SEARCH

SEARCH BY CITATION

Keywords:

  • aluminosilicates;
  • cracking;
  • heterogeneous catalysis;
  • mesoporous materials;
  • zeolites

The field of hierarchical zeolites in general and mesoporous zeolites in particular has seen tremendous growth over the past decade, as part of the ongoing efforts to alleviate the diffusional limitations that are imposed by the micropores of conventional zeolites. Many new approaches, herein classified as either “bottom-up” or “top-down”, have been reported and are briefly surveyed in this Review. Zeolite Y is the most widely used zeolite in catalysis and the developments in mesoporous Y reflect the general “landscape” of mesoporous zeolites. The preparation of mesoporous Y, the materials’ properties, and their catalytic application in fluid catalytic cracking (FCC) and hydrocracking are critically reviewed. Finally, the scale-up and use of mesostrutured zeolite Y on an industrial scale are described, which constitute the first commercial application of hierarchical zeolites.

1. Introduction

  1. Top of page
  2. 1. Introduction
  3. 2. Preparation Strategies of Mesoporous Zeolites
  4. 3. Mesoporous Zeolite Y
  5. 4. Summary and Outlook
  6. Acknowledgements
  7. Biographical Information
  8. Biographical Information
  9. Biographical Information

Zeolites are crystalline microporous aluminosilicates that are composed of TO4 tetrahedra (T=mostly Al or Si) as primary building units that are connected through the corner-sharing of oxygen atoms to form 3D frameworks that encompass pores of diameters of molecular dimensions (micropores of diameters smaller than 2 nm).1 The framework structure and chemical composition determine the unique properties of a specific zeolite and its use in important areas, such as water treatment, adsorption and separation, and catalysis. Among the different properties of zeolites, their pore structure, that is, the size, shape, and interconnectivity of the pores, have an overarching effect across all applications. On one hand, they impart different zeolites with unique and important capabilities, such as molecular sieving and shape/size selectivity.2, 3 On the other hand, they not only exclude certain large molecules, such as some of the larger molecules in heavy petroleum feed, from accessing the active sites that are located within the micropores of a zeolite, but they also impose significant diffusional limitations to molecules that can only tightly fit within the pores, for example, reactants, intermediates, and products in many reactions that are catalyzed by zeolites (Figure 1 a). The diffusion of tightly fitting molecules into the micropores of zeolites, referred as “configurational diffusion”, is often the rate-limiting step of a catalyzed reaction because, as the size of the molecules approaches the dimension of the pores, the molecular diffusivity drops sharply to orders of magnitude lower than, for example, the Knudsen-diffusion (often the dominating diffusion mechanism in mesopores) and molecular-diffusion regimes (Figure 1 b).4

thumbnail image

Figure 1. a) Schematic representation of the effect of pore size on the diffusion of large (red) and small (black) molecules. b) Effects of pore diameter on molecular diffusivity (D) and of the energy of activation (Ea) on diffusion. Adapted and reprinted with permission from Ref. 4. Copyright 1995 American Chemical Society.

Download figure to PowerPoint

Since the early days of utilizing zeolites in catalytic processes, there have been ongoing efforts to alleviate the diffusion challenges/limitations through a wide variety of approaches. They typically align with one of the following three directions: 1) The synthesis of zeolites with larger micropores;59 2) reduction of the crystal size of the zeolite down to nanoscales in at least one dimension;1013 3) introduction of additional porosities of larger sizes, typically mesopores or even macropores, into the crystals of microporous zeolites or into the particles that comprise the zeolite crystals. Whilst impressive progress has been made in the first two approaches, technical challenges, including the use of expensive templates, low hydrothermal stability, and difficulty in handling of the nanosized materials, have limited their large-scale commercial application. In this Review, we first provide a critical overview of the main strategies for introducing mesoporosity into zeolites and then undertake a more-detailed comparison of the different approaches to mesoporous zeolite Y in particular, as well as their catalytic applications and commercialization prospects. Notably, the term “hierarchical zeolite” encompasses any zeolite with at least a secondary pore-structure system and, therefore, “mesoporous zeolite” should be considered as a subclass of the former because it defines the size of the additional porosity as falling within mesopore range, that is, between 2 and 50 nm.

2. Preparation Strategies of Mesoporous Zeolites

  1. Top of page
  2. 1. Introduction
  3. 2. Preparation Strategies of Mesoporous Zeolites
  4. 3. Mesoporous Zeolite Y
  5. 4. Summary and Outlook
  6. Acknowledgements
  7. Biographical Information
  8. Biographical Information
  9. Biographical Information

Since the beginning of the 21st century, much progress has been made in the synthesis, characterization, and application of mesoporous zeolites.1419 Typically, these methods can be divided into two categories: “Top-down” or post-synthetic modification, which may or may not involve the use of templates, and “bottom-up” or primary syntheses, which mostly involve the use of templates (Figure 2).

thumbnail image

Figure 2. “Bottom-up” and “top-down” approaches to mesoporous zeolites as categorized herein.

Download figure to PowerPoint

2.1. “Bottom-up” approaches

2.1.1. Hard-templating

Jacobsen et al. published pioneering work on the preparation of ZSM-520 (and TS-1)21 single crystals with intracrystalline mesoporosity (Figure 3) as an extension of their earlier efforts2224 in the synthesis of nanosized zeolites ZSM-5, beta, X, and A in the confined space within the pores of carbon black. This approach, typically referred to as “hard-templating”, was quickly adopted by many researchers. In addition to carbon black, many other hard templates, such as carbon nanotubes or nanofibers,25, 26 ordered mesoporous carbons,27, 30 non-ordered carbon aerogels or mesoporous carbons,3134 pyrolyzed wood or carbonized rice husk,35, 36 CaCO3 nanoparticles,37 and polystyrene (and other polymeric) microspheres,3840 have been utilized to synthesize various mesoporous (or macroporous in a few cases) zeolites with various topologies, including MFI, MEL, MWT, BEA, AFI, and CHA. Because all of these strategies start from typical or modified zeolite-synthesis gels, they are all classified as “bottom-up” approaches (Figure 2).

thumbnail image

Figure 3. Growth of zeolite crystals around carbon particles as hard templates. Adapted with permission from Ref. 20. Copyright 2000 American Chemical Society.

Download figure to PowerPoint

2.1.2. Soft-templating

“Bottom-up” approaches also include “soft-templating” methods, which involve the use of cationic surfactants,41, 42 mesoscopic cationic polymers,43 silylated polymers or surfactants,4448 polymeric aerogels,49 starch,50, 51 bacteria,52 etc.53 Many of the early efforts at synthesizing mesoporous zeolites, often referred to as “dual-templating” methods, involved the simultaneous use of both mesopore templates, such as surfactants, and structure-directing agents (SDA, for example, short-chain alkylammonium salts), to form certain zeolite phases whilst introducing mesoporosity. In some earlier works, this approach often led to physical mixtures of amorphous mesoporous material and conventional (non-mesoporous) zeolite crystals.5456 More recently, mesoporous zeolites with MFI, BEA, FAU, MOR, LTA, etc., topologies have been obtained by building both the mesopore template and the SDA function into the same molecule. For example, Na et al.41 designed and synthesized a gemini surfactant that contained both a mesopore-templating function and a structural moiety that was similar to molecular SDAs that are typically used in zeolite synthesis (Figure 4). By having the long aliphatic chain and the SDA functions in the same molecule (18-N3-18), phase segregation was avoided and nanocrystalline zeolite with hexagonally ordered intracrystalline mesoporosity was obtained.

thumbnail image

Figure 4. Dual templating: A)  18-N3-18 surfactant (white spheres: hydrogen, gray spheres: carbon, red spheres: nitrogen). B) SEM, C), D) TEM, and E) XRD patterns of the mesostructured ZSM-5. Insets in (C) and (D): Fourier diffractograms. For structural comparison, an MFI framework model is shown in the bottom-right inset of (D). Reprinted with permission from Ref. 41. Copyright 2011 American Association for the Advancement of Science.

Download figure to PowerPoint

2.1.3. Zeolitization of mesoporous materials or the hierarchical assembly of nanozeolites

Also included in the “bottom-up” approaches, as shown in Figure 2, are the approaches that involve either converting the amorphous pore walls of mesoporous-silica-containing materials (such as silicates or aluminosilicates, silica monoliths, diatomite) into zeolites or assembling protozeolitic units into mesoporous materials with nanocrystalline pore walls.5776 Some of these approaches appear to be quite simple and versatile, for example, the TUD-C and TUD-M approaches only use TPAOH (tetrapropylammonium hodoxide) as a template for both micropores and mesopores in the preparation of composites of ZSM-5 nanocrystals that are embedded into well-connected mesoporous matrices.7476

Increased acidity and hydrothermal stability, relative to materials such as MCM-41 and SBA-15 with amorphous pore walls, are typically associated with these materials; however, only intercrystalline mesoporosity (in between the nanozeolite crystals) can be introduced by using these approaches. The pore walls often can only be partially crystallized, with very few exceptions.58 Therefore, the acidity of these materials are, in general, inferior to those of conventional zeolites.15

The “bottom up” approaches mostly involve the use of organic or inorganic templates in the zeolite crystallization process. While a main advantage of this approach could be the synthesis of mesoporous zeolites with identical (or similar) chemical compositions to conventional zeolites, the key to this general approach is to optimize the interactions between the templates and the aluminosilicate species in the reaction mixture to avoid phase segregation during the hydrothermal crystal-growth process. How well the templates are dispersed in the synthesis gels and become occluded during the hydrothermal synthesis will affect the phase purity and morphology of the zeolite crystals, as well as the location, distribution, shape, size, and interconnectivity of the mesopores. The cost of the templates would be of concern to the scalability and large-scale application of mesoporous zeolites that are prepared by using these approaches, especially if, for example, ordered mesoporous carbons are used, which are typically formed from tedious processes that involve the use of another template, for example, either SBA-15 mesoporous silica or carefully arranged colloidal silica particles.

2.2. “Top-down” approaches

2.2.1. Dealumination

Before the turn of the century, mesoporosity in zeolites was typically generated in a limited number of ways, which mostly involved dealumination through post-synthetic calcination, hydrothermal treatment (steaming), acid leaching, or chemical treatments.77 Originally, dealumination treatments were performed to control the concentration and strength of the acid sites by increasing the Si/Al ratio of low-silica zeolites. However, it was observed that, during the hydrothermal dealumination process, secondary pores at mesoscale may form. Because dealumination unavoidably alters the Si/Al ratio, the acidic properties of the zeolite will also be different from those of the original zeolite.78, 79 Nowadays, the hydrothermal dealumination techniques, alone or in combination with dealumination by acid, are widely practiced industrially, mainly to manufacture ultrastable Y (USY) zeolite for FCC and hydrocracking applications.77 Their limitations and drawbacks, discussed in more detail below, are the driving forces for the discovery of the new methods described in this Review.

2.2.2. Desilication

Desilication is another well-known demetalation approach for the creation of mesoporosity. The selective hydrolysis of Si[BOND]O[BOND]Si bonds can be traced back to as early as the 1960s. Young and Linda80 observed that, at high silica-to-alumina ratios, MOR zeolites could be treated in aqueous caustic solution to leach out a minor portion of the structural silica, thereby improving their adsorption capacity. At that point, the mechanism of desilication was not clearly understood. Mao et al.81 studied the desilication of ZSM-5 zeolites by using NaOH and Na2CO3 and they reported that the resulting zeolite had a homogeneous pore system of 0.56 nm and a higher density of ion-exchange sites.

Ogura et al.82 first described how the desilication of ZSM-5 zeolite in NaOH could generate intracrystalline mesoporosity. By using a 0.2 M NaOH aqueous solution, they observed mesopores of 4 nm in the zeolite and a loss of microporosity compared to the original zeolite.83 They also reported that a large portion (about 40 %) of the zeolite dissolved during the alkali treatment.83 Later on, they described how the NaOH concentration, desilication temperature, and treatment time could greatly affect the mesoporosity, and structural and acidic characteristics of the resulting zeolite. The mesopores size was later corrected to be a broad distribution approximately 10 nm.84

Groen et al.85 identified that the presence of Al gradients in the zeolite crystals and, specifically, the concentration of the framework Al species played key roles in the mechanism of mesopore formation in MFI (ZSM-5) zeolites in alkaline medium. They showed that the presence of high Al concentrations in the MFI zeolite framework (Si/Al<20) prevented the extraction of Si and, thus, limited the pore formation by desilication, whereas highly siliceous zeolites (Si/Al>50) showed excessive and unselective Si dissolution, thus leading to the creation of relatively large pores. A framework Si/Al ratio of 25–50 was optimal for obtaining substantial intracrystalline mesoporosity and generally preserved Al centers (Figure 5). A mechanism for the directing role of the framework aluminum species has also been proposed.86 Besides the concentration of the framework aluminum, the nature of these species and their location can impact desilication by alkaline treatment. Thus, alkaline treatment of a steam-calcined ZSM-5 zeolite (which contained a high degree of non-framework aluminum) led to minor silicon extraction and, consequently, limited mesoporosity development. The same group demonstrated the advantages of mesoporous ZSM-5 derived from desilication by using neopentane-adsorption experiments.87 They further discovered that the use of organic bases, such as TPAOH or TBAOH (tetrabutylammonium hydroxide) instead of NaOH, or use of tetraalkylammonium salts (the so-called “pore-growth moderators”) together with NaOH helped to better preserve microporosity.88 The same group also reported a few other variations to the typical NaOH-desilication process, for example, NaAlO2 treatment followed by acid washing to remove debris aluminum species89 and the addition of aluminum nitrate as a pore-directing agent to the NaOH solution.90 By using these different approaches, they effectively expanded the Si/Al ratio from about 10 to ∞ for the generation of mesoporosity in ZSM-5 by desilication (Figure 6).91

thumbnail image

Figure 5. Schematic representation of the influence of Al content on the desilication of MFI zeolites in NaOH solution and the associated pore-formation mechanism. Reprinted with permission from Ref. 85. Copyright 2004 American Chemical Society.

Download figure to PowerPoint

thumbnail image

Figure 6. Schematic representation of the procedures for the desilication of MFI zeolites with different framework Si/Al ratios. Adapted with permission from Ref. 91. Copyright 2011 Royal Society of Chemistry.

Download figure to PowerPoint

The desilication approach has also been applied to other zeolite topologies, such as MOR, BEA, FER, FAU, and CHA.15 Although both the chemical composition and structure of the zeolites seem to dictate the optimal desilication conditions and the properties of the resultant mesoporous zeolites, this approach clearly shows high versatility. However, there are issues that are associated with desilication. First, owing to the high usage of NaOH, material loss can be very significant, in addition to a loss of microporosity. Silica leaching from zeolite crystals and the less-crystalline regions in between the zeolite crystals (which typically serve as a binder to hold a few zeolite crystals together as larger particles) could cause a significant decrease in the size of the zeolite crystals and disintegration of the particles, which may lead to significant difficulties in filtration and further loss of yield.

2.2.3. Surfactant templated “top-down” approaches

A surfactant-templated method to introduce mesoporosity into zeolites was first filed as a patent application in 2004 and has been described in a few recent publications by Garcia-Martinez and co-workers.9294 This approach is based on the use of significantly milder conditions (e.g., dilute NH4OH solution) than those for desilication. Surfactants such as cetyltrimethylammonium bromide or chloride (CTAB or CTAC) are used to introduce well-controlled mesoporosity into various zeolites (e.g., Y, mordenite, ZSM-5). The technique does not suffer from the typical drawbacks of the desilication approach, i.e. significant loss of silica or damage of the zeolite crystals.

About the same time, Ivanova et al.95 published a preparation of composite micro/mesoporous mordenite that also involved the use of CTAB as a mesopore template, which was a follow-up of a procedure reported by Goto et al.96 There is an important difference between this approach and the surfactant-templating approach by Garcia-Martinez. The former process involves two steps: 1) Partial-to-total dissolution of the zeolite in NaOH solution, followed by, 2) hydrothermal treatment in the presence of CTAB at lower pH values, to produce a zeolite/mesoporous molecular-sieve composite (ZMC) of desilicated mesoporous zeolite and another mesoporous silica-rich amorphous phase that is deposited onto the surface of the crystals. The latter approach is a single-step process that combines surfactant templating with base treatment to produce single-phase mesostructured zeolites with intracrystalline mesoporosity. These two different approaches lead to the formation of markedly different materials, as illustrated in Figure 7, which will be discussed in details in the third section.

thumbnail image

Figure 7. Schematic representation of the differences between the materials prepared by using single-step and two-step approaches with cationic surfactants and NaOH; blue “snowflakes” represent the surfactant micelles, large red rectangle represents a zeolite crystal, small red squares represent silicate/aluminosilicate species that may contain zeolitic subunits.

Download figure to PowerPoint

In 2006, Pacheo-Malagón et al.97 reported an interesting depolymerization-recrystallization (DR) approach to hierarchical zeolites that first involved the depolymerization of zeolites (USY or ZSM-5) in glycerol at about 200 °C to form an amorphous gel (by X-ray Diffraction, XRD) and subsequent recrystallization of the amorphous gel in the presence of tetramethylammonium or tetrapropylammonium cations (TMA+ or TPA+) under hydrothermal conditions. Hierarchical zeolites that exhibited Y or ZSM-5 diffraction patterns were composed of recrystallized zeolite nanocrystals that were embedded in a mesoporous phase. Later, this approach was extended to silicalite-198 and Y zeolites.99 CTAB was also used in the preparation of hierarchical Y zeolite;100 however, nitrogen-adsorption/desorption analysis did not show the mesopore-templating effect of the surfactant.

2.3. Summary

Many new synthetic approaches for the generation of mesoporous zeolites have been discovered over the past decade or so and they can generally be categorized as either “bottom-up” or “top-down” approaches (Figure 2). It is also clear that “not all mesopores are created equal”, that is, qualitatively different mesoporosity results from different preparation methods. In some cases, tiny variations in the process conditions could result in quite different materials. The differences in material properties could lie in the amount (measured by the pore volumes or surface areas), location (intra- or inter-crystalline), size distribution, or interconnectivity (with adjacent mesopores and micropores) of the secondary mesopores. In addition, different methods also have different effects on the remaining microporosity, crystallinity, acid-site strength and distribution, and hydrothermal stability (of both the meso- and microporosity), because porosity is clearly not the only change to the original zeolites during the different chemical processes. Some efforts have attempted to make more-generalized comparisons of materials that were prepared by using different techniques. Perez-Ramirez et al.101 defined the hierarchical factor (HF) from the porosity results as the product of Smeso/Stotal and Vmicro/Vtotal (Figure 8), which was suggested as a tool to look at how effectively the mesopore surface area was generated by using different approaches at the expense of micropore volume, regardless of zeolite type. However, cross-laboratory comparisons should be treated very carefully, because HF is sensitive to how gas-adsorption experiments and the BET, t-plot, and total-pore-volume analyses are performed. Although there is some indication that HF can be used to rank the catalytic performance of different mesoporous zeolites,101 its generality is not yet confirmed.18 The optimal balance between the degree of mesoporosity and microporosity may be very different for different applications and the assumption that a higher HF will lead to better catalytic performance is questionable.

thumbnail image

Figure 8. Contour plot of the hierarchical factor (HF) of different zeolite types prepared by using different methods. Reprinted with permission from Ref. 101.

Download figure to PowerPoint

3. Mesoporous Zeolite Y

  1. Top of page
  2. 1. Introduction
  3. 2. Preparation Strategies of Mesoporous Zeolites
  4. 3. Mesoporous Zeolite Y
  5. 4. Summary and Outlook
  6. Acknowledgements
  7. Biographical Information
  8. Biographical Information
  9. Biographical Information

Much progress has also been made in the field of hierarchical zeolite Y over the last decade. Owing to the great relevance of zeolite Y to industrial catalysis, this section will try to provide a detailed comparison of the different preparation approaches, the catalytic applications that have been demonstrated in laboratories, and the most up-to-date progress in the commercialization efforts.

Zeolite Y is a synthetic zeolite with the faujasite (FAU) framework structure, which features 3D interconnected micropores. The pore openings are delimited by 12 tetrahedrally coordinated Si or Al atoms (linked through O atoms), which afford a relatively large diameter of about 7.4 Å. The inner cavity, which is surrounded by 10 sodalite cages, has a diameter of about 12 Å. Because of its large pore size, strong Brønsted acidity, and high hydrothermal stability, zeolite Y quickly superseded zeolite X in FCC in the early 1960s and still remains the active component of FCC and many other petroleum processing catalysts, for example, hydrocracking catalysts.1

3.1 Dealumination

Despite the supremacy of Y zeolites in petroleum refining, soon after its debut in FCC catalysis, diffusion limitations were recognized as a major problem that needed to be resolved to unleash its full potential.102105 Post-synthetic (“top-down”) modifications of zeolite Y, such as steaming and acid or chemical dealumination techniques, as briefly mentioned in the previous section, have been used to improve the hydrothermal stability of zeolite Y and may also have generated secondary mesoporosity. The ultrastabilization of ammonium-ion-exchanged zeolite Y to prepare USY is a well-known example of such a process. Heating NH4Y in the presence of steam (either added or evolved from the adsorbed water in the zeolite) causes hydrolysis of the Al[BOND]O[BOND]Si bonds. Then, Al atoms are expelled from the framework and atom-sized vacancies are created. Subsequently, some of these vacancies are filled (or “healed”) by mobile Si atoms from the amorphous region of zeolite crystals or particles, while the other vacancies coalesce to form larger cavities or pores, which typically appear as “mesopores” by N2 or Ar gas-adsorption analysis (Figure 9).106 Although some of the microporosity and framework Al species are lost during the ultrastabilization process, the resultant USY is much more hydrothermally stable. The extra-framework Al species are deposited onto the internal and external surfaces of USY zeolite, which can be removed by mild acid leaching using either inorganic or organic acids to enhance the porosity of the USY zeolite. Although this process is not typically used in the making of USY zeolite for FCC catalysis, repetitive steaming and acid leaching have been commonly utilized to prepare USY zeolites with much higher mesoporosity for other applications, such as hydrocracking. The extra pore volume and surface area serve as favorable support for the noble-metal catalyst particles that provide hydrogenation function of the bi-functional hydrocracking catalyst.107109

thumbnail image

Figure 9. Schematic representation of the formation of mesopores in a hydrothermal dealumination process. Adapted with permission from Ref. 106.

Download figure to PowerPoint

Although there are indications that mesoporosity generated by steaming may lead to enhanced accessibility and higher conversions of larger molecules,110 electron-tomography studies,111113 as well as a combination of nitrogen-physisorption and mercury-intrusion studies, have shown that some of the mesoporosity generated by steaming was, in fact, cavities that were entrapped within the zeolite crystals and were only connected to the crystal surface through micropores. Such meso-scale cavities do not help to improve the intracrystalline diffusion of molecules through the crystals. Furthermore, pulsed field gradient (PFG)-NMR studies showed that the intracrystalline diffusion of n-octane and 1,3,5-triisopropylbenzene through USY were not affected by the presence of the “mesopores” that were generated by steaming.114 The discrepancy between these studies may lie in the subtle differences in the steaming processes and the following acid-leaching processes that are used for the preparation of samples that lead to different types of mesoporosity (open mesopores or entrapped meso-scale cavities), which suggests that steaming techniques may not be a reliable approach to the desirable open mesopores.

More recently, there have been efforts to combine framework desilication with dealumination by ammonium hexafluorosilicate (AHFS) or steaming to generate USY zeolite with defect-guided mesoporosity.115, 116 The cracking of 1,3,5-triisopropylbenzene and vacuum gas oil has shown improvements in product selectivity, which was attributed to the introduced mesoporosity.

3.2 Surfactant-templated Y with intrachrystalline mesopores

As briefly mentioned in the previous section, surfactant-templated post-synthetic modification is a unique “top-down” approach that involves the single-step treatment of zeolites with a surfactant in mildly basic solution.92, 117 Specifically, the original invention described the hydrothermal treatment of a commercial USY with a Si/Al ratio of about 15, that is, Zeolyst CBV720, with a solution of CTAB in 0.37 M aqueous NH4OH (or 0.09 M TMAOH, that is, tetramethylammonium hydroxide) at 150 °C for 10–20 h. Removal of the surfactant templates by careful calcination first in nitrogen then in air exposes the uniformly distributed intracrystalline mesopores, as shown in the TEM images of the treated zeolite (Figure 10).93 Nitrogen-physisorption isotherms also show distinct uptake slightly below P/P0=0.4, which corresponds to a sharp Barrett–Joyner–Halenda (BJH) pore-size distribution118 at about 4 nm (the size of CTAB micelles).93 The validity of both the synthesis and the intracrystalline nature of the mesoporosity were later verified by an independent group.119 The direct introduction of well-controlled mesoporosity into zeolite crystals may happen through a crystal-rearrangement mechanism (Figure 11).93 This structural reorganization is only possible if a cationic surfactant is present when the base opens the Si[BOND]O[BOND]Si bonds to form negatively charged Si[BOND]O species. This process allows the needed interactions between the surfactant and the zeolite and prevents the dissolution of the crystals. Typically, almost complete recovery is observed during this mesostructuring process. SEM images (Figure 12 and the Supporting Information, Figure S1, in Ref. 93) showed that there was only one phase and no notable morphology changes were observed. The co-existence of both mesoporosity and crystallinity within the crystal boundaries is evident in the TEM images (Figure 10 and Figure 12). The remarkably high hydrothermal stability of the mesoporosity also supports the intracrystalline nature of the surfactant-templated mesopores.93

thumbnail image

Figure 10. TEM images of a) zeolite CBV720 and b) mesostructured zeolite Y. The higher-magnification micrographs of various mesostructured zeolite-Y crystals clearly show intracrystalline mesoporosity and crystal lattice fringes. Reprinted with permission from Ref. 93. Scale bars: a) 100 (left) and 20 nm (right), b) 500 (center) and 50 nm. Copyright 2012 Royal Society of Chemistry.

Download figure to PowerPoint

thumbnail image

Figure 11. Schematic representation of the speculated mechanism for the formation of surfactant-templated mesopores in zeolite: a) Original zeolite Y; b) Si[BOND]O[BOND]Si bond-opening/reconstruction in basic media; c) crystal rearrangement to accommodate the surfactant micelles; and d) removal of the template to expose the introduced mesoporosity. Reprinted with permission from Ref. 93. Copyright 2012 Royal Society of Chemistry.

Download figure to PowerPoint

thumbnail image

Figure 12. a) SEM and b–f) TEM images of the mesostructured Y zeolites. In (c), a single crystal of mesostructured zeolite shows crystalline lattice fringes and mesoporosity (holes) and, in (d), an ultramicrotomed mesostructured zeolite crystal shows both crystallinity and mesoporosity. e, f) Two TEM images of the same area of a mesostructured zeolite as obtained at two different foci to better visualize the two features of this material, that is, crystallinity and mesoporosity. Scale bars: a) 1 µm, b) 500 nm, c, d) 50 nm, e) 20 nm, f) 20 nm. Reprinted with permission from Ref. 93. Copyright 2012 Royal Society of Chemistry.

Download figure to PowerPoint

The cracking of triisopropylbenzene over the proton-form mesoporous zeolite Y (Meso-HY),92, 117 the starting CBV720 (HY), and Al-containing MCM-41 (AlMCM-41) suggested that, whereas both Meso-HY and HY showed much higher activities than AlMCM-41, owing to their higher zeolitic acidity, Meso-HY maintained higher activity for a longer period of time than HY, because the extra mesoporosity that was introduced by surfactant templates helped to slow down the deactivation by coke build-up. The products selectivity (Meso-HY versus HY) was also shifted to higher yields of 1,3-diisopropylbenzene and lower yields of cumene and benzene, which was attributed to the suppression of over-cracking by allowing more of the larger product (1,3-disiopropylbenzene) to diffuse out of the zeolites through the open mesopores before they were cracked further into the lower-molecular-weight products (cumene and benzene).

Strong interest in the catalytic and separation applications of mesoporous zeolites that were developed by using the surfactant-templated post-synthetic modification approach prompted the foundation of Rive Technology, Inc. in 2006 to scale up and commercialize this new technology (referred to as “molecular highway” technology).94 Since then, much progress has been made. Firstly, the technique has been extended from high Si/Al-ratio USY zeolite (and mordenite, ZSM-5) to NaY,120, 121 NaX, and NaA122 zeolites with much lower Si/Al ratios (1–3) by the incorporation of a pre-treatment step with a mild acid (with either organic or inorganic acids or mixtures of both). Careful acid treatment to selectively open some of the Al[BOND]O bonds with limited removal of Al atoms from the zeolite framework creates some defects and weakens the rigidity of the structure. Subsequent treatment of the acid-treated zeolite with cationic surfactants such as CTAB in a basic solution (e.g., NH4OH or NaOH solution) at elevated temperatures (typically below 100 °C) for as short as 1 h affords mesoporous NaY (or NH4Y) with very similar characteristics to Meso-HY (Figure 10 and Figure 12).93 Electron-diffraction patterns and field-emission scanning electron microscopy (FE-SEM) images further confirmed the co-existence of mesoporosity within the zeolite crystals and the well-controlled size and shape of the mesopores, as well as their uniform distribution throughout the crystals (Figure 12 and 13). Remarkably, the mesoporous zeolite Y showed high hydrothermal stability that was very similar to that of conventional zeolite Y. After steam treatment at 788 °C and 100 % steam for 4 h, the mesoporous zeolite Y still retained about 63 % of its original micropore volume and the total mesopore volume remained the same at 0.16 cc g−1 (Table 1).93 Temperature-programmed ammonia desorption (TPAD) confirmed that the mesoporous USY (as prepared by ultrastabilization of the ammonium-exchanged meso-NaY) had a total acidity (1.18 mmol) that was very close to that of conventional USY (CBV500, 1.25 mmol).93

thumbnail image

Figure 13. FE-SEM images of the untreated NaY zeolite (top) and the mesostructured Y-zeolite crystals (bottom). Scale bars: a) 1 µm, b, c) 500 nm, d) 200 nm.

Download figure to PowerPoint

Table 1. Micropore volume, mesopore volume, and unit-cell size of the samples described in Figure 8 and Figure 9 of ref. 93. Adapted with permission from Ref. 93. Copyright 2012 Royal Society of Chemistry.
Micropore volumeMesopore volume[a]BET surfaceExternal surfaceUnit-cell
(pore size 0–20 Å) [cc g−1](pore size 20–135 Å) [cc g−1]area [m2 g−1]area [m2 g−1]size [Å]
  1. [a] The mesopore range 20–135 Å was chosen to capture the characteristic mesoporosity that was introduced by using this technique. [b] The zeolites contained about 5 % rare-earth oxides. [c] Steaming was performed at 1450 °F (788 °C) under 100 % steam for 4 h.

zeolite NH4Y0.380.039702224.70
mesostructured zeolite Y[b]0.370.1691624324.67
mesostructured zeolite USY[b]0.270.1681215224.55
steamed mesostructured zeolite USY[b,c]0.240.1666113624.35
conventional USY (CBV500)0.320.048577524.55

One of the significant advantages of the surfactant-templating approach over other methods, such as desilication, is that the degree of mesoporosity can be increased without significantly compromising some important zeolite features, such as the Si/Al ratio, mesopore-size distribution, and the physical integrity of zeolite crystals, with retention of high recovery yields. The introduced mesopores have a very sharp mesopore-size distribution, the modal size of which varies in excellent accordance with the size of the surfactants that are used (Figure 14). Non-ordered mesostructures are typically observed in the highly crystalline meso-NaY (and the meso-USY derivative).93 However, if enough mesoporosity is introduced for the CTAB micelles to self-assemble, low-angle X-ray diffraction peaks that are similar to those observed in MCM-41 can be observed in the low 2θ range (1.5–5°), in addition to the typical diffraction peaks of zeolite Y at higher angles (Figure 15).92, 117

thumbnail image

Figure 14. a) Nitrogen-adsorption isotherms, b) nonlocal density functional theory (NLDFT) pore-size distributions (with normalized peak heights to better demonstrate the correspondence between the modal mesopores size and surfactant size), and c) a linear correlation between the modal mesopore diameters and the carbon numbers of the long alkyl chains in the surfactants that were used to prepare a series of mesostructured NH4-Y zeolites under the same reaction conditions. C8, C10, C12, C14, C16, and C18 denote octyl, decyl, dodecyl, tetradecyl, hexadecyl, and octadecyl trimethylammonium bromides, respectively. Adapted with permission from Ref. 93. Copyright 2012 Royal Society of Chemistry.

Download figure to PowerPoint

thumbnail image

Figure 15. XRD pattern of a mesostructured CBV720 (the one with the lowest micropore volume shown in the inset), which shows the (100), (110), and (210) peaks that are owing to hexagonal ordering of the mesopores (2θ<5°) and peaks that are characteristic of zeolite Y (2θ>5°). Inset shows that the microporosity/mesoporosity parameters in the surfactant-templated mesoporous Y zeolite can be fine-tuned in a linear trend by varying the mesostructuring conditions.

Download figure to PowerPoint

Subsequently, the process of producing the mesostructured zeolite Y by the surfactant-templated method was scaled up from the gram level to the kilogram level in a pilot catalyst-development plant and the mesostructured USY was also formulated in a FCC catalyst matrix (that comprised a binder and a clay, representative of commercial FCC catalysts) by using the spray-drying technique to create the first FCC catalyst with mesostructured zeolite Y. Then, the catalysts were deactivated by fluidized steaming at 788 °C for 8 h to simulate deactivation in a FCC unit. The unit-cell constants of the steam-deactivated catalysts that contained conventional and mesostructured USY zeolites were almost identical (24.26 and 24.27 Å, respectively). However, the latter material had a higher external surface area, as estimated by t-plot analysis (70 versus 51 m2 g−1). The additional 20 m2 g−1 external surface area came from the mesopores in the mesostructured zeolite because both catalysts had the same formulation (zeolite, binder, and matrix contents). The catalytic performance of the FCC catalyst was tested against a FCC catalyst that was made from conventional USY and deactivated under the same conditions for the cracking of vacuum gas oil (VGO) in an Advanced Catalyst Evaluation (ACE) test unit. Significantly higher yields of the valuable transportation fuels (gasoline and diesel) were obtained by using the FCC catalyst with the mesostructured USY zeolite although the yields of undesirable coke and unconverted bottoms were reduced (Figure 16).93, 123 These results are in line with what would be expected from the alleviated diffusion limitation in the microporous structure of the conventional Y zeolite. Thus, the open mesoporous network of the new zeolite provided better accessibility for heavy hydrocarbon molecules, thereby allowing them to diffuse in and out of the pores faster and more easily, increasing their conversion into more-valuable intermediate-molecular-weight hydrocarbons, and decreasing their deposition as coke.

thumbnail image

Figure 16. Catalyst evaluation of two FCC catalysts that were prepared and deactivated under the same conditions (788 °C in 100 % steam for 8 h), one of which contained a conventional zeolite USY (), whereas the other contained a mesostructured zeolite USY (- - - -- - - -). The catalyst evaluation was performed in an ACE unit at 527 °C by using a VGO feedstock. The lines were fitted by a kinetic lump model. Adapted with permission from Ref. 93. Copyright 2012 Royal Society of Chemistry.

Download figure to PowerPoint

Thirdly, the process has been successfully scaled up further to the tonnage level in a commercial zeolite manufacturing plant by using existing equipment, starting from NaY zeolite.123 So far, about 500 tons of FCC catalysts with mesoporous zeolite Y have been made in a commercial catalyst manufacturing plant over three separate production runs.94, 124, 125 Concurrently, major improvements have been made to both the raw materials and operating cost structures of the surfactant-based process. A larger than 10-fold decrease in manufacturing costs has been realized without capital investment. As a result, past reservations77, 126 about the high costs of the surfactant-based process were overcome. In addition, research at Rive Technology also showed that it was possible to recover and reuse the surfactant to further decrease the cost.127 Notably, the flexibility of this process allowed easy optimization of the zeolite properties for targeted applications. The first two batches of FCC catalyst with mesoporous zeolite Y were supplied to two refineries and showed successful operation in their FCC units, which confirmed both the hydrothermal and mechanical stability of the mesoporous zeolite and the yields of the cracking products observed in the lab catalytic testing.

Specifically, during 2011, the new mesostructured zeolite-Y-based FCC catalysts were successfully introduced into a North American refinery.94, 123, 124 Comparing the performance of the incumbent equilibrium catalyst (containing conventional zeolite) before the trial and the equilibrium catalyst in the unit after the catalyst inventory was replaced with the new catalyst containing the mesostructured Y zeolite at 66% change-out, an increase in bottoms upgrading and significant coke reduction were observed. The projected economic uplift for the refinery was estimated to be between $0.60–1.17/bbl (barrel) of the FCC feed.

Recently, a second refinery trial of a second-generation mesostructured zeolite Y was completed with outstanding success.125 Prior to this commercial trial, both the fresh incumbent catalyst and the FCC catalyst with the mesostructured Y zeolite were impregnated with same levels of nickel and vanadium and then deactivated by cyclic propylene steaming (CPS)128 and then tested in an ACE unit under identical conditions using the refinery FCC feed. The results at a constant conversion of 75 wt. % are shown in Table 2.125 The mesostructured zeolite Y was able to: 1) Increase the yields of gasoline and diesel [light cycle oil (LCO)]; 2) increase the production of valuable light olefins (propylene and butenes); and 3) remarkably decrease the yield of coke. Most importantly, these advantages were associated with almost no penalty in catalyst activity (as evidenced by the very close catalyst-to-oil ratios that were necessary to achieve 75 % conversion). The estimated economic uplift based on the laboratory test results as shown in Table 2 was about $2.00/barrel of FCC feed for the specific refinery. At the end of the commercial trial, the additional revenue that was delivered to the refinery by replacing the incumbent catalyst with the FCC catalyst that contained the new mesostructured zeolite Y, was estimated to be over $2.50/bbl of the FCC feed (Figure 17).125

Table 2. Comparison of the FCC catalyst (MH-1) with the new mesostructured zeolite Y to the incumbent catalyst that is used in a commercial refinery. Reprinted with permission from Ref. 125. Copyright 2013 American Fuels & Petrochemical Manufacturers.
CatalystIncumbent catalystRive catalyst (MH-1)
C/O ratio6.26.4
Conversion [wt. %]75.075.0
Yield [wt. %]
dry gas3.373.19
liquefied petroleum gas15.6415.94
 propane0.830.83
 propylene4.674.77
 butanes3.853.95
 butenes6.296.39
gasoline50.3951.33
LCO19.0919.14
bottoms5.915.86
coke5.604.54
thumbnail image

Figure 17. Observed trends during a trial at Alon’s Big Spring, Texas refinery: a) Increased feed rate by 700 BPSD (barrel per stream day); b, c) increased production of gasoline and LCO (the big spikes in the plant data were owing to process interruptions and not to the catalyst); d) an incremental value uplift owing to the change-out of the incumbent catalyst for Rive’s MH-1 catalyst that contained mesostructured Y zeolite. Adapted with permission from Ref. 125. Copyright 2013 American Fuels & Petrochemical Manufacturers. •) Rive catalyst, ♦) conventional catalyst.

Download figure to PowerPoint

In April 2013, Rive Technology began the ongoing supply of a commercial FCC catalyst with mesoporous zeolite Y that was made by surfactant-templated post-synthetic modification, which represents the first large-scale industrial catalytic application of hierarchical zeolites.

3.3 Surfactant-templated zeolite/mesophase composites

As mentioned in the first section, another surfactant-templated post-synthetic modification approach was first reported by Goto et al. in 200296 and later by Ivanova et al. in 2004.95 This process involves two steps: 1) Partial or complete destruction of the ZSM-5 or mordenite structure, followed by 2) hydrothermal treatment with CTAB and NaOH over a longer period of time, during which a pH adjustment was performed. Compared to the single-step process described above, the materials that resulted from the two-step process were very different. As suggested by the proposed mechanism of the two-step approach (Figure 18),129 the dissolved zeolite species are re-condensed and re-assembled around the surface of the surfactant micelles and are deposited back onto the crystal surface as a separate mesoporous molecular sieve phase, if the pH value of the reaction mixture is adjusted to about 8.5. The first step also generates some mesoporosity in the zeolite crystals by desilication, similar to the other desilicated mesoporosity as described in the previous section. SEM images show the co-existence of both zeolite crystals and a mesoporous phase that is templated by the surfactant (Figure 19).129 The bimodal mesoporosity, as suggested by the two distinct nitrogen uptakes at P/P0≈0.35 and 0.95 in the isotherms (Figure 20), is also consistent with the existence of two types of mesoporosity, that is, well-controlled mesopores with sizes of about 4 nm (templated by the surfactant) in the deposited mesophase and broad mesopores that are created by desilication in the remaining zeolite.95 Presumably, the surfactant-templated mesophase is similar to mesoporous materials that are formed by condensation from the filtrate that is obtained from the alkaline dissolution of the zeolite (first step), which may contain fragments of the zeolite and, therefore, shows enhanced Brønsted acidity compared to conventional Al-MCM-41.130 The depolymerization/recrystallization (DR) approach (as mentioned in Section 2) shares many characteristics with the dissolution/recrystallization approach described above, except that glycerol is used for the depolymerization of zeolite instead of NaOH. Another major difference is that, if a surfactant, such as CTAB, was used in the hydrolysis step of DR approach before recrystallization, no mesopore-templating effect was observed. The hierarchical Y zeolite showed a broad mesopore-size distribution from 10–100 nm.100 TEM analysis clearly showed the composite nature of the materials, that is, a mixture of recrystallized Y-zeolite nanocrystals and an amorphous mesoporous phase.100

thumbnail image

Figure 18. Schematic representation of the recrystallization procedure that leads to different types of materials. Reprinted with permission from Ref. 129. Copyright 2013 Royal Society of Chemistry.

Download figure to PowerPoint

thumbnail image

Figure 19. Morphology and texture of recrystallized zeolites a) RMOR-1, b) RMOR-2, and c) RMOR-3 by using TEM and of d) RMFI-1, e) RMFI-2, and f) RMFI-3 by using SEM. Reprinted with permission from Ref. 129. Copyright 2013 Royal Society of Chemistry.

Download figure to PowerPoint

thumbnail image

Figure 20. Nitrogen-adsorption/desorption isotherms at 77 K over starting (◊, ○, □) and recrystallized (♦, •, ▪) mordenites. Reprinted with permission from Ref. 95. Copyright 2004 IUPAC.

Download figure to PowerPoint

3.4 Desilicated zeolite Y

There has also been progress in the template-free desilication approach to introduce mesoporosity into zeolite Y. Desilication had only been applicable to zeolites such as ZSM-5, mordenite, and ferrite, with optimal Si/Al ratios of between 25–50. Later, the Si/Al range was expanded to 10–∞. USY zeolites typically have Si/Al ratios within this range, for example, Zeolyst CBV720 (about 15), CBV760 (about 30), and CBV780 (about 40). In 2010, de Jong et al.131 published the first examples of mesoporous zeolite Y that were prepared by desilication. Starting from a commercial USY (CBV760) at a Si/Al ratio of 30 (named HY-30), room-temperature treatment of the zeolite with dilute NaOH solution (0.05 M and 0.10 M) for a short period of time, followed by quenching of the reaction by pH neutralization with 1 M H2SO4 solution, yielded the desilicated mesoporous Y zeolite. However, it is not clear whether pH neutralization with H2SO4 caused re-precipitation of the dissolved silicates (similar to the re-deposition of dissolved silicates/aluminosilicates in the two-step approach of Ivanova et al. as described above). The dried zeolite was further subjected to mild steam calcination, the purpose and effects (synergistic or antagonistic) of which on the desilicated zeolite properties were not described. Nonetheless, the final products exhibited significantly enhanced mesoporosity over the starting HY-30. Bimodal mesoporosity was evident from both nitrogen-physisorption characterization and 3D TEM (or electron tomography, ET) analysis. Desilication generated more small mesopores (about 3 nm) than larger mesopores (about 30 nm), which led to a significant increase in the mesopore surface area from 213 m2 g−1 in the starting HY-30 to 339 m2 g−1 in HY-A and 443 m2 g−1 in HY-B, at the expense of significant drops in micropore volume and, possibly more so, of the crystallinity (by XRD; Table 3 and Figure 21).131 Desilication also caused a decrease in the Si/Al ratio from about 28 to 25 and 21, respectively, which suggested that there was between 10 % and 25 % yield loss to the solution, respectively, assuming that only silica dissolved from the zeolite and that no smaller crystals were lost during the filtration process. The unit-cell constant a0 increased from 24.28 to 24.30 and 24.31 Å, suggesting that some silica had dissolved from the framework. Notably, the total acidity, as measured by TPAD, stayed the same as that of the starting HY-30 at about 0.30 mmol.

Table 3. Textural properties of the zeolite Y samples. Reprinted with permission from Ref. 131.
SampleSmeso[a]Vmicro[b]Vmeso[c]Vs-meso[d]Vl-meso[e]Vtotal[f]Pore diameter[g]
[m2 g−1][cm3 g−1][cm3 g−1][cm3 g−1][cm3 g−1][cm3 g−1][nm]
SmallLarge
  1. [a] Mesopore surface area. [b] Micropore volume. [c] Mesopore volume (2–50 nm pores). [d] Volume of the small mesopores (2–8 nm). [e] Volume of the large mesopores (8–50 nm pores). [f] Total pore volume. [g] Determined from the pore-size distribution (PSD); see the Supporting Information, Experimental Section of Ref. [131].

HY-302130.210.160.070.090.4528
HY-A3390.160.250.140.110.512.727
HY-B4430.070.370.230.140.553.127
thumbnail image

Figure 21. XRD patterns of the parent HY-30, base-leached HY-A (0.05 M NaOH), and HY-B (0.10 M NaOH). Reprinted with permission from Ref. 131.

Download figure to PowerPoint

To establish the hydrocracking performance of the zeolites, HY-30 and HY-A were loaded with 0.3 wt. % Pt and used to hydrocrack the model compound n-hexadecane. Over mesoporous Pt/HY-A, the product yields showed an ideal C6/C10 ratio of about 1:1 over the whole temperature range tested, whereas, for conventional Pt/HY-30, the ratio deviated quickly away from 1, owing to overcracking. Hydrocracking of another model compound, squalane (branched C30 alkane), at 230 °C showed that better product symmetry was achieved over Pt/HY-A than over Pt/HY-30, thus also suggesting that secondary cracking was alleviated, possibly through improved product diffusion out of the zeolite before overcracking into the lighter products (Figure 22).131 The authors took another step closer to real applications by forming NiMoS2/HY-A/alumina catalyst extrudates and testing them with a pretreated commercial vacuum gas oil feed. Compared to a state-of-the-art commercial catalyst, the catalyst that was made from the HY-A zeolite yielded dramatically higher amounts of middle distillates (diesel and kerosene), with significantly smaller amounts of the less-desirable naphtha and coke. The activity of the catalyst with mesoporous zeolite was only slightly lower than that of the commercial catalyst.

thumbnail image

Figure 22. Catalytic testing of the parent HY-30 and base-leached HY-A that were loaded with 0.3 wt. % Pt. a) Conversion and b) selectivity for the hydrocracking of n-hexadecane over Pt/HY-30 (×) and Pt/HY-A (▴). c) Hydrocracking of squalane over Pt/HY-30 (▪) and Pt/HY-A (▪). Reprinted with permission from Ref. 131.

Download figure to PowerPoint

In early 2011, Garcia-Martinez et al.132 filed a patent application for the desilication of zeolite Y with a Si/Al ratio of below 10 by incorporating a pre-treatment step with mild acid prior to desilication. Examples included the desilication of NaY (Zeolyst CBV100) zeolite with a Si/Al ratio of about 2.5. Mild dealumination by citric acid slightly increased the Si/Al ratio of the zeolite to about 3–5 and subsequent desilication restored the Si/Al ratio to almost the original level. Significant mesoporosity (with good retention of microporosity) was observed by TEM analysis and argon-physisorption (Figure 23 and Figure 24), albeit with much larger and less-controlled mesopore sizes compared to the surfactant-templated mesoporosity described above.

thumbnail image

Figure 23. TEM images of mesoporous zeolite Y that was treated with citric acid and then desilicated. Scale bars: a) 200 nm, b) 50 nm, c) 20 nm.132

Download figure to PowerPoint

thumbnail image

Figure 24. Ar-sorption isotherms of starting NaY (▪), surfactant-templated mesoporous NaY (▴), and acid-treated/desilicated mesoporous NaY (▪).132 The latter two samples were prepared under the same reaction conditions (e.g., temperature, time, NaOH dose), except for the presence of CTAB in the former but not in the latter.

Download figure to PowerPoint

Later that year, Verboekend et al.133 also published the desilication of zeolite Y after first performing an acid treatment on the starting material. A series of post-synthetic treatments were applied to pristine zeolite NH4Y (Zeolyst CBV300, named “P-Y”) and three dealuminated USY zeolites (CBV500, named “USY1-P”; CBV720, named “USY2-P”; and CBV760, named “USY3-P”) by using a variety of chemicals, such as H4EDTA (EDTA=ethylenediaminetetraacetic acid), Na2H2EDTA, citric acid, HCl, NaOH, and TPAOH, as well as the sequential combination of some of these treatments. Their main findings were that, for pristine Y zeolite, sequential dealumination by the acid, followed by desilication in alkaline solution, resulted in the introduction of mesoporosity at the expense of a decrease in XRD crystallinity and material loss; however, the micropore volumes stayed relatively high, thus suggesting that the treatments disrupted the long-range ordering of the atoms in the zeolite crystals, but retained much of the coordination of the local zeolitic subunits, for example, sodalite cages and super cages. Subsequent mild treatment of the mesoporous Y zeolite with Na2H2EDTA solution helped to remove the Al-rich debris, thereby freeing the porosity and enhancing crystallinity (Figure 25).133 The authors also reported that dealuminated USY zeolites were more sensitive to the desilication conditions and that severe acid treatments of the USY zeolites allowed the introduction of more mesoporosity into the USY zeolites. However, the loss of crystallinity and microporosity were significant. The addition of tetraalkylammonium salts to the alkaline solution helped to preserve more micropore volume.

thumbnail image

Figure 25. Strategies for the design of hierarchical FAU zeolites by using post-synthetic modifications. After the desilication of Al-rich zeolites, the removal of any remaining debris by washing with mild acid is crucial. On the other hand, upon alkaline treatment of Si-rich zeolites, the inclusion of pore-growth moderators is highly beneficial. Reprinted with permission from Ref. 133.

Download figure to PowerPoint

Catalytic evaluation of the proton forms of the parent zeolite, termed “P-H”, and two modified zeolites, termed “P-DA4-AT1-H” and “P-DA4-AT1-AW1-H”, was performed by the liquid-phase alkylation of benzyl alcohol with toluene, which showed that, upon the introduction of mesoporosity (P-DA4-AT1-H), there was a slight increase in activity. After the final AW1 wash, to remove the Al-rich debris that possibly blocked the mesopores, the activity increased by about 55 % (from 53 % to 84 % conversion) over 40 min reaction time. Pyrolysis of low-density polyethylene was also performed over the modified USY zeolites to demonstrate the beneficial effects of additional mesoporosity and also the importance of maintaining adequate activity (Figure 26).

thumbnail image

Figure 26. Catalytic evaluation of Y and USY zeolites in a) the alkylation of toluene with benzyl alcohol (BA) and b) the pyrolysis of low-density polyethylene (LDPE). Inset in (b) shows derivative of the thermogravimetric (DTG) profiles. Reprinted with permission from Ref. 133.

Download figure to PowerPoint

Verboekend et al.134 also studied the effects of the so-called “pore-directing agents” (PDAs) on the desilication of a USY zeolite, that is, CBV760 (named “USY30” by the authors). Whereas negatively charged surfactants were found to be ineffective as PDAs, non-ionic surfactants, in particular amines, were able to enhance mesoporosity whilst preserving more of the microporosity of the modified zeolites. Alkylammonium cations, including the CTA+ cation that was used in the surfactant-templated approach described above, were found to be most effective as pore-directing agents. This result was attributed to the strong electrostatic affinity of the positively charged head groups with the negatively charged zeolite surface and the soluble species in the alkaline reaction mixture, which had also been recognized as one of the driving forces for the formation of surfactant-templated mesopores (Figure 20).93 Among the many alkylammonium salts that were tested, two stood out in the PDA screening experiments: TPA+ and CTA+ showed the highest indexed hierarchical factors (IHFs) of 0.67 and 0.54, respectively. Further studies revealed that, at proper concentrations, both TPA+ and CTA+ not only helped to preserve the microporosity whilst dramatically enhancing mesoporosity, but they also maintained the Si/Al ratio of the products and improved the yields. Such beneficial effects of CTA+ had previously been observed in the surfactant-templated approach described by Garcia-Martinez. Notably, the authors demonstrated the possibility of using a continuous process to achieve much higher productivity than the batch-wise lab preparations. In 2013, the catalytic cracking of vacuum gas oil over dealuminated-desilicated mesoporous Y zeolite on a fixed-bed Microactivity Test (MAT)135 reactor was reported.126 After ammonium exchange, the as-prepared mesoporous MY zeolite showed a micropore volume of 0.163 cc g−1, which was significantly lower than that (0.30 cc g−1) of the starting CBV300, and a mesoporosity (0.218 cc g−1) that was only marginally higher than that (0.209 cc g−1) of an equivalent conventional USY, that is, CBV760. Then, steaming at 750 °C for 5 h and 100 % steam was performed on MY. After further ammonium exchange and calcination at 500 °C for 3 h to convert the zeolite into its proton form, the zeolite was steamed again at 750 °C to afford HMY-S750C for testing. HMY-S750C showed almost zero microporosity (0.028 cc g−1) and similar mesoporosity (0.214 cc g−1) to MY, owing to the very severe hydrothermal treatments. The lanthanum-exchanged MY (with about 3.5 wt. % La2O3) was first converted into the proton-form zeolite by double calcination at 500 °C for 3 h and then steamed at 780 °C and 800 °C for 5 h in 100 % steam to result in LaHMY-S780 and LaHMY-S800, both of which showed slightly higher micropore volumes (0.053 and 0.045 cc g−1, respectively) than HMY-S750, possibly owing to the stabilization effect of lanthanum in the zeolite. However, the unit-cell constants (a0) of LaHMY-S780 (24.55 Å) and LaHMY-S800 (24.54 Å) were significantly higher than those of HMY-S750 (24.31 Å) and CBV760 (24.26 Å), which were also higher than that of a typical steam-deactivated rare-earth stabilized conventional USY (about 24.4 Å). It is unknown whether these unusually high steamed unit-cell constants for the LaHMY zeolites are the result of measurement issues, lower-than-believed deactivation severity, or of implausibly high steam stability for the desilicated mesoporous Y zeolites.

MAT testing of the above-described zeolite samples represents the first published example of evaluation of the desilicated Y zeolites for catalytic cracking. Although the over 20 % gap in mass balance in the reported MAT test results (Table 4 of ref. [126]) raises some questions about the quality of the tests, taken at face value, the data corroborate the advantages in bottoms upgrading and improved selectivity, such as higher diesel yield, less gases, and higher olefinicity in the liquefied petroleum gas fraction, that have been attributed to the presence of mesoporosity within zeolite crystals. The results also illustrate the well-known effects of rare-earth ion exchange (i.e., higher steamed zeolite unit-cell parameter) on the quality and yield of LCO (diesel). Prior unpublished work at Rive Technology by using more-realistic evaluation conditions on the desilicated mesoporous zeolite Y described elsewhere132 demonstrated similar trends in product yields, plus significant improvement in coke selectivity.

3.5 “Bottom-up” approaches to mesoporous Y

Since early 2000, there has also been some progress in the templated “bottom-up” approaches to mesoporous zeolite Y. In 2003, Tao et al.31 reported a synthesis of mesoporous zeolite Y by using carbon aerogels as templates, prepared first from the pyrolysis of the supercritically dried resorcinol-formaldehyde (RF) polymer aerogels. The synthesis of the zeolite involves three steps: 1) Infiltration of the precursors into the mesopores of the carbon aerogels; 2) synthesis of zeolite Y inside the inert mesopores of the templates; and 3) removal of the carbon aerogel templates by calcination to expose the mesoporous zeolite Y (Figure 27).32

thumbnail image

Figure 27. Growth of zeolite crystals in the uniform mesopores of carbon aerogel that consist of interconnected uniform carbon particles. The mesopores are large enough to allow the gel to be sufficiently concentrated and to allow growth to continue until the mesopores are filled. Adapted with permission from Ref. 32. Copyright 2003 American Chemical Society.

Download figure to PowerPoint

In 2010, Gu et al.136 reported another “bottom-up” approach to mesoporous zeolite Y or, more accurately, mesoporous silica that contains zeolite Y or sodalite fragments. By using a mixture of CTAB, tert-butanol (TBA), and trimethylbenzene (TMB) as the templates, the authors assembled nanocrystals with the zeolite-Y structure around the swollen micelles of CTAB to form a mesoporous silica that contained zeolite-Y or sodalite fragments (Figure 28). Fine-tuning of the synthesis conditions, such as silica source, TMB/CTAB molar ratio, and the amount of TBA, is crucial for the formation of the mesostructure. Although the results show that this method is applicable to a relatively wide range of bulk Si/Al ratios, the XRD crystallinity and micropore volumes of the synthesized materials are significantly lower than those of conventional NaY. Interestingly, the results of catalytic dehydration of 2-propanol and the TPAD data suggest much stronger acidity of one of the mesoporous samples than conventional NaY.

thumbnail image

Figure 28. Proposed route for the synthesis of hierarchical mesoporous zeolites. Reprinted with permission from Ref. 136. Copyright 2010 American Chemical Society.

Download figure to PowerPoint

More recently, in 2011, Xiao and co-workers published a synthesis of mesoporous zeolite Y by using a silylated quaternary ammonium surfactant, that is, N,N-dimethyl-N-octadecyl-(3-triethoxy-silylpropyl)ammonium bromide (TPOAB), as template.137 TPOAB was found to be well-dispersed in the synthesis mixture, owing to strong interactions of the silyl groups and the quaternary ammonium groups with the soluble aluminosilicate species in the crystal-growth gel. This is critical to the success of the “bottom-up” approach to mesoporous zeolite Y. Mesoporous material NaY-M showed very good crystallinity and micro-/mesoporosity. The TEM images showed the mesopores templated by the silylated surfactant (Figure 29). Then, palladium nanoparticles of similar size were loaded onto the H-form zeolites (mesoporous HY-M, conventional HY, mesoporous HBeta-M, and HZSM-5M) and the hydrodesulfurization (HDS) of a bulky model compound, 4,6-dimethyldibenzothiophene (4,6-DM-DBT), over the Pd-loaded zeolites (or γ-Al2O3) was tested on a fixed-bed continuous-flow reactor. Pd/HY-M showed the highest activity, 97.3 % conversion at 6 h versus 61.8 % conversion over Pd/HY and 34.1 % conversion over a conventional Pd/γ-Al2O3 catalyst. The higher activity of Pd/HY-M compared to those of Pd/HBeta-M and Pd/HZSM-5M was attributed to the larger micropore size of Y zeolite, which allowed the bulky sulfur-containing compound to access the acid sites inside the mesopores.

thumbnail image

Figure 29. a) XRD pattern, b) N2 isotherm, c) SEM image, and d) TEM image of NaY-M; the mesoporosity in the crystal is marked by red lines and white arrows. Reprinted with permission from Ref. 137. Scale bars: c) 1 µm, d) 20 nm. Copyright 2011 American Chemical Society.

Download figure to PowerPoint

4. Summary and Outlook

  1. Top of page
  2. 1. Introduction
  3. 2. Preparation Strategies of Mesoporous Zeolites
  4. 3. Mesoporous Zeolite Y
  5. 4. Summary and Outlook
  6. Acknowledgements
  7. Biographical Information
  8. Biographical Information
  9. Biographical Information

The discovery of surfactant-templated amorphous mesoporous materials, such as MCM-41, about 20 years ago raised many expectations because their open and tunable structures are ideal for the fast diffusion and processing of large molecules. However, their weak acidity and limited hydrothermal stability has restricted their use in applications for which such properties are required, such as in FCC and other important refining processes. Many efforts have been made to improve their acidity and hydrothermal stability by using various approaches and yet only limited success has been achieved.

Since the turn of the 21st century, a tremendous amount of effort has been devoted to introducing larger pores, typically mesopores of 2–50 nm in diameter, into conventional zeolites. The main motivation is to alleviate the diffusional limitation of molecules through the micropore system, in which strong acid sites catalyze many important reactions, such as the cracking of petroleum, isomerization, and desulfurization. Many preparation approaches have emerged, which can be generally categorized as either “bottom-up”/primary syntheses or “top-down”/post-synthetic modifications (Figure 2). Most “bottom-up” approaches involve the use of mesoscale templates of some sort, either hard templates, such as carbon black, carbon nanotubes or nanofibers, ordered or non-ordered mesoporous carbon, and carbon aerogels, or soft templates, such as cationic surfactants, silylated surfactants or polymers, cationic polymers, polymer aerogels, starch, and bacteria. Optimal interactions between the templates and the silicate or aluminosilicate species in the zeolite crystal-growth gels are essential to the success of the “bottom-up” approaches. Efforts to either convert preformed amorphous mesoporous silicate or aluminosilicate materials (or other precursors) into zeolitic materials or the assembly of protozeolitic sub-units or nanozeolite crystals into mesostructures have also achieved some success. In the “top-down” category, hydrothermal treatment and chemical dealumination have been known for about half a century. However, recent studies have suggested that not all of the “mesopores” are open to the external surface of the zeolite crystals, thus limiting the usefulness of these treatments to enhance the diffusion through zeolites. Desilication of zeolites with suitable Si/Al ratios could generate intracrystalline mesoporosity at the penalty of significant material loss, compromised zeolite integrity, and difficulty in material handling. Some “top-down” approaches also involve the use of templates, such as cationic surfactants. The single-step surfactant/alkaline treatments lead to single-phase mesoporous zeolites with intracrystalline mesoporosity, possibly through a crystal-rearrangement mechanism, whereas the two-step processes lead to the formation of a composite of desilicated mesoporous zeolite with amorphous surfactant-templated mesostructures through dissolution/re-deposition or recrystallization mechanism. The single-step approach bridges the gap between conventional zeolites and surfactant-templated amorphous mesoporous materials, which realizes the long-sought-after goal of combining the desirable properties of both types of materials, that is, large and tunable mesopores, high crystallinity, strong Brønsted acidity, and excellent hydrothermal stability, within a single-phase mesoporous zeolite.

The developments in the field of mesoporous zeolite Y nicely represent the “big picture” described above. Although many studies have shown the beneficial effects of secondary mesoporosity on the catalytic applications of mesoporous zeolites Y, most of which remain as academic research. Owing to the great importance of zeolite Y in industrial catalysis, significant developments and commercialization work have been performed at Rive Technology and its partners to commercialize surfactant-templated mesoporous zeolite Y (and other zeolites, such as ZSM-5, mordenite, X, and A) for applications in petroleum cracking, as well as other catalytic and non-catalytic applications. The zeolite-modification process has been scaled up to the commercial scale. FCC catalysts with proper chemical compositions, as well as physical and mechanical properties, have also been manufactured on the commercial scale, supplied to two separate commercial refineries, and have undergone successful trials that delivered significantly improved product yields and increased economic value to the refiners. Rive Technology began the ongoing supply of the very first commercial FCC catalyst with surfactant-templated mesoporous zeolite Y in early April 2013. The era of mesostructured zeolites in industrial catalysis has arrived!

Acknowledgements

  1. Top of page
  2. 1. Introduction
  3. 2. Preparation Strategies of Mesoporous Zeolites
  4. 3. Mesoporous Zeolite Y
  5. 4. Summary and Outlook
  6. Acknowledgements
  7. Biographical Information
  8. Biographical Information
  9. Biographical Information

The authors express their sincere appreciation to Dr. Barry Speronello (Rive Technology) for the many helpful discussions during the preparation of this Review.

Biographical Information

  1. Top of page
  2. 1. Introduction
  3. 2. Preparation Strategies of Mesoporous Zeolites
  4. 3. Mesoporous Zeolite Y
  5. 4. Summary and Outlook
  6. Acknowledgements
  7. Biographical Information
  8. Biographical Information
  9. Biographical Information

Dr. Kunhao Li joined Rive Technology, Inc. as a Project Leader in late 2008. Since then, he has been heavily involved in the improvement of Rive’s core technology in zeolite mesostructuring processes, zeolite and fluid catalytic cracking catalyst characterization, testing, and evaluation, as well as exploration and extension of the application areas of Rive’s mesoporous zeolites in chemical separation and other petroleum cracking and petrochemical processes. He obtained his doctorate in chemistry in 2006 at The George Washington University in Washington, D.C., USA. He then moved to Rutgers University, New Brunswick, USA, for his postdoctoral research on microporous metal-organic framework materials for applications such as hydrogen storage, gas adsorption and separation, and chemical sensing. His work at The George Washington University, Rutgers, and Rive has resulted in many publications in the form of original papers and reviews, book chapters, technical reports, patent applications, and patents.

Thumbnail image of

Biographical Information

  1. Top of page
  2. 1. Introduction
  3. 2. Preparation Strategies of Mesoporous Zeolites
  4. 3. Mesoporous Zeolite Y
  5. 4. Summary and Outlook
  6. Acknowledgements
  7. Biographical Information
  8. Biographical Information
  9. Biographical Information

Dr. Julia Valla is an Assistant Professor in the Chemical and Biomolecular Engineering Department at the University of Connecticut. Prior to her position in academia, she worked as a Project Leader at Rive Technology, Inc. on the development and catalytic evaluation of novel zeolites with ordered mesoporous structure for refinery applications. Her studies were focused on the diffusion limitations of zeolites and the kinetics and reaction pathways of heavy hydrocarbons in the micro/mesopore network within zeolites. She received her PhD in Chemical Engineering in 2005 from the Aristotle University of Thessaloniki in Greece in the field of in situ sulfur reduction in gasoline and diesel in the fluid catalytic cracking unit. Today, her research focuses on the modification of zeolite structure and the application of hierarchical pore zeolites in catalysis, adsorption, and energy. She has authored several patents and scientific papers in peer-reviewed journals and book chapters.

Thumbnail image of

Biographical Information

  1. Top of page
  2. 1. Introduction
  3. 2. Preparation Strategies of Mesoporous Zeolites
  4. 3. Mesoporous Zeolite Y
  5. 4. Summary and Outlook
  6. Acknowledgements
  7. Biographical Information
  8. Biographical Information
  9. Biographical Information

Dr. Javier Garcia-Martinez is the founder and Chief Scientist of Rive Technology, Inc. (Boston, USA), a venture capital-funded Massachusetts Institute of Technology (MIT) spin-off commercializing hierarchical zeolites for refining applications. He is also Director of the Molecular Nanotechnology Lab and a faculty member at the University of Alicante, Spain. He has published extensively in the areas of zeolites and nanotechnology and is co-inventor of more than 25 patents. His latest books are Nanotechnology for the Energy Challenge (2010, Wiley-VCH), The Chemical Element: Chemistry’s Contribution to Our Global Future (2011, Wiley-VCH) and Mesoporous Zeolites (2014, Wiley-VCH). He received the Europe Medal in 2005, the Silver Medal of the European Young Chemist Award in 2006, and the TR35 Award from MIT’s Technology Review magazine. Since 2010, he has been a member of the Council on Emerging Technologies of the World Economic Forum. He is a Fellow of the Royal Society of Chemistry and member of the Global Young Academy and IUPAC Bureau.

Thumbnail image of