Role of Voluminous Substituents in Controlling the Optical Properties of Disc/Planar-Like Small Organic Molecules: Toward Molecular Emission in Solid State.

Inspired by the role of coadsorbents in dye-sensitized solar cells, a pathway to disfavor aggregation in disclike luminophores was studied to enhance solid-state emission. By restricting the intense π-π stacking using a multicyclic aliphatic ring system, we brought the lithocholic ring system as bulky side substitution into the fluorophore design. Compared to the small-size cyclohexyl substitution in BC-CY6, which exhibited a bathochromic shift in solid-state emission owing to the intermolecular interactions, lithocholic-substituted BC-LTH had reduced intense intermolecular interactions. This very bulky/voluminous side substitution (lithocholic unit) helped us extract intermolecular interaction-free molecular emission in solid state. The cyclohexyl substitution provided solid-state emission, and the broad and high Stokes shift provided an insight into stacking interactions. Face-to-face stacking-originated dimerlike species was observed in the crystal packing, which was studied by theoretical geometry optimization. The dimer species exhibited an intermolecular distance of 3.5 Å. The molecular sizes of the developed chromophores were estimated by geometry optimization, and it was concluded that the dimeric interactions in BC-LTH may not be formed owing to the voluminous nature of the side substitution present. Hence, we have been able to successfully establish through molecular level understanding the role of lithocholic functionality in tuning the optoelectronic properties of various emissive materials for different applications.

Introduction

The concern about the depleting fossil fuels has intensified in the scientific community, and this issue has guided research toward renewable energy sources. This approach is one of the many possible ways to solve the increasing global energy demand. But at the same time, it is necessary to develop efficient devices/techniques, which can work upon minimum energy with maximum output. In this context, the development of organic electronics has emerged as a useful tool, which offers excellent energy efficiency. Organic photovoltaics (conversion of solar energy to electric energy) is considered to be an excellent clean energy source, and the technologies that can use this energy efficiently are highly desired. In this regard, the development of organic light-emitting diodes (OLEDs) represents both energy-efficient and cost-effective technology. The “organic semiconducting material” (OSC) is the key behind the development of these technologies. The organic fluorophores are capable of generating light efficiently with the expense of very little electrical energy.[1,2] The journey of fluorophore development has great influence in making OLED devices, the efficient light-producing devices. The conventional fluorophores usually have disclike shapes, and when the fluorophore molecules are in close proximity, as in solid state, they undergo strong face-to-face pi–pi stacking interactions.[3] In this aggregate state, owing to the strong intermolecular interactions, the nonradiative relaxation pathways for photoexcited molecules become active, and hence fluorophores loose the emission property. This phenomenon has been discussed widely with many examples and is known as aggregation-caused quenching (ACQ).[3] As a result, the flat fluorophores are not efficient solid-state emitters, which limit the application of many efficient fluorpohores such as pyrene, perylene, and anthracene directly into thin-film applications. As a solution to ACQ designs, the twisted/nonplanar geometries are generally adopted, which prevent formation of the close face-to-face pi–pi stacking interactions when they are in close proximity. The twisted fluorophore design provides efficient light-emitting materials such as aggregation-induced emission (AIE) and thermally activated delayed fluorescence (TADF) materials.[4−7] TADF materials are current state-of-art materials as they have surpassed the theoretical upper limit of the fluorescence quantum efficiency.[8,9] Because of the design limitations, flat luminophores have been left out of the thin-film application field. The diversion of research toward the TADF and AIE materials has hampered the process of material and application development aspects related to flat luminophores. There has been many design strategies introduced to prevent the face-to-face pi–pi stacking interactions, so that the solid-state emission could be generated in the flat luminophores.[3,7,10−14] As far as the modern fluorophore design is concerned, the strategies capable of understanding the photophysical processes and developing real world applications from the conventional fluorophore are equally important as the twisted molecular design is not reliable for every application. The introduction of alkyl long chains and three-dimensional (3D) rigid cyclic systems into the flat molecular design has been utilized to prevent pi–pi stacking and to generate solid-state emission.[13,15] The alkyl substitution provides solubility to the molecular structure, and thus the spin-coating technique becomes feasible. Moreover, such a modification does not alter the optical properties of the central emitting core because of the noninvolvement nature of the sp3-hybridized aliphatic system into the conjugated emitting core. In this context, dye-sensitized solar cell (DSSC) technology also utilizes the flat molecular design but uses a very simple technique, which prevents deterring close interactions in the molecules by diluting the organic material with the aliphatic units known as coadsorbent molecules.[16−21] A similar strategy has never been employed to promote solid-state emission in flat fluorophores as it may affect dye loading every time and hence the device efficiency. The covalently linked coadsorbent dye molecules and their effect on pi–pi stacking interactions to help originate solid-state emission in flat luminophores has never been studied. Our continuous efforts to understand the molecular packing and emission efficiency resulted in the synthesis of BC-CY6 and BC-LTH, where molecular design was introduced with monocyclic (cyclohexyl) and multicyclic (acetyl derivative of lithocholic acid) derivatives of flat nonemissive .coumarin luminophores (see Scheme ). The tailored materials gained AIE and were utilized as active materials in green emissive OLED devices.

Scheme 1

Synthesis Protocol Used to Obtain Compounds BC-CY6 and BC-LTH

Results and Discussion

Photophysical Studies

The synthesis of the parent coumarin derivative was completed using a previous reported procedure, and then the required aliphatic substations were introduced at the hydroxyl functionality of coumarin (see Scheme ). The synthesized compounds were obtained in moderate yields after column chromatography purification and were then characterized using 1H and 13C nuclear magnetic resonance (NMR) spectroscopy, high-resolution mass spectroscopy (HRMS), and Fourier transform infrared (FTIR) spectroscopy. The steady-state ultraviolet–visible (UV–vis) and fluorescence studies were completed on the dilute solutions [5 × 10–6 M in tetrahydrofuran (THF)] of the compounds. All compounds showed vibronic absorption features with a short wavelength absorption maximum around 286 nm and a long wavelength absorption around 405 nm (see Figure ). The presence of a small cyclohexyl ring or bulky lithocholic derivative ring system did not enforce any visible effect on the absorbing behavior of the compounds. Similarly, the excitation wavelength nondependent photoluminescence (PL) emission maxima lie around 460 nm (λex = 341 nm) irrespective of the size of the substituent. The quantum yields and lifetime data in the DCM solution of BC-LTH and BC-CY6 have been tabulated in Table S1 (see Supporting Information). Both compounds showed similar light emission efficiency and lifetime values. The lifetime decay (see Figure S1) showed biexponential decays for both compounds. These results validated that all compounds have identical absorbing and emitting units contained in them (see Figure ). The PL emission in solid state showed variability as the PL maximum lied around 470 and 493 nm (see Figure ), with the emission of BC-CY6 more bathochromically shifted than that of BC-LTH. BC-CY6 exhibited a broad emission at 493 nm (ΦPL,solid = 14%) accompanying a bathochromic shift of 33 nm from its solution emission maximum. On the other hand, lithocholic-substituted BC-LTH showed emission maxima at 477 nm with a red shift of only 17 nm compared to its solution emission (ΦPL,solid = 23%). Comparing the size of substitution in these compounds, it is easy to understand that the broad and bathochromic emission in the former compound is due to the presence of strong intermolecular interactions, whereas in the latter compound, the bulky substitution may have prevented the close face-to-face pi–pi stacking of molecules. The observed broadness in the tail region of the PL spectra of BC-LTH might be due to the presence of different degrees of short-range interactions existing between neighboring molecules.

Figure 1

UV–vis absorption and PL profiles for BC-CY6 and BC-LTH in solution and solid state.

The behavior of steady-state UV–vis absorption and PL emission was also studied as a function of solvent polarity (see Figure S1a–d). As shown in Figure S1, neither the absorption nor the PL emission profile was shifted significantly with the change in the solvent system from dichloromethane (DCM) to dimethylsulfoxide (DMSO). The UV–vis spectra were unaffected completely, whereas a small red shift of 17 nm was observed in PL spectra of both compounds in DMSO. The UV–vis absorption and PL spectra of parent coumarin has been provided in the Supporting Information for comparison (Figure S1e,f). This small red shift validated the existence of weak intramolecular charge-transfer property of the compounds upon photoexcitation. Also, the bathochromic shift only in the PL profile of both the compounds suggested that, first the absorption occurs on the vibrational level in the excited state, which then stabilized to the low-energy state owing to the charge-transfer character, and then finally decayed radiatively to the ground state.[22]

The planar aromatic molecules are efficient fluorescent emitters in their respective solutions but the aggregation of the flat emitters debilitates the emission property in the solid state. The monocyclic (BC-CY6) and multicyclic (BC-LTH) aliphatic ring substitutions resulted in an increment in solid-state quantum yields. It must be noted that the coumarin core without any side substitution is weakly fluorescent in the solid state owing to the flat design and pi–pi stacking interactions.[13] Also, BC-CY6 showed broadness and a bathochromic shift in the PL spectra compared to BC-LTH. Thus, it urges us to study the aggregation process, which governs the origin of emission from the closely packed state. In the tetrahydrofuran (THF)–water experiment, BC-CY6 showed an increment of fluorescence intensity up to 60% addition of water into THF, and then suffered a huge decrease in the PL intensity on further addition of water. Similarly, the fluorescence increment occurred with 20% addition of water for BC-LTH, which then increased up to 60% of water addition. Further addition of water reduced the fluorescence intensity of the solution (see Figure ). Thus, the origin of solid-state emission in both compounds could be attributed to the aggregation of the molecules in the solid state.[23,24]

Figure 2

PL behavior of (a) BC-CY6 and (b) BC-LTH in THF–water mixtures.

The THF–water experiment was further assembled with the measurement of the particle size upon aggregation using the dynamic light scattering (DLS) measurement.[25] First, the particle sizes of neat THF solutions of both BC-CY6 and BC-LTH were measured to be around 20 nm [polydispersity index (PDI) = 0.127] and 9.2 nm (PDI = 0.168), respectively (see Figure S2). Then, the sizes were measured for the solutions containing 90% water as the highest aggregation was observed for these solutions. Interestingly, the particle size for BC-LTH was found to be around 178 nm (PDI = 0.172), whereas BC-CY6 showed a huge particle size in the range of 860 nm or 0.8 μm (PDI = 0.362). These studies also suggested the high aggregation tendency of the planar chromophore with smaller cyclohexyl peripheral substitution as compared to the smaller aggregates with the lithocholic unit.

Molecular Packing

To understand the molecular organization in the solid state, crystallization of the compounds was attempted. Needlelike crystals of BC-CY6 were obtained from the THF/DCM (1:1) solution, whereas BC-LTH failed to give crystals in various solvent systems. BC-CY6 crystallized in the triclinic space group P1̅ with DCM molecules trapped in the crystal structure (see Figure S3). The crystal refinement data are given in Table S1. The phenyl ring exhibited a torsional angle of 47° with the naphthalene ring plane (see Figure S4a). The molecules attained a sandwich staircase-type packing of the stacked molecules (see Figure b). The sandwiched layers exhibited an interplanar distance of 3.5 Å between the stacked molecules, and the molecules were slipped by an angle of 16° from the other (see Figure b). The next sandwiched layer was slipped by an angle of 51.8° from the first layer. Each layer had an antiparallel arrangement of the dye molecules to reduce the intermolecular interactions/steric repulsions between similar groups, which also led to the close packing of the molecules with 30% π-overlap of the naphthalene rings in closely packed adjacent molecules (see Figure S5). Also, these molecules exhibited short-range communications with the carbonyl C=O in the coumarin ring of one molecule and a similar carbonyl group of the other low-lying molecule. The molecules in the one-dimensional (1D) layer showed communication through the carbonyl C=O···H–C (naphthalene, H9) contacts. The closely packed molecules with an interplanar distance of 3.5 Å could be expected to undergo the dimer interactions, which explained the red-shifted emission to 495 nm in the solid state of BC-CY6 (see Figure S6). We believe that the presence of bulky lithocholic substitution into the BC-LTH molecular design would not allow the close intermolecular/dimer interactions upon aggregation to shift the emission to longer wavelengths.[26,27] This is also proved by the solid-state emission of BC-LTH, which showed a minimum shift from the emission observed in the solution-state that signified the absence of strong intermolecular interactions upon aggregation in BC-LTH.[28]

Figure 3

(a) Oak Ridge thermal ellipsoid plot (ORTEP) diagram and (b) crystal packing of BC-CY6 showing the interplanar distance d and slip angles.

Cyclic Voltammetry (CV) and Thermal Studies

Electrochemical analysis of both compounds in DCM was found to exhibit irreversible reduction peaks within the scanned potential window. The reduction voltage was used to calculate the lowest unoccupied molecular orbital (LUMO), and the highest occupied molecular orbital (HOMO) energy levels were obtained by adding the bandgap values to the LUMO level. The reduction onset values of −0.62 and −0.51 V provided the LUMO energy levels at −3.68 and −3.73 eV for BC-CY6 and BC-LTH, respectively (see Figure ).[29,30] From the UV–vis absorption onset of 432 nm, the optical bandgap values were calculated to be 2.88 eV for both compounds, which provided the HOMO levels at −6.55 and −6.60 eV for BC-CY6 and BC-LTH, respectively.[13]

Figure 4

(a) CV curves showing reduction potentials and (b) thermogravimetric analysis (TGA) data, for BC-CY6 and BC-LTH.

Thermal stabilities were gauged by TGA in the heating scan up to 700 °C at a heating rate of 5 °C per minute (see Figure b). BC-CY6 showed better thermal stability up to 290 °C with a decomposition temperature (5% weight loss temperature) of 335 °C. The weight loss exhibited by BC-LTH around 100 °C is due to the loss of solvent and trapped water molecules and thus lost 5% weight around 150 °C; but still, after that compound continued to lose mass, it took another 100 °C to lose further 5% (at 265 °C) of the mass of the compound. The better thermal stability of BC-CY6 is due to the close packing and presence of dimerlike interactions in the compound, which are absent in BC-LTH.

Theoretical Studies

Density functional theory (DFT) studies were carried out to understand the distribution of the frontier molecular orbital (FMO) in the designed compounds. Gaussian 09 suit of program was used to carry out the computational studies.[31] In terms of the basis set, B3LYP-6-31g(d,p) was used to optimize the geometries. Once the structures were optimized, the TD-DFT approach was utilized to calculate the respective singlet–singlet transition energies. The FMO analysis revealed that the electron density of the HOMO lies on the naphthalene ring and the coumarin ring, whereas that of the LUMO lies toward the coumarin ring with a little extension into the phenyl substituent (see Figure ). The similar distribution of the FMOs could be ascribed to the similar molecular backbone in both compounds. It is also clear that the aliphatic substitutions (cyclohexyl ring and lithocholic ring) did not contribute to the HOMO and LUMO, and thus no effect on the bandgap was observed for both compounds, which also explained the emission in the similar wavelength region. For the compound BC-CY6, the cyclohexyl ring attained the classical chair conformation and thus could be expected to have ineffective steric bulkiness around the molecules to prevent the close π–π stacking upon aggregation. The single-crystal X-ray diffraction (SC-XRD) studies also proved the presence of a small intermolecular distance in the molecules, and the dimerlike stacking was also observed (see Molecular Packing section). The dimer form was also optimized, where the intermolecular distance was found to be 3.6 Å (see Figure S6), which was very close to the experimental value (3.5 Å).

Figure 5

Optimized geometries of BC-CY6 and BC-LTH and the corresponding HOMO and LUMO energy levels.

The TD-DFT also provided insights into the excitation process, and the simulated UV–vis absorption profiles were well-matched with the experimental data (see Figures S8 and S9). The main electronic transitions and the involved molecular orbitals have been shown in the Supporting Information. For BC-CY6, the lowest lying electronic transition was attributed to the HOMO-to-LUMO transition and had a charge-transfer nature. Because of the similar molecular structures of BC-CY6 and BC-LTH, the electronic transition behavior of the latter was almost identical to that of the former.

In the absence of the crystal structure of BC-LTH, optimized geometries proved helpful and provided the structural aspects of the lithocholic functionalization of the coumarin core. As the blue-shifted solid-state PL emission spectrum of BC-LTH, compared to BC-CY6, already provided the insight into the reduced intermolecular interactions, the optimized geometry also supported the observed behavior. When the respective sizes of the side substitution in BC-LTH and BC-CY6 were compared (see Figure S10), it was observed that the lithocholic unit (17 Å) was much longer than the cyclohexyl unit (5.2 Å), whereas the coumarin core had the same size (10.9 Å). Thus, the aggregation is expected to be highly affected in BC-LTH. The bigger size of the lithocholic unit might have induced longitudinal/transverse displacements of the neighboring molecules and thus prevented the face-to-face stacking.[28] Previous efforts explored the effect of linear aliphatic chains and 3D aliphatic rings on the coumarin core but did not overcome the intermolecular interactions to that extent. But the lithocholic functionalization of planar coumarin resulted in the significant decrease in the face-to-face π–π interactions and induced a systematic blue shift in the emission wavelength compared to BC-CY6 in the solid state.[28]

Electroluminescence (EL)

Having explored these compounds for solution and solid-state photophysical properties, the compounds were then utilized as the OLED-emitting layer to study the EL properties. As the compounds BC-CY6/BC-LTH failed to give EL when utilized as neat emitters in the OLED devices, the doped device architecture was used as a solution to obtain emission from the compounds.[32,33] CBP was picked as the host matrix in which the dye molecules BC-CY6/BC-LTH were doped, and the utilized device structure is shown in Figure . The effect of varying the doping percentage of guest molecules in the host matrix was also studied by varying the dye percentage between 1 and 10% in the host matrix. The layers were coated using a spin-coater, which provides the solution-processed devices with these small molecules. The devices 1-i and 2-i with 1% of BC-LTH and BC-CY6 showed anomalous EL with an intense peak around 400 nm and a broad peak around 500 nm. When compared with the solution/solid-state PL spectra, it was concluded that the emission did not have its origin from the active emitting layers (BC-CY6/BC-LTH). The blue emission around 400 nm must be coming from the host CBP, and the greenish emission around 510 nm (see Figure ) was not explainable as both compounds did not near 500 nm. Only the emission from BC-CY6 was near 495 nm, but the similar emission from both devices aroused the suspiciousness in the EL spectra. Further, as the dye percentage was increased in devices, the blue emission diminished completely and the origin around 525–538 nm was observed for 1-iv, 1-V, 2-iv, and 2-V (7–10% doping).Because the behavior of EL was similar in both the devices utilizing BC-CY6/BC-LTH, it was, for sure, expected that the emission originated from the similar type of interactions in the OLED devices. Moreover, the shift to greenish emission (device 1-V, 2-V) from the combination of blue and greenish emission (device 1-i, 2-i) is clearly a case of exciplex emission.[34,35]

Figure 6

Doped OLED device architecture used for guest molecules (BC-CY6 and BC-LTH) in host 4,4′-bis(N-carbazolyl)-1,1′-biphenyl (CBP).

Figure 7

EL spectra for the OLED devices using BC-CY6 and BC-LTH with various doping concentrations.

It is possible that the emission originated at BC-CY6/CBP and BC-LTH/CBP interfaces because of the relaxation of the electron from the LUMO of the dye to the HOMO of the CBP layer.[36−38] We expect that the electronic relaxation at the LUMO level of the dye may not be possible because of the deep-lying LUMO of the dyes. The respective I–V–L data are shown in Figure and Table . Device 1-ii showed an operating voltage (OV) of less than 5 V with an external quantum efficiency (EQE) of 1.0%, whereas device 2-ii showed a little high OV of 5.8 V with EQE of 0.9%; but the EQE dropped rapidly for devices 2-iv and 2-V compared to 2-iii, whereas device 1-V showed an EQE of 0.5%. In brief, the functionalized BC-LTH dye showed better device performance than BC-CY6 when used as a guest emitting material in OLED devices.

Figure 8

Current density–voltage (I–V) and luminance–current density (L–I) plots of the solution-processed OLEDs by using dopants BC-LTH and BC-CY6 at various doping concentrations.

Table 1

Effects of Doping Concentrations on the OV, Power Efficiency (PE), Current Efficiency (CE), EQE, CIE Coordinates, and Maximum Luminance of the Solution-Processed Devicesa

dopant	device	dopant (wt %)	OV (V)	PE (lm/W)	CE (cd/A)	EQE (%)	CIE (x,y)	Max. Lum. (cd/m2)
@100/1000 cd/m2
BC-LTH	1-i	1	4.7/6.1	1.6/0.7	2.4/1.3	1.8/-	(0.23, 0.33)/-	1552
	1-ii	3	4.7/6.0	1.7/1.0	2.6/1.9	1.0/1.0	(0.26, 0.42)/(0.23, 0.36)	2357
	1-iii	5	4.9/6.4	1.3/0.9	2/1.8	0.7/0.7	(0.28, 0.45)/(0.25, 0.40)	2462
	1-iv	7	6.1/7.4	0.8/0.6	1.5/1.4	0.5/0.6	(0.28, 0.44)/(0.24, 0.38)	2104
	1-v	10	6.7/8.4	0.6/0.5	1.3/1.2	0.5/0.5	(0.30, 0.45)/(0.26, 0.40)	1957
BC-CY6	2-i	1	5.2/6.7	1.3/0.6	2.2/1.3	1.0/0.9	(0.24, 0.34)/(0.21, 0.28)	1703
	2-ii	3	5.8/7.1	1.3/0.7	2.4/1.7	0.9/0.8	(0.27, 0.41)/(0.24, 0.36)	2265
	2-iii	5	5.9/7.3	1.4/0.8	2.6/1.9	0.9/0.8	(0.29, 0.45)/(0.25, 0.39)	2458
	2-iv	7	5.9/8.2	0.8/0.5	1.4/1.1	0.5/0.5	(0.30, 0.45)/(0.26, 0.40)	1841
	2-v	10	5.3/7.6	0.3/0.5	0.5/1.1	0.2/0.4	(0.32, 0.46)/(0.27, 0.41)	1771

No light came out from the neat film.

Conclusions

The concept of coadsorbent molecules in preventing the molecular dyes from getting aggregated in solid state and attaining nonefficient PL behavior was accepted. As a result, the covalently bonded molecular dyes with a bulky coadsorbent unit (modified lithocholic derivative, BC-LTH) were designed and synthesized, and their effect on controlling stacking interactions on the molecular level of a planar/disc-type coumarin system was studied. The results of the multicyclic aliphatic coadsorbent substituent were also compared with the monocyclic cyclohexyl-substituted compound (BC-CY6). It was observed that the bulkier lithocholic ring system was able to prevent stacking interactions and also provided a small Stokes shift in solid-state emission compared to the BC-CY6 compound. The small Stokes shift signifies that there exist minimum interactions between neighboring molecules with enhanced intermolecular distances. Excimer or dimer formation was also not observed with BC-LTH and also showed enhanced efficiency both in solid-state emission and OLED device configuration. Thus, the present manuscript provides a strategy to prevent solid-state intense packing and may help in designing noble planar/disc-type solid-state emitters for practical applications.

Experimental Section

General

1H and 13C NMR spectra were recorded on a JEOL ECX NMR spectrometer. FTIR spectra of neat samples were recorded on a Carry-660 spectrophotometer. HRMS-electrospray ionization (ESI) spectra were recorded on a Bruker maXis impact HD instrument. UV–vis and fluorescence spectra were recorded on a Shimadzu UV-2450 and a Cary Eclipse spectrophotometer, respectively. Solid-state emission was recorded by preparing a thin film of compounds on a quartz plate using the drop-cast method. Solvents and chemicals were purchased from commercial sources and used without further purification. Spectroscopic grade solvents were used for photophysical studies. Melting points were recorded using a Stuart melting point instrument. TGA was performed on a PerkinElmer Pyris 1 and NETZSCH STA449 F1 Jupiter instrument under a nitrogen atmosphere at a heating rate of 10 °C per minute. The temperature of degradation (Td) was correlated with a 5% weight loss. SC-XRD studies were performed on an Agilent Technologies X-ray diffractometer. 3D structure visualization and exploration of crystal packing were done using Mercury software. The absolute quantum yields (both solid and solution state) were calculated using an integrating sphere on a Fluorolog instrument (Horiba Scientific). DCM solutions of compounds were used to calculate the absolute quantum yield. The lifetime data were measured on an Agilent lifetime instrument.

Computational Details

Gaussian 09 suit of program was used to carry out the computational studies. In terms of the basis set, B3LYP-6-31g(d,p) was used to optimize the geometries. Once the structures were optimized, the TD-DFT approach was utilized to calculate the respective singlet–singlet transition energies. The geometries were optimized in the gaseous state, and no solvent interactions were included in the studies. The dimer form of BC-CY6 was optimized using the geometry in the SC-XRD data. Structure visualization was done using GaussView and ChemCraft software.

CV

CV was carried out in nitrogen-purged DCM (oxidation scan) at room temperature on a Metrohm Autolab electrochemical workstation. Tetrabutylammonium hexafluorophosphate (TBAPF6) (0.1 M) was used as the supporting electrolyte. The conventional three-electrode configuration consists of a platinum disc working electrode, a platinum wire auxiliary electrode, and an Ag/AgCl reference electrode, with ferrocenium–ferrocene (Fc+/Fc) as the external standard. Cyclic voltammograms were obtained at scan rate of 100 mV s–1 in DCM solutions.

Device Fabrication

All devices were fabricated on a precleaned glass substrate coated with a 125 nm indium tin oxide (ITO) layer. The fabrication process initially involved spin-coating of an aqueous solution of poly(3,4-ethylenedioxythiophene)–poly(styrenesulfonate) (PEDOT–PSS) at 4000 rpm for 20 s to form a 35 nm hole-injection layer on the ITO anode. The emissive layer consisted of host CBP doped with dopant BC-LTH or BC-CY6. The following emissive layer solution (ELS) was prepared by dissolving the host and guest materials in the solvent THF and then sonicated for 30 min. The resulting EML solution was spin-coated on the hole-injection layer at 2500 rpm for 20 s under nitrogen-purging conditions. A 32 nm electron-transporting layer of 1,3,5-tris(N-phenylbenzimidazol-2-yl)benzene (TPBi), a 1 nm electron injection layer of lithium fluoride (LiF), and a 130 nm layer of aluminum (Al) as cathode were deposited by using the thermal evaporation method in a high-vacuum chamber at rates of 0.3, 0.1, and 10 Å s–1, respectively. The current density–voltage and luminance (I–V–L) characteristics of the resultant devices were measured through a Keithley 2400 electrometer with a Minolta CS-100A luminance meter, whereas the spectrum and CIE color chromatic coordinates were measured using a PR-655 spectroradiometer. The emission area of the devices was 25 mm2, and the luminance in the forward direction was measured.