Synthesis, photophysical characterization and DFT studies on fluorine-free deep-blue emitting Pt(II) complexes

2.1 Synthesis and characterization

The ligand precursors L1, L2 and L3 are commercially available; the synthesis of L4 and of the intermediate E1 were carried out by adapting known procedures (Schemes 1 and 2) [66–68]. Compound L4 was prepared by heating 4-methoxypyridine-2-carbonitrile and sodium azide in a mixture of n-butyl alcohol and acetic acid. After addition of water, the product precipitated as a white powder and no further purification was necessary.

Scheme 2:

Preparation of the binuclear Pt(II) complex E1.

The binuclear Pt(II) complex E1 was synthesized by refluxing tris(2,4-di-tert-butylphenyl) phosphite and potassium tetrachloroplatinate (II) in 2-methoxyethanol, followed by a two-step purification procedure. Compound E1 is a mixture of cis and trans isomers (only the trans isomer is shown in Scheme 2 for clarity). The general synthesis of the complexes C1–C4 is described in Scheme 1. The corresponding chromophoric ligand precursors L1–L4 were reacted overnight with the binuclear complex E1 at 50°C in a 1:1 mixture of dichloromethane (DCM) and methanol. The crude product was then purified by column chromatography on silica gel with dichloromethane-methanol as eluent to give the desired pure compound.

2.2 Molecular structures in the crystal

To verify the structure of the two classes of complexes, the coordination geometries were studied by X-ray diffraction. C2 and C3 were dissolved in dichloromethane-methanol (1:1) and the solvent was slowly evaporated at –18°C, yielding crystals suitable for X-ray diffractometric analysis. The relevant crystal data can be found in Table 1. Figures 2 and 3 clearly indicate that the nitrogen atom of the pyridine from the luminophore [N1 (C2) or N5 (C3)] is coordinated trans to the cyclometallating carbon (C36) of the ancillary ligand, whereas the anionic moiety of the luminophore [O4 (C2); N1 (C3)] is coordinated trans to the phosphorus of the ancillary ligand (P1). The bite angle of the bidentate luminophoric and ancillary ligands cause a distorted square-planar coordination geometry for the structures of both complexes [C2: O4–Pt1–N1 78.8(1)°, C36–Pt1–P1 80.7(1)°; C3: N1–Pt1–N5 77.6(2)°, C36–Pt1–P1 81.2(2)°]. The Pt–C bond lengths [C2: 2.003(4) Å; C3: 2.015(7) Å] are significantly shorter than the Pt–N bonds [C2: 2.102(3) Å; C3: 2.134(6) Å], indicating a strong trans effect of the cyclometallating, in-plane coordinated phenyl ring due to its strong σ donor character. The same is valid for the second pair of coordinative bonds trans to each other. The carboxylic acid oxygen atom as well as the nitrogen atom of the tetrazole are much closer to the metallic center [C2: 2.070(3) Å; C3: 2.084(6) Å] than the phosphorus atom [C2: 2.142(1) Å; C3: 2.165(2) Å]. The observed metal-ligand bonds and angles are in good agreement with those described for related complexes [51].

Fig. 2:

Molecular structure of compound C2 in the crystal. Displacement ellipsoids are drawn at the 30% probability level, H atoms were omitted for clarity.

Fig. 3:

Molecular structure of compound C3 in the crystal. Displacement ellipsoids are drawn at the 30% probability level, H atoms were omitted for clarity.

2.3 Photophysical characterization

The absorption spectra of C1–C4 are shown in Fig. 4. Each complex shows two absorption bands, one centered at 225 nm and another band in the range between 320 and 350 nm. Additionally, all complexes show a shoulder around 275 nm.

Fig. 4:

Absorption spectra of complexes C1–C4 in dichloromethane at r. t.

The bands at high energies and the shoulder at 275 nm can be attributed to ligand-centered (LC) π-π* transitions of the phosphite ligand, whereas the band at 320–350 nm can be assigned to π-π* transitions arising from the chromophoric ligand. The latter is significantly red-shifted upon replacement of the carboxylic moiety by a tetrazole, as the size of the chromophoric system is increased. Notably, introduction of a OCH₃ substituent on the chromophoric ligand does not significantly change the absorption spectrum of C2 as compared to C1 which lets us assume that the OCH₃ group does not significantly affect the frontier orbital gaps for the NˆO luminophore. In contrast, the absorption of C4 compared to C3 is significantly blue shifted, which points to a shift in the HOMO-LUMO gap upon introduction of the OCH₃ moiety.

Emission at room temperature in degassed dichloromethane was not measureable for the complexes, pointing towards fast radiationless deactivation processes such as coupling with solvent molecules or thermal population of dark states. Emission spectra of C1–C4 in frozen glassy matrices of dichloromethane-methanol (1:1) at 77 K are shown in Fig. 5 along with the calculated vibrationally resolved emission spectra. In Table 2, the key photophysical properties of all complexes are shown. The measured spectra show broad, unstructured emission profiles, which can be attributed to excited states with charge-transfer character, whereas the microsecond-range lifetimes are indicative of a clear ³MP-LC character. The emission band of C1, which possesses an electron-poor carboxylate group on the small pyridine-based chromophoric ligand, appears blue-shifted as compared to that of C3 carrying a tetrazolate ring with extended delocalization. However, upon introduction of the OCH₃ substituent on C1 to yield C2, no significant shift is observed (Table 2). In contrast, introduction of the OCH₃ substituent to C3 yielding C4, leads to the bluest emission of all complexes, with a maximum peaking at 411 nm. This shows again that the influence of the OCH₃ groups is solely found in complexes with extended luminophores. On the other hand, the introduction of the OCH₃ substituent decreases Φ_L, which can be related to the blue-shift that brings the emissive state energetically closer to dark metal-centered states, which in turn become thermally more accessible.

Fig. 5:

Normalized emission spectra at 77 K (solid line) and vibrationally resolved calculated emission spectra of complexes C1–C4 at 77 K (dashed line).

All the complexes are weakly emissive in doped PMMA, as neat films and as powders, with the notable exception of C3 which shows an impressively high Φ_L in the amorphous solid state (CIE coordinates: 0.196, 0.251; Fig. 6). It is clear that the luminophore needs to have a significant delocalization, as provided by the tetrazolate unit, and that shorter emission wavelengths are associated with highly energetic excited states that favor the thermal population of dark metal-centered states.

Fig. 6:

Normalized emission spectrum of compound C3 (powder, at room temperature).

2.4 Density functional theory studies

We have determined the character of the vertical T₁ → S₀ transition by analyzing the eigenvector of the T₁ de-excitation at the TDDFT optimized T₁ geometry. For all complexes, the transition has an almost pure HOMO–LUMO character, with a weight of at least 81% in all cases, independently of the different ligands and their substituent. Figure 7 also shows that the shape of the HOMO and LUMO orbitals is practically invariant for C1–C4.

Fig. 7:

Visualization of the calculated HOMO and LUMO orbitals of complexes C1–C4 together with the respective orbital energies and HOMO-LUMO gaps.

The HOMO is associated with the cyclometallating phenyl group and the Pt(II) center, whereas the LUMO is mainly related to the luminophoric ligand. These findings are in agreement with the spectroscopically observed charge-transfer character of the emissive state (vide supra), which can now be confidently ascribed to an inter-ligand charge-transfer excitation mixed with a metal-to-ligand charge-transfer portion that explains the observed phosphorescence. Moreover, the calculations also show that the HOMO energy is only marginally affected by the changes on the luminophoric ligand, whereas the LUMO is destabilized upon replacement of tetrazolate moieties by carboxylate donors and upon insertion of OCH₃ groups, which in turn widens the HOMO-LUMO gap. The calculated blue-shift is larger for tetrazolate-bearing ligands, as already observed in the absorption and emission spectra (vide supra). Actually, no spectroscopic shift is experimentally observed when OCH₃ units are inserted on picolinate chelators, which points toward the fact that the emissive states cannot be simply described as monoelectronic excitations, even though the calculated trends are in good agreement with the experimental results. Finally, the lack of vibrational progression in the emission spectra at 77 K can be ascribed to the frozen solvent matrix, as opposed to the calculated spectra showing defined vibrational shoulders.

4.1 Synthesis and characterization

As starting materials tris(2,4-di-tert-butylphenyl) phosphite (98%; Sigma-Aldrich, Munich, Germany), K₂PtCl₄ (99.95%; ChemPur, Karlsruhe, Germany), and 4-methoxypyridine-2-carbonitrile (98%; abcr, Karlsruhe, Germany) were used as received. The luminophoric ligand precursors L1 (2-picolinic acid, 99%; Sigma-Aldrich, Munich, Germany), L2 (4-methoxy-pyridine-2-carboxylic acid Sigma-Aldrich, Munich, Germany) and L3 (5-(2-pyridyl)-1H-tetrazole, 98%; Alfa Aesar, Karlsruhe, Germany), were commercially available. Column chromatography was performed with silica gel 60 (particle size 35–70 μm, 230–400 mesh, Merck). NMR and mass spectra were measured at the Department of Organic Chemistry, University of Münster. NMR spectra were recorded on an Avance 400 Spectrometer from Bruker Analytische Messtechnik (Karlsruhe, Germany) for ligand precursor L4 and on an DD2 600 spectrometer from Agilent for the Pt(II) complexes. The ¹H NMR chemical shifts (δ) of the signals are given in parts per million (ppm) and referenced to residual protons in the deuterated solvents [D₆]DMSO (2.50 ppm) and [D₂]DCM (5.32 ppm). The ¹⁹⁵Pt NMR chemical shifts are referenced to Na₂PtCl₆, whereas the ³¹P NMR chemical shifts are referenced to phosphoric acid as an external standard. The signal multiplicities are abbreviated as follows: s, singlet; d, doublet; t, triplet; q, quartet; m, multiplet. All coupling constants (J) are given in Hertz. High resolution mass spectrometry (HRMS) was performed via electrospray ionization (ESI) on a Bruker Daltonics MicroTof spectrometer with loop injection.

4.2 Synthesis of 4-methoxy-2-(1H-tetrazol-5-yl)pyridine (L4)

4-Methoxypicolinonitrile (776 mg, 5.79 mmol) and NaN₃ (526 mg, 8.11 mmol) were suspended in n-butanol (14 mL) and acetic acid (1.37 mL, 23.15 mmol) was added dropwise. The mixture was stirred for 2 h at 80°C. After the mixture had cooled down, the same amount of acetic acid was added, followed by stirring at 80°C overnight. After cooling down, water was added and a white precipitate was formed, which was filtered off, washed with water and dried. The compound was again dissolved in ethyl acetate and absorbed onto silica gel by removing the solvent under reduced pressure. This was loaded onto a column packed with silica gel to perform a column chromatography separation with hexane-ethyl acetate as eluent to get the pure compound L4 as a white solid (748 mg, 73%). – ¹H NMR (400 MHz, [D₆]DMSO₆): δ = 8.59 (d, J = 5.8 Hz, 1H), 7.73 (d, J = 1.8 Hz, 1H), 7.21 (dd, J = 5.9, 1.6 Hz, 1H), 3.96 (s, 3H). – ¹³C NMR (101 MHz, [D₆]DMSO): δ = 166.58, 154.79, 151.08, 145.21, 112.21, 108.42, 55.99. – HRMS ((+)-ESI; MeOH): m/z = 178.07185 (calcd. 178.07234 for C₇H₈N₅O, [M+H]⁺), 201.05717 (calcd. 201.05644 for C₇H₇N₅ONa, [M+Na]⁺).

4.3 Synthesis of [{Pt-(μ-Cl){κ²-P,C-P(OC₆H₂-2,4-^tBu₂)(OC₆H₃-2,4-^tBu₂)₂}}₂] (E1)

Tris(2,4-di-tert-butylphenyl) phosphite (934 mg, 1.44 mmol) and K₂PtCl₄ (600 mg, 1.44 mmol) were combined in 2-methoxyethanol (48 mL) and heated to 140°C under nitrogen overnight. The clear solution was cooled to room temperature and the solvent was removed under reduced pressure. The off-white residue was dissolved in dichloromethane and filtered through a pad of celite. The filtered solution was then concentrated under reduced pressure and the product was precipitated from dichloromethane with ethanol to yield a white solid. The supernatant was removed and the solid dried under vacuum (1.27 g 50%). – HRMS ((+)-ESI; MeOH): m/z = 1791.72042 (calcd. 1791.72005 for C₈₄H₁₂₆Cl₂KO₆P₂Pt₂, [M+K]⁺).

4.4 General synthetic procedure for the preparation of complexes C1–C4

The corresponding ligand precursors L1–L4 (0.228 mmol) and NaOMe (0.228 mmol) were dissolved in dichloromethane-methanol (1:1, 25 mL) and stirred for 30 min. Next, the reaction mixture was degassed intensely before the Pt precursor E1 (0.114 mmol) was added. After the reaction mixture was stirred at 50°C for 12 h in a nitrogen atmosphere, it was cooled down. A small amount of silica gel was added into the reaction vessel and the crude product was absorbed onto silica gel by removing the solvent under reduced pressure. This was loaded onto a column packed with silica gel, and a column chromatographic separation was performed using dichloromethane-methanol as the eluent to give the final product C1–C4 as white solid.

4.5 X-ray diffractometry

The crystallographic data of C2 and C3 (Table 1) were collected on a rotating anode Nonius κ-CCD diffractometer with Montel mirror-monochromatized MoK_α radiation (λ = 0.71073 Å) by using an ω-φ scan method at 223(2) K. Programs used: data collection Collect [69]; data reduction Denzo-smn [70]; absorption corrections were applied using multi-scan techniques, Denzo [71]; the structures were solved by Direct Methods and refined by full-matrix least-squares techniques using the program Shelxl-97 [72]; graphics were done with Xp [73]. The non-hydrogen atoms of the compounds were refined anisotropically. The hydrogen atoms attached to carbon atoms were generated geometrically.

Exceptions and special features: Compound C2 contains in the asymmetric unit tree tert-butyl groups and two methanol molecules disordered over two positions. For compound C3 four tert-butyl groups were found to be disordered over two positions. Several restraints (SADI, SAME, ISOR and SIMU) were used in order to improve refinement stability. A badly disordered (probably) methanol molecule was found in the asymmetrical unit and could not be satisfactorily refined. The routine SQUEEZE in Platon [74] was therefore used to remove mathematically the effect of the solvent. The quoted formula and derived parameters do not include the squeezed solvent molecule.

CCDC 1435528 (C2) and 1435529 (C3) contain the supplementary crystallographic data for this paper. These data can be obtained free of charge from the Cambridge Crystallographic Data Centre viawww.ccdc.cam.ac.uk/data_request/cif.

4.6 Photophysical characterization

Absorption spectra were measured on a Varian Cary 5000 double-beam UV/Vis-NIR spectrometer and baseline-corrected. Steady-state excitation and emission spectra were recorded on a FluoTime300 spectrometer from PicoQuant equipped with a 300 W ozone-free Xe lamp (250–900 nm), a 10 W Xe flash lamp (250–900 nm, pulse width < 10 μs) with repetition rates of 0.1–300 Hz, an excitation monochromator (Czerny-Turner 2.7 nm mm^–1 dispersion, 1200 grooves mm⁻¹, blazed at 300 nm), diode lasers (pulse width < 80 ps) operated by a computer-controlled laser driver PDL-820 (repetition rate up to 80 MHz, burst mode for slow and weak decays), two emission monochromators (Czerny-Turner, selectable gratings blazed at 500 nm with 2.7 nm mm¹ dispersion and 1200 grooves mm⁻¹, or blazed at 1250 nm with 5.4 nm mm⁻¹ dispersion and 600 grooves mm⁻¹), Glan-Thompson polarizers for excitation (Xe lamps) and emission, a Peltier-thermostatized sample holder from Quantum Northwest (–40 to 105°C), and two detectors, namely a PMA Hybrid 40 (transit time spread FWHM < 120 ps, 300–720 nm). Steady-state and fluorescence lifetimes were recorded in TCSPC mode by a PicoHarp 300 (minimum base resolution 4 ps). Emission and excitation spectra were corrected for source intensity (lamp and grating) by standard correction curves. Lifetime analysis was performed using the commercial FluoFit software. The quality of the fit was assessed by minimizing the reduced chi squared function (χ²) and visual inspection of the weighted residuals and their autocorrelation. Luminescence quantum yields were measured with a Hamamatsu Photonics absolute PL quantum yield measurement system (C9920-02) equipped with a L9799-01 CW Xenon light source (150 W), monochromator, C7473 photonic multi-channel analyzer, integrating sphere and employing U6039-05 Φ_L measurement software (Hamamatsu Photonics, Ltd., Shizuoka, Japan). All solvents used were of spectrometric grade.

4.7 Density functional theory

The DFT calculations were performed with the Gaussian 09 package [75] using the PBE0 hybrid exchange-correlation functional [76]. The SDD basis set [77] was employed for all elements with the exception of phosphorus, which was described by the 6-31G* basis set. Vibrationally resolved emission spectra at 77 K were calculated with the Franck-Condon method [78]. For this purpose, the S₀ ground state and T₁ excited state geometries were optimized using standard Kohn-Sham DFT with multiplicity 0 and 1, respectively. To plot the emission spectrum, each vibronic line was represented by a Gaussian with half width at half maximum of 300 cm⁻¹. To determine the character of the T₁ → S₀ transition, TDDFT calculations were carried out at the TDDFT-optimized T₁ geometry. The molecular orbitals at the T₁ geometry were visualized using the VMD program [79]