The Ti-O phase diagram in [Massalski2] was adopted from [1987Mur]. [2001Oka] introduced two versions of the Ti-O phase diagram calculated by [1997Fis] and by [1999Wal].

A new thermodynamic assessment was carried out on this system by [2007Can] to refine the phase diagram of [1999Wal], as shown in Fig. 1 and 2.

Fig. 1
figure 1

Ti-O phase diagram

Full size image
Fig. 2
figure 2

Enlargement of Fig. 1 from 58 to 68 at.% O

Full size image

The lack of reliable experimental phase boundary data caused difficulty in confirming the validity of each phase diagram in [2001Oka]. This problem still persists because few experimental data have been added since then. Therefore, as stated in [2001Oka], the differences among the phase diagrams of [1987Mur], [1997Fis], [1999Wal], and [2007Can] can only suggest the areas where further experimental work is needed. Furthermore, [2007Vil] lists quite a few Ti-O compounds not shown in Fig. 1. Some of these may be equilibrium phases. It is also needed to examine the stability of phases expressed by \(\hbox{Ti}_{n}\hbox{O}_{2n-1}\; (n = 2, 3, 4 {\ldots})\) in Fig. 2 because these phases are so closely neighbored that their stability over a wide temperature range is questionable [1993Oka]. Experimental confirmation is needed in this regard also.

Table 1 shows Ti-O crystal structure data adopted from [1987Mur] with modifications for consistency with Fig. 1 and additional information from [2007Vil].

Table 1 Ti-O crystal structure data
Full size table