA-15-a-16, () SBA-15-a-9.0 and () SBA-15-a-2.0.

Figure 2.8: Representative TEM images of (a) SBA-15-a-2.0, (b) SBA-15-a-9.0 and (c) SBA-15-a-16.

(38)

In line with the physisorption and XRD data, SBA-15-a-2 appears as particles of 100 nm possessing small equally sized but randomly ordered mesopores. SBA-15-a-9, on the other hand, consists of much smaller particles with a clear pore structure. Hence the macroporosity is a result of the voids between the agglomerated particles. SBA-15-a-16 looks very similar to SBA-15-a-9. This material appears as an agglomerate of smaller silica particles and its porosity appears to come from the interparticle voids. Our results show that the sample with the highest Al content resembles to some extent the material produced by Su’s group [34], although the structural order in our sample is smaller. This order becomes lower when the Al content is decreased and this is undoubtedly due to the addition of TMOS.

27_{Al MAS NMR spectra of the as-synthesized and calcined SBA-15-a series catalysts}

are shown in figure 2.9. The spectra are very similar to the SBA-15-SSMP samples. Before calcination all three SBA-15-a catalysts produce peaks of tetrahedral (~53 ppm) and octahedral (~0 ppm) Al. Within the series the relative amount of octahedrally coordinated aluminium increases with the Al content. Another similarity is the appearance of the signal ~28 ppm due to distorted four- or five-coordinated Al.

Figure 2.9: 27_{Al-MAS-NMR spectra of the as synthesized (left) and calcined SBA-15-a}

series catalysts.

(39)

2.3.2 Acidity characterization

The acid catalytic activity of the Pd-loaded aluminosilicates was tested in the hydroconversion of n-heptane. The Pd content is chosen such that the rate is determined by the acid catalyzed reaction of olefins [30]. There is general agreement that strong BAS are required for the isomerization and cracking reactions [36,37]. The results are summarized in Table 2.3. This table also contains the results for three reference aluminosilicates, namely an amorphous silica-alumina with Si/Al = 9.5, a stabilized faujasite zeolite (Si/Al = 4.8) and HZSM-5 zeolite (Si/Al = 20). The acid activity is given as the first-order reaction rate constant compared to the rate constant of an ultrastabilized faujasite (USY) reference zeolite.

Table 2.3: Number of strong Brønsted acid sites (NBAS) as determined by H/D exchange

FTIR and kinetic parameters for the catalytic activity in n-heptane hydroconversion.

Catalyst NBAS (µmol/g) T401 (°C) Eact2 (kJ/mol) ln (k/kref) SBA-15-w-60 0.7 324 135 _-6.22 SBA-15-w-30 3.0 306 145 _-5.77 SBA-15-w-14 3.6 307 135 _-5.42 SBA-15-SSMP-10 1.0 323 128 _-5.85 SBA-15-SSMP-4 1.3 323 130 _-5.94 SBA-15-SSMP-1.3 < 0.1 330 128 _-6.15 SBA-15-a-16 1.2 307 137 _-5.50 SBA-15-a-9 1.0 324 130 _-5.99 SBA-15-a-2 < 0.1 323 152 _-6.95 ASA4 _2.6 ₃₄₃ ₁₃₀ -6.79 HZSM-55 ₇₃₀ ₂₂₁ ₁₅₆ -0.63 USY6 ₁₂₅₀ ₂₃₀ ₁₁₃ _0.00

1_{Temperature required for n-heptane conversion of 40 %.}2_{Apparent activation energy.}3

Not determined. 4_{Amorphous silica-alumina with Si/Al = 9.5.}5_{Si/Al = 19.4.}6_Steam

calcined faujasite zeolite with Si/Al = 4.1.

(40)

Two conclusions immediately obtrude themselves concerning the acidity of Al/ SBA-15: (i) it is much lower than HZSM-5 and USY zeolites and (ii) it is of the same order as the acidity of the amorphous silica-alumina reference sample. Within the set of SBA-15 samples it is seen that the conventional procedure to synthesize Al/ SBA-15 gives the most acidic supports. For the SBA-15-w set the acid activity increases with the Al content. The other two series prepared with a higher Al content show a decrease of the activity with the Al content. Expectedly, the acid activity of SBA-15-w without Al is very low (ca. 50x lower than the activity of SBA-15-w-14).

Figure 2.10: (left): FT-IR OH region of the SBA-15-w series catalysts. Figure 2.10 (right): FT-IR OD region spectra of SBA-15-w-14 exchanged for: (a) 10 min, 30o_{C, (b) 20 min,}

30o_{C, (c) 30 min, 30}o_{C (d) 30 min, 50}o_{C and (e) 30 min, 80}o_C.

The possible presence of strong BAS of several of these Al/SBA-15 samples was then probed by selective H/D exchange FTIR spectroscopy. This method is based on the exchange of the strongly acidic hydroxyl groups by deuterium by exposure to perdeuterobenzene and has been shown to be particularly useful to quantify the very small amount of BAS next to a larger number of weakly and non-acidic silanol groups as typically found in (partially) amorphous aluminosilicas. Figure 2.10 shows the hydroxyl region of the IR spectra of the calcined SBA-15-w samples. All spectra contain only one intense silanol band centered at 3740 cm-1_{as is typically}

(41)

observed for amorphous silica-aluminas [31]. This confirms that the OH stretching region of the FTIR spectra of these aluminosilicas is not suitable for determination of strong BAS. Gora-Marek and Datka found evidence for vibrational bands of bridging hydroxyl groups in aluminium-containing MCM-41 and MCM-48 [40,41], but it should be mentioned that these materials were prepared using a modified synthesis method involving structure directing agents for zeolite growth.

As an example the FTIR OD region spectra of SBA-15-w-14 after exchange are shown in Figure 2.10 Three bands can clearly be distinguished: two rather weak bands at 2632 cm-1_{and 2682 cm}-1_{from the strong Brønsted acidic sites and a silanol}

feature at 2759 cm-1_{. Upon exposure to perdeuterobenzene at 30 °C the band at}

2682 cm-1_{appears immediately, whereas the band at 2632 cm}-1_{becomes visible only}

for longer exposure times. Similar to the findings for amorphous silica-aluminas, the exchange of strong Brønsted acidic sites reaches its maximum after 30 min of exposure at 50 °C. This was chosen as a reference condition for acidity determination based on extensive studies with zeolites, clays and amorphous silica-aluminas [31]. Longer exposure leads to an increased intensity of the band at 2632 cm-1_{, but now the band at 2682 cm}-1_{becomes obscured by the broad}

background.

Figure 2.11: FT-IR OD region after H-D exchange for the w series (left), SBA-15-ssmp series (middle) and the SBA-15-a (right) series catalysts.

(42)

The FTIR spectra of the exchanged SBA-15 catalysts are shown in Figure 2.11, the corresponding acid site densities are listed in Table 2.3. For the SBA-15-w samples the acid site density increases with the Al content in accordance with the acid activity trend. Increasing the Al content by use of double alkoxide precursor results in a decrease of the Brønsted acidity. The FTIR method did not evidence the presence of strong acidic sites in SBA-15-SSMP-1.3, even after H/D exchange at 80 °C. Broadly speaking, the materials synthesized in acetone show a similar behaviour as the SBA-15-SSMP materials. With increasing Al content the amount of strongBrønsted acidic sites decreases and the sample with the highest Al content, SBA-15-a-2, does not contain any strong BAS that can be identified by H/D exchange FTIR.

Figure 2.12: Acid activity in hydroconversion of n-heptane and the concentration of acid sites determined by H/D exchange FTIR for (■) zeolites, () a clay, () amorphous silica-aluminas and () Al/SBA-15.

Figure 2.12 shows the relation between the acid activity in hydroconversion and the acid site density for the Al/SBA-15 samples, a set of amorphous silica-aluminas, a clay and the two zeolites [31]. There is quite some apparent scatter, but it appears safe to conclude that the acidity of Al/SBA-15 and amorphous

silica-Brønsted acidity of Aluminum rich SBA-15... 2

Hierarchical ZSM-5 zeolite catalysts for the selective oxidation of benzene