Analysis of the data in Fig b shows that the

Analysis of the data in Fig. 3b shows that the shape of the (222) reflection virtually does not change and corresponds to the instrumental resolution at all temperatures. The right shoulder of the (310) peak exhibits slight differences from the instrumental resolution even at high temperatures, which can be connected with the presence of the embryos of another phase. At T = 259K (curve 3 in Fig. 3b), there are significant differences from the instrumental resolution near the 2θ values corresponding to the angular positions of the reflections of the tetragonal phase (the decomposition of the reflections from the tetragonal and the cubic phases is shown in Fig. 4a). Importantly, the observed effects for the distortion of the lineshape cannot be linked to the stresses caused by the defects that emerged while the sample was being fabricated, because, firstly, the tested samples were annealed, and, secondly, such stresses would have also manifested at high temperatures. Therefore, the most likely cause of the observed effects is the appearance of the tetragonal phase. This also explains the invariance of the lineshape of the (hhh) peaks at temperature decrease, as it is known that these peaks are not split at the transition from the cubic to the tetragonal phase. By refining the structure taking into account the presence of two phases, it was possible to achieve a reduction of R-factors (compared to the description within the cubic phase) and to adequately describe the profile of the experimental diffraction pattern T = 259K. The contribution from the cubic phase does not disappear with the emergence of the contribution from the tetragonal phase, indicating that these two phases coexist. In view of the above, we can conclude that at T = 259K the sample with the composition x = 0.2 is in a two-phase state (tetragonal+cubic), and the phase transition starts in the temperature range T = 259–287K. Fig. 5 shows the temperature squalene epoxidase of the (311) and (222) peaks for the sample with the composition x = 0.30, but at higher temperatures. As in the case of x = 0.2, the (222) reflection does not change its shape and virtually coincides with the instrumental resolution (see Fig. 5b). The (311) peak has a deviation from the resolution function in the form of extended shoulders at large angles. These deviations exist even at high temperatures and increase under cooling. A possible cause for such deviations is similar to that described for the 0.2 composition. At T = 285K, there is a clear splitting of the (311) peak (Fig. 4b), but a small contribution from the cubic phase remains, which indicates the transition of the sample into the two-phase state. Subsequent full-profile analysis revealed that about 95% of the sample is in the tetragonal phase at this temperature. Based on the above, we can conclude that the temperature at which the phase transition starts in the solid (1–x)PFW–(x)PT solution for x = 0.3 is in the 285–335K range. A full-profile analysis of the diffraction patterns was performed for the temperatures corresponding to the high-temperature cubic phase of the samples, refining the unit cell parameters, the coordinates of the atoms, and the thermal factors. The cubic lattice parameter for the samples with the x = 0.2 and 0.3 compositions increases linearly with temperature growth above room temperature (Fig. 6). The point in Fig. 6a is the value of the cell parameter a = 3.9734Å at room temperature, taken from Ref. [4]. The analysis revealed that the model based on the perovskite structure yields anomalously high values of the Debye–Waller factor for lead. It is known that the lead ion is not in its (0, 0, 0) crystallographic position in lead-containing relaxors [9] and, in particular, in PFW [10]. Therefore, similar to [10], we used a model of the multiple-well potential where the lead is equiprobably displaced from the (000) position by a fixed distance in one of the 12 equivalent [110] directions. Fig. 7 and Table 1 show the values of these static displacements in the samples with the x = 0, 0.20 and 0.30 compositions. It can be seen that the displacements decrease with an increase in PbTiO3 concentration, which corresponds to the transition of PFW–PT from the relaxor state to the ferroelectric one, and is consistent with the dielectric spectroscopy data presented in Ref. [4].