Thermal decomposition of ammonia-borane under pressures up to 600 bar

Article abstract of DOI:10.1016/j.tca.2005.07.002

The thermal decomposition of ammonia-borane BH₃NH₃ in the temperature range up to 450 K has been studied by differential scanning calorimetry (DSC) and volumetric analysis of the released volatile decomposition products. Measurements were performed in a transitiometer ST6-VI under pressures up to 600 bar and in a DSC C-80 in the pressure range 1-100 bar hydrogen. Above 360 K ammonia-borane undergoes an exothermic decomposion, which proceeds in two steps with rising temperature. The decomposition is accompanied by hydrogen release. Formation of further volatile products, beside hydrogen, seems to be negligible. The heat evolution and hydrogen release terminates near 430 K. The final amount of released hydrogen is approximately equal to 2 mol H₂/mol ammonia-borane. Variation of pressure does not influence significantly the reaction enthalpy and hydrogen release. The transitiometer ST6-VI is well-suitable for the monitoring of solid-gas reaction under high-pressure conditions. This instrument enables a reliable determination of the reaction heat and the amount of gas release/gas uptake.

Full text of DOI:10.1016/j.tca.2005.07.002

Thermochimica Acta 445 (2006) 121–125
Thermal decomposition of ammonia–borane under
pressures up to 600 bar
a,∗
a
b
c
d
F. Baitalow , G. Wolf , J.-P.E. Grolier , F. Dan , S.L. Randzio
a
Department of Physical Chemistry, Freiberg University of Mining and Technology, Leipziger Str. 29, D-09596 Freiberg, Germany
b
Laboratory of Thermodynamics of Solutions and Polymers, Blaise Pascal University, 24 Avenue des Landais, 63177-Aubiere Cedex, France
c
Department of Macromolecular Chemistry, Gh. Asachi Technical University, 71A Mangeron Avn., 6600-Iasi, Romania
d
Institute of Physical Chemistry, Polish Academy of Science, ul. Kasprzaka 44/52, 01-224 Warzaw, Poland
Abstract
The thermal decomposition of ammonia–borane BH
3
3
NH in the temperature range up to 450 K has been studied by differential scanning
calorimetry (DSC) and volumetric analysis of the released volatile decomposition products. Measurements were performed in a transitiometer
ST6-VI under pressures up to 600 bar and in a DSC C-80 in the pressure range 1–100 bar hydrogen. Above 360 K ammonia–borane undergoes
an exothermic decomposion, which proceeds in two steps with rising temperature. The decomposition is accompanied by hydrogen release.
Formation of further volatile products, beside hydrogen, seems to be negligible. The heat evolution and hydrogen release terminates near
2
430 K. The final amount of released hydrogen is approximately equal to 2 mol H /mol ammonia–borane. Variation of pressure does not
influence significantly the reaction enthalpy and hydrogen release. The transitiometer ST6-VI is well-suitable for the monitoring of solid–gas
reaction under high-pressure conditions. This instrument enables a reliable determination of the reaction heat and the amount of gas release/gas
uptake.
©
2005 Elsevier B.V. All rights reserved.
Keywords: Ammonia–borane; Thermal decomposition; DSC; Volumetric measurements
1
. Introduction
age capacity. But it is often of great interest to obtain addi-
tionally values of the reaction enthalpy both for the hydrogen
uptake and the hydrogen release direction. Calorimetry is the
method of choice for obtaining such enthalpy values in a wide
range of pressures. Furthermore, the calorimetric monitoring
is a well-suitable tool for a quick obtaining of preliminary
information about the occurrence of several reaction steps in
case of a more complicated reaction pathway.
The aim of the presented work is to illustrate the per-
formance of the high-pressure transitiometer ST6-VI from
BGR-TECH (Poland) for reaction monitoring. It is con-
structed according to the principles of scanning transitiom-
etry, previously described [1]. This instrument enables the
simultaneous monitoring of the calorimetric signal and the
volume change of the reaction system. From the volume
change the released or consumed amount of gaseous reac-
tion products can be determined. Beside usual temperature
Decomposition reactions, which proceed under uptake or
release of gaseous reaction products, are in principle influ-
enced by the pressure of the gas phase in the reaction system.
The variation of pressure might lead to a larger or smaller
shift of the reaction equilibrium, depending on the corre-
sponding thermodynamic and kinetic parameters for the reac-
tion. Example for such reactions is the reversible hydrogen
uptake/hydrogen release in hydrogen storage materials.
The most common detection methods in studying the
reversible hydrogen exchange under different pressures
involve volumetric or gravimetric measurements. These
methods deliver important data regarding the hydrogen stor-
∗
Corresponding author. Tel.: +49 3731 39 2137; fax: +49 3731 39 3588.
E-mail address: felix.baitalow@chemie.tu-freiberg.de (F. Baitalow).
0
040-6031/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.tca.2005.07.002

1
22
F. Baitalow et al. / Thermochimica Acta 445 (2006) 121–125
scans, the transitiometer allows the performance of pressure
scans, which might be a favorable experimental tool for the
monitoring of reversible processes in solid–gas systems.
This paper presents results of investigations of the ther-
mal decomposition of solid ammonia–borane BH3NH3, also
known as borazane. Measurements were made with the tran-
sitiometer ST6-VI in the high-pressure range and with a DSC
C-80 device from Setaram in the low-pressure range. The
thermal decomposition of ammonia–borane at ambient pres-
sure is already well investigated [2–7]. Due to the exothermic
decomposition enthalpy, it could be expected that an increase
of pressure in the range below 1000 bar does not influence
significantly the reaction enthalpy and amount of released
hydrogen. Thus, ammonia–borane seems to be a viable cali-
bration substance for a performance check of the transitiome-
ter ST6-VI.
a stainless steel high-pressure capillary tube to the piston
pump. The piston pump is driven by a computer-controlled
stepper motor through a gear box.
Mercury was used as pressure transmitter. Thanks to
the mercury column volatile reaction products, which were
released during the decomposition of ammonia–borane,
remain inside the calorimetric vessel. Because the pressure
detector and the stepper motor are connected to the computer,
any of the three variables – temperature, pressure or volume
– can be changed stepwise or scanned continuously. Thus,
beside the usual temperature scans at isobaric conditions,
the transitiometer ST6-VI allows the performance of pres-
sure scans at isothermal conditions, which could be of spe-
cial interest during the investigation of solid–gas reactions.
Changes of the volume in the system are determined simulta-
neously to the calorimetric output by counting and recording
the number of steps of the stepper motor controlling the pis-
ton displacement of the pump. The resolution of volumetric
−
6
3
2
. Experimental
measurementsis(5.24 ± 0.04) × 10 cm /motorstep. How-
−
4 3
ever, the practical resolution was 7 × 10 cm due to the
sensitivity of the pressure detection. A detailed description
of the transitiometer ST6-VI is given in [8]. Measurements of
the thermal decomposition of ammonia–borane were carried
out in the pressure range up to 600 bar at a heating rate of
0.15 K min . Typical sample mass was about 50 mg. Exper-
iments were performed on a ST6-VI device placed in the
Blaise Pascal University in Clermont-Ferrand, France.
In the DSC C-80 from Setaram ammonia–borane was
decomposed under hydrogen atmosphere in the pressure
2
.1. Instruments
A detailed scheme of the transitiometer ST6-VI is repre-
sented in Fig. 1. It consists of two main parts, a calorimetric
block and a high-pressure piston pump connected to the reac-
tion vessel. The instrument is equipped with high-pressure
vessels made from 316 stainless steel. The ammonia–borane
sample was placed in a Teflon tubing sealed at both ends
with Teflon plugs. The reaction vessel is connected through
−
1
Fig. 1. A diagram of the transitiometric technique used in the present study.

F. Baitalow et al. / Thermochimica Acta 445 (2006) 121–125
123
range 1–100 bar at a heating rate of 0.15 K min 1. High-
pressure gas circulation vessels made from hastelloy 78206
alloy were used. The reaction vessel was connected through
a capillary tube to a pressure detector placed outside the
DSC C-80. The amount of released gaseous products was
calculated from the variation of pressure after the correction
for pressure change due to the thermal expansion. Volatile
decomposition products, which condense at room temper-
ature, were not detected by this equipment. The internal
volume of the reaction system was calibrated using hydro-
gen feed with a definite gas flow rate. Typical sample mass
at experiments was about 50 mg.
−
2
.2. Materials
Fig. 2. Temperature dependence of the heat flow (–) and the amount
of released volatile products (– –) at the thermal decomposition of
ammonia–borane BH3NH3 (transitiometer ST6-VI, 600 bar, mercury as
Ammonia–borane BH3NH3 (+90%, KatChem, Czech
−
1
pressure transmitter, heating rate 0.15 K min ).
Republic) was re-crystallized from diethyl ether by Soxthlet
extraction. All purification steps and the sample preparation
for calorimetric measurements were performed under argon
using Schlenk technique.
The purity of the used ammonia–borane sample was
proved by elemental analysis, powder X-ray diffraction anal-
ysis and H NMR, B NMR measurements in acetonitrile.
Boroncontentwasdeterminedphotometrically, nitrogencon-
tent by means of an Herraeus Analyser CHN-O-Rapid instru-
ment, the content of hydridic hydrogen by volumetric deter-
mination of the gas release at the acidic hydrolysis (1 M
aqueous HCl).
products as shown in Fig. 2. The amount of released products
was calculated from the detected variation of volume after
correction for thermal expansion. Just as in case of calorimet-
ric monitoring of the ammonia–borane decomposition, both
steps of gas evolution are not clearly separated. Gas evolution
is terminated near 430 K, the final value amounts to approx-
imately 2.0 mol volatile products/mol ammonia–borane. In
each decomposition step roughly 1 mol volatile products/mol
ammonia–borane was released.
1
11
It is well known that the thermal decomposition of
ammonia–borane is accompanied by release of hydrogen
[4–7]. The amount of released hydrogen was exclusively
3
. Results and discussion
recorded by means of gas-volumetric measurements at atmo-
spheric pressure. It was shown that the amount of released
hydrogen is not significantly influenced by experimental con-
ditions, for example it is nearly independent on the heating
rate used. The final amount of released hydrogen, which
was achieved at temperatures near 473 K in earlier investiga-
tions, is approximately equal to 2.2 ± 0.1 mol hydrogen/mol
ammonia–borane [5,6].
As has been shown by means of mass spectrometry and
FT-IR spectroscopy [6], in addition to hydrogen some further
gaseous products, such as monomeric aminoborane BH2NH2
and borazine (BHNH)3 (which is the B–N-analog of ben-
zene), were released during the thermal decomposition of
ammonia–borane. Contrary to the release of hydrogen, the
final amount of these additional volatile products strongly
increases with rising heating rate.
Thevariationofthevolumeofthereactionsystemdetected
in the transitiometer ST6-VI includes contributions of all
volatile reaction products. Due to the use of mercury as
pressure-transmitter, all volatiles remain in the hot reaction
area inside the calorimetric vessel. The great advantage of
this technique is that both the thermal behavior of the solid
sample and the formed gaseous products could be monitored.
The final amount of volatile products detected at the ther-
mal decomposition of ammonia–borane in the transitiometer
3
.1. High-pressure transitiometer
Ammonia–borane BH3NH3 was used to check the perfor-
mance of the high-pressure transitiometer ST6-VI. Obtained
results for the thermal decomposition of ammonia–borane
under 600 bar in the temperature ranges up to 453 K were rep-
resented in Fig. 2. Decomposition temperature was increased
−
1
with a heating rate of 0.15 K min . Fig. 2 shows the heat
flow data obtained after the subtraction of the constructed
baseline from the recorded calorimetric curve. First thermal
event starts near 360 K. There are two significant exothermic
events assigned to two decomposition steps. Both exothermic
events are not completely separated due to the low reac-
tion rate. The appearance of two decomposition steps is in
agreement with known investigations of the thermal decom-
position of ammonia–borane [4–6]. As expected, no signs of
an endothermic melting process were obtained. This fact con-
firms earlier observations [6,7] that at sufficiently low heating
rates ammonia–borane is completely decomposed already
below the apparent melting point reported in the literature
[
2,3].
Both exothermic steps of the thermal decomposition of
ammonia–borane are accompanied by release of volatile

1
24
F. Baitalow et al. / Thermochimica Acta 445 (2006) 121–125
Fig. 3. Comparison between the temperature dependence of the heat flow
obtained at the thermal decomposition of ammonia–borane BH3NH3 at
Fig. 5. Comparison between the temperature dependence of the heat flow
obtained at the thermal decomposition of ammonia–borane BH3NH3 in the
2
50 bar (–) and 600 bar (– –) (transitiometer ST6-VI, mercury as pressure
DSC C-80 at 50 bar hydrogen (–) and in the transitiometer ST6-VI at 600 bar
−1
transmitter, heating rate 0.15 K min ).
−1
(
– –) (heating rate 0.15 K min ).
ST6-VI is nearly equal to the release of hydrogen, which was
obtained in [6] by gas-volumetric measurements in a sim-
ilar temperature range. Consequently, hydrogen should be
the prevailing volatile reaction product in experiments per-
formed in the transitiometer ST6-VI. The amount of further
volatile products seems to be negligible. This conclusion was
confirmed by the gravimetrically determined mass loss of the
solid residue, which was recovered from the reaction vessel
after the decomposition experiment in the transitiometer was
finished.
In order to prove the consistency of the results obtained
in the transitiometer ST6-VI at the monitoring of the thermal
decomposition of ammonia–borane, isobaric measurements
were performed at different pressures. As shown in Fig. 3, the
variation of pressure from 250 to 600 bar does not influence
significantly the detected heat flow data. The same conclu-
sion is given for the comparison of the amount of released
volatile products, obtained at 250 and 600 bar (Fig. 4). It
should be noted that the variation of volume, which is caused
by the release of volatile products, became smaller at greater
pressure, while the variation of volume caused by thermal
expansion of mercury remains similar. Nevertheless, a reli-
able monitoring of the release of gaseous products is still
possible at higher pressures due to the high resolution of vol-
umetric measurements in the transitiometer ST6-VI.
3.2. DSC C-80
In order to make an additional inspection of the per-
formance of the transitiometer ST6-VI for reaction mon-
itoring, we have measured the thermal decomposition
of ammonia–borane in a DSC C-80 under pressures up
to 100 bar. Heating rate, preparation and mass of the
ammonia–borane sample were the same, as in case of the
transitiometer ST6-VI.
Comparisons of experimental plots, which were obtained
in a typical measurement in the DSC C-80, with results
obtained for the decomposition of ammonia–borane in the
transitiometer ST6-VI under 600 bar were represented in
Figs. 5 and 6. The agreement between data detected in the
Fig. 4. Comparison between the temperature dependence of the amount
of released volatile products measured at the thermal decomposition of
ammonia–borane BH3NH3 at 250 bar (–) and 600 bar (– –) (transitiometer
Fig. 6. Comparison between the temperature dependence of the amount
of released volatile products obtained at the thermal decomposition of
ammonia–borane BH3NH3 in the DSC C-80 at 50 bar hydrogen (–) and
−
1
−1
ST6-VI, mercury as pressure transmitter, heating rate 0.15 K min ).
in the transitiometer ST6-VI at 600 bar (– –) (heating rate 0.15 K min ).

F. Baitalow et al. / Thermochimica Acta 445 (2006) 121–125
125
Table 1
4. Conclusions
Reaction enthalpy and final amount of released volatile products at the ther-
mal decomposition of ammonia–borane in the transitiometer ST6-VI and
DSC C-80 (heating rate 0.15 K min
The thermal decomposition of ammonia–borane BH3NH3
in the temperature range up to 453 K was monitored in
the high-pressure transitiometer ST6-VI under pressures up
to 600 bar. Calorimetric measurements show two partially
overlapping exothermic events assigned to two decompo-
sition steps. Both steps of the thermal decomposition are
accompanied by a stepwise release of volatile reaction prod-
ucts. Gas evolution is terminated near 430 K, the final
value amounts to approximately 2 mol gaseous products/mol
ammonia–borane. In principle, the used transitiometry tech-
nique allows the monitoring of the thermal behavior of the
released gaseous reaction products. But the formation of
further volatile products, beside hydrogen, seems to be neg-
ligible. Possibly, this is caused by the low heating rate of
−1
)
Pressure
ꢀRH
kJ mol 1)
mol product/mol
ammonia–borane
−
(
Transitiometer ST6-VI
6
00 bar mercury as pressure
transmitter
50 bar mercury as pressure
transmitter
50 bar mercury as pressure
−50
−65
−70
2.0
2.0
1.8
2
2
transmitter
DSC C-80
80 bar H2
60 bar H2
50 barH2
10 bar H2
1 bar H2
−70
−65
−55
−60
−60
2.1
2.2
2.3
2.2
2.2
−
1
0.15 K min only.
Experimental results obtained in the transitiometer ST6-
VI agree with general conclusions made from earlier investi-
gations of the thermal decomposition of ammonia–borane. In
order to perform a quantitative comparison between experi-
ments in different devices and under different pressures, the
thermal decomposition of ammonia–borane was additionally
investigated in a DSC C-80 in the pressure range 1–100 bar
and with the same ammonia–borane sample and at the same
heating rate of 0.15 K min , as in case of experiments in the
transitiometer ST6-VI. There is a good agreement between
the results obtained for the ammonia–borane decomposi-
tion in the transitiometer ST6-VI (under pressures 250 and
DSC C-80 and in the transitiometer is sufficiently high, both
in case of the reaction heat flow and the amount of released
gaseous products.
It should be noted that the technique used in the DSC C-80
is quite different from transitiometer technique. The amount
of released gaseous products was calculated from the vari-
ation of pressure in a reaction system with definite volume.
In order to compensate the low resolution of the pressure
sensor used in these experiments, we have chosen a reaction
system with a low internal volume and, consequently, with
a great pressure variation during the thermal decomposition.
The range of pressure variation was usually 5–20 bar.
Due to the fact that the pressure detector was placed out-
side the DSC C-80, the release of gaseous products, which
condenseatroomtemperature, doesnotcontributetothepres-
sure variation detected in the C-80 experiments. The above-
mentioned monomeric aminoborane BH2NH2 and borazine
−
1
600 bar) and in the DSC C-80 (under pressures 1–100 bar).
Thus, the good performance of the transitiometer ST6-VI was
proved using ammonia–borane as calibration substance. This
instrument is well suitable for the monitoring of solid–gas
reaction under high-pressure conditions. It enables a reliable
determination of the reaction heat and the amount of gas
release/gas uptake.
(
BHNH)3 are such products. Borazine is a volatile liquid
with a boiling point at 328 K. Monomeric aminoborane is
an unstable gas at room temperature and undergoes a rapid
oligomerization under formation of solid deposits [9]. From
this point of view, the obtained good agreement between the
final amount of gaseous products detected in the transitiome-
ter ST6-VI and in the DSC C-80 confirms the conclusion that
the thermal decomposition of ammonia–borane at a heating
rate as low as 0.15 K min is accompanied by the release
of negligible amount of further gaseous decomposition prod-
ucts, beside hydrogen.
References
[
[
1] S.L. Randzio, Chem. Soc. Rev. 25 (1996) 383.
2] M.G. Hu, R.A. Geanangel, W.W. Wendlandt, Thermochim. Acta 23
(
1978) 249.
[3] M.G. Hu, J.M. van Paaschen, R.A. Geanangel, J. Inorg. Nucl. Chem.
9 (1977) 2147.
−
1
3
[
4] F.P. Hoffmann, G. Wolf, L.D. Hansen, Advances in Boron Chemistry,
Royal Society of Chemistry, Cambridge, UK, 1997, p. 514.
[
5] G. Wolf, J. Baumann, F. Baitalow, F.P. Hoffmann, Thermochim. Acta
The values of the decomposition enthalpy and the final
amount of released volatile decomposition products, which
were obtained in different experiments in the transitiometer
ST6-VI (under pressures 250 and 600 bar) and in the DSC
C-80 (under initial pressures ranged from 1 to 80 bar) were
listed in Table 1. There is a good agreement between different
devices and pressures.
3
43 (2000) 19.
[6] F. Baitalow, J. Baumann, G. Wolf, K. Jaenicke-R o¨ ␤ler, G. Leitner,
Thermochim. Acta 391 (2002) 159.
7] J. Baumann, F. Baitalow, G. Wolf, Thermochim. Acta 430 (2005) 9.
8] S.L. Randzio, Ch. Stachowiak, J.-P.E. Grolier, J. Chem. Thermodyn.
[
[
35 (2003) 639.
[
9] M.C.L. Gerry, W. Lewis-Bevan, A.J. Merer, N.P.C. Westwood, J. Mol.
Spectrosc. 110 (1985) 153.