A novel ‘build-bottle-around-ship’ method to encapsulate metalloporphyrins in zeolite-Y. An efficient biomimetic catalyst

One of the most attractive properties of zeolites are their well organized nanopores and nanochannels which serve readily as supporting hosts for various molecules. Encapsulation of catalytically active transition metal complexes inside the nanopores of zeolites, often referred to as ‘ship-in-a-bottle’ systems,1 has been believed to be one of the most promising strategies in the development of viable industrial catalysts.1–6 Extensive effort has been devoted to the synthesis and catalytic properties of zeolite entrapped metallocomplexes with such ligands as bipyridine, salen, polyamines and phthalocyanines.1–6 In most of these syntheses, a so-called ‘assemble-shipinside-bottle’ approach has usually been adopted in which the desired catalytic metallocomplex is synthesized inside the nanopores/channels of the zeolite in the presence of an excess of ligand or its synthetic precursors. As a consequence, what was usually obtained is a mixture of the desired complex with unreacted free ligand as well as side-products. For example, the synthesis of metalloporphyrin (MP) inside zeolites often leads to a significant portion of undesired polymerization product and the unmetallated free base porphyrin ligand.7–9 We report here a novel, efficient and quantitative method to synthesize high purity MPs incorporated in faujasite-Y at a controllable loading concentration. The novel aspect of this method is that an electrostatic interaction was introduced between the host (anionic aluminosilicate species) and the guest (cationic peripheral substituents on MPs) in a ‘build-bottlearound-ship’ approach, namely to synthesize the nanocages of zeolite around the high purity cationic MPs. The selected guest molecules have excellent solubility in aluminosilicate gel. In a typical synthesis, 240 mg of MTMAnPCl5 was added into an aluminosilicate gel, freshly prepared by mixing silicate and aluminate solutions containing 4.6 g of silica, 6.2 g of NaOH, 3.2 g of NaAlO2 and 80 ml of H2O. The gel was then crystallized at 95 ± 2 °C under static and autogeneous conditions in a stainless steel bomb (250 ml) for 48 h. After cooling to room temp., a solid product was recovered by filtration. The complexes adsorbed on the exterior surfaces were removed by a thorough extraction with distilled water, methanol, pyridine (2%)–methanol, and methanol again, respectively. The removed complexes can be fully recovered for the next round of synthesis. The crystals were then dried at 60 °C for 24 h. Thermogravimetry (TG) of the composite indicates that 1.51 mass% is attributable to the confined guest, corresponding to ca. one MP complex for every forty supercages or a supercage occupancy of 2.5%. The loading concentration of MTMAnP4+ is adjustable by controlling its concentration in the aluminosilicate gel before crystallization. The largest loading of ca. 5 mass% of the guest molecule can be achieved if 800 mg of MTMAnPCl5 is added to ca. 94 g of aluminosilicate gel. We speculate that the main driving force for the very successful nanoinclusion of MTMAnP4+ cations in the zeolitic supercages in such a readily controllable manner is the electrostatic interaction between the anionic aluminosilicate species and the cationic peripheral substituents on the porphyrin macrocycle. The excellent aqueous solubility of the cationic MP may contribute to the enhanced loading concentration in the zeolite, but this factor alone does not guarantee successful encapsulation. This assertion is supported by the following experimental facts: (a) in addition to MTMAnP4+, we have also successfully encapsulated other metalloporphyrins with cationic peripheral substituents into zeolitic cages such as metallo–tetrakis(N-methyl4-pyridyl)porphyrins (M = transition metal ion); (b) metalloporphyrins with anionic peripheral substituents, such as metallo–tetra(4-sulfonatophenyl)porphyrin (MTPPS42), could not be entrapped into the supercages of zeolite by this method; and (c) metalloporphyrins with neutral peripheral substituents, such as metallo–tetraphenylporphyrin (MTPP), could only be incorporated into the zeolite in trace amounts independent of the initial concentration of the porphyrin in the gel. On the other hand, it is believed that the first step in the construction of the zeolitic framework is the formation of anionic aluminosilicate species. X-Ray powder diffraction (XRD) patterns of the unloaded zeolite and zeolite entrapped FeTMAnP4+ and MnTMAnP4+ are in excellent agreement with the calculated simulation of the XRD pattern for faujasite zeolite.10 The excellent signal to noise ratio in the XRD indicates that our synthesized faujasite crystals are of high quality. X-Ray induced fluorescence (XIF) analysis indicates that all the samples have a Si/Al ratio of ca. 1.6. This suggests that the framework around the guest molecule MTMAnP4+ is faujasite-Y (denoted MTMAnP@NaY) (M = FeIII, MnIII etc.). The encapsulation of the cationic MP inside the zeolite nanopores is also supported by TG, UV–VIS DRS, surface area measurements, and resonance microRaman spectroscopy (mRR, vide infra). This observation suggests that the occluded MTMAnP4+ cations fit well into the supercages of faujasite-Y. In view of the 13 Å diameter of supercage of faujasite, ca. 18 Å MTMAnP4+ cation [across two opposite trimethylanilinium (TMAn) substituents assuming a planar macrocycle] must distort itself somewhat in order to fit into a supercage. The most plausible distortion is that each of the four peripheral TMAn groups swings toward one of the four channel windows (ca. 7 Å in diameter) tetrahedrally located in a supercage, leading to an overall ruffling of the macrocycle, a distortion commonly observed in the X-ray structures of similar MPs.11 Resonance Raman techniques are very powerful in the study of hemeproteins and zeolite confined molecules.12,13 Here, we applied mRR to study the synthesized zeolite–porphyrin composites. Laser excitation at 632.8 nm, in close resonance with the Q0 absorption band of MTMAnPCl5, was used. Fig. 1 shows the mRR spectra of MnTMAnPCl5 and its faujasite-Y confined composite MnTMAnP@NaY. The normal modes of MnTMAnPCl5 and other similar compounds have been thoroughly studied and assigned.13 Fig. 1 clearly shows that the basic RR features of MnTMAnPCl5 are all retained in the spectrum of MnTMAnP@NaY, indicating that neither decom-