You are on page 1of 8

ARTICLE IN PRESS

Journal of Physics and Chemistry of Solids 69 (2008) 967–974
www.elsevier.com/locate/jpcs

Formability of ABO3 cubic perovskites
L.M. Feng, L.Q. Jiang, M. Zhu, H.B. Liu, X. Zhou, C.H. Li
College of Material Science and Engineering, Shanghai University, Shanghai 200072, China
Received 29 September 2006; received in revised form 30 October 2007; accepted 3 November 2007

Abstract

In this study, 223 binary oxide systems (of which, 34 systems can form cubic perovskites) are collected to explore the regularity of cubic
perovskites formability. It is found that the octahedral factor (rB/rO) take the same important role as the tolerance factor (t) to form
cubic perovskites in complex oxide system. Regularities governing cubic perovskites formability are obtained by using empirical structure
map constructed by these two parameters, on this structure map, sample points representing systems of forming (cubic structure) and
non-forming are distributed in distinctively different regions. Prediction criteria for the formability of cubic perovskites are squeezed out,
which may be applied to design new substrate or buffer materials with cubic perovskite structure in compound semiconductor epitaxy.
r 2007 Elsevier Ltd. All rights reserved.

Keywords: A. Oxide; A. Inorganic compounds; D. Crystal structure; D. Phase equilibria; D. Phase transitions

1. Introduction quality of their heteroepitaxial film, due to lattice mismatch
and thermal expansion mismatch between the film and the
Perovskite is named after a Russian geologist, Count Lev growth substrate or buffer materials. As one of the most
Aleksevich von Perovski [1]. It is one of the most frequently abundant and widely investigated minerals, perovskites
encountered structures in solid-state inorganic compound compounds are widely studied as the candidate materials of
[2]. Cubic perovskite structure is an ideal perovskite substrate or buffer materials. For example, cubic perovs-
structure, which has ABX3 stoichiometry and cubic crystal kites has recently gained widespread usage as a substrate or
structure, as seen in Fig. 1, where, A is a bigger cation (such buffer material for the heteroepitaxial growth of high
as Na1+, K1+, Ca2+, Sr2+, Ba2+), B is a smaller cation temperature superconductors and other oxide thin films [6].
(such as Ti4+, Nb5+, Mn4+, Zr4+), and X is an anion A perovskite-type crystal (orthorhombic distorted perovs-
(such as O2, F1, Cl1) [3–5]. Cubic perovskites are kites), NdGaO3, was reported as a succeeded candidate for
important crystal structure due to their diverse physical/ the GaN substrate [7]; (La,Sr)(Al,Ta)O3 (LSAT) [8] and
chemical properties over a wide temperature range. For (RE,Sr)(Al,Ta)O3 (RE, rare earth elements) [9], the mixed
example, BaCeO3, SrZrO3 have been known as protonic perovskite-type crystal, have very good characteristics to
conductor; SrSnO3 and BaSnO3 are sensitive to humidity be a substrate since it has a fairly small lattice mismatch
and oxygen gas, so they can be used as humidity or oxygen with GaN. In general, the semiconductor, which has blende
sensor. Furthermore, some cubic perovskites exhibit structure, can grow on the substrate or buffer layers with
ferromagnetic (such as SrCoO3), ferroelectrics, and di- cubic structure [10]. The recent success of fabricating large
electric properties [1–4]. GaAs MESFETs on Si substrates using an SrTiO3 (cubic
Compound semiconductors of III–V group, like GaAs, perovskites) as buffer layer increases the interest of
GaN, and InP, attract increasing attentions because of investigating cubic perovskites [11].
their high carrier mobility, wide and direct band gap. Therefore, lots of researchers try to design and
However, it is still a challenge to fabricate large and high synthesize new cubic perovskites used as buffer materials.
Obviously, if new cubic perovskites compounds and their
Corresponding author. Tel./fax: +86 21 56332934. lattice constants can be predicted, it is helpful to design
E-mail address: chli@staff.shu.edu.cn (C.H. Li). new substrate or buffer materials with cubic perovskite

0022-3697/$ - see front matter r 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.jpcs.2007.11.007
ARTICLE IN PRESS
968 L.M. Feng et al. / Journal of Physics and Chemistry of Solids 69 (2008) 967–974

ovskites, where rA, rB, and rO are the ionic radii of A, B,
and O, respectively. Geometrically, for the ideal perovs-
kites, the ratio of D(A–O) (the bond lengthpffiffiffi of A–O) to
D(B–O) (the bond length of B–O bond) is 2 : 1. Thus, if the
bond length is roughly assumed to be the sum of two ionic
radius, the t value of a cubic perovskites should be equal to
1.0. Goldschmidt found that, t values of cubic perovskites
are in the range of 0.8–0.9. Goldschmidt’s tolerance factor t
has been widely accepted as a criterion for the formation of
the perovskite structure, many researchers have used it to
discuss the cubic perovskites stability, and therefore, it is
an important factor for the stability of cubic perovskites, so
tolerance factor t constructs one axe of the structure map.
However, up to now, it seems that t is not a sufficient
condition for the formation of the cubic perovskite
structure, for example, in many systems whose t are even
with in the range (0.8–0.9), no cubic perovskite structure is
stable, such as, LaVO3, DyMnO3, and CaMoO3, as seen in
Table 1. So, another important factor governing form-
ability of cubic perovskites should be introduced.
As seen in Fig. 1, the cubic perovskite structure is
Fig. 1. Ideal perovskite structure illustrated for ABO3. composed of a three-dimensional framework of corner-
sharing BO6 octahedron. The A-site cation fills the 12
structure. In our previous study [12], a prediction model of coordinate cavities formed by the BO3 network and is
lattice constant in cubic perovskites ABX3 was obtained by surrounded by 12 equidistant anions. So, the octahedron
using two parameters, the sum of ionic radius of B and X BO6 is the basic mosaic or unit for cubic perovskites. It is
atoms and the ‘‘tolerance factor’’ (t ¼ (rA+rX)/ well known that coordination polyhedron has been used to
[21/2(rB+rX)]), the lattice constant can be predicted by this studying the stability of ionic compounds in solid-state
model with an average error of 0.63%. So, it is of practical inorganic chemistry. In general, if one cation (M) and six
and emergent to find out regularities governing the anions (X) form an octahedron MX6, the value of the ratio
formation of cubic perovskite-type compounds in order of their ionic radius rM/rX (rM and rX are ionic radius of
to seek for new buffer materials or substrates for cation M and anions X, respectively) should be varied with
compound semiconductor direct growth. the limited range 0.414–0.732 (we called this factor as the
Muller and Roy [2] proposed to plot ‘‘structural map’’, octahedron factors) [4], so the octahedron factor (rB/rO) is
which took the ionic radius of A and B as coordinates to as important as the tolerance factor to form cubic
study the distribution of different structures for many ternary perovskites. As mentioned above, in our previous work,
structural families. Furthermore, the schematic distribution the tolerance factor and the octahedron factor are used to
map of different crystal structure for A1+B5+O3, A2+B4+O3, span two-dimension structural map to study the form-
and A3+B3+O3 systems separately, were given by the same ability of perovskites, systems of forming perovskites are
method [2,13]. However, the criterion for cubic perovskites successfully distinguished with ones of non-forming per-
formability was not discussed, possibly due to the lack of ovskites. This success encourages us to use this kind of
accurate data of crystal structure of some ABO3 compounds structural map to investigate the formability of cubic
at that time. In our previous research, we have investigated perovskites.
the regularities governing perovskites formation by using
empirical structure map methods and a total 197 binary 3. Data collection
oxide systems in 2003 [14]. It is found that the octahedral
factor (rB/rO) is as important as the tolerance factor A total of 223 binary oxide systems are collected, as seen
(t ¼ (rA+rO)/[21/2(rB+rO)]), with regard to the formability in Table 1, of which, 34 systems (denote ‘‘y’’) are found to
of perovskites. Prediction criterions for the formability of have cubic perovskite structure, 189 systems (denote ‘‘n’’)
perovskites are obtained by using these two parameters. In cannot form cubic perovskite structure. It is well known, a
this study, the same method will be used to find the crystal structure can change with variation of temperature
regularities governing cubic perovskites formability. and/or pressure. Compounds with other structure can,
therefore, transform into cubic perovskite structure, and
2. trB/rO structural map method vice verse at different temperature and/or pressure. So, the
criterion for classifying a forming system is that a
In 1926, Goldschmidt had used ‘‘tolerance factor’’ compound of cubic perovskites is stabilized at room
(t ¼ (rA+rO)/[21/2(rB+rO)]) to study the stability of per- temperature and one atmosphere pressure. In the 189
ARTICLE IN PRESS
L.M. Feng et al. / Journal of Physics and Chemistry of Solids 69 (2008) 967–974 969

Table 1
Formability, ionic radius, tolerance factor, and octahedral factor of 223 binary oxides systems

No. Systems Formabilitya rA (Å) rB (Å) t rB/rO Ref.

1 Cs2O–I2O5 y 1.67 0.95 0.924 0.679 [3,5]
2 K2O–Ta2O5 y 1.38 0.64 0.964 0.457 [3,5]
3 BaO–MoO2 y 1.35 0.65 0.949 0.464 [3,5]
4 BaO–SnO2 y 1.35 0.69 0.93 0.493 [3,5]
5 CaO–CeO2 y 1.00 0.87 0.748 0.621 [3]
6 BaO–ZrO2 y 1.35 0.72 0.917 0.514 [3,5]
7 SrO–TiO2 y 1.18 0.61 0.908 0.436 [3,5]
8 SrO–SnO2 y 1.18 0.69 0.873 0.493 [3,5]
9 BaO–PbO2 y 1.35 0.78 0.892 0.557 [3,5]
10 BaO–PuO2 y 1.35 0.86 0.86 0.614 [3,5]
11 BaO–PrO2 y 1.35 0.85 0.864 0.607 [3,5]
12 BaO–CeO2 y 1.35 0.87 0.857 0.621 [3,5]
13 Rb2O–I2O5 y 1.52 0.95 0.879 0.679 [5]
14 K2O–I2O5 y 1.38 0.95 0.836 0.679 [5]
15 SrO–ZrO2 y 1.18 0.72 0.861 0.514 [5]
16 SrO–RuO2 y 1.18 0.62 0.903 0.443 [5]
17 Rb2O–U2O5 y 1.52 0.76 0.956 0.543 [5]
18 K2O–U2O5 y 1.38 0.76 0.91 0.543 [5]
19 Rb2O–Pa2O5 y 1.52 0.78 0.947 0.557 [5]
20 K2O–Pa2O5 y 1.38 0.78 0.902 0.557 [5]
21 BaO–NbO2 y 1.35 0.68 0.935 0.486 [5]
22 BaO–HfO2 y 1.35 0.71 0.922 0.507 [5]
23 SrO–VO2 y 1.18 0.58 0.921 0.414 [5]
24 SrO–NbO2 y 1.18 0.68 0.877 0.486 [5]
25 BaO–TbO2 y 1.35 0.76 0.9 0.543 [5]
26 BaO–AmO2 y 1.35 0.85 0.864 0.607 [5]
27 BaO–NpO2 y 1.35 0.87 0.857 0.621 [5]
28 BaO–PaO2 y 1.35 0.9 0.845 0.643 [5]
29 SrO–MoO2 y 1.18 0.65 0.89 0.464 [5]
30 SrO–CoO2 y 1.18 0.65 0.89 0.464 [5]
31 SrO–FeO2 y 1.18 0.59 0.917 0.421 [5]
32 SrO–HfO2 y 1.18 0.71 0.865 0.507 [5]
33 BaO–ThO2 y 1.35 0.94 0.831 0.671 [5]
34 EuO–TiO2 y 1.17 0.61 0.904 0.436 [5]
35 CaO–VO2 n 1.00 0.58 0.857 0.414 [15]
36 BaO–FeO2 n 1.35 0.59 0.977 0.421 [16]
37 CaO–ZrO2 n 1.00 0.72 0.8 0.514 [17]
38 La2O3–V2O3 n 1.03 0.64 0.842 0.457 [18]
39 Dy2O3–Mn2O3 n 0.91 0.58 0.825 0.414 [19]
40 CaO–ThO2 n 1.00 0.94 0.725 0.671 [3]
41 CdO–ThO2 n 0.74 0.94 0.647 0.671 [3]
42 CdO–CeO2 n 0.74 0.87 0.667 0.621 [3]
43 MgO–CeO2 n 0.72 0.87 0.66 0.621 [3]
44 SrO–ThO2 n 1.18 0.94 0.78 0.671 [3]
45 Ce2O3–Ga2O3 n 1.01 0.62 0.844 0.443 [3]
46 Ce2O3–Fe2O3 n 1.01 0.55 0.874 0.393 [3]
47 Gd2O3–Mn2O3 n 0.94 0.58 0.836 0.414 [3]
48 BaO–UO2 n 1.35 0.89 0.849 0.636 [3]
49 Eu2O3–Cr2O3 n 0.95 0.62 0.823 0.443 [3]
50 La2O3–Rh2O3 n 1.03 0.67 0.83 0.479 [3]
51 La2O3–Ti2O3 n 1.03 0.67 0.83 0.479 [3]
52 Nd2O3–Co2O3 n 0.98 0.55 0.863 0.393 [3]
53 Nd2O3–Mn2O3 n 0.98 0.58 0.85 0.414 [3]
54 Sm2O3–V2O3 n 0.96 0.64 0.818 0.457 [3]
55 PbO–ZrO2 n 1.19 0.72 0.864 0.514 [3]
56 Ce2O3–V2O3 n 1.01 0.64 0.835 0.457 [3]
57 Ce2O3–Cr2O3 n 1.01 0.62 0.844 0.443 [3]
58 Pu2O3–Mn2O3 n 1.00 0.58 0.857 0.414 [2]
59 Pr2O3–Mn2O3 n 0.99 0.58 0.854 0.414 [2,3]
60 BaO–TiO2 n 1.35 0.61 0.967 0.436 [2,3]
61 CaO–TiO2 n 1.00 0.61 0.844 0.436 [2,3]
62 Ce2O3–Al2O3 n 1.01 0.54 0.878 0.386 [2,3]
63 Eu2O3–Al2O3 n 0.95 0.54 0.857 0.386 [2,3]
ARTICLE IN PRESS
970 L.M. Feng et al. / Journal of Physics and Chemistry of Solids 69 (2008) 967–974

Table 1 (continued )

No. Systems Formabilitya rA (Å) rB (Å) t rB/rO Ref.

64 Eu2O3–Fe2O3 n 0.95 0.55 0.852 0.393 [2,3]
65 Gd2O3–Al2O3 n 0.94 0.54 0.853 0.386 [2,3]
66 Gd2O3–Cr2O3 n 0.94 0.62 0.819 0.443 [2,3]
67 Gd2O3–Fe2O3 n 0.94 0.55 0.849 0.393 [2,3]
68 K2O–Nb2O5 n 1.38 0.64 0.964 0.457 [2,3]
69 La2O3–Al2O3 n 1.03 0.54 0.886 0.386 [2,3]
70 La2O3–Cr2O3 n 1.03 0.62 0.851 0.443 [2,3]
71 La2O3–Fe2O3 n 1.03 0.55 0.881 0.393 [2,3]
72 La2O3–Ga2O3 n 1.03 0.62 0.851 0.443 [2,3]
73 Na2O–Ta2O5 n 1.02 0.64 0.839 0.457 [2,3]
74 Nd2O3–Al2O3 n 0.98 0.54 0.867 0.386 [2,3]
75 Nd2O3–Cr2O3 n 0.98 0.62 0.833 0.443 [2,3]
76 Nd2O3–Fe2O3 n 0.98 0.55 0.863 0.393 [2,3]
77 Pr2O3–Al2O3 n 0.99 0.54 0.871 0.386 [2,3]
78 Pr2O3–Cr2O3 n 0.99 0.62 0.837 0.443 [2,3]
79 Pr2O3–Fe2O3 n 0.99 0.55 0.867 0.393 [2,3]
80 Pr2O3–Ga2O3 n 0.99 0.62 0.837 0.443 [2,3]
81 Pr2O3–V2O3 n 0.99 0.64 0.828 0.457 [2,3]
82 Sm2O3–Al2O3 n 0.96 0.54 0.86 0.386 [2,3]
83 Sm2O3–Co2O3 n 0.96 0.55 0.856 0.393 [2,3]
84 Sm2O3–Fe2O3 n 0.96 0.55 0.856 0.393 [2,3]
85 Y2O3–Al2O3 n 0.90 0.54 0.838 0.386 [2,3]
86 Y2O3–Cr2O3 n 0.90 0.62 0.805 0.443 [2,3]
87 Y2O3–Fe2O3 n 0.90 0.55 0.834 0.393 [2,3]
88 Na2O–Nb2O5 n 1.02 0.64 0.839 0.457 [14]
89 Na2O–U2O5 n 1.02 0.76 0.792 0.543 [14]
90 Na2O–V2O5 n 1.02 0.54 0.882 0.386 [14]
91 Ag2O–V2O5 n 1.15 0.54 0.929 0.386 [14]
92 Ag2O–Ta2O5 n 1.15 0.64 0.884 0.457 [14]
93 Ag2O–Nb2O5 n 1.15 0.64 0.884 0.457 [14]
94 Ag2O–Sb2O5 n 1.15 0.6 0.902 0.429 [14]
95 Tl2O–I2O5 n 1.50 0.95 0.873 0.679 [14]
96 CaO–MoO2 n 1.00 0.65 0.828 0.464 [14]
97 CaO–RuO2 n 1.00 0.62 0.84 0.443 [14]
98 CaO–SnO2 n 1.00 0.69 0.812 0.493 [14]
99 SrO–CeO2 n 1.18 0.87 0.804 0.621 [14]
100 SrO–PbO2 n 1.18 0.78 0.837 0.557 [14]
101 CaO–UO2 n 1.00 0.89 0.741 0.636 [14]
102 CaO–HfO2 n 1.00 0.71 0.804 0.507 [14]
103 PbO–CeO2 n 1.19 0.87 0.807 0.621 [14]
104 PbO–TiO2 n 1.19 0.61 0.911 0.436 [14]
105 CaO–PbO2 n 1.00 0.78 0.778 0.557 [14]
106 CaO–MnO2 n 1.00 0.53 0.879 0.379 [14]
107 Sm2O3–Cr2O3 n 0.96 0.62 0.826 0.443 [14]
108 Er2O3–Fe2O3 n 0.89 0.55 0.83 0.393 [14]
109 Er2O3–V2O3 n 0.89 0.64 0.794 0.457 [14]
110 Ho2O3–Fe2O3 n 0.89 0.55 0.83 0.393 [14]
111 La2O–Mn2O3 n 1.03 0.58 0.868 0.414 [14]
112 Nd2O3–Ni2O3 n 0.98 0.56 0.859 0.4 [14]
113 Nd2O3–Ti2O3 n 0.98 0.67 0.813 0.479 [14]
114 Nd2O3–V2O3 n 0.98 0.64 0.825 0.457 [14]
115 Pr2O3–Ni2O3 n 0.99 0.56 0.862 0.4 [14]
116 Sm2O3–Ni2O3 n 0.96 0.56 0.851 0.4 [14]
117 Sm2O3–Ti2O3 n 0.96 0.67 0.806 0.479 [14]
118 Tb2O3–Fe2O3 n 0.92 0.55 0.841 0.393 [14]
119 Tb2O3–V2O3 n 0.92 0.64 0.804 0.457 [14]
120 Tm2O3–Fe2O3 n 0.88 0.55 0.827 0.393 [14]
121 Tm2O3–V2O3 n 0.88 0.64 0.79 0.457 [14]
122 Yb2O3–Fe2O3 n 0.86 0.55 0.82 0.393 [14]
123 Yb2O3–V2O3 n 0.86 0.64 0.783 0.457 [14]
124 Lu2O3–Al2O3 n 0.86 0.54 0.824 0.386 [14]
125 Y2O3–Ti2O3 n 0.90 0.67 0.786 0.479 [14]
126 Gd2O3–Ti2O3 n 0.94 0.67 0.799 0.479 [14]
127 Yb2O3–Al2O3 n 0.86 0.54 0.824 0.386 [14]
ARTICLE IN PRESS
L.M. Feng et al. / Journal of Physics and Chemistry of Solids 69 (2008) 967–974 971

Table 1 (continued )

No. Systems Formabilitya rA (Å) rB (Å) t rB/rO Ref.

128 Bi2O3–Fe2O3 n 1.03 0.55 0.881 0.393 [14]
129 La2O3–Co2O3 n 1.03 0.55 0.881 0.393 [14]
130 La2O3–Cu2O3 n 1.03 0.73 0.807 0.521 [14]
131 La2O3–Ni2O3 n 1.03 0.56 0.877 0.4 [14]
132 Dy2O3–Fe2O3 n 0.91 0.55 0.838 0.393 [14]
133 Lu2O3–Fe2O3 n 0.86 0.55 0.82 0.393 [14]
134 Bi2O3–Al2O3 n 1.03 0.54 0.886 0.386 [14]
135 Dy2O3–Cr2O3 n 0.91 0.62 0.809 0.443 [14]
136 Er2O3–Cr2O3 n 0.89 0.62 0.802 0.443 [14]
137 Yb2O3–Cr2O3 n 0.86 0.62 0.791 0.443 [14]
138 Ho2O3–Cr2O3 n 0.89 0.62 0.802 0.443 [14]
139 Tm2O3–Al2O3 n 0.88 0.54 0.831 0.386 [14]
140 La2O3–Y2O3 n 1.03 0.9 0.747 0.643 [14]
141 Tm2O3–Cr2O3 n 0.88 0.62 0.798 0.443 [14]
142 Lu2O3–Cr2O3 n 0.86 0.62 0.791 0.443 [14]
143 Er2O3–Al2O3 n 0.89 0.54 0.835 0.386 [14]
144 Dy2O3–Al2O3 n 0.91 0.54 0.842 0.386 [14]
145 Nd2O3–Ga2O3 n 0.98 0.62 0.833 0.443 [14]
146 Gd2O3–Ga2O3 n 0.94 0.62 0.819 0.443 [14]
147 Eu2O3–Ga2O3 n 0.95 0.62 0.823 0.443 [14]
148 Ho2O3–Al2O3 n 0.89 0.54 0.835 0.386 [14]
149 Li2O–Nb2O5 n 0.76 0.64 0.749 0.457 [14]
150 Li2O–Ta2O5 n 0.76 0.64 0.749 0.457 [14]
151 Li2O–V2O5 n 0.76 0.54 0.787 0.386 [14]
152 Rb2O–Nb2O5 n 1.52 0.64 1.012 0.457 [14]
153 Li2O–As2O5 n 0.76 0.46 0.821 0.329 [14]
154 Rb2O–Ta2O5 n 1.52 0.64 1.012 0.457 [14]
155 Cs2O–Nb2O5 n 1.67 0.64 1.064 0.457 [14]
156 K2O–As2O5 n 1.38 0.46 1.057 0.329 [14]
157 K2O–V2O5 n 1.38 0.54 1.013 0.386 [14]
158 Na2O–As2O5 n 1.02 0.46 0.92 0.329 [14]
159 Na2O–P2O5 n 1.02 0.38 0.961 0.271 [14]
160 Cs2O–V2O5 n 1.67 0.54 1.119 0.386 [14]
161 Cu2O–P2O5 n 0.77 0.38 0.862 0.271 [14]
162 K2O–P2O5 n 1.38 0.38 1.104 0.271 [14]
163 Li2O–P2O5 n 0.76 0.38 0.858 0.271 [14]
164 Tl2O–Sb2O5 n 1.5 0.6 1.025 0.429 [14]
165 Na2O–Sb2O3 n 1.02 0.6 0.856 0.429 [14]
166 Li2O–Sb2O5 n 0.76 0.6 0.764 0.429 [14]
167 K2O–Sb2O5 n 1.38 0.6 0.983 0.429 [14]
168 Na2O–Bi2O5 n 1.02 0.76 0.792 0.543 [14]
169 Li2O–Bi2O5 n 0.76 0.76 0.707 0.543 [14]
170 Ag2O–Bi2O5 n 1.15 0.76 0.835 0.543 [14]
171 BaO–GeO2 n 1.35 0.53 1.008 0.379 [14]
172 CaO–SiO2 n 1.00 0.4 0.943 0.286 [14]
173 CoO–TiO2 n 0.65 0.61 0.721 0.436 [14]
174 MgO–GeO2 n 0.72 0.53 0.777 0.379 [14]
175 MgO–TiO2 n 0.72 0.61 0.746 0.436 [14]
176 MnO–TiO2 n 0.83 0.61 0.785 0.436 [14]
177 ZnO–TiO2 n 0.74 0.61 0.753 0.436 [14]
178 FeO–TiO2 n 0.61 0.61 0.707 0.436 [14]
179 NiO–TiO2 n 0.69 0.61 0.735 0.436 [14]
180 FeO–SiO2 n 0.61 0.4 0.79 0.286 [14]
181 ZnO–SiO2 n 0.74 0.4 0.841 0.286 [14]
182 MnO–GeO2 n 0.83 0.53 0.817 0.379 [14]
183 MgO–SnO2 n 0.72 0.69 0.717 0.493 [14]
184 SrO–SiO2 n 1.18 0.4 1.014 0.286 [14]
185 EuO–SiO2 n 1.17 0.4 1.01 0.286 [14]
186 SmO–SiO2 n 1.19 0.4 1.017 0.286 [14]
187 BaO–SiO2 n 1.35 0.4 1.08 0.286 [14]
188 CoO–SiO2 n 0.65 0.4 0.805 0.286 [14]
189 SnO–PbO2 n 0.93 0.78 0.756 0.557 [14]
190 NiO–SiO2 n 0.69 0.4 0.821 0.286 [14]
191 PbO–SiO2 n 1.19 0.4 1.017 0.286 [14]
ARTICLE IN PRESS
972 L.M. Feng et al. / Journal of Physics and Chemistry of Solids 69 (2008) 967–974

Table 1 (continued )

No. Systems Formabilitya rA (Å) rB (Å) t rB/rO Ref.

192 MgO–SiO2 n 0.72 0.4 0.833 0.286 [14]
193 SrO–GeO2 n 1.18 0.53 0.945 0.379 [14]
194 PbO–GeO2 n 1.19 0.53 0.949 0.379 [14]
195 CdO–GeO2 n 1.10 0.53 0.916 0.379 [14]
196 CaO–GeO2 n 1.00 0.53 0.879 0.379 [14]
197 BaO–MnO2 n 1.35 0.53 1.008 0.379 [14]
198 CoO–MnO2 n 0.65 0.53 0.751 0.379 [14]
199 NiO–MnO2 n 0.69 0.53 0.766 0.379 [14]
200 La2O3–B2O3 n 1.03 0.23 1.054 0.164 [14]
201 Sc2O3–B2O3 n 0.75 0.23 0.933 0.164 [14]
202 Sm2O3–B2O3 n 0.96 0.23 1.024 0.164 [14]
203 Al2O3–B2O3 n 0.54 0.23 0.842 0.164 [14]
204 Ga2O3–Al2O3 n 0.62 0.54 0.736 0.386 [14]
205 Eu2O3–Ln2O3 n 0.95 0.8 0.755 0.571 [14]
206 Dy2O3–B2O3 n 0.91 0.23 1.002 0.164 [14]
207 Er2O3–B2O3 n 0.89 0.23 0.993 0.164 [14]
208 Eu2O3–B2O3 n 0.95 0.23 1.019 0.164 [14]
209 Gd2O3–B2O3 n 0.94 0.23 1.015 0.164 [14]
210 Ho2O3–B2O3 n 0.89 0.23 0.993 0.164 [14]
211 Y2O3–B2O3 n 0.90 0.23 0.998 0.164 [14]
212 Yb2O3–B2O3 n 0.86 0.23 0.98 0.164 [14]
213 Bi2O3–Sm2O3 n 1.03 0.96 0.728 0.686 [14]
214 V2O3–Cr2O3 n 0.64 0.62 0.714 0.443 [14]
215 V2O3–Al2O3 n 0.64 0.54 0.744 0.386 [14]
216 As2O3–B2O3 n 0.58 0.23 0.859 0.164 [14]
217 Gd2O3–Y2O3 n 0.94 0.9 0.719 0.643 [14]
218 Sm2O3–Y2O3 n 0.96 0.9 0.726 0.643 [14]
219 Tm2O3–B2O3 n 0.88 0.23 0.989 0.164 [14]
220 In2O3–Cr2O3 n 0.80 0.62 0.77 0.443 [14]
221 In2O3–Fe2O3 n 0.80 0.55 0.798 0.393 [14]
222 K2O–Bi2O5 n 1.38 0.76 0.91 0.543 [21]
223 Sc2O3–Al2O3 n 0.75 0.54 0.784 0.386 [14]
a
‘‘y’’ represents that binary oxide systems can form cubic perovskite; ‘‘n’’ represents that binary oxide systems cannot form cubic perovskite.

systems which cannot form cubic perovskites, 114 systems perovskites formability. As shown in Fig. 2, all cubic
(from no. 35 to no. 148) have distorted perovskite structure perovskites and non-cubic perovskites are located in two
(14 structures resulted by the BO6 octahedron tilting), 75 different regions, and a clear border between two kinds of
systems (from no. 149 to no.223) are not perovskite compounds is identified. The criterion of cubic perovskites
structure and represent at least one of following three formability is, then, expressed by the following equations:
condition: (1) there is no new ternary compound; (2) there rB
are new ternary compounds but their is no oxide of the ¼ 0:414, (1)
rO
chemical formula ABO3 and (3) there is at least one ABO3
compound but it is not a cubic perovskite structure. rA þ rO
Two hundred and twenty-three binary oxides systems pffiffiffi ¼ 0:815, (2)
2ðrB þ rO Þ
with their formability, ionic radius of constituent ions A
and B, tolerance factor, the octahedral factor (the ratio of rA þ rO
radius of the small cation B over the radii of anion O) and pffiffiffi ¼ 0:964, (3)
2ðrB þ rO Þ
references are listed in Table 1. The ionic radius used here
are the values for six coordination number (shannon’s rB rA þ rO
value), which are complemented from Crystal chemistry [5] ¼ 2:0645 pffiffiffi þ 2:29. (4)
rO 2ðrB þ rO Þ
and Perovskites and high Tc superconductor [3], the ionic
radii of O2 is 1.4 Å. For all the 34 cubic perovskites compounds, only one
compound, CaCeO3, is wrongly discriminated as a ‘‘non-
4. Results and discussions cubic perovskites’’. It is not clear why our model does not
work for it.
A structure map is constructed by the tolerance factor t The tolerance factor is a widely used parameter in
and the octahedral factor (rB/rO) to study the cubic perovskites study, which takes all the ionic radii into
ARTICLE IN PRESS
L.M. Feng et al. / Journal of Physics and Chemistry of Solids 69 (2008) 967–974 973

0.8
CaCeO3 BaUO3 TlIO3 PbZrO3 KBiO3 PbTiO3 KNbO3

0.7

0.6

Octahedral factor
0.5

0.4

0.3

0.2

0.1
cubic perovskites can not for mcubic perovskites
0
0.6 0.7 0.8 0.9 1 1.1 1.2
Tolerance factor

Fig. 2. Classification of cubic perovskite oxides.

consideration. It can be defined as mentioned above. However, BaUO3, TlIO3, KBiO3, PbZrO3,
rA þ rO and PbTiO3 have a structure close to cubic perovskite
t ¼ pffiffiffi . (5) structure. BaUO3 and PbZrO3 have pseudocubic perovskite
2ðrB þ rO Þ
structure [3]; TlIO3 has rhombohedral perovskite structure
And it is known that almost all perovskites have a t value with a ¼ 89.341 (very close to 901) [3]; PbTiO3 has tetragonal
ranging from 0.75 to 1.00. According to Goldschmidt’s point, perovskite structure at room temperature, a ¼ 3.896,
t values of cubic perovskites are in the range of 0.8–0.9. c ¼ 4.136 [3], and a phase transition to cubic perovskite
However, t values of 17 cubic perovskites oxides, half of all structure can occurs at 590 1C [20]. KBiO3 does not form the
cubic perovskites, are not in the range of 0.8–0.9, as seen in perovskite structure but rather forms a cubic structure first
Table 1. From Fig. 2, it is indicated that t values of cubic observed in KSbO3 (space group Im3̄ no. 204) [21].
perovskites, except CaCeO3, are in the range of 0.815–0.964, In Fig. 2, one point, representing KNbO3, just inside the
which is wider than Goldschmidt’s range. However, boundary, which has distorted perovskite structure at
t ¼ 0.815–0.964 is a necessary but not a sufficient condition room temperature, can transfer to cubic perovskite
for the formation of the cubic perovskite structure. The 102 structure at 510 1C [5]. Furthermore, some points near
systems, the t values of which are in the range (0.815–0.964), the boundary between ‘‘cubic perovskites’’ and ‘‘non-cubic
cannot form cubic perovskite structure. 82 systems among perovskites’’, as seen in Fig. 2, including CaTiO3, AgNbO3,
them have distorted perovskite structure, while 20 systems and AgTaO3, can undergo distorted perovskites to cubic
cannot form perovskites, as seen in Table 1. perovskites transformation. AgTaO3, AgNbO3, and Ca-
Octahedral factor (rB/rO) is introduced into prediction of TiO3 can change into cubic perovskite structure at 530,
cubic perovskites formation. In our structure map (Fig. 2), 610, and 1361 1C, respectively [5,22].
the lowest limit of the octahedral factor (rB/rO) for cubic In this research, the cubic structure is not found in
perovskites formation is 0.414, and the highest rB/rO value A3+B3+O3 perovskites, and there are no A3+B3+O3 com-
of cubic perovskites is 0.6785. It is well known that rM/rX pounds in the ‘‘cubic perovskites’’ region, as seen in Fig. 2.
value of octahedron MX6 is ranging from 0.414 to 0.732 According to Galasso’s report [3], DyMnO3 and LaVO3,
[4], the values of rB/rO for all 34 cubic perovskites are in which belong to A3+B3+O3 perovskites, have cubic
this range. However, octahedral factor (rB/rO value is not perovskite structure previously, however, the later reports
less 0.414) also is a necessary but not a sufficient condition indicate that LaVO3 and DyMnO3 have orthorhombic
for the formation of the cubic perovskite structure. The 100 structure at room temperature [18,19]. This is agreement
systems in the range (rB/rO is not less 0.414) cannot form with our result, and may imply other criterion for cubic
cubic perovskite structure, as seen in Table 1, of which, 72 perovskites formability.
systems have distorted perovskite structure and 28 systems Our model gives a simple and effective prediction
cannot form perovskite structure. criterion for cubic perovskites formability. New cubic
The criteria of cubic perovskites formability are shown in perovskite structure compounds can be predicted by using
Fig. 2 constructed with the tolerance factor t and the this model, and their lattice constant can be predicted by a
octahedral factor (rB/rO). From this figure, it can be seen that prediction model of lattice constant in cubic perovskites,
only five systems of all 189 ‘‘non-cubic perovskites’’, including which was obtained in our previous study [12]. Using these
BaUO3, TlIO3, KBiO3, PbZrO3, and PbTiO3, are wrongly two models, we can seek for new cubic perovskites
classified in ‘‘cubic perovskites’’ region by the criterion which lattice constant is close to compound semiconductor.
ARTICLE IN PRESS
974 L.M. Feng et al. / Journal of Physics and Chemistry of Solids 69 (2008) 967–974

This is useful to design new buffer materials or substrates [2] O. Muller, R. Roy, The Major Ternary Structural Families, Springer,
for compound semiconductor direct growth. New York, 1974.
[3] F.S. Galasso, Perovskites and High Tc Superconductors, Gordon and
Breach, New York, 1990.
5. Conclusions [4] Z.L. Wang, Z.C. Kang, Functional and Smart Materials:
Structural Evolution and Structure Analysis, Plenum Press,
In this research, 223 binary oxide systems are collected, of New York, 1998.
which, only 34 systems can form cubic perovskites, 114 [5] H.C. Chen, Crystal Chemistry, Shandong Education Press, Jinan,
1985 (In Chinese).
systems have distorted perovskites, and 75 systems cannot [6] A. Gupta, B.W. Hussey, T.M. Shaw, Mater. Res. Bull. 31 (1996)
form perovskites. Using these samples, regularities of 1463–1470.
formability of cubic perovskites are investigated by an [7] H. Takahashi, J. Ohta, H. Fujioka, M. Oshima, Thin Solid Films 407
empirical two-dimension structural map, which is spanned (2002) 114–117.
by octahedral factor (rB/rO) and the tolerance factor. Through [8] K. Shimamura, H. Tabata, H. Takeda, V.V. Kochurikhin,
T. Fukuda, J. Cryst. Growth 194 (1998) 209–213.
this study, the following conclusions are obtained: [9] M. Ito, K. Shimamura, D.A. Pawlak, T. Fukuda, J. Cryst. Growth
235 (2002) 277–282.
1. Octahedral factor (rB/rO) is as important as the [10] Y.G. Tian, M.S. Cao, C.B. Cao, Introduction of Advanced
tolerance factor (t ¼ (rA+rO)/[21/2(rB+rO)]) for cubic Materials, Haerbin Institute of Technology Press, Haerbin, 2005
perovskites formability. (In Chinese).
[11] K. Eisenbeiser, R. Emrickm, R. Droopad, Z. Yu, J. Finder,
2. Both tolerance factor and octahedral factor are a S. Rockwell, J. Holmes, C. Overgaard, W. Ooms, IEEE Electron
necessary but not sufficient condition for cubic per- Devices Lett. 23 (2002) 300–302.
ovskites formability. Using these two factors, the cubic [12] L.Q. Jiang, J.K. Guo, H.B. Liu, M. Zhu, X. Zhou, P. Wu, C.H. Li,
perovskites formability can be reliably predicted. J. Phys. Chem. Solids 67 (2006) 1531–1536.
3. In the structural map that is drawn by the tolerance [13] A.S. Bhalla, R.Y. Guo, R. Roy, Mater. Res. Innovations 4 (2000)
3–26.
factor (t ¼ (rA+rO)/[21/2(rB+rO)]) and octahedral fac- [14] C.H. Li, K.C.K. Soh, P. Wu, J. Alloys Compd. 372 (2004) 40–48.
tor (rB/rO), the points for cubic perovskites and those for [15] H. Falcon, J.A. Alonso, M.T. Casais, M.J. Martinez-Lope,
non-cubic perovskites are located in different zones via J. Sanchez-Benitez, J. Solid State Chem. 177 (2004) 3099–3104.
clear boundary defined by Eqs. (1)–(4). These equations [16] K. Mori, T. Kamiyama, H. Kobayash, K. Itoh, T. Otomo, S. Ikeda,
form the criteria for cubic perovskites formability. This Physica B 329–333 (2003) 807–808.
[17] N.L. Ross, T.D. Chaplin, J. Solid State Chem. 172 (2003)
simple model may be applied to design new substrate or 123–126.
buffer materials with cubic perovskites structure in [18] R.T.A. Khan, J. Bashir, N. Iqbal, M.N. Khan, Mater. Lett. 58 (2004)
compound semiconductor epitaxy. 1737–1740.
[19] T. Mori, K. Aoki, N. Kamegashira, T. Shishido, T. Fukuda, Mater.
Lett. 42 (2000) 387–389.
[20] Y. Kuroiwa, S. Aoyagi, A. Sawada, Phys. Rev. Lett. 87 (2001)
References 217601(1)–217601(4).
[21] T.N. Nguyen, D.M. Giaquinta, W.M. Davis, Chem. Mater. 5 (1993)
[1] H. Tanaka, M. Misono, J. Curr. Opin. Solid State Mater. Sci. 5 1273–1276.
(2001) 381–387. [22] R. Ali, M. Yashima, J. Solid State Chem. 178 (2005) 2867–2872.