Stable Luminescent Radicals and Radical-Based LEDs with Doublet Emission

Open AccessCCS ChemistryMINI REVIEW1 Aug 2020Stable Luminescent Radicals and Radical-Based LEDs with Doublet Emission Zhiyuan Cui, Alim Abdurahman, Xin Ai and Feng Li Zhiyuan Cui State Key Laboratory of SupramolecularStructure and Materials, College of Chemistry, Jilin University, Changchun 130012 , Alim Abdurahman State Key Laboratory of SupramolecularStructure and Materials, College of Chemistry, Jilin University, Changchun 130012 , Xin Ai State Key Laboratory of SupramolecularStructure and Materials, College of Chemistry, Jilin University, Changchun 130012 and Feng Li *Corresponding author: E-mail Address: [email protected] State Key Laboratory of SupramolecularStructure and Materials, College of Chemistry, Jilin University, Changchun 130012 https://doi.org/10.31635/ccschem.020.202000210 SectionsAboutAbstractPDF ToolsAdd to favoritesDownload CitationsTrack Citations ShareFacebookTwitterLinked InEmail Radical-based light-emitting diodes are an innovative type of organic light-emitting diodes (OLEDs), which adopt luminescent radicals as emitters, aiming at improving the external quantum efficiency (EQE) of OLEDs. Research on luminescent radicals and the corresponding devices is a multidisciplinary field involving organic chemistry, solid-state physics, photochemistry, electronics, and others. The relevant theories have been established step-by-step in recent years. Considering the rapid development tendency, it is time to systematically summarize and discuss the doublet emission mechanism in devices, as well as the design principles of stable luminescent radicals. It is also of great importance to point out future developments in this area. In this minireview, the relevant advancements on radical-based OLEDs and stable radicals, especially those possessing luminescent properties, are summarized, respectively. Besides, the doublet emission mechanism is elaborated systematically. Some novel findings on stable luminescent radicals are also introduced. Finally, future developments and challenges in this field are presented. Download figure Download PowerPoint Introduction Organic light-emitting diodes (OLEDs) have been regarded as one of the most competitive products for panel display and solid-state lighting applications due to their superior properties such as flexible, light weight, easy to process, and so on.1–3 In OLEDs, much attention has been paid to the exploitation of luminescent materials, which determine the performance of OLEDs to a large extent. For conventional closed-shell luminescent materials, the electrical excitons are divided into 25% singlet excitons and 75% triplet excitons, according to the spin-statistics.4 It is well known that only singlet excitons are spin-allowed for radiative decay, while with most triplet excitons, energy is dissipated as heat, leading to an upper limit of 5% external quantum efficiency (EQE) after considering the light outcoupling efficiency ∼20%.5 Thus, numerous solutions have been proposed to utilize the nonemissive triplet excitons such as noble-metal-complex-based phosphorescent OLEDs (PhOLEDs),6–8 triplet–triplet annihilation (TTA),9,10 hybridized local and charge-transfer (HLCT) states,11,12 and thermally activated delayed fluorescence (TADF).13–15 The relevant photophysical processes are displayed in Figure 1. The internal quantum efficiency (IQE) of OLEDs was successfully improved to almost 100%, thanks to these great efforts. Figure 1 | (a) Generation process and the proportion of singlet and triplet excitons under electrical excitation. (b) Energy diagram of conventional closed-shell molecules. Photophysical processes of various approaches to utilize triplet excitons are marked in red. Abs, F, VR, NR, P, IC, RISC represent absorption (Abs), fluorescence (F), vibrational relaxation (VR), nonradiative relaxation (NR), phosphorescence (P) internal conversional (IC), and reverse intersystem crossing (RISC), respectively. Download figure Download PowerPoint Different from closed-shell molecules, luminescent monoradicals belong to a family of open-shell molecules. The outermost molecular orbital of monoradicals is occupied only by one electron,16 whose spin configuration is doublet. The terms of singlet, doublet, triplet, etc. refer to the spin degeneracy of the research objects. Examples, considering a closed-shell molecule whose excited state contains two unpaired electrons, one lies on the highest occupied molecular orbital (HOMO) and the other lies on the lowest unoccupied molecular orbital (LUMO), the two unpaired electrons become the research object, as follows: if the two electrons possess opposite spin directions, the total spin quantum number is 0, which has one projection value (0) along the Z direction (the direction of an external magnetic field, the blue line in Figure 2). Thus, the spin configuration is singlet; if the two electrons possess the same spin directions, the total spin quantum number is 1; it has three projection values (1, 0, and −1) along the Z direction. Hence, the spin degeneracy of this system is three, termed as triplet. Similarly, considering a monoradical, both the ground and first excited states have one unpaired electron, its spin quantum number is 1/2, which has two projection values (1/2 and –1/2) along the Z direction, the spin degeneracy is two, termed as doublet, as shown in Figure 2. Since both the ground and first excited states of monoradicals are doublet, there is no spin-forbidden transition. This kind of fluorescence coming from the doublet–doublet transition is termed as doublet emission. Figure 3 shows a representative sketch of doublet emission under photoexcitation and electrical excitation. For the photoexcitation, the electron on the HOMO-β is excited into the singly occupied molecular orbital (SOMO)-β, which is also termed as singly unoccupied molecular orbital (SUMO), and the fluorescence corresponds to a reverse process. For the electrical excitation, the electrons and holes are injected into SOMO and HOMO, respectively, forming doublet excitons. Then the radiative decay of doublet excitons produces fluorescence. It should be noted that, in addition to monoradicals, there exists biradicals, another kind of open-shell molecules, in which each molecule contains two unpaired electrons. For biradicals, the spin configurations of the ground and excited states are singlet or triplet because two unpaired electrons need to be considered, and the total spin quantum number is either 0 or 1. Figure 2 | Schematic diagram of the excited-state spin configurations of closed-shell molecules (a) and monoradicals (b). m represents the projection value of the total spin quantum number along the direction of the external magnetic field. Download figure Download PowerPoint Figure 3 | Doublet emission under photoexcitation (a) and electrical excitation (b). Abs, PL, EL represent absorption (Abs), photoluminescence (PL), and electroluminescence (EL), respectively. Download figure Download PowerPoint In 2007, there was a report about a triplet−triplet-transition fluorescence from a trimethylenemethane (TMM) biradical.17 In that work, the excited triplet states of TMM biradicals were created through γ-irradiation to break the covalent bond of a methylenecyclopropane. The emission originated from the triplet–triplet transitions. Although an OLED based on a methylenecyclopropane was shown, it was hard to confirm the circulation between bond breaking and bond recovery of the emitter under electrical excitation. The first doublet electroluminescence was demonstrated by our group,18,19 in which a neutral π monoradical (TTM-1Cz) was adopted as the emitter. OLEDs based on TTM-1Cz show a deep-red emission. To date, the EQE value of radical-based OLEDs has been dramatically improved by continuous molecular design and device optimization.20 Figure 4 shows the milestones of stable luminescent radicals and their applications in OLEDs. Figure 4 | Milestones in the development of stable luminescent radicals and doublet-emission OLEDs. The first triphenylmethyl radical was reported by Gomberg21 in 1900. In 1970, Ballester et al.22 synthesized the more stable radical PTM. In 1978, Armet et al.23 first synthesized TTM through the Friedel–Crafts reaction. In 2004, Heckmann et al.24 reported the first neutral organic mixed-valence compound, a PTM radical derivate. In 2006, TTM-1Cz with excellent stability as well as a high PLQE (53%) was reported by Gamero et al.25 In 2014, PyBTM (PLQE = 81% at 77 K) was reported by Hattori et al.26 In 2015, the first doublet-emission OLED was reported by Peng et al.19 The EQE value was dramatically improved by Ai et al.20 in 2018. In the same year, a novel diphenylmethyl luminescent radical CzBTM was synthesized by Ai et al.27 In 2019, non-Aufbau electron-structure radical PTM-3NCz was reported by Guo et al.28 In the same year, Abdurahman et al.29 reported the first luminescent polymer radical, PS-CzTTM. Download figure Download PowerPoint Although HLCT, TADF, and doublet-emission OLEDs have the similar result of promoting the IQE to 100%, the fundamental mechanisms are different. Therefore, it is meaningful and vital to study HLCT, TADF, doublet emission, and others to fulfill one's curiosity to gain insights into new mechanisms and their potential applications in the future. The study of luminescent radicals and corresponding devices have attracted much attention due to their low synthetic cost, short fluorescence life (nanosecond scale),and potential advantage in deep-red and near-infrared (NIR) emissions due to the natural narrow band gap. Some new-concept devices might come into being, based on luminescent radicals with integrated optical, electrical, and magnetic properties. In the following sections, we would introduce a cognition on stable organic radicals initially. Then the progress on various stable luminescent radicals [tris-2,4,6-trichlorophenylmethyl (TTM), perchlorotriphenyl methyl (PTM), and biphenylmethyl (BTM)] and OLEDs based on these radicals would be elaborated, respectively. Some academic valued findings and empirical summaries would be presented at the same time. Further, luminescent multiradical systems and other kinds of stable luminescent radical compounds would be introduced concisely. Finally, future developments and challenges in this field would be indicated. Cognition of Stable Radicals Stable radicals are unique kinds of functional materials, which have attracted more and more interest in various fields such as quantum teleportation,30 conductive polymers,31 biochemistry sensors,32,33 organic magnets,34,35 accelerating chemical reactions,36 and so on. The development of stable organic radicals could be traced as far back as 1900, during which Gomberg21 reported the first triphenylmethyl radical, which opened the door of the stable organic radical study. Generally, organic radicals are produced during the process of organic reactions, possessing high reactivity and challenging to isolate and purify. However, by rational molecular design, the stability of organic radicals could be improved dramatically and even be stored for an extended period under ambient conditions. After decades of development, now it is widely accepted that steric and delocalization effects usually play critical roles in tuning the stability of organic radicals.37,38 The steric effect can prohibit the dimerization of free radicals effectively, so does the reaction with other molecules such as oxygen or water, which are realized by introducing large protecting groups as halogen atoms or alkyls. Besides, for polyaromatic radical systems, the spin delocalization effect is beneficial to narrow the energy gap, decreasing the entire energy and reactivity of radical systems. Sometimes, those two effects coexist. Figure 5 shows the molecular structures of some common stable organic radicals. Figure 5 | Chemical structures of some common stable organic radicals. Download figure Download PowerPoint For a long time now, it is generally considered that stable radicals are not emissive and usually quench the emission of other luminous materials. For example, nitroxide radicals are excellent fluorescent quenching agents for aromatic hydrocarbons,39,40 which could be attributed to the electron exchange interaction or energy transfer process between the singlet excited state of the fluorophores and the doublet ground state of the nitrogen radicals. However, there exist rare stable organic radicals, which possess a bright doublet luminescence at room temperature, most of them are triaryl methyl radicals. To date, doublet-emission OLEDs are mainly confined in three kinds of luminescent radicals: TTM, PTM, and BTM. In the following sections, we would elaborate on the process systematically. TTM Radicals TTM belongs to the triphenylmethyl radical class, which is also the most investigated class of luminescent radicals. In 1978, Armet et al.23 first synthesized the TTM radical through the Friedel–Crafts reaction using 1,3,5-trichlorobenzene and chloroform, followed by dehydrogenation progress. This general synthetic method is still applied today. The structural formula and steric configuration of TTM are displayed in Figure 6. The Sp2 hybridization of the central carbon atom is beneficial to spin delocalization, the inside six chlorine atoms act as steric shields, and the outside three chlorine atoms prevent the formation of dimers.41 These factors make for the chemical stability of TTM. However, the photostability of TTM is rather poor, reflected by the poor half-time t1/2 of the luminous intensity attenuation, which is ∼200 s under 370 nm light irradiation in various solvents.26 Besides, the photoluminescence (PL) quantum efficiency (PLQE) of pure TTM is as low as 0.8% in carbon tetrachloride.42 Since 1994, Julia et al.43–45 reported the successive creations of a series of TTM derivates by replacing a peripheral chlorine atom with other substituents such as methoxyl, amino, nitro, and so on, aiming to tune the physicochemical properties of TTM radicals. No breakthrough was made until in 2006 by Gamero et al.25, who creatively introduced carbazole into TTM, forming TTM-1Cz radical, with significant improvement of the radical photostability, the t1/2 of the TTM-1Cz radical increased at least tenfold under the same test conditions. Besides, this radical also exhibited outstanding red-emission performance, with the PLQE reaching up to 53% in cyclohexane. After that, many functional groups such as thiophene and phenylene were introduced to TTM to improve the electron or hole transport abilities.46–48 These excellent works lay a foundation for the doublet-emission OLEDs. Figure 6 | Chemical structure and steric configuration of TTM. Download figure Download PowerPoint Based on the progress of TTM derivatives, the first doublet-emission OLED was reported by us.19 The device structure and performance are shown in Figure 7. TTM-1Cz doped 4,4′-bis(N-carbazolyl)-1,1′-biphenyl (CBP) was used as the luminescent layer to avoid the aggregation-caused quenching (ACQ).49 The electroluminescence (EL) peak wavelength lies at 692 nm, and 2.4% EQE was obtained. The unique doublet emission was proved by a magnetoelectroluminescence measurement, which demonstrated the feasibility of doublet emission to circumvent the triplet-excitons-harvesting problem in traditional closed-shell molecules. According to the radiative transition probability calculation formula,4 shown in Equation (1). R ij 2 ∝ | 〈 ψ ei | M | ψ ej 〉 | 2 | 〈 χ vi | χ vj 〉 | 2 | 〈 ψ si | ψ sj 〉 | 2 (1)where ψei and ψej correspond to the electron wave functions before and after the electron transition. M means the dipole moment operator. χvi and χvj refer to the vibrational state wave functions. ψsi and ψsj denote the spin wave functions. The three terms correspond to the dipole selection rule, Franck–Condon factor, and the spin selection rule, respectively. The electron transition will be forbidden if any of the three terms is close to zero. For open-shell radicals, the electron-spin transition is totally spin-allowed. Thus, to enhance the optical performance of luminescent radicals, much attention has been paid to promote the first and second terms, which are strongly associated with molecular structures. By rational molecular design and successively device optimization, the performance of OLEDs based on TTM radicals was gradually improved.50–53 An EQE of 27% @ 5.2 × 10−4 mA cm−2 was obtained and the EQE values maintained higher than 10% over a wide range (5.2 × 10−4–1.2 mA cm−2; 7.6 × 10−4–6.9 × 10−1 W Sr−1 m−2), which could serve as evidence that doublet-emission OLEDs break through the spin-statistic limit of traditional fluorescent OLEDs, and the charge-transfer doublet emission was confirmed.20 The progress of OLEDs based on TTM radicals is summarized in Table 1. Figure 7 | Device performance based on TTM-1Cz. (a) Device architecture diagram. (b) EL spectrum (7 V) of the OLEDs accompanied by the PL spectra of the doped thin film. The inset shows a photograph of the TTM-1Cz-based OLEDs operating at 7 V. (c) The EQE of the OLEDs as a function of voltage. The inset shows the lifetime of the excited states of TTM-1Cz in a toluene solution. (d) The J–V–L characteristics of the OLEDs. Reprinted with permission from ref 19. Copyright 2015 Wiley-VCH. Download figure Download PowerPoint Table 1 | Summary of the OLEDs Performance Based on TTM Series Radicals Radical Molecular structure λPL (nm)a λEL (nm)b PLQE (%)a EQEmax (%)c CIE (x, y) References TTM-1Cz 678 692 58 2.4 (0.69, 0.29) 19 and 50 TTM-2ID 620d 650 22 2.4 (0.67, 0.30) 51 TTM-2Bi 603 600 39 5.4 (0.60, 0.38) 52 TTM-3Bi 606 605 36 4.1 (0.62, 0.37) 52 TTM-DACz 605 608 57 10.6 (0.62, 0.36) 53 TTM-3PCz 695 703 60 17.0 (0.72,0.29) 20 TTM-3NCz 708 710 86 27.0 (0.72, 0.28) 20 aRadicals doped into CBP/NPB/TPBi films. bThe maximum EL wavelength. cThe maximum EQE. dMeasured in n-hexane solution. The empirical summaries of TTM radicals are listed as follows: (1) Compared with the localized state, the charge-transfer (CT) state is proved beneficial to acquire a more efficient luminescence for TTM radicals. That requires a construction of donor–acceptor (D–A) radical system, realized by connecting some functional substitute groups to TTM. One recent work has given the reason that D–A type radicals have higher emission efficiency. It is breaking the alternate symmetry of molecular structures that lifts the energy degeneracy between HOMO–SOMO and SOMO–LUMO, which causes higher D1 transition dipole moment leading to roughly 20-fold increases in PLQE.54 (2) In terms of a D–A radical system, the radical center usually acts as an electron acceptor. According to the density functional theory (DFT) calculation, the excitation process corresponds to an electron transfer from the connected substituent groups (D) to the center carbon (A). (3) The electron-donating capacity of the connected substituent groups largely determines the stability and optical property of TTM radicals.51,52,55 The increase in the degree of electron-donation results in a more stable radical system, narrowed band gap, and red-shifted emission. On the contrary, electron-withdrawing groups usually go against the stability of TTM radicals. (4) The geometric symmetry of the connected substituent groups influences the molecular PLQE to a certain extent.55 Global symmetry is found to be unfavorable for the radiative transition, reflected in the low radiative transition rate kr and high nonradiative transition knr, leading to a relatively low PLQE, while the asymmetry has a positive effect. (5) The precursors of TTM radicals can be used as the host material because they have similar molecular structures,50 which are conducive to the dispersion of dopants and the combination of electrons and holes in the emission layer to improve the device performance. PTM Radicals Compared with TTM, PTM was studied a little earlier, which possessed six more chlorine atoms. In 1970, Ballester et al.22 reported the synthesis of PTM through the perchlorination of triphenylmethane with reagent BMC (SO2Cl2, AlCl3, S2Cl2). Pure PTM radical possesses a weak orange-red light in nonpolar solvents, and the PLQE is very low, only 1.5% in cyclohexane,56 which is similar to that of TTM. The initial studies on PTM radicals were mainly focused on the reversible redox properties and photophysical properties of the anion and cation counterparts. In 2004, Heckmann et al.24 reported the first neutral organic mixed-valence (MV) compound by introducing a triphenylamine (TPA) derivate to the PTM radical. Subsequently, they investigated optically induced electron-transfer processes of a series of PTM radical derivates in detail in 2007.57 It should be noted that PTM radicals are not suitable for the vacuum evaporation method used to prepare OLED devices. The main reason is that the overmuch chlorine atoms cause loss of chlorine atoms during vacuum evaporation. Besides, it is a common phenomenon that PTM radicals are easy to degrade under high-intense light conditions.58 The plausible degradation process is illustrated in Figure 8. Figure 8 | Plausible degradation process of PTM radical. Download figure Download PowerPoint One advantage of PTM is that it is easier to change one or two of the peripheral chlorine atoms into bromine atoms by chemical means than TTM, making it conducive to modify the molecular structure with various substituent groups with higher yields and then exploring their applications and other aspects.59,60 Recently, one crucial discovery of the PTM derivatives is that the Aufbau principle could be violated if the substitute groups introduced to the PTM possess a rather strong electron-donating ability.28 In 1923, the Aufbau principle was proposed by Niels Bohr, who emphasized that "a maximum of two electrons are put into orbitals in the order of increasing orbital energy."61,62 However, some molecules violating the Aufbau principle were reported from time to time, and a common characteristic was the tremendously enhanced stability.63–65 This phenomenon has been observed with stable radicals for a long time but did not receive enough attention. For non-Aufbau electron-structure radicals, the energy level of the SOMO orbital lies below the HOMO, which is doubly occupied.66 This unique electronic structure reduces the chemical activity of the single electron reasonably, promising high photochemical stability. In 2018, by introducing different electron-rich groups, Guo et al.28 synthesized some non-Aufbau electron-structure TTM and PTM radicals successfully. The inversion of the orbital levels between SOMO and HOMO was confirmed by the cyclic voltammetry (CV), ultraviolet photoelectron spectroscopy (UPS) measurements, and DFT calculation. OLED device based on one of those radicals, PTM-3NCz, was fabricated through the spin-coating method. Compared with PTM, PTM-3NCz possesses an excellent photostability and an up to 54% PLQE in cyclohexane. The chemical structure, frontier molecular orbitals distribution of PTM-3NCz, as well as the photostability tests, are displayed in Figure 9. The OLEDs based on PTM-3NCz showed a deep-red/NIR emission with a maximal EQE of 5.3%. We noticed that there was a similar work about the non-Aufbau electron-structure TPA-R• radical reported by Tanushi et al.67 almost at the same time. These findings provide a practical strategy to synthesize highly stable luminescent radicals. Figure 9 | Photophysical properties of PTM-3NCz. (a) Chemical structure. (b) Frontier molecular orbitals. (c) Photostability comparison between PTM and PTM-3NCz. (d) Photographs of PTM and PTM-3NCz in dilute cyclohexane solution under a 365 nm UV lamp as a function of time. Download figure Download PowerPoint BTM Radicals BTM refers to a novel class of BTM luminescent radicals. In 2014, Hattori et al.26 tactfully replaced one benzene ring of TTM radical by a para-substituted pyridine, forming PyBTM radical. The relevant molecular structures of PyBTM radicals68–72 are summarized in Figure 10. Figure 10 | Chemical structures of some reported PyBTM series radicals. Download figure Download PowerPoint Strictly speaking, PyBTM still belongs to the category of triaryl methyl radicals, because its chemical structure has a similar skeleton to TTM radical. However, the physical and chemical properties are significantly different. The most direct manifestation is the improvement in its stability. Although at room temperature, the PLQE of PyBTM in dichloromethane is only 2%, it could reach up to 81% in EPA solvent at a low temperature (77 K).26 Different halogen-substituted PyBTM radicals were also synthesized and lucubrated by Hattori et al.69 in 2015. With the change of F–Cl–Br, the fluorescence emission of the radicals varies from yellow to orange then to red (Figure 11) and the photostability is improved too. Due to the introduction of pyridine, PyBTM series radicals have shown their potential applications in proton and metal ions response. A series of PyBTM metal complexes were prepared in recent years, and the unusual magnetic properties also intrigued some researchers' interests.71 Figure 11 | Absorption and photoluminescence spectra of different halogen atom substituted PyBTM radicals. Reprinted with permission from ref 69. Copyright 2015 Royal Society of Chemistry. Download figure Download PowerPoint A breakthrough on BTM radical was achieved by Ai et al.27 in 2018. A novel luminescent BTM radical, CzBTM, was synthesized by a different reaction, shown in Figure 12. One of the benzene rings of the TTM skeleton was first replaced by a carbazole. Research on CzBTM is still at an initial stage; the reaction mechanism is not precise. CzBTM exhibits a deep-red to NIR emission in various solvents, and the PLQE is relatively low, which is 2.0% in cyclohexane, and the performance of OLEDs based on CzBTM is unsatisfactory. Figure 12 | Synthetic process of CzBTM. Download figure Download PowerPoint In the same year, Abdurahman et al.73 synthesized PyID-BTM by replacing carbazole with a beta-carboline. The PLQE and the OLEDs' performance were dramatically enhanced although there was only one atom difference between the CzBTM and PyID-BTM. The primary cause is that the introduction of nitrogen atom changes the transition moments of molecular orbitals, resulting in a higher radiative transition rate kr and a lower nonradiative rate knr, which are necessary for highly efficient luminescence. The comparison of the photophysical properties of the two radicals is displayed in Figure 13 and Table 2. A supplementary work was carried out by modifying CzBTM with different halogens on the third and sixth substituted positions of carbazole, but no obvious distinction was observed, though the optical properties appeared to have some regularity.74 Figure 13 | Absorption and photoluminescence spectra of CzBTM and PyID-BTM in cyclohexane solution (10−6 M). The relevant chemical structures are displayed in the right. Download figure Download PowerPoint Table 2 | Photophysical Properties and Device Performances of CzBTM and PyID-BTM. Radical λabs/nm(ɛ/M−1cm−1)a λPL (nm)a PLQE (%)b τ (ns)c λEL (nm)d EQEmax (%)e CIE (x, y) References CzBTM 284(1.42 × 104)/387(8.99 × 103)514(2.24 × 103)/554(2.47 × 103) 697 2.0 4.0 700 0.66 (0.732, 0.261) 27 PyID-BTM 260(4.36 × 105)/383(2.11 × 104)510(5.17 × 103)/550(5.29 × 103) 664 19.5 12.8 660 2.8 (0.649, 0.317) 73 aMeasured in cyclohexane at room temperature. bMeasured by an integrating spectrophotometer. cMeasured by Edinburgh fluorescence spectrometer (FLS980) under laser excitation at 378.8 nm. dThe maximum EL wavelength. eThe maximum EQE. Luminescent Stable Bi-, Tri-, Multi-, and Polymer Radicals To date, research on stab