Perylene-Based Coordination Polymers: Synthesis, Fluorescent J-Aggregates, and Electrochemical Properties

The incorporation of electroactive organic building blocks into coordination polymers (CPs) and metal–organic frameworks (MOFs) offers a promising approach for adding electronic functionalities such as redox activity, electrical conductivity, and luminescence to these materials. The incorporation of perylene moieties into CPs is, in particular, of great interest due to its potential to introduce both luminescence and redox properties. Herein, we present an innovative synthesis method for producing a family of highly crystalline and stable coordination polymers based on perylene-3,4,9,10-tetracarboxylate (PTC) and various transition metals (TMs = Co, Ni, and Zn) with an isostructural framework. The crystal structure of the PTC-TM CPs, obtained through powder X-ray diffraction and Rietveld refinement, provides valuable insights into the composition and organization of the building blocks within the CP. The perylene moieties are arranged in a herringbone pattern, with short distances between adjacent ligands, which contributes to the dense and highly organized framework of the material. The photophysical properties of PTC-Zn were thoroughly studied, revealing the presence of J-aggregation-based and monomer-like emission bands. These bands were experimentally identified, and their behavior was further understood through the use of quantum-chemical calculations. Solid-state cyclic voltammetry experiments on PTC-TMs showed that the perylene redox properties are maintained within the CP framework. This study presents a simple and effective approach for synthesizing highly stable and crystalline perylene-based CPs with tunable optical and electrochemical properties in the solid state.


General methods and materials
All reagents and solvents employed in the syntheses were of high purity grade and were purchased from Sigma-Aldrich Co. and TCI. 1 H liquid-state NMR spectra were recorded on a Bruker AVANCE 300 spectrometer (300 MHz). Dimethylsulfoxide-d6 (DMSO-d6) was used as solvent. Tetramethylsilane (TMS) was used as internal reference. Chemical shifts (δ) are quoted in ppm from TMS and the coupling constants (J) are given in Hz. 1 H and 13 C solid-state NMR spectra were recorded on a 9.4 T Bruker Avance III 400 spectrometer using a 4 mm Bruker magic-angle spinning (MAS) probe. Chemical shifts are quoted in ppm from TMS using as secondary references solid adamantane. Infrared spectra were recorded in an ATR FT-IR GALAXY SERIES FT-IR 7000 (Mattson Instruments) spectrometer in the 4000-400 cm -1 range using powdered samples. Raman spectra were recorded in a RFS 100/S (Bruker) spectrometer equipped with Nd:YAG laser (1064 nm). Thermogravimetric analysis (TGA) was carried out with a Shimadzu TGA 50 equipment in the 25-600 ºC temperature range under a 5 ºC min -1 scan rate and a N2 flow of 20 mL·min -1 . Powder X-ray diffraction patterns were recorded using an Empyrean PANalytical diffractometer (Cu Kα1,2 X-radiation, λ1 = 1.540598 Å; λ2 = 1.544426 Å), equipped with an PIXcel 1D detector and a flat-plate sample holder in a Bragg-Brentano para-focusing optics configuration (45 kV, 40 mA).
The UV-vis absorption spectra were measured with a Jasco UV-660 Spectrophotometer (Jasco International, Tokyo). Fluorescence spectra were recorded with a Fluorolog 3-22 Spectrofluorimeter (Horiba Jobin Yvon, USA) with a 450 W xenon lamp. The fluorescence decay curves were recorded by the single-photon timing technique, by excitation at 285nm and 570 nm, using the vertical polarized light of the second harmonic of a Coherent Radiation Dye laser 700 series (laser dye DCM, 610-680 nm, 130 mW, 5 ps, 4 MHz). The emission was collected at the magic angle using a Jobin Yvon HR320 monochromator (Horiba Jovin Ivon Inc.). The instrument response functions (35-80 ps FWHM) were generated by the scattering of colloidal silica water dispersions. Decay curves were stored in 1024 channels with an accumulation of at least 20k counts in the peak channel. The photoluminescence decay curves were fitted by a non-linear least-squares reconvolution method using the TRFA DP software by SSTC (Scientific Software Technologies Center, Belarusian State University, Minsk, Belarus). 1 S3

Powder X-ray Diffraction and Rietveld refinement
Powder X-Ray Diffraction (PXRD) data for all compounds were collected at ambient temperature on an Empyrean PANalytical diffractometer, with a working wavelength of λ1 = 1.540598 Å and λ2 = 1.544426 Å (Cu Kα1,2 X-radiation), equipped with a PIXcel 1D detector, a capillary sample holder, and an Incident beam PreFIX module with elliptical X-ray mirror for Cu radiation (45 kV, 40 mA). Intensity data were collected by the step-counting method, in continuous mode, in the ca. 4.0º ≤ 2θ ≤ 50º range.
A fine powdered sample was placed inside a Hilgenberg borosilicate glass capillary (ca. 1.0 mm of diameter) which was spun during data collection to improve powder averaging over the individual crystallites, ultimately removing eventual textural effects such as preferential orientation.
The collected powder X-ray diffraction patterns were indexed using the LSI-Index algorithm implemented in TOPAS-Academic V5, 2-3 and a whole-powder-pattern Pawley fit permitted to unequivocally confirm the orthorhombic Pbam space group as the most suitable for the compound, in good agreement with the reported crystal structure for isotypical materials. 4 The crystal structures were also determined in TOPAS-Academic V5 2 by using a simulated annealing approach. In a first stage of the crystal solution, the metallic centers and the coordinated water molecules were allowed to converge to their optimal positions within the unit cell while using a battery of anti-bump restraints to ensure chemically reasonable coordination geometries. Despite the good quality of the collected patterns, the location of the atoms composing the crystallographically independent organic component proved to be of extreme difficulty, even when using distance restraints just like those employed for the inorganic backbone. The derivation of the most suitable location for the organic linker was performed in a second stage by using a Fenske-Hall Z-matrix for half of this chemical entity and treating the ligand as a rigid body inside the unit cell. We note that this strategy greatly facilitates the mobility of this chemical entity inside the unit cell boundaries during the global optimization processes. It does not, however, take into account the conformational flexibility associated with the mutual rotations of pendant moieties. This was taken into consideration by comparing the structure with the data previously published.
A Rietveld structural refinement 5 was performed with TOPAS-Academic V5 2 using either a Chebychev polynomial throughout the entire angular range to model the background contribution, or fixed background points. The peak shapes for the powder patterns were S8 described using the fundamental parameters approach, 6 with preferential orientation effects being modelled using a 4 th order spherical harmonics approach.

UV/visible absorption and diffuse reflectance:
The UV-vis-NIR absorption and diffuse reflectance spectra of the samples were measured using a Lambda 950 dual-beam spectrometer (PerkinElmer) and Reflectance FLEX Pack (Sarspec). The diffuse reflectance spectra are reported as the Kubelka-Munk transform, where F(R) = (1−R) 2 /2R. The direct optical band gaps of these materials were determined from respective Tauc plots.

Theoretical calculations
Molecular calculations of H4PTCA in its neutral form in gas phase were performed under the density functional theory (DFT) framework using the Gaussian-16.A03 suite of programs. 7 Minimum-energy structure and electronic structure calculations were obtained at the PBE0/6-31G(d,p) level of theory. 8 Quantum-chemical calculations in periodic boundary conditions were performed within the density functional theory (DFT) framework as implemented in the all-electron full-potential FHI-AIMS electronic structure code package. 9 The minimum-energy structures of PTC-TM CP materials were obtained, starting from the experimental X-ray data, upon full lattice and ionic relaxation using the GGA-type PBEsol functional 10 and the numeric atom-centered orbital light tier-1 basis set. Dispersion forces were included by means of the vdW Hirshfeld correction as described by Tkatchenko and Scheffler. 11 The electronic band structure and density of states (DOS) were calculated using the hybrid HSE06 functional. 12 A full k-path in the Pbam first Brillouin zone of Γ-X-S-Y-Γ-Z-U-R-T-Z-Γ and a 3×3×3 k-grid were employed. Effective masses for hole and electron along the k-path segments in the valence band maximum (VBM) and conduction band minimum (CBM) were calculated according to the parabolic approximation by using the aims_effect_mass.py utility included in FHI-AIMS. The absorption spectra of the PTC-TM CPs were computed under the dielectric function approximation as implemented in FHI-AIMS, using the hybrid HSE06 functional and the light tier-1 basis set.
Time-Dependent DFT (TD-DFT) calculations 13 were performed at the PBE0/6-31G(d,p) level of theory for the lowest-lying excited states of H4PTCA monomer and representative dimers, as extracted from the crystal structures, in gas phase by using the Gaussian-16A.03 software. 7 The minimum-energy optimized geometry of the low-lying bright state was obtained for H4PTCA, and its harmonic frequencies were calculated at the TD-DFT/PBE0/6-31G(d,p) level of theory. Vibrational resolution of the H4PTCA S0→S1 electronic transition was calculated by means of the Franck-Condon principle as implemented in Gaussian-16.A03. To simulate the charge-transfer state and compute the excimer-like band energy, the internal structure of the PTC ligands in the dimers was replaced by the minimum-energy structure of a PTC cation and a PTC anion, maintaining the intermolecular disposition unchanged.
The excitonic coupling between the PTC ligands in the MOF was estimated for the three representative dimers (A, B, and C, Figure 7 in the main text) extracted from the minimumenergy crystal structure of PTC-Zn, PTC-Ni, and PTC-Co, and using the electronic energy transfer (EET) analysis as coded in  The crystalline geometries, spin densities, and frontier crystal orbitals were displayed using the software VESTA. 14 The molecular orbital topologies were plotted by means of the Chemcraft software. 15 Table S1. Relative energy of the different spin configurations optimized at the PBEsol/light tier1 level and single-point calculated at the HSE06/light tier 1 level for the PTC-TM materials. FM, AFM, LS, and HS stem from ferromagnetic, antiferromagnetic, low spin, and high spin, respectively. Two AFM configurations are calculated in each case: config. 1 with the closest TM atoms arranged in an antiparallel configuration, and config. 2 with the farthest TM atoms arranged in antiparallel.        Figure S23. Vibrational resolution of the lowest-lying singlet excited state S1 calculated for the H4PTCA ligand at the PBE0/6-31G(d,p) level of theory. Figure S24. Simulated absorption spectra calculated for PTC-TM crystals via the linear macroscopic dielectric function approximation at the HSE06/light tier-1 level of theory. Figure S25. Monoexcitation that describes the charge transfer nature of the lowest-lying singlet excited state S1 in dimers A, and B, and C.

Cyclic voltammetry
Electrode preparation: The powdered materials (2 mg) were mixed in 2 mL of Nafion and ethanol (1:3). 100 µL were deposited on a 3 mm diameter glassy carbon disc working electrode, which was previously polished with 0.3, 0.1, and 0.05 µm alumina powders. Afterwards, the solvent was evaporated at room temperature.

Equipment:
The electrochemical experiments were performed using an Autolab electrochemical workstation (PGSTAT302N with FRA32M Module) connected to a personal computer that uses Nova 2.1 electrochemical software. A typical three-electrode experimental cell equipped with a platinum wire as the counter electrode and a silver wire as the pseudoreference electrode was used for the electrochemical characterization of the working electrodes. The electrochemical properties were studied measuring the cyclic voltammogram at different scan rates in previously N 2 purged 0.1 M TBAPF 6 /CH 3 CN solution. Ferrocene was added as an internal standard upon completion of each experiment. All potentials are reported in V versus Ag/AgCl. Figure S26. Cyclic voltammetry (CV) of H4PTCA ligand in DMF using TBAPF 6 0.1 M as electrolyte at 0.1 V/s scan rate. A platinum wire was used as the counter electrode and a silver wire as the pseudoreference electrode. Ferrocene was added as internal standard. All potentials are reported versus Ag/AgCl. Figure S27. Solid-state cyclic voltammetry (CV) of PTC-Co CP in CH 3 CN using TBAPF 6 0.1 M as electrolyte at different scan rates. A platinum wire was used as the counter electrode and a silver wire as the pseudoreference electrode. Ferrocene was added as internal standard. All potentials are reported versus Ag/AgCl. The inset shows the linear relationship of cathodic peak current vs. scan rate. Figure S28. Solid-state cyclic voltammetry (CV) of PTC-Ni CP in CH 3 CN using TBAPF 6 0.1 M as electrolyte at different scan rates. A platinum wire was used as the counter electrode and a silver wire as the pseudoreference electrode. Ferrocene was added as internal standard. All potentials are reported versus Ag/AgCl. The inset shows the linear relationship of cathodic peak current vs. scan rate. Figure S29. Solid-state cyclic voltammetry (CV) of PTC-Zn CP in CH 3 CN using TBAPF 6 0.1 M as electrolyte at different scan rates. A platinum wire was used as the counter electrode and a silver wire as the pseudoreference electrode. Ferrocene was added as internal standard. All potentials are reported versus Ag/AgCl. The inset shows the linear relationship of cathodic peak current vs. scan rate.