Dependence of the Color Tunability on the H₂Pc Thickness in DC-Voltage-Driven Organic Light-Emitting Diodes

Abstract

Dependence of the color tunability on the metal free Phthalocyanine (H₂Pc) layer thickness in DC-voltage-driven organic light-emitting diodes (OLEDs) was investigated. A H₂Pc layer was employed as a blue/red emission layer, which was prepared on an Alq₃ green emission layer. The thickness of the H₂Pc layer varied from 5 to 30 nm, with a step of 5 nm. The fabricated color-tunable OLEDs (CTOLEDs) were subjected to a thermal treatment layer for 2 min at a temperature of 120 °C to improve the interface properties, especially between H₂Pc and Alq₃. The current density–voltage–luminance characteristics and Commission Internationale de L’Eclairage (CIE) coordinates of the CTOLEDs with and without thermal treatment were measured, and their energy band diagrams were analyzed with respect to the H₂Pc thin film thicknesses. In addition, the recombination rates at the interfaces between the hole transport layer and Alq₃ and the H₂Pc/electron transport layer of the CTOLEDs with and without thermal treatment were theoretically investigated using a technology–computer-aided design (TCAD) program. The experimental and theoretical results showed that the emission color temperature from cool white to warm white at a low voltage can be controlled by adjusting the thickness of the H₂Pc layer in the CTOLED. It was verified that the thermally treated H₂Pc thin film layer acted as a barrier that prevented electrons from being transferred to the Alq₃ at low applied voltages, resulting in white color emission with temperature tunability. The CTOLED with a 20 nm of H₂Pc layer demonstrated the best stable interface state and stability, resulting in the lowest driving voltage, relatively high luminance, and optimal light emission uniformity, respectively.

1. Introduction

Organic light-emitting diodes (OLEDs) have been extensively studied for several decades owing to their unique properties including full vivid colors, rapid response times, wide angles of view, self-luminosity, and so on. In the display industry, OLEDs have gradually replaced conventional liquid-crystal displays (LCDs) and plasma display panels and have been already commercialized, especially in large area TVs, cell phones, and monitors [1,2,3]. As per the other application areas of OLEDs, they have also been considered as a next generation lighting source that can replace conventional incandescent bulbs and fluorescent lighting along with light-emitting diodes [4]. In particular, in lighting technology, many studies on emotional lighting that could change color according to a human being’s mood have been reported on [5,6].

Therefore, there have been several reports concerning shifting emission color by an applied voltage [7,8,9,10]. Accordingly, various types of color-tunable OLED (CTOLED) structures and their characteristics have been reported thus far [7,8,9,10]. Up to now, the reported conventional CTOLEDs’ structures were fabricated by using a polymer blend [7] or have multi-tandem structures with multiple electrodes [8], which might increase manufacturing costs and require complex manufacturing processes. Therefore, to be used as emotional lighting by utilizing the unique advantages of OLED, it should require stable operation and simple structures of the CTOLED lighting sources with a DC driving voltage [9,10].

Among them, we recently reported on DC-voltage-dependent CTOLED lighting without additional electrodes by inserting a H₂Pc thin layer into a green OLED, employing tris (8-hydroxyquinoline) aluminum (III) (Alq₃) as an emission layer (EML) [11]. In the previous study, the thermal treatment effect of a CTOLED employing a metal-free phthalocyanine (H₂Pc) thin film with a fixed thickness of 5 nm on the Alq₃ layer was investigated. The CTOLED without any thermal treatment emitted green dominantly at the Alq₃/HTL interface and emitted blue and infrared with relatively low intensities at the H₂Pc/HBL interface in the entire voltage regime due to unstable H₂Pc thin film by interface traps. Even though the CTOLED without any thermal treatment had an unstable H₂PC interface, it could transfer very few electrons beyond the H₂PC thin film thicker than 5 nm in CTOLEDs, which may lead to blue emission dominantly at the H₂Pc/HBL interface with relatively low bias. The dominant color conversion from blue to green and to white was observed from the thermally treated CTOLED device with a H₂Pc thickness of 5 nm, according to DC voltage driving. Despite undergoing thermal treatment, a thin film of several nanometers may act as an unstable factor in the device’s performance during operation. Therefore, studies to obtain the optimal thickness of a H₂Pc layer in CTOLEDs with and without thermal treatment are necessary for a more stable device operation and color conversion.

In this study, we have investigated the dependence on the H₂Pc layer thickness of the color tunability in CTOLEDs with and without thermal treatment to obtain an optimal thickness of the H₂Pc layer for stable color conversion properties. The thickness of the H₂Pc thin film layer varied from 5 to 30 nm, with a step of 5 nm. The H₂Pc layer was deposited on the Alq₃ layer in CTOLEDs, which was employed as the EML. The thickness of the H₂Pc thin film was optimized by analyzing the electroluminescence (EL) spectra and current density–voltage–luminance (J-V-L) characteristics obtained from the CTOLEDs. From the observation, the thickness of H₂Pc layer was optimized with a stable color shift.

2. Experiment and Simulation Method

2.1. Experiment

Figure 1 demonstrates a schematic of the CTOLED lighting device [11]. Organic materials were deposited on a pre-patterned 150 nm thick indium tin oxide (ITO) anode layer with a sheet resistance of 10 ohm/square using a thermal evaporation system at a base pressure of 3 × 10⁻⁶ torr, and the deposition rate was precisely controlled by adjusting the tooling factor. The ITO-coated glass substrate was cleaned via sonication in acetone, isopropyl alcohol, and deionized water, followed by UV–ozone irradiation. After preparing all the organic materials, a 150 nm thick Liq/Al layer was e-beam-evaporated on the organic layer as a cathode. A typical glass encapsulation method was used to protect the organic layer from atmospheric water vapor and oxygen. The active emission area, defined by the shadow mask, was 2 × 2 mm². More details about device’s structure and experimental procedures were reported on in a previous report [11].

The following various types and thicknesses of the organic layers were used: 30 nm KHI-001, 70 nm KHT-001, 30 nm Alq₃, H₂Pc with a variable thickness, 10 nm bathocuproine (BCP), and 30 nm LG-201, which were incorporated as the hole injection layer (HIL), hole transport layer (HTL), two EMLs, hole blocking layer (HBL), and electron transport layer (ETL), respectively. The H₂Pc and Alq₃ layers were employed as the EMLs. The thickness of the Alq₃ layer was fixed to 30 nm, whereas that of H₂Pc layer varied from 5 to 50 nm to investigate the thickness effect of the CTOLED on the device’s performance. The fabricated CTOLEDs were subjected to thermal treatment for 120 s at the temperature of 120 °C [12].

The EL spectra and typical J-V-L characteristics were tested using a J-V-L Measurement system M6100, which consisted of a spectroradiometer CS-2000, Keithley 2400 source meter, and Keithley 2100 digital multimeter [11].

2.2. Simulation Method

The electrical and optical characteristics in the CTOLED were investigated by using a technology–computer-aided design (TCAD) simulation program [13]. The current in the organic was calculated as a space–charge-limited current (SCLC) by the Poisson equation and the current continuity equation at a high voltage [14]. Material parameters included the electron affinity (LUMO), band gap energy (LUMO + band gap = HOMO), relative permittivity, and density of state. Carrier transport in organic materials was represented by Poole–Frenkel [15] and hopping mobility models [16]. The recombination rate for the exciton generation rate was calculated by the Langevin model, as follows [17]:

$$R=\frac{qnp}{\epsilon }\left[{\mu }_h\left(F\right)+{\mu }_e\left(F\right)\right],$$

(1)

where μ_e and μ_h are the mobilities of electrons and holes, respectively, and F is the electric field. The rate of single exciton generation is given by [18]:

$$\frac{\partial S\left(x,y,t\right)}{\partial t}=R_{st}{R}_{np}+\frac{R_{st}}{1+R_{st}}{K}_{TT}{T}^2-K_{ISC}S-K_{ST}ST-K_{SP}S\left(n+p\right)-\frac{K_{SS}}{2}{S}^2-K_{NRS}S-\frac{S}{{\tau }_S}+\nabla \left(D_S\nabla S\right),$$

(2)

where S(x,y,t) is the spatial and temporal singlet exciton density, R_st is the fraction of singlets formed during Langevin recombination, R_np is the Langevin recombination rate, KTT is the triplet–triplet constant, KISC is the intersystem crossing constant, KST is the singlet-triplet constant, KSP is the singlet–polaron constant, KSS is the singlet–singlet constant, KNRS is the singlet non-radiative decay constant, τ_S is the singlet radiative decay lifetime, and D_s is the singlet diffusion constant. D_s is represented as

$$D_S=\frac{L_{DS}}{{\tau }_S},$$

(3)

where L_DS is the singlet diffusion length. Additionally, the trap tunneling model [19] was included. A localized defect density of the state on the interface [20] was used for the defect model, and a total trap density of acceptor-like states with a Gaussian distribution, having an average energy level of 3.2 eV and a Gaussian width of 0.6 eV at the interface of the H₂Pc and Alq₃ layer, were set as 1 × 10¹¹ cm⁻³ in the defect model [13,21].

3. Results and Discussion

An energy band diagram of the CTOLED analyzed in this study is presented in Figure 2 [11]. In the proposed CTOLEDs, excitons are formed at the H₂Pc and Alq₃ layer, thus blue/weak red colors are emitted from H₂Pc and green colors from Alq₃, respectively [22,23]. It is well known that there are two energy band gaps in H₂Pc material: the Soret (B)-band and Q-band (visible). Excitons generated in the B-band and Q-band of the H₂Pc can be created and dissipated independently of each other and can play different roles in the recombination process [24,25]. In Figure 2, the white-dashed line denotes the energy bandgap of the Q-band of the H₂Pc.

Since the lowest unoccupied molecular orbital (LUMO) of the Alq₃ layer is higher than those of H₂Pc and HTL as shown in energy band diagram (see Figure 2), the quantum well structure can be formed. H₂Pc layer also acts as a thin potential barrier against the electrons injected from the cathode through the ETL, together with a blue/weak red emission layer by the B-band and Q-band. Thus, providing that energy states in the potential well are matched to incident electrons, electrons can be confined to potential well through potential barrier by quantum mechanical tunneling, resulting in resonant-tunneling-phenomenon-assisted electroluminescence [26,27].

3.1. Experimental Results

Figure 3a,b demonstrate the two main Gaussian peaks extracted by the multipeak fitting analysis [28] of the EL spectra and CIE coordinates of the CTOLED without any thermal treatment [11] at an applied low voltage of 10 V, with respect to the H₂Pc thin film thicknesses of 5, 10, 15, 20, and 30 nm, respectively. As the H₂Pc thin film thickness increases, the peak emission intensity at 453 nm corresponding to blue increases, whereas that at 518 nm corresponding to green decreases, and the color coordinates shift from a single green to single blue. Based on the analysis of the EL spectrum and color coordinates, when the H₂Pc thin film is sufficiently thick, it becomes stable, as in the result of the thermal treatment [11].

Figure 4a,b demonstrate the three main Gaussian peaks extracted by the multipeak fitting analysis [28] of the EL spectra and CIE coordinates of the CTOLED with thermal treatment at the applied low voltage of 10 V, with respect to the H₂Pc thin film thicknesses of 5, 10, 15, 20, and 30 nm, respectively. As the H₂Pc thin film thickness increases, the peak emission intensities at 453 nm corresponding to blue are nearly the same, and those at 518 nm corresponding to green are also nearly the same, and all color coordinate values are the same as a single blue. The peak emission at 710 nm is attributed to the Q-band of H₂Pc [29]. The emission intensity at 710 nm is strongest at an H₂Pc thickness of 10 nm. When the H₂Pc thickness is higher than 10 nm, the EL intensity decreases. When the H₂Pc thickness is 30 nm, the EL intensity is very weak. The reason why the EL intensity of 710 nm decreases as the H₂Pc thickness decreases can be interpreted as quenching due to the H₂Pc molecule-aggregation tendency [29]. It can be seen that as the H₂Pc thickness increases above 10 nm, the EL intensity decreases as more molecules of H₂Pc can be quenched. It was verified that the H₂Pc thin film layer acts as a barrier that prevents electrons from being transferred to the Alq₃ at a low applied voltage, resulting in white color emission. It can be analyzed that the interfaces are stable, with almost no traps due to the thermal treatment [11].

3.2. Simulation Results

The recombination positions and ratios of CTOLEDs with H₂Pc thin film thicknesses of 5, 10, 15, 20, and 30 nm were investigated for considering and ignoring interface traps using the TCAD simulation program [13].

Figure 5 and Figure 6 show the simulated recombination rates of CTOLEDs considering and ignoring traps at the Alq₃/HTL and H₂Pc/HBL interfaces, with respect to H₂Pc thin film thicknesses at the applied voltage of 10 V before the color conversion, respectively. In the TCAD simulation, it is impossible to consider two bands at the same time; thus, it is the result considering only B-bands. Filled green squares and empty blue circles denote the simulated recombination rates at the Alq₃/HTL and H₂Pc/HBL interfaces, respectively. The left and right insets in Figure 5 and Figure 6 show the recombination rates of the CTOLED with H₂Pc thicknesses of 5 and 30 nm, respectively. In Figure 5, as the thickness of H₂Pc thin film increases, the simulated recombination rates at the Alq₃/HTL interface decrease. On the other hand, those at the H₂Pc/HBL interfaces increase. Since the LUMO level of H₂Pc (2.27 eV) is lower than that of HBL (3.5 eV), most electrons are accumulated at the H₂Pc/HBL interface, causing the H₂Pc thin film to act as a thin potential barrier, which interferes with the flow of electrons moving to the Alq₃ layer. The holes are fully collected into the Alq₃ and H₂Pc layers because the HBL can also act as a potential barrier. Accordingly, the excitons can be generated by the recombination between the holes in the Alq₃ and H₂Pc layers and the electrons reaching the H₂Pc/HBL interface. The recombination rate is in proportion to the number of excitons [30]. However, the CTOLED without any thermal treatment may have traps (defects) in the H₂Pc and at its interface; thus, electrons arriving at the H₂Pc/HBL interface can be accumulated at the Alq₃/HTL interface beyond the potential barrier of the thin H₂Pc film by thermionic emission, resonant tunneling, and trap-assisted tunneling [26,31]. As shown in Figure 5, in the case that 5 nm of H₂P_C layer was employed, the recombination rate of the Alq₃/HTL interface is greater than that of the H₂Pc/HBL interface, but, in the other case that 30 nm H₂Pc layer was introduced, the recombination rate of the Alq₃/HTL interface is much less than that of the H₂Pc/HBL interface. The thinner the H₂Pc, the easier the electrons can move into the Alq₃ layer due to the influence of the trap. As the H₂Pc thickened, it was revealed that it was difficult for the electrons to move into the Alq₃ layer, even with the help of the trap. Accordingly, the recombination rate of the Alq₃/HTL interface significantly reduced, and the emission color was dominated by H₂Pc. Even when the trap was distributed on the interface, the number of electrons moving to Alq₃ adjusted through the thickness of H₂Pc. The experimental results (see Figure 3) of the CTOLED without any thermal treatment and simulation results (see Figure 5) of the CTOLEDs considering traps confirmed that the emission color at a low voltage before color conversion can be controlled by adjusting the H₂Pc thin film thickness for CTOLEDs. In Figure 6, as the H₂Pc thin film thickness increases, the simulated recombination rates at the Alq₃/HTL and H₂Pc/HBL interfaces are almost the same. Irrespective of the thickness of the H₂Pc film, most of the electrons are accumulated at the H₂Pc/HBL interface for recombination. The H₂Pc film operates as a potential barrier, making it difficult for electrons to reach the Alq₃ layer without the help of traps. If the H₂Pc layer is thin below 10 nm, some electrons can move to the Alq₃ layer due to thermionic emission or trap-assisted tunneling. At a low bias, it was calculated that electrons could not reach the Alq₃ layer when the thickness of the H₂Pc thin film was more than 10 nm. If traps were not distributed at the interface, few electrons moved to Alq₃ due to thermionic emission, and most of them were accumulated in H₂Pc and emitted blue dominantly. The experimental results (see Figure 4) of the CTOLED with thermal treatment and theoretical results (see Figure 6) of the CTOLEDs ignoring traps confirmed that the emission color at a low voltage before color conversion could be controlled by reducing the H₂Pc interface traps due to the thermal treatment of the CTOLED.

3.3. Optimal H₂Pc Thickness

Figure 7a,b demonstrate the J-V and L-V characteristics of the CTOLED with different H₂Pc thin film thicknesses, respectively. The CTOLEDs were thermally treated [11]. The black squares, red circles, blue triangles, green diamonds, and purple stars denote the J-L-V characteristics of the CTOLED with H₂Pc thin film thicknesses of 5, 10, 15, 20, and 30 nm, respectively. Figure 7c presents the photographs of the three CTOLEDs with H₂Pc thin film thicknesses of 10, 20, and 30 nm. The photographs, shown in Figure 7c, were taken by using a Nikon D700 camera with maximum resolution of 12.1 megapixels, and the distance between the samples and its lens was about 7 cm. Since it was important to obtain the photos that were the closest to the actual images, we had selected a camera model that had excellent color reproducibility. In this figure, color conversion according to the change in the applied voltage was observed in all the three CTOLEDs. Blue emission was observed at an applied voltage of 11 V from all three CTOLEDs and persisted around an applied voltage of 14 V. When the applied voltage exceeded 14 V, color conversion from blue to white started to occur with H₂Pc thin film thicknesses of 10, 20, and 30 nm. The applied voltages at the time of the color conversion were 14.8, 14.4, and 14.2 V, respectively. Considering the J-V characteristics of all the CTOLEDs, peak-to-valley (PV) current patterns that typically occurred when the flow of electrons was controlled by resonant tunneling [26,27] were observed. PV ratios of the CTOLED with H₂Pc thin film thicknesses of 5, 10, 15, 20, and 30 nm were approximately 6.86, 6.66, 9.47, 5.04, and 1.9, respectively. Figure 8 shows applied voltages and luminances at the first peak luminance point, which were extracted from L-V characteristics, shown in Figure 7b. When the first peak points of the current density or luminance were reached, a color conversion from blue to green occurred [11], as shown in the photographs in Figure 7c. As the H₂Pc thin film thickness increased up to 20 nm, the driving voltages decreased; however, other than the 20 nm thickness of the H₂Pc layer, the driving voltage increased. As the H₂Pc thin film thickness increased, the luminances tended to decrease, but a small peak existed only at a H₂Pc thickness of 20 nm. When the CTOLED with a H₂Pc thin film thickness of 20 nm was thermally treated, the driving voltages were the lowest, and relatively high luminances and uniform emissions were observed.

It can be thought that the quality of the prepared H₂Pc thin film layer, which was employed as an EML, plays a critical role in the color conversion and J-V characteristics. As shown in Figure 7c, when the H₂Pc thin film thickness was 20 nm, the driving voltage was the lowest and the light emission uniformity was the best, further indicating that a thin film with a stable interface state between H₂Pc and Alq₃ was formed.

Figure 9a–d show the CIE coordinates of the CTOLEDs with thermal treatment at different applied voltages for t_H2Pc = 10, 15, 20, and 30 nm, respectively. Color was tuned from blue (at applied voltage of 10 V) after going through white emission (at applied voltages of 13.7~14 V) to green (at applied voltage of above 14 V), as shown in Figure 7c. The CIE coordinate of each CTOLED slightly varied from cool white to warm white, which denoted that the color temperature of the CTOLEDs could be controlled by adjusting the thickness of the H₂Pc layer. The blue emission at low voltages of 10 V was changed at the voltages above 14 V, which meant that the emission color could be controlled as well as the color temperature in the case of white emission. A similar phenomenon was observed from the fabricated CTOLEDs, regardless of the thickness of H₂Pc, although there was a slight difference in CIE coordinates. It was evident that CTOLEDs, which could be tuned in color and color temperature, could be applied to the emotional OLEDs lighting industry.

4. Conclusions

In conclusion, the influence of the thickness of the H₂Pc layer as the EML prepared on one Alq₃ layer on the CTOLEDs was investigated. The J-V-L characteristics, energy band diagrams, and CIE coordinates of the CTOLEDs with and without thermal treatment were characterized and analyzed, with respect to the H₂Pc thin film thickness. As the H₂Pc thin film thicknesses varied, the recombination position and ratio of the CTOLEDs considering and ignoring interface traps were investigated by using the TCAD simulation program. From the TCAD simulation, considering interface traps, we observed that the recombination rates at the Alq₃/HTL interface decreased as the H₂Pc thin film thickness increased, whereas those at the H₂Pc/HBL interface increased. From the TCAD simulation, ignoring interface traps, as the H₂Pc thin-film thickness increased, the recombination rates at the Alq₃/HTL and H₂Pc/HBL interfaces were nearly the same. Those experimental and simulation results confirmed that the traps were significantly reduced by thermal treatment, and thin H₂Pc, which is thin enough to be 5 nm thick, effectively acted as a potential barrier that prevented electrons from being transferred to Alq₃ at low driving voltages. As the thickness of H₂Pc layer varied, the color temperature of the white emission also slightly varied from cool to warm white, which meant our study could be applied to the emotional lighting industry. In case that the CTOLED with a H₂Pc thin film thickness of 20 nm was thermally treated, a stable interface state was formed; thus, the driving voltages were the lowest, and uniform emissions and relatively high luminances were observed.