Reaction at interfaces:: The silicoaluminophosphate molecular sieve CAL-1

This work reports on the study to determine the best synthesis conditions for the preparation of CAL-1, a silicoaluminophosphate with a chabazite-like structure, prepared from a lamellar aluminophosphate, ALPO-ntu (herein called ALPO-kanemite), and using hexamethyleneimine (HMI) as the structure directing agent (SDA). These studies showed that optimum reaction conditions to obtain CAL-I depend on the concentration of silicon desired in the structure. For low silicon concentrations, SiO2/Al2O3 <= 0.4, no set of adequate conditions was found; the samples prepared were always contaminated with the starting lamellar materials. At SiO2/Al2O3 molar ratios of 0.8 and 1.2, crystalline samples were obtained after 24 h. For silicon concentrations still larger, an optimum set of reaction conditions could not be determined. The best sample with SiO2/Al2O3 = 1.6 in the gel was obtained with HMI/Al2O3 = 1.5 only after 48 h of reaction time, and even then, the crystallinity of the sample thus prepared was poor. The combination of Si-29 magic-angle spinning NMR and in situ infrared spectroscopy indicated that, at CAL-1 samples with the lower amount of silicon (chi(Si) = 0.21), protons are found either at the borders of silica patches/islands or inside the aluminosilicate domains. Characterization showed that HMI was always accompanied by the n-butylammonium cation, the SDA for the ALPO-kanemite, and showed that larger amounts of HMI led to the contamination of the samples. A mechanism involving the full dissolution of ALPO-kanemite to serve as the reactant for the formation of CAL-1 would not explain the different characteristics of this material in relation to SAPO-34, a structurally similar microporous molecular sieve obtained by hydrothermal synthesis of a gel precursor. Therefore, a small range, molecular level rearrangement of the lamella of the ALPO-kanemite, brought up by the silicon, is proposed to explain the formation of this material.