Synergistic use of Knudsen effusion quadrupole mass spectrometry, solid-state galvanic cell and differential scanning calorimetry for thermodynamic studies on lithium aluminates

Article abstract of DOI:10.1016/j.jssc.2008.03.003

Three ternary oxides LiAl₅O₈(s), LiAlO₂(s) and Li₅AlO₄(s) in the system Li-Al-O were prepared by solid-state reaction route and characterized by X-ray powder diffraction method. Equilibrium partial pr

Full text of DOI:10.1016/j.jssc.2008.03.003

ARTICLE IN PRESS
Journal of Solid State Chemistry 181 (2008) 1402– 1412
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
Synergistic use of Knudsen effusion quadrupole mass spectrometry,
solid-state galvanic cell and differential scanning calorimetry for
thermodynamic studies on lithium aluminates
Ã
S.K. Rakshit , Y.P. Naik, S.C. Parida, Smruti Dash, Ziley Singh, B.K. Sen, V. Venugopal
Product Development Section, Radiochemistry and Isotope Group, Bhabha Atomic Research Centre, Trombay, Mumbai 400085, India
a r t i c l e i n f o
a b s t r a c t
Article history:
Three ternary oxides LiAl5O8(s), LiAlO2(s) and Li5AlO4(s) in the system Li–Al–O were prepared by solid-
state reaction route and characterized by X-ray powder diffraction method. Equilibrium partial pressure
of CO2(g) over the three-phase mixtures {LiAl5O8(s)+Li2CO3(s)+5Al2O3(s)}, {LiAl5O8(s)+5LiAlO2(s)+2Li2
CO3(s)} and {LiAlO2(s)+Li5AlO4(s)+2Li2CO3(s)} were measured using Knudsen effusion quadrupole mass
spectrometry (KEQMS). Solid-state galvanic cell technique based on calcium fluoride electrolyte was
used to determine the standard molar Gibbs energies of formations of these aluminates. The standard
molar Gibbs energies of formation of these three aluminates calculated from KEQMS and galvanic cell
measurements were in good agreement. Heat capacities of individual ternary oxides were measured
from 127 to 868 K using differential scanning calorimetry. Thermodynamic tables representing the
values of DfH0(298.15 K), S0(298.15 K) S0(T), Cp0(T), H0(T), {H0(T)–H0(298.15 K)}, G0(T), DfH0(T), DfG0(T) and
free energy function (fef) were constructed using second law analysis and FACTSAGE thermo-chemical
database software.
Received 31 December 2007
Received in revised form
5 March 2008
Accepted 9 March 2008
Available online 16 March 2008
Keywords:
Breeder material
Lithium aluminate
Knudsen effusion method
Solid-state galvanic cell
Differential scanning calorimetry (DSC)
Heat capacity
& 2008 Elsevier Inc. All rights reserved.
Thermodynamic properties
1. Introduction
have studied the vaporization and thermo-chemical stability of
lithium aluminates using high-temperature Knudsen effusion
mass spectrometry. Many researchers have reported the enthalpy
and heat capacity values of LiAlO2 from 298 to 1700 K [8].
Kleykamp [8] has reported the heat capacity of LiAlO2(s). In this
study, thermodynamic properties of LiAl5O8(s), LiAlO2(s) and
Li5AlO4(s) were determined using Knudsen effusion quadrupole
mass spectrometry (KEQMS), solid-state galvanic cell based on
CaF2 solid electrolyte and differential scanning calorimetry (DSC).
Lithium–aluminum–oxygen system has four ternary oxides:
LiAl5O8(s), LiAlO2(s), Li3AlO3(s) and Li5AlO4(s); however, Li3
AlO3(s) is reported to be unstable above 670 K [1,2]. Among these
oxides, LiAlO2(s) has gained importance for its potential use as
breeder material in the irradiation blanket for future nuclear
fusion reactors due to its chemical and thermal stability as well as
less radiation damage [3]. This oxide exists in three possible
allotropes, hexagonal a-LiAlO2 up to 673 K, monoclinic b-LiAlO2
from 673 to 1073 K and tetragonal g-LiAlO2, the most stable form
at temperature greater than 1073 K. This aluminate is also used as
an inert and non-conductive ceramic matrix to contain molten
carbonate electrolyte between the anode and the cathode of
molten carbonate fuel cells due to its high mechanical and
thermal stability [3–5]. This aluminate is also expected to be less
reactive with cladding materials presently used in fusion reactors
due to lower vapor pressures and higher melting points than solid
Li2O. Many researchers have carried out the preparation, decom-
position and thermal studies of LiAlO2(s) [6,7] but studies on
other aluminates in Li–Al–O system are very scarce. Ikeda et al. [1]
2. Experimental
2.1. Materials preparation
Ternary oxides, LiAl5O8(s), LiAlO2(s) and Li5AlO4(s), were
prepared using conventional solid-state reaction route using
pre-heated powder samples of Li2CO3(s) and Al2O3(s) (LEICO
Industries Inc., mass fraction 0.9999). The individual powders
were weighed according to their stoichiometric ratios and mixed
homogeneously using an agate mortar and pestle and the
resultant powder samples were pelletized using a steel die at a
pressure of 100 MPa. The pellets were initially heated at 900 K for
50 h in air in a re-crystallized alumina crucible and then cooled,
reground and again pelletized. These pellets were then heated at
Ã
Corresponding author. Fax: +9122 2550 5151.
E-mail addresses: swarupkr@barc.gov.in, swarup_kr@rediffmail.com
(S.K. Rakshit).
0022-4596/$ - see front matter & 2008 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2008.03.003

ARTICLE IN PRESS
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
1403
1150 K for 48 h. The resultant samples were characterized by X-ray
powder diffraction (XRD) technique using DIANO XRD with Cu-Ka
radiation and graphite monochromator and found to be pure
crystalline phases of LiAl5O8(s), g-LiAlO2(s) and b-Li5AlO4(s).
Three-phase mixtures {LiAl5O8(s)+Li2CO3(s)+5Al2O3(s)}, {LiAl5
O8(s)+2Li2CO3(s)+5LiAlO2(s)} and {LiAlO2(s)+2Li2CO3(s)+Li5AlO4(s)}
for KEQMS were prepared by homogeneously mixing the
individual pre-heated compounds in stoichiometric ratios and
pelletized using a steel die and then sintered at 700 K to remove
moisture. The sintered pellets were broken into small pieces and
loaded inside the Knudsen cell.
According to the phase relations in the Li2O–Al2O3 pseudo-
binary system, phase mixtures: {2LiAl5O4(s)+5Al2O3(s)+2LiF(s)},
{5LiAlO2(s)+LiAl5O4(s)+4LiF(s)} and {Li5AlO4(s)+LiAlO2(s)+4LiF(s)}
were also prepared by homogenously mixing the pre-heated
individual powders and pelletized using a steel die at a pressure of
100 MPa and sintered at 800 K for 10 h under moisture and
hydrogen-free oxygen atmosphere. These sample pellets were
then used for solid-state galvanic cell experiments. Powder
samples of LiAl5O8(s), LiAlO2(s) and Li5AlO4(s) were used for DSC.
come into the path of the molecular flow. The shutter isolates the
Knudsen effusion cell while recording the background signal.
The detected signal (Ii+) measured using Faraday cup detector
is related to the partial pressure of the vapor species (pi) by
pi ¼ KinstIþi T=ðsiaiÞ
(1)
where Kinst is the instrumental constant, Ii+ is the measured ion
current in ampere, T is the absolute temperature near the Knudsen
cell, si is the electron impact cross-section and ai isotopic
abundance of the specific ion. Eq. (1) can be represented as
ln pi ¼ ln Kinst þ lnðIþi TÞ ꢀ ln si ꢀ ln ai
(2)
For permanent gaseous species such as CO2 at mass, m ¼ 44,
ln s ¼ ꢀ45.52 at 30 eV [14] and the isotopic abundance as 100%,
Eq. (2) can be expressed as
ln Pi ¼ ln Kinst þ lnðIþi TÞ þ 45:52 ðfor i ¼ CO2Þ
(3)
2.2.1. Calibration of KEQMS
Prior to calibration of the instrument, the background signals
were monitored by heating the Knudsen chamber with empty
Knudsen cell at different temperatures from ambient to 1161 K at
pressure level ꢁ1 ꢂ10ꢀ5 Pa. The background signals as a function
of temperature are shown in Fig. 1. It is evident from the figure
that the background signals corresponding to H+2, N2+, CO+ and CO2+
do not change appreciably with change in temperature. During
experiments, the actual signals were obtained by subtracting the
ion intensities due to background.
The instrument calibration constant (Kinst) was determined by
measuring the ion intensities of CO+2 over the phase mixtures of
{CaCO3(s)+CaO(s)}, {SrCO3(s)+SrO(s)}, {BaCO3(s)+BaO(s)} and {Li2
CO3(s)+Li2O(s)}. Three different ionization energies (30, 50 and
70 eV) were used to measure the ion intensities of CO+2 to check
the linearity of pressure measurements. However, 30 eV is
sufficient to ionize all types of gaseous molecules; hence, the
actual experiments were carried out at ionization energy of 30 eV.
Prior to actual measurement, a particular phase mixture (mass
ꢁ1 g) was loaded inside the Knudsen cell and then heated at 700 K
for 4 h under high vacuum to remove the moisture and other
unwanted gaseous species.
2.2. Knudsen effusion quadrupole mass spectrometry (KEQMS)
The Knudsen effusion mass spectrometric technique is one of
the most informative methods for vaporization processes and
thermodynamic properties of high-temperature systems. Gener-
ally, for thermodynamic measurements, traditional magnetic
sector mass spectrometer attached to Knudsen effusion system
is preferable among quadrupole and time-of-flight mass spectro-
meters. Murray et al. [9] have shown that thermodynamic data
obtained by Knudsen effusion technique using magnetic sector
spectrometer and quadrupole mass spectrometer are in good
agreement for pure chromium and chromium–silicon samples.
However, quadrupole mass spectrometer has no significant
advantages over magnetic mass spectrometer but they are very
compact and relatively inexpensive. Stolyarova et al. [10] have
reported that quadrupole mass spectrometer coupled to Knudsen
cell can effectively be used for thermodynamic studies at high
temperature.
In this study, a residual gas analyzer (RGA) coupled to Knudsen
effusion system was used for equilibrium partial pressure
measurements. An RGA is a quadrupole mass spectrometer in
which the ionizer is immersed in the gas to be analyzed, and the
ionizer is characterized by an open construction in which the gas
may enter and leave in all directions. It is assumed that the gas is
homogenous and that changes in the gas density with time occur
slowly enough such that the instrument is always in equilibrium
with the gas. This instrument can be used to identify the kind of
molecules present in the gaseous phase and, when calibrated, can
be used to determine concentrations or partial pressures [11–13]
of individual species.
The KEQMS used in this study is an in-house designed Knudsen
vacuum chamber and arranged in such a way that it allows
reciprocally perpendicular molecular beam from the Knudsen
effusion cell. The Knudsen vacuum chamber was heated to the
desired temperature using resistance heater. The temperature
near the Knudsen cell was measured using a pre-calibrated (ITS-
90) chromel–alumel thermocouple. The Knudsen cell used was
made of 15 mol% calcia stabilized zirconia (CSZ) with a thin
cylindrical orifice of dia 0.8 mm and height 0.2 mm at the centre of
the lid. This setup is used only for partial pressures measurements
of permanent gaseous species such as CO, O2, N2, CO2, etc. and not
for condensable vapor species. A shutter is placed between the
ionizer and the Knudsen effusion chamber such that it does not
2.2.2. Partial pressure measurements of CO2(g) over equilibrium
phase mixtures
Huang et al. [15] have reported the thermodynamic data of
Na4Fe6O11(s) by measuring the partial pressure of CO2(g) over
{2Na2CO3(s)+3Fe2O3(s)} phase mixture using Knudsen effusion
mass spectrometry from 918 to 1013 K. Similar approach was
adopted in this study to determine the Gibbs energies of
formation of LiAl5O8(s), LiAlO2(s) and Li5AlO4(s) by measuring
the partial pressure of CO2(g) over the equilibrium phase mixtures
{LiAl5O8(s)+Li2CO3(s)+5Al2O3(s)}, {LiAl5O8(s)+2Li2CO3(s)+5LiAlO2(s)}
and {LiAlO2(s)+2Li2CO3(s)+Li5AlO4(s)}.
The ion intensities of CO2+ over these equilibrium phase
mixtures were measured using KEQMS. For each measurement,
two sets of experiments were carried out and the ion intensities
for other gaseous species were in background level during the
measurement. Subsequently, partial pressures of carbon dioxide,
p(CO2) over the phase mixture were obtained using Eq. (3). After
the mass spectrometric measurements, the resultant phase
mixtures were analyzed by XRD technique and found to be the
mixture of corresponding lithium aluminate, lithium carbonate
and alumina. Therefore, it was assumed that the following
equilibrium reactions were established inside the Knudsen cell
under experimental conditions:
Li2CO3ðsÞ þ 5Al2O3ðsÞ ¼ 2LiAl5O8ðsÞ þ CO2ðgÞ
(4)

ARTICLE IN PRESS
1404
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
1.00x10-12
7.50x10-13
5.00x10-13
T = 1161 K
2.50x10-13
0.00
1.00x10-12
7.50x10-13
5.00x10-13
T = 675 K
2.50x10-13
0.00
1.00x10-12
7.50x10-13
5.00x10-13
2.50x10-13
0.00
H2+
H2O+
T = 299 K
N2+/CO+
CO2+
O2+
0
5
10
15
20
25
30
35
40
45
50
m / e
Fig. 1. Background spectrum of the mass spectrometer with blank Knudsen cell as a function of temperature.
LiAl5O8ðsÞ þ 2Li2CO3ðsÞ ¼ 5LiAlO2ðsÞ þ 2CO2ðgÞ
LiAlO2ðsÞ þ 2Li2CO3ðsÞ ¼ Li5AlO4ðsÞ þ 2CO2ðgÞ
(5)
(6)
1
2
Therefore, the measured p(CO2) corresponds to the equilibrium
partial pressures of the above reactions.
3
2.3. Solid-state galvanic cell technique with CaF2 electrolyte
4
6
The experimental setup and the cell assembly used in this
study have been explained in details by Rakshit et al. [16]. A
schematic diagram of the fluoride cell used in this experiment is
shown in Fig. 2. Optical grade single crystal of CaF2(s) pellet of
6 mm diameter and 3 mm thickness (supplied by Solon Technol-
ogies, Inc., USA) was used as fluoride ion conducting electrolyte. It
is a single compartment cell with provisions for passing purified
oxygen gas during the experiment and to measure the tempera-
ture of the cell near the electrode/electrolyte interface. High-
purity oxygen gas at one atmospheric pressure was allowed to
pass through successive traps of silica gels, molecular sieves,
oxidized form of BTS catalyst and anhydrous magnesium
perchlorate for removal of traces of H2(g) and moisture. The
reference electrode, the electrolyte and the sample electrode
stacked one over the other was kept in the isothermal tempera-
5
7
8
9
13
14
10
11
ture zone of
a Kanthal wire wound furnace. The furnace
15
temperature was controlled within 71 K using a PID temperature
controller. The cell was standardized using phase mixtures of
{CaO(s)+CaF2(s)} and {MgO(s)+MgF2(s)} as two standard electro-
des. The cell can be represented as
12
Cell I : ðꢀÞPt; O2ðg; 101:3 kPaÞ=fCaOðsÞ þ CaF2ðsÞg==CaF2ðsÞ==fMgOðsÞ
þ MgF2ðsÞg=O2ðg; 101:3 kPaÞ; PtðþÞ:
Fig. 2. Schematic diagram of the fluoride cell. (1) Pt lead wires, (2) alumina
pressing tube, (3) thermocouple, (4) stainless-steel flange, (5) gas inlet, (6) gas
outlet, (7) spring, (8) quartz tube; (9) quartz tube, (10) alumina cup, (11) Pt discs,
(12) kanthal wire wound furnace, (13) reference electrode, (14) CaF2(s) electrolyte
and (15) sample electrode.
After standardization, the reversible emf’s of the following
solid-state galvanic cells were measured as
temperature.
a function of
Cell IV : ðꢀÞPt; O2ðg; 101:3 kPaÞ=fLi5AlO4ðsÞ þ LiAlO2ðsÞ
þ LiFðsÞg==CaF2ðsÞ==fCaOðsÞ þ CaF2ðsÞg=O2ðg; 101:3 kPaÞ; PtðþÞ:
Cell II : ðꢀÞPt; O2ðg; 101:3 kPaÞ=fCaOðsÞ þ CaF2ðsÞg==CaF2ðsÞ
==fLiAl5O8ðsÞ þ Al2O3ðsÞ þ LiFðsÞg=O2ðg; 101:3 kPaÞ; PtðþÞ;
The cell temperature close to the electrodes was measured
using a pre-calibrated (ITS-90) chromel–alumel thermocouple.
The cell emf (70.02 mV) was measured by using a Keithley 614
Cell III : ðꢀÞPt; O2ðg; 101:3 kPaÞ=fCaOðsÞ þ CaF2ðsÞg==CaF2ðsÞ==fLiAlO2ðsÞ
þ LiAl5O8ðsÞ þ LiFðsÞg=O2ðg; 101:3 kPaÞ; PtðþÞ;

ARTICLE IN PRESS
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
1405
electrometer (input impedance 41014 O). At low temperatures,
stable values of emf were obtained approximately after 72 h
whereas at successive higher temperatures, stability in emf values
was observed within 5–6 h. The reversibility of the solid-state
electrochemical cells was evaluated by micro-coulometric titra-
tion in both directions. The electrode pellets after the emf
measurements were re-examined by XRD analysis and the phase
compositions were found unchanged.
pressure. The sensitivity of the experimental system was checked
by plotting the detected signals (ampere) as a function of
literature values of partial pressure of CO2(g) over the respective
carbonate-oxide phase fields and shown in Fig. 3. This figure
shows that the detected ion currents varies linearly as a function
of partial pressure of CO2(g) throughout the measurement range.
3.1.2. Equilibrium partial pressures over the ternary phase mixtures
3.1.2.1. The phase mixture {LiAl5O8(s)+Li2CO3(s)+5Al2O3(s)}. The ion
intensities of CO2+ peak over {LiAl5O8(s)+Li2CO3(s)+5Al2O3(s)}
phase mixture was measured at 30 eV ionization energy in the
temperature range 614–750 K. Partial pressures of CO2(g), p(CO2),
at different temperatures for two different runs were calculated
using the measured ion intensities, Eq. (3) and the calibration
constant from Eq. (7) and the values are listed in Table 2. The
variation of logarithmic values of p(CO2) as a function of reciprocal
of temperature follows linear relationship as shown in Fig. 4 and
can be expressed as
2.4. Measurement of heat capacity using differential scanning
calorimetry
Molar heat capacity measurements were carried out using a
heat flux-type DSC (131, Setaram Instrumentation, France). The
temperature and energy calibrations and the methods of heat
capacity measurements by continuous heating mode were
described in details by Rakshit et al. [16]. In order to check the
accuracy of the measurement, heat capacity of Fe2O3 (mass
fraction 0.9999, Alfa Aesar, USA) was measured in the temperature
range from (i) 130 to 320 and (ii) 310–860 K. The values of heat
capacity of Fe2O3(s) were found to be within 72% compared to the
literature values [17]. The pre-heated powder samples of
LiAl5O8(s), LiAlO2(s) and Li5AlO4(s) were used for heat capacity
measurements.
lnfpðCO2 ðPaÞÞ ¼ ꢀ18412:1ðꢃ260:2Þ=ðT ðKÞÞ
þ24:44ðꢃ0:38Þ
(8)
The enthalpy change associated with reaction (4) at the average
temperature of the measurement was estimated to be
D
r(4)H0m(682 K) ¼ (153.172.2) kJ molꢀ1. The standard molar Gibbs
energy of formation (DfGm0 ) of LiAl5O8(s) from the elements was
calculated from Eqs. (4) and (8) and the values of DfG0m(T) for
Li2CO3(s), Al2O3(s) and CO2(g) from the literature [17] and can be
represented as
3. Results and discussion
3.1. KEQMS technique
Df G0 ðLiAl5O8; s; TÞ kJ molꢀ1 ¼ ꢀ 4522:2 þ 0:8725 ðT ðKÞÞ
m
ðꢃ2:4Þ ð614pT ðKÞp750Þ
(9)
3.1.1. Calibration of KEQMS (Kinst
)
The ion intensities of CO+2 over the phase mixtures {CaCO3(s)+
CaO(s)}, {SrCO3(s)+SrO(s)}, {BaCO3(s)+BaO(s)} and {Li2CO3(s)+
Li2O(s)} were recorded in ampere using Faraday cup detector as
a function of temperature for two different runs. The values of
ln(Kinst) were calculated as a function of temperature using Eq. (3),
literature values of ln{p(CO2)/Pa} [17] and experimentally mea-
sured ln(Ii+T) values and represented in Table 1. The calibration
constant calculated at 30 eV for {Li2CO3(s)+Li2O(s)} phase mixture
was used for further calculation and is expressed as
3.1.2.2. The phase mixture {LiAl5O8(s)+2Li2CO3(s)+5LiAlO2(s)}. The
ion intensities of CO+2 peak over the phase mixture {LiAl5O8(s)+
2Li2CO3(s)+5LiAlO2(s)} was measured at 30 eV ionization energy
in the temperature range 604–723 K. Partial pressures of CO2(g),
p(CO2), at different temperatures for two different runs were
calculated using the measured ion intensities, Eq. (5) and the cali-
bration constant from Eq. (7) and the values are listed in Table 2.
The variation of logarithmic values of p(CO2) as a function of re-
ciprocal of temperature follows linear relationship as shown in
Fig. 4 and can be expressed as
lnðKinstÞ ¼ 3711:9=ðT ðKÞÞ ꢀ 47:05 ð614pT ðKÞp750Þ
(7)
An ideal instrument has a sensitivity that is constant and
independent of pressure. The ion current reaching the detector of
such an instrument should, therefore, increase linearly with
lnfpðCO2 ðPaÞÞ ¼ ꢀ 18064:8ðꢃ365:5Þ=ðT ðKÞÞ
þ 25:27ðꢃ0:55Þ
(10)
Table 1
Ion intensities of CO+2 at different ion energies over different phase mixtures
Phase
Ionization energy (eV)
Combined ln(Ii+T) of two runs
ln(Pi/atm) literature [15]
ꢀ21093.3/T+18.53
Temperature range (K)
CaCO3(s)+CaO(s)
30
50
70
ꢀ25191.1/T+18.44
ꢀ26090.3/T+21.10
ꢀ25452.4/T+20.28
604–772
617–764
622–781
SrCO3(s)+SrO(s)
BaCO3(s)+BaO(s)
Li2CO3(s)+Li2O(s)
30
50
70
ꢀ28649.7/T+16.16
ꢀ30706.4/T+19.48
ꢀ29566.4/T+19.37
ꢀ28792.7/T+19.85
ꢀ31664.9/T+19.57
ꢀ25882.3/T+17.64
832–951
833–978
835–939
30
50
70
ꢀ33203.1/T+17.64
ꢀ32296.9/T+18.28
ꢀ32531.8/T+19.17
821–1055
822–1056
838–1043
30
50
70
ꢀ29594.2/T+19.17
ꢀ26523.2/T+16.78
ꢀ26677.2/T+17.40
688–887
676–885
686–889

ARTICLE IN PRESS
1406
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
10-9
10-10
10-11
10-12
10-9
Li2CO3(s) = Li2O(s) + CO2(g)
CaCO3(s) = CaO(s) + CO2(g)
10-10
10-11
10-12
10-13
10-14
T = 690 - 870 K
T = 616 - 747 K
10-13
10-9
10-7
10-6
10-5
10-4
10-9
10-8
10-7
10-6
10-5
10-9
10-10
10-11
10-12
10-13
SrCO3 (s) = SrO(s) + CO2 (g)
BaCO3(s) = BaO(s) + CO2(g)
10-10
10-11
10-12
10-13
T = 736 - 939 K
T = 821 - 1029 K
10-9
10-8
10-7
10-6
10-5
10-4
p(CO
10-9
10-8
10-7
10-6
10-5
10-4
2) / bar
Fig. 3. Ion intensities of CO+2 peak over various carbonate—oxide phase mixtures as a function of literature values of p(CO2).
The enthalpy change associated with reaction (5) at the average
temperature of the measurement was estimated to be
D
energy of formation (DfGm0 ) of Li5AlO4(s) from the elements was
calculated from Eqs. (6) and (12) and the values of DfG0m(T) for
Li2CO3(s), Al2O3(s) and CO2(g) from the literature [17] and DfGm0
(LiAlO2, s, T) from Eq. (11) and can be represented as
r(5)H0m(664 K) ¼ (150.273.0) kJ molꢀ1. The standard molar Gibbs
energy of formation (DfG0m) of LiAlO2(s) from the elements was
calculated from Eqs. (5) and (10) and the values of DfG0m(T) for
Li2CO3(s), Al2O3(s) and CO2(g) from the literature [17] and
DfG0m(LiAl5O8, s, T) from Eq. (9) and can be represented as
Df G0 ðLi5AlO4; s; TÞ kJ molꢀ1 ¼ ꢀ 2341:3 þ 0:4201 ðT ðKÞÞ
m
ðꢃ5:5Þ ð672pT ðKÞp806Þ
(13)
Df G0 ðLiAlO2; s; TÞ kJ molꢀ1 ¼ ꢀ 1171:9 þ 0:2423 ðT ðKÞÞ
m
3.2. Emf studies on cells (I)– (IV) using solid-state galvanic cell
technique
ðꢃ4:0Þ ð604pT ðKÞp723Þ
(11)
3.2.1. Standardization of solid-state galvanic cell (I)
3.1.2.3. The phase mixture {LiAlO2(s)+2Li2CO3(s)+Li5AlO4(s)}. The
ion intensities of CO2+ peak over {LiAlO2(s)+2Li2CO3(s)+Li5AlO4(s)}
phase mixture was measured at 30 eV ionization energy in the
temperature range 614–750 K. Partial pressures of CO2(g), p(CO2),
at different temperatures for two different runs were calculated
using the measured ion intensities, Eq. (6) and the calibration
constant from Eq. (7) and the values are listed in Table 2. The
variation of logarithmic values of p(CO2) as a function of reciprocal
of temperature follows linear relationship as shown in Fig. 4 and
can be expressed as
The reversible emf values obtained at different experimental
temperatures for cell (I) are listed in Table 3 and the variation of
emf with temperature is shown in Fig. 5. The emf data were least-
squares fitted to yield the following linear relation:
E ðVÞðꢃ0:0002Þ ¼ 0:3956 ꢀ 2:1073 ꢂ 10ꢀ5 ðT ðKÞÞ
ð921pT ðKÞp1150Þ
(14)
The half-cell reactions at each electrode can be represented as
MgF2ðsÞ þ 2eꢀ þ 1=2O2ðgÞ ¼ MgOðsÞ þ 2Fꢀ
ðat þ ve electrodeÞ
lnfpðCO2 ðPaÞÞ ¼ ꢀ 28166:5ðꢃ544:7Þ=ðT ðKÞÞ
(15)
(16)
þ 34:97ðꢃ0:74Þ
(12)
and
The enthalpy change associated with reaction (6) at the average
temperature of the measurement was estimated to be
CaOðsÞ þ 2Fꢀ ¼ CaF2ðsÞ þ 2eꢀ þ 1=2 O2ðgÞ
ðat ꢀ ve electrodeÞ:
D
r(6)H0m(739 K) ¼ (234.274.5) kJ molꢀ1. The standard molar Gibbs

ARTICLE IN PRESS
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
1407
The net virtual cell reaction can be represented as
where ‘n’ is the total number of electrons involved in the half-cell
reactions and ‘F’ is the Faraday’s constant (F ¼ 96486.4 C molꢀ1
)
MgF2ðsÞ þ CaOðsÞ ¼ CaF2ðgÞ þ MgOðsÞ:
(17)
and ‘E’ is the net cell emf in volts. The values of Dr(17)G0(T) as a
function of temperature can be calculated using Eqs. (14) and (18)
(n ¼ 2) and is represented by the following expression:
The Gibbs free energy change for the net cell reaction is
calculated from the general relation:
Drð17ÞG0ðTÞ kJ molꢀ1ðꢃ0:1Þ ¼ ꢀ 76:3 þ 0:0041 ðT ðKÞÞ
DrG0 ¼ ꢀnFE
(18)
ð921pT ðKÞp1150Þ
(19)
The values of Dr(17)G0(T) obtained in this study are in good
agreement (72.0 kJ molꢀ1) with those calculated using the values
of standard molar Gibbs free energy of formations for CaF2(s),
MgF2(s), MgO(s) and CaO(s) from the literature [17].
Table 2
Partial pressures of CO2(g) over {LiAl5O8+Li2CO3+5Al2O3}, {LiAl5O8+5LiAlO2+2Li2-
CO3} and {LiAlO2+Li5AlO4+2Li2CO3} determined from KEQMS as a function of
temperature
Reaction (4)
Reaction (5)
Reaction (6)
3.2.2. DfG0(T) for LiAl5O8(s)
T (K)
p(CO2) (Pa)
T (K)
p(CO2) (Pa)
T (K)
p(CO2) (Pa)
The reversible emf values obtained at different temperatures
from 785 to 1036 K for cell (II) are listed in Table 3 and the
variation of emf with temperature is shown in Fig. 5. The half-cell
reactions at each electrode can be represented as
Run 1
625
639
655
665
678
688
698
709
718
730
738
750
0.00614
0.01224
0.02622
0.03991
0.06389
0.09048
0.13369
0.20509
0.27984
0.40107
0.52235
0.77532
610
623
634
644
655
671
685
701
713
723
733
750
0.01612
0.02957
0.04840
0.07466
0.11849
0.22572
0.38709
0.69819
1.06802
1.50569
2.10282
3.63553
680
690
700
710
720
730
740
750
760
770
780
790
800
0.00128
0.00234
0.00420
0.00741
0.01287
0.02202
0.03713
0.06173
0.10127
0.16401
0.26237
0.41474
0.64813
5Al2O3ðsÞ þ 2LiFðsÞ þ 2eꢀ þ 1=2O2ðgÞ ¼ 2LiAl5O8ðsÞ þ 2Fꢀ
ðat þ ve electrodeÞ
(20)
CaOðsÞ þ 2Fꢀ ¼ CaF2ðsÞ þ 1=2O2ðgÞ þ 2eꢀ
ðat ꢀ ve electrodeÞ
(21)
(22)
Hence, the net cell reaction can be written as
5Al2O3ðsÞ þ CaOðsÞ þ 2LiFðsÞ ¼ 2LiAl5O8ðsÞ þ CaF2ðsÞ
The emf data were least-squares fitted to yield the following
linear relation:
Run 2
614
629
649
665
676
688
699
709
720
732
743
0.00441
0.00747
0.01987
0.03991
0.06510
0.09513
0.15801
0.24541
0.37522
0.58085
0.89149
604
616
626
636
642
652
662
669
679
692
697
706
719
0.00874
0.01548
0.02451
0.03825
0.04963
0.07579
0.11426
0.15118
0.22331
0.36459
0.43811
0.60579
0.96763
672
684
696
706
716
726
736
746
756
766
776
786
796
806
0.00118
0.00245
0.00498
0.00883
0.01540
0.02645
0.04476
0.07470
0.12298
0.19985
0.32073
0.50856
0.79710
1.23550
Cell ðIIÞ : E ðVÞ ðꢃ0:0020Þ ¼ 0:1820 ꢀ 1:5799 ꢂ 10ꢀ4 ðT ðKÞÞ:
ð785pT ðKÞp1036Þ
(23)
The Gibbs free energy change for the equilibrium reaction (22)
can be calculated using Eqs. (18) (n ¼ 2) and (23) and is
represented as
Drð22ÞG0ðTÞ kJ molꢀ1 ðꢃ0:02Þ ¼ ꢀ 35:1 þ 0:0305 ðT ðKÞÞ
ð785pT ðKÞp1036Þ
(24)
The standard molar Gibbs energy of formation DfG0m (LiAl5O8, s,
T) was obtained by using Eqs. (22) and (24) and values of DfG0m(T)
10
1
5 Al2O3(s) + Li2CO3(s) = 2 LiAl5O8(s) + CO2(g)
LiAl5O8(s) + 2 Li2CO3(s) = 5 LiAlO2(s) + 2 CO2(g)
LiAlO2(s) + 2 Li2CO3(s) = Li5AlO4(s) + 2 CO2(g)
0.1
0.01
0.001
10
1
0.1
0.01
0.001
10
1
0.1
0.01
1E-3
1E-4
1.20
1.25
1.30
1.35
1.40
1.45
1.50
1000 K / T
Fig. 4. Partial pressure of CO2(g) for reactions (4)–(6) as a function of temperature for two experimental runs with same phase mixtures.

ARTICLE IN PRESS
1408
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
for Al2O3(s), CaO(s), LiF(s) and CaF2(s) from literature [17] and
represented as
emf with temperature is shown in Fig. 5. The half-cell reactions at
each electrode can be represented as
Df G0mðTÞ kJ molꢀ1 ðꢃ2:9Þ ¼ ꢀ 4545:0 þ 0:8818 ðT ðKÞÞ
4LiFðsÞ þ LiAl5O8ðsÞ þ 4eꢀ þ O2ðgÞ ¼ 5LiAlO2ðsÞ þ 4Fꢀ
ðat þ ve electrodeÞ
ð785pT ðKÞp1036Þ.
(25)
(26)
3.2.3. DfG0(T) for LiAlO2(s)
The reversible emf values obtained at different experimental
temperatures for cell (III) are listed in Table 3 and the variation of
2CaOðsÞ þ 4Fꢀ ¼ 2CaF2ðsÞ þ O2ðgÞ þ 4eꢀ
ðat ꢀ ve electrodeÞ
(27)
(28)
Hence, the net cell reaction can be written as
Table 3
Variation of emf as a function of temperature for cells (I)– (IV)
LiAl5O8ðsÞ þ 2CaOðsÞ þ 4LiFðsÞ ¼ 5LiAlO2ðsÞ þ 2CaF2ðsÞ
Cell (I)
Cell (II)
Cell (III)
Cell (IV)
The emf data were least-squares fitted to yield the following
linear relation:
T (K)
E (V)
T (K)
E (V)
T (K)
E (V)
T (K)
E (V)
921
941
0.3762
0.3758
0.3754
0.3749
0.3746
0.3741
0.3736
0.3733
0.3730
0.3727
0.3724
0.3720
0.3717
0.3714
785
793
803
813
826
833
842
854
864
873
885
894
905
913
0.0591
0.0565
0.0544
0.0529
0.0535
0.0496
0.0494
0.0465
0.0458
0.0437
0.0395
0.0407
0.0369
0.0382
0.0349
0.0362
0.0330
0.0330
0.0300
0.0328
0.0265
0.0300
0.0200
0.0266
0.0191
0.0141
719
734
748
760
771
780
790
801
811
819
829
837
850
866
871
882
890
900
908
920
929
0.0107
0.0157
0.0169
0.0217
0.0244
0.0260
0.0289
0.0329
0.0338
0.0365
0.0381
0.0437
0.0430
0.0469
0.0504
0.0505
0.0534
0.0548
0.0553
0.0576
0.0591
734
747
756
769
777
789
795
807
819
832
843
0.2077
0.2068
0.2043
0.2038
0.2011
0.2000
0.1991
0.1974
0.1953
0.1935
0.1927
Cell ðIIIÞ : E ðVÞ ðꢃ0:0020Þ ¼ ꢀ 0:1569 þ 2:3522 ꢂ 10ꢀ4 ðT ðKÞÞ.
ð719pT ðKÞp929Þ
(29)
960
982
1000
1019
1042
1060
1072
1088
1101
1122
1134
1150
The Gibbs free energy change for the equilibrium reaction (28)
can be calculated using Eqs. (18) (n ¼ 4) and (29) and is
represented as
Drð28ÞG0ðTÞ kJ molꢀ1ðꢃ0:7Þ ¼ 60:6 ꢀ 0:0908 ðT ðKÞÞ
ð719pT ðKÞp929Þ.
(30)
The standard molar Gibbs energy of formation DfG0m (LiAlO2, s,
T) was obtained by using Eqs. (28) and (30) and values of DfG0m(T)
for LiAl5O8(s) from Eq. (25), CaO(s), LiF(s) and CaF2(s) from
literature [17] and represented as
923
933
942
956
961
971
984
993
1000
1009
1015
1036
Df G0mðTÞ kJ molꢀ1ðꢃ4:5Þ ¼ ꢀ 1157:8 þ 0:2118 ðT ðKÞÞ
ð719pT ðKÞp929Þ.
(31)
3.2.4. DfG0(T) for Li5AlO4(s)
The reversible emf values obtained at different experimental
temperatures for cell (IV) are listed in Table 3 and the variation of
0.380
0.375
0.370
0.20
0.15
0.10
0.05
0.00
650
700
750
800
850
900
950
T / K
1000 1050 1100 1150 1200
Fig. 5. Emf values of cells (I)–(IV) as a function of temperature.

ARTICLE IN PRESS
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
1409
emf with temperature is shown in Fig. 5. The half-cell reactions at
each electrode can be represented as
DfG0m(T) are in close agreement and can be safely extrapolated to
the entire temperature range. Thus, simultaneous use of KEQMS
and solid-state galvanic cell study increases the experimental
temperature range.
CaF2ðsÞ þ 1=2O2ðgÞ þ 2eꢀ ¼ CaOðsÞ þ 2Fꢀ
ðat þ ve electrodeÞ
(32)
3.4. Measurement of heat capacities of ternary oxides of the system
Li– Al– O
Li5AlO4ðsÞ þ 4Fꢀ ¼ LiAlO2ðsÞ þ 4LiFðsÞ þ O2ðgÞ þ 4eꢀ
ðat ꢀ ve electrodeÞ
(33)
(34)
Hence, the net cell reaction can be written as
The isobaric molar heat capacities of LiAl5O8(s), LiAlO2(s) and
Li5AlO4(s) as a function of temperature were measured from (i)
127 to 308 and (ii) 308–868 K. The variation of heat capacities of
these ternary oxides in per gram-atom were plotted as a function
of temperature and shown in Fig. 7. The figure shows that heat
capacity for LiAlO2(s) in per gram-atom is more compared to other
two ternary oxides and can be effectively used as blanket material
for breeding in nuclear reactors. Kleykamp [8] has already
reported the heat capacity data of LiAlO2(s) from 298 to 1700 K.
Heat capacity values of LiAlO2(s) determined from DSC studies are
in close agreement with that of literature [8]. However, literature
on heat capacity values of LiAl5O8(s) and Li5AlO4(s) and low-
temperature heat capacity values of ternary aluminates are not
available. The individual values of heat capacities were fitted as a
function of temperature in low and high-temperature ranges and
are represented as
Li5AlO4ðsÞ þ 2CaF2ðsÞ ¼ LiAlO2ðsÞ þ 2CaOðsÞ þ 4LiFðsÞ
The emf data were least-squares fitted to yield the following
linear relation:
Cell ðIVÞ : E ðVÞ ðꢃ0:0006Þ ¼ 0:3145 ꢀ 1:4517 ꢂ 10ꢀ4 ðT ðKÞÞ.
ð734pT ðKÞp843Þ
(35)
The Gibbs free energy change for the equilibrium reaction (34)
can be calculated using Eqs. (18) (n ¼ 4) and (35) and is
represented as
Drð34ÞG0ðTÞ kJ molꢀ1ðꢃ0:2Þ ¼ ꢀ 121:4 þ 0:0560 ðT ðKÞÞ
ð734pT ðKÞp843Þ.
(36)
The standard molar Gibbs energy of formation DfG0m (Li5AlO4, s,
T) was obtained by using Eqs. (34) and (36) and values of DfG0m(T)
for LiAlO2(s) from Eq. (31) and that of CaO(s), LiF(s), and CaF2(s)
from literature [17] and represented as
(a) 127–298 K
Cop;mðLiAl5O8Þ J Kꢀ1 molꢀ1 ¼ ꢀ 130:2 þ 1:6994 ðT ðKÞÞ
Df G0mðTÞ kJ molꢀ1ðꢃ5:6Þ ¼ ꢀ 2341:2 þ 0:4240 ðT ðKÞÞ
2
ð734pT ðKÞp843Þ.
(37)
ꢀ 0:0018ðT ðKÞÞ
(38)
(39)
3.3. Comparison of Gibbs energies of formation of LiAl5O8(s),
Cop;mðLiAlO2Þ J Kꢀ1 molꢀ1 ¼ ꢀ 45:2 þ 0:6948 ðT ðKÞÞ
LiAlO2(s) and Li5AlO4(s) determined using KEQMS and solid-state
galvanic cell
2
ꢀ 0:0010ðT ðKÞÞ
The values of standard molar Gibbs energies of formation of
LiAl5O8(s), LiAlO2(s) and Li5AlO4(s) determined from KEQMS and
solid-state galvanic cell techniques are plotted as a function of
temperature and shown in Fig. 6. The figure shows that values of
Cop;mðLiAl5O4Þ J Kꢀ1 molꢀ1 ¼ ꢀ 48:2 þ 1:2035 ðT ðKÞÞ
2
ꢀ 0:0016 ðT ðKÞÞ
(40)
-1900
Li5AlO4
-2000
-2100
-2200
640
660
680
700
720
740
760
780
800
820
840
860
950
-900
-950
LiAlO2
-1000
-1050
-1100
550
600
650
700
750
800
850
900
-3600
-3800
-4000
-4200
LiAl5O8
600
650
700
750
800
850
T / K
900
950
1000
1050
Fig. 6. Comparison of DfG0m of ternary oxides determined from KEQMS and solid-state galvanic cell techniques: (O) KEQMS, (’) solid-state galvanic cell and solid line:
combined fit of both the experimental data.

ARTICLE IN PRESS
1410
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
0.45
0.40
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
LiAlO2(s)
Li5AlO4(s)
LiAl5O8(s)
100
200
300
400
500
600
700
800
900
T / K
Fig. 7. Specific heat of ternary oxides in per gram-atom as a function of temperature.
Table 4
DfH0m(298.15 K) of lithium aluminates
Compound
DfH0m(298.15 K) kJ molꢀ1
KEQMS (second law)
KEQMS (third law)
Galvanic cell (second law)
Literature
LiAl5O8(s)
LiAlO2(s)
Li5AlO4(s)
ꢀ4517.2
ꢀ1190.5
ꢀ2360.2
ꢀ4487.8
ꢀ1187.2
ꢀ2417.1
ꢀ4535.9
ꢀ1176.4
ꢀ2360.1
ꢀ4577.5 [18]
ꢀ1190.0 [17]
ꢀ2391.9 [18]
that of Guggi et al. [19] and Ikeda et al. [1] and tabulated in
Table 4.
(b) 308–868 K
Cop;mðLiAl5O8Þ J Kꢀ1 molꢀ1 ¼ 258:3 þ 0:0734 ðT ðKÞÞ
ꢀ 5749075:9 ðT ðKÞÞꢀ2
(41)
(42)
(43)
3.6. Construction of thermodynamic table for ternary oxides of
Li– Al– O system
Cop;mðLiAlO2Þ J Kꢀ1 molꢀ1 ¼ 82:1 þ 0:0344 ðT ðKÞÞ
ꢀ 184270:1 ðT ðKÞÞꢀ2
The Gibbs energies of formation of LiAl5O8(s), LiAlO2(s) and
Li5AlO4(s) obtained from KEQMS {Eqs. (11)–(13)} and galvanic cell
techniques {Eqs. (25), (31) and (37)} were used for second law
analysis to determine the standard molar enthalpy of formation,
DfH0m(298.15 K) and the standard molar entropies S0m(298.15 K).
Isobaric molar heat capacities from Eqs. (41)–(43) and standard
molar Gibbs energies of formation for LiAl5O8(s), LiAlO2(s) and
Li5AlO4(s) from Eqs. (11)–(13) and (25), (31) and (37) were used as
primary data for second law analysis and extrapolated to 1000 K
as required. The partial pressure data obtained from KEMQS
technique at each experimental temperature were treated by the
third law method to derive the values of DfH0(298.15 K). It was
observed that for LiAl5O8(s), LiAlO2(s), the values of DfH0(298.15 K)
obtained by second and third law treatment are in good
agreement with that of literature [1,18]. However, for Li5AlO4(s),
the values obtained by the third law method are more negative
compared to second law value. For construction of thermody-
namic tables for ternary oxides, average values of DfH0(298.15 K)
obtained from second law analysis of this study and that of
Cop;mðLi5AlO4Þ J Kꢀ1 molꢀ1 ¼ 161:2 þ 0:1462 ðT ðKÞÞ
ꢀ 2851348:0 ðT ðKÞÞꢀ2
3.5. Thermo-chemical stabilities of lithium aluminates
Guggi et al. [19] have discussed the thermal stability of these
ternary aluminates and reported the heats of formation from the
constituent oxides for one mole of Li2O. Ikeda et al. [1] have
determined the heats of formation of these oxides from the
elements at 298.15 K, DfH0(298.15 K), using their high-tempera-
ture vaporization studies. In this study, same has been calculated
using solid-state galvanic cell and Knudsen effusion mass
spectrometric data. These values are in close agreement with

ARTICLE IN PRESS
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
1411
Table 5
Thermodynamic functions for the compound LiAl5O8(s)
Ã
T (K)
C0p (J Kꢀ1 molꢀ1
)
H0 (kJ molꢀ1
)
G0 (kJ molꢀ1
)
S0 (J Kꢀ1 molꢀ1
)
H0T– H0298.15 (J molꢀ1
)
fef (J Kꢀ1 molꢀ1
)
DfH0 (kJ molꢀ1
)
DfG0 (kJ molꢀ1
)
298.15
300
350
400
450
500
550
600
650
700
750
800
850
900
950
1000
216.4
216.5
237.1
251.7
262.9
272.0
279.7
286.4
292.4
297.9
303.1
308.0
312.7
317.3
321.7
325.9
ꢀ4541.6
ꢀ4541.2
ꢀ4529.8
ꢀ4517.6
ꢀ4504.7
ꢀ4491.3
ꢀ4477.5
ꢀ4463.4
ꢀ4448.9
ꢀ4434.2
ꢀ4419.1
ꢀ4403.8
ꢀ4388.3
ꢀ4372.6
ꢀ4356.6
ꢀ4340.4
ꢀ4587.9
ꢀ4588.4
ꢀ4597.1
ꢀ4607.6
ꢀ4619.6
ꢀ4633.1
ꢀ4648.0
ꢀ4664.1
ꢀ4681.4
ꢀ4699.9
ꢀ4719.3
ꢀ4739.8
ꢀ4761.4
ꢀ4783.8
ꢀ4807.0
ꢀ4831.1
155.2
157.4
192.4
225.1
255.4
283.6
309.9
334.5
357.7
379.6
400.3
420.0
438.9
456.9
474.1
490.7
0
400
155.2
156.1
158.7
165.1
173.4
183.0
193.4
204.2
215.1
226.2
237.0
247.8
258.5
269.1
279.4
289.5
ꢀ4541.6
ꢀ4541.7
ꢀ4543.6
ꢀ4545.1
ꢀ4546.2
ꢀ4550.2
ꢀ4550.9
ꢀ4551.6
ꢀ4552.1
ꢀ4552.7
ꢀ4553.1
ꢀ4553.6
ꢀ4554.2
ꢀ4555.0
ꢀ4609.1
ꢀ4609.2
ꢀ4292.5
ꢀ4291.2
ꢀ4249.2
ꢀ4207.1
ꢀ4164.8
ꢀ4122.0
ꢀ4079.2
ꢀ4036.3
ꢀ3993.3
ꢀ3950.4
ꢀ3907.3
ꢀ3864.2
ꢀ3821.2
ꢀ3778.1
ꢀ3733.8
ꢀ3687.8
11,800
24,000
36,900
50,300
64,100
78,200
92,700
107,400
122,500
137,800
153,300
169,000
185,000
201,200
Ã
Estimated values of Cp.
Table 6
Thermodynamic functions for the compound Li5AlO4(s)
Ã
T (K)
C0p (J Kꢀ1 molꢀ1
)
H0 (kJ molꢀ1
)
G0 (kJ molꢀ1
)
S0 (J Kꢀ1 molꢀ1
)
H0T– H0298.15 (J molꢀ1
)
fef (J Kꢀ1 molꢀ1
)
DfH0 (kJ molꢀ1
)
DfG0 (kJ molꢀ1
)
298.15
300
350
400
450
500
550
600
650
700
750
800
850
900
950
1000
168.4
173.6
189.3
202.1
213.1
223.1
232.4
241.2
249.7
257.9
265.9
273.9
281.7
289.5
297.1
304.8
ꢀ2379.8
ꢀ2379.5
ꢀ2370.4
ꢀ2360.6
ꢀ2350.2
ꢀ2339.3
ꢀ2327.9
ꢀ2316.1
ꢀ2303.8
ꢀ2291.1
ꢀ2278.0
ꢀ2264.5
ꢀ2250.6
ꢀ2236.3
ꢀ2221.7
ꢀ2206.6
ꢀ2418.1
ꢀ2418.4
ꢀ2425.6
ꢀ2434.1
ꢀ2443.9
ꢀ2454.9
ꢀ2467.0
ꢀ2480.2
ꢀ2494.3
ꢀ2509.4
ꢀ2525.5
ꢀ2542.4
ꢀ2560.3
ꢀ2578.8
ꢀ2598.3
ꢀ2618.5
128.6
129.6
157.6
183.8
208.2
231.2
252.9
273.5
293.1
311.9
330.0
347.4
364.3
380.6
396.4
411.9
0
300
128.6
128.6
130.7
135.8
142.4
150.2
158.5
167.3
176.2
185.2
194.3
203.3
212.3
221.3
230.0
238.7
ꢀ2379.8
ꢀ2379.9
ꢀ2381.3
ꢀ2382.4
ꢀ2383.5
ꢀ2399.6
ꢀ2400.2
ꢀ2400.4
ꢀ2400.1
ꢀ2399.4
ꢀ2398.4
ꢀ2397.1
ꢀ2395.4
ꢀ2393.4
ꢀ2401.8
ꢀ2399.0
ꢀ2244.0
ꢀ2243.2
ꢀ2220.3
ꢀ2197.2
ꢀ2174.0
ꢀ2149.1
ꢀ2124.0
ꢀ2098.9
ꢀ2073.8
ꢀ2048.7
ꢀ2023.7
ꢀ1998.7
ꢀ1973.9
ꢀ1949.1
ꢀ1924.3
ꢀ1899.3
9400
19,200
29,600
40,500
51,900
63,700
76,000
88,700
101,800
115,300
129,200
143,500
158,100
173,200
Ã
Estimated values of Cp.
literature were used as primary data. Thermodynamic table
includes the basic functions such as: DfH0(298.15 K), S0(298.15 K)
S0(T), Cp0(T), H0(T), {H0(T)–H0(298.15 K)}, G0(T), DfH0(T), DfG0(T) and
free energy function (fef) which were calculated using ‘FACTSAGE
thermo-chemical database’ software [18]. The molar heat capacity
values of Li(s), Al(s) and O2(g) required for the second law analysis
have been taken from the ‘FACTSAGE thermo-chemical database’
[18] and the thermodynamic values of the most stable phases
were used. After calculation of all the thermodynamic functions,
the values obtained at selected temperatures from 298 to 1000 K
are tabulated and are given in Tables 5 and 6 for LiAl5O8(s) and
Li5AlO4(s), respectively. Thermodynamic table for LiAlO2(s) is
already available in literature [18] hence it was not constructed in
this study.
only suitable for partial pressure measurements of permanent
gaseous species.
Acknowledgment
The authors gratefully acknowledge the help of Dr. K. Krishnan
from Fuel Chemistry Division, BARC for X-ray diffraction analysis.
References
[1] Y. Ikeda, H. Ito, G. Matsumoto, J. Nucl. Mater. 97 (1981) 47.
[2] A. La Ginestra, M. Lo Jacono, P. Porta, J. Thermal Anal. 4 (1972) 5.
[3] W. Heyi, Z. Jumke, L. Yangming, J. Nucl. Mater. 208 (1994) 195.
[4] J.N. Singh, J.T. Dusek, J.W. Sim, Ceram. Bull. 60 (6) (1981) 629.
[5] F.M. Gray, in: Polymer Electrolyte, R.C.S. Materials Monographs, The Royal
Society of Chemistry, Cambridge, 1977.
4. Conclusion
[6] F. Okasuzomer, S. Naci Koc, I. Boz, M. Ali Gurkayank, Mater. Res. Bull. 39
(2004) 715.
[7] R.A. Ribeiro, G.G. Silva, N.D.S. Mohallem, J. Phys. Chem. Solids 62 (2001)
857.
[8] H. Kleykamp, J. Nucl. Mater. 270 (1999) 372.
[9] G.A. Murray, R.A. Kematick, C.E. Myers, High Temp. Sci. 26 (1990) 4.
[10] V.L. Stolyarova, D.U. Sichen, S. Seetharaman, Vacuum 46 (8–10) (1995)
871.
[11] J.A. Basford, M.D. Boeckmann, R.E. Ellefson, A.R. Filippelli, D.H. Holkeboer,
L. Lieszkovszky, C.M. Stupak, J. Vac. Sci. Technol. A 11 (3) (1993) A22.
Standard molar Gibbs energies of formation of LiAl5O8(s),
LiAlO2(s) and Li5AlO4(s) evaluated from KEQMS and solid-state
galvanic cell techniques. This study shows that RGA based on
quadrupole mass spectrometer can effectively produce reliable
thermodynamic data using Knudsen effusion mass spectrometric
technique on proper calibration. However, the present system is

ARTICLE IN PRESS
1412
S.K. Rakshit et al. / Journal of Solid State Chemistry 181 (2008) 1402–1412
[12] M.C. Cowen, W. Allison, J.H. Batey, J. Vac. Sci. Technol. A 12 (1) (1994)
228.
[17] M.W. Chase Jr., JANAF thermochemical tables, 4th ed, J. Phys. Chem. Ref. Data
23 (1995) Monograph No. 9.
[13] M.G. Rao, C. Dong, J. Vac. Sci. Technol. A 15 (3) (1997) 1312.
[14] J.B. Mann, J. Chem. Phys. 46 (1967) 1646.
[18] FactSage, version 5.3.1, Thermo-Chemical Database Software, Thermfact, GTT
Technologies, Germany, 1976–2004.
[15] J. Huang, T. Furkawa, K. Aoto, J. Chem. Thermodyn. 38 (2006) 1.
[16] S.K. Rakshit, S.C. Parida, S. Dash, Z. Singh, B.K. Sen, V. Venugopal, J. Solid State
Chem. 180 (2007) 523.
[19] D. Guggi, H.R. Ihle, A. Neubert, in: Proceedings of the 9th Symposium on
Fusion Technology, Garmisch-Partenkirchen, Fed. Rep. Germany, Pergamon,
Oxford, 1976.