RSC Advances

COMMUNICATION

Tuning the color of thermally activated delayed

fluorescent properties for spiro-acridine derivatives
Cite this: RSC Adv., 2015, 5, 18588
by structural modification of the acceptor
Received 24th November 2014 fragment: a DFT study†
Accepted 2nd February 2015
Jing Lu, Yiying Zheng and Jingping Zhang*
DOI: 10.1039/c4ra15155k

www.rsc.org/advances

A theoretical investigation on the relationship between the electronic HOMO and LUMO, an effective method for steric separation is
structures and emission properties of thermally activated delayed to introduce a spiro-junction between the acceptor and donor
fluorescent materials based on donor–acceptor type spiro-acridine fragments.7 Adachi and co-workers have reported a series of
derivatives has been performed. Efficient color tuning over the efficient TADF materials based on spiro structures, such as a
whole visible range has been achieved via structural modification of spirobiuorene derivative6a and a spiro-acridine derivative
the acceptor fragment. (ACRFLCN),8 whose emitting colors resided in the yellow region.
Therefore, it is eagerly anticipated that the emission color of
spiro-based TADF materials can be tuned from blue to red by
modifying the electron donor or acceptor fragment.
In this communication, using ACRFLCN as the parent
Introduction molecule, a series of derivatives were designed by utilizing
“CH”/N substitution and changing the p-conjugation degree on
Recently, thermally activated delayed uorescent (TADF) mate-
the acceptor fragment in order to tune the emitting color.9 As
rials as a promising alternative solution for the conversion of
shown in Fig. 1, an A series containing 1a–c and 2a–c, where 1
triplet into singlet excitons for highly efficient OLEDs were
reported,1 corresponding to a high external quantum efficiency
of close to 20% for TADF.2 Pure organic TADF materials will be
useful for the realization of the rare-metal free, low-cost fabri-
cation of at-panel displays and general lighting.3 TADF is
signicant only when the quantum yields of the triplet forma-
tion (4T) and singlet formation (4S) are both high. This in turn
implies that a small energy gap (DEST) between the rst singlet
excited state (S1) and the triplet excited state (T1), a long T1
lifetime, and a temperature that is not too low, are necessary for
TADF.4 The DEST is proportional to the exchange integral (JLU)
between the spatial wave functions of a highest occupied
molecular orbital (HOMO) and lowest unoccupied molecular
orbital (LUMO).5 Thus, TADF materials must have an effective
HOMO and LUMO separation in order to obtain a near zero
DEST, and the effective intermolecular charge transfer from the
p to p* orbital, while maintaining a reasonable high radiative
decay rate of uorescence.6 To achieve the separation of the

Faculty of Chemistry, Northeast Normal University, Changchun 130024, China.

E-mail: [email protected]; Fax: +86 431 85099521; Tel: +86 431 85099372
† Electronic supplementary information (ESI) available: Methodology in detail,
molecular geometries in the ground and excited states, frontier molecular
orbital analysis, density of states (DOS), transition density matrix, and
constrained density functional theory (C-DFT) calculations. See DOI: Fig. 1 Molecular structures of the investigated spiro-acridine
10.1039/c4ra15155k derivatives.

18588 | RSC Adv., 2015, 5, 18588–18592 This journal is © The Royal Society of Chemistry 2015
Communication RSC Advances

and 2 refer to mono- and di-substitutions, respectively, and a B resources. All of the calculations are performed with the
series corresponding to the reduced conjugation derivatives 2d Gaussian 09 package.17
and 2e. The calculated emitting wavelengths cover the blue to For the asymmetric mono-substitution of “CH”/N deriva-
red region, giving a high quantum emission yield. An in depth tives, the dihedral angles between the acceptor and donor
theoretical understanding of the delayed uorescent properties fragments are 89.9 , 95.4 , and 99.6 for 1a–1c in the S1 state,
of the designed spiro-acridine derivatives could be revealed. respectively. While the symmetric “CH”/N di-substituted deriv-
atives 2a–2e in S1 state have an orthogonal orientation between
the donor and acceptor fragments with the distortion angles
Results and discussion being 90.0 (Table S1†). This might be more favourable for the
steric separation of the frontier molecular orbital (FMO). For all
A crucial step in theoretical investigations employing the of the designed molecules, the largest difference between the
density functional method (DFT) and the time dependent bond lengths in the S0 and S1 states was found to be between the
density functional method (TD-DFT) is the choice of an appro- N and its neighbouring C atom with a length of about 0.03 Å
priate exchange correlation functional. The ground state and (Fig. S1†). The bond lengths and dihedral angles of these spiro-
excited state geometries of the studied compounds were opti- acridine derivatives change slightly between the T1 and S1 states
mized using the DFT and TD-DFT methods with the restricted to within 0.02 Å and 1 , respectively, indicating that it is easy to
hybrid-type Perdew–Burke–Ernzerhof exchange correlation get an efficient up-conversion from triplet to singlet excitons.
functional (PBE0), a functional which uses a 25% exchange and Furthermore, the FMOs for all of these derivatives in the S0 state
75% correlation weighting.10 On the basis of the optimized S1 show p character (Fig. S2†). The electron density distribution of
structure, we have chosen different functionals for the compu- the HOMO is located on the acridine donor fragment, while that
tations of delay wavelengths (lTADF) using the 6-31G (d) and 6- of LUMO is located on the acceptor fragment. Hence, all of the
311+G (d, p) basis sets, respectively. The calculated lTADF values complexes show a great separation of the HOMO and LUMO
for ACRFLCN together with the experimental data are shown in because of the spiro-conjugation. We also give the density of
Fig. 2. The lTADF value calculated at the PBE0/6-31G (d) theo- state (DOS) analysis to support the effect of the separation of the
retical level (540 nm) is quite close to the experimental value HOMO and LUMO (Fig. S3†). Meanwhile, the distribution of the
(530 nm), with the deviation being 10 nm. The calculated lTADF electron density plots for the T1 states (Fig. S4†) are similar to
values using the traditional B3LYP11 and newly developed those for the S1 states. It is also of benet for the reverse
TPSSh12 functionals are larger than the experimental value. The intersystem crossing (RISC) from the T1 to S1 states.
hybrid functional (M062X)13 and long range corrected func- Moreover, we also analyzed the DEST values for all of the
tionals (CAM-B3LYP14 and wB97XD15) are also unacceptable for derivatives in Table 1. The smaller the DEST value at a given
the current investigation, and overestimated the transition temperature, the easier it is for RISC from the T1 to S1 states.
energy between the LUMO and HOMO.16 Meanwhile, the basis Firstly, we discussed the substituent position effect on the DEST
set effect was also considered in this study. The lTADF value values. The calculated DEST value for ACRFLCN is 0.0102 eV at
calculated using the 6-31G (d) basis set is similar to that of the 6- the PBE0/6-31G (d) level, which is close to the Adachi et al.
311+G (d, p) basis set, which was augmented with a diffuse calculated value of 0.0083 eV at the B3LYP/6-31G (d) level of
function. Therefore, the 6-31G (d) functional was selected for theory.8 This small difference arises from the use of different
the following calculations in order to save computational density functionals. For the asymmetric mono-substitution of
the “CH”/N derivatives 1a–c, “CH”/N substitution at the Z-
position does not signicantly affect the DEST value, whereas
the DEST values of the “CH”/N substituted derivatives at the Y-
and X-positions are increased to 0.0164 eV and 0.0157 eV. For
the symmetric di-substitution of “CH”/N derivatives, the order

Table 1 The vertical excitation energy (ES1 and ET1) and vertical singlet-
triplet energy gap (DEST) of these studied spiro-acridine derivatives at
the TD-PBE0/6-31G (d) level (in eV)

ES1 ET1 DEST

ACRFLCN 2.7277 2.7175 0.0102

1a 2.6010 2.5907 0.0103
2a 2.4737 2.4628 0.0109
1b 2.6807 2.6643 0.0164
2b 2.6261 2.6093 0.0168
1c 2.2792 2.2635 0.0157
Fig. 2 The calculated lTADF values for ACRFLCN with different func- 2c 2.0834 2.0770 0.0064
tionals and basis sets on the basis of the resulting optimized S1 state at 2d 3.2893 3.2786 0.0107
the TD-PBE0/6-31G* level, together with the experimental result. 2e 2.7629 2.7530 0.0099

This journal is © The Royal Society of Chemistry 2015 RSC Adv., 2015, 5, 18588–18592 | 18589
RSC Advances Communication

of the DEST values is 2b (0.0168 eV) > 2a (0.0109 eV) > 2c (0.0064 values of 2c and 2e with the smaller DEST values should be
eV). 2c, i.e. with a di-substitution at the X, X0 -positions, has the higher than for those of other investigated derivatives.
smallest DEST value, which can be traced back to the more It is known that the photophysical properties of the
effective separation between its HOMO and LUMO, and donor/ complexes depend strongly on the character of the FMOs. The
acceptor compositions (%) of the LUMO in the S0 state that are electron density plots of the HOMOs and LUMOs and the energy
0.7 : 99.3 (Table S2 and Fig. S5†). gaps (DEH–L) of the studied spiro-acridine derivatives in the S1
The slight discrepancies in the DEST values between the state are depicted in Fig. 3. It is noted that the HOMOs are
symmetric di-substituted and the asymmetric mono- located on the donor fragments and the LUMOs reside in the
substituted derivatives for a and b are 0.0006 eV and 0.0004 acceptor fragments in the S1 states. The DEH–L values for the A
eV, respectively, i.e. the asymmetric mono-substitution and series are reduced compared to the value for ACRFLCN due to
symmetric di-substitution of “CH”/N substituted derivatives at the “CH”/N substitution effect. Compared with the asymmetric
the Z-position and Y-position do not signicantly affect the DEST mono-substituted derivatives at the same position, the DEH–L
value due to a slight change in the optimized structural values for these symmetric di-substituted derivatives decreased,
parameters for the S0 states. However, for the X-position and the emission wavelengths are predicted to red-shi. The
substituted derivatives, the di- and mono-substitution exhibi- HOMO energy levels in 2a and 2b are decreased, whereas that in
ted a signicant effect on the DEST value. The twisted angle 2c is increased. On the other hand, the LUMO energy levels in
between the donor and acceptor fragments of 1c in the ground 2a–2c are lower than those in ACRFLCN. A larger change in the
state is 79.7 , which is smaller than that for 2c (90.0 ). The LUMO energy levels than those in the HOMOs is observed,
electronic separation between the donor and acceptor moieties because the LUMO is mostly located on the “CH”/N substituted
of 2c is more effective than that for 1c due to the larger twisted derivative acceptor fragment. In contrast, the LUMO energy
angles of 2c, indicating that the DEST value for 2c is smaller levels in the B series are higher than that in ACRFLCN, and
compared to the value for 1c. Thus, we only considered the consequently the DEH–L values are enlarged due to a decrease in
“CH”/N di-substitution at the XB, XB0 -positions for the B series. the degree of p-conjugation. Therefore, we anticipate that blue-
The calculated DEST values for 2d (0.0107 eV) and 2e (0.0099 eV) and red-shied emission wavelengths could be observed for 2d
are similar to that for the parent molecule ACRFLCN. The DEST and 1(2)a–c, respectively, compared with that for ACRFLCN.
value for 2e, which is the derivative with “CH”/N di-substitution The time dependent density functional theory (TD-DFT)
at the XB, XB0 -positions, is smaller than that for the unsub- calculated delayed uorescent properties of these spiro-
stituted 2d. Thus, smaller DEST values were obtained via “CH”/N acridine derivatives are listed in Table 2. The FMO distribu-
di-substitution at the X, X0 -positions for spiro-junction TADF tion shows that the delayed uorescent band for all of the
materials. It is instructive to estimate the RISC rate constant derivatives can be assigned to a p* / p transition. Similar to
(kRISC) using the equation kRISC ¼ A exp(DEST/RT). It is known the order of the DEH–L values, the singlet vertical emission
that high values of kRISC are favored by a small DEST at a given energies of the asymmetric and symmetric derivatives follow the
temperature in the absence of oxygen. Consequently, the kRISC same tendency: 1c < 1a < 1b < ACRFLCN, 2c < 2a < 2b <
ACRFLCN z 2e < 2d, respectively. For the symmetric di-

Fig. 3 The electron density plots and energy levels of the HOMOs and LUMOs for these investigated spiro-derivations of the S1 states.

18590 | RSC Adv., 2015, 5, 18588–18592 This journal is © The Royal Society of Chemistry 2015
Communication RSC Advances

Table 2 The delayed fluorescence emission wavelength (lTADF) and Education for New Teachers (20120043120008) are gratefully
phosphorescence emission wavelength (lph), corresponding to the acknowledged.
absorption wavelength (labs) and Stokes shift of the explored
complexes
Notes and references
labs (nm) lTADF (nm) lph (nm) Stokes shi
1 (a) A. Endo, M. Ogasawara, A. Takahashi, D. Yokoyama,
ACRFLCN 455 540 540 85 Y. Kato and C. Adachi, Adv. Mater., 2009, 21, 4802; (b)
1a 477 571 575 94
2a 501 611 615 110
X.-L. Chen, R. Yu, Q.-K. Zhang, L.-J. Zhou, X.-Y. Wu,
1b 463 560 565 97 Q. Zhang and C.-Z. Lu, Chem. Mater., 2013, 25, 3910; (c)
2b 472 567 571 95 S. Wu, M. Aonuma, Q.-S. Zhang, S.-P. Huang, T. Nakagawa,
1c 544 723 731 179 K. Kuwabara and C. Adachi, J. Mater. Chem. C, 2014, 2,
2c 595 766 770 171 421; (d) H. Nakanotani, K. Masui, J. Nishide, T. Shibata
2d 377 432 430 55
2e 449 536 535 87
and C. Adachi, Sci. Rep., 2013, 3, 2127; (e) K. Goushi,
K. Yoshida, K. Sato and C. Adachi, Nat. Photonics, 2012, 6,
253; (f) D.-D. Zhang, L. Duan, Y. Li, D.-Q. Zhang and
Y. Qiu, J. Mater. Chem. C, 2014, 2, 8191; (g) H. Wang,
substituted derivatives 2a–2c, the lTADF values (611 nm, 567 nm, L. Xie, Q. Peng, L. Meng, Y. Wang, Y. Yi and P. Wang, Adv.
and 766 nm) are bathochromically-shied by 71 nm, 27 nm, and Mater., 2014, 26, 5198.
226 nm, respectively, which is caused by the “CH”/N substitu- 2 (a) H. Uoyama, K. Goushi, K. Shizu, H. Nomura and
tion at different positions. The derivative 2c with the largest C. Adachi, Nature, 2012, 492, 234; (b) Y. Im and J. Y. Lee,
Stokes shi among the symmetric di-substituted derivatives Chem. Mater., 2014, 26, 1413; (c) F. B. Dias,
exhibits a red emission. However, it is interesting to note that K. N. Bourdakos, V. Jankus, K. C. Moss, K. T. Kamtekar,
2d exhibits a blue emission centred at 432 nm with the smaller V. Bhalla, J. Santos, M. R. Bryce and A. P. Monkman, Adv.
Stokes shi, which might result from a decrease in the degree of Mater., 2013, 25, 3707.
p-conjugation. We also calculated the phosphorescence emis- 3 J. Lee, K. Shizu, H. Tanaka, H. Nomura, T. Yasuda and
sion wavelength (lph) using the TD-PBE0 method on the basis of C. Adachi, J. Mater. Chem. C, 2013, 1, 4599.
the optimized T1 states. The lph data are similar to the lTADF 4 (a) K. Masui, H. Nakanotani and C. Adachi, Org. Electron.,
data, and the small variation between the lTADF and lph values is 2013, 14, 2721; (b) S. Huang, Q. Zhang, Y. Shiota,
of benet for achieving the TADF phenomenon. T. Nakagawa, K. Kuwabara, K. Yoshizawa and C. Adachi, J.
Chem. Theory Comput., 2013, 9, 3872.
5 A. Endo, K. Sato, K. Yoshimura, T. Kai, A. Kawada,
Conclusion H. Miyazaki and C. Adachi, Appl. Phys. Lett., 2011, 98, 083302.
6 (a) Q. Zhang, J. Li, K. Shizu, S. Huang, S. Hirata, H. Miyazaki
In summary, we have successfully demonstrated and elucidated
and C. Adachi, J. Am. Chem. Soc., 2012, 134, 14706; (b)
the broad range color tuning of spiro-acridine TADF materials
H. Tanaka, K. Shizu, H. Nakanotani and C. Adachi, Chem.
via “CH”/N substitution and conjugation degree control on the
Mater., 2013, 25, 3766.
acceptor fragment. All of the designed TADF materials possess
7 (a) K. Nasu, T. Nakagawa, H. Nomura, C.-J. Lin, C.-H. Cheng,
small DEST values, which may provide an efficient up-
M.-R. Tseng, T. Yasuda and C. Adachi, Chem. Commun., 2013,
conversion from the T1 to S1 levels. The calculated delayed
49, 10385; (b) T. Nakagawa, S.-Y. Ku, K.-T. Wong and
emission wavelengths cover the whole visible region. The lTADF
C. Adachi, Chem. Commun., 2012, 48, 9580.
values for the “CH”/N substituted derivatives are
8 G. Méhes, H. Nomura, Q. Zhang, T. Nakagawa and C. Adachi,
bathochromically-shied compared with that for the parent
Angew. Chem., Int. Ed., 2012, 51, 11311.
compound. The derivative 2c, which is “CH”/N substituted at
9 M. Sun, B. Niu and J. Zhang, J. Mol. Struct.: THEOCHEM,
the X, X0 -positions, exhibits a red emission, and shows the
2008, 862, 85.
highest kRISC value, a factor that contributes to an increase in
10 C. Adamo and V. Barone, J. Chem. Phys., 1999, 110, 6158.
the emission quantum yield, observed among all of the inves-
11 A. D. Becke, J. Chem. Phys., 1993, 98, 5648.
tigated derivatives. The derivative 2d exhibited a blue emission
12 V. N. Staroverov, G. E. Scuseria, J. Tao and J. P. Perdew, J.
centred at 432 nm with the smallest Stokes shi due to a
Chem. Phys., 2003, 119, 12129.
decrease in the degree of p-conjugation. Our theoretical studies
13 Y. Zhao and D. Truhlar, Theor. Chem. Acc., 2008, 120, 215.
provide hints for the design of efficient spiro-conjugated TADF-
14 T. Yanai, D. P. Tew and N. C. Handy, Chem. Phys. Lett., 2004,
OLED emitting materials with higher kRISC values as blue and
393, 51.
red emission materials in the future.
15 J.-D. Chai and M. Head-Gordon, Phys. Chem. Chem. Phys.,
2008, 10, 6615.
Acknowledgements 16 F.-Q. Bai, J. Wang, B.-H. Xia, Q.-J. Pan and H.-X. Zhang,
Dalton Trans., 2012, 41, 8441.
The nancial support from NSFC (21173037 and 21203021) and 17 G. W. T. M. J. Frisch, H. B. Schlegel, G. E. Scuseria,
Specialized Research Fund for the Doctoral Program of Higher M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone,

This journal is © The Royal Society of Chemistry 2015 RSC Adv., 2015, 5, 18588–18592 | 18591
RSC Advances Communication

B. Mennucci, G. A. Petersson, H. Nakatsuji, M. Caricato, S. S. Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam,

X. Li, H. P. Hratchian, A. F. Izmaylov, J. Bloino, G. Zheng, M. Klene, J. E. Knox, J. B. Cross, V. Bakken, C. Adamo,
J. L. Sonnenberg, M. Hada, M. Ehara, K. Toyota, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski,
O. Kitao, H. Nakai, T. Vreven, J. A. Montgomery Jr, R. L. Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth,
J. E. Peralta, F. Ogliaro, M. Bearpark, J. J. Heyd, P. Salvador, J. J. Dannenberg, S. Dapprich, A. D. Daniels,
E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobayashi, O. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski, and
J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.

18592 | RSC Adv., 2015, 5, 18588–18592 This journal is © The Royal Society of Chemistry 2015