Manipulating Refractive Index in Organic Light-Emitting Diodes - ACS

Mar 1, 2018 - It is known that reducing the refractive indices of the constituent layers in an OLED can enhance light extraction. Among all of the lay...
5 downloads 0 Views 2MB Size
Download PDF
Research Article Cite This: ACS Appl. Mater. Interfaces 2018, 10, 9595−9601

www.acsami.org

Manipulating Refractive Index in Organic Light-Emitting Diodes Amin Salehi,† Ying Chen,‡ Xiangyu Fu,§ Cheng Peng,§ and Franky So*,‡ †

Department of Physics and ‡Department of Material Science and Engineering, North Carolina State University, Raleigh, North Carolina 27695, United States § Department of Material Science and Engineering, University of Florida, Gainesville, Florida 32611, United States

Downloaded via UNIV OF TOLEDO on September 27, 2018 at 06:25:24 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: In a conventional organic light-emitting diode (OLED), only a fraction of light can escape to the glass substrate and air. Most radiation is lost to two major channels: waveguide modes and surface plasmon polaritons. It is known that reducing the refractive indices of the constituent layers in an OLED can enhance light extraction. Among all of the layers, the refractive index of the electron transport layer (ETL) has the largest impact on light extraction because it is the layer adjacent to the metallic cathode. Oblique angle deposition (OAD) provides a way to manipulate the refractive index of a thin film by creating an ordered columnar void structure. In this work, using OAD, the refractive index of tris(8hydroxyquinoline)aluminum (Alq3) can be tuned from 1.75 to 1.45. With this low-index ETL deposited by OAD, the resulting phosphorescent OLED shows nearly 30% increase in light extraction efficiency. KEYWORDS: organic light-emitting diodes, oblique angle deposition, spectroscopic ellipsometry, refractive index, electron transport layer

■

INTRODUCTION

layers of an OLED can significantly enhance the outcoupling efficiency. The refractive index of thin films can be reduced by pore inclusion using oblique angle deposition (OAD).5−7 In the OAD process, substrates are positioned at an oblique angle with respect to the incident vapor flux. Upon adsorption of the initial vapor species on the substrate and the formation of the first islands, self-shadowing of these islands prevents the vapor flux from reaching the shadowed regions, leading to an ordered, porous, and columnar growth.8 Because the pore features are much smaller than the wavelength of the light, the resulting film does not scatter light.5,6,8 Instead, it behaves as a homogeneous medium of void and solid material, and the refractive index of the resulting film deposited by OAD can be significantly lower than that of the solid film.5,7,8 The refractive index can be determined from the film porosity using an effective medium approximation (EMA) model,9,10 and the porosity can be tuned by the angle of incidence of the vapor flux. A larger angle leads to a larger self-shadowing and hence a larger porosity; therefore, OAD is a powerful tool to control the film porosity and hence the refractive index. This technique to deposit lowrefractive index films is applicable to most evaporable materials because it is a physical method and it merely depends on the surface diffusion of the adsorbed species.8 Low-refractive index

In organic light-emitting diodes (OLEDs), only a small fraction of light escapes from the device. A significant amount of the radiation is coupled to the bound photonic modes in the multilayered device and cannot be extracted to air or the substrate. The bound modes can either propagate in the bulk of the constituent thin films as waveguide (WG) modes or propagate along the metallic electrode interface as surface plasmon polaritons (SPPs). The density of both “leaky” and bound photonic modes in any optical multilayered stack is strongly dependent on the refractive indices of the organic as well as the indium tin oxide (ITO) layers.1 We will show later that lowering the refractive index of the entire OLED stack to 1.5 can result in an extraction efficiency of almost 90% into the substrate and air. As for the individual layers in an OLED stack, lowering the refractive index of each layer can have a different impact on the specific loss channels. For instance, the SPP photonic mode dispersion is a strong function of the refractive index of the electron transport layer (ETL) which is the layer adjacent to the metal cathode.2,3 A low-refractive index ETL can significantly increase the light outcoupling efficiency because of the reduced loss to SPPs.3,4 In addition to the ETL, lowering the refractive index of the other organic and ITO layers can reduce light coupling to the thin-film guided modes and suppress Fresnel reflection from the boundaries. Therefore, the ability to manipulate the refractive index of the constituent © 2018 American Chemical Society

Received: December 5, 2017 Accepted: March 1, 2018 Published: March 1, 2018 9595

DOI: 10.1021/acsami.7b18514 ACS Appl. Mater. Interfaces 2018, 10, 9595−9601

Research Article

ACS Applied Materials & Interfaces Table 1. Maximum Achievable Air (ηA) and Air + Substrate (ηA+S) EQEsa case refractive index of anode (ITO) refractive index of HTL (TAPC) refractive index of EML (CBP) refractive index of ETL (Alq3) max achievable air (ηA) max achievable air + substrate (ηA+S)

reference device

device A

device B

2.1 1.68 1.8 1.75 31.6% 63.1%

1.5 1.5 1.5 1.5 32.5% 87.8%

2.0 2.0 2.0 2.0 21.0% 43.4%

device C device D device E 2.1 1.68 1.8 1.5 38.4% 73.2%

2.1 1.68 1.8 2.0 25.2% 50.7%

1.5 1.68 1.8 1.75 26.6% 69.4%

device F device G 2.0 1.68 1.8 1.75 31.0% 63.2%

2.1 1.5 1.8 1.75 36.6% 63.0%

device H 2.1 2.0 1.8 1.75 25.0% 51.7%

device I device J 2.1 1.68 1.5 1.75 28.5% 59.4%

2.1 1.68 2.0 1.75 32.5% 63.7%

a Maximum achievable air (ηA) and air + substrate (ηA+S) for optically optimized OLED devices. All devices have the following structures: ITO (100 nm)/TAPC (optically optimized thickness)/CBP−Irppy2acac (20 nm)/Alq3 (optically optimized thickness)/aluminum (100 nm). For the reference device, the following indices of refraction are used: nITO = 2.1, nTAPC = 1.68, nCBP = 1.8, and nAlq3 = 1.75 at 525 nm. To evaluate the effect of refractive index, ηA and ηA+S are calculated using different values of indices of refraction (under lined in the table) for different cases.

thin films deposited using OAD have been reported for different transparent oxide materials, such as SiO2 and ITO,10,11 and antireflection coatings using graded-index ITO thin films have been demonstrated in GaInN light-emitting diodes.6 Although the use of OAD for OLEDs has been reported, the focus of the previous works was only limited to creating scattering surfaces to extract WG modes11−13 and manipulating the refractive index of the organic layers for light extraction has not been explored. Here, we show that the low-index ETL deposited by OAD can increase the external quantum efficiency (EQE) of a green phosphorescent OLED by nearly 30%.

guided modes and Fresnel reflections and the suppression of the SPP mode due to the low-index ETL. In the cases where we only changed the refractive index of the ETL in devices C and D, we found that decreasing the ETL’s refractive index from 2 to 1.5 increases ηA and ηA+S from 25.2 and 50.7 to 38.5 and 73.2%, respectively. Compared to those of the other constituent layers, the refractive index of the ETL has the largest impact on light extraction. An ETL with a low refractive index can dramatically increase the light extraction efficiency through suppressing the loss to SPPs, following the SPPs dispersion equation and the Fermi’s golden rule.3,4,14 In the cases where we only changed the refractive index of the ITO layer in devices E and F, reducing the refractive index of ITO decreases the air contribution (ηA) but increases the overall air + substrate (ηA+S) contribution. The results are consistent with the previous results that using poly(3,4ethylenedioxythiophene)-polystyrene sulfonate (PEDOT:PSS) (a refractive index of 1.5) as an anode can increase ηA+S to more than 60%.15−18 In an OLED, the transverse electric waveguide (TEWG) mode centers at the ITO layer and we will show later that reducing the refractive index of the ITO layer will extract this WG mode into the substrate mode. In devices G and H, where we changed the refractive index of the HTL, decreasing the refractive index of this layer from 2.0 to 1.5 increases ηA and ηA+S from 25.0 and 51.7 to 36.6 and 63.0%, respectively. However, comparing device G with the reference device, whereas ηA increases from 31.6 to 36.6%, there is no change in ηA+S. The lack of changes in ηA+S implies that lowering the refractive index of the HTL will increase air contribution at the expense of a smaller substrate contribution. Finally, in devices I and J where we changed the refractive index of the EML, we found that the refractive index of the EML has a very small effect on light extraction. However, unlike other layers, decreasing the refractive index of the EML has a negative effect on light extraction. When the refractive index of the EML is smaller than that of the ETL, the entire dipole radiation will be coupled into the ETL, resulting in a slightly stronger coupling to the SPP loss modes of the metallic cathode. When the refractive index of the EML is larger than the ETL, there is a total internal reflection at the EML/ETL boundary and a portion of the dipole radiation is not coupled into the ETL because of the kinematic restriction of the wavevector. As the refractive index of the ETL has the largest impact on light extraction, in this work, we will attempt to reduce the refractive index of the ETL in an OLED device by OAD. We chose Alq3 as the electron transport material for our study.

■

RESULTS AND DISCUSSION A conventional OLED is composed of an aluminum cathode, organic active layers including an ETL, an emissive layer (EML), a hole transport layer (HTL), and an ITO transparent electrode on a glass substrate. Because reducing the refractive indices of the organic layers and the transparent anode can result in light extraction enhancements,3,4,14,15 we did optical simulations for an OLED to evaluate the effect of refractive indices of the constituent layers on light extraction. The device used for this optical study has the following structure: ITO (100 nm)/TAPC (optically optimized thickness)/CBP− Irppy2acac (20 nm)/Alq3 (optically optimized thickness)/ aluminum (100 nm). Here, TAPC is di[4-(N,N-di-p-tolylamino)-phenyl]cyclohexane, CBP is 4,4′-bis(carbazol-9-yl)biphenyl, and Irppy2acac is bis[2-(2-pyridinyl-N)phenyl-C](acetylacetonato)iridium(III). We varied the refractive indices of the constituent layers and calculated the maximum achievable EQE extracted to “air” (ηA) and to “air + substrate” (ηA+S) for 11 different cases, as presented in Table 1. In the case of the reference device, we calculated its light extraction efficiency using the actual refractive indices of the organic as well as ITO layers: nITO = 2.1, nTAPC = 1.68, nCBP = 1.8, and nAlq3 = 1.75. In the second and third cases, all layers have the same refractive index. For the rest of the cases, we kept the actual refractive indices of all of the layers unchanged and only varied the refractive index of one constituent layer at a time to evaluate its effect on light extraction. For the reference device, the maximum achievable ηA and ηA+S values are 31.6 and 63.1%, respectively. In devices A and B, where the indices are the same, as the refractive indices of all organic and ITO layers are reduced from 2.0 to 1.5, ηA and ηA+S values increase from 21 and 43.4 to 32.5 and 87.8%, respectively. The results show that when the OLED stack is a homogeneous medium with an index of refraction of 1.5, ηA+S reaches almost 90% because of the elimination of thin-film 9596

DOI: 10.1021/acsami.7b18514 ACS Appl. Mater. Interfaces 2018, 10, 9595−9601

Research Article

ACS Applied Materials & Interfaces

Figure 1. (A) Optical constants of the Alq3 film deposited at 80°. (B) Optical constants of Alq3 deposited normally. (C) Refractive index of Alq3deposited thin films as a function of angle of deposition at 525 nm, from ref 9. (D) Cross-sectional SEM image of ∼600 nm of Alq3 deposited at 80° on top of 90 nm B3PYMPM deposited at 0° on top of 300 nm of SiO2.

room temperature. This is due to the much lower glasstransition temperature of TAPC and CBP compared to Alq3. According to the Thornton’s structure zone model (SZM),22 the ratio between the substrate temperature (Ts) and the melting temperature (Tm) of the evaporated material must be smaller than 0.2, that is, Ts/Tm < 0.2, for a columnar porous film to form in OAD. The SZM is an empirical model explaining the film morphology of mainly metal thin films8,23 and has been modified for other materials, such as oxides, composites, and so forth.24,25 Although the melting temperature of Alq3 (441 °C)26 is much higher than those of TAPC (186 °C) and CBP (283 °C),27 all three materials have Ts/Tm < 0.2 in room temperature, that is, Ts ≈ 20 °C, and are expected to form a porous columnar structure according to the original SZM. However, as we mentioned earlier, the SZM is an empirical model observed for metals. If we consider the glasstransition temperatures (Tg) of Alq3 (175 °C),28 TAPC (82 °C),27 and CBP (62 °C) instead of the melting temperature,29 then only Alq3 has the Ts/Tg < 0.2 criteria and can satisfy the condition for forming a columnar porous structure in room temperature. Porous columnar TAPC, CBP, or any other organic thin films using OAD are attainable by lowering the temperature of the substrate. Fortunately, Alq3 is an electrontransporting material, and as we have shown above, the refractive index of the ETL has the largest impact on light extraction. For the OLEDs studied in this work, we used TAPC as the HTL. TAPC is chosen for its high triplet energy (T1 = 2.9 eV)30 that can effectively block exciton diffusion. TAPC also has one of the lowest refractive indices among the commonly used HTL materials (n = 1.68 at 525 nm). The refractive index of some of the common materials used in OLEDs is presented in Table 2. The dispersive refractive indices and extinction coefficients of all of the materials used in this study are presented in Figure S2. CBP is chosen as the host for its good charge balance when used in combination with our green phosphorescent emitter, Irppy2acac.31 Here, a thin layer of B3PYMPM is used as the hole blocking layer, and Alq3 is used as the ETL as mentioned before. We made two devices with the

Alq3 has a moderate isotropic refractive index of 1.75 at 525 nm. It has been previously shown that the refractive index of Alq3 can be tuned from 1.75 to 1.2 using OAD.9 In our experiment, our OAD-deposited Alq3 film at 80° showed a reduced refractive index of 1.45 at 525 nm as determined by variable angle spectroscopic ellipsometry (VASE). A crosssectional scanning electron microscopy (SEM) image of the sample used for ellipsometry measurement is shown in Figure S1. A 1 mm thick Si wafer with a 300 nm thermal oxide layer was used as a substrate. To measure the refractive index of the solid Alq3 film as well as the porous Alq3 film, first, a 200 nm thick solid Alq3 film was deposited at 0° on the substrate and then a 700 nm thick Alq3 film was deposited at 80° on top of the solid Alq3 film, following Szeto et al. approach.9 A general oscillator model based on Gaussian oscillators was used to determine the refractive index of the bottom solid Alq3 layer. A Bruggman EMA model based on the refractive index of the bottom solid Alq3 layer and void was then employed to determine the refractive index of the top porous Alq3 film.9 Figure 1A,B shows the dispersive optical constants of Alq3 when deposited at 0° (normally) and at 80°, respectively. Figure 1C shows the refractive index of Alq3 at 525 nm as a function of the angle of deposition. In OAD, organic molecules can form an initial solid wetting layer when deposited at an oblique angle.9,19−21 The formation of the wetting layer is due to the polar nature of organic molecules and polarity of the underlying surface and can be completely avoided by rendering the underlying surface apolar through surface passivation.20,21 In our OLED devices, Alq3 is directly deposited on top of 4,6-bis(3,5-di-3-pyridylphenyl)-2methylpyrimidine (B3PYMPM). Subsequently, to measure Alq3’s wettability on B3PYMPM, we deposited Alq3 at 80° on top of a solid B3PYMPM film and measured the thickness of the wetting layer using cross-sectional SEM and VASE. We found that Alq3 forms approximately a 20 nm thick wetting layer on top of B3PYMPM when deposited at 80°. The crosssectional SEM image is shown in Figure 1D. We also attempted OAD of CBP and TAPC. However, the films did not show any porosity when OAD was carried out at 9597

DOI: 10.1021/acsami.7b18514 ACS Appl. Mater. Interfaces 2018, 10, 9595−9601

Research Article

ACS Applied Materials & Interfaces Table 2. Refractive Indices of Some of the Commonly Used Materials in OLED Devices at 550 nm material

complete chemical name

common role

nob

nec

nd

refs

3TPYMB Alq3 AZOa B3PYMPM B4PYMPM BALq BCP BCzVBi Bphen C60 CBP CDBP CzPA DPVBi HAT-CN HfO ITO IZOa mCBP mCP mMTDATA MoO3 NBPhen NiO NPB P3HT PEDOT:PSSa PO-T2T sDPVBi T2T TAPC TcTa TmPyPB TPBi TrisPCZ ZnOa

tris(2,4,6-trimethyl-3-(pyridin-3-yl)phenyl)borane tris(8-hydroxy-quinolinato)aluminum aluminum zinc oxide 4,6-bis(3,5-di(pyridin-3-yl)phenyl)-2-methylpyrimidine 4,6-bis(3,5-di(pyridin-4-yl)phenyl)-2-methylpyrimidine bis(2-methyl-8-quinolinolate)-4-(phenylphenolato)aluminum 2,9-dimeth + B8:B29 + B8:B28yl-4,7-diphenyl-1,10-phenanthroline 4,4′-bis(9-ethyl-3-carbazovinylene)-1,1′-biphenyl 4,7-diphenyl-1,10-phenanthroline (C60-Ih)[5,6]fullerene 4,4′-bis(carbazol-9-yl)biphenyl 4,4′-bis(carbazol-9-yl)-2,2′-dimethylbiphenyl 9-[4-(10-phenylanthracen-9-yl)phenyl]carbazole 4,4′-Bis(2,2-diphenylethenyl)-1,1′-biphenyl dipyrazino[2,3-f:2′,3′-h]quinoxaline-2,3,6,7,10,11-hexacarbonitrile haffnium oxide indium tin oxide indium zinc oxide 3,3-di(9H-carbazol-9-yl)biphenyl 1,3-bis(carbazol-9-yl)benzene 4,4′,4″-tris(N-3-methylphenyl-N-phenyl-amino)triphenylamine molybdenum(VI) oxide 2,9-bis(naphthalen-2-yl)-4,7-diphenyl-1,10-phenanthroline nickel oxide N,N′-bis(naphthalen-1-yl)-N,N′-bis(phenyl)-benzidine poly(3-hexylthiophene-2,5-diyl) poly(3,4-ethylenedioxythiophene)-poly(styrenesulfonate) 2,4,6-tris[3-(diphenylphosphinyl)phenyl]-1,3,5-triazine 4,4′-bis(2,2′-diphenylvinyl)-1, 1′-spirobiphenyl 2,4,6-tris(biphenyl-3-yl)-1,3, 5-triazine di[4-(N,N-di-p-tolyl-amino)-phenyl]cyclohexane 4,4′,4″-tris(carbazol-9-yl)triphenylamine 1,3,5-tri[(3-pyridyl)-phen-3-yl]benzene 1,3,5-tris(1-phenyl-1H-benzimidazol-2-yl)benzene 9,9′-diphenyl-6-(9-phenyl-9H-carbazol-3-yl)-9H,9′H-3,3′-bicarbazole zinc oxide

ETL ETL anode ETL ETL ETL ETL emitter ETL ETL host host host host HILe dielectric anode anode host host HTL HIL ETL HIL HTL HIL HIL ETL host HTL ETL ETL ETL ETL HTL ETL

N/A N/A N/A 1.84 1.84 N/A N/A 1.99 N/A N/A 1.8 1.75 1.79 N/A 1.81 N/A N/A N/A N/A N/A 1.77 N/A 1.87 N/A N/A N/A 1.52 N/A N/A 1.67 N/A 1.81 1.77 1.75 1.89 N/A

N/A N/A N/A 1.59 1.54 N/A N/A 1.71 N/A N/A 1.74 1.71 1.76 N/A 1.76 N/A N/A N/A N/A N/A 1.71 N/A 1.69 N/A N/A N/A 1.61 N/A N/A 1.83 N/A 1.69 1.7 1.72 1.65 N/A

1.65 1.73 1.59 N/A N/A 1.69 1.71 N/A 1.73 2.23 N/A N/A N/A 1.78 N/A 2.04 2.10 2.12 1.74 1.75 N/A 1.99 N/A 2.79 1.87 1.89 N/A 1.71 1.78 N/A 1.68 N/A N/A N/A N/A 1.67

3 measured measured measured measured measured 38 measured measured 38 measured measured measured measured measured measured measured 38 4 4 measured measured 3 38 measured 38 measured 4 38 3 measured measured measured measured measured measured

Solution-processed films. bno stands for ordinary refractive index. cne stands for extraordinary refractive index. dn stands for isotropic refractive index. eHIL stands for hole injection layer. Dispersive refractive indices of all measured materials are shown in Figure S2.

a

Figure 2. Device characteristics of the OAD device and the control device.

same structures and layer thicknesses: an OAD device with the Alq3 layer deposited at 80° and a control device with the Alq3 layer deposited at 0°. The device structure for the both devices was ITO (100 nm)/TAPC (50 nm)/CBP−Irppy2acac 1:0.07 (20 nm)/B3PYMPM (10 nm)/Alq3 (10 nm)/Alq3−Cs2CO3 1:0.05 (50 nm)/Cs2CO3 (1 nm)/Al. We doped the last 50 nm of the Alq3 film with Cs2CO3 in both devices with the same concentration to increase the ETL conductivity.32 The OAD

device achieved a maximum EQE of 33.8%, which shows nearly 30% increase in efficiency compared to the control device with a maximum EQE of 26.4%. The current, voltage, and luminance characteristics and the EQE of the devices are presented in Figure 2. The EQE was calculated from the measured current efficiency and angular dependence of electroluminescence of the devices. The experimental and simulated angular-dependent 9598

DOI: 10.1021/acsami.7b18514 ACS Appl. Mater. Interfaces 2018, 10, 9595−9601

Research Article

ACS Applied Materials & Interfaces

Figure 3. (A) Power dissipation as a function of refractive index of the ETL and in-plane wavevector, at λ = 525 nm. Region 1 presents “air mode”, region 2 presents “substrate mode”, region 3 presents “WG mode”, and region 4 presents “evanescent mode”. (B) Relative mode contribution calculated from the power dissipation. The star marks indicate the experimental maximum EQE achieved by the OAD device and the control device. The calculations are based on the structure of the OAD device and the control device. The refractive index of the last 40 nm of the ETL (Alq3) layer was varied from 1.4 to 2.

Figure 4. (A) Experimental electroluminescence spectrum obtained from the OAD device and the control device. (B) Simulated electroluminescence spectrum of the OAD device and the control device.

We calculated the power dissipation versus the in-plane wavevector (kin) as a function of refractive index of the ETL at λ = 525 nm. The calculation was done using the actual refractive indices for all the layers except for the last 40 nm of the ETL. The refractive index of the last 40 nm of the ETL was varied from 1.4 to 2. The results are shown in Figure 3A. In region 4 of Figure 3A corresponding to the evanescent region of the power dissipation plot, we observe a major loss corresponding to the SPP peak. The SPP peak is given by

electroluminescence from the OAD and the control devices are shown in Figure S3. To better understand the device optics, we did optical simulation using the SETFOS simulation software.33 In our model, excitons are regarded as damped forced dipoles, driven by a reflected electromagnetic field from the metallic cathode.2,33,34 On the basis of this model, we can calculate the power dissipation of the dipoles as a function of the inplane wavevector of the generated electromagnetic waves. Depending on the magnitude of the in-plane wavevector (kin), radiation can be categorized in four different regions.34 In the first region, light with 0 < kin < 2πnair/λ will escape the glass substrate and “leak” to air. Here, 2πnair/λ is equal to 0.0119 nm−1 at λ = 525 nm. The amount of radiation escaping to the air, that is, air mode, can then be calculated by integrating the radiation from 0 < kin < 0.0119 nm−1. When kin becomes larger than 2πnair/λ, light will be totally internally reflected back into the glass substrate and for 2πnair/λ < kin < 2πnglass/λ, light is trapped in the glass substrate, this second region corresponds to the substrate mode. In the third region, where kin becomes larger than 2πnglass/λ, light will be totally internally reflected from the glass/ITO boundary and for 2πnglass/λ ≤ kin < 2πnEML/λ, radiation is coupled to the ITO-organic WG mode. In the fourth region, where kin becomes larger than the total magnitude of the wavevector in the organic thin film, that is, kin > 2πnEML/λ, the wavevector acquires an imaginary part and becomes evanescent.35 Therefore, light with kin > 2πnEML/λ is regarded as the evanescent mode.

the SPP dispersion equation, k in =

2π λ

(

εcathode·εETL εcathode + εETL

1/2

)

,2 where

εcathode is the dielectric constant of the metallic cathode and εETL is the dielectric constant of the ETL. When the refractive index of the ETL is reduced, the radiation is shifted from the evanescent region toward the outcoupled region following the SSP dispersion equation, leading to a reduction of the evanescent loss and an increased light outcoupling efficiency.3,4 Another peak in the evanescent region is the TEWG peak, with less intensity compared to the SPP peak. The TEWG peak appears in the evanescent region because it centers at the ITO layer and the refractive index of ITO is larger than the organic EML. The effect of the ETL refractive index is minimal on TEWG because TEWG is centered at the ITO layer. However, reducing the refractive index of ITO can weaken or eliminate this peak altogether. It is worth noting that the refractive index of ITO can also be reduced using OAD.6 Figure S4 shows the TE-polarized dipole power dissipation as a function of refractive index of ITO with the TEWG peak merging into 9599

DOI: 10.1021/acsami.7b18514 ACS Appl. Mater. Interfaces 2018, 10, 9595−9601

ACS Applied Materials & Interfaces

■

the substrate mode when the refractive index of ITO is reduced to 1.5. We determined the relative contributions of different modes based on the calculated power dissipation in Figure 3A. The different mode contributions are shown in Figure 3B as a function of the refractive index of the ETL. As we explained above, reducing the refractive index of the ETL increases the EQE because of reduced coupling to SPPs. The star symbols in Figure 3B are the experimental maximum EQEs achieved by the OAD device and control device, indicating that lowering the ETL refractive index from 1.75 to 1.45 increases the outcoupling efficiency by 30%, which is in excellent agreement with our experimental results. Given the optical constants and thicknesses of all the constituent layers and knowing the photoluminescence spectrum of the emitter, we can simulate the electroluminescent spectrum of an OLED device using transfer matrix formalism.33,36 To further verify the effect of the low-refractive index ETL, we simulated the electroluminescent spectra of the OAD device and control device. In our simulation, we used the Irppy2acac photoluminescence spectrum, and the actual measured refractive indices and thicknesses of the constituent layers in the OAD device and control device as the input parameters. Both the experimental and simulated electroluminescent spectra are shown in Figure 4. The experimental electroluminescent spectra show a small blue shift on the right shoulder of the spectrum, going from the control device to the OAD device. The same shift in the simulated spectra can be seen when the refractive index of the last 40 nm of the Alq3 layer is lowered from 1.75 to 1.45, further verifying the effect of the low-refractive index ETL on the device performance.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.7b18514. VASE modeling and measurement details; Dispersive optical constants of measured materials shown in Table 2; and TE-polarized power dissipation as a function of refractive index of ITO (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Franky So: 0000-0002-8310-677X Notes

The authors declare no competing financial interest.

■ ■

ACKNOWLEDGMENTS The authors acknowledge the support of CYNORA GmbH, Germany for this work. REFERENCES

(1) Yeh, P. Optical Waves in Layered Media, 2nd ed.; WileyInterscience, 2005. (2) Barnes, W. L. Fluorescence near Interfaces: The Role of Photonic Mode Density. J. Mod. Opt. 1998, 45, 661−699. (3) Salehi, A.; Ho, S.; Chen, Y.; Peng, C.; Yersin, H.; So, F. Highly Efficient Organic Light-Emitting Diode Using A Low Refractive Index Electron Transport Layer. Adv. Opt. Mater. 2017, 5, 1700197. (4) Shin, H.; Lee, J.-H.; Moon, C.-K.; Huh, J.-S.; Sim, B.; Kim, J.-J. Sky-Blue Phosphorescent OLEDs with 34.1% External Quantum Efficiency Using a Low Refractive Index Electron Transporting Layer. Adv. Mater. 2016, 28, 4920−4925. (5) Xi, J.-Q.; Schubert, M. F.; Kim, J. K.; Schubert, E. F.; Chen, M.; Lin, S.-Y.; Liu, W.; Smart, J. A. Optical Thin-Film materials with Low Refractive Index for Broadband Elimination of Fresnel Reflection. Nat. Photonics 2007, 1, 176−179. (6) Kim, J. K.; Chhajed, S.; Schubert, M. F.; Schubert, E. F.; Fischer, A. J.; Crawford, M. H.; Cho, J.; Kim, H.; Sone, C. Light-Extraction Enhancement of GaInN Light-Emitting Diodes by Graded-RefractiveIndex Indium Tin Oxide Anti-Reflection Contact. Adv. Mater. 2008, 20, 801−804. (7) Schubert, M. F.; Poxson, D. J.; Mont, F. W.; Kim, J. K.; Schubert, E. F. Performance of Antireflection Coatings Consisting of Multiple Discrete Layers and Comparison with Continuously Graded Antireflection Coatings. Appl. Phys. Express 2010, 3, 082502. (8) Barranco, A.; Borras, A.; Gonzalez-Elipe, A. R.; Palmero, A. Perspectives on Oblique Angle Deposition of Thin Films: From Fundamentals to Devices. Prog. Mater. Sci. 2016, 76, 59−153. (9) Szeto, B.; Hrudey, P. C. P.; Gospodyn, J.; Sit, J. C.; Brett, M. J. Obliquely Deposited tris(8-Hydroxyquinoline) Aluminium (Alq3) Biaxial Thin Films with Negative in-Plane Birefringence. J. Opt. A: Pure Appl. Opt. 2007, 9, 457−462. (10) Zhong, Y.; Shin, Y. C.; Kim, C. M.; Lee, B. G.; Kim, E. H.; Park, Y. J.; Sobahan, K. M. A.; Hwangbo, C. K.; Lee, Y. P.; Kim, T. J. Optical and Electrical Properties of Indium Tin Oxide Thin Films with Tilted and Spiral Microstructures Prepared by Oblique Angle Deposition. J. Mater. Res. 2008, 23, 2500−2505. (11) Slootsky, M.; Forrest, S. R. Enhancing Waveguided Light Extraction in Organic LEDs Using an Ultra-Low-Index Grid. Opt. Lett. 2010, 35, 1052−1054. (12) Saxena, K.; Mehta, D. S.; Srivastava, R.; Kamalasanan, M. N. Effect of Oblique Angle Deposition of α-Naphthylphenylbiphenyl Diamine on the Performance of Organic Light-Emitting Diodes. J. Phys. D: Appl. Phys. 2008, 41, 015102.

■

CONCLUSIONS In this work, we first analyze and quantify the effect of the refractive index of each layer in an OLED stack on light extraction. We show that among all constituent layers, the refractive index of the ETL has the largest impact on light extraction. OAD provides a way of reducing a film’s refractive index through controlled pore inclusion. We show that the refractive index of Alq3 can be reduced to 1.45 when deposited at 80° using OAD. We then employ the low-refractive index Alq3 layer in an OLED stack and achieve a 30% increase in efficiency, compared to the control device. To verify our experimental data, we did optical simulation to study the effect of a low-refractive index ETL on the evanescent loss and showed that the experimental data are consistent with our simulation results.

■

Research Article

EXPERIMENTAL SECTION

Substrates were cleaned using a ultrasonic bath, in acetone and isopropanol solutions. Vacuum pressure was maintained under 2 × 10−6 Torr during evaporation. A Woollam M2000 variable angle spectroscopic ellipsometer was used for measuring the optical constants. WVASE software was used for ellipsometry modeling. All of the organic materials were purchased from Lumtec incorporation. A field emission scanning electron microscope, FEI Verios 460L, was used to obtain SEM images. The EQE was measured from current efficiency, assuming a Lambertian distribution of emission. FLUXiM SETFOS 4.4 software package was used for optical calculations and simulations. Refractive index of all constituent layers was imported from real measured values using VASE. The Irppy2acac photoluminescence spectrum was used in simulations.37 An internal quantum efficiency of 95% and a horizontal orientation of 73% were assumed for the emitter, as reported elsewhere.31 9600

DOI: 10.1021/acsami.7b18514 ACS Appl. Mater. Interfaces 2018, 10, 9595−9601

Research Article

ACS Applied Materials & Interfaces (13) Kumar, A.; Srivastava, R.; Kamalasanan, M. N.; Mehta, D. S. Enhancement of Light Extraction Efficiency of Organic Light Emitting Diodes Using Nanostructured Indium Tin Oxide. Opt. Lett. 2012, 37, 575−577. (14) Callens, M. K.; Yokoyama, D.; Neyts, K. Anisotropic Materials in OLEDs for High Outcoupling Efficiency. Opt. Express 2015, 23, 21128−21148. (15) Lu, C.-Y.; Jiao, M.; Lee, W.-K.; Chen, C.-Y.; Tsai, W.-L.; Lin, C.Y.; Wu, C.-C. Achieving Above 60% External Quantum Efficiency in Organic Light-Emitting Devices Using ITO-Free Low-Index Transparent Electrode and Emitters with Preferential Horizontal Emitting Dipoles. Adv. Funct. Mater. 2016, 26, 3250−3258. (16) Cai, M.; Ye, Z.; Xiao, T.; Liu, R.; Chen, Y.; Mayer, R. W.; Biswas, R.; Ho, K.-M.; Shinar, R.; Shinar, J. Extremely Efficient Indium-Tin-Oxide-Free Green Phosphorescent Organic Light-Emitting Diodes. Adv. Mater. 2012, 24, 4337−4342. (17) Koh, T.-W.; Choi, J.-M.; Lee, S.; Yoo, S. Optical Outcoupling Enhancement in Organic Light-Emitting Diodes: Highly Conductive Polymer as a Low-Index Layer on Microstructured ITO Electrodes. Adv. Mater. 2010, 22, 1849−1853. (18) Lee, J.; Han, T.-H.; Park, M.-H.; Jung, D. Y.; Seo, J.; Seo, H.-K.; Cho, H.; Kim, E.; Chung, J.; Choi, S.-Y.; Kim, T.-S. Synergetic Electrode Architecture for Efficient Graphene-Based Flexible Organic Light-Emitting Diodes. Nat. Commun. 2016, 7, 11791. (19) Hrudey, P. C. P.; Westra, K. L.; Brett, M. J. Highly Ordered Organic Alq3 Chiral Luminescent Thin Films Fabricated by GlancingAngle Deposition. Adv. Mater. 2006, 18, 224−228. (20) Tsoi, S.; Szeto, B.; Fleischauer, M. D.; Veinot, J. G. C.; Brett, M. J. Control of Alq3 Wetting Layer Thickness via Substrate Surface Functionalization. Langmuir 2007, 23, 6498−6500. (21) Brinkmann, M.; Biscarini, F.; Taliani, C.; Aiello, I.; Ghedini, M. Growth of Mesoscopic Correlated Droplet Patterns by High-Vacuum Sublimation. Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 61, R16339−R16342. (22) Thornton, J. A. The Microstructure of Sputter-deposited Coatings. J. Vac. Sci. Technol., A 1986, 4, 3059−3065. (23) Müller, K.-H. Dependence of Thin-Film Microstructure on Deposition Rate by Means of a Computer Simulation. J. Appl. Phys. 1985, 58, 2573−2576. (24) Mirica, E.; Kowach, G.; Du, H. Modified Structure Zone Model to Describe the Morphological Evolution of ZnO Thin Films Deposited by Reactive Sputtering. Cryst. Growth Des. 2004, 4, 157− 159. (25) Anders, A. A Structure Zone Diagram Including Plasma-Based Deposition and Ion Etching. Thin Solid Films 2010, 518, 4087−4090. (26) Wang, M. H.; Sawada, Y.; Saito, K.; Horie, S.; Uchida, T.; Ohtsuka, M.; Seki, S.; Kobayashi, S.; Arii, T.; Kishi, A.; Takahashi, T. Thermal Change of Alq3, tris(8-Hydroxyquinolinato) aluminum(III) Studied by TG and XRD-DSC. J. Therm. Anal. Calorim. 2007, 89, 363−366. (27) Vamvounis, G.; Aziz, H.; Hu, N.-X.; Popovic, Z. D. Temperature Dependence of Operational Stability of Organic Light Emitting Diodes Based on Mixed Emitter Layers. Synth. Met. 2004, 143, 69−73. (28) Naito, K.; Miura, A. Molecular Design for Nonpolymeric Organic Dye Glasses with Thermal Stability: Relations between Thermodynamic Parameters and Amorphous Properties. J. Phys. Chem. 1993, 97, 6240−6248. (29) Tsai, M.-H.; Hong, Y.-H.; Chang, C.-H.; Su, H.-C.; Wu, C.-C.; Matoliukstyte, A.; Simokaitiene, J.; Grigalevicius, S.; Grazulevicius, J. V.; Hsu, C.-P. 3-(9-carbazolyI)carbazoles and 3,6-Di(9-Carbazolyl)carbazoles as Effective Host Materials for Efficient Blue Organic Electrophosphorescence. Adv. Mater. 2007, 19, 862−866. (30) Lee, J.; Chopra, N.; Eom, S.-H.; Zheng, Y.; Xue, J.; So, F.; Shi, J. Effects of Triplet Energies and Transporting Properties of Carrier Transporting Materials on Blue Phosphorescent Organic Light Emitting Devices. Appl. Phys. Lett. 2008, 93, 123306. (31) Sim, B.; Moon, C.-K.; Kim, K.-H.; Kim, J.-J. Quantitative Analysis of the Efficiency of OLEDs. ACS Appl. Mater. Interfaces 2016, 8, 33010.

(32) Tyagi, P.; Srivastava, R.; Kumar, A.; Tuli, S.; Kamalasanan, M. N. Effect of Doping of Cesium Carbonate on Electron Transport in Tris(8-Hydroxyquinolinato) Aluminum. Org. Electron. 2013, 14, 1391−1395. (33) FlUXiM. SETFOS 4.4 Manual; Fluxim AG: Feusisberg, Switzerland, 2016. (34) Nowy, S.; Krummacher, B. C.; Frischeisen, J.; Reinke, N. A.; Brütting, W. Light Extraction and Optical Loss Mechanisms in Organic Light-Emitting Diodes: Influence of the Emitter Quantum Efficiency. J. Appl. Phys. 2008, 104, 123109. (35) Penninck, L.; Mladenowski, S.; Neyts, K. The Effects of Planar Metallic Interfaces on the Radiation of Nearby Electrical Dipoles. J. Opt. 2010, 12, 075001. (36) Brütting, W.; Frischeisen, J.; Schmidt, T. D.; Scholz, B. J.; Mayr, C. Device Efficiency of Organic Light-Emitting Diodes: Progress by Improved Light Outcoupling. Phys. Status Solidi A 2013, 210, 44−65. (37) Kim, K.-H.; Moon, C.-K.; Lee, J.-H.; Kim, S.-Y.; Kim, J.-J. Highly Efficient Organic Light-Emitting Diodes with Phosphorescent Emitters Having High Quantum Yield and Horizontal Orientation of Transition Dipole Moments. Adv. Mater. 2014, 26, 3844−3847. (38) FLUXiM. SETFOS Database, 2017.

9601

DOI: 10.1021/acsami.7b18514 ACS Appl. Mater. Interfaces 2018, 10, 9595−9601