You are on page 1of 2

Abstract

When decreasing the temperature of (Sn,Sb) electrolytic cells we have obtained


accurate information concerning the liquids line in the binary (Sn,Sb) phase
diagram and strong evidence that the presently accepted phase diagram is
doubtful in the solid domains: the e.m.f. behavior in the so-considered nonstoichiometric monophase domain is not monophase but can be interpreted
by three different solid/solid equilibria between stoichiometric compounds
66 at.% Sb, 52 at.% Sb. Probably the phase, which forms at 597 K
according to the literature, is also stoichiometric 50 at.% Sb.

The e.m.f. equtions for the three observed (solid+solid) domains, (Sb) +
E/mV = 59.58 0.0464T/K. + E/mV = 61.63 0.0527T/K. +
E/mV = 67.26 0.0695 T/K, have been established with an experimental
uncertainty of 0.5 mV. Two equations of the liquidus line, above and below
the T = 696.5 K peritectic equilibrium were also established. We have adjusted
a subregular Redlich-Kister development for the Gibbs excess function of the
liquid phase; the integral mixing enthalpy at 800 K is Hmmim (liquid) = x(1x)|
5272 1431)12x)| and the integral excess entropy Smss/J mol1K = x(1x)
[1.779 2.314(12x)]. The thermodynamic formation functions of the (Sb)
primary solution and of the , , , and compounds have been evaluated.

A structural discussion completes this thermodynamic approach. The Sb and


phases exhibit many structural similarities. Both crystallographic cells can be
described as hexagonal non-compact packing by the periodic stacking of six
planes in the c-direction. The structure is ordered in the c-direction with,
alternately, an Sb atom or an Sn atom occupancy, which strongly reduces the
c-axis The two extreme structural designs are constituted by (Sb,Sb,Sb) tripleplane packing, with two different distances in the c-direction, or by an
(Sb,Sn,Sb) sandwich, with two equal distances in the c-direction. We can
imagine many stacking variants for the intermediate compounds. The X-ray
patterns of all the , , , or structures are very similar and close to the
f.c.c. NaCl type. The thermodynamic driving force between any biphase system
of two stacking variants (with unequal stacking periods) is very close to any
other, which is proved by the fact that the formation enthalpies of all the
intermediate compounds, versus composition, are very close to a straight line.

The Sb-Sn phase diagram in [Massalski2] was redrawn


from [1971Pre].
This phase diagram was updated experimentally and theoretically
by [1995Oht], as introduced by [1998Oka]. [2008Man]
also got a similar result by thermodynamic modeling.
Figure 1 shows the Sb-Sn phase diagram proposed by
[2008Che]. The phase diagram was determined by thermodynamic
modeling based on experimental data available in
the literature supplemented by their own DTA, micrograph,
and x-ray diffraction data.
This phase diagram differs from the earlier ones most
noticeably with regard to the Sb2Sn3 phase. In [Massalski2],
this phase exists only in a limited temperature range (250320 C). This phase exists at 57 at.% Sn, not at the stoichiometric
composition. Its crystal structure is still unknown.
[1997Vas] doubted the existence of the b phase having a
broad solubility range and proposed the existence of three
stoichiometric phases in this system at 34, 48, and probably
50 at.% Sn. There has been no supporting report on this
feature since then.