Mechanistic Understanding of Improved ... - ACS Publications

*E-mail: [email protected] (Y.P.)., *E-mail: [email protected] (J.-S.L.)., *E-mail: [email protected] (N.S.C.). Cite this:ACS Appl. Mater. Interfac...
0 downloads 0 Views 984KB Size
Subscriber access provided by University of South Dakota

Organic Electronic Devices

Mechanistic understanding of improved performance of graphene cathode inverted organic light emitting diodes by photoemission and impedance spectroscopy Jaehyun Moon, Hyunsu Cho, Min-Jae Maeng, Kwangmin Choi, ##ng Thành Nguyen, Jun-Han Han, Jin-Wook Shin, Byoung-Hwa Kwon, Jonghee Lee, Seungmin Cho, Jeong-Ik Lee, Yongsup Park, Jong-Sook Lee, and Nam Sung Cho ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.8b07751 • Publication Date (Web): 16 Jul 2018 Downloaded from http://pubs.acs.org on July 17, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Mechanistic understanding of improved performance of graphene cathode inverted organic light emitting diodes by photoemission and impedance spectroscopy Jaehyun Moon§, Hyunsu Cho§, Min-Jae Maeng†, Kwangmin Choi†, Đăng Thành Nguyen║, Jun-Han Han§, Jin-Wook Shin§, Byoung-Hwa Kwon§, Jonghee Lee§, Seungmin Cho‡ , Jeong Ik Lee§, Yongsup Park*†, Jong-Sook Lee*║, Nam Sung Cho*§ Yongsup Park: [email protected] Jong-Sook Lee: [email protected] Nam Sung Cho: [email protected]

§

Reality Device Research Section, Electronics and Telecommunications Research Institute

(ETRI), Daejeon 34129, Republic of Korea †

Department of Physics and Research Institute for Basic Sciences, Kyung Hee University,

Seoul 02447, Republic of Korea ║

School of Materials Science and Engineering, Chonnam National University, Gwangju

61186, Republic of Korea ‡

Hanwha Techwin R&D Center, Seongnam 13488, Republic of Korea

Keywords: Graphene, inverted organic light emitting diode, energy alignment diagram; impedance spectroscopy

Abstract Modification of multilayer graphene films was investigated for a cathode of organic light emitting diodes (OLEDs). By doping the graphene/electron transport layer (ETL) interface with Li, the driving voltage of the OLED was reduced dramatically from 24.5 V to 3.2 V at the luminance of 1000 cd/m2. The external quantum efficiency was also enhanced from 3.4 % to 12.9 %. Surface analyses showed that the Li doping significantly lowers the lowest unoccupied molecular orbital (LUMO) level of the ETL, thereby reducing the electron injection barrier and facilitating electron injection from the cathode. Impedance spectroscopy 1

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

analyses performed on electron-only-devices (EODs) revealed the existence of distributed trap states with well-defined activation energy, which is successfully described by the Havriliak-Negami capacitance functions and the temperature-independent frequency dispersion parameters. In particular, the graphene EOD showed a unique high frequency feature as compared to the ITO one, which could be explained by an additional parallel capacitance element.

1. Introduction Graphene is a two dimensional carbon material in which carbon atoms are arranged in a planar hexagonal lattice. Multilayered polycrystalline graphene films grown by a chemical vapor deposition (CVD) method possess high optical transparency (T~ 83 %) in the visible range and acceptable sheet resistance (~ 150 Ω/Sq.), which is suitable for optoelectronic applications.1,2 Presumably, since graphene’s work function (~4.6 eV) is similar to that of indium tin oxide (ITO), it has been predominantly investigated as a transparent anode in organic light emitting diodes (OLEDs).3-8 In this work, we investigate the technical possibility of using graphene as a cathode material. Our motivation stemmed from the burgeoning active matrix (AM)-OLED display applications in TV, mobile display and virtual reality gears. Moreover, because the reflectance of graphene is low compared to that of the metal, cavity effect is expected reduce by using graphene cathodes, opening a way to suppress detrimental spectral distortion of light emission.9 Due to their highly uniform performance across the thin film transistors (TFT) backplane, n-type oxide TFTs are widely chosen in large area OLED displays. For this reason inverted type OLEDs are common in AM-OLED displays.10 Thus, a cathode, the electron recipient OLED electrode, must possess suitable surface property to transport electrons effectively to the adjacent electron transport layer (ETL).11 The high work function of graphene (~ 4.6 eV) and the resulting energy level misalignment to common electron transport layer (ETL) makes graphene a bad choice for a cathode material, which would require prohibitively high voltage for electron injection and meaningful luminance.12,13 Therefore, to realize a graphene cathode OLED, there is a technical need to reduce the electron injection barrier. In inverted OLEDs with ITO cathode, thin metal oxides have been inserted between to resolve these difficulties.14 Application of similar method to graphene cathode, however, does not seem to be easy.15 2

ACS Paragon Plus Environment

Page 2 of 29

Page 3 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

In this work, as a way to attain high performance in OLED with graphene cathode, we have modified the electron transport layer (ETL)/graphene cathode interface by Li doping the entire ETL.16,17 We could significantly lower the operating voltage of the OLED and achieve high efficiency. Because the use of interfacial modification brought forth increase in the external quantum efficiency of approximately 3.8 times. In order to elucidate the interfacial characteristics quantitatively, we conducted in-situ ultraviolet photoelectron spectroscopy (UPS) and impedance spectroscopy (IS) analyses. Quantitative UPS analyses reveal that the electron injection barrier at the ETL/graphene interface was lowered from 1.49 eV to 0.24 eV by Li doping to ETL. 2. Experiments In this work, we used a by a chemical vapor deposition (CVD) method on Cu foils to grow graphene. Methane (CH4) was used as the carbon source. Monolayer graphene was sequentially transferred by a dry method to form a four-layered graphene film. Each monolayer graphene film was treated by a benzimidazole solution for delamination and doping.18 Graphene was isolated from the Cu foil using a H2O diluted mixture of sulfuric acid (H2SO4) and peroxide (H2O2). In order to facilitate the isolation and achieve doping effect benzimidazole (C7H6N2) was added to the etchant. The concentration of benzimidazole was 0.06 M. Our four-layered graphene film has been treated layer-by-layer. The direct transmittance of our film was 83 % at a wavelength of 550 nm. The sheet resistance of fourlayered graphene film was 65 Ω/Sq.. The surface roughness (Ra) of our graphene film was 0.75 nm. The atomic force microscope (AFM, XE-100, Park System) image of our graphene film can be found in the Supporting Information (Figure S1). Inverted phosphorescent green OLEDs were fabricated using a vacuum deposition method. Our OLEDs have a stack structure of Graphene film/BmPyPB (30 nm, undoped or 5 % Li doped)/BmPyPB (20 nm)/ 26DCzPPy: Ir(ppy)3 (7%) 10 nm/ TcTa: Ir(ppy)3 7%/ TAPC(40 nm) HAT-CN(10 nm)/ TAPC(40 nm)/ HAT-CN(10 nm)/ TAPC(40 nm) HAT -CN(10 nm)/Al (100 nm). The full chemical names of organics and their functions are summarized in Table 1. The ETL material in our work was BmPyPB, which was originally synthesized as an electron transport material to improve the carrier balance in Ir-complex phosphorescent blue dopants containing OLEDs.19 The highest occupied molecular orbital (HOMO) and lowest 3

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

unoccupied molecular orbital (LUMO) energies of BmPyPB are 6.7 eV and 2.7 eV, respectively. Furthermore, an excellent mobility and hole-blocking capacity makes it widely applicable in OLED as an ETL material.20 To modify the interface between the graphene cathode and the ETL, we have doped the ETL with Li. Li metal (SAES Getters, Italy) was thermally co-evaporated at approximately 900°C to achieve doping into the ETL. In order to spatially confine the excitons the emission layer was designed to consist of two layers.20 Our hole transport layer (HTL) had an alternating structure of HAT-CN and TAPC, which has been proven to be useful in enhancing the hole transport and stabilizing the device operation.21,22 For comparison, we fabricated OLEDs with ITO cathode. ITO thickness was 70 nm. All organics and Al anode were vacuum deposited below a pressure of 6.66 × 10-5 Pa. The current density-voltage (J-V) and luminescence-voltage (L-V) characteristics were measured with a current/voltage source/measure unit (Keithley 238) and a spectroradiometer (CS-2000, Minolta), respectively. In conjunction with the interfacial characteristics of ETL/graphene interface, we have performed in-situ UPS analyses. ETL/ITO interfaces were also examined for comparison. The measured UPS spectra were compiled to construct quantitative energy diagrams.23 The UPS measurements were performed using a He I (21.22 eV) ultraviolet light source and a modified KRATOS AXIS-165 photoelectron spectroscopy system. The energy resolution of the measured spectra is approximately 0.1 eV and a sample bias of -20 V was applied. In-situ analyses and thermal deposition were carried out in separate but connected ultra-high vacuum chambers, which enabled sample transfers without exposing them to the ambient condition.24 The base pressure of the deposition chamber was 6.6 × 10-9 Pa, while that of the analysis chamber was 2.6 × 10-9 Pa. We measured HOMO by UPS and estimated the LUMO from the optical band gap in the literature. In addition, we fabricated electron only devices (EODs) to perform electrical analyses. We have adopted a simplified EOD structure of Graphene/Li-doped BmPyPB (30nm) /BmPyPB (70 nm)/Al(100 nm) and ITO (70 nm) / Li-doped BmPyPB (30nm) /BmPyPB (70 nm)/Al(100 nm). DC current (I) – voltage (V) and impedance were measured using a source-meter (Keithley 2400) and an impedance analyzer (HP4284A, Agilent), respectively. The samples were loaded in a cryostat (CS-400/202, Janis) to control the temperature in 4

ACS Paragon Plus Environment

Page 4 of 29

Page 5 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

vacuum. The impedance spectra from 1 MHz to 20 Hz with 100 mV AC and without DC bias were measured first on cooling from 300 K to 240 K and heating back to 300 K at a rate of 1 K/min and additionally at 330 K. I-V curves were obtained at 270 K, 300 K, and 330 K within a voltage range of -5V~ 5V at scan rate 10 mV/s. 3. Results and discussion Figure 1 shows the device characteristics of the fabricated OLEDs. The OLEDs with the ITO cathode show slightly better performance than those with graphene cathode. Commonly, without Li doping the required applied voltage to reach identical levels of Js and Ls is much higher than the cases with Li doping. For the undoped cases, noticeable J (> 1 mA/cm2) is observed at applied voltages higher than 20 V. In the case of graphene cathode with undoped BmPyPB, the voltage required to achieve L of 1000 cd/m2 is high as 24.5 V. Such voltage is prohibitively high to operate OLEDs in a stable manner. On the other hand, the same level of luminance is achieved with only 3.2 V by doping the ETL. These results strongly indicate the usefulness of Li doping for efficient electron injection from the cathode into ETL. Benefitted from the Li doping, the external quantum efficiencies (EQEs) and power efficacies (PE) show significant enhancements. Without Li doping, the EQE and PE of graphene cathode OLEDs do not exceed 3.5 % and 1.5 lm/W. With Li doping into ETL, the corresponding values become much bigger. The highest EQE and PE were achieved at a L level of 40 cd/m2 as 13.8 % and 55 lm/W, respectively. At L of 1000 cd/m2 the EQE and PE were 13 % and 41 lm/W, respectively. Moreover, Li doping does not alter the electroluminescence (EL) spectra characteristics. All EL spectral lines almost superimpose and no shift in the main peak (λ= 517 nm) is observed (see Fig 1. (d)). These results show us that Li doping only affects the electrical characteristic but not the optical ones. Therefore, we conclude that the performance enhancement is due to the improved electrical characteristics. Because the EL spectra are preserved, there is no need to change the OLED structure. Before proceed to the next part, it is worthwhile to add comments on the optical issue of graphene OLED. Due to the thinness of the organic layers, internal interference or microcavity can influence the efficiency of OLEDs. The efficiency can be enhanced by adjusting the total thickness of organics to be an integer multiple of the characteristic wavelength of the emitter.6 To induce microcavity enhancement effect, reflectance of the organic/electrode strongly matters. If the reflectance is 5

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

low, it is difficult to induce microcavity effect. The reflectance of graphene/organic interface is fairly low. Thus optical method is fairly limited in graphene electrode bearing OLEDs. method and strategies for improving the electrical transport emerge as an important topic. In order to investigate the cathode/ETL interface quantitatively, we performed in-situ UPS analyses. Figure 2 shows the UPS spectra of HOMO region with increasing thickness of BmPyPB. The BmPyPB was directly deposited on graphene or ITO surface by a thermal evaporation method. Li doping was achieved by co-deposition with the ETL layer. The concentration of Li doping was approximately 5 %. As seen in Figs. 2 (a) and (c), when pristine BmPyPB was deposited, the HOMO onset clearly appeared at 2.70 eV at a thickness of 4.0 nm for both graphene and ITO substrates. Below this thickness, the onsets could be observed between 2.5 ~ 2.6 eV but they are not very clear. The onset positions did not significantly change up to 8.0 nm but increased to ~ 2.8 eV at 15.0 nm. For both graphene and ITO substrates the effect of Li doping was quite dramatic that the HOMO onsets moved to a higher binding energy by more than 1.0 eV as seen in Figs. 2(b) and (d). For Li-doped BmPyPB on both graphene and ITO, the onsets were observed around 3.80 eV at 4.0 nm. Although they changed slightly with increasing thickness, it remained between 3.7 ~ 3.9 eV up to the thickness of 15.0 nm. In addition, for Li-doped cases, a new peak at ~ 2.5 eV is clearly observable, which is absent in pristine BmPyPB layers. The new states are presumably created by the donation of charges from Li dopants to the ETL molecules.16, 25 Therefore, it does not affect the electron injection properties at the cathode interface. The difference of about 1.00 ~ 1.10 eV in HOMO onsets for both graphene and ITO substrates will be directly translated into the locations of the LUMO, which should reduce the electron injection barrier (EIB) by the same amount. The UPS spectra of high binding energy regions or secondary electron cut-off (SECO) regions are shown in Supporting Information (Figure S2). Figure 3 (a) and (b) shows the evolution of energy level positions derived from the HOMO onset positions in Fig. 2 and the measured values of work functions. The LUMO positions were deduced from the optical band gap of 4.05 eV for BmPyPB.26 These figures clearly show that the Li doping changes the work functions as much as 2.6 eV, which is much greater than the shift of HOMO onset positions, implying the formation of interface dipoles. This trend is more apparent in graphene than in ITO. Since the Fermi level (EF) is the reference, 6

ACS Paragon Plus Environment

Page 6 of 29

Page 7 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

the LUMO levels shown in the figure are the EIBs. But one should be cautioned that while the HOMO was measured directly by UPS, the LUMO position was estimated from the optical band gap as described in the experimental section. Therefore, these values underestimate the actual EIBs for the carrier injection at the cathode. The difference would be exciton binding energies. For more accurate estimation of EIBs, the transport band gap of the BmPyPB should be measured, which can be accomplished by direct LUMO onset position measurement by inverse photoemission spectroscopy, which has not been done for BmPyPB.27 In Figs. 3(a) and (b), we show the quantitative energy levels alignment (ELA) diagram at the ETL thickness of 2.0 nm for clear comparison between different cases. By Li doping, the EIB at the ETL/graphene interface was lowered from 1.35 eV to 0.23 eV. In the case of ETL/ITO, the EIB was lowered from 1.40 eV to 0.44 eV. The absolute values of the EIB among different cases vary depending on the thickness but the fact that Li doping dramatically lowers the EIBs is clear for both graphene and ITO cathodes. The positions of the vacuum levels or equivalently work functions are slightly different when the undoped ETL was deposited on different surfaces. Their values were 4.19 eV and 4.26 eV at the thickness of 2.0 nm for graphene and ITO, respectively, as seen in Figs. 3(a) and (b). The electron affinity (EA = Evac – |LUMO|) and ionization potential (IE = Evac + |HOMO|) also show similar differences, where Evac refers to the vacuum energy level. The EA values were 2.84 eV and 2.86 eV for graphene and ITO, respectively. The corresponding IE values were 6.89 eV and 6.91 eV. However, these values change dramatically upon Li-doping. This indicates that the interfacial dipole formation upon doping is a surface dependent characteristic. Thus, the actual ELA of doped surface cannot easily be deduced from the molecular parameters of the organic materials alone. In short, our UPS analysis results clearly indicate that the Li doping into BmPyPB dramatically lowers the EIBs, which enables the use of graphene as a transparent cathode. These results are in accordance with the J-V characteristics of Fig. 1(a). Having quantitatively probed the energy level alignments of cathode/ETL interfaces, we turn our attention to the electrical properties. The difference between the Li doped ETL/graphene and Li doped ETL/ITO cathodes were investigated using EODs which mostly exhibited capacitive behavior at zero DC bias. Fig. 4 (a) shows the I-V characteristics of the ITO(-)/Lidoped ETL/ETL/Al(+) sample. Three samples were examined and very similar electrical 7

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 29

behaviors were observed. At applied voltages of -2 V, -1.5 V, and -1 V at 270 K, 300 K, and 330 K, the current drastically increases, corresponding to the forward bias for the OLED emitting light shown above. The voltage value is close to the turn-on or built-in voltage ()

for OLED emitting. However, the EOD exhibits low-resistance upon reverse bias with the transition voltage () around 2 V at 270 K and 1 V at 300 K and 330 K. As previously suggested with a difference of ( − ), interface charge (QIF) is given as the following.

28-

30

 =



( −  )

(1)

With values of  ≈ 3.5,  = 4mm , and  = 100nm one obtains QIF of a few nano

Coulombs. This may be the interface dipole at the graphene (or ITO) and organic interface

evidenced by the UPS. The corresponding capacitance-voltage characteristics are shown in Fig. 4(b). Characteristic voltage dependence matching the transitions I-V characteristics around  and  at low frequencies around 1 kHz is similar to the reported for OLEDs

for 200 Hz.31 The large frequency dispersion in the leaky reverse bias of 2.5 V, which was not

observed in Ref. 31, is due to the electrode polarization. Strong decrease in capacitance with increasing frequency above 100 kHz was observed (Fig. 4(b)) which is due to the series spread resistance. It should be noted that capacitance at high frequencies around 100 kHz decreases at a bias larger than V .

Fig. 5 shows the Mott-Schottky analysis of the capacitance at 100kHz. Although the capacitance magnitude around 1 nF at 100 kHz is close to the geometric capacitance for the organic layer thickness of 100 nm, the variation upon reverse bias above  follows the Mott-Schottky relation for the voltage range of 1 ~ 3 V in a satisfactory way (Fig. 5).



!"

 =  (#$#

%& )

(2)

Flat-band potential () of -6 V and dopant concentration of 8.35×1015 cm-3 were estimated for the organic semiconductor. This  is larger than  around 1.5 V. It may be due to the inhomogeneity in the organic layer composed of the 30nm doped and 70nm undoped part. 8

ACS Paragon Plus Environment

Page 9 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Note that without Li-doping turn-on voltages for achieving L of 1000 cd/m2 are as high as 24 V vs. 3.2 V for Li-doped OLEDs (Fig. 1(a)). The top-left graphs of Fig. 6(a) and (b) show the capacitance-frequency plots or Bode plots for ITO/Li-doped ETL/ETL/Al and graphene/Li-doped ETL/ ETL /Al, as a function of temperature, respectively. The magnitude of temperature-dependent capacitance relaxations is measured as 0.6~0.7 nF with a common base value of 1~1.6 nF. The latter fits with the geometric capacitance of the organic layers. The low frequency capacitance relaxations of two electrodes are almost similar. Similar capacitance variation is also reported for the full OLEDs.28-31 Capacitance-dominated impedance responses are also shown in the complex plane impedance plots of the respective top right graphs of Fig. 6(a) and (b). The bottom right graphs are the magnified impedance plots of high frequency region. The temperatureindependent finite series resistances were estimated from the high frequency limit as approximately 500 Ω and 200 Ω for the graphene and ITO electrodes, respectively. For the graphene electrode, an additional high frequency impedance arc of 500 Ω is indicated, which is temperature-independent. Corresponding complex capacitance behavior is shown at the bottom left of Fig. 6(a) and (b). It should be noted that some graphene electrodes with finite, unsystematic DC resistance values, exhibit the capacitance values around 0.3 nF, similar to the high frequency behavior of Fig. 6(b), supporting the high frequency response is characteristic of the graphene electrode. The unsystematic DC resistance is thought to have its origin in the short-circuit. The present observation may be in line with the one where additional high-frequency low-capacitance component with the graphene electrode of an organic semiconductor vs. PEDOT:PSS/ITO electrode.32 The electrical observation may be related to the difference in the near interface UPS spectra of the graphene electrode from ITO, implying the influence of substrate on the interfacial capacitance. Conventionally, similar to other electronic and electrochemical applications, the serially connected impedance arcs represented by RC parallel circuits have been attempted in the impedance spectroscopy for OLEDs or organic semiconductors.28-32 However, the impedance characteristics in Fig. 6 can be most succinctly represented by a capacitance relaxation on top of the base capacitance function or a parallel network of capacitance functions. In such cases, the dielectric relaxations can be represented by a Havriliak-Negami function, which is a generalized form including the Cole-Cole or Davidson-Cole dielectric relaxations. 9

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

*

∗ '( = (-.(/01+,

+, )

Page 10 of 29

(3)

2 )3

Complex capacitance plane presentations for different β and γ are shown in Fig. 7(a). Note that Debye relaxation for β=γ=1 can be represented by the resistor-capacitor series network but no circuit analogs are present otherwise. Much employed constant-phase-element (CPE) ∗ model,4∗ = (56)7$- , corresponds to the high frequency limit of '( which can be

represented by the electrical capacitance only when α=1. CPE models have been used as ‘generalized capacitors’ in the conventional impedance analysis based on the parallel RC lump theoretical models. However, CPE cannot provide the well-defined capacitance information. As indicated in Fig. 6(a) for ITO cathode and (b) for graphene cathodes the impedance data can be mainly represented by the additive capacitance effects such as 8 , - , and  . The

series resistance (9: ) and inductance elements (L) are needed for the high frequency response

and also low frequency or DC loss is described by a resistor element in parallel to the capacitance functions (9;< ). From the preliminary fittings, all the capacitance strengths and

exponents β and γ are fixed to the representative values indicated in the insets of Fig. 6(a) and (b), respectively. As shown in Fig. 6(a), for ITO electrode in parallel to the capacitance of

1.03 nF (8 +- ), a Havriliak-Negami response of the magnitude 0.63 nF ( ) with

β=0.5

and γ =0.8. The simulation of the two capacitance effects is shown in bold solid lines. The

deviation at the high frequency can be explained by the series 9: and L of the magnitudes

approximately 150 Ω and 6 µH, independent of temperature.

The response with the

graphene electrode can be described by the parallel network of 8 in 0.30 nF, a temperatureindependent Havrilak-Negami element of the capacitance magnitude 1.33 nF (- ) with β- =1

and γ- =1, i.e. a Debye response, and a temperature-dependent relaxation of the magnitude

0.69 nF and with β =1.6 and γ =0.6. The latter corresponds to the Havriliak-Negami

response of ITO electrode. The parallel network of the three capacitance effects is simulated in bold solid lines in Fig. 6(b). The deviation in the high frequency data is satisfactorily explained by the series 9: and L., similarly as ITO case, but about 3 times high values as

and 400 Ω and 20 µH. The 9: values correspond to the high frequency intercept indicated in 10

ACS Paragon Plus Environment

Page 11 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

the bottom right impedance plots of Figs. 6(a) and (b). It should be emphasized that the high frequency impedance arc of the approximately magnitude of 500 Ω, as shown in the magnified impedance plane graph in the bottom right corner of Fig. 6(b), characteristic of the graphene electrode, is successfully described by the ideal capacitance a Debye response with the capacitance strength - connected with 9 = τ- /-, rather than 9 parallel circuit in

series to the lower frequency response. In the present analysis No CPE parameter with

arbitrarily fitted α exponent is used as in the conventional impedance analysis. Consistent with the temperature-independent high frequency response, the fitted relaxation time τ- is

also almost temperature-independent as shown in Fig. 7(b).

The strongly temperature dependent low frequency relaxation τ were found to follow the Arrhenius form of the activation energy of 0.50 eV (Fig. 7(b)). Since similar activation

energy is observed for undoped samples, it is suggested that the electronic states responsible for this trapping process may be considered to be the intrinsic properties of the pristine organic layer. The capacitance relaxations are fitted into differently skewed HavriliakNegami functions as shown in Fig. 7(a). Note that β can be larger than 1 as long as βγ is smaller than 1. This leads to a positively skewed response, in contrast to the negatively skewed response when both β and γ are smaller than 1. to the peak frequency τ@ does not coincide with τ

The relaxation time corresponding

in Eq. (3) for β≠γ≠1, as previously

reported.33 There, previously reported equation was employed but accurate relaxation time

(τ@ ) is obtained as follows.34

FGH I(GJK) Fγ CDEI(GJK)

CDE

A@ = A'( B

-/γ

L

(4)

Although the τ is different by one order of the magnitude, the relaxation times corresponding to the peak frequencies are comparable in Fig. 7(b), which is consistent with

the experimental results presented in Fig. 6 and discussed above. The exponents of β and γ may reflect the characteristics of the individual samples. They also depend on the other model elements in the equivalent circuits. In any case, the exponents can be fixed as temperatureindependent for the given physical processes. With arbitrarily variable exponents β and γ, the impedance analysis cannot satisfactorily provide physically significant parameters, as in the 11

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

conventional analysis with CPEs with arbitrary α parameter. An activation energy of 0.5 eV was found for the low-voltage conductance of an OLED with ITO and Al electrodes.35 Nguyen et al. reported on the influence of deep traps on the transient current-voltage characteristics of OLEDs where the trap levels are suggested to be 0.5~0.7 eV.36 The distributed trap states in the organic semiconductors are essential elements in the organic electronic devices such as OLEDs and organics photovoltacis.37-41 Transmission-line model has been suggested to describe the distributed nature.40 The present work suggests that the application of the Havriliak-Negami capacitance functions can successfully describe the energetics of the trap states characterized by the well-defined activation energy and almost the temperature-independent frequency dispersion parameter β and γ. This is very similar to the generalized form of a parallel R-C circuit known for the trapping process in semiconductors.39 The distributed nature of the trap states was discussed in depth by Bisquert

but the corresponding circuit model was suggested to be CPE, C4∗ = Q(jω)Q$- .42 On the

other hand, the Havriliak-Negami model can represent the well-defined limiting capacitance of a physical significance, which is indeed experimentally observed, without the introduction of the resistance parameter. Recently, parallel combination of the Havriliak-Negami

capacitance functions has been shown to be far superior description of the AC behavior of polycrystalline solid electrolytes with grain boundary blocking effects against the conventional R-CPE parallel circuit models.43-46 The DC leakage current was successfully estimated by the parallel resistance element to all these capacitance functions in the present work as indicated in the equivalent circuits in Fig. 6. The parameter describes nicely the low frequency impedance behavior as shown in the impedance spectra of top right Figs. 6(a) and (b). An activation energy of approximately 0.6 eV was observed above 270 K. Similar activation energy values between the leakage current and the capacitance relaxation times suggest the trap-dominated transport in DC characteristics of OLEDs. Consistently in two different cells and also other samples, an insulating transition was observed around 270 K. This is likely to originate from the electronic properties of the organic materials. It should be mentioned that the extrapolated low frequency resistance from the conventional analysis using an R-CPE parallel circuit model with arbitrarily adjusted α parameter cannot provide a reliable estimation of DC limit resistance as found in this work. 12

ACS Paragon Plus Environment

Page 12 of 29

Page 13 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

4. Conclusion Owing to the n-type characteristics of TFTs, inverted type OLEDs are dominantly being used in AM-OLEDs. In this study, we probed the possibility of using graphene films as the cathode in inverted OLEDs. By modifying the graphene/ETL interface with Li, it was possible to reduce the prohibitively high voltage (24.5 V) to 3.2 V for achieving 1000 cd/m2. Benefited greatly by the interfacial modification the EQE increased from 3.4 % to 12.9 % at 1000 cd/m2. In-situ UPS analyses revealed that the Li doping into ETL brings forth significantly lowered LUMO level of the ETL, providing acceptably low EIB. The smallsignal as well as DC characteristics of EODs indicate the transition voltage (), which is due

to the interface charge or trap states, and the voltage ( ) for OLED emitting. The leaky DC

reverse bias with Mott-Schottky behavior of the high frequency capacitances yields the

dopant concentration and the flat-band voltage ( ) of the pristine organic materials. The

capacitive zero-bias small-signal characteristics can be successfully described by the Havriliak-Negami dielectric relaxations of the well-defined capacitance strength around

0.6~0.7 nF and the relaxation times with activation energy of 0.5 eV, which can be ascribed to the distributed trapping levels in the organic layer. Graphene electrodes showed a high frequency feature as compared to the ITO electrodes, which is successfully described by the additional capacitance element in parallel. Parallel capacitance network is suggested to be a superior description for OLED devices than the conventional series network of impedance arcs. From the practical view point, our approach suggests a readily applicable method for achieving graphene cathode inverted OLEDs, and hence, implies the possibility of using graphene cathode in AM-OLED applications. Surface analyses reveal the importance of engineering the LUMO level of ETL to achieve high electron transport in OLEDs. Our electrical analyses suggest a systematic procedure for not only interpreting the electrical components present but also extracting quantitative parameters. We believe that our work can be widely used to design and analyze graphene electrode including organic electrical devices. In addition, because our Li modification is approach is compatible to the existing OLED processing, it can be readily implemented to resolve the issue of the high energy barrier problem at the cathode/ETL interface inverted graphene cathode OLEDs. Associated content 13

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

-

Supporting information

The Supporting Information is available free of charge on the ACS Publications website at DOI: xxx/acsami.xxx. Figure S1. Graphene film surface morphology. Figure S2. Secondary cutoff region of UPS spectra. (a) BmPyPB/Graphene, (b) Li dopedBmPyPb/Graphene, (c) BmPyPB/ITO, (d) Li doped-BmPyPB/ITO. Notes The authors declare no competing financial interest. Acknowledgements This work was supported by the Ministry of Trade, Industry and Energy/Korea Evaluation Institute of Industrial Technology (Development of basic and applied technologies for OLEDs with Graphene, MOTIE/KEIT 10044412), Korea. This work was supported by the National Research Foundation of Korea (NRF) grants funded by the Ministry of Science and ICT (NRF-2014R1A2A2A04004950 and NRF-2017R1A2B4009260). References [1] Geim A.K. Graphene: Status and Prospects. Science 2009, 324, 1530-1534. [2] Bonaccorso F.; Sun Z.; Hasan T.; Ferrari A.C. Graphene Photonics and Optoelectronics. Nature Photon. 2010, 4, 611-622. [3] Song S.M.; Park J.K.; Sul O.J.; Cho B.J. Determination of Work Function of Graphene under a Metal Electrode and its Role in Contact Resistance. Nano Lett. 2012, 12, 3887-3892. [4] Moon J.; Shin J.-W.; Cho H.; Han J.-H.; Cho N.S.; Lim J.T.; Park S.K.; Choi H.K.; Choi S.-Y.; Kim J.-H.; Maeng M.-J.; Seo J.; Park Y.; Lee J.-I. Technical Issues in Graphene Anode Organic Light Emitting Diodes. Dia. Related Mat. 2015, 57, 68-73. [5] Cho H.; Shin J.-W.; Cho N.S.; Moon J.; Han J.-H.; Kwon Y.-D.; Cho S.; Lee J.-I. Optical Effects of Graphene Electrodes on Organic Light Emitting Diodes. IEEE J. Selected Topics in Quant. Elec. 2016, 22, 7230237. [6] Moon J.; Hwang J.; Choi H.K.; Kim T.Y.; Choi S.-Y.; Joo C.W.; Han J.-H.; Shin J.-W.; Lee B.J.; Cho D.-H.; Huh J.W.; Park S.K.; Cho N.S.; Chu H.Y.; Lee J.-I. Large Area Organic 14

ACS Paragon Plus Environment

Page 14 of 29

Page 15 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Light Emitting Diodes with Multilayered Graphene Anodes. Proc. SPIE. 2012, 8476, 84760U. [7] Park, S.; Hu, Y.; Hwang, J.; Lee, E.-S.; Casabianca, L. B.; Cai, W.; Potts, J. R.; Ha, H.-W.; Chen, S.; Oh, J.; Kim, S.O.; Kim, Y.-H.; Ishii, Y.; Ruoff, R.S. Chemical Structures of Hydrazine-treated Graphene Oxide and Generation of Aromatic Nitrogen Doping. Nat. Comm. 2012, 3, 638. [8] Hwang, J. O.; Park, J. S.; Choi, D. S.; Kim, J. Y.; Lee, S. H.; Lee, K. E.; Kim, Y.-H.; Song, M. H.; Yoo, S.; Kim, S.O. Workfunction-Tunable, N-Doped Reduced Graphene Transparent Electrodes for High-Performance Polymer Light-Emitting Diodes. ACS Nano 2012, 6, 159167. [9] Lim J.; Lee H.; Cho H.; Kwon B.-H.; Cho N.S.; Lee B.; Park J.; Kim J.; Han J.-H.; Yang J.-H.; Yu B.-G.; Hwang C.-S.; Lim S.; Lee J.-I. Flexion Bonding Transfer of Multilayered Graphene as a Top Electrode in Transparent Organic Light Emitting Diodes. Sci. Rep. 2015, 5, 17748. [10] Jeong J.K.; Jeong J.H.; Choi J.H.; Im J.S.; Kim S.H.; Yang H.W.; Kang K.N.; Kim K.S.; Ahn T.K.; Chung H.-J.; Kim M.; Gu B.S.; Park J.-S.; Mo Y.-G.; Kim H.D.; Chung H.K. 12.1-inch WXGA AMOLED Display driven by Indium-gallium-zinc oxide TFTs Array. Digest of Technical Papers - SID Int. Symp. 2008, 39, 1-4. [11] Shirota Y.; Kageyama H. Charge Carrier Transporting Molecular Materials and their Applications in Devices. Chem. Reviews. 2007, 107, 953-1010. [12] So F.; Kido J.; Burrows P. Organic Light Emitting Diodes for Solid-state Lighting. MRS Bull. 2008, 33, 663-669. [13] Sasabe H.; Kido J. Multifunctional Materials in High-performance OLEDs: Challenges for Solid-state Lighting. Chem. Mat. 2011, 23, 621-630. [14] Zhao Y.; Zhang J.; Liu S.; Gao Y.; Yang X.; Leck K.S.; Abiyasa A.P.; Divayana Y.; Mutlugun E.; Tan S.T; Xiong Q.; Demir H.V.; Sun X.W. Transition Metal Oxides on Organic Semiconductors. Org. Elec. 2014, 15, 871-877. [15] Kim J.-B.; Lee J.-H.; Moon C.-K.; Kim S.-Y.; Kim J.-J. Highly enhanced Light Extraction from Surface Plasmonic Loss Minimized Organic Light Emitting Diodes. Adv. Mat. 2013, 25, 3571-3577. [16] Parthasarathy G.; Shen C.; Kahn A.; Forrest S.R. Lithium Doping of Semiconducting Organic Charge Transport Materials. J. Appl. Phys. 2001, 89, 4986-4992. 15

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

[17] Lussem B.; Riede M.; Leo K. Doping of Organic Semiconductors. Phys. Stat. Sol. (A) 2013, 210, 9-43. [18] Kim S.J.; Ryu J.; Son S.; Yoo J.M.; Park J.B.; Won D.; Lee E.-K.; Cho S.-P.; Bae S.; Cho S.; Hong B.H. Simultaneous Etching and Doping by Cu-stabilizing Agent for Highperformance Graphene-based Transparent Electrodes. Chem. Mat. 2014, 26, 2332-2336. [19] Sasabe H.; Gonmori E.; Chiba T.; Li Y.-J.; Tanaka D.; Su S.-J.; Takeda T.; Pu Y.-J.; Nakayama K.-I.; Kido J. Wide-energy-gap Electron-transport Materials containing 3,5dipyridylphenyl Moieties for an Ultrahigh Efficiency Blue Organic Light Emitting Diode. Chem. Mat. 2008, 20, 5951–5953. [20] Su S.-J.; Gonmori E.; Sasabe H.; Kido J. Highly Efficient Organic Blue-and White-lightemitting Devices having a Carrier- and Exciton-confining Structure for Reduced Efficiency Roll-off. Adv. Mat. 2008, 20, 4189–4194. [21] Joo, C.W.; Moon, J.; Han, J.-H.; Huh, J. W.; Shin, J.-W.; Cho, D.-H.; Lee, J.; Cho, N.S.; Lee, J.-I. White transparent organic light-emitting diodes with high top and bottom color rendering indices. J. Info. Display 2015, 16, 161-168. [22] Jeon, W.S.; Park, J.S.; Li, L.; Lim, D.C; Son, Y.H.; Kwon, J.H. High Current Conduction with High Mobility by Non-radiative Charge Recombination Interfaces in Organic Semiconductor Devices. Org. Elec. 2012, 13, 939-944. [23] Kim, D. Y.; Cho, S.; Park, Y. Photoelectron Spectroscopy Study of the Interfaces between Al and tris-(8-hydroxyquinoline) Aluminum with a LiF Interlayer. J. Kor. Phys. Soc. 2000, 37, 598-604. [24] Kim, J.-H.; Seo, J.; Kwon, D.-G.; Hong, J.-A.; Hwang, J.; Choi, H. K.; Moon, J.; Lee, J.I.; Jung, D. Y; Choi, S.-Y.; Park, Y. Carrier Injection Efficiencies and Energy Level Alignments of Multilayer Graphene Anodes for Organic Light Emitting Diodes with different Hole Injection Layers. Carbon. 2014, 79, 623-630. [25] Kim, G. W.; Son, Y. H.; Yang, H. I.; Park, J. H.; Ko, I. J.; Lampande, R.; Sakong, J.; Maeng, M.-J.; Hong, J.-A.; Lee, J. Y.; Park, Y.; Kwon, J. H. Diphenanthroline Electron Transport Materials for the Efficient. Chem. Mater. 2017, 29, 8299−8312. [26] Nghia, N. V.; Park, S.; An, Y.; Lee, J.; Jung, J.; Yoo, S.; Lee, M. H.; Impact of the Number of o-carboranyl Ligands on the Photophysical and Electroluminescent Properties of Iridium(III) Cyclometalates. J. Mat. Chem. C. 2017, 5, 3024-3034 16

ACS Paragon Plus Environment

Page 16 of 29

Page 17 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

[27] Yoshida, H.; Yamada, K.; Tsutsumi, J.; Sato, N.; Complete Description of Ionization Energy and Electron Affinity in Organic Solids: Determining Contributions from Electronic Polarization, Energy Band Dispersion, and Molecular orientation. Phy. Rev. B 2015, 92, 075145. [28] Nowy S.; Ren W.; Wagner J.; Weber J.A.; Br̈ utting W. Impedance Spectroscopy of Organic Hetero-layer OLEDs as a Probe for Charge Carrier Injection and Device Degradation. Pro. SPIE. 2009, 7415, 74150G. [29] Nowy S.; Ren W.; Elschner A.; Lövenich W.; Brütting W. Impedance Spectroscopy as a Probe for the Degradation of Organic Light Emitting Diodes. J. Appl. Phys. 2010, 107, 054501. [30] Brütting W.; Riel H.; Beierlein T.; Riess W. Influence of Trapped and Interfacial Charges in Organic Multilayer Light-emitting Devices. J. Appl. Phys. 2001, 89, 1704. [31] Mo. H.-W.; Lo M.-F.; Yang Q.-D.; Ng T.-W.; Lee C.-S. Multi-alternating Organic Semiconducting Films with High Electric Conductivity. Adv. Func. Mat. 2014, 24, 53755379. [32] Kim C.-H.; Hlaing H.; Yang S.; Bonnassieux Y.; Horowitz G.; Kymissis I. Impedance Spectroscopy on Copper Phthalocyanine Diodes with Surface-induced Molecular Orientation. Org. Elec. 2014, 15, 1724-1730. [33] Moon J.; Park J.-A.; Lee S.-J.; Lee J.-I.; Zyung T.; Shin E.-C.; Lee J.-S. A Physicochemical Mechanism of Chemical Gas Sensors using an AC Analysis. Phys. Chem. Chem. Phys. 2013, 15, 9361-9374. [34] Díaz-Calleja R. Comment on the Maximum in the Loss Permittivity for the Havriliak−Negami Equation. Macromol. 2000, 33, 8924–8924. [35] Juhasz P.; Nevrela J.; Micjan M.; Novota M.; Uhrik J.; Stuchlikova L.; Jakabovic J.; Harmatha L.; Weis M. Charge Injection and Transport Properties of an Organic Light Emitting Diode. Beilstein J. Nanotech. 2016, 7, 47-52. [36] Nguyen P.H.; Scheinert S.; Berleb S.; Brüiting W.; Paasch G. The influence of Deep Traps on Transient Current-voltage Characteristics of Organic Light Emitting Diodes. Org. Elec. 2001, 2, 105-120. [37] Xu G.-F.; Sun J.-X.; Jin K.; Cai L.-C. Semi-analytic Formulae of Impedance Spectroscopy in Organic Layers with Gaussian Traps. Appl. Phys. A 2014, 116, 1637-1646. 17

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

[38] Ray B.; Baradwaj A.G.; Boudouris B.W.; Alam M.A. Defect Characterization in Organic semiconductors by forward bias capacitance-voltage (FB-CV) analysis. J. Phys. Chem. C. 2014, 118, 17461-17466. [39] Okachi T.; Nagase T.; Kobayashi T.; Naito H. Determination of localized-state distributions in organic light-emitting diodes by impedance spectroscopy. Appl. Phys. Lett. 2009, 94, 043301. [40] Garcia-Belmonte G.; Munar A.; Barea E.M.; Bisquert J.; Ugarte I.; Pacios R.; Charge Carrier Mobility and Lifetime of Organic Bulk Heterojunctions Analyzed by Impedance Spectroscopy. Org. Elec. 2008, 9, 847-851. [41] Sah C.; The Equivalent Circuit Model in Solid-state Electronics. Sol. State Elec. 1970, 13, 1547-1575. [42] Bisquert J.; Beyond the Quasistatic Approximation: Impedance and Capacitance of an Exponential Distribution of Traps. Phys. Rev. B 2008, 77, 235203. [43] Lee J.-S. A Superior Description of AC Behavior in Polycrystalline Solid Electrolytes with Current-constriction Effects. J. Korean Ceramic Soc. 2016, 53, 150-161. [44] Moon S.-H.; Kim Y.H.; Cho D.-C.; Shin E.-C.; Lee D.; Im W.B.; Lee J.-S. Sodium Ion Transport in Polymorphic Scandium NASICON Analog Na3Sc2(PO4)3 with New Dielectric Spectroscopy Approach for Current-constriction Effects. Sol. State Ion. 2016, 289, 55-71. [45] Kim J.-H.; Shin E.-C.; Cho D.-C.; Kim S.; Lim S.; Yang K.; Beum J.; Kim J.; Yamaguchi S; Lee J.-S. Electrical Characterization of polycrystalline sodium β″- alumina: Revisited and Resolved. Sol. State Ion. 2014, 264, 22-35. [46] Nguyen, H. T.; Tran, T. L.; Nguyen, D. T.; Shin, E.-C.; Kang, S.-H.; Lee J.-S. Full Parametric Impedance Analysis of Photoelectrochemical Cells: Case of a TiO2 Photoanode. J. Korean Ceram. Soc. 2018, 55, 244-260.

18

ACS Paragon Plus Environment

Page 18 of 29

Page 19 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure and Table captions. Fig.1 (a) The JVL characteristics, (b) EQE as function of L, (c) PE as function of L and (d) EL spectra. Fig. 2 UPS spectra of (a) BmPyPB/Graphene interface, (b) Li doped-BmPyPB/Graphene interface, (c) BmPyPB/ITO interface and (d) Li doped-BmPyPB/ITO interface. Fig. 3 Energy level positions and work functions as a function of BmPyPB thickness. (a) BmPyPB and Li doped BmPyPB/Graphene and (b) BmPyPB and Li doped BmPyPB/ITO. Fig. 4 (a) I-V characteristics at different temperature and (b) capacitance-voltage characteristics at different frequencies at 300 K of ITO(-)/Li doped ETL/ETL/Al(+). Fig. 5 Mott-Schottky analysis of the capacitance at 100 kHz of Fig. 2 (b). Fig. 6 (a) Impedance characteristics of ITO(-)/organic(Li)/Al(+). Capacitance Bode plots and complex plane plots (left) and complex plane impedance graphs where the lower one magnifies the high frequency range (right). The bold lined capacitance plots are simulation without series resistance (RS) and inductance (L). Note that the complex plane projections of the H-N element at different temperature are overlapped.

(b) Impedance characteristics of

graphene(-)/organic(Li)/Al(+). Capacitance Bode plots and complex plane plots (left) and complex plane impedance graphs where the lower one magnifies the high frequency range (right).

The bold lined capacitance plots are simulation without series resistance (RS) and

inductance (L). Fig. 7 (a) Havriliak-Negami type relaxations for the graphene and ITO electrodes. (b) Temperature dependence of the capacitance relaxation times of graphene(-)/Li doped ETL/ETL/Al(+) and the limiting resistance of graphene(-)/Li doped ETL/ETL/Al(+) and ITO(-)/Li doped ETL/ETL/Al(+). Table 1. The abbreviations of organic materials, their full chemical names and functions.

19

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 29

Table 1. Abbreviation

Full chemical name

Function

HAT-CN

1,4,5,8,9,11-hexaazatriphenylene-hexacarbonitrile

Charge generation layer

TAPC

1,1-bis[(di-4-tolylamino)phenyl]cyclohexane

Hole transport layer

26DCzPPy

2,6-bis(3-(carbazol-9-yl)phenyl)pyridine

Host in emissive layer

TcTa

Tris(4-carbazoyl-9-ylphenyl)amine

Host in emissive layer

Ir(ppy)3

Tris(2-phenylpyridine)Iridium

BmPyPB

1,3,5-bi[(3-pyridyl)-phen-3-yl]benzene

Dopant

in

emissive

(phosphorescent green) Electron transport layer

20

ACS Paragon Plus Environment

layer

Page 21 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Table of Contents Graphic

21

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

(a)

(b)

(c)

(d)

ACS Paragon Plus Environment Fig.1 (a) The JVL characteristics, (b) EQE as function of L, (c) PE as function of L and (d) EL spectra.

Page 22 of 29

ACS Applied Materials & Interfaces

(c) Intensity (Arb. Unit)

(b) Intensity (Arb. Unit)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Intensity (Arb. Unit)

(a)

(d) Intensity (Arb. Unit)

Page 23 of 29

Fig. 2 UPS spectra of (a) BmPyPB/Graphene interface, (b) Li doped-BmPyPB/Graphene interface, (c) BmPyPB/ITO interface and (d) ACS Li doped-BmPyPB/ITO Paragon Plus Environment interface.

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

(a)

(b)

Fig. 3 Energy level positions and work functions as a function of BmPyPB thickness. (a) BmPyPB and Li doped BmPyPB/Graphene and (b) BmPyPB and Li doped BmPyPB/ITO . ACS Paragon Plus Environment

Page 24 of 29

Page 25 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

(a)

ACS Applied Materials & Interfaces

(b)

Fig. 4 (a) I-V characteristics at different temperature and (b) capacitance-voltage characteristics at different frequencies at 300 K of ITO(-)/Li doped ETL/ETL/Al(+).

ACS Paragon Plus Environment

ACS Applied Materials & Interfaces 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

Fig. 5 Mott-Schottky analysis of the capacitance at 100 kHz of Fig. 2 (b). ACS Paragon Plus Environment

Page 26 of 29

Page 27 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

ACS Applied Materials & Interfaces

(a)

Fig. 6 (a) Impedance characteristics of ITO(-)/organic(Li)/Al(+). Capacitance Bode plots and complex plane plots (left) and complex plane impedance graphs where the lower one magnifies the high frequency range (right). The Paragon Plus Environment bold lined capacitance plots are simulation ACS without series resistance (RS) and inductance (L). Note that the complex plane projections of the H-N element at different temperature are overlapped.

(b) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

ACS Applied Materials & Interfaces

Fig. 6 (b) Impedance characteristics of graphene(-)/organic(Li)/Al(+). Capacitance Bode plots and complex plane plots (left) and complex plane impedance graphs where the lower one magnifies the high ACS Paragon Plus Environment frequency range (right). The bold lined capacitance plots are simulation without series resistance (R S) and inductance (L).

Page 28 of 29

Page 29 of 29 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41

(a)

ACS Applied Materials & Interfaces

(b)

Fig. 7 (a) Havriliak-Negami type relaxations for the graphene and ITO electrodes. (b) Temperature dependence of the capacitance relaxation times of graphene(-)/Li doped ETL/ETL/Al(+) and the limiting resistance of graphene(-)/Li doped ETL/ETL/Al(+) and ITO(-)/Li doped ETL/ETL/Al(+). ACS Paragon Plus Environment