This note is related to the above-mentioned paper (1) and uses for its conclusions the data, interpretation, and results reported therein. The original paper deals with Gd-doped CaTiO perovskite in which the charge compensation was interpreted to be a reduction of titanium from Ti> to Ti> for samples fired at 1550°C in air. This paper stated that the charge compensation ''takes place via the formation of one formula unit of Ti> per formula unit of Gd.''
The following methods and data were used in Ref. 1 to prove the existence of Ti> in Ca \V Gd V TiO fired in air for 1 week at 1550°C: atomic ratios were obtained from quantitative analysis using SEM and TEM techniques, X-ray photoelectron spectroscopy (XPS), X-ray absorption nearedge spectroscopy (XANES), unit cell parameters from Xray powder diffraction (XRD), cathodoluminescence, and density measurements by helium gas pycnometry.
It is important to note that although XANES, cathodoluminescence, and density measurements showed clearly negative results for the presence of Ti>, the authors of Ref. 1 interpreted them as inconclusive. In addition, electron paramagnetic resonance (EPR) spectroscopy, which could give direct information about the presence of Ti>, was not employed.
Let us look more closely at the reported data for the unit cell parameters determined by XRD, the measured Ti 2p XPS spectra, and the quantitative analysis by SEM.
The unit cell parameters of orthorhombic Gd-substituted perovskite were determined by XRD. The results were interpreted in terms of changes in volume of the unit cells: 223.8 A s for pure CaTiO , 224.2 A s for the sample with 7.5% Gd, and 224.7 A s for the sample with 15% Gd. The difference in volume changes between samples with maximum Gd loadings and pure CaTiO , 224.7!223.8" 0.9 A s or 0.4% (&13 times less than the volume of one oxygen ion), was interpreted as ''a small but significant increase in unit cell volume.'' The radius of Gd> is less than 355