Application of n-dodecyltrimethylammonium chloride for the oxidation of intermetallic phases

Article abstract of DOI:10.1515/znb-2011-0304

The thermal decomposition products of ionic liquids based on n-dodecyltrimethylammonium chloride (DTAC) were used for the preparation of the metastable allotrope Ge(cF136) by oxidation of Na₁₂Ge₁₇ in gas-solid reactions. This method of preparation provides a promising low-temperature route for the synthesis of intermetallic phases and elemental modifications. In order to explore the reaction mechanism, we investigated the thermal decomposition of DTAC as well as of the ionic liquids DTAC/MgCl ₂ and DTAC/AlCl₃ by in-situ mass spectrometry and by powder X-ray diffraction. The results have revealed HCl, CH₃Cl and 1-chlorododecane to act as oxidizing agents in the gas-solid redox reactions.

Full text of DOI:10.1515/znb-2011-0304

Application of n-Dodecyltrimethylammonium Chloride
for the Oxidation of Intermetallic Phases
Bodo Bo¨hme, Stefan Hoffmann, Michael Baitinger, and Yuri Grin
Max-Planck-Institut fu¨r Chemische Physik fester Stoffe, No¨thnitzer Straße 40, 01187 Dresden,
Germany
Reprint requests to B. Bo¨hme. Fax: +49 351 46 46 40 02. E-mail: bodo.boehme@cpfs.mpg.de
Z. Naturforsch. 2011, 66b, 230 – 238; received November 6, 2010
The thermal decomposition products of ionic liquids based on n-dodecyltrimethylammonium chlo-
ride (DTAC) were used for the preparation of the metastable allotrope Ge(cF136) by oxidation of
Na12Ge17 in gas-solid reactions. This method of preparation provides a promising low-temperature
route for the synthesis of intermetallic phases and elemental modifications. In order to explore the re-
action mechanism, we investigated the thermal decomposition of DTAC as well as of the ionic liquids
DTAC/MgCl2 and DTAC/AlCl3 by in-situ mass spectrometry and by powder X-ray diffraction. The
results have revealed HCl, CH3Cl and 1-chlorododecane to act as oxidizing agents in the gas-solid
redox reactions.
Key words: DTAC, Ionic Liquid, Oxidation, Ge(cF136), Ge(hR8)
Introduction
and DTAC-based ionic liquids, and the gaseous prod-
ucts formed thereby. We explain how the decomposi-
tion products can be used for an optimized synthesis
of the allotrope Ge(cF136) via a gas-solid preparative
route.
Intermetallic phases are usually obtained from
the melt or by diffusion-controlled reactions in the
solid state at sufficiently high reaction temperatures.
However, phases with low temperature stability or
metastable phases are hardly accessible this way. A
more promising method for low-temperature syntheses
is the controlled oxidation of intermetallic precursors
by reactive agents. By this approach, a metastable ger-
manium allotrope Ge(cF136) with clathrate-II struc-
ture was obtained when heating the dispersion of
Na12Ge17 in the ionic liquid (IL) DTAC/AlCl3 [1]. The
metastable clathrate-II K8.6Ge136 was similarly pre-
pared from K4Ge9 [2]. However, in addition to these
crystalline products, considerable amounts of X-ray-
amorphous by-products were obtained. Recently, the
problem of X-ray-amorphous by-products in the prepa-
ration of Ge(cF136) has widely been overcome when
only the gaseous thermal decomposition products
of DTAC-based ILs were used for the oxidation of
Results and Discussion
Reaction of Na12Ge17 with the IL DTAC/AlCl3
The allotrope Ge(cF136) was originally obtained
by reacting a precursor of composition NaGe2.25 with
◦
the IL DTAC/AlCl3 at 300 C (Fig. 1a). The reaction
was found to comprise a Hofmann elimination of the
n-DTA+ (n-dodecyltrimethylammonium) cations. The
precursor was considered as a base for the abstraction
of the acidic β protons of n-DTA+, which are finally
reduced to H2 during the reaction [1]. As the precursor
consisted essentially of Na12Ge17 [5], the overall reac-
tion to Ge(cF136) (i. e. ꢀ24Ge136) can be written as:
8Na12Ge17 + 96[C12H25N(CH3)3]Cl −→
ꢀ24Ge136 + 96NaCl+ 48H2
(1)
◦
Na12Ge17 at 280 C [3]. The resulting Ge(cF136)
product was particularly suited as a starting material
for the high-pressure transformation into the allotrope
+96(C10H21)CH = CH2 + 96N(CH3)3
The liquid-solid reaction yields crystalline
Ge(hR8) [4]. However, the improved preparative route Ge(cF136) and α-Ge as a by-product. Moreover,
to Ge(cF136) and the nature of the gaseous oxidiz- the presence of X-ray-amorphous phases is evidenced
ing agents have not been reported yet. In this work we by a significant background in the powder X-ray
have investigated the thermal decomposition of DTAC diffraction patterns (Fig. 2a).
c
0932–0776 / 11 / 0300–0230 $ 06.00 ꢀ 2011 Verlag der Zeitschrift fu¨r Naturforschung, Tu¨bingen · http://znaturforsch.com
Unauthenticated
Download Date | 5/9/19 1:02 PM

B. Bo¨hme et al. · Oxidation of Intermetallic Phases
231
gaseous oxidizing agents. The powder X-ray diffrac-
tion pattern of a Ge(cF136) product obtained under
optimized conditions using DTAC/MgCl2 (2 : 1 molar
ratio, 280 ◦C, 300 h) is shown in Fig. 2b. Because NaCl
was found in all products of the gas-solid reactions, the
gas phase must have contained chlorides which acted
as oxidizing agents. NaCl was easily separated from
Ge(cF136) by washing with water.
Formation of DTAC-based ionic liquids as sources of
oxidizing agents
Fig. 1. Setup for the oxidation of intermetallic precursors
(a) in a liquid-solid reaction with the IL DTAC/AlCl3 (1 : 1
molar ratio) and (b) in a gas-solid reaction with the gaseous
decomposition products of the IL.
Upon mixing DTAC with AlCl3, a clear and slightly
yellowish melt was formed. A slight temperature in-
crease of the mixture indicated an exothermic reaction.
Upon cooling to r. t., the melt solidified gradually. The
melting point of the 1 : 1 molar mixture was estimated
to be 50 ◦C. The melt of DTAC/AlCl3 contains quater-
nary ammonium cations n-DTA+ and [AlCl4]− anions
in equilibrium with [Al2Cl7]− and Cl− (Eqs. 2, 3) [7].
DTAC+ AlCl3 −→ n-DTA+[AlCl4]−
(2)
−AlCl
3
−
−
− −−−−→
[Al2Cl7]
[AlCl4] ꢁ AlCl3 + Cl
(3)
←−−−−
+AlCl
Fig. 2. Powder X-ray diffraction patterns of Ge(cF136) prod-
ucts obtained by oxidation of Na12Ge17 after washing in
(a) the liquid-solid reaction with the IL DTAC/AlCl3 (1 : 1
molar mixture) [1]; and in (b) the gas-solid reaction with
gaseous decomposition products of DTAC/MgCl2 (2:1 mo-
lar mixture) [3, 4]. Asterisks mark reflection of α-Ge.
3
The 2 : 1 molar mixture of DTAC/MgCl2 was found
to have a di◦stinctly higher melting point of approxi-
mately 150 C. A colorless and clear IL was formed
above this temperature. In the IL DTAC/MgCl2, chloro
complexes [MgCl4]2− or [MgCl3]− should exist as
well, but their stability is expected to be lower than
that of [AlCl4]− due to the reduced Lewis acidity of
Mg2+ compared with Al3+. Therefore, the Cl− con-
centration in the melt should be higher in the neu-
tral IL DTAC/MgCl2 (2 : 1) than in the neutral IL
DTAC/AlCl3 (1 : 1). Among the anions in the melt,
Cl− is both the most nucleophilic and the most basic
species, which is important for the decomposition re-
actions discussed below.
The pure IL DTAC/AlCl3 (1 : 1 molar mixture) de-
veloped a significant vapor pressure at elevated tem-
peratures due to its thermal decomposition. This led to
the question, whether the gaseous decomposition prod-
ucts of the IL contribute to the oxidation of Na12Ge17
to Ge(cF136). In order to react the precursor Na12Ge17
only with the gaseous decomposition products of the
IL, a setup was chosen in which the IL and the pre-
cursor are arranged in a spatially separated fashion
(Fig. 1b)∗. A gas-solid reaction of the gaseous de-
composition products of the IL with Na12Ge17 yielded
Ge(cF136) as in the liquid-solid route, but the amount
of X-ray-amorphous by-products was significantly re-
duced. The reaction could also be performed with
The thermal decomposition of the IL DTAC/AlCl3
investigated by in-situ mass spectrometry
Thermal decomposition of quaternary ammonium
DTAC/MgCl2 and with pure DTAC as sources for ions has been investigated for a variety of salts and
solvents [8–13]. In the case of DTAC, thermal decom-
∗
A similar setup has already successfully been used for the ox-
position has been studied in organic solvents [14] and
idation of the silicides Na4Si4, K4Si4 and Na2BaSi4 with gaseous
at the surface of clay minerals [15]. From these results,
HCl to clathrate-I phases Na6.2Si46, K7.0Si46, and Na2Ba6Si46 [6].
different reaction paths can be expected to occur for the
Na12Ge17 was obtained by heating the elements in sealed Ta am-
poules to 1100 ◦C [3].
n-DTA+ ions upon heating a DTAC-based system. The
Unauthenticated
Download Date | 5/9/19 1:02 PM

232
B. Bo¨hme et al. · Oxidation of Intermetallic Phases
Fig. 3. Reaction paths of the Cl−-induced thermal decomposition of n-DTA+ ions: (a), (b) nucleophilic substitutions with
SN2 mechanism and (c) Hofmann elimination with E2 mechanism.
decomposition is initiated by the attack of Cl− anions In our case the unambiguous identification of CH3Cl
at different positions of the n-DTA+ cations (Fig. 3).
(Fig. 3a) was successful because no significant sig-
A nucleophilic attack (SN2 mechanism) of Cl− on nal overlap at m/z = 50/52 occurred. For N(CH3)3,
an α-C atom of the N-methyl groups leads to the which is an expected product of the mechanisms in
formation of CH3Cl and N,N-dimethyldodecylamine Figs. 3b and c, the direct assignment of the molec-
(Fig. 3a), while an attack on the α-C atom of the ular ion peak (m/z = 59) was not possible because
dodecyl group yields 1-chlorododecane and N(CH3)3 the radical cation [N(CH3)3]+ is also a characteris-
(Fig. 3b). If the Cl− anion acts as a base, it favorably tic fragment of N,N-dimethyldodecylamine. In the ex-
attacks the acidic protons in β position of the dode- perimental spectra an almost time-constant ratio of
cyl chain. This initiates a Hofmann elimination (E2 I(59)/I(213) ≈ 4 was detected. However, the fragmen-
mechanism), yielding 1-dodecene, N(CH3)3 and HCl tation of N,N-dimethyldodecylamine should only lead
(Fig. 3c).
to an intensity ratio of I(59)/I(213) = 1.85 [16]. The
observed distinctly higher value indicates the presence
of N(CH3)3 in the gas mixture. The detection of HCl,
which is an expected product of the Hofmann elimina-
tion (Fig. 3c), is hampered by the high Ar content in
the experimental setup. Even the argon isotopes 36Ar
and 38Ar with low natural abundance produced strong
ion peaks.
Upon heating the◦ IL DTAC/AlCl3 at the constant
temperature of 280 C, characteristic products for all
three reaction paths were found in the gas phase by
TG-MS (Fig. 4). The decomposition started already
◦
at ≈ 160 C as indicated by a significant mass loss
and by the detection of gaseous decomposition prod-
◦
ucts (Fig. 4). At 280 C, the following decomposi-
tion products were directly identified by their molec-
The markedly higher intensity of the CH3Cl molec-
ular ion peaks: CH3Cl (main contribution for the ular ion peaks compared to the most intense peaks of
signal m/z = 50/52) and N,N-dimethyldodecylamine 1-chlorododecane confirms the expectation [8–10] that
(m/z = 213), indicative of the mechanism in Fig. 3a, the nucleophilic substitution by the Cl− ions proceeds
1-chlorododecane (m/z = 204/206), indicative of the on the more electrophilic N-methyl α-C atoms of n-
mechanism in Fig. 3b, and 1-dodecene (main con- DTA+ (Fig. 3a), rather than on the α-C atoms of the
tribution for the signal m/z = 168), indicative of dodecyl chain (Fig. 3b). However, a quantitative inter-
the mechanism in Fig. 3c. Characteristic fragment- pretation of the signal intensities in the mass spectra
ion peaks were identified as well. In the case of 1- would require detailed knowledge about volatility and
chlorododecane, the molecular ion peak became very ion yield of the different compounds [12].
small in intensity with progressing reaction time, but
characteristic fragment-ion peaks (main contributions
Signals at m/z = 17/18 observed during the heati◦ng
process disappeared when the temperature of 280 C
of the signals m/z = 105/107 and m/z = 91/93) re- was reached. We assign these peaks to layers of water
mained visible during the whole experiment. For the adsorbed on the apparatus.
other compounds mentioned above the molecular-ion
Thermal decomposition in closed ampoules
peaks also remained clearly visible.
The identification of molecules with low molecular
The preparation of Ge(cF136) by oxidation of
mass is often hampered by the overlap of the respec- Na12Ge17 with DTAC-based ILs was performed in
tive molecular ion peaks with signals of fragment ions. sealed glass ampoules (Fig. 1b) [3,4]. Temperature and
Unauthenticated
Download Date | 5/9/19 1:02 PM

B. Bo¨hme et al. · Oxidation of Intermetallic Phases
233
Fig. 4. Time-resolved combined thermogravimetric and mass spectrometric investigation of the decomposition of the IL
DTAC/AlCl3 at 280 ◦C: sample mass normalized to the initial mass of the specimen (top), sample temperature (middle), mass
spectra with ion current expressed in grey scale (bottom). Reference mass spectra [16] of 1-chlorododecane, 1-dodecene,
N,N-dimethyldodecylamine, CH3Cl and N(CH3)3 are plotted on the right.
Unauthenticated
Download Date | 5/9/19 1:02 PM

234
B. Bo¨hme et al. · Oxidation of Intermetallic Phases
Table 1. Thermal decomposition of DTAC-based systems at 280 ◦C in sealed ampoules.
Concluded major
decomposition
Relative
decomposition
rate
Crystalline compounds identified in
the deposit at the top of the
ampoules
Assigned gaseous products of the
System
thermal decomposition of n-DTA+ a mechanism of
n-DTA+
CH3Cl, 1-chlorododecane,
Fig. 3a, b
DTAC
high
N(CH3)4Cl, DTAC
N(CH3)3, N,N-dimethyldodecylamine
HCl, N(CH3)3, possibly
Fig. 3c, a
DTAC/ MgCl2 (2 : 1)
middle
low
HN(CH3)3Cl, N(CH3)4Cl, NH4Cl
NH4Cl
CH3Cl, [HCl, CH3Cl, NH3]
DTAC/ AlCl3 (1 : 1)
a
HCl, N(CH3)3, [HCl, CH3Cl, NH3] Fig. 3c
Oxidizing agents for intermetallic phases are printed bold. The products of secondary reactions during the decomposition are listed in
brackets.
Table 2. Experimental findings during thermal decomposition of pure DTAC in sealed ampoules.
T (◦C) t (h) Observations
Viscous colorless liquid and a colorless solid in the reaction crucible.
Small amounts of a colorless solid at the top of the ampoule.
24
185
Only a colorless solid residue remained in the reaction crucible.
The amount of colorless deposit at the top of the ampoule had increased.
140
Colorless smoke immediately rose from the reaction crucible.
0.1 Reflux of a colorless liquid.
Deposition of a colorless solid starts at the top of the ampoule.
280
Only a small amount of a colorless oily liquid remained in the reaction crucible.
Colorless solid at the top of the ampoule containing crystalline N(CH3)4Cl and DTAC. The decomposition products had an
amine-like odor.
1
The solid is partially soluble in water forming a basic solution.
Table 3. Experimental findings during thermal decomposition of DTAC/MgCl2 in sealed ampoules.
T (◦C) t (h) Observations
0.1 Colorless and clear liquid in the reaction crucible.
Colorless solid precipitated from the liquid in the reaction crucible.
Droplets of a colorless liquid started to form at the top of the ampoule, the amount of which increased with time.
185
0.5
140 Besides droplets, a colorless solid had formed at the top of the ampoule.
Colorless melt in the reaction crucible transforming into a solid within few minutes.
A colorless liquid started to reflux at the top of the ampoule.
0.1
24
A colorless solid started to deposit at the top of the ampoule.
280
Strong reflux of colorless liquid.
The amount of solid at the top of the ampoule had increased, containing crystalline HN(CH3)3Cl, and smaller amounts of
170 NH4Cl and N(CH3)4Cl.
A solid remained inside the reaction crucible containing unidentified crystalline phases. A solution in water was weakly
basic.
pressure conditions in this setup differ from those in sition products also allows to draw conclusions on
a Knudsen-like arrangement in TG-MS, which might the constituents of the gas phase and hence, on the
influence the decomposition behavior of the n-DTA+ decomposition mechanisms of the n-DTA+ cations
cations. Hence, we studied the thermal decomposition (Table 1).
When heated up to 185 ◦C, the IL samples were en-
of DTAC/AlCl3, DTAC/MgCl2, and pure DTAC al◦so
in sealed ampoules at temperatures of T = 185 C
and T = 280 C typically used for the oxidation re-
tirely liquid. Pure DTAC, which has been reported to
decompose without melting at 237 C [19], also par-
◦
◦
actions. Inside the sealed ampoules, the constituents
of the gas phase could not be investigated in situ.
However, the ex-situ identification of the decompo-
tially reacted to a liquid phase within 24 h. For all sys-
tems, decomposition products condensed at the top of
the reaction ampoules on a time scale of several hours
Unauthenticated
Download Date | 5/9/19 1:02 PM

B. Bo¨hme et al. · Oxidation of Intermetallic Phases
235
Table 4. Experimental findings during thermal decomposition of DTAC/AlCl3 in sealed ampoules.
T (◦C) t (h) Observations
0.1 Yellowish clear liquid in the reaction crucible turning to orange after several minutes. Gas bubbles were visible.
185
1
The liquid in the reaction crucible was dark red. The color persisted also after 24 h.
The liquid in the reaction crucible had turned to yellow.
Colorless droplets had condensed at the top of the ampoule.
140
0.1 The initially yellowish melt turned to dark red within few minutes, and gas bubbles were observed.
Reflux of colorless liquid increasing significantly within the next hours. The liquid in the reaction crucible had turned to
yellow.
After cooling down to r. t., the reflux collected as a colorless liquid outside the reaction crucible.
The liquid inside the reaction crucible solidified to an X-ray amorphous gel (Fig. 5a, b).
1
The reflux collected outside of the reaction crucible was yellow and became brown after longer reaction times.
The liquid phase inside the reaction crucible darkened as well and solidified upon cooling. The residue showed reflections of
an unknown phase in PXRD (Fig. 5c).
72
280
The amount of the liquid in the reaction crucible was decreased markedly.
A colorless solid started to deposit at the top of the ampoule.
140
170
On cooling two immiscible liquids collected outside of the reaction crucible. The phase with the higher density was highly
viscous and dark brown, the one with lower density was yellow and clear.
Only a small amount of a black solid residue remained in the reaction crucible. In PXRD it showed the same reflections as the
300 product in the reaction crucible after 72 h, but low in intensity. By hydrolysis with air moisture, AlCl3 ·6 H2O was formed.
The colorless solid at the top of the ampoule contained NH4Cl and unknown crystalline phases.
to days (Tables 2 – 4). In case of DTAC/AlCl3, distinct
color changes were also observed.
◦
At 280 C, the decomposition rates strongly in-
creased: For all systems, reflux of a liquid at the
top of the ampoules was observed within a few min-
utes, along with rapid color changes in the case
of DTAC/AlCl3. PXRD patterns of the cooled-down
residues inside the reaction crucible differed from that
of the solidified IL (Fig. 5). For pure DTAC, the sub-
stance almost completely disappeared from the reac-
tion crucible within 1 h.
The gaseous decomposition products of n-DTA+
were assigned on the basis of the identified crys-
talline deposition products at the top of the am-
poules (Table 1). In the case of pure DTAC, the de-
posit contained N(CH3)4Cl and DTAC. The presence
of N(CH3)4Cl proves N(CH3)3 and CH3Cl as con-
stituents of the gas phase. CH3Cl should be formed
in the nucleophilic substitution reaction according
to Fig. 3a, N(CH3)3 in the nucleophilic substitu-
tion according to Fig. 3b. Hence, the gas phase
should also contain N,N-dimethyldodecylamine and
Fig. 5. PXRD patterns of the solidified ionic liquid
DTAC/AlCl3 (a) and of the resi◦dues in the reaction crucible
after thermal treatment at 280 C for 1 h (b) and 72 h (c).
The patterns (b) and (c) possibly show the same product with
different crystallinity.
nucleophilic substitution mechanisms is expected for
the nucleophilic Cl− anions in pure DTAC [8–13].
For the decomposition of MgCl2/DTAC, the deposit
1-chlorododecane. The presence of DTAC in the de- contained HN(CH3)3Cl as the main phase, as well as
posit consequently results from the reverse reactions. NH4Cl and N(CH3)4Cl. The presence of HN(CH3)3Cl
The presence of N(CH3)3 might also point to a Hof- proves HCl and N(CH3)3 in the gas phase, which
mann elimination of the n-DTA+ cations (Fig. 3c). points to Hofmann elimination (Fig. 3c) as the domi-
However, in this case N(CH3)3 and HCl should form nating decomposition mechanism of n-DTA+. The oc-
HN(CH3)3Cl, which was not detected in the deposit. currence of NH4Cl and N(CH3)4Cl can be attributed
The observed decomposition of n-DTA+ cations via to the decomposition of HN(CH3)3Cl. Thus, due to
Unauthenticated
Download Date | 5/9/19 1:02 PM

236
B. Bo¨hme et al. · Oxidation of Intermetallic Phases
Fig. 6. Reaction schemes for the oxidation of intermetallic precursors with different oxidizing agents: (a) and (b) Wurtz-
analogous and competing reactions of the organic chlorides, and (c) reaction of HCl under formation of H2.
secondary processes, the gas phase should addition- above. Because any deposit at the top of the ampoule
ally contain CH3Cl, di- and monomethylamine [17], was only observed after 140 h of heat treatment, we
and NH3. N(CH3)4Cl might also be formed by the re- conclude that the evaporation of Al complexes starts
action of N(CH3)3 with CH3Cl, which originates from only after longer heat treatment.
the decomposition of n-DTA+ according to the nucle-
ophilic substitution mechanism depicted in Fig. 3a.
For the system DTAC/AlCl3, the deposit at the
top of the ampoule was a multiphase product, in
The reaction of gaseous decomposition products of
DTAC with intermetallic phases
The gaseous decomposition products of n-DTA+
cations in DTAC or DTAC-based ILs explain the ox-
idation of intermetallic precursors such as Na12Ge17
in an experimental setup according to Fig. 1b. Upon
heating, the oxidizing agents HCl, CH3Cl and 1-
chlorododecane pass into the gas phase (Table 1). In
the course of the redox reactions, protons of HCl are
reduced to H2, and the organic chlorides can react
in Wurtz-analogous reactions [8] or competing pro-
cesses (Fig. 6a – c). The precursor is oxidized to the
clathrate phase, and NaCl is formed as co-product.
Well-crystallized Ge(cF136) was favorably obtained
in a reaction with the IL DTAC/MgCl2 (2 : 1 molar
which only NH4Cl could be identified. The presence
of NH4Cl might point to a Hofmann elimination of
the n-DTA+ cations to HCl, N(CH3)3 and 1-dodecene
(Fig. 3c). In a secondary process, HCl and N(CH3)3
should form HN(CH3)3Cl, which finally decomposes
to NH4Cl [17]. However, the presence of AlCl3 in the
system complicates any conclusion on the decomposi-
tion mechanisms of n-DTA+. AlCl3 is known to cat-
alyze proton- or methyl-transfer reactions, so that sub-
sequent reactions of the original decomposition prod-
ucts of n-DTA+ are enhanced. This might lead to an
absence of the above-mentioned methylated ammo-
nium chlorides expected to deposit (Fig. 3). Also, the
occurrence of a black tar-like residue in the reaction
crucible after long reaction times can be explained
by the catalytic effect of AlCl3 in isomerization and
polymerization reactions. The fragmentation of the n-
DTA+ cation and its decomposition products should
lead to the formation of different amine complexes
with AlCl3. Such Lewis acid-base complexes should
be present both in the residues inside the reaction cru-
cible and, if volatile [18], in the deposit at the top of
the ampoule.
◦
ratio) as gas source at 280 C after 12 d. By us-
ing DTAC/AlCl3 under the same conditions, the re-
action was incomplete [3]. Such differences cannot
be explained by the kind of oxidizing agents formed
from the respective DTAC-based system (Table 1),
since they are all highly reactive towards Na12Ge17.
In fact, the reaction rates are influenced by the to-
tal in-situ pressure of the oxidizing agents. Under the
same conditions, the in-situ pressure of the oxidiz-
ing agents is determined by the decomposition rate
of n-DTA+, which decreases in the order r(DTAC) >
r(DTAC/MgCl2) > r(DTAC/AlCl3). This result can
It is remarkable that no hints for AlCl3 were found
in the gas phase in the TG-MS experiment described
Unauthenticated
Download Date | 5/9/19 1:02 PM

B. Bo¨hme et al. · Oxidation of Intermetallic Phases
237
be attributed to the concentration of free Cl− anions In-situ investigation of the thermal decomposition of
since all decomposition paths of n-DTA+ follow sec-
DTAC/AlCl by TG-MS
3
ond order mechanisms (Fig. 3). The decomposition
For an in-situ investigation by mass spectrometry the mix-
rate of n-DTA+ in the ILs is lower than the one of
pure DTAC due to chloro complexes formed in the
IL, diminishing the concentration of free Cl− ions.
Comparing Lewis acidity, chloro complexes should be
more stable with AlCl3 than with MgCl2, a fact which
explains why DTAC/AlCl3 decomposes more slowly
than DTAC/MgCl2.
ture of the components for the IL DTAC/AlCl3 was homog-
◦
enized at T = 100 C under an argon atmosphere. Subse-
quently, the IL (m = 11.17 mg) was transferred with a dry
glass pipette to an Al2O3 crucible (d = 5 mm, V = 85 mm3).
The crucible was covered by a cap with a small hole in or-
der to resemble a Knudsen arrangement. The crucible was
transferred under Ar to the TG-MS setup which was oper-
Other experimental parameters such as reactor vol- ated in an Ar-filled glove box. The measurement was carried
out under a constant He stream (50 mL min−1). The reaction
ume and temperature gradients can also influence the
cell was heated in the oven of a thermobalance (Netzsch STA
409 CD) at a rate of 10 K min−1 up to 280 ◦C and kept at this
temperature. The mass loss during the decomposition reac-
tion was recorded. During the whole heat treatment, the gas
phase was continuously directed to the ionization chamber
of a quadrupole mass spectrometer (Pfeiffer Quadstar QMS
422, electron impact ionization at 75 eV) by a skimmer lo-
cated directly above the hole of the reaction cell. In order to
remove the strong peaks of the purging gas, the mass spectra
were corrected by subtracting a spectrum initially recorded at
r. t. Nonetheless, Arn+ and He2+ signals were detected dur-
actual concentration of oxidizing gases. This is ob-
viously the case, when decomposition products of
DTAC deposit from the gas phase at the colder
top of the reaction ampoule. Different kinds of de-
posits become especially important for longer reaction
times.
Conclusion
The oxidation of Na12Ge17 to Ge(cF136) by the use
of DTAC-based ionic liquids in gas-solid reactions is ing the whole measurement.
explained: The oxidizing agents HCl, CH3Cl and 1-
chlorododecane are formed by Cl−-induced thermal
Decomposition experiments in sealed ampoules
decomposition of n-DTA+ cations via different reac-
tion paths. The specific decomposition behavior of a
DTAC-based system has a significant effect on the oxi-
dation process. The knowledge of the thermal stability
of the DTAC-based systems and their gaseous decom-
position products allowed an optimized preparation of
the allotrope Ge(cF136). The direct use of the identi-
fied gaseous oxidizing agents will facilitate the appli-
cation of this preparative method to other intermetallic
compounds.
The decomposition of DTAC was investigated on samples
of 0.6 g (2.3 mmol); for the preparation of the ILs the re-
spective amounts of 0.303 g AlCl3 (2.3 mmol) or 0.108 g
MgCl2 (1.15 mmol) were added to 0.6 g (2.3 mmol) of
DTAC. Thermal treatment was performed in glass ampoules
(V ≈ 50 cm3, di = 26 mm, l ≈ 90 mm) sealed under Ar.
The ampoules were placed vertically in a block thermostat
◦
(Gebr. Liebisch GmbH, Bielefeld, Germany) kept at 185 C
or 280 ◦C. In this arrangement, the top of the ampoules had a
lower temperature than the bottom. The same setup was used
for the oxidation of intermetallic compounds (Fig. 1). To ter-
minate an experiment, the ampoule was removed from the
thermostat and cooled in air to r. t. The products were further
investigated under Ar atmosphere. The experimental obser-
vations are summarized in Tables 2 – 4.
Experimental Section
Preparation of the ionic liquids
Before usage, the DTA◦C (Merck) was dried for 2 d under
vacuum conditions at 25 C, until the pressure dropped be-
low 5×10−3 mbar. A mass loss of up to 4 % was observed.
AlCl3 (Merck, anhydrous powder, sublimed) was sublimed
at 160 ◦C under Ar in a sealed glass ampoule. MgCl2 (Fluka,
anhydrous purum, pure according to PXRD) was used as pur-
chased. For the preparation of the ionic liquids, the respective
Caution: During the reaction, high pressure develops in
the ampoules and appropriate safety precautions must be
taken.
Powder X-ray diffraction
Diffraction experiments on powdered samples of the de-
components were mixed in an Ar-filled glove-box in glass composition products of DTAC and the ILs were performed
crucibles (Duran). The IL DTAC/AlCl3 was prepared at a in air using a Huber G670 image plate Guinier camera
˚
molar ratio of 1 : 1, the IL DTAC/MgCl2 at 2 : 1. Both com- (transmission setup, CuKα1 radiation, λ = 1.540598 A, Ge
positions correspond to neutral ILs [7].
(111) monochromator, 5◦ ≤ 2θ ≤ 100◦, ∆2θ = 0.005◦).
Unauthenticated
Download Date | 5/9/19 1:02 PM

238
B. Bo¨hme et al. · Oxidation of Intermetallic Phases
To prevent hydrolysis, the samples were fixed under Ar Acknowledgements
atmosphere between two polyimide foils (7.5 µm, Kap-
ton, Chemplex) on the sample holder. Data for an opti-
mized evaluation of the background intensity originating
from X-ray amorphous constituents were collected with
a Stoe STADI P diffractometer using a zero-background
sample holder (Bragg-Brentano geometry, CuKα1 radiation,
We sincerely thank S. Scharsach for performing the
TG-MS measurement, the Kompetenzgruppe Struktur for
PXRD patterns, and Dr. J. Hunger for the discussion of the
manuscript. B. B. and Y. G. gratefully acknowledge funding
by the European Union and the Free State of Saxony (SAB
project 13853/2379), M. B. and Y. G. gratefully acknowledge
financial support by the Deutsche Forschungsgemeinschaft
(SPP 1415, “Kristalline Nichtgleichgewichtsphasen (KNG)
– Pra¨paration, Charakterisierung und in-situ-Untersuchung
der Bildungsmechanismen“).
◦
˚
λ = 1.540598 A, Ge (111) monochromator, 5 ≤ 2θ ≤ 120 ,
∆2θ = 0.02◦).
[12] M. Hesse, H. Meier, B. Zeeh, Spektroskopische Metho-
den in der organischen Chemie, Georg Thieme Verlag,
Stuttgart, 2005.
[13] M. Hesse, W. Vetter, H. Schmid, Helv. Chim. Acta
1965, 48, 674 – 689; M. Hesse, Fortschr. Chem. Forsch.
1967, 8, 608 – 651; R. E. Verrall, J. A. Burns, Can. J.
Chem. 1974, 52, 3438 – 3443; H. J. Veith, Mass Spec-
trom. Rev. 1983, 2, 419 – 446.
[1] A. M. Guloy, R. Ramlau, Z. Tang, W. Schnelle,
M. Baitinger, Y. Grin, Nature 2006, 443, 320 – 323.
[2] A. M. Guloy, Z. Tang, R. Ramlau, B. Bo¨hme,
M. Baitinger, Y. Grin, Eur. J. Inorg. Chem. 2009, 2455 –
2458.
[3] B. Bo¨hme, Neue Pra¨parationswege fu¨r intermetallische
Verbindungen, Logos, Berlin, 2010.
[4] U. Schwarz, A. Wosylus, B. Bo¨hme, M. Baitinger,
M. Hanfland, Y. Grin, Angew. Chem. 2008, 120, 6895 –
6898; Angew. Chem. Int. Ed. 2008, 47, 6790 – 6793.
[14] S. Takano, C. Takasaki, K. Kunihiro, M. Yamanaka,
J. Am. Oil. Chem. Soc. 1977, 54, 139 – 143.
¨
[15] W. Xie, Z. Gao, W. -P. Pan, D. Hunter, A. Singh,
R. Vaia, Chem. Mater. 2001, 13, 2979 – 2990; W. Xie,
Z. Gao, K. Liu, W.-P. Pan, R. Vaia, D. Hunter, A. Singh,
Thermochim. Acta 2001, 367 – 368; ibid. 339 – 350.
[16] S. E. Stein in NIST Chemistry WebBook, NIST Stan-
dard Reference Database Number 69 (Eds.: P. J. Lin-
strom, W. G. Mallard), National Institute of Stan-
dards and Technology, Gaithersburg MD, 2010,
http://webbook.nist.gov; N. Wasada in SDBSWeb, Na-
tional Institute of Advanced Industrial Science and
Technology 2009, http://riodb01.ibase.aist.go.jp/sdbs/.
[17] C. Vincent, C.R. Hebd. Acad. Sci. 1877, T84, 1139 –
1141; C. Vincent, J. Chem. Soc. Abstr. 1878, 34, 400 –
4011; C. Vincent, C.R. Hebd. Acad. Sci. 1877, T85,
667 – 671; C. Vincent, J. Chem. Soc. Abstr. 1878, 34,
25.
[5] W. Carrillo-Cabrera, R. Cardoso Gil, M. Somer, O. Per-
sil, H. G. von Schnering, Z. Anorg. Allg. Chem. 2003,
629, 601 – 608.
[6] B. Bo¨hme, A. Guloy, Z. Tang, W. Schnelle,
U. Burkhardt, M. Baitinger, Y. Grin, J. Am. Chem.
Soc. 2007, 129, 5348 – 5349; B. Bo¨hme, U. Aydemir,
A. Ormeci, W. Schnelle, M. Baitinger, Y. Grin, Sci.
Technol. Adv. Mater. 2007, 8, 410 – 415.
[7] C. M. Gordon, M. J. Muldoon in Ionic Liquids in Syn-
thesis (Eds.: P. Wasserscheid, T. Welton), Wiley-VCH,
Weinheim, 2003, chapter 2.1, p. 14.
[8] H. R. Christen, F. Vo¨gtle, Organische Chemie - Von den
Grundlagen zur Forschung, Band I, Otto Salle, Frank-
furt am Main; Sauerla¨nder, Aarau, 1992.
[9] A. Cope, E. R. Trumbull in Organic Reactions, Vol. 11
(Ed.-in-chief: A Cope), John Wiley&Sons Inc., New
York, London, 1960.
[18] H. Jacobs, B. No¨cker, Z. Anorg. Allg. Chem. 1993, 619,
73 – 76; H. Jacobs, B. No¨cker, Z. Anorg. Allg. Chem.
1992, 614, 25 – 29.
[10] E. D. Hughes, C. K. Ingold, C. S. Patel, J. Chem. Soc.
1933, 526 – 530.
[19] DTAC Safety Data Sheet, Merck KGaA, Darmstadt,
2004.
[11] W. Hanhart, C. K. Ingold, J. Chem. Soc. 1927, 997 –
1020.
Unauthenticated
Download Date | 5/9/19 1:02 PM