Sn-卟啉

Thermally Activated Delayed Fluorescence from Sn4þ–Porphyrin Complexes and Their Application to Organic Light-Emitting Diodes — A Novel Mechanism for Electroluminescence By Ayataka Endo, Mai Ogasawara, Atsushi Takahashi, Daisuke Yokoyama, Yoshimine Kato, and Chihaya Adachi* In recent years, there have been high expectations of the use of organic light-emitting diodes (OLEDs) in flat-panel displays and general lighting applications, owing to the characteristic features triplet annihilation by the use of thermally activated delayed fluorescence (TADF),[7,8] as shown in Figure 1. For TADF materials, heat accelerates the reverse intersystem-crossing (ISC) C O M M U N IC A T IO N www.advmat.de ism: carrier cesses. The highlighted. and recom- ratio; hPL: iency; ISC: .) 744 Motooka 4802 DOI: 10.1002/adma.200900983 Figure 1. Schematic view of the electroluminescence mechan injection, transport, recombination; and radiative decay pro thermally activated delayed fluorescence (TADF) process is (g : ratio of holes and electrons in carrier injection, transport, bination processes; hr: singlet and triplet exciton formation photoluminescence efficiency; hhn: light out-coupling effic intersystem crossing; and RISC: reverse intersystem crossing Nishi, Fukuoka 819-0395 (Japan) Dr. A. Takahashi[+] Sogo Pharmaceutical Co., Ltd. Hebinami 28-3, Johban Shimofunao Iwaki, Fukushima 972-8312 (Japan) [+] Present address: Mitsubishi Chemical Group Science and Technology Research Center, Inc., Display Project, Research and Development Division, 1000 Kamoshida-cho, Aoba-ku, Yokohama 227-8502, Japan. E-mail: adachi@cstf.kyushu-u.ac.jp M. Ogasawara, Prof. Y. Kato Department of Materials Science and Engineering Kyushu University Nishi, Fukuoka 819-0395 (Japan) � 2009 WILEY-VCH Verlag Gmb of these organic semiconductingmaterials, such as flexibility over large-area, low-cost fabrication and high-performance optical and electrical properties. In order to enhance the electroluminescence (EL) efficiency of OLEDs, various fluorescent and phosphorescent materials have been widely developed.[1–5] AlthoughOLEDs using fluorescent materials have achieved a high reliability practically, the external EL efficiencies (hEL) are intrinsically limited to approximately 5%, due to limitations of their singlet-exciton production efficiency (hexciton� 25%) under electrical excita- tion.[4] In contrast, OLEDs that use phosphorescent materials have achieved hexciton of almost 100%, and a significant roll-off of the hEL has been observed with an increase of the current density, owing to the rather long lifetime of the triplet excited states, which results in acceleration of various triplet exciton annihilation processes.[6] Furthermore, the selection of practically useful phosphorescent materials has been limited to Ir and Pt complexes. Therefore, since both fluorescence- and phosphor- escence-based OLEDs have both advantages and disadvantages, the use of a novel light-emitting mechanism has been hoped for to avoid the shortcomings. We have proposed potential mechanisms to achieve compat- ibility of harvesting both singlet and triplet excitons and avoiding [*] Prof. C. Adachi, A. Endo, Dr. D. Yokoyama Center for Future Chemistry Kyushu University 744 Motooka from a triplet excited state (T1) to a singlet excited one (S1), thus leading to an increase of the fluorescence intensity. Therefore, in principle, when TADF materials are used in OLEDs, heating the device accelerates reverse ISC, and thus makes it possible to provide OLEDs with high EL efficiency, even with a fluorescence decay rate. Furthermore, the roll-off characteristics of the hEL can be improved if the rate constant of the reverse ISC (kRISC) is significantly larger than the phosphorescence decay rate (kP). However, the efficiencies of reverse ISC in the previously examined TADF materials were quite low, and we experienced difficulties in verifying the effectiveness of the TADFmaterials for a novel EL mechanism. In this Communication, we report the first observation of TADF under electrical excitation. Although the overall EL efficiency is still very low, the observations are convincing that TADF is a possible pathway for high EL efficiency, even using fluorescent materials. Furukawa and Igarashi reported that a complex of tin(IV) chloride and coproporphyrin III (SnCl2–Copro III) exhibited strong TADF, and it was applied to an oxygen sensor, based on the dependence of the TADF lifetime on the oxygen concentration.[11] On the basis of this previous report, we have searched for more efficient TADF materials and found six types of tin(IV) H & Co. KGaA, Weinheim Adv. Mater. 2009, 21, 4802–4806 C O M M U N IC A T IO N www.advmat.de while weak room temperature phosphores- cence was also observed around 700 nm. With an increase of temperature from 30 to 100 8C, the emission intensity significantly increased, which indicates an acceleration of the reverse ISC from T1 to S1 excited states caused by heat activation. The photophysical characteristics of each SnF2–porphyrin complex are summar- ized in Table 1; all derivatives exhibited intense TADF. The S1–T1 thermal energy gap, which is necessary in order to induce the reverse ISC, was estimated to be 37–40 kJ mol�1 (ca. 0.4 eV) from the peak wavelengths of the fluorescence fluoride–porphyrin complexes (SnF2–porphyrin): octaethylpor- phine (OEP), etioporphyrin I (Etio I), hematoporphyrin IX (Hemato IX), protoporphyrin IX (Proto IX), mesoporphyrin IX (Meso IX), and Copro III (Fig. 2). These SnF2–porphyrin complexes exhibited rather strong TADF at room temperature, and also displayed a clearly visible increase in fluorescence intensity with increasing temperature. Figure 2. Molecular structures of the six Sn(IV)–porphyrin complexes. The photoluminescence spectra of the SnF2–Meso IX complex dispersed on filter paper are shown in Figure 3. Strong fluorescence and TADF were observed around 550–650 nm, no phosphorescen T¼ 400K. The TA with the previous reverse ISC cause of each process in using an integrat system. A promp observed between slightly above T¼ similar manner, (FPHOS) displayed other hand, the (FTADF) increased Figure 3. Photoluminescence spectra for the SnF2–Meso IX complex at T¼ 0 8C (a), 30 8C (b), and 100 8C (c) on a filter paper. Inset: Energy diagram for SnF2–OEP. Table 1. Photophysic paper dispersion). Porphyrin l OEP Etio I Hemato IX Proto IX Meso IX Copro III [a]Fluorescence and TAD [c] [c]Singlet and triplet Adv. Mater. 2009, 21, 4802–4806 � 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Wein ce was observed at the elevated temperature of DFand phosphorescence behavior is consistent report[9] of TADF due to the acceleration of the d by heat activation. The quantum efficiencies SnF2–OEP (Fig. 5) were carefully estimated ed sphere photoluminescence measurement t fluorescence efficiency (FF) of 0.6% was T¼ 5K and 300K, which then decreased 300K, resulting inFF¼ 0.4% at T¼ 400K. In a the phosphorescence quantum efficiency a gradual decrease above T¼ 300K. On the quantum efficiency of delayed fluorescence steeply from 0.6% at T¼ 300K to 2.4% at al characteristics of SnF2–porphyrin complexes (filter F [a] [nm] lP [b] [nm] DE [c] [kJ mol �1] 571 701 38.8 569 701 39.5 576 698 36.5 579 708 37.6 570 703 38.8 571 701 38.9 Fpeak wavelengths. [b] [b]Phosphorescence peak wavelength. exciton energy difference. and phosphorescence spectra. These intense TADF characteristics were anticipated to indicate suitability of the material as the dopant in the OLED architec- ture; therefore, polymer-dispersed films were fabricated that contained the SnF2–OEP com- plex for use as a dopant in the emitting layer. A polymer dispersion film of 2 wt% SnF2–OEP:PVCz was prepared, where PVCz is poly(vinylcarbazole), and the emission spectra and streak images at T¼ 200, 300, 350, and 400K are shown in Figure 4. The streak images provide visual indications of the intensities of the prompt and delayed fluorescence components. The intense emissions around t¼ 0 s correspond to the prompt component and the long tail corresponds to the delayed components, TADF and phosphorescence. Although only a weak TADF component was observed at low temperatures, below T¼ 200K, intense TADF emission was observed at T¼ 400K. On the other hand, phosphorescence was only observed at lower temperatures and heim 4803 C O M M U N IC A T IO N www.advmat.de lm res 4804 T¼ 400K. Thus, the overall photoluminescence quantum efficiency was significantly enhanced above T¼ 300K. TADF was also confirmed in the ITO/PEDOT/2 wt% Figure 4. Streak images and emission spectra of 2wt% SnF2–OEP:PVCz fi 300, and 200 K. Photoluminescence spectra are resolved into prompt fluo phosphorescence components. SnF2–OEP:PVCz/MgAg/Ag device (ITO is indium tin oxide and PEDOT is poly(3,4-ethylene dioxythiophene)) under electrical excitation. The device required a slightly higher driving voltage, owing to insufficient optimization of device parameters such as film thickness and carrier injection barriers, which resulted in a rather high onset of current injection around 10V, requiring 29V at J¼ 100mA cm�2. Figure 6 shows the temperature dependence of the transient EL spectra with streak images at T¼ 300, 350, and are independent temperature depe of 5� 101 s–1 at T kP¼ 0.14 s–1 is sig appreciable TADF further enhanced However, the kRI up-conversion. Sin kB T), the Arrhen shown in the in estimated to be difference in the inset of Figure 3. T have a smaller en This study cle pathway to prom enhance overall E cence efficiency purpose, some g molecules should deactivation proce heavy atoms, met mutual intersyste heavy-atom effects the most critical i could be accompli Figure 5. Dependence of the total photoluminescence (FPL), fluorescence (FF), phosphorescence (FPHOS), and TADF (FTADF) quantum efficiencies on temperature. � 2009 WILEY-VCH Verlag GmbH & Co. KGaA, Weinhe are the fluorescence and phosphorescence , respectively. It was found that kISC, kR, and kP of temperature, while kRISC exhibited strong ndence. Although kRISC has a very low value ¼ 300K, the phosphorescence decay rate of nificantly lower than kRISC, which resulted in an , even at T¼ 300K. At T¼ 400K, kRISC was up to 5.5� 102 s–1, resulting in intense TADF. SC value is still too small to induce efficient ce kRISC should be proportional to exp(�DEST/ ius plot of kRISC versus 1/T was examined, as set of Figure 7. The activation energy was DEST¼ 0.24 eV, which corresponds to the energy levels of T1 and S1, as indicated in the hus, in order to increase kRISC, it is necessary to ergy gap. arly demonstrated that TADF is a possible ote T1 to S1 states. However, in order to L efficiency, molecules having high fluores- must be provided with high kISC. For this uidelines have been determined. Firstly, the have a rigid structure to minimize thermal sses. Secondly, the complex should contain als or halogens, to promote single and triplet m exchange. It is probable that external may also be effective in TADF. Furthermore, ssue is the preparation of a small DEST, which shed by considering the correlation of electron 400K under short-pulse excitation. This is a clear first demonstration of both prompt and delayed EL components. At T¼ 400K, the ratio of the TADF component to the prompt component significantly increased, resulting in a TADF intensity approximately 5 times higher than that at T¼ 300K. Finally, we discuss the rate constant of reverse ISC (kRISC) in SnF2-OEP. Figure 7 summarizes the temperature dependence of the rate constants kISC (intersystem crossing from S1 to T1), kR (radiative decay rate from S1), kRISC (reverse ISC from T1 to S1), and kP (phosphorescence decay rate), as well as DEST (activation energy, inset) of kRISC, as expressed by kISC ¼ FP tFkPtP (1) kR ¼ FF tF (2) kRISC ¼ FTADF kISCt2FtPkR (3) kP ¼ FP tP (4) where tF and tP transient lifetimes s at T¼ 400, 350, cence, TADF, and im Adv. Mater. 2009, 21, 4802–4806 C O M M U N IC A T IO N www.advmat.de orbitals in the triplet state, that is, by minimizing electro- n–electron repulsion due to the electron exchange term (J). The magnitude of J is given by the matrix element J ¼ ZZ f1 � ðr1Þf2ðr2Þ e2 4p"0 r1 � r2j jf2ðr1Þf1 � ðr2Þdr1dr2 (5) where f1 and f2 represent ground and excited-state wave functions, respectively, e is the charge, and e0 is the dielectric constant. Therefore, a small overlap integral of f�1 f2j � � is necessary to obtain a smallDEST. For example, a smallDESTcan be realized by the introduction of nonbonding molecular orbitals, such as n–p* transitions.[10] In summary, TADF has been demonstrated for six Sn4þ–porphyrin complexes. The TADF intensities significantly increased with temperature, due to acceleration of the reverse ISC Figure 6. Streak images and electroluminescence spectra of TADF in the ITO/PEDOT/2wt% SnF2–OEP:PVCz/MgAg/Ag device at T¼ 300, 350, and 400 K. Adv. Mater. 2009, 21, 4802–4806 � 2009 WILEY-VCH Verlag G from triplet to singlet excited states by heat activation. Under application of short electrical pulse excitation, prompt and delayed EL components were clearly observed. The delayed component was composed of both TADF and phosphorescence, and the TADF component significantly increased with an increase Figure 7. Temperature dependence of the decay rates: kISC: intersystem crossing, kF: radiative decay from S1, kISC: reverse intersystem crossing, and kP: radiative decay from T1. of temperature. It is thought that TADF will provide a novel pathway for efficient EL through the design of appropriate molecules that fulfill the discussed requirements. Experimental All SnF2–porphyrin complexes were synthesized using a modification of Furukawa’s method [11]: tin(II) fluoride and the porphyrin were dissolved in dimethyl sulfoxide (DMSO) and then heated at 150 8C for 5 h. Tin(II) was oxidized to tin(IV) through the complexation reaction. The mixture was then cooled and poured into 1% aqueous HF solution; the precipitated complex was collected by filtration. The crude complex was dried under reduced pressure and recrystallized fromMeOH/CH2Cl2 to produce a pure SnF2–porphyrin complex. For rapid confirmation of solid-state TADFmeasurement, the porphyrin complex (10mg) was dissolved in 10% MeOH/CH2Cl2 (10mL), and this solution (100mL) was dripped onto a filter paper and dried. Molecularly doped polymer films composed of the 2 wt%-SnF2–OEP complex as a dopant and PVCz as a host matrix were also prepared. Fluorescence, phosphorescence, and TADF characteristics were measured under vacuum using a streak camera system (C4334, Hamamatsu Co.). A nitrogen gas laser with an excitation wavelength of 337 nm was used. For electrical excitation, an OLED was prepared, which was composed of ITO (100nm)/ poly(3,4-ethylene dioxythiophene)-poly(styrenesulfonate) (PEDOT-PSS) (40 nm)/2wt% SnF2–OEP:PVCz (100nm)/MgAg (100nm)/Ag (10 nm) with SnF2OEP as an emitter. Here, we used a guest–host system with a dopant concentration of 2 wt% to minimize the concentration quenching. The details of device fabrication and measurement procedures have been previously reported [12]. In the present experiment, short-pulse electrical mbH & Co. KGaA, Weinheim 4805 C O M M U N IC A T IO N www.advmat.de excitation with a pulse width of 100ms was employed, and both the prompt and delayed electroluminescence were analyzed using a streak camera system. Acknowledgements We sincerely thank the Inamori Foundation and the Global COE (Centers of Excellence) Program ‘‘Science for Future Molecular Systems’’ of the Ministry of Education, Culture, Sports, Science and Technology (MEXT) for their financial support. Received: March 23, 2009 Published online: August 12, 2009 [1] J. Kalinowski, J. Phys. D: Appl. Phys. 1999, 32, R179. [2] M. A. Baldo, D. F. O’Brien, Y. You, A. Shoutstikov, S. Sibley, M. E. Thompson, S. R. Forrest, Nature 1998, 395, 151. [3] M. A. Baldo, S. Lamansky, P. E. Burrows, M. E. Thompson, S. R. Forrest, Appl. Phys. Lett. 1999, 75, 4. [4] C. Adachi, M. A. Baldo, S. R. Forrest, M. E. Thompson, J. Appl. Phys. 1999, 11, 285. [5] M. Ikai, S. Tokito, Y. Sakamoto, T. Suzuki, Y. Taga, Appl. Phys. Lett. 2001, 79, 156. [6] M. A. Baldo, C. Adachi, S. R. Forrest, Phys. Rev. B 2000, 62, 10967. [7] B. Valeur, Molecular Fluorescence, Wiley-VCH, Weinheim 2002, p. 41. [8] Y. Nichikawa, K. Hiroaki, Y. Onoue, K. Nishikawa, Y. Yoshitake, T. Shige- matsu, Bunseki Kagaku 1983, E115, 32. [9] S. K. Lam, D. Lo, Chem. Phys. Lett. 1997, 281, 35. [10] N. J. Turro, Modern Molecular Photochemistry, University Science Books, Sausalito, CA 1991. [11] M. Furukawa, S. Igarashi, presented at the 81st Spring Mtg. of the Chem. Soc. Jpn. 2002, No. 2F7-30. [12] J. A. Mikroyannidis, L. Fenenko, M. Yahiro, C. Adachi, J. Polym. Sci, Part A: Polym. Chem. 2007, 45, 4661. 4806 � 2009 WILEY-VCH Verlag Gmb H & Co. KGaA, Weinheim Adv. Mater. 2009, 21, 4802–4806