Synthesis of 1 - 有機電化學發光元件之高能隙離子性材料及載子平衡研究

1.5 1.0 0.5 0.0 -0.5-1.0-1.5-2.0-2.5-3.0-3.5 -0.03

-0.02 -0.01 0.00 0.01 0.02

-2.25 -2.50 -2.75 -3.00

I (mA)

E (V vs. Fc/Fc

⁺

)

Figure 2-1 Cyclic voltammogram of compound 1. All potentials were recorded versus ferrocene/ferrocenium (Fc/Fc⁺) (saturated) as a reference electrode. Inset:

Differential pulse voltammetry (DPV) of the reduction region.

250 300 350 400 450 500 550 600

Figure 2-2 Absorption (left-hand axis) and PL (right-hand axis) spectra of 1 in acetonitrile solution (10^–5 M) and in neat films in the presence and absence of [BMIM⁺][PF₆^–] (10 wt%).

400 500 600 700 800

Figure 2-3 EL spectra of device I under 3.8 V (circle) and device II under 3.4 V

(square) after 0.5-hour (open symbol) and 5-hour (solid symbol) operation. PL spectra of the emissive layers are presented for comparison. Inset: CIE coordinates of the EL and PL spectra.

0 100 200 300 400 500 600

Figure 2-4 Brightness (solid symbols) and current density (open symbols) plotted with respect to time under a constant bias voltage of (a) 3.4–4.2 V for device I and (b) 3.2–3.6 V for device II.

0 100 200 300 400 500 600

Figure 2-5 EQE (solid symbols) and power efficiency (open symbols) plotted with respect to time under a constant bias voltage of (a) 3.4–4.2 V for device I and (b) 3.2–3.6 V for device II.

Chapter 3 UV Light-Emitting Electrochemical Cells Based on An Ionic 2, 2′- Bifluorene Derivative

3.1 Introduction

Solid state LECs possess solution-processable simple device architecture and high compatibility with air-stable electrodes such as gold, silver, and aluminum.[1] In LEC devices, electrochemically doped regions induced by spatially separated ions under an applied voltage generate ohmic contacts with electrodes, resulting in balanced carrier injection and low operating voltages and, consequently, high power efficiencies. These promising advantages render this type of electroluminescent device competitive with conventional OLEDs as a cost-effective alternative for display and lighting.[94-97]

LECs based on CTMCs have shown several advantages including (i) high compatibility with ionic electrolytes imparted by intrinsic ionicity of CTMCs; (ii) high EL efficiencies due to phosphorescent nature, and (iii) tunable emission colors. Particularly, the most widely used CTMCs for single-component LEC devices are based on cationic iridium(III) complexes owing to their facile synthetic pathways and a full coverage of emission wavelength in the visible region by tailoring the structures of their chelating ligands.[8,22,23,26,30,34,62,86,99,100-103] However, intrinsic narrow energy

gaps and accessible non-radiative ligand field states of cationic iridium complexes substantially restrict their practical use for achieving blue-emitting LEC devices.[64-66] To achieve saturated blue-emitting LEC devices, we have successfully demonstrated so far the bluest electroluminescence from the LEC based on ionic terfluorene derivative.[104] The ionic fluorescent emitter was achieved by covalent tethering of 1-methylimidazolium moieties as pendant groups, rendering a hydrophobic terfluorene core bearing movable anions to form homogeneous films through spin-coating. The efficient and successful strategy for accomplishing blue-emitting LEC propelled us to further explore the molecules with larger energy gaps, which is highly desired for host materials or excitation light source for light-emitting devices.

The development of wide-gap materials is rather limited to a few functionalized structural features such as materials containing carbazoles[105]

and silane[8] moieties in order to obtain a confined conjugation length. The emissions of these wide-gap materials are generally above 400 nm, i.e., violet-blue emission. However, limited -conjugation considerably complicates the molecular design since it also constraints the molecular size and coplanarity which are related to the molecular morphological stability and luminescent properties. It is rather difficult to construct a molecule which meets most general

requirements for optoelectronic materials such as high thermal stability, good carrier mobility, and ionic (for LEC application) in a limited conjugation and dimension. As a result, with meticulous selection on the wide-gap chromophores, we have demonstrated several UV OLEDs based on 2,2′-bifluorene derivatives with remarkable short emission wavelength centered at 370 nm and external quantum efficiency up to 3.6%.[111-112] Distinguish and efficient UV emission below 400 nm is rarely reported for organic materials and has never been done for LEC application. Herein, we present the first UV LEC achieved by 2,2′-bifluorene derivative with ionic pendant 1-methylimidazolium moieties, achieving the EL as short as 386 nm. The successful demonstration of UV LEC once again proved that our judicious molecular design strategy for ionic emitters is much suitable for LEC application.

3.2 Results and Discussion

3.2.1 Materials and Experimental Methods Materials

Scheme 3-1 outlines the synthetic route of the ionic bifluorene (1). This ionic bifluorene (1) was synthesized and provided by Prof. Wong’s group of chemistry department in National Taiwan University. They obtained the bromo-substituted 2,2′-bifluorene 1-Br after homo-coupling via modified

one-pot Suzuki coupling[81] of 2-bromo-9,9-bis(6-bromohexyl)fluorene (2) in 46% yield. After treating 1-Br with excess amount of 1-methylimidazole followed by ion-exchange with potassium hexafluorophosphate, they isolated compound 1 in 81% yield.

Figure 3-1 displays the electrochemical characteristics of 1, as probed by C.

A quasi-reversible oxidation potential [at +1.04 V (vs. Fc/Fc⁺)] in acetonitrile (0.1 M n-Bu₄NPF₆ as supporting electrolyte) and a quasi-reversible reduction potential [at -2.66 V (vs. Fc/ Fc⁺)] in acetonitrile (0.1 M n-Bu₄NClO₄ as supporting electrolyte) were distinctly observed for 1. This figure was measured and provided by Prof. Wong’s group of chemistry department in National Taiwan University. They unambiguously assign both the oxidation and reduction peaks to electron transfer processes that occurred on the bifluorene backbone, consistent with the published bifluorene derivatives.[111-112] The reduction peak is less reversible in terms of a lower peak current in the reverse scan as compared with that of oxidation. It is presumably due to the pendent imidazolium units interacting with the reduced species and, consequently, retarding the diffusion toward the electrode.[86] 1 possesses a more positive oxidation and a less negative reduction than those of neutral bifluorene derivatives (where the oxidation and reduction potentials are in a range of +0.89

to +0.99 V and -2.74 to -2.94 V, respectively, vs. Fc/ Fc⁺)[111], suggesting a substantial inductive effect given by methylimidazolium units.

Experimental Methods

1H and ¹³C NMR spectra of compounds were collected on a 400 MHz spectrometer at room temperature. Photophysical characteristics of complexes in solutions were collected at room temperature by using 10^-5 M acetonitrile solutions of all compounds, which were carefully purged with nitrogen prior to measurements. The thickness of spin-coated films was ~200 nm, as measured by profilometry. UV-visible absorption spectra were recorded on a spectrophotometer (HITACHI U2800A). PL spectra were measured with a fluorescence spectrophotometer (HITACHI F9500). PLQYs for solution and thin-film samples were determined with a calibrated integrating sphere system (HAMAMATSU C9920). Oxidation and reduction potentials of all complexes were determined by CV at a scan rate of 100 mV/s in acetonitrile solutions (1.0 mM). A glassy carbon electrode and a platinum wire were used as the working electrode and the counter electrode, respectively. All potentials were recorded versus the Ag/AgCl (sat’d) reference electrode. Oxidation CV was performed using 0.1 M TBAPF₆ in acetonitrile. For reduction CV, 0.1 M TBAP in acetonitrile was used as the supporting electrolyte.

ITO-coated glass substrates were cleaned and treated with UV/ozone prior to use. A PEDOT:PSS layer was spin-coated onto the ITO substrate in air and then the structure was baked at 150 °C for 30 min. For device I, the emissive layer (ca. 400 nm) was spin-coated from an acetonitrile solution of 1. For device II, the emissive layer was spin-coated from a mixed acetonitrile solution

containing a mass ratio of 1/poly(methyl methacrylate) (PMMA) = 0.9:0.1. All solution preparing and spin-coating processes were carried out under ambient conditions. After spin-coating the emissive layer, the samples were baked at 70

°C for 10 h in a nitrogen glove box (oxygen and moisture levels below 1 ppm) and the subjected to thermal evaporation of a 100-nm-thick Ag top contact in a vacuum chamber (ca. 10^–6 torr). The electrical and emission characteristics of the LEC devices were measured using a source-measurement unit and a Si photodiode calibrated with a Photo Research PR-650 spectroradiometer. All device measurements were performed under a constant bias voltage in a nitrogen glove box. EL spectra of LEC devices were recorded using a calibrated CCD spectrograph.

3.2.2 Photophysical Properties

Figure 3-2 depicts the UV–Vis absorption and PL spectra of 1 in solution (acetonitrile, 10^–5 M) and in the form of neat films. Table 1 summarizes the

photophysical data. To increase the film quality, we also dispersed compound 1 into inert PMMA which can efficiently fill out the defect sites of the film to diminish the current leakage of the device.[14] Thus, the emission properties of compound 1 dispersed in PMMA (10 wt%) is also examined. In solution, 1 exhibits an absorption signal centered at 328 nm, which we assign to the lowest

–* transition of the bifluorene backbone. In its PL spectrum, 1 reveals intense fluorescence with an emission maximum centered at 373 nm and an excellent quantum yield of 100%. These values correspond with the published results of bifluorene derivatives.[111-112] Both absorption and PL spectra of 1 in neat films are red-shifted by ca. 10 nm as compared with those in solutions, possibly due to a certain degree of intermolecular interactions ( – stacking) in the solid state. Interestingly, by adding 10 wt% PMMA in the films of 1, no significant change is observed for the absorption and PL spectra as compared to that without PMMA, indicating that 1 is still in the form of small aggregates instead of completely dispersion in PMMA. The amphiphilic property of 1, imparted by solvophilic imidazolium and solvophobic bifluorene backbone, renders a substantial hydrophobic effect to potentially organize nano-scale architectures by non-covalent interactions.[114-115] The addition of PMMA is speculated to give certain extent of separation to these small aggregates, evidenced by the

identical absorption and PL spectra for 1 with and without the presence of PMMA, respectively. The triplet energy of 1 (2.33 eV) was determined from the highest energy vibronic sub-band of the phosphorescence spectra recorded in ethanol (EtOH) at 77 K. The low E_T value indicates a significant exchange stabilization energy of the bifluorene chromophore.

3.2.3 Electroluminescence

To examine the electroluminescence properties of 1, LEC devices based on 1 were fabricated and the device characteristics are summarized in Table 3-2.

The device structures are glass/ITO (120 nm)/PEDOT:PSS (30 nm)/emissive layer (400 nm)/Ag (100 nm) (where the emissive layer was a thin film of 1 for Device I and a thin film of 1 containing 10 wt% PMMA for Device II). In device II, PMMA was added to improve film quality and thus to decrease the leakage current during device operation, enhancing device efficiency.[12]

Measurements of both device properties were performed under constant bias voltages. Figure 3-3 compares the EL spectra of the LEC devices under 4.2 V and the PL spectra of their emissive layers. The similarity of the EL spectra of both devices indicates that the addition of PMMA did not significantly alter the EL of 1. Slight discrepancies in intensities of EL vibronic peaks between the two devices may be attributed to altered molecular potential energy surfaces of 1 in

media with different polarities[116] under applied electric fields. Notably, the signals in the EL spectra are slightly broadened at 420-500 nm relative to those in the PL spectra. It is primarily related to thermal rearrangement of the molecules and, consequently, the enhanced formation of intermolecular excimer species.[90] Since such effect is not significant in LEC devices based on 1, both devices exhibited EL emissions centered at UV region (λ_max = 386 and 388 nm for device I and device II, respectively), which are bluest EL emissions ever to have been reported for LECs. Therefore, 1 is a promising candidate for use as a UV emitting or a high-gap host material in LECs.

Figure 3-4(a) and 3-4(b) presents the time-dependent brightness and current densities when operated under 4.2 and 4.6 V for devices I and II, respectively.

The driving voltages are chosen to be close to the energy gap of 1 (3.7 eV in solutions, Table 3-1) to improve the device stability.[34] Furthermore, the UV EL emissions of both devices exhibited poor overlap with the luminosity function[117] and thus lead to relatively lower brightness (Table 3-2). Both LEC devices exhibited similar electrical characteristics. The brightness and device current first increased with time after the bias was applied, reaching maximum values before undergoing gradual decreases over time. The time required for the brightness to reach its maximum value decreased as the bias voltage increased

(Table 3-2), presumably because a higher accumulation rate of mobile ions facilitated the formation of doped regions under a higher electric field. Both devices showed similar time required to reach maximum brightness under the same bias voltage (e.g., 5.0 and 4.8 min for device I and II, respectively, at 4.2 V). This result indicates that addition of inert PMMA could not significantly alter the ionic mobility of the emissive layer possibly due to a high density of mobile ions in the films of 1 (4 PF₆^- anions per molecule). However, device II exhibited a lower maximum device current density under the same bias voltage as compared to device I (e.g., 2.27 and 1.51 mA cm^-2 for device I and II, respectively, at 4.2 V). Since the emissive layers of both devices have similar thicknesses, lowered device current density of device II may be attributed to impeded charge hopping between dispersed 1 in PMMA matrix. Furthermore, improved film quality by adding PMMA may also reduce the leakage current of the emissive layer, rendering a lower device current. The decreases in brightness and current densities over time after reaching the maximum values are associated with degradation of the emissive material during LEC operation. The maximum brightness and current density obtained in the first measurement were not fully recoverable in subsequent measurements, even under the same driving conditions. Under a constant bias, the lifetime of each device, defined as the

time required for the brightness of the device to decay from the maximum value to half of the maximum value, decreased upon increasing the bias voltage. For instance (Table 3-2), the lifetime of device I (device II) decreased from 5.8 to 3.3 min (4.9 to 3.0 min) upon increasing the bias voltage from 4.2 to 4.6 V. It results from that higher current density induced by a higher bias voltage led to a higher rate of irreversible multiple oxidation and subsequent decomposition of the emissive material, thereby accelerating the degradation of the LEC devices.[84,93]

Figure 3-5(a) and 3-5(b) presents the time-dependent EQEs and corresponding power efficiencies operated under 4.2 and 4.6 V for devices I and II, respectively. Both devices exhibited similar time evolutions of their EQEs.

Immediately after a forward bias was applied, the EQE was rather low because of unbalanced carrier injection. During the formation of the doped regions near the electrodes, the balance of the carrier injection was improved and, accordingly, the EQE of the device increased rapidly. The peak EQE and peak power efficiency were 0.15% and 0.06 lm W^–1, respectively, for device I under 4.2 V and 0.66% and 0.23 lm W^–1, respectively, for device II under 4.2 V. The device efficiency of device II is much higher than that of device I under the same bias voltage. Since the PLQYs of 1 in the films with and without PMMA

are similar (0.41 and 0.45, respectively, Table 1), enhanced device efficiency of LECs based on 1 containing 10 wt% PMMA cannot be attributed to reduced self-quenching of 1 dispersed in PMMA matrix. Possible reason for this phenomenon would come from reduced device current due to impeded charge hopping between dispersed molecules in PMMA matrix. Lower current density suppresses collision-induced exciton dissociation and consequently reduces efficiency roll-off. Thus, a higher device efficiency would be obtained at a lower device current density. In addition, reduced leakage current of the emissive layer due to improved film quality by adding PMMA may also enhance device efficiency. However, the maximum EQEs (0.66%) of LEC devices based on 1 are lower than the upper limit (~2 %) that one would expect from the PLQYs in the films of 1 containing 10 wt% PMMA (0.41, Table 3-1), when considering fluorescent spin statistics (ca. 25%) and an optical out-coupling efficiency of ca.

20% from a typical layered light-emitting device structure. It may mainly result from imperfect carrier balance of 1 in the films containing 10 wt% PMMA.

These results shown above reveal that LEC devices based on 1 can exhibit UV EL emissions and moderate device efficiencies, making them potential candidates of UV emitting materials or high-gap host materials for use in LECs.

3.3 Summary

Prof Wong’s group has synthesized a UV-emitting ionic bifluorene derivative, 1, that realized the unprecedented UV-emitting LECs. They achieved the essential ionic character required for LECs by chemically tethering methylimidazolium moieties as pendent groups to the bifluorene. Incorporating the inert PMMA in the emitting layer of device II impeded charge hopping between dispersed molecules as well as improved the film quality, leading to a reduced current density and current leakage. The highest peak EQE was over four times larger in the device with PMMA (device II) than that without PMMA (device I). The EQE and power efficiency of device II were 0.66% and 0.23 lm W^-1, respectively, at 4.2 V. Both devices I and II exhibited UV EL emissions at 386 and 388 nm, respectively, which are the first example of UV LEC ever reported.

Table 3-1 Physical properties of 1

max, PL (nm)^b PLQY^c

E1/2ox(V)^d E1/2red(V)^e 　E1/2(V)^f ET(eV) Solution^a Film Solution^a Film

1 373 385 1.00 0.45 +1.04^g –2.66^h 3.70 2.33 1 with 10 wt%

PMMA － 385 － 0.41 －－－

[a] Measured in acetonitrile (10^–5 M) at room temperature. [b] PL peak wavelength. [c]

Photoluminescence quantum yields. [d] Oxidation potential vs ferrocene/ferrocenium redox couple. [e] Reduction potential vs. ferrocene/ferrocenium redox couple. [f]The electrochemical gap E1/2 is the difference between E1/2ox and E1/2red. [g] 0.1 M TBAPF6 in acetonitrile. [h] 0.1 M TBAP in acetonitrile.

Table 3-2 LEC device characteristics

[a] Device structures: Glass/ITO (120 nm)/PEDOT:PSS (30 nm)/emissive layer (Device I:

100 wt% C6, Device II: 90 wt.% C6 + 10 wt.% PMMA) (400 nm)/Ag (100 nm). [b] EL peak wavelength. [c] Time required to reach the maximal brightness. [d] Maximal device current density achieved at a constant bias voltage. [e] Maximal brightness achieved at a constant bias voltage. [f] Maximal external quantum efficiency achieved at a constant bias voltage. [g]

Maximal power efficiency achieved at a constant bias voltage. [h] The time for the brightness of the device to decay from the maximum to half of the maximum under a constant bias voltage.

Scheme 3-1 Synthesis of 1.

Figure 3-1 Cyclic voltammogram of compound 1. All potentials were recorded versus Ag/AgCl (saturated) as a reference electrode.

Figure 3-2 Absorption (left-hand axis) and PL (right-hand axis) spectra of 1 in

acetonitrile solution (10-5 M) and in neat film or dispersed in PMMA film (10 wt%) and the phosphorescence (Phos, right-hand axis) spectrum of 1 in EtOH solutions at 77 K.

250 300 350 400 450 500 550 600

0.0

Figure 3-3 EL spectra of Device I and Device II under 4.2 V. PL spectra of the emissive layers are presented for comparison.

300 400 500 600 700

0.0 0.2 0.4 0.6 0.8

1.0

EL (Device I)

EL (Device II) PL (Neat films)

PL (Films with 10 wt% PMMA)

Inte ns it y ( a .u.)

Wavelength (nm)

Figure 3-4 Brightness (solid symbols) and current density (open symbols)

plotted with respect to time under a constant bias voltage of 4.2 and 4.6 V for (a) device I and (b) device II.

Figure 3-5 EQE (solid symbols) and power efficiency (open symbols) plotted

with respect to time under a constant bias voltage of 4.2 and 4.6 V for (a) device I and (b) device II.

Chapter 4 Improving the Balance of Carrier Mobilities of Host-Guest Solid-State Light-Emitting Electrochemical Cells

4.1 Introduction

Solid-state LECs possess several advantages over conventional OLEDs. In LECs, electrochemically doped regions induced by spatially separated ions under a bias form Ohmic contacts with electrodes, giving balanced carrier injection, low operating voltages, and consequently high power efficiencies.[1,59] As such, LECs generally require only a single emissive layer, which can be easily processed from solutions and can conveniently use air-stable electrodes, while OLEDs typically require more sophisticated multilayer structures and low-work-function cathodes.[60-61] Compared with conventional polymer LECs that are usually composed of an emissive conjugated polymer, a salt and an ion-conducting polymer,[1,59] LECs based on CTMCs show several further advantages and have attracted much attention in recent years.[4-5,9-12]

In such devices, no ion-conducting material is needed since these CTMCs are intrinsically ionic. Furthermore, higher EL efficiencies are expected due to the phosphorescent nature of CTMCs.

In general, LECs are composed of neat films of emissive materials, which

molecules. Many efforts have been made to enhance device efficiencies of LECs based on CTMCs by reducing self-quenching of the emissive materials.

Modifying the molecular structures, such as adding bulky substituents on ligands[14,38-39] or utilizing bulky auxiliary ligands[34] have been shown to suppress interchromophore interaction to some degree, improving device efficiencies of LECs. However, self-quenching is still significant in neat films even composed of materials with bulky molecular structures and thus limits device efficiencies. To further reduce self-quenching and increase EL efficiency, one feasible approach is to spatially disperse an emitting guest into a transporting host matrix, as previously reported for conventional OLEDs[61]

and solid-state LECs.[ 18,23,39,62,99,119,126] Among the reported host-guest LECs, CTMCs were the most commonly used host materials and high EQE up to 10.4% photon/electron has been demonstrated in host-guest LECs based on CTMCs.[62] To optimize device efficiencies, sophisticated molecular design for CTMC-based host materials with balanced carrier mobilities is generally

在文檔中有機電化學發光元件之高能隙離子性材料及載子平衡研究 (頁 56-200)