Thermodynamic properties and decomposition of lithium hexafluoroarsenate, LiAsF6

Article abstract of DOI:10.1023/A:1022102914631

The heat capacity of lithium hexafluoroarsenate is determined in the temperature range 50-750 K by adiabatic and differential scanning calorimetry techniques. The thermodynamic properties of LiAsF₆ under standard conditions are evaluated: C_p⁰ (298.15 K)= 162.5 ± 0.3 J/(K mol), S⁰(298.15 K) = 173.4 ± 0.4 J/(K mol), Φ⁰(298.15 K) = 81.69 ± 0.20 J/(K mol), and H ⁰(298.15 K) - H⁰(0) = 27 340 ± 60 J/mol. The C _p(T) curve is found to contain a lambda-type anomaly with a peak at 535.0 ± 0.5 K, which is due to the structural transformation from the low-temperature, rhombohedral phase to the high-temperature, cubic phase. The enthalpy and entropy of this transformation are 5.29 ± 0.27 kJ/mol and 10.30 ± 0.53 J/(K mol), respectively. The thermal decomposition of LiAsF₆ is studied. It is found that LiAsF₆ decomposes in the range 715-820 K. The heat of decomposition, determined in the range 765-820 K using a sealed crucible and equal to the internal energy change ΔU _r(T), is 31.64 ± 0.08 kJ/mol.

Full text of DOI:10.1023/A:1022102914631

Inorganic Materials, Vol. 39, No. 2, 2003, pp. 175–182. Translated from Neorganicheskie Materialy, Vol. 39, No. 2, 2003, pp. 227–235.
Original Russian Text Copyright © 2003 by Gavrichev, Sharpataya, Gorbunov, Golushina, Plakhotnik, Goncharova, Gurevich.
Thermodynamic Properties and Decomposition
of Lithium Hexafluoroarsenate, LiAsF6
K. S. Gavrichev*, G. A. Sharpataya*, V. E. Gorbunov*, L. N. Golushina*,
V. N. Plakhotnik**, I. V. Goncharova**, and V. M. Gurevich***
Kurnakov Institute of General and Inorganic Chemistry, Russian Academy of Sciences,
Leninskii pr. 31, Moscow, 119991 Russia
*
*
* Dnepropetrovsk State Technical University of Transport,
ul. Akademika Lazaryana 2, Dnepropetrovsk, 49010 Ukraine
*
** Vernadsky Institute of Geochemistry and Analytical Chemistry, Russian Academy of Sciences,
ul. Kosygina 19, Moscow, 117975 Russia
e-mail: gavrich@igic.ras.ru
Received August 21, 2002
Abstract—The heat capacity of lithium hexafluoroarsenate is determined in the temperature range 50–750 K
by adiabatic and differential scanning calorimetry techniques. The thermodynamic properties of LiAsF under
6
0
p
0
standard conditions are evaluated: C (298.15 K) = 162.5 ± 0.3 J/(K mol), S (298.15 K) = 173.4 ± 0.4 J/(K mol),
0
0
0
Φ (298.15 K) = 81.69 ± 0.20 J/(K mol), and H (298.15 K) – H (0) = 27340 ± 60 J/mol. The C (T) curve is found
p
to contain a lambda-type anomaly with a peak at 535.0 ± 0.5 K, which is due to the structural transformation
from the low-temperature, rhombohedral phase to the high-temperature, cubic phase. The enthalpy and entropy
of this transformation are 5.29 ± 0.27 kJ/mol and 10.30 ± 0.53 J/(K mol), respectively. The thermal decompo-
sition of LiAsF is studied. It is found that LiAsF decomposes in the range 715–820 K. The heat of decompo-
6
6
sition, determined in the range 765–820 K using a sealed crucible and equal to the internal energy change
∆
U (T), is 31.64 ± 0.08 kJ/mol.
r
INTRODUCTION
LOW-TEMPERATURE MEASUREMENTS
Given that lithium hexafluoroarsenate, LiAsF , is a
The low-temperature heat capacity of a 2.7895-g
6
promising material for lithium batteries, its thermody- lithium hexafluoroarsenate sample was measured in a
3
namic properties are of considerable interest. This work vacuum adiabatic calorimeter with a 2-cm caloric cell
continues our studies of the thermodynamic properties [4]. To improve heat exchange in the calorimeter, we
of alkali hexafluoroarsenates. Earlier, we investigated
the heat capacity of rubidium and cesium hexafluoro-
arsenates in a wide temperature range [1–3].
used helium gas (p = 3 kPa). The heat capacity data for
LiAsF in the range 51–331 K (72 data points) are sum-
6
2
2
50
00
SAMPLE PREPARATION
Lithium hexafluoroarsenate was prepared by react-
ing lithium metaarsenate and a 70% aqueous HF solu-
tion, with rapid water removal. The reaction product
was then purified by recrystallization from anhydrous
HF. According to analytical data, the sample contained
150
00
1
2
3
1
9
9.7 wt % LiAsF , 0.19 wt % hydroxylated species (in
6
5
0
particular, LiAsF OH and LiAsF (OH) ), 0.07 wt %
5
4
2
HF, and 0.04 wt % water. No insoluble arsenates were
detected. The pH of aqueous LiAsF solutions was 3.2.
6
0
100
200
300
400
T, K
The sample was stored in a sealed silica tube. All
manipulations, including the loading of crucibles and
calorimetric cells, were carried out in a dry nitrogen
atmosphere.
Fig. 1. Heat capacity of LiAsF6 in the range 0–450 K:
1) adiabatic calorimetry data, (2) differential scanning cal-
orimetry data, (3) extrapolation to absolute zero.
(
0
020-1685/03/3902-0175$25.00 © 2003 MAIK “Nauka/Interperiodica”

1
76
GAVRICHEV et al.
Table 1. Low-temperature heat capacity, J/(K mol), of converted into those of the fractal dimension (Fig. 2) by
LiAsF6
the procedure described earlier [9].
T, K
1.61
Cp
T, K
Cp
T, K
Cp
The smoothed heat capacity data and thermody-
namic functions of lithium hexafluoroarsenate are sum-
marized in Table 2. The thermodynamic properties at
standard conditions are
5
5
5
6
6
6
6
7
7
7
8
8
8
8
9
9
9
32.40
34.59
37.01
39.29
41.50
43.63
45.69
47.69
49.81
51.85
53.89
55.88
57.30
59.59
61.71
63.78
65.89
68.03
70.13
72.20
74.62
76.53
78.55
80.78
123.07
126.98
130.92
134.85
138.82
142.78
146.68
83.44 219.84 135.9
86.07 224.00 137.5
88.71 228.17 139.4
91.26 232.34 140.9
93.75 236.58 142.6
96.42 240.80 144.1
98.68 245.51 145.9
4.43
7.67
0.85
3.88
6.86
9.84
2.78
5.70
8.57
1.44
4.23
6.28
9.26
2.21
5.14
8.11
0
C (298.15 ä) = 162.5 ± 0.3 J/(K mol),
p
S0(298.15 K) = 173.4 ± 0.4 J/(K mol),
0
Φ (298.15 K) = 81.69 ± 0.20 J/(K mol),
0
0
H (298.15 K) – H (0) = 27340 ± 60 J/mol.
150.65 101.0
154.63 103.3
158.63 105.6
162.63 107.9
166.69 110.1
170.76 112.3
174.87 114.6
178.97 116.7
183.04 118.9
187.11 120.9
191.17 122.9
195.27 124.9
199.39 126.8
203.52 128.6
207.53 130.5
211.55 132.3
215.69 134.1
250.71 147.7
255.92 149.4
261.13 151.1
266.37 152.9
271.61 154.6
276.84 156.3
282.10 157.9
287.31 159.4
292.50 160.9
296.77 162.0
299.40 162.9
304.66 164.4
309.94 165.7
315.20 167.2
320.47 168.8
325.75 170.4
331.03 171.3
In evaluating the error, we assumed (based on earlier
experimental C data for lithium compounds with com-
p
plex anions [10–14]) that lithium hexafluoroarsenate
undergoes no structural phase transitions below 50 K.
HIGH-TEMPERATURE MEASUREMENTS
The thermal properties of lithium hexafluoroarsen-
ate were studied using a SETARAM DSC 111 differen-
tial scanning calorimeter (precision in temperature
measurements, ±1 K; uncertainty in temperature scale,
±
1 K; maximum sensitivity, 15 µW). Since LiAsF is
6
1
1
1
1
1
1
1
01.07
04.02
06.97
09.98
12.94
15.91
19.27
very hygroscopic, we used Ni-sealed stainless steel cru-
3
cibles (0.15-cm volume).
First, several LiAsF samples were subjected to the
6
following temperature program: heating from 290 to
6
1
23–673 K at a rate of 3 K/min, isothermal hold for
0 min, and cooling at 3 or 5 K/min (Fig. 3). Thermal
effects were evaluated using a SETARAM program for
the graphic integration of DSC peaks. The integration
error was 0.5–2%. As the peak temperature we took
either the temperature at which the auxiliary line
through the ascending peak slope intersected the lin-
marized in Table 1 and Fig. 1 (the molecular weight of
early extrapolated initial baseline (T ) or the maximum
e
LiAsF was taken to be 195.853). No anomalies were
6
peak temperature (Tp).
detected in the C (T) curve in the temperature range
p
studied. The C (T) data were smoothed using an appli-
p
The heating curves of LiAsF show an endotherm at
6
cation program incorporated in the IVTANTERMO 495–560 K, which is obviously due to the structural
Database [5]. The discrepancy between the measured transformation of LiAsF [15, 16]. The enthalpy of this
6
and smoothed data in the range 50–330 K was no transformation is ∆ H(endo) = 3.728 ± 0.075 kJ/mol;
tr
greater than 0.2%.
T = 534.6 ± 0.3 K, T = 538.7 ± 0.4 K (average of four
e
p
scans).
To extrapolate the data to absolute zero, we used a
technique which takes into account that compounds
similar in the structure of the complex anion and crystal
The cooling curves of LiAsF show an exotherm in
the range 540–485 K, with an enthalpy of transforma-
structure have similar temperature dependences of the tion –∆ H(exo) = 3.742 ± 0.056 kJ/mol; T 531.4 ±
6
tr
e
fractal dimension [6]. Lithium hexafluorophosphate
and lithium hexafluoroarsenate have an LiSbF -type
0
.8 K, T = 526.1 ± 0.2 K (average of three scans). The
p
6
estimated uncertainty in ∆ H includes the errors relat-
tr
rhombohedral structure (R ) and are close in lattice ing to peak integration and ∆ H averaging over differ-
1
tr
parameters [7]. In view of this, the temperature depen- ent samples. Our results demonstrate that the structural
dences of heat capacity for LiAsF and LiPF [8] were transformation of LiAsF is reversible and involves no
6
6
6
INORGANIC MATERIALS Vol. 39 No. 2 2003

THERMODYNAMIC PROPERTIES AND DECOMPOSITION
177
4
3
2
1
2
1
1
0
5
0
5
0
5
526.1
EXO
1
2
Cooling
–
–
–
–
–
10
15
20
25
Heating
ENDO
538.7
3
50 400 450 500 550 600 650 700
0
50
100
150
200
T, K
T, K
Fig. 2. Fractal dimension of (1) LiPF6 and (2) LiAsF6
extracted from heat capacity data.
Fig. 3. Heating–cooling cycle around the structural phase
transition of LiAsF6.
thermal hysteresis. No other thermal events were the sample took up some water while it was transferred
detected between 290 and 673 K.
to the container).
In the range 440–570 K, the heating curve shows
two endothermic peaks and a weak exothermic peak at
Thermal Decomposition
of Lithium Hexafluoroarsenate
T
p
= 620.9 K. The temperature of the first endotherm is
93.1 K. The parameters of the main endothermic peak
512–570 K) are T = 535.3 K, T = 541.3 K, and
4
(
As shown earlier, experimental conditions have a
e
p
significant effect on LiAsF decomposition. According
6
∆trH(endo) = 3.648 ± 0.045 kJ/mol. The cooling curve
to pyrolytic mass spectrometry data, the decomposition
process begins at 553 K in vacuum and near 623 K in a
sealed glass tube filled with dry air [15]. Ippolitov et al.
shows one exotherm in the range 530–455 K, with
−
∆ H(exo) = 3.773 ± 0.101 kJ/mol, T = 526.1 K, and
t
r
e
Tp = 521.1 K. These findings provide conclusive evi-
[
16] reported that LiAsF dissociates into LiF and AsF
6
4
dence that the endothermic peak in the range 512–
at 623 K, without melting.
5
70 K is not due to the decomposition process
In our studies of LiAsF decomposition, we used
6
sealed steel cells and special calorimetric containers
ensuring the removal of gaseous decomposition prod-
LiAsF (cr) = LiF(cr)+ AsF (g),
6
5
ucts. The measuring technique was described in detail because, under the conditions of that experiment, at
in a previous report, dealing with LiPF [17]. All of the least part of the AsF released would be removed, and
6
5
thermal effects in DSC curves were reduced to the same the exotherm in the cooling curve would be smaller in
sample weight in order to provide a comparison of pro- magnitude than the endotherm in the heating curve,
cesses occurring in the same temperature range under which was not the case.
different conditions.
LiAsF decomposition with the removal of gas-
6
4
0
0
0
eous decomposition products. As pointed out above,
the heating curves of the samples contained in sealed
tubes in a dry nitrogen atmosphere show no thermal
events other than the phase transition up to 673 K. To
ascertain that this event is indeed a phase transforma-
tion, rather than decomposition, the sample was heated
in a container ensuring the removal of gaseous reaction
products. To this end, the pressure in the system was
slightly reduced, to 200 Pa; that is, the experiment was
performed at an almost normal pressure.
5
21.1
EXO
2
Cooling
Heating
–20
Figure 4 shows the first scan of an LiAsF sample
0.1004 g), which includes heating from 298 to 723 K
541.3
6
ENDO
(
–
40
3
at 5 K/min, isothermal hold for 30 min, and cooling
from 723 K at 5 K/min. The small peak in the heating
curve at 330–390 K may be due to the melting of
LiAsF · 3H O and LiAsF · H O impurities (probably,
00
400
500
600
700
T, K
Fig. 4. First heating–cooling cycle of LiAsF6 with the
removal of gaseous decomposition products.
6
2
6
2
INORGANIC MATERIALS Vol. 39 No. 2 2003

1
78
GAVRICHEV et al.
Table 2. Low-temperature thermodynamic properties of
Next, the same sample, without intermediate weigh-
ing, was heated again from 323 to 793 K at a rate of
LiAsF6
1
0 K/min. The heating curve showed a peak due to the
0
p
S0(T)
0
0
0
C (T)
Φ (T) H (T) – H (0) phase transition (525–560 K) with T = 533.0 K and
e
T, K
Tp = 540.9 K. Decomposition began at about 720 K and
J/(K mol)
J/mol
6.405
did not reach completion at 793 K. At this temperature,
the experiment was ceased because the container mate-
rial would not withstand higher temperatures. Unfortu-
nately, the weight loss due to sample decomposition
could not be determined since the gaseous decomposi-
tion products attacked the container.
1
2
2
3
3
4
4
5
6
7
8
9
5
0
5
0
5
0
5
0
0
0
0
0
1.708
4.669
8.822
13.63
18.32
22.81
27.06
31.06
38.57
45.72
52.86
60.09
67.31
74.43
81.36
88.06
94.51
100.6
106.5
112.0
117.3
122.3
127.1
131.6
136.0
140.0
143.8
147.5
150.9
154.1
157.2
160.2
162.5
163.0
165.9
168.6
171.2
0.569
1.458
2.922
4.955
7.411
10.15
13.09
16.15
22.48
28.97
35.54
42.18
48.89
55.64
62.41
69.19
75.96
82.69
89.37
95.99
102.5
109.0
115.4
121.7
128.0
134.1
140.1
146.1
151.9
157.7
163.3
168.9
173.4
174.4
179.8
185.1
190.3
0.1420
0.3485
0.7069
1.239
1.941
2.793
3.773
4.857
7.263
9.900
12.69
15.60
18.59
21.65
24.77
27.92
31.11
34.33
37.56
40.80
44.05
47.30
50.55
53.79
57.02
60.23
63.44
66.62
69.79
72.94
76.07
79.18
81.69
82.26
85.32
88.35
91.36
22.19
55.39
111.5
191.5
294.4
419.2
564.6
913.2
1335
LiAsF decomposition in a sealed crucible. In the
6
first run, an LiAsF sample (0.030 g) sealed in a cruci-
6
ble was subjected to the following temperature pro-
gram: heating from 298 to 693 K at 10 K/min, isother-
mal hold for 10 min, heating from 693 to 873 K at
5
K/min, and cooling from 873 K at 5 K/min (Fig. 5).
As in the case of the removal of gaseous reaction
products, the heating curve recorded between 298 and
6
93 K at a rate of 10 K/min shows two endothermic
1828
peaks. The first, small endotherm (460–510 K, T =
p
2392
482.5 K) differs in magnitude from the exotherm in the
range 605–640 K (T = 621.2 K) by no more than 5%.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
00
10
20
30
40
50
60
70
80
90
00
10
20
30
40
50
60
70
80
90
3029
p
The parameters of the main peak (512–575 K), related
3738
to the phase transition, are T = 532.3 K, T = 540.6 K,
e
p
4517
and ∆ H(endo) = 3.640 ± 0.018 kJ/mol.
tr
5365
The heating curve recorded between 693 and 873 K
6278
at 5 K/min shows an endothermic peak due to LiAsF6
decomposition. At this heating rate, thermal decompo-
sition occurs in the range 765–820 K, and the endot-
7254
8290
herm comprises two unresolved peaks with T (I) =
p
7
75.3 K and T (II) = 796.8 K, with a shoulder on the
9382
p
high-temperature side (800.6 K) of the latter peak. As
the temperature of decomposition onset, we took
T (I) = 760.5 K. Since the AsF released during LiAsF
10530
11730
12970
14270
15610
16990
18410
19860
21350
22880
24440
26020
27340
27640
29280
30960
32660
e
5
6
decomposition in a sealed crucible was not removed
from the reaction zone, the thermal effect of decompo-
sition was determined at a constant volume and varying
(
rising) pressure. Consequently, the measured heat of
reaction is equal to the internal energy change ∆U (T).
r
Unfortunately, the ∆U (T) data cannot be converted into
r
∆
H (T) since the AsF pressure in the sealed crucible is
r 5
unknown. The experimentally determined value of
∆U (T) is 31.64 ± 0.08 kJ/mol.
r
In the range 805–700 K, the cooling curve shows an
exotherm due to the reverse reaction, LiAsF forma-
6
tion, with ∆U = 31.65 ± 0.09 kJ/mol. The main peak is
r
observed at 796.4 K, with a shoulder at 781.0 K. T =
e
2
98.15
799.9 K can be taken as the onset of the reverse reac-
tion.
3
00
10
20
30
The parameters of the exotherm due to the reverse
3
transformation (530–500 K) are T = 528.7 K, T =
e
p
3
3
5
25.6 K, and –∆ H(exo) = 2.563 ± 0.010 kJ/mol.
tr
Thus, lithium hexafluoroarsenate decomposes at
notably higher temperatures than does lithium hexaflu-
Note: The data in the range 15–50 K were obtained be extrapolat-
ing the experimental C (T) data to absolute zero.
orophosphate [17]. As in the case of LiPF , the reverse
p
6
INORGANIC MATERIALS Vol. 39 No. 2 2003

THERMODYNAMIC PROPERTIES AND DECOMPOSITION
60
179
reaction (LiAsF formation) reaches completion in a
6
7
96.4
sealed crucible. Figure 6 compares the DSC curves of
the processes run in a sealed crucible and with the
removal of gaseous reaction products.
Comparison of the present and earlier results dem-
onstrates that the parameters of the thermal decomposi-
tion of lithium hexafluoroarsenate depend significantly
on both the process conditions and sample quality.
5
4
3
2
0
0
0
0
EXO
781.0
5
25.6
10
0
Cooling
7
69.5
–
10
II
I
Heating
ENDO
Heat Capacity of LiAsF at Elevated Temperatures
–20
6
800.6
5
40.6
7
96.8
The heat capacity Cp,step(T) of a 0.1725-g sample
–30
7
75.3
(
the crucible almost entirely filled with the substance)
–40
3
00
400
500
600
700
800 T, K
was measured from 298 to 748 K in the course of inter-
mittent heating: each cycle included heating at a rate of
Fig. 5. Heating–cooling cycle of LiAsF6 in a sealed
crucible.
1
.5 K/min for 200 s and isothermal hold for 400 s.
Thus, the temperature was raised by 5 K per cycle. Each
series of measurements consisted of 20–40 cycles (data
points), and its temperature range overlapped with
those of the preceding and subsequent series. After
each series of measurements, the sample was cooled in
the calorimeter at 10–20 K/min and stored at room tem-
perature until the subsequent series. Heating to the set
temperature was carried out at 5–10 K/min.
0
–
–
10
20
30
2
Previous work showed that, in the case of thick-
walled steel cells, the measured heat capacity should be
corrected using heat capacity data for a references sub-
stance (e.g., corundum). In connection with this, we
also measured the heat capacity of powdered synthetic
corundum (α-Al O ) crystals using the same tubes and
analogous series of steplike heating. The procedure
used to correct heat capacity data was described previ-
ously [8].
1
541
2
3
–
–
ENDO
40
3
The corrected heat capacity data for LiAsF in the
6
00
400
500
600
700
800 T, K
range 298–748 K (3 series of measurements, 95 data
points) are displayed in Table 3. With a rare exception,
the discrepancy between the measured and smoothed
values within a series does not exceed 0.5% (standard
deviation of 0.15–0.23%). For the convenience of ther-
Fig. 6. Heating curves of lithium hexafluoroarsenate in
(1) a container ensuring the removal of gaseous decomposi-
tion products and (2) a sealed crucible.
modynamic computations, the C values in Table 3 are
p
thermodynamic properties of LiAsF .) Quasi-equilib-
given with one extra significant digit. In the tempera-
ture range 300–330 K, the heat capacity determined by
DSC is systematically lower compared to the adiabatic
calorimetry data (Table 1). For this reason, prior to the
optimization and simultaneous processing of the low-
and high-temperature heat capacity data, the DSC val-
ues were increased by 4.1 J/(K mol).
6
rium conditions near 530–540 K were only reached in
series V (heating for 200 s and holding for 1800 s in
each measurement cycle). The transition temperature
was determined as the average over three measure-
ments (series II, IV, and V) in which the highest anom-
alous heat capacity was obtained: T (α
β) =
tr
5
35.0 ± 0.5 K. In series V, the heat capacity at 535 K
The C (T) curve in Fig. 7 shows a lambda-type
p
was 1720 J/(K mol), whereas that in series II was
anomaly (350–600 K) due to the transition from the α
phase (rhombohedrally distorted NaCl structure) [7] to
the β phase (cubic NaCl structure). The most drastic
7
22 J/(K mol).
In evaluating the thermodynamic properties of lith-
changes in C occur in the range 480–550 K. To more
p
ium hexafluoroarsenate, the temperature range 0–
accurately determine the transition temperature, we 748 K was divided into three portions:
carried out two additional series of measurements (IV
and V) in the range 510–550 K at a heating rate of
.6 K/min, raising the temperature by 2 K per cycle.
The data obtained in series IV and V are not included
in Table 3 because they were not used in assessing the
0
–350 K (α-LiAsF6);
0
(
3
50–600 K (α
β transition;
600–748 K (β-LiAsF6.
INORGANIC MATERIALS Vol. 39 No. 2 2003

1
80
GAVRICHEV et al.
Table 3. Heat capacity, J/(K mol), of LiAsF at elevated temperatures
6
T, K
Cp
T, K
Cp
T, K
Cp
T, K
Cp
Series I
419.8
424.8
429.8
434.8
439.9
444.8
449.8
454.8
459.8
464.8
192.6
192.4
195.5
195.5
197.3
199.0
201.1
203.1
204.3
205.2
529.9
534.9
539.9
545.0
549.9
554.9
559.9
564.9
569.9
574.9
579.9
584.9
589.9
594.9
599.9
604.9
609.8
614.8
619.9
624.8
629.8
634.8
639.8
644.8
259.9
722.1*
233.0*
205.0
201.6
202.4
202.5
203.6
204.4
204.3
205.2
204.1
204.8
204.1
204.2
204.4
204.7
205.5
206.7
205.9
206.9
207.0
208.3
208.3
Series III
2
99.8
04.8
09.8
14.8
19.7
24.7
29.8
34.8
39.8
44.8
49.8
54.8
59.9
64.8
69.8
74.9
79.9
84.9
89.9
94.9
99.9
04.8
09.8
159.0
160.4
161.7
163.0
164.4
165.7
167.1
168.9
169.4
171.3
171.3
173.4
175.3
176.3
177.2
180.3
179.8
182.7
182.9
184.9
185.1
187.1
187.9
635.1
639.9
644.9
649.9
654.9
659.9
664.8
669.8
674.8
679.8
684.8
689.8
694.7
699.7
704.9
709.8
714.8
719.9
724.8
729.8
734.9
739.9
208.9
208.2
208.2
208.5
209.8
209.7
210.0
211.2
211.5
211.8
213.2
213.5
213.6
214.0
213.7
215.4
216.3
217.6
218.3
219.1
218.9
221.1
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
4
4
Series II
460.0
465.0
470.0
475.0
480.0
485.0
490.0
495.0
500.0
505.0
510.0
515.0
519.9
204.2
205.7
205.9
207.3
209.1
209.1
212.2
215.0
216.4
221.1
223.6
226.8
234.7
*
Obtained under nonequilibrium experimental conditions.
The low- and high-temperature heat capacity data derived from the adiabatic calorimetry data, to within
were smoothed jointly using programs described previ- Ӎ0.5%.
ously [18]. For the range 0–350 K, smoothed heat
The heat capacity data for β-LiAsF in the range
600–715 K were smoothed using the DE equation:
6
capacity data for α-LiAsF were obtained by the DEK
6
equation,
C (T) = 8.304D(287 K) + 7.250D(1364 K)
p
C (T) = 7.40D(211 K) + 7.25D(810 K)
p
+
8.000D(1364 K) + 2.250E(1215 K).
+
8.00D(1884 K) + 1.00E(503 K)
1.25K(490, 567 K),
The thermodynamic properties of LiAsF in the
6
range 300–715 K evaluated by the above equations are
+
summarized in Table 4. As follows from the heat capac-
ity data for β-LiAsF , this compound begins to decom-
where D, E, and K are the Debye, Einstein, and Kiffer
functions, respectively. The thermodynamic properties
of α-LiAsF under standard conditions determined
6
pose above 715 K.
The enthalpy and entropy of transformation were
6
using this equation and the jointly processed DSC and evaluated by graphically integrating the Cp,an(T) and
adiabatic calorimetry data coincide with those above, Cp,an(T)/T plots in the range 350–600 K. The anomalous
INORGANIC MATERIALS Vol. 39 No. 2 2003

THERMODYNAMIC PROPERTIES AND DECOMPOSITION
181
C , J/(K mol)
p
2
2
2
1
1
80
40
00
60
5
35 K
1
2
3
20
2
00
300
400
500
600
700 T, K
Fig. 7. Heat capacity of LiAsF at elevated temperatures: (1) adiabatic calorimetry data, (2) DSC data, (3) normal component in the
6
region of the transformation.
heat capacity was found as Cp,an(T) = Cp,meas(T) –
The fact that the enthalpy of transformation is close
β structural phase tran-
C (T). To determine the normal component of heat to Rln3 suggests that the α
p,n
capacity, the low- and high-temperature portions of the sition of lithium hexafluoroarsenate involves order–dis-
C (T) curve lying beyond the transformation region order processes which increase the number of possible
p
6
–
were extrapolated to T = 535 K. In this way, we equilibrium orientations of the AsF ion in going from
tr
obtained ∆ H(α
β, step) = 5.29 ± 0.27 kJ/mol and the low-temperature, ordered phase to the high-temper-
tr
∆trS(α
β, step) = 10.30 ± 0.53 J/(K mol) ≈ Rln3. ature, disordered phase by a factor of three.
The indicated uncertainties arise from the scatter in
experimental data.
Table 4. Thermodynamic properties of LiAsF from 300 to
6
7
15 K
Since, in evaluating ∆ H(α
β, step) and
β, step), we used the data points at 534.9 and
39.9 K (series II), which were obtained under non-
tr
∆trS(α
5
0
p
S0(T)
0
0
0
C (T)
Φ (T) H (T) – H (0)
T, K
equilibrium experimental conditions, we also measured
heat capacity Cp,scan(T) during continuous heating from
J/(K mol)
J/mol
4
53 to 573 K at a rate of 1 K/min. The enthalpy of trans-
3
3
00
50
163.2
175.8
186.0
194.4
201.2
205.3
198.7
201.1
207.8
213.4
218.1
219.3
173.1
199.2
223.4
245.8
266.6
280.3
290.7
296.2
314.0
341.2
357.2
361.8
81.07
96.11
110.5
124.3
137.5
146.4
146.4
150.9
163.7
178.1
190.3
193.9
27600
36080
45140
54650
64550
71660
76950
79950
90180
106000
116800
120100
formation, ∆ H'(α β, scan) = 3.98 ± 0.06 kJ/mol,
was evaluated by graphically integrating (SETARAM
program) the endothermic peak between 485 and
tr
400
5
47 K. For comparison, ∆ H'(α
β, scan) was cor-
tr
4
5
50
00
rected for the anomalous enthalpy in the ranges 350–
4
∆trH''(α
tal error (5.29 ± 0.27 and 5.03 ± 0.24 kJ/mol, respec-
tively), indicating that the α β phase transition is
accompanied by no isothermal contribution and is,
therefore, second-order. Note that the enthalpy of trans-
formation evaluated by integrating the peak in heating
curves is substantially lower. The reason is that the
anomalous enthalpy increment in the early stage of the
very broad transformation is included in the baseline of
the DSC curve (see, e.g., [19]).
85 and 547–600 K. ∆ H(α
β, step) and
tr
β, scan) coincide to within the experimen-
535(α)
5
5
6
6
35(β)
50
00
50
700
15
7
INORGANIC MATERIALS Vol. 39 No. 2 2003

1
82
GAVRICHEV et al.
CONCLUSION
Heat capacity measurements were used to evaluate
Lithium Hexafluorophosphate, Zh. Neorg. Khim., 2002,
vol. 47, no. 7, pp. 1048–1051.
9
. Shebershneva, O.V., Izotov, A.D., Gavrichev, K.S., and
Lazarev, V.B., A Method for Treating Low-Temperature
Calorimetry Data with Regard to the Multifractality of
Atomic Vibrations, Neorg. Mater., 1996, vol. 32, no. 1,
pp. 36–40 [Inorg. Mater. (Engl. Transl.), vol. 32, no. 1,
pp. 28–32].
the thermodynamic properties of lithium hexafluoro-
arsenate from liquid helium temperatures to the decom-
position onset and to assess the characteristics of the
structural phase transition in this compound. By study-
ing the high-temperature thermal properties of LiAsF6,
we determined the parameters of its decomposition,
which were found to depend significantly on both the
experimental conditions and sample quality.
1
1
0. Gorbunov, V.E., Gavrichev, K.S., and Bakum, S.I., Ther-
modynamic Properties of LiAlH from 12 to 320 K, Zh.
4
Neorg. Khim., 1981, vol. 26, no. 2, pp. 311–314.
1. Gorbunov, V.E., Gavrichev, K.S., Zalukaev, V.L., et al.,
Low-Temperature Heat Capacity of Hexagonal (α) and
Tetragonal (β) Lithium Iodate Polymorphs, Zh. Fiz.
Khim., 1982, vol. 56, no. 11, pp. 1121–1124.
ACKNOWLEDGMENTS
This work was supported by the Russian Foundation
for Basic Research, grants no. 03-99-32727 and
no. 00-15-97394.
1
1
1
1
1
2. Zalukaev, V.L., Gorbunov, V.E., Sharpataya, G.A., et al.,
Heat Capacity and Thermodynamic Properties of Lith-
ium Perchlorate, Zh. Neorg. Khim., 1981, vol. 26, no. 4,
pp. 899–901.
REFERENCES
3. Gorbunov, V.E., Gavrichev, K.S., Zalukaev, V.L., et al.,
Heat Capacity and Phase Transition of Lithium Borohy-
dride, Zh. Neorg. Khim., 1984, vol. 29, no. 9,
pp. 2333−2337.
1
2
3
. Gavrichev, K.S., Gorbunov, V.E., Golushina, L.N., et al.,
Low-Temperature Heat Capacity of RbAsF , Neorg.
6
Mater., 1996, vol. 32, no. 7, pp. 878–880 [Inorg. Mater.
(Engl. Transl.), vol. 32, no. 7, pp. 778–780].
4. Gorbunov, V.E., Gavrichev, K.S., Totrova, G.A., et al.,
Heat Capacity and Thermodynamic Properties of Lith-
ium Borofluoride from 10 to 340 K, Zh. Neorg. Khim.,
1993, vol. 38, no. 2, pp. 217–219.
. Gavrichev, K.S., Gorbunov, V.E., Golushina, L.N., et al.,
Low-Temperature Heat Capacity of Cesium Hexafluoro-
arsenate, Zh. Neorg. Khim., 1997, vol. 42, no. 1,
pp. 96−98.
. Gavrichev, K.S., Gorbunov, V.E., and Sharpataya, G.A.,
Calorimetric Study of Cesium and Rubidium Hexafluo-
roarsenates, 14th IUPAC Conf. on Chemical Thermody-
namics, Osaka, 1996, p. 217.
. Gorbunov, V.E., Gurevich, V.M., and Gavrichev, K.S.,
Adiabatic Microcalorimeter with an Aneroid-Type Cry-
ostat, Zh. Fiz. Khim., 1982, vol. 56, no. 1, pp. 235–237.
. Iorish, V.S., Belov, G.V., and Yungman, V.S., Windows
Compatible IVTANTERMO Package for Applied Ther-
modynamic Analysis, Preprint of Joint Inst. of High
Temperatures, Russ. Acad. Sci., Moscow, 1998,
no. 8-415.
5. Lawless, E.W., Wiegand, C.J.W., Mizumoto, Yu., and
Weis, C., Lithium Hexafluoroarsenate and Hexafluoro-
arsenic Acid, Inorg. Chem., 1971, vol. 10, no. 5,
pp. 1084–1086.
6. Ippolitov, E.G., Maifat, M.A., and Zhigarnovskii, B.M.,
Properties of Alkali Hexafluorophosphates and Hexaflu-
oroarsenates, V Vsesoyuznyi simpozium po khimii neor-
ganicheskikh ftoridov (VAll-Union Symp. on the Chem-
istry of Inorganic Fluorides), Dnepropetrovsk, 1978,
p. 132.
4
5
1
1
7. Gavritchev, K.S., Sharpataya, G.A., Smagin, A.A., et al.,
Calorimetric Study of Thermal Decomposition of Lith-
ium Hexafluorophosphate, J. Therm. Anal. Cal. (in
press).
6
7
. Lazarev, V.B., Izotov, A.D., Gavrichev, K.S., and She-
bershneva, O.V., Fractal Model of Heat Capacity for
Substances with Diamond-like Structures, Thermochim.
Acta, 1995, vol. 269/270, pp. 109–116.
8. Gurevich, V.M., Gorbunov, V.E., Gavrichev, K.S., and
Khodakovskii, I.L., Calorimeter for Heat Capacity Mea-
surements from 50 to 300 K and the Low-Temperature
Heat Capacity of Kogarkoite, Na SO F, Geokhimiya,
. Kemmitt, R.D.W., Russell, D.R., and Sharp, D.W.A.,
The Structural Chemistry of Complex Fluorides of Gen-
3
4
1
999, no. 4, pp. 423–434.
I
V
eral Formula A B F , J. Chem. Soc., 1963, vol. 9,
6
1
9. Parsonage, N. and Staveley, L., Disorder in Crystals,
Oxford: Oxford Univ. Press, 1978. Translated under the
title Besporyadok v kristallakh, Moscow: Mir, 1982,
part 1.
pp. 4408–4413.
. Gavrichev, K.S., Sharpataya, G.A., Golushina, L.N.,
et al., Heat Capacity and Thermodynamic Properties of
8
INORGANIC MATERIALS Vol. 39 No. 2 2003