Record‐Low Compressibility in [N(C₂H₅)₃CH₃]FeCl₄ Expands Phase Engineering Horizons in Hybrid Molecular Ferroelectrics

ABSTRACT

Advances in hybrid organic–inorganic ferroelectrics (HOIFs) plateau; high‐pressure studies offer powerful approaches to deepen understanding of structure‐property relationships and enable rational phase engineering of lead‐free HOIFs as viable inorganic alternatives. To date, high‐pressure studies focus on metal–organic frameworks (MOFs) and lead‐based HOIFs, constrained by toxicity, stability, and limited pressure ranges. Structure‐stability‐compressibility correlations prove essential for understanding lead‐free HOIFs under ultrawide pressure conditions, yet remain scarcely explored. We present high‐pressure studies of the lead‐free HOIF [N(C₂H₅)₃CH₃]FeCl₄ (EMAFC), which stays stable and mechanochromic up to 51.5 GPa with a reversible P6₃ mc‐to‐P1 phase transition at 0.75 GPa. Pressure‐triggered synchrotron powder X‐ray diffraction, Raman, UV–vis, dielectric, and second‐harmonic‐generation switching data reveal coupling between structural changes and bandgap modulation. With a bulk modulus of K ₀ = 42.0(5) GPa, EMAFC sets a record for compressibility among HOIFs. Beyond low compressibility, the tunability of EMAFC manifests through reversible retention of the SHG “on” state up to 9.5 GPa and transition to “off” by 20.0 GPa. Our results show that halide choice and lattice dynamics govern the compressibility and functional properties of HOIFs. EMAFC exhibits the lowest compressibility reported, establishing key structure‐compressibility relationships and enabling advanced phase and property control in hybrid ferroelectrics.

High‐pressure studies of the lead‐free hybrid organic–inorganic ferroelectric [N(C₂H₅)₃CH₃]FeCl₄ demonstrate record low compressibility, establishing it as the least compressible molecular ferroelectric known. In situ diamond anvil cell experiments identify halide selection and inorganic sublattice tuning as key compressibility controls, advancing green chemistry principles toward eco‐friendly ferroelectric alternatives.

1. Introduction

The ferroelectric materials sector is expected to experience significant growth, driven by their increasing use in energy harvesting applications for electronic and innovative technologies [1]. The intrinsic phase transition behavior and tunable physical properties of these materials underpin their suitability for use in switches and information storage devices. Although inorganic ferroelectrics are known for their high spontaneous polarization, piezoelectric coefficients, and low or anomalous compressibilities, their widespread application is restricted by high processing temperatures, mechanical rigidity, high production costs, and the use of toxic heavy metals such as lead. In contrast, the development of hybrid organic–inorganic ferroelectrics (HOIFs) over the past decade has introduced new design paradigms [2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15]. Solution‐processable HOIFs offer several advantages over conventional inorganic counterparts, including lower toxicity, reduced weight, easy processability, and greater compositional and structural diversity [15, 16, 17, 18, 19]. Significant progress has been made in optimizing key functional parameters of these molecular ferroelectrics, including Curie temperature (T _c), piezoelectric coefficient (d ₃₃), and pyroelectric coefficient [20, 21]. For example, You et al. (2017) reported that (C₄H₁₁Cl₂N)MnCl₅, exhibits a large piezoelectric coefficient (d ₃₃ = 185 pC/N) and a high phase‐transition temperature (T _c = 406 K), surpassing that of BaTiO₃ (T _c = 390 K, d ₃₃ = 190 pC/N) [9]. Similarly, Liao et al. (2019) demonstrated that [Me₃NCH₂F]_x[Me₃NCH₂Cl]_1‐xCdCl₃ possesses an exceptional d ₃₃ value of 1540 pC/N, comparable to commercial Pb(Zr,Ti)O₃ [21]. These favorable properties have enabled the application of HOIFs in energy harvesting [22], self‐powered mechanical sensing [8, 23], biological interfaces [24, 25, 26, 27], and, more recently, advanced information technologies for anticounterfeiting and encryption [28]. The versatility of HOIFs arises from their tunable frameworks, in which metal cations are coordinated by bridging ligands (organic or inorganic), and the cavities are occupied by strategically selected cations [9, 10, 11, 12, 13, 14, 15, 29] Namely, the choice of inorganic anion has a pronounced influence on the lattice and electronic structure, while the organic cation modulates lattice parameters, dielectric constants, hydrogen bonding interactions with halide anions, and octahedral distortions. Selective alteration of organic cations, combined with the construction of different inorganic anionic frameworks via intermolecular interactions, confers multifunctionality at the molecular level. Although the functional properties of HOIFs represent viable alternatives that may surpass traditional inorganic ferroelectrics, further performance enhancements through current synthetic design alone are becoming challenging. It is well‐known that the functional properties of HOIFs are exquisitely sensitive to their molecular and crystal structure. The application of external pressure provides a direct pathway to modify these properties and explore new functionalities. Pressure can induce changes in carrier lifetime, lattice disorder, compressibility, bandgap narrowing, octahedral tilting and shrinkage, and enhanced electrical tunability [12, 30, 31, 32, 33, 34, 35]. Additionally, HOIFs can exhibit pressure‐induced phenomena such as color changes, crystalline‐to‐amorphous transitions, insulator‐to‐metal transitions, and both reversible and irreversible structural alterations, resulting in dramatic and tunable changes in functional properties. This demonstrates the potential of compositional engineering for pressure‐tunable functionalities. Although most devices are engineered to operate optimally under ambient conditions, one may question the rationale for subjecting materials to extreme environments such as high pressure. Our motivation lies in the current challenges facing the field–particularly the performance gap between lead‐free HOIFs and their lead‐based counterparts. A key objective is to expand understanding of the structure‐property relationship in lead‐free HOIFs and to establish robust correlations among material stability, compressibility, and detailed structural information. Achieving long‐term stability, compressibility responsiveness, and high switchable performance in lead‐free hybrid materials with non‐toxic or less toxic components remains a major challenge in modern photovoltaics.

Recently, we conducted comprehensive in situ high‐pressure measurements on [N(C₂H₅)₃CH₃]FeBr₄ (EMAFB), revealing its remarkable flexibility in photophysical response under varying pressure conditions [12]. Beyond the well‐studied perovskites, which often suffer from intrinsic stability issues, this lead‐free HOIF presents a promising alternative for hybrid optoelectronic materials. For comparison, Zhang et al. (2019) reported pressure‐induced broadband emission in the 2D hybrid lead‐based perovskite (C₆H₆C₂H₄NH₃)₂PbBr₄; however, such materials remain highly susceptible to degradation from oxygen, moisture, light, and heat [36, 37, 38]. On top of everything, structural phase transitions are the most extensively studied phenomena in MAPbX₃‐type HOIFs under compression, while research on materials with abnormal compressibility has focused mainly on metal–organic frameworks (MOFs) and all‐inorganic frameworks [39, 40, 41, 42, 43, 44]. MOFs, with diverse organic linkers, display flexible mechanical responses such as large negative and near‐zero linear compressibility. However, their softness limits these effects to pressures below 10 GPa, restricting their practical application. Designing these materials requires explicit structure‐property correlations across an ultrawide pressure range. Despite breakthroughs in phase engineering, the precise roles of both organic and inorganic ions in determining the compressibility of HOIFs remain poorly understood, particularly for powder samples, due to the scarcity of systematic research to date. These observations highlight the critical need to achieve both exceptional ambient stability and robust switchable performance in lead‐free HOIFs, while simultaneously advancing our understanding of their structure‐property interplay under ambient and non‐ambient conditions. Although substantial research has focused on the synthesis of novel HOIF structures and the investigation of their optoelectronic properties, systematic studies examining the mechanical behavior of HOIFs remain comparatively scarce. Motivated by the pronounced instability of Pb‐and Sn‐based HOIFs and the utility of pressure as a versatile tuning parameter for simultaneously modulating structural, electronic, magnetic, optical, and elastic properties, we report the intricate pressure‐induced mechanochromic behavior of the environmentally friendly lead‐free compound triethylmethylammonium tetrachloroferrate(III), [N(C₂H₅)₃CH₃]FeCl₄ (EMAFC). This HOIF exhibits exceptionally low compressibility – an uncommon characteristic among molecular ferroelectrics–and demonstrates remarkable long‐term ambient stability alongside outstanding functional properties, as established in previous studies [11].

In this work, we systematically investigate the physical response of EMAFC to mechanical compression using a diamond anvil cell (DAC). We elucidate the interplay between structural transitions and bandgap evolution through a combination of in situ synchrotron high‐pressure powder X‐ray diffraction (SP‐XRD), Raman spectroscopy, absorption, and emission spectroscopy, dielectric measurements, and second‐harmonic generation (SHG) “on‐off” switching studies. We demonstrate that EMAFC exhibits exceptionally low compressibility, due to uniform lattice resistance over an ultrawide pressure range of 0.01–51.5 GPa–markedly deviating from behavior reported for similar HOIFs to date. Together, our high‐pressure investigations of EMAFC, in comparison with recent results on the isostructural analogue EMAFB, underscore the potential for pressure‐tuned functionalities in molecular ferroelectrics and provide new perspectives for the engineering of multifunctional materials designed to perform under extreme conditions. The remarkable properties exhibited by EMAFC–including low compressibility, reduced permittivity, and substantial resistance to pressure‐induced distortion–render it highly suitable for advanced communication technologies, robust insulating coatings, and a broad spectrum of applications where both electrical and mechanical integrity are essential. Low‐compressibility organic–inorganic materials are driving advances in technologies that require exceptional resistance to deformation under stress–particularly in next‐generation optical, electronic, and mechanical systems–and are recognized as contributing to the clean transition.

2. Results and Discussion

2.1. Structural Evolution Under High Pressure and RT

Only in the case of the sample at ambient pressure and after decompression (0.01 GPa) we carried out with success the ab‐initio structure solution process using EXPO2014 software. The structure crystallized under ambient conditions in the hexagonal system, s. g. P6₃ mc (no. = 186), and the refined structure model was strongly overlaying with that one described in our previous work (Figure 1a,b and Table 1). Specifically, under ambient conditions, the refined hexagonal unit cell parameters were a = 8.1427(7) Å, c = 13.145(2) Å, V = 754.77(1) ų, and the corresponding weighted‐profile agreement factor was R _wp = 3.22%. Besides the hexagonal EMAFC phase, featuring a completely disordered [N(C₂H₅)₃CH₃]⁺ cation within the lattice and a [FeCl₄]⁻ anion exhibiting typical six‐fold symmetry of a slightly distorted trigonal pyramid, no other phases were detected. SP‐XRD data confirm that the inorganic framework comprises [FeCl₄]⁻ anions charge‐balanced by [N(C₂H₅)₃CH₃]⁺ cations. Both ions lie on the hexagonal axis, similar to the isostructural ferroelectrics EMAFB and [N(C₂H₅)₄]FeCl₄ [10, 12, 45]. The final refined structural model obtained from diffraction data collected under ambient conditions was very similar to that reported in [11]. The two structures showed a root‐mean‐square deviation (RMSD) of 0.018 Å, indicating excellent structural similarity, where RMSD = sqrt(Σ_i d _i ²/N _au), that is, the square root of the average squared distances between pairs of corresponding atoms in the two compared models, with N _au being the number of atoms in the asymmetric unit. At a low pressure of 0.75 GPa, a new phase (Figure 1b) emerges with unit cell lattice metrics: a = 8.627(2) Å, b = 7.709(2) Å, c = 13.073(2) Å, α = 91.894(4)°, β = 84.623(5)°, γ = 119.846(3)°, V = 750.8(3) ų, and a lower‐symmetry space group P1 (Figure 1b). P6₃ mc and P1 follow the group‐subgroup hierarchy, G > H (P6₃ mc > P1), where G is the supergroup and H is the subgroup. In conventional ferroelectric materials at ambient pressure, the high‐temperature paraelectric phase–characterized by centrosymmetric crystal structure–transforms upon cooling into a low‐temperature ferroelectric phase with non‐centrosymmetric symmetry, resulting in a well‐defined symmetry‐breaking transition. This process is a hallmark of ferroelectric behavior, underpinning the emergence of spontaneous electric polarization in the low‐temperature phase [7]. Herein the one‐step hexagonal‐triclinic phase transition is accompanied with Aizu notation 6/mmmF1 [46, 47]. Hexagonal hybrid molecular materials predominantly exhibit one‐or two‐step ferroelastic transitions, with multistep transitions representing exceptionally rare occurrences in this materials class [11, 12, 48, 49, 50, 51, 52]. The observed phase transition arises from mechanisms analogous to those described by Fu and Ye (2017), involving both the dynamic reorganization of organic cations–which serve to buffer compressive stresses within the hybrid framework–and the cooperative displacement of metal and halogen ions throughout the crystal lattice [48, 53]. Remarkably, the triclinic phase exhibited exceptional stability up to 51.5 GPa, as evidenced by SP‐XRD measurements, which revealed no significant changes in symmetry. However, a systematic shift toward higher °2θ values under increasing hydrostatic pressure indicated progressive unit cell contraction and reduced interplanar spacing. Strikingly, EMAFC exhibits markedly enhanced pressure stability compared to isostructural EMAFB, whose monoclinic high‐pressure phase persists only up to 16.1 GPa. This difference highlights the critical influence of distinct binding mechanisms within their electronic structures. Our findings underscore the urgent need for in‐depth investigation of the binding mechanisms between inorganic frameworks and organic constituents to advance phase engineering of next‐generation HOIF‐based materials. Despite significant progress, the precise binding mechanism between organic cations and anions in organic–inorganic hybrid systems, appears unclear and, in some cases, seems contradictory. Recent work by Liu et al. has demonstrated that the bond strength between the A⁺ cation and BX₃ ⁻ anion can serve as a descriptor linking a decreasing atomic number at the X site with enhanced electronic structure stability [54]. Both the Coulomb effect and orbital polarization play roles in modulating this bond strength, though their individual contributions differ. For example, as the atomic number of the halogen increases from fluorine to iodine, the electron cloud around the halogen atom becomes more diffuse, resulting in altered charge distributions within the ABX₃ framework. In accordance with Coulomb's law, this results in a progressive weakening of the Coulomb interaction between hydrogen and the halogen. Specifically, the electron accumulation on the halogen atoms decreases in the sequence F > Cl > Br > I, corresponding to a gradual reduction in the Coulomb effect. Taken together, stronger bond strength is associated with a larger Coulomb effect, a smaller perovskite unit cell volume, and a narrower energy gap between the lowest unoccupied molecular orbital of the organic cation and the highest occupied molecular orbital of the anion. These insights offer valuable guidance for designing highly stable organic–inorganic hybrid systems. In parallel, the elastic properties of molecular ferroelectrics–particularly promising multiferroics such as EMAFC and EMAFB–are critically important for technological applications, as they directly affect the durability and reliability of devices during operation [12, 55, 56, 57]. Despite their growing technological relevance, there have been almost no investigations specifically focused on analyzing the elastic properties of this class of molecular ferroelectrics. Remarkably, only a very limited number of studies have provided comprehensive examinations of the elastic characteristics of molecular perovskite ferroelectrics [58, 59, 60]. A complete ferroelectric/ferroelastic 2F1‐type phase transition was studied in depth by Xiong and Zhang (2022) [61]. Taking into account the space group P1 and the cell parameters preliminarily refined by EXPO2014 and further optimized by GSAS‐II [62], and estimating intensities using the Le Bail algorithm–without refining or applying a structural model for reflection intensity calculations–we find that a straightforward point‐by‐point mapping of the observed powder pattern intensities to each reflection results in progressively improved profile fitting (Figure 2a–c). In particular, at a pressure of 0.75 GPa, changes in the diffraction pattern, including an increased number of peaks, indicated that a transition to a lower‐symmetry phase had occurred. Unlike the data collected under ambient conditions, the ab initio structure solution process failed for the diffraction pattern measured at 0.75 GPa, likely due to peak broadening, which prevented reliable indexing. By relaxing the hexagonal symmetry and assuming space group P1, a refinement of the hexagonal unit cell parameters determined from ambient condition data was performed, using these parameters as starting values for Rietveld refinement with EXPO. The parameters provided by EXPO were further optimized by the Rietveld method using GSAS‐II. In the absence of a structural model, the Rietveld refinement was performed by optimizing only the profile parameters and assigning the intensities to reflections via the Le Bail method. The assumption of symmetry lowering at the pressure point of 0.75 GPa was supported by the metrics of the final refined unit cell, characterized by a not negligible differentiation of the six cell parameters (i.e., a≠b≠c and α≠β≠γ≠90°, see Table 1). A similar trend was observed in case of SP‐XRD patterns collected above 0.75 GPa, during both pressure increase or decrease, for which the refinement in P1 is carried out using, as starting values for the Rietveld refinement, the final refined unit cell parameters obtained in case of the pattern collected at the nearest pressure point, without the need of a structure model to be used for calculating the reflection intensities, that were cyclically estimated and assigned via the Le Bail method. Table 1 lists the unit cell parameters resulting from: (i) the Rietveld refinement of EMAFC under ambient conditions before pressure was applied, (ii) the Rietveld refinement coupled with Le Bail fitting of the SP‐XRD high‐pressure data with respect to the s. g. P1 and (iii) the Rietveld refinement coupled with Le Bail fitting of the powder pattern collected from the relaxed EMAFC sample at 0.01 GPa.

FIGURE 1.

(a) The evolution of SP‐XRD patterns for EMAFC is presented in a waterfall‐style plot, with critical pressure points highlighted in the inset at the top, while selected SP‐XRD patterns, shown in offset mode at the bottom, illustrate the reversible transition between hexagonal and triclinic phases. (b) Observed (red circles) and calculated (solid blue line) SP‐XRD profiles obtained from the Rietveld refinement against the SP‐XRD data of EMAFC at ambient conditions before compression (top) and the Rietveld refinement coupled with Le Bail fitting following gradual release to 0.01 GPa (bottom). The difference between observed and calculated profiles is represented by the lower solid gray line, while purple tick marks indicate the reflection positions of EMAFC. The fitted background contribution is shown by the lower solid green lines. For enhanced clarity, the diffraction patterns at the top and bottom have been enlarged twice in the region from 12.5° to 28°.

TABLE 1.

A summary of the unit cell metrics and reliability factors obtained by GSAS‐II from Rietveld refinement of SP‐XRD data for EMAFC under ambient conditions (λ = 0.6199 Å), as well as Le Bail intensity extractions from high‐pressure SP‐XRD data at RT.

Pressure point	P (GPa)	a (Å)	b (Å)	c (Å)	α (°)	β (°)	γ (°)	V (Å3)	s. g.	R wp (%)
P0 a	0.01	8.1427(7)	8.1427(7)	13.145(2)	90.00	90.00	120.00	754.77(1)	P63 mc	3.22
P1	0.75	8.627(2)	7.709(2)	13.073(2)	91.894(4)	84.623(5)	119.846(3)	750.8(3)	P1	1.56
P2	1.78	8.638(3)	7.686(3)	12.950(3)	92.212(5)	84.310(5)	120.028(6)	740.7(4)	P1	1.27
P3	2.70	8.652(6)	7.613(4)	12.812(4)	92.191(8)	84.479(6)	119.659(9)	729.9(6)	P1	1.06
P4	3.60	8.636(6)	7.601(5)	12.763(5)	92.202(8)	84.363(7)	119.805(9)	723.4(7)	P1	1.18
P5	4.70	8.483(6)	7.498(5)	12.681(6)	91.869(9)	84.326(9)	119.25(1)	700.2(8)	P1	0.68
P6	5.80	8.455(7)	7.443(7)	12.634(8)	92.09(1)	84.19(4)	119.29(1)	689.8(9)	P1	0.93
P7	6.70	8.468(9)	7.433(7)	12.530(8)	92.14(1)	84.23(1)	119.58(2)	682.3(8)	P1	0.97
P8	8.40	8.394(9)	7.365(8)	12.49(1)	92.37(2)	84.08(1)	119.42(2)	669.1(1)	P1	0.94
P9	10.0	8.37(1)	7.350(8)	12.452(9)	92.60(2)	84.17(2)	119.38(2)	664.0(1)	P1	1.11
P10	12.3	8.20(1)	7.30(1)	12.31(1)	93.54(2)	83.36(2)	117.95(3)	647.3(6)	P1	1.13
P11	14.8	8.15(3)	7.30(2)	12.27(2)	93.13(4)	83.97(3)	118.03(6)	640.6(2)	P1	0.77
P12	17.7	7.90(2)	7.22(1)	12.14(1)	94.09(2)	82.67(2)	116.78(5)	612.7(4)	P1	0.87
P13	22.4	7.743(7)	7.222(5)	11.951(7)	93.78(1)	82.71(1)	116.52(2)	593.1(7)	P1	1.11
P14	25.2	7.57(2)	7.14(2)	11.82(2)	94.57(5)	82.35(4)	116.07(5)	568.9(8)	P1	0.55
P15	28.9	7.436(7)	7.109(8)	11.662(7)	94.14(2)	82.84(2)	116.21(2)	548.7(8)	P1	0.82
P16	37.4	7.41(2)	6.99(2)	11.52(2)	94.17(5)	82.69(5)	116.19(5)	531.0(7)	P1	0.52
P17	43.7	7.34(1)	6.98(1)	11.48(1)	94.00(3)	82.92(3)	116.17(3)	523.4(4)	P1	0.58
P18	47.0	7.302(8)	7.003(8)	11.44(1)	93.79(2)	82.53(2)	116.38(2)	519.4(3)	P1	0.45
P19	51.5	7.24(1)	7.06(1)	11.43(2)	93.79(2)	82.28(3)	116.51(3)	517.8(5)	P1	0.43
P20R	39.6	7.243(8)	7.09(1)	11.54(1)	93.84(2)	82.25(2)	116.55(2)	525.1(3)	P1	0.68
P21R	33.7	7.42(1)	7.15(1)	11.55(1)	93.53(2)	83.06(3)	116.93(3)	542.6(4)	P1	0.66
P22R	24.0	7.72(1)	7.16(1)	11.90(2)	93.89(3)	82.40(1)	116.26(3)	585.0(3)	P1	0.42
P23R	10.0	8.36(1)	7.455(8)	12.45(1)	92.35(2)	84.39(2)	120.3(3)	667.1(8)	P1	0.83
P24R	0.70	8.61(1)	7.761(9)	13.29(1)	90.83(3)	86.78(2)	120.58(2)	763.0(7)	P1	1.21
P25R b	0.01	8.134(1)	8.134(1)	13.170(3)	90.00	90.00	120.00	754.5(4)	P63 mc	1.32

^a Rietveld refinement was conducted under ambient conditions prior to pressure application.

^b 0.01 GPa pressure point was obtained by carefully releasing the screws from the DAC.

FIGURE 2.

(a) The results of Rietveld refinement applied to an indexed powder pattern coupled with the Le Bail method for selected SP‐XRD patterns with respect to P1 symmetry (λ = 0.6199 Å). The observed and calculated powder XRD profiles are depicted as red dots and solid blue lines, respectively, while the fitted background is shown as lower solid green lines. The difference between the observed and calculated profiles is indicated by lower solid black lines, and the calculated positions of the Bragg peaks for the high‐pressure phase are marked by vertical purple ticks. The evolution of the (b) lattice constants and (c) lattice angles for EMAFC during compression and decompression with transparent solid lines serving as visual guides to highlight trends. Open symbols in panels (b,c) indicate the unit cell metric values observed during decompression.

The triclinic phase exhibits positive compressibility across all crystallographic directions, resulting in denser packing as pressure increases–a behavior typical of most materials. However, what stands out is the notably large value of the bulk modulus, which differs markedly from values commonly observed in both, MOFs, HOIFs and non‐ferroelectric hybrid organic–inorganic perovskites (HOIPs). Specifically, the bulk modulus of EMAFC (K ₀ = 42.0(5) GPa) is significantly higher than that of its isostructural counterpart EAMFB (K ₀ = 2.3(12) GPa) and also exceeds the values typically reported for HOIFs such as [CH₃NH₃]PbI₃ (K ₀ = 10.4 GPa) [63], [CH₃NH₃]SnI₃ (K ₀ = 12.3 GPa) [39], [(NH₂)₂CH]PbI₃ (K ₀ = 11.8 GPa) [64], and (F₂BTa)₂PbI₄ (K ₀ = 12.3 GPa) [65]. Generally, most HOIFs display bulk moduli below 20 GPa and often show pronounced mechanical anisotropy, which stands in stark contrast to the much higher bulk moduli–typically ranging from 100 to 200 GPa–and low isotropic behavior observed in inorganic ferroelectrics [19].

Table S1 compares experimental/theoretical bulk moduli and optical bandgaps of EMAFC with key HOIFs (MAPbI₃, EMAFB), selected non‐ferroelectric HOIPs, MOFs, and inorganic ferroelectrics (BaTiO₃, PbTiO₃, and KnbO₃). No molecular ferroelectric or hybrid organic–inorganic perovskite reported to date has exhibited a bulk modulus as high as that observed here. The presence of weak hydrogen bonds and halogen interactions in hybrid materials allows the lattice to deform without breaking chemical bonds, whereas inorganic ferroelectrics are composed of rigid ionic or covalent networks that limit compressibility. In organic–inorganic systems, mechanical properties are largely determined by internal structure, as well as hydrogen and metal‐halide bonds. When the metal‐halide bond is weaker, the material tends to be more elastically soft. These distinctive properties make HOIFs particularly valuable for flexible electronics and customizable device applications, although inorganic ferroelectrics remain the preferred choice for environments subjected to high mechanical stress. In our specific comparison between EMAFC and EMAFB, the effect of bromide and chloride ions on compressibility is clearly apparent and arises from their unique bonding characteristics and structural interactions, with electronegativity and ionic radius playing crucial roles [66, 67]. Bromide ions have a larger ionic radius and lower electronegativity than chloride ions. The comparatively lower electronegativity and greater ionic radius in ABX₃ compounds contribute to the formation of longer, more flexible B─Br bonds, which lead to a moderate bulk modulus, as observed for EMAFB (Figure 3a). In contrast, the higher electronegativity of chloride ions strengthens the B─Cl bonds, resulting in increased bond stiffness and a correspondingly higher bulk modulus for EMAFC [66, 68]. HOIFs have been extensively studied for their intriguing mechanical properties. Nevertheless, prior research has primarily focused on Pb‐or Sn‐based HOIFs, which often exhibit instability or toxicity. Sun et al. examined the elastic and plastic behavior of MAPbX₃ single crystals (MA = methylammonium; X = Cl, Br, I) [63]. Their experimental results showed that the Young's modulus (E) decreases in the order E _Cl > E _Br > E _I, which aligns well with the theoretical bulk modulus trends reported by Giorgi et al. [69]. This pattern reflects the decreasing electronegativity of the halide ions from Cl to I, leading to weaker Pb─X bonds and consequently lower E values. Interestingly, hardness (H) exhibited the opposite trend, increasing in the order H _Cl <H _Br < H _I, likely due to the reduction in structural symmetry (Cl > Br > I), which facilitates the activation of more dislocation systems under indentation. This interpretation is supported experimentally by nanoindentation studies conducted by Rakita et al. on APbX₃ (A = Cs, MA; X = I, Br) and theoretically by Feng, who investigated MABX₃ (B = Sn, Pb; X = Cl, Br, I) using first‐principles calculations [70, 71]. More recently, Sun and co‐workers investigated the mechanical properties of FAPbX₃ analogues (FA = formamidinium; X = Br, I), both of which crystallize in the cubic phase [63]. The effect of halide electronegativity was again evident: elongation of the Pb─X bond in FAPbBr₃ compared to FAPbI₃ led to reduced elastic moduli and hardness. Notably, replacing the smaller MA cation with the larger FA cation unexpectedly weakened the lattice rigidity. For molecular ferroelectrics, a higher bulk modulus usually indicates greater hardness, reflecting stronger bonding and denser structures. However, as molecular ferroelectrics can exhibit significant anisotropy and weaker intermolecular forces, the correlation is not always straightforward or as strong as in inorganic crystals. Detailed experimental or computational analysis is typically required to quantify this relationship for specific systems. The structural effects that depend on the choice of halide enable the design and control of compressibility properties for specialized applications such as pressure sensing and adaptive energy storage, with chloride ions increasing lattice stiffness and bromide ions introducing distinctive, anisotropic mechanical behavior [55, 66, 67].

FIGURE 3.

(a) Isothermal (293 K) pressure dependence of the unit cell for the triclinic phase of EMAFC and the monoclinic phase of EMAFB as pressure increases (left), alongside a pressure expansivity indicatrix diagram for the monoclinic phase of EMAFC, derived from in situ variable‐pressure SP‐XRD data (right). The transparent blue and green lines fitted to the data points correspond to third‐order semi‐empirical Birch‐Murnaghan equations of state for the high‐pressure phases of EMAFC and EMAFB, respectively. (b) Illustration of the application of high pressures to EMAFC using DAC (left) and overlays of [FeCl₄]⁻ arrangement in EMAFC at 0 GPa (prior to pressure application) and in relaxed sample at 0.01 GPa.

One method for quantifying the difference between the structure at ambient conditions (before pressure application) and the structure relaxed at 0.01 GPa is to compute RMSD of atoms between the two models. The EMAFC structural models acquired before (i.e. P0 data) and after decompression (i.e. P25 data) show an RMSD of 0.560 Å. Although this value indicates general structural similarity, it is slightly higher than the typical benchmark RMSD values (∼0.1–0.3 Å) reported for strongly comparable small‐molecule crystal structures. The material's hybrid organic–inorganic composition and potential molecular flexibility result in an RMSD of 0.560 Å, indicating strong structural similarities but notable variations. Therefore, the structures can be described as structurally related rather than virtually identical. This approach recognizes the complexity and intrinsic conformational flexibility of hybrid crystals, which may contribute to the observed differences. For example, a recent machine‐learning‐based crystal structure prediction study for 2D hybrid organic–inorganic perovskites found RMSD values of approximately 0.2–0.4 Å between predicted and experimentally confirmed structures, indicating good, though not perfect, agreement due to structural flexibility in these systems [72]. Superimposing the structures before and after the pressure change allows direct visualization of phase purity, confirms reversibility and supports the refinement models (Figure 3b, left panel), bridging the gap between the diffraction patterns and the actual changes at the atomic level.

2.2. Pressure‐Driven Raman Measurements

To back up our interpretation, we used solid‐state Raman band assignments together with X‐ray diffraction, which helps distinguish between crystal forms by measuring subtle differences in the energies of external and internal crystal lattice vibrational modes. These differences result from changes in molecular interactions. The low‐frequency external vibrational modes in the Raman spectrum are highly sensitive to the crystal form and molecular arrangements within the unit cell. Thus, the Raman signature complements the XRD features by providing additional details of the structural changes observed.

As pressure increases, a blueshift is observed in all Raman modes, resulting from the continuous contraction of the hexagonal EMAFB structure (Figure 4a–d). Analysis of the low‐frequency Raman spectra reveals that nearly all vibrational modes exhibit positive pressure coefficients throughout the 0–43.0 GPa range, suggesting that compression predominantly induces tilting of the [FeCl₄]⁻ anions. This interpretation is consistent with structural insights obtained from SP‐XRD measurements. The vibrational modes of the [FeCl₄]⁻ anion show up in the 70–400 cm⁻¹ wavenumber range at ambient pressure. The bands above 400 cm⁻¹ are caused by the vibrational modes of the cationic part ([N(C₂H₅)₃CH₃]⁺) (Figure 4a–d) [73, 74]. The prominent Raman bands at 120.5 and 139.7 cm⁻¹ are signature features, identified as the symmetric and antisymmetric bending modes–δ _s[FeCl₄]⁻ and δ _as[FeCl₄]⁻–respectively. The exceptionally strong peak at 334.0 cm^{− 1} is the fingerprint of the asymmetric stretching vibration, ν _as[FeCl₄]⁻. Particularly intriguing are the low‐energy lattice vibrations, below 150 cm⁻¹, which serve as ultra‐sensitive probes for shifts in crystal structure. Even subtle changes in this region can rapidly trigger local structural transformations, paving the way for dramatic phase transitions. The electronic spectrum of [FeCl₄]⁻ exhibits weak features below 50 cm⁻¹, identified as spin‐forbidden and electrodipole‐forbidden bands. The emergence of new Raman modes within the 0.6–3.9 GPa range (Figure 4a,c,d) provides evidence for the proposed hexagonal‐to‐triclinic phase transition in EMAFC, reflecting a reduction in symmetry. Moreover, during the transition, the broad Fe─Cl bending modes evolve, while Fe─Cl stretching distorts, indicating a pressure‐induced distortion of the [FeCl₄]⁻ polyhedra. Even at modest pressures, intriguing changes appear in the Raman spectra: not only do the intensities of the active modes shift, but new, faint maxima gradually appear. For example, a subtle yet distinct broad peak emerges at 51.5 cm^{− 1} under 0.6 GPa. Below 3 GPa, the fundamental modes δ _s[FeCl₄]⁻ and δ _as[FeCl₄]⁻ display pronounced oscillations, giving rise to new maxima at 1.5 GPa. These dynamic shifts can be tracked throughout both compression and decompression, revealing a complex landscape of structural response to pressure. Raman changes at ∼15 GPa reflect local [FeCl₄]⁻ tilting, which is undetectable as a long‐range symmetry shift in PXRD, where only peak broadening is observed. As EMAFC is compressed and decompressed under a 633 nm laser, up to 31.0 GPa, each of the three fundamental Raman modes elegantly shifts gradually and traceably (Figure 4d). This smooth evolution stands in sharp contrast to the behavior observed in EMAFB, where–once the pressure exceeds 6.0 GPa–an intense laser‐induced fluorescence rapidly develops, eventually overwhelming all Raman signals beyond 8.2 GPa. The observed Raman modes serve as precise pressure sensors, with their shifts responding acutely to decreasing intermolecular distances. Progressive Raman shifts continue up to 31.8 GPa, followed beyond 37.0 GPa by abrupt peak broadening and intensity collapse (Figure 4c), indicating pressure‐induced local distortions in the organic and inorganic components. The disappearance of the modes above 40.0 GPa confirms significant unit cell distortions, likely associated with amorphization of the sample [55]. Crucially, decompression from 43.0 GPa triggers spectral re‐emergence at 33.2 GPa, culminating in near‐total recovery at 0.2 GPa. This exceptional reversibility correlates precisely with SP‐XRD results.

FIGURE 4.

(a,b) Selected pressure‐dependent Raman spectra of EMAFC in the low‐frequency region, collected using a 633 nm excitation laser during both compression and decompression. (c) The evolution of Raman spectra in a waterfall‐style plot demonstrates the reversible structural transition to the high‐pressure phase. (d) The pressure dependence of Raman shifts–after the phase transition at 0.6 GPa (red dotted line), low‐intensity maxima emerge and fluctuate over the pressure range up to 5 GPa, highlighting changes in the vibrational modes of the [FeCl₄]⁻ anion, as indicated by blue and green dotted lines. White circles mark the Raman shift values observed during decompression. (e) The pressure‐dependent UV–vis absorption spectra of EMAFC show a redshift. (f) Optical micrographs capturing the transformation of EMAFC as it undergoes compression to 24.0 GPa, and then returns to 0.01 GPa within the DAC chamber. The images reveal the dynamic color shifts and reversible piezochromic behavior, showcasing its remarkable adaptability under extreme pressure. (g) The evaluation of the bandgap energy of EMAFC as a function of applied pressure is presented. The inset in (g) displays representative UV–vis absorption spectra measured under ambient conditions, plotted using the Kubelka‐Munk function.

2.3. Pressure‐Induced UV–Vis Absorption Spectroscopy Data Collection

To probe the electronic drivers of structural evolution, we tracked pressure‐responsive UV–vis absorption. EMAFC exhibits a distinct blueshifted absorption profile at 420 nm with a reduced slope intensity compared to isostructural EMAFB (Figure 4e). The absorption edge–characterized by a broad 450 nm peak superimposed with minor maxima at 430 and 440 nm–progressively broadens under compression. While stable below 1.9 GPa, the edge undergoes gradual redshifting up to 13.7 GPa, where a pronounced absorption increase indicates electronic reorganization. Further compression drives continuous edge broadening toward longer wavelengths, ultimately spanning the visible to near‐infrared range. Strikingly, EMAFC maintains traceable edge evolution at 24.0 GPa, contrasting sharply with the signal suppression observed for EMAFB at 16.0 GPa. This exceptional mechanochromism is visually evident (Figure 4f): the symmetry collapse from hexagonal to triclinic induces a dynamic color trajectory–transparent yellow (ambient) → yellow‐orange (13.7 GPa) → brownish red (15.0 GPa) → deep umber (>20 GPa) – with full reversion upon decompression. Critically, low‐frequency Raman modes correlate with these transitions, and, unlike the brown‐block crystals of EMAFB, the EMAFC sample retains chromatic integrity throughout. Such a difference in trends between EMAFB and EMAFC arises from the charge transfer process and the different electronegativity of the halogens. Specifically, the [FeCl₄]⁻ in EMAFC absorbs in the UV region (∼420 nm) via ligand‐centred transitions and symmetry‐allowed 3d‐4p hybridization, while [FeBr₄]⁻ in EMAFB absorbs in the visible range due to bromine‐driven charge transfer and spin‐orbit effects, with absorption extending to higher wavelengths (∼420 nm). This divergence underscores the role of halogens in tuning electronic structures: chlorine favors higher‐energy transitions, whereas bromine promotes redshifted, charge‐transfer‐dominated spectra. Stronger spin‐orbit coupling in bromine broadens spectral features and enhances charge delocalization. Absorption in the visible region indicates greater ligand‐to‐metal charge transfer compared to [FeCl₄]⁻ [75, 76]. These results confirm the superior binding affinity of chlorine for iron relative to bromine, which likely stabilizes the high‐pressure phase through stronger Fe─Cl bonding. Furthermore, the enhanced Fe─Cl interaction may dominate reversible processes–including proton transfer, symmetry‐lowering transitions, and electronic reconfigurations–under compression. The optical bandgap (E _g) of EMAFC was determined by extrapolating the linear region of the Kubelka‐Munk function [F(R)] plots, as shown in Figure 4g. Under ambient conditions, the indirect optical bandgap of EMAFC was found to be 3.01 eV, indicating a wide bandgap semiconductor [12]. This value remains nearly constant up to 5.6 GPa, beyond which a pronounced decrease occurs. Above 3.2 GPa, the bandgap decreases gradually up to 12.1 GPa, with a closure rate of −38 meV/GPa. Between 12.1 and 16.1 GPa, a marked reduction is observed, corresponding to an accelerated closure rate of −127 meV/GPa. Changes at ∼15 GPa, similar to Raman observations, suggest pressure‐induced orbital hybridization (e.g., Fe─Cl overlap) and octahedral compression, indicating an electronic transition independent of symmetry change. This behavior mirrors HOIPs such as MAPbBr₃ where Pb─Br bond shortening reduces Eg without a phase transition [77]. Beyond 16.1 GPa, the rate decreases, continuing up to 24.0 GPa, where the bandgap closes at an approximate rate of −44 meV/GPa. This reversal in slope indicates a pressure‐induced structural reorganization, mirroring the reduced‐dimensionality perovskites [78]–all modes show pronounced blueshifts under compression and serve as a signature of the response to molecular confinement. The bandgap narrowing is attributed to two competing structural distortions: tilting of the Fe─Cl bond and reduction of the Cl─Fe─Br angle. These distortions enhance the overlap of the metal‐halide orbitals and thus alter the optoelectronic properties, although the bandgap remains above the Shockley‐Queisser limit of 1.22 eV, unlike in the case of EMAFB. For comparison, EMAFC has a higher bandgap than EMAFB, as the incorporation of bromine in EMAFB reduces the bandgap to 2.03 eV at 0.01 GPa. Generally, the E _g values observed in most molecular ferroelectrics fall within the range of 2.3–2.9 eV, which are substantially lower than those reported for conventional inorganic ferroelectrics such as Nb₂O₅, ZnO, and GaN (>3.1 eV) (see Table S1). Despite exhibiting strong absorption in the UV region, no emission was detected from EMAFC under high pressure. The absence of emission in EMAFC can be attributed to several factors. First, following UV excitation, the relaxation of photoexcited carriers predominantly proceeds via nonradiative pathways, such as phonon‐assisted or defect‐mediated recombination, rather than radiative emission. Second, pronounced electron–phonon coupling and structural disorder, commonly observed in molecular ferroelectrics containing transition metal complexes like [FeCl₄]⁻, facilitate efficient nonradiative decay channels that further suppress emission. Third, the presence of quenching centres–arising from intrinsic defects, impurities, or the [FeCl₄]⁻ framework itself–can trap excited carriers and inhibit radiative recombination. Finally, the d‐d transitions in Fe(III) complexes are typically spin‐forbidden, resulting in intrinsically low luminescence quantum yields. Collectively, these factors suggest that, although EMAFC exhibits significant UV absorption owing to its wide bandgap, photoluminescence is effectively quenched due to dominant non‐radiative relaxation mechanisms and the electronic structure of the Fe(III) centre. The Commission Internationale de l'Éclairage (CIE)colorimetric analysis quantitatively correlates the observed color evolution with applied pressure (Table 2). This reveals a clear relationship between pressure‐induced bandgap narrowing, and color progression (yellow → yellow–orange → brownish‐red → deep umber).

TABLE 2.

Correlation of applied pressure, CIE Lab coordinates, color descriptions, bandgap values, and optical micrographs for EMAFC.

P (GPa)	CIE L a	CIE a b	CIE b c	Color description	E g (eV)	Optical image
0.001	96.3	−0.9	3.7	transparent slightly yellowish	2.99
5.6	91.6	4.5	11.7	light cream yellowish	2.93
10.2	77.6	9.3	66.8	very light saturated yellow	2.79
15.0	45.6	36.8	54.8	saturated yellow–orange	2.34
24.0	3.3	12.8	5.8	dark desaturated red–orange	1.78

^a L (Lightness): ranges from 0 (dark/black) to 100 (light/white).

^b a (red–green axis): positive values indicate red; negative values indicate green.

^c b (yellow–blue axis): positive values indicate yellow; negative values indicate blue.

2.4. In Situ SHG Features Under High Pressure

Figure 5a–d shows the phase transition‐driven SHG “on‐off” effect. To confirm the authenticity of the SHG signal, the intensity of the SHG signal at the same position of EMAFC under different laser powers was measured at atmospheric pressure. The SHG intensity has a linear relationship with the fourth power of the laser current, indicating that the detected signal is true and originates from the SHG effect of EMAFC. The excellent fit, indicated by an R ² value of 0.99, strongly supports the expected quadratic relationship for SHG. During compression, EMAFC remains in the SHG “on” state up to 9.5 GPa, with no apparent intensity attenuation. Above 9.5 GPa, the SHG strength of EMAFC decreases significantly, consistent with the pressure‐induced phase transition. Distinct changes in bandgap, Raman, and SHG intensity near 15 GPa suggest distortion of the [FeCl₄]⁻ framework, or even pressure‐induced aggregation or organization of organic and inorganic components into distinct microscopic or nanoscale regions, known as domains [19, 79]. The organic cation primarily acts as a counterion for charge balance and physical stabilization, yet its templating influence on the orientation of the inorganic framework significantly affects the electronic structure. In EMAFC, the asymmetric [N(C₂H₅)₃CH₃]⁺ interacts with [FeCl₄]⁻, tilting under pressure through electrostatic interactions and partially through hydrogen bonding, thereby maintaining non‐centrosymmetry (P1 phase) and inducing SHG intensity changes at 15 GPa due to increased distortion. The intensified organic–inorganic interactions result in greater structural distortion, which is directly manifested in the optical properties [80]. Above 20 GPa, EMAFC switches to the SHG “off” state, indicating no SHG activity in the high‐pressure phase of EMAFC. The pressure‐dependent SHG response shows significant hysteresis during decompression. Specifically, the SHG “off‐on” transition exhibited pronounced pressure hysteresis, with the “on” state restored only when the pressure was reduced to 0.6 GPa. This behavior is attributable to residual stress effects or partial amorphization induced under high‐pressure conditions, suggesting that structural recovery upon decompression is incomplete until near ambient pressure is reached. To date, SHG switching has predominantly been reported in materials undergoing temperature‐induced phase transitions, including molecular ferroelectrics [9, 10, 15], organic–inorganic hybrid perovskites [11, 12, 13], and inorganic salts [16, 17, 18]. When comparing the SHG response of two non‐centrosymmetric space groups, as in our case, both can support SHG; however, their magnitude, orientation, and efficiency can vary considerably due to differences in their symmetry elements and crystal structures. Nonetheless, it remains crucial to continue exploring SHG‐switchable materials that respond to stimuli other than temperature, thereby expanding their applicability across a wider range of conditions [5, 6, 81, 82, 83].

FIGURE 5.

(a) Laser current dependence of the SHG intensity of powder EMAFC at 0 GPa. (b) Pressure dependence of the SHG intensity of EMAFC during compression and decompression. (c) SHG polar plots of EMAFC under different pressures during compression. (d) SHG polar plots of EMAFC under different pressures during decompression.

2.5. Dielectric Studies Under High Pressure

Figure 6a shows the temperature dependence of the real part of the dielectric permittivity, ε′(T), obtained for the EMAFC pellet at f = 500 kHz. An anomaly in ε′(T) is observed above 340 K during heating, corresponding to the order‐disorder phase transition reported in our earlier paper [11]. In the case of a classical first‐order phase transition, the T _c is typically estimated from the position of the peak maximum in the ε′(T). However, determining of the phase transition temperature becomes ambiguous in the case of a diffuse phase transition or when the maximum in the ε′(T) dependence is poorly defined. Therefore, we proposed an additional method to determine the temperature associated with the phase transition. We determine the phase transition temperature as the intersection of two straight lines corresponding to the start and end of the visible bump anomaly. The obtained T ₀ = 361.5 ± 1 K at ambient pressure shows good agreement with the structural phase transition detected by the classical method and XRD results. Hence, we use the same procedure to determine T ₀ under high‐pressure conditions (Figure 6b,c). The phase transition anomaly becomes less visible at increasingly higher pressures. To determine the extent to which the hydrostatic pressure affects the phase transition temperature, a phase diagram was constructed (Figure 6d). The phase transition temperature, T ₀, decreases linearly with pressure, exhibiting a slope of −131 K/GPa. Literature reports examples of materials with even greater sensitivity to pressure; for instance, [CH₃NH₂NH₂]₂PbI₄ shows a shift toward higher transition temperatures under increased pressure, with a notably high positive slope of 196 K/GPa [84]. It is important to note that a decrease in phase transition temperature under high pressure is characteristic of ferroelectric perovskites [85]. In this context, PbTiO₃ has been reported to exhibit a decrease in T ₀ with a slope of −84 K/GPa [86].

FIGURE 6.

Temperature dependence of the real part of dielectric permittivity ε′(T) at different pressures (a) ambient, (b) 75 mPa, (c) 150 mPa), and (d) the T ₀(p) phase diagram (transition temperature vs. pressure) for the EMAFC sample.

To gain deeper insight into the complex behavior of EMAFC under stress, high‐pressure isothermal measurements were conducted at RT. As shown in Figure 7, the real part of the permittivity remains nearly stable up to 1 GPa, after which it begins to decrease. This behavior corresponds with the XRD results, where an enantiomorphic, lower symmetry space group P1 was observed above 0.75 GPa. However, for lower frequencies, we can observe the weak change in ε' around 0.75 GPa marked by the arrow and the greatest changes above 1 GPa. This effect could be assigned to the crossover between the two ferroelectric phases. When the pressure increased, the phase transition became more visible, and phase P6₃ mc finally disappeared. This transition is typically attributed to the suppression of ferroelectric order and a decrease in polarizability as the crystal structure becomes less favorable for spontaneous polarization under compression.

FIGURE 7.

Pressure dependence of the real part of the dielectric permittivity at 300 K for (a) 100 kHz and (b) 500 kHz. Dashed lines guide the trends.

High pressure often restricts the motion of molecular dipoles and the flexibility of hydrogen bonds, leading to diminished dielectric response [87]. The phase transition may also convert the material into a paraelectric or antiferroelectric phase, both of which have lower permittivity than the ferroelectric phase [88]. Lower permittivity in a material increases signal propagation speed and decreases capacitance in electronic devices. Materials with reduced permittivity enable electrical signals to move more quickly and allow for smaller, lower‐capacitance capacitors. This property is especially important in microelectronics, where high speed and miniaturization are essential. The high‐pressure dielectric response of EMAFC reveals its tunability, encompassing temperature, frequency, and anisotropy effects that are crucial for advanced applications in sensors, capacitors, and multifunctional devices.

3. Conclusion

Our high‐pressure investigation of the lead‐free HOIF, EMAFC, establishes a critical structure‐compressibility signature that extends the design space of HOIFs. The unprecedented compressibility, with a bulk modulus of K ₀ = 42.0(5) GPa, coupled with reversible pressure‐induced mechanochromism and tunable SHG switching, underscores the role of halide chemistry and lattice dynamics in governing multifunctional responses. By demonstrating a reversible hexagonal‐to‐triclinic phase transition and bandgap modulation up to approximately 50 GPa, this work highlights how strategic chemical selection enables phase engineering for optoelectronic and mechanical resilience. The choice of inorganic anion (e.g., [FeCl₄]⁻ vs. [FeBr₄]⁻) significantly affects the bulk modulus, as demonstrated by our discussion and direct comparison with an isostructural equivalent. The inorganic component–the metal centre and, more importantly, the specific halide element (e.g., F, Cl, Br, I)–strongly influences this property. Selecting the appropriate halide is an effective way to modify the compressibility of a hybrid inorganic–organic molecular system. This makes EMAFC the least compressible molecular ferroelectric discovered to date, opening new avenues for studying and manipulating the phases and properties of this class of materials. These findings establish performance benchmarks for compressible HOIFs and provide design guidelines to advance the development of environmentally sustainable materials capable of operating reliably in future devices exposed to both ambient and extreme thermal, optical, and mechanical conditions.

4. Experimental

4.1. Synthetic Procedures

Dark yellow polycrystalline samples and single crystals of EMAFC were synthesized at room temperature (RT) by combining aqueous solutions of FeCl₃ and (C₂H₅)₃N(Cl)CH₃, following the procedure reported in our earlier work [11].

4.2. Pressure‐Induced Synchrotron Powder X‐Ray Diffraction (SP‐XRD) Data Collection

The SP‐XRD measurements were conducted at RT up to 51.5 GPa in axial geometry using a high‐resolution powder diffractometer. Experiments were performed at the SPring‐8 facility (BL12B2, Hyogo, Japan) with an incident beam of 50 µm diameter and a wavelength of λ = 0.6199 Å. Polycrystalline EMAFC sample, together with a ruby sphere, were loaded into a Mao‐type symmetric DAC equipped with 400 µm culets. The sample chamber was fully filled with the sample, and Daphne oil was employed as the pressure‐transmitting medium (PTM). Pressure calibration at each measurement point was achieved via monitoring the shift in ruby fluorescence emission. To ensure a high signal‐to‐noise ratio and minimize radiation damage, five exposures of 60 s each were collected and summed for each pressure point. The resulting 2D diffraction images were integrated using IPAnalyzer, and subsequent data analysis was carried out employing the EXPO2014 suite [89].

4.3. Pressure‐Driven Raman Measurements

Raman spectroscopy measurements under both increasing and decreasing pressure were conducted using a MonoVista CRS+ system (Spectroscopy and Imaging) equipped with a 633 nm excitation laser operating at 5 mW. A grating with 1200 grooves/mm was employed throughout the experiments, and the laser spot size was maintained at 20–30 µm. A single‐crystal of EMAFC, along with a ruby sphere, was loaded into the sample chamber of a Mao‐type symmetric DAC fitted with 300 µm culet low‐fluorescence anvils. Silicone oil served as the PTM. The pressure at each measurement point was determined by calibrating the shift in the ruby fluorescence lines.

4.4. Pressure‐Induced UV–Vis Absorption Spectroscopy Measurements

Optical and fluorescence imaging, as well as UV–vis absorption spectra, were acquired using a custom‐built spectroscopy system (Gora‐UVN‐FL, Ideaoptics, Shanghai) equipped with an integrated micro‐angle spectroscopy module. UV–vis absorption spectra were collected in the 300–1100 nm range using a Xe light source (SLS201L, Thorlabs, USA). The excitation beam was focused onto the sample surface with a 20× objective (0.42 NA, Mitutoyo, Japan), yielding a spot size of approximately 40 µm in diameter. Laser power was modulated using a neutral‐density filter. Single crystals of EMAFC (60 × 50 × 10 µm³) were mounted on a 400 µm culet‐sized, low‐fluorescence flat diamond surface. Background signals from the diamond were recorded and subtracted from sample measurements. Absorption spectra were acquired in transmission mode. The optical bandgap (E _g) of EMAFC was determined by extrapolating the linear portion of the (α ^1/2) vs. hν plot, where α is the absorption coefficient, h is Planck's constant, and ν is the photon frequency. Fluorescence images were captured under UV irradiation with an exposure time of 200 µs. The CIE colorimetric analysis was performed alongside UV–vis spectroscopy across the full pressure range. CIE color coordinates and CIELAB L*, a*, and b* values were determined using Color Calculator software (version 7.77, Corporate Innovation Advanced Technologies Application Solutions Design, OSRAM SYLVANIA, Inc.). To derive the perceived color of transmitted light under illumination from the sample's absorption spectrum, we first converted it to transmittance using the equation:

$$T\left(\lambda \right)={10}^{-A}\left(\lambda \right)$$

where T(λ) is the transmittance at wavelength λ.

4.5. Second‐Harmonic Generation (SHG) Switching Measurements at High Pressure

Fixed‐wavelength, high‐pressure, polarization‐dependent SHG “on‐off” switching measurements were performed on EMAFC powder using a custom‐built optical setup (Ideaoptics, China). The excitation source was a 1064 nm, 20 mHz fiber laser (NPI LASER Co., Ltd.), with polarization precisely controlled by rotating a polarizer positioned in the beam path. A 20× objective lens was used to focus the laser to an 40 µm spot on the crystal surface. The generated SHG signal was detected with a high‐sensitivity photomultiplier tube (PMT1000, Thorlabs, Inc.), enabling accurate data acquisition.

4.6. Dielectric Measurements at High Pressure

Two distinct apparatus setups were employed for the high‐pressure measurements: one operating under isobaric conditions and the other under isothermal conditions. High‐pressure isobaric measurements were performed using a system manufactured by UNIPRESS, consisting of an MV1‐30 high‐pressure chamber and an MP5 micropump. During measurements, the sample was enclosed in a Teflon capsule sealed with a steel cork and immersed in Julabo Thermal HL90 silicone oil, which served as the pressure‐transmitting medium, secured by a steel clamp. High‐pressure isothermal measurements were carried out in a stainless‐steel high‐pressure chamber. The pellet under study was placed in a Teflon capsule filled with Julabo Thermal HL90 silicone oil. Hydrostatic pressure inside the chamber was generated using a hydraulic press. Temperature control for both measurement setups was maintained via a Julabo Presto thermostatic bath. Isobaric measurements were conducted over a temperature range of 310–383 K, while the isothermal measurements at RT covered a pressure range of 50–1300 mPa in increments of 100 mPa. Dielectric properties were measured using a Novocontrol Alpha impedance analyzer following a stabilization period of 30 min. Silver paste was applied to the electrodes, and the sample was dried in an oven at 350 K for 8 h prior to measurement.

Conflicts of Interest

The authors declare no conflict of interest.

Supporting information

Supporting File: smll72728‐sup‐0001‐SuppMat.docx.

Data Availability Statement

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