Ferroelastic phase transition and dielectric anomalies in 2,4,6-trimethylanilinium perchlorate

Article abstract of DOI:10.1039/c2jm30581j

The crystal structure of the crystalline 1:1 complex of 2,4,6-trimethylaniline with perchloric acid, (C₉H₁₄N) ⁺·ClO₄^- (1), was determined at 25 °C and 110 °C, respectively. The crystal structural analysis of both temperatures show that 1 undergoes a reversible phase transition from triclinic P1 (No.2) to monoclinic P2₁/m (No.11) with a distinct change of the cell parameters. It is a ferroelastic phase transition with an Aizu notation of 2/mF1. DSC (differential scanning calorimetry) measurement confirms that this compound undergoes a reversible first-order solid-state structural phase transition at about 48 °C with a thermal hysteresis of 3 °C. Usually, phase transition is accompanied by an anomaly of the dielectric constant near phase transition temperature (T_c). The DSC measurement, temperature-dependent single-crystal X-ray diffraction and dielectric studies reveal that 1 displays structural phase transitions and distinctly step-like dielectric anomalies at 48 °C probably supporting that this phase transition is of the first-order.

Full text of DOI:10.1039/c2jm30581j

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links <
C
Journal of
Materials Chemistry
Cite this: J. Mater. Chem., 2012, 22, 9841
www.rsc.org/materials
PAPER
Ferroelastic phase transition and dielectric anomalies in
2,4,6-trimethylanilinium perchlorate†
a
Yi Zhang, Kunio Awaga, Hirofumi Yoshikawa and Ren-Gen Xiong
b
b
a
*
Received 31st January 2012, Accepted 22nd March 2012
DOI: 10.1039/c2jm30581j
The crystal structure of the crystalline 1 : 1 complex of 2,4,6-trimethylaniline with perchloric acid,
(C9H14N)+$ClO4ꢀ (1), was determined at 25 ꢁC and 110 ꢁC, respectively. The crystal structural analysis
ꢀ
of both temperatures show that 1 undergoes a reversible phase transition from triclinic P1 (No.2) to
monoclinic P21/m (No.11) with a distinct change of the cell parameters. It is a ferroelastic phase
ꢀ
transition with an Aizu notation of 2/mF1. DSC (differential scanning calorimetry) measurement
confirms that this compound undergoes a reversible first-order solid-state structural phase transition at
about 48 ꢁC with a thermal hysteresis of 3 ꢁC. Usually, phase transition is accompanied by an anomaly
of the dielectric constant near phase transition temperature (Tc). The DSC measurement, temperature-
dependent single-crystal X-ray diffraction and dielectric studies reveal that 1 displays structural phase
ꢁ
transitions and distinctly step-like dielectric anomalies at 48 C probably supporting that this phase
transition is of the first-order.
Hooke’s law is not valid in cellular structures like wood and in
1. Introduction
elastomers like rubber. Whatever, the recoverable strain in
materials of either the linear or the nonlinear type is usually
a single-value function of the small applied stress.11 However, the
strain in the ferroelastic crystals is a double-value function of
applied stress and show hysteresis loops, which are similar to
ferroelectric hysteresis loops. Such hysteresis is the basic char-
acter of the ferroelastic effect which is defined as the mechanical
switching between either different crystal structures (e.g. cubic to
tetragonal) or different orientations (a ‘twin’ phase). Although
the atomic displacement mechanism captures the essential reason
why a hysteresis can be observed, the actual switching mecha-
nism is still elusive. For example, the energy stored in one
hysteresis loop is much smaller than the energy required to switch
all atoms from one configuration to another, which indicates that
not all atoms take part in the switching process simultaneously.12
The local ferroelastic switching may produce a high-energy
domain state, leading to a relaxation of the smaller (switched)
domain back to the original ferroelastic state.13,14 Domain
structure is an important ingredient in functionality of ferroe-
lastic materials. It has impact on their nonlinear optical prop-
erties, dielectric permittivity, and polarization switching
phenomena. Since domain boundaries in ferroelastic materials
can simultaneously play the role of the ferroelastic walls, such
domain walls also strongly influence the electromechanical
material properties: they facilitate switching of spontaneous
polarization and spontaneous deformation. The domain struc-
ture also provides additional degrees of freedom for tuning the
material properties. Further development of domain engineering
strategies requires deeper understanding of the physics of the
ferroelastic domain wall itself.1
For the last decade, special attention has been paid to ferroelastic
materials due to their promising potential applications. Ferroe-
lastics are potentially useful materials for nanotechnological
applications as the ferroelastic twin patterns occur exactly on the
nanometer scale and ferroelastic crystals may hold the key for
templating electronic nanostructures.1,2 The shape memory effect
and superelasticity are also manifestations of ferroelasticity.3–5
Nitinol (nickel titanium), a common ferroelastic alloy, can
display either superelasticity or the shape-memory effect at room
temperature, depending on the nickel/titanium ratio. Particu-
larly, multiferroics, where (anti-) ferromagnetic, ferroelectric and
ferroelastic order parameters coexist,6–10 have become one of the
most popular research topics, which enable manipulation of
magnetic ordering and/or polarization ordering by an external
stress through switching of the ferroelastic state.1
Ferroelasticity is a phenomenon in which a material may
exhibit a spontaneous strain while the strain can be reversed by
stress. When stresses are applied to objects, a linear elastic
response is often observed for a large number of crystal, glass and
many cross-linked polymers, which is called Hooke’s law.
aOrdered Matter Science Research Center, Southeast University, Nanjing
211189, P. R. China. E-mail: xiongrg@seu.edu.cn; Fax: (+86)-25-
52090626
bResearch Center for Materials Science and Department of Chemistry,
Nagoya University, Furo-cho, Chikusa-ku, Nagoya 464-8602, Japan
† Electronic supplementary information (ESI) available: IR, PXRD,
crystal data and hydrogen-bond geometry of compound 1. CCDC
reference numbers 865148–865149. For ESI and crystallographic data
in CIF or other electronic format see DOI: 10.1039/c2jm30581j
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 9841–9845 | 9841

View Article Online
The ferroelastic properties are sensitive to thermodynamic
parameters such as temperature T, pressure P or chemical
composition N. Most ferroelastic materials show phase transi-
tions from a paraelastic phase to a ferroelastic phase with
decreasing temperature. However, the control of such ferroe-
lastic switching in a single-phase system has been a significant
challenge as elastic interactions tend to destabilize small switched
volumes.13,14 As it is rather a difficult experimental task to
measure ferroelastic hysteresis with any acceptable degree of
accuracy, it has become customary to call a material ‘ferroelastic’
if a phase transition occurs (or may be thought to occur) which
may conceivably generate ferroelasticity.12
automatic impedance TongHui 2828 Analyzer in the frequency
range 10–1000 kHz. The measuring AC voltage was 1 V.
2.4 DSC measurement
DSC measurements were performed by heating and cooling the
samples (20.8 mg) in the temperature range of 30–110 ꢁC on
PerkinElmer Diamond DSC instrument. The measurements were
carried out under nitrogen at atmospheric pressure in aluminum
crucibles with a heating rate of 5 C minꢀ1
.
ꢁ
2.5 Specific heat measurement
The measurement procedures can be referred to in the paper on
powdered samples or polycrystals.10
2. Experimental
2.1 Synthesis
3. Results and discussion
The compound 1 was prepared by the dropwise addition of
concentrated perchloric acid to a solution of 2,4,6-trimethylan-
line (1.35 g, 0.01 mol) in absolute ethanol (10 ml). The resulting
white precipitate was filtered. High quality single crystals were
obtained by evaporation from aqueous solution at room
temperature. In the IR spectra (see Supporting Information,
Fig. 1S†), several peaks at approximately 1580 cmꢀ1 and 1600
cmꢀ1 indicate the presence of benzene ring, while the peak at
3.1 Thermal analysis
DSC measurement is usually used to detect whether the
compound displays a reversible phase transition triggered by
temperature or not. The DSC measurement for 1 is performed in
ꢁ
the temperature region of 30–110 C. The heating and cooling
curves of 1 show one pair of reversible peaks, an exothermic peak
ꢁ
ꢁ
at 44.7 C and an endothermic peak at 47.7 C (Fig. 1a). These
observed peaks represent a reversible phase transition with
a thermal hysteresis of 3 ꢁC. The sharp peaks reveal the
discontinuous character of the phase transition, being indicative
of a first-order phase transition. Thus, for warming process,
DH ¼ 13.52 J gꢀ1; for cooling process, DH ¼ 12.05 J gꢀ1, the
averaged DH is estimated to be 12.79 J gꢀ1. For warming process,
TC ¼ 320.7 K, DS ¼ (12.79/320.7) J gꢀ1 Kꢀ1 ¼ 3.988 ꢂ 10ꢀ2 J gꢀ1
Kꢀ1 while for the cooling process, TC ¼ 317.7 K, DS ¼
3450 cmꢀ1 may be attributed to the –NH3 and the peak at
+
1375 cmꢀ1 may be attributed to the –CH3. The anions show
strong infrared bands at 1100 cmꢀ1 (ClO4ꢀ). The phase purity of
1 at room temperature was verified by powder XRD (PXRD)
pattern (Fig. 2S†).
2.2 Single-crystal X-ray crystallography
X-Ray diffraction experiments were carried out on the
compound using a Rigaku Saturn 924 diffractometer with Mo-
(12.79/317.7) J gꢀ1 Kꢀ1 ¼ 4.026 ꢂ 10ꢀ2 J gꢀ1 Kꢀ1
.
It is well-known that the shape of the specific heat plot will
further confirm whether there is a phase transition occurring and
the phase transition character. Fig. 1b shows that a peak appears
at ca. 48 ꢁC, surely manifesting the presence of a phase transition
as a typical second-order one (non-l-shape), like that of trigly-
cine sulfate (TGS). However, it is very strange that the temper-
ature-dependence of Cp goes down not like normal going up.
This phenomenon will be under study (experimental results
repeated 2–3 times, reproducible).
ꢁ
Ka radiation (l ¼ 0.71073 A) at various temperatures. Data
processing including empirical absorption correction was per-
formed using the Crystalclear software package (Rigaku, 2005).
The structures were solved using direct methods and successive
Fourier difference synthesis (SHELXS-97), and refined using the
full-matrix least-squares method on F2 using the SHELXLTL
software package (Sheldrick, 2008). Non-H atoms were refined
anisotropically using all reflections with I > 2s(I). All H atoms
were placed in calculated positions and refined using ‘‘riding’’
model for aromatic ring H atoms with Uiso ¼ 1.2 Ueq (C), methyl
H atoms with Uiso ¼ 1.5 Ueq (C) and ammonium H atoms with
Uiso ¼ 1.5 Ueq (N). The asymmetric units and the packing views
were drawn with DIAMOND (Brandenburg and Putz, 2005)
Visual Crystal Structure Information System Software. Crys-
tallographic data and details of the data collection and refine-
3.2 Crystal structure
The phase transition was confirmed by the determination of
variable-temperature crystal structures of the compound 1. In
ꢁ
ꢁ
ment for 25 C and 110 C are summarized in Table 1 (Table
1S†).
2.3 Dielectric measurement
For the complex dielectric measurement the sample was made
with the powder sample-pressed pellet of the compound 1. The
pressed-powder pellet deposited with carbon conducting glue
was used for dielectric studies. Complex dielectric permittivity
3 (3 ¼ 30 ꢀ i300) was performed on pure compound 1 powder using
Fig. 1 (a)DSCmeasurementresultof1. (b)Specificheat(Cp)diagramof1.
This journal is ª The Royal Society of Chemistry 2012
9842 | J. Mater. Chem., 2012, 22, 9841–9845

View Article Online
Fig. 2 (a) Compound 1 with the atomic numbering scheme at 25 ꢁC. (b)
Packing diagrams of 1 at 25 ꢁC.
Fig. 4 (a) Compound 1 with the atomic numbering scheme at 110 ꢁC
(symmetry codes: x, 0.5 ꢀ y, z). (b) Packing diagrams of 1 at 110 ꢁC.
Fig. 3 The 2,4,6-trimethylanilinium ions stack head-to-tail with the C–
H/p interactions at 25 ꢁC.
order to understand the structural changes of 1, the room-
temperature (25 ꢁC) structure and the high-temperature (110 ꢁC)
structure have been measured, using the same crystal. The room
temperature phase (RTP) structure is triclinic with a space group
Fig. 5 Symmetry breaking process in 1 during temperature cooling
down and warming up processes.
ꢀ
ꢁ
ꢁ
ꢁ
of P1, and cell parameters of a ¼ 7.1581(14) A, b ¼ 8.5937(17) A,
ꢁ
ꢁ
ꢁ
c ¼ 9.0642(18) A, a ¼ 88.74(3) , b ¼ 79.19(3) , g ¼ 87.02(3) , v ¼
3
ꢁ
546.90(19) A . The high temperature phase (HTP) structure,
carried out at 110 ꢁC, is monoclinic and its space group turns to
ꢁ
P21/m, with cell parameters of a ¼ 8.9084(18) A, b ¼ 7.2253(18)
ꢁ
ꢁ
ꢁ
3
ꢁ
ꢁ
ꢁ
A, c ¼ 9.172(3) A, a ¼ 90 , b ¼ 95.55 , g ¼ 90 , v ¼ 587.6(3) A .
The unit cell parameters of 1 at the two different temperatures
change abruptly and remarkably, indicating that 1 displays
a reversible first-order phase transition, which matches well with
the results of the DSC measurements.
3.2.1 Crystal of the RTP structure. The asymmetric unit of
the RTP structure at 25 ꢁC contains a protonated 2,4,6-trime-
thylanilinium cation, (C9H14N)+, and an isolated tetrahedral
Scheme 1 The simulation strain–stress hysteresis loop in compound 1
showing that the relationship between strain and stress is nonlinear under
Hooke’s law (S ¼ ls where l is the elastic constant).
ꢀ
ꢀ
perchlorate anion, ClO4 (Fig. 2a). The ClO4 anion shows
ꢁ
typical Cl–O bond lengths [1.403(4)–1.422(4) A] and the O–Cl–O
angles range from 107.8(2)ꢁ to 111.0(3)ꢁ.
In compound 1, –NH3+ groups of the 2,4,6-trimethylanilinium
cation form three hydrogen bonds to oxygen atoms of the
adjacent three ClO4ꢀ anions, respectively (Table 2S†). The Cl–O
bond length for Cl atoms involved in hydrogen bonding is
1.403(4) and 1.419(4) A. The N–H/O bonds between the 2,4,6-
ꢁ
trimethylanilinium cations and perchlorate anions result in
a noteworthy one-dimensional ribbon-like structure along the
a-axis (Fig. 2b). This ribbon motif (Fig. 2b) is the dominant
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 9841–9845 | 9843

View Article Online
paraelastic phase, i.e. its maximal non-isomorphic sub-groups
ꢀ
containing Pm, P21 and P1, respectively. Furthermore, spatial
symmetric operation numbers decreases from 4 [(1) 1 (2) 2(0, 1/2,
ꢀ
ꢀ
0) 0, y, 0 (3) 1 0, 0, 0 (4) m, x, 1/4, z] to 2 [(1) 1 (2) 1 0, 0, 0] during
symmetry breaking process in good agreement with macroscopic
symmetry breaking.
Furthermore, we can use Landau theory to mimic the strain–
stress hysteresis loop according to the equation of
1
1
1
DG ¼ aðT ꢀ TcÞS2 þ bS4 þ gS6 ꢀ Ss, as shown in
Scheme 1.
2
4
6
Fig. 6 Dielectric constants (30) of compound 1 as a function of
temperature (the measured frequency is fixed at 1 MHz).
3.3 Dielectric properties
3.3.1 Dielectric constant and dielectric loss. As usual, phase
transition is accompanied by an anomaly of the dielectric
constant near the structural phase transition point. The phase
transition of 1 is characterized by dielectric measurements.
Because of the difficulty in obtaining large crystals, the powder-
pressed pellet of 1 was used in the dielectric measurements. The
real part of the dielectric constant of 1 was measured in the
temperature range of 27 ꢁCꢁto 100 ꢁC at 1 MHz (Fig. 6). In
ꢁ
the temperature range of 27 C to 42 C, the dielectric constant
almost remains unchanged (from ca. 4.41 to 4.47). Then it
sharply increases toꢁca. 5.15 at about 46 ꢁC, and then it is almost
unchanged till 100 C. The dielectric loss of 1 a1so displays an
abrupt change (from ca. 0.22 to 0.16) at about 40 ꢁC. The
dielectric anomaly was observed during the measured tempera-
ture ranges, suggesting that there probably is a phase transition.
Fig. 7 Temperature dependence of the AC conductivity of compound 1
at the frequency of 1 MHz.
hydrogen-bonding motif in the structure of 1, and the structure
propagates via the R4 (12) ring. The ammonium cations stack
4
3.3.2 AC conductivity. The AC conductivity sAC was
obtained from the imaginary part of the dielectric permittivity
3 (3 ¼ 30 ꢀ i300, where 30 and 300 are the real and imaginary parts of
the complex dielectric constant 3) by using the relation:
head-to-tail with the C–H/p interactions (Fig. 3).
3.2.2 Crystal of the HTP structure. The asymmetric unit of
the HTP structure at 110 ꢁC consists of one half 2,4,6-trime-
thylanilinium cation and one half perchlorate anion (Fig. 4a). In
sAC ¼ u30030
ꢀ
the ClO4 anion, the lengths of the Cl–O bond is 1.317(7) to
ꢁ
1.365(11) A, which are shorter than those in RTP. In contrast to
where u is the angular frequency 2pf, and 30 the permittivity of
free space (30 ¼ 8.854 ꢂ 10ꢀ12 F mꢀ1). Experimental values of the
AC conductivity depends upon the frequency f. Fig. 7 shows the
temperature dependence of the AC conductivity at the frequency
1 MHz. A drastic change of the conductivity of 1 occurred at
about 40 ꢁC. The observed behavior agreed with the results of the
dielectric measurement.
the RTP structure, the O1 atom of the perchlorate anion is
disordered, which may result in the formation of higher
symmetry. In compound 1, the two components both lie on
a common crystallographic mirror plane.
The N–H/O hydrogen bonds interaction are the main
molecular interactions (Table 2S†). In 1, the 2,4,6-trimethylani-
ꢀ
linium cations and the ClO4 anions are stacked alternating
along the b-axis, forming a ribbon-like structure parallel to the
b-axis. The structure of 1 also propagates via the R4 (12) ring.
4. Conclusion
4
For the reason of the splitting of the O1 atom, the HTP structure
has more hydrogen bonds than the RTP structure, resulting in
the greater stability of the structure. The aromatic rings are
parallel to the ac-plane and the adjacent aromatic rings are
In summary, we have discovered one organic–inorganic hybrid
compound 2,4,6-trimethylanilinium perchlorate (1) undergoing
ꢁ
first-order solid-state phase transitions at ca. 48 C with abrupt
changes in the cell parameters. The present work may open up
a new avenue for the exploration of smart functional materials
with phase transitions.
ꢁ
separated by a centroid-to-centroid distance of 5.59 A, which is
too long for p–p stacking interactions (Fig. 4b).
Symmetry breaking phenomenon occurs during the transition
from the HTP to LTP, a total symmetry decreasing from four
symmetric elements (E, C2, i, sh) to two symmetric elements
(E, i). Thus, this phase transition may be ferroelastic with an
Acknowledgements
This work was supported by the National Natural Science
Foundation of China (21101025, 20931002 and 90922005 as well
as 21111140013), Jiangsu Planned Projects for Postdoctoral
Research Funds (1101010B).
15
ꢀ
Aizu notion of 2/mF1. Spatial symmetric operations change is
shown in Fig. 5. Curie symmetry principle tells us that the space
group at ferroeleastic phase should be the sub-space group at the
9844 | J. Mater. Chem., 2012, 22, 9841–9845
This journal is ª The Royal Society of Chemistry 2012

View Article Online
S. D. Huang, J. Am. Chem. Soc., 2008, 130, 10468; (c) H.-Y. Ye,
D.-W. Fu, Y. Zhang, W. Zhang, R.-G. Xiong and S. D. Huang, J.
Am. Chem. Soc., 2009, 131, 42; (d) W. Zhang, L.-Z. Chen,
R.-G. Xiong, T. Nakamura and S. D. Huang, J. Am. Chem. Soc.,
2009, 131, 12544; (e) W. Zhang, H.-Y. Ye, H.-L. Cai, J.-Z. Ge,
R.-G. Xiong and S. P. D. Huang, J. Am. Chem. Soc., 2010, 132,
7300; (f) T. Hang, W. Zhang, H.-Y. Ye and R.-G. Xiong, Chem.
Soc. Rev., 2011, 40, 3577; (g) H.-L. Cai, W. Zhang, J.-Z. Ge,
Y. Zhang, K. Awaga, T. Nakamura and R.-G. Xiong, Phys. Rev.
Lett., 2011, 107, 147601; (h) D.-W. Fu, W. Zhang, H.-L. Cai,
Y. Zhang, R.-G. Xiong, S. D. Huang and T. Nakamura, Angew.
Chem., Int. Ed., 2011, 50, 11947; (i) D.-W. Fu, W. Zhang,
H.-L. Cai, Y. Zhang, J. Z. Ge, R.-G. Xiong and S. D. Huang, J.
Am. Chem.Soc., 2011, 133, 12780; (j) D.-W. Fu, W. Zhang,
H.-L. Cai, J.-Z. Ge, Y. Zhang and R.-G. Xiong, Adv. Mater., 2011,
23, 5658; (k) G.-C. Xu, W. Zhang, X.-M. Ma, Y.-H. Chen,
L. Zhang, H.-L. Cai, Z.-M. Wang, R.-G. Xiong and S. Gao, J. Am.
Chem. Soc., 2011, 133, 14948; (l) W. Zhang and R.-G. Xiong,
Chem. Rev., 2012, 112, 1163.
Notes and references
1 S. H. Baek, H. W. Jang, C. M. Folkman, Y. L. Li, B. Winchester,
J. X. Zhang, Q. He, Y. H. Chu, C. T. Nelson, M. S. Rzchowski,
X. Q. Pan, R. Ramesh, L. Q. Chen and C. B. Eom, Nat. Mater.,
2010, 9, 309.
2 V. Nagarajan, A. Roytburd, A. Stanishevsky, S. Prasertchoung,
T. Zhao, L. Chen, J. Melngailis, O. Auciello and R. Ramesh, Nat.
Mater., 2002, 2, 43.
3 M. C. Gallardo, J. Manchado, F. J. Romero, J. Cerro, E. K. H. Salje,
A. Planes, E. Vives, R. Romero and M. Stipcich, Phys. Rev. B, 2010,
81, 174102.
4 M. D. Hollingsworth, M. L. Peterson, J. R. Rush, M. E. Brown,
M. J. Abel, A. A. Black, M. Dudley, B. Raghothamachar,
U. Werner-Zwanziger, E. J. Still and J. A. Vanecko, Cryst. Growth
Des., 2005, 5(6), 2100.
5 Y. Yamada, Phys. Rev. B, 1992, 46, 5906.
6 N. A. Spaldin and M. Fiebig, Science, 2005, 309, 391.
7 W. Eerenstein, N. D. Mathur and J. F. Scott, Nature, 2006, 442, 759.
8 T. Kimura, T. Goto, H. Shintani, K. Ishizaka, T. Arima and
Y. Tokura, Nature, 2003, 426, 55.
11 V. K. Wadhawan, Bull. Mater. Sci., 1984, 6, 733.
12 E. K. H. Salje, Contemp. Phys., 2000, 41, 79.
13 H. Schmid, J. Phys.: Condens. Matter, 2008, 20, 4201.
14 M. P. Cruz, Y. H. Chu, J. X. Zhang, P. L. Yang, F. Zavaliche, Q. He,
P. Shafer, L. Q. Chen and R. Ramesh, Phys. Rev. Lett., 2007, 99,
7601.
9 N. Hur, S. Park, P. A. Sharma, J. S. Ahn, S. Guha and S. W. Cheong,
Nature, 2004, 429, 392.
10 (a) D.-W. Fu, Y.-M. Song, G.-X. Wang, Q. Ye, R.-G. Xiong,
T. Akutagawa, T. Nakamura, P. W. H. Chan and S. D. Huang, J.
Am. Chem. Soc., 2007, 129, 5346; (b) W. Zhang, R.-G. Xiong and
15 K. Aizu, Phys. Rev. B, 1970, 2, 754.
This journal is ª The Royal Society of Chemistry 2012
J. Mater. Chem., 2012, 22, 9841–9845 | 9845