Organic Light-Emitting Diode Microcavities from Transparent

LETTER pubs.acs.org/NanoLett

Organic Light-Emitting Diode Microcavities from Transparent Conducting Metal Oxide Photonic Crystals Daniel P. Puzzo,†,§ Michael G. Helander,†,§ Paul G. O’Brien,† Zhibin Wang,† Navid Soheilnia,‡ Nazir Kherani,† Zhenghong Lu,*,† and Geoffrey A. Ozin*,‡ † ‡

Department of Materials Science and Engineering, University of Toronto, 184 College Street, Toronto, Ontario, Canada M5S 3E4 Department of Chemistry, University of Toronto, 80 St. George Street, Toronto, Ontario, Canada, M5S 3H6

bS Supporting Information ABSTRACT: We report herein on the integration of novel transparent and conducting one-dimensional photonic crystals that consist of periodically alternating layers of spin-coated antimony-doped tin oxide nanoparticles and sputtered tindoped indium oxide into organic light emitting diode (OLED) microcavities. The large refractive index contrast between the layers due the porosity of the nanoparticle layer led to facile fabrication of dielectric mirrors with intense and broadband reflectivity from structures consisting of only five bilayers. Because our photonic crystals are easily amenable to large scale OLED fabrication and simultaneously selectively reflective as well as electronically conductive, such materials are ideally suited for integration into OLED microcavities. In such a device, the photonic crystal, which represents a direct drop-in replacement for typical ITO anodes, is capable of serving two necessary functions: (i) as one partially reflecting mirror of the optical microcavity; and (ii) as the anode of the diode. KEYWORDS: Organic light emitting diodes, optical microcavities, transparent and conducting oxides, photonic crystals evelopment of organic light-emitting diodes (OLEDs)1
3 for display applications represents a very active area of academic and industrial research. For a given display, it is desired that as broad a color gamut be spanned by each pixel, which can only be realized if each of the subpixel sources provide pure and saturated colors. Unfortunately, organic emitters exhibit poor color purity as they emit light with very broad spectral bandwidth owing to vibronic sidebands and strong inhomogeneous broadening of their electronic transitions. One route to overcome the problem of the poor color purity associated with such characteristic broad emission involves making use of an optical microcavity.4 By incorporating the OLED into a microcavity architecture and coupling the device emission to a cavity mode, the spectral line width of the device emission can be narrowed, and the directionality of the emission as well as the emission light intensity can be improved effectively leading to purer color generation and enhanced-efficiency LEDs.5
17 Planar microcavities,4 consist of two mirrors (either metallic or dielectric) sandwiching a medium with a thickness on the order of optical wavelengths. In an OLED microcavity, the optical medium between the two mirrors constitutes the active layers of the OLED (i.e., emitter, electron transport layer (ETL), hole transport layer (HTL), and one or more electrodes). The most common OLED microcavity architectures consist of (i) two metal mirrors made of the same metal but deposited with different thicknesses to afford one mirror that is only partially reflective and the other almost completely reflective (to reflect the Fabry
Perot nature of optical cavity); and (ii) one mirror being a dense dielectric one-

D

r 2011 American Chemical Society

dimensional photonic crystal (1D PC), also referred to as distributed Bragg reflectors (DBRs), consisting of periodic alternating layers of SiO2 and TiO2 or SiO2 and SixNy and the other mirror (which also serves as the cathode of the LED) being a low work function metal. With regards to the former, excessive use of metals are unfavored in such excitonic applications as they readily engage in energy transfer with excitons of the organic emitter which in turn decreases the quantum efficiency of the LED. With regards to the latter, that is microcavities made with dielectric mirrors, because neither SiO2 nor TiO2 nor SixNy are electrically conductive, a layer of ITO (or another suitably conductive and transparent oxide or metal) must be deposited atop the dielectric mirror to serve as the anode (hole injector) of the OLED (following ITO deposition, the remainder of the OLED structure is fabricated). Although necessary, it is preferable that the deposition of such an additional layer be avoided as it adds an additional step to the fabrication process and it modulates the Bragg reflectivity of the 1D PC. We recently reported on the development of novel transparent and conducting one-dimensional photonic crystals (TCPCs) that consisted of periodically alternating layers of spin-coated antimony-doped tin oxide (ATO) nanoparticles and sputtered tin-doped indium oxide (ITO).18 The large refractive index contrast (Figure 1c) between the layers due to the porosity of Received: November 18, 2010 Revised: January 25, 2011 Published: March 21, 2011 1457

dx.doi.org/10.1021/nl104036c | Nano Lett. 2011, 11, 1457–1462

Nano Letters

LETTER

Figure 1. (a) Reflectance spectra of a blue, green, yellow, and orange 1D TCPCs (ATO nanoparticle based) that peak at 475, 520, 565, and 625 nm. A photograph of each 1D TCPC is shown to the right of the reflectance spectra. (b) Cross-sectional SEM of TCPC consisting of periodically alternating layers of sputtered ITO and spin-coated ATO nanoparticles. (c) Refractive index dispersions for sputtered ITO (solid line) and ATO NPs (dashed line).

the ATO nanoparticle layer led to facile fabrication of dielectric mirrors with intense and broadband reflectivity from structures consisting of only five bilayers. Because our TCPCs are easily amenable to large scale OLED fabrication and simultaneously selectively reflective as well as electronically conductive, such 1D photonic materials are ideally suited for integration into OLED microcavities. In such a device, the 1D PC, which represents a direct drop-in replacement for typical ITO anodes, is capable of serving for the first time two necessary functions, (i) as one partially reflecting mirror of the optical microcavity; and (ii) as the anode of the diode. Although the reflectivity intensity and bandwidth of our previously reported TCPCs could be easily modulated, their significant surface roughness inhibited OLED integration and fabrication. The appreciable roughness was attributed to the poor film-forming properties of the ATO nanoparticle dispersion employed. Thus, for the application described herein, it was necessary to employ an alternative nanoparticle dispersion with suitable refractive index, optical transparency, electrical conductivity, and improved film forming properties. We found that the SnO2 nanoparticles (with sheet resistances of ∼105 Ω/0 following sintering at 450
C in air for 30 min, see experimental section in Supporting Information) offered by Nyacol Inc. or the ATO nanoparticles (103 Ω/0) reported by Muller et al.19 both satisfied

the aforementioned criteria and thus afforded easily processable 1D PCs (Figure 1a) capable of OLED microcavity fabrication. The central wavelength, intensity, and bandwidth of the reflectance peak of the 1D TCPCs can be tuned throughout the visible spectrum by appropriately modulating the lattice constant of the 1D PC that in turn can be modulated by manipulating the thicknesses of either of the layers (i.e., either ITO or ATO or both simultaneously). The reflectance spectra of four 1D TCPCs consisting of periodically alternating layers of ATO nanoparticles with thicknesses in the range of 100
130 nm and sputtered ITO in the range of 80
100 nm are shown in Figure 1a. The Bragg reflectance peaks of the 1D TCPCs peak at 475, 520, 565, and 625 nm and thus appear blue, green, yellow, and orange in color, respectively. A photograph of each of the four 1D TCPCs are included to the right of the reflectance spectra from which it is evident that the 1D TCPCs can be prepared with uniformity and excellent optical quality (see Supporting Information that includes optical micrographs of the surface of such TCPCs), possessing virtually no striations and only a small concentration of “comets”. In our experiences, comet formation can be almost entirely prevented by working in meticulously clean environments (i.e., a clean room) as well as by filtering dispersions at the dispensing stage of the spin-coating process through 0.22 μm pore Nylon syringe filters.20 1458

dx.doi.org/10.1021/nl104036c |Nano Lett. 2011, 11, 1457–1462

Nano Letters

LETTER

Figure 2. (a) Device structure of an OLED microcavity containing a TCPC as both the anode as well as partially reflecting mirror. (b) Molecular structures of organic molecules employed in the OLED microcavity. (c) Measurement setup employed for the analysis of the EL properties of TCPC containing OLED microcavity. (d) Green and black plots correspond to the reflectance of a green-reflecting TCPC and the microcavity containing the green-reflecting TCPC and device structure shown in (a), respectively. (e) Reflectance spectra of cavities bearing equal cavity thicknesses but fabricated from three different TCPCs, one exhibiting blue reflectance, green reflectance, and orange reflectance.

To evaluate the performance of our TCPC materials as both the partially reflecting mirror and anode of an OLED microcavity, devices were fabricated directly onto the TCPC with an overall structure of (TCPC)/poly(ethylenedioxythiophene)/ poly(styrene sulfonate) (PEDOT/PSS)/N,N-di(naphthalene1-yl)-N,N-diphenyl-benzidine (NPB)/tris(8-hydroxyquinoline) aluminum (Alq3)/LiF/Al as illustrated in Figure 2a. The molecular structures of the hole-injecting layer, PEDOT/PSS, the hole transport layer, NPB, and the emitter/electron transport layer, Alq3, are also included in Figure 2c. The use of a transparent and conducting oxide photonic crystal in this way creates now a cavity region that consists purely of organic materials. This is advantageous as it should simultaneously allow one to control the properties of the cavity by simply manipulating the thickness

of the organic layers as well as enhance the desired microcavity effect as no high refractive index conducting oxide is warranted here to serve as the anode. The reflectance spectra of a green-reflecting TCPC and a microcavity of composition (green TCPC)/PEDOT/PSS (60 nm)/NPB (70 nm)/Alq3 (45 nm)/LiF/Al correspond to the green and black plots, respectively of Figure 2d. The most notable feature of the reflectance spectrum of the TCPC is again the observance of a photonic stopband in the visible spectrum centered at 535 nm. With regards to the reflectance of the OLED microcavity containing the green reflecting TCPC, what is observed as expected is intense broadband reflectivity over a range of 450
700 nm with a narrow transmission dip with peak transmissivity of approximately 60% centered at 540 nm. The 1459

dx.doi.org/10.1021/nl104036c |Nano Lett. 2011, 11, 1457–1462

Nano Letters

LETTER

Figure 3. (a,b) (left) Normalized EL spectra from a noncavity OLED (dotted black) and green-TCPC and orange-TCPC OLED microcavities, respectively, under a 10 V bias. Note here that by orange-TCPC, the authors are not referring to the color of the electroluminescence of the OLED microcavity, which correlates with the wavelength of the cavity mode (evident from the reflectivity spectra of the microcavities provided) but rather for simplicity, the color of the reflectance of the TCPC employed to fabricate the microcavity of interest, such colors of which are evident from the photographs of the TCPCs shown in Figure 1. (a,b) (middle) Normalized EL spectrum and cavity reflectance overlaid for both microcavity structures considered herein. (a,b) (right) Photographs of OLED of green-TCPC and orange-TCPC OLED microcavities under a 10 V bias.

narrow transmission dip is a direct manifestation of the resonant cavity mode that represents a standing optical wave arising from the particular spacing between the two mirrors of the microcavity. The high transmissivity (∼60%) and small full width at half-maximum (21 nm) of the cavity mode are a testament of the quality of the fabricated optical microcavity. The peak wavelength of the resonant cavity mode is dependent on the optical thickness of the cavity region and the Bragg peak wavelength of the TCPC. Thus, to vary the position of the cavity mode, we altered solely the Bragg peak wavelength and held the optical thickness of the cavity region constant. The effect of manipulating the Bragg peak is shown in Figure 2e, which includes the reflectance spectra of cavities bearing equal cavity thicknesses but fabricated from three different TCPCs, one exhibiting blue reflectance (Bragg peak at 475 nm), green reflectance (548 nm), and orange reflectance (614 nm). The cavity modes thus arise at 498, 535, and 583 nm for the cavities bearing blue, green, and orange-reflecting TCPCs, respectively. Thus, by employing TCPCs with peak reflectivities that span the range of 494
614 nm, it is possible to achieve cavity reflectances which span a range of 498
583 nm. This range is particularly well suited for the commonly employed OLED green-emitter Alq3 as such exhibits an electroluminescence fwhm over a spectral range of ∼488
571 nm. Thus, because of the good overlap of the Alq3 emission with the cavity modes of the TCPC microcavities, the emission of Alq3 can be simultaneously narrowed and amplified over it’s entire EL range using the TCPC microcavity system.

Owing to the aforementioned, we employed both the greenand orange-reflecting TCPCs in two different OLED microcavities to gauge the optical effect of the TCPC microcavities on the EL properties of Alq3. Again, as mentioned above, it is expected that (i) the peak Alq3 electroluminescence overlap the peak transmissivity (i.e., the cavity mode) of the microcavities; (ii) that the spectral line width of the emission of the Alq3 embedded within the microcavity be significantly narrowed compared to the Alq3 EL from a noncavity-based OLED; and (iii) that the peak intensity of the EL emission from the cavity be greater than the peak intensity of the EL emission from a noncavity OLED. The observation of such phenomena is indeed indicative of a resonant effect between the cavity mode of the microcavity and the emission of the Alq3 emitter. The aforementioned can be explained by considering Fermi’s golden rule that relates the transition rate, Wiff, of a transition at a particular frequency to the photon density of states at that frequency.5 Wi f f ¼

2π Æ f jH 0 jiæ2 DðωÞ h2

where Æf|H0 |iæ is the atom-vacuum matrix element between electronic wave functions ψf and ψi, and D(ω) is the effective photon density of states (PDOS) of the electromagnetic field. Thus the microcavity effect and the Alq3 emission enhancement originate from a reorganization and an overall concentration of the PDOS in the cavity region (i.e., available to the emitter Alq3) to the cavity mode. Thus, in an ideal microcavity, as the PDOS tends to zero everywhere outside of the cavity mode and is large 1460

dx.doi.org/10.1021/nl104036c |Nano Lett. 2011, 11, 1457–1462

Nano Letters

LETTER

Figure 4. (a) I
V curve of green-TCPC containing OLED microcavity. The inset includes a measurement of brightness versus current density. (b) EL spectra of both the green-TCPC OLED microcavity and a noncavity OLED bearing identical cavity thicknesses at 10 V. (c) Demonstrations of photolithographic patterning of a blue-reflecting TCPC.

at the cavity mode, Alq3 emission can only proceed at frequencies commensurate with cavity mode frequencies. From a comparison of the EL properties of both microcavity devices bearing green- and orange-TCPCs and a noncavity-based OLED employing an anode fabricated with the same sputtered ITO employed in the TCPCs (Figure 3) and each OLED consisting of equal cavity thicknesses, it is evident that (i) listed above is satisfied as the EL maxima overlap with complete registry the cavity modes of the green- and orange-TCPC based microcavities. In addition, the fwhm of the cavity modes are nearly identical to the Alq3 cavity based emission. As shown in Figure 3, the EL spectrum of the orange-TCPC OLED microcavity, in addition to an intense narrow peak commensurate with the cavity mode, also consists of a low intensity sideband. Such EL arises at wavelengths outside of the TCPC Bragg peak and thus the microcavity is incapable of completely suppressing Alq3 emission beyond the cavity mode. In addition, statement (ii) is also observed in our TCPC microcavity systems as the fwhm of the Alq3 emission is reduced from 90 nm in the noncavity OLED to 24 and 17 nm in the green- and orange-TCPC OLED microcavities, respectively. The obtained values are comparable to previous literature reports on OLED microcavities; however, it is expected that such values can be even further reduced by reducing the roughness of the NP layers and by increasing the reflectivity of the TCPC (which can be simply achieved by increasing the number of bilayers in the TCPC structure). In addition, at an applied bias of 10 V, as shown in Figure 4a, the EL emission of the microcavity is more intense than the noncavity OLED confirming that the microcavity is indeed capable of enhancing Alq3 emission at cavity frequencies.

Additional TCPC microcavity OLED characteristics are provided in Figure 4a. The I
V plot of the OLED microcavities display obvious diode behavior. The inset of Figure 4a displays how the luminance of the green-TCPC OLED microcavity varies with current density. From such, a maximum luminance of 12 030 cd/m2 is achieved at a current density of 450 mA/cm2. For such TCPCs to be integrated into commercial OLED displays, it must be possible that the entire photonic crystal be patternable via photolithography. As a proof-of-concept, as shown in Figure 4c, we patterned a transistor pattern and lines into a blue-reflecting TCPC. The patterning proceeded by first spin-coating a commercial photoresist onto the surface of the TCPC. It is relevant to note here that the uppermost layer of the TCPC is a dense layer of ITO and thus, the resist does not penetrate the material. Following spin-coating, a suitable mask was placed over the photoresist and the coated PC exposed to UV light. The resist was then developed in toluene and the exposed regions subsequently etched with Zn powder and concentrated HCl. The etch, at least on the macroscale, was observed to proceed from the top-down, i.e., anisotropically (surprisingly owing to the significant porosity of the nanoparticle layers) by simply patterning the surface and we are currently investigating what is the lower limit of this patterning process. In conclusion, owing to their high conductivity as well as tunable optical properties, we demonstrated the integration of 1D TCPCs into OLED microcavities to simultaneously serve as the anode of the diode and as a one of the two mirrors of the optical cavity. The aforementioned represents the first example of such a capability as well as the first example of ITO-based PC integration in OLEDs. We observed an impressive cavity effect as 1461

dx.doi.org/10.1021/nl104036c |Nano Lett. 2011, 11, 1457–1462

Nano Letters the fwhm of the OLED emitting species, Alq3, was reduced to 24 and 17 nm, in green- and orange-TCPC microcavities, respectively. We are currently pursuing the incorporation of our TCPCs into other devices in the areas of lasing (potentially a good anode for an organic injection laser), electrochemistry (an electrode material), optical telecommunication, and multijunction photovoltaics for the end-purpose of efficiency enhancement.

LETTER

(17) Qiu, X. J.; Tan, X. W.; Wang Z, Z.; Y.Liu, Z. G.; Xiong, H. J. Appl. Phys. 2006, 100, No. 074503. (18) O’Brien, P. G.; Puzzo, D. P.; Chutinan, A.; Bonifacio, L. D.; Ozin, G. A.; Kherani, N. P. Adv. Mater. 2010, 22, 611. (19) Muller, V.; Rasp, M.; Stefani, G.; Ba, J.; Gunther, S.; Rathousky, J.; Niederberger, M.; Fattakhova-Rohlfing, D. Chem. Mater. 2009, 21, 5229. (20) Puzzo, D. P.; Bonifacio, L. D.; Oreopoulos, J.; Yip, C. M.; Manners, I.; Ozin., G. A. J. Mater. Chem. 2009, 19, 3500.

’ ASSOCIATED CONTENT

bS

Supporting Information. Additional information. This material is available free of charge via the Internet at http://pubs. acs.org.

’ AUTHOR INFORMATION Corresponding Author

*E-mail: (Z.L.) [email protected]; (G.A.O.) gozin@ chem.utoronto.ca. Author Contributions §

D.P.P. and M.G.H. contributed equally to this work.

’ ACKNOWLEDGMENT Z. Lu and G.A.O. are Government of Canada Research Chairs in Organic Electronics and Materials Chemistry and Nanochemistry, respectively. They are both deeply indebted to the Natural Sciences and Engineering Research Council NSERC for strong and sustained support for research. ’ REFERENCES (1) Tang, C. W.; VanSlyke, S. A. Appl. Phys. Lett. 1987, 51, 913. (2) Burroughes, J. H.; Bradley, D. D. C.; Brown, A. R.; Marks, R. N.; Mackay, K.; Friend, R. H.; Burns, P. L.; Holmes, A. B. Nature 1990, 347, 539. (3) Braun, D.; Heeger, A. J. Appl. Phys. Lett. 1991, 58, 1982. (4) Brorson, S. D.; Skovgaard, P. M. W. In Optical Processes in Microcavities; World Scientific Publishing Co. Pte.: Singapore, 1996. (5) Takada, N.; Tsutsui, T.; Saito, S. Appl. Phys. Lett. 1993, 63, 2032. (6) Wittmann, H. F.; Gruner, J.; Friend, R. H.; Spencer, G. W. C.; Moratto, S. C.; Holmes, A. B. Adv. Mater. 1995, 7, 541. (7) Huang, Q.; Walzer, K.; Pfeiffer, M.; Lyssenko, V.; He, G. F.; Leo, K. Appl. Phys. Lett. 2006, 88, No. 113515. (8) Chen, C. W.; Hsieh, P. Y.; Chiang, H. H.; Lin, C. L.; Wu, H. M.; Wu, C. C. Appl. Phys. Lett. 2003, 83, 5127. (9) Lin, C. L.; Lin, H. W.; Wu, C. C. Appl. Phys. Lett. 2005, 87, No. 021101. (10) Nakayama, T.; Itoh, Y.; Kakuta, A. Appl. Phys. Lett. 1993, 63, 594. (11) Dodabalapur, A.; Rothberg, L. J.; Miller, J. M.; Kwock, E. W. Appl. Phys. Lett. 1994, 64, 2486. (12) Dodabalapur, A.; Rothberg, L. J.; Miller, T. M. Appl. Phys. Lett. 1994, 65, 2308. (13) Tsutsui, T.; Takada, N.; Saito, S.; Ogino, E. Appl. Phys. Lett. 1994, 65, 1868. (14) Suzuki, M.; Yokoyama, H.; Brorson, S. D.; Ippen, E. P. Appl. Phys. Lett. 1991, 58, 998. (15) Osche, A.; Lemmer, U.; Deussen, M.; Feldmann, J.; Greiner, A.; Mahrt, R. F.; Bassler, H.; Feldmann, J. F.; Gobel, E. O. Mol. Cryst. Liq. Cryst. 1994, 256, 335. (16) Jeong, S. M.; Takanishi, Y.; Ishikawa, K.; Nishimura, S.; Suzaki, G.; Takezoe, H. Opt. Commun. 2007, 273, 167. 1462

dx.doi.org/10.1021/nl104036c |Nano Lett. 2011, 11, 1457–1462