C−H Borylation/Cross‐Coupling Forms Twisted Donor–Acceptor Compounds Exhibiting Donor‐Dependent Delayed Emission

Abstract Benzothiadiazole (BT) directed C−H borylation using BCl3, followed by B−Cl hydrolysis and Suzuki–Miyaura cross‐coupling enables facile access to twisted donor–acceptor compounds. A subsequent second C−H borylation step provides, on arylation of boron, access to borylated highly twisted D−A compounds with a reduced bandgap, or on B−Cl hydrolysis/cross‐coupling to twisted D‐A‐D compounds. Photophysical studies revealed that in this series there is long lifetime emission only when the donor is triphenylamine. Computational studies indicated that the key factor in observing the donor dependent long lifetime emission is the energy gap between the S1/T2 excited states, which are predominantly intramolecular charge‐transfer states, and the T1 excited state, which is predominantly a local excited state on the BT acceptor moiety.


S3
THF. 2 µL of sample solution and 20 µL of matrix solution were thoroughly mixed and 1 µL of this solution was spotted onto a well with no dopant and 1 µL spotted by a layered method with the NaI. The solvent was allowed to evaporate before being placed in the spectrometer. Samples were run in positive polarity mode in either linear or reflection mode. High resolution mass spectra (HRMS) were recorded on a Waters QTOF mass spectrometer.
All UV-vis absorption spectra were recorded on a Varian Cary 5000 UV-vis-NIR spectrometer at room temperature in spectroscopic grade solvents. Solution phase emission spectra were recorded on a Varian Cary Eclipse Fluorimeter at room temperature in spectroscopic grade solvents, exciting at their relative absorbance maxima. Absolute quantum yield values were recorded on an Edinburgh Instruments FP920 Phosphorescence Lifetime Spectrometer equipped with a 5 watt microsecond pulsed xenon flashlamp (with single 300 mm focal length excitation and emission monochromators in Czerny Turner configuration) and a red sensitive photomultiplier in peltier (air cooled) housing and determined using a calibrated Edinburgh Instruments integrating sphere. Solid phase samples were also measured using the FP920 and integrating sphere, with steady state spectra excited via a 450 W Xe lamp.
Cyclic voltammetry was performed using a CH-Instrument 1110C Electrochemical/Analyzer potentiostat under a nitrogen flow. Measurements were made using a 1 mM analyte solution with 0.1 M tetra n butylammonium hexafluorophosphate (Fluka ≥99.0 %) as the supporting electrolyte in DCM that had been degassed prior to use and obtained from a dry solvent system. A glassy carbon electrode served as the working electrode and a platinum wire as the counter electrode. An Ag/AgNO 3 non-aqueous reference electrode was used. All scans were calibrated against the ferrocene/ferrocenium (Fc/Fc + ) redox couple, which in this work is taken to be 5.1 eV below vacuum. S3 The half-wave potential of the ferrocene/ferrocenium (Fc/Fc + ) redox couple (E 1/2 , Fc,Fc+ ) was estimated from E 1/2, Fc,Fc+ = (E ap + E cp )/2, where E ap and E cp are the anodic and cathodic peak potentials, respectively.
Calculations were performed using the Gaussian 09 (Revision D.01) suite of programmes. S4 Structures were optimised with DFT method at the PBE0/6-31G(d,p)/PCM(toluene) level of theory. S5 In all cases, structures were confirmed as S4 minima by frequency analysis and the absence of imaginary frequencies. The S 1 geometries were optimised using time-dependent DFT and the T 1 geometries were optimised using triplet ground state. In-house Fortran 77 codes were used for calculations of SOC for k (R)ISC , HOMO/LUMO and HONTO/LUNTO absolute overlap percentages, and ICT/LE characters. SOC values were obtained based on the equations provided by Gao et al. S6 Absolute overlap integrals were numerically calculated based on Becke's grid-based integration. S7 ICT/LE contribution percentages for each excited state were obtained via the Löwdin population analysis of NTOs; see refs S8 and S9 for more information. Full Cartesian coordinates of the optimised ground state geometries are provided in the computational section below. S7 1 (279 mg, 0.70 mmol) was dissolved in anhydrous DCM (3 mL) and BCl 3 (1M in DCM) (1 mL, 1 mmol) was added to the solution where a colour change from yellow to dark purple was observed. The solution was then stirred at ambient temperature for 0.5 hours under the dynamic flow of nitrogen. The solvent and other volatiles were then removed under reduced pressure and the resulting purple residue was dissolved in non-anhydrous THF (10 mL). H 2 O (2 mL) was then added to the reaction mixture which was stirred overnight at ambient temperature were a colour change from purple to orange was observed. 9-(4-Bromophenyl)-9H-carbazole (235 mg, 0.75 mmol) was added to the reaction mixture which was then degassed (bubble N 2 ). Pd(PPh 3 ) 4 (40 mg, 0.035 mmol) was added to the degassed reaction mixture followed by the addition of K 3 PO 4 2M (aq.) (1.75 mL, 3.50 mmol). The reaction mixture was then stirred for 10 hours at 75 o C. The reaction mixture was diluted with ethyl acetate (50 mL) followed by the addition of brine (10 mL) and deionised water (30 mL). The organic layer was isolated using a separating funnel and dried (MgSO 4 ). The solvent was evaporated under reduced pressure and the resulting residue was purified using silica gel chromatography [eluent = 1:9 DCM: petroleum ether graduated to 2:8 DCM: petroleum ether]. The desired product was then isolated as a yellow/green solid. Yield: 375 mg, 84 %.  7, 153.5, 143.6, 143.2, 141.0, 140.6, 135.8, 134.5, 133.7, 133.3, 133.0, 131.1, 130.7, 130.6, 130.4, 129.0, 128.7, 127.8, 127.3, 126.2, 125.8, 123.2, 120.2, 119.8, 109.5, 35.5, 35.4, 33.5, 22.5, 22.3, 14.0, 13.9; Compound 5. the dynamic flow of nitrogen. The solvent and other volatiles were then removed under reduced pressure and the resulting purple residue was dissolved in non-anhydrous THF (30 mL). H 2 O (3 mL) was then added to the reaction mixture which was stirred overnight at ambient temperature were a colour change from purple to orange was observed. 10-(4bromophenyl)-10H-phenoxazine (286 mg, 0.85 mmol) was added to the reaction mixture which was then degassed (bubble N 2 ). Pd(PPh 3 ) 4 (46 mg, 0.04 mmol) was added to the degassed reaction mixture followed by the addition of K 3 PO 4 2M (aq.) (2.00 mL, 4.00 mmol). The reaction mixture was then stirred for 12 hours at 75 o C. The reaction mixture was diluted with ethyl acetate (50 mL) followed by the addition of brine (10 mL) and deionised water (30 mL). The organic layer was isolated using a separating funnel and dried (MgSO 4 5, 153.5, 143.8, 143.6, 143.3, 142.3, 140.7, 137.0, 134.5, 134.2, 133.8, 133.4, 133.2, 131.8, 131.0, 130.6, 130.2, 129.9, 129.0, 128.7, 128.0, 127.2, 123.1, 121.1, 115.3, 113.0, 35.5, 35.4, 33.6, 33.5, 22.6, 22.4, 14.0, 14.0; Compound 6 4,7-Bis(9,9-dioctyl-9H-fluoren-2-yl)-2,1,3-Benzothiadiazole (compound 2, 294 g, 0.32 mmol) was dissolved in anhydrous DCM (5 mL) and BCl 3 (1M in DCM) (0.8 mL, 0.8 mmol) was added to the solution where a colour change from yellow to dark purple was observed. The solution was then stirred at ambient temperature for 3 hours under the dynamic flow of nitrogen. The solvent and other volatiles were then removed under reduced pressure and the resulting purple residue was dissolved in non-anhydrous THF (10 mL). H 2 O (1 mL) was then added to the reaction mixture which was stirred overnight at ambient temperature were a colour change from purple to orange was observed. 4-Bromotriphenylamine (115 mg, 0.35 mmol) was added to the reaction mixture which was then degassed (bubble N 2 ). A solution of Pd( t Bu 3 P) 2 (17 mg, 0.033 mmol) in THF (3 mL) was added to the degassed reaction mixture followed by the addition of K 3 PO 4 2M (aq.) (0.80 mL, 1.60 mmol). The reaction mixture was then stirred overnight at ambient temperature. The reaction mixture was diluted with ethyl acetate (50 mL) followed by the addition of brine (10 mL) and deionised water (30 mL). The organic layer was isolated using a separating funnel and dried (MgSO 4 ). The solvent was evaporated under reduced pressure and the resulting residue was purified using silica gel chromatography [eluent = 1:9 DCM: petroleum ether]. The desired product was then isolated as a yellow solid. Yield: 320 mg, 86 %.

Delayed Emission Studies
Compound 3 showed a decrease in the fluorescence intensity by ~23 % after the addition of air ( Figure S11).
Emission is observed at delay times after excitation of >0.1 ms with decreasing intensity as the delay time increases. The spectra are essentially identical to that of the prompt fluorescence. Additionally, complete quenching of the delayed fluorescence is observed under Air saturated conditions. Therefore, the origin of the delayed emission is assigned to reverse intersystem crossing from the triplet state ( Figure S12). S38 Figure S12: Normalised emission intensity of compound 3 at different delay times under Argon (Ar) and after 0.1 ms under Air. Figure S15: Normalised photoluminescence spectra in (toluene) of 9 under Air, of 9 under Argon (sample prepared in a glovebox in a sealed cuvette using degassed solvent).
Compound 9 showed a decrease in the fluorescence intensity by ~17 % after addition of air S40 Figure S16: Normalised emission intensity of compound 9 at different delay times under Argon (Ar) and after 0.1 ms under Air. Emission is observed at delay times of >0.1 ms with decreasing intensity as the delay time increases. The spectra are essentially identical to that of the prompt fluorescence. Additionally, complete quenching of the delayed fluorescence is observed under Air saturated conditions. ( Figure S16).
Figure S17: 9 shows an exponential decrease in emission intensity with increasing delay time S41 Figure S18: Normalised photoluminescence spectra in (toluene) of 2 in Air, of 2 under Argon (sample prepared in a glovebox in a sealed cuvette using degassed solvent), of 6 in Air, and 6 under Argon (sample prepared in a glovebox in a sealed cuvette using degassed solvent).
Compound 6 showed a decrease in the fluorescence intensity by ~10 % after the addition of O 2 ( Figure S18). Compound 6 in dilute toluene solution demonstrates a quantum yield value of 46.7 % under argon which reduces to 42.0 % under air ( Figure S19).  Figure S21: Normalised photoluminescence spectra in (toluene) of 11 in Air, of 11 under Argon (sample prepared in a glovebox in a sealed cuvette using degassed solvent), of 10 in Air, and 10 under Argon (sample prepared in a glovebox in a sealed cuvette using degassed solvent).
Compound 10 showed a decrease in the fluorescence intensity by ~20 % after the addition of O 2 ( Figure S21). Emission is observed at delay times of >0.1 ms with decreasing intensity as the delay time increases. The spectra are essentially identical to that of the prompt fluorescence. Additionally, complete quenching of the delayed fluorescence is observed under Air saturated conditions. Therefore, the origin of the delayed emission is assigned to reverse intersystem crossing from the triplet state ( Figure S22). S44 Figure S22: Normalised emission intensity of compound 10 at different delay times under Argon (Ar) and after 0.1 ms under Air. Emission is observed at delay times of >0.1 ms with decreasing intensity as the delay time increases. The spectra are essentially identical to that of the prompt fluorescence. Figure S23: An exponential decrease in emission intensity with increasing delay time for 10 S45 Figure S24: The emission spectra of compound 10 (under Argon) remains essentially identical at delay times up to 1 ms ( Figure 17).
The effects of the triphenylamine unit ortho-to the benzothiadiazole unit can be observed as the linear 7,7' substituted isomer of 10 (compound 11) shows no appreciable decrease in emission intensity upon the addition of O 2 and no delayed emission is observed (Figure 25). For comparison compound DPS-PXZ, which is a well-studied thermally activated delayed fluorescence (TADF) material published by Adachi et. al., S10 was synthsized using the published procedure and in dilute toluene solution under an argon atmosphere DPS-PXZ shows intense fluorescence and weak emission after a 0.2 ms delay. Upon the addition of oxygen a significant decrease in the fluorescence intensity by ~70% (reported 73% decrease in quantum yield upon addition of O 2 ) 1 is observed and the delayed emission is absent ( Figure S27). Figure S27: Emission of DPS-PXZ in toluene under argon and air.
A related compound DPS-TCb published by Monkman et. al.,S11 shows emission at delay times up to 600 µs with decreasing intensity as the delay time increases. The spectra are essentially identical to that of the prompt fluorescence. Additionally, complete quenching of the delayed fluorescence is observed under Air saturated conditions ( Figure S28).
These observations on two other compounds are closely comparable to the photophysical properties observed for the triphenylamine functionalised compounds studied in this work. Figure S28: The delayed fluorescence emission in DPS-TCb decays within 600 µs with a single exponential time constant. Taken from reference S11

Crystallographic Details
Crystallographic data for compound 3-BPh 2 were recorded on an Agilent Supernova diffractometer, at 150 K with Mo Kα radiation (mirror monochromator, λ =0.7107). The CrysAlisPro software package was used for data collection, cell refinement and data reduction. S12 Data for compounds 3 and 4 were collected on an Oxford Diffraction Xcalibur 2 at 150 and 293 K, respectively, with Mo Kα radiation (mirror monochromator, λ =0.7107). Synchotron X-ray data for compound 5 were collected at beamline I19 (λ = 0.6889 Å) of Diamond Light Source at a temperature of 100 K and measured using GDA suite of programs. S13 The CrysAlisPro5 (3, 3-BPh 2 , 4) and dials (5) software packages were used for data collection, cell refinement and data reduction. For all data sets the CrysAlisPro software package was used for empirical absorption corrections, which were applied using spherical harmonics, implemented in SCALE3 ABSPACK scaling algorithm. All further data processing was undertaking within the Olex2 software. S14 The molecular structures all compounds were solved with ShelXT S15 structure solution program using Intrinsic Phasing. All structures were refined with the SHELXL S16 refinement package using Least Squares minimisation against F2. Non-hydrogen atoms were refined anisotropically. Hydrogen atoms were all located in a difference map and repositioned geometrically.