Phase transitions and photoluminescence switching in hybrid antimony(III) and bismuth(III) halides

Nannan Shen a, Zeping Wang a, Jiance Jin ab, Liaokuo Gong a, Zhizhuan Zhang ac and Xiaoying Huang *a
aState Key Laboratory of Structure Chemistry, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou 350002, China. E-mail: xyhuang@fjirsm.ac.cn
bUniversity of Chinese Academy of Sciences, Beijing 100049, China
cCollege of Chemistry and Materials Science, Fujian Normal University, Fuzhou 350002, China

Received 15th January 2020 , Accepted 30th March 2020

First published on 31st March 2020


Main group metal ions antimony(III) and bismuth(III) with external electron shell of s2 configuration belong to the so-called mercury-like ions. Recently, luminescent antimony(III) and bismuth(III) hybrid halide compounds have aroused great research interest. In particular, the aggregation/crystallization induced emission in hybrid antimony(III) and bismuth(III) halides endows them with molecular packing dependent emissive properties. As a result, they exhibit great potential for achieving reversible phase transformations and switchable photoluminescence (PL) under external stimuli. In this review article, recent advances in this context are highlighted with the aid of illustrative examples, and the structure–optical property relationships are further emphasized. The present challenges and future prospects of smart hybrid antimony(III) and bismuth(III) halides are also demonstrated.


Introduction

Photoluminescent (PL) materials possessing PL switching properties in response to external stimuli, such as light, temperature, pressure, mechanical force and vapours, have an attractive prospect owing to their potential applications in sensors,1,2 memory chips,3–6 and security inks.7 In the year 2001, Tang and co-authors established the concepts of aggregation/crystallization-induced emission (AIE/CIE), that is, emission occurred only in the aggregated or crystalline states in contrast to monomeric or amorphous states.8 In AIE/CIE, the involved non-covalent interactions such as hydrogen bonding, anion–π, and C–H⋯π interactions in aggregated and crystalline states are considered to be the main factor for enhancing luminescence via restricting intramolecular motions and shutting off the non-radiative decay channels.9–12 Moreover, the luminescence properties, e.g., emission colour, lifetimes, quantum yields, are closely related to the packing modes of AIE/CIE molecules. Since non-covalent interactions are weak and flexible, they could be formed or broken under ambient conditions. Therefore, it becomes possible to tune the solid state emissions of AIE/CIE materials via modulating molecular packing modes through non-covalent routes.13,14 Furthermore, reversible structural transformation and PL-switching under external stimuli may be achieved. Thus far, most of the reported AIE/CIE phenomena were observed in complex organic molecules or precious metal based PL materials, which may be faced with the complicated preparation and purification processes and high costs. Study on new types of AIE/CIE materials requires continuous exploration.

Main groups ions with external s2 electron configuration (M = e.g., Tl+, Sn2+, Pb2+, Sb3+, Bi3+, Te4+), known as mercury-like ions, have been considered to possess efficient luminescence. As early as 1940, Pringsheim and Vogels et al. described the photoluminescence of Tl+, Sn2+, and Pb2+ in aqueous solution.15–17 Emissive s2 ions have been used as efficient activators doped into various solid host lattices.18–20 Although investigations on the photophysics and photochemistry of non-aqueous solution and solid state of halide compounds based on main group s2 ions can be traced back to the year 1990, most of the studies on these compounds were limited to the crystal structures, the nonlinear optical properties, and the spectroscopic characterization.21–29 Until recently, the booming development in low-cost and easy-to-make lead(II) halide perovskites has renewed the interest in studying the luminescence of inorganic–organic hybrid metal halides (IOMHs) based on main group s2 ions.30–34 In lead halide perovskites, the electronic dimensionality could be engineered by tuning the crystal size (e.g., nanocrystals), or by structurally controlling the connectivity of lead halide polyhedra. Particularly, low dimensional lead halides exhibit remarkable emission due to strong quantum confinement and site isolation.35 Thus, the luminescence properties of halide compounds based on Sb(III) and Bi(II) ions, as the same s2 element, are worth studying.

Generally, IOMHs of main group s2 ions can be classified into two types, i.e., neutral metal halide molecules with directly coordinated organic ligands and ionic compounds composed of organic cations (A) and [MyXn]z anionic sublattices (X = F, Cl, Br and I). Among them, Sb(III)/Bi(III)-IOMHs are prominent in many aspects. For instance, ionic compounds based on Sb(III) could possess the near unity PL quantum yield, which is hardly achievable for crystalline Pb-IOMHs.30,36,37 It is worth noting that Sb(III) and Bi(III) may adopt different coordination numbers and geometry distortion in [Sb(III)/Bi(III)Xn], which affects their emissive properties. In particular, an environment full of non-covalent interactions in IOMHs would easily alter the packing mode of the molecules or ions, thus achieving diversified PL properties.38–41 Thus, the conversion between different crystalline polymorphs and transformation from crystalline into amorphous states, along with the changing of emissive properties could be easily triggered by external stimuli.38,42 Indeed, recent studies reveal that many Sb(III)/Bi(III)-IOMHs exhibit AIE/CIE properties and interesting structural and PL transformations under stimuli.

The primary scope of this review article is to provide a concise summary of the stimuli-responsive phase transitions and PL switching in Sb(III)/Bi(III)-IOMHs, including recent contributions from our own research, with a special emphasis on Sb(III)/Bi(III)-IOMHs containing flexible ionic liquid (IL) cations and their PL switchable properties. Toward this end, a necessary introduction into the luminescence characteristics of Sb(III)/Bi(III)-IOMHs is included. Finally, the present challenges and further perspectives concerning this area are highlighted.

Structure and luminescence characteristics of metal halides based on main group s2 ions

As mentioned above, IOMHs of main group s2 ions can be classified into two types. For Sb(III)/Bi(III)-IOMHs, the metal centres could adopt five-coordinated square pyramidal and six-coordinated octahedral configurations, in which the s2 lone electron pairs are stereo-chemically active and inactive, respectively.43 Such mononuclear coordination polyhedra can be further joined together by sharing common corners, edges, or faces to form polymeric anionic sublattices with diverse stoichiometries, defined as type I anions (Fig. 1). For instance, antimony(III) halide anions exist in the forms of [SbX4], [SbX5]2−, [SbX6]3−, and [Sb2X9]3−. In the case of [SbX4], octahedra or square pyramids were connected to form infinite polymeric chains and di-, tri-, or tetramers. The anionic sublattice can be composed of mono-, di-, and tetramers or infinite polymeric chains in the case of [SbX5]2− stoichiometry and dimers or infinite polymeric chains in the case of [Sb2X9]3− stoichiometry. For the [SbX6]3− stoichiometry, the anion is formed by isolated octahedra. For bismuth(III) halide compounds, the types of anionic sublattices are more complex and diverse: the BiX6 octahedra can not only polymerizes into zero dimensional di-, tri-, tetra-, penta-, hexa-, hepta-, and octamers, but also aggregates into infinite chains with the stoichiometries of [BiX5]2−, [Bi2X8]2−, [Bi2X9]3−, [Bi3X11]2−, [Bi4X20]2−, [Bi6X20]2− and two dimensional [Bi2X9]3− layers.44,45 While in type II anions, nitrogen or oxygen containing ligands could replace some halogen ions, forming organically decorated metal halide anions.46
image file: d0ce00057d-f1.tif
Fig. 1 Structural diagrams for type I [MyXn]z− (M = Sb3+, Bi3+; X = F, Cl, Br and I) anionic sublattices.

Previous studies indicate that the luminescence of s2 halide complexes of type I mainly originates from the s → p transitions of free s2 ions: the ground state is 1S0, and the excited states are 3P0, 3P1, 3P2, and 1P1. The transition from 1S0 to 3P0 is strictly forbidden. The remaining three kinds of transitions 1S03P1, 1S03P2, 1S01P1 are denoted as A, B, and C bands, respectively. The transition 1S03P2 (B band) is forbidden but can be induced by lattice vibrations. The transition from 1S0 to 3P1 is partially allowed by spin–orbit coupling, governing the luminescence of s2 halide compounds.16,47 At present, the emissions from self-trapped excitons have been fully discussed.30,48 For antimony(III) and bismuth(III) halide compounds, the valence band consists of the hybridization between the occupied Sb-5s/Bi-6s and p orbits of halogen ions, while the conduction band primarily derives from Sb-5p/Bi-6p states. In general, the emission of s2 halide compounds has the following features:

(1) The geometry of the anion sublattice, in particular the degree of asymmetry in the s2 ion coordination polyhedron, has significant influences on the emission properties. Compared with ground states, metal halide coordination polyhedra in the excited states undergo structural deformation induced by the second-order Jahn–Teller effect of s2 lone-pair electrons, forming a more symmetrical geometry. Therefore, the luminescence of these compounds usually features relatively large Stokes shift and broad bands. Nevertheless, excitation energy dissipation also occurs during the structural reorganization processes, which in some cases may lead to luminescence quenching. Generally, compared with isolated mononuclear clusters, polymeric structures exhibit greater distortion and provide non-radiation energy dissipation channels. Hence, low dimensional compounds of type I, especially compounds composed of isolated MX5 or MX6 anions, usually exhibit excellent luminescence performance. Note that except for isolated MX5 or MX6 anions, compounds based on other anions can also emit light.15,16,18–21

(2) Different types of halogens do affect the energy of the excited 3P1-level and the basic 1S0-level. The transitions from chloride to bromide and iodide complexes result in a considerable bathochromic shift of the A-band in the excitation and emission spectra. Heavy halogens in turn facilitate the overlapping of the 3P1 and 1S0 states and the increased relaxation of the electron excitation energy, thus leading to the drop of luminescence intensity or emission quenching, especially in iodide complexes, relative to that in chloride or bromide compounds.49

(3) Outer-sphere organic cations could also work as potential luminescence sources in s2 halide compounds. They can either sensitize or quench the luminescence of s2 ions, which depends on the energy levels of HOMO and LUMO orbits in organic parts and s2 anion sublattices.24,27

For Sb(III)/Bi(III)-IOHMs of type II, in the organically-decorated metal halide anions, the low energy unoccupied orbits on ligands promote the charge transfer from inorganic parts (metal and halogen ions) to organic parts.46 Moreover, in both cases, the kinetic lability, variable coordination numbers, the sp orbital mixing and low structural symmetry as imposed by the lone electron pairs further complicate the spectroscopic identification and characterization.

Phase transitions and PL switching in antimony(III) and bismuth(III)-IOHMs

Influenced by different factors, diverse aggregated states in antimony(III) and bismuth(III) hybrid halides could be formed, such as polymorphs, solvates, and amorphous phases with different molecular arrangements. (1) Crystalline polymorphs: a competition between kinetic and thermodynamic crystallization processes may lead to the formation of polymorphs with different molecular packing modes; for ionic compounds, the combination of rotationally flexible organic cations with antimony(III) and bismuth(III) halide anions could also lead to luminescent polymorphs. (2) Amorphous states: crystalline compounds could be ground or heated into amorphous states. (3) Solvates: during crystallization, solvent molecules may enter the crystal lattices, affecting the interactions within host molecules. (4) The kinetic lability and variable coordination numbers of the antimony(III)/bismuth(III) halide coordination polyhedra may lead to the transformation between different compounds. In this section, we will try to integrate the phase transitions and PL switchable properties of hybrid antimony(III)/bismuth(III) halides according to the above-mentioned types. Special efforts will be made to exhibit how the structural aspects, such as the variations of packing modes, conformations, and morphologies, influence light-emitting behaviours.

Polymorphism dependent emission and PL switching

By assembling 2,2′-bipyridine (2,2′-bpy), 4,4′-bipyridine (4,4′-bpy) and their derivatives with inorganic BiX3 units, a series of hybrid bismuth(III) halide coordination complexes with efficient luminescence were obtained by Mercier and co-authors.50,51 Among them, three polymorphs α-, β-, and γ-[BiBr3(bp2mo)2] (1) were synthesized by solvothermal reactions from a mixture of BiBr3 and ditopic ligand N-oxide-2,2′-bipyridine (bp2mo) in acetonitrile with slight differences in the stoichiometry of reagents and temperature programs.38 The crystal structures of the three phases were refined in the space groups P[1 with combining macron], P21/n and P21/c, respectively, and featured chiral complex units based on one Bi3+ ion coordinated by three Br ions and two bp2mo molecules. The two bp2mo molecules were connected to Bi3+via their pyridyl-N and pyridyl-N–O parts. The dihedral angle between the two pyridyl rings (30–46°) resulted from both the N–Bi bond, with Bi being in the plane defined by the pyridyl ring, and the (N)O–Bi coordination involving a N–O–Bi bond angle close to 120°. The geometrical characteristics of the chiral complex units in the three phases (three complex units in the asymmetric units of 1 and only one unit in 1 and 1) were somewhat different in the dihedral angles and the Bi–N or Bi–O bond distances. The overall structures of the three polymorphs differed by the packing of these complex units, as depicted in Fig. 2b and c.
image file: d0ce00057d-f2.tif
Fig. 2 (a) Structures of the chiral [BiBr3(bp2mo)2]. The packing structures of crystalline polymorphs 1 (b), 1 (c), and 1 (d).

[BiBr3(bp2mo)2] polymorphs were aggregation induced phosphorescence (AIP) active. In solutions, [BiBr3(bp2mo)] exhibited an extremely weak blue emission (λmax = 442 nm, QY = 0.01%) characterized by a very short lifetime (258 ps), indicative of emission from a singlet excited state. While in the solid state, the polymorphs 1 and 1 displayed bright greenish-yellow emissions which were principally from triplet excited states and characterized by different PL efficiencies and longer lifetimes (for 1, λmax = 525 nm, QY = 17%, τ = 4.8 μs; for 1: λmax = 503 nm, QY = 5%, τ = 1.0 μs); polymorph 1 showed a very weak broad emission (λmax = 516 nm, QY = 0.5%). DFT/TDDFT calculations revealed that in the optimized geometries of lower energy singlet (S1) and triplet (T1) excited states, one bp2mo ligand underwent partial planarization compared to the ground state S0 geometry, forming a pseudo-quinoid structure and stronger Bi–O and Bi–N bonds, while the other ligand essentially preserved its distortion. At both optimized S1 and T1 states, the LUMO was localized only on the more planar bp2mo ligand and the HOMO essentially preserved its distribution on the inorganic part of the complex (Bi and Br). In the solid states, the torsional motion around the bond connecting the two pyridine moieties was greatly prevented by the non-covalent bonds and the presence of the heavy atom (Bi) induced the occurrence of intersystem crossing from S1 to the low-lying T1 triplets, finally leading to bright phosphorescence emission. However, in solution, the long lived phosphorescence was completely quenched by free intramolecular motion, but the solution still preserved the weak fluorescence from the short lived singlet state. The different emission properties in the three crystal polymorphs (the stronger emission of polymorph 1 compared to those of 1 and, to a less extent, 1) can be ascribed to the different weak interactions in the three phases. C–H⋯Br interactions were similar in the three structures, and ‘stronger’ C–H⋯O, and C–H⋯π hydrogen bonds were found in 1 and 1, which may explain their higher luminescence compared to 1. On the other hand, the higher QY of 1 compared to that of 1 could be explained by the lower number of intermolecular interactions found in the latter structure.

Besides the crystalline polymorphs constructed from the different packing of neutral molecules, the rotationally flexible organic cations could also direct the formation of polymorphism in ionic compounds. Ionic liquids (ILs), consisting of organic cations and anions, are a subset of molten salts with the melting points at or below 100 °C. Nowadays, ionic liquids have become the hot research topic among multidisciplinary areas, especially in the construction of functional materials.52–54 Through modification of the cations and anions, the properties of ILs can be altered. Such great tunability of cations and anions gives the possibility to design an IL according to a specific purpose or task, hence the commonly coined term “task-specific ionic liquids”.55,56 Our group has developed a new class of luminescent materials by combining common ionic liquid cations and metal halide complex anions of s2 metal ions. Compared with traditional organic-amine cations, the use of ionic liquid cations provides unique opportunities for tuning and switching luminescence properties.

For instance, one interesting but overlooked characteristic of ILs is the rotational isomerism of their cations that contain flexible alkyl chains, such as imidazolium, pyridinium, and pyrrolidinium cations. The combination of such IL cations with appropriate anions provides an effective strategy for the construction of luminescent polymorphs. [Bmim]+ (1-butyl-3-methyl imidazolium) and [Bmmim]+ (1-butyl-2,3-dimethyl imidazolium) are two typical ionic liquid cations that exhibit conformational isomerism.57–60 As demonstrated by Schottenberger and Siehl, four different conformations (TT, TG, GT, GG) could be represented by butyl chains when the [Bmmim]+ cation was combined with different anions, as listed in Fig. 3.60T and G respectively refer to trans and gauche conformations, which were determined by the torsion angle of the C(α)–C(β) and C(β)–C(γ) bonds. In T conformation, the torsion angle was in the range of 170–179° while in G conformation, the torsion angle ranged from 63° to 90°.


image file: d0ce00057d-f3.tif
Fig. 3 The conformation of [Bmmim]+ cations in different salts. Reproduced from ref. 60 with permission from the American Chemical Society, Copyright 2012.

Based on this, our group obtained two kinds of luminescent polymorphs by assembling [BiCl4(2,2′-bpy)] (2,2′-bpy = 2,2′-bipyridyl) anions with rotationally isomeric [Bmim]+ and [Bmmim]+ cations, namely α/β-[Bmim][BiCl4(2,2′-bpy)] (2)39 and α/β-[Bmmim][BiCl4(2,2′-bpy)] (3),40 respectively. Because of the structural disorder of the butyl chains, the conformational differences between the cations in 2 and 2 were not distinct, and the emission properties of 2 (λmax = 530 nm, QY = 26.07%, lifetime = 8.045 μs) and 2 (λem = 524 nm, QY = 36.59%, lifetime = 12.56 μs) were similar. While in 3 and 3, the n-butyl chains exhibited TG and GT conformations, respectively (Fig. 4). Thus, two polymorphs crystallized in different space groups: 3 belonged to the triclinic space group P[1 with combining macron], and 3 crystallized in the monoclinic space group P21/c.


image file: d0ce00057d-f4.tif
Fig. 4 The asymmetric unit (top) and the π–π contacts between adjacent [BiCl4(2,2′-bpy)] (bottom) in compounds 3 (a) and 3 (b). Reproduced from ref. 40 with permission from the American Chemical Society, Copyright 2019.

Under 365 nm UV light irradiation, 3 and 3 showed greenish-blue and greenish-yellow emission with maximum emission wavelengths of 490 and 520 nm (Fig. 5a–c). Single-crystal structure and Hirshfeld surface analyses disclosed that the polymorphism-dependent emission was attributed to the different weak interactions, especially to the weak π–π interactions between adjacent [BiCl4(2,2′-bpy)] anions in the two compounds. As depicted in Fig. 4, in 3, the centre-to-centre distance of the pyridine ring was 3.873(3) Å with a slippage of 1.905 Å, while in 3, the slippage was up to 3.024 Å, and the π–π contacts (more than 4.5 Å) can be ignored. The stronger interactions in 3 would increase the structure rigidity at excited states and reduce the energy reduction caused by intramolecular vibrations, resulting in higher quantum yields and longer lifetimes (3: QY = 44.98%, lifetime = 190.01 μs; 3: QY = 36.36%, lifetime = 38.78 μs) (Fig. 5d). Interestingly, 3 could spontaneously be transformed into 3 under ambient conditions. The transformation may be triggered by the moisture in the air (Fig. 5e).


image file: d0ce00057d-f5.tif
Fig. 5 (a) Solid-state emission images of 3 and 3 under 365 nm UV light. (b) Excitation (dashed lines) and emission (solid lines) spectra of 3 (cyan) and 3 (light green): λex = 370 nm for 3 and 3, λem = 490 nm for 3, and λem = 520 nm for 3. (c) CIE (1931) chromaticity diagrams. (d) Time-resolved PL spectra of 3 and 3. (e) Images under daylight (top) and 365 nm UV light (bottom). Reproduced from ref. 40 with permission from the American Chemical Society, Copyright 2019.

Crystalline to amorphous state transformation and emission switching

Since the emissive properties are strictly related to the detailed weak interactions, the peculiar categories of AIE or AIP materials have great potential to exhibit PL-switching towards mechanical forces, which commonly causes the amorphisation of samples. As shown in Fig. 6, upon grinding the samples of 1 (λmax = 525 nm) and 1 (λmax = 503 nm), the green phosphorescence gradually became orange-red (1-α-g, λmax = 611 nm; 1-γ-g, λmax = 614 nm), exhibiting a quite large PL shift (100 nm). The mechanical force induced solid state structure changes from crystalline to amorphous states, as demonstrated by PXRD analysis. The processes could be reversible by different treatments. Surprisingly, whatever the ground sample, either 1-α-g or 1-γ-g, was used, heating and fuming the ground samples in H2O or ACN, involving slow recrystallization processes allowed the recovery of the thermodynamic phase 1, while the fast recrystallization of both 1-α-g and 1-γ-g by adding a few drops of ACN resulted in the kinetic phase 1.42
image file: d0ce00057d-f6.tif
Fig. 6 (a) PL spectra of 1 and 1-α-g (left). Photos of 1 and 1 before and after grinding under UV light (right). PXRD patterns of ground samples 1-α-g and 1-γ-g after heat treatment (1-α-g, Δ and 1-γ-g, Δ) (b) and exposure to water (1-α-g, H2O(v) and 1-γ-g, H2O(v)) or ACN vapours (1-α-g, ACN(v) and 1-γ-g, ACN(v)) (c) showing that polymorph 1 is systematically recovered. (d) PXRD patterns of ground samples 1-α-g and 1-γ-g after a few drops of ACN deposited on them and rapidly evaporated (1-α-g, ACN(s) and 1-γ-g, ACN(s)) showing that polymorph 1 is obtained as the main phase (* is for lines belonging to 1). Reproduced from ref. 42 with permission from the Royal Society of Chemistry, Copyright 2016.

(TBA)[BiBr4(bp4mo)] (4, TBA = tetrabutylammonium, bp4mo = N-oxide-4,4′-bipyridine) and [BiBr3(bp4mo)2] (5) were another two hybrid bismuth(III) halide compounds that exhibited aggregation induced emission and mechanochromic luminescence.33 In these two compounds, the inorganic coordination polyhedra were bridged by another ditopic bp4mo ligand, forming 1D coordination polymers (CPs) (Fig. 7a and b). Efficient solid-state phosphorescence peaks at 540 and 516 nm with quantum yields of 85% and 15% were determined for 4 and 5. The lifetimes were determined to be 18 and 1 μs, respectively. Upon grinding, the luminescence changed from yellow (λmax = 540 nm, 4) to orange (λmax = 585 nm, 4-g) and from greenish-yellow (λmax = 516 nm, 5) to orange-red (λmax = 622 nm, 5-g) (Fig. 7c), accompanying the transformation from crystalline to amorphous states. The wavelength shift of 106 nm for 5 was one of the largest ever reported for MCL complexes (Fig. 7d). However, by heating and fuming the ground samples with saturated water atmosphere, reversible emission changes were clearly seen by the naked eye for both 4-g and 5-g (Fig. 7e and f). Nevertheless, the reversibility was only partial for 5-g. It is assumed that the key parameter for the change in luminescence could be ascribed to the conformational change of the N–O–Bi moiety upon grinding. Such reversible PL switching in hybrid halobismuthate CPs indicated their potential applications in the fields of organic compound detection, and humidity and thermal sensing.


image file: d0ce00057d-f7.tif
Fig. 7 General views of 4 (a) and 5 (b) showing adjacent 1D coordination polymers. (c) PL spectra of 4 and 4-g (left) and 5 and 5-g (right). Inset: Photographs under UV light. (d) Grinding of 5 in a mortar, photographs under UV light. (e) PXRD patterns of 4, 4-g, 4-g, Δ, and 4-g, H2O (left) and DSC curves (right) of 4 and 4-g. (f) Photographs under ambient light (left) and UV light (right) of letters written on a 4-g sample with a heating pen. Reproduced from ref. 33 with permission from the Wiley-VCH, Copyright 2016.

Ionic liquids are characterized with low melting points and easy processability. Sb(III)/Bi(III)-IOMHs containing IL cations also possess the above properties. In 2015, our group reported the first example of antimony(III)-containing ionic liquid, namely [Bmim]2SbCl5 (6, Bmim = 1-butyl-3-methyl imidazolium).366 was crystalized at room temperature from the homogeneous solution formed by heating the mixture of [Bmim]Cl and SbCl3 at 70 °C. The crystallographic asymmetric unit was composed of six [Bmim]+ cations and three isolated [SbCl5]2− anions. Every Sb3+ ion was five-coordinated by Cl ions, forming square-pyramidal configuration (Fig. 8a). 6 exhibited bright yellow light emission when exposed to UV light irradiation. PL measurements and DFT calculations indicated that 6 had two different luminescent centres with the emission peaks at 460 and 583 nm, respectively. The emission band peaking at 583 nm originated from [SbCl5]2− anions, corresponding to a wide range of excitation wavelengths from 250 to 400 nm, while the emission band around 460 nm came from the intra-ligand charged transfer of [Bmim]+ cations and could be excited by 303 nm independently. By varying the excitation wavelengths from 250 to 450 nm, the emission colour of 6 can be changed from yellow to white, and finally to yellow again (Fig. 7b). In other words, both emission from cations and anions could be exhibited in 6 by tuning excitation wavelengths. This phenomenon was different from that of the previous compounds, in which only one kind of luminescent centre worked.


image file: d0ce00057d-f8.tif
Fig. 8 (a) Photographs of the polycrystalline compound 6 and its crystal structural diagram. (b) Emission spectra of 6 at different excitation wavelengths. Inset: Excitation spectra of 6 at different emission wavelengths (left); CIE coordinates of the emission colours and corresponding PL images of 6 at different excitation wavelengths (right). (c) Photographs of UV-LED lamp and the LED coated with a thin layer of compound 6 in the states of turned off (left) and on (middle), and the photographs of the film coated on a piece of glass in daylight lamp and under UV-LED lamp (right), respectively. Reproduced from ref. 36 with permission from the Royal Society of Chemistry, Copyright 2015.

Significantly, 6 was characterized with a high quantum yield of 86.3% (λex = 370 nm) in the crystalline state at room temperature. Such efficient luminescence was directly related to its crystalline structures. In 6, [SbCl5]2− anions were completely isolated from each other and periodically doped in the wide band gap matrix of [Bmim]+ cations (Fig. 8a). Such perfect host–guest systems have been confirmed to be beneficial for the intrinsic luminescence of the individual metal halide species. Additionally, the rigid structure in the crystalline state built by weak interactions (hydrogen bonds and anion–π interactions) between cations and anions could efficiently reduce the energy dissipation caused by thermal vibration. Owing to the introduction of ionic liquid, 6 was characterized with a low melting point of 349 K. As demonstrated by the CIE properties, the bright yellow emission could be quenched absolutely when it was heated to form an amorphous molten state; while after the temperature decreased below 349 K, the luminescence could be regenerated accompanied by the recrystallization. Based on this, 6 could be easily fabricated into temperature sensing thin films and LEDs (Fig. 8c).

The hybrid chlorobismuthates(III) 2 and 3 mentioned above also exhibited similar phase transitions from crystalline to amorphous states upon heating (melting points: about 110 °C for 2 and 3).39,40 Additionally, crystalline polymorphs of α/β-[Bmmim][BiCl4(2,2′-bpy)] also underwent transformation into the amorphous state when exposed to NH3 vapour (Fig. 9a). As demonstrated by the PXRD patterns (Fig. 9b), the crystallinity of the samples was severely weakened under the treatment of NH3 vapour for a period of time, along with the formation of NH4Cl and amorphous powders. Because the two title compounds possessed the nature of crystallization-induced emission, their PL was then completely quenched (Fig. 9a), while after the removal of NH3 vapour, the samples could be recrystallized to their original states with the recovery of luminescence.40


image file: d0ce00057d-f9.tif
Fig. 9 (a) Images under daylight and UV light showing the luminescence quenching and recovery of compounds 3 and 3. (b) PXRD pattern changes of compound 3 before and after exposure to NH3 vapor. Reproduced from ref. 40 with permission from the American Chemical Society, Copyright 2019.

Solvent effects

Solvent molecules could also affect the luminescence through non-covalent interactions. 4,4-Bipyridinium salt of binuclear chlorobismuthate complex [bipyH2]2[Bi2Cl10]·2H2O (7) synthesized by Adonin displayed yellow luminescence (λmax = 530 nm) in the solid state. The solvate composition strongly affected the luminescence by blocking or altering cation–anion interactions based on hydrogen bonds. After dehydration, (bipyH2)2[Bi2Cl10] (8) showed orange luminescence with the maximum emission wavelength of 560 nm.61

Biwu Ma and co-authors observed the solvent effects in [Ph4P]2[SbCl5] (9).48 Similar to compound 6, owing to the excited state structural reorganization of the individual [SbCl5]2− species, strong Stokes shifted broadband emission bands (λex = 375 nm, λem = 648 nm) with photoluminescence quantum efficiencies of up to near-unity (87%) were determined. A rapid crystal growth process within minutes in the synthesis of 9 could yield yellow emission (λmax = 600 nm), which kinetically favoured a metastable solvate with a PLQE of around 100%. Such metastable species could turn into the red emitting thermodynamically stable crystals slowly at room temperature or quickly upon thermal treatment (Fig. 10). The transformation process also occurred on the thin films. The variation of emission was attributed to the removal of solvent molecules (e.g. DMF) in crystals and thin films.


image file: d0ce00057d-f10.tif
Fig. 10 (a) View of the crystal structure of 9 (top) and its crystal growth pathways (bottom). (b) In situ photoluminescence spectra of bulk crystals during the rapid crystal growth of 9. Reproduced from ref. 48 with permission from the American Chemical Society, Copyright 2018.

Combined effects of CIE and the kinetically lability of coordination polyhedra

Besides the PL switching induced by transformation between crystalline and molten states or solvent effects, the combined features of IOMHs including kinetically lability of antimony(III) halide coordination anions and easy assembly between ionic liquid cations and antimony halide anions provided unique access to multi-stimuli responsive PL switching materials. By introducing antimony(III) into the IL, [Bzmim]Cl (Bzmim = 1-benzyl-3-methylimidazolium) and controlling the crystallization processes, thermally induced triple-mode reversible PL switching materials based on two hybrid chloroantimonates(III), namely [Bzmim]3SbCl6 (10) and [Bzmim]2SbCl5 (11), were realized.37 The Sb3+ ions in 10 were six-coordinated by Cl ions, forming disordered octahedral configuration (Fig. 11a), while the [SbCl5]2− anions in 11 adopted a square pyramidal geometry (Fig. 11b). The abundant weak interactions including hydrogen bonds and π–π interactions allowed 10 and 11 to remain rigid. Under UV light irradiation, 10 exhibited green emission originating from [SbCl6]3− anions with the maximum wavelength at 525 nm (λex = 342 nm) (Fig. 11c). For 11, dual emission bands peaking at 483 and 600 nm, which were attributed to [Bzmim]+ cations and [SbCl5]2− anions respectively (Fig. 11d), could be detected by varying the excitation wavelengths, similar to that of 6. Owing to the second-order Jahn–Teller effect induced by the stereochemically active s2 lone pair electrons in [SbCl5]2−, the emission of 11 appeared at a long wavelength compared with that of 10.
image file: d0ce00057d-f11.tif
Fig. 11 Crystal structures of 10 (a) and 11 (b); dotted lines are used to differentiate the secondary Sb–Cl bonds from the regular ones and π–π interactions between the imidazole ring and benzene ring. PL spectra of 10 (c) and 11 (d). Reproduced from ref. 37 with permission from the Wiley-VCH, Copyright 2019.

Both compounds possessed the properties of crystallization induced emission. The melting points for 10 and 11 were measured to be 410 K (Tm1) and 348 K (Tm2), respectively. By utilizing the difference in melting points and controlling the crystalline processes, the reversible structural and PL transformation between 10 and IL@11 (a mixture of [Bzmim]Cl with 11 with the stoichiometric ratio of 1[thin space (1/6-em)]:[thin space (1/6-em)]1) could result in the formation of firstly reported thermally induced triple-mode reversible PL switching materials, as depicted in Fig. 12. On–off mode: when the temperature was above Tm1, the green emission of 10 could be quenched owing to the collapse of the crystalline structures and the formation of a molten state. Such a molten state could be fabricated into thin films by a silk-screen printing method (Fig. 12a and b). By controlling the cooling processes of the molten thin films, the luminescence of 10 and 11 could be turned on separately: the green-emission crystalline 10 could be obtained by rapidly cooling and scratching the molten films (Fig. 12f, process III), while red-emission IL@11 can be obtained via naturally cooling the molten films under ambient conditions (Fig. 12d, process II), indicating a competition between kinetic and thermodynamic crystallization processes. On–off–on mode: when the temperature was in the range of Tm1 > T > Tm2, IL@11 could melt along with the quenching of red emission (process V, Fig. 12e). Then the green emission could appear because the molten state of IL@11 was gradually solidified into crystalline 10 (process VI, Fig. 12f). Such reversible PL switching from red to non-emission and finally to green was firstly reported. Colour switching mode: when the temperature was below Tm2, the transformation from IL@11 to 10 occurred (process VIII), and the PL switched from red to green. Interestingly, 10 could convert back to IL@11 by the triggering of water vapour (process IX).


image file: d0ce00057d-f12.tif
Fig. 12 Schematic description for the thermally induced triple-mode reversible PL switching between 10 and IL@11. (a) Images of the molten state of 10 at 410 K under visible light. (b) Molten 10 fabricated on a piece of glass at 410 K under UV light. (c) Solid-state thin film at RT under visible light. (d)–(f) The digital pictures of the on–off–on PL switching process from the on state (red-emitting thin film) to the off state (molten film), and finally to the on state (green emission thin film). Procedures: I) silk-screen printing, II) naturally cooling the molten thin film under ambient conditions, III) rapidly cooling to 353 K, IV) rapidly cooling and scratching, V) heating the film at 353 K for 10 seconds, VI) continually heating the film at 353 K for 20 minutes, VII) naturally cooling the molten thin film under ambient conditions, VIII) heating at 330 K for 30 minutes, and IX) keeping in a moist environment for six hours. Δ indicates heating the film to 410 K. Reproduced from ref. 37 with permission from the Wiley-VCH, Copyright 2019.

Such reversible solid-state structural transformation between 10 and IL@11 was closely related to their structures. Compared with 10, 11 featured a looser crystal structure and the cations and anions were more dispersive (Fig. 13a and b). Upon heating IL@11, [Bzmim]+ and Cl might be inserted in the structure. In the structure of 10, four of the six crystallographic independent [Bzmim]+ cations could interact with the nearest one through π–π interactions, which were formed between the imidazole and benzene rings. While for the remaining two cations (marked in yellow), the distances within them were too large to form π–π interactions, indicating that they were unconstrained (Fig. 13b). Moreover, one of the six Cl ions in the octahedral [SbCl6]3− anion was slightly attached to Sb3+ ions, forming secondary Sb–Cl bonds (Fig. 11a). Upon exposure to moisture, H2O molecules absorbed in the surface of crystals of 10 could extract the unconstrained [Bzmim]+ cations and slightly bonded Cl ions via strong dipole–ion interactions, forming IL@11 without breaking the crystals (Fig. 13c). Such unique thermally induced triple-mode reversible PL switching as well as moisture- and excitation wavelength responsive colour switching implied the application of 10 and 11 in high-level anti-counterfeiting technology. As depicted in Fig. 14, by coating the filter paper with polycrystals of 10, ink and mask free rewritable PL paper was achieved. Under UV light, the coated paper emitted green light. The images or letters could be engraved on the paper by laser. The paper would be initialized completely for rewriting after aging at 80 °C for one hour. This study further proved that the combination of ionic liquid cations and chloroantimonate(III) anions to form a hybrid material is an efficient strategy to generate materials with novel properties. The dynamic insertion/extraction of ILs in hybrid materials may shed light on exploring new functional stimuli–response materials.


image file: d0ce00057d-f13.tif
Fig. 13 Diagram of the crystal structural transformation between 11 (a) and 10 (b). (c) In situ photographs of the crystals upon fuming with water vapour and heating under 390 nm UV light and natural light (inset); from top to bottom are crystals of 10, half-converted crystals triggered by moisture, full-converted crystals triggered by moisture and crystals of 10 converted back from the crystals with red emission. Reproduced from ref. 37 with permission from the Wiley-VCH, Copyright 2019.

image file: d0ce00057d-f14.tif
Fig. 14 (a) Demonstration of the laser printing process. Photograph of the filter paper coated with polycrystals of 20 under visible light (b) and 390 nm UV light (c)–(e). Procedures: I) printing the paper with a laser engraving machine, II) keeping the paper in a moist environment for blacking out, III) aging the paper at 353 K for initialization, IV) heating the paper at 353 K, and V) keeping the paper in a moist environment for six hours. Reproduced from ref. 37 with permission from the Wiley-VCH, Copyright 2019.

Summary and outlook

In summary, here we provide a brief review on the recent developments of AIE/CIE-active antimony(III)/bismuth(III)-IOMHs, with emphasis on their phase transitions and PL-switchable properties. Considering the luminescence characteristics of main group s2 ions, rigid environments are highly beneficial to the emission of antimony(III)/bismuth(III)-IOMHs, leading to their AIE/CIE properties. Incorporating the flexible IL cations purposely into IOMHs increases the possibilities of forming different solid states such as crystalline polymorphs, amorphous states, and solvates, and therefore offers a feasible method to alter the PL properties of IOMHs via changing the weak interactions within a structure that can be triggered by a stimulus. However, for this rapidly growing research topic, there are still many issues that need to be further addressed, exemplified as follows.

(1) The effect of the structure of IL cations (for example, the number, type and length of the substituents in imidazolium cations) on the structure of halometallate anions, and the relationship between the packing mode/supramolecular interactions in IOMH crystals and the PL properties need to be investigated in-depth. The further study would help in realizing the controlled synthesis of crystalline polymorphs and clarifying the structural–optical property relationship, especially the origin of stimulus-responsive PL.

(2) It would be very interesting but challenging to use task-specific ILs to achieve the designing of task-specific PL Sb(III)/Bi(III)-IOMH materials that can respond to one special stimulus or multiple stimuli.

(3) The effect of small molecules such as H2O, NH3 and HCl on the composition, structure and PL properties of Sb(III)/Bi(III)-IOMHs has been only reported in a few compounds. This topic deserves further study when considering the construction of stable Sb(III)/Bi(III)-IOMHs for practical applications.

Undoubtedly, there is still a huge expanding research area in constructing stimuli responsive and PL switchable IOMH materials by combining IL cations and haloantimonate(III)/halobismuthate(III) anions. Practical applications of such kinds of PL materials also need to be further explored. Finally, the strategy can be applied to construct luminescent IOMHs based on other main group s2 ions such as Sn2+ and Te4+.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (Grants 21671187).

Notes and references

  1. T. L. Mako, J. M. Racicot and M. Levine, Chem. Rev., 2019, 119, 322–477 CrossRef CAS PubMed.
  2. Y. M. Zhang, S. Yuan, G. Day, X. Wang, X. Y. Yang and H. C. Zhou, Coord. Chem. Rev., 2018, 354, 28–45 CrossRef CAS.
  3. W. P. Lin, Q. Zhao, H. B. Sun, K. Y. Zhang, H. R. Yang, Q. Yu, X. H. Zhou, S. Guo, S. J. Liu and W. Huang, Adv. Opt. Mater., 2015, 3, 368–375 CrossRef CAS.
  4. Q. Z. Zhang, P. C. Xue, P. Gong, G. H. Zhang, J. Peng and R. Lu, J. Mater. Chem. C, 2014, 2, 9543–9551 RSC.
  5. S. Hirata and T. Watanabe, Adv. Mater., 2006, 18, 2725–2729 CrossRef CAS.
  6. M. Irie, T. Fukaminato, T. Sasaki, N. Tamai and T. Kawai, Nature, 2002, 420, 759–760 CrossRef CAS PubMed.
  7. A. Kishimura, T. Yamashita, K. Yamaguchi and T. Aida, Nat. Mater., 2005, 4, 546–549 CrossRef CAS PubMed.
  8. J. D. Luo, Z. L. Xie, J. W. Y. Lam, L. Cheng, H. Y. Chen, C. F. Qiu, H. S. Kwok, X. W. Zhan, Y. Q. Liu, D. B. Zhu and B. Z. Tang, Chem. Commun., 2001, 1740–1741 RSC.
  9. Y. N. Hong, J. W. Y. Lam and B. Z. Tang, Chem. Soc. Rev., 2011, 40, 5361–5388 RSC.
  10. J. Mei, N. L. C. Leung, R. T. K. Kwok, J. W. Y. Lam and B. Z. Tang, Chem. Rev., 2015, 115, 11718–11940 CrossRef CAS PubMed.
  11. J. Qian and B. Z. Tang, Chem, 2017, 3, 56–91 CAS.
  12. J. Yang, Z. G. Chi, W. H. Zhu, B. Z. Tang and Z. Li, Sci. China: Chem., 2019, 62, 1090–1098 CrossRef CAS.
  13. S. Varughese, J. Mater. Chem. C, 2014, 2, 3499–3516 RSC.
  14. D. P. Yan and D. G. Evans, Mater. Horiz., 2014, 1, 46–57 RSC.
  15. A. Vogler, A. Paukner and H. Kunkely, Coord. Chem. Rev., 1990, 97, 285–297 CrossRef CAS.
  16. A. Vogler and H. Nikol, Pure Appl. Chem., 1992, 64, 1311–1317 CAS.
  17. P. Pringsheim and H. Vogels, Physica, 1940, 7, 225–240 CrossRef CAS.
  18. E. W. J. L. Oomen, W. M. A. Smit and G. Blasse, J. Phys. C: Solid State Phys., 1986, 19, 3263–3272 CrossRef CAS.
  19. E. W. J. L. Oomen, G. J. Dirksen, W. M. A. Smit and G. Blasse, J. Phys. C: Solid State Phys., 1987, 20, 1161–1171 CrossRef CAS.
  20. E. W. J. L. Oomen and G. J. Dirksen, Mater. Res. Bull., 1985, 20, 453–457.20 CrossRef CAS.
  21. G. Blasse and A. Vogler, Inorg. Chim. Acta, 1990, 170, 149–150 CrossRef CAS.
  22. N. V. Petrochenkova, T. Storozhuk, A. G. Mirochnik and V. E. Karasev, Russ. J. Coord. Chem., 2002, 28, 468–472 CrossRef CAS.
  23. T. V. Storozhuk, A. A. Udovenko, A. G. Mirochnik, N. V. Petrochenkova and V. E. Karasev, Russ. J. Coord. Chem., 2002, 28, 175–182 CrossRef CAS.
  24. T. V. Storozhuk, A. G. Mirochnik, N. V. Petrochenkova and V. E. Karasev, Opt. Spectrosc., 2003, 94, 920–923 CrossRef CAS.
  25. A. G. Mirochnik, P. A. Zhikhareva, T. V. Sedakova and V. E. Karasev, Russ. J. Phys. Chem. A, 2007, 81, 297–300 CrossRef CAS.
  26. T. V. Sedakova, A. G. Mirochnik and V. E. Karasev, Opt. Spectrosc., 2008, 105, 517–523 CrossRef CAS.
  27. T. V. Sedakova, A. G. Mirochnik and V. E. Karasev, Russ. J. Phys. Chem. A, 2009, 83, 308–310 CrossRef CAS.
  28. T. V. Sedakova and A. G. Mirochnik, Russ. J. Phys. Chem. A, 2017, 91, 791–795 CrossRef CAS.
  29. A. G. Mirochnik, N. V. Petrochenkova, T. V. Storozhuk and L. I. Lifar, Russ. J. Coord. Chem., 2001, 27, 783–785 CrossRef CAS.
  30. C. K. Zhou, H. R. Lin, Y. Tian, Z. Yuan, R. Clark, B. H. Chen, L. J. van de Burgt, J. C. Wang, Y. Zhou, K. Hanson, Q. J. Meisner, J. Neu, T. Besara, T. Siegrist, E. Lambers, P. Djurovich and B. W. Ma, Chem. Sci., 2018, 9, 586–593 RSC.
  31. C. K. Zhou, Y. Tian, M. C. Wang, A. Rose, T. Besara, N. K. Doyle, Z. Yuan, J. C. Wang, R. Clark, Y. Y. Hu, T. Siegrist and S. C. Lin, Angew. Chem., Int. Ed., 2017, 56, 9018–9022 CrossRef CAS PubMed.
  32. A. Khan, A. Zeb, L. N. Li, W. C. Zhang, Z. H. Sun, Y. Y. Wang and J. H. Luo, J. Mater. Chem. C, 2018, 6, 2801–2805 RSC.
  33. O. Toma, M. Allain, F. Meinardi, A. Forni, C. Botta and N. Mercier, Angew. Chem., Int. Ed., 2016, 55, 7998–8002 CrossRef CAS PubMed.
  34. N. N. Shen, M. L. Cai, Y. Song, Z. P. Wang, F. Q. Huang, J. R. Li and X. Y. Huang, Inorg. Chem., 2018, 57, 5282–5291 CrossRef CAS PubMed.
  35. H. R. Lin, C. K. Chen, Y. Tian, T. Siegrist and B. W. Ma, ACS Energy Lett., 2018, 3, 54–62 CrossRef CAS.
  36. Z. P. Wang, J. Y. Wang, J. R. Li, M. L. Feng, G. D. Zou and X. Y. Huang, Chem. Commun., 2015, 51, 3094–3097 RSC.
  37. Z. P. Wang, Z. Z. Zhang, L. Q. Tao, N. N. Shen, B. Hu, L. K. Gong, J. R. Li, X. P. Chen and X. Y. Huang, Angew. Chem., Int. Ed., 2019, 58, 9974–9978 CrossRef CAS PubMed.
  38. O. Toma, N. Mercier, M. Allain, A. Forni, F. Meinardi and C. Botta, Dalton Trans., 2015, 44, 14589–14593 RSC.
  39. N. N. Shen, J. R. Li, Z. F. Wu, B. Hu, C. C. Cheng, Z. P. Wang, L. K. Gong and X. Y. Huang, Chem. – Eur. J., 2017, 23, 15795–15804 CrossRef CAS PubMed.
  40. N. N. Shen, J. Li, G. Li, B. Hu, J. R. Li, T. Q. Liu, L. K. Gong, F. Q. Huang, Z. P. Wang and X. Y. Huang, Inorg. Chem., 2019, 58, 8079–8085 CrossRef PubMed.
  41. S. M. Parke and E. Rivard, Isr. J. Chem., 2018, 58, 915–926 CrossRef CAS.
  42. O. Toma, N. Mercier and C. Botta, J. Mater. Chem. C, 2016, 4, 5940–5944 RSC.
  43. R. A. Walton, Spectrochim. Acta, Part A, 1968, 24, 1527–1530 CrossRef CAS.
  44. L. M. Wu, X. T. Wu and L. Chen, Coord. Chem. Rev., 2009, 253, 2787–2804 CrossRef CAS.
  45. S. A. Adonin, M. N. Sokolov and V. P. Fedin, Coord. Chem. Rev., 2016, 312, 1–21 CrossRef CAS.
  46. H. Kunkely, A. Paukner and A. Vogler, Polyhedron, 1989, 8, 2937–2939 CrossRef CAS.
  47. H. Nikol and A. Vogler, J. Am. Chem. Soc., 1991, 113, 8988–8990 CrossRef CAS.
  48. C. K. Zhou, M. Worku, J. Neu, H. R. Lin, Y. Tian, S. J. Lee, Y. Zhou, D. Han, S. Y. Chen, A. Hao, P. I. Djurovich, T. Siegrist, M. H. Du and B. W. Ma, Chem. Mater., 2018, 30, 2374–2378 CrossRef CAS.
  49. T. V. Sedakova and A. G. Mirochnik, Opt. Spectrosc., 2016, 120, 268–273 CrossRef CAS.
  50. O. Toma, N. Mercier and C. Botta, Eur. J. Inorg. Chem., 2013, 1113–1117 CrossRef CAS.
  51. O. Toma, N. Mercier, M. Allain and C. Botta, CrystEngComm, 2013, 15, 8565–8571 RSC.
  52. T. Torimoto, T. Tsuda, K. Okazaki and S. Kuwabata, Adv. Mater., 2010, 22, 1196–1221 CrossRef CAS PubMed.
  53. P. Wasserscheid and W. Keim, Angew. Chem., Int. Ed., 2000, 39, 3772–3789 CrossRef CAS PubMed.
  54. S. A. Forsyth, J. M. Pringle and D. R. MacFarlane, Aust. J. Chem., 2004, 57, 113–119 CrossRef CAS.
  55. J. J. H. Davis, Chem. Lett., 2004, 33, 1072–1077 CrossRef CAS.
  56. H. R. Li, H. F. Shao, Y. G. Wang, D. S. Qin, B. Y. Liu, W. J. Zhang and W. D. Yan, Chem. Commun., 2008, 5209–5211 RSC.
  57. J. D. Holbrey, W. M. Reichert, M. Nieuwenhuyzen, S. Johnston, K. R. Seddon and R. D. Rogers, Chem. Commun., 2003, 1636–1637 RSC.
  58. R. Ozawa, S. Hayashi, S. Saha, A. Kobayashi and H. Hamaguchi, Chem. Lett., 2003, 32, 948–949 CrossRef CAS.
  59. S. Hayashi, R. Ozawa and H. Hamaguchi, Chem. Lett., 2003, 32, 498–499 CrossRef CAS.
  60. G. Laus, G. Bentivoglio, V. Kahlenberg, K. Wurst, G. Nauer, H. Schottenberger, M. Tanaka and H. U. Siehl, Cryst. Growth Des., 2012, 12, 1838–1846 CrossRef CAS.
  61. S. A. Adonin, M. N. Sokolov, M. E. Rakhmanova, A. I. Smolentsev, I. V. Korolkov, S. G. Kozlova and V. P. Fedin, Inorg. Chem. Commun., 2015, 54, 89–91 CrossRef CAS.

This journal is © The Royal Society of Chemistry 2020