Ryoga
Hojo
*,
Don M.
Mayder
and
Zachary M.
Hudson
*
Department of Chemistry, The University of British Columbia, 2036 Main Mall, Vancouver, British Columbia V6T 1Z1, Canada. E-mail: zhudson@chem.ubc.ca; Fax: +1-604-822-2847; Tel: +1-604-822-3266
First published on 12th May 2022
Three luminescent donor–acceptor compounds were prepared based on the Dimroth rearrangement of tris(triazolo)triazines (TTT). In comparison to the non-rearranged TTT isomers, the Dimroth isomer (TTTD) exhibits a substantial blue-shift in emission while maintaining thermally activated delayed fluorescence (TADF) properties. Out of the series of emitters, TTTD-3HMAT exhibits deepest blue emission with CIE(x, y) < (0.16, 0.03) and unity quantum yield in toluene, a narrower emission band, and highest two-photon cross-section of 1001 GM owing to the nature of the planarized HMAT donor. TTTD-3tBu also exhibits unity quantum yield and deep-blue emission with CIE(x, y) < (0.16, 0.05) in toluene. Finally, TTTD-3ACR exhibits TADF with blue-shifted emission and prolonged delayed lifetime due to the slightly larger ΔEST of 0.26 eV in comparison to the non-rearranged isomer. Overall, this work demonstrates a practical strategy to convert TTT-based donor–acceptor materials to their Dimroth isomers, opening the door to deeper blue-emitting TADF materials with TTT-type acceptors.
TADF behaviour is achieved by reducing the energy gap between a molecule's lowest excited singlet and triplet states (ΔEST), which can be achieved by lowering the electron exchange energy through the spatial separation of the HOMO and LUMO.1,20 In terms of donor–acceptor design, this process of minimizing ΔEST can be achieved by inducing a large dihedral angle between the donor–acceptor moieties.1,21 If ΔEST is sufficiently small (<0.3 eV), thermal energy enables upconversion from the triplet to singlet excited state, allowing for RISC. Despite this seemingly simple design strategy, minimizing ΔEST is often done at the cost of broad and red-shifted emission, due to the strong charge-transfer (CT) character of this donor–acceptor design.22,23 For this reason, designing efficient deep-blue emitting TADF materials remains one of the biggest challenges in this field.2,24,25
In recent years [1,2,4]-triazolo-[1,3,5]-triazine (TTT) has attracted significant attention in the realm of blue-emitting TADF materials due to their weak electron-withdrawing ability.26–30 The synthesis for this unique heterocyclic motif features cyclization of aryl tetrazoles onto cyanuric chloride under basic conditions.31,32 To date, several derivatives of TTT-based TADF materials have been successfully synthesized with respectable device performance.26–30 Our group has also revealed the remarkable two-photon cross-section of TTT-based emitters using hexamethylazatriangulene (HMAT) donors, which has great potential for biological imaging using two-photon excited fluorescence (2PEF).33
Following the synthesis of TTT reported in the original paper by Hoffman et al.,34 Tartakovsky and co-workers31 reported the thermal isomerization process of TTT derivatives, uncovering the Dimroth isomer, TTTD. Similarity in 2018, Detert explored the TTTD isomer in the design of discotic liquid crystals.35 The Dimroth rearrangement involves thermal ring-transformation of 1,2,3 triazoles,36,37 whereas for TTT derivatives, the process involves a threefold ring-opening and ring-closing mechanism (Scheme 1).31,35 More recently, the You group demonstrated a mono-fused (triazolo)triazine derivative undergoing the Dimroth rearrangement to produce efficient TADF emitters with minimal ΔEST (Scheme S1A, ESI†).38 Inspired by this work, we hypothesized that the Dimroth rearrangement would provide a unique opportunity to observe the effect of thermal isomerization on TTT compounds; and precisely observe the change in TADF behavior.
Scheme 1 The isomerization of TTT to TTTD.35 |
Herein we report the synthesis of three compounds, TTTD-3HMAT, TTTD-3tBu, and TTTD-3ACR, based on the Dimroth rearrangement of donor-acceptor materials bearing a TTT core. TTTD-3HMAT and TTTD-3tBu exhibit unity photoluminescence quantum yields (PLQYs) in both solution and the solid-state, with TTTD-3HMAT exhibiting deeper-blue emission and a higher two-photon cross-section (σ2) in toluene relative to TTTD-3tBu. Finally, TTTD-3ACR retains TADF character with blue-shifted emission compared to the non-rearranged TTT emitter. Indeed, TTTD-3ACR represents the deepest blue emitter displaying TADF behaviour to date based on the TTT acceptor motif.
Our method contrasts with the synthetic route reported by You and co-workers involving a Pd-catalyzed cross-coupling on the pre-assembled and Dimroth-rearranged (triazolo)triazine (Scheme S1A, ESI†). Our results demonstrate that thermal rearrangement can reliably occur after the donors are appended to the acceptor, providing a direct synthetic route without the use of palladium. Moreover, we also attempted a Pd-catalyzed coupling using a triiodide-functionalized Dimroth-rearranged TTT acceptor, for cross-coupling with amine-based donors; the insolubility of tris(p-iodophenyl)-TTTD made this method unfeasible and challenging to characterize (Scheme S1C, ESI†). We also attempted to synthesize and characterize TTTD derivatives bearing 10H-phenoxazine (TTTD-3POX) or 10H-phenothiazine donors (TTTD-3PTZ) by thermal rearrangement. In these cases, however, the products were highly insoluble in common organic solvents, likely due to extensive π-stacking (Scheme S1C, ESI†). No glass transitions could be observed for the three isolated compounds between 30–200 °C by differential scanning calorimetry (Fig. S16, ESI†).
The UV-visible absorption and photoluminescence (PL) spectra of TTTD emitters in toluene are given in Fig. 1A. The absorption maxima ranged from 298 nm to 391 nm, with TTTD-3HMAT and TTTD-3tBu exhibiting small Stokes shifts of 25 nm and 58 nm, respectively (Table 1). PLQYs of 1.0 were observed for both TTTD-3HMAT and TTTD-3tBu, while TTTD-3HMAT exhibited deeper blue emission (416 nm) and smaller FWHM (42 nm) in toluene, attributed to the restricted intermolecular motion of the planarized donors. In contrast, a weak charge-transfer absorption band is observed for TTTD-3ACR, with a broader emission profile and an emission maximum at 461 nm in toluene. In addition, an increase in PLQY from 0.26 under air to 0.41 in N2-sparged solution is observed for TTTD-3ACR in toluene, suggesting the likely involvement of triplet states in the emission mechanism. The prompt fluorescence PL decays in toluene (Fig. S7A, ESI†) were also measured using time-correlated single-photon counting (TCSPC): the lifetime for TTTD-3HMAT and TTTD-3tBu ranged from 2.1 to 2.6 ns, while the lifetime for TTTD-3ACR is slightly longer at 7.6 ns (Table 1). Multi-channel scaling (MCS) was then used to analyze the longer-lived PL decays in solution (Fig. S8A, ESI†). While TTTD-3HMAT and TTTD-3tBu show no observable delayed lifetime in both ambient and inert conditions, TTTD-3ACR exhibits a delayed lifetime of 19 μs in oxygen-free conditions, likely arising from delayed fluorescence.
Entry | Solution | 3 wt% in PMMA | ||||||||
---|---|---|---|---|---|---|---|---|---|---|
λ abs (nm) ε (104 cm−1 M−1) | λ em/FWHM (nm) | σ 740 2 (GM) | Φ air/ΦN2b/c | τ aira | τ N2 | λ em/FWHM (nm) | Φ air | τ aird | τ vac | |
a Measured in toluene under air at 0.01 mg mL−1. b Absolute photoluminescence quantum yields determined using an integrating sphere. c Measured in toluene solutions sparged with N2. d Measured under air at 298 K. e Measured under vacuum at 298 K. f Amplitude-weighted lifetime measurement via multiexponential tail fitting. | ||||||||||
TTTD-3HMAT | 298 (11), 391 (11) | 416/42 | 1001 | 1.00/1.00 | 2.1 ns | 2.3 ns | 425/56 | 1.00 | 2.7 ns | 2.7 ns |
TTTD-3tBu | 298 (4.1), 370 (10) | 428/58 | 845 | 1.00/1.00 | 2.6 ns | 2.6 ns | 432/60 | 1.00 | 3.5 ns | 3.5 ns |
TTTD-3ACR | 363 (0.5) | 461/74 | 83 | 0.26/0.41 | 7.6 ns | 11 ns/19 μs | 453/130 | 0.14 | 9.2 ns/107 μsf | 9.2 ns/13 msf |
Aggregation-induced emission (AIE) and solvatochromism experiments were conducted to investigate the nature of the CT properties of these three materials. AIE experiments were conducted using water/THF mixtures, with water fractions (fw) ranging from 0% to 99%. The emission intensity increased as the percentage of water increased for TTTD-3ACR, with an overall 170% enhancement at fw = 70% Fig. S9, ESI†). In contrast, TTTD-3HMAT and TTT-3tBu lack AIE character. A positive solvatochromic shift is observed for all emitters, from least to most polar solvents (Fig. S10, ESI†). However, a minimal solvatochromic shift is observed with TTTD-3HMAT and TTT-3tBu. In contrast, the significant solvatochromic shift observed for TTTD-3ACR is evidence of strong CT character, consistent with its TADF behaviour (Fig. S10, ESI†).21,39,40
The two-photon cross-sections (σ2) for TTTD compounds were examined over the range of 710–850 nm using 2PEF measurements (Fig. 1B). Using (E)-4,4′-(ethene-1,2-diyl)bis(N,N-diphenylaniline) in dichloromethane as a blue-emitting reference dye standard, all measurements were periodically checked to verify a square dependence of signal intensity with excitation power. The highest σ2 was observed with TTTD-3HMAT (σ7402 = 1001 GM), followed by TTTD-3tBu (σ7402 = 845 GM) and TTTD-3ACR (σ7402 = 83 GM). In agreement with previous reports, the structural constraints and planar nature of HMAT donors allow for superior two-photon cross-sections, where a high σ2 value are enhanced with higher planarity and symmetry of the molecule (Table 1).33,41–43
Photophysical properties of the emitters in the solid-state were analyzed by preparing poly(methyl methacrylate) (PMMA) films doped with 3 wt% of emitter. Blue-shifted emission is observed with TTTD-3ACR (453 nm) relative to the emission maximum in toluene (461 nm). Red-shifted emission is observed with TTTD-3HMAT and TTTD-3tBu, which can arise from aggregation due to greater π-stacking in these less-twisted materials.26 Moreover, PLQY values of unity were observed for TTTD-3HMAT and TTTD-3tBu in the solid state, while that of TTTD-3ACR reached 0.14. Under vacuum, the emission intensity of TTTD-3ACR rises 2.0-fold compared to an air atmosphere, while intensities for TTTD-3HMAT and TTTD-3tBu remain essentially unchanged (Fig. S13, ESI†).
PL decays were measured for PMMA-doped films using TSCPC to measure prompt fluorescence (Fig. S1B, ESI†), and MCS was used to analyze the longer-lived PL decays (Fig. S2B, ESI†). Lifetimes of TTTD-3HMAT and TTTD-3tBu were consistent under air or vacuum at 298 K, only exhibiting prompt fluorescence in each case. In contrast, TTTD-3ACR has a long-lived PL lifetime of 107 μs in air, further lengthened to 13 ms under vacuum (Table 1). To elucidate if the long-lived emission of TTTD-3ACR was due to TADF, temperature-dependent lifetime measurements were conducted in the range of 77 to 298 K. As predicted, suppression of delayed fluorescence was observed upon cooling from 298 K to 77 K (Fig. 2). However, the emergence of long-lived phosphorescence is also observed below 200 K, which can be attributed to the restriction of roto-vibrational motion and dominance of phosphorescence emission when reaching cryogenic temperatures. Time-gated and steady-state emission measurements at 77 K of TTTD-3ACR in 2-methyltetrahydrofuran further support TADF, with a measured value for ΔEST of 0.26 eV (Fig. S14, ESI†). Conversely, the larger values of ΔEST measured for TTTD-3HMAT (0.51 eV) and TTTD-3tBu (0.35 eV) likely preclude RISC.
The determination of EHOMO and ELUMO were identified using cyclic voltammetry (CV) and Tauc plots (Fig. S11, ESI†). For all emitters, quasi-reversible one-electron oxidation is observed, which is attributed to each donor moiety. An irreversible reduction event is shown for TTTD acceptors within the solvent window of o-difluorobenzene. Therefore, Tauc plots were necessary to obtain optical gaps (Egap) from UV-Vis absorption measurements, where the ELUMO ranged from −1.85 eV to −1.92 eV. The trend for Egap is in slight disagreement with the calculated values using DFT (Fig. S12, ESI†). Nevertheless, all emitters display a large Egap around 3.1 eV, expected for deep-blue emitting chromophores (Table 2). Finally, we provide a comprehensive comparison of the photophysical and electrochemical properties of these TTTD compounds and their TTT isomers in Table S8 (ESI†).
Entry | HOMO/LUMOab (eV) | E DFTgapc (eV) | E optgapd (eV) | ΔEST (eV) | |
---|---|---|---|---|---|
Calc.c | Exp.e | ||||
a HOMO = −(Eox1/2 + 4.8 eV). b LUMO = (Eoptgap + HOMO). c Theoretical calculations from TDA-DFT at the B3LYP/6-31g(d) level. d Calculated using a Tauc plot of the UV-Vis spectrum in toluene. e Determined from onsets of time-gated phosphorescence and fluorescence spectra measured in 2-MeTHF at 77 K. | |||||
TTTD-3HMAT | −4.97/−1.91 | 3.42 | 3.06 | 0.38 | 0.51 |
TTTD-3tBu | −5.01/−1.85 | 3.03 | 3.16 | 0.46 | 0.35 |
TTTD-3ACR | −4.98/−1.92 | 3.01 | 3.06 | 0.006 | 0.26 |
Density functional theory (DFT) at the B3LYP/6-31g+(d) level was used further to analyze the photophysical and electronic properties of the emitters (Fig. 3). Structural optimization was performed at the B3LYP/6-31g+(d) level, followed by the use of the Tamm–Dancoff approximation (TDA) for time-dependent calculations at the B3LYP/6-31g(d) level. TDA-DFT was selected for theoretical calculations, as it has been shown to be suitable for larger molecules that suffer from the triplet instability problem.44 While diffuse orbitals were used during structural optimization at the B3LYP/6-31g+(d) level, they were not included for TDA-DFT due to the higher computational demand for the large molecules presented here.
Fig. 3 Pictorial representation of optimized S0 geometries for TTTD emitters. The weight of the transition between HONTO and LUNTO is indicated above the arrow. |
In general, theoretical calculations yielded trends in agreement with experimental observations. The calculated ΔEST for TTTD-3ACR is the smallest at 0.006 eV, while both TTTD-3HMAT and TTTD-3tBu have larger calculated ΔEST values above 0.38 eV. Molecular orbital visualization through natural transition orbitals (NTOs) indicates poor spatial separation of frontier molecular orbitals of the donor and acceptor in TTTD-3HMAT and TTTD-3tBu. Consequently, the higher calculated oscillator strengths for the S0 to S1 transition support the stronger observed CT absorption, and subsequently the high PLQY for these materials as well (Table S1, ESI†). In contrast, the near-orthogonal predicted dihedral angle between the ACR donor and TTTD acceptor in TTTD-3ACR induces strong spatial separation between the highest-occupied NTO (HONTO) and lowest-unoccupied NTO (LUNTO). This likely explains the TADF behaviour of TTTD-3ACR, yet also limits strong absorption from S0 to S1 as indicated by the low oscillator strength of this transition (Fig. 3 and Table S1, ESI†).
Interestingly, a general comparison between TTTD and TTT isomers indicates that each isomer is of similar total energy, with a minimal predicted change in the energy of the S1 state between each isomer pair (Fig. S15 and Table S1, ESI†). In all three cases, the predicted energies of the S1 states of the TTT isomers were within 50 meV of the TTTD isomers, when simulated in toluene using a polarized continuum model (PCM).
Footnote |
† Electronic Supplementary Information (ESI) available: A synthetic scheme towards aryl tetrazoles; 1H and 13C{1H} NMR spectra, solvatochromic and AIE properties, nanosecond PL decay curves, cyclic voltammetry data, Tauc plots, time-resolved emission spectra and computational results for TTTD-3HMAT, TTTD-3tBu, TTTD-3ACR. See DOI: https://doi.org/10.1039/d2tc01153k |
This journal is © The Royal Society of Chemistry 2022 |