SMARTER crystallography of the fluorinated inorganic-organic compound Zn3Al2F12·[HAmTAZ]6

Article abstract of DOI:10.1039/c2dt30100h

We present in this paper the structure resolution of a fluorinated inorganic-organic compound - Zn₃Al₂F₁₂· [HAmTAZ]₆ - by SMARTER crystallography, i.e. by combining powder X-ray diffraction crystallography, NMR crystallography and chemical modelling of crystal (structure optimization and NMR parameter calculations). Such an approach is of particular interest for this class of fluorinated inorganic-organic compound materials since all the atoms have NMR accessible isotopes (¹H, ¹³C, ¹⁵N, ¹⁹F, ²⁷Al, ⁶⁷Zn). In Zn₃Al₂F ₁₂·[HAmTAZ]₆, ²⁷Al and high-field ¹⁹F and ⁶⁷Zn NMR give access to the inorganic framework while ¹H, ¹³C and ¹⁵N NMR yield insights into the organic linkers. From these NMR experiments, parts of the integrant unit are determined and used as input data for the search of a structural model from the powder diffraction data. The optimization of the atomic positions and the calculations of NMR parameters (²⁷Al and ⁶⁷Zn quadrupolar parameters and ¹⁹F, ¹H, ¹³C and ¹⁵N isotropic chemical shifts) are then performed using a density functional theory (DFT) based code. The good agreement between experimental and DFT-calculated NMR parameters validates the proposed optimized structure. The example of Zn ₃Al₂F₁₂·[HAmTAZ]₆ shows that structural models can be obtained in fluorinated hybrids by SMARTER crystallography on a polycrystalline powder with an accuracy similar to those obtained from single-crystal X-ray diffraction data.

Full text of DOI:10.1039/c2dt30100h

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
Dalton
Transactions
Cite this: Dalton Trans., 2012, 41, 6232
www.rsc.org/dalton
PAPER
SMARTER crystallography of the fluorinated inorganic–organic compound
Zn3Al2F12·[HAmTAZ]6†
Charlotte Martineau,*a Amandine Cadiau,‡b Boris Bouchevreau,a Jürgen Senker,c Francis Taulellea and
Karim Adil*b
Received 15th January 2012, Accepted 16th March 2012
DOI: 10.1039/c2dt30100h
We present in this paper the structure resolution of a fluorinated inorganic–organic compound—
Zn3Al2F12·[HAmTAZ]6—by SMARTER crystallography, i.e. by combining powder X-ray diffraction
crystallography, NMR crystallography and chemical modelling of crystal (structure optimization and
NMR parameter calculations). Such an approach is of particular interest for this class of fluorinated
inorganic–organic compound materials since all the atoms have NMR accessible isotopes (1H, 13C, 15N,
19F, 27Al, 67Zn). In Zn3Al2F12·[HAmTAZ]6, 27Al and high-field 19F and 67Zn NMR give access to the
inorganic framework while 1H, 13C and 15N NMR yield insights into the organic linkers. From these
NMR experiments, parts of the integrant unit are determined and used as input data for the search of a
structural model from the powder diffraction data. The optimization of the atomic positions and the
calculations of NMR parameters (27Al and 67Zn quadrupolar parameters and 19F, 1H, 13C and 15
N
isotropic chemical shifts) are then performed using a density functional theory (DFT) based code. The
good agreement between experimental and DFT-calculated NMR parameters validates the proposed
optimized structure. The example of Zn3Al2F12·[HAmTAZ]6 shows that structural models can be obtained
in fluorinated hybrids by SMARTER crystallography on a polycrystalline powder with an accuracy
similar to those obtained from single-crystal X-ray diffraction data.
calculations has recently emerged as an efficient way to over-
1. Introduction
come the intrinsic difficulties of powders. This so-called
SMARTER crystallography (structure elucidation by combining
magnetic resonance, computational modeling and diffractions)
encompasses many structural analyses using the combination of
such methods. It takes advantage of NMR atom resolved spec-
troscopy with a more local character to assist, improve and
perform structure determination together with powder diffrac-
tion. The information extracted from solid-state NMR data can
be used at different stages of the structure resolution process,
ranging from the determination or validation of a space group
over building of a structural model up to the structure refinement.
One (1D) and two-dimensional (2D) NMR spectra indeed reflect
the number, nature and multiplicity of the crystallographically
inequivalent atoms or block of atoms in the integrant unit (the
integrant unit—IU—is the first multiple of the asymmetric unit
that has integer crystallographic multiplicities for all atoms in the
unit cell)2 as well as their relative positions, and the combination
of the measurement of NMR parameters (chemical shift, quadru-
polar or scalar tensors…) with their ab initio calculations has
been shown to improve the accuracy of the atomic coordinates
initially determined from diffraction data. The use of NMR also
allows getting insights in various sub-networks that are usually
difficult to access from diffraction measurements, like ionic
mobility, the localization of organic templates in organic–inor-
ganic hybrid compounds, the distribution of iso-electronic atoms
Although both diffraction equipment and computing methods
have greatly improved over the past decade, ab initio structure
solution still remains highly challenging for powders,1 in par-
ticular for compounds lacking long-range order. A novel
approach combining diffraction with high-resolution solid-state
nuclear magnetic resonance (NMR) and quantum mechanical
aTectospin – Institut Lavoisier de Versailles, CNRS UMR 8180,
Université de Versailles Saint-Quentin en Yvelines, 45 Avenue des États-
Unis, 78035 Versailles Cedex, France. E-mail: charlotte.martineau@
chimie.uvsq.fr; Tel: +33139254260
bLUNAM Université, Université du Maine, CNRS UMR 6283, Institut
Moléculaire du Mans, Avenue Olivier Messiaen, 72085 Le Mans, Cedex
9, France. E-mail: karim.adil@univ-lemans.fr; Tel: +33243833352
cAnorganische Chemie III, Universität Bayreuth, Universitätsstr. 30,
95447 Bayreuth, Germany. E-mail: juergen.senker@uni-bayreuth.de;
Tel: +49921552538
†Electronic supplementary information (ESI) available: Conditions of
X-ray data collection, experimental and calculated XRPD diagram,
instruction and output files for the space group determination, differ-
ences between the fractional atomic positions of the DFT-optimized
structure and of the model determined from single-crystal X-ray diffrac-
tion of Zn3Al2F12·[HAmTAZ]6. CCDC reference number 871623. For
ESI and crystallographic data in CIF or other electronic format see DOI:
10.1039/c2dt30100h
‡Current address: CICECO, Complexo de Laboratórios Tecnológicos
Campus Universitário de Santiago 3810-193, Aveiro, Portugal.
6232 | Dalton Trans., 2012, 41, 6232–6241
This journal is © The Royal Society of Chemistry 2012

View Article Online
2.2. Thermal analysis
or group of atoms (Si/Al, F/OH…). SMARTER crystallography
has provided structural models of a wide variety of material
classes, ranging from small organic molecules3–5 and
pharmaceuticals6–8 over semiconductors for optical appli-
cations,9,10 inorganic fluorides,11–15 to porous systems like
zeolites,16–21 or metal–organic-framework (MOFs),22–29 which
have been described with a high degree of accuracy.
Differential thermal analysis (DTA) and thermogravimetric
analysis (TGA) curves of Zn3Al2F12·[HAmTAZ]6 were recorded
on a TA Instruments SDT-Q600 apparatus under argon, for temp-
eratures up to 800 °C, with a 5 °C min−1 heating rate.
Fluorinated hybrid materials, in particular, fluorinated metal–
organic frameworks (F-MOFs), sometimes exhibit enhanced
thermal stability, low surface tension and improved physico-
chemical performances compared to fully hydrogenated
MOFs,30–34 in particular in the presence of water.35–38 However,
to date, only a few materials built up from fluorinated inorganic
frameworks have been reported in the literature.39–41 Recently,
we have shown that F-MOFs, resulting from the inclusion of alu-
minium with the possibility of generating cationic linkers by
association of Zn2+ cations with triazolate molecules, could be
obtained by hydrothermal synthesis.42 In this paper, we present a
new fluorinated hybrid compound obtained from hydrothermal
synthesis with 3-aminotriazole (AmTAZ) organic linker,
Zn3Al2F12·[HAmTAZ]6.
2.3. X-ray diffraction
The powder X-ray diffraction pattern of Zn3Al2F12·[HAmTAZ]6
has been recorded at room temperature under air in a Bragg–
Brentano geometry with a PANalytical MPD-PRO diffractometer
using Cu Kα radiation in the 4–99° 2θ range and a 0.017° inter-
polated step. The Rietveld43 method using the Fullprof44
program was used for the structural refinement.
The single-crystal X-ray intensity data were collected on a
Bruker APEX II Quazar diffractometer (4-circle Kappa gonio-
meter, Iμs microfocus source, CCD detector) at 173 K. Empirical
absorption corrections were applied. The structure solution was
solved by direct methods (SHELXS-97),45 extended by succes-
sive difference Fourier syntheses and refined by full-matrix least-
square on all F2 data using SHELXL-97; these programs are
included in the WinGX46 package. All non-hydrogen atoms were
refined with anisotropic thermal parameters. All hydrogen atoms
were generated and refined isotropically.
The structure elucidation of this sample represents a case
study for SMARTER crystallography since all atoms, both in the
inorganic framework (27Al, 67Zn, 19F) and the organic linkers
(1H, 13C, 15N) can be measured by solid-state magic angle-spin-
ning (MAS) NMR. X-ray powder diffraction (XRPD) and high-
resolution one-dimensional (1D) 13C, 15N, 1H, 19F, 67Zn and
2.4. Solid-state NMR
1
27Al and 2D H NMR data are used to select the space group
and partially determine the integrant unit by identifying blocks
of atoms as sub-units. Emphasis is given on NMR experimental
issues (quantitative measurements, spectral resolution…) related
to each nuclei probed in this study. The search for a structural
model is then carried out by a Monte Carlo approach in direct
space, using parts of the sub-integrant units as input data. The
optimization of the atomic positions and the calculations of
NMR parameters (27Al and 67Zn quadrupolar parameters and
The 27Al single-pulse MAS (8 kHz) NMR spectrum of
Zn3Al2F12·[HAmTAZ]6 was recorded from powdered sample on
an Avance 500 Bruker spectrometer (B0 = 11.6 T, Larmor fre-
quency = 130.3 MHz) using a 2.5 mm probe, a 1 μs pulse
length, a recycle delay of 3 s and 19F 64-step small-phase incre-
mental alternation (SPINAL-64)47 decoupling (radio-frequency
field corresponding to a nutation frequency of 70 kHz). 256 tran-
sients were accumulated.
19F, H, 13C and 15N isotropic chemical shifts) are done by ab
The 67Zn static NMR spectrum was taken on an Avance 750
Bruker (Larmor frequency = 42.9 MHz) using a 4 mm probe. A
Hahn-echo (inter-pulse delay of 100 μs) sequence was used,
with 90° pulse length of 3.5 μs, and the full echo was recorded.
The recycle delay was set to 0.5 s and ∼110 000 transients were
accumulated (∼15 hours).
1
initio quantum calculations. The structural model proposed for
Zn3Al2F12·[HAmTAZ]6 is validated and its accuracy assessed by
comparing the experimental and DFT-calculated NMR par-
ameters. An independent structural model was also obtained
from single-crystal diffraction data. We show that the structural
model provided for Zn3Al2F12·[HAmTAZ]6 from powder diffrac-
tion data by SMARTER crystallography has an accuracy similar
to that of single-crystal X-ray diffraction data, including the
localization of the protons.
The 13C NMR spectra were recorded on an Avance 500
Bruker (Larmor frequency = 125.8 MHz) using 3.2 and 4 mm
probes. Cross-polarization polarization inversion (CPPI)48 curves
were recorded at a spinning frequency of 8 kHz, using CP49 con-
ditions that fulfill the n = +1 Hartmann–Hahn50 condition (50
kHz RF pulse on 13C) and 1 ms contact time. 13C Hahn-echo
spectra were recorded at a MAS frequency of 20 kHz using
various inter-pulse delays synchronized with 1 to 4 rotor periods,
3.3 μs 90° pulse length, 400 s recycle delay and 160 transients
for each spectrum. The 15N cross-polarization (CP) MAS (5
kHz) NMR spectrum was recorded on an Avance II 300 Bruker
spectrometer (B0 = 7 T, Larmor frequency = 30.4 MHz) using a
2. Experimental
2.1. Synthesis
Zn3Al2F12·[HAmTAZ]6 has been synthesized from a mixture of
ZnO (Merk), Al(OH)3 (Merk), 3-amino-1,2,4-triazole (Aldrich),
hydrofluoric acid solution (40% HF, Prolabo) and water. The
hydrothermal reaction has been performed in a Parr Teflon®
enclosure system at 160 °C by classical heating for 48 hours.
The obtained solid polycrystalline powder has been washed with
water and dried at room temperature. A single-crystal of suffi-
cient size could be extracted from this powder.
1
7 mm probe. The CP transfer was done using 50 kHz RF on H
and fulfilling the n = +1 Hartmann–Hahn condition (νnut(15N) =
νnut(1H) − νrot). The contact time was set to 7 ms, the recycle
delay to 15 s and ∼16 000 transients were accumulated. In all
1
13C and 15N NMR spectra, H SPINAL-64 decoupling with a
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6232–6241 | 6233

View Article Online
nutation frequency of 80 kHz was applied during the acquisition
period.
structure solution by SMARTER crystallography from powder
diffraction and solid-state NMR data and quantum mechanical
computations.
1
The 19F and H NMR spectra were recorded on an Avance III
800 Bruker spectrometer (B0 = 18.8 T, Larmor frequency =
800.1 MHz for 1H, 752.9 MHz for 19F) using a 1.3 mm ultra-fast
MAS probe. The 19F and 1H NMR spectra were recorded at MAS
frequency of 60 kHz. The recycle delay was set to 20 s and 30 s
3.2.1. Determination of the space group and integrant units.
The determination of a structural model of such fluorinated
hybrid starts with the indexing of the XRPD diagram to extract
unit cell parameters and possible space groups. In the case of
Zn3Al2F12·[HAmTAZ]6, the indexing of the XRPD diagram
using the McMaille64 software indicates a rhombohedral cen-
tered hexagonal cell, with refined parameters (Le Bail
method):65 a = 12.583(7) Å and c = 17.649(3) Å. Systematic
1
for 19F and H, respectively. 16 transients were accumulated. The
2D double-quantum single-quantum (DQ–SQ) NMR spectrum
51,52
was recorded at MAS 62.5 kHz using the R122
recoupling
pulse sequence. The DQ build-up curves were constructed based
on several 2D spectra with recoupling times ranging from 25 to
85 μs. In the 1D NMR experiments, the DEPTH53 pulse sequence
synchronized with the rotor period was applied to suppress exist-
ent 19F or 1H background. Phase sensitive detection in the indirect
dimension was obtained using the States-TPPI method.54
ˉ
line extinctions indicate possible space groups R3, R32, R3m,
ˉ
R3m and R3. The second step is to (i) reduce the number of
possible space groups; (ii) determine the largest possible part(s)
of the integrant unit, including both inorganic framework and
organic linkers, to ease the search for an initial structural model.
Those stages can be assisted by solid-state NMR.
Cationic framework: 27Al, 67Zn and 19F NMR. The central tran-
sition of the 27Al (nuclear spin I = 5/2) MAS NMR spectrum of
Zn3Al2F12·[HAmTAZ]6 shows a single shapeless resonance
(Fig. 2), whose NMR parameters have been determined by
reconstruction of the whole spinning sideband pattern: isotropic
The 1H, 13C, 15N, 19F, 67Zn and 27Al chemical shifts were
referenced to proton and carbon signals in TMS, nitromethane,
CFCl3, a 1 M solution of Zn(NO3)2 and a 1 M solution of Al
(NO3)3, respectively. The NMR spectra were reconstructed using
the Dmfit55 software. The CPPI curves were fitted using a home-
made routine running in MATLAB.56 The 13C echo-decay
curves were fitted to a mono-exponential decay.
2.5. DFT calculations
All calculations were conducted with the Kohn–Sham57 density
functional theory (DFT) using the CASTEP58,59 program in the
Materials Studio 5.0 environment.60 For the structure optimiz-
ation, ultrasoft pseudopotentials were employed, with a plane-
wave cut-off energy of 500 eV and a 2 × 2 × 2 Monkhorst–
Pack61 k-point sampling grid. During the structure optimization,
the cell parameters were kept constant. The Perdew, Burke and
Ernzerhof (PBE)62 functionals were used in the generalized gra-
dient approximation (GGA) for the exchange correlation energy.
Magnetic properties were computed using the projector-augmen-
ted wave method (GIPAW).63 For the calculation of the 27Al and
67Zn electric field gradient (EFG) tensor values, a 2 × 2 × 2
Monkhorst–Pack k-point grid was used to sample the Brillouin
zone, with a plane-wave basis set expanded to kinetics energy
lower than 500 eV. For the calculation of the 19F, 13C, 15N and
1H shielding tensor components, a 2 × 2 × 3 Monkhorst–Pack k-
point grid was used to sample the Brillouin zone, with a plane-
wave basis set expanded to kinetics energies lower than 500 eV.
Fig. 1 DTA/TGA curves of Zn3Al2F12·[HAmTAZ]6 in the temperature
range 25–800 °C.
3. Results and discussion
3.1. Thermal analysis
The thermal analysis curves (Fig. 1) show that Zn3Al2F12·[HAm-
TAZ]6 is stable up to 250 °C. The X-ray diffraction pattern of the
powder obtained at 800 °C corresponds to a mixture of α-AlF3
and ZnCN2 (theoretical loss 50.3%, experimental loss 48.3%),
which validates the chemical composition proposed.
3.2. SMARTER crystallography
Because all atoms have NMR active nuclei (1H, 13C, 15N, 27Al,
67Zn), Zn3Al2F12·[HAmTAZ]6 represents a case study for
Fig. 2 Experimental and calculated 27Al MAS NMR spectrum of
Zn3Al2F12·[HAmTAZ]6.
6234 | Dalton Trans., 2012, 41, 6232–6241
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 4 19F NMR spectrum of Zn3Al2F12·[HAmTAZ]6, on which lines
are labelled, recorded at ultra-fast MAS (60 kHz) and high-field
(18.8 T). Stars indicate non-identified impurities.
Fig. 3 Experimental and calculated 67Zn static Hahn-echo NMR spec-
trum of Zn3Al2F12·[HAmTAZ]6. The two individual contributions are
shown below.
Table 1 27Al and 67Zn line label, line intensity, experimental and
calculated from the DFT-optimized structure (in italic) isotropic
chemical shift δiso, quadrupolar coupling constant CQ and asymmetry
parameter ηQ and line assignment in Zn3Al2F12·[HAmTAZ]6
its upper-left part indicating the presence of two inequivalent
fluorine sites. Despite the high-resolution conditions employed,
the resolution of the two sites is poor, which indicates that the
fluorine atoms have very close chemical environments. The 19F
isotropic chemical shifts ∼−145 ppm are characteristic of F
atoms shared between one aluminum and one zinc atoms.42 The
AlF6 octahedra are thus isolated from each other.
Line
label
Intensity
(%)
δiso/ppm
CQ/MHz
ηQ
Assignment
Al1
27Al
1
The 19F, 67Zn and 27Al NMR data indicates that in
Zn3Al2F12·[HAmTAZ]6 the cationic network is built up from
AlF6 octahedra, isolated from each other, but which share
fluorine atoms with the Zn polyhedra. By analogy to ZnAlF5·
[TAZ],41 the first coordination shell of the Zn is assumed to be
completed by N atoms from the AmTAZ molecules (ZnF6−xNx
octahedra).
100
1.5 ( 0.5) 0.2 ( 0.1) 0 ( 0.1)
1.0 −0.23 0.0
67Zn
1
33 ( 1) 83 ( 2)
2.9 ( 0.1) 0 ( 0.1)
2.20 0.0
7.5 ( 0.1) 0 ( 0.1)
6.61 0.0
Zn1
Zn2
72
67 ( 1) 87 ( 2)
82
2
Organic linkers: 1H, 15N and 13C NMR. The 1D 1H MAS (62.5
kHz) NMR spectrum of Zn3Al2F12·[HAmTAZ]6 (Fig. 5a) shows
three resonances of relative intensity 24%, 28% and 48%. The
line at 7.7 and 6.7 ppm are at positions characteristic of protons
attached to a carbon atom and of protons from a NH2 group,
respectively. The 1H resonance at much higher chemical shift
(13.8 ppm) indicates that one nitrogen atom of the AmTAZ mol-
ecules is protonated, i.e. HAmTAZ in the final compound. A 2D
1H–1H NMR spectrum of Zn3Al2F12·[HAmTAZ]6 (Fig. 5b) was
chemical shift δiso = −1.5 ppm, characteristic of an Al atom in
six-fold fluorinated coordination, a small quadrupolar coupling
constant CQ ∼ 250 kHz and the asymmetry parameter ηQ ∼ 0
characteristic of only a slightly distorted AlF6 octahedron. A
single set of parameters has been used for the reconstruction of
this NMR spectrum, indicating the absence of distribution of the
27Al quadrupolar parameters and therefore the absence of F/OH
substitution in the compound.
67Zn (nuclear spin I = 7/2) solid-state NMR is challenging
because of the very low sensitivity of this nuclide associated
with a large quadrupolar moment66,67 Q = 0.15 × 10−28 m2 that
recorded using the symmetry-based homonuclear dipolar recou-
5,51,52
pling sequence R121
a scheme that can be used under
ultra-fast MAS conditions. On such a 1H double-quantum
single-quantum (DQ–SQ) NMR correlation spectrum, dipolar-
coupled inequivalent nuclei will generate a pair of off-diagonal
peaks, dipolarly coupled equivalent nuclei will generate a peak
on the diagonal, while non-coupled spins will be filtered out by
1
broadens the NMR lines. Moreover, the presence of H in the
structure of Zn3Al2F12·[HAmTAZ]6 contributes to a strong
decrease of the 67Zn non-refocusable transverse relaxation time
T2, making the use of refocusing signal enhancement techniques
like Carr–Purcell–Meiboom–Gill (CPMG)68,69 difficult. There-
fore, the 67Zn NMR spectrum of Zn3Al2F12·[HAmTAZ]6 was
recorded at high-field (17.6 T), under static condition using a
Hahn-echo pulse sequence. The 67Zn NMR spectrum (Fig. 3)
1
the pulse sequence. The 2D H NMR spectrum of Zn3Al2F12·
[HAmTAZ]6 shows intense cross-peaks between the protons
from the CH and NH groups, between the protons from the NH2
and NH groups, between the protons from the CH and NH2
groups as well as a strong auto-correlation peak for the two
protons of the NH2 group. Diagonal peaks are also present for
the NH and CH, which must be due to correlations between two
neighbouring amines. This is confirmed by the DQ build-up
curves (Fig. 5c) of the two protons from the NH2 group which
present a maximum for a recoupling time ∼65 μs, and decays
rapidly afterwards. The auto-correlation peaks for CH and NH
shows two Zn resonances: line 1 at δiso = 83 ppm, with CQ
=
2.9 MHz and ηQ = 0, and line 2 at δiso = 87 ppm, with a larger
CQ = 7.5 MHz and ηQ ∼ 0. The relative intensity of lines 1 and
2 are approximately 1 : 2, respectively (Table 1).
The 19F MAS NMR spectrum of Zn3Al2F12·[HAmTAZ]6
(Fig. 4), recorded at high-magnetic field (B0 = 18.8 T) and ultra-
fast MAS (62.5 kHz), shows one broad peak with a shoulder on
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6232–6241 | 6235

View Article Online
1
Fig. 5 (a) 1H ultra-fast MAS (62.5 kHz) and high-field (18.8 T) NMR spectrum of Zn3Al2F12·[HAmTAZ]6. Lines are assigned. (b) 2D H MAS
DQ–SQ correlation NMR spectrum. The top spectrum, on which lines are assigned, is the full projection onto the horizontal dimension. Dash lines
indicate proton–proton correlations. Thick line is the DQ diagonal (slope of 2). The yellow peak has a negative amplitude. (c) DQ build-up curves for
the auto-correlation peaks NH–NH, CH–CH and NH2–NH2.
have a slower build up, indicating longer CH–CH and NH–NH
distances (as expected between protons from neighboring
amines). Finally, one can notice on the 2D DQ–SQ NMR spec-
trum a peak of negative amplitude with no corresponding peak
across the diagonal (Fig. 5b). This peak appears at the δiso of the
NH in the horizontal dimension and at the sum of 2 δiso of NH2
in the vertical dimension and therefore it originates from a
relayed magnetization transfer from one NH2 to another NH2
through the NH.70
The 15N CPMAS NMR spectrum of Zn3Al2F12·[HAmTAZ]6
(Fig. 6) shows two resonances located at −118 and −163 ppm of
relative ratio 1 : 2. The aminotriazole molecule contains four
different nitrogen atoms, thus at least four 15N lines were
expected on the NMR spectrum of Zn3Al2F12·[HAmTAZ]6.
Because of the low natural abundance of 15N (below 1%), its
Fig. 6 15N CPMAS NMR spectrum of Zn3Al2F12·[HAmTAZ]6.
1
low magnetogyric ratio (∼1/10 of that of H) and usually long
6236 | Dalton Trans., 2012, 41, 6232–6241
This journal is © The Royal Society of Chemistry 2012

View Article Online
Fig. 7 (a) 13C CPMAS NMR spectrum and (b) CPPI curves of Zn3Al2F12·[HAmTAZ]6.
Table 2 19F, 1H, 15N and 13C line label, line intensity, experimental
and calculated from the DFT-optimized structure δiso and line
assignment in Zn3Al2F12·[HAmTAZ]6
represent the intensity of the lines for various CP contact times.
The behavior of the two resonances are different and the slower
build-up rate for the resonance at 156 ppm confirms its assign-
ment to the quaternary carbon atom of the amine. The strong
decay of intensity occurring at longer contact time is due to
spin–lattice relaxation of the protons in the rotating frame T1ρ.
Because of these differences, the 13C CPMAS NMR spectrum of
Zn3Al2F12·[HAmTAZ]6 is not quantitative under the experimen-
tal conditions employed in our study. To improve this quantitat-
ive aspect of 13C NMR, direct observation of 13C was done. A
Hahn-echo sequence was used to remove unwanted signals from
the probe, and the inter-pulse delays were synchronized with one
rotor period (50 μs). Quantitative measurements in this sequence
depends on the transverse relaxation rate T2 of the nuclei. In the
case of Zn3Al2F12·[HAmTAZ]6, the two 13C resonances have
distinct T2, as shown by the different lines of intensity decay of
the two resonances upon increase of the inter-pulse delay
(Fig. 8b). Quantitative relative line intensity (close to 1 : 1) was
therefore extracted by extrapolating the decay curves to an initial
time t = 0.
This NMR study indicates that the integrant unit in
Zn3Al2F12·[HAmTAZ]6 is built up from alternating AlF6 and
ZnF6−xNx octahedra. From these NMR data and the chemical
composition, the crystallochemical formula2,73 of the compound,
i.e. the chemical formula in which each atom type is split into its
inequivalent crystallographic sites, can be derived:
Zn2Zn1Al2F6F6[HAmTAZ]6. Possible space groups are then
those that possess the adequate Wyckoff positions that can
embed this crystallochemical formula.2 In agreement with dif-
fraction data, all hexagonal space groups were automatically
tested (see ESI† for the input data of the program: the number
and relative ratio of the NMR resonances, the chemical formula,
and if known the number of asymmetric units per unit cell).
Results indicate that only three of them are compatible with the
Line Relative
label intensity δiso,exp ( 1)/ppm
δiso,cal/ppm
Assignments
19F
1
1
1
−144
−145
−146
−144
F2
F1
2
1H
1
1
1
2
13.8
7.7
6.7
15.6
8.5
7.3 and 8.9
H4
H2
2
3
H3a and H3b
15N
1
2
1
−118
−163
−115.7
−161.9
−221.3
−336.6
N2
N1
N4
N3
2
3
4
13C
1
1
0.9
156
141
156.3
143.9
C2
C1
2
spin–lattice relaxation times, direct observation of 15N signals is
precluded. Therefore, the CPMAS49 technique, which consists of
transferring the magnetization from the surrounding sensitive
protons to the nitrogen atoms, has been employed. The augmen-
tation of the 15N spin response is however strongly dependent on
the dynamics occurring during the CP transfer, and is usually
non-uniform, which may explain why only two of the four
expected N signals are observed.
The 13C CPMAS NMR spectrum of Zn3Al2F12·[HAmTAZ]6
(Fig. 7a) shows two resonances located at δiso = 141 ppm and
156 ppm (Table 2). In order to identify the nature of the two
carbon sites, CPPI experiment was carried out. In such an exper-
iment, the rate of the polarization decay mostly depends on the
C–H dipolar interaction. The curve of the line located at
141 ppm (Fig. 7b) decays mono-exponentially, characteristic of a
quaternary carbon atom. In contrast, the line at 156 ppm exhibits
a bi-exponential decay with turning point for the normalized
intensity at zero, characteristic of a carbon atom from a CH
group.71 The relative multiplicity of the two resonances is differ-
ent from the 1 : 1 ratio expected for the amino-triazole molecule.
Quantitativity in CPMAS is difficult to control because of the
complex dynamics involved in the CP transfer.72 In Fig. 8a are
shown the CP-build up curves of the two resonances, which
ˉ
crystallochemical formula of Zn3Al2F12·[HAmTAZ]6: R3, R32
ˉ
ˉ
and R3m (see ESI†). In the R3m and R32 space groups, all
atoms would be in a special position. Since no such special pos-
itions are observed on the NMR data, the search for a structural
ˉ
model has been done in the centrosymmetrical space group R3
(no. 148).
3.2.2. Structural model. Once a space group is selected and
part of the IU defined, an initial structural model has to be
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6232–6241 | 6237

View Article Online
Fig. 8 13C (a) CP build-up curves and (b) Hahn-echo decay curves for the two 13C resonances in Zn3Al2F12·[HAmTAZ]6.
Fig. 9 (a) Projection of the structure of Zn3Al2F12·[HAmTAZ]6 in the rhombohedric unit cell along the [001] axis. (b) Representation of a cluster
showing the environments of the Al3+ and Zn2+ cations. (c) Representation of the environment of an amine: N1 and N2 are bonded to the Zn1 and
Zn2 atoms by iono-covalent bonds. N3 and N4 form hydrogen bonds (dash lines) with two fluorine atoms from a neighboring cluster. For the sake of
clarity, only one amine is shown.
found. This can be done, for example, by using Monte Carlo-
based software like FOX74,75 or Espoir.76 Partial or complete
knowledge of the IU at this stage is of great importance since it
reduces the number of independent atomic coordinates to be
determined. According to the NMR experiments, the input data
in the software FOX for Zn3Al2F12·[HAmTAZ]6 are an AmTAZ
molecule, an AlF6 octahedron and 2 independent Zn atoms. The
search, carried out using the XRPD as cost function, converged
to a structural model, which was then refined by the Rietveld
method from the powder diffraction data (Rp = 10.3%, Rwp
12.2%, RBragg = 5.63%, see ESI†).
=
The structure of Zn3Al2F12·[HAmTAZ]6 is built up from iso-
lated clusters (Fig. 9). Each cluster contains two crystallographi-
cally inequivalent Zn atoms and one six-fold fluorinated
coordinated aluminium atom. Zn(1) is surrounded by six amines
(dZn(1)–N = 2.18 Å), half of them being also connected to Zn(2)
(dZn(2)–N = 2.03 Å). Three fluorine atoms complete the coordi-
nation sphere of Zn(2). The Al and Zn polyhedra share a face.
6238 | Dalton Trans., 2012, 41, 6232–6241
This journal is © The Royal Society of Chemistry 2012

View Article Online
Table 3 Atom labels, Wyckoff positions, atomic coordinates (x,y,z) of
Because the Zn(1) ion sits on a −3 symmetry axis, the ZnN6 octa-
hedron is regular with all six Zn–N distances equal to 2.18 Å. On
the contrary, the ZnN3F3 is much more distorted, with Zn(2)–N
distances (2.03 Å) shorter than the Zn(2)–F distances (2.17 Å).
Similarly, in the AlF6 octahedron, the Al–Fbridging distances
(1.84 Å) are longer than the Al–Fnon-bridging distances (1.77 Å).
The amines are bonded to two Zn ions through iono-covalent
bonds and also form strong hydrogen bonds (average N–F
distance of 2.70 Å) with the F atoms from two neighbouring
clusters, yielding a three-dimensional character to the solid
network.
the
DFT-optimized
structure
and
of
the
structure
of
Zn3Al2F12·[HAmTAZ]6 determined from single-crystal diffraction data
(in italic). Only the DFT-optimized positions are given for the protons.
Uncertainties are given in brackets
Atom
Zn1
Zn2
Al1
F1
Wyckoff
x
y
Z
1a
2c
2c
6f
6f
6f
6f
6f
6f
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0.2030
0.2055(2)
0.3634
0.36462(5)
0.4197
0.41860(6)
0.2834
0.29831(6)
0.1488
0.1484(1)
0.0705
0.0702(1)
0.0505
0.0520(1)
0.1726
0.1723(1)
0.9936
0.1117
0.1132(1)
0.2834
0.2606
0.2444
0.2441(1)
0.1090
0.04717
0.0854(1)
0.1853
0.1290(1)
0.1138
0.1117(2)
0.0967
0.0981(2)
0.1641
0.1628(2)
0.1928
0.1860(2)
0.1708
0.2260
0.2187(2)
0.1925
0.2574
0.2380
0.2263(2)
0.2894
0.1925
0.1356(1)
0.0683
0.0479(1)
0.1592
0.1575(1)
0.1645
0.1638(2)
0.2742
0.2788(2)
0.2817
0.2708(2)
0.3186
0.3533
0.3484(2)
0.2372
0.3942
0.3069
0.3027(2)
0.4509
F2
3.3. Structure validation and optimization: DFT calculation of
NMR parameters
N1
N2
Validation of a structural model can be done by comparing NMR
parameters (shielding, electric field gradient tensors) determined
experimentally with parameters calculated ab initio from the
structural model. In inorganic fluorides, geometry optimization is
often required to improve the agreement between experimental
and calculated parameters,14,15,42,77,78 mostly because the pos-
itions of the light F and H atoms can be difficult to determine
from X-ray diffraction data only (powder or single-crystal).
Optimization of the structure of Zn3Al2F12·[HAmTAZ]6 was
done using DFT-based code CASTEP package,58,59 keeping the
cell parameters unchanged. The optimized atomic coordinates
are given in Table 3. Single crystals of Zn3Al2F12·[HAmTAZ]6
were also obtained. An independent structure solution could thus
be obtained from single crystal X-ray diffraction measurements
(Table 3). The structure model extracted from SMARTER crys-
tallography of the polycrystalline powder of Zn3Al2F12·[HAm-
TAZ]6 is close to the single-crystal model (atomic fractional
C1
C2
H2
N3
6f
6f
H3a
H3b
N4
6f
6f
6f
H4
6f
of Zn3Al2F12·[HAmTAZ]6, including the positions of the
protons, which could not be obtain from diffraction data (even
from single-crystal).
coordinate differences below 0.06, see Table 3 and ESI†).
1
The NMR parameters (27Al and 67Zn EFGs, H, 13C, 15N, 19
F
4. Conclusions
δiso) were calculated from the optimized structure (Tables 1 and
2). They agree rather well with the experimental values, which
validates the proposed structural model. The low 27Al quadrupo-
lar coupling constant CQ as well as the difference in CQs
between the two Zn atoms determined experimentally are well
reproduced by the calculations. The 19F δiso values calculated for
the two F sites are very close to each other (difference below
1 ppm), explaining why they were not resolved on the 19F MAS
NMR spectrum (Fig. 4), despite the high-magnetic field and
ultra-fast MAS conditions employed. Because the closeness in
δiso stands within the accuracy of the DFT calculations, the pro-
posed line assignment is not completely certain. The calculated
15N δiso values (Table 2) shows that the two lines observed on
the 15N CP-MAS NMR spectrum (Fig. 6) arise from the nitrogen
atoms N(1) and N(2) linked to the Zn atoms. The calculated 13C
δiso (Table 2) are close to the experimental values and confirm
the line assignment deduced from the 13C CPPI experiment
(Fig. 7b). Finally, the calculated chemical shifts of the protons
(Table 2) are close to the experimental values, which indicates
that the protons have been well positioned during the geometry
optimization. The good agreement between six independent sets
of experimental and calculated NMR parameters, one for each
type of nucleus in the compound (27Al, 67Zn, 19F, 15N, 13C and
1H), is a strong support for the reliability of the optimized model
SMARTER crystallography structure resolution of the fluorinated
hybrid sample Zn3Al2F12·[HAmTAZ]6 has been presented. Com-
bination of powder X-ray diffraction, solid-state NMR data and
quantum computation (structure optimization and NMR par-
ameter calculations) have yielded an accurate structural model
for Zn3Al2F12·[HAmTAZ]6. This class of samples is of particular
interest since all the atoms have NMR accessible isotopes. In
Zn3Al2F12·[HAmTAZ]6, 27Al and high-field 19F and 67Zn NMR
give access to the inorganic part of the framework while H, 13
C
1
and 15N NMR yield insights into the organic linkers. From these
experiments, parts of the integrant unit have been determined
and taken as input data for the search of a structural model from
the powder diffraction data. The optimization of the atomic pos-
itions and the calculations of NMR parameters (27Al and 67Zn
quadrupolar parameters and 19F, 1H, 13C and 15N isotropic
chemical shifts) has been done using DFT code. In this methodo-
logical approach, validation has also been obtained for
Zn3Al2F12·[HAmTAZ]6 with the structural model obtained inde-
pendently from single-crystal diffraction data, as well as with the
good agreement between six independent sets of experimental
and calculated NMR parameters. Through the example of
Zn3Al2F12·[HAmTAZ]6, we have shown that by SMARTER
crystallography, structural models could be obtained from
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6232–6241 | 6239

View Article Online
27 J. Y. Jiang, J. Huang, S. Marx, W. Kleist, M. Hunger and A. Baiker,
J. Phys. Chem. Lett., 2010, 1, 2886.
28 R. Hajjar, C. Volkringer, T. Loiseau, N. Guillou, J. Marrot, G. Férey,
I. Margiolaki, G. Fink, C. Morais and F. Taulelle, Chem. Mater., 2011,
23, 39.
powder X-ray diffraction data with NMR and modelling, with a
quality similar to that obtained from single-crystal diffraction
measurements. This approach also allows us to go even further
by providing the localization of the protons.
29 J. P. S. Mowat, S. R. Miller, A. M. Z. Slawin, V. R. Seymour, S.
E. Ashbrook and P. A. Wright, Microporous Mesoporous Mater., 2011,
142, 322.
30 C. Yang, X. Wang and M. A. Omary, J. Am. Chem. Soc., 2007, 129,
15454.
Acknowledgements
31 C. A. Fernandez, R. K. Thallapally, R. K. Motkuri, S. K. Nune, J.
C. Sumrak, J. Tian and J. Liu, Cryst. Growth Des., 2010, 10, 1037.
32 T. Devic, P. Horcajada, C. Serre, F. Salles, G. Maurin, B. Moulin,
D. Heurtaux, G. Clet, A. Vimont, J.-M. Grenèche, B. Le Ouay,
F. Moreau, E. Magnier, Y. Filinchuk, J. Marrot, J.-C. Lavalley, M. Daturi
and G. Férey, J. Am. Chem. Soc., 2010, 132, 1127.
33 P. Horcajada, F. Salles, S. Wuttke, T. Devic, D. Heurtaux, G. Maurin,
A. Vimont, M. Daturi, O. David, E. Magnier, N. Stock, Y. Filinchuk,
D. Popov, C. Riekel, G. Férey and C. Serre, J. Am. Chem. Soc., 2011,
133, 17839.
Financial support from the TGE RMN THC FR3050 for con-
ducting the research is gratefully acknowledged. CM thanks Dr
Franck Fayon (CEMHTI Orléans, France) and Dr Julien Trébosc
(UCCS Lille, France) for their help in the high-field NMR
measurements. AC and KA thank Prof. Marc Leblanc (IMMM
Le Mans, France) for helpful discussions.
34 P. Pachfule, Y. Chen, J. Jian and R. Banerjy, Chem.–Eur. J., 2012, 18,
688.
References
35 C. Yang, X. Wang and M. A. Omary, J. Am. Chem. Soc., 2007, 129,
15454.
36 C. A. Fernandez, R. K. Thallapally, R. K. Motkuri, S. K. Nune, J.
C. Sumrak, J. Tian and J. Liu, Cryst. Growth Des., 2010, 10, 1037.
37 Z. Hulvey, D. A. Sava, J. Eckert and A. K. Cheetham, Inorg. Chem.,
2011, 50, 403.
38 C. Yang, U. Kaipa, Q. Z. Mather, X. Wang, V. Nesterov, A. F. Venero and
M. A. Omary, J. Am. Chem. Soc., 2011, 133, 18094.
39 M. A. Goforth, C.-Y. Su, R. Hipp, R. B. Macquart, M. D. Smith and H.-
C. Loye, J. Solid State Chem., 2005, 178, 2511.
40 D. L. Rogow, G. Zapeda, C. H. Swanson, X. Fan, C. F. Campana, A.
G. Oliver and S. R. J. Oliver, Chem. Mater., 2007, 19, 4658.
41 L. Liu, X. Wang and A. J. Jacobsen, Dalton Trans., 2010, 39, 1722.
42 A. Cadiau, C. Martineau, M. Leblanc, V. Maisonneuve, A. Hémon-
Ribaud, F. Taulelle and K. Adil, J. Mater. Chem., 2011, 21, 3949.
43 H. M. Rietveld, J. Appl. Crystallogr., 1969, 2, 65.
44 J. R. Carjaval, FULLPROF Program, Rietveld Pattern Matching Analysis
of Powder Patterns, ILL, Grenoble, 1990.
1 A. Le Bail, L. M. D. Cranswick, K. Adil, A. Altomare, M. Avdeev,
R. Cerny, C. Cuocci, C. Giacovazzo, I. Halasz, S. H. Lapidus, J.
N. Louwen, A. Moliterni, L. Palatinus, R. Rizzi, E. C. Schilder, P.
W. Stephens, K. H. Stone and J. van Mechelen, Powder Diffr., 2009, 24,
254.
2 F. Taulelle, Fundamental Principles of NMR Crystallography, in NMR
Crystallography, ed. R. K. Harris, R. Wasylishen and M. Duer, J. Wiley
& Sons Ltd, Chichester UK, 2009, p. 245.
3 J. Senker and J. Lüdecke, Z. Naturforsch., 2001, 56b, 1089.
4 J. Senker, L. Seyfarth and J. Voll, Solid State Sci., 2004, 6, 1039.
5 L. Seyfarth, J. Sehnert, N. E. A. El-Gamel, W. Milius, E. Kroke, J. Breu
and J. Senker, J. Mol. Struct., 2008, 889, 217.
6 R. K. Harris, in NMR Crystallography, ed. R. K. Harris,
R. D. Wasylishen and M. J. Duer, John Wiley & Sons Ltd, Chichester,
2009, ch. 1, p. 1.
7 R. K. Harris, P. Hodgkinson, V. Zorin, J.-N. Dumez, B. Elena-Herrmann,
L. Emsley, E. Salager and R. S. Stein, Magn. Reson. Chem., 2010, 48,
S103.
45 G. M. Sheldrick, Appl. Crystallogr., 2008, A64, 112.
46 L. J. J. Farrugia, J. Appl. Crystallogr., 1999, 32, 837.
47 B. M. Fung, A. K. Khitrin and K. Ermolaev, J. Magn. Reson., 2000, 142,
97.
8 X. Filip, G. Borodi and C. Filip, Phys. Chem. Chem. Phys., 2011, 13,
17978–17986.
9 L. Seyfarth, J. Seyfarth, B. V. Lotsch, W. Schnick and J. Senker, Phys.
Chem. Chem. Phys., 2010, 12, 2227.
48 X. Wu and K. W. Zilm, J. Magn. Reson., Ser. A, 1993, 102, 205.
49 J. Schaefer and E. O. Stejskal, J. Am. Chem. Soc., 1976, 89, 1031.
50 S. R. Hartmann and E. L. Hahn, Phys. Rev., 1962, 128, 2042.
51 M. Carravetta, M. Eden, X. Zhao, A. Brinkmann and M. H. Levitt,
Chem. Phys. Lett., 2000, 321, 205.
52 L. Mafra, R. Siegel, C. Fernandez, D. Schneider, F. Aussenac and
J. Rocha, J. Magn. Reson., 2009, 199, 111.
53 D. G. Cory and W. M. Ritchey, J. Magn. Reson., 1988, 80, 128.
54 D. Marion, M. Ikura, R. Tschudin and A. Bax, J. Magn. Reson., 1989,
85, 393.
10 L. Seyfarth and J. Senker, Phys. Chem. Chem. Phys., 2009, 11, 3522.
11 J. Dutour, N. Guillou, C. Huguenard, F. Taulelle, C. Mellot-Draznieks
and G. Férey, Solid State Sci., 2004, 6, 1059.
12 M. Body, G. Silly, C. Legein, J.-Y. Buzaré, F. Calvayrac and P. Blaha, J.
Solid State Chem., 2005, 178, 3655.
13 C. Martineau, F. Fayon, C. Legein, J.-Y. Buzaré, F. Goutenoire and
E. Suard, Inorg. Chem., 2008, 47, 10895.
14 C. Martineau, F. Fayon, C. Legein, J.-Y. Buzaré, M. Body, D. Massiot
and F. Goutenoire, Dalton Trans., 2008, 6150.
15 C. Martineau, F. Fayon, M. R. Suchomel, M. Allix, D. Massiot and
F. Taulelle, Inorg. Chem., 2011, 50, 2644.
16 H. Gies and B. Marler, Zeolites, 1992, 12, 42.
17 D. H. Brouwer, J. Am. Chem. Soc., 2008, 130, 6306.
18 D. H. Brouwer, J. Magn. Reson., 2008, 194, 136.
19 D. H. Brouwer, R. J. Darton, R. E. Morris and M. H. Levitt, J. Am.
Chem. Soc., 2005, 127, 10365.
20 S. Cadars, D. H. Brouwer and B. F. Chmelka, Phys. Chem. Chem. Phys.,
2009, 11, 1825.
21 D. H. Brouwer, I. L. Moudrakovski, R. J. Darton and R. E. Morris,
Magn. Reson. Chem., 2010, 48, S113.
22 L. Beitone, C. Huguenard, A. Gansmüller, M. Henry, F. Taulelle,
T. Loiseau and G. Férey, J. Am. Chem. Soc., 2003, 125, 9102.
23 L. Beitone, J. Marrot, T. Loiseau, G. Férey, M. Henry, C. Huguenard,
A. Gansmüller and F. Taulelle, J. Am. Chem. Soc., 2003, 125, 1912.
24 T. Ahnfeldt, N. Guillou, D. Gunzelmann, I. Margiolaki, T. Loiseau,
G. Férey, J. Senker and N. Stock, Angew. Chem., Int. Ed., 2009, 48,
5163.
55 D. Massiot, F. Fayon, M. Capron, I. King, S. Le Calvé, B. Alonso, J.
O. Durand, B. Bujoli, Z. Gan and G. Hoatson, Magn. Reson. Chem.,
2002, 40, 70.
56 The language of technical computing: MATLAB; Maths Works Inc., 24
Prime Park Way, Natick, MA 01760-150, Copyright 1984–2004.
57 W. Kohn and L. Sham, Phys. Rev., 1965, 140, 1133.
58 M. D. Segall, P. L. D. Lindan, M. J. Probert, C. J. Pickard, P. J. Hasnip,
S. J. Clark and M. C. Payne, J. Phys.: Condens. Matter, 2002, 14, 2717.
59 S. J. Clark, M. D. Segall, C. J. Pickard, P. J. Hasnip, M. J. Probert,
K. Refson and M. C. Payne, Z. Kristallogr., 2005, 220, 567.
60 http://accelrys.com
61 H. J. Monkhorst and J. D. Pack, Phys. Rev. B: Solid State, 1976, 13,
5188.
62 J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett., 1996, 77,
3865.
63 C. J. Pickard and F. Mauri, Phys. Rev. B: Condens. Matter, 2001, 63,
245101.
64 A. Le Bail, Powder Diffr., 2004, 19, 249.
65 A. Le Bail, H. Duroy and J. L. Fourquet, Mater. Res. Bull., 1988, 23,
447.
66 P. Pyykkö, Mol. Phys., 2001, 99, 1617.
25 M. Haouas, C. Volkringer, T. Loiseau, G. Férey and F. Taulelle, Chem.–
Eur. J., 2009, 15, 3139.
26 T. Ahnfeldt, D. Gunzelmann, T. Loiseau, D. Hirsemann, J. Senker,
G. Férey and N. Stock, Inorg. Chem., 2009, 48, 3057.
6240 | Dalton Trans., 2012, 41, 6232–6241
This journal is © The Royal Society of Chemistry 2012

View Article Online
67 P. Pyykkö, Mol. Phys., 2008, 106, 1965.
68 T. Vosegaard, F. H. Larsen, H. J. Jakobsen, P. D. Ellis and N. C. Nielsen,
J. Am. Chem. Soc., 1997, 119, 9055.
69 F. H. Larsen, H. J. Jakobsen, P. D. Ellis and N. C. Nielsen, J. Magn.
Reson., 1998, 131, 144.
72 S. Ding, C. A. McDowell and C. Ye, J. Magn. Reson., Ser. A, 1994, 109,
6.
73 F. Taulelle, Solid State Sci., 2004, 6, 1053.
74 V. Favre-Nicolin and R. Cerny, J. Appl. Crystallogr., 2002, 35, 734.
75 http://objcryst.sourceforge.net
70 R. R. Ernst, G. Bodenhausen and A. Wokaun, Principles of Nuclear
Magnetic Resonance in One and Two Dimensions, Oxford University
Press, Oxford, UK, 1994.
76 A. Le Bail, Powder Diffr., 2005, 20, 316.
77 M. Body, G. Silly, C. Legein, J.-Y. Buzaré, F. Calvayrac and P. Blaha,
J. Solid State Chem., 2005, 178, 3655.
71 C. Gervais, F. Babonneau, J. Maquet, C. Bonhomme, D. Massiot,
E. Framery and M. Vaultier, Magn. Reson. Chem., 1998, 36, 407.
78 A. Zheng, S.-B. Liu and F. Deng, J. Phys. Chem. C, 2009, 113,
15018.
This journal is © The Royal Society of Chemistry 2012
Dalton Trans., 2012, 41, 6232–6241 | 6241