SOLID-STATE EQUILIBRIA AND THERMODYNAMIC PROPERTIES OF COMPOUNDS
1077
First, we synthesized Bi2Te3, BiTe, Bi2Te, BiTeI,
Bi2TeI, Bi4TeI1.25, BiI3, and BiI by melting stoichio-
metric high-purity elemental mixtures in silica ampules
sealed off under a vacuum of ~10–2 Pa. Bismuth tellu-
rides were synthesized at ~900 K and then slowly
cooled. Given the high iodine vapor pressure, bismuth
iodides and bismuth telluride iodides were synthesized
in an inclined two-zone furnace. The temperature of the
upper, “cold” zone was ~370 K, and that of the lower,
“hot” zone was 20–30 K above the melting point of the
compound.
The first equilibrium emf values were measured
after ~30 h; subsequently, measurements were taken
every 4–5 h after the temperature had stabilized.
RESULTS AND DISCUSSION
From DTA results for thoroughly homogenized
Bi2TeI and Bi4TeI1.25 (first obtained by Savilov et al.
[13] using chemical vapor deposition), we determined
their melting points and melting behavior. These com-
pounds melt incongruently at 723 and 688 K, respec-
tively. According to powder XRD data, the lattice
parameters of BiTeI are a = 4.311 Å and c = 6.831 Å
(Z = 1), and those of Bi2TeI are a = 7.585 Å, b = 4.381 Å,
c = 17.740 Å, and β = 98.20°, in agreement with earlier
results [2, 13].
After most of the iodine entered into the reaction,
the ampule was lowered to the hot zone, and the reac-
tion was thoroughly stirred. Next, the ampule was
slowly (~2–3 K/min) cooled. In this way, the congru-
ently melting compounds Bi2Te3, BiI3, and BiTeI were
obtained in a homogeneous state. To fully homogenize
the incongruently melting compounds BiTe, Bi2Te, BiI,
Bi2TeI, and Bi4TeI1.25, they were annealed 30–50 K
below the corresponding peritectic temperature for
500–800 h (the peritectic decomposition temperatures
of Bi2TeI and Bi4TeI1.25 were determined to be 723 and
688 K, respectively). Every 200 h of annealing, the
alloys were reground and pressed into pellets. The com-
pletion of reactions was checked by differential thermal
analysis (DTA) and x-ray diffraction (XRD) using ear-
lier results [13, 17, 18].
The present experimental data in combination with
earlier data for the constituent binaries Bi–Te [17], Bi–
I [18], and Te–I [18] and the system Bi2Te3–BiI3–I–Te
[14–16] were used to construct the 300-K section of the
Bi–Te–I phase diagram (figure). The system contains
three ternary compounds: BiTeI (I), Bi2TeI (II), and
Bi4TeI1.25 (III).
BiI3, the most thermodynamically stable compound
in the system, plays a key role in determining the phase
compatibility diagram. BiI3 is in equilibrium with Te,
TeI4, TeI, and the three ternary compounds.
In addition, Bi–Bi2Te3–BiI3 alloys were synthesized
by vacuum-melting the constituent phases, followed by
annealing to the above schedule and then at 500 K for
an additional 200 h.
Since bismuth, bismuth tellurides, and the ternary
compounds have similar XRD patterns, the phase fields
in the Bi–BiI3–Bi2Te3 region are difficult to identify by
XRD, but the problem can easily be solved using emf
measurements. In the figure, we indicate the 300-K emf
(mV) of cell (1) for some of the phase fields in the
Bi−Te–I system. The emf is composition-independent
within each three-phase field and changes jumpwise in
going from one three-phase field to another. For exam-
ple, the emf in the BiI3–III–II region differs from that in
BiI3–II–I by 73 mV (the accuracy in our emf measure-
ments was 0.2 mV).
The alloys were characterized by DTA (NTR-72
pyrometer, Chromel–Alumel thermocouples), XRD
(DRON-2 diffractometer, CuKα radiation), and emf
measurements using concentration cells of the type
(–)Bi(s) | glycerol + KI + BiI3 | Bi–Te–I(s)(+). (1)
Such concentration cells are widely used in thermo-
dynamic studies of ternary chalcogenides [19].
From emf data for cell (1), we were able to evaluate
the thermodynamic properties of the ternary com-
pounds. To this end, we used emf data for the three-
phase regions BiI3–I–Te, BiI3–I–II, and BiI3–II–III,
which correspond to potential-forming reactions
involving the ternary compounds [19, 20].
The left electrode in cell (1) was metallic bismuth,
and the right electrode was an equilibrium alloy of the
system under investigation. The composition of the
right electrode was chosen using phase-diagram data as
described elsewhere [19]. The electrolyte used was a
saturated KI solution in glycerol with the addition of
0.1 wt % BiI3.
The temperature dependences of the emf in these
phase regions were found to be almost linear. This con-
firms that the phases coexisting in these regions have
constant compositions in the temperature range of our
emf measurements and allows the partial entropy and
enthalpy to be evaluated from the temperature coeffi-
cient of emf [19, 20]. The emf data were represented by
least squares linear equations of the form [21, 22]
The procedures used to prepare the electrolyte and
electrodes and to set up the electrochemical cell were
similar to those described earlier [19, 20]. The cell was
mounted in a vertical tubular furnace. The emf of cell
(1) was measured in the temperature range 305–400 K
by a compensation technique using a V7-34A digital
voltmeter. The temperature was measured by a
Chromel–Alumel thermocouple and a mercury ther-
mometer with 0.2°C divisions.
1/2
S2E
n
E = a + bT t ----- + S2b(T – T)2
,
(2)
INORGANIC MATERIALS Vol. 44 No. 10 2008