Achieving Extreme Utilization of Excitons by an Efficient Sandwich

Research Article www.acsami.org

Achieving Extreme Utilization of Excitons by an Efficient SandwichType Emissive Layer Architecture for Reduced Efficiency Roll-Off and Improved Operational Stability in Organic Light-Emitting Diodes Zhongbin Wu,† Ning Sun,‡ Liping Zhu,† Hengda Sun,† Jiaxiu Wang,† Dezhi Yang,† Xianfeng Qiao,† Jiangshan Chen,† Saad M. Alshehri,§ Tansir Ahamad,§ and Dongge Ma*,†,§

Downloaded via UNIV OF WINNIPEG on October 23, 2018 at 15:49:42 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

State Key Laboratory of Polymer Physics and Chemistry, Changchun Institute of Applied Chemistry, University of Chinese Academy of Sciences, Changchun 130022, People’s Republic of China ‡ The Organic Photonics and Electronics Group, Department of Physics, Umeå University, SE-901 87 Umeå, Sweden § Department of Chemistry, King Saud University, Riyadh, Kingdom of Saudi Arabia S Supporting Information *

ABSTRACT: It has been demonstrated that the efficiency roll-off is generally caused by the accumulation of excitons or charge carriers, which is intimately related to the emissive layer (EML) architecture in organic light-emitting diodes (OLEDs). In this article, an efficient sandwich-type EML structure with a mixed-host EML sandwiched between two single-host EMLs was designed to eliminate this accumulation, thus simultaneously achieving high efficiency, low efficiency roll-off and good operational stability in the resulting OLEDs. The devices show excellent electroluminescence performances, realizing a maximum external quantum efficiency (EQE) of 24.6% with a maximum power efficiency of 105.6 lm W−1 and a maximum current efficiency of 93.5 cd A−1. At the high brightness of 5 000 cd m−2, they still remain as high as 23.3%, 71.1 lm W−1, and 88.3 cd A−1, respectively. And, the device lifetime is up to 2000 h at initial luminance of 1000 cd m−2, which is significantly higher than that of compared devices with conventional EML structures. The improvement mechanism is systematically studied by the dependence of the exciton distribution in EML and the exciton quenching processes. It can be seen that the utilization of the efficient sandwich-type EML broadens the recombination zone width, thus greatly reducing the exciton quenching and increasing the probability of the exciton recombination. It is believed that the design concept provides a new avenue for us to achieve highperformance OLEDs. KEYWORDS: efficiency roll-off, operational stability, sandwich-type, organic light-emitting diodes, extreme utilization of nanoseconds,14,15 the triplet excitons suffer from being quenched by other triplets, singlets, polarons, electrical field, even the thermal effect, leading to the efficiency roll-off. Among, triplet−triplet annihilation (TTA) and triplet-polaron quenching (TPQ) have been considered as the dominant mechanisms responsible for the efficiency roll-off in phosphorescent OLEDs.16−21 To achieve high efficiency and reduce efficiency roll-off in the phosphorescent OLEDs, an effective exciton confinement in EML is indispensable. This is often realized through the use of charge and exciton blocking layers with specific properties, such as a shallow lowest unoccupied molecular orbital (LUMO) energy level to block the electrons, a deep highest occupied molecular orbital (HOMO) energy level to block the holes, or a large energy gap with a high triplet

1. INTRODUCTION Over recent years, organic light-emitting diodes (OLEDs) have achieved enormous progress emerging as the most promising technology for next generation solid-state lighting because of its high efficiency and attractive features such as the homogeneous large area, color tunability, and flexibility.1−5 Because of the spin−orbital coupling (SOC) interaction, phosphorescent OLEDs are able to achieve high efficiency by converting nearly 100% of the injected electrons and holes into photons via the utilization of both singlet and triplet excitons.6 At present, much effort has been undertaken to increase the device efficiency with a maximum external quantum efficiency (EQE) gradually close to the theoretical limitation.7,8 However, for the case of phosphorescent OLEDs, the efficiency reduction at high brightness levels, an effect known as “efficiency roll-off”, still remains unsolved.9−13 Because of the long natural exciton lifetime of triplets in the range of microseconds by comparison with the singlets on the order © 2016 American Chemical Society

Received: November 3, 2015 Accepted: January 21, 2016 Published: January 21, 2016 3150

DOI: 10.1021/acsami.5b10532 ACS Appl. Mater. Interfaces 2016, 8, 3150−3159

Research Article

ACS Applied Materials & Interfaces

external quantum efficiency (EQE) are 93.5 cd A−1, 105.6 lm W−1 and 24.6%, respectively, and remain as high as 92.4 cd A−1, 93.6 lm W−1 and 24.3% at 1000 cd m−2, 88.3 cd A−1, 71.1 lm W−1, and 23.3% at 5000 cd m−2, 84.2 cd A−1, 60.1 lm W−1 and 22.2% at 10000 cd m−2, which should be among the best values reported for bis(2-phenylpyridine)iridium acetylacetonate (Ir(ppy)2(acac))-based OLEDs. By using 0.05 nm-thick ultrathin sensing layer, the exciton distribution profiles of various EMLs including the S-EML, D-EML, M-EML and sandwich-type emissive layer in the devices were systematically probed. It is clearly shown that there exists a rather broad exciton recombination zone in the sandwich-type EML. We also compared the lifetimes of fabricated OLEDs with these EMLs, and thereby confirmed that the device with the efficient sandwich-type EML holds more potential to achieve both high efficiency and operational stability than the devices with the conventional S-EML, D-EML, and M-EML.

level to confine excitons inside the EML (see Table 1). Although the blocking layers can strongly confine the charges Table 1. HOMO, LUMO, Singlet Energy, and Triplet Energy of the Used Materials 36

TAPC TCTA36 Bepp230,35 Ir(ppy)2(acac)10 BmPyPB36

HOMOa (eV)

LUMOb (eV)

S1c (eV)

T1d (eV)

−5.5 −5.7 −5.7 −5.2 −6.4

−2.1 −2.4 −2.6 −2.5 −2.7

3.26 3.12 2.60

2.90 2.70 2.45 2.40 2.78

3.45

a

Highest occupied molecular orbital energy level. bLowest unoccupied molecular orbital energy level. cSinglet state energy. dTriplet state energy.

and excitons inside the EML, to some extent these confined charges and excitons also tend to accumulate at the interfaces. As a result, the triplet excitons are prone to be quenched via the TTA and TPQ processes, resulting in the efficiency roll-off at high brightness. In the design of phosphorescent OLEDs, double-emissive layer (D-EML) and mixed-emissive layer (M-EML) are two main structures to adjust charge balance and exciton distribution.22−28 The D-EML consists of two distinct layers of a hole-transporting material (HTM) and an electrontransporting material (ETM) with an emissive guest uniformly doped throughout both layers, whereas the M-EML contains a uniformly mixed layer of HTM and ETM doped by an emissive guest. Although they broaden the recombination region and increase the exciton recombination probability, thus reducing the exciton quenching to a certain extent, and finally further enhancing the device efficiency and mitigating the efficiency roll-off at high brightness compared with the conventional single-emissive layer (S-EML) structure-based devices, the efficiency roll-off still exists in the above two EML structure devices. As we can see in our studies below, the exciton accumulation in the emissive region cannot be fully eliminated, even there yet are certain excitons to diffuse into the transport layers. Increasing the thickness of EML seems an effective method to ensure the exciton recombination zones to be completely inside this EML; however, the increase in EML thickness will lead to the rise of operational voltage, resulting in the decrease of power efficiency and operational stability. A recent study reported by Erickson et al. demonstrated that a single-layer OLED with a graded-emissive layer (G-EML), where a preferentially hole transporting material was doped into an electron-transporting material at a concentration that changes continuously across the thickness of the EML, exhibited a broad exciton recombination zone and resulted in a lower efficiency roll-off compared to the device with the DEML.29 However, the graded single-layer structure exhibits low device efficiency, and the concentration gradient layer is complicated to operate, which is unsuitable for the commercial application. Therefore, to accomplish the high-efficiency/ reduced efficiency roll-off/operational stability trade-off, much work still needs to be further carried out in the device architectures. In this article, we design an efficient sandwich-type EML architecture, where a mixed-host EML is resided between the two single-host EMLs. An extremely high-performance OLED with nearly 100% internal quantum efficiency (IQE) was realized. The maximum current efficiency, power efficiency and

2. EXPERIMENTAL SECTION The fabricated devices were grown on clean glass substrates precoated with a 180 nm thick layer of ITO with a sheet resistance of 10 Ω per square. The ITO surface was treated with oxygen plasma for 2 min, followed a degrease in an ultrasonic solvent bath, then it was dried at 120 °C before it was loaded into an evaporator. All layers were grown in succession by thermal evaporation without breaking vacuum (∼5 × 10−4 Pa). The evaporation rates were monitored by a frequency counter and calibrated by a Dektak 6 M profiler (Veeco). The typical evaporation rate for the organic layer, Li2CO3, and Al are 1−2, 0.1− 0.2, and 10−20 Å/s, respectively. The evaporation rate for ultrathin phosphorescent dyes is 0.01−0.02 Å/s. The thickness for the ultrathin nondoped sensing layer is the effective or average thickness as such thin film should be around or below monolayer thickness. The overlap between ITO and Al electrodes was 4 mm × 4 mm as the active emissive area of the devices. The Current−voltage−brightness characteristics were measured by using a Keithley source measurement unit (Keithley 2400 and Keithley 2000) with a calibrated silicon photodiode. The EL spectra were measured by a Spectrascan PR650 spectrophotometer. EQEs were calculated from the luminance, current density, and EL spectrum, assuming a Lambertian distribution. The PL measurement was carried out using fluorescence spectrophotometer (HITACHI, F-7000). Excitation wavelengths were 330 nm. Solid films were prepared by vacuum thermal evaporation at a thickness of 40 nm. All the measurements were carried out in ambient atmosphere.

3. RESULTS AND DISCUSSION 3.1. Device Structure and Performance. Figure 1 shows the device architecture and the molecular structures used in this study. 1,1′-bis[4-(di-p-tolylamino)phenyl]cyclohexane (TAPC) doped with 20 wt % molybdenum oxide (MoO3) and bis[2-(2hydroxy-phenyl)-pyridine]beryllium (Bepp2) doped with 4 wt % lithium carbonate (Li2CO3) separately function as the hole and electron-transporting layers to achieve a low driving voltage and improve the device power efficiency. Pure TAPC layer and 1,3-bis(3,5-dipyrid-3-yl-phenyl)benzene (BmPyPB) layer act as the electron and hole blocking layer, respectively, to confine charge carriers and excitons in the EML. Table 1 shows the energy level parameters of the materials used in this study. The well-known green dopant Ir(ppy)2(acac) was selected as the emissive guest. The green OLEDs (Devices D1-D4) with various EMLs were elaborately fabricated and their structural details were shown in Figure 1a. Apart from varying the EML structure, all other parameters of these devices were kept unchanged to allow for optimum comparability, the thickness of EML was fixed by 10 nm in the devices. As shown in Figure 2, there is a large overlap between the absorption spectrum of 3151


Research Article


S-EML, D-EML, and M-EML, which are corresponding to the devices D1, D2 and D3, respectively. The S-EML is a hole preferentially transporting host doped with the guest. Although this design is easy and convenient to operate, it may cause the exciton and charge carrier aggregation with serious TTA and TPQ. The D-EML consists of two distinct emissive layers with an emissive guest doped into a hole-transport material (TCTA) and an electron-transport material (Bepp2), respectively. The D-EML structure can effectively confine the excitons and charge carriers near the EML1/EML2 interface to ensure the exciton recombination zone away from the interface between the EML and the blocking layer. And the related narrow recombination zone likely results in a much lower excitonpolaron density overlap under electrical pumping, which in turn reduces the TPQ.29 However, the TTA process is still serious, because of the dense excitons located at the EML1/EML2 interface, detrimental to relieve the efficiency roll-off. For the case of M-EML, which is composed of a mixed-host emissive layer with an emissive guest uniformly doped throughout the cohost TCTA and Bepp2, although its exciton recombination zone can be adjusted by varying the mixed ratio of TCTA: Bepp2, and is further broadened, as described above, the MEML structure also has its shortcoming, which should be further improved. Therefore, we propose a sandwich-type emissive layer structure and construct the device D4, where the cohost emissive layer is sandwiched between two S-EMLs. It can be seen that the utilization of the two S-EMLs not only broadens the exciton recombination zone, but also ensure almost all of excitons reside in the EML away from the interfaces between the EML and the blocking layers, greatly reducing the TTA and TPQ, thus further mitigating the device efficiency roll-off at high luminance. The comparison of device performances including the current density−luminance−voltage (J−L−V), the current efficiency and power efficiency versus luminance characteristics, and the EQE−luminance characteristics is depicted in Figure 3. The details of the performance are summarized in Table 2. Figure 3a portrays the current density−luminance−voltage (J− L−V) characteristics. The turn-on voltages of devices D1-D4 are 2.7, 2.5, 2.5, and 2.5 V, respectively. Clearly, the turn-on voltage of device D1 is still close to the triplet energy gap (2.4 eV) of the phosphorescent emitting dopant, Ir(ppy)2(acac) (V = hυ/q),12 but the sandwich-type EML structure we proposed, similar to the D-EML and M-EML, does not increase the device operational voltage, suggesting the small influence of the EML structure in electrical property. The current efficiency, power efficiency and EQE-luminance characteristics of the fabricated OLEDs are shown in Figure 3b, 3c and 3d. The inferior performance of D1 is well-accounted for the obvious exciton leakage into the blocking layer with nonradiative decay at the high brightness, indicating that the exciton recombination zone locates at the EML/HBL interface. It can be nicely seen that the device D4 with the sandwich-type EML shows the best performance, revealing a maximum EQE of 24.6%, a maximum current efficiency of 93.5 cd A−1, and a maximum power efficiency of 105.6 lm W−1, which are among the highest values reported so far for the Ir(ppy)2(acac)-based green OLEDs without any out-coupling enhancement technology. Considering the out-coupling efficiency limited to ηout ≈ 25%, a very high charge balance factor of 98.4% is estimated in device D4. Furthermore, a low efficiency roll-off in device D4 was also obtained, achieving a EQE of 24.3% at 1000 cd m−2, and 23.3% at 5000 cd m−2, as shown below, the reduced efficiency roll-off

Figure 1. (a) Device structure of devices D1−D4 with different emissive layers. (b) Molecular structures used in the devices.

Figure 2. Normalized absorption spectrum of Ir(ppy)2(acac) in CHCl3 solvent and photoluminescence spectra of pure TCTA, pure Bepp2, and TCTA:Bepp2 (1:1) mixture solid films.

Ir(ppy)2(acac) and the emission spectra of TCTA and Bepp2, furthermore, the PL spectra show that TCTA:Bepp2 film does not show the long wavelength emission by exciplex, so it can rule out the exciplex formed between the hole-transporting material TCTA and electron-transporting material Bepp2. As stated previously, the EML architectures have obvious impact on the efficiency roll-off. We compare the influence of 3152


Research Article


Figure 3. Comparison of the EL performances of devices D1−D4: (a) current density−luminance−voltage characteristics; (b) current efficiency versus luminance characteristics; (c) power efficiency versus luminance characteristics; (d) EQE−luminance characteristics.

Table 2. Summary of EL Performance of Devices D1−D4 CEc (cd A−1)

EQEb (%)

PEd (lm W−1)

device

Vona (V)

max

@1000 cd/m2

@5000 cd/m2

max

@1000 cd/m2

@5000 cd/m2

max

@1000 cd/m2

@5000 cd/m2

D1 D2 D3 D4

2.7 2.5 2.5 2.5

16.1 22.5 21.8 24.6

15.8 22.3 21.6 24.3

12.6 21.0 20.6 23.3

61.0 86.1 83.6 93.5

56.2 85.5 83.1 92.5

47.5 80.8 78.9 88.3

68.4 97.2 96.5 105.6

55.2 89.5 86.9 93.7

37.3 70.4 68.8 71.1

Operating voltage at the brightness of 1 cd m−2. bForward-viewing external quantum efficiency (EQE). cCurrent efficiency (CE). dPower efficiency (PE). a

sensing layer-position dependence, indicating that the inclusion of the sensing strip does not vastly influence the charge transport. Furthermore, the photons emitted from Ir(bt)2(acac) molecules in the devices originated from the bimolecular recombination energy transfer rather than trap-assisted recombination, meaning that there exists negligible charge carrier trapping in the ultrathin sensing layer, which has been proved in previous report.28 As shown in Figure 5, the electroluminescence (EL) spectra of devices D1−D4 with sensing layer translated across the EML at a current density of 10 mA cm−2 are depicted to probe the exciton distribution. We can notice that there is intense orange emission observed when placing the sensing layer into the BmPyPB layer at x = 11 and 13 nm, which should be attributed to that the leaking excitons can be utilized by the ultrathin sensing layer for emission. Meanwhile because of the deeper HOMO of BmPyPB (−6.4 eV) than that of TCTA (−5.7 eV), the possibility of holes penetrating into the BmPyPB layer

should be attributed to broadening the exciton recombination zone and minimizing the possible quenching processes. 3.2. Dependence of Exciton Distribution on the Emissive Layer Structure. To determine the shape of exciton distribution in recombination zones in devices D1− D4, we fabricated a series of phosphorescent OLEDs with a 0.05 nm thick ultrathin sensing layer of bis(2-phenyl benzothiozolato-N,C 2 ′ )iridium(acetylacetonate) (Ir(bt)2(acac)), whose relative emission intensity can provide information about the spatial distribution of the triplet excitons, along with the control devices without the sensing layer. Owing to the narrow sensing layers, their presence in EMLs should not substantially impact the charge carrier transport and recombination properties, which is proved in Figure 4. The voltages of devices with an ultrathin sensing layer and control devices at 10 mA cm−2 versus the position of sensing layer are plotted. For each of the architectures, the voltage fluctuation is derived from the minor sample-to-sample variation differences but not the 3153


Research Article


Figure 4. Operating voltages of OLEDs with ultrathin sensing layer and control OLEDs without the sensing layer versus the position of sensing layer at an applied current density of 10 mA cm−2, (a) for device D1, (b) for device D2, (c) for device D3, and (d) for device D4. The error bars are calculated from the voltage variation of the devices fabricated in different runs.

Bepp2: Ir(ppy)2(acac) in device D2 will also cause certain exciton quenching, making the efficiency roll-off at high brightness. In Figure 6c, device D3 consisting of the mixed host of TCTA and Bepp2 exhibits much wider exciton recombination zone by contrast with devices D1 and D2. As the mixed-host can simultaneously transport both electrons and holes, the excitons widely distribute in the EML. Meanwhile, according to the plotted exciton distribution profile, it is shown that the peak exciton density locates in the center of the cohost emissive layer. Furthermore, Figure 6d shows the profile of the exciton recombination zone in device D4. It can be seen that the main exciton recombination zone is further broadened, the excitons distribute across the whole EML and most excitons distribute in the cohost emissive layer. When increasing the current density, the exciton-polaron quenching correspondingly decreases and hence leads to extremely low efficiency roll-off, thereby the unique sandwich-type EML structure is the main reason for the excellent performance in device D4. 3.3. Device Operational Stability. The introduction of sandwich-type EML structure provides not only a significant enhancement in EQE and a reduction in efficiency roll-off but also an enhancement of device operational stability under electrical excitation. We compared the stability of devices D1− D4, as shown in Figure 7, the operation lifetime of devices D1− D4 was measured at an initial brightness of 5000 cd m−2 under a constant current. To predict LT50 of the resulting devices with various EMLs at initial luminance of 1000 cd m−2, we estimated an

should be very low. According to the relative luminous intensity of the ultrathin sensing layer, we get the exciton density profiles in the EMLs in devices D1−D4, which are shown in Figure 6. In device D1, the significant exciton accumulation occurs at the EML/BmPyPB interface. As a result of the narrow exciton recombination zone and the loss of exciton leakage into the blocking layer at high brightness, device D1 exhibits very low efficiency and large efficiency roll-off, which can be attributed to the exciton−exciton annihilation and exciton-polaron quenching, whereas for device D2, the utilization of double EML makes the recombination zone away from the EML/BmPyPB interface and also broadens the main exciton recombination zone width to some extent, thus greatly enhancing the efficiency and relieving the efficiency roll-off. As shown in Figure 6b, the exciton recombination zone is pinned near the TCTA:Ir(ppy)2(acac)/Bepp2:Ir(ppy)2(acac) interface, and further penetrating into the Bepp2: Ir(ppy)2(acac) layer. This is because the mobility of holes in TCTA is slightly larger than that of electrons in Bepp2 and the highest occupied molecular orbital (HOMO) level of TCTA (−5.83 eV) is deeper than that of Bepp2 (−5.7 eV),30,31 the injected holes are prior to reach the interface more quickly, and then combine with the injected electrons at the side of the Bepp2: Ir(ppy)2(acac). Additionally, the triplet energy level of the host TCTA (2.76 eV) is higher than that of the host Bepp2 (2.45 eV),30,31 and the triplet excitons on the TCTA molecules can diffuse into the Bepp2, resulting in the vivid peak of relative exciton density located at the side of Bepp2: Ir(ppy)2(acac) in device D2. However, the relative exciton density located at the side of 3154


Research Article


Figure 5. EL spectra of devices with sensing layer situated at different positions, distances to the pure TAPC through whole EML at an applied current density of 10 mA cm−2, (a) for device D1, (b) for device D2, (c) for device D3, (d) for device D4.

architecture is combined with the best possible combination of surrounding materials. 3.4. Analysis of Exciton-Density Driven Quenching Mechanism Models. The normalized EQEs of devices D1-D4 as a function of current density and the fitting according to the triplet−triplet annihilation (TTA model) (dot line), the tripletpolaron quenching (TPQ model) (dash line) and the unified model (solid line) integrating the TTA with TPQ mechanism are shown in Figure 8. The TTA model follows the established biexcitonic quenching mechanism proposed by Baldo et al.16 The equation for the TTA fitting is expressed as

acceleration factor of 1.7 for each device from the lifetime measurements using the following well-known equation32 LT × L0n = constant

(1)

Where n is the acceleration factor, LT is the operational lifetime. On the basis of eq 1, LT50 is predicted 422, 1399, 1467, and 1975 h, respectively, at the initial luminance of 1000 cd m−2. The results indicate an increasing trend in LT50 from device D1−D4. Interestingly, the trend is almost consistent with the width of exciton distribution profile in the EMLs in devices D1−D4. The significant dependence of the operational lifetime on the EML structures is due to the broadened exciton recombination zone, which minimizes the interactions between triplets on dopant and polarons on host molecules that lead to molecular decomposition. The shortest lifetime of device D1 should be related to the serious polaron accumulation at the EML/BmPyPB interface.33 Devices D2 and D3 show longer lifetime than device D1, which should be attributed to the reduction of exciton-polaron quenching in the recombination zone away from the transport layers. And owing to the relatively wide exciton recombination zone, the lifetime of device D3 is further improved. It is nicely seen that device D4 with the sandwich-type EML architecture shows much more excellent operational stability compared to other devices. As known, the device lifetime is closely related to the surrounding materials such as the host and transporting layer materials. It is believed that much more improvements would be achieved in the operational stability when the sandwich-type EML

ηEQE η0

=

⎞ J0 ⎛ ⎜ 1 + 8J − 1⎟ ⎟ 4J ⎜⎝ J0 ⎠

(2)

Where η0 represents the maximum EQE in the absence of TTA, J0 is the critical density density, which represents the current density at which EQE drops to half of its maximum value, J is the current density of OLEDs. The TPQ fitting equation follows the quenching model proposed by Reineke et al.17 which is represented as ηEQE η0

=

1 1 + τKTPCJ1/ l + 1

(3)

Where τ is the decay lifetime of the triplet, C is a constant related to the parameters such as the dielectric constant and carrier mobility, KTP is the TPQ rate constant, l is taken to be unity. 3155


Research Article


Figure 6. Normalized relative exciton density as a function of the position for different distances to the pure TAPC at an applied current density of 10 mA cm−2, (a) for device D1, (b) for device D2, (c) for device D3, (d) for device D4.

Where W represents the width of exciton recombination zone, kTT is the TTA rate constant, the other parameters are the same as the mentioned above. The detailed calculation process is also shown in Supporting Information. The TTA and TPQ, considered as the main quenching mechanisms responsible for the efficiency roll-off in devices D1D4, are proved by the fitting results in Figure 8. The normalized EQEs as a function of the current density for all the devices show a good fitting by the unified model. It is found that the TPQ has a more dominant role compared with the TTA in device D1. Because the host TCTA is a holetransporting material and the deeper HOMO of blocking material BmPyPB (−6.4 eV, whereas that of TCTA is −5.7 eV), the injected charges would largely aggregate at the interface between the EML and blocking layer BmPyPB, the massive accumulating charges would also seriously quench the formed excitons at this interface. On the contrary, in devices D2-D4, the TTA model plays a more dominant role compared with the TPQ. Compared to device D1, the recombination zone of devices D2−D4 is located far away from the interface, reducing the TPQ, and the formed excitons are gathered at the region, leading to the increased TTA. In other words, in device D1, the efficiency roll-off is mainly due to the triplet-polaron quenching, which agrees with the result reported by Dandan Song et al.,34 whereas in devices D2−D4, the roll-offs are dominated by the TTA. In addition, we calculated the TTA and TPQ rate constants, which are summarized in Table 3. For device D1, the rate constant of triplet-polaron quenching (kTP) is much larger than that of triplet−triplet annihilation (kTT), indicating that the TPQ plays a dominant role in device D1, whereas for devices D2−D4, kTT is much larger than kTP, suggesting that the TTA dominates devices D2−D4. Moreover,

Figure 7. Normalized luminance of devices D1−D4 as a function of operating time at initial luminance of approximately 5 000 cd m−2.

The unified model proposed by Reineke et al.17 incorporates the TTA and TPQ quenching mechanisms, which is shown as following ηEQE η0

⎛ eW ⎜ = ⎜ τJ ⎜ ⎝ −

1 τ

2

( 1τ + k TPCJ1/(l+1))

+

2J k eW TT

2 k TT

⎞ + k TPCJ1/(l + 1) ⎟ ⎟ k TT ⎟ ⎠

(4) 3156


Research Article


Figure 8. Normalized EQEs of devices D1−D4 as a function of current density. The lines represent fitting curves according to the TTA model (dot lines), TPQ model (dash lines) and unified model (solid lines) combining TTA and TPQ mechanisms, (a) for device D1, (b) for device D1, (c) for device D3, (d) for device D4.

effectively utilize all the electrically generated both triplet and singlet excitons to obtain high efficiency and low efficiency rolloff in OLEDs. All the aspects mentioned above demonstrate that the efficient sandwich-type EML architecture holds great potential for the realization of high efficiency, operational stability, and low efficiency roll-off for OLEDs. It is believed that the design concept provides a new avenue for us to achieve high-performance OLEDs.

Table 3. Summary of TTA and TPQ Rate Constants in Devices D1−D4 device D1

device D2

device D3

device D4

kTTa

0.8 ± 0.3

2 ± 0.6

2.6 ± 1.5

1.4 ± 0.8

kTPb

2.4 ± 0.5

0.8 ± 0.5

0.7 ± 0.2

0.5 ± 0.3

(× 10−12 cm3 s−1) (× 10−12 cm3 s−1)

a

■

The TTA rate constant. bthe TPQ rate constant.

ASSOCIATED CONTENT

S Supporting Information *

the kTT of device D4 shows much lower than those of the other devices, which is due to the broader exciton recombination zone.

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsami.5b10532. Detailed calculation process for rate constants of TTA and TPQ (PDF)

■

CONCLUSION We have successfully designed an efficient sandwich-type EML architecture to realize high efficiency and low efficiency roll-off in OLEDs. The realization of high efficiency is well-accounted for the maximization of the charge carrier balance and exciton recombination in the EML. Whereas, the low efficiency roll-off is mainly due to the broadening of exciton recombination zone. On one hand, the two single-host EMLs in the sandwich-type EML architecture can strongly confine the excitons inside of the EML away from the interface between the emissive layer and the blocking layer. On the other hand, they can effectively harness the diffused excitons from the main exciton recombination zone and broaden the recombination zone. According to the analysis of exciton-density driven quenching mechanism models, we found that the TTA is dominant in the device with the sandwich-type emissive structure. It can be seen that the sandwich-type emissive structure can be used to

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Author Contributions

The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors gratefully acknowledge the National Natural Science Foundation of China (51333007, 91433201) and 3157


Research Article


Emitting Diodes by Introducing Reduced Carrier Injection Barriers. J. Appl. Phys. 2010, 108, 064516. (19) van Eersel, H.; Bobbert, P. A.; Janssen, R. A. J.; Coehoorn, R. Monte Carlo Study of Efficiency Roll-Off of Phosphorescent Organic Light-Emitting Diodes: Evidence for Dominant Role of TripletPolaron Quenching. Appl. Phys. Lett. 2014, 105, 143303. (20) Zhu, L. P.; Wu, Z. B.; Chen, J. S.; Ma, D. G. Reduced Efficiency Roll-Off in All-Phosphorescent White Organic Light-Emitting Diodes with an External Quantum Efficiency of Over 20%. J. Mater. Chem. C 2015, 3, 3304−3310. (21) Giebink, N. C.; Forrest, S. R. Quantum Efficiency Roll-Off at High Brightness in Fluorescent and Phosphorescent Organic Light Emitting Diodes. Phys. Rev. B: Condens. Matter Mater. Phys. 2008, 77, 235215. (22) Wang, Q.; Aziz, H. Degradation of Organic/Organic Interfaces in Organic Light−Emitting Devices due to Polaron−Exciton Interactions. ACS Appl. Mater. Interfaces 2013, 5, 8733−8739. (23) Kondakova, M. E.; Pawlik, T. D.; Young, R. H.; Giesen, D. J.; Kondakov, D. Y.; Brown, C. T.; Deaton, J. C.; Lenhard, J. R.; Klubek, K. P. High-Efficiency, Low-Voltage Phosphorescent Organic LightEmitting Diode Devices with Mixed Host. J. Appl. Phys. 2008, 104, 094501. (24) He, G.; Pfeiffer, M.; Leo, K.; Hofmann, M.; Birnstock, J.; Pudzich, R.; Salbeck, J. High-Efficiency and Low-Voltage Electrophosphorescent Organic Light-Emitting Diodes with Double-Emission Layers. Appl. Phys. Lett. 2004, 85, 3911. (25) Lee, J.; Wu, C.; Liu, S.; Huang, C.; Chang, Y. Mixed Host Organic Light-Emitting Devices with Low Driving Voltage and Long Lifetime. Appl. Phys. Lett. 2005, 86, 103506. (26) Sasabe, H.; Takamatsu, J.; Motoyama, T.; Watanabe, S.; Wagenblast, G.; Langer, N.; Molt, O.; Fuchs, E.; Lennartz, C.; Kido, J. High-Efficiency Blue and White Organic Light-Emitting Devices Incorporating a Blue Iridium Carbene Complex. Adv. Mater. 2010, 22, 5003−5007. (27) Zhou, X.; Qin, D. S.; Pfeiffer, M.; Blochwitz-Nimoth, J.; Werner, A.; Drechsel, J.; Maennig, B.; Leo, K.; Bold, M.; Erk, P.; Hartmann, H. High-Efficiency Electrophosphorescent Organic Light-Emitting Diodes with Double Light-Emitting Layers. Appl. Phys. Lett. 2002, 81, 4070. (28) Zhu, L. P.; Zhao, Y. B.; Zhang, H. M.; Chen, J. S.; Ma, D. G. Using an Ultra-thin Non-Doped Orange Emission Layer to Realize High Efficiency White Organic Light-Emitting Diodes with Low Efficiency Roll-Off. J. Appl. Phys. 2014, 115, 244512. (29) Erickson, N. C.; Holmes, R. J. Engineering Efficiency Roll-Off in Organic Light-Emitting Devices. Adv. Funct. Mater. 2014, 24, 6074− 6080. (30) Fukagawa, H.; Shimizu, T.; Kamada, T.; Kiribayashi, Y.; Osada, Y.; Hasegawa, M.; Morii, K.; Yamamoto, T. Highly Efficient and Stable Phosphorescent Organic Light-Emitting Diodes Utilizing Reverse Intersystem Crossing of the Host Material. Adv. Opt. Mater. 2014, 2, 1070−1075. (31) Su, S.; Gonmori, E.; Sasabe, H.; Kido, J. Highly Efficient Organic Blue-and White-Light-Emitting Devices Having a Carrier- and Exciton-Confining Structure for Reduced Efficiency Roll-Off. Adv. Mater. 2008, 20, 4189−4194. (32) Féry, C.; Racine, B.; Vaufrey, D.; Doyeux, H.; Ciná, S. Physical Mechanism Responsible for the Stretched Exponential Decay Behavior of Aging Organic Light-Emitting Diodes. Appl. Phys. Lett. 2005, 87, 213502. (33) Zamani Siboni, H.; Luo, Y.; Aziz, H. Luminescence Degradation in Phosphorescent Organic Light-Emitting Devices by Hole Space Charges. J. Appl. Phys. 2011, 109, 044501. (34) Song, D.; Zhao, S.; Luo, Y.; Aziz, H. Causes of Efficiency RollOff in Phosphorescent Organic Light Emitting Devices: Triplet-Triplet Annihilation versus Triplet-Polaron Quenching. Appl. Phys. Lett. 2010, 97, 243304. (35) Chen, Y.; Zhao, F.; Zhao, Y.; Chen, J.; Ma, D. Ultra-Simple Hybrid White Organic Light-Emitting Diodes with High Efficiency

Ministry of Science and Technology of China (973 program 2013CB834805) for the support of this research. Dongge Ma extends his appreciation to the Distinguished Scientist Fellowship Program (DSFP) at King Saud University, Riyadh, Kingdom of Saudi Arabia for financial support.

■

REFERENCES

(1) Tang, C. W.; VanSlyke, S. A. Organic Electroluminescent Diodes. Appl. Phys. Lett. 1987, 51, 913−915. (2) Reineke, S.; Lindner, F.; Schwartz, G.; Seidler, N.; Walzer, K.; Lüssem, B.; Leo, K. White Organic Light-Emitting Diodes with Fluorescent Tube Efficiency. Nature 2009, 459, 234−238. (3) Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R. Very High-Efficiency Green Organic Light-Emitting Devices Based on Electrophosphorescence. Appl. Phys. Lett. 1999, 75, 4. (4) Sun, Y. R.; Forrest, S. R. Enhanced Light Out-Coupling of Organic Light-Emitting Devices Using Embedded Low-Index Grids. Nat. Photonics 2008, 2, 483−487. (5) Uoyama, H.; Goushi, K.; Shizu, K.; Nomura, H.; Adachi, C. Highly Efficient Organic Light-Emitting Diodes from Delayed Fluorescence. Nature 2012, 492, 234−238. (6) Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Nearly 100% Internal Phosphorescence Efficiency in an Organic LightEmitting Device. J. Appl. Phys. 2001, 90, 5048−5051. (7) Helander, M. G.; Wang, Z. B.; Qiu, J.; Greiner, M. T.; Puzzo, D. P.; Liu, Z. W.; Lu, Z. H. Chlorinated Indium Tin Oxide Electrodes with High Work Function for Organic Device Compatibility. Science 2011, 332, 944−947. (8) Lee, C. W.; Lee, J. Y. Above 30% External Quantum Efficiency in Blue Phosphorescent Organic Light-Emitting Diodes Using Pyrido[2,3-b]indole Derivatives as Host Materials. Adv. Mater. 2013, 25, 5450−5454. (9) Murawski, C.; Leo, K.; Gather, M. C. Efficiency Roll-Off in Organic Light-Emitting Diodes. Adv. Mater. 2013, 25, 6801−6827. (10) Sun, N.; Wang, Q.; Zhao, Y. B.; Chen, Y. H.; Yang, D. Z.; Zhao, F. C.; Chen, J. S.; Ma, D. G. High-Performance Hybrid White Organic Light-Emitting Devices without Interlayer between Fluorescent and Phosphorescent Emissive Regions. Adv. Mater. 2014, 26, 1617−1621. (11) Zhang, D. D.; Duan, L.; Zhang, D. Q.; Qiu, Y. Towards Ideal Electrophosphorescent Devices with Low Dopant Concentrations: the Key Role of Triplet Up-Conversion. J. Mater. Chem. C 2014, 2, 8983− 8989. (12) Park, Y.-S.; Lee, S.; Kim, K.-H.; Kim, S.-Y.; Lee, J.-H.; Kim, J.-J. Exciplex-Forming Co-Host for Organic Light-Emitting Diodes with Ultimate Efficiency. Adv. Funct. Mater. 2013, 23, 4914−4920. (13) Wang, Q.; Ding, J. Q.; Ma, D. G.; Cheng, Y. X.; Wang, L. X.; Wang, F. S. Manipulating Charges and Excitons within a Single-Host System to Accomplish Efficiency/CRI/Color-Stability Trade-off for High-Performance OWLEDs. Adv. Mater. 2009, 21, 2397−2401. (14) Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Sibley, S.; Thompson, M. E.; Forrest, S. R. Highly Efficient Phosphorescent Emission from Organic Electroluminescent Devices. Nature 1998, 395, 151−154. (15) Holmes, R. J.; Forrest, S. R.; Tung, Y. J.; Kwong, R. C.; Brown, J. J.; Garon, S.; Thompson, M. E. Blue Organic Electrophosphorescence Using Exothermic Host−Guest Energy Transfer. Appl. Phys. Lett. 2003, 82, 2422. (16) Baldo, M. A.; Adachi, C.; Forrest, S. R. Transient Analysis of Organic Electrophosphorescence. II. Transient Analysis of TripletTriplet Annihilation. Phys. Rev. B: Condens. Matter Mater. Phys. 2000, 62, 10967. (17) Reineke, S.; Walzer, K.; Leo, K. Triplet-Exciton Quenching in Organic Phosphorescent Light-Emitting Diodes with Ir-Based Emitters. Phys. Rev. B: Condens. Matter Mater. Phys. 2007, 75, 125328. (18) Setoguchi, Y.; Adachi, C. Suppression of Roll-Off Characteristics of Electroluminescence at High Current Densities in Organic Light 3158


Research Article

ACS Applied Materials & Interfaces and CRI Trade-Off: Fabrication and Emission-Mechanism Analysis. Org. Electron. 2012, 13, 2807−2815. (36) Kim, B. S.; Lee, J. Y. Engineering of Mixed Host for High External Quantum Efficiency above 25% in Green Thermally Activated Delayed Fluorescence Device. Adv. Funct. Mater. 2014, 24, 3970− 3977.

3159


Achieving Extreme Utilization of Excitons by an Efficient Sandwich

Recommend Documents