An add-on organic green-to-blue photon-upconversion layer for organic light emitting diodes†

Lorenz Graf von Reventlow ^ab, Matthias Bremer ^c, Bernd Ebenhoch ^ab, Martina Gerken ^c, Timothy W. Schmidt ^d and Alexander Colsmann *^ab
aLight Technology Institute, Karlsruhe Institute of Technology (KIT), Engesserstrasse 13, 76131 Karlsruhe, Germany. E-mail: alexander.colsmann@kit.edu
^bMaterial Research Center for Energy Systems, Karlsruhe Institute of Technology (KIT), Strasse am Forum 7, 76131 Karlsruhe, Germany
^cChair for Integrated Systems and Photonics, Faculty of Engineering, Kiel University, Kaiserstr. 2, 24143, Kiel, Germany
^dARC Centre of Excellence in Exciton Science, School of Chemistry, University of New South Wales, Sydney, NSW 2052, Australia

First published on 15th March 2018

---

---

Various concepts have been investigated to produce deep-blue and efficient organic light emitting diodes (OLEDs) for displays, and to ultimately generate white light for general lighting applications. Among them, fluorescent emitters have been widely employed in industrial OLEDs, but such fluorescent OLEDs discard the energy available in triplet excitons.¹ Internal triplet–triplet-annihilation can produce a higher quantum efficiency by generating additional fluorescent emission.^2,3 Phosphorescent emitters enable the radiative decay of triplet excitons, yielding an about fourfold efficiency over purely fluorescent emitters.⁴ Thermally activated delayed fluorescence (TADF) harvests triplets by reverse intersystem crossing.⁵

In this work, we demonstrate the emission of deep-blue photons from a quasi-solid state upconversion layer that is applied atop a high-brightness green phosphorescent OLED. Whereas, formerly, efficient photon-upconversion has been realized mostly in solution,⁶ here, its integration into OLEDs calls for solid-state upconversion concepts that can be applied to thin-films.

For the upconversion of green, non-coherent photons, we utilized the concept of triplet–triplet-annihilation (TTA), as illustrated in Fig. 1a. Upon absorption of green photons, concurrent excitation of the singlet state S₁ and subsequent intersystem crossing (ISC), the triplet states T₁ on the organic sensitizer molecules are populated. These T₁ states can then transfer to the T₁ states of a second molecular species, the emitter, via triplet-energy-transfer (TET). Two T₁ states on two neighboring emitter molecules can then undergo TTA, combining both T₁ energies to populate the higher energy singlet state S₁ of one emitter molecule. Eventually, this S₁ state then relaxes by fluorescence to its ground state, emitting a blue photon. Since two green photons are required for the generation of one blue photon, the theoretical maximum quantum efficiency of an optically excited external upconversion layer is 50%. As TTA is a bimolecular annihilation process, a quadratic dependence of the TTA rate on the T₁ population is observed which, to be efficient, requires strong irradiance and close proximity of the sensitizer–emitter and emitter–emitter pairs. At high irradiance, TTA becomes the predominant process of triplet decay, the upconversion efficiency maximizes and becomes independent of the irradiance, and hence the emission of the upconversion system converges into a linear dependence. The transition point between the quadratic and the linear dependence of the upconverter emission on the pump-irradiation intensity defines the upconversion threshold.⁷

TET and TTA are usually Dexter-type energy transfer processes, so that the distance between two involved molecules has to be rather small, i.e. typically <3 nm.⁸ In solution, the sensitizer and emitter molecules can freely diffuse and eventually collide, allowing best energy transfer and best photon-upconversion efficiencies.⁶ The integration of photon-upconverters into OLEDs, however, calls for solid-state concepts that can be realized in thin-films.⁹ Recently, Sripathy et al. presented the idea of using photon-upconversion gels that, despite forming a quasi-solid medium, allow the molecules to diffuse.¹⁰ Building on this idea, we incorporated the green-absorbing sensitizer palladium tetraphenylporphryrin (PdTPP) and the fluorescent blue emitter 9,10-diphenylanthracene (DPA) with a predicted maximum TTA efficiency of 45% into an organogel matrix of 1,3:2,4-bis(3,4-dimethylbenzylidene) sorbitol (DMDBS). The chemical structures of the materials are depicted in Fig. 2a. The detailed preparation of the gelled upconverter is described in the ESI.† The absorption spectrum of PdTPP and the emission spectrum of DPA sensitized by PdTPP are displayed in Fig. 2b. Since the absorption spectrum of the sensitizer PdTPP and the emission spectrum of DPA show some overlap, we expect some self-absorption of the upconverter system in the Soret band of PdTPP (see ESI†). The emission of an Ir(ppy)₃-OLED matches nicely with the absorption of PdTPP. Pumping the upconverter with this type of OLED would result in an anti-Stokes shift of the peak emission wavelength of about 65 nm.

To achieve efficient upconversion, an OLED with high brightness, high external quantum efficiency and a low efficiency roll-off is required. To reduce the current density in an OLED while maintaining high luminance we designed the high-brightness tandem-OLED that is depicted in Fig. 1b. The OLEDs were fabricated in a home-built cluster tool by thermal evaporation. The detailed fabrication and characterization of the tandem-OLEDs is described in the ESI.† In such tandem architectures, each injected electron–hole pair can generate two photons because the two light-emitting units are connected in series via a tunnel contact which is often referred to as charge generation layer. Here, the architecture consists of two identical green-emitting OLEDs comprising the stable, efficient and commercially available green emitter Ir(ppy)₃, embedded in a tri-layer mixed-host matrix which was optimized for high-efficiency at high luminance and which we discuss in more detail in the ESI.† The tandem-OLED achieved an external quantum efficiency of 19% at a maximum radiance of L_max = 148 W (m² sr)⁻¹, which corresponds to a current efficiency of 68 cd A⁻¹ at a luminance of L_V,max = 71000 cd m⁻². The maximum emission wavelength of Ir(ppy)₃ at λ_max = 515 nm is close to the maximum absorption in the Q-band of PdTPP at λ_Q,PdTPP = 525 nm (Fig. 2b) which enables efficient energy transfer. Fig. 3 displays the gelled upconverter in a vial atop the tandem-OLED to demonstrate the green-to-blue photon-upconversion. Under the green illumination from Ir(ppy)₃, the deep-blue emission is clearly visible.

Yet, the integration of photon-upconversion into OLEDs requires thin-films rather than bulky vials, which provides the impetus to advance from former liquid photon-upconversion to solid-state systems. To prevent oxidation of the molecules, we prepared a 1.1 mm-thick glass-encapsulated film of the gelled upconverter that we fixed atop the green OLED. Although a monolithic device architecture with the upconverter being directly deposited onto the OLED would eventually facilitate the device fabrication process, we have deliberately chosen this manual assembly that allows us to investigate further modifications as detailed below. It also allowed the comparison of the device performance with and without the upconverter at any time during our study. Since the device can provide reference to itself, for example, we can validate that blue photons are only emitted by the upconverter and not by any wide-bandgap charge carrier transport layer within the OLED. Fig. 4a shows the emission spectra of a typical OLED with and without upconversion layer on a logarithmic scale, measured in 2π-configuration in an integrating sphere connected to a calibrated spectrometer (CAS140CT-151, Instrument Systems). The spectra were normalized to the emission at 600 nm, where the upconverter is fully transparent and hence where the normalized emission of the OLED with and without upconverter matches. About half of the OLED photon flux is absorbed by the upconversion layer. At 445 nm, the deep-blue emission of DPA is visible which alone yields CIE1931XY color coordinates of (x, y) = (0.16, 0.05) as detailed in the ESI.† We observed an increasing relative DPA emission with increasing OLED irradiance. This indicates that the photon-upconversion system operates below its threshold. This is confirmed by the double-logarithmic plot of the DPA fluorescence versus the OLED irradiance in Fig. 4b, where we find a slope of 2.35, indicating rather quadratic dependence and possibly influence from competing oxygen reactions from trace levels. Reasons for the below-threshold operation of the upconverter despite the excellent performance of the OLED include incomplete absorption of the photons that are emitted by the OLED on the sensitizer due to reflections at the glass encapsulation of the upconverter and the rather narrow absorption band of the sensitizer. To calculate the photon conversion efficiency, we determined the flux of the emitted photons Φ_e of the DPA around 450 nm and the flux of the absorbed photons Φ_a around 525 nm from the difference of the emission spectra in Fig. 4a. Then the photon conversion efficiency can be calculated from the fluxes’ ratio Φ_e/Φ_a. We obtained a maximum quantum yield of the photon-upconversion system of 0.11% at an irradiance of 196 W m⁻².

To further improve the upconversion efficiency by increasing the light absorption on the sensitizer, we applied a dichroic short-pass (SP) filter with a high reflectivity in the green and a high transmissivity in the blue (with a sharp edge at 470 nm) on top of the OLED/upconverter stack. This filter transmits most of the blue DPA emission and reflects the green emission from the OLED back into the upconverter. As a further measure to improve the upconversion efficiency, we placed a reflective dichroic long-pass (LP) filter with an edge at 480 nm between the OLED and the upconverter to avoid re-absorption of upconverted blue photons inside the OLED. The principal device set-up is illustrated in the inset of Fig. 5. Further specifications of the dichroic filters (470FDS and 480FDL, Knight Optical) are available in the ESI.† The emission spectra with and without filters are shown in Fig. 5 at different operation current densities of the OLED. They were measured perpendicular to the surface with a spectrometer (Horiba iHR320) through an optical fiber (NA = 0.22). Using the combination of LP and SP filters clearly enhances the blue-light emission of the upconverter at 445 nm by 8–11% (rel.). We note that, above 450 nm and in absence of filters, the emission is dominated by the OLED. In contrast, with the long pass filter applied, the OLED emission above 450 nm is effectively suppressed. We further note that the emission of the upconversion-OLED was measured perpendicular to the surface of the OLED to avoid any ambiguities in the data interpretation that would stem from the dependence of the transmission and reflection of the dichroic filters on the viewing angle.

In conclusion, we have presented a new concept to convert green light that is emitted from an OLED to deep-blue light. Importantly, the quasi-solid state upconverter is compatible with thin-film device concepts which will facilitate fabrication, e.g., by additive manufacturing in roll-to-roll printing. Future advantages of blue-light generation by photon-upconversion may be found in lower driving voltages, the use of more long-term stable fluorescent molecules, low-cost emitters or the generation of broad-band white-light, employing only one spectrally limited emitter in the OLED. Using only materials with low triplet energies may lead to enhanced device lifetimes compared to state-of-the-art TADF or phosphorescent OLEDs. Via TTA¹¹ and triplet–polaron-annihilation¹² in the latter two, high-energy states are excited that may break molecular bonds upon relaxation.¹³ In contrast, low triplet-energy materials as employed here, cannot affect the molecular integrity and therefore may enable improved lifetimes in future devices. The yield of blue photons could increase upon developing upconverter systems with lower thresholds, higher conversion efficiencies and higher absorption coefficients. Today's moderate upconversion efficiencies may be improved in the future by templating the molecules for best assembly (sensitizer–emitter and emitter–emitter) and thus to enhance the transfer rate of excited states between molecules. Ideally, such templates will enable the fabrication of solid-state layers that will allow a full integration of the upconversion layer into OLEDs. First approaches towards higher efficiencies and lower thresholds include self-assemblies of the sensitizer and emitter molecules¹⁴ or their incorporation into metal–organic frameworks.¹⁵

The authors are deeply indebted to Holger Röhm for setting up the cluster tool and acknowledge fruitful discussions with Stefan Höfle and Rowan MacQueen. We thank Janek Buhl for assistance with the filter measurements. A. C. thanks the Helmholtz Association for support through the Program Science and Technology of Nanosystems (STN). T. W. S. acknowledges the Australian Research Council for a Future Fellowship (FT130100177). This work was supported by the Australian Research Council Centre of Excellence in Exciton Science (CE170100026). M. B. and M. G. acknowledge funding by the European Research Council (ERC) within the “PhotoSmart” project (Starting Grant Agreement 307800).

Conflicts of interest

There are no conflicts to declare.

Notes and references

1. M. Zhu and C. Yang, Chem. Soc. Rev., 2013, 42, 4963–4976 RSC .
2. Y.-H. Chen, C.-C. Lin, M.-J. Huang, K. Hung, Y.-C. Wu, W.-C. Lin, R.-W. Chen-Cheng, H.-W. Lin and C.-H. Cheng, Chem. Sci., 2016, 7, 4044–4051 RSC .
3. D. Di, L. Yang, J. M. Richter, L. Meraldi, R. M. Altamimi, A. Y. Alyamani, D. Credgington, K. P. Musselman, J. L. MacManus-Driscoll and R. H. Friend, Adv. Mater., 2017, 29, 1605987 CrossRef PubMed .
4. T. Fleetham, G. Li, L. Wen and J. Li, Adv. Mater., 2014, 26, 7116–7121 CrossRef CAS PubMed .
5. L.-S. Cui, H. Nomura, Y. Geng, J. U. Kim, H. Nakanotani and C. Adachi, Angew. Chem., Int. Ed., 2017, 129, 1571–1575 CrossRef PubMed .
6. T. N. Singh-Rachford and F. N. Castellano, Coord. Chem. Rev., 2010, 254, 2560–2573 CrossRef CAS .
7. A. Monguzzi, J. Mezyk, F. Scotognella, R. Tubino and F. Meinardi, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 78, 195112 CrossRef .
8. A. Monguzzi, R. Tubino and F. Meinardi, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 77, 155122 CrossRef .
9. C. E. McCusker and F. N. Castellano, Top. Curr. Chem., 2016, 374, 19 CrossRef PubMed .
10. K. Sripathy, R. W. MacQueen, J. R. Peterson, Y. Y. Cheng, M. Dvořák, D. R. McCarney, N. D. Treat, N. Stingelin and T. W. Schmidt, J. Mater. Chem. C, 2015, 3, 616–622 RSC .
11. N. C. Giebink, B. W. D’Andrade, M. S. Weaver, R. B. MacKenzie, J. J. Brown, M. E. Thompson and S. R. Forrest, J. Appl. Phys., 2008, 103, 044509 CrossRef .
12. S. Reineke, K. Walzer and K. Leo, Phys. Rev. B: Condens. Matter Mater. Phys., 2008, 75, 125328 CrossRef .
13. S. Schmidbauer, A. Hohenleutner and B. König, Adv. Mater., 2013, 25, 2114–2129 CrossRef CAS PubMed .
14. T. Ogawa, N. Yanai, A. Monguzzi and N. Kimizuka, Sci. Rep., 2015, 5, 10882 CrossRef CAS PubMed .
15. M. Oldenburg, A. Turshatov, D. Busko, S. Wollgarten, M. Adams, N. Baroni, A. Welle, E. Redel, C. Wöll, B. S. Richards and I. A. Howard, Adv. Mater., 2016, 28, 8477–8482 CrossRef CAS PubMed .

---

Footnote

† Electronic supplementary information (ESI) available: Design of tandem OLEDs, emission color, filter characteristics. See DOI: 10.1039/c7tc05649d

This journal is © The Royal Society of Chemistry 2018