Open access peer-reviewed chapter

Superelastic Behaviors of Molecular Crystals

Written By

Takuya Taniguchi

Submitted: 22 May 2023 Reviewed: 27 May 2023 Published: 20 October 2023

DOI: 10.5772/intechopen.1001971

From the Edited Volume

Shape Memory Alloys - New Advances

Mohammad Asaduzzaman Chowdhury and Mohammed Muzibur Rahman

Chapter metrics overview

49 Chapter Downloads

View Full Metrics

Abstract

Molecular crystals have medium mechanical properties between inorganic alloys and organic polymers. The material category of molecular crystals has recently shown unique mechanical responses induced by external stimuli such as light, heat, and force. This review explores the superelasticity of molecular crystals, a phenomenon first discovered by Takamizawa et al. in 2014. Molecular crystals can manifest superelasticity by much smaller stresses than typical shape memory alloys, reflecting weaker intermolecular interactions of molecular crystals. A novel photo-responsive occurrence of superelastic deformation was observed in a chiral salicylideneamine crystal, exhibiting photoisomerization and phase transition. This process, involving torsional bending and superelastic deformation within a single crystal, could offer new functionalities in photo-responsive materials. Furthermore, it was found that superelasticity is prevalent across the molecular space by an informatics approach. As data accumulate, materials informatics may unveil the underlying relationship between superelasticity and the structures of molecular crystals, potentially enabling innovative material design.

Keywords

  • superelasticity
  • molecular crystals
  • mechanical property
  • actuation
  • finite element analysis
  • materials informatics

1. Introduction

Superelasticity, the phenomenon of material’s returning to the original shape even after large deformation beyond the elastic limit, has been observed only in some specific shape memory alloys. Superelasticity is a unique behavior unlike the normal elastic and plastic deformation occurring in all materials (Figure 1a). Shape memory materials deform irreversibly in the temperature range below the martensitic transition, and then the original shape is recovered upon heating due to the transition to the austenite phase. This phenomenon is known as the shape memory effect. On the other hand, in the temperature range above the martensitic transition temperature, even if a load beyond the elastic limit is applied to the material, the material of austenite phase deforms largely, and then immediately returns to its original shape upon unloading due to stress-induced martensitic phase (Figure 1b). This phenomenon is known as superelasticity. For example, nickel-titanium (NiTi) alloys are well-known as shape memory alloys, with elastic moduli ranging from 30 to 80 GPa and strain magnitudes from 0.5 to 5% (Figure 2) [2, 3]. Compared to soft organic polymers with a small modulus of 1 MPa ∼ 1 GPa and a large strain of 1 ∼ 100%, shape memory alloys are more rigid.

Figure 1.

Schematic illustration of the stress–strain curve of (a) typical elastic and plastic deformations and (b) superelastic deformation.

Figure 2.

Relationship between Young’s modulus and strain of actuation materials. Reprinted with permission from [1]. Copyright 2019 John Wiley and Sons.

Molecular crystals have medium mechanical properties between those of hard inorganic materials and soft polymers (Figure 2) [1, 4]. The elastic modulus is about 1 ∼ 20 GPa, and the strain is about 0.1 ∼ 1% [5]. These mechanical properties originate from the ordered nature of crystals and the weak intermolecular interactions of organic molecules. The periodic arrangement of organic molecules in the structure leads to the ordered formation of intermolecular interactions, resulting in a large elastic modulus. On the other hand, the intermolecular interactions of organic molecules are composed of hydrogen bonds, π-π interactions, and van der Waals forces, so the interactions are smaller than those of inorganic materials [6]. Molecular crystals are attracting attention as novel stimuli-responsive materials called soft crystals because they take advantage of these moderate mechanical properties to undergo structural changes in response to relatively weak stimuli [7, 8]. For example, crystal deformations, such as bending, twisting, and jumping, appear upon light irradiation or temperature change [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]. The driving mechanism is the crystal structure change through photoisomerization or structural phase transition, and the deformation mode depends on the structural difference before and after the change. In addition, various types of crystals have been discovered, including elastic crystals that deform greatly under loading [20, 21, 22, 23, 24, 25, 26] and plastic crystals that quickly develop plastic deformation in a very narrow elastic region [27, 28, 29, 30]. Furthermore, in this decade, superelastic phenomena, observed only in shape memory alloys, have been discovered in molecular crystals, attracting attention as a new type of shape memory material.

The superelasticity of molecular crystals is based on structural phase transition or twinning transition. These mechanisms also relate to other mechanical phenomena: ferroelasticity, actuation, shape-memory effect, and self-healing (Figure 3). The mechanical responses can be utilized for applications such as sensing, optics, soft robot, and catheter. Although the application to devices requires other specifications, such as the cost and the ease of integration with current technologies, researchers are trying to construct devices using molecular crystals, as explained in a later section.

Figure 3.

Typical mechanical responses of molecular crystals induced by external stimuli and examples of potential applications.

This review briefly summarizes the superelastic phenomenon in molecular crystals. Various superelastic crystals have been found since their discovery in 2014 [31]. In addition, crystals that exhibit a unique phase transition, named the photo-triggered phase transition, exhibit superelasticity during crystal actuation. The last section discussed the potential applications of superelastic molecular crystals and the structural features of molecules that exhibit superelasticity in molecular crystals. As a prospect, materials informatics may construct strategies for designing superelastic molecular crystals.

Advertisement

2. Discovery of superelasticity in molecular crystals

The superelasticity of molecular crystals was discovered by Takamizawa et al. in 2014 [31]. The molecule terephthalamide (Figure 4a) yields plate-like crystals with the crystal structure of space group P1¯. The crystal is in the α phase before loading, and when a load is applied to the (010) side face, the crystal deforms by creating a new crystal phase, as seen by the phase boundary (Figure 4b). Upon unloading, the new crystal phase gradually vanished, accompanying the movement of the phase boundary (Figure 4b). The new crystalline phase was assigned as the metastable β phase by X-ray diffraction measurement (Figure 4c). This molecular crystal was the first example of superelasticity.

Figure 4.

Superelasticity of terephthalamide crystal. (a) the molecular structure of terephthalamide. (b) Pictures during superelastic deformation. The crystal is fixed at the left side, and loaded from the below right. (c) Assignment of crystal phases and face indices during superelastic deformation. Reprinted with permission from [31]. Copyright 2014 John Wiley and Sons.

Measuring the stress–strain curve during the crystal deformation is crucial to characterize the superelastic property. Takamizawa et al. measured the stress–strain curve and simultaneously observed crystal deformation under a polarized microscope (Figure 5). Initially, the platelet terephthalamide crystal fixed on a stand was the α phase. Upon loading, the β phase was created, as seen in different colors due to the change of molecular orientation and birefringence property (Figure 5a). During the observation, the time profile of the applied force was monitored, corresponding to snapshots of the crystal deformation (Figure 5b). The stress–strain curve was obtained from this measurement and displayed hysteresis of stress, indicative of superelasticity (Figure 5c).

Figure 5.

(a) Snapshots of crystal deformation under a polarized microscope. Due to the simultaneous measurement, each snapshot was assigned to the time profile. (b) Time profile of applied force. Inset is the result of 100 cycles. (c) Stress–strain curve. Inset is the repetition result of 100 cycles. Reprinted with permission from [31]. Copyright 2014 John Wiley and Sons.

This deformation is due to a structural phase transition: loading on the (010) plane of the α-phase induces shear stress, which induces a transition to the metastable β-phase. Comparing the superelasticity of terephtalamide molecular crystal and typical NiTi alloy is essential to characterize the differences in superelastic properties. The energy storage of terephtalamide crystal was 62 kJ m−3, 226 times smaller than 14 MJ m−3 of typical NiTi alloy [31, 32]. This smaller energy storage of the molecular crystal is consistent with the smaller lattice energy compared with alloys. In other words, superelastic molecular materials can produce large deformations by small energy inputs. Thus, it was demonstrated that noncovalent interactions can produce superelasticity with great precision and can be controlled more precisely than in metallic alloys.

After this discovery, various superelastic molecular crystals have been found [33, 34, 35, 36, 37, 38, 39, 40]. The primary mechanism of expressing superelasticity was based on mechanical twinning. Crystal twinning generally occurs during crystal growth. It occurs when two or more adjacent crystals of the same molecule are symmetrically oriented to share some of the same crystal lattice points, and two separate crystals grow tightly bound. The plane on which the twin lattice points are shared is called the twinning plane. Twinning is also manifested by shear stress on the crystal, which is the deformation mechanism of some superelastic molecular crystals (Figure 6). Sasaki summarized molecular structures which have been reported to manifest superelasticity or ferroelasticity induced by mechanical twinning [41]. Ferroelasticity is also the mechanism for large deformation like superelasticity, but the deformation does not return just by removing the load. Such ferroelasticity was often found in molecular crystals [42, 43]. Figure 6 displays that there are diverse molecules that exhibit superplasticity and ferroelasticity. In addition, some molecular crystals manifest both effects by twinning mechanism.

Figure 6.

Representative examples of superelastic molecular crystals by mechanical twinning.

Advertisement

3. Superelasticity of actuating crystal

3.1 Actuation by a photo-triggered phase transition

In the superelastic phenomena described in the previous section, a force is input, and superelastic deformation occurs as output. On the other hand, in actuation, energy other than force is the input, and force is the output extracted from material deformation. Thus, although superelasticity and actuation are essentially different phenomena, it was found that superelastic deformation occurs during actuation in chiral salicylideneamine crystal (Figure 7a) [44, 45].

Figure 7.

(a) Molecular structure of chiral salicylideneamine. (b) Photograph and illustration of crystal shape and face index. Scale bar is 1 mm. (c-e) Molecular packing and energy framework, viewed from (c) (100) side face, (d) (010) cross-section face, and (e) (001) top face. Adapted from [44] under a creative commons attribution 4.0 international license (http://creativecommons.org/licenses/by/4.0/).

Two relatively strong intermolecular interactions are formed in the plate-like crystal of the chiral salicylideneamine (Figure 7b): the CH-π interaction and the CH-O hydrogen bond. CH-π interactions are mainly formed by the face-to-face stacking of naphthyl rings along the a- and b-axes, and CH-O interactions are formed between the OH group of one molecule and the tert-butyl group of the adjacent molecule along the b-axis (Figure 7c). The rigid interaction layers are stacked along the c-axis via weaker van der Waals interactions between bulky tert-butyl substituents. The energy framework was calculated to evaluate the strength of this intermolecular interaction, and it was confirmed that the strong and weak interaction layers (Figure 7c). Also, when viewed from the (010) plane, the strong and weak interaction layers are arranged alternately; these layers are spread two-dimensionally along the a- and b-axes (Figure 7d). When viewed from the (001) plane, only the stronger interactions are seen due to the layer-by-layer stacking (Figure 7e).

Light irradiation on the (001) plane of the crystal caused a unique phase transition called the photo-triggered phase transition (Figure 8). Ultraviolet (UV) light irradiation of the crystal induces photoisomerization from the enol to the trans-keto form, which generates stress in the crystal structure. When the generated stress accumulates to a certain amount, a phase transition of the crystal structure is induced to reduce the stress, and finally, a new crystal phase is formed (Figure 8). Here, it should be noted that photoisomerization occurs only in some parts of the bulk crystal near the irradiated surface, and a small number of the trans-keto form (approximately 5%) induce the phase transition of the entire crystal.

Figure 8.

Proposed mechanism of the photo-triggered phase transition and the measured lattice angles. Chiral salicylideneamine molecules in green and yellow reflect Z′ = 2, and the trans-keto form is shown in red. Adapted from [44] under a creative commons attribution 4.0 international license (http://creativecommons.org/licenses/by/4.0/).

Because photoirradiation of the crystal causes photoisomerization and photo-triggered phase transitions, UV irradiation of the (001) plane leads to complicated crystal deformation. The deformation is divided into three steps (Figure 9a). The first step is a simple bending toward the light source due to photoisomerization. The second step is twisted bending due to the progression of the photo-triggered phase transition. The third stage is a simple bend toward the light source due to photoisomerization after the completion of the transition. Finally, when the light is stopped, the shape returns to its original shape reversibly.

Figure 9.

(a) Photographs of typical deformation of the chiral salicylideneamine crystal, fixed on a glass plate, irradiated by UV light (365 nm). Scale bar is 1 mm. (b) Definition of torsion angle θ and displacements δ1 and δ2. The dotted lines are the initial position. (c) Cross-section view when irradiated on the (001) face, and (d) time profile of θ, δ1, and δ2. (e) Cross-section view when irradiated on (001¯) face, and (f) time profile of θ, δ1, and δ2. Scale bars in (c) and (e) are 0.5 mm. The regions highlighted in purple represent UV light at 180 mW cm−2. Adapted from [44] under a creative commons attribution 4.0 international license (http://creativecommons.org/licenses/by/4.0/).

A cross-section view can quantify complicated actuation (Figure 9b). Displacements at both edges are represented as δ1 and δ2, and the torsional angle as θ. Snapshots showed the actuation behavior observed from the cross-section view, and the time profile of displacements and torsional angle were measured (Figure 9c-f). Notably, the twisted bending finished within 1 second upon light irradiation, indicating the photo-triggered phase transition started and finished during that time.

3.2 Simulation of crystal deformation

Torsional bending is attributed to the photo-triggered phase transition, which arises during the initial photo process. For controlling the actuation behavior and understanding the actuation mechanism, it is crucial to simulate and replicate the material mechanics. Finite element analysis (FEA), commonly used for mechanical simulation, is suitable for the purpose and requires a simplified actuation model. The chiral salicylideneamine crystal displayed a stepwise actuation pattern: simple bending, torsional bending, and then simple bending again. Hence, it is considered a dynamic multilayer model comprising h1 as the photoisomerization layer, h2 as the transition layer, and h3 as the remaining layer (Figure 10a). The crystal thickness is denoted as h0. Considering light irradiation from the top surface, h1 and h2 are assumed to originate from the top surface. As the maximum photoproduct generation is 5%, h1 can be much smaller than h0, and the maximum value of h2 can be equal to h0. FEA was conducted based on this model with the primary objective of reproducing simple and torsional bending. Once achieved, the effect of h1 and h2 thickness on deformation was examined to understand the progression of photoisomerization and the photo-triggered phase transition within the crystal.

Figure 10.

FEA-simulation of crystal deformation. (a) Dynamic multi-layer model. (b) Independent assumed deformations of h1 and h2 layers. The original dimensions of the plate object are 4.0 mm in length and 0.94 mm in width. Deformation is enhanced 6.4 times because raw deformation is much smaller than the object size. (c) Simulated typical deformations. (d, e) simulated dependence of (d) torsion angle and (e) maximum displacement on the thicknesses of h1 and h2 layers. Blue dots are the simulated points, and the response surfaces are drawn by a polynomial function. Red lines are the estimated route that reproduces the observed torsion angle and displacement. (f, g) comparison of the simulation and the torsion angle in the (f) photo-process and (g) relaxation process. (h, i) comparison of the simulation and the displacement in the (h) photo-process and (i) relaxation process. Reprinted from [44] under a creative commons attribution 4.0 international license (http://creativecommons.org/licenses/by/4.0/).

For FEA, the material’s dimensions were 4.0 mm in length, 0.96 mm in width, and 50 μm in thickness (h0), approximately equivalent to the crystal shown in Figure 9. Then, pure effects of photoisomerization and photo-triggered phase transition were replicated: h1 undergoes longitudinal shrinkage (−0.8%), and h2 undergoes shear deformation (Δ1°) with slight shrinkage (−0.05%) (Figure 10b). These pure effects were replicated based on the results of X-ray diffraction measurements and implemented by thermal and piezoelectric effects because photo effects cannot be directly incorporated into FEA. Then, FEA successfully reproduced the simple bending, torsional bending, and consecutive simple bending by changing h1, h2, and h3 (Figure 10c).

Then, the influence of h1 and h2 on deformation was evaluated. The values of h1 and h2 were systematically varied within the range of 0 to 5 μm and 0 to 50 μm, respectively. Polynomial regression was then employed to construct response surfaces illustrating the relationship between the torsion angle, maximum displacement, and the two thickness parameters (Figure 10d,e). Although this response function does not incorporate a time factor, assuming that h1 is proportionate to the progression of the photoisomerization reaction, it can be expressed using the following equation

h1t=h1,max1exptτ1pwhent0E1

Here, h1,max represents the maximum depth of the photoisomerization layer, t denotes the duration of light irradiation, and τ1p represents the time constant. The progression of h2 is also influenced by photoisomerization, which induces a phase transition. Therefore, it is posited that h2 can be described by a similar exponential function, characterized by an independent time constant and a certain delay, tdelay, from t = 0.

h2t=h2,max1exptτ2pttdelayE2

where h2,max denotes the maximum depth of the transition, and τ2p represents the time constant. As h2,max can be equivalent to h0, the optimization process involves determining the optimal values for the remaining parameters: h1,max, τ1p, τ2p, and tdelay. To account for the relaxation process in the analysis, time-related factors are introduced under the assumption that h1 and h2 depend on the reverse chemical reaction as follows

h1t=h1,maxexptτ1rt0E3
h2t=h2,maxexptτ2rttdelayE4

where τ1r and τ2r represent the relaxation time constants. In Eq. (3) and (4), τ1r, τ2r, and tdelay are the parameters to be optimized, as the others have been determined in the photo process.

Following parameter optimization, the FEA-based simulations demonstrated a good fit with the observed torsion angle and displacement of the crystal during both the photo and relaxation processes (Figure 10fi), albeit with some discrepancies in displacement. Notably, this simulation outperformed the classical Stoney’s bimorph model in predicting displacement behavior (Figure 10h,i).

The optimized results shed light on the progression of the h1 and h2 layers within the crystal (depicted by the red lines in Figure 10d,e). During the photo-process, h1,max attained a value of 4.7 μm, while the h2 layer commenced with a delay time of 0.38 s at a rate ten times faster than the h1 layer, resulting in the maximum torsion angle occurring when h2 = h0/2. Subsequently, the h2 layer reached h0 = 50 μm, followed by a subsequent increase in the h1 layer up to h1,max. In the relaxation process, only h1 initially decreased, and then h2 began to decrease when h1 reached approximately 1.5 μm. The relaxation speeds of h1 and h2 were found to be equal based on the fitting. Notably, there exists hysteresis between the photo and relaxation processes. Considering that photoproducts are responsible for inducing stress, which in turn triggers the phase transition, and that the release of stress without UV light leads to the reverse transition, this hysteresis can be considered analogous to the stress–strain curve of superelasticity. Thus, it can be concluded that superelasticity manifests in the photo-triggered phase transition due to the stress induced by photoproducts.

Advertisement

4. Other mechanical effects: shape memory and self-healing

Structural phase transition relates to superelasticity and shape memory effects of molecular crystals. It is reported that terephthalic acid crystals can undergo a mechanically induced phase transition without delamination upon bending while retaining their overall crystal integrity [46, 47]. Plastically bent crystals exhibit bimorph behavior, and their phase uniformity can be restored thermally by raising the crystal’s temperature above the phase transition point. This thermal treatment recovers the original straight shape, and a reverse thermal treatment can induce shape memory effects similar to those observed in certain metal alloys and polymers. It is anticipated that similar memory and restorative effects are common in other molecular crystals with metastable polymorphs. This example insists on the advantage of utilizing intermolecular interactions to achieve mechanically adaptive properties in organic solids.

Self-healing of molecular crystals is less related to superelasticity but is crucial in the recovery of mechanical integrity even after appearing the crystal fracture. The first finding of self-healing in molecular crystals was discovered by Commins et al. using crystals of dipyrazolethiuram disulfide [48]. They broke a piece of single crystal into two pieces and then compressed it so that the cut surfaces of the two pieces touched each other. A healing degree of 6.7% was observed after mild compression of these crystals. It was hypothesized that the self-healing property of the material is attributed to the disulfide shuffling mechanism. The crystal structure has three close S − S contacts, which could form new bonds at the interfaces of two crystals.

Another mechanism also achieved self-healing. Bhunia et al. found that piezoelectric molecular crystals were autonomously self-healed in milliseconds with crystallographic precision [49]. This self-healing was very fast due to the unnecessity of long-time compression. The self-healing mechanism arises from stress-induced electrical charges on fracture surfaces, leading to the precise recombination of the broken pieces through an electrostatically driven, diffusionless self-healing process. Although the self-healing of molecular crystals has yet to encounter many examples, such crystals will improve the durability and robustness of devices if other self-healing crystals are found.

Advertisement

5. Prospects

5.1 Applications of superelastic molecular crystals

As briefly explained in the introduction, molecular crystals have various mechanical properties and may lead to applications. For example, superelastic and ferroelastic organic semiconductor crystals can be used in a flexible device [50, 51]. Flexible single-crystal electronic devices were achieved by leveraging the bending-induced ferroelastic transition of an organic semiconductor crystal. These devices can withstand strains of over 13% while preserving the charge carrier mobility of unstrained crystals. This advancement lays the groundwork for high-performance ultra-flexible single-crystal organic electronics, opening doors to their utilization in sensors, memories, and robotic applications. In addition, superelasticity in the actuating crystal may work for lifting and moving objects, photoswitches, and micromechanical devices.

Superelasticity and shape memory effect in nanoparticles also have the potential for future applications. Experimental studies have demonstrated that the shape memory properties of ceramics can be enhanced by reducing the density of grain boundaries. For example, Zhang et al. observed a superelastic strain of 8.3% in single-crystal nanoparticles of zirconia-based ceramics, which could be fully recovered by simply removing the compressive load [52]. However, they also confirmed that the shape memory behavior deteriorated as the number of loading-unloading-heating–cooling cycles increased. This degradation of shape memory properties was attributed to the formation of an amorphous phase, its accumulation around the load contact area, and an increase in surface roughness. If the decline in function of nanoparticles is controlled, nanoparticle-based materials may be implemented or replaced with the current technologies. A similar strategy using nanoparticles will be applied to molecular crystals to improve the superelasticity and other mechanical properties.

5.2 Materials informatics for molecular design

As mentioned above, many molecular crystals have been reported to exhibit superelasticity, but the theoretical science behind this phenomenon has not yet been established. Therefore, it is crucial to find the structural characteristics of the molecules in which superelasticity is observed.

In order to understand the structural characteristics of molecules, it is necessary to represent molecules mathematically and compare the closeness between molecules. Therefore, we collected information from the literature on molecules that exhibit superelasticity, ferroelasticity, thermal phase transitions, and mechanochromic luminescence in crystals. The mechanochromic molecular crystals are summarized in [53], and thermal phase transitions are summarized in [54]. Then, the molecular structures were vectorized by Mordred descriptor [55]. Each dimension of the Mordred vectors was standardized to have the mean 0 and variance 1 and then reduced to two dimensions by a data embedding method, uniform manifold approximation and projection (UMAP). The molecular dataset and Python code are available via [56].

The scatter plot shows that molecules exhibiting superelasticity, ferroelasticity, and thermal phase transition have similar distributions and are widely distributed in the 2-dimensional manifold space (Figure 11). This result indicates that molecules exhibiting superelasticity, ferroelasticity, and thermal phase transition are widely distributed in the material space with structural diversity. On the other hand, the molecules exhibiting mechanochromic luminescence are relatively tightly distributed, indicating that their molecular structures are similar. This difference can be attributed to the need for molecular structures involved in luminescence and that studies have been conducted with similar molecular structures with different substituents.

Figure 11.

Scatter plot of Mordred vectors embedded by UMAP. In the legend, thermal PT is a thermal phase transition, tFE is ferroelasticity by twinning, tSE is superelasticity by twinning, and mechanoCML is mechanochromic luminescence.

On the other hand, the relationship between superelasticity and molecular structure is still unclear. Therefore, it is necessary to find hidden patterns in the data, and this approach is called materials informatics (MI). Academia and many chemical and materials companies are developing materials utilizing MI, mainly focusing on industrially important inorganic solid materials and polymeric materials [57, 58, 59]. Although there are not many examples of MI research for molecular crystals compared to those materials, the number of reported cases is increasing, such as using machine learning for predicting polymorphism in pentacene crystals by Musil et al. [60]. As for superelasticity, if more data becomes available for machine learning, MI may be able to discover hidden relationships between molecular and crystal structures and develop them efficiently. Furthermore, MI can be applied to other mechanical properties, such as Young’s modulus and maximum strain. Since materials’ space is vast for human labor, MI-assisted materials design will benefit material research.

Advertisement

6. Conclusions

This review deals with the superelasticity of molecular crystals. The superelasticity of molecular crystals was discovered by Takamizawa et al. in 2014, and many superelastic and ferroelastic crystals have been discovered. It has been quantitatively evaluated that superelasticity develops at stresses as small as about 1/200th of NiTi, a typical shape memory alloy, showing the unique mechanical properties of molecular crystals. It was also found that superelastic deformation occurs during actuation in crystals in which photo-triggered phase transitions occur upon light irradiation. By observing and simulating the torsional deformation of chiral salicylideneamine crystal and clarifying the deformation mechanism, we have found that light-induced superelastic deformation occurs for the first time. This result is a unique photo-responsive phenomenon in which torsional deformation and superelastic deformation occur in a single crystal and may lead to new functionalities in photo-responsive materials. Other mechanical responses, shape memory, and self-healing, of molecular crystals were also described briefly. Finally, future applications and the possibility of materials informatics for molecular crystals were summarized. It was found that the molecules exhibiting superelasticity are widely distributed in the material space. As the number of data increases, the hidden relationship between superelasticity and molecular crystals and crystal structures can be captured by materials informatics and may be used for new material design.

Advertisement

Acknowledgments

The author thanks Mr. Daisuke Takagi and Mr. Ryo Fukasawa at Waseda University for curating molecular data for structural feature analysis. This study was financially supported by JSPS Grant-in-Aid (20H04677, 22 K14747).

Advertisement

Conflict of interest

The authors declare no conflict of interest.

References

  1. 1. Koshima H, Taniguchi T, Asahi T. Mechanically responsive crystals by light and heat. In: Koshima H, editor. Mechanically Responsive Materials for Soft Robotics. New York: Wiley; 2019. pp. 57-87. DOI: 10.1002/9783527822201.ch3
  2. 2. Inamura T, Hosoda H, Wakashima K, Miyazaki S. Anisotropy and temperature dependence of Young’s modulus in textured TiNbAl biomedical shape memory alloy. Materials transactions. 2005;46:1597-1603. DOI: 10.2320/matertrans.46.1597
  3. 3. Namazu T, Hashizume A, Inoue S. Thermomechanical tensile characterization of NiTi shape memory alloy films for design of MEMS actuator. Sensors and Actuators A: Physical. 2007;139:178-186. DOI: 10.1016/j.sna.2006.10.047
  4. 4. Taniguchi T, Asahi T, Koshima H. Photomechanical azobenzene crystals. Crystals. 2019;9:437. DOI: 10.3390/cryst9090437
  5. 5. Wang C, Sun CC. The landscape of mechanical properties of molecular crystals. CrystEngComm. 2020;22:1149-1153. DOI: 10.1039/C9CE01874C
  6. 6. Desiraju GR. Supramolecular synthons in crystal engineering—A new organic synthesis. Angewandte Chemie International Edition. 1995;34:2311-2327. DOI: 10.1002/anie.199523111
  7. 7. Kato M, Ito H, Hasegawa M, Ishii K. Soft crystals: Flexible response systems with high structural order. Chemistry–A European Journal. 2019;25:5105-5112. DOI: 10.1002/chem.201805641
  8. 8. Kato M, Ishii K, editors. Soft crystals: Flexible response systems with high structural order. Singapore: Springer Nature; 2023. p. 265. DOI: 10.1007/978-981-99-0260-6
  9. 9. Naumov P, Chizhik S, Panda MK, Nath NK, Boldyreva E. Mechanically responsive molecular crystals. Chemical Reviews. 2015;115:12440-12490. DOI: 10.1021/acs.chemrev.5b00398
  10. 10. Naumov P, Karothu DP, Ahmed E, Catalano L, Commins P, Mahmoud Halabi J, et al. The rise of the dynamic crystals. Journal of the American Chemical Society. 2020;142:13256-13272. DOI: 10.1021/jacs.0c05440
  11. 11. Kitagawa D, Nishi H, Kobatake S. Photoinduced twisting of a photochromic diarylethene crystal. Angewandte Chemie. 2013;125:9490-9492. DOI: 10.1002/ange.201304670
  12. 12. Kitagawa D, Tsujioka H, Tong F, Dong X, Bardeen CJ, Kobatake S. Control of photomechanical crystal twisting by illumination direction. Journal of the American Chemical Society. 2018;140:4208-4212. DOI: 10.1021/jacs.7b13605
  13. 13. Al-Kaysi RO, Tong F, Al-Haidar M, Zhu L, Bardeen CJ. Highly branched photomechanical crystals. Chemical Communications. 2017;53:2622-2625. DOI: 10.1039/C6CC08999B
  14. 14. Tong F, Xu W, Guo T, Lui BF, Hayward RC, Palffy-Muhoray P, et al. Photomechanical molecular crystals and nanowire assemblies based on the [2+ 2] photodimerization of a phenylbutadiene derivative. Journal of Materials Chemistry C. 2020;8:5036-5044. DOI: 10.1039/C9TC06946A
  15. 15. Taniguchi T, Fujisawa J, Shiro M, Koshima H, Asahi T. Mechanical motion of chiral azobenzene crystals with twisting upon photoirradiation. Chemistry–a. European Journal. 2016;22:7950-7958. DOI: 10.1002/chem.201505149
  16. 16. Hagiwara Y, Taniguchi T, Asahi T, Koshima H. Crystal actuator based on a thermal phase transition and photothermal effect. Journal of Materials Chemistry C. 2020;8:4876-4884. DOI: 10.1039/D0TC00007H
  17. 17. Taniguchi T, Sugiyama H, Uekusa H, Shiro M, Asahi T, Koshima H. Walking and rolling of crystals induced thermally by phase transition. Nature Communications. 2018;9:538. DOI: 10.1038/s41467-017-02549-2
  18. 18. Ishizaki K, Sugimoto R, Hagiwara Y, Koshima H, Taniguchi T, Asahi T. Actuation performance of a photo-bending crystal modeled by machine learning-based regression. CrystEngComm. 2021;23:5839-5847. DOI: 10.1039/D1CE00208B
  19. 19. Taniguchi T, Kubota A, Moritoki T, Asahi T, Koshima H. Two-step photomechanical motion of a dibenzobarrelene crystal. RSC advances. 2018;8:34314-34320. DOI: 10.1039/C8RA06639F
  20. 20. Ahmed E, Karothu DP, Naumov P. Crystal adaptronics: Mechanically reconfigurable elastic and superelastic molecular crystals. Angewandte Chemie International Edition. 2018;57:8837-8846. DOI: 10.1002/anie.201800137
  21. 21. Worthy A, Grosjean A, Pfrunder MC, Xu Y, Yan C, Edwards G, et al. Atomic resolution of structural changes in elastic crystals of copper (II) acetylacetonate. Nature Chemistry. 2018;10:65-69. DOI: 10.1038/nchem.2848
  22. 22. Thompson AJ, Orué AIC, Nair AJ, Price JR, McMurtrie J, Clegg JK. Elastically flexible molecular crystals. Chemical Society Reviews. 2021;50:11725-11740. DOI: 10.1039/D1CS00469G
  23. 23. Annadhasan M, Basak S, Chandrasekhar N, Chandrasekar R. Next-generation organic photonics: The emergence of flexible crystal optical waveguides. Advanced Optical Materials. 2020;8:2000959. DOI: 10.1002/adom.202000959
  24. 24. Hayashi S, Koizumi T. Mechanically induced shaping of organic single crystals: Facile fabrication of fluorescent and elastic crystal fibers. Chemistry–A European Journal. 2018;24:8507-8512. DOI: 10.1002/chem.201801965
  25. 25. Hayashi S, Koizumi T. Elastic organic crystals of a fluorescent π-conjugated molecule. Angewandte Chemie. 2016;128:2751-2754. DOI: 10.1002/ange.201509319
  26. 26. Hayashi S, Ishiwari F, Fukushima T, Mikage S, Imamura Y, Tashiro M, et al. Anisotropic Poisson effect and deformation-induced fluorescence change of elastic 9, 10-Dibromoanthracene single crystals. Angewandte Chemie International Edition. 2020;59:16195-16201. DOI: 10.1002/anie.202006474
  27. 27. Das S, Mondal A, Reddy CM. Harnessing molecular rotations in plastic crystals: A holistic view for crystal engineering of adaptive soft materials. Chemical Society Reviews. 2020;49:8878-8896. DOI: 10.1039/D0CS00475H
  28. 28. Das S, Saha S, Sahu M, Mondal A, Reddy CM. Temperature-reliant dynamic properties and Elasto-plastic to plastic crystal (rotator) phase transition in a metal oxyacid salt. Angewandte Chemie International Edition. 2022;61:e202115359. DOI: 10.1002/anie.202115359
  29. 29. Mondal A, Bhattacharya B, Das S, Bhunia S, Chowdhury R, Dey S, et al. Metal-like ductility in organic plastic crystals: Role of molecular shape and dihydrogen bonding interactions in Aminoboranes. Angewandte Chemie International Edition. 2020;59:10971-10980. DOI: 10.1002/anie.202001060
  30. 30. Saha S, Mishra MK, Reddy CM, Desiraju GR. From molecules to interactions to crystal engineering: Mechanical properties of organic solids. Accounts of Chemical Research. 2018;51:2957-2967. DOI: 10.1021/acs.accounts.8b00425
  31. 31. Takamizawa S, Miyamoto Y. Superelastic organic crystals. Angewandte Chemie. 2014;126:7090-7093. DOI: 10.1002/ange.201311014
  32. 32. Pieczyska E, Gadaj S, Nowacki WK, Hoshio K, Makino Y, Tobushi H. Characteristics of energy storage and dissipation in TiNi shape memory alloy. Science and Technology of Advanced Materials. 2005;6:889-894. DOI: 10.1016/j.stam.2005.07.008
  33. 33. Takamizawa S, Takasaki Y. Superelastic shape recovery of mechanically twinned 3, 5-Difluorobenzoic acid crystals. Angewandte Chemie. 2015;127:4897-4899. DOI: 10.1002/ange.201411447
  34. 34. Takasaki Y, Sasaki T, Takamizawa S. Temperature-diversified anisotropic superelasticity and ferroelasticity in a 3-methyl-4-nitrobenzoic acid crystal. Crystal Growth & Design. 2020;20:6211-6216. DOI: 10.1021/acs.cgd.0c00964
  35. 35. Mutai T, Sasaki T, Sakamoto S, Yoshikawa I, Houjou H, Takamizawa S. A superelastochromic crystal. Nature Communications. 2020;11:1824. DOI: 10.1038/s41467-020-15663-5
  36. 36. Sasaki T, Sakamoto S, Engel ER, Takamizawa S. An Organosuperelastic mechanism with bending molecular chain bundles. Crystal Growth & Design. 2021;21:2920-2924. DOI: 10.1021/acs.cgd.1c00089
  37. 37. Sakamoto S, Sasaki T, Sato-Tomita A, Takamizawa S. Shape Rememorization of an Organosuperelastic crystal through Superelasticity–Ferroelasticity interconversion. Angewandte Chemie. 2019;131:13860-13864. DOI: 10.1002/ange.201905769
  38. 38. Sasaki T, Nishizawa K, Takamizawa S. Versatile Organosuperelastic deformability by multiple mechanical twinning. Crystal Growth & Design. 2021;21:2453-2458. DOI: 10.1021/acs.cgd.1c00054
  39. 39. Sasaki T, Ranjan S, Takamizawa S. Perpendicularly oriented dual Organosuperelasticity correlated with molecular symmetry. Crystal Growth & Design. 2021;21:3902-3907. DOI: 10.1021/acs.cgd.1c00210
  40. 40. Sasaki T, Ranjan S, Takamizawa S. A photoluminescent organosuperelastic crystal of 7-amino-4-methylcoumarin. CrystEngComm. 2021;23:5801-5804. DOI: 10.1039/D1CE00359C
  41. 41. Sasaki T. Mechanical twinning in organic crystals. CrystEngComm. 2022;24:2527-2541. DOI: 10.1039/D2CE00089J
  42. 42. Mir SH, Takasaki Y, Takamizawa S. An organoferroelasticity driven by molecular conformational change. Physical Chemistry Chemical Physics. 2018;20:4631-4635. DOI: 10.1039/C7CP07206F
  43. 43. Seki T, Feng C, Kashiyama K, Sakamoto S, Takasaki Y, Sasaki T, et al. Photoluminescent ferroelastic molecular crystals. Angewandte Chemie International Edition. 2020;59:8839-8843. DOI: 10.1002/anie.201914610
  44. 44. Taniguchi T, Ishizaki K, Takagi D, Nishimura K, Shigemune H, Kuramochi M, et al. Superelasticity of a photo-actuating chiral salicylideneamine crystal. Communications Chemistry. 2022;5:4. DOI: 10.1038/s42004-021-00618-8
  45. 45. Taniguchi T, Sato H, Hagiwara Y, Asahi T, Koshima H. Photo-triggered phase transition of a crystal. Communications Chemistry. 2019;2:19. DOI: 10.1038/s42004-019-0121-8
  46. 46. Karothu DP, Weston J, Desta IT, Naumov P. Shape-memory and self-healing effects in mechanosalient molecular crystals. Journal of the American Chemical Society. 2016;138:13298-13306. DOI: 10.1021/jacs.6b07406
  47. 47. Ahmed E, Karothu DP, Warren M, Naumov P. Shape-memory effects in molecular crystals. Nature Communications. 2019;10:3723. DOI: 10.1038/s41467-019-11612-z
  48. 48. Commins P, Hara H, Naumov P. Self-healing molecular crystals. Angewandte Chemie International Edition. 2016;55:13028-13032. DOI: 10.1002/anie.201606003
  49. 49. Bhunia S, Chandel S, Karan SK, Dey S, Tiwari A, Das S, et al. Autonomous self-repair in piezoelectric molecular crystals. Science. 2021;373:321-327. DOI: 10.1126/science.abg3886
  50. 50. Park SK, Sun H, Chung H, Patel BB, Zhang F, Davies DW, et al. Super- and ferroelastic organic semiconductors for ultraflexible single-crystal electronics. Angewandte Chemie. 2020;132:13104-13112. DOI: 10.1002/ange.202004083
  51. 51. Sun H, Park SK, Diao Y, Kvam EP, Zhao K. Molecular mechanisms of superelasticity and ferroelasticity in organic semiconductor crystals. Chemistry of Materials. 2021;33:1883-1892. DOI: 10.1021/acs.chemmater.1c00080
  52. 52. Zhang N, Zaeem MA. Superelasticity and shape memory effect in zirconia nanoparticles. Extreme Mechanics Letters. 2021;46:101301. DOI: 10.1016/j.eml.2021.101301
  53. 53. Ito S. Recent advances in mechanochromic luminescence of organic crystalline compounds. Chemistry Letters. 2021;50:649-660. DOI: 10.1246/cl.200874
  54. 54. Takagi D, Ishizaki K, Asahi T, Taniguchi T. Molecular screening for solid–solid phase transition by machine learning. ChemRxiv. DOI: 10.26434/chemrxiv-2022-8z976-v2. [Preprint]
  55. 55. Moriwaki H, Tian YS, Kawashita N, Takagi T. Mordred: a molecular descriptor calculator. Journal of Cheminformatics. 2018;10:1-14. DOI: 10.1186/s13321-018-0258-y
  56. 56. Github. Available from: https://github.com/takuyhaa/Superelasticity-manifold/
  57. 57. Butler KT, Davies DW, Cartwright H, Isayev O, Walsh A. Machine learning for molecular and materials science. Nature. 2018;559:547-555. DOI: 10.1038/s41586-018-0337-2
  58. 58. Himanen L, Geurts A, Foster AS, Rinke P. Data-driven materials science: Status, challenges, and perspectives. Advanced Science. 2019;6:1900808. DOI: 10.1002/advs.201900808
  59. 59. Schmidt J, Marques MR, Botti S, Marques MA. Recent advances and applications of machine learning in solid-state materials science. npj Computational Materials. 2019;5:83. DOI: 10.1038/s41524-019-0221-0
  60. 60. Musil F, De S, Yang J, Campbell JE, Day GM, Ceriotti M. Machine learning for the structure–energy–property landscapes of molecular crystals. Chemical Science. 2018;9:1289-1300. DOI: 10.1039/C7SC04665K

Written By

Takuya Taniguchi

Submitted: 22 May 2023 Reviewed: 27 May 2023 Published: 20 October 2023