Abstract
Mechanically responsive molecular crystals that reversibly change shape triggered by external stimuli are invaluable for the design of actuators for soft robotics, artificial muscles and microfluidic devices. However, their strong deformations usually lead to their destruction. We report a fluorenone derivative (4-DBpFO) showing a strong shear deformation upon heating due to a structural phase transition which is reproducible after more than hundred heating/cooling cycles. Molecular dynamic simulations show that the transition occurs through a nucleation-and-growth mechanism, triggered by thermally induced rotations of the phenyl rings, leading to a rearrangement of the molecular configuration. The applicability as actuator is demonstrated by displacing a micron-sized glass bead over a large distance, delivering a kinetic energy of more than 65 pJ, corresponding to a work density of 270āJākgā1. This material can serve as a prototype structure to direct the development of new types of robust molecular actuators.
Similar content being viewed by others
Introduction
Thermosalient crystals are characterized by thermally induced solid-state phase transitions accompanied by a sudden anisotropic lattice expansion that induces a jumping, bending, deformation or rotation of the crystals upon heating1,2,3,4,5. They belong to a large class of dynamic crystals that respond to external stimuli, such as heat, light, or pressure, with various switching effects of their electronic6,7, magnetic8,9, optical10, and mechanical11,12,13 properties. These phase transitions are generally considered to proceed through cooperative molecular motions inside the crystal3,5, but their understanding is still in its infancy. It is not evident how the classifications and phase transition mechanisms in inorganic crystals or crystals with much less molecular complexity14,15 translate to molecular crystals, with weaker interactions and more steric hindrance. The large anisotropic changes in crystal shape and size are difficult to accommodate for the often brittle organic crystals and the mechanical effects are usually accompanied by crystal cracking, splitting or even explosion1,2. Therefore, new model crystal structures need to be developed that combine large thermomechanical effects with robustness of the crystal.
The known thermosalient materials were classified by Sahoo et al.2 into three classes: two classes of layered materials, one containing flat, rigid molecules with weak interactions between them, the other containing molecules with strong interconnecting hydrogen bonds within the layer and both with weak interactions between the layers; the third class contains bulky flexible molecules with hindered rotations. Here we present a thermosalient material based on 2,7-di([1,1ā²-biphenyl]-4-yl)-fluorenone molecules (4-DBpFO)16, which does not fall in any of these classes. The transition appears to be triggered by small conformational changes in the molecules, which propagate through the crystal and result in a large crystal shape change2,3. Practical application of 4-DBpFO is demonstrated by its ability to displace glass plates and a micron-sized glass bead over a considerable distance upon a slight temperature increase.
Results
Structural phase transition
4-DBpFO is constituted of five flat segments: a fluorenone center with two phenyl rings at each side linked by single bonds (Fig.Ā 1). The small rigid planes can rotate around the single bonds with a large degree of freedom (dihedrals Īø and Ļ) when the molecules are heated in solution, similar to the molecular motors designed by Feringaās lab17,18. We have earlier reported that this compound crystallizes in a non-centrosymmetric crystal structure at room temperature with a Cmc21 space group (referred to as the Ī±ā²-phase), useful for optical second harmonic generation16. Here, we show another room temperature phase of 4-DBpFO described by a centrosymmetric Pnma space group (referred to as the Ī±-phase), crystallized from a mixed solvent of chloroform and heptane (see TableĀ 1 for cell parameters). The morphology of the as-grown crystals is a parallelogram-like thin platelet (Fig.Ā 1a) with {010} basal planes and {101} or {10-1} side faces (Fig.Ā 1b). The crystal structure shows an antiparallel molecular organization along the crystallographic b-axis (Fig.Ā 1c), resulting in a layered structure parallel to the (010) plane (Fig.Ā 1b, c). The corner angles of the parallelogram are 91Ā° and 89Ā°, in accordance with those calculated from the cell parameters at room temperature.
When a thin, almost square, parallelogram Ī±-phase single crystal is heated to around 189āĀ°C, it transforms to a new phase (referred to as the Ī²-phase) with a more oblique morphology (Fig.Ā 1a, Supplementary MovieĀ 1). Depending on crystal qualityĀ the transition temperature may vary. This dramatic shape change can be understood as a shear deformation in the (010) planes, parallel to theā<ā101ā>āorā<ā10-1ā>ādirections. To exclude thermal expansion effects on the cell parameters, the unit cell parameters of the Ī±-phase at 170āĀ°C (right below the phase transition temperature) and the Ī²-phase at 178āĀ°C (right above the phase transition temperature) were determined by single crystal X-Ray Diffraction (SXRD) (TableĀ 1). To facilitate comparison with the Ī±-phase, the high temperature Ī²-phase is described in a non-standard setting. During the Ī±- to Ī²-phase transition, the b-axis stays almost the same; the a-axis shrinks and the c-axis expands, while the corner angles of the crystal change drastically (TableĀ 1), leading to the observed crystal shear deformation in the (010) plane19. This unit cell change corresponds very well with the shape change observed in optical microscopy20 (Supplementary Fig.Ā 1). During this in-plane anisotropic expansion (Fig.Ā 1b), the crystals still keep the layered structure along their normal direction (Fig.Ā 1c). This shear deformation is very pronounced compared to that found in photosalient crystals. For example, Kobatake et al. reported a diarylethene crystal which can undergo a similar shear deformation by light irradiation, but the corner angles only changed from 88o and 92o to 82o and 98o.
The changes observed in our crystal are so large and abrupt that, when the crystal was placed sideways on one of its {101} or {10-1} facets, it jumped far away upon heating (Supplementary Fig.Ā 2). When lying on its basal plane a distinct and coherent phase boundary between the Ī±- and Ī²- phases was clearly observed during the phase transition, starting from one of the {101} or {10-1} facets, as shown in Fig.Ā 2a. Remarkably, this phase boundary propagated through the entire crystal upon continuous heating, being always parallel to one of the crystal sides and staying macroscopically undistorted during migration through the crystal. There is no strong interaction between the molecules within the layers and this allows the phase boundary to be perpendicular to the layer plane, in contrast with two of the classes by Sahoo et al.2. The third class includes transitions with hindered rotation of bulky groups. Although torsional rotation plays a role in the present transition, it is not hindered since the dihedrals do not change sign. This resembles the characteristics of martensitic transformations in metals21, which also proceeds by the movement of an invariant phase boundary, called the habit plane. Generally, the structural transformation in thermosalient organic crystals will rapidly transform the whole crystal once initiated, which makes it hard to control. However, we found that the migration of the phase boundary of 4-DBpFO could be pinned at a certain position by keeping the crystal at a constant temperature, and its propagation direction could be even reversed upon cooling (Supplementary Fig.Ā 3). This is presumably due to the presence of defects or dislocations22. As the (101) and (10ā1) faces are alike due to pseudosymmetry (symmetry operation āz, āy, āx), the phase boundary can in principle appear in both directions of the same crystal, which was indeed observed in some crystals (Fig.Ā 2a, b). The phase transition proceeding via the two different paths resulted in the same final crystal shape, in a mirrored orientation (Fig.Ā 2c, d).
Shear deformation and durability
The observed shear deformation can be understood from the relationship between the crystallographic orientations of the two phases during the phase transition (Fig.Ā 2e, f). The anisotropic lattice expansion in the (010) plane generates a shear force parallel to the unit cell diagonals, resulting in a huge macroscopic shear deformation parallel to the crystal sides of more than 18% (Fig.Ā 2f and Supplementary Fig.Ā 4 for method of calculation). This large strain is two times larger than that of thermally activated shape memory alloys (~8.5%)23,24. Although the cell lengths of the two axes in the (010) plane change prominently (~10%) during the phase transition, the diagonal lengths of the in-plane unit cell do not change and stay at about 10.0āĆ . Accordingly, the macroscopic lengths of the parallelogram-shaped crystal sides do not change. For martensitic phase transitions, the lattice parameters usually change prominently in all three directions, so an additional lattice invariant shear in the new phase is needed to maintain the invariant plane of the phase boundary25,26. This lattice invariant shear results in the formation of substructures in metals or alloys27,28 which can be plastically deformed by twinning or slipping29 (Supplementary Fig.Ā 5). For brittle organic crystals, this lattice invariant shear often results in breakage, splitting or even explosion. For 4-DBpFO, the phase boundary is parallel to the (101) or (ā101) plane during the migration (Fig.Ā 2). As the lengths of the unit cell diagonal in the (010) plane and the normal b-axis stay the same, the phase boundary is an invariant plane and thus the phase transition does not need any extra lattice invariant shear30. This explains the full reversibility of the phase transition: the high temperature phase crystal returned completely to its original shape when cooled below 178āĀ°C. This reversibility is also apparent from the differential scanning calorimetry (DSC) measurements (Fig.Ā 3a): the change in enthalpy estimated from the DSC curve is 1.02ā2.10ākJāmolā1 for both cooling and heating. More importantly, the crystal showed full reversibility of the shape deformation after more than 100 cycles without any visible damage (Fig.Ā 3b), demonstrating excellent fatigue durability of 4-DBpFO as a thermosalient material. Another reason for the good reversibility may be related to the fact that the in-plane anisotropic lattice expansion results from the in-plane rotation of the benzene ring planes in the molecules, while the rigid fluorenone center does not rotate. As shown in Fig.Ā 3c, the dihedral angles Īø and Ļ change, but their signs stay the same after the phase transition (please refer to TableĀ 1 as well). The molecular configurations become more planar after the phase transition, which may be the reason that the cell volume of the Ī²-phase is 2.7% smaller than the Ī±-phase. In fact, crystal splitting parallel to the (010) plane was observed in a few crystals (two crystals out of 50+ crystals studied) which may be because of the weak intermolecular force between the layers along the normal direction; upon cooling, their shape still completely recovered, but not the damage due to the splitting (Supplementary Fig.Ā 6).
Molecular dynamics simulations of phase transition
To gain more insight into the microscopic origin of the strong shape deformation of the 4-DBpFO crystal, classical molecular dynamics simulations were performed using LAMMPS. The simulation cell consists of four 4-DBpFO layers in the y-direction, each with four dihedral angles per molecule. The system was first equilibrated in the Ī±-phase at 27āĀ°C and then kept at this temperature for 60āps. As can be seen in Fig.Ā 4b, the cell lengths of the Ī±-phase were well retained, as was its orthorhombic character, the molecular conformation (Fig.Ā 4c, d) and the crystal packing, giving confidence to the applied force field. The system was subsequently heated during 100āps to 177āĀ°C (vertical dashed lines and color gradient background). The crystal showed roughly isotropic expansion during this heating, as is visible in the graph. After heating to 177āĀ°C, a drastic change in the corner angles (Fig.Ā 4a) and lattice parameters (Fig.Ā 4b) can be observed. This is accompanied by changes in the dihedral angles of the molecules, as visualized in the bottom two panels. At elevated temperatures, the dihedral fluctuations of the outer phenyl rings increase and trigger a conformational change (phase transition).
Judging by the packing, the cell parameters and the molecular conformation, the system has transformed to the Ī²-phase, in agreement with the experimental findings. We would like to stress here, that this transformation appeared in our simulations spontaneously: we did not steer the system by adding a bias, fixing the lattice parameters to the new form or changing the molecular conformation. Only, the volume was allowed to change and the molecules were treated flexibly while heating the system. A movie as well as snapshots of the onset of the transformation (Supplementary MovieĀ 2) shows that within the simulation cell, a nucleation point of a few molecules can be identified after which the transformation spreads through the rest of the crystal in all three directions. In 15āps the whole cell is transformed to the new structure. The general consensus in the literature is that thermosalient materials transform through a fast, cooperative mechanism2,31. Although the length scale of this cooperative movement is typically not specified, it is generally assumed to be on a larger scale than the 54āĆā108āĆā58āĆ scale of our simulation cell. The nucleation-and-growth mechanism that follows from our molecular dynamics simulations has the following consequences, in agreement with our experimental observations: (i) there is a well-defined transition temperature with little hysteresis and (ii) the transition is more gentle, making it less destructive and allowing it to be cycled many times without destroying the crystal.
Temperature controlled movement of load
To demonstrate the applicability of the 4-DBpFO thermosalient crystal, we used it to move a micron-sized glass bead by fixing the crystal on the substrate on one side and following the position of the bead with a CCD camera with a frame rate of 30 fps (see Fig.Ā 5 and Supplementary MovieĀ 3). Heating or cooling the crystal (200āĆā200āĆā50āĪ¼m) through the transition both gave rise to a strong ākickā to the bead which subsequently was displaced several centimeters. From the measured average velocity (29āmmāsā1) and mass (0.15āmg) of the glass bead, the kinetic energy generated by the deformation could be estimated to be >6.5āĆā10ā11āJ. This amount of energy would allow the crystal, when standing on one of its side faces, to jump 3āmm high, which can indeed explain that our crystal can jump over a long distance. The work density (output mechanical energy per unit volume or mass of material) in this process is >270āJākgā1, which is two orders of magnitude larger than the typical performance of most MEMS (microelectromechanical systems)23,32 or the recently discovered autonomous actuator33. Actuators are typically characterized by strain34 and strain rate, i.e. the change in shape per second35. From the time-dependent microscopy observations we can estimate that the largest strain rate for some micron-sized 4-DŠpFO crystal is larger than 180%āsā1 (Supplementary Fig.Ā 7), comparable with thermally activated shape memory alloys (ā<ā300%āsā1) but much larger than that of piezoelectric actuators (ā<ā12%āsā1)23 (Supplementary TableĀ 1). In fact, the strain rate was too large to be determined for many of the smaller, high quality crystals (that thus likely had fewer pinning dislocations or impurities), as their phase boundary can hardly be observed because of the high speed of the phase transition. To estimate the generated force during the phase transition, a crystal was placed between two parallel glass plates on a flat substrate. Here, the measured pushing force onto the glass plates by the crystal is equivalent to the maximum static friction between the glass plates and the substrate. A small crystal (160āĆā160āĆā60āĪ¼m) can generate a pushing force as large as 530āĀµN at its phase transition temperature, corresponding to 104 times its gravitational force (Supplementary Fig.Ā 8b and MovieĀ 4). When the glass plates were too heavy, the molecular movement became distorted and incoherent phase boundaries appeared during the phase transition (Supplementary Fig.Ā 8c and MovieĀ 5).
As the crystal lattice completely returns to its initial state through the reversible phase transition, the āshear strainā of the crystal, which is as high as 0.18 during the transformation (Supplementary Fig.Ā 4), is completely elastic. Although the optimum strain of some metals could be more than 0.10 in theory36, the real maximum elastic strain only reaches 0.01 in most polycrystalline and even nanocrystalline alloys37 due to the defects in the crystals38. Moreover, elastically deformable metals can only generate unidirectional forcesāwhen the deformed lattice returns to its stable state. Here, the lattices of both phases are stable, so both the heating and cooling induced shape change can be exploited to generate a force. That is, the lattice change during heating is endothermic, converting thermal energy into kinetic energy while storing thermal energy in the crystal in the form of chemical energy. This stored chemical energy is subsequently released and converted into kinetic energy during cooling, due to the reversible shape change.
Discussion
In summary, we have fabricated an organic thermosalient crystal 4-DBpFO that can be applied as a molecular microactuator. The crystal shows a reversibly major shear deformation in the (010) plane, which proceeds along two orthogonal crystal sides with excellent reversibility for more than 100 cycles of the temperature induced structural phase transition without visible wear. The applicability of this controllable and strong deformation was demonstrated by its ability to displace a load of more than 100 times its own mass for at least 5 times. The remarkable reversibility and robustness of this transition arises from in-plane molecular rotations through a nucleation-and-growth mechanism as demonstrated by Molecular Dynamics simulations. This is contrary to many cases in the literature where some thermosalient behavior is explained by cooperative motion on much larger length scales. This molecular structure can be used as a guidance to design new types of controllable and resilient self-assembly molecular microactuator that may operate in desired temperature ranges.
Methods
Materials
The synthesis of the 4-DBpFO compound can be found in reference 15. The Ī±-phase crystal was grown through a solution-diffusion method. Heptane was dropped gently on the top of a saturated solution of chloroform, and then heptane will diffuse into the chloroform slowly. Well-defined parallelogram-shaped micro-crystals with side lengths varying from 2 to 50āĪ¼m were formed after 24āh. After several weeks of diffusion, larger single crystals with side lengths ranging from 100 to 200āĪ¼m could be obtained, suitable for single crystal XRD analysis.
Optical microscopy
Fluorescence images were acquired with a Leica-Microsystems DM2500 microscope. Optical microscope images were obtained with a Zeiss Axioplan 2 microscope, equipped with a MediaCybernetics Evolution VF digital camera. For images of the shape change this microscope was equipped with a Linkam LTS420 thermal stage under nitrogen atmosphere, using heating rates ranging from 3 to 10āĀ°Cāminā1. The movie of the ākickingā glass bead was taken with a Leica WILD M10 microscope, equipped with an Evolution MP5.0 Leica DMC2900 digital camera with frame rate of 30 fps and a Linkam LTS420 thermal stage under nitrogen atmosphere.
Differential scanning calorimetry
Differential scanning calorimetry (DSC) measurements were carried out with a Mettler Toledo DSC1 calorimeter equipped with a high sensitivity sensor (HSS8), in combination with LN2 liquid nitrogen cooling, a sample robot and STARe software 13.00a. As-grown microcrystals (about 0.45āmg) were heated and cooled in a sealed aluminum pan with a rate of 10āKāminā1 in the temperature range of 50ā205āĀ°C. The heat flow was measured by comparing to an empty reference pan as a function of temperature. The calorimeter was calibrated with the melting points of indium (TInā=ā429.5āK and ĪHā=āā28.13āJāgā1) and zinc (TZnā=ā692.85āK and ĪHā=āā104.77āJāgā1), both supplied by Mettler Toledo. The enthalpy change of the phase transition was calculated by integrating the exothermic peak and the endothermic peak, respectively.
Single crystal X-ray diffraction
Single crystal X-ray diffraction (SXRD) measurements were collected on a Bruker D8 Quest diffractometer with sealed tube (MoKa radiation) and Triumph monochromator (Ī»ā=ā0.71073āĆ ). The software package Saint was used for the intensity integration39. Absorption correction was performed with SADABS40.The structures were solved with direct methods using SHELXT41. Least-squares refinement was performed with SHELXL-201442 against \(\left| {F_h^o} \right|^2\) of all reflections. Non-hydrogen atoms were refined freely with anisotropic displacement parameters. Hydrogen atoms were placed on calculated positions or located in difference Fourier maps. All calculated hydrogen atoms were refined with a riding model. For high temperature SXRD, the crystal was sealed in a capillary and an Oxford Cryostream 700 plus was used to heat the crystal to the desired temperature.
Molecular dynamic simulation method
Molecular dynamic (MD) simulations were performed using LAMMPS43 (version 16-02-2016). A simulation cell was build out of 8āĆā2āĆā8 crystallographic unit cells of the Ī±-phase (512 molecules). The GAFF2.1 force field was used for the intra- and intermolecular interactions44 in combination with ESP-derived point charges (DFT, B3LYB cc-pVDZ). The latter were calculated with the help of Molpro and Molden45,46. Two modifications were applied to the GAFF2.1 force field. The contribution for the torsional change in dihedral Īø was made asymmetric in accordance with DFT calculations, and the H atom sites were chosen between the nucleus and the maximum in the electron density as determined by XRD. For all interactions, the dispersion contribution is cut-off at 10.0āĆ and the Coulombic part of the potential is computed using Ewald summation with a relative RMS error in per-atom force of 10ā6. The simulations were performed in the triclinic NPT ensemble with a time step of 0.5āfs. The thermostat and barostat parameters were set to 40 and 400āfs, respectively.
Estimation of actuator performance
Strain: Crystal strain at the phase transition temperatures: The strain during the phase transition was calculated from the lattice parameters of the two phases. The linear strain can be calculated from:
where ālā is the cell length, and āĪlāāis the corresponding expansion length along the cell axes after the phase transition. An elastic compressive strain of 0.094 along the a-axis and an elastic tensile strain of 0.069 along the c-axis can be reached during the phase transition from the Ī±-phase to the Ī²-phase by heating, while an elastic tensile strain as large as 0.10 along the a-axis and an elastic compressive strain of 0.064 along the c-axis can be reached during the reversed phase transition by cooling. In engineering, shear strain is equal to the length of deformation at its maximum divided by the perpendicular length in the plane of force application. By this definition, the shear strain here is calculated from:
where āĪxā is the displacement along the lattice diagonal and y is the cell length perpendicular to the displacement direction (SupplementaryĀ Fig. 4). The shear strain was calculated to be as large as 0.18.
Kinetic energy: The capability of delivering kinetic energy of the crystals was estimated by moving a micron-sized glass bead. The mass of the crystal shown in Fig.Ā 5 and in Supplementary MovieĀ 3 was ~2.5āĀµg, which was calculated from the relative molecular mass and the size of the crystal (200āĆā200āĆā50āĪ¼m). The bead moved over 0.96āmm in 0.033ās, which corresponds to an average speed of 29āmmāsā1. The kinetic energy (6.5āĆā10ā11āJ) of the bead obtained from the shape change of the crystal was calculated from:
where āmā is the mass of the bead (0.15āmg), and āvā is the speed of the bead induced by the movement of the crystal.
Maximum pushing force: The maximum pushing force of the microcrystals was estimated by the ability of moving glass plates (as shown in Supplementary Fig.Ā 8). Crystals were placed between two parallel glass plates on a larger flat glass substrate. The parallel glass plates can be pushed apart by the crystals at the phase transition temperature, due to the shape change of the crystals. The force acting on the glass plates generated by the shape change (F) was calculated from:
where ā\(f_s^{{\mathrm{max}}}\)ā is the maximum static friction between the glass plates and the glass substrate; āĪ¼sā is the static friction coefficient between two clean and dry glass surfaces, which is 0.9ā1.0;47 āmā is the mass of a single glass plate and āgā is the gravity acceleration. The glass plates in Supplementary Fig.Ā 8b each weigh 60āmg, which is ~3āĆā104 times heavier than the crystal (160āĆā160āĆā60āĪ¼m), so the corresponding static friction between the glass plates and the glass substrate is at least 0.9āĆā60āmgāĆā9.8āNākgā1āāā530āĀµN, which is ~3āĆā104 times stronger thanĀ its gravitational force. When each glass plate weighs 260āmg, which is ~105 heavier than the crystal (200āĆā200āĆā50āĪ¼m), the glass plates can still be pushed apart by the crystal, but the crystal becomes ādamagedā during the shape change (Supplementary Fig.Ā 8c). During this process, the static friction between the glass plates and substrate can be as large as 2300āĀµN, which is 105 times the gravitational force of the crystal.
Data availability
Crystal data of the Ī±-phase and Ī²-phase are available from the Cambridge Crystallographic Data Centre with reference number of CCDC-1862516 and CCDC-1920480, respectively. The source data underlying Figs.Ā 3a and 4 are provided as a Source Data file. All data needed to evaluate the conclusions in the paper are present. Additional data related to this paper may be requested from the authors.
References
Naumov, P., Chizhik, S., Panda, M. K., Nath, N. K. & Boldyreva, E. Mechanically responsive molecular crystals. Chem. Rev. 115, 12440ā12490 (2015).
Sahoo, S. C., Panda, M. K., Nath, N. K. & Naumov, P. Biomimetic crystalline actuators: structure-kinematic aspects of the self-actuation and motility of thermosalient crystals. J. Am. Chem. Soc. 135, 12241ā12251 (2013).
Panda, M. K. et al. Colossal positive and negative thermal expansion and thermosalient effect in a pentamorphic organometallic martensite. Nat. Commun. 5, 4811 (2014).
Commins, P., Desta, I. T., Karothu, D. P., Panda, M. K. & Naumov, P. Crystals on the move: mechanical effects in dynamic solids. Chem. Commun. 52, 13941ā13954 (2016).
Chung, H. et al. Rotator side chains trigger cooperative transition for shape and function memory effect in organic semiconductors. Nat. Commun. 9, 278 (2018).
Fu, D.-W. et al. Diisopropylammonium bromide is a high-temperature molecular ferroelectric crystal. Science 339, 425ā428 (2013).
Horiuchi, S. et al. Above-room-temperature ferroelectricity in a single-component molecular crystal. Nature 463, 789ā792 (2010).
Shepherd, H. J. et al. Molecular actuators driven by cooperative spin-state switching. Nat. Commun. 4, 2607 (2013).
Fujita, W. & Awaga, K. Room-temperature magnetic bistability in organic radical crystals. Science 286, 261ā262 (1999).
Setaka, W. & Yamaguchi, K. Thermal modulation of birefringence observed in a crystalline molecular gyrotop. Proc. Natl Acad. Sci. USA 109, 9271ā9275 (2012).
Kobatake, S., Takami, S., Muto, H., Ishikawa, T. & Irie, M. Rapid and reversible shape changes of molecular crystals on photoirradiation. Nature 446, 778ā781 (2007).
Yao, Z. S. et al. Molecular motor-driven abrupt anisotropic shape change in a single crystal of a Ni complex. Nat. Chem. 6, 1079ā1083 (2014).
Ikeda, T. & Ube, T. Materials science: a superelastic organic crystal. Nature 511, 300ā301 (2014).
Qu, H., Louhi-Kultanen, M. & Kallas, J. Additive effects on the solvent-mediated anhydrate/hydrate phase transformation in a mixed solvent. Cryst. Growth Des. 7, 724ā729 (2007).
Brandel, C. et al. Mechanisms of reversible phase transitions in molecular crystals: case of ciclopirox. Chem. Mater. 27, 6360ā6373 (2015).
Duan, Y. et al. Aggregation induced enhancement of linear and nonlinear optical emission from a hexaphenylene derivative. Adv. Funct. Mater. 26, 8968ā8977 (2016).
Kudernac, T. et al. Electrically driven directional motion of a four-wheeled molecule on a metal surface. Nature 479, 208ā211 (2011).
Å tacko, P. et al. Locked synchronous rotor motion in a molecular motor. Science 356, 964ā968 (2017).
Goodwin, A. L. Organic crystals: packing down. Nat. Mater. 9, 7ā8 (2010).
Matvienko, A. A. et al. Structural aspects of displacive transformations: what can optical microscopy contribute? Dehydration of Sm2(C2O4)3Ā·10H2O as a case study. IUCrJ 4, 588ā597 (2017).
Porter, D. A., Easterling, K. E. & Sherif, M. Phase Transformations in Metals and Alloys (Revised Reprint). (CRC Press, 2009).
Zheng, H. et al. Observation of transient structural-transformation dynamics in a Cu2S nanorod. Science 333, 206ā209 (2011).
Mirvakili, S. M. & Hunter, I. W. Artificial muscles: mechanisms, applications, and challenges. Adv. Mater. 30, 1704407 (2018).
Dye, D. Shape memory alloys: towards practical actuators. Nat. Mater. 14, 760ā761 (2015).
Wang, Y. & Khachaturyan, A. G. Three-dimensional field model and computer modeling of martensitic transformations. Acta Mater. 45, 759ā773 (1997).
ZĆ”rubovĆ”, N., Ge, Y., Heczko, O. & Hannula, S. P. In situ TEM study of deformation twinning in NiāMnāGa non-modulated martensite. Acta Mater. 61, 5290ā5299 (2013).
Huang, M. & Brinson, L. C. A Multivariant model for single crystal shape memory alloy behavior. J. Mech. Phys. Solids 46, 1379ā1409 (1998).
Song, Y., Chen, X., Dabade, V., Shield, T. W. & James, R. D. Enhanced reversibility and unusual microstructure of a phase-transforming material. Nature 502, 85ā88 (2013).
Wang, J. et al. In situ atomic-scale observation of twinning-dominated deformation in nanoscale body-centred cubic tungsten. Nat. Mater. 14, 594ā600 (2015).
Bhadeshia, H. K. D. H. Worked Examples in the Geometry of Crystals. (Institute of Materials, London 2006).
LonÄariÄ, I. et al. Reversible thermosalient effect of Nā²-2-Propylidene-4-hydroxybenzohydrazide accompanied by an immense negative compressibility: structural and theoretical arguments aiming toward the elucidation of jumping phenomenon. Cryst. Growth Des. 17, 4445ā4453 (2017).
Phinney, L. Baker, M. & Serrano, J. Thermal Microactuators. (BoD ā Books on Demand, 2011).
Arazoe, H. et al. An autonomous actuator driven by fluctuations in ambient humidity. Nat. Mater. 15, 1084ā1089 (2016).
Acome, E. et al. Hydraulically amplified self-healing electrostatic actuators with muscle-like performance. Science 359, 61ā65 (2018).
Pelrine, R., Kornbluh, R. & Kofod, G. High-strain actuator materials based on dielectric elastomers. Adv. Mater. 12, 1223ā1225 (2000).
Roundy, D., Krenn, C. R., Cohen, M. L. & Morris, J. W. Ideal Shear Strengths of fcc Aluminum and Copper. Phys. Rev. Lett. 82, 2713ā2716 (1999).
Meyers, M. A., Mishra, A. & Benson, D. J. Mechanical properties of nanocrystalline materials. Prog. Mater. Sci. 51, 427ā556 (2006).
Banerjee, A. et al. Ultralarge elastic deformation of nanoscale diamond. Science 360, 300ā302 (2018).
Bruker AXS Inc., (Madison, 2001)
Sheldrick, G. M. SADABS and TWINABS: Area-Detector Absorption Correction. (UniversitƤt Gƶttingen, 1999).
Sheldrick, G. M. SHELXTāintegrated space-group and crystal-structure determination. Acta Crystallogr. A. Found. Adv. 71, 3ā8 (2015).
Sheldrick, G. M. Crystal structure refinement with SHELXL. Acta Crystallogr. C. Struct. Chem. 71, 3ā8 (2015).
Plimpton, S. Fast parallel algorithms for short-range molecular dynamics. J. Comput. Phys. 117, 1ā19 (1995).
D. A. Case, et al. AMBER 2016. (University of California, San Francisco, 2016)
Schaftenaar, G., Vlieg, E. & Vriend, G. Molden 2.0: quantum chemistry meets proteins. J. Comput. Aided Mol. Des. 31, 789ā800 (2017).
Schaftenaar, G. & Noordik, J. H. Molden: a pre- and post-processing program for molecular and electronic structures. J. Comput. Aided Mol. Des. 14, 123ā134 (2000).
Blau, P. J. Friction Science and Technology: From Concepts to Applications 2nd edn. (CRC Press, 2008).
Acknowledgements
We thank K. Li from the Center for High Pressure Science and Technology Advanced Research (HPSTAR) in Beijing for help with the high temperature SXRD, and T. Toonen and E. Ronde from Radboud University for their technical supports. Y. D gratefully acknowledges the financial supports from the China Scholarship Council. Part of this work was supported by the Netherlands Organization of Scientific Research (NWO SPI63.256) and National Natural Science Foundation of China (21773168). Beamline BL17B of National Facility for Protein Science Shanghai (NFPS) at Shanghai Synchrotron Radiation Facility (SSRF) is acknowledged for providing assistance during data collection.
Author information
Authors and Affiliations
Contributions
Y.D., J.X., S.S., and T.R conceived the study. T.R, H.C., and J.X supervised the work. Y.D. carried out the experimental studies. P.T performed the SXRD measurements. H.C performed the MD simulation. The paper was initially drafted by Y.D. and T.R., and the molecular dynamic simulation was drafted by H.C with support from J.X, S.S, and P.T. All authors made comments in editing of the final paper and Supplementary material.
Corresponding authors
Ethics declarations
Competing interests
The authors declare no competing interests.
Additional information
Peer review information Nature Communications thanks Tomce Runcevski, Satoshi Takamizawa and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisherās note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Source Data
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the articleās Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the articleās Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Duan, Y., Semin, S., Tinnemans, P. et al. Robust thermoelastic microactuator based on an organic molecular crystal. Nat Commun 10, 4573 (2019). https://doi.org/10.1038/s41467-019-12601-y
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-019-12601-y
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.