A new insight into the unique magneto‐optical effect of layered perovskite (C6H5C2H3FNH3)2MnCl4

Magnet‐optical materials embracing coupled magnetic and photoluminescent properties in single phase are promising in microelectronics and optoelectronic devices. However, the current research mainly focuses on traditional inorganic materials, and there are few reports on molecule materials. Recently, we synthesized an organic–inorganic hybrid complex (C6H5C2H3FNH3)2MnCl4 (1) with perovskite structure. Physical measurements show that 1 not only behaves as an antiferromagnet with spin canting but also exhibits unusual fluorescent properties. Importantly, under the magnetic field at different temperatures, the luminous intensity of 1 changed, and a red‐shift occurred with obviously optical hysteresis. These phenomena directly prove the existence of magneto‐optical coupling in 1. More interestingly, the optical hysteresis can be observed in both low and high field, which is unprecedented in other molecular materials. Even in traditional inorganic materials, it can only be observed in strong field. This special function provides the possibility for the application of low energy consumption optoelectronic devices.


INTRODUCTION
Proper control of photoluminescence (PL) properties, such as emission peaks, intensity and lifetime, are highly desirable for widespread applications like laser mediums, optical waveguides, display lighting and biomedicine. [1][2][3][4] Applying magnetic field serves as a facile and feasible method to regulate the physical properties of materials. Magneto-optical interaction, including Zeeman effect, Faraday effect, Kerr magneto-optical effect and Cotton-Murton effect, has also been widely utilized to regulate the PL of materials. These effects all originate from the magnetization of the materials and reflect the relationship between light and magnetism of the materials. In molecular complex family, they have only been observed in the organic radicals, in which the spins originate from s or p orbitals. [5][6][7][8] Moreover, the magnetic field applied in these complexes can only change their fluorescence intensity. [8][9][10] The change in other optical phenomena, such as red-or blue-shift of the peak and optical hysteresis, has not been found in the reported complexes. Nevertheless, these abnormal optical phenomena can be realized in 3d or 4f inorganic compounds. [11][12][13][14] Qiu and the co-workers found that Zeeman splitting caused by magnetic field can change the intensity of rare earth fluorescence and shows hysteresis phenomenon under high field. [15] However, this kind of example is quite exceptional in complexes. [16,17] The energy level of the paramagnetic metal ion will split due to the Zeeman splitting when a magnetic field is applied, which will lead to the splitting of the emission spectrum and the decrease of the intensity. [13,18] Meanwhile, the local magnetic order of the spin centre will be induced in high field. If the magnetic field is removed, the local magnetization hysteresis could cause the optical hysteresis behaviour under high field. [15] In this line, magneto-optical effect may be more easily realized in a three-dimensional (3D)ordered magnetic system benefiting from its magnetic order or remnant magnetization.
It is known that organic-inorganic hybrid layered perovskite complexes, containing paramagnetic metal, can form long-range magnetic order more easily than discrete molecules. [19] In addition, compared with 3D pure inorganic compounds, it is easier to regulate their structures during the synthesis. For example, A 2 BX 4 (A = organic cation, B = Mn, Cu, Fe, Pb and Sn, and X = Cl, Br and I) complexes can show a variety of interesting properties, such as photoelectric effect, [20] PL, [21] magnetism [22] and ferroelectricity. [23,24] Particularly for the dual nature of photosensitivity and magnetism, these A 2 BX 4 complexes would be good candidates for the investigation of magneto-optical properties. [25] In the crystal structure of A 2 BX 4 complexes, there is a dense layer with transition metal ions bridged by X − ions. When the metal ions contain unpaired electrons, it is easy to generate magnetic ordering in this layer. If the magnetic interaction between the layers is strong enough, a 3D-ordered magnet could be formed. At the same time, this kind of complexes easily show the fluorescence properties due to d-d transition or the mixing of ground state and excited state spectra terms of transition metal ion.
In this work, we synthesized a layer-like complex, namely (C 6 H 5 C 2 H 3 FNH 3 ) 2 MnCl 4 (1). The magnetic measurements show that 1 behaves as an antiferromagnet with spin canting. Its magnetic ordering provides us the chance to investigate the magneto-optical properties. Under different temperatures and applied magnetic field, complex 1 exhibits the interesting unusual fluorescence properties. Some of these special phenomena are observed experimentally for the first time in molecular complexes. Herein, we report the synthesis, structure, and magneto-optical properties with the possible structure-activity relationship.

RESULTS AND DISCUSSION
Magneto-optical effect refers to various optical phenomena caused by the interaction between magnetized compounds and light. Zeeman effect is one of important magneto-optical effects as well as an important magnetic phenomenon. If Zeeman effect is reasonably applied to the design and synthesis of magneto-optical effect compounds, the possibility of obtaining this kind of material will be greatly improved.
In other words, only when matter has both magnetic and optical properties, can we more easily study this effect. It has been reported that Zeeman effect can change the light intensity, and the use of remnant magnetization can produce optical hysteresis. [13,15,18] In magnetic compounds, only 3D magnets (especially hard magnets) have remanence. Therefore, our goal is to find compounds with optical properties in 3D magnets. However, so far, the coexistence of 3D magnetic ordering and luminescence is very rare. On the basis of literature research, we found that perovskite compounds are usually sensitive to light and often used as raw materials for solar cells. [26][27][28] Meanwhile, manganese chloride is a well-known luminescent material, [29,30] but pure manganese chloride is not a 3D-ordered ferromagnet. This means that we cannot obtain remanence in manganese chloride. However, manganese chloride complexes with 2D-layered structure can exhibit weak ferromagnetic behaviour due to spin canting, [31][32][33] which provides an important clue for us to obtain possible magneto-optic materials. Based on the above considerations, we designed and synthesized an organicinorganic hybrid perovskite ABX 4 complex. It not only contains inorganic layer, which is conducive to the formation of magnetic order but also has the unique luminescence properties of manganese chloride due to the existence of d-d transition. All these conditions are the basis for us to obtain magneto-optical effect. Facts have proved that our thinking is correct. We have successfully obtained magneto-optical materials with optical properties regulated by magnetic field. The crystal structure of complex 1 was determined at room temperature. At 293 K, complex 1 crystallizes in the monoclinic space group C2/m, with a = 39.881(2) Å, b = 5.1182(3) Å, and c = 5.1204(3) Å. As shown in Figure 1, the molecular structure of 1 is composed of 2D {[MnCl 4 ] 2− } n inorganic layers and m-fluorophenylethylamine cations locating between the layers. In the inorganic layer, each manganese atom is surrounded by six chloride ions, four of which behave as the bridges between the adjacent Mn ions. The corner-sharing octahedra forming 2D network parallel to the bc plane, which stacks along the a-axis ( Figure S3). Notably, the m-fluorophenylethylamine cations are equally disordered over two different crystallographic sites associated with the mirror plane. And, the plane m bisects the mfluorophenylethylamine cation by C3, C7 and N atoms. The swing of m-fluorophenylethylamine cation causes no obvious  (Table S2).
The 2D network lies in the bc plane, and the distance between the nearest layers is 20 Å. Thus, the magnetic anisotropy was determined by measuring the magnetic properties of the single crystal along with two orientations under an applied magnetic field parallel (H || , a axis) or perpendicular (H ┴ ) to the Mn-Cl (bc) layer.
As shown in Figure 2A, the room-temperature χ M T values of complex 1 are 4.10 and 4.00 cm 3 K mol −1 for H ┴ and H || directions, respectively, which are slightly lower than the theoretical value of 4.375 cm 3 K mol −1 for single Mn(II) ion (S = 5/2 and g = 2), suggesting strong antiferromagnetic (AFM) coupling between spins. Upon lowering the temperature from 300 to 100 K, the χ M T values decrease smoothly. Fitting the experimental magnetic data to the Curie-Weiss law with χ M = C/(T − θ) in this temperature range gave the following parameters: C = 4.67 cm 3 K mol −1 and θ = −272.67 K for H ┴ direction, C = 4.46 cm 3 K mol −1 and θ = −273.01 K for H || direction ( Figure 2B). The very large negative Weiss constants demonstrate strong AFM interactions between the Mn(II) ions in this complex. Through continuous cooling, a maximum value of χ M was observed at T max = 76 K, which corresponds to the short-range AFM ordering of spins within the layer. This phenomenon is common in 2D magnetic hybrids. [34] However, the interlayer magnetic interaction is not strong enough to generate longrange AFM order in this temperature because of the thermal dominance. Upon further cooling, complex 1 undergoes a magnetic ordering transition (short-to long-range ordering) at T N = 45 K ( Figure 3B). The abrupt increase of χ M T indicates that complex 1 is not a pure antiferromagnet. It is obvious that there are net spins arranged in an orderly manner and magnetic domains formed. Meanwhile, below T N , the magnetic properties of complex 1 show the significant anisotropy and the existence of a preferred orientation of spins along different axis. In other words, the susceptibility perpendicular to the inorganic layer (bc plane), χ ┴ decreased smoothly and approached zero as the temperature decreased, corresponding to the easily anti-parallel alignment of spins (along a axis). This alignment is the same as within the layers of (C n H 2n+1 NH 3 ) 2 MnCl 4 (n = 1, 2 and 3) reported by Achiwa, [33] but the inorganic layer is in ab plane. In contrast, the magnetic susceptibility parallel to the layer, χ || , rises sharply below 45 K (T N ) and goes to 0.071 cm 3 mol −1 the maximum of χ || at 2 K. This dissimilarity indicates that the net spins along the external field are retained and form weak ferromagnet. Ferromagnetic concerns can be attributed to spin canting. A perfect anti-parallel arrangement of spins on adjacent metal ions within the AFM layer is not achieved, but net spins are produced within the bc plane. Spin canting has been ascribed to antisymmetric Dzyaloshinsky-Moriya interactions [35,36] in this family of hybrids, characterized by coupling among the spin directions and the tilting of the MnCl 6 octahedra. [31][32][33] Field-dependent magnetizations of complex 1 were measured in the whole field along the two magnetic field orientations at temperatures 1.8 K, which is below ordering temperature (T N ). The measurements show the magnetic behaviour along the two crystal orientations was completely different (Figure S4.). When applying a magnetic field perpendicular to the Mn-Cl layer (H ┴ ), the critical point H sf = 3.5 T is observed rather obviously owing to the spinflop (SF) transition, [37,38] which further confirmed the spins of AFM alignment perpendicular to the Mn−Cl layer (along a axis). Furthermore, the magnetic field parallel to the Mn-Cl layer (H || ) produces hysteresis around zero magnetic field, and the magnetization increases more quickly than along H ┴ , indicating the easily magnetized axis on net spins within the Mn−Cl layer. The M(H) increases linearly until 7.0 T unsaturated due to the strong AFM coupling. This behaviour is typical for weak ferromagnets. Namely, the sharp increase in the low field reflects spontaneous magnetization and linear high field behaviour due to the main AFM interaction. This also indicates that the canted net spin parallels the Mn−Cl layer. The hysteresis loop observed at 1.8 K shows a coercive field (H c ) of 400 Oe ( Figure 2C). The remnant magnetization (M r ) along the measuring direction is 0.0032 μ B . The canting angle is estimated to be 0.037 • according to the equation sin α = M r /M s .
In order to confirm spin canting and 3D ordering behaviour, the alternating current (AC) susceptibilities were measured for complex 1. The in-phase susceptibility (χ M ′) data in the perpendicular direction, H ┴ ( Figure S5), are close to zero, which is consistent with an easy axis for AFM alignment. In contrast, the in-phase in parallel direction, H || ( Figure 2D) shows an apparent peak of about 43 K, which corresponds to the AFM ordering. At the same time, the signals of both in-phase and out-of-phase show a wide peak in the temperature range from 20 to 40 K, suggesting that the domain walls of weak ferromagnet are moving with temperature. The AC susceptibility data indicate that the antiparalleled spins are almost completely oriented along the a axis below T N , while the spin tilt occurs within the Mn−Cl layer.
From the above direct current and AC magnetic measurements, it could be concluded that complex 1 exhibits a well-isolated two-dimensional Heisenberg AFM property with characteristics that enable to achieve axis anisotropy with a temperature lower than T N due to spin canting. The magnetic anisotropy indicates the presence of the competing coupling along two direction of H ┴ and H || due to the layered structure in 1. The Cl atoms mediate the intralayer strong AFM coupling, and the interlayer dipole-dipole weak interaction leads to the ferromagnetic interaction. The latter interaction is so weak that the net spins are quite few. The very small values of AC susceptibility also reflects this situation, even no out-of-phase signals were observed in the similar system reported by Huh. [39] In addition to magnetic properties, complex 1 also exhibits intriguing PL. As can be seen in Figure 3A, the square crystals of complex 1 are light-pink and somewhat transparent under natural light. When exposed to UV light, they show orange-red PL ( Figure 3B). In order to deeply understand the optical properties and luminescence mechanism, we performed the measurement of absorption and PL spectra at room temperature. As shown in Figure 3C, two successively intense peaks (336 and 358 nm) owing to 6 A 1g (S) → 4 E g (D) and 6 A 1g (S) → 4 T 2g (D) transitions of Mn 2+ ions, respectively, which can be observed in the UV region. Besides the absorption in the UV range, complex 1 also shows strong and representative absorption in the visible range (400-760 nm) centred at around 423 and 508 nm. They could be ascribed to the transitions of 6 A 1g (S) → 4 A 1g (S), 4 E g (G), and 6 A 1g (S) → 4 T 1g (G), respectively, which are in agreement with the previous reports. [40,41] The PL decay kinetics were investigated using time-resolved PL measurements of complex 1 ( Figure  S6B). The PL lifetime is 2.96 μs, and absolute quantum yield (QY) is 2.87%, which is owing to the spin forbidden d-d transition in manganese ions.
Both magnetic ordering and PL in complex 1 provides us with opportunities to study magneto-optical properties. To investigate the effect of magnetic field on the luminescent spectra, we further measured the luminescent spectra below and above the AFM ordering temperature (76 K) and ferromagnetic phase (45 K) transition temperature under the action of pulsed magnetic field (0-40 T, Figure S7) using a single crystal of complex 1. Upon cooling, the fluorescent spectrum occurs blue-shifts below 76 K ( Figure S6A) under zero field due to the crystal lattice contraction at this temperature. [42] The magnetic dependence of luminescent spectrum of complex 1 was measured at different temperatures (8,40,50 and 77 K) and in the whole magnetic field range (0-40 T). As shown in Figure 4 and Figure S7A, the position of the maximum emission peak shows no dependence on the magnetic field when the measuring temperature is higher than 76 K. But, an obvious dependence of the position of emission peak on the magnetic field is found when the measuring temperature is lower than 76 K. The maximum emission peak has an obvious red-shift with the increasing magnetic field, which is probably due to the change of bond length caused by magnetostriction resulting from magnetic ordering. This phenomenon is different from the magneto-optical effect reported previously. In the reported compounds, the application of the magnetic field could only result in the change of the intensity of the peak. [8][9][10]16,17] However, for complex 1, it is interesting that the magnetic field cannot only change the intensity but also produce the optical hysteresis at 8 K due to the magnetic order ( Figure S7D). With the temperature falling, the full width at half maximum (FWHM) of fluorescent spectrum decreases gradually, which is caused by the weakening of vibration ( Figure S7C). At the same temperature, the FWHM almost does not depend on magnetic field.
Comparing the spectral intensities at 77, 50, 40 and 8 K under magnetic field, we observed that there is a trough in the strong magnetic field, which is caused by the spin-flip of complex 1 due to the decouple effect in the strong magnetic field (Figure 4 and Figure S7B). Noteworthily, when the temperature (77 and 50 K) is higher than the temperature of magnetic phase transition (43 K), the intensity of luminescence decreases with the increasing magnetic field. However, when the temperature is lower than the magnetic phase transition temperature (45 K), it can be observed that the intensity of luminescence increases with the increasing magnetic field. Meanwhile, a weak peak appears at 4.5 T near to the magnetic field (3.5 T) of SF in the M versus H curve. When the magnetic field increases continuously, the intensity decreases until the magnetic field is sufficient for the spin-flip and then becomes stronger. Comparing the curves of luminescent intensity versus magnetic field for complex 1 at 40 and 50 K in the low field, we find that this difference might originate from the broken spin canting ordering. Furthermore, by control of the spin direction using magnetic field, not only the intensity of the emission spectrum would change, but also the red-shift of the luminescent spectrum could be caused.

CONCLUSION
In conclusion, we investigated the magneto-optical effect of layered organic-inorganic hybrid complex {[MnCl 4 ] 2− } n . The magnetic measurements show the AFM order between Mn ions in the layer occurs below 76 K. The interlayer ferromagnetic interaction leads to the magnetic phase transition and the spin canting behaviour below 43 K. The luminescence studies of complex 1 exhibit that it emits red light. When we used a strong magnetic field to control the spin direction, the luminescence spectrum of complex 1 shows not only change in intensity but also red-shifted with the increasing magnetic field. What's more, we have observed the phenomenon of optical hysteresis in both low and high fields, which is not possessed by other materials. This result proves that optical hysteresis can be realized in 3D-ordered magnetic materials in low field, which provides the possibility for the application of low energy consumption optoelectronic devices.  Figure S1A). The phase purity was verified by powder X-ray diffraction (PXRD, Figure S1B), and the thermal stability was measured by TGA ( Figure S2). Due to the different powder sizes in the PXRD experimental data collection process, there are some differences between the simulated and experimental figures in the reflection intensity.

C O N F L I C T O F I N T E R E S T
The authors declare no conflict of interest.