Direct population of triplet excited states through singlet–triplet transition for visible-light excitable organic afterglow

Direct population of triplet states via singlet-to-triplet absorption red-shifts the excitation wavelength and improves the organic afterglow efficiency under ambient conditions.


Synthesis and characterization
Chemicals and solvents purchased from Aldrich or Acros are of analytical grade and were used without further purification. Unless otherwise noted, reactions were carried out under a dry nitrogen atmosphere using standard Schlenk techniques. 1 H and 13 C-nuclear magnetic resonance (NMR) spectra were recorded on a Bruker Ultra Shield Plus 400 MHz instrument with CDCl 3 as the solvent and tetramethylsilane (TMS) as the internal standard. Chemical shifts (δ) are given in ppm in Hz. Splitting patterns were designed as follows: s (singlet), d (doublet) and m (multiplet). Mass spectra were obtained using a Shimadzu GCMS-QP2010 instrument. Elemental analyses were performed on an Elementar Vario MICRO elemental analyzer.

9-Phenyl-9H-carbazole (PhCz):
To a 250 mL round bottom flask charged with a stir bar was added 9H-carbazole (3.34 g, 20.0 mmol), copper (0.56 g, 8.8 mmol), potassium carbonate (11.17 g, 80.8 mmol), bromobenzene (3.2 mL, 4.70 g, 30.0 mmol), and 50 mL dry nitrobenzene. The mixture under nitrogen protection was stirred and reacted for 24 hours at 180°C. After cooling to room temperature, the solvent of nitrobenzene was removed by vacuum distillation. 1 The resulting solid was dissolved in 120 mL dichloromethane (DCM) and washed with brine (60 mL). The mixture was then extracted with DCM for three times. The organic phase was collected and dried over MgSO 4 . After removing the solvent under reduced pressure, the crude product was purified by column chromatograph (using petroleum ether: DCM = 10:1 as the eluent) and recrystallized from DCM/hexane for several times to obtain a colorless crystal.

Single crystal X-ray analysis
Single crystals of pBrPhCz, mBrPhCz and DBrPhCz were grown from a mixture of DCM and hexane. X-ray diffraction data of these single crystals were collected on a Bruker Smart Apex CCD area detector diffractometer using graphite-monochromated

S8
Mo-Kα radiation (λ = 0.71073 Å) at 100 K. A narrow-frame method with scan widths of 0.30° in angular velocity (ω) was applied during the data collection. Cell parameters were retrieved using SMART software and refined using SAINT on all observed reflections. Structures were solved by direct methods using the program SHELX-97 program package. Non-hydrogen atoms were found using alternating difference Fourier syntheses and least-squared refinement cycles and, during the final cycles, were refined anisotropically. Hydrogen atoms were placed in calculated positions and refined as riding atoms with a uniform value of U iso . Thus obtained crystallographic parameters of pBrPhCz, mBrPhCz and DBrPhCz were summarized in Table S1 and their CCDC reference numbers are 1573611, 1573612, and 1573614, respectively. 12 Figure S9. Single-crystal unit cells of PhCz, pBrPhCz, mBrPhCz and DBrPhCz.

Thermophysical and electrochemical property measurements
Differential thermal analysis (DTA) and thermogravimetric analysis (TGA) were conducted on a PERKIN-ELMER Diamond TG/DTA under a heating rate of 10ºC/min and a nitrogen flow rate of 100 mL/min.

S10
Cyclic voltammetry (CV) measurements were carried out on a CHI660E system in a typical three-electrode cell with a working electrode (glass carbon), a reference electrode (Ag/Ag + ), and a counter electrode (Pt wire) in an acetonitrile solution of Bu4NPF6 (0.1 M) at a sweeping rate of 100 mV s -1 at room temperature. 14 The highest occupied molecular orbital (HOMO) energy level (E HOMO ) of the material deposited on the surface of the glass carbon working electrode can be estimated from the onset potential of the electrochemical oxidation wave based on the reference energy level of ferrocene/ferrocenium (-4.8 eV), according to Equations S1: : 3.71 a : T m is the melting temperature; b : T d is the thermal decomposition temperature.

Photophysical property investigations
Ultraviolet-visible (UV-Vis) spectra were obtained using a SHIMADZU UV-3600 UV-VIS-NIR spectrophotometer. Steady-state and time-resolved photoluminescence and excitation spectra, photoluminescence lifetime decay curves, and emission quantum yields were measured on an Edinburgh FLSP920 fluorescence spectrophotometer. A xenon arc lamp (Xe900) was used as the excitation source for the steady-state photoluminescence and excitation spectra, photophysical kinetic and S11 quantum yields measurements. The temperature of the samples was controlled by Optistat DN2 and the emission quantum yields were measured by an integrating sphere.
The phosphorescence spectra were obtained at 77 K with a 10 ms delay time after the excitation of a microsecond flash-lamp (uF900). The uF900 flash lamp produces short, typically a few μs, and high irradiance optical pulses for phosphorescence measurements in the range from microseconds to 10 seconds. Therefore, the uF900 flash lamp was also used in the measurements of the organic ultralong room temperature phosphorescence (OURTP) for organic afterglow. Excitationphosphorescence mapping was performed on Hitachi F-4600 with an internal 5 ms delay time under ambient conditions. The lifetimes (τ) of the luminescence were obtained by fitting the decay curve with a multi-exponential decay function of where A i and τ i represent the amplitudes and lifetimes, respectively, of the individual components for multi-exponential decay profiles. 9 The quantum yields (η) of the emission were the absolute ones done by photon counting from the excitation source into an integration sphere with the ratio of photons emitted, as descripted in the following equation.
(S4) = em abs N N  In the equation, N em is the number of emitted photons and N abs is the number of absorbed photons. The quantum efficiency (QE) of OURTP is determined through peakdifferentation-imitating analysis from the steady-state PL spectra.
The photographs and videos were recorded by a Nikon D90 camera at room temperature.

Steady-state photophysical properties
These Br-substituted PhCz compounds have very similar UV-Vis absorption spectra to that of PhCz in dilute solutions, thin films, and crystals ( Figure S11), exhibiting carbazole-dominated π→π* transition for the first strong absorption band S12 around 290 nm in solution and 295 nm in solid and n→π* transition for the second weak absorption band around 340 nm in solution and 345 nm in solid at room temperature. It should be noted that the π→π* transition is slighted influenced by the Br substituents, while the n→π* transition can be significantly modified, leading to the blue-shifted absorption bands after Br substitution due to the electron-withdrawing effects of Br (Table S4)    DBrPhCz (d) crystals for their OURTP peaks at 77 K.

Time-resolved photoluminescence properties
The OURTP emission excited by either 295 or 400 nm is very stable in both strength and lifetime at different atmospheres of nitrogen, air and oxygen ( Figure S17).
However, the 400 nm excitation needs more time (~0.5 s) to populate T 1 for steady OURTP emissions than 295 nm excitation (0.1 s), indicating also that different photophysical procedures occur under the two different excitation wavelength ( Figure   S18): the one-step S 0 →T 1 absorption to populate T 1 through visible-light excitation is slower than the traditional multi-step UV-light excited process, including S 0 →S n absorption, internal conversion (IC), and ISC. Therefore, when the excitation intensity increases, the steady-state and afterglow emissions by 295 nm excitation can be strengthened linearly, but they will be saturated at high intensities of 400 nm excitation      The lifetime of an excited state is determined by the relaxation rates of all its decay channels, but if the lifetime of its population source is longer, the lifetime of the luminophor will be dominated by the population source. 15 When the lifetime of T n * is longer than that of the intrinsic lifetime of T 1 * at 77 K, longer organic afterglow emission lifetime is resulted under 295 nm excitation at 77 K, since the long-lived T n * participates the population process of T 1 * for afterglow emission and dominates the afterglow lifetime under 295 nm excitation (Table S5). The intrinsic lifetime of T 1 * can be measured under 400 nm excitation to exclude the influence of T n * (Table S5). At the S22 room temperature, the lifetime of T n * is significantly decreased, leading to close lifetime under 295 or 400 nm excitation; the lightly shorter lifetime under 295 nm excitation could be due to the enhanced non-radiative decay rates of T 1 *, since the organic afterglow compounds upon 295 nm excitation would result in larger molecular geometry changes for looser molecular packing than that under 400 nm excitation with small geometry variation. 16 Table S5. Emission peaks (λ) and lifetimes (τ) of the OURTP emission after excitation of 295 and 400 nm light at 77 K.

Control experiment
Control experiments were also performed on a previously reported afterglow molecule of DNCzP ( Figure S24a-b) to check the unique S 0 →T 1 absorption in these heavy atom and heteroatom incorporated molecules. DNCzP has a similar phosphorescent spectrum ( Figure S24c) to pBrPhCz, but it cannot be excited by the visible light (λ ex >355 nm) for either phosphorescent emission ( Figure S24d).

Theoretical calculations
Density functional theory (DFT) and time-dependent DFT (TD-DFT) calculations were carried out at the B3LYP/6-31G(d) level using on Gaussian 09 package. 18 The ground state (S 0 ) geometry was fully optimized with B3LYP/6-31G(d) and the optimized stationary point was further characterized by harmonic vibration frequency analysis to ensure that real local minima had been found. The excitation energies in the n-th singlet (S n ) and n-th triplet (T n ) states were obtained using the TD-DFT method based on an optimized molecular structure at ground state (S 0 ). Spin-orbit coupling (SOC) matrix elements between the singlet and triplet excited states are calculated with quadratic response function methods using the Dalton program at the optimized geometry of the lowest singlet excited state (S 1 ) using B3LYP functional and 6-31G(d) basis set. 19,20 The SOC between the ground state and the lowest triplet excited state (T 1 ) S25 was calculated using the same method but was based on the optimized ground state geometry.  radiative decay rate constants of S 1 respectively; k ISC is the ISC rate constant; kT 1 r and kT 1 nr are the radiative and non-radiative decay rate constants of T 1 ; kT 1 * rand kT 1 * nr are the radiative and non-radiative decay rate constants of stabilized T 1 (T* 1); kT 1 ET and kTn ET are the energy trapping rate constants of T 1 and T n ; kT n * rand kT n * nr are the radiative and non-radiative decay constants of stabilized T n (T* n); k IC and k* IC are the rate constants of internal conversion rate constants of T n and T* n respectively.
Theoretically, the concentration decay of S 1 , T n , T 1 , T* n, and T* 1 after the removal of the UV-excitation source at 295 nm can be described in Equations S5-9 under consideration of their all decay channels as illustrated in Figure S28.

S28
The solutions of these differential equations can be expressed in biexponential forms as (S10)   - Under visible light (400 nm) excitation, the concentration decay of T 1 and T* 1 after removing the excitation source can be written in Equations S20-21 ( Figure S28).
The solutions of these differential equations are as follows in biexponential forms According to this photophysical model, multi-lifetimes should be observed from the luminescence decay curves of phosphorescence and OURTP. However, in most cases, shorter lifetime cannot be fitted from the experimental decay curves, which should be due to the large difference between these lifetimes and the larger one generally dominates the luminescence decay measured by time-correlated single photon counting (TCSPC) technology using a microsecond flash-lamp.

Aggregation structure analysis
Aggregation structures influence significantly on the photophysical properties of organic optoelectronic materials. Especially, H-aggregation plays an important role in stabilizing the triplet excitons to elongate their lifetime for the realization of organic afterglow at room temperature. To probe the existence of H-aggregation, the aggregation structure analysis was performance on the single crystal structures of these OURTP molecules. According to the molecular exciton theory, the exciton splitting energy (Δε) in dimer is given by 9 : where M is the electric dipole transition moment, α is the angle between the transition moments of the two molecules in the dimer, and θ 1 and θ 2 are the angles between transition moments of the two molecules and the interconnection of the centers respectively, r uv is the distance between the molecular pair. And, when Δε > 0, it belongs to H-aggregation, and when Δε < 0, it is J-aggregation.

Flexible pattern encryption application
Time-resolved and color-encoded encrypted images were fabricated as follows. On a flexible polyethylene terephthalate (PET) substrate, a thin film of a blue short-lived thermally activated delayed fluorescence (TADF) emitter of bis [4-(9,9- Thirdly, the pattern encryption device was turned over. Finally, the 'IAM' logo was printed on the back surface of the PET substrate using the organic afterglow molecule of pBrPhCz dispersed in ALOE VERA.