Light emission mechanism of mixed host organic light-emitting diodesWook Song and Jun Yeob LeeCitation: Applied Physics Letters 106, 123306 (2015); doi: 10.1063/1.4916549View online: Table of Contents: 06/12?ver pdfcovPublished by the AIP PublishingArticles you may be interested inOrganic light-emitting diodes containing multilayers of organic single crystalsAppl. Phys. Lett. 96, 053301 (2010); 10.1063/1.3298558Organic light-emitting diode with liquid emitting layerAppl. Phys. Lett. 95, 053304 (2009); 10.1063/1.3200947Luminescence enhancement and emission color adjustment of white organic light-emitting diodes with quantumwell-like structuresJ. Appl. Phys. 105, 113105 (2009); 10.1063/1.3138810Long-lifetime, high-efficiency white organic light-emitting diodes with mixed host composing double emissionlayersAppl. Phys. Lett. 89, 243521 (2006); 10.1063/1.2408663A top-emission organic light-emitting diode with a silicon anode and an Sm Au cathodeAppl. Phys. Lett. 85, 5406 (2004); 10.1063/1.1823601This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: Downloaded to IP: On: Wed, 22 Apr 2015 00:28:38

APPLIED PHYSICS LETTERS 106, 123306 (2015)Light emission mechanism of mixed host organic light-emitting diodesWook Song and Jun Yeob Leea)School of Chemical Engineering, Sungkyunkwan University, 2066 Seobu-ro, Jangan-gu, Suwon,Gyeonggi 440-746, South Korea(Received 30 December 2014; accepted 19 March 2015; published online 27 March 2015)Light emission mechanism of organic light-emitting diodes with a mixed host emitting layerwas studied using an exciplex type mixed host and an exciplex free mixed host. Monitoring of thecurrent density and luminance of the two type mixed host devices revealed that the light emissionprocess of the exciplex type mixed host was dominated by energy transfer, while the light emissionof the exciplex free mixed host was controlled by charge trapping. Mixed host composition wasalso critical to the light emission mechanism, and the contribution of the energy transfer processwas maximized at 50:50 mixed host composition. Therefore, it was possible to manage the lightC 2015emission process of the mixed host devices by managing the mixed host composition. VAIP Publishing LLC. []Lifetime and quantum efficiency are the most importantparameters of organic light-emitting diodes (OLEDs) andhost materials in the emitting layer played a key role ofenhancing the lifetime and efficiency of OLEDs.1–5 The hostmaterials manage holes and electrons balance in the emittinglayer, energy transfer from the host to emitters and recombination zone of OLEDs.6–8 Therefore, various host materialshave been developed to meet the requirement of the hostmaterials for high efficiency and long lifetime OLEDs.There have been several approaches to develop theappropriate host materials for OLEDs, and the most effectivemethod was to introduce a mixed host instead of commonunipolar or bipolar host materials. In general, the mixed hostsystem has two host materials with different charge transportproperties such as a combination of a hole transport typehost and an electron transport type host.9–11 The mixed hostsystem is superior to other host material system in thatcharge balance and recombination zone can be easily tunedby changing the host composition of the mixed host.12Improved efficiency and long lifetime in the OLEDs werereported using the mixed host system.13,14 In particular, themixed host system was popular in the phosphorescentOLEDs rather than fluorescent OLEDs.The mixed host of the phosphorescent OLEDs can beclassified into exciplex type and exciplex free mixed hosts.The exciplex type mixed hosts are composed of strong holeand electron transport type host materials which can formexciplex by strong intermolecular interaction, while the exciplex free mixed hosts are made up of weak hole and electrontransport type host materials which do not generate the exciplex.13,14 Exciplex free mixed host was a main focus of thedevelopment of the host materials for phosphorescentOLEDs, but recent developments of the mixed host are beingdirected to devise exciplex type host material because ofmerits of the exciplex type host such as low driving voltageand high quantum efficiency.15 However, little has beenknown about the difference of light emission mechanismbetween the exciplex type mixed host and exciplex freea)E-mail: [email protected] Tel.: 82-31-8005-3585. Fax: 82-31-8005-3585.0003-6951/2015/106(12)/123306/4/ 30.00mixed host. It was reported that the main emission mechanism of the exciplex type mixed host is energy transfer, butthere was no report about the systematic comparison of theexciplex type and exciplex free mixed hosts.16 Therefore, itis necessary to study the light emission process of the exciplex type and exciplex free mixed hosts.Herein, we describe the light emission mechanism studyof exciplex type and exciplex free mixed hosts by monitoringthe current density and luminance of the mixed host devicesaccording to driving voltage of the devices. It was demonstrated that energy transfer dominates the light emission process of the exciplex type mixed host, while charge trappingcontrols the light emission process of the exciplex free mixedhost.The fabrication of the mixed host devices was carriedout on an indium tin oxide (ITO, 120 nm) substrate cleanedwith isopropanol, acetone, and deionized water. 4,40 ne) (TAPC, 75 nm)/4,40 ,400 -tris(N-carbazolyl)triphenylamine (TCTA, 10 nm) doublehole transport layer was deposited by vacuum thermal evaporation at a deposition rate of 0.1 nm/s. Mixed host emittinglayer of TCTA: 1,3,5-tris(N-phenylbenzimidazole-2-yl)benzene (TPBI): iridium (III) tris(2-phenylpyridine) (Ir(ppy)3)and 4,40 -di(9 H-carbazol-9-yl)biphenyl (CBP):TPBI:Ir(ppy)3were deposited by thermal evaporation. The composition of themixed host was controlled by changing relative deposition rateof two host materials. Doping concentration of Ir(ppy)3 was8%. After the emitting layer deposition, BmPyPb was formedon the emitting layer as an electron transport layer followed byLiF and Al deposition. Chemical structures of the materials areshown in Figure 1. After the cathode formation, devices wereprotected from oxygen and moisture by encapsulating thedevice with a glass lid, CaO getter, and an epoxy adhesive. Alldevice performance measurements were carried out afterencapsulation. Current density-voltage relationship was measured with Keithley 2400 source measurement unit andluminance-voltage relationship was obtained using CS 2000spectroradiometer and Keithley 2400 source measurement unit.It has been known that TCTA and TPBI form exciplexby excitation, while CBP and TPBI do not form exciplex by106, 123306-1C 2015 AIP Publishing LLCVThis article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: Downloaded to IP: On: Wed, 22 Apr 2015 00:28:38

123306-2W. Song and J. Y. LeeAppl. Phys. Lett. 106, 123306 (2015)FIG. 1. Chemical structure of materials used in the device fabrication.excitation.17 To confirm the exciplex formation of theTCTA:TPBI and CBP:TPBI mixed hosts, PL emission spectra of TCTA:TPBI and CBP:TPBI were compared with thatof TCTA, TPBI, and CBP hosts.18 PL emission of theTCTA:TPBI mixed host was shifted to long wavelengthcompared to that of TCTA and TPBI, suggesting exciplexformation between TCTA and TPBI. The emission energy ofthe TCTA and TPBI exciplex (2.83 eV) was similar to thedifference of the highest occupied molecular orbital(HOMO) of TCTA and the lowest unoccupied molecular orbital (LUMO) of TPBI (2.9 eV), supporting the exciplex formation between TCTA and TPBI. However, the CBP:TPBImixed host did not show any red-shifted PL emission compared to PL emission of CBP and TPBI, indicating that noexciplex is generated between CBP and TPBI. The exciplexwas only generated in the TCTA:TPBI mixed host becauseof strong electron donating aromatic amine unit of TCTA.CBP has only moderately electron donating carbazole unitand no exciplex formation was observed. Therefore, thecomparison of the TCTA:TPBI and CBP:TPBI can revealthe difference of light emission mechanism between exciplextype mixed host and exciplex free mixed host. The twomixed host showed extensive overlap between the emissionof the mixed host and absorption of Ir(ppy)3 as shown in supporting information.18Green phosphorescent OLEDs were fabricated by doping Ir(ppy)3 as a green triplet emitter. Figure 2(a) showsplots of current density and luminance of TCTA:TPBI(50:50) and CBP:TPBI (50:50) devices according to drivingvoltage of the devices. The current density and luminance ofthe green OLEDs were high in the TCTA:TPBI (50:50) device because of better hole injection and transport propertiesof TCTA than CBP. The driving voltage at 1000 cd/m2 of theTCTA:TPBI (50:50) device was 3.74 V compared to 3.95 Vof CBP:TPBI (50:50) devices. Turn-on voltage of the devices was 2.5 V.Quantum efficiency-power efficiency-luminance relationship of the TCTA:TPBI (50:50) and CBP:TPBI (50:50)FIG. 2. Current density–voltage-luminance curves of TCTA:TPBI (50:50)and CBP:TPBI (50:50) devices (a) and quantum efficiency–luminance–power efficiency curves of TCTA:TPBI (50:50) and CBP:TPBI (50:50)devices (b).devices is presented in Figure 2(b). The quantum efficiencyof the two devices was similar, while the power efficiencywas slightly high in the TCTA:TPBI (50:50) device. As thetwo mixed host devices possessed both hole transport typeand electron transport type host materials, both holes andelectrons are efficiently injected, which lead to the similarquantum efficiency. The rather high power efficiency ofthe TCTA:TPBI (50:50) device is due to relatively lowdriving voltage of the TCTA:TPBI device. Maximumquantum efficiency of the TCTA:TPBI (50:50) device was20.5% and that of the CBP:TPBI (50:50) device was21.2%.To investigate the light emission mechanism of the exciplex type TCTA:TPBI and exciplex free CBP:TPBI mixedhosts, the current density and luminance around turn-on voltage of the devices were measured at an interval of 0.01 V. Ingeneral, two emission processes, energy transfer from host todopant and direct charge trapping by dopant, compete in thephosphorescent OLEDs, and the two processes can be differentiated by current density-voltage-luminance relationship.The current density and voltage relationship of a diode canbe described by Shockley diode equation in the diffusion regime, which is expressed asThis article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: Downloaded to IP: On: Wed, 22 Apr 2015 00:28:38

123306-3W. Song and J. Y. Lee qVJ ¼ J0 expgkTAppl. Phys. Lett. 106, 123306 (2015) 1 ;(1)where J is a current density, Jo is a saturation current density,k is a Boltzmann constant, T is a temperature, q is a unitcharge, V represents a voltage, and n is an ideality factor.16,19The ideality factor is also called recombination parameter oremission coefficient and varies from 1 to 2 depending on thecarrier recombination process of the diodes. Langevinrecombination process which corresponds to energy transferfrom host to dopant in the emitting layer shows the idealityfactor of one, while trap dominated recombination processexhibits the ideality factor of two.20,21 Therefore, the idealityfactor tells the recombination process of the phosphorescentOLEDs. The ideality factor was calculated by rearrangingEq. (1) as shown below kT @ ln J 1:g¼q @V(2)The ideality factor calculated from Eq. (2) was plotted inFigure 3 according to driving voltage of the device. Theideality factors for the TCTA:TPBI and CBP:TPBI deviceswere 1.45 and 1.80, respectively. Comparing the ideality factor of the two devices, the CBP:TPBI device exhibited largeideality factor, which implies that trap dominated recombination is dominant in the recombination process of theCBP:TPBI device and Langevin recombination is more prevalent in the TCTA:TPBI device than in the CBP:TPBI device. In other words, the Langevin recombination processplays a more important role in the exciplex type host than inthe exciplex free host, while the trap dominated recombination plays a key role in the exciplex free host.22 This resultcan be schematically described by the exciton formation process in Figure 4. In the case of the TCTA:TPBI mixed hostdevice, holes are mostly injected into TCTA host and electrons are dominantly injected into TPBI host because ofenergy barrier for charge injection. The energy barrier forhole injection from TAPC to TCTA is only 0.2 eV, but theenergy barrier for hole injection from TAPC to TPBI is0.6 eV. Therefore, holes are mostly injected from TAPC toTCTA. For the same reason, electrons are injected fromBmPyPb to TPBI rather than from BmPyPb to TCTA.Therefore, positive polarons and negative polarons areFIG. 3. Ideality factor–voltage curves of TCTA:TPBI (50:50) and CBP:TPBI(50:50) devices.FIG. 4. Emission mechanism of exciplex type TCTA:TPBI device (a) andemission mechanism of exciplex free type CBP:TPBI device (b).mostly generated in the TCTA and TPBI, respectively,which recombine to form excitons via exciplex formationbetween TCTA and TPBI. The exciplex between TCTA andTPBI emits light and then the exciplex emission is absorbedby Ir(ppy)3 through energy transfer process. Although holesand electrons are partially injected from hole and electrontransport layers to Ir(ppy)3 directly as indicated by the ideality factor of 1.45, dominant emission mechanism is exciplexformation and subsequent energy transfer to Ir(ppy)3(Langevin recombination). This result agrees with the workreported for other exciplex type host.16The emission process of exciplex free CBP:TPBI deviceis different from the exciplex type TCTA:TPBI. In the caseof the CBP:TPBI device, holes are mostly injected fromTAPC to CBP, and electrons are typically injected fromBmPyPb to TPBI. The hole injection forms positive polaronsin the CBP, and the electro injection makes negative polarons in the TPBI. As there is no exciplex formation betweenCBP and TPBI, the host materials cannot generate excitons.Therefore, the positive polarons of CBP and negative polarons of TPBI are trapped by Ir(ppy)3 due to the deep LUMOThis article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: Downloaded to IP: On: Wed, 22 Apr 2015 00:28:38

123306-4W. Song and J. Y. LeeAppl. Phys. Lett. 106, 123306 (2015)exciplex type TCTA:TPBI host, while charge trapping wasthe main light-emission process of exciplex free CBP:TPBIdevices. Additionally, the control of the mixed host composition changed the light-emission process of the exciplex typeTCTA:TPBI host, but it had little effect on the light emissionprocess of the exciplex free CBP:TPBI host. From this work,the detailed light emission process of the exciplex and exciplex free mixed host devices was revealed, and it can beused to design mixed hosts in the future phosphorescentOLED development.FIG. 5. Ideality factor–voltage curves of TCTA:TPBI devices with differentmixed host compositions.and shallow HOMO level of Ir(ppy)3 compared to theHOMO and LUMO of the CBP and TPBI host materials,resulting in direct exciton formation in the Ir(ppy)3 emitter.Therefore, the light emission process of the CBP:TPBI device can be explained by direct charge trapping, which isreflected in the high ideality factor of 1.80 in the CBP:TPBIdevice. It was reported that the main emission process ofCBP:Ir(ppy)3 device is charge trapping rather than energytransfer16 and similar result was obtained in the CBP:TPBImixed host. This work demonstrated that the exciplex andexciplex free hosts behave differently in the light emissionprocess. Although the charge trapping is the dominant lightemission process, the excitons are also formed in the CBPand TPBI hosts, contributing to the light emission processvia energy transfer from each host to Ir(ppy)3.The light emission mechanism of the exciplex typemixed hosts was further studied by changing the mixed hostcomposition of the host materials.18 Figure 5 shows theideality factor of the TCTA:TPBI mixed host devices withdifferent mixed host compositions. TPBI contents in themixed host were 25%, 50%, and 75%. The ideality factor ofthe TCTA:TPBI mixed host devices was 1.45 at a TPBI content of 50%, but it was increased to 1.75 and 1.51 at 25%and 75% of TPBI content. This result can be correlated withthe degree of exciplex formation at different mixed hostcompositions. As the exciplex formation would be maximized when the relative ratio of TCTA and TPBI is 50:50,the ideality factor was small in the TCTA:TPBI (50:50)device. However, excess amount of TCTA or TPBI whichdid not form exciplex in the TCTA:TPBI (25:75) andTCTA:TPBI (75:25) devices induced charge trapping andincreased the ideality factor.In conclusion, light-emission mechanism of exciplextype and exciplex free mixed host devices was investigatedby monitoring the ideality factor of the mixed host devices.Energy transfer dominated the light emission process ofThis work was supported by the Development of CoreTechnologies for Organic Materials Applicable to OLEDLighting with High Color Rendering Index, development ofHigh Performance and Long Term Stable Deep BluePhosphorescence OLED Materials and high efficiency andlong lifetime in green phosphorescent organic light-emittingdiodes using host and hole transport materials funded byMOTIE.1M. A. Baldo, S. Lamansky, P. E. Burrows, M. E. Thompson, and S. R.Forrest, Appl. Phys. Lett. 75(1), 4–6 (1999).2J. C. Scott, S. Karg, and S. A. Carter, J. Appl. Phys. 82, 1454 (1997).3C. Adachi, M. A. Baldo, M. E. Thompson, and S. R. Forrest, J. Appl.Phys. 90, 5048 (2001).4M. A. Baldo, M. E. Thompson, and S. R. Forrest, Pure Appl. Chem. 71,2095 (1999).5X. Zhou, D. S. Qin, M. Pfeiffer, J. Blochwitz, A. Werner, J. Crechsel, B.Maennig, K. Leo, M. Bold, P. Erk, and H. Hartmann, Appl. Phys. Lett. 81,4070 (2002).6R. H. Parmenter and W. J. Ruppel, Appl. Phys. 30, 1548 (1959).7G. G. Malliaras and J. C. Scott, J. Appl. Phys. 85, 7426 (1999).8A. J. M akinen, I. G. Hill, and Z. H. Kafafi, “Vacuum level alignment in organic guest-host systems,” J. Appl. Phys. 92, 1598–1603 (2002).9S. Kim, W. Jeong, C. Mayr, Y.-S. Park, K.-H. Kim, J.-H. Lee, C.-K.Moon, W. Br utting, and J.-J. Kim, Adv. Funct. Mater. 23, 3896(2013).10Y.-S. Park, S. Lee, K.-H. Kim, S.-Y. Kim, J.-H. Lee, and J.-J. Kim, Adv.Funct. Mater. 23, 4914 (2013).11J. Y. Lee, J. Inf. Display 15, 139 (2014).12J.-H. Lee, J.-I. Lee, J.-Y. Lee, and H.-Y. Chu, Appl. Phys. Lett. 94,193305 (2009).13B. J. Chen, X. W. Sun, and K. R. Sarma, Appl. Phys. Lett. 88, 243505(2006).14C. Hsiao, Y. Chen, T. Lin, C. Hsiao, and J. Lee, Appl. Phys. Lett. 89,163511 (2006).15Y. Seino, H. Sasabe, Y.-J. Pu, and J. Kido, Adv. Mater. 26, 1612 (2014).16J.-H. Lee, S.-H. Lee, S.-J. Yoo, K.-H. Kim, and J.-J. Kim, Adv. Funct.Mater. 24, 4681 (2014).17B. S. Kim and J. Y. Lee, Adv. Funct. Mater. 24, 3970 (2014).18See supplementary material at forphotoluminescence and absorption spectra of the mixed host, and deviceperformances of the TCTA:TPBI mixed hosts.19W. Shockley, Bell Syst. Tech. J. 28, 435 (1949).20M. P. Langevin, Ann. Chim. Phys. 28, 433 (1903).21M. Pope and C. E. Swenberg, Electronic Processes in Organic Crystals(Clarendon Press, Oxford, 1982).22J. Kalinowski, Electronic Phenomena in Organic Solids, edited by J.Kahovec (Wiley-VCH, Weinheim, 2004), Vol. 25, p. 212.This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: Downloaded to IP: On: Wed, 22 Apr 2015 00:28:38 On: Wed, 22 Apr 2015 00:28:38 Wook Song and Jun Yeob Lee a) School of Chemical Engineering,