1. Introduction

Top-emitting organic light-emitting devices (TOLEDs) based on fluorescent and phosphorescent have many future potential commercial applications, such as large area solid illumination, panel displays used in smartphones and digital cameras, because of their higher aperture ratio and display image quality [1, 2]. Noble-metal phosphors and purely organic thermally activated delayed fluorescent (TADF) emitters used in TOLEDs allow their theoretically 100% internal quantum efficiency (IQE) [3–5]. However, significant power loss arising from reflection and surface plasmon–polariton (SPP) modes exists at the metallic electrode/organic interface [6, 7]. It has been predicted by theory analysis that up to about 40% of the power lost into nonradiative SPP modes in typical TOLEDs [7, 8]. Hence, how to increase the optical extraction efficiency of TOLEDs is crucial for improving the external quantum efficiency (EQE). It is well known that SPP is transverse magnetic (TM) polarized electromagnetic surface mode occurred at the metal/dielectric interface [9] and nonradiative in nature on a flat interface, which can be excited and becomes radiative mode by employing metallic nanostructure in TOLEDs, such as metallic nanoparticles and periodically nanopatterned metal film [10–12]. Therefore, improved light extraction would be expected by exciting the SPP modes in the TOLEDs.

SPP-mediated light emission through metal films has been discussed by others in OLEDs [13]. However, it mainly focused on the enhanced photoluminescence (PL) effect. On the other hand, the cross coupling between the excited SPP modes and microcavity modes in TOLEDs with the periodic corrugation can effectively release the trade-off between the efficiency and device stability [14], which requires the SPP modes excited at the two interfaces of the metallic electrode and coupled to each other, meanwhile the wavelength of the SPP resonant should match that of microcavity modes. Taking into account the period and thickness of the corrugated metallic cathode are the main factors that affect the SPP resonant wavelength and the SPP coupling at two interfaces of the metallic cathode. In this work, we systematically optimized the period and thickness of the corrugated Ag electrode in the TOLEDs. The experimental and simulated results demonstrated that the optical extraction efficiency was obviously improved as the high efficient cross coupling between the SPP and microcavity modes occurred at the optimal period and thickness of the corrugated Ag cathode.

2. Experiment

The TOLEDs with corrugated structure were fabricated as shown in Fig. 1(a). A holographic lithography technique combined with the vacuum thermal evaporation process provides a simple approach for the fabrication of the corrugated TOLEDs with high period precision and reproducibility [14]. Two continuous interference laser beams with 325 nm wavelength (Kimmon Koha CO., LTD.) was used as an irradiance light source for holographic lithography. The periodic corrugation was fabricated on a photoresist layer on the SiO₂/Si substrate. The corrugation amplitude can be tuned by adjusting the thickness of photoresist film and the laser exposure time. The period was defined by the angle of two laser beams. A Ag anode (100 nm), a MoO₃ anodic modification layer (5 nm), a hole-transporting layer of N,N’-diphenyl-N,N’-bis(1,1’-biphenyl)-4,4’-diamine (NPB, 55 nm), an emitting layer of tris-(8-hydroxyquinoline) aluminum (Alq₃, 55 nm), and a multilayer cathode film of LiF (1 nm)/Al (2 nm)/Ag (x nm, x = 20 nm, 30 nm, 40 nm, 50 nm) were deposited on the corrugated substrate by thermal evaporation in a high vacuum system with the pressure of 5 × 10⁻⁴ Pa, sequentially. The top surface morphology of the corrugated Ag cathode layer on the TOLEDs with the period of 500 nm was characterized by an atomic force microscopy (AFM) as shown in Fig. 1(b). The groove depth of the corrugation is around 50 nm [15]. It indicates that the profile of the deposited layers essentially replicated the underlying substrate, formed a periodic corrugation throughout the structure. For comparison, a flat TOLED was also fabricated on the SiO₂/Si substrate. The active area of the device was 2 mm × 2 mm, and all of the measurements were conducted in air at room temperature.

 

Fig. 1 (a) Schematic structure of the corrugated TOLEDs. (b) AFM image of the surface morphology of the corrugated Ag cathode layer on the TOLED with the period of 500 nm.

Download Full Size | PDF

3. Result and discussion

For efficient cross coupling between the SPP and microcavity modes, the wavelength of SPP resonances should be coincided with that of the microcavity modes by adjusting the period of the corrugation. Hence, the electroluminescence (EL) spectra were measured from the experimental corrugated TOLEDs with the periods of 400, 450 and 500 nm at several observation angles off the surface normal, respectively, as shown in Fig. 2(a), 2(b) and 2(c). The EL spectra were measured by the fiber optic spectrometer, and the TOLEDs were placed on a rotational stage with grooves parallel to the rotation axis. It can be seen that the resonant wavelength of the microcavity modes determined by the thickness of the organic layer between the two metallic electrodes is around 525 nm, which coincides with the wavelength of the PL peak of Alq₃ [16]. The optical modes appeared as additional emission peaks shift in wavelength as the angle varies. Meanwhile, at the same observation angle, the additional emission peak red shifts with the period increased. When the period of the corrugation is 450 nm, the additional emission peak overlaps the microcavity emission peak at the observation angle of $0^{\circ }$. It indicates an efficient outcoupling of far-field light modes from the optical modes [14, 17]. To establish the optical modes in the EL spectra, the absorption spectra of the corrugated TOLEDs with the periods of 400, 450 and 500 nm are simulated in condition of TM modes by employing a finite difference time domain (FDTD) method, where the FDTD codes are in-house generated [14, 18, 19]. The corrugation with a sinusoidal cross section and a fill factor of 50% was employed and the refractive index of the organic materials employed in the TOLEDs was measured by the ellipsometry for the simulation. The structure of the TOLED in the FDTD simulation is Ag anode (400 nm)/MoO₃ (5 nm)/NPB (55 nm)/Alq₃ (55 nm)/Ag cathode. The grid size is set to 2 nm. The simulated dispersion maps for the corrugated TOLEDs are shown in Fig. 2 (d), (e) and (f), in which the absorption intensity appears as a function of both incident angles and absorption wavelength. The wavelength of measured emission peaks at the different observed angles extracted from the EL spectra in Fig. 2 (a)-(c) are also plotted (circles). It can be seen that there is an excellent agreement between the numerical calculated dispersion and experimental measurement.

 

Fig. 2 (a)-(c) EL spectra of the corrugated TOLEDs with the periods of 400, 450 and 500 nm at different observation angle off the surface normal, respectively. (d)-(f) Numerical calculated dispersion relation for the wavelength versus incident angle of the corrugated TOLEDs with the periods of 400, 450 and 500 nm for TM polarization, respectively. The wavelength of the measured emission peaks extracted from the EL spectra (circles) are also plotted in (d)-(f).

Download Full Size | PDF

To observe the optical modes, the spatial magnetic field (Hy) distribution across the device structure as a function of position were simulated for the corrugated TOLED with the period of 450 nm. Figures 3(a) and 3(b) show the field distribution at the illumination wavelengths of 560 and 510 nm in TM polarization with an observation angle of${10}^{\circ }$. It indicates that the emission peak at 560 nm originates from the SPP modes associated with the interface of the Ag cathode/air and Ag anode/ NPB layer, whose intensity decay exponentially into both media and reach a maximum value at the interface (Fig. 3(a)). The field at the wavelength of 510 nm is mainly confined within the microcavity, which is assigned to the standing-wave microcavity mode (Fig. 3(b)). The two modes meet at the observation angle of $0^{\circ }$. The efficient cross coupling between the two modes requires the thickness of Ag cathode comparable to or smaller than the decay length of SPP modes in order to realize a multilayer-coupled SPP on the adjacent interfaces of Ag [20–22]. Hence, the field distributions of the 450-nm periodic corrugated TOLEDs with different thicknesses of Ag cathode were simulated at the wavelength of 525 nm as shown in Fig. 3 (c)-(f). It can be seen that the field intensity of the SPP at the Ag cathode/air interface is enhancing with the thickness of the Ag cathode increasing. For the corrugated TOLED with 20-nm thick Ag cathode, the SPP modes mainly distribute at Ag anode/organic layer interface. The Ag cathode with the thickness of 20 nm is not ideal to excite the SPP mode due to the poor property of the surface smoothness and film continuity [14], [17] as shown in Fig. 3(c). For the corrugated TOLED with the 50-nm thick Ag cathode, the SPP modes mainly distribute at the Ag cathode/air interface. The intensity of SPP modes at the Ag anode /organic layer/Ag cathode interface was very weak, because the 50-nm Ag cathode is too thick and unsuitable to realize a high efficient multilayer-coupled SPP on the adjacent interfaces of Ag [22, 23]. As the thickness of the Ag cathode is 40 nm, the field intensity of the SPP modes at the two interfaces of Ag cathode are both high and exhibit an efficient coupling. Together with the cross coupling between the SPP and microcavity modes, the multilayer-coupled SPP enables the energy transfer across the cathode and out of the microcavity, which contributes to the much enhanced light extraction [12, 14, 20, 22]. Meanwhile, the Ag cathode with the appropriate thickness (40 nm in this paper) not only allows high EQE but also ensuring the device stability [14].

 

Fig. 3 Distribution of the magnetic field intensity in the corrugated TOLEDs with 30-nm Ag cathode at the wavelength of incident TM polarized light of 560 nm at ${10}^{\circ }$ (a), 510 nm at ${10}^{\circ }$ (b), and with 20, 30, 40 and 50 nm Ag cathode at the wavelength of incident TM polarized light of 525 nm at $0^{\circ }$ from (c) to (f), respectively.

Download Full Size | PDF

To verify the effect of the cross coupling between SPP and microcavity modes on the EL performance of the TOLEDs, the TOLEDs with different cathode thicknesses and with/without corrugation are fabricated. The EL performance of the TOLEDs are compared and shown in the Fig. 4. The current density-luminance characteristics were measured by Keithley 2400 programmable voltage-current source and Photo Research PR-655 spectrophotometer. The TOLEDs fabrication was repeated more than 10 times, and the performance exhibited good reproducibility. It can be seen that the luminance and current efficiency exhibit significant differences for the flat or corrugated TOLEDs with the cathode thickness increased from 20 nm to 50 nm. For the flat devices, the cathode thickness was the main factor affecting the EL performance, because the transmittance of the Ag film is very sensitive to the film thickness, and quickly decreased from 30% for 20 nm Ag to 5% for 45 nm Ag layers [14]. Therefore, for the flat TOLEDs, the device with 20 nm Ag cathode exhibits the maximum efficiency of 4.29 cd A⁻¹. For the corrugated TOLEDs, besides the period, the multilayer-coupled SPP at the adjacent interface of the Ag cathode was another factor affecting the cross coupling between the SPP and microcavity modes, which determined the light extraction enhancement of the TOLEDs. Comparing to the flat devices, all the corrugated TOLEDs with period of 450 nm exhibit higher EL performance, and the enhancement of the efficiency were 9.4%, 12.5%, 38.6% and 11.4%, for the devices with 20, 30, 40 and 50 nm Ag cathode at the current density of 100 mA cm⁻², respectively. Due to the high efficient cross coupling between the SPP and microcavity modes, the corrugated TOLED with 40 nm Ag cathode exhibits the maximum enhancement of the efficiency, which shows excellent agreement with the numerical calculated results. Meanwhile, the TOLEDs with thicker metallic cathode perform a higher stability.

 

Fig. 4 (a) Current density–voltage–luminance, (b) current density–efficiency characteristics of the corrugated and planar TOLEDs.

Download Full Size | PDF

4. Conclusion

In summary, the period and thickness of the corrugated Ag cathode are the key factors that affect the cross coupling between SPP and microcavity modes. By optimizing the period and thickness of the corrugated Ag cathode, the SPP modes realized a multilayer coupling at the adjacent interfaces of Ag cathode and the resonance wavelength of the SPP modes coincided with that of the microcavity modes, which allowed efficient cross coupling between the two modes. As a result, the efficiency of the TOLED with the optimal values of the period and thickness exhibits 38.6% enhancement.