Bromide anion-triggered visible responsive metallogels based on squaramide complexes

Abstract

Driven by the cooperation of metal coordination and multiple hydrogen bonds, a squaramide-based bisimidazole ligand forms structurally novel metallogels involving one-dimensional (1D) nanofibrils with linearly and infinitely extended Cl–Cu or Br–Cu bonds wrapped up with organic ligands. The resulting soft material enables us to discriminate bromides from other salts and, of special significance, from other halides both in aqueous solutions and in powder forms by the naked-eye. Raman spectroscopy clearly unravels the discrimination mechanism associated with a key step of anion ligand exchange, which has witnessed the exceptional ability of Raman spectroscopy in the investigation of metallogels.

---

Introduction

Coordination polymer gels, also known as metallogels, have drawn extensive attention as a class of functional soft materials owing to their fascinating properties including metal-related photophysical, magnetic, catalytic, and redox properties as well as self-healing ability and host–guest interactions.¹ Many interesting metallogels have been reported,^2–4 including the gels driven by the coordination between metals and bisimidazole ligands L1–3 developed by us (Scheme 1).³ As a ligand to construct metallogels, the squaramide building block, a structural analogue of urea,^2,5 has the following advantages.⁶ Firstly, squaramide has equal numbers of H-bond donors (N–H) and acceptors (C=O) and thus excels in the formation of multiple and self-complementary H-bonds. Secondly, the squaramides can pack with each other through π–π interactions between cyclobutenedione rings due to the planarity and aromaticity. More attractively, the two types of non-covalent interactions are fixed perpendicularly to each other by the square geometry of the cyclobutenedione ring, which would be useful to build anisotropic 3D architectures. However, despite extensive applications in catalytic, biological and medicinal realms, the squaramide derivatives remain underrepresented in metallogels. In this work, by combining a squaramide motif with two imidazole units, we assemble the ligand L4 that could have three types of intermolecular interaction capabilities including H-bonding, π–π stacking and metal coordination (Scheme 1).

Scheme 1 Molecular structures of ligands L1–5.

Experimental

General remarks

¹H and ¹³C NMR spectra were recorded by using a Bruker AV-400 or a Varian Inova spectrometer. The chemical shifts were calibrated using CDCl₃, TMS or DMSO-d₆ as the internal reference (for ¹H, CDCl₃: δ = 7.26 ppm; DMSO-d₆: δ = 2.50 ppm; for ¹³C CDCl₃: δ = 77.16 ppm; DMSO-d₆: δ = 39.52 ppm). The powder X-ray diffraction (PXRD) was performed on a Shimadzu XRD-6100 diffractometer using Cu K_α radiation (where λ is 1.5406 Å) operated at 30 mA and 40 kV. The samples were mounted onto a zero background sample holder and the surface of the powders was leveled using a glass slide. A scan range of 2θ between 5° and 35°, a scan speed of 10° min⁻¹ and a sampling width of 0.02° were employed. The single crystals were easy to lose the crystallized solvent molecules, and thus were covered by Paratone-N oil during data collection. A New Gemini (Dual, Cu at zero) EosS2 diffractometer was used to collect single crystal data at 293(2) K. The structure was solved by Olex2 ⁷ with the Superflip program⁸ and refined with the ShelXL⁹ package (least squares minimization). Due to the unresolvable disorder, residual electron densities possibly corresponding to the solvent molecules and the counterions of the framework were squeezed out.

Transmission electron microscopy (TEM) was performed by using a Hitachi H-600, operating at 75 kV. The metallogel was dispersed in EtOH. A drop of this dispersion was placed on a carbon-coated copper grid, stained negatively with aqueous phosphotungstic acid, dried at room temperature and then subjected to TEM observation. A JSM-5900LV scanning electron microscope (SEM) was employed for imaging at 5.0 kV. The xerogel sample for SEM was prepared as follows. The gel was frozen in liquid nitrogen for ca. 10 min and then evaporated under vacuum at −55 °C for 20 h. The obtained xerogel was shielded with gold and then examined. Absorption spectra were not available using the gel sample either sandwiched between quartz plates or scattered in solvents. Diffuse reflectance UV-vis spectroscopy (DR-UV-vis) was performed by using a UV2100 using gel samples sandwiched between two quartz plates. Raman scattering spectra were recorded on a gel sample by using a Horiba HR800 spectrometer with a 532 nm argon ion laser as the exciting light. Elemental analyses were recorded on an EA Flash 1112 analyzer. The inductively coupled plasma-atomic emission spectroscopy (ICP-AES) measurements were performed on an IRIS advantage ER/S.

Chemicals

Unless otherwise noted, reagents were purchased and used without further purification. Diethyl squarate was prepared according to the literature.¹⁰L4 and L5 were synthesized via the route presented in Scheme 2.

Scheme 2 Synthetic routes for L4 and L5.

2-(1H-Imidazol-1-yl)ethanamine. Sodium hydroxide (3.60 g, 90 mmol) was added into a solution of imidazole (2.04 g, 30 mmol) in THF (150 mL). The mixture was stirred vigorously for 30 min and then tetrabutylammonium bromide (TBAB, 0.48 g, 0.15 mmol) and 2-chloroethylamine hydrochloride (3.48 g, 30 mmol) were added. The reaction mixture was stirred at 110 °C for 20 h. After being cooled to ambient temperature, the obtained suspension was filtered. The filtrate was concentrated under reduced pressure and the resulting residue was purified by column chromatography on silica gel (CH₂Cl₂/MeOH = 8 : 1, v/v) to provide 2-(1H-imidazol-1-yl)ethanamine as a pale yellow oil (2.0 g, 60%). ¹H NMR (400 MHz, CDCl₃): δ = 1.22 (s, 2H), 3.06 (t, J = 6.0 Hz, 2H), 4.01 (t, J = 6.0 Hz, 2H), 6.97 (s, 1H), 7.09 (s, 1H), 7.53 (s, 1H) ppm. ¹³C NMR (100 MHz, CDCl₃): δ = 42.1, 49.5, 118.4, 128.7, 136.7 ppm.

3,4-Bis(2-(1H-imidazol-1-yl)ethylamino)cyclobut-3-ene-1,2-dione (L4). Diethyl squarate (1.00 g, 5.9 mmol) in EtOH (20 mL) was added into a solution of 2-(1H-imidazol-1-yl)ethanamine (1.11 g, 10 mmol) in EtOH (30 mL) under an N₂ atmosphere. The solution was stirred at 30 °C for 26 h. The obtained precipitate was filtered off to give the pure product L4 as a white solid (0.78 g, 52%). Further purification was performed by column chromatography on silica gel (CH₂Cl₂/MeOH = 5/1 to 3/1, v/v) if necessary. M.p.: 223–226 °C. ¹H NMR (400 MHz, DMSO-d₆): δ = 3.38 (s, 4H), 4.13 (s, 4H), 6.90 (s, 2H), 7.14 (s, 2H), 7.37 (br. s, 2H), 7.59 (s, 2H) ppm. ¹³C NMR (100 MHz, DMSO-d₆): δ = 44.0, 47.1, 119.5, 128.6, 137.6, 167.8, 182.8 ppm. HRMS (ESI⁺): calcd for C₁₄H₁₇N₆O₂ [M + H]⁺ 301.1413, found 301.1414.

1,3-Bis(2-(1H-imidazol-1-yl)ethyl)urea (L5). A solution of triphosgene (0.35 g, 1.17 mmol) in CH₂Cl₂ (10 mL) was added dropwise into a solution of 2-(1H-imidazol-1-yl)ethanamine (0.78 g, 7 mmol) and Et₃N (1 mL) in CH₂Cl₂ (20 mL) at 0 °C. The mixture was stirred at 30 °C for 36 h. After being concentrated under reduced pressure, the resulting residue was purified by column chromatography on silica gel (CH₂Cl₂/MeOH = 5/1 to 3/1, v/v) to provide 1,3-bis(2-(1H-imidazol-1-yl)ethyl)urea L5 as a white solid (0.27 g, 31%). M.p.: 52–54 °C. ¹H NMR (400 MHz, DMSO-d₆): δ = 3.30 (q, J = 6.0 Hz, 4H), 3.97 (t, J = 6.0 Hz, 4H), 6.06 (t, J = 5.6 Hz, 2H), 6.90 (s, 2H), 7.13 (s, 2H), 7.58 (s, 2H) ppm. ¹³C NMR (100 MHz, DMSO-d₆): δ = 40.4, 46.4, 119.5, 128.2, 137.4, 157.8 ppm. HRMS (ESI⁺): calcd for C₁₁H₁₇N₆O [M + H]⁺ 249.1464, found 249.1460.

Synthesis of the complex [Cu-L4₂]Cl₂. A MeOH/H₂O (v/v = 10 : 4, 5.0 mL) solution of L4 (0.40 mmol, 120.1 mg) was added into a MeOH/H₂O (v/v = 10 : 4, 5.0 mL) solution of CuCl₂·2H₂O (0.2 mmol, 34.1 mg) at 50 °C to afford a turbid mixture. The mixture was kept at the same temperature and a MeOH/H₂O (v/v = 10 : 4) solvent was added until the mixture became clear. After standing overnight at ambient temperature, the crystalline precipitation was filtered and dried under reduced pressure to obtain a green powder (101.6 mg, 69%). NMR spectra were not obtained due to the existence of the paramagnetic Cu(II) center. Elemental analysis and ICP analysis for C₂₈H₃₂Cl₂CuN₁₂O₄: Calc. C 45.75, H 4.39, N 22.87, Cu 8.65%. Found C 45.10, H 4.83, N 23.35, Cu 8.79%.

The complex [Cu-L4₂]Br₂ was obtained from CuBr₂ (0.20 mmol, 44.7 mg) and L4 (0.40 mmol, 120.1 mg) similar to the complex [Cu-L4₂]Cl₂ as a lavender crystalline powder (111.1 mg, 67%). NMR spectra were not obtained due to the existence of the paramagnetic Cu(II) center. Elemental analysis and ICP analysis for C₂₈H₃₂Br₂CuN₁₂O₄: Calc. C 40.81, H 3.91, N 20.40, Cu 7.71%. Found C 40.81, H 3.98, N 21.28, Cu 7.61%.

Single crystal X-ray diffraction (SXRD)

Single crystals of [Cu-L4₂]Cl₂ were isolated as the precipitate from a slowly cooled solution of L4 (0.2 mmol, 120.1 mg) and CuCl₂·2H₂O (0.2 mmol, 34.1 mg) in MeOH/H₂O (v/v = 10 : 4). Crystal data: orthorhombic, Pnma, a = 27.9361(13), b = 25.4180(14), c = 6.0524(3) Å, V = 4297.7(4) ų, Z = 4, T = 293.15 K, R(ref) = 0.0924(3221), wR₂(ref) = 0.2698(3922), CCDC 1402219.

Single crystals of [Cu-L4₂]Br₂ were isolated as the precipitate from a slowly cooled solution of L4 (0.40 mmol, 120.1 mg) and CuBr₂ (0.20 mmol, 44.7 mg) in MeOH/H₂O (v/v = 10 : 4). Crystal data: orthorhombic, Pcmn, a = 6.06932(11), b = 25.5373(6), c = 27.9983(6) Å, V = 4339.57(16) ų, Z = 4, T = 293.15 K, R(ref) = 0.0980(3374), wR₂(ref) = 0.3134(3967), CCDC 1402142.

Results and discussion

Gelation behavior

Gelation was initially observed after shaking the mixture of the methanol solutions (0.50 mL) of L4 (1.0 mL, 0.050 mmol) and CuCl₂·2H₂O with molar ratios of 1 : 0.6–1.4 at 50 °C. A green opaque gel (Gel-Cl) was yielded and no visible collapse was observed over several months at ambient temperature (Fig. S1 and S2†). In contrast, the urea derivative L5 gave rise to precipitation under the same conditions, possibly due to the absence of the π–π interactions between the planar and aromatic cyclobutenedione rings and/or the self-complementary H-bond donor and acceptor. L4 itself did not undergo gelation in various solvents including methanol, ethanol, water, dimethyl formamide (DMF), dimethyl sulphoxide (DMSO) and dimethylacetamide (DMAc), demonstrating that the coordination serves as one of the main driving forces of gelation. The testing of Cu(II) salts revealed that CuBr₂ formed a dark purple gel (Gel-Br), whereas CuSO₄·5H₂O, Cu(OAc)₂·H₂O, Cu(NO₃)₂·3H₂O, Cu(OOCCF₃)₂·H₂O and Cu(OTf)₂ could not provide any metallogels, suggesting that the gel formation is dependent on the nature of the anions. Gelation proceeded well in mixture solvents with methanol. A maximum ratio of 1 : 10 (v/v) of H₂O : MeOH still led to gelation (Table S1†). Gel-Cl and Gel-Br used for the following measurements were prepared by mixing L4 and CuCl₂·2H₂O or CuBr₂ (molar ratio 1 : 0.6) in MeOH, respectively. Gelators are 1.53 wt% for Gel-Cl and 1.68 wt% for Gel-Br, respectively.

Morphology of the metallogel

The morphology of Gel-Cl was next investigated by TEM and SEM (Fig. 1a and b). Tape-like fibers with widths of approximately 70–200 nm were observed in the TEM image. The SEM image of the xerogel depicts well-grown fibrillar networks with lengths of over ten micrometers. Interestingly, a majority of the fibers show parallel packing, illustrating an ordered assembly on a micrometer-scale.

Fig. 1 (a) TEM image of Gel-Cl, stained with phosphotungstic acid. (b) SEM image of the xerogel of Gel-Cl. (c) PXRD pattern of the xerogel of Gel-Cl and the simulated pattern from single crystal XRD data of [Cu-L4₂]Cl₂. Star-marked peaks are at 6.33°, 6.96° and 14.64°, respectively. The packing structure of [Cu-L4₂]Cl₂ in crystals viewed along (d) the b-axis, rotated by 20° along the c-axis and (e) the c-axis. The Cu : Cl molar ratio of 1 : 1 in the framework of single crystals suggests that there is another equivalent of Cl⁻ dissociated from the framework. Residual electron densities possibly corresponding to the counterions were squeezed out due to the unresolvable disorder. H atoms are partially omitted for clarity. For details, see Fig. S3 and S4.†

Gelation mechanism

To better understand the gelation mechanism, we tried to isolate the complex of L4 with CuCl₂·2H₂O under conditions similar to the gelating environment. Fortunately, the single crystals of [Cu-L4₂]Cl₂ were obtained in H₂O/MeOH (v/v = 4 : 10). The powder X-ray diffraction (PXRD) experiments were subsequently employed to investigate the freeze-dried Gel-Cl (xerogel) (Fig. 1c). Compared with the PXRD pattern simulated from single crystal XRD data, despite the broadening of the xerogel pattern due to less order and smaller particle size in the xerogel than in crystals,^2,11 2θ peaks such as those at 6.33°, 6.96° and 14.64° similarly appear in both the patterns, implying the resemblance of molecular packing in the gel and in crystals.

The gelation mechanism of Gel-Cl is next proposed. Two L4 molecules bend over their flexible alkyl chains to bidentately coordinate the Cu center to four imidazolyl motifs, forming a central symmetric, figure eight-shaped mononuclear complex (Cu-L4₂)²⁺. Four chelating nitrogen atoms and a cupric center are situated in the crystallographic ab plane with a typical and identical N–Cu distance of 2.015 Å.¹² Two Cl atoms axially coordinate to Cu with unusually long distances of 2.984 and 3.069 Å, forming an axially elongated octahedral configuration (Fig. 1d, e and S3, 4†).¹¹ By sharing apical Cl atoms, CuN₄Cl₂ octahedra line up in a row along the crystallographic c-axis. The Cl–Cu bonds linearly and infinitely extend to form a one dimensional (1D) Cl–Cu line wrapped up with organic ligand L4, which is rarely reported.¹² It could also be described as a Cl-bridged dicopper(II) complex with a short intermetallic distance of 6.053 Å, which may be appealing in the emerging fields of molecular spintronics and quantum computation.¹³ Vertically situated squaramides stretch their hydrogen donors (N–H) and acceptors (C=O) along the c-axis to form four equivalent H-bonds with a N⋯O distance of 2.759 Å. The cooperation of metal coordination and multiple H-bonds is robust enough to give rise to 1D-fibril formation. A number of 1D-fibrils (approximately 35–140) parallelly aggregate to form 2D-sheets through the π–π interaction of squaramides with a distance of 3.667 Å between the centroids (Fig. 1d). 2D-sheets then pack into a 3D-framework through van der Waals interactions, solidifying solvents to form a gel (Fig. 2). Single crystals of [Cu-L4₂]Br₂ were also successfully isolated for comparison. Resemblance is found in two single crystals with a similar intermetallic distance (6.069 Å in [Cu-L4₂]Br₂), 1D-fibrils, 2D-sheets and 3D-frameworks (Fig. 3 and S5–7†). Different from the exceptional length of the Cl–Cu coordination bond,¹² those of Br–Cu (3.010 and 3.059 Å) are consistent with the similar systems in the literature.¹⁴

Fig. 2 Schematic representation of the proposed gelation mechanism of Gel-Cl.

Fig. 3 The packing structure of [Cu-L4₂]Br₂ in crystals viewed along the b-axis, rotated by 20° along the a-axis (a) and the a-axis (b). The Cu : Br molar ratio of 1 : 1 in the framework of single crystals suggests that there is another equivalent of Br⁻ dissociated from the framework. Residual electron densities possibly corresponding to the counterions were squeezed out due to the unresolvable disorder. H atoms are partially omitted for clarity. For details, see Fig. S6 and S7.†

Bromide anion-triggered visible responses

Despite the environmental harmfulness, biological toxicity and wide distribution of bromide anions in aqueous solutions and in powder forms in the biosphere, a convenient method to discriminate bromide salts from other halides remains less developed.¹⁵ The challenge mainly lies in the chemical similarity of halides, which often show close chemical behaviors. Considering that two metallogels display structural resemblance and visually distinguishable color difference, we tested the feasibility of discriminating the Br⁻ anion. First, aqueous solutions of various sodium salts were investigated. Gladly, a pale purple color immediately appeared (within ca. 4 seconds) upon the addition of a NaBr solution into Gel-Cl, whereas other anions led to blue color (Fig. 4a and S8–11†). The purple color could remain over three months, even after further adding chloride or drying in air (Fig. S12 and 13†). The generality of the Br⁻ source is elucidated by adding a variety of bromide salts including inorganic, ammonium, phosphonium and imidazolium salts. To further expand the application scope, bromide powders were directly added into Gel-Cl. To our delight, the stimuli-responsive color change occurred and was even more easily observed than in solutions, which demonstrates an exceptional advantage of a wet and soft material. All these features make Gel-Cl a smart soft material for rapid and selective discrimination of bromides from other salts and, of special significance, from other halides.

Fig. 4 (a) Samples of Gel-Cl upon addition of aqueous sodium salts (2.4 mol L⁻¹, 70 μL), aqueous bromide salts (2.4 mol L⁻¹, 70 μL) and bromide powders (10.0 equiv.). TBAB: tetrabutylammonium bromide; BMIMB: 1-butyl-3-methylimidazolium bromide; PDIMB: 1-propyl-2,3-dimethylimidazolium bromide; MTPB: methyl triphenyl phosphonium bromide; BEAB: 2-bromoethyl-1-ammonium bromide; TEAB: tetra-ethylammonium bromide; TOAB: tetra-n-octylammonium bromide; HTMAB: hexadecyl trimethyl ammonium bromide; DMIMB: 1-dodecyl-3-methylimidazolium bromide; EPIMB: 1-ethyl-3-(4-N,N-dimethylphenyl)imidazolium bromide; and EDTA²⁻: ethylene diamine tetraacetate. (b) Raman and (d) DR-UV-vis spectra of Gel-Br and Gel-Cl before and after adding Br⁻. Inserted in (d) is a photo of Gel-Br. (c) Schematic representation of anion ligand exchange. For experimental details, see Fig. S8.†

Mechanism for bromide anion-triggered responses

As illustrated in Fig. 1d, the crystal structure of [Cu-L4₂]Cl₂ shows the exceptionally long Cu–Cl bonds, suggesting a relatively weak Cu–Cl coordination.¹² Considering the structural similarity between [Cu-L4₂]Cl₂ and [Cu-L4₂]Br₂ and the rarely observed gel-to-gel transition of Gel-Cl after adding bromides, we envisage that the exogenous bromide anion could enter into the framework of a 1D-fibril and replace the Cl⁻ anion that axially coordinates to the cupric center. It is then necessary to find evidence to confirm the coordination interaction between Br⁻ and the Cu center in the gel. However, despite significant progress, it remains a huge challenge to understand the structure of the gel at the molecular level mainly because common techniques that are often used to investigate bond connections in the solid state and in solution are not applicable for metallogel samples.⁴ For example, the NMR technique could be invalidated by paramagnetic metal centers, including often used Cu(II), Fe(III), Mn(II) and Co(II). Raman spectroscopy is a useful tool to distinguish the Cu–Br coordination bond from the electrostatic interaction between Cu²⁺ and Br⁻ as well as to detect the difference between Cu–Br and Cu–Cl coordination bonds.¹⁶ Thus, Raman spectroscopy was employed to gain insight into the discrimination mechanism. The sample of Gel-Cl + Br⁻ for the Raman test was prepared by mixing solid NaBr (10 equiv. relative to Cl species in Gel-Cl) with Gel-Cl and aging overnight. As depicted in Fig. 4b, the addition of Br⁻ into Gel-Cl leads to the disappearance of the peak at 227 cm⁻¹ corresponding to the vibration of the Cu–Cl coordination bond and meanwhile causes a new peak at 172 cm⁻¹ corresponding to the vibration of the Cu–Br coordination bond, which is similar to the pattern of Gel-Br.¹² These observations suggest that the bromide-triggered color switching of Gel-Cl could mainly be attributed to the Cl⁻ ligand exchange with Br⁻ rather than the extra electrostatic interaction between dissociated bromide ions and frameworks (Fig. 4c).

The diffuse reflectance UV-visible (DR-UV-vis) test was performed upon Gel-Cl samples overnight after adding increasing equivalents of solid NaBr. As shown in Fig. 4d, the addition of bromide into Gel-Cl gives rise to a substantial red-shift of absorption band edges from 435 to 560 nm, very close to that of Gel-Br. Based on the mechanism of anion ligand exchange, we propose that color switching could originate from either the changed d–d transition energies ascribed to the modification of the coordination environment of the cupric center or the ligand-to-metal charge transfer (LMCT) from different ligands (Cl⁻ or Br⁻). Density functional theory (DFT) calculations were thus performed to show that the energy gaps between the HOMO and LUMO reduce dramatically when the ligand changes from Cl⁻ to Br⁻, explaining the color change (Fig. S14 and S15†).

Conclusions

In conclusion, structurally novel metallogels are constructed by the assembly of squaramide-based bisimidazole ligand with copper(II) halides. The gelation mechanism involves the formation of one-dimensional (1D) nanofibrils with linearly and infinitely extended Cl–Cu or Br–Cu bonds wrapped up with organic ligands. Such 1D linear assemblies centered with a Cu(II)-containing line are rarely reported, especially in metallogels. Gel-Cl is selectively responsive to bromides to display purple color, which proceeds in a way of rarely reported gel-to-gel transition. The stability in the gel-to-gel transition is dependent on the structural similarity between the two frameworks. Based on this property, we develop a soft material for convenient and selective discrimination of bromide anions from other anions, especially for other halides. Raman spectroscopy offers insights into the recognition mechanism with a key step of anion ligand exchange. We believe that Raman spectroscopy would be an ideal tool to better understand the gelation mechanism.

Acknowledgements

This work was supported by grants from the National NSF of China (no. 21432005 and 21321061). We acknowledge the Comprehensive Training Platform of Specialized Laboratory, College of Chemistry, Sichuan University for measurements.

Notes and references

1. For selected reviews, see: N. M. Sangeetha and U. Maitra, Chem. Soc. Rev., 2005, 34, 821; F. Fages, Angew. Chem., Int. Ed., 2006, 45, 1680; M.-O. M. Piepenbrock, G. O. Lloyd, N. Clarke and J. W. Steed, Chem. Rev., 2010, 110, 1960; G. R. Whittell, M. D. Hager, U. S. Schubert and I. Manners, Nat. Mater., 2011, 10, 176; A. Y.-Y. Tam and V. W.-W. Yam, Chem. Soc. Rev., 2013, 42, 1540; J. Zhang and C.-Y. Su, Coord. Chem. Rev., 2013, 257, 1373; J. H. Jung, J. H. Lee, J. R. Silverman and G. John, Chem. Soc. Rev., 2013, 42, 924; T. Tu, W. Fang and Z. Sun, Adv. Mater., 2013, 25, 5304; S. Dong, B. Zheng, F. Wang and F. Huang, Acc. Chem. Res., 2014, 47, 1982.
2. For selected samples of metallogels, see: J. B. Beck and S. J. Rowan, J. Am. Chem. Soc., 2003, 125, 13922; M. Vondrova, T. M. McQueen, C. M. Burgess, D. M. Ho and A. B. Bocarsly, J. Am. Chem. Soc., 2008, 130, 5563; A. Y.-Y. Tam, K. M.-C. Wong and V. W.-W. Yam, J. Am. Chem. Soc., 2009, 131, 6253; X. Chen, Z. Huang, S.-Y. Chen, K. Li, X.-Q. Yu and L. Pu, J. Am. Chem. Soc., 2010, 132, 7297; H. Lee, S. H. Jung, W. S. Han, J. H. Moon, S. Kang, J. Y. Lee, J. H. Jung and S. Shinkai, Chem. – Eur. J., 2011, 17, 2823; T. Tu, W. Fang, X. Bao, X. Li and K. H. Dötz, Angew. Chem., Int. Ed., 2011, 50, 6601; L. Meazza, J. A. Foster, K. Fucke, P. Metrangolo, G. Resnati and J. W. Steed, Nat. Chem., 2013, 5, 42; L. Li, S. Xiang, S. Cao, J. Zhang, G. Ouyang, L. Chen and C. Su, Nat. Commun., 2013, 4, 1774; X. Yan, T. R. Cook, J. B. Pollock, P. Wei, Y. Zhang, Y. Yu, F. Huang and P. J. Stang, J. Am. Chem. Soc., 2014, 136, 4460; W. Fang, X. Liu, Z. Lu and T. Tu, Chem. Commun., 2014, 50, 3313; X.-Q. Wang, W. Wang, G.-Q. Yin, Y.-X. Wang, C.-W. Zhang, J.-M. Shi, Y. Yu and H.-B. Yang, Chem. Commun., 2015, 51, 16813; A. V. Zhukhovitskiy, M. Zhong, E. G. Keeler, V. K. Michaelis, J. E. P. Sun, M. J. A. Hore, D. J. Pochan, R. G. Griffin, A. P. Willard and J. A. Johnson, Nat. Chem., 2015, 8, 33; S. C. Grindy, R. Learsch, D. Mozhdehi, J. Cheng, D. G. Barrett, Z. Guan, P. B. Messersmith and N. Holten-Andersen, Nat. Mater., 2015, 14, 1210; M. Xue, Y. Lü, Q. Sun, K. Liu, Z. Liu and P. Sun, Cryst. Growth Des., 2015, 15, 5360; M. Martínez-Calvo, O. Kotova, M. E. Möbius, A. P. Bell, T. McCabe, J. J. Boland and T. Gunnlaugsson, J. Am. Chem. Soc., 2015, 137, 1983; P. Chen, Q. Li, S. Grindy and N. Holten-Andersen, J. Am. Chem. Soc., 2015, 137, 11590; J. Sun, Y. Liu, L. Jin, T. Chen and B. Yin, Chem. Commun., 2016, 52, 768.
3. S. Zhang, J. Lan, Z. Mao, R. Xie and J. You, Cryst. Growth Des., 2008, 8, 3134; S. Zhang, S. Yang, J. Lan, S. Yang and J. You, Chem. Commun., 2008, 6170; S. Zhang, S. Yang, J. Lan, Y. Tang, Y. Xue and J. You, J. Am. Chem. Soc., 2009, 131, 1689; L. Yang, L. Luo, S. Zhang, X. Su, J. Lan, C.-T. Chen and J. You, Chem. Commun., 2010, 46, 3938.
4. J. Brandt, K. K. Oehlenschlaeger, F. G. Schmidt, C. Barner-Kowollik and A. Lederer, Adv. Mater., 2014, 26, 5758.
5. J. W. Steed, Chem. Soc. Rev., 2010, 39, 3686.
6. R. I. Storer, C. Aciro and L. H. Jones, Chem. Soc. Rev., 2011, 40, 2330; J. Alemán, A. Parra, H. Jiang and K. A. Jørgensen, Chem.−Eur. J., 2011, 17, 6890.
7. O. V. Dolomanov, L. J. Bourhis, R. J. Gildea, J. A. K. Howard and H. Puschmann, J. Appl. Crystallogr., 2009, 42, 339.
8. L. Palatinus and G. Chapuis, J. Appl. Crystallogr., 2007, 40, 786; L. Palatinus and A. van der Lee, J. Appl. Crystallogr., 2008, 41, 975; L. Palatinus, S. J. Prathapa and S. van Smaalen, J. Appl. Crystallogr., 2012, 45, 575.
9. G. M. Sheldrick, Acta Crystallogr., Sect. A: Fundam. Crystallogr., 2008, 64, 112.
10. H.-B. Zhou, J. Zhang, S.-M. Lü, R.-G. Xie, Z.-Y. Zhou, M. C. K. Choi, A. S. C. Chan and T.-K. Yang, Tetrahedron, 2001, 57, 9325.
11. M.-O. M. Piepenbrock, N. Clarke and J. W. Steed, Langmuir, 2009, 25, 8451.
12. Cu–Cl bond lengths are normally about 2.3–2.9 Å in CuN₄Cl₂ octahedra. See: M. Lagrenee, S. Sueur, J. P. Wignacourt, B. Mernari and A. Boukhari, J. Chim. Phys. Phys.-Chim. Biol., 1991, 88, 2075; P. Comba, S. M. Luther, O. Maas, H. Pritzkow and A. Vielfort, Inorg. Chem., 2001, 40, 2335; S. Banerjee and P. Dastidar, CrystEngComm, 2013, 15, 9415; R. Atencio, K. Ramirez, J. A. Reyes, T. Gonzalez and P. Silva, Inorg. Chim. Acta, 2005, 358, 520.
13. J. Gómez-Herrero and F. Zamora, Adv. Mater., 2011, 23, 5311; M. Castellano, R. Ruiz-García, J. Cano, J. Ferrando-Soria, E. Pardo, F. R. Fortea-Pérez, S.-E. Stiriba, M. Julve and F. Lloret, Acc. Chem. Res., 2015, 48, 510.
14. Cu–Br bond lengths are normally about 2.8–3.2 Å in CuN₄Br₂ octahedra. See: C. Nather, M. Wriedt and I. Jess, Acta Crystallogr., Sect. E: Struct. Rep. Online, 2002, 58, m107; S. Bieller, A. Haghiri, M. Bolte, J. W. Bats, M. Wagner and H.-W. Lerner, Inorg. Chim. Acta, 2006, 359, 1559; V. Kubicek, I. Rehor, J. Havlickova, J. Kotek, I. Cisarova, P. Hermann and I. Lukes, Eur. J. Inorg. Chem., 2007, 3881.
15. F. Qu, N. B. Li and H. Q. Luo, Anal. Chem., 2012, 84, 10373; L. Xu, Y. Xu, W. Zhu, Z. Xu, M. Chen and X. Qian, New J. Chem., 2012, 36, 1435.
16. L. V. Stepakova, M. Y. Skripkin, L. V. Chernykh, G. L. Starova, L. Hajba, J. Mink and M. J. Sandström, Raman Spectrosc., 2008, 39, 16; W. J. Gee and S. R. Batten, Chem. Commun., 2012, 48, 4830.

---

Footnote

† Electronic supplementary information (ESI) available. CCDC 1402219 and 1402142. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6qi00386a

---