Exceptionally fast radiative decay of a dinuclear platinum complex through thermally activated delayed fluorescence†

A novel dinuclear platinum(ii) complex featuring a ditopic, bis-tetradentate ligand has been prepared. The ligand offers each metal ion a planar O^N^C^N coordination environment, with the two metal ions bound to the nitrogen atoms of a bridging pyrimidine unit. The complex is brightly luminescent in the red region of the spectrum with a photoluminescence quantum yield of 83% in deoxygenated methylcyclohexane solution at ambient temperature, and shows a remarkably short excited state lifetime of 2.1 μs. These properties are the result of an unusually high radiative rate constant of around 4 × 105 s−1, a value which is comparable to that of the very best performing Ir(iii) complexes. This unusual behaviour is the result of efficient thermally activated reverse intersystem crossing, promoted by a small singlet–triplet energy difference of only 69 ± 3 meV. The complex was incorporated into solution-processed OLEDs achieving EQEmax = 7.4%. We believe this to be the first fully evidenced report of a Pt(ii) complex showing thermally activated delayed fluorescence (TADF) at room temperature, and indeed of a Pt(ii)-based delayed fluorescence emitter to be incorporated into an OLED.


Theory
To assist the interpretation of the experimental results, we have performed density functional theory (DFT) and time-dependent density functional theory (TD-DFT) simulations on the diand mono-nuclear complexes 5 and 6, respectively using the ORCA quantum chemistry software 1 . All simulations were performed within the approximation of the LC-BLYP ( =0.13 a0 -1 ) exchange and correlation functional. 2,3 A def2-SVP basis set was used for all atoms except Pt, for which a def2-ECP basis was used. Spin-orbit coupling (SOC) calculations were performed using the Amsterdam Density Functional (ADF) code. [4][5][6] The SOC matrix elements (SOCME) were computed with the perturbative approach developed by Ziegler and Wang. 7 A DZP basis set was used for all atoms except Pt, for which a TZP basis set was used. Scalar relativistic effects were accounted for using a zeroth-order relativistic approximation (ZORA). 8,9 Electrochemistry Cyclic voltammetry was conducted in a three-electrode, one-compartment cell. All measurements were performed using 0.1 M Bu4NBF4 (99%, Sigma Aldrich, dried) solution in dichloromethane (ExtraDry AcroSeal®, Acros Organics). Solutions were nitrogen-purged prior to measurement and the measurement was conducted in a nitrogen atmosphere. Electrodes: working (Pt disc d = 1 mm), counter (Pt wire), reference (Ag/AgCl calibrated against ferrocene). All cyclic voltammetry measurements were performed at room temperature with a scan rate of 50 mV s -1 .
The ionization potential (IP) and electron affinity (EA) are obtained from onset redox potentials; these figures correspond to HOMO and LUMO values, respectively. The ionization potential is calculated from onset oxidation potential IP = Eox CV + 5.1 and the electron affinity is calculated from onset reduction potential EA = Ered CV + 5.1. 10,11,12,13 An uncertainty of ±0.02 V is estimated for the electrochemical onset potentials.

Photophysics
Absorption spectra of 10 -6 -10 -5 M solutions were recorded with UV-3600 double beam spectrophotometer (Shimadzu). Photoluminescence (PL) spectra of solutions and films were recorded using a QePro compact spectrometer (Ocean Optics) or FluoroLog fluorescence spectrometer (Jobin Yvon). Phosphorescence decays were recorded using nanosecond gated luminescence and lifetime measurements (from 400 ps to 1 s) using the third harmonic of a high-energy pulsed Nd:YAG laser emitting at 355 nm (EKSPLA). The emitted light was focused onto a spectrograph and detected with a sensitive gated iCCD camera (Stanford Computer Optics) having sub-nanosecond resolution. Time-resolved measurements were performed by exponentially increasing gate and integration times. Further details are available in reference 14 . Time-resolved decays in solution were recorded with a Horiba DeltaFlex TCSPC system using a 330 nm SpectraLED light source. Temperature-dependent experiments were conducted using a liquid nitrogen cryostat VNF-100 (sample in flowing vapour, Janis Research) under nitrogen atmosphere, while measurements at room temperature were recorded under vacuum in the same cryostat. Solutions were degassed using five freeze-pump-thaw cycles. Thin films in Zeonex ® and OLED host matrix were obtained from toluene solution while polystyrene solid films were deposited from chloroform. The films were fabricated through spin-coating and dried under vacuum at room temperature. Solid state emission spectra and photoluminescence quantum yield were obtained using an integrating sphere (Labsphere) coupled with a 365 nm LED light source and QePro (Ocean Optics) detector.

OLED devices
OLEDs were fabricated by spin-coating / evaporation hybrid method. The hole injection layer (Heraeus Clevios HIL 1.3N), electron blocking/hole transport layer (PVKH), and emitting layer (TPD:PBD + dopant) were spin-coated, whereas the electron transport layer (TPBi) and cathode was spun-coated and annealed onto a hotplate at 200 ˚C for 3 min to give a 45 nm film. Electron blocking/hole transport layer (PVKH), was spun from chloroform:chlorobenzene (95:5 v/v) (3 mg/mL) and annealed at 50 ˚C for 5 min to give a 10 nm film. Emitting layer was spun from toluene solution of TPD:PBD (60:40 w/w) with total concentration of host 10 mg/mL. The dopant was dissolved in the solution of blend host in order to obtain final 5% concentration of the emitting layer. The solution was spun onto the PVKH layer and then annealed at 50 ˚C for 5 min giving 30 nm film. All solutions were filtrated directly before application using a PVDF or PTFE syringe filter with 0.45 µm pore size. All other electron transport and cathode layers were thermally evaporated using Kurt J. Lesker Spectros II deposition system at 10 -6 mbar. All organic materials and aluminium were deposited at a rate of 1 Å s -1 . The LiF layer was deposited at a rate of 0.1-0.2 Å s -1 . Characterisation of OLED devices was conducted in 10 inch integrating sphere (Labsphere) connected to a Source Measure Unit and coupled with a spectrometer USB4000 (Ocean Optics). Further details are available in reference 15 .
[1,1′-Bis(diphenylphosphino)ferrocene]dichloropalladium(II) (0.0986 g, 0.121 mmol) was added and the reaction mixture was further deairated for 10 min. The reaction mixture was stirred at 90°C for 18 h. The reaction mixture was allowed to cool to room temperature, diluted with water (50 mL) and extracted with diethyl ether (3X 50 mL). The organic phase was washed with water (2X30 mL), dried with MgSO4 and filtered. The solvent was evaporated to dryness under reduced pressure. The product was purified by column chromatography (silica gel, gradient elution with hexane:ethyl acetate 100:0 to 100:10) to give compound 2 as a brown oil. Yield: 1.

Compound 3
The boronic derivative 2 (4.38 g, 7.72 mmol) was added to a dry round-bottomed flask flushed with argon. Anhydrous 1,4-dioxane (100 mL) was added and the reaction mixture was stirred. 4,6-Dichloropyrimidine (0.460 g, 3.09 mmol) and 2M K2CO3 solution (2.56 g, 18.5 mmol, 9.50 mL) were added and the reaction mixture was degassed under argon for 10 min. Tetrakis(triphenylphosphine)palladium(0) (0.214 g, 0.185 mmol) was added and the reaction mixture degassed for a further 10 min. The reaction mixture was heated at 95°C for 24 h. The reaction mixture was cooled to room temperature and extracted with ethyl acetate. The organic phase was washed with water, separated and dried with MgSO4. The solvent was evaporated to dryness. The crude product was purified by column chromatography (silca gel, gradient elution with hexane:ethyl acetate) to give the product as a pale yellow oil  Compound 3 (0.360 g, 0.375 mmol) was added to a roundbottomed flask. Pyridine hydrochloride (0.867 g, 7.50 mmol) was added and the reaction mixture was heated to 250°C under argon for 12 h. The reaction mixture was cooled to room temperature. Water was added and the resulting precipitate was filtered in vacuo and washed with water to give the desired product as a grey solid. Yield: 0.243 g, 0.261 mmol, 70%

Compound 5
The proligand 4 (0.103 g, 0.110 mmol) was added to a round-bottomed flask. Potassium tetrachloroplatinate (0.114 g, 0.275 mmol) was added. A 9:1 mixture of acetic acid and chloroform (50 mL) was added and the reaction was heated to reflux for 3 days. The mixture was cooled to room temperature and neutralised with a saturated solution of Na2CO3. The reaction mixture was then extracted with dichloromethane. The organic phase was separated and dried with MgSO4. The solvent was evaporated to dryness and the crude red residue purified by column chromatography (silica gel, DCM 100:0). The solid obtained was further purified by recrystallisation from methanol to give the desired product as a dark red solid. Yield: 0.0301 g, 0.0228 mmol, 21 %

X-ray diffraction analysis
The X-ray single crystal data for compound 5 have been collected using λMoKα radiation    Table S4. 1 The energies of the ground and low-lying excited states of complex 5 and 6 calculated at the ground state (S0) and singlet excited state (S1) optimised geometries. Note the S0 energy at S1 geometry is relative to the S0 energy at S0 geometry.

6
Ground state geometry, eV S1 excited state geometry, eV

Photophysics a) Solution state
It is worth noting that methylcyclohexane is a bad solvent for 5, thus not very effective in solubilizing 5 molecules and avoiding aggregation completely. Although the emission spectra at 10 -6 -10 -5 M remain identical (Figure S5.1), a significant change in the absorption spectrum is observed in this concentration range (Figure S5.2). The excitation spectrum of 5 emission at 605 nm closely resembles the absorption at low concentrations (Figure S5.3). This inevitably indicates that despite the large 83% PLQY in MCH, 5 still forms some kind of aggregate states in this solvent that are non-or hardly luminescent.

c) Studies in an OLED host material
Studies of 5 dispersed in TPD:PBD were carried out using the same composition as that of thin films used in devices. The emission spectrum in this host is shown in Figure S5.13; the ΦPL = 0.31 ± 0.05 is smaller than in a solution, probably due to aggregation. The photoluminescence decay (Figure S5.11) is significantly more complex than in solution, Zeonex or polystyrene, which is probably also caused by the higher emitter concentration when compared to those used in the former photophysical studies that were aimed at characterising predominantly isolated molecules (5 % vs 0.1%). The decay shows a weak short-lived fluorescence from the host, followed by longer-lived photoluminescence of 5. This is observed at every temperature and indicates that the Förster energy transfer from host to guest is not fully completea strong indication that a lower dopant concentration is not suitable for OLEDs using this host. The decay of 5 emission is characterized by two main decay components indicating a more heterogenous environment for the emitter: τ1 = 65 ± 11 ns (75 %), τ2 = 0.7 ± 0.1 μs (25 %) at 295 K and τ1 = 0.46 ± 0.08 μs (59 %), τ2 = 8 ± 1 μs (41 %) at 80 K. Overall, the shorter  in the OLED host in relation to the polymer film and solution may indicate that additional quenching processes are active. Note that the host fluorescence lifetime varies from 2.5 ± 0.2 ns at 295 K to 6.1 ± 0.7 ns at 80 K, and thus does not interfere with the luminescence decay of 5. A third, long-lived component is present at lower temperatures, which may be assigned to traps also occurring in other bicomponent (donor + acceptor) blend hosts 21,22 and in exciplexes. 23,24 The decay does vary with T in a manner qualitatively consistent with TADF but, due to the complicated nature of the dopant emission, lifetime data were not subject to analysis towards determination of the activation energy.   The most likely cause for the observed spectral shifts is the variation in relative orientation of host-guest dipoles or intermolecular interactions of dopant molecules that occur in the ground state, which may affect the charge-transfer excited state of 5. 26 In contrast with the low-concentration studies, the emission spectrum in TPD:PBD {TPD -

N,N'-bis(3-methylphenyl)-N,N'-bis(phenyl)-benzidine; PBD -2-(4-biphenyl)-5-(4-tert-
butylphenyl)-1,3,4-oxadiazole} thin film is not a superposition of just two emissions originating from S1 and T1 states, respectively. Instead, the emission attributed to the S1 state shows a broad distribution of energy, while the emission originating from the T1 state appears to remain constant in time. This is in agreement with the S1 and T1 behaviour in the monometallic analogue 27 6, as it appears that both complexes 5 and 6 show an environment-sensitive S1 (i.e., substantial CT character) and significantly less sensitive T1 (lower CT character decaying with different lifetimes, as seen in Figure S5.12, is related to molecules with a higher energy singlet (thus experiencing larger ΔEST) that preferentially decay through phosphorescence, while molecules with lower energy S1 states receive sufficient thermal energy to preferentially decay through the singlet state via TADF. This is consistent with the reduction of the energy of the delayed fluorescence onset at lower temperatures. 6. Electrochemistry 5 shows a typical electrochemical behaviour to many other Pt(II) complexes (Figure S6.1).
Oxidation of the complex is irreversible due to dPt(II) orbital admixtures of the central atom to the HOMO. Thus, the metal gives up electron(s) reaching higher oxidation states. I.e. the onset oxidation potential of the complex Eonset ox = 0.41 V is a result of the ligand not being strongly electron-rich. By using electron-rich ligands HOMO is destabilised leading to a quasireversible oxidation. 28 On the other hand the ligand stabilises LUMO of the complex, mostly due to its strong electron-withdrawing character that originates from presence of pyrimidine and pyridine moieties. The presence of pyrimidine linker is a crucial factor differentiating 5 from its monometallic analogue 27 as the ligand in the latter has significantly weaker electronwithdrawing properties. This as an effect gives a Eonset red = -1.45 V. Consequently, the ionization potential (IP) and electron affinity (EA) of the compound equal to IP = 5.51 eV, EA = 3.65 eV.
Thus, electrochemical energy gap of the material Eg el = 1.86 eV.

OLED devices
Exciplex hosts, such as TPD:PBD, have been successfully and commonly used for other metal complexes. [29][30][31] Moreover, the TPD:PBD mixture is also readily soluble in toluene exhibiting excellent film-forming properties. However, as PBD is an electron-transporting and holeblocking material, TPD does transport holes, but does not block electrons. This results in very low-efficiency devices if an additional electron-blocking layer is not used.
The percentage doping of the emitter varies usually between 1-20% in typical OLED devices.
5 % doping concentration has found to be a successful approach in solution-processed devices of various kinds. 22,25,28,32 Specifically, in the case of 5, which was found to form non-emissive aggregates in methylcyclohexane and in solid film (see discussion of the photophysics in film and solution), we aimed the doping level to be maintained as low as possible. However, at a doping level below 5 % there is insufficient amount of dopant to trap all charge carriers, promoting recombination in the host.