Solid state molecular dynamic investigation of an inclusion ferroelectric: [(2,6-diisopropylanilinium)([18]crown-6)]BF4

Article abstract of DOI:10.1021/ja503344b

Many order-disorder-type phase transitions in molecule-based ferroelectrics are related to changes of molecular dynamics. If the molecular motions do not involve reorientations of dipole moments, their ordering fails to contribute directly to spontaneous electric polarization. For understanding ferroelectric mechanisms in these systems, it is important to clarify how such molecular dynamics changes induce structurally symmetry-breaking phase transitions and thus the appearance of spontaneous electric polarization. Systematic characterization of an [18]crown-6 based host-guest inclusion compound, [(DIPA)([18]crown-6)]BF₄ (DIPA = 2,6-diisopropylanilinium), shows it is an excellent ferroelectric with a large dielectric anomaly, significant pyroelectricity, and SHG response, and rectangular polarizaiton-electric field hysterisis loops. By the combination of variable-temperature single-crystal structural determination and solid-state NMR observation, it is found that the slowing down of the rotation of the [18]crown-6 molecule and the tumbling of the BF₄ anion causes the symmetry breaking, while the spontaneous polarization is induced by the relative displacement between the cationic and anionic sublattices. This investigation will contribute to a deeper understanding of the structure-property relationship in the emerging molecular ferroelectrics.

Full text of DOI:10.1021/ja503344b

Article
pubs.acs.org/JACS
Solid State Molecular Dynamic Investigation of An Inclusion
Ferroelectric: [(2,6-Diisopropylanilinium)([18]crown-6)]BF4
Heng-Yun Ye,†,§ Shen-Hui Li,‡,§ Yi Zhang,† Lei Zhou,‡ Feng Deng,‡ and Ren-Gen Xiong*,†
†Ordered Matter Science Research Center, Southeast University, Nanjing 211189, P. R. China
‡State Key Laboratory Magnetic Resonance and Atomic and Molecular Physics, Wuhan Center for Magnetic Resonance, Wuhan
Institute of Physics and Mathematics, Chinese Academy of Sciences; Wuhan 430071, P. R. China
S
* Supporting Information
ABSTRACT: Many order−disorder-type phase transitions in
molecule-based ferroelectrics are related to changes of
molecular dynamics. If the molecular motions do not involve
reorientations of dipole moments, their ordering fails to
contribute directly to spontaneous electric polarization. For
understanding ferroelectric mechanisms in these systems, it is
important to clarify how such molecular dynamics changes
induce structurally symmetry-breaking phase transitions and
thus the appearance of spontaneous electric polarization.
Systematic characterization of an [18]crown-6 based host−
guest inclusion compound, [(DIPA)([18]crown-6)]BF4
(DIPA = 2,6-diisopropylanilinium), shows it is an excellent
ferroelectric with a large dielectric anomaly, significant pyroelectricity, and SHG response, and rectangular polarizaiton−electric
field hysterisis loops. By the combination of variable-temperature single-crystal structural determination and solid-state NMR
observation, it is found that the slowing down of the rotation of the [18]crown-6 molecule and the tumbling of the BF4 anion
causes the symmetry breaking, while the spontaneous polarization is induced by the relative displacement between the cationic
and anionic sublattices. This investigation will contribute to a deeper understanding of the structure−property relationship in the
emerging molecular ferroelectrics.
salts,5 dimethylammonium metal formate salts6 and crown-
INTRODUCTION
■
ether-based rotator-stator-type compounds.7
Ferroelectrics are an important class of materials and are widely
In the rotator-stator-type ferroelectrics, the crown-ether
used in memory applications (because of their polarization
molecule acts as a host which affords lone-pair electrons of
switching), capacitors (because of their high dielectric
the O atoms to anchor a dipolar ammonium cation through
constant), sensors (because of their piezoelectricity), and
hydrogen bonding interactions. At high temperatures, the
applications based on their optical effects (such as electro-optic
dipolar organic ammonium cation may perform a rotational/
hopping motion like a molecular rotator, and accordingly, the
crystal does not have a net dipole moment because they cancel
each other out. At low temperatures, the rotation is frozen,
effect, nonlinear optic effect, and anomalous photovoltaic effect,
and so on).1,2 A ferroelectric usually undergoes a structural
phase transition. Above the transition temperature Tc, it is in
the paraelectric state, that is, an applied electric field induces an
leading to long-distance orientation of the dipole moments and
electric polarization, which goes to zero when the field is
removed.1 Below Tc, it is in the ferroelectric state with a
spontaneous electric polarization (Ps) whose direction can be
switched by an applied external electric field. Transitions
between paraelectric states to ferroelectric states are induced by
the changes of atomic positions. For example, in the ionic
compound BaTiO3, Ps derives largely from the electric dipole
moment created by the off-center shift of the Ti ion from the
center of the TiO6 octahedron. In molecular compounds, the
changes of atomic positions accompany deformations and/or
reorientations of molecules.3 This has been utilized as an
important element in design of order−disorder-type molecular
ferroelectrics by introducing dynamically disordered molecules
into crystal lattices, such as pyridinium salts,4 imidazolium
thus appearance of Ps. This has been evidenced in
[(C7H10NO)([18]crown-6)][BF4],8 [(C7H10NO)([18]crown-
6)][ReO4]9 (C7H10NO = 4-methoxyanilinium) and (m-FAni)-
(DB[18]crown-6)[Ni(dmit)2]10 (m-FAni = m-fluoroanilinium;
dmit2− = 2-thioxo-1,3-dithiole-4,5-dithiolate). However, the
ferroelectric mechanism becomes complex if the rotating
molecule does not carry a dipole moment perpendicular to
the rotating axis because its ordering will not lead to long-
distance orientation of dipole moments, as has been observed
in our recently reported ferroelectric compound, [(DIPA)
([18]crown-6)]ClO4 (DIPA = 2,6-diisopropylanilinium), in
Received: April 9, 2014
Published: June 23, 2014
© 2014 American Chemical Society
10033
dx.doi.org/10.1021/ja503344b | J. Am. Chem. Soc. 2014, 136, 10033−10040

Journal of the American Chemical Society
Article
which the order−disorder change of the [18]crown-6 molecule
and perchlorate anion is probably responsible for the
ferroelectric phase transition.11 To disclose the ferroelectric
mechanism induced by motions not involving reorientations of
dipole moments remains a challenge and is currently an
important subject of science since the uses of molecular
ferroelectrics are quickly expanding.
MATERIALS AND METHODS
■
Synthesis. All reagents and solvents in the synthesis were purchased
from Aladdin Chemistry Co., Ltd. (Shanghai, China). They were of
reagent grade and used without further purification. 1 was obtained as
crystals by slow evaporation of an acetone solution (50 mL)
containing 2,6-diisopropylaniline (1.77 g, 10 mmol), [18]crown-6
(2.64 g, 10 mmol) and HBF4 (0.88 g, 10 mmol) at room temperature.
Large colorless crystals (up to 1 × 1 × 2 cm3) were formed after about
3 weeks. IR (Figure S1, Supporting Information) (cm−1): 3126 (s,
νNH), 2893, 2920 (s, νCH), 1617 (m, νCC), 1109 (s, νBF), 1028 (m,
νC−O). The purity and stability of the bulk phase was confirmed by
variable-temperature powder X-ray diffraction (PXRD) (Figure S2,
Supporting Information).
Variable-temperature X-ray structure determination is one of
the most important methods in understanding ferroelectric
mechanisms. It can provide much information on structural
phase transitions for molecular compounds, such as the changes
of molecular orientations and order−disorder transitions.
However, a structure determined from the diffraction pattern
is the spatial average over the whole crystal, and thus the
dynamics in the above-mentioned system cannot be revealed
sufficiently. For example, it is not reliable to distinguish static
disorder from dynamic disorder.12 Especially, it cannot identify
the system lacking orientational disorder such as the “molecular
merry-go-round” motion of the disk-shaped [18]crown-6.13
The solid-state NMR technique is a well-established tool for
investigating a local structure and clarifying dynamics of
molecular motions in material sciences.14−16 Several NMR
parameters such as line-width, chemical shift anisotropy, and
spin relaxation rates, are considerably sensitive to molecular
motions, which can serve as an indicator for molecular
dynamics. It may play an important role in disclosing
ferroelectric mechanisms for systems involving dynamic
molecules. However, it is usually insufficient to resolve the
detail of molecular orientations and/or configurations in a
crystal. The combination of the two techniques will facilitate
probing the detail of molecular motions in solids.
Measurement Methods. Variable-temperature X-ray single
crystal diffraction analysis was carried out using the same crystal.
The instrument is a Rigaku Saturn 724+ CCD diffractometer equipped
with a Rigaku low-temperature gas-spray cooler with Mo Kα radiation
(λ = 0.71073 Å). Data collection, cell refinement, and data reduction
were performed using Rigaku Crystalclear 1.3.5. The structures of 1
were solved by direct methods and refined by the full-matrix method
based on F2 with the use of the SHELXTL software package. All non-
hydrogen atoms were refined anisotropically, and the positions of all
hydrogen atoms were generated geometrically. X-ray powder
diffraction was measured on a Rigaku DMX/2000 X-ray diffraction
instrument. Specific heat (Cp) measurements were carried out on a
Quantum Design PPMS. Differential scanning calorimetry (DSC)
experiments were performed on a NETZSCH DSC 200 F3 instrument
under a nitrogen atmosphere in aluminum crucibles with the heating
and cooling rate of 10 K min−1. For dielectric, pyroelectric, and P−E
hysteresis loops measurements, thin single-crystal plates with about 10
mm2 in area and 1 mm thickness were cut perpendicular to the polar
axis of LTP. Silver conductive paste deposited on the plate surfaces
was used as the electrodes. Complex dielectric permittivity was
measured with a Tonghui 28 impedance analyzer at the frequency
range from 20 Hz to 1 MHz with an applied electric field of 0.5 V. The
P−E hysteresis loops were recorded on a Sawyer−Tower circuit,
Precision Premier II (Radiant Technologies, Inc.). Pyroelectric
property was measured with an electrometer/high resistance meter
(keithley 6517B) with the cooling rate of 10 K/min. For second
harmonic generation (SHG) experiments, an unexpanded laser beam
with low divergence (pulsed Nd:YAG at a wavelength of 1064 nm, 5
ns pulse duration, 1.6 MW peak power, 10 Hz repetition rate) was
used. The instrument model is FLS 920, Edinburgh Instruments, and
the low temperature system (10−325 K) is DE 202, while the laser is
Vibrant 355 II, OPOTEK. The numerical values of the nonlinear
optical coefficients for SHG have been determined by comparison with
a KDP reference. Solid-state NMR experiments were carried out on a
Varian Infinity plus-300 spectrometer with a 4 mm double-resonance
MAS probe at resonance frequencies of 299.8, 282.0, 75.4, and 96.2
MHz for 1H, 19F, 13C, and 11B, respectively. 13C CP/MAS NMR
spectra were acquired with a recycle delay of 2 s under 6 kHz MAS.
Typically 240 scans were accumulated in the 13C CP/MAS
experiments. 13C−1H dipolar couplings were measured using DIP-
SHIFT experiments14h with a recycle delay of 2 s and at a spinning
speed of 6 kHz. The time domain data were fit to give the apparent
dipole coupling strengths in the DIPSHIFT experiments. The true
13C−1H dipolar couplings were calculated on the basis of apparent
dipole coupling strengths and PMLG14i scaling factor. Both the 19F
and 11B MAS NMR spectrum were obtained through direct
polarization without heteronuclear decoupling at a spinning rate of 5
kHz. Scans (8, 16, and 64) were accumulated to obtain the 19F MAS,
19F static, and 11B MAS NMR spectra, respectively. The recycle delays
were set to 5 and 1 s for 11F and 11B NMR measurements. Typical
radio frequency field strengths were 43−63 kHz for 13C, 50−85 kHz
for 1H, 50 kHz for 19F, and 80 kHz for 11B channel. The 13C chemical
shifts were externally referenced to that of an adamantane CH2 signal
at 38.5 ppm. 11B and 19F NMR chemical shifts were referenced to a 0.1
M aqueous solution of H3BO3 at 19.5 ppm, and a solution of
trifluoroacetic acid, respectively. The sample temperatures were
As a continuation of search for new molecular ferroelectrics,
we have synthesized [(DIPA)([18]crown-6)]BF4 (1) (Scheme
1). As an analogue of ferroelectric [(DIPA)([18]crown-
Scheme 1. Structure Formula of Compound 1
6)]ClO4,11 it crystallizes in isomorphic forms, undergoes
similar phase transitions (Ibam ↔ Pbcn ↔ Pna21 in the
cooling run) and has the similar ferroelectric properties, such as
nearly equal Ps. In [(DIPA)([18]crown-6)]ClO4, the structural
phase transitions are ascribed to the order−disorder transitions
of the crown-ether molecules and/or the ClO4 anions by X-ray
single-crystal structure determinations. However, the dynamics
of the disordered moieties, especially whether or not the
apparently ordered moieties are really static, remains unknown.
By introducing BF4 anion into the system, we can make full use
of solid-state 11B and 19F NMR to determine the dynamics of
the anions. Using variable-temperature solid-state 13C, 11B, and
19F NMR, we obtained more detailed information on the
dynamics, and found it is the slowing down of the motions that
leads to the ferroelectric transition. Here, we report the
excellent ferroelectric properties and the ferroelectric mecha-
nism in 1 by combination of variable-temperature solid-state
NMR and X-ray structure analysis.
10034
dx.doi.org/10.1021/ja503344b | J. Am. Chem. Soc. 2014, 136, 10033−10040

Journal of the American Chemical Society
Article
calibrated by the chemical shifts of Pb(NO3)2 under the similar
condition in the variable-temperature NMR experiments.
RESULTS AND DISCUSSION
■
The synthesis of 1 is similar to that of the ferroelectric [(DIPA)
([18]crown-6)]ClO4. The single crystals are stable over the
investigated temperature range 100−350 K. No decomposition
or deliquescence was seen at room temperature over about six
months. As is known, ferroelectricity is highly dependent on a
certain structure, it may be affected significantly by a minor
structural modification. For example, KD2PO4 (DKDP) has a
Tc 90 K higher than that of KH2PO4 (KDP).1 Another example
is that BaTiO3 is the well-known ferroelectric, whereas CaTiO3
and SrTiO3 are the classic quantum paraelectrics at ambient
pressure.17 For molecular ferroelectrics, preparation of
analogues is one of the simplest methods to search for new
ferroelectricity and to understand the underlying ferroelectric
mechanisms. Beside this, the use of the BF4 anion in this case
facilitates the utilization of solid-state 11B and 19F NMR in the
determination of the dynamics of the anion which has been
believed to play an important role in the ferroelectric transition.
Calorimetric measurements are effective in the study of phase
transitions especially for crystals involving dynamically
disordered molecules. Results of DSC and Cp measurements
are presented in Figure 1. The DSC curves in the heating and
Figure 2. Asymmetric units of 1 in (a) the high-temperature phase
(HTP, 333 K), (b) the intermediate-temperature phase (ITP, 193 K),
and (c) the low-temperature phase (LTP, 93 K). The parts with rose
bonds in (a) and (b) are generated by symmetry operation. H atoms
bonded to the C atoms were omitted for clarity.
Figure 1. DSC and Cp curves for 1, revealing two reversible phase
transitions.
cooling runs clearly reveal two couples of reversible anomalies
at around Tc1 = 305 K and Tc2 = 120 K, respectively. The peak
shapes and the narrow thermal hystereses reveal the continuous
characteristics of the transition. The Cp measurements are
consistent with the DSC measurements. The entropy changes
(ΔS) are 4.416 and 3.622 J/(mol K) for the transitions at Tc1
and Tc2, respectively. From the Boltzmann equation, ΔS = R ln
N, where R is the gas constant and N is the ratio of the numbers
of respective geometrically distinguishable orientations, we
obtain N1 = 1.7 and N2 = 1.5, respectively, indicating that the
transitions are of ordered-disorder type.
Supporting Information Figure S4, ). The asymmetric units
projected along the common b-axis are illustrated in Figure 2.
In the three structures, the DIPA cation appears to be ordered,
and the geometrical parameters are unexceptional and similar.
The DIPA cation makes N−H···O hydrogen bonds to the
[18]crown-6 molecule, forming a supramolecular cation of
DIPA···crown-ether.
The crown-ether has a boat-like shape because of the steric
hindrance between the skeleton of the crown-ether and the two
diisopropyl groups, different from the common 2D symmetrical
disk-shaped structure. Thus, it has a dipole moment nearly
along the C−N bond of the DIPA cation. At 333 K, it appears
to be positionally disordered. Crown-ether molecules in crystals
tend to show rotary motions and thus orientational disorder
because of the high symmetry and difficulty to form relatively
strong intermolecular interactions.7 In 1 at 333 K, such disorder
is probably of continuous (dynamic) positional disorder,12 due
to the molecular rotation, because no obvious molecular
orientation can be determined from the Fourier difference map.
We have modeled the disorder with two mirror-related
orientations which can satisfy the requirement of the mirror
symmetry. Accordingly, the two orientations have equal
populations (0.5:0.5). In ITP and LTP, the crown-ether
molecules are well ordered. Their molecular geometries (Figure
Variable-temperature single-crystal structure determinations
at 333, 193, and 93 K confirm the two sequential structural
phase transitions with space group Ibam in the high-
temperature phase (HTP), Pbcn in the intermediate-temper-
ature phase (ITP), and Pna21 in the low-temperature phase
(LTP), respectively.18 A common origin and similar axial
directions are chosen for HTP and ITP. The relationship
between the two structural cells of ITP and LTP is a93K
≈
−c193 K, b93K ≈ b193 K and c93K ≈ a193 K. The crystal structure
consists of DIPA···crown-ether supramolecular cations and BF4
counteranions (Figure 2). The supramolecular cations stack in
columns along the three cell axes respectively, BF4 anions
occupy the cavities enclosed by the columns (Figure 3 and
10035
dx.doi.org/10.1021/ja503344b | J. Am. Chem. Soc. 2014, 136, 10033−10040

Journal of the American Chemical Society
Article
the orientations change. One remains to be disordered over two
orientation with one common B−F and equal populations
(0.5:0.5). The other becomes ordered with the intramolecular
2-fold rotation axis superimposed with the crystallographic 2-
fold axis. In LTP, all BF4 anions are well ordered. As can be
seen in Figure 2c, the BF4 (I) has a B−F bond along the b-axis,
and thus the 2-fold axis symmetry along the b-axis is broken. It
appears the order−disorder transitions of the crown-ether
molecules and/or BF4 anions play an important role in the
phase transitions. In HTP and ITP, the 2-fold axis and/or
mirror symmetry are satisfied by their orientational disorder.
Their ordering leads to the breaking of the symmetries,
accompanying, for the transition from ITP to LTP, a relative
displacement of 0.02 Å between the centers of the positive and
negative charges within a structural unit cell (assuming that
they are located on the sites of N atoms and B atoms,
respectively). It allows for the appearance of a net dipole-
moment in each structural unit cell. As can be seen from Figure
3 and Supporting Information, Figure S4, the lattices at three
temperatures are similar except for the differences of the
orientational states of the BF4 anions and/or the crown-ether
molecules. For ellipsoid drawings of the structures and their
revealed dynamic information, see Figure S3 and the
supplementary structural analysis in the Supporting Informa-
tion.
In terms of polarization response, the two transitions from
Ibam to Pbcn to Pna21 are of paraelectric-to-paraelectric and
paraelectric-to-ferroelectric transitions, respectively. For the
former, there should be a small dielectric anomaly at Tc1, while
for the latter, there should be a larger one at Tc2. We confirmed
this by measurements of variable-temperature dielectric
constants. The real part (ε′) of the complex constant as a
function of temperature, measured on a single-crystal sample
along the polar c-axis of LTP, is illustrated in Figure 4. As
Figure 3. Comparison of the packing diagrams projected along the
common b-axis of 1 at (a) 333, (b) 193, and (c) 93 K, showing the
similarities of the lattices and the differences of the orientational states
of the crown-ether molecules and the BF4 anions.
Figure 4. Real part of the complex dielectric constant of 1 as a
function of temperature at several frequencies, measured along the
polar c-axis of LTP. Insert: plot of 1/ε′ vs temperature in the vicinity
of Tc2.
S4, Supporting Information) and orientations are essentially
identical in the two phases. The two crystallographically
independent crown-ethers in LTP have the pseudo 2-fold
rotation symmetry in the b-axis direction. Their dipole
moments contribute little to Ps, in that their components
along the polar c-direction are opposite and cancel each other
out. The same is true of the DIPA cations.
The tetrahedral BF4 anion behaves like the crown-ether
molecule. In HTP, it is located on the crystallographic 2-fold
axis with equal orientational disorder populations (0.5:0.5). In
ITP, the two asymmetric half ions still lie on the 2-fold axes, but
expected, the anomaly at around 120 K is much larger than that
at around 305 K. The large dielectric anomaly indicates that the
ferroelectric transition is proper, that is, the large dielectric
anomaly is due to the appearance of Ps. Here, we focus on the
dielectric behavior in the vicinity of the paraelectric-to-
ferroelectric transition temperature (Tc2). In the vicinity of
Tc2, the temperature dependence of ε′ shows Curie−Weiss
behavior, ε′ = Cpara/(T − T0) (T > Tc2) or Cferro/(T0′ − T) (T
< Tc2). The fitted Curie−Weiss constants at 500 Hz are Cferro
=
29.6 K, Cpara = 34.2 K, Curie−Weiss temperatures T0′ = 120.5
10036
dx.doi.org/10.1021/ja503344b | J. Am. Chem. Soc. 2014, 136, 10033−10040

Journal of the American Chemical Society
Article
and T0 = 119.7 K for the ferroelectric LTP and paraelectric ITP,
respectively (the inset of Figure 4). The relative small Curie−
Weiss constants reveal the order−disorder feature of the
transition.19 The ratio Cpara/Cferro of 1.16 is smaller than 4.0,
suggesting a second-order ferroelectric transition.20 The
Curie−Weiss temperatures T0 and T0′ are close to Tc2, also
indicating that the ferroelectric transition is continuous.
Different from other polar materials, ferroelectrics must have
switchable Ps. This behavior is usually recorded as polar-
ization−electric field (P−E) hysteresis loops. The loops for a
crystal of 1 recorded at various temperatures are shown in
Figure 7. Above 128 K, the response is almost linear, consistent
Since the structural analysis and dielectric measurements
reveal a centrosymmetric-to-noncentrosymmetric transition at
Tc2, there should be a temperature-dependent second harmonic
generation (SHG) response, that is, the second-order nonlinear
optical coefficient (χ(2)) should be zero in ITP, while χ(2)
should be nonzero in LTP (because only solids with
noncentrosymmetric space group display frequency doubling
behavior). As shown in Figure 5, no frequency doubling occurs
Figure 7. Ferroelectric hysteresis loops of 1 measured along the polar
c-axis of LTP at various temperatures.
with the behavior for a paraelectric phase. At 118 K, a flat
hysteresis loop appears. Ps and remnant polarization (Pr)
gradually increase with temperature decreasing. A perfect
rectangular loop is observed at 100 K with Ps and Pr reaching
about 0.3 and 0.25 μC/cm2, respectively, and a coercive field
being of the order of 12 kV/cm. The observation of the
hysteresis loops provides direct evidence for the ferroelectricity
in 1.
To provide an insight into the order−disorder mechanism,
we performed solid-state NMR investigation on the dynamics
of 1. Figure 8 shows the variable-temperature 11B and 19F MAS
NMR spectra of 1. The 11B and 19F chemical shift of the BF4
anions is centered at −1 and −149 ppm, respectively, which are
comparable to those in the relevant compounds.21 As shown in
Figure 8a and Supporting Information Figure S6, two kinds of
Figure 5. Variable-temperature second-order nonlinear optical
coefficients of the powdered sample of 1. Inset: relative intensity of
SHG signal as a function of wavelength at various temperatures.
above Tc2, indicating that 1 in ITP is centrosymmetric. As
temperature decreases, χ(2) shows a detectable value at around
Tc2, suggesting that the crystals becomes noncentrosymmetric
in LTP. The gradual increase of χ(2) from Tc2 is consistent with
the character for a second-order transition.
According to the Landau phenomenological theory on
proper ferroelectric transitions, both the SHG and the dielectric
response in the vicinity of Tc2 are essentially associated with the
appearance of Ps. The measured Ps in a cooling process as a
function of temperature is shown in Figure 6. As expected, Ps is
zero in ITP. It begins to gradually increase at around 130 K,
and reaches 0.35 μC/cm2 at 80 K. The behavior of gradual
increase of Ps further confirms the continuous character of the
ferroelectric transition.
Figure 6. Ps of 1 as a function of temperature by integrating
pyroelectric current, measured along the polar c-axis of LTP.
Figure 8. (a) Variable-temperature 11B and (b) 19F MAS NMR spectra
of 1. Asterisks denote spinning sidebands.
10037
dx.doi.org/10.1021/ja503344b | J. Am. Chem. Soc. 2014, 136, 10033−10040

Journal of the American Chemical Society
Article
11B NMR signals could be clearly resolved above 312 K, being
indicative of the rapid tumbling of the BF4 anions with different
motion rates in HTP. The sharp 11B NMR signal can be
associated with the fast motion of the BF4 anions, whereas the
broad component having spinning sidebands corresponds to
the intermediate motion of the BF4 anions. As temperature
varies from 302 to 244 K, 11B NMR line-width (Figure S5a,
Supporting Information) gradually increases, indicating the
slowing down of BF4 anion motion. No significant change can
be discerned upon further decreasing the temperature to 121 K.
The 19F spectra (Figure 8b) are analogous to the 11B spectra
except that there is a pronounced change in line-width (Figure
S5b, Supporting Information) at below 128 K, indicating that
the slow motion of BF4 anions is largely frozen below Tc2. The
continuous resonance peak broadening is ascribed to the
enhancement of the homo- and heteronuclear dipolar
interactions resulting from the slowing down of the tumbling
motion of the BF4 anion. Moreover, the variable-temperature
19F static NMR spectra also reveal the similar dynamics of the
BF4 cation. (Figure S5c, Supporting Information).
To reveal the dynamics of the DIPA and [18]crown-6, we
carried out 13C CP/MAS NMR experiments. Additionally,
DIPSHIFT experiments are further employed to quantitatively
characterize the molecular motions of the DIPA and [18]-
crown-6 at various temperatures. In the DIPSHIFT experi-
ments, the simulated one-bond C−H coupling strength relative
to rigid limit (23.9 kHz), namely order parameter, can directly
manifest the regional molecular motion. The order parameter
adopts a value ranging from 0 to 1. Figure 9 shows the variable-
temperature 13C CP/MAS NMR spectra and the obtained
order parameters for three designated carbon sites. The
chemical shifts assignments are indicated in Figure 9a.
As shown in Figure 9a, no significant change can be found for
the aromatic 13C signal over the whole examined temperature
range. The serious peak congestion in the aromatic carbon
region prevents us from extracting a separate aromatic 13C
signal for data analysis in the DIPSHIFT experiments. Thus, we
chose the 13C signals of the side-chain CH group and terminal
methyl groups to investigate the dynamic behavior of the DIPA.
As shown in Figure 9b, the order parameters of the CH group
are determined to be ca. 0.92 with a slight fluctuation from 341
to 121 K. It reveals that the CH group keeps mostly rigid over
the whole temperature range, due to steric hindrance between
the skeleton of the crown-ether and the two diisopropyl groups.
This is quite different from that previously reported for 4-
methoxyanilinium perrhenate [18]crown-6, in which the guest
4-methoxyanilinium cation performs a pendulum-like motion
whose slowing down leads to a ferroelectric phase transition.9
For the 13C signal of terminal methyl groups, a significant
change is the appearance of two discrete resonance peaks at
138 K, which is probably due to the doubling of the number of
the crystallographically independent DIPA in LTP relative to
that in ITP. The jump of the order parameter of the CH3 peak
at 22 ppm (Figure 9c) from ca. 0.41 in the range 341−138 K to
0.9 at below 128 K indicates the dynamic change from the
classic three-site jump to a motion restricted state. The other
peak at 25 ppm reveals the similar dynamic behavior at below
138 K.
The most significant changes in the 13C CP/MAS NMR
spectra lies in the region for the CH2 group of the [18]crown-6.
In the temperature range 341−312 K, the narrow line-width
and small order parameter (less than 0.24) indicate a fast
“merry-go-round” (in-plane) rotation of the [18]crown-6 in
Figure 9. (a) Variable-temperature 13C CP/MAS NMR spectra of 1
and (b, c, d) temperature-dependent order parameters for the CH and
CH3 group of the DIPA and the CH2 group of the [18]crown-6.
10038
dx.doi.org/10.1021/ja503344b | J. Am. Chem. Soc. 2014, 136, 10033−10040

Journal of the American Chemical Society
Article
HTP. The signal becomes very broad at 302 K and almost
indiscernible at 293−264 K, which is a classic signature for the
intermediate time-scale motion of the [18]crown-6 molecule.
Obviously, the [18]crown-6 undergoes a transition from the
fast motion to the intermediate motion at around Tc1. The
order parameters at 293−264 K cannot be accurately
determined due to the almost invisible signals, while the
motion frequency is normally comparable to the C−H dipolar
coupling and proton decoupling strength (20 000 to 62 500
s−1).14f,g Upon further decreasing the temperature to 244 K, the
CH2 signal comes back with a greater order parameter of 0.81,
which can be ascribed to the transition of the intermediate
time-scale motion into a slow motion of the [18]crown-6.
According to the method22 proposed by deAzevedo et al., the
in-plane rotation rates of the [18]crown-6 were estimated to be
in the range of 1 × 104 to 5 × 103 s−1 at 244−225 K (Figure S7,
Supporting Information). As temperature decreases to 225−
196 K, the order parameter approaches the rigid limit value,
indicating considerable rigidity of the [18]crown-6 (Figure 9d).
Further decreasing the temperature to 177 K leads to signal
attenuation. It probably suggests that it comes into a transition
state similar to that with intermediate time-scale motion.
Finally, as the temperature decreases to 138 K, the CH2 signal
restores and is split into three distinct peaks. The chemical
shifts distribution of the CH2 group could be probably ascribed
to the symmetry breaking in the ferroelectric phase, that is, the
nonequivalent shielding impacts from neighboring 19F nuclei to
13C sites in [18]crown-6 result in the chemical shifts
distribution of CH2 group in LTP. The order parameters
reaching the limit at 138−121 K indicate the rigid feature of the
[18]crown-6.
motions. At the first stage (293−264 K), both the [18]crown-6
and BF4 anion demonstrate a kinds of intermediate motion.
Below 244 K, both simultaneously come into a slow motion.
The BF4 anion retains the slow motion until the phase
transition at Tc2, whereas the [18]crown-6 undergoes a
transition state similar to that with intermediate time-scale
motion from 177 to 147 K, which might be favorable to a
smooth structural transition from ITP to LTP. Their
synchronous change in the dynamic behavior in a wide range
341−244 K indicates that there is a synergetic interaction
between BF4 cation and [18]crown-6 in facilitating the phase
transitions. Below Tc2, the motions appear to be frozen, leading
to a completely ordered structure. As shown in Figure S3b,c
(Supporting Information), the single orientation of the BF4 (I)
cannot satisfy the 2-fold rotation symmetry, giving rise to a
structurally symmetry-breaking phase transition.
CONCLUSIONS
■
We have constructed and systematically characterized an
[18]crown-6-based host−guest inclusion ferroelectric com-
pound [(DIPA)([18]crown-6)]BF4. By a combination of
variable-temperature solid-state NMR and X-ray single-crystal
structure deteminations, we disclosed that the two sequential
phase transitions are induced by the changes of the dynamics of
[18]crown-6 molecules and BF4 anions from the fast motion in
HTP to the intermediate and slow motion in ITP and then to a
rigid state in LTP. The molecular motions do not involve
reorientation of the molecular dipole moments. However, their
slowing down still leads to the structurally symmetry-breaking
phase transitions. The transition at Tc2 allows for a relative
displacement between the cationic and anionic sublattices, and
thus, the appearance of a net dipole-moment in each structural
unit cell. The observation of such a symmetry transition
deepens the understanding of the ferroeletric origin in the
crown-ether-based host−guest-type ferroelectrics that was
previously believed to be related to rotating molecular dipoles.
Combining the single-crystal structural analysis and solid-
state NMR observations, we summarize the dynamics of 1 and
the phase transition mechanism as follows. The DIPA cation
does not perform reorientational motions over the whole
temperature range, though the terminal methyl groups show a
three-site jump motion in HTP and ITP. Above Tc1 in HTP,
the [18]crown-6 molecule performs the fast “merry-go-round”
rotation among two or more orientations (Figure 10). The
ASSOCIATED CONTENT
■
S
* Supporting Information
Supplementary Figures S1−S7 and discussion for this article.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
■
Corresponding Author
xiongrg@seu.edu.cn
Author Contributions
§H.-Y.Y. and S.-H.L. contributed equally to this work.
Figure 10. Schematic drawing of the dynamics of the [18]crown-6
revealed by variable-temperature solid-state NMR measurement and
X-ray single-crystal structure determinations.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
rotation does not involve a reorientation of the molecular
dipole moment because it is along the rotation axis. The BF4
anions tumble including fast and intermediate motions between
at least two orientations. In ITP, the motions slow down,
leading to a crystallographic single orientation of the [18]-
crown-6 and accordingly the breaking of the mirror symmetry
(Figure S3a,b, Supporting Information), and the coexistence of
orientational disorder and order of the BF4 anions in the ratio
of 1:1. These orderings do not preclude the molecular/anionic
This work was supported by the Project 973 (2014CB93103)
and NSFC (21290172, 91222101 and 21371032).
REFERENCES
■
(1) Lines, M. E.; Glass, A. M. Principles and Application of
Ferroelectrics and Related Materials; Oxford University Press:
Clarendon, UK, 1977.
(2) Lallart, M., Ed. FerroelectricsApplications; InTech: Rijeka,
Croatia, 2011.
10039
dx.doi.org/10.1021/ja503344b | J. Am. Chem. Soc. 2014, 136, 10033−10040

Journal of the American Chemical Society
Article
(3) (a) Jarowski, P. D.; Houk, K. N.; Garcia-Garibay, M. A. J. Am.
Chem. Soc. 2007, 129, 3110. (b) Gould, S. L.; Tranchemontagne, D.;
Yaghi, O. M.; Garcia-Garibay, M. A. J. Am. Chem. Soc. 2008, 130, 3246.
(c) Karlen, S. D.; Reyes, H.; Taylor, R. E.; Khan, S. I.; Hawthorne, M.
F.; Garcia-Garibay, M. A. Proc. Natl. Acad. Sci. U.S.A. 2010, 107, 14973.
(d) Lemouchi, C.; Vogelsberg, C. S.; Zorina, L.; Simonov, S.; Batail,
P.; Brown, S.; Garcia-Garibay, M. A. J. Am. Chem. Soc. 2011, 133, 6371.
(e) Rodríguez-Molina, B.; Farf’an, N.; Romero, M.; M’endez-Stivalet,
J. M.; Santillan, R.; Garcia-Garibay, M. A. J. Am. Chem. Soc. 2011, 133,
7280. (f) Kuzmanich, G.; Simoncelli, S.; Gard, M. N.; Spanig, F.;
Henderson, B. L.; Gudi, D. M.; Garcia-Garibay, M. A. J. Am. Chem. Soc.
2011, 133, 17296. (g) Lemouchi, C.; Vogelsberg, C. S.; Zorina, L.;
Simonov, S.; Batail, P.; Brown, S.; Garcia-Garibay, M. A. J. Am. Chem.
Soc. 2011, 133, 13765.
(17) Lemanov, V. V.; Sotnikov, A. V.; Smirnova, E. P.; Weihnacht,
M.; Kunze, R. Solid State Commun. 1999, 110, 611.
(18) Crystal data for 1 at 333 K: [(C12H20N)(C12H24O6)](BF4), Mr
= 529.41, orthorhombic, Ibam, a = 17.125(17) Å, b = 17.685(16) Å, c
= 19.071(16) Å, V = 5776(9) Å3, Z = 8, Dc = 1.218 g cm−3, R1 (I >
2σ(I)) = 0.1144, wR2 (all data) = 0.2982, μ = 0.101 mm−1, S = 1.205.
At 193 K: orthorhombic, Pbcn, a = 16.868(5) Å, b = 17.468(5) Å, c =
18.823(5) Å, V = 5546(3) Å3, Z = 8, Dc = 1.268 g cm−3, R1 (I > 2σ(I))
= 0.0781, wR2 (all data) = 0.1772, μ = 0.105 mm−1, S = 1.160. At 93 K:
orthorhombic, Pna21, a = 18.757(5) Å, b = 17.392(5) Å, c = 16.703(4)
Å, V = 5449(2) Å3, Z = 8, Dc = 1.291 g cm−3, R1 (I > 2σ(I)) = 0.0485,
wR2 (all data) = 0.1076, μ = 0.107 mm−1, S = 1.078.
(19) Nakamura, E.; Mitsui, T.; Furuichi, J. J. Phys. Soc. Jpn. 1963, 18,
1477.
(20) Draegert, D. A.; Singh, S. Solid State Commun. 1971, 9, 595.
(21) (a) Lee, S.-I.; Saito, K.; Kanehashi, K.; Hatakeyama, M.; Mitani,
S.; Yoon, S.-H.; Korai, Y.; Mochida, I. Carbon 2006, 44, 2578.
(4) (a) Vujosevic, D.; Muller, K.; Roduner, E. J. Phys. Chem. B 2006,
110, 8598. (b) Czarnecki, P. Phys. Rev. B 1998, 57, 3326.
(5) (a) Przesławski, J.; Czapla, Z. J. Phys.: Condens. Matter. 2006, 18,
5517. (b) Jakubas, R.; Piecha, A.; Pietraszko, A.; Bator, G. Phys. Rev. B
(b) Pillinger, M.; Nunes, C. D.; Vaz, P. D.; Valente, A. A.; Gonca
̧
lves, I.
S.; P. Ribeiro-Claro, J. A.; Rocha, J.; Carlos, L. D.; Kuhn, F. E. Phys.
̈
2005, 72, 104107. (c) Pajak
Małuszynska, H.; Fojud, Z. Phys. Rev. B 2004, 69, 132102. (d) Czapla,
Z.; Dacko, S.; Kosturek, B.; Waskowska, A. Phys. Status Solidi b 2005,
242, R122.
́
̧ , Z.; Czarnecki, P.; Szafranska, B.;
Chem. Chem. Phys. 2002, 4, 3098. (c) Wiench, J. W.; Michon, C.;
Ellern, A.; Hazendonk, P.; Iuga, A.; Angelici, R. J.; Pruski, M. J. Am.
Chem. Soc. 2009, 131, 11801. (d) Chifotides, H. T.; Giles, I. D.;
Dunbar, K. R. J. Am. Chem. Soc. 2013, 135, 3039.
́
́
(6) (a) Sante, D. D.; Stroppa, A.; Jain, P.; Picozzi, S. J. Am. Chem. Soc.
2013, 135, 18126. (b) Fu, D.-W.; Zhang, W.; Cai, H.-L.; Zhang, Y.;
Ge, J.-Z.; Xiong, R.-G.; Huang, S. P. D.; Nakamura, T. Angew. Chem.,
Int. Ed. 2010, 49, 6608. (c) Xu, G.-C.; Ma, X.-M.; Zhang, L.; Wang, Z.-
M.; Gao, S. J. Am. Chem. Soc. 2010, 132, 9588. (d) Jain, P.;
Ramachandran, V.; Clark, R. J.; Zhou, H. D.; Toby, B. H.; Dalal, N. S.;
Kroto, H. W.; Cheetham, A. K. J. Am. Chem. Soc. 2009, 131, 13625.
(7) Akutagawa, T.; Nakamura, T. Dalton Trans. 2008, 6335.
(8) Fu, D.-W.; Zhang, W.; Cai, H.-L.; Zhang, Y.; Ge, J.-Z.; Xiong, R.-
G.; Huang, S. P. D. J. Am. Chem. Soc. 2011, 133, 12780.
(9) Fu, D.-W.; Cai, H.-L.; Li, S.-H.; Ye, Q.; Zhou, L.; Zhang, W.;
Zhang, Y.; Deng, F.; Xiong, R.-G. Phys. Rev. Lett. 2013, 110, 257601.
(10) Akutagawa, T.; Koshinaka, H.; Sato, D.; Takeda, S.; Noro, S.-I.;
Takahashi, H.; Kumai, R.; Tokura, Y.; Nakamura, T. Nat. Mater. 2009,
8, 342.
(22) deAzevedo, E. R.; Saalwachter, K.; Pascui, O.; de Souza, A. A.;
Bonagamba, T. J.; Reichert, D. J. Chem. Phys. 2008, 128, 104505.
(11) Zhang, Y.; Ye, H.-Y.; Fu, D.-W.; Xiong, R.-G. Angew. Chem., Int.
Ed. 2014, 53, 2114.
(12) Muller, P.; Herbst-Irmer, R.; Spek, A.; Schneider, T.; Sawaya, M.
Crystal Structure RefinementA Crystallographer’s Guide to SHELXL;
Oxford University Press: New York, 2006; pp 56−65.
(13) Ratcliffe, C. I.; Ripmeester, J. A.; Buchanan, G. W.; Denike, J. K.
J. Am. Chem. Soc. 1992, 114, 3294.
(14) (a) Fyfe, C. A. Solid State NMR for Chemists; C. F. C. Press,
Ontario, Canada, 1983. (b) Yan, S.; Suiter, C. L.; Hou, G.; Zhang, H.;
Polenova, T. Acc. Chem. Res. 2013, 46, 2047. (c) Tang, M.; Comellas,
G.; Rienstra, C. M. Acc. Chem. Res. 2013, 46, 2080. (d) Hong, M.;
Zhang, Y.; Hu, F. Annu. Rev. Phys. Chem. 2012, 63, 1. (e) Brown, S. P.;
Spiess, H. W. Chem. Rev. 2001, 101, 4125. (f) Cady, S. D.; Goodman,
C.; Tatko, C. D.; DeGrado, W. F.; Hong, M. J. Am. Chem. Soc. 2007,
129, 5719. (g) Rothwell, W. P.; Waugh, J. S. J. Chem. Phys. 1981, 74,
2721. (h) Munowitz, M. G.; Griffin, R. G.; Bodenhausen, G.; Huang,
T. H. J. Am. Chem. Soc. 1981, 103, 2529. (i) Vinogradov, E.; Madhu, P.
K.; Vega, S. Chem. Phys. Lett. 1999, 314, 443.
(15) (a) Kuzmanich, G.; Vogelsberg, C. S.; Maverick, E. F.; Netto-
Ferreira, J. C.; Scaiano, J. C.; Garcia-Garibay, M. A. J. Am. Chem. Soc.
2012, 134, 1115. (b) Lemouchi, C.; Meziere, C.; Zorina, L.; Simonov,
S.; Rodriguez-Fortea, A.; Canadell, E.; Wzietek, P.; Auban-Senzier, P.;
Pasquier, C.; Giamarchi, T.; Garcia-Garibay, M. A.; Batail, P. J. Am.
Chem. Soc. 2012, 134, 7880. (c) Vogelsberg, C. S.; Garcia-Garibay, M.
A. Chem. Soc. Rev. 2012, 41, 1892. (d) de Loera, D.; Stopin, A.; Garcia-
Garibay, M. A. J. Am. Chem. Soc. 2013, 135, 6626. (e) Rodriguez-
Molina, B.; Perez-Estrada, S.; Garcia-Garibay, M. A. J. Am. Chem. Soc.
2013, 135, 10388.
(16) (a) Buchanan, G. W.; Kirby, R. A.; Ripmeester, J. A.; Ratcliffe,
C. I. Tetrahedron Lett. 1987, 28, 4783. (b) Magdalena, H. Nature 2004,
427, 597.
10040
dx.doi.org/10.1021/ja503344b | J. Am. Chem. Soc. 2014, 136, 10033−10040