Non-equivalent water molecules in [Ni(H2O)6](ClO 4)2 and in [Ni(D2O)6](ClO 4)2 in the thermogravimetric investigations

Article abstract of DOI:10.1007/s10973-006-7497-2

The thermal decomposition of [Ni(H₂O)₆](ClO ₄)₂ and [Ni(D₂O)₆](ClO ₄)₂ were studied by thermogravimetric analysis (TG) and simultaneous differential thermal analysis (SDTA) at a constant heating rate. The gaseous products of the decomposition were on-line identified by a quadrupole mass spectrometer (QMS). In both cases the process of decomposition starts at ca. 410 K and is connected with removal of water molecules in a stepwise way; at the beginning the first water molecule is lost, then the second and at higher temperature the third one. The rest of the water molecules are lost in the temperature region of ClO ₄ ^- decomposition. The energy of activation of the process was calculated in both cases.

Full text of DOI:10.1007/s10973-006-7497-2

Journal of Thermal Analysis and Calorimetry, Vol. 89 (2007) 1, 277–281
NON-EQUIVALENT WATER MOLECULES IN [Ni(H2O)6](ClO4)2 AND IN
[Ni(D2O)6](ClO4)2 IN THE THERMOGRAVIMETRIC INVESTIGATIONS
Malgorzata Rachwalska1* and Dorota Majda2
1Department of Chemistry, Jagiellonian University, Ingardena 3, 30-060 Cracow, Poland
2Regional Laboratory of Physicochemical Analysis and Structural Research, Jagiellonian University
Ingardena 3, 30-060 Cracow, Poland
The thermal decomposition of [Ni(H2O)6](ClO4)2 and [Ni(D2O)6](ClO4)2 were studied by thermogravimetric analysis (TG) and si-
multaneous differential thermal analysis (SDTA) at a constant heating rate. The gaseous products of the decomposition were on-line
identified by a quadrupole mass spectrometer (QMS). In both cases the process of decomposition starts at ca. 410 K and is con-
nected with removal of water molecules in a stepwise way; at the beginning the first water molecule is lost, then the second and at
higher temperature the third one. The rest of the water molecules are lost in the temperature region of ClO–4 decomposition.
The energy of activation of the process was calculated in both cases.
Keywords: DTA, hexaaquanickel(II) perchlorates, QMS, TG, thermal decomposition
Introduction
both compounds. The effect of non equivalence of
water molecules in octahedral [Ni(H2O)6]2+ or
[Ni(D2O)6]2+ is probably connected with hydrogen
bonds’ net existing in the sample till the temperature
of the phase transition III–II [9] (ca. 312 K). Some ar-
ticles [10, 11] show that such nonequivalent water
molecules can exist in other perchlorate hexa-
hydrates, e.g., in CoHP or in MnHP. It seems to us
that one can easily confirm the existence of water
molecules bounded on different way in NiHP and
NiDP when investigating the process of dehydration
in those substances by the thermogravimetric method.
Dehydration takes place in the temperature region
above 365 K, i.e. above the phase transition II–I.
Therefore, such a measurement has been undertaken.
It was the main aim of the work. We wanted also to
show that the phase transitions: III–II, II–I are revers-
ible (using DSC method and comparing directly the
results for those two compounds on one picture) and
to analyze the process of decomposition using TG and
QMS method on the possible way. Using TG method
we wanted to estimate energy of activation for dehy-
dration. It was also an aim of the work.
[Ni(H2O)6](ClO4)2 (hereafter abbreviated to NiHP)
has three solid phases in the temperature range
77–320 K: a low-temperature phase (Phase III), an in-
termediate-temperature phase (Phase II) and a
high-temperature phase (Phase I). Evidence for the
existence of these phases was obtained through spe-
cific heat measurements carried out using adiabatic
calorimetry [1] and differential scannning calorimetry
(DSC) methods for NiHP and [Ni(D2O)6](ClO4)2
(NiDP) [2]. The calorimetric experiments showed that
Phase III–Phase II phase transition appears in NiDP at
a temperature higher by about 3 K than in NiHP,
namely at 314.6 K (NiDP) vs. 311.3 K (NiHP).
Phase II–Phase I transition appears at approximately
the same temperature, i.e., 362.4 K (NiDP) and 362.8
(NiHP), in the DSC method [2]. Both
Phase III–Phase II and the Phase II–Phase I phase
transitions have a dynamical [3–5] and structural
character [6, 7]. A small change of structure is exhib-
ited at the III–II phase transition [7], i.e., a symmetry
reducing the deformation of the water-ligand octa-
hedron [6], and a greater structural change is exhib-
ited at the II–I phase transition [7]. Low-frequency vi-
brations and molecular motions of H2O molecules in
NiHP and of D2O in NiDP were investigated by IINS
method [8] and by FTIR one [9]. Evidence of 3 types
of water molecules was presented using those meth-
ods [8, 9] in Phase III showing three groups of fre-
quencies of molecular motions for water molecules in
Experimental
Thermogravimetric measurements were carried out
on a derivatograph, type TGA/SDTA 851e (Mettler
Toledo), in an argon atmosphere (flow rate:
80 cm3 min–1) at a heating rate b=10 K min–1, starting
*
Author for correspondence: rachwals@chemia.uj.edu.pl
1388–6150/$20.00
Akadémiai Kiadó, Budapest, Hungary
Springer, Dordrecht, The Netherlands
© 2007 Akadémiai Kiadó, Budapest

RACHWALSKA, MAJDA
Table 1 Parameters of NiHP – thermal analysis; sample mass=40.3 mg
Stage
Temperature range/K
Mass loss/%
Mass after decomposition
Products of the decomposition
I
300–440
440–490
490–520
520–700
4.9
4.9
38.3 mg
36.3 mg
34.4 mg
8.5 mg
95.1%
90.2%
85.3%
21.0%
[Ni(H2O)5](ClO4)2
[Ni(H2O)4](ClO4)2
[Ni(H2O)3](ClO4)2
NiO
II
III
IV
4.9
64.3
Table 2 Parameters of NiDP – thermal analysis; sample mass=33.5 mg
Stage
Temperature range/K
Mass loss/%
Mass after decomposition
Products of the decomposition
I
300–440
440–490
490–520
520–700
5.3
5.3
5.3
63
31.7 mg
20.0 mg
28.2 mg
7.1 mg
94.7%
89.4%
84.1%
21.1%
[Ni(D2O)5](ClO4)2
[Ni(D2O)4](ClO4)2
[Ni(D2O)3](ClO4)2
NiO
II
III
IV
from room temperature up to 700 K. NiHP and NiDP
were prepared on the way described carefully in [2].
The number of hydrogen atoms (resulting mainly
from using 94% D2O) in the obtained compound was
estimated by NMR method to be not greater than 6%.
Samples of masses given in Tables 1 and 2 were
placed in 150 mL platinum crucibles.
A simultaneous differential thermal analysis
curve (SDTA) and a DSC curve were also recorded to
confirm evidence of the above mentioned phase tran-
sitions and also to follow the process of decomposi-
tion. Additionally, on-line quadrupole mass spec-
trometry (QMS), via a Balzer GSD 300T apparatus,
was used to allow easier interpretation of all results.
Temperature was measured by a Pt–Pt/Rh thermocou-
ple with an accuracy of ±0.5 K.
cule is lost. It can be seen from the QMS results (Figs 1
and 2) that subsequent water molecules are lost in a
stepwise process which is now associated with the re-
lease of oxygen and Cl2. This stage starts at ca. 550 K
and ends at ca. 700 K. The SDTA curve shows us also
the thermal effects connected with heating of our sam-
ples in investigated temperature region. We do not
want to analyze the curve from room temperature till
ca. 400 K because we have got more detailed DSC re-
sults in the temperature region. On DSC curve (Fig. 5)
Results and discussion
Thermal methods
Figure 1 presents TG, DTG, SDTA and QMS curves
for NiHP, and Fig. 2 presents the same results for
NiDP. At the very beginning of both experimental runs
(for NiHP and NiDP) we can see the mass increasing,
distinctly shown on DTG curves (Figs 3 and 4). We
should state that the effect could be treated as an arti-
fact. Probably it is connected with the influence of a
magnetic field, emanating from an active heating sys-
tem surrounding the sample, on the sample’s paramag-
netic ions. Our results distinguish 4 stages of decompo-
sition in NiHP as well as in NiDP. The results are
presented in Tables 1 and 2, and graphically displayed
in Figs 1 and 2. The results are shown more distinctly,
as they pertain to the first stage of dehydration, on
Figs 3 and 4. In the first stage the first water molecule
is lost, in the second stage the second water molecule is
lost and in the third stage, again, another water mole-
Fig. 1 TG, DTG, SDTA and QMS curves for NiHP in the
range of 330–700 K, heated at a constant heating rate
of 10 K min–1
278
J. Therm. Anal. Cal., 89, 2007

NON-EQUIVALENT WATER MOLECULES IN [Ni(H2O)6](ClO4)2 AND IN [Ni(D2O)6](ClO4)2
Fig. 4 TG and DTG curves for NiDP in the temperature region
where the first, the second and the third water molecule
is removed
DSC curve above ca. 520 K is the sign that our cruci-
ble is not tight. Therefore, we had to stop our DSC
experimental run at ca. 550 K. Figure 5 shows that the
process of heating of our samples is connected only
with endo- thermic effects till 520 K. The first two
peaks are connected with phase transitions in the
solid-state at ca. 313 and 363 K. They confirm well
our previous DSC results [2]. A comparison of DSC
results for NiHP and NiDP samples is shown in
Fig. 6. These results correspond well to our previous
results [2]. They show distinctly that phase transitions
in the solid modification at ca. 313 and 363 K are re-
versible, which was not so carefully investigated in
[2]. One can easily see a shift of phase transition tem-
peratures connected with deuteration.
Fig. 2 TG, DTG, SDTA and QMS curves for NiDP in the
range of 330–700 K, heated at a constant heating rate
of 10 K min–1
A comparison of the process of dehydration for
NiHP and NiDP is of interest. From Fig. 7 it can be
seen that the process by which the first water molecule
Fig. 3 TG and DTG curves for NiHP in the temperature region
where the first, the second and the third water molecule
is removed
we can see a large endothermic split peak, (about
470 K) which is in correspondence with a distinctly
shown peak on the SDTA curve at the temperature.
The effect is connected with losing of two water mol-
ecules (as we can see from Tables 1 or 2). No smooth
Fig. 5 DSC curve for NiHP. At ca. 450 K one can see that the
process of NiHP decomposition starts
J. Therm. Anal. Cal., 89, 2007
279

RACHWALSKA, MAJDA
tion of the total decomposition of our samples is diffi-
cult because of its decomposition with explosion well
seen as an exothermic effect at ca. 600 K (SDTA) and
on QMS curve as great oxygen peaks. In connection
with the effect of explosion an estimation of intermedi-
ate decomposition’s products is difficult. The curve
signed by Cl2 on QMS results does not shows without a
doubt that chlorine is one of decompositions’ products.
Chlorine can appear as a product of decomposition of
ClO2 in our samples or as a result of ionization of ClO2
and of ionization of some another compounds com-
posed of oxygen and chlorine in QMS spectrometer.
Speaking about NiO we should sign rather the chemi-
cal formula as NiOx. The value of x calculated from TG
results is equal to 1.14 for NiHP and 1.31 for NiDP.
Because x is not exactly equal to 1 we can state that ob-
tained NiO compound is not stoichiometric one.
Calculation method
Based on the results from TG, the integral method of
Coats and Redfern [12] was used to study the kinetics
of dehydration of NiHP and NiDP. The kinetic pa-
rameters can be derived using a modified Coats–Red-
fern equation [15]:
Fig. 6 DSC curves for NiHP and NiDP when heating, cooling
and again heating and cooling these samples. In the
lowest part of the figure the thermal program is shown
is lost starts at a temperature for NiDP comparable to
that for NiHP. This further suggests that the activation
energy of the process of dehydration for NiDP is close
to that for NiHP. However, it should be noted that a
main extreme on the DTG curve (Fig. 7), correspond-
ing to release of the first water molecule, appears at a
distinctly lower temperature for NiHP than for NiDP.
Total decomposition of those samples seems to go on
the following way: the product of three first steps of
dehydration transforms in the following way:
[Ni(H2O)3](ClO4)2®NiO+Cl2+4O2+3H2O. On the
QMS curve (Figs 1 and 2) we can see signs of oxygen
and water even at the highest temperature of the mea-
surements i.e. at ca. 650 K. More detailed interpreta-
g(a)
T 2
AR
E
é
ù
ln
= ln
–
qE RT
ê
ú
ë
û
where a is the degree of transformation defined as
W0 – W
a =
W0 – W¥
where W is the mass of the solid and the subscripts 0
and ¥ refer to the initial and residual amounts, respec-
tively, q refers to the heating rate, and g(a) is a func-
tion, the expression of which depends on the kinetic
model of the specific reaction. If a correct g(a) is used,
a plot of ln[g(a)/T 2] vs. 1/T should give a straight line,
from which some values of the activation energy E and
the pre-exponential factor A in the Arrhenius equation
can be calculated. The formal expressions of the func-
tions g(a) depend on the conversion mechanism and its
mathematical model [13]. The latter usually represents
the limiting stage of the reaction – the chemical reac-
tions; random nucleation and nuclei growth; phase
boundary reaction or diffusion. If the correct g(a) is
used, the corresponding linear dependence should give
the highest correlation coefficient from the linear re-
gression analysis. The most common kinetic models
and their algebraic expressions one can find in [13].
Figure 8 presents Coats and Redfern plots of dehydra-
tion of NiHP and Fig. 9 presents the dehydration of
NiDP. The energy of activation value obtained for de-
hydration (the first step of the process) of NiHP is
equal to 77.74±0.24 kJ mol–1 and for NiDP is equal to
Fig. 7 Comparison of TG and DTG curves for NiHP and NiDP
280
J. Therm. Anal. Cal., 89, 2007

NON-EQUIVALENT WATER MOLECULES IN [Ni(H2O)6](ClO4)2 AND IN [Ni(D2O)6](ClO4)2
water molecules are lost in the time of perchlorate
ions’ decomposition (clearly seen on QMS results).
These results are compatible with those obtained in
[8, 9] showing three groups of frequencies for water
molecules in each of both compounds. The effect of
non equivalence of water molecules in octahedral
[Ni(H2O)6]2+ or [Ni(D2O)6]2+ is probably connected
with hydrogen bonds’ net existing in the sample till
the temperature of the phase transition III–II [9].
• Total decomposition of both samples is connected
with losing of 3 and then losing of rest of water
molecules and also with decomposition of per-
chlorate anion to O2, Cl2. NiO is the solid remain-
der of the process.
Fig. 8 Coats and Redfern plot of NiHP dehydration (the first
water molecule is removed)
• The value of energy of activation obtained for dehy-
dration (the first step of the process) of NiDP is com-
parable to that for NiHP and is equal ca. 77 kJ mol–1.
References
1 M. Rachwalska and T. Stanek, Phys. Stat. Sol. A,
48 (1978) 297.
2 M. Godlewska and M. Rachwalska, J. Thermal Anal.,
45 (1995) 1073.
3 J. M. Janik, A. MigdaÓ-Mikuli, E. Mikuli and T. Stanek,
Acta Phys. Pol., A 59 (1981) 599.
4 J. A. Janik, J. M. Janik, K. Otnes and T. Stanek,
Acta Phys. Pol., A 59 (1981) 815.
5 C. NÞldeke, B. Asmussen, W. Press, H. Buttner,
G. Kearley, R. E. Lechner and B. Ruffle, J. Chem. Phys.,
113 (2000) 3219.
Fig. 9 Coats and Redfern plots of NiDP dehydration (the first
water molecule is removed)
76.26±0.13 kJ mol–1. Thus the energy of activation for
NiDP is comparable to that for NiHP. The best model
used for experimental values of a was F3/2 i.e. reaction
model: three-halves order kinetics
6 A. K. Jain and G. C. Upreti, J. Phys. C, 13 (1980) 5177.
7 M. GoÓ·b and M. Rachwalska, has been getting ready for
publication.
8 M. Rachwalska and I. Natkaniec, Physica B,
348 (2004) 371.
f(a)=(1– a)3/2
9 M. Rachwalska, E. ŸciesiÕska and J. SciesiÕski,
Z. Phys. Chem., 220 (2006) 1021.
a
10 M. B. Patel, S. Patel, D. P. Khandelwal and H. D. Bist,
Chem. Phys. Lett., 101 (1983) 93.
da
g(a) =
= kt = 2[(1– a)–1/ 2 –1]
ò
f (a)
0
11 K. C. Moss, D. R. Russel and D. W. A. Sharp,
Acta Crystallogr., 14 (1961) 330.
An important conclusion from those measure-
ments is that there are at least 3 kinds of water mole-
cules in NiHP and NiDP. This result agrees with our
previous results [8, 9]. We can admit that the problem
discussed in the article from the main point of view is
comparable with similar problems considered in an-
other publications for example [14, 15].
12 A. W. Coats and J. P. Redfern, Nature (London),
201 (1964) 68.
13 J. J. œrf±o and F. G. Martins, Thermochim. Acta,
390 (2002) 195.
14 P. Masset, J. Y. Poinso and J. C. Poisnet, J. Therm. Anal. Cal.,
83 (2006) 457.
15 L. T. Vlaev, V. G. Georgieva and S. D. Genieva,
J. Therm. Anal. Cal., 83 (2006) 421.
Conclusions
Received: January 8, 2006
Accepted: June 9, 2006
• The two-phase transitions in solid NiHP (311.3,
362.8 K) and NiDP (314.6, 362.4 K) are reversible.
• When heating those two samples lose water mole-
cules and then they decompose. Losing of 3 first
molecules is realized in three separate steps. The rest
OnlineFirst: October 20, 2006
DOI: 10.1007/s10973-006-7497-2
J. Therm. Anal. Cal., 89, 2007
281