Low-Turn-On-Voltage, High-Brightness, and Deep-Red Light-Emitting Electrochemical Cell Based on a New Blend of [Ru(bpy)3]2+ and Zn-Diphenylcarbazone.

Deep red light-emitting electrochemical cells were prepared based on a blend of [Ru(bpy)3]2+, a cationic complex, and a neutral Zn(II)-complex based on diphenylcarbazone ligands, named Zn(DPCO). The crystal structure of the Zn(DPCO)2 (bpy)] molecule revealed that the DPCO ligand has been deprotonated to form DPCO- and coordinated to the Zn center metal through the C=O and N=N moieties of DPCO. From the cyclic voltammetry results and controlled potential coulometry data of the diphenylcarbazide (DPC) ligand, it is possible to establish that DPC is oxidized in an irreversible process at +0.77 V, giving DPCO and later oxidized at a higher potential (+1.32 V) to produce diphenylcarbadiazone (DPCDO). A detailed assignment of UV-vis spectra futures to determine the origin of ground- and excited-state transitions was achieved by time-dependent density functional theory calculations, which showed good agreement with the experimental results. Using a simple device architecture, we obtained deep red electroluminescence (EL) with high brightness (740 cd m-2) and luminous efficiency of 0.39 cd/A at a low turn-on voltage of 2.5 V. The favorable configuration of the cell consists of only a blend of complexes of indium tin oxide as the anode electrode and molten alloy cathode (Ga/In) without any polymer as the transporting layer. The comparison between [Ru(bpy)3]2+ and [Ru(bpy)3]2+/Zn(DPCO) demonstrates a red shift in the EL wavelength from 625 to 700 nm in the presence of Zn(DPCO), revealing the importance of using blends for future systems.

Introduction

Organic light-emitting diodes (OLEDs) were introduced for lowering the costs, producing high-brightness emissions, and easily tuning the color of electroluminescence (EL) through a variety of emitters.[1] However, the fabrication of OLEDs simplifies the sequential deposition of several layers, some of which consist of instable interlayers and electrodes, implying a rigorous encapsulation of the device and, thus, posing a great challenge for low-cost commercialization. In the past decade, scientists proposed a single-layer light-emitting device called light-emitting electrochemical cells (LEECs)[2] that overcame some of the limitations of OLEDs. LEECs generally consist of a single layer of ionic transition-metal complexes (iTMCs) sandwiched between two electrodes. LEECs can also include a bilayer device, with a thin layer of conductive polymer such as poly(3,4-ethylenedioxythiophene)polystyrene sulfonate (PEDOT:PSS) located between the indium tin oxide (ITO) glass and the emitter in order to improve the hole mobility and smooth the electrode surface.[3] Because of the ionic nature in LEEC, which is the fundamental condition to operate the device, a redistribution of ions under an applied voltage allows the maintenance of a low turn-on voltage (the typical voltage required to turn on the LEEC device) without the use of injecting layers when the material’s ionic conductivity is enough,[4] resulting in a very simple architecture that substantially reduces the cost of fabrication. The highest brightness and external quantum efficiency (EQE) were achieved for LEECs based on iTMCs (iTMC–LEECs) as the emitter layer.[5] As the first LEEC employing a polymer in the ionic media was introduced in 1996,[6] several studies have been carried out in order to enhance the performance of the device.[7] [Ru(bpy)3]2+ as a benchmark luminescent material attracts great interest in both research and industry carriers because of its outstanding photophysical and photochemical properties such as extremely long excited-state lifetime (τ ≈ 1 μs), high photoluminescence quantum yields (Φ ≈ 0.095) in solutions at room temperature, and intense absorption band in the visible region.[8] As an emitting layer and electron-transport material, aluminum trisquinoline (AlQ3) was the first and most studied efficient low-molecular-weight OLED material. It attracted considerable attention because of its remarkable emission properties such as high brightness, high EQE, and fine color-tuning.[9] The quinoline ligand of AlQ3 acts as a N^O-donor ligand that can be coordinated to metals such as Al3+, Zn2+, or Pt2+. Here, we introduced a blend of [Ru(bpy)3]2+ and a Zn(II) complex containing a π-extended derivate of quinoline named diphenylcarbazone (DPCO) as the emitter layer in an LEEC for the first time. Similar to the quinoline derivate, carbazides act as N^O-donor ligands. DPCO has interesting properties such as easy coordination to metal, a π-extended system, and pH sensitivity that modulate the emission color. The application of a blend of an organic polymer as the host material is an important method for obtaining desirable EL behavior. Recently, a tris-chelated complex of Zn(II) with a ligand of 2,2′-bipyridyl, [Zn(bpy)3]2+, was successfully employed as a promising additive to the cyclometalated–iridium complex emitter to improve the LEEC features.[10] However, there is a gap in the literature about the influence of blending two low-weight molecules on the EL behavior of LEECs.

In this study, we focused on the DPCO ligand and its correlation with the benchmark tris(2,2′-bipyridine)ruthenium(II) complex named [Ru(bpy)3]2+ and the blend of [Ru(bpy)3]2+/DPCO, highlighting the importance of EL in a new class of the blend of the [Ru(bpy)3]2+ emitter.

Results and Discussion

The complexes were synthesized with an ca. 80% yield, substituting the N^N of [Zn(N^N) (DPCO)3–] by the corresponding ancillary polypyridinic ligand phen for Zn(LH1) and bpy for Zn(LH3), when x = 1 and its homoleptic complex (x = 0) for Zn(LH2) with three DPCO moieties. The complexes were characterized by Fourier transform infrared spectroscopy (FT-IR) (see Supporting Information, Figure S1). Two obvious important frequencies around 1630 and 120 cm–1 in the FT-IR spectra of the complexes can be assigned to (C=O) and (N=N) bands, respectively, because of the coordination of ketonic oxygen and nitrogen in the N=N group to the metal.[11] Moreover, the 1H NMR spectra in a low-temperature condition clearly showed the formation of complexes (Supporting Information, Figure S2). Characterization by CHN micro analysis confirmed the successful synthesis of the complexes. In addition, a single crystal of one complex, which was suitable for X-ray single-crystal analysis, was analyzed to determine the accurate chemical structure (Supporting Information, Tables S1 and S2, and Figure S3). The challenge still remains for determining the coordination modes of the keto–enol form of the DPCO ligand and the number of metal-coordinated atoms. Some reports on the coordination of monodentate and bidentate modes of N-amide and N^O were also reported.[12] Here, the crystal structure indicated the coordination of DPCO through the N^O mode to the metal (Figure ). The structure of C36H30N10O2Zn, Zn(LH3) was a mononuclear six-coordinated octahedral zinc(II) complex of the P21/c space group. The ZnII ion was coordinated by one N atom and one O atom from every DPCO ligand, and two N atoms from the 2,2-bipyridyl ligand. The Zn–O(1) and Zn–N(3) bond distances were estimated to be 2.037(4) and 2.222(5) Å, respectively, which are typical for most Zn–N^O complexes. The bond distances of N(3)–N(4) and N(7)–N(8) equaled 1.281(7) and 1.273(7), respectively, indicating the double bond between two N atoms, whereas N(5)–N(6) and N(9)–N(10) were, respectively, 1.342(8) and 1.335(7), which could be assigned to the single bond between two N atoms. Furthermore, the bond lengths of 1.286(7) and 1.288 (7) for O(1)–C(17) and O(2)–C(30), respectively, indicated the formation of C=O. All these crystallographic data confirm that the DPCO ligand was deprotonated to form DPCO, resulting in Zn2+ in Zn(LH3). This evidence was proved from the difference density map. An overview of the studies on the crystal structure of the DPCO ligand and its complexes proved that there is no coordination mode including N=N and C=O of diphenylcarbazide (DPC) similar to Zn(LH1)–Zn (LH3).

Figure 1

Oak Ridge Thermal Ellipsoid Polt (ORTEP)  of Zn(LH3). Thermal ellipsoids are at the 30% probability level.

The UV–vis spectra of the ligand and complexes are shown in Figures and 3, respectively. As previously reported in the literature, DPCO has one absorption band in the visible region, which can be attributed to the equilibrium of keto–enol forms.[13] On the basis of Scheme , ligand DPC has four acidic hydrogens that can deprotonate at different values of pH. Interestingly, the color of the solution of DPCO changed from pale pink to dark cherry when increasing the pH from 7 to 12, respectively (upper section of Figure ). In pH levels above 12, there was no change in the UV–vis spectrum. Moreover, the intensity of the maximum absorption band in the visible region increased with increasing the pH. These changes indicate the deprotonation of the N–H groups in the basic medium.[14] The UV–vis spectra for the complexes are similar to those of the DPCO ligand, along with a red shift of the maxima in the visible band and significantly enhancing the molar extinction coefficient (ε) associated with the coordination of DPCO to the metal.

Figure 2

UV–vis absorption spectra of the DPC ligand in varied pH levels in the mixture of the 10–5 M H2O/EtOH (50%:50%) solution. The inset of the figure is the correlation between epsilon in different wavelengths and pH. The color variation through different pH levels is also shown in the upper section.

Figure 3

Electronic absorption spectra of complexes in CH3OH.

Scheme 1

Chemical Structures of Zn(LH1)–Zn(LH3) Complexes

To explore the Ox/Red properties of the ligands and complexes, we also performed an electrochemical analysis (Figure and Supporting Information, Figure S4).

Figure 4

Cyclic voltammetry of Zn(LH1)–Zn(LH3) in acetonitrile (ACN). The employed electrodes and electrolyte including the working electrode = platinum disc, counter electrode = platinum wire, reference electrode = Ag/AgCl and the supporting electrolyte = tetrabutylammonium perchlorate in ACN solutions (0.1 mol L–1).

A proved mechanism for the oxidation of DPC consists of two steps: first, DPC oxidizes to DPCO and then the oxidation of DPCO produces DPCDO in the potential of around 0.83 (V) and 1.5 (V), respectively.[15] In the reversed potential, the cathodic reduction of oxidation products of DPC was formed because of the abstraction of proton by triethylamine, which was observed in the form of one peak of −0.3 versus Saturated Calomel Electrode (SCE), corresponding to the DPCO reduction to DPC. The Ox/Red behavior of complexes is the same as that of the DPC ligand in the positive region except for a shifting of about 0.4–0.5 V to lower potentials. Nevertheless, some reduction processes appeared, indicating the influence of the polypyridyl ancillary ligand in the reversed potential condition. The π-conjugated system of the DPCO ligand allows for the easy transfer of electrons, revealing the electrochemical activity of DPCO (Supporting Information, Figure S5).

The half-wave of reduction of the complexes was relatively broad compared to their ligands because of the overlapping of the half-waves of ancillary ligands and DPCO. Because of the presence of assignable absorption bands in the UV–vis spectra and Ox/Red half-waves in the cyclic voltammetry (CV) experiments of the DPCO ligand in Zn(LH1)–Zn(LH3) complexes, the influence of the DPCO ligand on their different complexes was investigated. Specific data are given in Table. Controlled potential coulometry (CPC) was carried out to calculate the number of electrons that are exchanged during the Ox/Red process. To distinguish the origin of peak Ia (II) at the potential of 1.5 V, the applied potential was kept in that potential for 1600 s in the solution in varied amounts of DPC (Figure a). The number of moles of DPC has a linearity function with the variation of charge in the solution media along with the correlation coefficient of about 0.99, and the slope of the Q = f(t) formula is equal to t on F. According to the above explanation and formula, 4.07 ± 0.05 of electrons was achieved for the oxidation of DPC, which means that the DPC at the applied potential was probably transformed to DPCDO through the equilibrium of 4e–, 4H+. CPC was also conducted to determine the products in the reduction process of DPCO at the applied potential of −0.94 V for 1000 s by employing different amounts of the DPCO/DPC mixture (40:60). Figure b illustrates the family of curves obtained for these compounds. For this reaction, the number of electrons is 1.97 ± 0.05, which was obtained from the slope of the mentioned function, which equals nF.

Table 1

UV–Vis and Redox Properties of Zn(LH1)–Zn(LH3)

comp.	absorbance λmax(log ε)	Eoxa	Eredb
Zn(LH1)	228(5.03), 268(5.06), 530(3.95)	0.72, 1.23	–0.68, −1.13
Zn(LH2)	232(5.09), 271(5.01), 526(4.55)	0.70, 1.13	–0.17, −0.78, −1.26
Zn(LH3)	228(4.16), 280(4.23), 525(4.44)	0.63, 0.97	–0.41, −0.88
[Ru(bpy)3]2+	245(4.4), 290(4.91), 451(4.17)	1.34	–1.33

The oxidation potential values.

The reduction potential values

Figure 5

The time dependence of the charge involved in the controlled potential coulometry at (a) 1.4 V (SCE) and (b) −0.94 V (SEC) for different amounts of DPCO dissolved in 0.3 M triethylamine.

Time-dependent density functional theory calculation was employed to obtain a new insight into the ground- and excited-state transitions and subsequently simulated absorption spectra. The energy level along with the optimized structure of Zn(LH1)–Zn(LH3) are depicted in Figure . The calculated Highest Occupied Molecular Orbital (HOMO) and Lowest Unoccupied Molecular Orbital (LUMO) energies of Zn(LH1) and Zn(LH3) complexes were −4.68 and −2.62 eV, whereas those of the Zn(LH2) complex equaled −4.56 and −2.53 eV, respectively. The calculated HOMO–LUMO energy gaps were the same for Zn(LH1) and Zn(LH3) complexes (2.06), whereas it was 2.03 for Zn(LH2) (Supporting Information, Table S3).

Figure 6

Isosurfaces (isodensity contour = 0.03) for the HOMO and the LUMO of the complexes obtained from the DFT method through the B3LYP/LANL2DZ basis set.

It can be observed that the HOMO of the complexes has an amplitude on the Zn center and the DPCO ligand, whereas LUMO is localized on the DPCO ligand. The assignment of the calculated S0/T1 transition for these complexes is summarized in the Supporting Information, Table S4, and Figure demonstrates their Frontier orbitals involved in the main compositions of the S0/T1 transition.

For the Zn(LH1) complex, the T1 state originates mainly from HOMO → LUMO + 1, corresponding to electron promotion within DPCO and from Zn to DPCO. The main conclusion of the Frontier orbitals’ analysis is that the contribution of DPCO ligands in complexes is higher than that of another ligand, as a result contributing to electron transitions (Supporting Information, Table S4).[16] Also, there is good agreement between simulated UV–vis spectra obtained from time-dependent density functional theory calculation and experimental data (Supporting Information, Figure S6).

As the structure of the DPCO ligand is similar to Alq3 as the N^O donor ligand, we performed the LEEC test with the configuration of ITO/complex/Ga/In. The advantage of this configuration is its simplicity for two reasons: first, using only a single active layer between two electrodes without any polymer, even PEDOT:PSS; second, utilizing low-melting-point alloy cathode contacts that have the potential for LED production by printing and inject methods to escape vacuum evaporation techniques.

Moreover, a blend of two metal complexes, [Ru(bpy)3][ClO4]2 and Zn(LH1–3), was employed to achieve deep red EL emission. Different concentrations of the blend were tested and characterized. The I–V–L characteristics of Zn(LH3) in different W/W of [Ru(bpy)3][ClO4]2/Zn(DPCO) are summarized in Figure and Table .

Figure 7

Current–voltage and luminance–voltage plots of a single-layer LEEC using ITO/[Ru(bpy)3(ClO4)2]/Ga/In and ITO/Ru(bpy)3(ClO4)2/Zn(LH3)/Ga/In. W/W of Ru(bpy)32+/Zn(LH3) for 4:1, 4:2, and 4:3 named (A–C), respectively. The voltage scan rate was 0.05 V/s.

Table 2

EL Spectral Data of ITO/[Ru(bpy)3]2+–Zn(LH3)/Ga/In/Epoxy Devices

W/W = [Ru(bpy)3]2+/Zn(LH3)	Jmaxa	Vonb	Lmaxc	LEd
4:3	35	3.1	150	0.45
4:2	80	2.8	250	0.32
4:1	200	2.5	740	0.39

Maximum current density [A m–2] at the scan rate of 50 mV/s.

Turn-on voltage (the typical voltage required to turn on the LEEC device) (V).

Maximum luminance [cd m–2].

LE: luminous efficiency [cd A–1] at 4 V.

Regarding the optimization of doping concentration, we found a deep broad red EL for Zn(LH3) with 4/1 of W/W of [Ru(bpy)3]2+/Zn(DPCO) along with the high luminance of 740 cd/m2 and high luminous efficiency of 0.39 cd/A.[16,17] These values are close to those obtained with other Zn(LH) complexes in a variety of concentrations. Surprisingly, although the emission intensity of the blend was relatively decreased with increasing the amount of the Zn(LH) complex, the EL wavelength was significantly red-shifted to the near-infrared region (Figure ), achieving a deep red EL emission at approximately 700 nm with CIE coordinates of 0.710 and 0.280, regarding the EL wavelength of 625 nm for [Ru(bpy)3][ClO4]2 with CIE of 0.681 and 0.308 (white reference CIE(x, y) = (0.31; 0.33).[12] Another crucial point is that the applied turn-on voltage (Vturn-on) of the fabricated device is just 2.5 V, which is the lowest value among the reported Vturn-on for deep red light electrochemical cells to date.[5,7,17−19] Under an applied bias, ClO4– counter ion in [Ru(bpy)3]2+ complexes drift, leading to the accumulation of negative counter-ion and cationic Ru and Zn complexes in close proximity to holes and electrons, respectively.[20]Figure demonstrates the proposed schematic representation of charge transfer in the fabricated LEEC. However, because a barrier energy of approximately 1.5 eV is produced at the interface of [Ru(bpy)3]2+/Zn(DPCO) complexes, it is difficult for the holes to be injected into Zn(DPCO) complexes. Therefore, holes (electrons) will be blocked by Zn(DPCO) complexes and accumulate at the interface of [Ru(bpy)3]2+/Zn(DPCO) complexes.

Figure 8

EL spectra of [Ru(bpy)3(ClO4)2] and a blend of Ru(bpy)3 [ClO4]2/Zn(LH1)–Zn(LH3), W/W = 4:1.

Figure 9

Schematic representation of a state-of-the-art LEEC based on [Ru(bpy)3]2+–Zn(LH). The movement of ions in the single layer under an applied voltage allows for efficient charge carrier injection from air-stable electrodes.

In both sides, near the [Ru(bpy)3]2+/Zn(DPCO) complex interfaces, the electric field in the bulk is redistributed, and the electric field in the [Ru(bpy)3]2+/Zn(DPCO) layer moves higher than the one in the[Ru(bpy)3]2+ layer alone. This explanation is in good agreement with the observed red-shifted and broadening EL, which is routinely seen in Electroplex emission.[21]

As evidence to confirm the Electroplex emission, the amount of accumulated electrons and holes in the interface was gradually increased with increasing the applied voltage, which was observed in our fabricated LEEC cells as shown in Figure .

Figure 10

El spectra of Zn(LH3)/Ru(bpy)32+ at different applied voltages.

Conclusions

The X-ray crystal structure of the Zn(LH3) complex clearly indicated a rare coordination mode of the DPCO ligand to a metal center including the coordination of C=O and N=N of every DPCO ligand to Zn. The electron density map also showed that the DPCO ligand has been deprotonated and then coordinated to metal, confirmed by CV and CPC. The blend of [Ru(bpy)3]2+/Zn(DPCO) complexes confirms promising EL behavior when operated in a solid-state LEEC, even with a simple air-stable cathode, reaching the high luminescence of 740 cd·m–2 and luminous efficiency of 0.39 cd·A–1 at only 2.5 V. A red-shifted spectra from 625 to 700 nm was achieved through the modification of the W/W of blend compositions.

Experiments

We prepared a new family of Zn complexes of general formula [Zn(DPCO)2(X)] (Scheme ) where X = bipyridine (bpy), phenanthroline (phen), and DPCO = diphenylcarbazone as an N^O bidentate ligand with an extended aromatic system; fully characterized; and utilized as a blend of [Ru(bpy)3]2+ emitter in the single-layer LEEC (Supporting Information, S1). The schematic structures of the complexes Zn(LH1)–Zn(LH3) are shown in Scheme .

Synthesis and Characterization

[Zn(phen) (DPCO)2] = Zn(LH1)

The heteroleptic Zn(II) complexes were synthesized by standard procedures according to Scheme . In a typical two-pot synthesis, Zn(CH3COO)2·XH2O (0.03 g, 1 mmol) and phenenthroline (0.18 g, 1 mmol) were dissolved in 5 mL of methanol and refluxed under inert atmosphere for 8 h. Then, the DPC ligand (0.06 g, 2 mmol) was added to the solution and refluxed for 4 h under N2. Subsequently, the solvent was evaporated and the deep violet powder was collected. The product was then washed with water, acetone, and ether three times and dried in air at room temperature. Yield: 82% (84 mg). Anal. Calcd for C38H30N10O2Zn (%): C, 63.03; H, 4.18; N, 19.34. Found (%): C, 63.03; H, 4.19; N, 19.34 (MW, 724.12). 1H NMR (400 MHz, −50 °C, CD3OD): 8.32 (N-ortho CH (m), d, 1H), 8.26 (N-ortho CH of phen (n), d, 1H), 8.18 (C-ortho quadrupole CH of phen (j), s, 2H), 8.10 (C-para quadrupole CH of phen (k), d, 1H), 8.08 (N-para CH of phen (i), d, 1H), 7.82 (N-meta CH of phen (l), d, 1H), 7.80 (N-meta CH of phen (h), d, 1H), 7.57 (N-ortho CH of DPCO (a), d, 4H), 7.49 (N-ortho CH of DPCO (d), d, 4H), 7.39 (N-para CH of DPCO (c), t, 2H), 7.31 (N-meta CH of DPCO (b), t, 4H), 7.26 (N-para CH of DPCO (f), t, 2H),7.18 (N-meta CH of DPCO (e), t, 4H), 6.98 (NH of DPCO (a1), s, 2H). ESI-MS m/z: 721.18, [M – H]+.

[Zn(DPCO)3]Na = Zn(LH2)

The homoleptic Zn(DPCO) was synthesized in a typical one-pot synthesis. In a typical two-pot synthesis, Zn(CH3COO)2·XH2O (0.03 g, 1 mmol) and DPC (0.1 g, 3 mmol) were dissolved in 5 mL of ethanol and refluxed under inert atmosphere for 8 h. Finally, the solution was treated with the saturated aqueous solution of NaClO4 and gave a deep violet precipitate. It was washed several times with water to remove traces of salts, and then dried in air at room temperature. Anal. Calcd for C39H33N12O3NaZn (%): C, 58.11; H, 4.13; N, 20.85. Found (%): C, 58.02; H, 4.16; N, 20.93. 1H NMR (400 MHz, −50 °C, CD3OD): 7.53 (N-ortho CH of DPCO (a), d, 6H), 7.49 (N-ortho CH of DPCO (d), d, 6H), 7.39 (N-para CH of DPCO (c), t, 3H), 7.33 (N-meta CH of DPCO (b), t, 6H), 7.25 (N-para CH of DPCO (f), t, 3H),7.20 (N-meta CH of DPCO (e), t, 6H), 6.99 (NH of DPCO (a1), s, 3H). ESI-MS m/z: 780.18, [M – H]+.

[Zn(bpy) (DPCO)2] = Zn(LH3)

Zn(LH3) was prepared starting from Zn(CH3COO)2·XH2O (0.03 g, 1 mmol), bpy (0.039, 1 mmol), and DPC ligand (0.065 g, 2 mmol), using the same procedure as described for Zn(LH1) to yield the product. Anal. Calcd for C36H30N10O2Zn (%): C, 61.76; H, 4.32; N, 20.01. Found (%): C, 61.76; H, 4.33; N, 20.01. 1H NMR (400 MHz, −50 °C, CD3OD): 8.30 (C-ortho quadrupole CH of bpy (k), d, 1H), 8.25 (C-ortho quadrupole CH of bpy (j), d, 1H), 8.19 (N-ortho CH of bpy (n), d, 1H), 8.15 (N-ortho CH of bpy (g), d, 1H), 8.10 (N-para CH of bpy (l), t, 1H), 7.97 (N-para CH of bpy (i), t, 1H), 7.90 (N-meta CH of bpy (m), t, 1H), 7.87 (N-meta of bpy (h), t, 1H), 7.55 (N-ortho CH of DPCO (a), d, 4H), 7.50 (N-ortho CH of DPCO (d), d, 2H), 7.39 (N-para CH of DPCO (c), t, 2H), 7.34 (N-meta CH of DPCO (b), t, 4H),7.29 (N-para CH of DPCO (f), t, 2H), 7.23 (N-meta CH of DPCO (e), t, 4H), 6.98 (NH of DPCO (a1), s, 2H). ESI-MS m/z: 697.18, [M – H]+.