Calphad 61 (2018) 20–32

Contents lists available at ScienceDirect

Calphad
journal homepage: www.elsevier.com/locate/calphad

Experimental investigation of phase equilibria in the Ti-Fe-Zr system T

a b a,⁎ a a,⁎
Lijun Zeng , Guanglong Xu , Libin Liu , Weimin Bai , Ligang Zhang
a
School of Material Science and Engineering, Central South University, Changsha, Hunan 410083, PR China
b
Tech Institute for Advanced Materials & School of Material Science and Engineering, Nanjing Tech University, Nanjing, Jiangsu 211800, PR China

A R T I C L E I N F O A B S T R A C T

Keywords: The phase relations of the Ti-Fe-Zr system at 873 and 1173 K were experimentally investigated by using the
Bio-titanium alloys equilibrated alloys. Four ternary phases were determined stable at 1173 K. Two of them were line compounds
Ti-Fe-Zr with fixed Fe/(Ti, Zr) ratio, while the other two showed ellipse-shaped homogeneity. There were eight three-
Isothermal section phase equilibria and a continuous single phase field of β(Ti, Zr) in the isothermal section at 1173 K. The solu-
Phase equilibria
bility of Zr in Fe2(Ti, Zr) and Ti in (Fe, Ti)2Zr was measured to be 25.5 at% and 6.1 at%, respectively. The
Laves phase
remaining binary intermetallic phases hardly dissolved the third component. The phase relationships in the
isothermal section at 873 K were similar to those at 1173 K over the composition range of 0–50 at% Fe. In this
isothermal section, five three-phase equilibria were observed and they were all related with the newly found
ternary phases. The phase relations measured in this work are of significant difference from the existing ones
obtained with the diffusion couple.

1. Introduction
Owing to the combination of favorable properties such as high
specific strength, good biocompatibility, relatively low elastic modulus
and high corrosion resistance, titanium alloys have attracted significant
attention for their biomedical application as orthopedic implants [1,2].
Among them, as a typical biomedical titanium alloy, the Ti-6Al-4V alloy
with (α + β)-type microstructure has been commercially utilized for a
long period. However, the high Young's modulus of this alloy
(~110 GPa), the toxicity of vanadium to the human body and the
neurological disease potentially induced by aluminum [1] are dis-
advantageous from the biological point of view. In comparison with
other phase structures in Ti, the β-phase has the relatively low elastic
modulus [3–6]. Hence, biocompatible β-phase stabilizing alloying ele-
ments (e.g. Nb, Ta, Fe and Mo) are preferable in the development of
bio-titanium alloys. The addition of Zr into β-type Ti alloys has been
reported to improve alloy mechanical properties and corrosion re-
sistance, and also to inhibit the martensitic transformation in alloy,
regardless of the slight effect of Zr on the β-transus [7–9]. As a low cost
β-phase stabilizer, Fe is quite attractive to develop alloys of both high
strength and low Young's modulus [10]. As the phase diagram is an
important basis for the materials design and application [11–13], the
exploration of phase equilibria in the Ti-Fe-Zr system is of importance
in developing high performance-to-cost ratio bio-titanium alloys.
Phase diagrams of the three constitutive binary subsystems, i.e. Fe-
Ti, Fe-Zr and Ti-Zr, in the Ti-Fe-Zr ternary system are presented in
Fig. 1. Crystallographic data of all solid phases in these binary sub-
systems [14–19] are given in Table 1.
For the Fe-Ti system, lots of studies have been carried out [20–30].
Murry [25] conducted a critical review on the phase diagram, Jonsson
[27] presented an overview on the available thermodynamic data. Up
to now, the thermodynamic assessment has been implemented by dif-
ferent groups [25–30]. The phase diagram of the Fe-Ti system has been
given in Fig. 1(a) according to the latest thermodynamic description in
Ref. [30]. Two stable intermetallic phases (i.e. FeTi and Fe2Ti) are
present in phase diagram and both exhibit visible homogeneity ranges.
The homogeneity range of FeTi is nearly symmetric with respect to the
stoichiometric composition, while that of Fe2Ti leans significantly to-
wards the Fe-rich side.
For the Fe-Zr system, the intermetallic phases are listed in Table 1.
Although the phase diagram has been investigated experimentally by
many researchers [31–34], it still remains controversial that whether
the Fe23Zr6 or Fe2Zr with a hexagonal structure (denoted as hex.-Fe2Zr
in short) phase is stable in the Fe-rich side. A series of thermodynamic
assessments of this system have been conducted with the Fe23Zr6
[33,35,36] or hex-Fe2Zr phase [37,38] alternatively serving as a stable
phase. Due to the fact that the Fe23Zr6 phase was found in most ex-
perimental studies (e.g. the recent investigations of the Cu-Fe-Zr [39]
and Fe-Nb-Zr [40] systems), the Fe-Zr phase diagram assessed lately by
Lu et al. [36] is accepted in present work. As shown in Fig. 1(b), among
the four intermetallic phases, only the Fe2Zr phase has a noticeable
homogeneity range. For the other three phases, FeZr2 appears only

⁎
Corresponding authors.
E-mail addresses: pdc@csu.edu.cn (L. Liu), ligangzhang@csu.edu.cn (L. Zhang).

https://doi.org/10.1016/j.calphad.2018.02.005
Received 19 December 2017; Received in revised form 6 February 2018; Accepted 7 February 2018
0364-5916/ © 2018 Elsevier Ltd. All rights reserved.
L. Zeng et al. Calphad 61 (2018) 20–32

Fig. 1. Phase diagrams of constitutive binary systems in the Ti-Fe-Zr system, calculated based on the data given in: (a) Fe-Ti [30]; (b) Fe-Zr [36]; (c) Ti-Zr [45].

Table 1
Crystallographic data of all phases involved in the Ti-Fe-Zr system.

System Phase Strukturbericht designation Pearson symbol Space group Prototype Lattice parameter(s) (nm) Ref.

Fe α(Fe) A2 cI2 Im3m W a = b = c = 0.2887 [14,15]

γ(Fe) A1 cF4 Fm3m Cu a = b = c = 0.3660 [14,15]
δ(Fe) A2 cI2 Im3m W a = b = c = 0.2932 [14,15]
Ti α(Ti) A3 hP2 P 63/ mmc Mg a = b = 0.2950, c = 0.4681 [14,15]
β(Ti) A2 cI2 Im3m W a = b = c = 0.3311 [14,15]
Zr α(Zr) A3 hP2 P 63/ mmc Mg a = b = 0. 3252, c = 0. 5180 [15,16]
β(Zr) A2 cI2 Im3m W a = b = c = 0.3540 [15,16]
Fe-Ti Fe2Ti C14 hP12 P 63/ mmc Zn2Mg a = b = 0.4785, c = 0.7799 [14,15]
FeTi B2 cP2 Pm3m CsCl a = b = c = 0.2979 [14,15]
Ti2Fe* – cF96 Fd3m Ti2Ni a = b = c = 1.1328 [17]
Fe-Zr Fe23Zr6 D82 cF116 Fm3m Mn23Th6 a = b = c = 1.1690 [15,18]
Fe2Zr* C36 hP24 P 63/ mmc Ni2Mg a = b = 0.4954, c = 1.6304 [19]
Fe2Zr C15 cF24 Fd3m Cu2Mg a = b = c = 0.7070 [15,16]
FeZr2 C16 tI12 I 4/ mcm Al2Cu a = b = 0.6385, c = 0.5596 [15,16]
FeZr3 E112 oS16 Cmcm Re3B a = 0.3324, b = 1.0990, c = 0.8810 [15,16]
Ti-Zr – – – – – –

The phase marked with ‘* ’ symbol denotes metastable phase.

within the temperature range of 1215–1054 K, and FeZr3 is formed
through a peritectoid reaction.
For the Ti-Zr system, several thermodynamic descriptions [41–46]
have been performed based on the experimental data reported in stu-
dies [47–55]. The first thermodynamic assessment was performed by
Murry [41]. In comparison with other existing assessments, the one
recently conducted by Cui et al. [45] is believed more reliable due to
the consideration of the enthalpy of mixing of liquid alloys reported by
Thiedemann et al. [54]. In Fig. 1(c), the full isomorphous phase dia-
gram does not exhibit any intermetallic phases or invariant reactions, in
comparison with those of the Fe-Ti (Fig. 1(a)) and Fe-Zr (Fig. 1(b))
systems.
For the Ti-Fe-Zr ternary system, many studies have been conducted
on the phase relations [56–64]. By means of diffusion couples, two
groups constructed the isothermal sections at 1173 [56] and 1073 K
[57] after annealing for 1440 and 5760 h, respectively. In these two
isothermal sections, all binary intermetallic phases exhibited noticeable
solubility of the third component. Phase relations in Refs. [56,57] were
determined solely relying on the compositional measurement while
without crystal structure analysis. This may lead to some uncertainty in
measuring the limited solubility of the third component for the inter-
metallic phases, especially for the Fe2Ti and Fe2Zr (i.e. Fe2(Ti, Zr) and
(Fe, Ti)2Zr in this study) phases which have the same stoichiometric
ratio and may be misidentified without the structure analysis [56,57].
Ardisson et al. [58] annealed specimens at 1223 K for 60 h to determine
the solubility in FeZr2 and at 1073 K for 48 h for the solubility in FeZr3.
Nevertheless, Ti was found insoluble in above two phases. This result
was in contradiction with that in studies [56,57], in which notable

21
L. Zeng et al. Calphad 61 (2018) 20–32

solutions of the third component in the intermetallic phases have been
observed. The vertical section of TiFe2-ZrFe2 has been widely in-
vestigated in literatures [59–64]. Piegger and Craig [59] firstly in-
vestigated this section through as-cast specimens, and derived a con-
centration dependency of the lattice parameters of Fe2Ti and Fe2Zr
phases. This dependency was also observed in works [62–64]. Pet’kov
et al. [60] reported a peritectic reaction of Liquid + Fe2Zr → Fe2Ti
(measured via differential thermal analysis) occurring at 1743 ± 10 K
in this vertical section. The solubility of the third component in the
Fe2Ti and Fe2Zr phases were measured to be ~16 − 17 at% Zr at
1173 − 1743 K and 7.7 at% Ti at 1173 K [60]. The former solubility
was lower than those reported in studies [59] (18.6 at% Zr determined
from as-cast alloys) and [62] (20.0 at% Zr determined from alloys an-
nealed at 1073 − 1773 K), whereas the latter one was higher than that
(5.0 at% Ti) in study [59]. No stable ternary phase has been found in
this system up to now. Based on all above introduction, it is clear that a
comprehensive study of the Ti-Fe-Zr ternary system is still lacking.
In the present work, phase relations in the Ti-Fe-Zr ternary system at
873 and 1173 K were experimentally investigated in detail over the
whole composition triangle, by using the equilibrated alloys. Isothermal
sections obtained at these two temperatures were compared mutually
and also with sections obtained in existing studies. The experimental
data can be further utilized for the thermodynamic description.
2. Experimental procedure
To experimentally investigate the phase relations in the Ti-Fe-Zr
system, a total of 38 alloys with compositions given in Tables 2, 3 were
prepared. The selection of alloy compositions was partially referred to
literatures [56,57]. Several alloys were intentionally prepared to verify
the tentative three-phase regions reported in Refs. [56,57],
Pure elements of Ti (99.99 mass%), Fe (99.99 mass%) and Zr (99.99
mass%), supplied by China New Metal Materials Technology Co., Ltd.,
were weighed to the desired proportions to achieve various nominal
compositions of alloys. The weight of each sample was controlled
around 6 g with a error of ± 0.005 g. Alloy samples were prepared by
arc-melting under a high-purity argon atmosphere, using a non-con-
sumable tungsten electrode. Prior to the arc-melting, a pure Ti ball was
first melted to clean the slight oxygen and other contaminants in the
furnace. To ensure the composition and microstructure homogeneity,
all samples were melted at least four times and inverted after each
melting. The weight loss of each obtained button-shaped sample was
less than 1%.
Each sample was then divided into four pieces for different pro-
cesses. Two pieces for annealing at 873 and 1173 K were sealed in the
evacuated and argon-refilled quartz tubes, placed in a tube furnace and
then heated to the corresponding temperature (with a temperature
error of ± 3 K) for 150 days (3600 h). After annealing, samples were
quenched in cold water to retain the annealed microstructure. The
microstructure was observed with scanning electron microscopy (SEM,

Table 2
Constituent phases of the Ti-Fe-Zr ternary alloys annealed at 1173 K.

Alloys Phase equilibria Composition (at%)

No. Composition (at%) Phase 1/2/3 Phase 1 Phase 2 Phase 3

Ti Fe Zr Ti Fe Zr Ti Fe Zr Ti Fe Zr

A1 10.1 84.0 5.9 Fe2(Ti, Zr)/α(Fe) 18.8 71.2 10.0 3.3 96.7 0.0 – – –
A2 10.2 79.8 10.0 Fe2(Ti, Zr)/α(Fe) 14.2 71.1 14.7 2.2 97.7 0.1 – – –
A3 20.5 74.4 5.1 Fe2(Ti, Zr)/α(Fe) 23.6 71.4 5.0 4.1 95.9 0.0 – – –
A4 4.9. 72.8 22.3 Fe2(Ti, Zr)/α(Fe) 5.6 71.5 22.9 1.5 98.2 0.3 – – –
A5 1.8 76.5 21.7 Fe23Zr6/Fe2(Ti, Zr)/α(Fe) 0.3 78.9 20.8 3.1 71.5 25.4 0.9 99.1 0
A6 5.1 67.2 27.7 (Fe, Ti)2Zr/Fe2(Ti, Zr) 3.2 68.0 28.8 7.0 67.5 25.5 – – –
A7 36.4 58.6 5.0 Fe2(Ti, Zr)/FeTi 30.7 63.2 6.1 50.4 49.5 0.1 – – –
A8 25.6 63.3 11.1 Fe2(Ti, Zr)/FeTi 25.5 63.1 11.4 50.3 49.6 0.1 – – –
A9 31.3 49.9 18.8 τ1/FeTi 26.1 49.6 24.3 50.4 49.2 0.4 – – –
A10 20.0 55.0 25.0 τ1/FeTi 19.8 55.7 24.5 50.3 49.5 0.2 – – –
A11 20.5 49.6 29.9 τ2/τ1 20.8 33.6 45.6 20.5 49.8 29.7 – – –
A12 9.7 57.7 32.6 τ2/τ1/(Fe, Ti)2Zr 8.8 34.9 56.3 11.8 54.5 33.7 6.1 63.8 30.1
A13 5.0 50.0 45.0 τ2/(Fe, Ti)2Zr 5.0 35.0 60.0 4.9 62.7 32.4 – – –
A14 11.0 45.2 43.8 τ2/τ1 10.3 34.6 55.1 12.2 54.1 33.7 – – –
A15 30.2 45.1 24.7 τ2/τ1/FeTi 20.6 33.4 46.0 26.5 48.5 25.0 50.2 49.6 0.2
A16 2.5 33.8 63.7 FeZr2 /τ2/ (Fe, Ti)2Zr 0.4 32.2 67.4 3.9 34.5 61.6 4.3 63.0 32.7
A17 45.1 45.1 9.8 τ3/FeTi 35.7 39.0 25.3 50.2 49.5 0.3 – – –
A18 60.6 34.6 4.8 β(Ti, Zr)/FeTi/τ4 79.3 17.5 3.2 52.0 47.8 0.2 60.9 33.3 5.8
A19 50.0 39.9 10.1 τ3/τ4/FeTi 39.1 37.4 23.5 51.4 33.4 15.2 50.3 49.5 0.2
A20 60.1 29.7 10.2 τ3/τ4/β(Ti, Zr) 45.5 32.7 21.8 56.7 33.4 9.9 77.4 16.1 6.5
A21 61.3 20.4 18.3 τ3/β(Ti, Zr) 44.0 32.0 24.0 75.9 11.4 12.7 – – –
A22 45.1 34.3 20.6 τ3/τ4 41.0 35.6 23.4 50.4 33.2 16.4 – – –
A23 41.6 29.1 29.3 β(Ti, Zr)/τ3 57.3 5.4 37.3 40.1 30.9 29.0 – – –
A24 29.7 14.4 55.9 β(Ti, Zr)/τ3 26.4 2.0 71.6 33.9 34.3 31.8 – – –
A25 23.4 17.2 59.4 β(Ti, Zr)/τ2/τ3 17.6 1.9 80.5 22.8 33.2 44.0 31.4 36.3 32.3
A26 18.3 10.1 71.6 β(Ti, Zr)/τ2 17.8 3.0 79.2 22.3 33.2 44.5 – – –
A27 20.4 5.4 74.2 β(Ti, Zr)/τ3 20.5 3.2 76.3 31.7 35.8 32.5 – – –
A28 11.0 17.1 71.9 β(Ti, Zr)/τ2 6.5 3.8 89.7 15.9 34.2 49.9 – – –
A29 11.5 6.9 81.6 β(Ti, Zr)/τ2 10.7 3.4 85.9 19 34.1 46.9 – – –
A30 71.6 9.9 18.5 β(Ti, Zr) 71.6 9.9 18.5 – – – – – –
A31 61.2 8.1 30.7 τ3/β(Ti, Zr) 41.5 30.3 28.2 62.8 6.7 30.5 – – –
A32 50.8 5.8 43.4 β(Ti, Zr)/τ3 52.2 4.9 42.9 38.9 31.6 29.5 – – –
A33 88.8 5.5 5.7 β(Ti, Zr) 88.8 5.5 5.7 – – – – – –
A34 84.9 3.1 12.0 β(Ti, Zr) 84.9 3.1 12.0 – – – – – –
A35 9.1 9.1 81.8 β(Ti, Zr)/τ2 7.8 2.0 90.2 16.8 33.6 49.6 – – –
A36 5.0 25.0 70.0 β(Ti, Zr)/τ2 1.4 3.9 94.7 6.4 34.3 59.3 – – –
A37 41.3 14.5 44.2 β(Ti, Zr)/τ3 44.0 3.1 52.9 37.7 32.2 30.1 – – –

22
L. Zeng et al. Calphad 61 (2018) 20–32

Table 3
Constituent phases of the Ti-Fe-Zr ternary alloys annealed at 873 K.

Alloys Phase equilibria Composition (at%)

No. Composition (at%) Phase 1/2/3 Phase 1 Phase 2 Phase 3

Ti Fe Zr Ti Fe Zr Ti Fe Zr Ti Fe Zr

B20 60.1 29.7 10.2 τ4/β(Ti, Zr)/FeTi 56.4 32.9 10.7 81.8 11.3 6.9 51.6 48.3 0.1
B21 61.3 20.4 18.3 τ3/τ4/β(Ti, Zr) 42.2 30.1 27.7 51.4 32.9 15.7 77.4 7.8 14.8
B22 45.1 34.3 20.6 τ3/τ4/FeTi 40.4 33.5 26.1 50.6 33.0 16.4 50.4 49.1 0.5
B23 41.6 29.1 29.3 τ3/β(Ti, Zr) 41.0 29.7 29.3 73.3 6.8 19.9 – – –
B24 29.7 14.4 55.9 α(Ti, Zr)/τ3 21.7 0.5 77.8 36.8 30.3 32.9 – – –
B25 23.4 17.2 59.4 α(Ti, Zr)/τ2/τ3 13.2 0.6 86.2 25.4 33.0 41.6 34.8 33.5 31.7
B26 18.3 10.1 71.6 α(Ti, Zr)/τ2/τ3 13.2 0.5 86.3 25.4 33.1 41.5 34.4 33.8 31.8
B27 20.4 5.4 74.2 α(Ti, Zr)/τ3 18.6 0.3 81.1 36.3 31.4 32.3 – – –
B28 11.0 17.1 71.9 α(Ti, Zr)/τ2 4.5 0.3 95.2 17.3 33.1 49.6 – – –
B29 11.5 6.9 81.6 α(Ti, Zr)/τ2 9.7 0.2 90.1 21.6 33.0 45.4 – – –
B30 71.6 9.9 18.5 τ3/β(Ti, Zr) 41.7 29.7 28.6 75.5 6.9 17.6 – – –
B31 61.2 8.1 30.7 τ3/β(Ti, Zr) 41.1 29.6 29.3 63.6 5.3 31.1 – – –
B32 50.8 5.8 43.4 β(Ti, Zr)/τ3 51.3 2.5 46.2 40.2 29.7 30.1 – – –
B33 88.8 5.5 5.7 β(Ti, Zr)/α(Ti, Zr) 85.5 10.4 4.1 94.6 0.0 5.4 – – –
B34 84.9 3.1 12.0 β(Ti, Zr)/α(Ti, Zr) 81.3 5.6 13.1 88.9 0.0 11.1 – – –
B36 5.0 25.0 70.0 α(Ti, Zr)/τ2 0.2 0.8 99.0 6.7 33.6 59.7 – – –
B37 41.3 14.5 44.2 β(Ti, Zr)/τ3 42.7 2.0 55.3 38.7 30.4 30.9 – – –
B38 35.0 10.0 55.0 α(Ti, Zr)/β(Ti, Zr)/τ3 25.2 0.1 74.7 36.9 1.3 61.8 38.3 30.4 31.3

Fig. 2. The BSE images and XRD patterns for alloys annealed at 1173 K: (a) and (b) for A1; (c) and (d) for A5.

JEOL JSM 6000, equipped with energy dispersive X-ray spectrometer
(EDX)). The chemical compositions of individual phases were analyzed
by electron probe microanalysis (EPMA), performed on the JEOL JXA-
8800R instrument (with OXFORD INCA 500 wave-length dispersive X-
ray spectrometer) with an acceleration voltage of 20 kV and probe
current of 2 × 10−8 A. To confirm the phase constituents in all alloys,
powder X-ray diffraction (XRD) was carried out on the Rigaku D-MAX/
2500 VB + X-ray diffractometer with a Cu Kα radiation (40 kV and
250 mA), and the scanning range was 10°− 90° with a speed of 0.067°/
s.
Electron backscatter diffraction (EBSD) analysis was additionally
performed to further determine crystal structures of the ternary phases.
A detailed introduction to application of the EBSD technique can be
found in Ref. [65]. The structure identification of each tested phase was
realized based on the matching between the diffraction bands experi-
mentally obtained and the bands simulated with known structure types

23
L. Zeng et al.
Fig. 3. The BSE images and XRD patterns for alloys annealed at 1173 K: (a) and (b) for A9; (c) and (d) for A14.
and lattice parameters. We firstly re-polished the specimens after EPMA
examination in 0.05 µm colloidal silica suspensions, and then removed
the remaining minor deformation with a vibratory polisher (BUEHLER
VibroMet® 2) in 0.03 µm colloidal silica suspension to ensure the surface
quality of specimens. A field-emission scanning electron microscope
(QUANTA FEG 650), equipped with an EBSD detector (OXFORD Nor-
dlysMax2), was used to collect the Kikuchi patterns. The acceleration
voltage and emission current were 20 kV and 199.6 μA, respectively.
3. Results and discussion
The phase equilibria of the Ti-Fe-Zr system at 873 and 1173 K were
discussed based on the experimental data obtained through the com-
bination of SEM, EPMA, XRD and EBSD examination. The constituent
phases in all equilibrated alloys at these two temperatures and the
corresponding phase compositions are summarized in Tables 2, 3, re-
spectively. In the following context, the phase relations in several key
alloys were discussed in detail.
3.1. Phase equilibria at 1173 K
To experimentally verify the existence of the tentative three-phase
region of α(Fe) + Fe2(Ti, Zr) + (Fe, Ti)2Zr reported in [56,57], five
alloys A1 −A5 were intentionally designed at the Fe-rich corner. The
BSE images and XRD patterns of two representative alloys A1 and A5
are presented in Fig. 2. The microstructure of alloy A1 (Fig. 2(a)) was
composed of the black and gray phases. These two phases were iden-
tified as α(Fe) and Fe2(Ti, Zr) based on the EPMA and XRD results. By
conducting the same examination, the two-phase equilibrium of α(Fe)
+ Fe2(Ti, Zr) was also detected in alloys A2, A3 and A4 (see Table 2).
For alloy A5, three phases (i.e. the dark, gray and light phases) could be
24
Calphad 61 (2018) 20–32
readily seen from its microstructure in Fig. 2(c). However, only the
characteristic peaks of α(Fe) (the dark phase) and Fe2(Ti, Zr) (the gray
phase) phase were captured in Fig. 2(d). Compared with these two
phases, the volume fraction of the light phase having a composition
Ti.3Fe78.9Zr20.8 (at%) was much smaller (see Fig. 2(c)). In this case, the
characteristic peaks of the light phase were hard to capture. Con-
sidering the fact that the stoichiometric ratio of Fe/Zr in this phase was
extremely close to that in the binary Fe23Zr6 phase, the light phase in
alloy A5 was recognized as Fe23Zr6, which has also been observed in the
recent experimental studies of the Cu-Fe-Zr [39] and Fe-Nb-Zr [40]
systems. Based on the phase equilibira in above alloys, a broad two-
phase region of α(Fe) + Fe2(Ti, Zr) and a three-phase region of α(Fe)
+ Fe2(Ti, Zr) + Fe23Zr6 were obtained at the Fe-rich corner, instead of
the three-phase region of α(Fe) + Fe2(Ti, Zr) + (Fe, Ti)2Zr reported in
Refs. [56,57].
Fig. 3 illustrates the constituent phases in alloys A9 and A14. From
the microstructure of alloy A9 in Fig. 3(a), this alloy consisted of two
phases. The dark phase with a composition of Ti50.4Fe49.2Zr.4 was
identified as FeTi with the help of XRD data in Fig. 3(b). The gray one
having a composition Ti26.1Fe49.6Zr24.3 was speculated as an uni-
dentified ternary compound. Although the diffraction peaks of the gray
phase could not be matched by any PDF cards corresponding to the
crystal structure of stable solid phases in the ternary system, they were
indexed by a hexagonal crystal structure. A two-phase equilibrium was
also observed in alloy A14 (see Fig. 3(c)), i.e. the black strip phase and
the gray one. The chemical composition of the former phase was de-
tected as Ti12.2Fe54.1Zr33.7, which showed visible difference from that of
the gray one in A9. Nevertheless, the characteristic peaks of these two
phases were in coincidence (see Fig. 3(b) and (d)). Thus, the strip phase
in A14 and the gray phase in A9 were considered as the same phase,
denoted as τ1 here. The gray phase in A14 was measured with a
L. Zeng et al.
Fig. 4. The BSE images and XRD patterns for alloys annealed at 1173 K: (a) and (b) for A12; (c) and (d) for A15. (e) the BSE image of alloy A16.
composition of Ti10.3Fe34.6Zr55.1, which matched the formula of Fe(Ti,
Zr)2. According to the phase diagram in Ref. [56], this phase should be
the FeZr2 phase with a tetragonal structure. However, the diffraction
peaks of this phase suggested a cubic structure. Thus, the gray phase in
A14 was identified as another new phase, denoted as τ2.
The BSE image of alloy A12 in Fig. 4(a) exhibits the coexistence of
three phases. Among them, the light gray, gray and dark gray phases
were discriminated as τ2, τ1 and (Fe, Ti)2Zr, which were confirmed by
the corresponding XRD patterns in Fig. 4(b). Similarly, τ1 and τ2 phases
were detected in alloy A15 and presented in Fig. 4(c), besides the FeTi
phase. The determined three-phase regions of τ1 + τ2 + (Fe, Ti)2Zr in
A12 and τ1 + τ2 + FeTi in A15 were next to the two-phase tie-lines of
alloys A14 and A9 in the composition triangle (see the ongoing
Fig. 8(a)), respectively. The microstructure of alloy A16 given in
Fig. 4(e) shows a three-phase equilibrium of (Fe, Ti)2Zr + FeZr2 + τ2.
As given in Table 2, only 0.4 at% Ti was dissolved into the light gray
25
Calphad 61 (2018) 20–32
FeZr2 phase in A16 and 3.9 at% Ti into the matrix τ2 phase. The small
solubility of Ti in FeZr2 obtained in this work is close to the solubility
~2.0 at% Ti at 1223 K in Ref. [58], but largely differs from the value of
26.9 at% Ti at 1173 K in Ref. [56]. This large difference can be un-
derstood. Without the crystal structure analysis in Ref. [56], the τ2
phase could not be accurately discriminated from the FeZr2 phase, and
then the solubility of Ti in τ2 phase might be mistaken as that in FeZr2
phase.
According to the composition data listed in Table 2, the τ2 phase
showed different chemical compositions in various alloys, the ratio of
Fe/(Ti, Zr) in this phase nevertheless approximated 0.5. Thus, it can be
deduced that τ2 was a line compound, with a nearly constant Fe content
(34.0 at% on average) and a maximum Ti content in alloy A25 (22.8 at
%). Also from Table 2, noticeable amount of the third component Zr
could dissolve into Fe2(Ti, Zr), with a maximum of 25.5 at% in alloy A6.
Therefore, the Fe2(Ti, Zr) phase would extend its homogeneous range
L. Zeng et al.
Fig. 5. The BSE images and XRD patterns of alloys annealed at 1173 K: (a) and (b) for A17; (c) and (d) for A18; (e) and (f) for A22.
into the ternary system, with the extension direction parallel to the Ti-
Zr boundary. In contrast to the Fe2(Ti, Zr) phase, quite few amount of Zr
was dissolved into FeTi.
Fig. 5(a) and (b) represent the BSE image and XRD patterns of alloy
A17. Evidently, this alloy was in a two-phase equilibrium. The black
phase was readily identified as the FeTi phase, while the white phase
was determined with a hexagonal crystal structure. Although the
structure of this white phase was the same as that of the τ1 phase de-
tected in A15, the compositions of this unidentified phase and τ1
showed a significant difference. Moreover, due to the determination of
the three-phase equilibrium of τ1 + τ2 + FeTi in A15, this phase in A17
should be another ternary phase, denoted as τ3. As show in Fig. 5(c) and
(d), a matrix phase with a composition Ti60.9Fe33.3Zr5.8 and a cubic
structure was determined in alloy A18, besides the FeTi and β(Ti, Zr)
phases. In alloy A22, a dark phase with a composition Ti50.4Fe33.2Zr16.4
and also a cubic structure was found to coexist with τ3 (see Fig. 5(e) and
26
Calphad 61 (2018) 20–32
(f)). If only from the view of the chemical composition and crystal
structure, one might assume the unidentified matrix phase in A18 and
the dark phase in A22 as the τ2 phase, and then τ2 would posses a
homogeneity range up to 60.9 at% Ti. Obviously, this assumption was
invalid as a three-phase equilibrium of τ2 + τ3 + β(Ti, Zr) was found in
alloy A25. Hence, these two unidentified phases should be considered
as a new ternary phase τ4.
Fig. 6 presents the BSE images of alloys A19 and A20, which were
situated closely to alloys A18 and A22. It can be seen from Fig. 6(a) that
A19 was in a three-phase equilibrium of FeTi + τ3 + τ4, and A20 in a
three-phase equilibrium of β(Ti, Zr) + τ3 + τ4 from Fig. 6(b). The
chemical compositions of τ4 (see data of A18–20 and A22 in Table 2)
indicated that τ4 was a line compound with a nearly constant Fe content
(33.3 at% on average) and a homogeneity range of 50.4 − 60.9 at% Ti.
For the τ3 phase, an ellipse-shaped composition region could be de-
duced from the phase composition data in Table 2.
L. Zeng et al.
Fig. 6. The BSE images for (a) alloy A19 and (b) alloy A20.
To further verify the four ternary phases, i.e. τ1, τ2, τ3 and τ4,
electron back-scattering patterns (EBSPs) and mapping functions in
EBSD were applied for identifying the crystallographic structure of each
phase. Fig. 7 shows the original Kikuchi patterns and the indexed re-
sults. All the Kikuchi lines of the four ternary phases were well indexed.
The crystal structure and space group of the τ1 phase detected in alloy
A9 (see Fig. 7(a)) were measured to be hexagonal and P63/mm, using
the input lattice parameters a = 0.4954 nm and c = 1.6304 nm col-
lected for the binary compound hex.-Fe2Zr in literature [19]. For the τ3
phase in alloy A20, the crystal structure and space group were mea-
sured to be identical with those of the τ1 phase, using the same lattice
parameters as those of τ1. For the τ2 and τ4 phases in alloys A12 and
A20 (see Fig. 7(b) and (d)), a cubic structure with the space group Fd3m
was determined using parameter a = 1.1328 nm collected for Ti2Fe in
literature [17].
As introduced above, the line compounds τ2 and τ4 were both of a
cubic structure and located near the iso-concentration line of 33.3 at%
Fe. The stoichiometric ratio of Fe/(Ti, Zr) in them approximately
equaled that in the Ti2Fe phase also with a cubic structure (see Table 1).
For the hexagonal structure τ1 and τ3 phases, their chemical composi-
tions were distributed around the 28.0 at% Zr iso-concentration line
with relatively large scatter, the averaged stoichiometric ratios of (Fe,
Ti)/Zr in them were both close to that in the hex.-Fe2Zr phase. It has
been widely recognized that the addition of the third component can
stabilize the metastable binary phase in the interior part of ternary
system [66–70]. Based on the above analyses, one could deduce that the
ternary phase τ1 & τ3 and τ2 & τ4 should result from the stabilization of
the metastable hex.-Fe2Zr and Ti2Fe phases with the addition of Ti and
Zr components, respectively.
Fig. 8(a) illustrates the isothermal section of the Ti-Fe-Zr system at
1173 K over the whole composition range, constructed based on the
analyses of experimental results. Eight three-phase equilibria were ob-
served. Near the Ti-Zr boundary, a continuous single phase region of
β(Ti, Zr) was obtained. Noticeable amount of the third component was
dissolved into the binary intermetallic phases Fe2(Ti, Zr) and (Fe,
Ti)2Zr. The ellipse-shaped homogeneity regions were determined for
both τ1 and τ3, while τ2 and τ4 possessed a nearly constant Fe content,
on average, 34.0 at% for τ2 and 33.3 at% for τ4, respectively.
3.2. Phase equilibria at 873 K
For the phase equilibria to be investigated at 873 K, only partial
isothermal section was determined (as shown in Fig. 8(b)). This mainly
resulted from the fact that the Fe-rich alloys (with Fe composition
larger than ~50.0 at%) were difficult to reach the equilibrium state,
even after annealing for 3600 h. In this part, we only discussed the
phase relations in those equilibrated alloys, the corresponding con-
stituent phases in them are given in Table 3.
27
Calphad 61 (2018) 20–32
From the microstructure of alloy B25 in Fig. 9(a), three phases were
observed in this alloy, i.e. the tiny white, gray and dark gray phases.
With the help of EPMA, the first phase was abundant in Zr, the last two
were of compositions Ti25.4Fe33.0Zr41.6 and Ti34.8Fe33.5Zr31.7, respec-
tively. With the further analysis of XRD results shown in Fig. 9(b), the
white phase was readily confirmed as α(Ti, Zr) phase. Two sets of dif-
fraction patterns with hexagonal and cubic structures could be dis-
criminated in the remaining peaks of B25, and the peak positions co-
incided with those of τ2 and τ3 in A25. Coupled with the composition
analysis, the gray and dark gray phases were identified as τ2 and τ3,
respectively. In addition, τ2 was also found to be in equilibrium with
α(Ti, Zr) in alloys B28, B29 and B36. Based on the compositions of the
τ2 phase contained in all alloys in Table 3, one could infer that τ2 was a
line compound with a nearly constant content of Fe (33.1 at% on
average), which was consistent with the case at 1173 K. The solubility
range of Ti in τ2 phase was measured about 6.7 − 25.4 at%.
Fig. 9(c) illustrates the BSE image of alloy B21. Similarly to the
phase relation in A21 at 1173 K (see Table 2), this alloy was also in a
three-phase equilibrium. Through the EPMA analysis, the dark gray
phase was considered as the β(Ti, Zr) phase, which was confirmed by
the XRD results in Fig. 9(d). The gray and light gray phases were
measured with compositions of Ti51.4Fe32.9Zr15.7 and Ti42.2Fe30.1Zr27.7.
The chemical compositions of the latter two phases deviated slightly
from those of τ4 and τ3 phases identified at 1173 K (see Table 2).
Moreover, their characteristic peaks shown in Fig. 9(d) were in good
consistency with the peaks diffracted by τ4 and τ3 captured in alloy
A22. Thus, the gray and light gray phases in B21 could be easily
identified as τ4 and τ3. Further measurements revealed that τ4 also
coexisted with FeTi + β(Ti, Zr) in alloy B20 and with FeTi + τ3 in B22.
The chemical compositions of τ4 in the above three alloys (B20-B22)
suggested that this phase was a line compound, with a nearly constant
Fe composition (32.9 at% Fe on average) and a solubility range of
50.6 − 56.4 at% Ti. Fig. 9(e) and (f) give the BSE image and XRD
patterns of alloy B34 sampled at the Ti-rich corner. Clearly, a two-phase
equilibrium of α(Ti, Zr) + β(Ti, Zr) was observed in this alloy. In
combination of the β(Ti, Zr) compositions in various alloys, there
should be a long and narrow β(Ti, Zr) single phase field, which would
start from the Fe-Ti boundary. Moreover, the three-phase equilibrium of
τ2 + τ3 + α(Ti, Zr) in alloy B25 implied that this phase field could not
continuously extend into the Zr-rich corner. In order to obtain the
limited solubility of Zr in β(Ti, Zr), alloy B38 was additionally prepared
near the Ti-Zr boundary. The BSE image and XRD patterns are pre-
sented in Fig. 10. A three-phase equilibrium of τ3 + α(Ti, Zr) + β(Ti,
Zr) was confirmed in this alloy. The composition of β(Ti, Zr) phase
indicated that the limited Zr solubility in this phase was 61.8 at% Zr,
where the β(Ti, Zr) phase field terminated.
Based on the above experimental results and analyses, the iso-
thermal section of the Ti-Fe-Zr ternary system at 873 K covering the
L. Zeng et al. Calphad 61 (2018) 20–32

Fig. 7. The original and indexed Kikuchi patterns of phases (a) τ1 in A9; (b) τ2 in A12; (c) τ3 in A20; (d) τ4 in A20. For each phase, the corresponding pearson symbol and space group are
given under the indexed patterns.

28
L. Zeng et al.
Fig. 8. Comparison of the isothermal sections of the Ti-Fe-Zr system at different temperatures. (a) and (b) are sections experimentally determined in this work, (c) and (d) are from Refs.
[56,57].
composition range of 0–50 at% Fe was established and illustrated in
Fig. 8(b). As for the phase relations in the remaining composition re-
gion, they were tentatively derived on the basis of those relations ob-
tained at 1173 K and illustrated with dashed lines. Five three-phase
regions were determined. A single phase field of β(Ti, Zr) extending to
61.8 at% Zr was determined. Among the three detected ternary phases
τ2, τ3 and τ4, both τ2 and τ4 were line compounds with about 33.0 at%
Fe, while τ3 exhibited a solubility rang of 34.8 − 42.2 at% Ti.
3.3. Comparison of phase relations at different temperatures
In this part, a comparison between the phase relations at 873 and
1173 K was conducted. As shown in Fig. 8(a) and (b), an overall
agreement on the general topology of the isothermal sections at these
two temperatures can be noted within the composition range of
0 − 50 at% Fe. For example, at both temperatures, three ternary phase
τ2, τ3 and τ4 were found at the same compositional region and τ2 and τ4
were both line compounds. A total of eight three-phase equilibria were
observed in the isothermal section at 1173 K, in contrast to only five
ones at 873 K due to the limited number of alloys well equilibrated.
Differences between these two sections also existed near the Ti-Zr
boundary. For example, a three-phase equilibrium of τ2 + τ3 + β(Ti,
29
Calphad 61 (2018) 20–32
Zr) was observed at 1173 K, but an equilibrium of τ2 + τ3 + α(Ti, Zr) at
873 K; the β(Ti, Zr) phase field starting from the Fe-Ti boundary could
continuously extend into the Zr-rich corner at 1173 K, but it terminated
at about 61.8 at% Zr at 873 K.
The isothermal sections constructed in this work were further
compared with the reported ones at 1173 K [56] and 1023 K [57]. The
last two sections are presented in Fig. 8(c) and (d). Remarkable dif-
ferences could be found in the phase relations between the current work
and the studies [56,57]. Most evidently, the four ternary phases (i.e. τ1,
τ2, τ3 and τ4) were absent in their works. In contrast to current sections,
a much larger solubility of the third component in FeTi, (Fe, Ti)2Zr,
FeZr2 and FeZr3 nevertheless a smaller solubility in Fe2(Ti, Zr) were
observed in sections at 1173 and 1023 K [56,57]. The huge dis-
crepancies in isothermal sections presented in Fig. 8 could be partially
attributed to the experimental method adopted to investigate the phase
relations. In literatures [56,57], the diffusion couple rather than the
equilibrated alloys was used to determine the isothermal section. Al-
though it is well known that in comparison with the latter method, the
former one is of relative high-efficiency and low-labor, some of the
phases may be not formed by inter-diffusion reactions in diffusion
couple [71–73]. Moreover, the crystal structure analyses were not
performed for confirming the phases in Refs. [56,57]. This likely
L. Zeng et al.
Fig. 9. The BSE images and XRD patterns of ternary alloys annealed at 873 K: (a) and (b) for B25, (c) and (d) for B21, (e) an (f) for B34, respectively.
brought in more uncertainty to their results, e.g. in obtaining the lim-
ited solubility of the third component in Fe2(Ti, Zr) and (Fe, Ti)2Zr.
Thus, the results obtained in this work should be more reliable.
4. Conclusion
In this study, the isothermal sections of the Ti-Fe-Zr ternary system
at 1173 (covering the entire composition range) and 873 K (only cov-
ering the partial range) were established based on the experiment data
acquired from SEM, EPMA, XRD and EBSD. At 1173 K, four stable
ternary phases τ1, τ2, τ3 and τ4 were found, and all of them were formed
by stabilizing the metastable binary intermetallic phases with the ad-
dition of the third component. Among them, τ2 and τ4 were line com-
pounds. Eight three-phase equilibria were detected in the isothermal
section at 1173 K. A continuous β(Ti, Zr) phase field near the Ti-Zr
boundary was observed to cover a composition scope from the Ti-rich
30
Calphad 61 (2018) 20–32
corner to the Zr-rich corner. At 873 K, five three-phase equilibria were
determined, and the β(Ti, Zr) phase field could not continuously extend
into the Zr-rich corner. The isothermal section constructed at 873 K was
in a broad consistency with that at 1173 K. Phase relations of Ti-Fe-Zr
system measured in this work showed remarkable differences from the
reported ones obtained by means of the diffusion couple.
Acknowledgment
This work was financially supported by the National Key Research
and Development Program of China (Grant No. 2016YFB0701301), 973
(Grant No. 2014CB644002) and the National Natural Science
Foundation of China (Grant No. 51671218).
L. Zeng et al.
Fig. 10. Alloy B38 annealed at 873 K: (a) BSE image and (b) XRD patterns.
Appendix A. Supporting information
Supplementary data associated with this article can be found in the
online version at http://dx.doi.org/10.1016/j.calphad.2018.02.005.
References
[1] M. Geetha, A.K. Singh, R. Asokamani, A.K. Gogia, Ti based biomaterials, the ulti-
mate choice for orthopaedic implants – a review, Prog. Mater. Sci. 54 (2009)
397–425.
[2] M. Abdel-Hady Gepreel, M. Niinomi, Biocompatibility of Ti-alloys for long-term
implantation, J. Mech. Behav. Biomed. 20 (2013) 407–415.
[3] M. Niinomi, Mechanical properties of biomedical titanium alloys, Mater. Sci. Eng. A
243 (1998) 231–236.
[4] J. Nejezchlebová, M. Janovská, H. Seiner, P. Sedlák, M. Landa, J. Šmilauerová,
J. Stráský, P. Harcuba, M. Janeček, The effect of athermal and isothermal ω phase
particles on elasticity of β-Ti single crystals, Acta Mater. 110 (2016) 185–191.
[5] J. Sun, Q. Yao, H. Xing, W.Y. Guo, Elastic properties of β, α'' and ω metastable
phases in Ti-Nb alloy from first-principles, J. Phys.: Cond. Mat. 44 (2007) 6215.
[6] M. Tane, Y. Okuda, Y. Todaka, H. Ogi, A. Nagakubo, Elastic properties of single-
crystalline ω phase in titanium, Acta Mater. 61 (2013) 7543–7554.
[7] M. Geetha, A.K. Singh, A.K. Gogia, R. Asokamani, Effect of thermomechanical
processing on evolution of various phases in Ti–Nb–Zr alloys, J. Alloy. Compd. 384
(2004) 131–144.
[8] Y. Zhou, Y. Li, X. Yang, Z. Cui, S. Zhu, Influence of Zr content on phase transfor-
mation, microstructure and mechanical properties of Ti75−xNb25Zrx (x = 0–6) al-
loys, J. Alloy. Compd. 486 (2009) 628–632.
[9] M. Abdel-Hady, H. Fuwa, K. Hinoshita, H. Kimura, Y. Shinzato, M. Morinaga, Phase
stability change with Zr content in β-type Ti–Nb alloys, Scr. Mater. 57 (2007)
1000–1003.
[10] L.-F. Huang, B. Grabowski, J. Zhang, M.-J. Lai, C.C. Tasan, S. Sandlöbes, D. Raabe,
J. Neugebauer, From electronic structure to phase diagrams: a bottom-up approach
to understand the stability of titanium–transition metal alloys, Acta Mater. 113
(2016) 311–319.
[11] C. Mao, M. Tan, L. Zhang, D. Wu, W. Bai, L. Liu, Experimental reinvestigation and
thermodynamic description of Bi-Te binary system, Calphad 60 (2018) 81–89.
[12] R. Schmid-Fetzer, Phase diagrams: the beginning of wisdom, J. Phase Equilib. Diff.
35 (2014) 735–760.
[13] S. Huang, X. Zhang, Y. Jiang, Y. Jiang, C. Mao, D. Wu, L. Zhang, L. Liu,
Experimental investigation of Ti-Nb-Co ternary system at 1000 °C, Mater. Des. 115
(2017) 170–178.
[14] L. Eleno, J. Balun, G. Inden, C. Schön, Phase equilibria in the Fe–Rh–Ti system II.
CVM calculations, Intermetallics 15 (2007) 1248–1256.
[15] T.B. Massalski, H. Okamoto, P.R. Subramanian, L. Kacprzak, Binary Alloy Phase
Diagrams, 2nd ed., ASM International, Materials Park, OH, 1992.
[16] M.S. Granovsky, M. Canay, E. Lena, D. Arias, Experimental investigation of the Zr
corner of the ternary Zr–Nb–Fe phase diagram, J. Nucl. Mater. 302 (2002) 1–8.
[17] P. Villars, K. Cenzual, PAULING FILE in: Inorganic Solid Phases, SpringerMaterials.
〈http://materials.springer.com/isp/crystallographic/docs/sd_0377957〉.
[18] P.I. Kripyakevich, V.S. Protasov, E.E. Cherkashin, Crystal structure of the compound
ZrFe3, Zh. Neorg. Khim. 10 (1965) 151–152.
[19] P. Villars, K. Cenzual, PAULING FILE in: Inorganic Solid Phases. 〈http://materials.
springer.com/isp/crystallographic/docs/sd_0455401〉.
[20] Y. Murakami, H. Kimura, Y. Nishimura, An investigation on the titanium-iron-
carbon system, Trans. Nat. Res. Inst. Met. 1 (1959) 7–21.
[21] P.H. Booker, Ternary Phase Equilibria in the Systems Ti-Fe-C, Ti-Co-C and Ti-Ni-C,
Oregon Graduate Center, Beaverton, OR, 1979.
[22] J.L. Murray, The Fe−Ti (Iron-Titanium) system, J. Phase Equilib. Diff. 2 (1981)
320–334.
[23] P.P.J. Ramaekers, F.J.J. van Loo, G.F. Bastin, Phase relations, diffusion paths and
31
Calphad 61 (2018) 20–32
kinetics in the system Fe-Ti-C at 1273 K, Z. Metallkd. 76 (1985) 245–248.
[24] L. Zeng, L. Liu, S. Huang, L. Zhang, Experimental investigation of phase equilibria in
the Ti-Fe-Cr ternary system, Calphad 58 (2017) 58–69.
[25] J.L. Murray, Phase Diagrams of Binary Titanium Alloys, ASM International, 1987,
pp. 99–111.
[26] K.C. Hari Kumar, L. Dumitrescu, B. Sundman, P. Wollants, Thermodynamic as-
sessment of the Fe−Ti system with special emphasis on the modeling of the FeTi
(B2) phase, in: Proceedings of the Calphad XXVIII Conference. Grenoble, 1999.
[27] S. Jonsson, Assessment of the Fe-Ti system, Metall. Mater. Trans. B 29 (1998)
361–370.
[28] G. Cacciamani, J. De Keyzer, R. Ferro, U.E. Klotz, J. Lacaze, P. Wollants, Critical
evaluation of the Fe-Ni, Fe-Ti and Fe-Ni-Ti alloy systems, Intermetallics 14 (2006)
1312–1325.
[29] C. Guo, C. Li, X. Zheng, Z. Du, Thermodynamic modeling of the Fe–Ti–V system,
Calphad 38 (2012) 155–160.
[30] H. Bo, J. Wang, L. Duarte, C. Leinenbach, L. Liu, H. Liu, Z. Jin, Thermodynamic re-
assessment of Fe–Ti binary system, Trans. Nonferrous Met. Soc. China 22 (2012)
2204–2211.
[31] A.D. Pelton, L. Leibowitz, R.A. Blomquist, Thermodynamic analysis of phase equi-
libria in the iron-zirconium system, J. Nucl. Mater. 201 (1993) 218–224.
[32] C. Servant, C. Gueneau, I. Ansara, Experimental and thermodynamic assessment of
the Fe-Zr system, J. Alloy. Compd. 220 (1995) 19–26.
[33] D.P. Abraham, J.W. Richardson, S.M. Mcdeavitt, Formation of the Fe23Zr6 phase in
an Fe-Zr alloy, Scr. Mater. 37 (1997) 239–244.
[34] F. Stein, G. Sauthoff, M. Palm, Experimental determination of intermetallic phases,
phase equilibria, and invariant reaction temperatures in the Fe-Zr system, J. Phase
Equilib. Diff. 23 (2002) 480–494.
[35] M. Jiang, K. Oikawa, T. Ikeshoji, L. Wulff, K. Ishida, Thermodynamic calculations of
Fe-Zr and Fe-Zr-C systems, J. Phase Equilib. Diff. 22 (2001) 406–417.
[36] H.-J. Lu, N. Zou, X.-S. Zhao, J.-Y. Shen, X.-G. Lu, Y.-L. He, Thermodynamic in-
vestigation of the Zr-Fe-Nb system and its applications, Intermetallics 88 (2017)
91–100.
[37] C. Guo, Z. Du, C. Li, B. Zhang, M. Tao, Thermodynamic description of the Al–Fe–Zr
system, Calphad 32 (2008) 637–649.
[38] Y. Yang, L. Tan, H. Bei, J.T. Busby, Thermodynamic modeling and experimental
study of the Fe–Cr–Zr system, J. Nucl. Mater. 441 (2013) 190–202.
[39] W.L. Huang, Y. Yu, S.Y. Yang, C.P. Wang, X.J. Liu, R. Kainuma, K. Ishida,
Experimental investigation of phase equilibria in the Cu-Fe-Zr ternary system, J.
Phase Equilib. Diff. 34 (2013) 438–446.
[40] J. Liang, M. Zhang, Y. Ouyang, G. Yuan, J. Zhu, J. Shen, M.R. Daymond,
Contribution on the phase equilibria in Zr–Nb–Fe system, J. Nucl. Mater. 466
(2015) 627–633.
[41] J.L. Murray, The Ti-Zr (Titanium-Zirconiium) system, J. Phase Equilib. 2 (1981)
197–201.
[42] K.C. Hari Kumar, P. Wollants, L. Delaey, Thermodynamic assessment of the Ti-Zr
system and calculation of the Nb-Ti-Zr phase diagram, J. Alloy. Compd. 206 (1994)
121–127.
[43] M.A. Turchanin, P.G. Agraval, A.R. Abdulov, Thermodynamic assessment of the Cu-
Ti-Zr system. II. Cu-Zr and Ti-Zr systems, Powder Metall. Met. C 47 (2008) 428.
[44] Y. Gao, Thermodynamic Study on the Ti−Sn−(Cr,Zr) Alloy Systems, University of
Science and Technology Beijing, Beijing, 2010.
[45] J. Cui, C. Guo, L. Zou, L. Changrong, Z. Du, Experimental investigation and ther-
modynamic modeling of the Ti–V–Zr system, Calphad 55 (Part 2) (2016) 189–198.
[46] S. Sridar, R. Kumar, K.C. Hari Kumar, Thermodynamic modelling of Ti-Zr-N system,
Calphad 56 (2017) 102–107.
[47] J.P. Auffredic, E. Etchessahar, J. Debuigne, Remarques sur le diagramme de phases
Ti-Zr: étude microcalorimétrique de la transition α⇄ β, J. Less-Common Met. 84
(1982) 49–64.
[48] J. Blacktop, J. Crangle, B.B. Argent, The α→β transformation in the Ti-Zr system
and the influence of additions of up to 50 at% Hf, J. Less-Common Met. 109 (1985)
375–380.
[49] D. Chatterji, M.T. Hepworth, S.J. Hruska, On the system Ti-Zr, Metall. Trans. 2
L. Zeng et al.
(1971) 1271–1272.
[50] E. Etchessahar, J. Debuigne, Investigation of the allotropic transformation of the
equiatomic titanium zirconium alloy, Sci. Rev. Metall. 74 (1977) 195–205.
[51] P.A. Farrar, S. Adler, System titanium-zirconium, Trans. Met. Soc. AIME 236 (1966)
1061.
[52] M. Ruch, D. Arias, Comments on the equilibrium diagram of the Ti-Zr system, Scr.
Mater. 29 (1993) 533–538.
[53] E. Rudy, Compendium of phase diagram data, Part V (Technical Report), AFML-TR-
65-2 21 (1969) 127–130.
[54] U. Thiedemann, M. Rösner-Kuhn, K. Drewes, G. Kuppermann, M.G. Frohberg,
Mixing enthalpy measurements of liquid Ti-Zr, Fe-Ti-Zr and Fe-Ni-Zr alloys, Steel
Res. 70 (1999) 3–8.
[55] R.H. Zee, J.F. Watters, R.D. Davidson, Diffusion and chemical activity of Zr-Sn and
Zr-Ti systems, Phys. Rev. B 34 (1986) 6895–6901.
[56] G.J. Zhou, S. Jin, L.B. Liu, H.S. Liu, Z.P. Jin, Determination of isothermal section of
Fe-Ti-Zr ternary system at 1173 K, Trans. Nonferrous Met. Soc. China 17 (2007)
963–966.
[57] G. Zhou, D. Zeng, Z. Liu, Phase equilibria in the Fe-Ti–Zr system at 1023 K, J. Alloy.
Compd. 490 (2010) 463–467.
[58] J.D. Ardisson, R.A. Mansur, E.G. da Silva, A study of structural and electronic
properties of the alloy systems (Zr1−xTix)2Fe and (Zr1−xTix)3Fe in the range 0 ≤ x
≤ 0.2, Scr. Mater. 25 (1991) 1327–1331.
[59] E. Piegger, R.S. Craig, Structural and magnetic characteristics of TiFe2-ZrFe2 and
ZrCo2-ZrFe2 alloys, J. Chem. Phys. 39 (1963) 137–145.
[60] V.V. Pet’kov, Y.A. Kocherzhinskiy, V.Y. Markiv, Investigation of the polythermal
sections ZrFe2-TiFe2 and ZrCo2-TiCo2, Dop. Akad. Nauk Ukr. RSR, Ser. A 33 (1971)
942–944.
[61] V.N. Svechnikov, Y.A. Kocherzhinsky, V.Y. Markiv, V.V. Pet’kov, Laves phases in
transition metal systems of the IV-VIII groups of periodic system, Akad. Nauk Ukr.
32
Calphad 61 (2018) 20–32
SSR, Metallofiz. 46 (1973) 35–45.
[62] Z. Blazina, R. Trojko, On Friauf-Laves phases in the Zr1–xAlxT2, Zr1–xSixT2 and
Zr1–xTixT2 (T = Mn, Fe, Co) systems, J. Less-Common Met. 133 (1987) 277–286.
[63] M. Budzyn’ski, J. Sarzyn’ski, Z. Surowiec, M. Wiertel, Moessbauer and x-ray dif-
fraction studies of Zr1–xTixFe2 laves phase compounds, Acta Phys. Pol. A 100 (2001)
717–722.
[64] Z. Surowiec, M. Wiertel, A.I. Beskrovnyi, J. Sarzyn’ski, J.J. Milczarek, Investigations
of microscopic magnetic properties of the pseudo-binary system (Zr1–xTix)Fe2, J.
Phys.: Cond. Matter 37 (2003) 6403–6414.
[65] A.J. Schwartz, M. Kumar, B.L. Adams (Eds.), Electron backscatter diffraction in
materials science, kluwer Acacemic/Plenum Publishers, New York, 2000.
[66] S. De Negri, M. Giovannini, A. Saccone, Phase relationships of the La–Ni–Mg system
at 500°C from 0 to 66.7at% Ni, J. Alloy. Compd. 397 (2005) 126–134.
[67] S. De Negri, P. Solokha, A. Saccone, V. Pavlyuk, Isothermal section of the La–Ni–Zn
system from 16.7 to 100at% La at 400°C, Intermetallics 16 (2008) 168–178.
[68] S. De Negri, A. Saccone, S. Delfino, The isothermal section of the La–Si–Mg system
at 500 °C, Calphad 33 (2009) 44–49.
[69] Z. Malik, A. Grytsiv, P. Rogl, G. Giester, Phase relations and crystal structures in the
system Ce–Ni–Zn at 800 °C, J. Solid State Chem. 194 (2012) 80–90.
[70] F. Zhang, H. Xu, Y. Du, R. Schmid-Fetzer, T. Zhou, Phase equilibria of the
Mg–La–Nd system at 500°C, J. Alloy. Compd. 585 (2014) 384–392.
[71] S. Cao, J.-C. Zhao, Application of dual-anneal diffusion multiples to the effective
study of phase diagrams and phase transformations in the Fe–Cr–Ni system, Acta
Mater. 88 (2015) 196–206.
[72] J.-C. Zhao, Combinatorial approaches as effective tools in the study of phase dia-
grams and composition–structure–property relationships, Prog. Mater. Sci. 51
(2006) 557–631.
[73] J.-C. Zhao, M.R. Jackson, L.A. Peluso, Determination of the Nb–Cr–Si phase dia-
gram using diffusion multiples, Acta Mater. 51 (2003) 6395–6405.