Disch et al.
Experimental Section
All of the reagents were used as purchased: magnesium powder,
NH4Cl (99%), and (NH4)2SO4 (99%) from Fisher Scientific, NH4F
(98%) from Acros, NH4Br (99%) from Alfa Aesar, and NH4NO3
(98%) from Sigma Aldrich. Argon and methane gases were
purchased in UHP quality from Air Liquide.
Magnesium sesquicarbide was prepared according to published
methods.6 Magnesium powder was placed in an inert atmosphere
in the hot zone of a tube furnace. By heating the sample to 685 °C
in a methane gas stream, Mg2C3 was obtained in a high purity of
94%, on the basis of the crystalline fraction of the sample as refined
from X-ray powder diffraction (XRPD) data. Although graphite
reflections were observed in XRPD, this phase was neglected during
refinement because of the strong preferred orientation effects, which
would otherwise lead to a significant overestimation of the graphite
content.7
Coupled TGA/sDTA/MS measurements were performed in a
Mettler STARe Thermogravimetric Analyzer, TGA/sDTA 851e,
coupled to a Balzers ThermoStar mass spectrometer. Roughly
stoichiometric amounts of the reactants were ground to a homo-
geneous mixture inside an MBraun glove box and filled into 70
µL alumina crucibles. As the TGA is not placed inside the glove
box, the contact of the sample with air during loading could not be
avoided but was kept as short as possible (<30 s). The sample
was kept at a constant temperature for 2 min in advance of each
measurement to purge the sample chamber with nitrogen gas. The
measurements were carried out in a temperature range of 25-
600 °C with a heating rate of 10 °C/min and nitrogen as purge gas
(50 mL/min).
Figure 1. (a) Thermoanalytic data of the protolysis of Mg2C3 with NH4F.
TGA and sDTA data are presented as black and red lines, respectively. (b)
Corresponding MS diagram, m/z values for C3H4 (green, m/z ) 38, 39,
40), water (blue, m/z ) 18), and ammonia (orange and red, m/z ) 16, 17)
are shown.
For ex situ investigations on the protolysis reaction with NH4F,
samples of a stoichiometric mixture of Mg2C3 and NH4F were filled
in alumina boats, covered with alumina lids, and heated for 2 h to
different temperatures between 100 and 300 °C in an inert
atmosphere. XRPD data of the resulting samples were collected
on a Huber G670 diffractometer using Cu KR1 radiation.
High-temperature synchrotron powder investigations on the
protolysis reaction with NH4Cl were performed at the powder
diffractometer of beamline B2 of the Hamburg synchrotron facility
(HASYLAB) using the following setup: λ ) 0.6875 Å; position
sensitive imaging plate detector system (OBI8); STOE capillary
furnace. The quartz capillary (0.3 mm) containing the reacting
sample was sealed with vacuum grease to maintain the inert
atmosphere inside while avoiding a high pressure due to the gas
evolution during the reaction. The measurements were performed
at temperatures between 150 and 390 °C in steps of 10 °C while
each measurement took about 18 min. All of the diffractograms
were analyzed using the STOE software package Win XPOW.9 The
Rietveld refinements were performed using the GSAS software
package.10
are easier to handle and dose and therefore provide a better
interaction between the reagents. As the protolysis reaction
is expected to start upon decomposition of the ammonium
salt, it could be realized in a relatively narrow temperature
range, making it easily observable with in situ methods such
as TGA/sDTA/MS measurements and temperature-dependent
X-ray powder diffraction (XRPD).
Protolysis with NH4F. The protolysis of Mg2C3 with
NH4F as a precursor for HF is expected to yield MgF2 as a
solid residue besides NH3 and the desired C3H4 in the gas
phase (3).
Mg2C3(s) + 4NH4F(s)
9
∆8 C3H4(g) +
4NH3(g) + 2MgF2(s) (3)
After verifying MgF2 as a solid product of this reaction
(heating a mixture of Mg2C3 and NH4F at 350 °C),
thermoanalytic methods were chosen to monitor the pro-
tolysis reaction in situ and verify its gaseous products. On
the basis of TGA and sDTA measurements, the protolysis
with NH4F can be described as a three-step reaction (part a
of Figure 1).
The TGA experiment shows a large weight loss occurring
in three steps at 130, 200, and 270 °C due to the decomposi-
tion of the ammonium salt and the evolution of NH3 and
C3H4. In the sDTA measurement, three distinct, endothermic
signals can be observed simultaneously. To identify the
gaseous products responsible for the weight loss, mass
spectrometry measurements were carried out simultaneously
with the TGA/sDTA experiments. As can be seen in part b
Results and Discussion
Because inorganic acids are mostly hazardous, less ac-
cessible, or difficult to get in a constant gas flow, their
respective ammonium salts were chosen as precursors. These
(7) See Supporting Information.
(8) (a) Knapp, M.; Joco, V.; Baehtz, C.; Brecht, H. H.; Berghaeuser, A.;
Ehrenberg, H.; von Seggern, H.; Fuess, H. Nucl. Instrum. Methods
Phys. Res., Sect. A 2004, 521, 565. (b) Knapp, M.; Baehtz, C.;
Ehrenberg, H.; Fuess, H. J. Synchrotron Radiat. 2004, 11, 328.
(9) Win XPOW, version 1.10 (03-Jun-2002); Stoe & Cie GmbH: Darm-
stadt, Germany.
(10) Larson, A. C.; von Dreele, R. B. Los Alamos Laboratory, Rep. No.
LA-UR 1987, 86, 748; revised PC version of December 2006.
970 Inorganic Chemistry, Vol. 47, No. 3, 2008