The crystal structures of the room temperature and the low temperature phase of dimethylammonium trifluoromethanesulfonate

Article abstract of DOI:10.1002/zaac.200700138

Dimethylammonium trifluoromethanesulfonate 1 was synthesized by reaction of trifluoromethanesulfonic acid with an excess of dimethylamine. A temperature variable synchrotron measurement on the polycrystalline substance reveals that 1 passes through a phase transition below room temperature. The transition occurs in the temperature range of 282-285 K on heating and 272-280 K on cooling as determined by DSC. The room temperature phase crystallizes in space group Cmca (a = 11.031(6) A, b = 18.466(14) A, c = 8.173(9) A, V = 1665(2) A³, Z = 8) and the low temperature phase in space group P 2₁/c (a = 8.8717(18) A, b = 8.0838(16) A, c = 10.968(2) A, β = 92.128(4)°, V = 786.0(3) A³, Z = 4). The structures of both phases were determined by single crystal X-ray diffraction, but refinement did not yield satisfactory residuals for the low temperature phase because of twinning of the crystal. It was, therefore, independently solved from the synchrotron powder diffraction data using rigid body models of the constituent ions and ab-initio direct space methods. Both, the CF₃ group and the SO₃ group of the triflate ion, are rotationally disordered around the S-C bond, in the room temperature phase. In the lowtemperature phase, the triflate ion is well localized. Like in the alkali metal triflates, the triflate ions are arranged in double layers with the hydrophobic trifluoromethyl groups and the sulfonate groups, respectively, pointing towards each other. The dimethylammonium ion is located closer to the sulfonate group with contacts indicating hydrogen bonding. The packing in both phases is of the topological CsCl structure type.

Full text of DOI:10.1002/zaac.200700138

DOI: 10.1002/zaac.200700138
The Crystal Structures of the Room Temperature and the Low Temperature
Phase of Dimethylammonium Trifluoromethanesulfonate
Pascal D. C. Dietzela, Robert Dinnebierb, and Martin Jansenb,*
a Oslo/Norway, SINTEF Materials and Chemistry and University, Centre of Materials Science and Nanotechnology and Department of
Chemistry
b Stuttgart/Germany, Max-Planck-Institut für Festkörperforschung
Received March 22nd, 2007.
Abstract. Dimethylammonium trifluoromethanesulfonate 1 was
synthesized by reaction of trifluoromethanesulfonic acid with an
excess of dimethylamine. A temperature variable synchrotron me-
asurement on the polycrystalline substance reveals that 1 passes
through a phase transition below room temperature. The transition
occurs in the temperature range of 282Ϫ285 K on heating and
272Ϫ280 K on cooling as determined by DSC. The room tempera-
tron powder diffraction data using rigid body models of the consti-
tuent ions and ab-initio direct space methods. Both, the CF3 group
and the SO3 group of the triflate ion, are rotationally disordered
around the SϪC bond, in the room temperature phase. In the low
temperature phase, the triflate ion is well localized. Like in the al-
kali metal triflates, the triflate ions are arranged in double layers
with the hydrophobic trifluoromethyl groups and the sulfonate
groups, respectively, pointing towards each other. The dimethylam-
monium ion is located closer to the sulfonate group with contacts
indicating hydrogen bonding. The packing in both phases is of the
topological CsCl structure type.
˚
ture phase crystallizes in space group Cmca (a ϭ 11.031(6) A, b ϭ
3
˚
˚
˚
18.466(14) A, c ϭ 8.173(9) A, V ϭ 1665(2) A , Z ϭ 8) and the low
˚
temperature phase in space group P 21/c (a ϭ 8.8717(18) A, b ϭ
3
˚
˚
˚
8.0838(16) A, c ϭ 10.968(2) A, β ϭ 92.128(4)°, V ϭ 786.0(3) A ,
Z ϭ 4). The structures of both phases were determined by single
crystal X-ray diffraction, but refinement did not yield satisfactory
residuals for the low temperature phase because of twinning of the
crystal. It was, therefore, independently solved from the synchro-
Keywords: Trifluoromethanesulfonate; Dimethylammonium; Tri-
fluoromethyl group; Crystal structure; Disorder; Phase transitions
Introduction
the ions. In the related compound ammonium triflate [13,
14], in which the asphericity of the cation is less pro-
nounced, four phase transitions occur in the range of
113Ϫ523 K [15].
The alkali metal trifluoromethanesulfonates (triflates) have
been intensively studied recently. They warrant this atten-
tion by the aptitude of these compounds to behave as ion
conductors which is due to the large number of rotational
degress of freedom of the triflate ion and the resulting pos-
sibilities of cation movability within the crystal lattice. Ac-
cordingly, emphasis has been put on investigations of the
crystal structures [1Ϫ6], the ionic conductivities [4, 5, 7],
and discussion of the correlated mechanism of ion move-
ment by solid state NMR techniques [8Ϫ12].
In this paper, we report the synthesis and crystal struc-
ture determination of dimethylammonium triflate. The di-
methylammonium ion can be considered to be a larger
homologue of an alkali metal. The aspherical shape of the
dimethylammonium ion may result in a more interlocked
type of packing than is achieved with the spherical alkali
metal cations, which would inhibit the free movability of
Experimental Section
Synthesis. Dimethylammonium trifluoromethansulfonate was syn-
thesized by dropwise adding trifluoromethanesulfonic acid to di-
methylamine (ϳ10 mL) that was kept at 195 K under argon in one
leg of an H-shaped apparatus which contained a glass filter sieve
between the two legs of the vessel. When a precipitate started to
form on addition of the acid, the solution was filtered into the
other leg. The cooling was removed, whereupon the dimethylamine
destilled off leaving behind the colorless dimethylammonium tri-
fluoromethansulfonate. The purity of the product was checked by
comparing the X-ray powder diffractogram with the calculated
powder diffractogram of the room temperature phase.
Single crystal X-ray structure determination. Intensity data of crys-
tals of C3H8F3NO3S 1 (M ϭ 195.16 g molϪ1) was collected on a
Bruker Smart Apex CCD X-ray diffractometer at 293 K and 100 K.
The data of the crystal at 293 K was processed with the SAINT
software [16]. The structure was solved by direct methods using
SIR2004 [17]. Full-matrix least-squares refinement against ΗFΗ2 and
localization of the remaining non-hydrogen atoms from difference
fourier maps proceeded with the SHELXL-97 program [18]. All
non-hydrogen atoms were refined with anisotropic thermal dis-
* Prof. Dr. Dr. h. c. Martin Jansen
Max-Planck-Institut für Festkörperforschung
Heisenbergstr. 1
D-70569 Stuttgart, Germany
Fax: ϩ49-711-6891502
E-mail: m.jansen@fkf.mpg.de
1410
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2007, 633, 1410Ϫ1416

Room Temperature and the Low Temperature Phase of Dimethylammonium Trifluoromethanesulfonate
Table 1 Selected crystallographic data for the single crystal struc-
ture determination of [NH2(CH3)2][CF3SO3] 1 at 293 and 100 K.
T / K
293
100
space group
Cmca
P21/c
˚
a / A
11.031(6)
18.466(14)
8.173(9)
90
1665(2)
8
1.557
0.405
4.42-50.00
5689
778 (Rint ϭ 0.1251)
387
8.8717(18)
8.0838(16)
10.968(2)
92.128(4)
786.0(3)
4
1.649
0.429
4.60-56.00
10828
1876
˚
b / A
˚
c / A
β / °
3
˚
V / A
Z
dcalcd / g cmϪ3
µ(Mo Kα) / mmϪ1
2θ range /°
no. reflections measured
no. unique reflections
no. observed reflections
a)
8688
no. parameters / restraints
R indices (I > 2σ)
94 / 0
103 / 0
R1 ϭ 0.0510
wR2 ϭ 0.1001
R1 ϭ 0.0978
wR2 ϭ 0.1156
0.882
R1 ϭ 0.1063
wR2 ϭ 0.2857
R1 ϭ 0.1195
wR2 ϭ 0.2941
1.198
R indices (all data)
Figure 1 Rietveld refinement plots of dimethylammonium triflate
1 (crosses: observed pattern, line: calculated pattern, bottom line:
difference plot. Bragg peaks are indicated by tick marks).
GOF
Ϫ3
˚
max./min. peaks / e A
0.148/Ϫ0.181
1.708/Ϫ0.981
a) Observation criterion: I > 2σ(I).
Table 2 Crystallographic data for the Rietveld refinement of the
low-temperature phase of dimethylammonium trifluoromethane-
sulfonate 1.
placement parameters. The trifluoromethyl and the sulfonate group
were found to be disordered. They were refined in two split posi-
tions related by a rotation of 180°. The ratio of the site occupation
factors of these two possible orientations refined to ϳ2:1 for the
CF3 group and ϳ3:1 for the SO3 group. Hydrogen atoms of the
methyl groups were geometrically constructed at idealized positions
and refined with isotropic thermal parameters proportional to the
thermal parameter of the carbon atom to which they are attached.
The sole crystallographically unique hydrogen atom at the nitrogen
atom was localized from the difference fourier map. Its positional
parameters and isotropic thermal parameter were freely refined.
formula
C3H8F3NO3S
106.1
formula weight / g molϪ1
T / K
space group
137
P21/c
Z
a / A
b / A
c / A
β / °
V / A
λ / A
2θ range /°
Rp
Rwp
4
˚
˚
˚
8.8876(6)
8.0619(9)
10.9915(11)
92.556(6)
786.77(18)
0.92234
3.08-45.27
0.0334
0.0475
0.0231
0.13116
67
3
˚
˚
The crystal at 100 K was found to be twinned with two compo-
nents, and the data was accordingly integrated with SAINT utiliz-
ing two orientation matrices obtained with the help of the program
GEMINI [19]. The structure was then solved by direct methods
using SHELXS-97 [20] with an input file containing only the non-
overlapping reflections of one component. Full-matrix least-
squares refinement against ΗFΗ2 was performed using a reflection file
containing non-overlapping and overlapping reflections from both
components and the hklf 5 option of SHELXL-97 [18]. The non-
hydrogen atoms were refined anisotropically. Hydrogen atoms were
assigned to idealized positions and refined with isotropic thermal
parameters proportional to the thermal parameter of the atom to
which they are attached.
Rexp
R(F2)
no. of parameters
no. of restraints
47
tance, exact wavelength, and tilting angle of the image plate. The
sample was contained in a sealed lithiumborate glass capillary with
0.5 mm diameter. The sample was aligned such that the sample was
closest to the centre of the nozzle of the cold air blower at a
distance of approx. 5 mm. The sample was cooled at a speed of
63 K/h from room temperature down to T ϭ 100 K and reheated
at a speed of 190 K/h back to room temperature. During exposure,
the sample was spun in order to improve randomization of the
crystallites. An exposure time of 240 seconds was chosen depending
on the saturation intensity of the image plate. Together with a read-
out time of 80 seconds of the image plate reader, this led to rates
of 5.6 K/ scan for cooling and 16.9 K/ scan for heating. Integration
of the full-circle powder patterns was performed using the program
FIT2D [21, 22] resulting in diagrams of corrected intensities versus
the scattering angle 2θ (figure 1). It was observed that the diffracted
intensity was quite uniformly distributed over the Debye-Scherrer
rings, ruling out severe grain size effects and preferred orientation.
Low angle diffraction peaks had a typical FWHM of 0.18° 2θ.
Further experimental details can be found in table 2.
Temperature variable synchrotron powder diffraction. In-situ X-ray
powder diffraction data of dimethylammonium trifluoromethane-
sulfonate (DMA-triflate) at low temperature were collected in
transmission geometry with a Cryostream 600 cold air blower
(Oxford Cryosystems) on a motorized goniometer head at beamline
X7B at the National Synchrotron Light Source (NSLS) at Brook-
haven National Laboratory. X-rays of energy 13.5 keV were selec-
ted by a double flat crystal monochromator in ultra-high vacuum
located 8 meters from the source. The size of the beam was adjusted
to approximately 0.5 mm2. As detector, a MAR 345 image plate
reader was set up perpendicular to the beam path at a distance of
approx. 165 mm from the sample. LaB6 was used as an external
standard to determine the beam center, sample-to-detector dis-
Z. Anorg. Allg. Chem. 2007, 1410Ϫ1416
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
1411

P. D. C. Dietzel, R. Dinnebier, M. Jansen
Data reduction on all powder diffraction patterns was performed
using the Powder3D program [23]. For the structure determination
of the low temperature phase of DMA-Triflate, the data set at T ϭ
139 K was selected. Indexing with ITO [24] led to the lattice param-
eters given in table 2. The number of formula units per unit cell
could be determined from volume increments. The extinctions
found in the powder pattern indicated P21/c as the most probable
space group, which could later be confirmed by Rietveld [25] refine-
ments. The crystal structure was solved by global optimization in
direct space using the DASH [26] structure solution package [27].
The measured powder pattern at room temperature was subjected
to Pawley refinements [28] in all possible space groups in order to
extract correlated integrated intensities from the pattern. Good fits
to the data were obtained. An internal coordinate description of
the (SO3CF3)Ϫ moiety was constructed using bond lengths, angles
and torsion angles from lithium triflate [1]. The position and orien-
Table 3 Relative atomic coordinates and isotropic displacement
parameters (in A ) of the non-hydrogen atoms for the crystal struc-
ture determination of 1 at 293 K and 100 K.
3
˚
Atom
Wyck.
x
y
z
Uiso
293 K
S1
8f
8f
16g
8f
16g
8f
8f
16g
8f
16g
8e
16g
0.0000
0.0000
Ϫ0.1066(5)
0.0000
Ϫ0.1079(13)
0.0000
0.0000
Ϫ0.0890(11)
0.0000
Ϫ0.1040(19)
Ϫ0.2500
Ϫ0.3215(3)
0.08658(7)
0.0956(3)
0.0603(2)
0.0485(9)
0.0706(8)
0.1787(3)
0.1826(4)
0.2152(5)
0.2169(19)
0.1940(8)
0.0626(3)
0.1034(2)
0.14210(15)
Ϫ0.0326(5)
0.2004(9)
0.299(2)
0.0834(6)
0.122(3)
0.157(5)
0.139(10)
0.154(14)
0.1107(18)
0.173(11)
0.173(7)
O1A
O2A
O1B
O2B
C1
F1A
F2A
F1B
F2B
N1
0.063(2)
0.2120(9)
0.3717(9)
0.1566(14)
0.093(5)
0.293(2)
Ϫ0.2500
0.36(4)
0.179(18)
0.0983(14)
0.1284(16)
C2
Ϫ0.1369(4)
100 K single crystal solution
ϩ
tation of the divalent cation NH2(CH3)2 as well as the position,
orientation, and torsion of the (SO3CF3)Ϫ anion in the unit cell
were postulated and the trial structure was subjected to a global
optimization. The structure giving the best fit to the data in space
group P21/c was validated by Rietveld refinement of the fractional
coordinates obtained at the end of the simulated annealing run
using the GSAS [29] program. To stabilize the refinements of the
(SO3CF3)Ϫ anion, 17 soft constraints for the bond lengths (CϪS,
CϪF, SϪO; NϪH, NϪC, CϪH) and 30 soft constraints for the
bond angles (SϪCϪF, CϪSϪO; CϪNϪH, HϪNϪN, HϪCϪN)
were used. The peak profiles and precise lattice parameters were
determined by LeBail-type fits [30] using the program GSAS. The
background was modeled manually using GUFI [31]. The peak-
profile was described by a pseudo-Voigt function [32]. Lattice par-
ameters, and cell volume of [NH2(CH3)2][CF3SO3] as a function of
temperature were determined by LeBail fits for all data sets using
GSAS (figure 5).
S1
4e
4e
4e
4e
4e
4e
4e
4e
4e
4e
4e
0.17826(7)
0.1897(2)
0.1048(2)
0.1385(2)
0.3762(3)
0.3861(2)
0.4599(2)
0.4332(3)
0.1345(2)
0.2050(3)
0.2245(3)
0.34711(6)
0.5235(2)
0.2503(3)
0.3129(3)
0.2793(3)
0.1151(2)
0.3585(2)
0.3041(3)
0.7893(2)
0.6315(3)
0.8756(3)
0.53513(4)
0.51209(17)
0.44235(17)
0.65806(15)
0.5345(2)
0.55632(18)
0.61772(19)
0.42630(18)
0.23745(16)
0.1988(2)
0.01185(18)
0.0264(4)
0.0334(5)
0.0278(5)
0.0208(5)
0.0361(4)
0.0431(5)
0.0454(5)
0.0140(4)
0.0202(5)
0.0196(5)
O1
O2
O3
C1
F1
F2
F3
N1
C2
C3
0.3351(2)
100 K powder solution
S1
4e
4e
4e
4e
4e
4e
4e
4e
4e
4e
4e
0.18048(50)
0.18079(130)
0.10217(110)
0.12927(110)
0.37903(70)
0.38531(100)
0.45728(110)
0.43561(100)
0.1402(12)
0.34491(60)
0.52087(100)
0.25293(130)
0.3103(13)
0.27429(90)
0.11737(110)
0.3584(13)
0.29657(130)
0.79327(170)
0.6291(19)
0.5350(5)
0.5086(11)
0.4403(10)
0.6546(8)
0.5377(7)
0.5649(10)
0.6202(10)
0.4311(8)
0.2321(13)
0.2047(15)
0.3408(12)
0.0111(16)
0.0111(16)
0.0111(16)
0.0111(16)
0.0461(33)
0.0461(33)
0.0461(33)
0.0461(33)
0.018(4)
O1
O2
O3
C1
F1
F2
F3
N1
C2
C3
0.2052(17)
0.2157(5)
0.018(4)
0.018(4)
Selected crystallographic data are summarized in table 1. CCDC
600194, 600195, and 602816 contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge via www.ccdc.cam.ac.uk/datarequest/cif, by emailing
data_request@ccdc.cam.ac.uk, or by contacting The Cambridge
Crystallographic Data Centre, 12, Union Road, Cambridge CB2
1EZ, UK; fax: ϩ44 1223 336033.
0.8689(16)
2
˚
Table 4 Anisotropic displacement parameters (in A ) for the sin-
gle crystal structure determination of 1 at 293 K and 100 K.
Atom U11
U22
U33
U23
U13
U12
293 K
Results and Discussion
S1
C1
0.1089(12) 0.0663(7) 0.0749(8) Ϫ0.0100(6)
0.000
0.000
0.000
0.003(9)
0.000
0.087(7)
0.000
0.06(2)
0.000
Ϫ0.109(16)
0.024(3)
Ϫ0.003(2)
0.000
0.000
0.000
0.067(5)
0.000
Ϫ0.055(4)
0.000
0.033(10)
0.000
0.073(11)
0.000
0.174(7)
0.071(4)
0.108(5)
0.087(5)
0.067(5)
0.005(3)
Ϫ0.027(3)
Crystal structures
F1A 0.34(3)
F2A 0.236(11) 0.098(6)
O1A 0.181(9)
O2A 0.164(7)
F1B
F2B
O1B 0.21(3)
O2B 0.128(18) 0.165(13) 0.17(2)
N1
C2
0.185(11) Ϫ0.011(5)
0.119(4)
0.103(3)
0.076(14) 0.18(2)
0.064(3)
0.203(9)
Ϫ0.017(3)
Ϫ0.045(4)
0.084(17)
Ϫ0.055(16)
0.018(9)
Trifluoromethyl groups are often found to be rotationally
disordered in crystal structure solutions at room tempera-
ture [33]. In the case of the trifluoromethanesulfonate ion,
the SO3 groups can also rotate around the SϪC axis, and
there may even be a dynamic exchange of the CF3 and SO3
groups at sufficiently high temperatures [11, 12]. Thus, it
should be advantageous for the structure determination to
freeze the dynamic disorder by working at lower tempera-
tures. However, a single crystal study of 1 at 100 K revealed
that the sample was twinned which is strongly indicative
that a phase transition occurred on cooling. Although the
structure could still be solved, refinement resulted in unsat-
isfactory residual factors which are likely to be an artefact
of the twin integration of the diffraction data. We then con-
firmed that 1 passes through a phase transition by per-
0.82(12)
0.241(18) 0.128(9)
0.17(4)
0.093(11) 0.110(13)
Ϫ0.097(14)
0.000
Ϫ0.039(3)
0.103(4)
0.112(3)
0.080(3)
0.174(4)
0.111(3)
0.099(3)
0.019(3)
100 K
F1
S1
0.0222(9) 0.0162(8) 0.0697(12)
0.0084(7)
Ϫ0.0015(8)
0.00072(19)
0.0069(8)
Ϫ0.0136(8)
0.0078(7)
0.0020(10)
Ϫ0.0136(9)
0.0327(9)
Ϫ0.0017(7)
0.0019(9)
0.0047(7)
0.0022(2)
0.0060(8)
0.0153(10)
0.0056(9)
0.0007(9)
Ϫ0.0039(8)
0.0142(9)
0.0003(7)
0.0051(10)
Ϫ0.0012(9)
0.0146(3) 0.0111(3) 0.0099(3) Ϫ0.00165(17)
0.0287(11) 0.0145(8) 0.0363(10) 0.0043(7)
0.0286(11) 0.0376(11) 0.0330(10) Ϫ0.0244(9)
O1
O2
O3
C1
F2
F3
N1
C2
C3
0.0278(12) 0.0437(12) 0.0125(8)
0.0185(12) 0.0135(11) 0.0306(12)
0.0059(7)
0.0030(9)
0.0175(10) 0.0403(12) 0.0704(13) Ϫ0.0059(9)
0.0385(12) 0.0457(11) 0.0544(11)
0.0174(11) 0.0095(9) 0.0149(9)
0.0115(9)
0.0012(7)
0.0268(16) 0.0145(11) 0.0194(11) Ϫ0.0053(8)
0.0255(14) 0.0131(11) 0.0195(11) Ϫ0.0028(8)
Ϫ0.0088(9)
1412
www.zaac.wiley-vch.de
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2007, 1410Ϫ1416

Room Temperature and the Low Temperature Phase of Dimethylammonium Trifluoromethanesulfonate
˚
Table 5 Selected bond lengths /A and angles /° for the single cry-
stal structure determination of 1 at 293 K and 100 K.
293 K
S1ϪO2A
S1ϪO2B
S1ϪO1A
S1ϪO1B
S1ϪC1
1.359 (4)
1.388 (10)
1.437 (5)
1.464 (17)
1.794 (6)
C1ϪF1B
C1ϪF2A
C1ϪF1A
C1ϪF2B
N1ϪC2
1.20 (3)
1.274 (9)
1.307 (11)
1.354 (15)
1.430 (4)
100 K
F1ϪC1
S1ϪO2
S1ϪO3
S1ϪO1
1.351 (3)
C1ϪF2
C1ϪF3
N1ϪC3
N1ϪC2
1.321 (3)
1.323 (3)
1.485 (3)
1.489 (3)
1.4225 (19)
1.4333 (17)
1.4527 (19)
1.840 (3)
S1ϪC1
O2ϪS1ϪO3
O2ϪS1ϪO1
O3ϪS1ϪO1
O2ϪS1ϪC1
O3ϪS1ϪC1
O1ϪS1ϪC1
F2ϪC1ϪF3
116.38 (14)
116.67 (13)
111.99 (12)
104.21 (12)
102.38 (12)
102.66 (12)
108.8 (2)
F2ϪC1ϪF1
F3ϪC1ϪF1
F2ϪC1ϪS1
F3ϪC1ϪS1
F1ϪC1ϪS1
C3ϪN1ϪC2
108.8 (2)
106.5 (2)
111.42 (18)
110.82 (17)
110.39 (17)
112.80 (19)
Table 6 Hydrogen bonds in dimethylammonium triflate 1 at
293 K and 100 K.
˚ ˚ ˚
d(D؊H) / A d(H···A) / A d(D···A) / A Є(D؊H···A) / °
D؊H···A
293 K
Figure 2 Molecular structure of dimethylammonium triflate 1 at
293 K (a) and 100 K (b) with atom-numbering scheme. Displace-
ment ellipsoids are drawn at the 50 % probability level.
N1ϪH1···O2A 0.82(6)
2.10(7)
2.796(7)
142(6)
100 K
N1ϪH1A···O1 0.92
N1ϪH1B···O3 0.92
2.09
1.93
2.955(3)
2.840(3)
156
169
forming a temperature variable synchrotron diffraction
experiment on a powdered sample. To support the results
of the single crystal structure determination, the crystal
structure of the low temperature phase was independently
solved from the synchrotron data using rigid body models
of the constituent ions and ab-initio direct space methods.
Both structure solutions, from single crystal data and from
powder data, agree well with each other. The structure of
the room temperature phase was determined from single
crystal data. It was refined to satisfactory residuals even
though the I/σ(I) ratio falls below 2 for θ above ϳ19° which
is a typical observation for compounds containing disorder.
Selected crystallographic data for the single crystal struc-
ture determinations and the Rietveld refinement from
synchrotron data are given in tables 1 and 2. Atomic coor-
dinates and anisotropic displacement parameters are listed
in tables 3 and 4. Table 5 contains selected geometric par-
ameters.
mean that the triflate ion is additionally either subject to a
dynamic oscillating motion with a small amplitude along
[100] or a static displacement to both sides of the bc plane.
In contrast, both ions are well localized at 100 K (figure
2b). The conformation of the triflate ion is staggered as is
commonly found. The anions in both phases are arranged
in the typical double layer fashion also found in the triflate
of lithium, sodium, and cesium [1Ϫ3, 5, 6], in which the
trifluoromethanesulfonate ions are arranged such that the
fluoroorganic groups in adjacent layers face each other, and
the hydrophilic sulfonate groups do likewise. The dimeth-
ylammonium ions are located exclusively close to the sul-
fonate groups (figure 3 center and right), and the hydrogen
atoms of the nitrogen are involved in hydrogen bonds with
oxygen atoms of the sulfonate group (table 6).
Both, the trifluoromethyl as well as the sulfonate group,
are rotationally disordered around the CϪS bond in the
room temperature phase, and the absolute magnitude of the
displacement parameters of the dimethylammonium ion
also indicates a certain degree of positional freedom for the
cation (figure 2a). The shape of the anisotropic displace-
ment parameter of C1 and the CϪS bond length which is
slightly shorter than in the low temperature structure may
The space group Cmca of the room temperature phase is
a minimal non-isomorphic supergroup of the monoclinic
space group P 21/c of the low temperature phase. The unit
cell of the orthorhombic unit cell at room temperature is
related to the monoclinic one of the low temperature phase
by a ϭ cЈ, b ϭ 2aЈ, and c ϭ bЈ. In the room temperature
phase, the dimethylammonium and the triflate ions are lo-
cated on special positions. On passing through the phase
Z. Anorg. Allg. Chem. 2007, 1410Ϫ1416
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
1413

P. D. C. Dietzel, R. Dinnebier, M. Jansen
Figure 3 Comparison of the packing in the low temperature (a; view along a, c, and b, respectively, from left to right) and in the room
temperature phase (b; view along b, a, and c, respectively, from left to right) of dimethylammonium triflate 1. Only one orientation of the
disordered EX3 groups in the room temperature phase is shown for clarity. A second unit cell is drawn along the a axis for the low
temperature phase to emphasize the relationship between the two structures.
transition, not only does the rotation of the EX3 groups in
the triflate cease, but they leave the special positions and
alternatingly shift to both sides of the ab plane of the mon-
oclinic phase (figure 3 left). The same occurs to the dimeth-
ylammonium ion in respect to the ac plane (figure 3 center).
In addition, it is located on a glide plane in the orthorhom-
bic phase, and each quadrant of the projection of the unit
cell onto the ac plane contains the dimethylammonium ion
in two orientations (figure 3b left). In the monoclinic phase,
with the triflate ions being less uniformly positioned than
in the orthorhombic phase, there is only one preferred
orientation of the dimethylammonium ion per quadrant
(now in the bc plane, figure 3a left).
significantly sharper and more intense endothermic peak at
285.35 K which actually has a shoulder at the slightly lower
temperature of 284.6 K. (The peak positions shift slightly
by ϳ0.4 K towards lower temperatures when the scan rate
is reduced to 1 °C/min.) On cooling, a strong exothermic
peak is observed with a maximum at 279.8 K. The shoulder
is not clearly discernible, though the peak shape is not sym-
metric either. The maximum of the weaker exothermic peak
that follows is located in the range of 272.5Ϫ274 K. It is
judicious to ascribe the intense signals to the crystallo-
graphic phase transition from the orthorhombic to the
monoclinic phase and vice versa, but we can only surmise
that the broader weak signal is caused by the onset or freez-
ing in, respectively, of the rotation of the CF3 and SO3
groups. No further thermal effects are observed in the tem-
perature range of 103 to 313 K. This includes the absence
of further phase transitions. For comparison, ammonium
triflate passes through three phase transitions in the same
temperature interval [15].
The packing principle of both phases becomes apparent
when the ions are replaced by their center of gravity (figure
4). Each center of gravity is surrounded by eight counter-
˚
ions within 4.32Ϫ5.77 A in the monoclinic phase and
˚
4.53Ϫ5.76 A in the orthorhombic phase, respectively. The
˚
next closest anion-cation distance is more than 3 A distant.
The coordination polyhedron resembles a distorted cuboid.
The packing is then topologically of the cesium chloride
structure type.
Temperature variable synchrotron diffraction
The phase transition between the room temperature and the
low temperature phase occurs in thermodiffraction pattern
ϳ20Ϫ30 K below the temperature found in the DSC experi-
ment (figure 6). On cooling, all three lattice parameters of
the low temperature phase decrease with the temperature
with the unit cell volume showing a linear dependency. The
temperature range in which the room temperature phase
Differential scanning calorimetry
The DSC measurements indicate three closely spaced ther-
mal effects within a short temperature range (figure 5). On
heating, there first occurs a less intense endothermic peak
with a maximum between 281.7Ϫ282.7 K, followed by a
1414
www.zaac.wiley-vch.de
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2007, 1410Ϫ1416

Room Temperature and the Low Temperature Phase of Dimethylammonium Trifluoromethanesulfonate
Figure 4 Topological representation of the packing in the room
Figure 6 Powder diffraction patterns, lattice parameters, and cell
temperature (a) and low temperature (b) structure of dimethyl-
ammonium triflate 1 that illustrates the similarity to the cesium
chloride structure type. The molecular ions are represented by their
center of gravity (dimethylammonium: light grey, triflate: dark
grey). One full set of short contacts around one cation and anion
each is colored grey for emphasis.
volume of NH2(CH3)2[CF3SO3] as a function of temperature in the
range from T ϭ 298 K down to T ϭ 100 K (65 K/h) and up to T ϭ
298 K (190 K/h).
phase and, thus, it is largely responsible for the increase
in volume.
Conclusion
Dimethylammonium triflate 1 is obtained in quantitative
yield by reaction of trifluoromethanesulfonic acid with di-
methylamine. It is found to pass through one structural
phase transition in the investigated temperature range of
103 to 313 K. The number of phases is therefore substan-
tially reduced in comparison with ammonium triflate. The
less spherical dimethylammonium ion apparently facilitates
a more optimal packing arrangement which reduces the
number of degrees of motional freedom of both ions in the
crystal packing. The trifluoromethanesulfonate ion is ro-
tationally disordered around the SϪC bond for both EX3
groups in the orthorhombic room temperature phase. In the
monoclinic low temperature phase, the motion is frozen.
The crystal structure contains double layers of triflate ions,
in which groups of similar polarity face each other, and the
cations are located in the vicinity of the hydrophilic sulfon-
ate group. The packing principle of both phases derives
from the cesium chloride structure type.
Figure 5 DSC plot of dimethylammonium triflate 1 recorded with
a scan rate of 5 °C/min (inset: 1 °C/min).
could be observed is too small to allow for an analysis of
the lattice parameter trend. A discontinuous increase of the
unit cell volume is observed when the phase transition oc-
curs, which is in accordance with the additional space re-
quirements of the rotating EX3 groups. The b axis of the
room temperature phase is significantly more than twice as
large as the corresponding a axis of the low temperature
Acknowledgements. The synchrotron measurements were carried
out in part at the National Synchrotron Light Source at Brook-
haven National Laboratory, which is supported under contract
Z. Anorg. Allg. Chem. 2007, 1410Ϫ1416
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
www.zaac.wiley-vch.de
1415

P. D. C. Dietzel, R. Dinnebier, M. Jansen
DE-AC02-98CH10886 by the US Department of Energy, Division
of Chemical Sciences, Office of Basic and Energy Sciences. Finan-
cial support by the Max-Planck Society (MPG), the Bundesminis-
terium für Bildung und Forschung (BMBF) and the Fonds der
Chemischen Industrie (FCI) is gratefully acknowledged.
[17] M. C. Burla, R. Caliandro, M. Camalli, B. Carrozzini, G. L.
Cascarano, L. De Caro, C. Giacovazzo, G. Polidori, R.
Spagna, J. Appl. Cryst. 2005, 38, 381.
[18] G. M. Sheldrick, SHELXL97. Program for the Refinement of
Crystal Structures (Release 97-2). Universität Göttingen, Ger-
many, 1997.
[19] GEMINI. Program for twinning solution. Bruker AXS Inc.,
Madison, Wisconsin, USA, 2000.
[20] G. M. Sheldrick, SHELXS97. Program for the Solution of
Crystal Structures (Release 97-2). Universität Göttingen, Ger-
many, 1997.
[21] A. P. Hammersley, FIT2D, Vers. 9.129, ESRF Internal Re-
port ESRF98HA01T.
[22] A. P. Hammersley, S. O. Svenson, M. Hanfland, D. Hauser-
man, High Pressare Res. 1996, 14, 235.
References
[1] M. Tremayne, P. Lightfoot, M. A. Mehta, P. G. Bruce, K. D.
M. Harris, K. Shankland, C. J. Gilmore, G. Bricogne, J. Solid
State Chem. 1992, 100, 191.
[2] M. Bolte, H. Lerner, Acta Crystallogr. 2001, E57, m231.
[3] R. Dinnebier, N. Sofina, M. Jansen, Z. Anorg. Allg. Chem.
2004, 630, 1613.
[4] G. Korus, M. Jansen, Z. Anorg. Allg. Chem. 2001, 627, 1599.
[5] N. Sofina, E.-M. Peters, M. Jansen, Z. Anorg. Allg. Chem.
2003, 629, 1431.
[6] L. Hildebrandt, R. Dinnebier, M. Jansen, Z. Anorg. Allg.
Chem. 2005, 631, 1660.
[7] M. Pompetzki, L. van Wüllen, M. Jansen, Z. Anorg. Allg.
Chem. 2004, 630, 484.
[8] C. I. Rathcliffe, B. A. Dunell, J. Chem. Soc., Faraday Trans.
1981, 77, 2169.
[9] M. Mortimer, E. A. Moore, M. A. K. Williams, J. Chem. Soc.,
Faraday Trans. 1992, 88, 2393.
[10] M. Mortimer, E. A. Moore, M. A. K. Williams, J. Chem. Soc.,
Faraday Trans. 1996, 92, 5043.
[23] B. Hinrichsen, R. E. Dinnebier, M. Jansen, Z. Krist. Suppl.
(EPDIC IX proceeding) 2006, 23, 231.
[24] J. W. Visser, J. Appl. Crystallogr. 1969, 2, 89.
[25] H. M. Rietveld, J. Appl. Crystallogr. 1969, 2, 65.
[26] DASH Ϫ Crystal Structure Solution from Powder Diffraction
Data, Version 3.0. The Cambridge Crystallographic Data Cen-
tre, 2005.
[27] W. I. F. David, K. Shankland, N. Shankland, Chem. Commun.
1998, 931.
[28] G. S. Pawley, J. Appl. Crystallogr. 1981, 14, 357.
[29] A. C. Larson, R. B. von Dreele, GSAS, Vers. 2002, Los
Alamos National Laboratory Report LAUR 86-748, Los
Alamos, USA.
[11] L. van Wüllen, N. Sofina, M. Jansen, Chem. Phys. Chem.
2004, 5, 1906.
[12] L. van Wüllen, L. Hildebrandt, M. Jansen, Solid State Ionics
2005, 176, 1449.
[30] A. Le Bail, H. Duroy, J. L. Fourquet, Mater. Res. Bull. 1988,
23, 447.
[31] R. E. Dinnebier, L. Finger, Z. Krist. 1998, Supplement Issue
15, 148.
[13] F. A. Schröder, B. Gänswein, G. Brauer, Z. Anorg. Allg. Chem.
1972, 391, 295.
[32] P. Thompson, D. E. Cox, J. B. Hastings, J. Appl. Crystallogr.
1987, 20, 79.
[33] P. D. C. Dietzel, M. Jansen, Z. Anorg. Allg. Chem. 2004, 630,
1573.
[14] B. Gänswein, G. Brauer, Z. Anorg. Allg. Chem. 1975, 415, 125.
[15] B. Gänswein, V. Kraemer, Thermochim. Acta 1976, 16, 395.
[16] SMART32 and SAINT32. Bruker AXS Inc., Madison,
Wisconsin, USA, 2000.
1416
www.zaac.wiley-vch.de
 2007 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim
Z. Anorg. Allg. Chem. 2007, 1410Ϫ1416