Guest/host relationships in zeolite synthesis: ring-substituted piperidines and the remarkable adamantane mimicry by 1-azonio spiro [5.5] undecanes1
Introduction
The field of molecular sieve research has undergone some remarkable transformations during the career of Professor Lovat Rees. This is reflected in the fact that more researchers continue to enter the field and more powerful tools of computation and physical analysis enter the picture. Professor Rees has pioneered concepts in the study of molecular diffusion within molecular sieves, as well as having made significant contributions to such areas as zeolite synthesis and molecular modeling.
Over this same time period, just the number of molecular sieves which the researcher might choose to study has expanded greatly. Synthetic chemists continue to hit upon successful crystallizations of unknown crystalline materials. The boundaries of the dimensions of these materials continue to expand as well. In the last few years alone, two different laboratories have found routes to 14-ring silicate molecular sieves (or zeolites in these instances), a feat that had often been debated as to its very possibility 1, 2. The successful syntheses further served to illustrate the importance of guest structure-directing agents (SDAs), particularly in terms of the close fit within the eventual host framework. From a different standpoint, this very relationship is being explored computationally with a view that new molecular sieves can be designed `de novo' with the space-filling properties of the guest molecules a principal feature 3, 4.
In some instances, particularly when the organic synthetic resources are available, it has been useful to explore a family of organic guest molecules, each being a derivative based upon a parent theme. We have previously reported such explorations with respect to imidazoles [5], Diels–Alder polycyclic imines [6], and the tricyclo (5.2.1.0) decane family of derivatives [7]. Conversely, the same guest molecule can be used in changing inorganic environments to produce different attractive guest/host configurations 8, 9. We have recently attempted to predict the likely classes of silica-based molecular sieves to be formed as a function of both the size of the organo-cation guest and the degree of lattice substitution (principally Al or B for Si) 7, 10, 11. Five different regions for types of high silica zeolite crystallized are shown in Fig. 1. The mapping out of these five types can be considered to be a function of the lattice substitution for SiO2 (x-axis) and size of the organo-cation used in the synthesis (y-axis).
In regions of low lattice substitution, smaller guests (organo-cations) favor clathrate formation. Larger guests become too big to reside in a clathrate cavity and parallel, one-dimensional pore systems are generated. To date these have been pores comprised of openings circumscribed by tetrahedral rings of 10, 12, and now 14 T atoms. Both of these groups formed at low lattice substitution are typically rich in five-ring subunits in their framework construction.
With increasing lattice substitution, a greater number of four-rings are found in the zeolite structures. Parallel, one-dimensional pore systems and clathrates no longer form, giving way to multidimensional channel systems; these contain large pore openings when the guest molecules become too large to be stabilized by cavities with eight-ring openings (lower left region of map). The fifth group belongs to a less common guest/host system where the zeolite product is multidimensional and forms across the entire lattice substitution range. Tetrapropyl ammonium cations in the MFI structure is the best-known example.
In the recent modeling work in which the concept of `de novo' synthesis was developed, the guest molecule candidates (for a given zeolite framework and its void regions) are built up starting with the smallest alkane fragment, methane. In this study we have followed a related, albeit experimental, path, asking the question as to what zeolite structures will develop if we start with a simple ring and build up. Thus, we began with the heterocycle piperidine and then generated larger and more complex quaternary ammonium compounds.
Our interest was further stimulated during the course of the study owing to a rather unexpected result when one of us (YN) found a derivative capable of crystallizing a pure-phase MEL structure [12]. This synthesis established itself as the first route to producing ZSM-11 without an intergrowth of ZSM-5 being observed [13]. The organo-cation is shown in Fig. 2 and is a good example of both C and N substitution on the piperidine ring. The corresponding X-ray diffraction (XRD) pattern is in Fig. 3 and the weak, low angle peak described in XRD simulations can now be seen experimentally. The fit of this guest molecule was so favorable into ZSM-11 (the MEL structure) that the synthesis proved to be another example of the fifth type of crystallization relationship discussed above for Fig. 1. No lattice sensitivity was observed for this guest/host pair. Attempts to understand the beneficial interactions have recently come from van Konigsveld and co-workers using computational modeling [14]. These studies, and just the data confirming the high SDA occupancy per unit cell in the product, demonstrated that different components of the guest molecule provided `space stabilization' into different channels of the zeolite. More often, what we observe is the linear sequence of SDAs stabilizing a one-dimensional pore, one SDA molecule stacked behind the previous one down the pore [15], or the symmetric construction of cages stabilized by guests [16].
In this study, building up the size and position of both C and N ring substituents generated a variety of high silica zeolite structures. Some were novel structures. In other cases, the derivatives provided an unusual entry into high silica zeolites discovered previously using other types of SDA. Computational studies were carried out as well in an effort to bridge how different organo-cation conformations, or key structural details, may be decisive in directing the nucleation to a given zeolite structure. Here the hope was to go beyond the trends shown in Fig. 1 and ascertain more subtle control parameters for zeolite nucleation selectivity.
Section snippets
Template synthesis
Many template (or SDA) molecules are listed in tables which follow and which also list the zeolites made. These listings also provide references, and the exact organic syntheses can be found therein. The general method employed in most instances was to start with a C-substituted piperidine and then build up derivatization on N, usually through the addition of two substituents selected from methyl, ethyl, benzyl, etc. One set of derivatives has both positions linked by a methylene chain (–CH2–,
Single-ring derivatives including prior work in this area
If one starts with just piperidine and then builds up the size of the organic molecule from either what is placed on the nitrogen or by adding ring substituents at C, the gradation first forming ten-ring zeolites, then expanding into 12-ring products and finally into cavity-stabilized products with less obvious pore systems is what is generally observed. The large number of piperidine derivatives considered in this study is listed in Table 3. The derivatives will be described by their entry
Conclusions
The use of piperidine derivatives as SDAs or template molecules in zeolite synthesis has resulted in a rich production of varied high silica products. The extent of substitution, as well as location on the piperidine ring system, has proven important in determining the phases crystallized. Consistent with our previous concepts about the use of SDAs in high silica zeolite synthesis, the smaller derivatives and those with a long principal axis have led to clathrate and one-dimensional structures
Acknowledgements
The authors with to thank their colleagues I.Y. Chan, R.C. Medrud, and C.Y. Chen at Chevron Research and Technology Company who have worked on the various zeolites described in the paper. S.I. Zones also thanks Professor M.E. Davis and Katsu Tsuji at Caltech for much useful discussion on their results with the piperidine derivatives they have studied. The referees are gratefully acknowledged for their suggestions concerning the rewriting of the manuscript.
References (46)
- et al.
J. Mol. Catal. A: Chem.
(1997) Zeolites
(1989)- et al.
Micropor. Mater.
(1996) - et al.
Micropor. Mater.
(1997) - et al.
Micropor. Mater.
(1994) - et al.
Micropor. Mater.
(1997) - et al.
Nature
(1996) - P. Wagner, M. Yoshikawa, M. Lovallo, K. Tsuji, M. Tsapatsis, M.E. Davis, J. Chem. Soc. Chem. Commun. (1997)...
- et al.
Nature
(1996) - Y. Nakagawa, S.I. Zones, in: M.L. Occelli, H.E. Robson (Eds.), Synthesis of Microporous Materials, Van...
J. Am. Chem. Soc.
J. Phys. Chem.
Chem. Mater.
Micropor. Mater.
J. Am. Chem. Soc.
Cited by (65)
Structure-property relationships that influence platinum stability in all-silica or highly siliceous zeolites
2023, Microporous and Mesoporous MaterialsUnveiling the elusive role of tetraethyl orthosilicate hydrolysis in ionic-liquid-templated zeolite synthesis
2022, Materials Today ChemistryCitation Excerpt :Specifically, the isomers of tricyclodecane derivatives have shown preferential synthesis toward the zeolite framework of mordenite and ZSM-12, respectively, in the condition of Si/Al = 70 [16]. In addition, both experimental and molecular simulation studies have shown that cyclic and polycyclic quaternary/diquaternary ammonium compounds and piperidine derivatives led to unique synthesis toward zeolite frameworks (STF, AEI, MTT, TON, MTW, etc.) [17–22]. In addition to these conventional SDAs, ionic liquids (ILs) can serve as SDAs.
A review on recent advances in CO<inf>2</inf> separation using zeolite and zeolite-like materials as adsorbents and fillers in mixed matrix membranes (MMMs)
2021, Chemical Engineering Journal AdvancesSolvent-free synthesis of aluminosilicate SSZ-39 zeolite
2021, Microporous and Mesoporous MaterialsCitation Excerpt :For example, aluminosilicate SSZ-13 zeolite with 8-MR structure is an outstanding catalyst for NH3-SCR reaction, which has already been industrialized in the past decade [14–22]. More recently, it has been reported that aluminosilicate SSZ-39 zeolite with the AEI structure, which has a three-dimensional pore system with large cages (AEI cage, 0.85 × 0.99 nm) that are accessible by 8-MR windows (0.38 × 0.38 nm), exhibits the better catalytic performance than that of aluminosilicate SSZ-13 zeolite in NH3-SCR reaction, attracting the wide interest of many researchers [23–28]. In general, the aluminosilicate SSZ-39 zeolite would be prepared by interzeolite transformation of high-silica FAU zeolite under hydrothermal condition [25,29,30].
The structure-directing effect of organic amines in the multi-template/one-structure phenomenon of microporous crystal synthesis
2017, Microporous and Mesoporous MaterialsCitation Excerpt :Thus, it might be more appropriate to say that BDA, MPMD, and HDA play the same topological structure-directing role instead of the same structure-directing role [54]. In previous studies, non-bonding interactions (hydrogen bonding and van der Waals) between the inorganic host and organic guest have been used to estimate the templating effect or structure-directing effect in the synthesis of open-framework crystals [55–62]. The organic amines having similar non-bonding interactions to the inorganic species may result in the inorganic open-framework with the same topology.
- 1
Dedicated to Professor Lovat V.C. Rees in recognition and appreciation of his lifelong devotion to zeolite science and his outstanding achievements in this field.