Thermally activated delayed fluorescence (TADF) is also referred to as E-type delayed fluorescence. This mechanism was initially observed by Francis Perrin in 1924.1,2
Later, in 2012, Professor Chihaya Adachi along with colleagues at Kyushu University harvested triplet excitons in organic light emitting diodes (OLED) using the TADF mechanism and successfully developed a new kind of high-efficiency OLED that eliminates the use of heavy metals.3 Consequently, the TADF once again received a great deal of attention and was exposed to a broader set of audience. Since then, the TADF mechanism has become one the most preferred methods to harvest triplet excitons in OLEDs, and the latest TADF emitters with attractive color coordinates and excellent stability are being extensively studied in both industry and academia.
The Triplet Exciton Problem
As one of the most widespread display types for smartphones and televisions, OLEDs consume relatively less power and provide higher contrast ratios when compared to traditional LCD displays. In an OLED, layers of carbon-based organic semiconductors are tightly packed between a pair of electrodes, and the organic semiconductor is injected with electrons and holes under an applied bias. Upon encountering, the electrons and holes initially form Coulombically bound electron-hole pairs, referred to as excitons, which can subsequently recombine to produce light.
Owing to spin statistics, 75% of the excitons created will be in the triplet state (T1), while 25% will be in the singlet state (S1). In the initial OLED designs, in which fluorescent molecular emitters were used, the S1 state alone was emissive, while the T1 > S0 radiative transition was barred because of the conservation of spin angular momentum. As a result, these first-generation OLEDs were restricted to the highest internal quantum efficiency (IQE) of 25%.
In order to overcome this restriction, heavy metals, like iridium and platinum, were integrated into the molecular emitters to develop second-generation phosphorescent OLEDs. Due to the presence of heavy metals in the molecule, the strength of the spin-orbit coupling between the orbital angular momentum and the spin angular momentum is increased and the T1 > S0 radiative transition is permitted.
Although this method makes it possible to achieve IQEs of 100%, the use of heavy metals has a number of major disadvantages. Since the metals are costly and rare, they are not practical for large-scale production. Moreover, phosphorescent OLEDs have poor stability, especially in the blue, and so far, a stable deep blue phosphorescent emitter has not been identified yet.
Third-Generation TADF OLEDs
These limitations resulted in the fabrication of heavy metal-free, third-generation OLEDs, which employ the TADF mechanism for their operation. In the case of a TADF emitter, the S1 and T1 states are designed in a such way that they are robustly coupled and close in energy, which allows the excitons produced in the T1 state to go through a thermally assisted reverse intersystem crossing, or RISC, to the S1 state, in which they can radiatively decay to the S0 leading to delayed fluorescence emission (see Figure 1).
Figure 1. The operating mechanisms behind first, second, and third-generation OLEDs.
With the help of the TADF mechanism, IQEs of 100% can be realized without having to use heavy metals. As such, there is a need to develop a new generation of TADF emitters that has desirable color coordinates, good stability, and high quantum yields. At the time of development of the latest emitters, the emission characteristics have to be completely defined and these data should be utilized to fine-tune subsequent molecular designs. This article demonstrates the potential of the FS5 Spectrofluorometer for offering a comprehensive characterization of TADF emitters, and this is done by studying the newly published emitter CzDBA,4 and validating the presence of TADF emission.
Figure 2. Chemical structure of the CzDBA TADF emitter.4
Materials and Methods
CzDBA was first dissolved in toluene at 2 x 10–5 M concentration. Then, using freeze-pump-thaw, the solution was degassed and the cuvette was backfilled with nitrogen to inhibit oxygen ingress. The FS5 Spectrofluorometer — fitted with a PMT-900 detector, a 375 nm picosecond pulsed diode laser (EPL-375), a 5 W microsecond Xe flashlamp, a 150 W Xe lamp, and multichannel scaling (MCS) lifetime electronics — was used to make absorption and photoluminescence measurements. The SC-25 Thermoelectric Cuvette Holder Module was used to make room temperature measurements, while the SC-70 Liquid Nitrogen Dewar Module was used to make cryogenic measurements. Finally, the SC-30 Integrating Sphere Module was used to determine the PLQY.
Figure 3. FS5 spectrofluorometer.
Results and Discussion
Absorption and Emission Spectra
When it comes to the characterization of a TADF emitter, the first step is the precise determination of its emission and absorption spectra. A transmittance detector — included in the FS5 spectrofluorometer as standard — allows the measurement of emission and absorption spectra with the help of a single instrument (see Figure 4).
Figure 4. Absorption (black) and emission (red) spectra of degassed CzDBA in toluene. The small peak in the emission spectrum at 425 nm is the Raman scatter from the solvent. Absorption parameters: Δλex = 2 nm. Emission parameters: λex = 375 nm, Δλex = 2 nm, Δλex = 2 nm.
CzDBA absorption has maxima at 340 nm and 290 with a lengthy tail that extends out to 500 nm. The emission spectrum is featureless and wide, with a maximum emission at 550 nm and an FWHM of 95 nm corresponding to color coordinates of 0.37, 0.58 in CIE 1931 color space.
Oxygen Quenching of the Photoluminescence Quantum Yield
The measurement of the photoluminescence quantum yield in the presence and absence of oxygen is a preliminary check for the presence of TADF emission. Since RISC from the T1 is involved in TADF, any mechanism that eliminates the population from the T1 state will reduce the intensity of the TADF emission. Moreover, since molecular oxygen possesses a triplet ground state, it will quench the TADF emission by quickly removing the population from the T1 state via energy transfer. Figure 5 shows the change in emission intensity when the CzDBA solution is subjected to oxygen.
Figure 5. Emission spectra of degassed CzDBA solution (red) and non-degassed CzDBA (black). The quantum yield of the non-degassed solution was measured using the SC-30 Integrating Sphere module and the degassed quantum yield extrapolated based on the relative intensity of the emission. Experimental parameters: λex = 375 nm, Δλex = 5 nm, Δλem = 0.5 nm.
The SC-30 Integrating Sphere Module of the FS5 spectrofluorometer was used to measure the photoluminescence quantum yields. The considerable 50% reduction in quantum yield upon exposing the solution to oxygen indicates that the T1 state is not only involved in the emission process but is also the first indicator that emission can possibly contain TADF.
In order to decisively confirm the presence of TADF, the temporal response of the emission should be measured. The FS5 spectrofluorometer can be fitted with flashlamps, lifetime counting electronics, and a broad variety of LEDs and pulsed diode lasers for determining the phosphorescence, fluorescence, and delayed fluorescence of materials. The solution was excited with a pulsed diode laser and the emission decay of CzDBA was measured using multichannel scaling single photon counting (MCS) (Figure 6).
Figure 6. Emission decay of degassed CzDBA solution measured using MCS. Experimental Parameters: Source = EPL-375, Rep Rate = 50 kHz, λex = 375 nm, λem = 550 nm, Δλem = 20 nm.
MCS is a time-resolved method complementary to the more commonly recognized time-correlated single photon counting (TCSPC). TCSPC is the preferred technique for obtaining fast fluorescence decays at high laser excitation rates; however, if the emission is long (>μs) and if there is a need to lower the excitation, TCSPC becomes slow because of the condition that the detection count rate must be less than 5% of the excitation rate. In MCS acquisition, the detection window is divided into time intervals and all photons that reach within a definite time interval are counted; the photons in the next interval are then counted and this continues until the whole time range has been covered.
This multiple stop mode detection allows higher detection count rates to be used than in TCSPC, which decreases the acquisition time of longer decays and is, thus, the preferred technique for measuring delayed fluorescence. The FS5 spectrofluorometer can be changed effortlessly from TCSPC to MCS acquisition modes and vice versa using the Fluoracle software, which allows the optimum detection technique for the preferred time change to be selected. The decay of the CzDBA emission has a characteristic biexponential behavior containing a prompt constituent with a lifetime of 78 ns and a delayed constituent with a lifetime of 1487 ns.
The prompt constituent can be assigned explicitly as a fluorescence originating from the S1 > S0 transition (see Figure 7). The delayed component could either be delayed S1 > S0 fluorescence after repopulation of the S1 through RISC from the T1, or direct T1 > S0 phosphorescence. These mechanisms cannot be conclusively distinguished just with the time scale of the emission.
Figure 7. Simplified Jablonski diagram of TADF photophysics.
Time-Resolved Emission Spectrum (TRES)
Comparing the spectral shape of the delayed and prompt components by obtaining a time-resolved emission spectrum (TRES) is one technique to determine if the origin of the delayed component is phosphorescence or delayed fluorescence. In a TRES measurement, the emission decay is determined as a function of emission wavelength to create a 3D time-resolved spectrum (Figure 8a). The TRES data can subsequently be sliced to create a snapshot of the emission spectrum at a definite time after the laser flash.
Two TRES slices exhibiting the shape of the delayed component (4 μs after flash) and the prompt component (100 ns after flash) are illustrated in Figure 8b. It is obvious that the spectra of the prompt and delayed components are identical, which indicates that the delayed component is delayed fluorescence as phosphorescence from the T1 state would take place at a longer wavelength than the fluorescence because of the lower lying energy of the T1 state.
Figure 8. Time-resolved emission spectrum (TRES) of CzDBA solution measured using MCS. TRES color map (a) and TRES slices (b) of the prompt and delayed emission. Experimental Parameters: Source = EPL-375, Rep Rate = 50 kHz, λex = 375 nm, Δλem = 10 nm.
Temperature Dependence of the Emission Decay
The second technique to determine the source of the delayed component involves the measurement of the temperature dependence of the decay. The FS5 spectrofluorometer can be fitted with heating stages, a variety of thermoelectric cuvette holders, and cryostats that allow measurements to be performed over a temperature range of 196 °C (77 K) to 600 °C (873 K). The emission decays of CzDBA at 77 K and 300 K were determined with the help of the SC-80 Liquid Nitrogen Dewar Module and are illustrated in Figure 9.
Figure 9. Variation of the delayed fluorescence intensity (a) and phosphorescence intensity (b) with temperature. Experimental Parameters (a): Source = EPL-375, Rep Rate = 50 kHz, λex = 375 nm, λem = 550 nm, Δλem = 10 nm. Experimental Parameters (b): Source = μs Xe Flashlamp, Rep Rate = 100 Hz, λex = 375 nm, Δλex = 2 nm, λem = 550 nm, Δλem = 2 nm.
It can be noted from Figure 9a that the delayed component is greatly suppressed when the sample is cooled to 77 K, signifying that the delayed component originates from a thermally activated process. In addition, the decays were measured on a longer timescale with the help of the microsecond Xe flashlamp for excitation (see Figure 9b) to find if there is longer-lived millisecond phosphorescence emission. The longer-lived phosphorescence exhibits the opposite trend, where it is present at 77 K but suppressed when the sample is warmed to 300 K.
These temperature-dependent decays prove the assignment of the delayed component as thermally activated delayed fluorescence. At 300 K, the molecule has adequate thermal energy to overcome the energy gap, ΔEST, between the S1 and the T1. RISC advances and repopulates the S1 state, resulting in delayed fluorescence, and the depopulation of the T1 state suppresses the phosphorescence. When the temperature of the sample is lowered to 77 K, this situation reverses — the thermal energy of the molecule is no longer sufficient to overcome ΔEST and RISC is turned off, which leads to suppressed delayed fluorescence and improved phosphorescence.
The FS5 spectrofluorometer was used to study the emission and absorption properties of CzDBA, and the presence of TADF emission was also verified. In this article, the efficacy of the FS5 spectrofluorometer for investigating TADF emitters and its potential in characterizing the absorption, emission, quantum yield, and lifetime of new emitters in a single compact instrument have been demonstrated.
References and Further Reading
- F. Perrin, La fluorescence des solutions, Ann. Phys. 12 169–275 (1929).
- B. Valeur, M. N. Berberan-Santos, Introduction, Molecular Fluorescence: Principles and Applications, 2nd Ed. Wiley-VCH 01-25 (2012).
- H. Uoyama, K. Goushi, K. Shizu, H. Nomura and C. Adachi, Nature, 492, 234–238 (2012).
- T.-L. Wu, M.-J. Huang, C.-C. Lin, P.-Y. Huang, T.-Y. Chou, R.-W. C.-H. Chen, H.-W. Lin, R.-S. Liu & C.-H. Cheng, Nat. Photonics 12 235–240 (2018).
This information has been sourced, reviewed and adapted from materials provided by Edinburgh Instruments.
For more information on this source, please visit Edinburgh Instruments.