The aim of this paper is the characterization of pure tobelite synthesized in the system (NH4)(2)O-Al2O3-SiO2-H2O, which is a first, indispensable step for accurately interpreting the data obtained in more complex chemical systems. Tobelite has been synthesized at hydrothermal conditions (200 +/- 3 degrees C), using a solid:solution ratio (in weight) = 1:15, and reaction times from 1 to 90 days and has been characterized by X-ray diffraction; infrared and Raman spectroscopy; differential thermal analysis and thermogravimetry; and transmission-analytical electron microscopy. The average chemical composition of tobelite formed after 90 days run time, determined by combining the analytical electron microscopy data with nitrogen determination in the solid products of the reaction by elemental analysis, was [NH4Al1.98(Si3.07Al0.93)O-10(OH)(2)]. The refinement of the X-ray diffraction patterns indicated that the 1M polytype predominates, and the determined cell-parameters were: a = 5.219 (0.004) angstrom; b = 9.021 (0.007) angstrom; c = 10.556 (0.009) angstrom; beta = 101.41 degrees (0.08); V = 487.23 (0.69) angstrom(3). The infrared spectra indicated the presence of two v(4) bands (at 1403 and 1435 cm(-1)), which can be related to the presence of two main tobelite populations with mean particle sizes of 400 x 60 and 2000 x 200 angstrom. The thermal study by means of differential thermal analysis and thermogravimetry, combined with X-ray thermodiffraction suggested that the mechanisms of NH4 loss were different in the two differently-sized tobelite populations: the thinner particles showed a low thermal stability (between 380 degrees and 440 degrees C) and the NH4 loss was coupled with dehydroxylation, leading directly to the decrease of the tobelite content. Loss of NH4 from the thicker particles occurred between 440 degrees and 500 degrees C, causing a gradual contraction of the basal spacing, which was followed by dehydroxylation. There were no previous data of Raman spectroscopy of tobelite and this study presents its most characteristic features. (C) 2016 Elsevier B.V. All rights reserved.