Semitransparent Organic Light Emitting Diodes with Bidirectionally

Jun 16, 2016 - Institute of Materials for Electronics and Energy Technology (i-MEET), ... be integrated into building facades, suggesting colorful win...
1 downloads 2 Views 1MB Size
Subscriber access provided by Weizmann Institute of Science

Article

Semitransparent Organic Light Emitting Diodes with Bi-directionally controlled Emission Carina Bronnbauer, Andres Osvet, Christoph J. Brabec, and Karen Forberich ACS Photonics, Just Accepted Manuscript • DOI: 10.1021/acsphotonics.6b00234 • Publication Date (Web): 16 Jun 2016 Downloaded from http://pubs.acs.org on June 17, 2016

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 Photonics 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 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

ACS Photonics

Semitransparent Organic Light Emitting Diodes with Bidirectionally controlled Emission Carina Bronnbauer1,2*, Andres Osvet1, Christoph J. Brabec1,2,3 and Karen Forberich1

1

Institute of Materials for Electronics and Energy Technology (i-MEET), Friedrich-Alexander-

Universität Erlangen-Nürnberg, Martensstrasse 7, 91058 Erlangen, Germany 2

Erlangen Graduate School in Advanced Optical Technologies (SAOT) Friedrich-Alexander-

Universität Erlangen-Nürnberg, Paul-Gordan-Strasse 6, 91052 Erlangen, Germany 3

Bavarian Center for Applied Energy Research (ZAE Bayern), Haberstrasse 2a, 91058

Erlangen, Germany

Authors information A. O.:[email protected]; C.J.B.:[email protected]; K.F.:[email protected];

Correspondence should be addressed to: Carina Bronnbauer Martensstr. 7, 91058 Erlangen E-Mail: [email protected]. Telefon: +49 (0)9131 85-27730 Fax: +49 (0)9131 85-28495

1 ACS Paragon Plus Environment

ACS Photonics

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 23

ABSTRACT Semitransparent OLEDs are a candidate for large area eco-friendly light sources that can be integrated into building facades, suggesting colorful windows that become luminescent if the OLED is switched on. However, since the light is emitted into two directions, smart light engineering has to be implemented to direct the light into a preferred direction and to prevent for instance huge energetic losses to the outside of a building. We introduce an unprecedented device architecture, comprised of a dielectric mirror attached to a semitransparent OLED. Such a system features a dual functionality which depends on the viewing direction: changing the color perception or/(and) enhancing the light directionality while still preserving a high overall device transparency. First, we motivate the potential of this concept with a theoretical study, showing that broad modifications in the color gamut can be realized via device optimization and that the maximum possible emission enhancement of the OLED is only limited by the transparency of the interfacial layers and the electrodes. Then, experimental investigations with a semitransparent yellow OLED (transparency = 58.2%) in combination with six different dielectric mirrors validate the theoretical results. Retaining the same color perception, up to 80% of the total emitted light can be directed towards one side while the color is modified at the other side of the device stack. Here, modifications from yellow to purple, dark or light blue can be realized.

KEYWORDS: Color modification, Dielectric mirror, Fully-printed, Light management, OLED, Semitransparent

2 ACS Paragon Plus Environment

Page 3 of 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

ACS Photonics

Organic light emitting diodes (OLEDs) show a huge potential to be the next generation light source as they are thinner, more eco-friendly with respect to energy consumption and materials, easier to fabricate due to solution-processability, not restricted to any shape, more flexible or even stretchable and provide wider viewing angles than the present lighting systems. 1–6 Furthermore and in contrast to their inorganic counterpart, OLEDs enable the production of large area flat panel lighting systems without any light distribution elements.7,8 In principle, an OLED consists of an organic electroluminescent material sandwiched between two electrodes. Additional electron- and hole transporting layers at the anode and cathode interfaces can further improve charge carrier injection and thus enhance the overall device performance. Apparently, the interfacial layers as well as at least one electrode have to be semi-transparent to guarantee light extraction. OLEDs with two semitransparent electrodes have already been reported and thus allow the realization of window integration or transparent displays. 6,9–11 While the emitted color spectrum of OLEDs is mainly a function of the band gap of the emitter materials, their brightness depends on the applied voltage and the light extraction efficiency. The bandwidth of most emitter materials is relatively narrow and does not extend over the whole visible region. However, there are two main approaches for OLEDs that help to generate broad emission spectra, including white light. 12 One technique is based on the doping of a single organic electroluminescence material with emissive additives, also known as dyes. Then, the host material non-radiatively transfers its energy to the dye, which eventually emits light.13 The other technique is based on stacking different organic electroluminescence materials. In this case, white light is generated by intermixing three primary colors or two complementary colors.14–16 Another less often used concept is the deposition of a color tunable filter on top of a white topemitting OLED.17 Such approaches are not beneficial because of considerable energy loss due to the reduction of the bandwidth of the spectrum. However, by precisely adjusting the layer thicknesses of such a wavelength selective mirror (also known as Bragg mirror or dielectric mirror) as well as the layer thicknesses of the device itself, cavity resonators can be realized that 3 ACS Paragon Plus Environment

ACS Photonics

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 23

allow enhancing the luminance and a modification of the color of emission, eventually accomplishing higher color purity. 18–22 Inspired by the dielectric mirror approach for top emitting OLEDs, we investigated the combination of such a mirror with semitransparent OLEDs, creating an unprecedented device architecture featuring a dual functionality which depends on the viewing direction: changing the color perception or/and enhancing the light extraction. This suggests a window with bidirectionally controlled emission that is semitransparent when the OLED is switched off, but emits into a preferential direction while the OLED is switched on. First, we present theoretical simulations that help to predict and thus to optimize this architecture in terms of color modifications and relative emission enhancement. Secondly, experimental investigations based on a yellow emitting semitransparent OLED with 58.2% transparency and six different dielectric mirrors validate the theoretical findings.

RESULTS AND DISCUSSION Theoretical investigation Dielectric mirrors are wavelength selective mirrors which rely on constructive or destructive interference of thin layers. They consist of a stack of alternating low and high refractive index layers all satisfying the Bragg condition, nhigh*dhigh=nlow*dlow=λmirror/4

(1)

where n is the refractive index of the high and the low refractive index material, d is the respective layer thickness and λmirror is the wavelength of maximum reflectance at normal incidence. The spectral width as well as the height of the reflection peak of λmirror depends on the refractive indices of the material, the number of deposited layers, and the angle of incidence, 4 ACS Paragon Plus Environment

Page 5 of 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

ACS Photonics

see Figure S1. The combination of a semitransparent OLEDs with a dielectric mirror offers the ability to change the color perception or/and to systematically enhance the directed emission. Figure 1 sketches the general working principle of such a device architecture, differentiating between two general cases: top view (viewer faces the OLED, normal incidence) and bottom view (viewer faces the dielectric mirror, normal incidence). For the case of top view, the emitted light is wavelength dependently reflected at the backside of the OLED and hence enhances the light extraction at the front side. The relative light enhancement mainly depends on the overlap of the reflection spectrum of the dielectric mirror and the emission spectrum of the OLED. If the reflection spectrum does not fully overlap with the emission spectrum, slight variations on the color perception will occur. If the viewer faces the dielectric mirror (bottom view), losses in luminance as well as changes in color will be observed since the dielectric mirror modifies the emission spectrum by for example narrowing the emission line. As we are considering window applications, the impact of the dielectric mirror on the transmittance and thus on the color and the transparency of the full semitransparent stack has to be taken into account for the case that the OLED is switched off. Here, the optical response is independent on the viewing direction (top or bottom view, normal incidence) if the same illumination source is used. Theoretical investigations can predict the effect of the dielectric mirror on the optical characteristics of the OLED: the change in color perception as well as the enhancement in light extraction arising from highly reflective dielectric mirrors. The relative light enhancement (L.E.) with respect to the normalized human eye sensitivity curve during daylight (V) can be estimated by  

. . =

   ∗  ∗ ∗ ∗ 

  ∗ ∗

∗ 100,

(2)

where EOLED is the normalized emission spectrum, TOLED is the transmittance of the OLED and RDM is the reflectance of the dielectric mirror. A correction factor has to be included for OLEDs 5 ACS Paragon Plus Environment

ACS Photonics

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 23

that do not emit equally in both directions. While reflectanceDM can be simulated via the transfer matrix formalism, emissionOLED was modeled by a normalized Gaussian function with a certain variance σ² and mean λOLED.23 The CIE coordinates of the full device which is illuminated by different light sources can be determined via the respective transmittance spectrum. The relative light enhancement (top view), the transparency (OLED switched off) and the color perception (bottom view and top view) for variable λmirror and constant λOLED are shown in Figure 1 and Figure S2. For this particular case, the dielectric mirror stack configuration features eleven alternating layers with refractive indices of either ~1.3 or ~1.9, σ² is set to 15 and transmittanceOLED is 70% (constant for all wavelengths). σ²=15 was chosen since this value corresponds to a Gaussian with ~100 nm full width at half maximum - a value that is similar to common electroluminescence materials that are used for OLED fabrication. The architecture as well as the refractive indices of the simulated dielectric mirrors are identical to the ones used in the experimental part of this work. For a detailed study on the impact of all parameters on the optical characteristics of the OLED, see Figure S3, S4, S5. In addition, the dielectric mirror approach was simulated for a white OLED with a constant emission ranging from 420 to 720 nm, see Figure S6. The results indicate a huge potential in terms of color tuning for a white OLED which is attached to a dielectric mirror. However, OLEDs with thinner emission characteristics can achieve at high values of transparencies, higher values in relative light enhancement compared to white OLEDs. Considering the influence of λOLED on the color modifications in bottom view (Figure 1b), it is shown that the closer the emission maximum to the human eye sensitivity curve maximum, the larger the effect. If the electroluminescence material emits at wavelengths in or close to the nonvisible region, almost no changes in color perception can be observed because the emission spectrum is influenced only where the human eye is not sensitive. Regarding the case that the OLED is switched off, the changes in transmittance caused by the attachment of the dielectric mirror, tremendously influence the overall device color and transparency. The impact on the 6 ACS Paragon Plus Environment

Page 7 of 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

ACS Photonics

color, which is presented in Figure S7 depends only on the architecture of the mirror and the transmittance of the OLED itself and not on λOLED since the OLED is switched off in this particular case. The influence on the overall device transparency is represented on the x-axis of Figure 1a. The lowest transparencies are obtained for a mirror with λmirror close to the maximum of the human eye sensitivity curve, while the highest transparencies can be achieved for λmirror located outside the visible range. However, a transparency of 70% (assumed transparency of the OLED device) can never be accomplished due to the overlap of the side minima and maxima of the interference pattern with the human eye sensitivity curve. The relative light enhancement is maximum as soon as the reflection maximum of the dielectric mirror fully overlaps with the emission spectrum of the OLED. Accordingly, for fixed transmittanceOLED, σ², refractive indices and number of layers, the maximum value of relative light enhancement is constant (here 66%). A reduction in OLED device transparency and/or a decrease in the number of layers of the mirror or a smaller refractive index contrast, respectively, would lead to a lower value and vice versa. Generally and since equal values of transparencies are achievable for λmirror either located on the left side or the right side of the maximum of the human eye sensitivity curve, there are two solutions for the relative light enhancement for a certain value of transparency. The difference between those two values is mainly a function of λOLED (compare Figure S3, S4, S5). Besides the huge gain in relative light enhancement for top view, there is almost no change of the color perception (see Figure S2) due to not shortening but only partially enhancing the emission line. As an alternative to the dielectric mirror approach, a thin semitransparent layer of silver can also serve as a reflector that will cause the light to be emitted into a preferential direction. However, with such a layer (thickness was varied between 0 and 100 nm) it is hardly possible to influence the color perception because of the almost constant reflection characteristics all over the visible region (see Figure 1, grey and blue symbols). The increase in relative light enhancement is inversely proportional to the transparency. The slope rises with increasing λOLED because of 7 ACS Paragon Plus Environment

ACS Photonics

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 23

silver showing slightly higher reflection characteristics at longer wavelength compared to shorter wavelength, leading to a better overlap of emissionOLED and reflectionAg. Although 70% of maximum relative light enhancement can be accomplished at 0% of transparency, for higher values of transparency the dielectric mirror approach offers significantly better light enhancement than a thin layer of silver.

Experimental investigation After introducing the general concept of dielectric mirrors attached to semitransparent OLEDs in theory, experiments based on this approach will be discussed and compared to simulated results in this section of the manuscript. Figure 2 summarizes all OLED characteristics that needed to be investigated before providing a detailed proof of concept. The device architecture of the OLED (see Figure 2a) which was selected for the experimental study is: AgNWs (silver nanowires, ~100 nm) / PEDOT:PSS (poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate), ~40 nm) / SY (Super Yellow, ~22 nm) / zinc oxide (ZnO, ~30 nm) / AgNWs (~100 nm). Super yellow was chosen since it is already well established and commonly used in the literature and thus serves as an appropriate material for a general study. To guarantee almost equal emission from both sides of the OLED, the same material and layer thickness were used for anode and cathode. The detailed device fabrication is outlined in the supporting information. A digital image of the fabricated OLED without bias and under applied voltage (5 V) is shown in Figure 2b. The corresponding wavelength dependent emission and transmittance spectrum of the device is displayed in Figure 2c while the luminance as a function of the applied voltage is shown in Figure 2d. As it is already known from the literature, the general emission characteristics (main emission maximum at 550 nm and side maximum at 590 nm) are independent of the super yellow thickness and the applied voltage (see Figure S9) and lead to a bright yellow emission

8 ACS Paragon Plus Environment

Page 9 of 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

ACS Photonics

color with CIE X=0.47 and CIE Y=0.53. The luminance normal to the surface in top view and bottom view are comparable. The dielectric mirror architecture was selected based on the following considerations: First, from the theoretical point of view, the higher the refractive index contrast, the broader and the higher the main reflection peak. In addition, the magnitude of the reflection peak is also a function of the number of alternating layers. Therefore high refractive index contrasts at constant numbers of alternating layers will result in higher values for relative light enhancement at any given value of transparency and broader variations in the color perception (compare Figure S4). Moreover, increasing the number of alternating layers for constant refractive indices, leads to an increase in the relative light enhancement at a certain value of transparency and offers more possibilities in the modifications of the color gamut of the full device (compare Figure S5). Concluding, for accomplishing maximum light extraction and broad variations in color, many layers of materials showing high refractive index contrasts have to be incorporated. Secondly, from an experimental point of view and since we are aiming for fully printability, refractive indices of ~1.3 or ~1.9 are the highest or lowest values achieved up to now according to literature.24 Moreover, since there is almost no change between the use of eleven and fifteen layers, but a significant changes between the use of eleven and seven layers, hence we decided in favor for the eleven layer stack (compare Figure S5). Now, combining the OLED with different dielectric mirrors (for respective reflection spectra see Figure S10), once again it has to be distinguished between top and bottom view as well as OLED switched on or off. For all experimental studies, λmirror of the dielectric mirrors that were used within this study is 550 nm, 590 nm, 640 nm, 660 nm, 680 nm or 730 nm. First, Figure 3a,c and d show the results regarding changes in color perception and transparency for the previously characterized OLED that is switched off but illuminated by the sun (for experiments a white light flat panel with 15mW cm-² was used). By the attachment of different dielectric mirrors, 9 ACS Paragon Plus Environment

ACS Photonics

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 23

the yellow color of the pristine device can be modified to purple (λmirror=550 nm), dark blue (λmirror=590 nm) or light blue (λmirror=640 nm, λmirror=660 nm, or λmirror=680 nm). For λmirror outside the visible region (λmirror=730 nm). The respective experimentally determined as well as theoretically calculated CIE coordinates are displayed in Figure 3c and listed in Table 1. In an extension of the experimental results, simulations reveal that even orange and red colors can be realized by the utilization of λmirror smaller than 500 nm. In addition to the effect on the color, the overall device transmittance and thus the transparency are affected as well. Figure 3d shows the wavelength dependent transmittance of the device architectures that were realized experimentally. The respective values of transparencies are listed in Table 1, together with simulated data. By shifting λmirror outside the visible region, transparencies increase from 7.93% for λmirror=550 nm to 40.9% for λmirror=730 nm. The good agreement between experiment and theory for the evaluation of transparency and color perception verifies that simulations are a helpful tool to predict the impact of dielectric mirrors on OLEDs that are switched off. Small mismatches between both data sets arise from non-perfect transmittance characteristics of the fabricated dielectric mirrors caused by stochastic variations in layer thicknesses due to the particulate nature of the layers.24 Furthermore, there will be also changes in color perceptions for the case that the viewer is facing the illuminated OLED from the dielectric mirror side. Figure 3b displays the digital images for this particular situation. The corresponding experimentally determined as well as theoretically calculated CIE coordinates for different λmirror are shown in Figure 3e and listed in Table 2. Accordingly, colors such as green, red and orange can be realized by the combination of a yellow emitter with different dielectric mirrors. Nevertheless, it has to be mentioned that this change in color gamut comes along with a decrease in the luminance since the width of the emission band becomes narrower. Again, there is a good agreement between theory and experiments.

10 ACS Paragon Plus Environment

Page 11 of 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

ACS Photonics

Finally, the case for top view and OLED switched on will be discussed in this paragraph. Figure 4a shows the digital image of all particular dielectric mirror combinations. The respective CIE coordinates (Table 3) reveal that there is almost no modification in color perception for all given values of λmirror. However, the luminance is highly influenced. Figure 4b displays the measured as well as theoretically determined relative light enhancement as a function of λmirror. It has its maximum value at ~55% between 550 – 600 nm where the overlap between the emission spectrum of the OLED and the reflection response of the dielectric mirror is maximum and decreases for lower or higher λmirror values. For shorter or longer wavelengths, respectively, this overlap is reduced resulting in a decrease of the relative light enhancement. Finally, the relative light enhancement is plotted as a function of the transparency (see Figure 4c). This graph excellently summarizes the general applicability of a semitransparent OLED with super yellow as electroluminescence material for the dielectric mirror approach. The transparency is almost inversely proportional to the relative light enhancement. For example, at still relatively high transparencies of ~22%, a relative light enhancement of 51% for top view is accomplished However, the maximum enhancement of 61% is achieved with λmirror=590 nm at a transparency of 11%. Once again, it is shown that all experimentally determined values can be estimated by applying the transfer matrix formalism and Formula 2.

CONCLUSION In summary we introduced an unprecedented architecture, comprised of a dielectric mirror attached to a semitransparent OLED. Such a system allows to adjust the color perception of the emitted light as well as the color of the OLED itself for the case that it is switched off. Furthermore, it offers the possibility to direct the emission of the semitransparent OLED into a preferential direction. First, we provide a detailed theoretical study taking into account different emission characteristics (width and wavelength position of main emittance) and different 11 ACS Paragon Plus Environment

ACS Photonics

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 23

dielectric mirrors (main peak of reflection at perpendicular incidence, number of layers and refractive index contrast). These investigations reveal that by smart device design, we can achieve maximum emission enhancement of the OLED and broad modifications in the color gamut while still preserving high transparencies. Then, a step by step experimental investigation with a semitransparent yellow emitting OLED (transparency = 58.2%) in combination with different dielectric mirrors is presented that validates the theoretical results. For instance, we developed an OLED that mainly emits from one side (up to 80 % of the total emitted light), while the color is only modified at the other side of the device stack. Here, modifications from yellow to purple, dark or light blue can be realized. This architecture therefore suggest a window integrated light panel that directs light of the same color gamut mainly into the interior of the building, while the color perception at the outside of the building can be extensively varied. In addition, since the color of the OLED in its non-iluminated state is modified for white light illumination from the outside, this concept becomes very interesting for architectural design in terms of dynamical color changes of building facades throughout the day. Finally, since both, the dielectric mirror and the OLED are fully printed, this concept is applicable for further upscaling.

METHODS Device fabrication and materials: The preparation methods of the dielectric mirrors are described in detail in a previous report.24 Except the top electrode, all layers of the OLED were doctor bladed (doctor blade: Zehntner ZAA 2300; applicator: Zehntner ZUA 2009) on a float glass which was previously cleaned with acetone and IPA in an ultrasonic bath for 10 min each. For more detailed information, see supporting information Figure S7. The layer thickness of SY (toluene based ink provided by Siemens) was varied while the layer thicknesses of AgNWs (Cambrios, Clear Ohm), PEDOT:PSS (CLEVIOSTM, PVP AL4083, Heraeus) and ZnO (N-10, nanograde) were always kept constant. For the encapsulation process the adhesive Katiobond 12 ACS Paragon Plus Environment

Page 13 of 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

ACS Photonics

LP655 from DELO GmbH & Co KGaA was deposited with a dispenser robot I&J 4100-LF from i&J Fisnar Inc. at the corners of the device and a glass with a thickness of 150 µm was attached as a barrier. Finally the epoxy was UV-cured (UVACUBE 100 equipped with a UV 150 F lamp, Hönle UV Technologie) for two minutes. Similar to the dielectric mirror approach for organic photovoltaic, the dielectric mirror and the OLED device were taped together, resulting in a layer sequence of Air / Glass / AgNWs / PEDOT:PSS / SY / ZnO / AgNWs /Encasulation glue / Glass / Air / high refractive index material (HRIM) / low refractive index material (LRIM) / HRIM / LRIM / HRIM / LRIM / HRIM / LRIM / HRIM / LRIM / HRIM / Glass /Air.24 Device characterization: Layer thicknesses are measured with a profilometer (Tencor Alpha Step). Transmittance and reflectance over wavelength were determined with an UV-Vis-NIR spectrometer (Lambda 950 from Perkin Elmer) equipped with an integrating sphere. Thus, in this manuscript the term transmittance or reflectance refers to total transmittance or total reflectance as diffuse scattering is included. Current density-voltage characteristics were recorded with a source measurement unit (SMU) from BoTest. The emission spectra of the devices were measured with an integrating sphere attached to an iHR-320 monochromator (Jobin-Yvon), equipped with a Si CCD detector. The spectral sensitivity of the setup was determined with the help of a calibrated halogen lamp (StellarNet), leading to a wavelength dependent calibration factor. The absolute calibration for light power was done by measuring a green LED with a known optical power (32 µW, I=20 mA). The luminance of the OLED at perpendicular incidence was measured with a Botest LIV test system, which was previously calibrated with a luminance camera (Chroma meter CS-200, Konica Minolta Sensing) .The CIE coordinates were either determined from the simulated transmittance spectrum or directly measured with the luminance camera. The transmittance of samples was converted to the “visible” transparency, T, by %&' ()  !"# ∗ $+ 

 = *&' ()

(3)

13 ACS Paragon Plus Environment

ACS Photonics

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 23

ACKNOWLEDGEMENTS The authors would like to gratefully acknowledge the funding of the Erlangen Graduate School in Advanced Optical Technologies (SAOT) at the University of Erlangen-Nürnberg which is funded by the German Research Foundation (DFG) within the framework of its “Excellence Initiative”. The work was further supported by the service and facilities of the Energie Campus Nürnberg (EnCn) and financial support through the “Aufbruch Bayern” initiative of the state of Bavaria and the Cluster of Excellence “Engineering of Advanced Materials” (EAM) at the Universität Erlangen-Nürnberg. C.B. and K.F would like to acknowledge the EU-project SOLPROCEL (“Solution Processed high performance transparent organic photovoltaic cells”, Grant no. 604506).

Supporting Information Available: The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acsphotonics.XXXXXXX.

REFERENCES (1)

Thejo Kalyani, N.; Dhoble, S. J. Organic Light Emitting Diodes: Energy Saving Lighting Technology - A Review. Renew. Sustain. Energy Rev. 2012, 16, 2696–2723.

(2)

Bagher, A. M. OLED Display Technology. Am. J. Opt. Photonics 2014, 2, 32.

(3)

Gomez, E. F.; Steckl, A. J. Improved Performance of OLEDs on Cellulose/Epoxy Substrate Using Adenine as a Hole Injection Layer. ACS Photonics 2015, 2, 439–445.

(4)

Small, D. M.; Sanchez, W. Y.; Hickey, M. J.; Gobe, G. C. Review of Recent Progress in

14 ACS Paragon Plus Environment

Page 15 of 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

ACS Photonics

Multilayer Solution Processed Organic Light Emitting Diodes. J. Biomed. Opt. 2014, 19, 020901. (5)

White, M. S.; Kaltenbrunner, M.; Głowacki, E. D.; Gutnichenko, K.; Kettlgruber, G.; Graz, I.; Aazou, S.; Ulbricht, C.; Egbe, D. A. M.; Miron, M. C.; Major, Z.; Scharber, M. C.; Sekitani, T.; Someya, T.; Bauer, S.; Sariciftci, N. S. Ultrathin, Highly Flexible and Stretchable PLEDs. Nat. Photonics 2013, 7, 811–816.

(6)

Liang, J.; Li, L.; Niu, X.; Yu, Z.; Pei, Q. Elastomeric Polymer Light-Emitting Devices and Displays. Nat. Photonics 2013, 7, 817–824.

(7)

Chang, H. W.; Meng, H. F.; Horng, S. F.; Zan, H. W. ITO-Free Large-Area Top-Emission Organic Light-Emitting Diode by Blade Coating. Synth. Met. 2016, 212, 19–24.

(8)

Choi, K.-J.; Lee, J.-Y.; Park, J.; Seo, Y.-S. Multilayer Slot-Die Coating of Large-Area Organic Light-Emitting Diodes. Org. Electron. 2015, 26, 66–74.

(9)

Oliva, J.; Papadimitratos, A.; De La Rosa, E.; Zakhidov, A. Semi-Transparent Polymer Light Emitting Diodes with Multiwall Carbon Nanotubes as Cathodes. Phys. Status Solidi A 2014, 2832, 2828–2832.

(10)

Zhang, M.; Höfle, S.; Czolk, J.; Mertens, A.; Colsmann, A. All-Solution Processed Transparent Organic Light Emitting Diodes. Nanoscale 2015, 20009–20014.

(11)

Chang, J.-H.; Lin, W.-H.; Wang, P.-C.; Taur, J.-I.; Ku, T.-A.; Chen, W.-T.; Yan, S.-J.; Wu, C.-I. Solution-Processed Transparent Blue Organic Light-Emitting Diodes with Graphene as the Top Cathode. Sci. Rep. 2015, 5, 9693.

(12)

Mukherjee, S.; Thilagar, P. Organic White-Light Emitting Materials. Dye. Pigment. 2014, 110, 2–27.

15 ACS Paragon Plus Environment

ACS Photonics

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

(13)

Page 16 of 23

Thejokalyani, N.; Dhoble, S. J. Novel Approaches for Energy Efficient Solid State Lighting by RGB Organic Light Emitting Diodes - A Review. Renew. Sustain. Energy Rev. 2014, 32, 448–467.

(14)

Fröbel, M.; Schwab, T.; Kliem, M.; Hofmann, S.; Leo, K.; Gather, M. C. Get It White: Color-Tunable AC/DC OLEDs. Light Sci. & Appl. 2015, 4, e247.

(15)

Höfle, S.; Schienle, A.; Bernhard, C.; Bruns, M.; Lemmer, U.; Colsmann, A. Solution Processed, White Emitting Tandem Organic Light-Emitting Diodes with Inverted Device Architecture. Adv. Mater. 2014, 26, 5155–5159.

(16)

Xue, K.; Chen, P.; Duan, Y.; Sheng, R.; Han, G.; Zhao, Y. Improved Color Stability of White Organic Light-Emitting Diodes without Interlayer between Red, Orange and Blue Emission Layers. Opt. Commun. 2016, 362, 59–63.

(17)

Hou, J.; Wu, J.; Xie, Z.; Wang, L. Realization of Blue, Green and Red Emission from TopEmitting White Organic Light-Emitting Diodes with Exterior Tunable Optical Films. Org. Electron. physics, Mater. Appl. 2008, 9, 959–963.

(18)

Kim, H. K.; Cho, S. H.; Oh, J. R.; Lee, Y. H.; Lee, J. H.; Lee, J. G.; Kim, S. K.; Park, Y. Il; Park, J. W.; Do, Y. R. Deep Blue, Efficient, Moderate Microcavity Organic Light-Emitting Diodes. Org. Electron. physics, Mater. Appl. 2010, 11, 137–145.

(19)

Shiga, T.; Fujikawa, H.; Taga, Y. Design of Multiwavelength Resonant Cavities for White Organic Light-Emitting Diodes. J. Appl. Phys. 2003, 93, 19–22.

(20)

Brütting, W.; Frischeisen, J.; Schmidt, T. D.; Scholz, B. J.; Mayr, C. Device Efficiency of Organic Light-Emitting Diodes: Progress by Improved Light Outcoupling. Phys. Status Solidi 2013, 210, 44–65.

(21)

Agrawal, M.; Sun, Y.; Forrest, S. R.; Peumans, P. Enhanced Outcoupling from Organic 16 ACS Paragon Plus Environment

Page 17 of 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

ACS Photonics

Light-Emitting Diodes Using Aperiodic Dielectric Mirrors. Appl. Phys. Lett. 2007, 90, 241112. (22)

Riel, H.; Karg, S.; Beierlein, T.; Rieß, W.; Neyts, K. Tuning the Emission Characteristics of Top-Emitting Organic Light-Emitting Devices by Means of a Dielectric Capping Layer: An Experimental and Theoretical Study. J. Appl. Phys. 2003, 94, 5290–5296.

(23)

Centurioni, E. Generalized Matrix Method for Calculation of Internal Light Energy Flux in Mixed Coherent and Incoherent Multilayers. Appl. Opt. 2005, 44, 7532–7539.

(24)

Bronnbauer, C.; Hornich, J.; Gasparini, N.; Guo, F.; Hartmeier, B.; Luechinger, N. A.; Pflaum, C.; Brabec, C. J.; Forberich, K. Printable Dielectric Mirrors with Easily Adjustable and Well-Defined Reflection Maxima for Semitransparent Organic Solar Cells. Adv. Opt. Mater. 2015, 3, 1424–1430.

17 ACS Paragon Plus Environment

ACS Photonics

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 23

FIGURES

Figure 1: General study on bi-directional emission control. The color coordinates (OLED on, bottom view), the transparency (OLED off, top and bottom view) and the relative light enhancement (OLED on, top view) are calculated for a stack consisting of a dielectric mirror with nhigh~1.9, nlow~1.3 and eleven alternating layers and a Gaussian emission line with σ²=15. λOLED was kept constant to either 550 nm (purple) or 650 nm (blue) while λmirror was varied between 400 nm – 800 nm.

18 ACS Paragon Plus Environment

Page 19 of 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

ACS Photonics

Figure 2: OLED device characterization. a) Architecture of the full device including the dielectric mirror. b) Digital images of the OLED with 15mW cm-² back illumination (top) and under applied voltage (5 V). c) Wavelength dependent emission and transmittance characteristics of the OLED. d) Luminance as a function of voltage measured for top and bottom view.

19 ACS Paragon Plus Environment

ACS Photonics

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 23

Figure 3 Color modifications for a yellow emitting OLED attached to dielectric mirrors. Digital images of devices for the case OLED switched off (a) and OLED switched on, bottom view (b). d) and e) respective experimentally (red) and theoretically (black) CIE coordinates. c) Transmittance spectra of OLED switched off in combination with different mirrors.

Figure 4 Influences of a dielectric mirror attached to a yellow emitting OLED in top view. a) Digital images of the full device. Experimentally (red) and theoretically (black) determined relative light enhancement as a function of the main reflection peak of the dielectric mirror (b) or transparency (c).

TABLES Table 1 CIE coordinates of different dielectric mirrors attached to an OLED that is switched off but illuminated by a white light source. Simulated data

Experimental data 20 ACS Paragon Plus Environment

Page 21 of 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

ACS Photonics

λmirror

CIE x

CIE y

Transparency [%]

CIE x

CIE y

Transparency [%]

550

0.268

0.147

6.61

0.310

0.211

7.93

590

0.176

0.219

11.0

0.201

0.264

11.1

640

0.207

0.370

27.8

0.219

0.362

21.5

660

0.245

0.393

34.7

0.247

0.389

26.1

680

0.285

0.398

40.7

0.284

0.401

33.3

730

0.351

0.390

49.3

0.342

0.394

40.9

0.374

0.382

58.2

Pristine

Table 2 CIE coordinates of different dielectric mirrors attached to an OLED that is switched on (applied voltage: 5.0V). The viewer faces the dielectric mirror (bottom view). Simulated data

Experimental data

λmirror

CIE x

CIE y

CIE x

CIE y

550

0.538

0.439

0.577

0.422

590

0.390

0.556

0.387

0.590

640

0.289

0.678

0.335

0.633

660

0.323

0.655

0.338

0.639

680

0.371

0.613

0.379

0.607

730

0.450

0.539

0.432

0.559

Table 3 CIE coordinates of different dielectric mirrors attached to an OLED that is switched on (applied voltage: 5.0V). The viewer faces the OLED (bottom view). λmirror

CIE X

CIE Y

550

0.467

0.523

590

0.422

0.577

640

0.477

0.514

660

0.480

0.511

680

0.483

0.509

730

0.462

0.521

Pristine

0.463

0.527 21

ACS Paragon Plus Environment

ACS Photonics

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 23

22 ACS Paragon Plus Environment

OLED off

With dielectric mirror (DM) Without DM Page 23 of 23 ACS Photonics

OLED on

1 2 3 4 5 6

ACS Paragon Plus Environment