Kinetic investigations on the basis of isothermal DSC measurements of hydrogenation and dehydrogenation of magnesium hydride

Article abstract of DOI:10.1016/S0040-6031(97)00392-4

Results of kinetic investigations on the basis of isothermal DSC measurements by means of defined pressure changes are presented and interpreted for the Mg/MgH₂ system. From an evaluation of the experimentally obtained DSC-signals, the dependen

Full text of DOI:10.1016/S0040-6031(97)00392-4

Thermochimica Acta 310 (1998) 167±171
Kinetic investigations on the basis of isothermal DSC measurements of
1
hydrogenation and dehydrogenation of magnesium hydride
*
K. Bohmhammel, B. Christ , G. Wolf
Institute of Physical Chemistry, Technical University Freiberg, Leipziger Straûe 29, D-09596 Freiberg, Germany
Received 30 April 1997; received in revised form 25 June 1997; accepted 30 June 1997
Abstract
Results of kinetic investigations on the basis of isothermal DSC measurements by means of de®ned pressure changes are
presented and interpreted for the Mg/MgH system. From an evaluation of the experimentally obtained DSC-signals, the
2
0
dependence of reaction rate on hydrogen pressure p , temperature and sample morphology could be described. Under certain
conditions, it is possible to visualise individual rate-determining steps and, thus, to arrive at conclusions on the reaction
mechanism. # 1998 Elsevier Science B.V.
Keywords: Activation energy; DSC; Kinetics; Magnesium hydride
1
. Introduction
kinetic and thermodynamic interpretation of the
results obtained from hydrides of intermetallic com-
pounds of magnesium (e.g. La, Ce or Cu), it seems
The hydrogenation and dehydrogenation of metals
and metal alloys are simple heterogeneous reactions,
the kinetics of which are of scienti®c as well as
technological interest. Mg/MgH , as a typical repre-
sentative of high-temperature hydrides, gains much
attention in all respects, because it reaches the highest
hydrogen content (7.65 wt%) among all the compar-
able metal hydrides. Due to its poor absorption
kinetics, the high hydride stability and the low cycle
stability, however, it has not yet found large-scale
application in hydrogen or heat storage.
necessary to investigate the behaviour of MgH much
2
more in detail.
By means of isothermal DSC measurements an
extensive kinetic characterisation of the system should
be given, which implies formulation of statements on
the dependence of reaction rate on hydrogen pressure
p0 and temperature. These results enable conclusions
concerning the reaction mechanism.
2
So far as thermodynamics and kinetics of the
system is concerned, rather different results are found
in literature. In order to establish a basis for further
2
. Experimental
2
.1. Sample material
*
Corresponding author. Fax: +49 3731/39/3588.
1
Three different magnesium hydride samples were
investigated.
Presented at the Twelfth Ulm-Freiberg Conference, Freiberg,
Germany, 19±21 March 1997
0040-6031/98/$19.00 # 1998 Elsevier Science B.V. All rights reserved
P I I S 0 0 4 0 - 6 0 3 1 ( 9 7 ) 0 0 3 9 2 - 4

1
68
K. Bohmhammel et al. / Thermochimica Acta 310 (1998) 167±171
within short, and thus the reaction had been started
using time de®nition as a precondition for kinetic
evaluation.
ꢀ
ꢀ
Magnesium hydride from Th. Goldschmidt AG,
Essen [1] ± procedure of synthesis: autocatalyti-
cally produced MgH with a mean particle size
2
<53 mm
Magnesium hydride from Max-Planck-Institut f uÈ r
Kohleforschung, M uÈ hlheim (Dept. B. Bogdanovic)
3. Results and discussion
±
cally produced MgH with a mean particle size
procedure of synthesis: homogeneously catalyti-
The evaluation of DSC measuring curves has been
accomplished by means of a curve ®t on the basis of
the TIAN equation and under the assumption of a ®rst-
order reaction [2].
2
<60 mm
ꢀ
Magnesium hydride, prepared by the authors ±
procedure of synthesis: synthesis from the elements
(
magnesium powder (<45 mm) has been exposed in
an autoclave at 673 K and a hydrogen pressure of
5 bar (RT) for ca. 45 h)
While the absorption curve shows only one step of
reaction and the evaluation by means of curve approx-
imation is rather successful, the desorption curve
shows distinctly a three-step sequence of reaction.
In this case only a numerical evaluation is possible.
Fig. 1 presents a typical resulting DSC curve of
absorption and desorption of MgH2.
In order to analyse systematic and statistical uncer-
tainties, simulation measurements with electrically
produced heat effect were carried out, and the recon-
structed signal had been compared with the starting
signal. The following results were obtained:
3
2
.2. Calorimetric measurements
A differential scanning calorimeter of the type DSC
11 from SETARAM has been applied. The isother-
1
mal DSC measurements (Tˆ633, 653 and 663 K) with
de®ned pressure changes were carried out in such a
way that, in a gas storage tank of 1 l volume, a ®xed
hydrogen pressure was pre-set. By opening a valve
connecting the gas reservoir vessel with the high-
pressure crucible (sample, VreservoirꢁVcrucible), the
pressure in the measuring crucible had been changed
ꢀ The numerical evaluation is to be preferred to a
balanced calculation on the basis of a model func-
tion.
Fig. 1. DSC curve of hydrogenation and dehydrogenation of MgH
2
(G) at Tˆconst.

K. Bohmhammel et al. / Thermochimica Acta 310 (1998) 167±171
169
ꢀ
Half-life of heat production (reaction) should be
lower, at least 10 times lower than constant of the
measuring system.
in accordance with the reacted fractions obtained by
Stander [3].
3
.2. Kinetics of the dehydrogenation of
magnesium hydride
3
.1. Hydrogenation kinetics of magnesium
At a given temperature the DSC curve of dehydro-
The given evaluation of absorption kinetics at con-
stant temperature leads to ®rst-order rate constants
depending strictly linearly on pressure difference
genation has nearly the same shape and area. Due to
the multistage reaction sequence a modelling evalua-
tion is impossible, and a numerical evaluation has
been undertaken, by which the desorption curve, ®rst
of all, has been numerically reconstructed. The values
for the time constants were taken from electrical
calibration. The integration of the resulting perfor-
(
equilibrium pressure (Fig. 2).
p � p ), where p represents the thermodynamic
0
eq
eq
k ˆ k0ꢂp0 � peq†
The temperature dependence on k can be described
0
mance curve q_ ꢂt† and the reference to q leads to the
ges
by the Arrhenius equation. Apparently, the value of the
activation energy does not depend on the sample and is
reacted fraction-time curve ꢀ(t):
on average E ˆ100 kJ/mol (Fig. 3). In comparison,
t
A
Z
1
Stander [3,4] calculated an activation energy of
5.5 kJ/mol on the basis of isothermal gravimetric
ꢀ
ꢂt† ˆ
q_ ꢂt†dt
9
qges
0
and volumetric measurements. The resulting different
ordinate intersections are due to different sample
surfaces.
The obtained reacted fraction±time curve are abso-
lutely comparable to those curve behaviours known
from literature [4], which were gained from pressure±
time or volume±time measurements. The structure of
the desorption curve is completely lost, when these
integral method are used so that the importance of
The enthalpy change, Á H obtained under iso-
R
sp
thermal conditions are 1.47 kJ/g MgH at a maximum
2
independently of the sample. This value corresponds
to a reacted fraction of 50% reference to MgH and is
2
0 2
Fig. 2. Dependence of rate constants on pressure p at constant temperature of MgH (E).

1
70
K. Bohmhammel et al. / Thermochimica Acta 310 (1998) 167±171
0 2
Fig. 3. Temperature dependence of k for the absorption of MgH (E).
Fig. 4. Reacted fraction±time curves of MgH
2
(E) at different desorption pressures (Tˆ663 K).
isothermal DSC measurements should be emphasised
again.
temperature. An in¯uence on the reacted fraction
cannot be observed.
The effect of pressure on the desorption rate of
hydrogen can be taken from the reacted fraction±time
curves of MgH (E) in Fig. 4. The results show that the
4. Conclusions
2
decomposition reaction of MgH becomes slower with
2
On the basis of the presented results a detailed
interpretation of the reaction mechanism is still impos-
decreasing pressure difference (p � p ) at constant
eq
0

K. Bohmhammel et al. / Thermochimica Acta 310 (1998) 167±171
171
sible. Referring to the hitherto obtained results the
following conclusions can be made.
acterised by the formation of Mg-nuclei. The more
rapid crystal growth leads to the second maximum in
the performance curve. The third maximum can be
referred to the volume contraction, accompanying the
formation of the magnesium phase, which leads to
crack formation thus enabling additional options of
diffusion. Furthermore, detailed studies on pressure
dependence of desorption rate seem to be necessary in
order to make a still more exact statements concerning
the mechanism of reaction.
The hydrogenation rate is primarily determined by
the step of dissociative chemisorption of hydrogen.
This is supported by the pressure-proportional velo-
city.
With the growth of magnesium hydride from the
boundary into the grain, the diffusion possibilities of
hydrogen are minimised. Thus, an apparent order of
reaction results. The assumed ®rst order has only a
formal character.
Assuming that hydrogenation and dehydrogenation
of magnesium and magnesium hydride, respectively,
are based upon the mechanism of nucleation and
growth, an in¯uence of pressure difference (p � p )
References
0
eq
on the number of nuclei developing cannot be derived
as described by Vigeholm [5], because a dependence
of reacted fraction on pressure p0 could not be
observed.
[
1] G. Koerner, K.-D. Klein, W. Knott, Z. Naturforsch. 47b
(1992) 767±771.
[
2] K. Bohmhammel, B. Christ, G. Wolf, Thermochim. Acta 271
(
1996) 67±73.
[
3] C.M. Stander, Z. Phys. Chem. NF 104 (1977) 229±238.
4] C.M. Stander, J. Inorg. Nucl. Chem. 39 (1977) 221±223.
Obviously, the dehydrogenation of MgH2 starts
with a rapid formation of ꢀ-MgH at the surface.
The following incubation period is apparently char-
[
2
[5] B. Vigeholm, J. Kjoller, B. Larsen, A.S. Pedersen, J. Less-
Common Met. 89 (1983) 135±144.