Exciton-Induced Degradation of Carbazole-Based Host Materials and

Subscriber access provided by MacEwan University Libraries

Article

Exciton-induced Degradation of Carbazole-based Host Materials and its Role in the Electroluminescence Spectral Changes in Phosphorescent OLEDs with Electrical Aging Hyeonghwa Yu, Yingjie Zhang, Yong Joo Cho, and Hany Aziz ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b01432 • Publication Date (Web): 31 Mar 2017 Downloaded from http://pubs.acs.org on April 5, 2017

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 free 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 accessible to all readers and 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.

ACS Applied Materials & Interfaces 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 24

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

Exciton-induced Degradation of Carbazole-based Host Materials and its Role in the Electroluminescence Spectral Changes in Phosphorescent OLEDs with Electrical Aging Hyeonghwa Yu, Yingjie Zhang, Yong Joo Cho and Hany Aziz* Department of Electrical and Computer Engineering & Waterloo Institute for Nanotechnology, University of Waterloo, 200 University Avenue West, Waterloo, Ontario, Canada N2L 3G1 E-mail: [email protected]

Abstract We investigate the origins of the long wavelength bands that appear in the emission spectra of carbazolebased host materials and play a role in the electroluminescence spectral changes of phosphorescent organic light emitting devices (PhOLEDs) with electrical aging. 4,4′-bis- (carbazol-9-y)biphenyl, CBP is used as a model carbazole host material and is studied utilizing photoluminescence, electroluminescence and atomic force microscopy measurements under various stress scenarios in both single and bilayer devices and in combination with various electron transport layer (ETL) materials. Results show that exciton-induced morphological aggregation of CBP is behind the appearance of those long wavelength bands, and that complexation between the aggregated CBP molecules and ETL molecules plays a role in this phenomenon. Comparisons between the effects of exciton and thermal stress suggest that excitoninduced aggregation may be limited to short-range molecular ordering or pairing (e.g. dimer or trimer species formation) and versus longer range ordering (crystallization) in case of thermal stress. The findings provide new insights on exciton-induced degradation in wide band gap host materials and its role in limiting the stability of PhOLEDs. Keywords: spectral red shift, phosphorescent OLED, carbazole-based host materials, complexation, aggregation, degradation

ACS Paragon Plus Environment

1


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 2 of 24

Introduction Since the invention of the first bilayer Organic Light emitting Device (OLED) almost 30 years ago, the technology has undergone tremendous progress.1 This includes the ability to obtain very high efficiencies through the use of phosphorescent emitters.2 In phosphorescent OLEDs (PhOLEDs), efficient intersystem crossing allows harvesting photons from both singlet and triplet excitons, and thus the internal quantum efficiency theoretical limit is 100%.3,4 In these devices, the phosphorescent emitter is typically used as a guest in a host:guest systems. To ensure efficient host to guest energy transfer in these systems, it is required that the host material will have a higher triplet energy level relative to that of the phosphorescent guest. Host materials with relatively wide bandgaps are therefore generally required for PhOLEDs.5–8 To this end, carbazole-based compounds, which are generally characterized by their high triplet energy (~ 2.7-3.3 eV) are often used as host materials in these devices.9 Among these, 4,4′-bis- (carbazol-9y)biphenyl, CBP, is perhaps the most widely used candidate.10–12 Due to its wide usage, and relatively simple molecular structure, CBP is often also studied as a model compound for other carbazole-based host materials.

The limited electroluminescence (EL) stability of PhOLEDs, especially the blue-emitting ones, remains as one of the longest-standing challenges that hamper commercialization. In this regard, elucidating degradation mechanisms in carbazole-based PhOLEDs has been the focus of a number of studies.13–16 For example, studies by Kondakov et al. and by Scholz et al. have shown that homolytic cleavage of carbonnitrogen (C-N) bonds in CBP13 and Tris(4-carbazoyl-9-ylphenyl)amine (TCTA) can occur under prolonged electrical bias14, a phenomenon that may play an important role in device EL degradation. We recently discovered that interactions between excitons and polarons play a leading role in the degradation of these devices during electrical bias.15,16 We also found that the accompanying red shift that occurs in the EL spectra of the devices after aging is caused by the appearance of new longer wavelength emission bands that are likely the result of a molecular aggregation process that is caused by those interactions. As the aggregated species have an energy band that is smaller than the host monomer band gap, they can act as trap sites and non-radiative recombination centers. Results also showed that device EL stability will decrease as the rate of such aggregation process increases.16 The exact origin of those long wavelength bands is however still unclear. For example, long wavelength bands can arise from a variety of intermolecular species including various excimer, exciplex, electromer, and electroplex species.17–21 Light emission from excimers and electromers arises from the relaxation of excited states of bimolecular complexes involving the same kind of molecules, whereas that from exciplexes and electroplexes arises from the relaxation of excited states of bimolecular complexes involving different kinds of molecules.


2

Page 3 of 24

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


Excimer and exciplex complex emission can in general be observed in both photoluminescence (PL) and EL, whereas electromer and electroplex emission is detectable only in EL due to the fact that it involves charged species and hence are formed only during electrical bias. Due to the strong intermolecular interactions in these complexes, they can act as quenching centers and thus decrease the quantum efficiency of OLED.22,23 Considering that electron-hole recombination zone and exciton formation occurs in the vicinity of the emission layer (EML) /electron transport layer (ETL) interface in a PhOLED during operation, the role of the ETL materials should also be considered in this process.

In this study, we investigate the origin of those degradation-induced long wavelength bands that gradually emerge in the EL spectra of typical PhOLEDs on prolonged electrical bias. For this purpose, we use devices with neat (i.e. un-doped) EMLs, utilizing the widely used CBP as a model host material. We find that those bands depend on the choice of the ETL material, revealing that the latter influences the EL spectral shifts that occur in PhOLEDs with aging. Results show that both ETL molecules and electrical driving-induced CBP aggregates participate in complex formation, leading to the observed spectral shifts.


3


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 4 of 24

Experimental Section In this work, we utilize CBP as host and hole transport material, and 1,3,5-tris(N-phenyl-benzimidazol-2yl)-benzene (TPBi), 1,3-bis(3,5-di(pyridin-3-yl)phenyl)benzene (BmPyPb) or 4,7-diphenyl-1,10phenanthroline (Bphen) as electron transport materials. The materials are purchased from EM Index and Luminescence Technology Corp., and used as received without further sublimation. All devices are fabricated on pre-patterned ITO substrates (purchased from Kintec) via thermal deposition in vacuum (at a base pressure of below 5x10-6 Torr) using an Angstrom Engineering EvoVac system. All layers are deposited at a rate of 0.1-1.0 Å/s.

The luminescence spectra were measured using an Ocean Optics spectrometer. The PL spectra were collected under an excitation wavelength at 360 nm from 200 W Hg lamp equipped with an Oriel-77720 monochromator. For ultra violet (UV) irradiation, 365 nm illumination from a UV lamp at a power density of 2.3 mW/cm2 was used. All surface roughness measurements were obtained using a Veeco Dimension 3100 atomic force microscopy (AFM) in tapping mode. Fourier transfer infra-red (FTIR) was performed with Bruker Hyperion 3000 with mercury cadmium telluride detector in reflectance mode. All measurements were carried out in a nitrogen atmosphere.


4

Page 5 of 24

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


Results and Discussion We first study the EL characteristics of a device of the structure indium tin oxide (ITO)/MoO3 (5 nm)/ CBP (30 nm)/TPBi (40 nm)/LiF (1 nm)/Al (80 nm). In this device, CBP and TPBi are used as EML and ETL materials, respectively. Figure 1a shows the EL spectra collected before and after electrical driving the device at 20 mA/cm2 for 30 mins. The spectra are normalized to the same maximum height. The initial EL spectrum exhibits the characteristic CBP emission band with a peak at 410 nm, which is herein referred to as the “CBP monomer” band. The spectrum of the same device after electrical stress shows a second band at 488 nm, which can be attributed to changes induced by the electrical stress. Since both polarons and excitons are created by the electrical bias, and in order to understand which of these two species may be behind the mechanism producing this new band, the same device structure is subjected to UV irradiation for 72 hrs (without any electrical bias). In this case, only excitons are present in the device. Figure 1b shows the EL spectra before and after the UV irradiation. Similar to the electrically-stressed device, the initial EL spectrum shows the CBP monomer band at 400 nm, whereas a new band at 490 nm again emerges after the UV irradiation. The observation that both electrical stress and UV irradiation can induce the appearance of the new band suggests that excitons are the primary stimulus behind its appearance. The fact that the intensity of the new band is higher in case of the electrically-stressed device may suggest that the underlying mechanism, although driven by excitons, is accelerated by the presence of polarons, possibly due to exciton-polaron interactions.16 The higher intensity may also be simply due to the higher exciton density produced by the electrical bias in comparison to that produced by the UV irradiation.

According to Wang et al., the new longer wavelength band is likely associated with molecular aggregation of CBP.16 However, since excitons play a role in the EL spectra shift, and CBP molecules may undergo chemical (photo)-degradation, it is possible that the new band may instead be caused by species resulting from chemical decomposition of CBP. It has been shown by Kondokov et al. that C-N bond cleavage in CBP can occur under UV irradiation, resulting in a gradual decrease in its PL quantum yield.13 At 50% loss in PL yield, the fraction of decomposed CBP molecules was estimated to be around 21%, (estimated from changes in IR spectra in the 1300-1600 cm-1 range where C-N bond stretching modes exist).24 Therefore, to test for any chemical decomposition in our samples due to UV irradiation under the conditions used in Figure 1b, we use FTIR measurements. Figure 2 shows the IR absorption spectra from a 100 nm CBP film (coated on a KBr substrate) before and after UV irradiation for 72 hrs (i.e. the same irradiation conditions as that used with the device of Figure 1b. The PL quantum yield of the CBP drops by around 80% after the irradiation, suggesting (based on the above studies) that the fraction of decomposed molecules would exceed 21% and should therefore be detectable by FTIR. The


5


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 6 of 24

test results however show no detectable changes in the IR spectra. Particularly, no new peaks are detectable in the spectra after the UV irradiation. Thus, the FTIR results suggest that any chemical decomposition in CBP as a result of UV irradiation must be insignificant. It can therefore be concluded that the new emission bands in Figure 1 arise primarily from morphological changes rather than due to chemical changes in the CBP films.

To further prove that the new band in Figure 1 is indeed associated with CBP intermolecular interactions (i.e. aggregation), we test the effect of introducing TPBi into the CBP EMLs on the EL spectra of the devices and their evolution with time. Figure 3 exhibits the collected EL spectra during electrical driving at 20 mA/cm2 for 30 mins of devices of the structure ITO/MoO3 (5 nm)/CBP:TPBi (x%) (30 nm)/TPBi (40 nm)/LiF (1 nm)/Al (80 nm) with TPBi concentrations of 0, 5, and 25 (by volume). As expected, the device without TPBi mixing shows the CBP monomer band in its initial EL (exact bands vary among devices in the range of 400-420 nm which we can attribute to changes in microcavity effects due to small variations of device thicknesses). The new longer wavelength band at 488 nm emerged after 1 mins and got stronger over time as given in Figure 3a. On the other hand, the new band is much weaker in case of 5% TPBi mixing (Figure 3b), and is completely suppressed with 25% concentration (Figure 3c). The suppression of the new band can be attributed to the role of TPBi molecules in acting as molecular spacers that separate CBP molecules from each other and thus prevent their dimerization or aggregation. This observation further supports the conclusion that the new long wavelength band is associated with increased intermolecular interactions between CBP molecules.

We have now determined that CBP molecules aggregate and contribute to the emergence of the new band when they are exposed to excitons. Next, to gain more insight into the nature of the species behind the new band and whether ETL materials play a role in it, we study the effect of varying the ETL material on the changes in EL spectra of the devices during prolonged electrical bias. For this purpose, we investigate devices that have the same EML material but three different ETL materials. The device structure is ITO/MoO3 (5 nm)/CBP (30 nm)/ETL (40 nm)/LiF (1 nm)/Al (80 nm). In these devices, CBP is used as the EML material, and the three different ETL materials are TPBi, BmPyPb and Bphen. The devices are electrically-driven at 20 mA/cm2 for a period of 30 mins and their EL spectra were collected periodically during that time. Figure 4 presents the collected spectra after being normalized to the CBP monomer emission band. As can be seen, the initial EL spectra of all devices are similar, with peaks at around 405 nm which corresponds to the CBP monomer band. As time progresses, a new emission band emerges at a longer wavelength in each case. In case of the device with the TPBi ETL, the new band had a peak at 488 nm as before, and got stronger over time, resulting in significant EL spectra change with time. On the


6

Page 7 of 24

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


other hand, the peak wavelength of the new band was at 612 nm and 562 nm for the devices with the BmPyPb and Bphen ETLs, respectively. The fact that the wavelength of the new band spectra do not correspond to any emission from either CBP or ETL materials and varied with ETL materials suggests that the new bands may originate from some complex species involving both CBP and ETL molecules.25 We note that there is no clear correlation between the emission wavelength of such complex species and the lowest unoccupied molecular orbital (LUMO) level of these ETL materials. This suggests that factors other than the energy offset between the ETL LUMO and the CBP highest occupied molecular orbital (HOMO) play a role in forming this complex. The geometry of the ETL material molecules and the intermolecular separation between ETL and CBP molecules may perhaps be important factors in this regard.

In order to investigate if these new bands indeed arise from CBP/ETL complex, the EL characteristics of devices without an ETL were studied. For this purpose, single organic layer devices with the structure: ITO/MoO3 (5 nm)/CBP (30 nm)/LiF (1 nm)/Al (80 nm) were fabricated and tested. The devices were driven at a constant current density of 125 mA/cm2 for 18 hrs. (a 5 times higher current density than that used for aging the bilayer CBP/ETL devices was employed here due to the highly unbalanced electron/hole ratio in these devices which result in a significant fraction of hole current leakage to the counter electrode and a lower density of electron-hole recombination and thus both a lower EL efficiency and slower degradation under electrical bias). Figure 5 shows EL spectra collected from these devices after various intervals of time under the electrical stress. As can be seen from the figure, the normalized EL spectra show only the CBP monomer band at 403 nm, but none of the long wavelength bands observed in Figure 4. Moreover, the EL spectra show an additional emission band at 603 nm. The broad 603 nm band is a longer-wavelength band than that of the CBP monomer, and does not appear in PL spectra, suggesting that the band corresponds to an electromer band associated with a dimer excited state that gets formed only under electrical bias.18 The appearance of this band in CBP single layered devices is consistent with the formation of CBP-CBP dimer species under electrical driving. Unlike the CBP/ETL devices, however, the EL characteristics do not show the 488, 612 or 562 nm bands observed in the bilayer devices with the TPBi, BmPyPb and Bphen ETLs, respectively in case of Figure 4, even after prolonged electrical driving, proving that the appearance of those bands requires the presence of the ETL materials. The results therefore seem to further suggest, at least on first sight, that those bands indeed arose from CBP/ETL complexes.

Seeing that those long wavelength bands emerge only in devices that contain ETLs and that their peak position varies depending on the choice of the ETL material, we further investigate if they indeed arise


7


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 24

from CBP/ETL material complexes. Towards this end, we investigate the effect of intermixing the ETL material with the CBP in order to see if it would similarly induce the formation of such complex (and hence the appearance of the same bands) as would be expected in such cases. We therefore fabricate and test devices of the following structure ITO/MoO3 (5 nm)/CBP:TPBi(x%) (30 nm)/LiF (1 nm)/Al (80 nm) with TPBi concentrations of 0, 5, 25, and 50% (by volume). The EL spectra collected after various time intervals of electrical driving at 125 mA/cm2 are shown in Figure 6a-d. Figure 6e depicts the rate of growth of the new band in each device (defined as the ratio of the height of new band to the height of the monomer in each device after the various driving times). Figure 6a shows the EL spectra of the single CBP layered device without TPBi mixing. Similar to Figure 5, all EL spectra show the CBP monomer band at 403 nm and the electromer band at 603 nm. As can be seen from Figure 6b and c, the presence of TPBi molecules (at 5% and 25% by volume, respectively) in the single CBP layer grants the appearance of the 488 nm band after electrical driving, in agreement with that in Figure 1, pointing to the formation of the complex as expected. The increase in intensity of the new band with the TPBi concentration from 0% to 25% indicates that the growth of this new band becomes faster as the number of TPBi molecules, as given in Figure 6e, demonstrating that the band indeed arises from CBP/TPBi complex. Quite surprisingly however, further increasing TPBi to 50% leads to an opposite effect and prevents this band from appearing, suggesting that the production of the CBP/TPBi complex is suppressed by the high TPBi concentrations. This can be seen in Figure 6d where, in contrast with the low-TPBi concentration devices, the CBP:50% TPBi layer device only exhibits the 403 nm CBP monomer emission band. This unusual trend indicates that while TPBi is necessary for the appearance of this band, a high CBP concentration (> 50%) is also necessary for the production of this complex suggesting that the complex may in fact involve an aggregated (or dimerized) CBP species, i.e. CBPag/TPBi complex where CBPag represents the CBP dimer or aggregate. In this context, increasing TPBi concentration inhibits the formation of the CBPag species, resulting in no complexation and no new band. Also, as can be seen from the four figures, the CBP electromer band at 603 nm disappeared completely upon introducing the TPBi (appears only in Figure 6a, the device with no TPBi), suggesting that emission from CBP-CBP electromer is replaced by emission from CBPag/TPBi complex as TPBi is introduced in low concentrations. This CBPag/TPBi complex band shows a time-increasing property evident from the fact that it did not exist initially. The increase in its emission intensity with time reflects an increasing population of the complexes with time.23 This may perhaps be attributed to molecular reorganization and reorientation (that is necessary for the CBPag species formation) which is known to have a slow response time.26 To further test the complexation behaviour between the CBPag species and TPBi molecules, and to investigate the influence of excitons in inducing it, neat CBP layers are subjected to excitons via either


8

Page 9 of 24

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


electrical stress or UV irradiation, employing for this purpose single CBP layer devices of the structure ITO/MoO3 (5 nm)/CBP (30 nm)/LiF (1 nm)/Al (80 nm). After the stress, the LiF/Al layers are peeled off using Scotch tape inside a glove box (Since the adhesion between organics and LiF/Al is the weakest among all device inter-layer interfaces, the technique allows removing the LiF/Al contact without compromising the underlying layers.27,28), and TPBi (40 nm)/LiF (1 nm)/Al (80 nm) layers are then deposited on the stressed CBP layers. Thus, the final devices are bilayer devices that comprise prestressed CBP layers and pristine TPBi layers. (A device in which the CBP layer was not pre-stressed was also included to be used as a reference). The final devices are then electrically driven and their EL spectra are collected. Figure 7a depicts the various steps of this experiment, whereas Figure 7b shows the initial EL spectra collected from the final devices. The EL spectrum of the device with the pristine (i.e. unstressed) CBP layer shows the usual CBP monomer band without any other new bands. In the devices with the pre-stressed CBP layers, however, significant spectral broadening in the 500 nm region is clearly visible. This indicates that excitons induce the formation of aggregate species within the CBP layer, i.e. CBPag species (consistent with them being essentially CBP dimers or aggregates), but that the presence of such species becomes detectable (emissive) only when TPBi molecules are introduced (hence allows the emissive CBPag/TPBi complex species to be created, in this case across the CBP/TPBi interlayer interface). It should be noted that EL spectral measurements on the single layer devices (Figure 5) after electrical stress do not show such bands. That excitons can induce molecular dimerization or aggregation may perhaps be attributed to the increased electronic dipole moment of excited molecules which may promote intermolecular dipole-dipole interactions which, in turn, may induce molecular reorientation and reorganization.15

Knowing that excitons induce aggregation in CBP, and that the latter leads to complexation with ETL material molecules, thermal stress was applied and compared to electrical stress and UV irradiation in order to verify whether they perform similar effects on morphological changes of the CBP film. Single CBP layer devices with the structure ITO/MoO3 (5 nm)/CBP (30 nm)/LiF (1 nm)/Al (80 nm) were treated under different stress scenarios: (a) no stress; (b) electrical stress under a current density of 125 mA/cm2 for 18 hrs; (c) irradiation by UV light for 18 hrs; (d) thermal stress by heating at 100 oC for 1 hr (This temperature is above the glass transition temperature of CBP, 62oC,29 and thus would lead to CBP aggregation). Devices with the three pre-stressed CBP layers are compared to one with a non-stressed CBP layer, denoted to here by “pristine” CBP. As before, the LiF/Al layers were peeled off from the devices after the stress, and then TPBi (40 nm)/LiF (1 nm)/Al (80 nm) layers were deposited. Figure 8 shows the EL spectra collected from these devices under electrical driving of 20 mA/cm2 after different time intervals (like before the spectra are normalized to the CBP monomer band). As expected, the


9


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 10 of 24

pristine CBP/TPBi device shows a CBP monomer band at 420 nm initially, whereas the emergence of the 500 nm CBPag/TPBi complex band occurred over time (Figure 8a). The EL characteristics of the UV and the electrically stressed devices are almost identical, and have an EL trend similar to that of the pristine device (Figure 8b and c) except that the CBPag/TPBi complex bands grew much faster. The faster increase in the complex band in these devices is not surprising and can be attributed to the fact that CBP aggregation had already been initiated earlier during the stressing of the CBP film, and thus complexation with the TPBi happened relatively faster compared to that in the pristine CBP device. On the other hand, the EL spectra of the thermally stressed CBP/TPBi device (Figure 8d) exhibited only a slight increase in the complex band, much less than in the other devices. This suggests that thermal stress brings about certain morphological changes in CBP that may result in a decrease in its susceptibility to excitoninduced aggregation later. The different behaviour in the exciton-stressed devices versus the thermallystressed devices suggests that the aggregated species might be different in the two cases, where, for example, CBP aggregation induced by excitons is limited to short range molecular ordering or “pairing” that produces strongly interacting intermolecular species (e.g. dimer or trimer species), compared to the longer-range molecular ordering (crystallization) in case of thermal stress.

For a better understanding of CBP morphologies, the topographies of the CBP films under the same four different stress scenarios are given in Figure 9. As can be seen, the morphologies of the UV irradiation and the electrically stressed CBP films are almost identical to that of the pristine CBP film, whereas that produced by thermal stress is distinctly different. The surface roughness (Rms) 0.7 ±0.1 nm and peak-tovalley roughness (Rpv) 7.0 ± 1.0 nm are obtained for the pristine, the UV irradiation and the electrically stressed CBP films. In contrast, the thermally stressed CBP film presents crystalline features with many pin holes in the CBP film. Hence, the Rms, 1.5 nm, and Rpv, 15.8 nm are doubled, compared to that of the other CBP films. Clearly, the thermal stress clearly leads to long-range crystallization in CBP whereas both UV irradiation and electrical stress do not. This is consistent with a scenario where the CBPag species induced by excitons have only a short-range order, perhaps consisting of only a few CBP molecules each, and thus are below what can be detected by AFM. The EL characteristics and the AFM images therefore show that CBP aggregation induced by excitons differs from that induced by thermal stress.


10

Page 11 of 24

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


Conclusion In conclusion, the long wavelength bands that appear in the emission spectra of CBP are found to be caused primarily by molecular aggregation in the material, induced by excitons. Although the aggregation occurs primarily in the host material, complexation between these aggregates and the ETL material occurs and influences the location of this band and hence the spectral shifts that occur. Comparisons between the effects of exciton and thermal stress suggest that exciton-induced aggregation may be limited to shortrange molecular ordering or pairing (e.g. dimer or trimer species formation) versus longer range ordering (crystallization) in case of thermal stress. The findings provide new insights on exciton-induced degradation in wide band gap host materials and its role in limiting the stability of PhOLEDs.


11


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 12 of 24

Reference (1)

Tang, C. W.; VanSlyke, S. A. Organic Electroluminescent Diodes. Appl. Phys. Lett. 1987, 51 (12), 913.

(2)

Forrest, S. R.; Baldo, M. A.; O’Brien, D. F.; You, Y.; Shoustikov, A.; Sibley, S.; Thompson, M. E. Highly Efficient Phosphorescent Emission from Organic Electroluminescent Devices. Nature 1998, 395 (6698), 151–154.

(3)

Adachi, C.; Baldo, M. A.; Thompson, M. E.; Forrest, S. R. Nearly 100% Internal Phosphorescence Efficiency in an Organic Light-Emitting Device. J. Appl. Phys. 2001, 90, 5048.

(4)

Forrest, S. R.; Baldo, M. A.; Thompson, M. E. High-Efficiency Fluorescent Organic LightEmitting Devices Using a PhosphorescentSensitizer. Nature 2000, 403 (6771), 750–753.

(5)

Brunner, K.; van Dijken, A.; Börner, H.; Bastiaansen, J. J. A. M.; Kiggen, N. M. M.; Langeveld, B. M. W. Carbazole Compounds as Host Materials for Triplet Emitters in Organic Light-Emitting Diodes: Tuning the HOMO Level without Influencing the Triplet Energy in Small Molecules. J. Am. Chem. Soc. 2004, 126 (19), 6035–6042.

(6)

Bagnich, S. A.; Athanasopoulos, S.; Rudnick, A.; Schroegel, P.; Bauer, I.; Greenham, N. C.; Strohriegl, P.; Köhler, A. Excimer Formation by Steric Twisting in Carbazole and Triphenylamine-Based Host Materials. J. Phys. Chem. C 2015, 119 (5), 2380–2387.

(7)

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 (38), 5450–5454.

(8)

Seo, J.-A.; Jeon, S. K.; Gong, M. S.; Lee, J. Y.; Noh, C. H.; Kim, S. H.; Liao, L.-S.; Grigalevicius, S.; Grazulevicius, J. V.; Hsu, C.-P.; Forrest, S. R.; Brown, J. J. Long Lifetime Blue Phosphorescent Organic Light-Emitting Diodes with an Exciton Blocking Layer. J. Mater. Chem. C 2015, 3 (18), 4640–4645.

(9)

Yersin, H. Highly Efficient OLEDs with Phosphorescent Materials; Wiley, 2008.

(10)

O’Brien, D. F.; Baldo, M. a.; Thompson, M. E.; Forrest, S. R. Improved Energy Transfer in Electrophosphorescent Devices. Appl. Phys. Lett. 1999, 74 (3), 442.

(11)

Baldo, M. A.; Lamansky, S.; Burrows, P. E.; Thompson, M. E.; Forrest, S. R. Very HighEfficiency Green Organic Light-Emitting Devices Based on Electrophosphorescence. Cit. Appl. Phys. Lett. Appl. Phys. Lett 1999, 75, 4.

(12)

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 (15), 2422–2424.

(13)

Kondakov, D. Y.; Lenhart, W. C.; Nichols, W. F. Operational Degradation of Organic Light-


12

Page 13 of 24

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


Emitting Diodes: Mechanism and Identification of Chemical Products. J. Appl. Phys. 2007, 101, 24512. (14)

Scholz, S.; Walzer, K.; Leo, K. Analysis of Complete Organic Semiconductor Devices by Laser Desorption/Ionization Time-of-Flight Mass Spectrometry. Adv. Funct. Mater. 2008, 18 (17), 2541–2547.

(15)

Wang, Q.; Aziz, H. Exciton-Polaron-Induced Aggregation of Organic Electroluminescent Materials: A Major Degradation Mechanism in Wide-Bandgap Phosphorescent and Fluorescent Organic Light-Emitting Devices. Adv. Opt. Mater. 2015, 3 (7), 967–975.

(16)

Wang, Q.; Sun, B.; Aziz, H. Exciton-Polaron-Induced Aggregation of Wide-Bandgap Materials and Its Implication on the Electroluminescence Stability of Phosphorescent Organic LightEmitting Devices. Adv. Funct. Mater. 2014, 24 (20), 2975–2985.

(17)

Birks, J. B. Excimers. Reports Prog. Phys. 1975, 38 (8), 903–974.

(18)

Kalinowski, J.; Giro, G.; Cocchi, M.; Fattori, V.; Di Marco, P. Unusual Disparity in Electroluminescence and Photoluminescence Spectra of Vacuum-Evaporated Films of 1,1-Bis ((Di-4-Tolylamino) Phenyl) Cyclohexane. Appl. Phys. Lett. 2000, 76 (2000), 2352.

(19)

Granlund, T.; Pettersson, L. a. a.; Anderson, M. R.; Inganäs, O. Interference Phenomenon Determines the Color in an Organic Light Emitting Diode. J. Appl. Phys. 1997, 81 (12), 8097.

(20)

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 (39), 4914– 4920.

(21)

Yang, S.; Jiang, M. White Light Generation Combining Emissions from Exciplex, Excimer and Electromer in TAPC-Based Organic Light-Emitting Diodes. Chem. Phys. Lett. 2009, 484 (1–3), 54–58.

(22)

Naitou, R.; Ichikawa, M.; Suzuki, T.; Adachi, C.; Koyama, T.; Taniguchi, Y. Energy Relay Systems for Efficient Organic Light-Emitting Devices: Influence of Formation of Exciplex as Revealed by Photoexcitation. Mol. Cryst. Liq. Cryst. Sci. Technol. Sect. A. Mol. Cryst. Liq. Cryst. 2001, 370 (1), 31–34.

(23)

Wei, M.; Gui, G.; Chung, Y.-H.; Xiao, L.; Qu, B.; Chen, Z. Micromechanism of Electroplex Formation. Phys. status solidi 2015, 252 (8), 1711–1716.

(24)

Glaser, T.; Beck, S.; Lunkenheimer, B.; Donhauser, D.; Köhn, A.; Kröger, M.; Pucci, A. Infrared Study of the MoO3 Doping Efficiency in 4,4′-bis(N-Carbazolyl)-1,1′-Biphenyl (CBP). Org. Electron. 2013, 14 (2), 575–583.

(25)

Lu, H.-W.; Weng, H.-L.; Kao, P.-C.; Chu, S.-Y.; Juang, Y.-D. Fabrication of Color-Tunable BlueViolet Organic Light Emitting Diodes for White Light Source. ECS J. Solid State Sci. Technol.


13


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 14 of 24

2016, 5 (6), R104–R109. (26)

Zhang, J.; Chen, Z.; Liu, Y.; Huang, M.; Wei, Q.; Gong, Q. Improvement on the Photorefractive Performance of a Monolithic Molecular Material by Introducing Electron Traps. Appl. Phys. Lett. 2004, 85 (8), 1323.

(27)

Phatak, R.; Tsui, T. Y.; Aziz, H. Dependence of Dark Spot Growth on Cathode/organic Interfacial Adhesion in Organic Light Emitting Devices. J. Appl. Phys. 2012, 111 (5), 54512.

(28)

Liew, Y.-F.; Aziz, H.; Hu, N.-X.; Chan, H. S.-O.; Xu, G.; Popovic, Z. Investigation of the Sites of Dark Spots in Organic Light-Emitting Devices. Appl. Phys. Lett. 2000, 77 (17), 2650–2652.

(29) Tsai, M.-H.; Hong, Y.-H.; Chang, C.-H.; Su, H.-C.; Wu, C.-C.; Matoliukstyte, A.; Simokaitiene, J.; Grigalevicius, S.; Grazulevicius, J. V.; Hsu, C.-P. 3-(9-Carbazolyl)carbazoles and 3,6-Di(9Carbazolyl)carbazoles as Effective Host Materials for Efficient Blue Organic Electrophosphorescence. Adv. Mater. 2007, 19 (6), 862–866.


14

Page 15 of 24

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


Figure 1. EL spectra (normalized to the CBP monomer band intensity) of CBP/TPBi devices subjected to (a) electrical stress at electrical driving at 20 mAcm-2 for 30 mins and (b) UV stress for 72 hrs, collected before and after the stresses.


15


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 16 of 24

Figure 2. FTIR spectra of CBP film, collected before and after subjecting it to the UV stress for 72 hrs.


16

Page 17 of 24

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


Figure 3. EL spectra of CBP/TPBi devices where CBP layers mixed with TPBi ((a) 0, (b) 5 and (c) 25 % by volume), collected after electrical driving at 20mAcm-2 for certain periods of time.


17


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 18 of 24

Figure 4. EL spectra (normalized to the CBP monomer band intensity) of CBP/ETL devices with (a) TPBi, (b) BmPyPb, (c) Bphen ETLs, collected after electrical driving at 20mAcm-2 for certain periods of time.


18

Page 19 of 24

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


Figure 5. EL spectra (normalized to the CBP monomer band intensity) of single CBP organic layer device, collected after electrical driving at 125mAcm-2 for certain periods of time.


19


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 24

Figure 6. EL spectra of single organic layer devices with CBP mixed with TPBi ((a) 0, (b) 5, (c) 25 and (d) 50% by volume), collected after electrical driving at 20mAcm-2 for certain periods of time. (e) Change in 490nm to 410nm EL bands ratio in the devices ((a)-(d)) over time.


20

Page 21 of 24

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


Figure 7 (a) Illustration of conversion of single to double layer devices, in which single CBP layer devices were subjected to various stress scenarios, and then converted to double layer devices with TPBi/LiF/Al layers. (b) Initial EL spectra of the double layer devices in step 5.


21


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 22 of 24

Figure 8. EL spectra (normalized to the CBP monomer band intensity) of pre-treated CBP/TPBi devices ((a) pristine, (b) electrically-stressed, (c) UV-stressed and (d) thermally-stressed CBP layers), collected after electrical driving at 20mAcm-2 for certain periods of time.


22

Page 23 of 24

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


Figure 9. AFM images of the pre-treated CBP layers ((a) pristine, (b) electrically-stressed, (c) UVstressed, and (d) thermally-stressed) after peeling off LiF/Al layers. Rms and Ppv values are indicated on the AFM images.


23


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 24 of 24

Table of Contents (TOC) graphic


24

Exciton-Induced Degradation of Carbazole-Based Host Materials and

Recommend Documents