2
H.J. Seifert, S. Funke / Thermochimica Acta 320 (1998) 1±7
To avoid such misinterpretations we used another
2.3. Solution calorimetry
strategy when investigating the dehydration of the
hydrates of cerium (III) chloride [2]: CeCl3ÁxH2O
was heated in a drying cabinet and greater than
908C in an HCl stream to temperatures increasing
all 10 h at 58C. The samples ± about 10 g in ceramic
boats ± were weighed at each step to control the loss
of water. If the weight loss was as high as calculated
for the formation of an expected lower hydrate, the
composition of the sample was controlled by X-ray
diffraction (XRD). Thus, we could ®nd out if either
a new phase was formed or if we had a mixture of
the two phases. This decision was possible by com-
paring the pattern with that of analogous hydrates
of other lanthanide chlorides. Thus, we could pre-
pare the existing hydrates in amounts suf®cient for
measuring solution enthalpies. These measurements
are described in this paper; the ÁsolH0 values are
discussed depending on the ionic radii of the Ln3
ions.
The heats of solution were measured using an
isoperibolic underwater calorimeter ®rst described
in 1978 [4]. The solvent H2O was slightly acidi®ed
with HCl (10 2 mol l 1). In a silver vessel of 1.3 l
volume samples of 2±4 g in thin-walled glass
ampoules are cracked under water at the start of the
measurement. Thus, with a dissolution rate 1:15 000
virtually ideal solutions are formed. The change of
temperature ÁT against the surroundings ± a thermo-
stat at ca. 258C with a temperature constancy of
2Â10 48C ± generated by the solution enthalpy is
measured with a thermopile. The inevitable heat loss is
compensated according to Dickinson: The corrected
temperature change ÁT* is obtained by setting F1F2
in Fig. 1. The calibration was done by Joule's heat
giving a temperature change ÁTE. The heat of solution
QL is ®nally calculated from the equation
QL
QE
ÁTÃ ÁTE
For the solution enthalpy of KCl we found
1
17.67Æ0.05 kJ mol from the literature [5].
2. Experimental
2.1. Preparation of hexahydrates LnCl3Á6H2O
Lanthanide oxides (99.9%, Fa. Heraeus, Hanau)
were dissolved in concentrated hydrochloric acid
and evaporated to saturation at ca. 508C. Under cool-
ing with ice the chloride±hexahydrates were precipi-
tated with gaseous HCl, sucked off with a glass sinter
crucible, washed with ether and dried at ca. 508C.
They were identi®ed by comparing their X-ray pat-
terns with calculated intensities in s.g. P2/n, using the
site parameters of PrCl3Á6H2O [3]. The correct water
content was controlled by thermogravimetry (Ther-
mobalance TG 750, Stanton Redcroft), using dehy-
dration to the trihydrate. (Chemical analysis for Ln
and Cl was not applied because the results are not
suf®ciently sensitive.)
3. Results
In Table 1 the measured solution enthalpies are
compiled. Each value is the mean of two measure-
ments with the deviation for the last digit in brackets.
Below these values the preparation temperatures at
which the weight loss of ca. 10 g hydrate has become
greater than ca. 0.05 g in 10 h are given. Then, the
sample was kept at this temperature till the weight loss
reached the calculated value. Experience had proved
that at temperatures higher than ca. 708C the forma-
tion of oxide chlorides by hydrolysis became effective.
As a consequence only the dehydration of hexahy-
drates up to NdCl3Á6H2O could be performed in air,
using a drying cabinet.
2.2. X-ray powder patterns
One example for this method: From 10.00 g
HoCl3Á6H2O, 8.56 g HoCl2Á2H2O will be formed.
The temperature of a perceptible degree of dehydra-
tion was 708C. This procedure had to be performed in
an HCl atmosphere. After 30 h the formation of the
dihydrate was complete; this was controlled by an
XRD record.
Powder patterns were taken at ambient temperature
with a Philips PW 1050/25 goniometer equipped with
a proportional counter and a vacuum attachment.
During exposure (Cu Kꢀ radiation) the samples were
kept under a helium atmosphere.