Recently, we reported the formation of TiO2 mesoporous films by anodizing titanium in hot phosphate/glycerol electrolytes [1-3]. The resultant films showed interesting features, for example: (i) the pore size was as small as ~10 nm leading to a high surface area and did not change linearly with the formation voltage; and (ii) formation of the films at an 20 V generated the crystalline anatase phase without the requirement of post-annealing, whereas as prepared films anodized in the electrolytes containing fluorides were usually amorphous. However, the details of formation behavior of the crystalline TiO2mesoporous films have yet to be revealed. Therefore, in this study, we investigated the growth process of the films during the anodizing in hot alkaline glycerol electrolytes containing phosphate by cyclic voltammetry and changing electrolyte anion species. The specimens were 99.5% pure titanium plates, which were electropolished in 1.0 mol dm-3 NaCl/ethylene glycol solution. The electropolished specimens were anodized in stirred glycerol electrolytes containing selected amounts of phosphate such as K2HPO4 and K3PO4 and/or NOH, and 0.03 wt% H2O at 433 K. Anodizing by cyclic voltammograms (CVs) was performed using a three-electrode system with a platinum foil as a counter electrode and a platinum wire as a reference electrode. The potential sweep rate and stirring rate were 30 mV s-1and 330 rpm, respectively. Anodizing at constant voltage of 20 V was performed using a two-electrode system with the platinum foil as the counter electrode. Fig. 1 shows the CVs of titanium at the first cycle in the hot glycerol electrolyte containing 0.6 mol dm-3 K3PO4 + 0.2 mol dm-3 K2HPO4 (Fig. 1a), together with that in the ethylene glycol electrolyte containing fluoride (Fig. 1b). In the case of the anodizing in the electrolyte containing the phosphates, the current density became almost constant at potentials higher than 2 V vs Pt. Similar constant currents were obtained even during the negative sweep of the potential, whereas the current density decreased during the negative sweep in the electrolyte containing fluoride. This result suggested that the ionic transport in the barrier layer under the high electric field was not a rate-determining step during the anodizing in the hot phosphate/glycerol electrolyte. The Cottrell plot during anodizing in the phosphate/glycerol electrolyte at 3 V vs Pt reveals a linear correlation between the reciprocal of the current density and the square root of anodizing time. Thus, the anodizing process is diffusion-controlled. In addition, the limiting current became smaller with decreasing basicity of the electrolyte, implying some basic species, such as OH-, should be diffusing ones to grow the anodic TiO2films during anodizing in the hot phosphate/glycerol electrolyte. The crystallinity of the films anodizing the electrolyte containing phosphate or not was investigated by XRD. The results suggested that a small amount of the phosphate contained the anodic films may have important roles to crystallize during anodizing.  References [1] E. Tsuji, Y. Taguchi, Y. Aoki, T. Hashimoto, P. Skeldon, G.E. Thompson and H. Habazaki, Appl. Surf. Sci. 2014, 301, 500. [2] E. Tsuji, N. Hirata, Y. Aoki and H. Habazaki, Mater. Lett., 2013, 91,39. [3] Y. Taguchi, E. Tsuji, Y. Aoki and H. Habazaki, Appl. Surf. Sci. 2012, 258, 9810.  Figure caption Fig. 1 The cyclic voltammogram of titanium specimens during anodizing in (a) 0.6 mol dm-3 K3PO4 and 0.2 mol dm-3 K2HPO4 glycerol electrolytes containing 0.03 mass% water at 433 K and (b) 0.25 wt% NH4F /ethylene glycol electrolyte containing 1 vol% water at 293 K with stirring. The right axis shows the current density during anodizing in (a) and the left one shows that in (b), respectively. Figure 1