Chinese Chemical Letters  2019, Vol. 30 Issue (4): 933-936   PDF    
Polymorphism dependent triplet-involved emissions of a pure organic luminogen
Zihan He, Wenbo Li, Gan Chen, Yongming Zhang*, Wang-Zhang Yuan*     
School of Chemistry and Chemical Engineering, Shanghai Key Lab of Electrical Insulation and Thermal Aging, Shanghai Electrochemical Energy Devices Research Center, Shanghai Jiao Tong University, Shanghai 200240, China
Abstract: Polymorphism has been frequently used in tuning the singlet emissions of pure organic dyes. The modulation of triplet-involved emissions, particularly room temperature phosphorescence (RTP), however, is scarcely reported. Herein, polymorphism is reported to tune the triplet-involved emissions of 2CZBZL, a newly designed pure organic luminogen consisting of twisted benzil and two planar carbazole moieties. Other than the conventional modulation through changing molecular conformation and packing, vibration can also finely tune the triplet-involved emissions. Besides prompt fluorescence (PF), polymorph B with relatively extended conformation emits thermally activated delayed fluorescence (TADF), whereas the others (A, C-E) with similarly more twisted conformations generate predominant RTP or simultaneous DF and RTP. These results demonstrate the fascinating chance to regulate the tripletinvolved emissions through controlling conformation and vibration.
Keywords: Polymorphism     Room temperature phosphorescence     Pure organic luminogens     Thermally activated delayed fluorescence     Triplet-involved emissions    

Modulation of the emission of organic luminogens has obtained considerable attention due to its fundamental importance and the practical requirement for applications in organic light-emitting diodes (OLEDs), organic lasers, bioimaging, etc. [1-13]. Specifically, polymorphism of organic crystals is widely adopted to tune the emission of a single compound through the variation of conformation and molecular packing [5-10, 14-16]. Definite single crystal structures provide a visualized mode to uncover the relationship between conformation/packing and emission. In 2005, Araki and coworkers obtained two forms of crystal from 2, 2':6', 2"-terpyridine (tpy) with distinct photoluminescent (PL) properties [8]. Despite its efficacy in tuning the emission of pure organic luminogens, most endeavors have concentrated on singlet emissions, and much less attention has been devoted to modulating the triplet-involved emissions of delayed fluorescence (DF) and room temperature phosphorescence (RTP) [5, 7, 17]. Such situation might be ascribed to the difficulty of simultaneously achieving both efficient tripletinvolved emissions and polymorphic crystals [2, 3, 18-22]. Particularly for the efficient pure organic RTP, even though great progress have been attained in the past years [20, 19-38], it remains an outstanding challenge.

Previous results indicate metal-free thermally activated DF (TADF) materials are characterized by luminescent molecules with small energy gap (ΔEST) between the lowest singlet (S1) and triplet (T1) excited states [39-41], while RTP phosphors normally have much larger ΔEST. However, there is no definite boundary between these two competitive processes, as evidenced by their coexistence in the same states [42]. Fine control of such triplet processes via polymorphs is beneficial to the fundamental understanding on the underlying relationship between supramolecular structure and emission. In this contribution, we present a new pure organic polymorphic system consisting of benzil and two carbazole moieties, namely 2CZBZL (Scheme 1). The combination of twisting and planar segments is envisaged to impart the high potential to obtain crystal polymorphs through the change of the torsion angle of benzil and the packing mode of carbazole. Moreover, the presence of carbonyl groups can promote spin-orbital coupling (SOC) and consequent intersystem crossing (ISC), whereas carbazole is capable to generate efficient tripletinvolved emissions through facile decoration with electronaccepting units [20-22, 24, 29].

Download:
Scheme 1. Synthetic route to 2CZBZL.

Indeed, 2CZBZL readily generates remarkable triplet emissions in the crystalline states. And to our delight, so far, we have collected five polymorphs with varied emission colors and distinct triplet emission behaviors. Time resolved emission measurement reveals that triplet excitons can be radiatively deactivated through competitive channels of phosphorescence, DF, or concurrent RTP and DF in various polymorphs. Besides the conformation effect, the crucial role of vibration is also demonstrated herein. Such readily tunable triplet-involved emission based on conformation change and vibration is important for better understanding and regulation of triplet processes.

2CZBZL was prepared according to the synthetic route shown in Scheme 1. Palladium-catalyzed Buchwald coupling of 4, 4'-dibromobenzil (DBBZL) and carbazole (CZ) led to the target compound in 82.5% yield. It was characterized by 1H and 13C NMR spectra (Figs. S1 and S2 in Supporting information), as well as single crystal analysis (vide infra), with satisfactory results obtained.

Cultivation of single crystals of 2CZBZL in dichloromethane (DCM) and tetrahydrofuran (THF) successfully yields five needleor plate-like polymorphs (crst-A~E) (CCDC Nos. 1517342- 1517346), which exhibit different emission colors upon UV irradiation. To decipher the structure-emission relationship, we systematically investigated their photophysical properties, with focus on crst-A~D, however, photophysical properties of crst-E were not thoroughly characterized due to the little amount of single crystals obtained. As depicted in Fig. 1, under 365 nm UV irradiation, crst-A and B emit green and yellow lights with emission maxima (Em) at 486 and 565 nm, respectively, whereas crst-C generates greenish-yellow light with Ems at 465, 486, and 527 nm. Astonishingly, with almost identical emission profiles (Fig. 1 and Fig. S3 in Supporting information), crst-A and D demonstrate remarkable difference in emission colors of green and orange (Fig. 1), respectively. Previous investigations suggest a considerable influence of RTP on the apparent emission color of fluorescence-phosphorescence dual-emissive luminophores, albeit their low fractions [34]. It is thus assumed that such dramatic distinction in emitting colors is likely the consequence of their different RTP emissions.

Download:
Fig. 1. PL spectra and photographs taken under 365 nm UV irradiation of crst-A~D at room temperature (r.t.) and 77 K. Excitation wavelength (nm): 365 (crst-A), 430 (crst-B), 370 (crst-C and D). PL efficiencies of varying crystals at room temperature are also given.

Further insights into delayed and time-resolved emissions help us gain better understanding. For crst-A, with a delay time (td) of 0.1 ms (all signals of prompt fluorescence (PF) could be excluded with td > 0.1 ms), the original peak disappears, meanwhile, a maximal along with two shoulders at 560 and 527/605 nm (Fig. 1), are detected. Such peaks should be corresponded to RTP, which is further verified by its long lifetime (<τ>) of 18.05 ms at 560 nm (Fig. 2A, <τ> values are calculated with the data reported in Fig. 2). Notably, the emission at 486 nm demonstrates remarkable three lifetimes of 3.84 ns, 28.3 μs, and 10.76 ms in different timescales, indicating the simultaneous contribution of PF, DF, and RTP. Distinct from crst-A, the delayed emission of crst-B is merely slightly red-shifted from 565 nm to 568 nm, which is similar to those of DF emitters. Further time-resolved measurement at 554, 560, and 565 nm unambiguously discloses their nearly identical biexponential decay with lifetimes of 6.67 ns and 4.16 μs (Fig. 2B), thus confirming the DF nature of crst-B. Crst-C exhibits analogous delayed emission profile to that of A. Its original peaks (465, 486 nm) are vanished and two maxima at 527 and 570 nm corresponding to RTP are observed, whose <τ> values are 8.18 and 7.05 ms, respectively (Figs. 1A, C, 2A, C). However, other than the prompt lifetime of 2.33 ns, crst-C does not exhibit any significant long-lived species at 486 nm (Fig. S4 in Supporting information), indicative of the absence of DF. Namely, crst-C only possesses PF and RTP emissions. As for crst-D, it shows two delayed peaks at 571 (<τ> = 7.95 ms) and 618 nm (<τ> = 8.19 ms), together with a tail at around 527 nm, which are assignable to RTP. The much redder RTP emission of crst-D compared to that of A verifies our conjecture that the difference in their apparent emission color originates from their RTP fraction.

Download:
Fig. 2. Emission decay profiles of (A) crst-A, (B) crst-B, (C) crst-C, and (D) crst-D monitored at varying wavelengths.

We summarized some basic photophysical data for the crystals in Table S1 in Supporting information. Clearly, other than PF, crst-A exhibits DF and RTP, whereas crst-B demonstrates only DF, and polymorphs C and D display predominantly RTP. Notably, different polymorphs also possess varied PL efficiencies (Φ), whose values are 4.2%, 6.3%, 8.3%, and 3.9% for crst-A~D, respectively. Furthermore, except crst-B, short but visible persistent RTP (p-RTP) of the other three forms are noticed (Fig. S5 in Supporting information). The vivid afterglow contrast between crst-A and D is also detected. While the former is greenish orange, the latter is red, which are consistent with their delayed emission spectra (Fig. 1). Clearly, different polymorphs exhibit varying emission colors, PL spectra, lifetimes, and efficiencies, which strongly indicative of the high tuneability of emission, particularly triplet-involved emissions of 2CZBZL.

Single-crystal structure offers further insights into the polymorphism dependent emission of 2CZBZL. As depicted in Fig. 3, molecular conformations of crst-A, C, and D are almost identical to one another. They all crystalize according to the monoclinic space group C2/c (Z = 4), with other crystal unit cell parameters and density also in close similarity (Table S2 in Supporting information). Moreover, the short contacts and their distances are pretty much the same. Careful inspection reveals merely slight differences in such intermolecular interactions as C—H…H—C, C—H…π, and ππ short contacts, which are effective in restricting vibrational stretching of subgroups and thus rigidify the molecular conformations. There are two fewer C—H…π contacts in crst-C and D than those in A (Fig. 3 and Figs. S6–S9 in Supporting information). Actually, the contact distances are rather close (Fig. S9 in Supporting information) in each polymorph. It is quite surprising that such a nuance as several fewer short contacts and subtle variation of contact distance in single crystals can lead to the remarkable difference in their triplet-involved emissions. However, scrutinization of the prompt and delayed emissions reveals that major peaks for these spectra are alike, and the main difference lies in their relative intensity/fraction. More specifically, three prompt spectra share the main peak around 486 nm and two shoulders around 465 and 527 nm, despite the latter shoulders in polymorphs A and D are unconspicuous (Fig. 1 and Fig. S3). Moreover, polymorphs A, C and D share two evident delayed emission peaks at 527 and ~570 nm (crst-A: 560 nm), accompanied with another shoulder/peak at 605, 617, and 621 nm, respectively. These peaks/shoulders probably result from different vibrational energy levels of RTP, rather than monomer emission and partial ππ stacking induced excimer emissions, as comparable or even longer <τ> value at 527 nm is detected with comparison to that at 570 nm in polymorph C (Fig. 2C). It is well known that stretching C—H vibrational stretching is one of the highest energy consumable vibration (0.37 eV) in organic molecules [43], which exert significant impact on molecular vibrational energy levels. Hence, such C—H…π contacts can be crucial for determining molecular vibrational energy levels. Our example of crst-A, C, and D suggests that mere molecular interactions can be utilized in tuning triplet emission through regulating molecular vibrations.

Download:
Fig. 3. Single crystal structures of crst-A~D with denoted intermolecular interactions around one molecule.

In contrast to the almost identical crystal structure of polymorphs A, C, and D, polymorph B adopts utterly different molecular conformation with merely four C—H…π contacts around one molecule (Fig. 3 and Fig. S10 in Supporting information). As depicted in Fig. S11 in Supporting information, the dihedral angle between two carbazole planes in crst-B is 128.4°, which is far bigger than those of the other three (~91.3°) by 37.1°. Additionally, the dihedral angle between carbazole and its adjacent phenyl ring in crst-B (41.9°) is slightly smaller compared to those of the others (~43.4°). Furthermore, the central C16-C19-C19'-C16' torsion angle of crst-B (135.1°) is also much larger than those of the other polymorphs (~101.6°). These results indicate the more planar conformation of crst-B when compared to those of others.

DF rather than RTP or DF-RTP in crst-B might be ascribed to its effective reverse ISC (RISC) or triplet-triplet annihilation (TTA) process. To gain more information, its ΔEST was estimated. The emission spectra of the crystals at 77 K with td of 0 and 0.1 ms were measured. As can be seen from Fig. 1, crst-B shows an orange emission centered at 588 nm at 77 K, together with the delayed emission maxima at 552 and 577 nm, from which a ΔEST of 0.08 eV is derived for crst-B. Such small ΔEST will readily facilitate the RISC process, thus generating significant TADF [40-42]. For the other three polymorphs at cryogenic temperature, while they demonstrate finely structured prompt emission and long-lasting phosphorescence owing to further confinement of vibrational motions, all their delayed emission shows extremely similar profiles peaking at 527 nm (Fig. S12 in Supporting information). These results further verify the phosphorescence nature of the emission at 527 nm at room temperature. Moreover, both emission spectra and emission colors for polymorphs A and D become close at 77 K (Fig. S13 in Supporting information), which confirms our assumption that their sharp distinction in emission color at room temperature lies in their different RTP emissions, which stem from their varied vibrations.

It is also worth noting that other than above mentioned four crystal forms obtained from DCM solutions, we also acquired another green-emitting polymorph E from the THF solution. Crst-E holds similar cell parameters and conformation to those of A, C, and D (Table S2 and Fig. S14 in Supporting information), whose prompt and delayed emissions are peaking at 488 and 494/562/ 612 nm (Fig. 4), respectively. While the 494 nm peak is assignable to DF, the other two are attributable to RTP. We are also optimistic that 2CZBZL can produce additional polymorphs given enough attempts, thus affording more chances to tune the triplet-involved emissions.

Download:
Fig. 4. (A) Optical microscopy photograph taken under 330–380 nm light and (B) prompt/delayed PL spectra of crst-E at room temperature.

In conclusion, with the combination of twisting benzil and planar carbazole moieties, 2CZBZL with multiple polymorphism and finely tunable triplet relaxations was obtained. Other than PF, polymorph B with extended conformation exhibits TADF, whereas the other forms with more twisted and closely similar conformations display DF-RTP or solely RTP, indicating the conformationcontrollable triplet processes. Notably, albeit slight variations in intermolecular interactions, considerable changes in triplet relaxations and emission colors in crst-A, C, D, and E are identified, on account of their distinct vibration dissipations and the high susceptibility of triplet excitons to vibrations.

Acknowledgment

This work was financially supported by the National Natural Science Foundation of China (Nos. 51822303, 51473092).

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.cclet.2019.03.015.

References
[1]
Z. Song, W. Zhang, M. Jiang, et al., Adv. Funct. Mater. 26 (2016) 824-832. DOI:10.1002/adfm.v26.6
[2]
O. Bolton, K. Lee, H.J. Kim, K.Y. Lin, J. Kim, Nat. Chem. 3 (2011) 205-210. DOI:10.1038/nchem.984
[3]
W. Huang, L. Sun, Z. Zheng, J. Su, H. Tian, Chem. Commun. 51 (2015) 4462-4464. DOI:10.1039/C4CC09613D
[4]
C. Yuan, S. Saito, C. Camacho, et al., J. Am. Chem. Soc. 135 (2013) 8842-8845. DOI:10.1021/ja404198h
[5]
J. Yang, Z. Ren, B. Chen, et al., J. Mater. Chem. C 5 (2017) 9242-9246. DOI:10.1039/C7TC03656F
[6]
D. Yan, D.G. Evans, Mater. Horiz. 1 (2014) 46-57. DOI:10.1039/C3MH00023K
[7]
Y. Zhang, H. Ma, S. Wang, et al., J. Phys. Chem. C 120 (2016) 19759-19767. DOI:10.1021/acs.jpcc.6b05537
[8]
T. Mutai, H. Satou, K. Araki, Nat. Mater. 4 (2005) 685-687. DOI:10.1038/nmat1454
[9]
K. Wang, H. Zhang, S. Chen, et al., Adv. Mater. 26 (2014) 6168-6173. DOI:10.1002/adma.201401114
[10]
G. Zhang, J. Lu, M. Sabat, C.L. Fraser, J. Am. Chem. Soc. 132 (2010) 2160-2162. DOI:10.1021/ja9097719
[11]
W. Luo, Y. Zhang, Y. Gong, et al., Chin. Chem. Lett. 29 (2018) 1533-1536. DOI:10.1016/j.cclet.2018.08.001
[12]
X. Fang, D. Yan, Sci. China Chem. 61 (2018) 397-401. DOI:10.1007/s11426-017-9183-9
[13]
X. Li, S. Semin, L.A. Estrada, et al., Chin. Chem. Lett. 29 (2018) 297-300. DOI:10.1016/j.cclet.2017.11.001
[14]
C. Wang, B. Xu, M. Li, et al., Mater. Horiz. 3 (2016) 220-225. DOI:10.1039/C6MH00025H
[15]
M. Brinkmann, G. Gadret, M. Muccini, et al., J. Am. Chem. Soc. 122 (2000) 5147-5157. DOI:10.1021/ja993608k
[16]
H. Liu, X. Cheng, Z. Bian, K. Ye, H. Zhang, Chin. Chem. Lett. 29 (2018) 1537-1540. DOI:10.1016/j.cclet.2018.08.003
[17]
H. Wu, W. Chi, G. Baryshnikov, et al., Angew. Chem. Int. Ed. 58 (2019), doi: http://dx.doi.org/10.1002/anie.201900703.
[18]
El-Sayed M.A., Acc. Chem. Res. 1 (1968) 8-16. DOI:10.1021/ar50001a002
[19]
N.J. Turro, V. Ramamurthy, J.C. Scaiano, Modern Molecular Photochemistry of Organic Molecules, University Science Books, Sausalito, CA, 2010.
[20]
C. Wang, Y. Gong, W.Z. Yuan, Y. Zhang, Chin. Chem. Lett. 27 (2016) 1184-1192. DOI:10.1016/j.cclet.2016.05.026
[21]
S. Hirata, Adv. Opt. Mater. 5 (2017) 1700116. DOI:10.1002/adom.201700116
[22]
M. Baroncini, G. Bergamini, P. Ceroni, Chem. Commun. 53 (2017) 2081-2093. DOI:10.1039/C6CC09288H
[23]
W.Z. Yuan, X.Y. Shen, H. Zhao, et al., J. Phys. Chem. C 114 (2010) 6090-6099. DOI:10.1021/jp909388y
[24]
Y. Xie, Y. Ge, Q. Peng, et al., Adv. Mater. 29 (2017) 1606829. DOI:10.1002/adma.v29.17
[25]
W. Zhao, Z. He, J.W.Y. Lam, et al., Chemistry 1 (2016) 592-602. DOI:10.1016/j.chempr.2016.08.010
[26]
J. Wang, X. Gu, H. Ma, et al., Nat. Commun. 9 (2018) 2963. DOI:10.1038/s41467-018-05298-y
[27]
S. Hirata, J. Mater. Chem. C 6 (2018) 11785-11794. DOI:10.1039/C8TC01417E
[28]
M. Shimizu, R. Shigitani, M. Nakatani, J. Phys. Chem. C 120 (2016) 11631-11639. DOI:10.1021/acs.jpcc.6b03276
[29]
Z. An, C. Zheng, Y. Tao, et al., Nat. Mater. 14 (2015) 685-690. DOI:10.1038/nmat4259
[30]
L. Bian, H. Shi, X. Wang, et al., J. Am. Chem. Soc. 140 (2018) 10734-10739. DOI:10.1021/jacs.8b03867
[31]
H. Ma, Q. Peng, Z. An, W. Huang, Z. Shuai, J. Am. Chem. Soc. 141 (2019) 1010-1015. DOI:10.1021/jacs.8b11224
[32]
D. Li, F. Lu, J. Wang, et al., J. Am. Chem. Soc. 140 (2018) 1916-1923. DOI:10.1021/jacs.7b12800
[33]
L. Xu, G. Li, T. Xu, et al., Chem. Commun. 54 (2018) 9226-9229. DOI:10.1039/C8CC04734K
[34]
Y. Gong, G. Chen, Q. Peng, et al., Adv. Mater. 27 (2015) 6195-6201. DOI:10.1002/adma.201502442
[35]
W.X. Wu, H. Wang, W.J. Jin, Cryst. Growth Des. 18 (2018) 6742-6747. DOI:10.1021/acs.cgd.8b01021
[36]
Z. Yang, Z. Mao, X. Zhang, et al., Angew. Chem. Int. Ed. 55 (2016) 2181-2185. DOI:10.1002/anie.201509224
[37]
H. Wu, Y. Zhou, L. Yin, et al., J. Am. Chem. Soc. 139 (2017) 785-791. DOI:10.1021/jacs.6b10550
[38]
H. Wua, B. Wu, X. Yu, et al., Chin. Chem. Lett. 28 (2017) 2151-2154. DOI:10.1016/j.cclet.2017.08.002
[39]
H. Uoyama, K. Goushi, K. Shizu, H. Nomura, C. Adachi, Nature 492 (2012) 234-238. DOI:10.1038/nature11687
[40]
Y. Tao, K. Yuan, T. Chen, Adv. Mater. 26 (2014) 7931-7958. DOI:10.1002/adma.v26.47
[41]
Z. Yang, Z. Mao, Z. Xie, et al., Chem. Soc. Rev. 46 (2017) 915-1016. DOI:10.1039/C6CS00368K
[42]
J.S. Ward, R.S. Nobuyasu, A.S. Batsanov, et al., Chem. Commun. 52 (2016) 2612-2615. DOI:10.1039/C5CC09645F
[43]
M. Pope, C.E. Swenberg, Electronic Processes in Organic Crystals and Polymers, Oxford University Press, Oxford, 1999.