Martensitic phase transition and stimuli responsive effects in thermosalient cocrystal of 9,10-dimethylanthracene with F₂TCNQ

Abstract

Martensitic single-crystal-to-single-crystal phase transitions, being rare in organic crystals, can result in several phenomena with promising potential applications, including thermosalient effect, shape memory and self-healing. We report here the first charge-transfer cocrystal of 9,10-dimethylanthracene and 2,5-difluoro-7,7,8,8-tetracyanoquinodimethane, exhibiting a unique combination of dynamic properties stemming from a martensitic phase transition. This organic material demonstrates thermosalient and self-healing behavior, alongside shape recovery during heating and cooling cycles. These effects are driven by collective rotational and translational movements of rigid molecular frameworks, resulting in significant structural changes, while maintaining the process reversibility. Raman spectroscopy, combined with DFT calculations and electron density distribution analysis, provides insight into intermolecular interactions and the potential mechanism of the phase transition. Concurrently, the system displays characteristics of a narrow-gap semiconductor based on transport properties.

---

1. Introduction

Martensitic phase transitions well known in inorganic compounds,¹ such as alloys,² are rarely observed in organic crystals.³ Such phase transformations involve fast collective diffusionless movements of molecules in crystals and are closely related to mechanical properties, such as plasticity and elasticity.^4–7 The concerted movements of molecules in organic crystals can also result in several phenomena with promising potential applications, including thermosalient^8–11 and shape memory effects^12,13 as well as self-healing.^14–16 The possible applications may involve electrical fuses, which was exemplified¹⁷ by the thermosalient crystals 1,2,4,5-tetrabromobenzene. Optical waveguard switching materials¹⁸ with thermosalient properties and light organic single-crystal actuators are further examples of these applications.¹⁹

The feasibility of martensitic phase transitions relies on the anisotropic character of intermolecular interactions. Relatively strong interactions keep supramolecular fragments rigid and sustained, while weak interactions allow the rearrangements or shifts of these rigid fragments. Perfect example is a series of crystals of amino acids, containing bilayers formed via strong hydrogen bonds, while the van-der-Waals interactions between these bilayers are very weak, thus allowing collective shifts during martensitic phase transitions.^20,21 A similar mechanism underlies the martensitic phase transitions and thermosalient behavior observed in terephthalic acid single crystals.¹⁶ The thermosalient effect in 6-chloronicotinic acid single crystals has been attributed to structural rearrangements involving weak interactions between Cl atoms.¹⁹

An extraordinary combination of restorative behavior was observed in the single crystals of penciclovir, explained by the dynamic character of the hydrogen bonds rearranging upon phase transitions.²² An interesting example of dynamic properties at a molecular level and at a macroscopic scale is observed in the carbazole–DABCO complexes which exhibits thermosalient effect,²³ as a result of the phase transition which is accompanied with the reorientation of the DABCO in the cavity and the changes of the hydrogen bond parameters.

Usually the structural differences between polymorphic forms are small and the transformations are reversible. For example nickel(II) bis(diisopropyldithiocarbonate) can undergo up to 900 cycles of phase transitions²⁴ accompanied by conformational changes of the isopropyl group while the central core is rigid. The phase transition comes with a thermosalient behavior of crystals. Most of the crystals exhibiting thermosalient effect are composed of flexible molecules and the phase transitions are accompanied with subtle conformational changes^25–28 and can be very fast.²⁹ The larger structural differences usually result in irreversible phase transitions.^30–34

Thermosalient effect, being not common phenomenon in molecular crystals, is even more rare in multicomponent systems. When it does occur, phase transitions often involve dynamic conformational changes in the most flexible component of the cocrystals.^12,35–37 For example, a multicomponent single crystal of caffeine, 4-chloro-3-nitrobenzoic acid, and methanol exhibits outstanding shape recovery behavior, attributable to mobile solvent channels within the crystal.³⁸

Thermosalient effects in charge transfer cocrystals represent a particularly unique phenomenon. Currently, few reports detail this effect, and the underlying mechanism remains largely unexplored.³⁹ This is often due to instability or irreversibility of transformations, as seen in systems like pyrene/1,8-dinitroanthraquinone.⁴⁰ Studies of the coronene/tetrafluoro-1,4-benzoquinone system suggest that observed thermosalient behavior may result from component escape from the cocrystal.⁴¹ The combination of coronene with tetracyanobenzene gives a self-healing charge transfer cocrystal, though the phase transition in this system is not martensitic.⁴²

Herewith, we report the first charge transfer cocrystal composed of 9,10-dimethylanthracene (Me₂Ant) and 2,5-difluoro-7,7,8,8-tetracyanoquinodimethane (F₂TCNQ) which exhibits a unique combination of dynamic properties upon martensitic phase transition: thermosalient effect (jumping, breaking, exploding), self-healing effect and shape-recovery during heating and cooling cycles. Thermodynamic studies combined with theoretical models suggest a first-order martensitic phase transition governs these effects. Raman spectroscopy confirms the collective modes necessary for the system's concerted rotational–translational movements along the phase transition path.

2. Results and discussions

Microscopic structural changes and macroscopic crystal behavior during the phase transition

The cocrystal of Me₂Ant/F₂TCNQ shows a remarkable polymorphic behavior with reversible martensitic phase transition in the range close to ambient temperatures. The single-crystal-to-single-crystal (SCSC) temperature induced transformation is observed with onset temperature of 14 °C. Both polymorphs crystallize in P1̄ space group (Table S1) and have mixed stacks consisting of donor and acceptor molecules bound via π–π interactions. The low temperature (LT) phase has a perfect layered structure, while in the high temperature (HT) polymorph the neighbouring stacks are shifted respective to each other along the stack direction by 0.9 Å. The polymorphic phase transition is also accompanied by the changes in the orientation between the donor and acceptor molecules within the stacks (Fig. 1). Thus, the structural changes involve not only the in-plane rotational (by ≈10 deg.) and translational (by more than ≈2 Å) motion of donor molecules respective to the acceptor components, but also the shifts of the stacks in the direction orthogonal to molecular planes. Such significant structural changes are rarely reversible.

Fig. 1 Schematic representation of the LT to HT phase transition. (a) Fragments of crystal packing along the a-axes show stacking arrangement, (b) views orthogonal to molecular planes showing donor–acceptor overlap modes and unit cell parameters.

Usually reversible phase transitions in organic crystals include small rotational changes as in the cocrystal of anthracene with tetracyanoquinodimethane (TCNQ),⁴³ or in cocrystal of anthracene with 1,2,4,5-tetracyanobenzene⁴⁴ with the molecular centres of gravity retaining their positions, or small conformational changes at the periphery of molecules, as it is observed upon reversible SCSC phase transitions of rubrene induced by pressure changes.⁴⁵

The SCSC phase transition of the Me₂Ant/F₂TCNQ cocrystal is accompanied not only by remarkable structural changes but by a series of macroscopic effects associated with martensitic phase transitions: thermosalient behavior, self-healing and shape-recovery which are observed in various conditions. When the crystals are placed on a hot plate, they jump, break and explode while cooling and heating (Fig. 2).

Fig. 2 Thermosalient effect and habitus changes of single crystals on a hot stage of the microscope as a function of temperature during cooling and heating.

Being mounted on a loop of the goniometer head with the help of the vacuum grease, the single crystal undergoes a series of shape changes when the nitrogen flow temperature varies. These changes are dependent on the shape of the single crystal and the way how it is mounted (Fig. 3(a)). Multiple experiments with the crystals mounted on a cryoloop showed that the shape changes are accompanied by some cracks along or orthogonal to the stack direction, sometimes the crystals split into parts completely after cooling, as it is shown in Fig. 3(a) (bottom), but return to its original habitus after the heating to initial temperatures (Videos VS1 and VS2). Thus, for both crystals in Fig. 3(a) one can clearly see self-healing and shape-recovery effects.

Fig. 3 (a) The transformations of the habitus of Me₂Ant/F₂TCNQ single cocrystals mounted on a cryoloop, as a function of temperature. The nitrogen flow is approximately along the stacks direction. (b) Crystals of Me₂Ant/F₂TCNQ grown from the solution of methyl ethyl ketone and nonane.

The onset temperatures corresponding to thermosalient effect are different for experiments on a microscope stage and in the nitrogen stream of the diffractometer. Such a difference is likely due to a different placement of the temperature sensors with respect to the sample. The sensor is attached to the microscope stage in the first experiment and positioned near the nozzle of the nitrogen gas-flow temperature unit in the second experiment. Such macroscopic mechanical responses are usually induced by the release of strain accumulated in the crystal due to anisotropic changes of the crystal lattice.⁴⁶

In case of the Me₂Ant/F₂TCNQ cocrystal the largest changes are observed along the stacking directions, the a-axis changes by 11% while two other cell parameters change much less, only by 1%.

In order to study the phase transition in more detail the DSC experiments were carried out. They indicated that one reversible phase transition is observed in the range of −150 °C to 70 °C with onset temperature 14 °C, which is shown in Fig. 4. The heating and the cooling curves confirm the reversibility of the phase transition. In the first cycle (1b and 1a) the curves have a pronounced sawtooth profile typical to thermosalient crystals with martensitic phase transitions³⁴ which is related to crystal size distribution.

Fig. 4 DSC thermograms showing heating (red) and cooling (blue) curves, which are shifted along the ordinate axis for clarity.

The sample for DSC measurements could not be ground as by this it loses its crystallinity. Thus the sample consisted of crystals of different size which undergo phase transition at slightly different temperatures. The size of the crystals in the batch varied from 0.1 mm to 5 mm (Fig. 3(b)). The thermosalient effect upon phase transition is accompanied with crystal fragmentation, thus it results in more even distribution of the crystal sizes and, consequently, in smoothing the DSC curves and decreasing of the transformation enthalpy with each subsequent cycle (Fig. 4).

The phase homogeneity of the HT polymorph was proved by powder X-ray diffraction (Fig. S1).

Lattice dynamics from theoretical and spectroscopic data

To pinpoint modes responsible for the collective lattice dynamics in the phase transition regime, we calculated phonon spectra for both low-temperature and high-temperature phases. Combination of two phonon modes appear to align with the pathway of the phase transition: one with a calculated frequency of 40 cm⁻¹, corresponding to the rocking vibrations of the donor molecule around the center of symmetry, and another at 50 cm⁻¹, involving the sliding of the donor molecule relative to F₂TCNQ (Fig. 5 and Videos VS3, VS4).

Fig. 5 Displacement directions for two phonon modes ((a) rocking and (b) sliding) are well-aligned with the potential pathway of the phase transition. The screenshot captures the locality of the donor–acceptor pair. Hydrogen atoms were removed for simplicity. (c) Structures overlay (HT is red, LT is blue) shows structural changes induced by the phase transition.

Thermal excitation of these low-frequency lattice vibrations results in the phase transformation, involving concerted displacive molecular movements, the resulting structural changes are demonstrated in Fig. 5(c). These martensitic phase transitions become possible owing to the low symmetry of the cocrystal (P1̄ space group) which is the same for both phases.

Experimentally, we can address these modes with Raman spectroscopy. Relatively low crystalline symmetry gives only A_g and A_u symmetry types of normal modes, the former being Raman active. The spectra are therefore quite rich in features, as can be seen in the overview of the room-temperature Raman spectra of the Me₂Ant/F₂TCNQ crystal in the whole range of molecular vibrations in Fig. S2, where it is also compared to the Raman spectrum of F₂TCNQ. The low-frequency part of the Raman spectrum measured in different polarization geometries is shown in Fig. 6.

Fig. 6 Room-temperature Raman spectra of Me₂Ant/F₂TCNQ crystal (excitation at 532 nm) measured with different polarizations (XX, YY, XY, and YX; as two cross-polarized geometries gave identical spectra, only one is shown). Also shown are averaging of all four polarizations (“aver. pol.”), and depolarized spectrum (“depol.”). Vertical bars in the upper part of the panel denote DFT-computed frequencies of Raman-active normal modes with A_g symmetry.

While the intramolecular vibrations show only weak polarization dependence, several high-intensity modes below 100 cm⁻¹ exhibit strong variation with polarization geometry and are readily ascribed to lattice modes. Four experimental peaks identified at 37, 45, 53, and 62 cm⁻¹ closely match theoretical frequencies of A_g modes at 41, 45, 50, and 59 cm⁻¹, confirming that our computational approach gives reliable description of the lattice dynamics in the Me₂Ant/F₂TCNQ crystal. The lowest frequency mode predicted by theory is not observed in the Raman spectra, as it is too close to the measurement window.

Martensitic phase transitions often imply small barriers for atomic rearrangements. As we observe significant structural changes between low and high temperature phases, it is very challenging to define the transition state geometry, especially with periodic boundary conditions in place. This complexity arises when dealing with systems where unit cells have significantly different orientations, as is observed in this case. Nonetheless, we found that the relative energy change between two periodic phases is approximately 0.1 eV per pair. Assuming the Bell–Evans–Polanyi principle applies, this value provides an estimation for the activation energy. Consequently, we can conclude that the process is quite visible around room temperature.

In earlier work,⁴⁷ we studied pairwise interactions in a series of cocrystals of anthracene and its methyl-substituted derivatives with TCNQ and fluoro-substituted TCNQ acceptors. The calculations of donor–acceptor pairs showed that for each pair, several mutual orientations correspond to energy minima, some close to eclipsed or ring-over-ring, as is observed in the HT polymorph (Fig. 1), and several ring-over-bond orientations.

An important point is that the mutual orientation of donor and acceptor molecules in pairs does not depend on methyl substituents or the number of fluorines in the acceptors. The experimental crystal structures (of a total of 9 cocrystals) exhibited a whole range of D–A orientations close to ring-over-ring and ring-over-bond forms with small geometry variations, implying that they are all close in energy, and the barriers are small.

Intermolecular interactions in LT and HT polymorphs

Taking into account the importance of intermolecular interactions in martensitic phase transitions we have studied the experimental and theoretical electron density distribution in LT and HT polymorphs with a view to consider the directionality, the strength and the nature of intermolecular interactions (Tables S2–S5).

In the low-temperature (LT, Fig. 7) polymorph, the crystal is packed by a dense network of short contacts. Within the donor–acceptor stacks, the molecules exhibit optimal face-to-face π–π overlap (the central acceptor ring is positioned over a C=C bond of the donor, and vice versa), which Bader's AIM analysis assigns an interaction energy of ≈7.7 kcal mol⁻¹ per D–A pair. These contacts are complemented by Coulomb attractions between cyano-groups of neighboring F₂TCNQ molecules (≈5.9 kcal mol⁻¹ per acceptor) and a series of CN⋯H–C hydrogen bonds.

Fig. 7 Deformation electron density distribution illustrating the nature of π–π interactions in the LT phase. Planes are drawn perpendicular to molecular planes; sections are shown in the upper-middle figure, and atoms in front of and behind the planes are shown in the upper-left and upper-right figures. Contours are 0.05 e Å⁻³. Negative contours are red, positive contours are blue, and green indicates zero.

An additional 4.8 kcal mol⁻¹ of stabilization arises from C–H⋯π contacts involving donor methyl groups. The donor–acceptor planes are separated by only 3.31 Å, and the large horizontal shift of the donor relative to the acceptor leaves the cyano-groups free to engage in the electrostatic interactions that draw the stacks and layers closer together.

In the high-temperature (HT, Fig. 8) polymorph, the D–A stacking mode changes to an almost eclipsed arrangement, increasing the interplanar distance to 3.43 Å and reducing C–H⋯π contributions to only 0.6 kcal mol⁻¹. Despite the larger separation, AIM calculations give a marginally higher π–π interaction energy (≈8.3 kcal mol⁻¹) because the reduced horizontal shift allows a greater number of individual D–A contacts and more bond-critical points. The loss of the strong cyano-cyano Coulomb attractions that dominate the LT phase and the C–H⋯π interactions with the participation of methyl groups explains the overall weaker intermolecular binding in the HT crystal. The electron density distribution in planes showing π–π interactions is more complex (Fig. 8). The electron density accumulates on the central aromatic bond and the MeC–C bond of the acceptor, facing the ring, with a depletion of electron density elsewhere. In another section (Fig. 8, right), an attractive interaction is observed between the aromatic bond of a peripheral dimethylanthracene ring and the negative region of the deformation electron density between the quinoid ring of the acceptor and the cyano group.

Fig. 8 Deformation electron density distribution showing the nature of π–π interactions in the HT phase. The sections are drawn perpendicular to molecular planes and are shown in the upper-middle figure, the atoms in front and behind the planes are shown in the upper left and right figures. Contours are 0.05 e Å⁻³. Negative contours are red, positive contours are blue, green is for zero.

Summarizing the consideration of intermolecular interactions, one can conclude that π–π interactions within stacks, though possessing the highest energy among a plethora of intermolecular interactions in these crystals, are not directional. This is because they allow for numerous D–A overlap modes with different bond paths for the molecules in parallel planes, and the arrangements can differ by horizontal shifts and twist angles.

The strong directionality is inherent to the hydrogen bonding between acceptor molecules. Therefore, in both polymorphs, acceptors are doubly bridged by CN⋯H interactions and form planar 1D networks with similar geometrical parameters (Fig. 9). The hydrogen bonds in HT polymorph are stronger than in LT polymorph (3.54 and 3.10 kcal mole⁻¹) with the distances d(N⋯H) = 2.362 Å in HT and 2.405 Å in LT. These interactions render the acceptor subsystem rigid, while the donor subsystem is dynamic and experiences large-amplitude, low-frequency vibrations responsible for phase transitions. All other intermolecular interactions are diverse and of small energy.

Fig. 9 Rigid acceptor–acceptor hydrogen bonded chain (top) and deformation electron density distribution (bottom) showing C–H⋯N≡C interactions (contours are 0.05 e Å⁻³; negative contours are red, positive contours are blue, green is for zero).

These features of intermolecular interactions in cocrystal of Me₂Ant/F₂TCNQ explain the dynamic properties of the system and make reversible martensitic phase transitions possible despite the large structural differences between the two polymorphs. The anisotropic character of structural transformations, i.e. the shorter contacts in the molecular planes in the HT phase due to nearly eclipsed position of donors respective to acceptor components and stronger acceptor–acceptor interactions with respect to LT polymorph, and on the contrary much tighter stacking arrangement in the latter, leads to various macroscopic mechanical effects observed in this cocrystal.

In our previous study⁴⁷ we have found out that the cocrystal of 9-methylanthracene (MeAnt) with F₂TCNQ has similar layered structure at 100 K as the cocrystal of Me₂Ant/F₂TCNQ. Thus, we decided to check, whether this cocrystal has phase homogeneity in the temperature range 100–300 K. The DSC measurements revealed no phase transitions, while DFT modeling showed the absence of a sliding low-frequency vibration mode that is observed in Me₂Ant/F₂TCNQ and is important for the phase transition in that cocrystal. The absence of this vibration mode can be explained by the C2/c symmetry of MeAnt/F₂TCNQ cocrystal, which makes translational movement of the whole molecules impossible, as compared to P1̄ space group in Me₂Ant/F₂TCNQ cocrystal. In spite of the similar crystal packing (layered structure, similar overlap mode in the stacks) the different space group results in considerable difference of the cocrystals’ habitus: triclinic crystal has needle-like shape, while monoclinic crystallizes as rectangular prisms.

UV/Vis spectroscopy

The two polymorphs of Me₂Ant/F₂TCNQ were further characterized by UV/Vis spectroscopy. The cocrystal exhibits a charge transfer absorption band composed of two or more overlapping excitations (Fig. 10). The shape of the band is different for HT and LT polymorphs, but no variations were observed within the temperature limits of existence of each phase. The peaks of LT polymorph are more pronounced and the intensity is higher than for HT phase, also the relative intensity of two major excitations is changed though the peak positions are nearly the same. The absorption bands of individual compounds 9,10-dimethylanthracene and F₂TCNQ lie above 2 eV (Fig. S3).

Fig. 10 UV/Vis charge transfer band of Me₂Ant/F₂TCNQ cocrystal for LT and HT phase.

Transport properties

To assess the transport properties, we measured the temperature dependence of the electrical resistance R(T) of the cocrystals.

A representative example of the R(T) dependence for one of the samples with a needle-like crystal morphology measured along the needle axis is shown in Fig. 11. Due to the high resistance of the sample, which reaches several GΩ already at room temperature and increases strongly as the temperature decreases, it was possible to measure this dependence in the relatively narrow region around room temperature only. The resistance exhibits a typical semiconducting behavior, increasing exponentially with decreasing temperature. The R(T) dependence was modelled using the standard formula for an intrinsic semiconductor, whose conductivity is determined both by electrons thermally excited across the band gap E_g into the conduction band and by the corresponding holes created in the valence band: R(T) = R₀·exp(E_g/2k_BT), where k_B is the Boltzmann constant and R₀ is a temperature-independent prefactor.

Fig. 11 (a) Temperature dependence of the resistance of the co-crystal (closed circles) and its fit using the formula R(T) = R₀·exp(E_g/2k_BT) (solid line). (b) Same dependence plotted in the coordinates ln(R) versus 1/T (closed circles) and its fit using the formula ln(R) = ln(R₀) + (E_g/2)·(1/k_BT) (solid line).

The applicability of this approach becomes evident when plotting the logarithm of the resistance as a function of inverse temperature, as shown in Fig. 11. In this case, the experimental data points fall well on a straight line: ln(R) = ln(R₀) + (E_g/2)·(1/k_BT) whose slope is determined by the band gap value E_g. The fitting according to this formula yielded a band gap value of E_g = 0.51 eV which allows to classify the investigated cocrystal as a narrow-gap intrinsic semiconductor.

3. Conclusions

The 1 : 1 cocrystal composed of 9,10-dimethylanthracene and 2,5-difluoro-7,7,8,8-tetracyanoquinodimethane undergoes a reversible martensitic phase transition with an onset temperature of 14 °C. The DSC studies prove the reversibility, and the sawtooth profile of the curves indicates the martensitic character of the phase transition. Despite significant structural changes, including the sliding of donor–acceptor stacks relative to each other and a drastic change in the mutual donor–acceptor orientation within the stacks, the phase transition is reversible. This donor–acceptor system is the first charge transfer cocrystal that exhibits a unique combination of dynamic properties upon martensitic phase transition: a thermosalient effect (jumping, breaking, exploding), a self-healing effect after complete splitting of the single crystal into two parts, and shape recovery during heating and cooling cycles. The mechanism of phase transition, as proposed based on DFT calculations and Raman spectroscopic data, involves the collective movement of donor molecules relative to the rigid, hydrogen-bonded acceptor subsystem. Electron density distribution provides insight into the intermolecular interactions responsible for the phase transition. The π–π interactions within the stacks, though possessing the highest energy among the plethora of intermolecular interactions in these cocrystals, are not directional, as they allow for numerous D–A overlap modes for the molecules in the parallel planes. These arrangements can differ by horizontal shifts and twist angles. The strong directionality is inherent in the hydrogen bonding between the acceptor molecules; thus, in both polymorphs, acceptors are doubly bridged by CH⋯N hydrogen bonds and form rigid planar 1D networks with similar geometrical parameters. All other intermolecular interactions are diverse and small in energy, thus allowing structural rearrangements during phase transitions. Transport properties measured on the single crystals classify them as low-gap semiconductors.

4. Materials and methods

Materials

The following commercially available reagents were used: 9-methylanthracene (98 + %, TCI); 9,10-dimethylanthracene (98 + %, TCI); 2,5-difluoro-7,7,8,8-tetracyanoquinodimethane (98 + %, TCI).

Crystal growth

A mixture of F₂TCNQ (7.2 mg, 0.03 mmol) and (a) MeAnt (6.3 mg, 0.033 mmol) or (b) Me₂Ant (6.8 mg, 0.033 mmol) were placed in round glass jar (10 mL, 55 × 20 mm) and dissolved at heating in the mixture of 4 mL of methyl ethyl ketone and 2 mL of nonane. Cocrystals were grown by keeping the resulting solutions at room temperature with slow solvent evaporation (up to ≈1/4 of the original volume), then filtered off and dried in vacuo to give (a) 8.5 mg (65.6%) of the 1 : 1 MeAnt/F₂TCNQ complex and (b) 9 mg (67.3%) of the 1 : 1 Me₂Ant/F₂TCNQ complex. The modification of the method for preparing cocrystals compared to which we described earlier⁴⁷ is done to obtain the larger and better cocrystals suitable for precision X-ray analysis and resistivity measurements.

DSC measurements

The differential scanning calorimetry (DSC) measurements were performed using a DSC 1/700 (Mettler Toledo) in the temperature range −150 °C to 70 °C. The device was regularly calibrated with reference substances such as indium (T_m = 156.60 °C), gallium (T_m = 29.77 °C), and n-octane (T_m = −56.77 °C).

Detailed measurements in the temperature range −50 °C to 60 °C were carried out on a Netzsch DSC 204 F1 Phoenix differential scanning calorimeter (τ-sensor) in aluminum pans. The rate of heating and cooling was 10 °C min⁻¹. The mass of the samples amounted to approximately ≈2 mg and was controlled with Sartorius CPA 2P balance. Multiple heating and cooling cycles were repeated. Enthalpy values were determined from DSC data as integrals of the peak area relative to the baseline over the entire temperature range where the thermal effect of the process is observed.

X-ray crystallography

The high-resolution X-ray diffraction data for the single crystal of Me₂Ant/F₂TCNQ were collected on a Bruker AXS D8 Quest diffractometer at 100.0(5) K using Mo Kα radiation (λ = 0.71073 Å). Data collection was performed according to recommended strategies employing an ω/φ-scan mode. The programs used: APEX3⁴⁸ for data collection, SAINT⁴⁹ for data reduction, SADABS⁵⁰ for multi-scan absorption correction, SHELXT⁵¹ for structure solution, SHELXL⁵¹ for structure refinement by full-matrix least-squares against F². The details are given in Table S1.

Powder XRD

The powder XRD (X-ray diffraction) measurements were taken using a Bruker D8 Advance diffractometer (Cu Kα radiation λ = 1.5418 Å) with symmetric measurement geometry; the measurement time was 6 seconds per step and the diffraction angle range was of X°–Y° (Fig. S1).

The samples were not ground in order to retain crystallinity. Thus, the powder XRD pattern is affected by texture effects due to needle like shape of crystals with some reflections in experimental curves possessing zero intensity.

Multipole refinement against experimental data

The refinement for Me₂Ant/F₂TCNQ was similar to previously used.^52,53

Theoretical calculations of the crystal structure and electron density

Calculations for MeAnt/F₂TCNQ and Me₂Ant/F₂TNCQ were carried out applying the method described earlier⁵⁴ using the CRYSTAL17 software.⁵⁵ The starting geometry for Me₂Ant/F₂TCNQ was adopted from the XRD data, while for MeAnt/F₂TCNQ, which crystallizes in C2/c space group, the starting geometry was adopted from the data received from PLATON,⁵⁶ that was used for converting molecular structure from C2/c to P1̄ space group to avoid disorder.

The model refined against theoretical structure factors

The structure factors were computed using the XFAC module of CRYSTAL17 software, the theoretical multipole refinement was done by analogy with previously published procedures.^57,58

Electron density analysis

Analysis of ED ρ(r) of the model refined against theoretical structure factors was carried out using WinXPRO v3.4.30 and 3Dplot v2.5.18 software^59–61 by analogy with previously described procedures.^57,58

Raman spectroscopy

Micro-Raman measurements were performed at room temperature in back-scattering configuration using T64000 spectrometer (Horiba Jobin Yvon, USA), equipped with liquid nitrogen-cooled Symphony CCD detector and laser excitations at 532 nm.

DFT modelling of the electronic structure and lattice dynamics

In this work we used DFT with the PBE/PAW level of theory to obtain electronic structure details for the systems discussed in this manuscript. We expect GGA to underestimate bandgaps, due to its well-known shortcoming – lacking an exact exchange part which leads to excess delocalization and smaller band gaps and resulting in a larger charge transfer.

The low and high temperature phases were optimized systems at the DFT/PBE level of theory using the VASP package⁶² and projector augmented-wave method as implemented in the v5.5.4 code and the standard pseudopotential.^63,64 For each phase, unit cells were fixed to experimental values, since adjusting the cell would cause the system to drift significantly from its experimental characteristics. The optimizations were complicated on a G-centered mesh of 3 × 3 × 2 k-points sampled. Following this, electronic spectra as a function of momentum (k) were estimated along high symmetry lines (e.g. G–Z–T–Y–G–X–V–R–U).

The 2 × 2 × 2 supercell calculations at PBE level using VASP code were performed to estimate the Hessian matrix and phonon dispersion in the system. Consistent results were obtained employing density-functional perturbation theory and finite displacement approaches. We utilized the Phonopy library along with our in-house Python code for phonon analysis.^65,66

UV-Vis spectroscopy

The spectra were measured using a Bruker Vertex 80 V spectrometer. This spectrometer is equipped with a Hyperion microscope extension including a liquid nitrogen cooling stage. Measurements have been carried out at various temperatures in (77 K–300 K) using the LINKAM FTIR 600 Stage. The samples were prepared by rubbing a crystal onto a KBr substrate.

Transport measurements

Due to the high resistance of the crystals of the order of several GΩ at room temperature the measurements were done using a Tonghui TH2683A Insulation Resistance Meter, which is capable of measuring resistances up to 10 TΩ. The sample resistance was measured using a two-contact method. For that, a special dielectric high resistance (R > 10 TΩ) plate with two copper contact pads was used on which the studied crystal was placed. To ensure reliable contact between the sample and the contact pads, an aqueous suspension of fine graphite powder was used. After drying, a stable, low-resistance electrical contact was established between the crystal and the contacts, which did not affect the accuracy of measuring the high resistance of the sample. For temperature-dependent measurements, the plate with the sample was fixed inside a massive copper thermostatic cylinder, which was placed in a Dewar vessel with a small amount of liquid nitrogen at the bottom. The temperature was measured using a Fluke 51 II thermometer, with its sensor positioned in close proximity to the sample. The sample was cooled by vapours of liquid nitrogen, and gradual changing of temperature was achieved naturally.

Author contributions

K. I. and A. F. carried out single crystal XRD, K. M. performed synthesis and crystal growth, D. Z. and A. E. conducted DSC measurements, N. G. and A. Ka. carried out transport measurements, A. Ki. performed powder XRD, M. N. did spectroscopic and microscopic studies, S. A. conducted DFT calculations, S. S. and A. P. carried out Raman spectroscopy, V. K. and M. K. carried out interpretation of results, O. K. wrote the manuscript, and all authors reviewed and approved the final version.

Conflicts of interest

There are no conflicts to declare.

Data availability

Data supporting this article have been included in the supplementary information (SI). Supplementary information: video of single crystal temperature dependent behavior, video of crystal lattice dynamic modes, powder XRD pattern, Raman spectra, details of the crystal structure, topological properties. See DOI: https://doi.org/10.1039/d5qm00738k.

CCDC 2492283 and 2492284 contain the supplementary crystallographic data for this paper.^67a,b

Acknowledgements

The work was supported by the Russian Science Foundation, project no. 25-73-20029. The authors gratefully acknowledge the Spectral-Analytical Center “CSF-SAC FRC KSC RAS”.

Notes and references

1. G. Krauss, Martensite in Steel: Strength and Structure, Mater. Sci. Eng., A, 1999, 273–275, 40–57.
2. K. Otsuka and C. M. Wayman, Shape Memory Materials, Cambridge University Press, 1998, p. 298.
3. S. K. Park and Y. Diao, Martensitic Transition in Molecular Crystals for Dynamic Functional Materials, Chem. Soc. Rev., 2020, 49, 8287–8314.
4. C. Wei, L. Li, Y. Zheng, L. Wang, J. Ma, M. Xu, J. Lin, L. Xie, P. Naumov, X. Ding, Q. Feng and W. Huang, Flexible Molecular Crystals for Optoelectronic Applications, Chem. Soc. Rev., 2024, 53, 3687–3713.
5. D. P. Karothu, J. M. Halabi, E. Ahmed, R. Ferreira, P. R. Spackman, M. A. Spackman and P. Naumov, Global Analysis of the Mechanical Properties of Organic Crystals, Angew. Chem., Int. Ed., 2022, 61, e202113988.
6. S. K. Park, H. Sun, H. Chung, B. B. Patel, F. Zhang, D. W. Davies, T. J. Woods, K. Zhao and Y. Diao, Super- and Ferroelastic Organic Semiconductors for Ultraflexible Single-crystal Electronics, Angew. Chem., Int. Ed., 2020, 59, 13004–13012.
7. S. Varughese, M. S. R. N. Kiran, U. Ramamurty and G. R. Desiraju, Nanoindentation in Crystal Engineering: Quantifying Mechanical Properties of Molecular Crystals, Angew. Chem., Int. Ed., 2013, 52, 2701–2712.
8. W. Wu, K. Chen, T. Wang, N. Wang, X. Huang, L. Zhou, Z. Wang and H. Hao, Stimuli-responsive Flexible Organic Crystals, J. Mater. Chem. C, 2023, 11, 2026–2052.
9. P. Naumov, D. P. Karothu, E. Ahmed, L. Catalano, P. Commins, J. M. Halabi, M. B. Al-Handawi and L. Li, The Rise of the Dynamic Crystals, J. Am. Chem. Soc., 2020, 142, 13256–13272.
10. P. Commins, I. T. Desta, D. P. Karothu, M. K. Panda and P. Naumov, Crystals on the Move: Mechanical Effects in Dynamic Solids, Chem. Commun., 2016, 52, 13941–13954.
11. N. K. Nath, M. K. Panda, S. C. Sahoo and P. Naumov, Thermally Induced and Photoinduced Mechanical Effects in Molecular Single Crystals – a Revival, CrystEngComm, 2014, 16, 1850–1858.
12. P. Gupta, D. P. Karothu, E. Ahmed, P. Naumov and N. K. Nath, Thermally Twistable, Photobendable, Elastically Deformable, and Self-healable Soft Crystals, Angew. Chem., Int. Ed., 2018, 57, 8498–8502.
13. S. Takamizawa and Y. Takasaki, Superelastic Shape Recovery of Mechanically Twinned 3,5-Difluorobenzoic Acid Crystals, Angew. Chem., Int. Ed., 2015, 54, 4815–4817.
14. H. Chung, D. Dudenko, F. Zhang, G. D′Avino, C. Ruzié, A. Richard, G. Schweicher, J. Cornil, D. Beljonne, Y. Geerts and Y. Diao, Rotator Side Chains Trigger Cooperative Transition for Shape and Function Memory Effect in Organic Semiconductors, Nat. Commun., 2018, 9, 278.
15. P. Commins, H. Hara and P. Naumov, Self-healing Molecular Crystals, Angew. Chem., Int. Ed., 2016, 55, 13028–13032.
16. D. P. Karothu, J. Weston, I. T. Desta and P. Naumov, Shape-memory and Self-healing Effects in Mechanosalient Molecular Crystals, J. Am. Chem. Soc., 2016, 138, 13298–13306.
17. A. Khalil, E. Ahmeda and P. Naumov, Metal-coated Thermosalient Crystals as Electrical Fuses, Chem. Commun., 2017, 53, 8470–8473.
18. Q. Di, M. B. Al-Handawi, L. Li, P. Naumov and H. Zhang, A Thermosalient and Mechanically Compliant Organic Crystalline Optical Waveguide Switcher, Angew. Chem., Int. Ed., 2024, 63, e202403914.
19. J. Lin, Z. Guo, K. Zhang, P. Zhao, S. Wu, J. Xu, J. Gong and Y. Bao, Mechanical Motion and Modulation of Thermal Actuation Properties in a Robust Organic Molecular Crystal Actuator, Adv. Funct. Mater., 2022, 32, 2203004.
20. M. M. H. Smets, E. Kalkman, A. Krieger, P. Tinnemans, H. Meekes, E. Vlieg and H. M. Cuppen, On the Mechanism of Solid-state Phase Transitions in Molecular Crystals – the Role of Cooperative Motion in (Quasi)racemic Linear Amino Acids, IUCrJ, 2020, 7, 331–341.
21. J. Anwar, S. C. Tuble and J. Kendrick, Concerted Molecular Displacements in a Thermally-induced Solid-state Transformation in Crystals of DL-Norleucine, J. Am. Chem. Soc., 2007, 129, 2542–2547.
22. J. Meng, Y. Su, H. Zhu and T. Cai, Shape Memory and Self-healing in a Molecular Crystal with Inverse Temperature Symmetry Breaking, Chem. Sci., 2024, 15, 5738–5745.
23. A. Colin-Molina, D. P. Karothu, M. J. Jellen, R. A. Toscano, M. A. Garcia-Garibay, P. Naumov and B. Rodríguez-Molina, Thermosalient Amphidynamic Molecular Machines: Motion at the Molecular and Macroscopic Scales, Matter, 2019, 1, 1033–1046.
24. A. Angeloski, J. R. Price, C. Ennis, K. Smith, A. M. McDonagh, A. Dowd, P. Thomas, M. Cortie, D. Appadoo and M. Bhadbhade, Thermosalience Revealed on the Atomic Scale: Rapid Synchrotron Techniques Uncover Molecular Motion Preceding Crystal Jumping, Cryst. Growth Des., 2022, 22, 1951–1959.
25. H. Kato, Y. Horii, C. Watanabe, T. Sasaki, K. Ichiyanagi, M. Noguchi, H. Fujimori, T. Yamamoto, H. Suzuki, Y. Hirai, T. Ohmura, K. Yano, S. Hayashi and T. Kajiwara, Molecular Thermal Engine Based on a Highly Flexible Elastic Crystal, ChemRxiv, 2025, preprint DOI:10.26434/chemrxiv-2025-tpnjj.
26. A. Angeloski, P. Galaviz, R. A. Mole, R. O. Piltz, A. M. McDonagh, C. Ennis and D. Appadoo, Manipulating a Thermosalient Crystal Using Selective Deuteration, J. Am. Chem. Soc., 2025, 147, 8032–8047.
27. M. A. Uddin, R. Martín, S. Gámez-Valenzuela, M. Echeverri, M. C. R. Delgado, E. Gutiérrez Puebla, A. Monge and B. Gómez-Lor, Giant Thermosalient Effect in a Molecular Single Crystal: Dynamic Transformations and Mechanistic Insights, J. Am. Chem. Soc., 2024, 146, 27690–27700.
28. M. I. Tamboli, D. P. Karothu, M. S. Shashidhar, R. G. Gonnade and P. Naumov, Effect of Crystal Packing on the Thermosalient Effect of the Pincer-Type Diester Naphthalene-2,3-diyl-bis(4-fluorobenzoate) a New Class II Thermosalient Solid, Chem. – Eur. J., 2018, 24, 4133–4139.
29. E. Ahmed, D. P. Karothu, A. Slimani, J. M. Halabi, I. Tahir, K. Q. Canales and P. Naumov, Ultrafast, Light, Soft Martensitic Materials, Adv. Funct. Mater., 2022, 32, 2112117.
30. K. Hwang, G. Sin, M. Jang, Y. M. Choi, D. Moon, H. Park and S. K. Park, Amplifying Colossal Thermal Expansion of a Martensitic Molecular Crystal Through Interlayer Shear-induced Side-Chain Liberation, Angew. Chem., Int. Ed., 2025, 64, e202415821.
31. Y.-H. Yang, Y.-S. Chen, W.-T. Chuang and J.-S. Yang, Bifurcated Polymorphic Transition and Thermochromic Fluorescence of a Molecular Crystal Involving Three-Dimensional Supramolecular Gear Rotation, J. Am. Chem. Soc., 2024, 146, 8131–8141.
32. M. Dharmarwardana, B. S. Arimilli, M. A. Luzuriaga, S. Kwon, H. Lee, G. A. Appuhamillage, G. T. McCandless, R. A. Smaldone and J. J. Gassensmith, The Thermo-responsive Behavior in Molecular Crystals of Naphthalene Diimides and their 3D Printed Thermochromic Composites, CrystEngComm, 2018, 20, 6054–6060.
33. M. Dharmarwardana, R. P. Welch, S. Kwon, V. K. Nguyen, G. T. McCandless, M. A. Omary and J. J. Gassensmith, Thermo-mechanically Responsive Crystalline Organic Cantilever, Chem. Commun., 2017, 53, 9890–9893.
34. S. C. Sahoo, M. K. Panda, N. K. Nath and P. Naumov, Biomimetic Crystalline Actuators: Structure–Kinematic Aspects of the Self-actuation and Motility of Thermosalient Crystals, J. Am. Chem. Soc., 2013, 135, 12241–12251.
35. J. Meng, Y. Su, H. Zhu, J. Zhang and T. Cai, Dynamic Molecular Cocrystals with Alkyl Chain Dependent Thermosalient Phase Transitions, Adv. Sci., 2025, 12, 2502692.
36. Y. Takahashi, T. Kondo, S. Yokokura, M. Takehisa, J. Harada, T. Inabe, M. M. Matsushita and K. Awaga, Electric and Thermosalient Properties of a Charge-transfer Complex Exhibiting a Minor Valence Instability Transition, Cryst. Growth Des., 2020, 20, 4758–4763.
37. H. Rawat, R. Samanta, B. Bhattacharya, S. Deolka, A. Dutta, S. Dey, K. B. Raju and C. M. Reddy, Thermosalient Forms: Carryover of Thermosalient Behavior of Coformers from Single Component to Multicomponent Forms?, Cryst. Growth Des., 2018, 18, 2918–2923.
38. S. Ghosh and C. M. Reddy, Elastic and Bendable Caffeine Cocrystals: Implications for the Design of Flexible Organic Materials, Angew. Chem., Int. Ed., 2012, 51, 10319–10323.
39. J. Jin, S. Wu, Y. Ma, C. Dong, W. Wang, X. Liu, H. Xu, G. Long, M. Zhang, J. Zhang and W. Huang, Nucleation Control-Triggering Cocrystal Polymorphism of Charge-transfer Complexes Differing in Physical and Electronic Properties, ACS Appl. Mater. Interfaces, 2020, 12, 19718–19726.
40. M. Singh and D. Chopra, Diversity in Mechanical Response in Donor–acceptor Coupled Cocrystal Stoichiomorphs Based on Pyrene and 1,8-Dinitroanthraquinone Systems, Cryst. Growth Des., 2018, 18, 6670–6680.
41. Y. Chen, J. Li and J. Gong, Jumping Crystal Based on an Organic Charge Transfer Complex with Reversible ON/OFF Switching of Luminescence by External Thermal Stimuli, ACS Mater. Lett., 2021, 3, 275–281.
42. G. Liu, J. Liu, X. Ye, L. Nie, P. Gu, X. Tao and Q. Zhang, Self-healing Behavior in a Thermo-mechanically Responsive Cocrystal During a Reversible Phase Transition, Angew. Chem., Int. Ed., 2017, 56, 198–202.
43. S. Yokokura, Y. Takahashi, H. Nonaka, H. Hasegawa, J. Harada, T. Inabe, R. Kumai, H. Okamoto, M. M. Matsushita and K. Awaga, Switching of Transfer Characteristics of an Organic Field-Effect Transistor by Phase Transitions: Sensitive Response to Molecular Dynamics and Charge Fluctuation, Chem. Mater., 2015, 27, 4441–4449.
44. K.-H. Brose, T. Luty and C. J. Eckhardt, Statics and Dynamics of Organic Charge-Transfer Crystals: the Orientational Phase Transition in A-TCNB, J. Chem. Phys., 1990, 93, 2016–2031.
45. S. Bergantin, M. Moret, G. Buth and F. P. A. Fabbiani, Pressure-induced Conformational Change in Organic Semiconductors: Triggering a Reversible Phase Transition in Rubrene, J. Phys. Chem. C, 2014, 118, 13476–13483.
46. T. Seki, T. Mashimo and H. Ito, Anisotropic Strain Release in a Thermosalient Crystal: Correlation between the Microscopic Orientation of Molecular Rearrangements and the Macroscopic Mechanical Motion, Chem. Sci., 2019, 10, 4185–4191.
47. K. A. Ivshin, K. Metlushka, A. Fedonin, S. K. Latypov, V. V. Khrizanforova, Y. H. Budnikova, A. E. Vandyukov, A. G. Kiiamov, A. Laskin, S. M. Avdoshenko, M. Knupfer and O. Kataeva, Substituent Controllable Assembly of Anthracene Donors and TCNQ Acceptors in Charge Transfer Cocrystals, Cryst. Growth Des., 2023, 23, 954–964.
48. Bruker, APEX3 Crystallography Software Suite, Bruker AXS Inc, Madison, WI, USA, 2016.
49. Bruker, SAINT Crystallography Software Suite, Bruker AXS Inc, Madison, WI, USA, 2016.
50. L. Krause, R. Herbst-Irmer, G. M. Sheldrick and D. Stalke, Comparison of Silver and Molybdenum Microfocus X-Ray Sources for Single-crystal Structure Determination, J. Appl. Crystallogr., 2015, 48, 3–10.
51. G. M. Sheldrick, A Short History of SHELX, Acta Crystallogr., Sect. A: Found. Adv., 2008, 64, 112–122.
52. S. A. Shteingolts and R. R. Fayzullin, X-Ray Charge Density Study of the Drug Methimazole with Z′ = 2: Differences in the Electronic Structure of the Thiourea Core due to Crystal Packing Effects, Cryst. Growth Des., 2020, 20, 2074–2090.
53. O. Kataeva, M. Nohr, K. Ivshin, S. Hampel, B. Büchner and M. Knupfer, Understanding Intermolecular Interactions in a Tetracene–F₄TCNQ Cocrystal via its Electron Density Distribution and Topology, Cryst. Growth Des., 2021, 21, 471–481.
54. O. Kataeva, K. Ivshin, K. Metlushka, K. Nikitina, V. Khrizanforova, Y. Budnikova, R. R. Fayzullin, S. Latypov, S. Schiemenz, M. Bretschneider, A. Popov, S. Avdoshenko, Y. Krupskaya, B. Büchner and M. Knupfer, New Charge Transfer Cocrystals of F₂TCNQ with Polycyclic Aromatic Hydrocarbons: Acceptor–acceptor Interactions and their Contribution to Supramolecular Arrangement and Charge Transfer, Cryst. Growth Des., 2022, 22, 751–762.
55. R. Dovesi, A. Erba, R. Orlando, C. M. Zicovich-Wilson, B. Civalleri, L. Maschio, M. Rérat, S. Casassa, J. Baima, S. Salustro and B. Kirtman, Quantum-mechanical Condensed Matter Simulations with CRYSTAL, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2018, 8, e1360.
56. A. L. Spek, Single-crystal Structure Validation with the Program PLATON, J. Appl. Crystallogr., 2003, 36, 7–13.
57. S. A. Shteingolts, A. F. Saifina, L. F. Saifina, V. E. Semenov, G. K. Fukin and R. R. Fayzullin, X-Ray Charge Density Study of the 6-Methyluracil Derivative in the Crystal: Revealing, Consequences, and Multipole Refinement of Minor Static Disorder, J. Mol. Struct., 2021, 1228, 129724.
58. S. A. Shteingolts, J. K. Voronina, L. F. Saifina, M. M. Shulaeva, V. E. Semenov and R. R. Fayzullin, On the Transfer of Theoretical Multipole Parameters for Restoring Static Electron Density and Revealing and Treating Atomic Anharmonic Motion. Features of Chemical Bonding in Crystals of an Isocyanuric Acid Derivative, Acta Crystallogr., Sect. B: Struct. Sci., Cryst. Eng. Mater., 2021, 77, 871–891.
59. A. I. Stash and V. G. Tsirelson, Developing WinXPRO: a Software for Determination of the Multipole-model-based Properties of Crystals, J. Appl. Crystallogr., 2014, 47, 2086–2089.
60. A. I. Stash and V. G. Tsirelson, Modern Possibilities for Calculating Some Properties of Molecules and Crystals from the Experimental Electron Density, Crystallogr. Rep., 2005, 50, 177–184.
61. A. Stash and V. Tsirelson, WinXPRO: a Program for Calculating Crystal and Molecular Properties using Multipole Parameters of the Electron Density, J. Appl. Crystallogr., 2002, 35, 371–373.
62. G. Kresse and J. Hafner, Ab initio Molecular Dynamics for Liquid Metals, Phys. Rev. B: Condens. Matter Mater. Phys., 1993, 47, 558–561.
63. E. Zirngiebl, B. Hillebrands, S. Blumenröder, G. Güntherodt, M. Loewenhaupt, J. M. Carpenter, K. Winzer and Z. Fisk, New Crystal Field Level Scheme of CeB₆ Deduced from Raman and Neutron Spectroscopy (abstract), J. Appl. Phys., 1985, 57, 3769.
64. E. Zirngiebl, B. Hillebrands, S. Blumenröder, G. Güntherodt, M. Loewenhaupt, J. M. Carpenter, K. Winzer and Z. Fisk, Crystal-field Excitations in CeB₆ Studied by Raman and Neutron Spectroscopy, Phys. Rev. B: Condens. Matter Mater. Phys., 1984, 30, 4052–4054.
65. L. Chaput, A. Togo, I. Tanaka and G. Hug, Phonon-phonon Interactions in Transition Metals, Phys. Rev. B: Condens. Matter Mater. Phys., 2011, 84, 094302.
66. L. Chaput, A. Togo, I. Tanaka and G. Hug, First-principles Phonon Calculations of Thermal Expansion in Ti₃SiC₂, Ti₃AlC₂, and Ti₃GeC₂, Phys. Rev. B: Condens. Matter Mater. Phys., 2010, 81, 174301.
67. (a) CCDC 2492283: Experimental Crystal Structure Determination, 2025 DOI:10.5517/ccdc.csd.cc2pnf71; (b) 2492284: Experimental Crystal Structure Determination, 2025 DOI:10.5517/ccdc.csd.cc2pnf82.

---