Hydrogen-bond engineering induced ferroelastic phase transition in copper-based organic–inorganic hybrid thermochromic materials

Abstract

We report two organic–inorganic hybrid thermochromic crystals, [(2A5MP)₂CuCl₄] (2A5MP = 2-amino-5-methylpyridine) (1) and [(2A4MP)₂CuCl₄] (2A4MP = 2-amino-4-methylpyridine) (2). Through hydrogen bond engineering optimization, the ferroelastic material 2 undergoes an isomorphic phase transition with a large thermal hysteresis of 36 K at 397 K/361 K.

---

Ferroic materials, including ferromagnets, ferroelectrics, and ferroelastics, have been widely studied for their versatility in multiple fields.^1–3 Among them, ferroelastic materials exhibit spontaneous strain which can be switched under thermal stimulus or external stress.⁴ This characteristic endows them with application potential in the fields of information storage, energy conversion, and mechanical switches.⁵ To further enrich the application fields of the materials, the exploration of combining multiple physical properties into a single material is a long-term vision and pursuit,⁶ such as thermochromic properties, dielectricity, piezoelectricity, and so on.⁷ Among them, the thermochromic properties are expected to be potential properties for a variety of applications, including smart temperature-sensitive coatings and stealth materials.⁸ For example, a thermochromic ferroelastic (3,3-difluoropyrrolidium)₂CuCl₄ was reported recently, which exhibited significant reversible color changes, and was successfully used as an anti-counterfeiting ink for hand-painting.⁹ However, there is still a lack of an effective strategy for designing materials with both thermochromism and ferroelasticity.

In recent years, organic–inorganic hybrid (OIH) materials offer a unique combination of fascinating properties: they inherit the structural variability and flexibility of organic components and also exhibit the magnetic, electrical, and thermal properties characteristic of inorganic parts.¹⁰ This dual nature positions OIH materials as precious repositories for multifunctional materials.^11,12 The OIH materials also exhibit highly reconfigurable structures under external stimuli, resulting in color changes, phase transitions, domain changes, etc.,^13–15 which offer opportunities for structure design and function modulation by weak intermolecular forces such as hydrogen bonds.¹⁶ For instance, Fu et al. improved the phase transition temperature of (Me-Hdabco)Rb[BF₄]₃ by the introduction of new hydrogen bonds via replacing the cation [Medabco-F]²⁺ (Medabco-F = 1-fluoro-4-methyl-1,4-diazoniabicyclo[2.2.2]octane) with the cation [Me-Hdabco]²⁺ (Me-Hdabco = N-methyldabconium).¹⁷ Du et al. achieved modulation of photoluminescence in two OIH manganese halide compounds by adjusting the hydrogen bonding strength, thereby increasing the quantum yield.¹⁸ Therefore, hydrogen-bond engineering plays an important role in the regulation of multifunctional OIH materials, and may serve as an efficient approach to designing materials with both thermochromicity and ferroelasticity.

In this work, we successfully synthesized two OIH thermochromic crystals [(2A5MP)₂CuCl₄] (2A5MP = 2-amino-5-methylpyridine) (1) and [(2A4MP)₂CuCl₄] (2A4MP = 2-amino-4-methylpyridine) (2), respectively (Scheme S1, ESI†). In response to temperature variations, the color of 1 switches between yellow and orange, and 2 switches between green and yellow. By strategically altering the position of the methyl group on the pyridine ring, we can modulate the strength of the hydrogen bonding within the lattice. This adjustment facilitates the free rotation and tumbling of the cations and anions in 2, thereby giving rise to phase transition behaviors with a large thermal hysteresis of 36 K. Furthermore, we also found that 2 is a ferroelastic, and its ferroelastic domains can be switched under both temperature variations and stress stimulation. Not only the multi-channel physical switches based on 2 hold promise for applications in information processing and storage devices, but also this work demonstrates a strategy for the design and performance optimization of multifunctional materials.

The crystals of 2 and 1 exhibit reversible thermochromic properties from green to yellow and from yellow to orange, respectively (Fig. 1a and d). Solid-state UV-vis absorption spectra at different temperatures were obtained to study their color changes (Fig. 1b and e). As the temperature increases, the absorption edge of 2 shifts from 568 nm to 578 nm, and 1 shifts from 598 nm to 619 nm, which are consistent with their color changes. The band gap variations of compounds 1 and 2 during the heating process were determined using the Tauc equation. The results indicate a decrease in band gaps from 2.25 eV to 2.15 eV for compound 2 (Fig. 1c) and from 2.27 eV to 2.16 eV for compound 1 (Fig. 1f). These findings confirm that both compounds exhibit thermochromic properties. Furthermore, the gradual shifts in the absorption edges of compounds 1 and 2 suggest that both are potential indirect band gap semiconductors.¹⁹

Fig. 1 Color changes of 2 (a) and 1 (d) in the continuous heating and cooling processes. Variable-temperature UV-vis absorption spectra of 2 (b) and 1 (e). The extrapolated optical band gaps of 2 (c) and 1 (f).

To verify the occurrence of the phase transition behavior, thermal analysis and dielectric tests were performed on 1 and 2. For 1, differential scanning calorimetry (DSC) and temperature-dependent dielectric measurements reveal no anomaly peaks prior to the melting point, indicating that 1 did not undergo structural phase transition (Fig. S1a and b, ESI†). Fortunately, the DSC curves of 2 show a pair of thermal anomaly peaks at 397 K/361 K, respectively (Fig. 2a). Notably, a significant thermal hysteresis of 36 K is observed between the endothermic and exothermic peaks. This large discrepancy suggests that the phase transition is of the first order. In addition, the real part (ε′) of the dielectric constants of 2 at 100 kHz increases from 3.6 to about 39.5 upon heating (Fig. 2b). The dielectric anomalies are observed around 397 K/361 K upon heating and cooling, which are in good agreement with the DSC data. For convenience, the phase below the phase transition temperature is defined as the low-temperature phase (LTP), and the phase above the phase transition temperature is denoted as the high-temperature phase (HTP).

Fig. 2 (a) DSC curves of 2 with a scanning rate of 20 K min⁻¹. (b) Temperature-dependent ε′ of 2 at 100 kHz.

To further confirm its large thermal hysteresis behavior, the DSC curves of 2 at the scanning rates of 5 K min⁻¹ and 10 K min⁻¹ were also measured, respectively (Fig. S2a, ESI†). Through fitting analysis, it can be inferred that when the scanning rate is 0 K min⁻¹, the thermal hysteresis of 2 is 30.25 K (Fig. S2b, ESI†). The results indicate that the scanning rate does not exert a significant influence on the thermal hysteresis of compound 2. Even at a very slow scanning rate, the thermal hysteresis remains larger than 30 K. Furthermore, during the heating and cooling processes, 2 exhibits a relatively large average enthalpy change (ΔH) of 5518.8 J mol⁻¹, showing great application potential in the field of thermal energy storage.^20,21 Additionally, thermogravimetry analysis (TGA) shows that 2 and 1 have thermal stability up to 469 K and 485 K, respectively (Fig. S3a and b, ESI†).

For elucidation of the microscopic mechanism underlying the structural phase transition, variable-temperature single-crystal X-ray diffraction (XRD) characterization was performed. At 298 K, both 2 and 1 crystallize in the monoclinic centrosymmetric space group C2/c (point group 2/m), and detailed crystallographic data are shown in Table S1 (ESI†), which are consistent with previous reports.^22–24 As illustrated in Fig. 3a and Fig. S4a (ESI†), at 298 K, the asymmetric units of compounds 2 and 1 each consist of an aminomethylpyridine cation and half of a [CuCl₄]²⁻ anion at 298 K. Each central Cu atom is coordinated by four Cl atoms to form a [CuCl₄]²⁻ tetrahedron. The [CuCl₄]²⁻ tetrahedron exhibits significant deviations from the traditional tetrahedral geometry due to non-equivalent bond distances and angles, as detailed in Tables S2 and S3 (ESI†), which result in a notably twisted structure. As shown in Fig. 3b and Fig. S5a (ESI†), the crystal stacking diagram of 2 at 298 K shows the ABAABA sequenced arrangement of 2-amino-4-methylpyridine cations and [CuCl₄]²⁻ anions, forming a zero-dimensional stacking structure. The 2-fold axis parallel to the b-axis penetrates the center of each [CuCl₄]²⁻ tetrahedron, and the organic amine cations are arranged in a regular pattern around the 2-fold axis (Fig. S6, ESI†). Hydrogen bonds are formed between every two 2-amino-4-methylpyridine cations and one [CuCl₄]²⁻ anion. The detailed hydrogen bonding information is shown in Table S4 (ESI†). The average distance between hydrogen bonding donors and acceptors of 2 is 3.397 Å. The stacking diagram of 1 is shown in Fig. S4b (ESI†). The basic structural units of 1 are connected by N–H⋯Cl hydrogen bonds (hydrogen bonding information is shown in Table S5, ESI†). The distance between the donor and acceptor is 3.254 Å. Significantly, the hydrogen bond length of 2 is longer than that of 1, illustrating weaker hydrogen bond strength than that of 1. Therefore, the 2-amino-4-methylpyridine cations and the [CuCl₄]²⁻ anions of 2 are able to rotate and roll to a certain extent, which may be the main factor for triggering structural phase transition.

Fig. 3 Crystal structures of 2 at 298 K (a) and 410 K (c). Stacking diagrams of 2 along the b-axis at 298 K (b) and 410 K (d).

At 410 K, 2 crystallizes in monoclinic centrosymmetric space group C2/c (point group 2/m), with a subtle change of lattice parameters and average hydrogen bond length (from 3.397 Å to 3.347 Å), indicating an isomorphic phase transition at 397 K of 2 (Fig. 3c, d and Fig. S5b, Tables S1 and S7, ESI†).

As shown in Fig. S7a–c (ESI†), the powder X-ray diffraction (PXRD) spectra match well with the simulated PXRD patterns of the corresponding single crystal structures at 298 K, revealing good phase purity of both 1 and 2. For 2, variable temperature PXRD patterns were recorded within the temperature range from 299 K to 411 K (Fig. S7d, ESI†). As the temperature increases to 411 K, the stronger diffraction peak at 26° is shifted slightly to the left. In addition, the number of diffraction peaks at 14.55° and 27.07° decreases from two to one. This result means that the symmetry of the crystal structure undergoes a slight change, further proving the occurrence of isomorphic phase transition.

Hirshfeld surfaces and the related 2D fingerprint plots are effective tools for qualitative and quantitative studies on intermolecular interactions. From the Hirshfeld surface pattern (Fig. 4a and c), it can be seen that the N–H⋯Cl hydrogen bonds formed between anions and cations contribute a lot to the intermolecular forces of 2 and 1. From the 2D fingerprint plots (Fig. 4b and d), it can be directly seen that the overall contribution of H⋯Cl interatomic force to structure stacking in 2 (28.8% of the total weak interactions) is lower than that in 1 (31.3% of the total weak interactions). In other words, the total hydrogen bond strength of 2 is weaker than that of 1, facilitating cation tumbling and the occurrence of the phase transition of 2.

Fig. 4 Hirshfeld surfaces of 2 (a) and 1 (c) at 298 K. 2D fingerprint plots of 2 (b) and 1 (d) at 298 K.

Polarized light microscopy is an effective tool to observe ferroelastic domains. The thin films of 2 were observed in situ under a polarized light microscope. Surprisingly, at 303 K, the strip-shaped domains can be observed after first annealing (Fig. 5a). Subsequently, the second annealing was performed on the ferroelastic domains. As the temperature increases, the domains gradually weaken and then disappear as the temperature reaches 413 K (Fig. 5b). When the temperature reaches 313 K, the strip-shaped domains appear again (Fig. 5c). Moreover, the number of stripe domains is slightly different from that of the initial state. Ferroelastic domains of 2 can appear or disappear repeatedly with the rise and fall of temperature, which indicates that they can be switched under thermal stimulus. Because 2 undergoes an isomorphic phase transition at 397 K, it does not belong to the 94 species of ferroelastic phase transitions. Therefore, we speculate that the paraelastic phase occurs after the melting point.

Fig. 5 Ferroelastic domains of 2 at 303 K (a), 413 K (b), and 313 K (c) in the process of heating and cooling. The topography of 2 at 303 K (d), 413 K (e), and 313 K (f).

Except for the thermal stimulus, ferroelastic domains should be switchable under external stress stimulation. To verify the response of the ferroelastic domains of 2 under external stress stimulation, an appropriate force was applied to the thin film sample of 2. According to Fig. 6, it can be clearly seen that the number of local strip-shaped domains changes after being subjected to force (white box: from 5 to 7; yellow box: from 3 to 5). The domains of 2 respond to external stress stimulus, which proves the ferroelasticity of 2. In addition, nanoindentation tests were performed on crystal 2 for further exploration of its ferroelasticity. The nanoindentation technique is one of the powerful methods to test the mechanical properties of materials on the nano- and micro-scales.^25,26 Especially, when performing indentation tests on ferroelastics, the stress field imposed by the pyramid-shaped indentor could drive ferroelastic domain movements, giving rise to discontinuous serrations on both loading and unloading segments, namely “pop-in” and “pop-out”, respectively.²⁷ As shown in Fig. S8 (ESI†), the “pop-in” and “pop-out” serrations were observed on both loading and unloading curves, indicating the ferroelastic domain movement under external force, which further confirms that 2 is a ferroelastic.

Fig. 6 The evolution of ferroelastic domains of 2 under external stress stimuli observed using a polarized light microscope. (a) Initial state. (b) The state after applying external stress.

In summary, by adjusting the position of the methyl group in aminomethylpyridine, the strength of the N–H⋯Cl hydrogen bonding between the aminomethylpyridine cation and the [CuCl₄]²⁻ anion is regulated. We successfully obtained a copper-based OIH thermochromic ferroelastic 2. Compared to 1, 2 not only exhibits thermochromic properties but also ferroelasticity. The strip-shaped domains were observed under a polarized light microscope, and these domains can be switched under dual channels of temperature and mechanical stress simultaneously. Furthermore, the isomorphic phase transition of 2 at 397 K has a large thermal hysteresis of 36 K, which is expected to achieve thermal energy storage and release in a wide temperature range. As a potential candidate for multifunctional materials, it will expand the application range of intelligent switching materials while meeting the increasingly sought-after information security requirements.

This work was financially supported by the National Natural Science Foundation of China (22375082 and 22105094).

Data availability

The data that support the findings of this study are available from the corresponding author upon reasonable request.

Conflicts of interest

There are no conflicts to declare.

References

1. W. Luo, N. Wang, H.-K. Li, Z.-J. Xu, Y. Feng, X.-B. Fu, C. Shi, H.-Y. Ye and L.-P. Miao, Inorg. Chem. Front., 2023, 10, 5082–5088.
2. X.-X. Chen, X.-Y. Zhang, D.-X. Liu, R.-K. Huang, S.-S. Wang, L.-Q. Xiong, W.-X. Zhang and X.-M. Chen, Chem. Sci., 2021, 12, 8713–8721.
3. Q.-J. Zhou, L. He, K. Xu, X.-B. Han, T.-J. Yin, W. Zhang and Q. Ye, Adv. Opt. Mater., 2024, 12, 2302292.
4. T.-Y. Ju, C.-C. Fan, B.-D. Liang, C.-D. Liu, M.-L. Jin, C.-Y. Chai and W. Zhang, Adv. Funct. Mater., 2024, 34, 2316747.
5. J.-N. Chen, K.-X. Huang, P.-M. Cheng, M.-Q. Qi, H. Xu, J. Chen, Y. Duan, X.-J. Kong, L.-S. Zheng and L.-S. Long, J. Am. Chem. Soc., 2024, 146, 22913–22917.
6. Z. Li, C.-M. Ji, Y.-P. Fan, T.-T. Zhu, S.-H. You, J.-B. Wu, R.-Q. Li, Z.-K. Zhu, P.-P. Yu, X.-J. Kuang and J.-H. Luo, J. Am. Chem. Soc., 2023, 145, 25134–25142.
7. T.-M. Guo, F.-F. Gao, Y.-J. Gong, Z.-G. Li, F. Wei, W. Li and X.-H. Bu, J. Am. Chem. Soc., 2023, 145, 22475–22482.
8. Y.-P. Zeng, J.-C. Liu, L. Zhou, X. Deng, W.-L. Yang, X. Yan, Y.-L. Luo, X. Zhu, X.-Y. Huang, X.-J. Song and Y.-Y. Tang, Chin. Chem. Lett., 2023, 34, 108127.
9. S.-L. Jiao, Z. Tang, D. Li, C.-P. Xu, Z. Yang, P. Chen, X.-F. Sun, H.-L. Cai and X.-S. Wu, Chem. Eng. J., 2023, 466, 143188.
10. S.-G. Han, J. Bie, W. Fa, S. Chen, L.-W. Tang, W.-Q. Guo, H.-J. Xu, Y. Ma, Y. Liu, X.-T. Liu, Z.-H. Sun and J.-H. Luo, J. Am. Chem. Soc., 2024, 146, 8298–8307.
11. J.-H. Wu, B. Zhang, T.-K. Jiang, F. Kong and J.-G. Mao, Chin. J. Struct. Chem., 2023, 42, 100016.
12. J.-J. Fang, Z. Liu, Y.-P. Xie and X. Lu, Chin. J. Struct. Chem., 2023, 42, 100155.
13. L.-Y. Sheng, D.-C. Han, R.-K. Huang, L.-M. Cao, C.-T. He, Z.-Y. Du and T. Nakamura, J. Am. Chem. Soc., 2024, 146, 22893–22898.
14. Y.-Z. Tang, Y.-M. Yu, J.-B. Xiong, Y.-H. Tan and H.-R. Wen, J. Am. Chem. Soc., 2015, 137, 13345–13351.
15. X. Zhu, L. Zhou, X.-Y. Huang, Y.-L. Luo, X. Deng, X. Yan, Y.-J. Wang, Y. Qin and Y.-Y. Tang, Chin. J. Struct. Chem., 2024, 43, 100272.
16. K. Sambe, T. Takeda, N. Hoshino, W. Matsuda, K. Shimada, K. Tsujita, S. Maruyama, S. Yamamoto, S. Seki, Y. Matsumoto and T. Akutagawa, J. Am. Chem. Soc., 2024, 146, 8557–8566.
17. T. Zhang, K. Ding, J.-Y. Li, G.-W. Du, L.-L. Chu, Y. Zhang and D.-W. Fu, Chin. J. Chem., 2022, 40, 1559–1565.
18. L.-K. Gong, Q.-Q. Hu, F.-Q. Huang, Z.-Z. Zhang, N.-N. Shen, B. Hu, Y. Song, Z.-P. Wang, K.-Z. Du and X.-Y. Huang, Chem. Commun., 2019, 55, 7303–7306.
19. P. H. M. Andrade, C. Volkringer, T. Loiseau, A. Tejeda, M. Hureau and A. Moissette, Appl. Mater. Today, 2024, 37, 102094.
20. Y. Sui, G.-X. Zhang, W.-Q. Wang, F. Hu, C.-L. Liu, D. Luo and D.-S. Liu, ChemistrySelect, 2019, 4, 3921–3925.
21. G.-S. Wei, G. Wang, C. Xu, X. Ju, L.-J. Xing, X.-Z. Du and Y.-P. Yang, Renewable Sustainable Energy Rev., 2018, 81, 1771–1786.
22. R. H. Al-Far and B. F. Ali, Acta Crystallogr. E, 2008, 65, m73–m74.
23. S. Gatfaoui, N. Issaoui, O. Noureddine, T. Roisnel and H. Marouani, J. Iran. Chem. Soc., 2021, 18, 2331–2343.
24. H. Place and R. D. Willett, Acta Crystallogr. C, 1987, 43, 1050–1053.
25. W. C. Oliver and G. M. Pharr, J. Mater. Res., 1992, 7, 1564–1583.
26. S. Saha, M. K. Mishra, C. M. Reddy and G. R. Desiraju, Acc. Chem. Res., 2018, 51, 2957–2967.
27. N. Orlovskaya, Y. Gogotsi, M. Reece, B. Cheng and G. Ion, Acta Mater., 2002, 50, 715–723.

---

Footnote

† Electronic supplementary information (ESI) available: Fig. S1–S9 and Tables S1–S7. CCDC 2379925, 2379926, 2387503. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d4cc04373a

---