Masaki
Shimizu
*,
Takumi
Kinoshita
,
Ryosuke
Shigitani
,
Yusuke
Miyake
and
Kunihiko
Tajima
Faculty of Molecular Chemistry and Engineering, Kyoto Institute of Technology, 1 Hashikami-cho, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan. E-mail: mshimizu@kit.ac.jp
First published on 14th December 2017
We designed and characterized 1,4-diaroyl-2,5-bis(silylmethoxy)benzenes as precious-metal-free organic luminophores that efficiently phosphoresce at room temperature. The benzene derivatives in crystals emit green phosphorescence with quantum yields up to 0.45 under ambient conditions. The luminescence quantum yield increases with increasing number of intermolecular interactions in the crystal, such as hydrogen bonding and CH–π interactions. The luminescence lifetimes are inversely proportional to temperature over the −80 to 80 °C range, demonstrating the potential of the benzene derivatives as materials for temperature sensing. Poly(methyl methacrylate) films doped with these luminophores also exhibit intense green phosphorescence at room temperature under vacuum, while they emit very faint blue fluorescence under ambient conditions. Electron spin resonance spectroscopy of a UV-excited diphenylmethylsilyl-derivative in toluene at 77 K reveals a triplet diradical species, whose electronic distribution is similar to that of naphthalene, indicating that the triplet diradical is distributed over almost ten atoms.
To achieve efficient RTP, first, intersystem crossing (ISC) from the singlet excited state (S1) to T1 needs to be faster than internal conversion and fluorescence, both of which are spin-allowed transitions from S1 to S0. Since ISC is promoted by large spin–orbit coupling, the incorporation of heavy atoms such as iodine or bromine, precious metals such as iridium or platinum, or arylcarbonyl groups into luminophores effectively accelerates ISC and efficiently generates T1. While precious metals provide the additional advantage of accelerating the rate of phosphorescence, they are limited and expensive resources, and their complexes are toxic. Therefore, increasing attention is being paid to the development of precious-metal-free organic phosphors.4 However, as the rate of radiative decay (phosphorescence) of a metal-free phosphor is very slow, its RTP is easily quenched through non-radiative decay processes such as molecular vibrations that occur at room temperature, and collisions with the surrounding molecules. Accordingly, the construction of rigid environments that suppress possible intramolecular phosphor motion is the second requirement for achieving efficient RTP. Early approaches that satisfied this requirement involved the adsorption of phosphors onto solid substrates such as filter paper or silica gel, and the inclusion of phosphors into cyclodextrin or surfactants, although most of these studies did not disclose the luminescence efficiency of the resulting RTP.5 Recent approaches include the crystallization of phosphors6 and the co-crystallization of phosphors with low molecular weight molecules that fix the phosphor conformation through intermolecular interactions such as halogen- or hydrogen-bonding;7 these approaches have led to the development of several highly efficient RTP materials. The dispersion or cross-linking of phosphors in a polymer matrix is the third approach for facilitating efficient RTP.8 Considering that polymer films exhibit good processabilities that are not available to crystals and have found many bio- and chemical sensing applications,9 the development of phosphorescent polymers is also an intriguing subject of investigation from the viewpoint of practical applications. On the other hand, most phosphors that exhibit efficient RTP in crystals are non- or poorly phosphorescent when dispersed in polymer films. Therefore, the creation of a polymer film that exhibits efficient RTP with a metal-free phosphor as the dopant is an important frontier in the field of materials chemistry.
We recently demonstrated that crystals of 1,4-dibenzoyl-2,5-disiloxybenzenes 1 exhibit RTP with excellent quantum yields of 0.46–0.64 and persistent lifetimes of 76.0–98.3 ms (Scheme 1).10 Considering that T1 states with longer lifetimes, in theory, increase the chance of non-radiative de-excitation, the simultaneous realization of excellent luminescence efficiencies (Φ > 0.4) and long lifetimes (τ > 10 ms) during RTP is unusual. In other words, the efficient RTP of 1 indicates that the T1 state of 1 is very stable against radiationless relaxation. X-ray diffraction analysis of single crystals of 1 disclosed that each molecule in the crystal is locked through several intermolecular interactions, such as hydrogen bonding and CH–π interactions, leading to the inhibition of thermal vibrations. Based on electron spin resonance studies and time-dependent density functional theory (TD-DFT) calculations, we proposed that the T1 of 1 is generated as a triplet diradical species T1 (1), in which one radical develops over the oxygen of the silyl ether moiety, and is electronically stabilized by σ–n conjugation with the Si–C σ-bond attached to the ether oxygen. We then realized that reversing the arrangement of the Si and C atoms attached to the ether oxygen atom (see T1 (2) in Scheme 1) should also provide similar stabilization to the oxyradical in T1 by σ–n conjugation.11 Hence, we envisaged that 1,4-diaroyl-2,5-bis(silylmethoxy)benzenes 2 would serve as new RTP materials; indeed, this proved to be the case. Herein, we report the preparation, structures, and photophysical properties of 2, and demonstrate that crystals of 2 in air, and 2 doped in poly(methyl methacrylate) (PMMA) films under vacuum, exhibit green RTP with high quantum yields and persistent lifetimes.
Single crystals of 2a, 2b, and 2d, suitable for X-ray diffraction analysis, were obtained by recrystallization from a mixed solution of hexane and CH2Cl2;132a, 2b, and 2d adopt similar conformations in which the two Si–C–O linkages and the central benzene ring are arranged in an almost coplanar fashion, and the two aroyl groups are twisted in vertically opposite directions with respect to the central benzene ring. The molecular structure of 2d, as a representative example, is shown in Fig. 1; the structures of 2a and 2b are available in the ESI.†
Crystal packing diagrams of 2a, 2b, and 2d are shown in Fig. 2. Each molecule of 2a interacts with the surrounding molecules through eight hydrogen bonds (Fig. 2a), suggesting that the molecular motions/vibrations of 2a in the crystal are relatively restricted. A unit cell of the crystal of 2b contains two molecules, each of which forms only two hydrogen bonds (Fig. 2b), indicating that multiple molecular motion/vibration opportunities exist that result in the non-radiative decay of T1. Four hydrogen bonds and eight CH–π interactions are observed per molecule in the crystal of 2d (Fig. 2c); consequently, the conformation of 2d in the crystal is firmly locked. Thus, the number of intermolecular interactions which is an important factor in enhancing the conformational rigidity of the phosphor in the crystal varies according to the triorganosilyl group.
Fig. 2 Crystal packing diagrams of (a) 2a, (b) 2b, and (c) 2d. The dotted blue lines indicate intermolecular interactions. |
Crystals of 2a and 2c–2e are green luminescent under ambient conditions with good-to-high quantum yields (Table 1). On the other hand, no RTP was observed with crystalline 2b or 2f.15 It is remarkable that a quantum yield in excess of 0.4 and a lifetime longer than 10 ms are simultaneously exhibited by 2d. The excellent quantum yield (0.45) of 2d is similar to that (0.46) of 1 (SiR3 = SiPh3, Ar = C6H5) and much higher than that (0.14) of 1,4-dibenzoyl-2,5-dibromobenzenes.16 Accordingly, the Ph2MeSiCH2O group serves as a superior inducer of RTP in 1,4-diaroylbenzene-based π-conjugated systems. Fig. 4 displays a luminescence image of 2d in the crystal and the luminescence spectra of 2a and 2c–2e, which exhibit small bands at 426–440 nm and strong bands at 513–522 nm. Short and long luminescence lifetimes,17 which are of the order of nano- and milliseconds, respectively, are observed for each luminescent crystal. Similar photoluminescence features, involving two emission bands in the blue and green regions, with nano- and millisecond-order lifetimes, are observed in the luminescence spectra of 2 in 2-MeTHF at 77 K (Table 1 and Fig. S3, ESI†). Therefore, we conclude that the green emissions from crystals of 2 are due to phosphorescence, while the small bands at shorter wavelengths are due to prompt fluorescence. The quantum yields of crystalline 2a, 2b, and 2d, the crystal structures of which are described above, were determined to be 0.15, 0.02, and 0.45, respectively (Table 1). The magnitude of the quantum yield is related to the number of intermolecular interactions observed in the crystal-packed structures. This relationship clearly demonstrates that intermolecular interactions in the crystal are factors that suppress non-radiative decay processes, i.e., facilitation of efficient RTP.
2 | Crystals at 298 K | In 2-MeTHF at 77 K | In PMMA films at 298 K | |||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Under vacuum | In air | |||||||||||||
λ em (nm) | Φ | τ F (ns) | τ P (ms) | λ em (nm) | Φ | τ F (ns) | τ P (ms) | λ em (nm) | Φ | τ F (ns) | τ P (ms) | λ em (nm) | Φ | |
a Excited at 330 nm for 2a–2d, 320 nm for 2e, and 350 nm for 2f. Absolute quantum yields (Φ) were determined using a calibrated integrating sphere system. The luminescence lifetimes (τ) were determined as intensity-averaged lifetimes when the emission decay was multi-exponential. b NE: no emission. | ||||||||||||||
2a | 426, 513 | 0.15 | 0.1 | 12 | 429, 521 | 0.78 | 2.2 | 175 | 541 | 0.17 | 0.8 | 92 | 450 | 0.02 |
2b | 437, 517 | 0.02 | — | — | 423, 521 | 0.79 | 1.6 | 167 | 537 | 0.17 | 0.9 | 94 | 457 | 0.02 |
2c | 433, 517 | 0.24 | 0.2 | 32 | 426, 517 | 0.79 | 1.5 | 167 | 540 | 0.18 | 0.8 | 99 | 463 | 0.02 |
2d | 435, 516 | 0.45 | 0.1 | 58 | 430, 521 | 0.78 | 1.8 | 163 | 530 | 0.17 | 0.7 | 88 | 466 | 0.02 |
2e | 442, 522 | 0.22 | 0.4 | 26 | 420, 505 | 0.92 | 0.4 | 102 | 523 | 0.24 | 0.3 | 71 | 433 | 0.01 |
2f | NEb | — | — | — | 467, 533 | 0.64 | 6.3 | 141 | 549 | 0.13 | 0.9 | 69 | 468 | 0.02 |
With crystals of 2d exhibiting RTP with the highest quantum yield in this study, we investigated the relationship between the phosphorescence lifetime and temperature over the −80 to 80 °C range. As shown in Fig. 5, the lifetime is essentially inversely proportional to temperature, with a correlation coefficient (R2) of 0.99186. Hence, compounds 2 are potentially excellent materials for phosphorescence-based temperature sensing.18
Fig. 5 Transient luminescence decay curves of crystalline 2d at various temperatures. Inset: Temperature dependence of the phosphorescence lifetime. |
We next investigated the photoluminescence of 2 dispersed in PMMA films at 298 K. PMMA films doped with 1.0 wt% of 2 were prepared in a cuvette equipped with a J. Young high-vacuum valve. Under vacuum, all 2, including 2b and 2f whose crystals showed no emissions, exhibited efficient green RTP with long lifetimes (Table 1 and Fig. 6), along with weak prompt fluorescence in the blue region. The luminescence quantum yields of the doped films ranged from 0.13 to 0.24, which are as high as those of other polymer films doped with precious-metal-free phosphors.8 The brilliantly emissive behavior of 2 in polymer films contrasts sharply with that of 1, since 1 dispersed in polymer films was almost non-emissive under vacuum.19 When the green-emissive films of 2 under vacuum were exposed to air (oxygen), the green color immediately disappeared and a faint blue emission (λem = 433–468 nm and Φ = 0.01–0.02) was observed instead.20 The spectra of 2 (except for 2d) in PMMA films under vacuum and in air are shown in Fig. S4–S8 (ESI†). Evacuation of the blue-emissive cuvette restored the intense green emission. The spectra and luminescence images of a 2d-doped PMMA film are shown in Fig. 7 as typical examples. Hence, the RTP of 2 in PMMA films are very sensitive to molecular oxygen as anticipated from the intrinsic nature of their T1 states. The spectral differences in the absence and presence of molecular oxygen are significant. Accordingly, 2-doped polymer films may serve as useful media for oxygen sensing.2
Fig. 7 Photoluminescence spectra and luminescence images of 2d dispersed in a PMMA film under vacuum and in air at 298 K. |
The electron spin resonance (ESR) spectrum of 2d irradiated by UV was measured in toluene at 77 K to confirm the generation of a triplet diradical species (Fig. 8). There are six signals under magnetic fields of approximately 230, 260, 290, 350, 385, and 415 mT, which are derived from the ΔmS = ±1 transitions of the electron spins in the triplet state. The zero splitting parameters D and E were determined to be 0.088 and 0.010 cm−1, respectively. The intense signal near 166 mT is derived from the ΔmS = ±2 transition, and its g value was determined to be 3.95. As the D and E parameters are comparable to those of the triplet state of 1 (SiR3 = SitBuPh2 and Ar = C6H5) and naphthalene in biphenyl,21 the electron distributions of the diradicals in the triplet state of 2 are very similar to those of 1 and naphthalene, indicating that the distribution is spread over almost ten atoms. These results indicate that the silylmethoxy groups in 2 are essential components of the triplet diradical species in the same manner as the siloxy groups are in 1.
2a: M.p.: 180 °C. Td: 259 °C. TLC: Rf 0.27 (hexane/EtOAc 20:1). 1H NMR (CDCl3, 400 MHz): δ −0.26 (s, 18H), 3.46 (s, 4H), 7.14 (s, 2H), 7.45 (t, J = 7.6 Hz, 4H), 7.56 (t, J = 7.6 Hz, 2H), 7.83 (d, J = 7.6 Hz, 4H). 13C NMR (CDCl3, 100 MHz): δ −3.6, 62.1, 112.5, 128.4, 129.6, 131.4, 133.1, 138.0, 152.9, 196.7. IR: ν = 3059, 2951, 2899, 2880, 2826, 1659, 1599, 1582, 1485, 1451, 1398, 1240, 1190, 1015, 951, 845, 818, 756, 723, 692, 679 cm−1. HRMS (FAB): [M + H]+ calcd for C28H35O4Si2, 491.2074; found, 491.2066.
2b: M.p.: 224 °C. Td: 279 °C. TLC: Rf 0.46 (hexane/EtOAc 10:1). 1H NMR (CDCl3, 400 MHz): δ −0.35 (s, 12H), 0.71 (s, 18H), 3.53 (s, 4H), 7.13 (s, 2H), 7.44 (t, J = 7.6 Hz, 4H), 7.56 (tt, J = 7.6, 1.2 Hz, 2H), 7.84 (dd, J = 7.6, 1.2 Hz, 4H). 13C NMR (CDCl3, 100 MHz): δ −7.9, 16.1, 26.4, 59.9, 112.2, 128.4, 129.8, 131.4, 133.2, 137.6, 152.8, 196.7. IR: ν = 3051, 2949, 2926, 2882, 2855, 1665, 1593, 1580, 1487, 1468, 1449, 1400, 1239, 1196, 1017, 951, 835, 820, 799, 772, 725, 691, 683 cm−1. HRMS (FAB): [M + H]+ calcd for C34H47O4Si2, 575.3013; found, 575.3014.
2c: M.p.: 150 °C. Td: 332 °C. TLC: Rf 0.29 (hexane/EtOAc 10:1). 1H NMR (CDCl3, 400 MHz): δ −0.01 (s, 12H), 3.65 (s, 4H), 7.11 (s, 2H), 7.22–7.34 (m, 10H), 7.43 (t, J = 7.6 Hz, 4H), 7.56 (tt, J = 7.6, 1.2 Hz, 2H), 7.83 (dd, J = 7.6, 1.2 Hz, 4H). 13C NMR (CDCl3, 100 MHz): δ −5.2, 61.6, 112.5, 127.8, 128.4, 129.4, 129.7, 131.5, 133.2, 133.7, 136.2, 137.7, 152.7, 196.5. IR: ν = 3067, 2961, 2907, 2882, 2826, 1663, 1597, 1582, 1485, 1449, 1427, 1402, 1240, 1200, 1115, 1018, 951, 839, 818, 768, 723, 691, 681 cm−1. HRMS (FAB): [M]+ calcd for C38H38O4Si2, 614.2309; found, 614.2307.
2d: M.p.: 215 °C. Td: 358 °C. TLC: Rf 0.39 (toluene). 1H NMR (CDCl3, 400 MHz): δ 0.21 (s, 6H), 3.95 (s, 4H), 7.14 (s, 2H), 7.22–7.36 (m, 24H), 7.50 (t, J = 7.6 Hz, 2H), 7.77 (d, J = 7.6 Hz, 4H); 13C NMR (CDCl3, 100 MHz): δ −6.3, 60.9, 112.5, 127.9, 128.4, 129.7, 129.7, 131.6, 133.3, 134.2, 134.6, 137.5, 152.6, 196.3. IR: ν = 3067, 3050, 2999, 2953, 2895, 2820, 1657, 1595, 1582, 1483, 1451, 1427, 1398, 1234, 1190, 1111, 1020, 955, 812, 770, 723, 689, 681, 667 cm−1. HRMS (FAB): [M]+ calcd for C48H42O4Si2, 738.2622; found, 738.2629.
2e: M.p.: 215 °C. Td: 367 °C. TLC: Rf 0.45 (hexane/EtOAc 5:2). 1H NMR (CDCl3, 400 MHz): δ 0.29 (s, 6H), 3.84 (s, 6H), 3.96 (s, 4H), 6.81 (d, J = 8.4 Hz, 4H), 7.11 (s, 2H), 7.24 (t, J = 7.6 Hz, 8H), 7.29–7.35 (m, 12H), 7.76 (d, J = 8.4 Hz, 4H). 13C NMR (CDCl3, 100 MHz): δ −6.2, 55.4, 60.8, 112.3, 113.6, 127.8, 129.6, 130.5, 131.7, 132.2, 134.3, 134.6, 152.3, 163.7, 194.9. IR: ν = 3069, 3051, 3011, 2965, 2928, 2907, 2884, 2838, 1645, 1595, 1572, 1510, 1487, 1425, 1397, 1250, 1192, 1171, 1111, 1017, 955, 845, 833, 812, 779, 737, 698 cm−1. HRMS (FAB): [M + H]+ calcd for C50H47O6Si2, 799.2911; found, 799.2917.
2f: M.p.: 217 °C. Td: 366 °C. TLC: Rf 0.38 (hexane/EtOAc 4:1). 1H NMR (CDCl3, 400 MHz): δ 0.23 (s, 6H), 3.97 (s, 4H), 7.22 (s, 2H), 7.25–7.32 (m, 16H), 7.39–7.45 (m, 8H), 7.70 (d, J = 8.0 Hz, 4H). 13C NMR (CDCl3, 100 MHz): δ −6.2, 60.6, 113.0, 116.2, 118.1, 128.1, 129.5, 130.0, 131.0, 132.0, 133.6, 134.4, 140.4, 153.0, 194.6. IR: ν = 3071, 3021, 3005, 2959, 2903, 2882, 2824, 2230, 1663, 1485, 1427, 1406, 1279, 1234, 1200, 1173, 1109, 1011, 953, 858, 823, 785, 729, 714, 692 cm−1. HRMS (FAB): [M]+ calcd for C50H40N2O4Si2, 788.2527; found, 788.2523.
Footnote |
† Electronic supplementary information (ESI) available. CCDC 1570763, 1570764 and 1570805. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c7qm00524e |
This journal is © the Partner Organisations 2018 |