Theoretical Investigations of the Photophysical Properties of Star-Shaped π-Conjugated Molecules with Triarylboron Unit for Organic Light-Emitting Diodes Applications.

The density functional theory (DFT) and time-dependent DFT (TD-DFT) methodologies have been applied to explore on a series of star-shaped π-conjugated organoboron systems for organic light-emitting diode (OLED) materials. The compounds under investigation consist of benzene as π-bridge and different core units and triarylboron end groups. Their geometry structures, frontier molecular orbital (FMO) energies, absorption and fluorescence spectra, and charge transport properties have been investigated systematically. It turned out that the FMO energy levels, the band gaps, and reorganization energies optical are affected by the introduction of different core units and triarylboron end groups. The results suggest that the designed compounds are expected to be promising candidates for luminescent materials. Furthermore, they can also serve as hole and/or electron transport materials for OLEDs.

1. Introduction

Organic π-conjugated luminophors have recently stirred great attention because of their potential applications as molecular probes and switches, organic light-emitting diodes (OLEDs), and solid-state light sources [1,2,3,4,5,6,7]. Particularly, π-conjugated organoboron systems have attracted considerable interest for OLEDs due to their outstanding properties, such as their high photo and chemical stability, good charge transport property, and high fluorescence quantum yields [8,9,10]. Unfortunately, the efficiency of OLEDs is still comparatively low, a shortcoming that restricts their commercialization application. To promote the efficiency of OLEDs, the luminescence and charge transport properties have become the two top-priority issues for OLEDs materials [11]. It has therefore become an urgent need to find novel materials with high luminous efficiency and carrier mobility for use in OLEDs [12,13]. Since the vacant pπ orbital of boron and the extended π-system, the π-conjugated three-coordinate organoborons exhibited strong electron accepting ability [14] and good charge transport properties [15]. In addition, variations in the chromophore part of the molecule would influence the frontier molecular orbital (FMO) energy levels and thereby the color of emission [16,17]. Namely, the FMO energy levels and the color of emission can be modulated by the introduction of different groups in the molecules. Particularly, the introduction of an electron-donating group can improve the intense luminescence and carrier mobility. Therefore, π-conjugated organoboron systems are expected to be the promising candidates for bifunctional OLEDs materials with strong fluorescence and high carrier mobility [18,19]. Moreover, comparing with a linear one, π-conjugated materials with star-shaped architecture possess the prime virtues of their good charge transport property. This kind material can enhance the π–π stacking of conjugated arms because of their high structural symmetry and planarity. Furthermore, they can also cause no an ordered, long-range, and coplanar π–π stacking due to their steric hindrance [20,21]. On the other hand, theoretical investigation also plays an essential role in the design and synthesis of materials. A number of studies demonstrate the interplay between theory and experiment, which is capable of providing useful insights into the understanding of the nature of molecules [21,22,23]. Recently, it has been reported that triarylboron-systems π-conjugated three-coordinate organoborons possessed excellent optical and charge transfer properties for use in OLEDs [14].

In this contribution, with the aim to enhance the electronic, optical, and charge transfer properties of materials, we have designed new π-conjugated three-coordinate organoborons with benzene π-bridge fragments (BBs), different core fragments (CFs), and triarylboron end groups (TBGs) (Scheme 1). The purpose of this kind molecular structure is to investigate the relationship between topologic structure and electronic, optical, and charge transfer properties, providing a demonstration for the rational design of a novel candidate for luminescent and charge transport materials for OLEDs. By applying density functional theory (DFT) and time-dependent DFT (TD-DFT) methodology, we calculated the FMO (HOMO and LUMO) energies (EHOMO and ELUMO), the HOMO–LUMO gaps (Eg), the absorption and fluorescence spectra, and the reorganization energies of designed molecules.

Scheme 1

Molecular structures of the investigated molecules.

2. Results and Discussion

2.1. Frontier Molecular Orbitals

With the aim to investigate the optical and electronic properties, the FMOs of the designed molecules were examined. The distributions HOMOs and LUMOs of the designed molecules are shown in Figure 1. The EHOMO and ELUMO, Eg, and the contributions of individual fragments (in %) to the FMOs of the designed molecules are given in Table 1. From Figure 1, one can see that both HOMOs and LUMOs exhibit π characteristics. For 1a–c, 2b, 2c, and 3b, both the HOMOs and LUMOs are spread over the CFs, BBs, and TBGs of the molecules. However, the HOMOs are mainly localized on TBGs, while their LUMOs are spread over the whole molecule for 2a, 3a, and 3c. Inspection of Table 1 reveals clearly that the HOMOs of 1a–c, 2b, 2c, and 3b are mainly localized on CFs and BBs, with minor contributions from the TBGs. Their LUMOs are mainly centralized on BBs and TBGs, with minimal density on CFs. The sum contributions of CFs and BBs for HOMOs are larger than 87.6%, while the contributions of TBGs are within 15.7%. For 2a, 3a, and 3c, the contributions of TBGs for HOMOs are larger than 90.7%, while the sum contributions of CFs and BBs are within 9.3%. On the other hand, the sum contributions of BBs and TBGs of LUMOs are larger than 84.3%, while the contributions of CFs are within 12.4%. These results reveal that different configurations of the CF and TBG building blocks have obvious effects on the distribution of FMOs. Furthermore, the contributions of CFs and BBs for 1a, 1c, 2b, 2c, and 3b are decreased, while the corresponding contributions of TBGs are increased compared with those of HOMOs, respectively. The percentages of charge transfer from CFs and BBs to TBGs are 43.2, 53.9, 73.1, 53.0, and 55.2%, respectively. In contrast, the contributions of CFs and BBs for 2a, 3a, and 3c are increased, while the corresponding contributions of TBGs are decreased compared with those of HOMOs. The corresponding charge transfer from TBGs to CFs and BBs are 38.3, 65.5, and 52.7%, respectively. For 1b, the contribution of CFs is decreased, and the contributions of BBs and TBGs are increased compared with those of HOMOs, respectively. These results reveal that the excitation of the electron from the HOMOs to the LUMOs causes the electronic density to flow mainly from CFs and BBs to TBGs for 1a, 1c, 2b, 2c, and 3b. For 2a, 3a, and 3c, the electronic densities mainly flow from TBG fragments to CFs and BBs. The electronic densities mainly flow from CFs to BBs and TBGs for 1b. It suggests that CFs and BBs serve as donors and TBGs serve as acceptors for 1a, 1c, 2b, 2c, and 3b. The CFs serve as the donor, and BBs and TBGs serve as acceptors for 1b. In general, TBGs serve as electron acceptors. Interestingly, for 2a, 3a, and 3c, the TBGs serve as donors, and CFs and BBs serve as acceptors. It is well known that the photophysical properties of intramolecular charge transfer are highly dependent on the electron donor/acceptor strength [24,25]. For 2a and 3a, the reasons may be due to the electron donating of two mesitylene groups in TBGs, which strengthens the electron donating abilities of (Mes)2B compared with those of CFs and BBs, respectively. For 3c, the TBGs serve as donors because of the strongly mesomeric effect of fluorine atoms of two pentafluorobenzenes groups in TBGs, which increases the electron donating ability of (PFB)2B compared with those of CFs and BBs.

Figure 1

The electronic density contours of the frontier orbitals for the studied compounds at the PBE0/6-31G(d,p) level.

Table 1

The FMO energies EHOMO and ELUMO, Eg (eV), and HOMO and LUMO contributions (%) of the designed molecules.

Species	HOMO				LUMO				Eg
	EHOMO	CF a	BB b	TBG c	ELUMO	CF a	BB b	TBG c
1a	−5.483	23.1	64.4	12.5	−1.693	0.4	43.8	55.7	3.789
1b	−6.055	23.7	68.6	7.6	−2.838	0.2	76.3	23.5	3.217
1c	−6.270	23.1	64.5	12.4	−2.806	0.3	33.4	66.3	3.464
2a	−6.244	2.3	7.1	90.7	−1.754	14.3	33.3	52.4	4.490
2b	−6.842	46.9	47.3	5.9	−2.794	4.3	16.7	79.0	4.049
2c	−7.006	45.0	43.8	11.2	−2.736	9.0	26.8	64.2	4.270
3a	−6.279	0.1	4.6	95.3	−2.215	34.6	35.6	29.8	4.064
3b	−7.440	18.7	74.1	7.2	−3.011	15.7	21.9	62.4	4.429
3c	−7.395	0.0	0.2	99.8	−3.066	23.1	29.8	47.1	4.328

a CFs: core fragments; b BBs: benzene π-bridge fragments; c TBGs: triarylboron end groups.

It is commonly known that the EHOMO, ELUMO, and Eg are closely related to the optical and electronic properties of molecules. For the designed molecules with nitrogen as core, 1a–c, the EHOMO and ELUMO are in the order 1a > 1b > 1c and 1a > 1c > 1b, respectively. Therefore, the Eg values is in the sequence 1a > 1c > 1b. This shows that Molecule 1a with ((mesitylene)2B, (Mes)2B) TBGs possesses higher EHOMO and ELUMO and larger Eg in comparison to molecules with ((1,3,5-tris(trifluoromethyl)benzene)2B, (FMes)2B), or ((1,2,3,4,5-pentafluorobenzene)2B, (PFB)2B) as TBGs. The reason is that –CF3 and –F in TBGs (FMes)2B and (PFB)2B is electron-withdrawing in nature, while –CH3 in TBG (Mes)2B exhibit an electron-donating character. The electron-withdrawing groups –CF3 and –F are enable the electronic density to flow from donors to acceptors, while the electron-donating group –CH3 in TBG (Mes)2B is unfavorable to the charge transfer. Moreover, the mesomeric effect of –F in (PFB)2B strengthens the electron donating ability of (PFB)2B in 1c. Therefore, the electron-withdrawing strength of (PFB)2B is weaker than that of (FMes)2B in 1b. As a consequence, the electronic density to flow from donors to acceptors for 1b is more fluent than that of 1c. Thus, the EHOMO of 1b is higher than that of 1c, whereas the ELUMO of 1b is lower than that of 1c. Similar phenomena are found for the designed molecules with benzene as core, 2a–c. For the designed molecules with 1,3,5-triazine as core, 3a–c, the sequences of EHOMO and ELUMO are 3a > 3c > 3b and 3a > 3b >3c, respectively. As a result, the order of Eg is 3b > 3c > 3a. This shows that Molecule 3a with (Mes)2B end group possesses higher ELUMO and EHOMO values and thus a smaller Eg in comparison with (FMes)2B and (PFB)2B as the end group. These can be explained by the fact that the electron-donating group –CH3 in the (Mes)2B end group of 3a improves charge transfer, while the electron-withdrawing group 1,3,5-triazine core fragment of 3b may prevent the electronic density from flowing from donor to acceptor fluently. These results indicate that the electron-withdrawing or electron-donating characters of CFs and TBGs significantly influence the Eg values for the designed molecules.

2.2. Absorption and Fluorescence Spectra

The longest absorption wavelengths λabs, electronic transitions, main assignments, and the oscillator strength f for the designed molecules are listed in Table 2. The absorptions of 1a–c, 2a–c, and 3b are assigned to the S0 → S1 electronic transitions, while the absorption of 3a and 3c are assigned to the S0 → S2 and S0 → S3 electronic transitions, respectively. The longest absorption of 1a–c, 2c, and 3b essentially mainly originates from HOMO → LUMO transition, while the corresponding absorption of 2b and 3c are mainly attributed to HOMO → LUMO+1 transitions. However, the transitions of absorption for 2a and 3a correspond to HOMO−2 → LUMO+1 transitions. Moreover, the sequence of λabs is 1b > 1c > 1a > 3a > 2b > 2a > 2c > 3c > 3b. It indicates that the designed molecules with nitrogen as core show longer absorption wavelengths than those with benzene and 1,3,5-triazine as cores. Furthermore, 2c has the largest oscillator strengths in the designed molecules. 1a–c, 2b, and 3b have large oscillator strengths, while the oscillator strengths of 2a, 3a, and 3c are smaller than those of other molecules. This indicates that 1a–c, 2b, 2c, and 3b may exhibit larger absorption intensity than those of 2a, 3a, and 3c.

Table 2

The longest of absorption wavelengths λabs, the oscillator strength f, and assignments for the designed molecules at the TD-PBE0/6-31G(d,p)//PBE0/6-31G(d,p) level.

Species	Electronic Transitions	λabs	f	Assignment
1a	S0 → S1	399	0.68	HOMO → LUMO (0.67)HOMO → LUMO+1 (0.17)
1b	S0 → S1	469	0.50	HOMO → LUMO (0.70)
1c	S0 → S1	433	0.72	HOMO → LUMO (0.69)
2a	S0 → S1	348	0.07	HOMO−2 → LUMO (0.45)HOMO−1 → LUMO (0.33)
2b	S0 → S1	361	0.72	HOMO → LUMO+1 (0.38)HOMO → LUMO+2 (0.45)
2c	S0 → S1	344	1.14	HOMO → LUMO (0.38)HOMO → LUMO+1 (0.15)
3a	S0 → S2	374	0.07	HOMO−2 → LUMO (0.38)HOMO → LUMO+1 (0.30)
3b	S0 → S1	335	0.88	HOMO → LUMO (0.39)HOMO−1 → LUMO+1 (0.36)
3c	S0 → S3	343	0.02	HOMO → LUMO+1 (0.48)HOMO−1 → LUMO+1 (0.23)

The longest fluorescence wavelengths λfl, electronic transitions, main assignments, and the oscillator strength f of the designed molecules are given in Table 3. The fluorescence of 1a, 2b, 3a, and 3b are assigned to the S1 → S0 electronic transitions, while the fluorescence of 1b, 1c, 2a, and 2c are assigned to the S2 → S0 electronic transitions. For 3c, the fluorescence is assigned to the S3 → S0 electronic transition. The LUMO → HOMO transitions play a dominant role for the fluorescence of 1a, 2a, 2b, 3a, and 3b. However, the fluorescence’s of 1b, 1c, and 2c are mainly attributed to LUMO+1 → HOMO transitions. For 3c, the fluorescence is mainly attributed to a LUMO → HOMO−5 transition. Furthermore, 1a–c, 2a, and 2c have larger oscillator strengths than those of 3a–c. The oscillator strength value of 2b is equal to that of 2b. This implies that 1a–c, 2a, and 2c exhibit fluorescence spectra that is stronger than those of 3a–c. This suggests that the designed molecules have greater fluorescence intensity and are promising luminescent materials for OLEDs, particularly for 1a–c, 2a, and 2c.

Table 3

The strongest fluorescence wavelengths λfl, the oscillator strength f, and main assignments (coefficient) of the designed molecules at the TD-PBE0/6-31G(d,p)//TD-PBE0/6-31(d,p) level.

Species	Electronic Transitions	λfl	f	Assignment
1a	S1 → S0	456	0.25	LUMO → HOMO (0.70)
1b	S2 → S0	496	0.49	LUMO+1 → HOMO (0.70)
1c	S2 → S0	499	0.57	LUMO+1 → HOMO (0.70)
2a	S2 → S0	364	0.41	LUMO → HOMO−1 (0.70)LUMO → HOMO−3 (0.20)
2b	S1 → S0	613	0.03	LUMO → HOMO (0.69)LUMO → HOMO−2 (0.10)
2c	S2 → S0	389	0.13	LUMO+1 → HOMO (0.70)
3a	S1 → S0	486	0.02	LUMO → HOMO (0.69)LUMO+2 → HOMO (0.13)
3b	S1 → S0	528	0.03	LUMO → HOMO (0.70)
3c	S3 → S0	382	0.04	LUMO → HOMO−5 (0.68)LUMO+2 → HOMO−5 (0.14)

2.3. Reorganization Energy

It is noteworthy that the small reorganization energy could be beneficial for a significant charge transport [26,27]. The reorganization energies for hole and electron of the designed molecules are calculated and listed in Table 4. Usually, N,N′-diphenyl-N,N′-bis(3-methlphenyl)-(1,1′-biphenyl)-4,4′-diamine (TPD) (λh = 0.290 eV) [28] and tris(8-hydroxyquinolinato)aluminum(III) (Alq3) (λe = 0.276 eV) [29] are the typical hole and electron transport materials, respectively. From Table 4, one can find that the λh values of 1a, 1c, 2a, 2c, and 3a–c are smaller than that of TPD. However, the λh values of 1b and 2b are larger than that of TPD. This indicates that the hole transfer rates of 1a, 1c, 2a, 2c, and 3a–c may be higher, while the corresponding hole transfer rates of 1b and 2b might be lower than that of TPD. On the other hand, the calculated λe values of the designed molecules (0.082–0.263 eV) are smaller than that of Alq3. This suggests that the electron transfer rates of the designed molecules might be higher than that of Alq3. The λh values are predicted in the order 1b > 1a > 1c for 1a–c. The sequences of λh values are 2b > 2c > 2a and 3b > 3c > 3a for 2a–c and 3a–c, respectively. This suggests that the (Mes)2B and (PFB)2B end groups can increase the hole transfer rates, while the (FMes)2B end groups decrease the hole transfer rates for the designed molecules. For the λe, the order of 1a–c and 2a–c are 1a > 1b > 1c and 2a > 2b > 2c, respectively, while the sequences of 3a–c is 3a > 3c > 3b. This implies that molecules with (FMes)2B and (PFB)2B end groups have greater electron transfer rates than those with (Mes)2B end groups, respectively. Therefore, 1a, 1c, 2a, 2c, and 3a–c are expected to be promising candidates for hole as well as electron transport materials, whereas 1b and 2b can serve as electron transport materials only.

Table 4

The predicted λe and λh (both in eV) of the designed molecules at the B3LYP/6-31G(d,p) level.

Species	λh	λe
1a	0.200	0.263
1b	0.365	0.176
1c	0.024	0.143
2a	0.069	0.156
2b	0.369	0.122
2c	0.096	0.082
3a	0.073	0.273
3b	0.270	0.134
3c	0.109	0.190

3. Materials and Methods

Computational Methods

All calculations were performed using Gaussian 09 code [30]. Geometry optimizations in the ground state (S0) and the lowest singlet excited state (S1) were carried out using the PBE0 and TD-PBE0 functionals, respectively, together with the 6-31G(d,p) basis set. On the basis of the optimized S0 and S1 geometries, the absorption and fluorescence spectra were predicted with the TD-PBE0/6-31G(d,p) method. With the aim to attest to the validity of the selected approach, N-(4-(dimesitylboryl)phenyl)-N-phenylbenzenamine (DBPB) was taken as an example because its geometric structure is similar to those of the designed molecules. The molecular structure of DBPB is presented in Figure S1. The geometry optimizations of DBPB in S0 and S1 states were carried out via the DFT and TD-DFT method using the 6-31G(d,p) basis set, respectively. The absorptions and fluorescence spectra were predicted using TD-DFT with the 6-31G(d,p) basis set based on the optimized geometries in S0 and S1 states, respectively. The various functionals for all DFT and TD-DFT computations include B3LYP, PBE0, CAM-B3LYP, wB97XD, and M062X. The longest absorptions and fluorescence wavelengths of DBPB are listed in Table S1. The results displayed in Table S1 show that the TD-PBE0/6-31G(d,p) method provided a better agreement with the reported experimental observations [14] than those obtained with other methods. Furthermore, our previous work [31] and other reports in the literature [32] suggest that PBE0 appeared notably adapted to organoboron compounds. Therefore, geometry optimizations, band gaps Eg, and the absorption and fluorescence properties of the designed molecules were carried out with the PBE0/6-31G(d,p) method.

According to Marcus electron transfer theory [26,27], two key factors that determine the charge transfer rates are the charge transfer coupling integral V and the reorganization energy λ. The V values can be obtained through the crystal data. However, the crystal structures of the designed molecules are unavailable and may be non-crystal. Therefore, we investigate the charge transport property of the designed molecules using reorganization energies λ. The λ can be partitioned into internal and external contributions. The internal reorganization energy λint is induced by structural change between ionic and neutral states [33]. The external reorganization energy λext arises from the surrounding media in bulk materials. Generally, the predicted value of λext in pure organic phases is not only small but also much smaller than its λint [34,35,36]. Therefore, we focus on the λint exclusively. The λe and λh can be calculated with the following equations [37]: where and represent the energy of the neutral species with the optimized structure of the cation and anion species, while and are the energies of the cation and anion species with the optimized cation and anion structure, respectively. , , and are the energies of the cation, anion, and neutral species with the optimized neutral structure, respectively. Our results have thus been compared with results reported in the literature [36,37]. The λ for electron (λe) and hole (λh) were predicted at the B3LYP/6-31G(d,p) level.

4. Conclusions

In the study, a series of star-shaped π-conjugated organoboron compounds have been designed and systematically investigated for OLED applications. By applying density functional theory (DFT) and time-dependent DFT (TD-DFT) methodology, we calculated the FMO (HOMO and LUMO) energies, the HOMO–LUMO gaps, the absorption and fluorescence spectra, and the reorganization energies of designed molecules. It turned out that the optical, electronic, and charge transport properties are affected by the core fragments and TBGs. Our results suggest that the designed molecules can serve as luminescent materials for OLEDs. In addition, they are expected to be promising candidates for hole and/or electron transport materials.