<%BANNER%>

Down Conversion White Organic Light Emitting Diodes with Microcavities

Permanent Link: http://ufdc.ufl.edu/UFE0041147/00001

Material Information

Title: Down Conversion White Organic Light Emitting Diodes with Microcavities
Physical Description: 1 online resource (163 p.)
Language: english
Creator: Lee, Jae
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: down, led, microcavity, oled, organic, phosphor
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: DOWN CONVERSION WHITE ORGANIC LIGHT EMITTING DIODES WITH MICROCAVITIES There are several approaches to generate white light from organic light emitting diodes (OLEDs). One approach is to use blue emitting OLEDs in conjunction with down-conversion phosphors. This approach requires high efficiency and saturated blue emitters to provide excitation for the phosphors. For blue phosphorescent emitters such as FIrpic, the emission spectrum contains too much green light and the emitter does not excite the phosphors efficiently. In order to enhance the blue excitation for the down-conversion phosphors, we incorporated the microcavity structure with the blue OLEDs and demonstrated highly efficiency white OLEDs. We demonstrated highly efficient microcavity blue phosphorescent OLEDs (PHOLEDs) with power efficiency of 41 lm/W. With down-conversion phosphors incorporated in the blue micro-cavity PHOLEDs, a maximum luminous efficiency of 68 lm/W and a color rendering index (CRI) of 83 were achieved. With the macrolens attached to the phosphor film, a further enhancement of 46% in power efficiency was achieved. It is therefore expected that the resulting white emitting OLEDs should have a high luminous efficiency exceeding 99.3, 87 lm/W at at 30 and 100 cd/m2.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jae Lee.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: So, Franky.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041147:00001

Permanent Link: http://ufdc.ufl.edu/UFE0041147/00001

Material Information

Title: Down Conversion White Organic Light Emitting Diodes with Microcavities
Physical Description: 1 online resource (163 p.)
Language: english
Creator: Lee, Jae
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2009

Subjects

Subjects / Keywords: down, led, microcavity, oled, organic, phosphor
Materials Science and Engineering -- Dissertations, Academic -- UF
Genre: Materials Science and Engineering thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: DOWN CONVERSION WHITE ORGANIC LIGHT EMITTING DIODES WITH MICROCAVITIES There are several approaches to generate white light from organic light emitting diodes (OLEDs). One approach is to use blue emitting OLEDs in conjunction with down-conversion phosphors. This approach requires high efficiency and saturated blue emitters to provide excitation for the phosphors. For blue phosphorescent emitters such as FIrpic, the emission spectrum contains too much green light and the emitter does not excite the phosphors efficiently. In order to enhance the blue excitation for the down-conversion phosphors, we incorporated the microcavity structure with the blue OLEDs and demonstrated highly efficiency white OLEDs. We demonstrated highly efficient microcavity blue phosphorescent OLEDs (PHOLEDs) with power efficiency of 41 lm/W. With down-conversion phosphors incorporated in the blue micro-cavity PHOLEDs, a maximum luminous efficiency of 68 lm/W and a color rendering index (CRI) of 83 were achieved. With the macrolens attached to the phosphor film, a further enhancement of 46% in power efficiency was achieved. It is therefore expected that the resulting white emitting OLEDs should have a high luminous efficiency exceeding 99.3, 87 lm/W at at 30 and 100 cd/m2.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Jae Lee.
Thesis: Thesis (Ph.D.)--University of Florida, 2009.
Local: Adviser: So, Franky.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2011-12-31

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2009
System ID: UFE0041147:00001


This item has the following downloads:


Full Text

PAGE 1

1 DOWN CONVERSION WHITE ORGANIC LIGHT EMITTING DIODES WITH MICROCAVITIES By JAEWON LEE A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2009

PAGE 2

2 2009 Jaewon Lee

PAGE 3

3 To my family

PAGE 4

4 ACKNOWLEDGMENTS I have learned that I m nothing without others My entire achievements in this Ph D dissertation are based on the direct or indirect help of many people surround me as it was whether I noticed or not When I met Dr. Franky So as his first student in 2005, I was complete ly ignorant about organic electronics I am indebted in hi s earnes t guidance though 4 years of my doctoral period. I have learned not only the knowledge I have accumulated here but the insight which have made my life ripen I am grateful for all group members especially Dr. Doyoung Kim, Neetu C hopra, and Galileo Sarasq ueta. As setup members in Dr.So s group, we have always shared the very useful academic discussion and tried to make our group enjoyable. Dr. Kaushik Roy Choudhury, Dr. Subbiah Jegadesan, Cephas Small, Dongwoo Song, Michael Hartel, Song Chen, Pieter De Som er, Verena Giese, and Daniel S. Duncan, I was very happy to enjoy with you all. Y ou will be always in my mind as warm and shiny memory like Florida weather. I would like to thank for full guidance of Dr. Jiangeng Xue, Dr. Steaphen Pearton, Dr. David Norton Dr. Paul Holloway and Dr. John Reynolds to my dissertation. Thanks for all Korean friends in UF From the first da y to last da y in Florida, Dr. Seung Young Son Dr. Tak Geon Oh, Dr. Do Won Jung, Dr. Tae Gon Kim, Chan Woo Lee, Dr. Wan Tae Lim, Dr. Junghun Jang, Dong Jo Oh, Se Yeon Jung Dong Hwa Lee, and Moon Hee Kang, without them my life in Gainesville would be much boring. M y parents made all of this possible. Without their support I would never have gotten all opportunities throughout my life. I m st ill learning from their honest and plain life. Their true life itself has been the greatest lesson to me over 30 years

PAGE 5

5 Finally, I like to thank my wife, Jong A, for her continuous support and love. She makes my life plentiful After the first day I met m y wife, her presence beside me release d my soul from suffering by loneliness. The memory I got with my wife here is as priceless as the knowledge I learned here. I like to thank for the f inancial support for my dissertation provided by the Department of Energy, solid state lighting program.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LI ST OF FIGURE S ................................ ................................ ................................ ........ 10 LIST OF ABBREVIATIONS ................................ ................................ ........................... 15 ABSTRACT ................................ ................................ ................................ ................... 18 CHAPTER 1 INTROD UCTION ................................ ................................ ................................ ........ 19 1.1 The State of the Art of Organic Light Emitting D iodes ................................ ....... 19 1.1.1 The Need for Efficient White Light Illumination ................................ ........ 20 1.1.2 Opportunity for W hite O rganic L ight E mitting D iodes .............................. 20 1.1.3 The Status of E fficient W OLEDs ................................ ............................. 21 1.2 Physics of O rganic L ight E mitting D iodes ................................ ......................... 23 1.2.1. OLED Basics: Electrical P roperties ................................ ........................ 23 1.2.1.1 Orga nic vs. I norganic S emiconductor M aterials ............................. 23 1.2.1.2 Charge Injection ................................ ................................ ............. 24 1.2.1.3 Charge Transport ................................ ................................ ........... 26 1.2.1.4. Fluorescence and Phosphorescence ................................ ............ 28 1.2.1.5 Excitons ................................ ................................ ......................... 29 1.2.1.6 Excitonic Energy Transfer ................................ .............................. 29 1.2.1.7 The Definition of Brightness and Efficiency ................................ .... 30 1.2.2 OLED Basics : Optical P roperties ................................ ............................ 32 1.2.2.1 Ref lection, Ref raction, and A bsorption in O rganic M aterials .......... 33 1.2.2.2 Fabry Perot I nterference ................................ ................................ 36 1.2.2.3 Lambertian emission ................................ ................................ ...... 37 1.2.2.4 Light O ut coupling E fficiency (Light extraction efficiency) .............. 38 1.2.2.5 P hotopic and Scotop ic R esponse ................................ .................. 40 1.2.2.6 The CIE ................................ ................................ .......................... 41 1.2.2.7 The CRI ant CCT ................................ ................................ ........... 42 1.3 Disse rtation O utline ................................ ................................ ........................... 43 2 HIGH LY EFFICIENT BLUE PHOSPHORESCENT OLEDS ................................ ....... 62 2.1 H ighly Efficient Blue P hosphorescent OLEDs ................................ ................... 62 2.1.1 Effect of Triplet E nergy of T ransport L ayers for T riplet E xciton C onfinement ................................ ................................ ................................ .. 64 2.1.2 Host Materials ................................ ................................ ......................... 66

PAGE 7

7 2.1.3 Effect of C harge B alance ................................ ................................ ......... 68 2.1.4 Demonstration of H igh ly E fficient B lue PHOLEDs ................................ ... 71 2.2 Conclusion ................................ ................................ ................................ ........ 72 3 WEAK MICROCA V I T Y EFFECTS ON OLEDs ................................ ........................... 82 3.1 Introduction ................................ ................................ ................................ ....... 82 3.2 Experi ment ................................ ................................ ................................ ........ 82 3.3 Results and Discussions ................................ ................................ ................... 83 3.3.1 Weak M icrocavity E ffects o n t he S pontaneous E mission of OLEDs ....... 83 3.3.2 Weak M icrocavity E ffects on the L ight E xtraction E fficiency of OLEDs ... 84 3.3.3 Weak M icrocavity E ffects on A ngle D ependent E mission P attern of OLEDs ................................ ................................ ................................ ........... 85 3.4 Conclusion ................................ ................................ ................................ ........ 87 4 STRONG MICROCAVITY EFFECTS ON OLEDS ................................ ..................... 93 4.1 I ntr oduction ................................ ................................ ................................ ....... 93 4.2 Experiment ................................ ................................ ................................ ........ 95 4.3 Results and D iscussion ................................ ................................ ..................... 97 4.3.1 Stro ng M icrocavity E ffects on E fficiency of OLEDs ................................ 97 4.3. 2 Strong M icrocavity E ffects on E mission Profile of OLEDs ....................... 99 4.3.3 Strong M icro cavity E ffects on L ight E xtraction E fficiency of OLEDs ...... 100 4.3.4 Analysis of C avity M odes ................................ ................................ ...... 101 4.4 Conclusion ................................ ................................ ................................ ...... 104 5 DOWN CONVERSION WHITE OLED ................................ ................................ ...... 117 5.1 Introduction ................................ ................................ ................................ ..... 117 5.2 Concept of D own C onversion W hit e L ight I llumination with B lue M icrocavity OLEDs ................................ ................................ ................................ ............... 119 5.3 Experiment and discussion ................................ ................................ ............. 120 5.3.1 Down C onversion P hosphor Films ................................ ........................ 120 5.3.2 Fabrication of D own C onversion W hite OLEDs ................................ ..... 122 5.3.3 High E fficienc ies of D own C onversion W hite OLEDs ............................ 123 5.3.4 The CIE and CRI of D own C onversion WOLED ................................ .... 125 5.3.5 L ight E xtraction E fficiency on D own C onversion W OLED ..................... 126 5.4 Conclusion ................................ ................................ ................................ ...... 128 6 C ONCLUSION S AND F UTURE WORK ................................ ................................ ... 145 6.1 Conclusion s ................................ ................................ ................................ .... 145 6.2 Future Work ................................ ................................ ................................ .... 147 6.2.1 Challenges for White OLEDs ................................ ................................ 147 6.2.2 Future Work ................................ ................................ ........................... 148 APPENDIX A ................................ ................................ ................................ ............... 152

PAGE 8

8 LIST OF REFERENCES ................................ ................................ ............................. 154 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 163

PAGE 9

9 LIS T OF TABLES Table page 1 1 Review of recent w hite OLEDs ................................ ................................ ........... 44 1 2 Organic semiconductor vs. inorganic semiconductor materials. Modified from R ef. [38] ................................ ................................ ................................ .............. 45 1 3 Extraction efficiencies for emitter in a bare situation (top) or modified by a mirror. For an emitter lying close to a mirror (bottom), there is a negligible phas e change for rays emitted between normal incidence and critical angle, Modified from Ref. [39] ................................ ................................ ....................... 46 1 4 Schemes of enhancing the light extraction efficiency ................................ ......... 47 1 5 CIE, CCT, CRI and power efficiency of the representative white light sources from [4, 42] ................................ ................................ ................................ ......... 48 2 1 Energy levels, triplet energy and mobility parameters for d ifferent electron transport materials used in this study Mobility is the hole mobilities for NPD, TPD, and TAPC and the electron mobilities for BCP, BPhen, and 3TPYMB. ..... 74 4 1 Summary of the max imum current efficiency, luminous power efficiency, external quantum efficiency and CIE coordinate of Noncavity, 2QWS, and 4QWS devices. ................................ ................................ ................................ 106 5 1 Weight of components in different down conversio n thin film samples ............. 129 5 2 Summary of current and power efficiencies for down conversion of noncavity and 4QWS microcavity OLEDs at 30, 100, and 1000 cd/m 2 CIE and CRI are also given here. ................................ ................................ ................................ 130

PAGE 10

10 LIST OF FIGURE S Figure page 1 1 Display Technology (b). ................................ ................................ ...................... 49 1 2 Three main approaches for generating white light in OLEDs. (a) Multiple doping in emissive layer, (b) Multiple emissive layers, (c) down conversion with blue OLED ................................ ................................ ................................ ... 49 1 3 Recent progress es of white OLEDs. All power efficiency is without light out coupling scheme. ................................ ................................ ................................ 50 1 4 Metal semiconductor interface (a) before and (b) after contact for ideal ca se. Metal semiconductor interface (c) with interface dipole layer ............................ 51 1 5 Energy level diagram illustrating the energy level splitting between singlet and triplet states arising from the exchange interaction K. ................................ 51 1 6 Schematic representations of three types of excitons in a solid: (a) a Frenkel exciton, localized on a single molecule; (b) a charge transfer exciton, slightly delocalized over two or several adjacent molecules; and (c) a Wannier Mott exciton, which is highly delocalized with a radius much greater than the lattice constant Adapted from Ref. [20] ................................ ............................. 52 1 7 Specular, spread, and diffuse reflections from a surface. ................................ ... 53 1 8(a) Refractive index (n) of ITO, NPD, Alq 3 PEDOT:PSS, and SiO 2 Adapted from the data base of Setfos software [47]. ................................ ................................ 54 1 8(b) Extinction coefficient (k) of ITO, NPD, Alq3, PEDOT:PSS, and SiO 2. Adapted from data base of Setfos software [47] ................................ ............................... 54 1 9 me of light is reflected and some of light is refracted at the boundary between two mediums. ................................ .............. 55 1 10(a) Transmission of a light wave with electric field amplitude E 0 through a Fabry Perot cavity ................................ ................................ .............................. 56 1 10(b) Schematic illustration of allowed and disallowed optical modes in a Fabry Perot cavity consisting of two coplanar reflectors. ................................ .............. 56 1 11 Light emtting diodes with (a) planar, (b) hemispherical, and (c) parabolic surfaces. (d) Far field patterns of the different types of LEDs. At an ang le of Adapted from [23]. ................................ ....................... 57

PAGE 11

11 1 12 (a) Definition of the escape cone by the critical angle c (b) Area element dA. (c) Area of calotte shaped section of the sphere defined by radius r and angle c Adapted from [23]. ................................ ................................ ............... 58 1 13 3 optical modes in OLEDs. Extracted, substrate guided, and ITO/organic modes ................................ ................................ ................................ ................. 58 1 14 photopic vision regime. Also shown is the eye sensitivity function for the Adapted from [23]. ................................ ................................ ................................ ............ 59 1 15 CIE (1931) and CIE (1978) xyz color matching functions (CMFs). The y CMF the currently valid official standard. Adapted f rom [23]. ................................ ...... 60 1 16 CIE 1931 (x, y) chromaticity diagram. Mono chromatic colors are located on the perimeter. Color saturation decreases towards the center of the diagram. White light is located in the center. Also shown are the regions of distinct colors. The equal energy point is loc a ted at the center and has the coordinates (x, y) =(1/3, 1/3). Adapted from [23]. ................................ ............... 61 2 1 Device structures of blue PHOLEDs for studying effect of triplet energy confinement from HTL ................................ ................................ ........................ 75 2 2 Current efficiency and power efficiency of blue PHOLEDs by changing HTLs ... 75 2 3 Device structure used for studying effect of different electron transport layers (ET L s). 76 2 4 Current efficienc ies of blue PHOLEDs with different electron transpor t layers. .. 76 2 5 Current efficiency of blue PHOLEDs fabricated with different host materials. .... 77 2 6 Device structures for single carr ier devices used in this study ( a) Hole only device structure ( b) Electron only device structure. ................................ ............ 78 2 7 Current density voltage (J V) characteristics for single carrier devices shown in Figure 2 6 ................................ ................................ ................................ ..... 78 2 8 Devices fabricated for probing the recombination zone. a) Device doped on the entire emitting layer, b) Device doped only on interface of HTL/EML (left doped), c ) Device d oped only on interface of EML/ETL (right doped). ............... 79 2 9 ( a) LIV characteristics for devices in Figure 2 8 and control device. The filled symbols indicate the current density whereas open symbols indicate lu minance of the corresponding device. ( b) Current efficiency for these devices. ................................ ................................ ................................ .............. 80

PAGE 12

12 2 10 (a) T he current efficiency and luminous efficacy for highly efficient blue PHOLEDs with LiF or CsCO 3 layer a s a electron injection layer (b)LIV characteristics these two devices. Turn on voltage is reduced significantly. ...... 81 3 1 Device structure and measurement of substrate mode. The distance of the re combination zone from the cathode, L rec is also shown. ................................ 88 3 2 (a) Measured spectra by changing Alq3 thickness, (b) Simulation spectra by changing Alq3 thickness, (c) normalized spectra of (a), ( d) normalized spectra of (b) ................................ ................................ ................................ ..... 89 3 3 The ratio of the substrate guided mode to the sum of substrate and external modes ( I gel / I ext ) and substrate extraction efficiency versus Alq 3 buffer thickne ss. ................................ ................................ ................................ ........... 90 3 4 EL intensity as a function of angle for different Alq 3 buffer layer thicknesses .... 91 3 5 Simulated polar plot of OLED emis sion patterns for devices with different Alq 3 buffer layer thickne 1 2 are critical angles at the air/glass and glass/ITO interfaces, respectively. The emission pattern for a Lambertian emitter is also shown for comparison. ................................ ................................ 92 3 6 Simulated polar pl ots of OLED emission patterns for three different wavelengths and two different Alq 3 1 2 are critical angles of air/glass, glass/ITO, respectively. ................................ ................................ ... 92 4 1 Structure of blue Microcavity OLED and its 2, 4 quarter wave stacks .............. 107 4 2 Measured reflectivities of 2, 4 layers of quarter wave stacks with 50nm thick ITO and aluminum. Most of FIrpic emission is from 470 to 5 00nm. .................. 108 4 3 (a) The power efficiency and external quantum efficiency of noncavity, 2QWS, and 4QWS blue OLEDs with mCP host. EQE of 2QWS and 4QWS are calculated from measured angular dependent l ight intensity. ........................... 109 4 3 (b) Measured spectrum of noncavity, 2QWS, and 4QWS blue OLEDs at the 0.4 mA/cm 2 ................................ ................................ ................................ ............. 109 4 4 Measured angular depe ndent light intensity (the number of photons) of noncavity, 2QWS, and 4QWS blue OLEDs ................................ ...................... 110 4 5 (a) Current and power effic i ency of noncavity, 2QWS, and 4QWS blue OLEDs with UGH2 host. ................................ ................................ ............................... 111 4 5 (b) Measured spectr a of noncavity, 2QWS, and 4QWS blue OLEDs at the 0.4mA/cm 2 ................................ ................................ ................................ ........ 111

PAGE 13

13 4 6 (a) Measured angular dependent light intensity and (b) those measured spectrum at the normal direction (b) by changing the thickness of HTL from 40 to 45, 50nm on 2QWS PHOLED. Emission patterns of noncavity device and lambertian are shown together for the comparison. ................................ ... 112 4 7 (a) Measured angular dependent light intensity and (b) those measured spectrum at the normal direction (b) by changing the thickness of HTL from 45 to 50, 60nm on 4QWS PHOLED. Emission patterns of noncavity device and lambertian are shown together for the comparison. ................................ ... 112 4 8 The ratio of the substrate guided mode to the sum of substrate and extracted modes (I gel /I air ) versus HTL (TAPC) thickness for each noncavity, 2QWS and 4QWS devices. Peak of the 1 st resonant cavity wavelength is also shown in bottom axis of Figure Inset of Figure is measured and normalized spectrum of 4QWS with 4 different HTL thicknesses from 32.5 to 56nm. Normalized spectrum of 4QWS is also shown in the inset for comparison. ......................... 113 4 9 Schematic diagram of mode distribution of (a) noncavity, (b) 4QWS, and (c) 6QWS 1 2 are critical angles of TIR for substrate guided mode and ITO/organic mode. 1 st 2 nd and 3 rd cavity mode generated by strong cavity effect of 6QWS device is shown in Figure (c). Free space emission pattern escaping to air (extracted mo de) for each 3 devices are shown in upper diagram. Color changing in Figure (b) is to illustrate color spectrum dependence on the optical cavity length, which is correspondent to the outgoing angle of emitted light. ................................ ................................ ... 114 4 10 Schematic diagram of cavity mode distribution in 4QWS device with 3 different HTL thicknesses of 60 (a), 50 (b), and 45nm (c). ............................... 115 4 11 Measured angular dependent spectra (a) and its normalized spectra (b) of 4QWS device. A small fraction of the 2 nd cavity mode is observed at the glancing angle. ................................ ................................ ................................ 115 4 12 Unicad simulation of angular intensity at free spa ce (a) and at the oranic medium (b) for noncavity and 4QWS devices. ................................ .................. 116 5 1 Schematic diagram of down conversion W OLED with microcavity structure. ... 131 5 2 Quantum yield of yellow phosphor and normalized spectra of noncavity, 2QWS, 4QWS microcavity PHOLED. ................................ ............................... 132 5 3 (a) Photoluminescence and photoluminescence excitation spectra from yellow phosphor and (b) red phosphor ................................ ................................ ........ 132 5 5 (a) SEM micrograph of cross section of Y30S150 thin film sample (scale bar resolution SEM micrographs of phosphor particles incorporated in the ................................ ................................ ....... 134

PAGE 14

14 5 6 Device structure of blue microcavity PHOLED with down conversion phosphor. The phosphor film and m acrolens were attached with index matching gel. ................................ ................................ ................................ .... 135 5 7 (b) Normalized spectra of noncavity, 2QWS, and 4QWS PHOLEDs before applying phosphor films ................................ ................................ ................... 136 5 8 Current and power efficiency of blue noncavity PHOLED and its dow n converted efficiencies. Efficiency enhancement with macrolens is also shown here. Open symbols are power efficiency. ................................ ........................ 137 5 9 Current and power efficiency of blue 2QWS PHOLED and its down conve rted efficiencies. Efficiency enhancement with macrolens is also shown here. O pen symbols are power efficiency. ................................ ................................ 138 5 10 Current and power efficiency of blue 4QWS PHOLED and its down converted e fficiencies. Efficiency enhancement with macrolens is also shown here. Open symbols are power efficiency. ................................ ................................ 139 5 1 1 Normalized spectra of down conversion white OLEDs. 3 different thicknesses of yellow phosphor films and yellow:red phosphor mixture films are applied (a) noncavity with phosphor films, (b) 2QWS with phosphor films, (c) 4QWS with phosphor films ................................ ................................ .......... 140 5 1 2 CIE coordinate of noncavity device with incorporating to phosphor films (a), 4QWS device with incorporating to phosphor films (b). ................................ .... 141 5 1 3 Quantum yield of yellow phosphor and normalized spectrum of red emitting PQIr device ................................ ................................ ................................ ....... 1 42 5 1 4 Measured r elative photocurrent of photodiode for PQIr, noncavity, 2QWs, and 4QWS devices. Photocurrent without phosphor film is set as 1 for comparison. ................................ ................................ ................................ ...... 143 5 15 Measurement of the substrate guided mode. See Chapter 3 and 4. ................ 144 5 16 I gel /I air ratios for PQIr, blue noncavity, 2QWS, and 4QWS devices. .................. 144 6 1 Photo taken of (a) blue FIrpic doped PHOLED, (b) blue microcavity PHOLED with 4QWS, and (c) down conversion white OLED. ................................ .......... 151

PAGE 15

15 LIST OF ABBREVIATION S Alq 3 tri s(8 quinolinolato)aluminum BCP 2,9 dimethyl 4,7 diphenyl 1,10 phenanthroline BPhen 4,7 diphenyl 1,10 phenanthroline CBP 4, 4' Bis(carbazol 9 yl)biphenyl CDBP 4,4' Bis(9 carbazolyl) 2,2' dimethyl biphenyl CIE CRI Col or Rendering Index CT C harge T ransfer CsCO 3 Cesium Carbonate CuPc copper phthalocyanine FIrpic Iridium (III)bis [(4,6 di fluorophenyl) pyridinato N,C2] picolinate F i r6 difluorophenylpyridinato)tetrakis (1 pyrazolyl) borate EL Electro luminescence EML Emit ting Layer ETL Electron Transporting Layer

PAGE 16

16 EQE External Quantum Efficiency HBL Hole Blocking Layer HOMO H ighest O ccupied M olecular O rbital HTL Hole Transporting Layer LCD Liquid Crystal Display LED Light Emitting Diode LiF Lithium Fluoride LUMO L owest U noc cupied M olecular O rbital mCP dicarbazolyl 3, 5 benzene NPD NPD ) N, N' Bis(naphthalen 1 yl) N,N' bis(phenyl) 2,2' dimethylbenzidine OLED Organic Light Emitting Diode PEDOT:PSS poly(3,4 ethylenedioxythiophene) polystyrenesulfonic acid PL Photoluminescence PHOLED Phosphorescent Organic Light Emitting Diode PLED P olymer Light Emitting Diode PQIr I ridium ( III ) bis ( 2 phenylquinoly N C 2 ) dipivaloylmethane

PAGE 17

17 TAPC 1,1 bis[(di 4 tolylamino) phenyl]cyclohexane TPD 4,4 bis[N (p tolyl) N phenyl amino]biphenyl TIR Total Internal Reflection T 1 Triplet Energy UGH2 p bis(triphen ylsilyly)benzene WOLED White Organic Light Emitting Diode 3TPYMB 4,7 diphenyl 1,10 phenanthroline (BPhen) and tris[3 (3 pyridyl) mesityl]borane QWS Quarter Wave Stack

PAGE 18

18 Abstract of Dissertation Presented to the Graduate School of the University of Flor ida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy DOWN CONVERSION WHITE ORGANIC LIGHT EMITTING DIODES WITH MICROCAVITIES By Jaewon Lee December 2009 Chair: Franky So Major: Materials Science and Engine ering There are several approaches to generate white light from organic light emitting diodes (OLEDs). One approach is to use blue emitting OLEDs in conjunction with down conversion phosphors. This approach requires high efficiency and saturated blue emit ters to provide excitation for the phosphors. For blue phosphorescent emitters such as FIrpic, the emission spectrum contains too much green light and the emitter does not excite the phosphors efficiently. In order to enhance the blue excitation for the do wn conversion phosphors, we incorporated the microcavity structure with the blue OLEDs and demonstrated highly efficiency white OLEDs. We d emonstrated highly efficient microcavity blue phosphorescent OLEDs ( PHOLED s) with power efficiency of 41 lm/W. With d own conversion phosphors incorporated in the blue micro cavity PHOLEDs, a maximum luminous efficiency of 68 lm/W and a color rendering index (CRI) of 83 were achieved. W ith the macrolens attached to the phosphor film, a further enhancement of 46 % in power efficiency was achieved It is therefore expected that the resulting white emitting OLEDs should have a high luminous efficiency exceeding 9 9.3, 87 lm/W at at 30 and 100 cd/m 2

PAGE 19

19 CHAPTER 1 INTRODUCTION The purpose of this introductory Chapter is to provide the readers an introduction to the field of organic light emitting diodes and their fundamentals First of all, the status of OLEDs for applications of display and solid state lighting is presented. Electrical and optical fundamental properties are also g iven to help the understanding of microcavity OLED in later Chapter for readers. 1.1 The S tate of the A rt of O rganic L ight E mitting D iodes Organic electroluminescence using small molecule was firstly reported by Pope et al. in 1963 [1] and polymer organic electroluminescence was introduced by Partridge in 1983 [2] The invention of efficient organic electroluminescent diodes based on hetero junction structure by Tang et al [3] has initiated the tremendous strides of researches and developments for ef ficient and long lifetime OLEDs Nowadays, numerous OLED s for full color display applications by Sony, Samsung [4] Kodak, and LG so on have been commercialized out in the small scale mobile display and the large scale TV display (up to 11 inch) market. OLED technology has already matured enough for the mass production of OLED display products in Figure 1 1 (a). The only left issue seems to find the development of the cost effective manufacturing processes of large scale display However, the dev elopment of this low Dupont, and Cambridge display technology, are developing the wet fabrications of polymer OLED based on the roll to roll and inkjet printing processes [5 7] Cambridge display technology Co ltd. demonstrated 14 inch OLED TV using inkjet printing technique in Figure 1 1 (b).

PAGE 20

20 In the late 2000s, power efficiencies of white OLEDs have overwhelmed those of incandescent bulb (~15 lm/W) using harvesting of phosphorescence from OLEDs [8] In the very recent, power efficiencies of white OLEDs (90 lm/W at 1000 cd/m 2 ) [9] with light extraction even higher than that of fluorescent lamps (70~80 lm/W) were reported. Now white OLEDs are one of major optio ns for ultimate white illumination source White OLEDs will be reviewed in the section 1.1.3 1.1.1 The N eed for E fficient W hite L ight I llumination Homes and commercial buildings use energy for lighting. The average home spends 25 percent of its electric bill for lighting. Schools, stores, and businesses use about 60 percent of their electricity for lighting. In these ways, the United States consumes ~765 TWh of electricity a year for only lighting, which corresponds to 18% of total building energy consump tion. The cost for consumers to light their homes, offices, streets, and factories amounts to almost $58 billion a year [10] A California lawmaker is low efficient incandescent lightbulbs as part of California's groundbreaking initiatives to reduce energy use and greenhouse gases blamed for global warming [11] These big movements to save ener gy give the highest importance on the alternative efficient lighting sources for tremendous energy and cost savings. 1.1.2 Opportunity for W hite O rganic L ight E mitting D iodes Initially, WOLEDs are of interest due to its ultimate display instead of the liq uid crystal display (LCD), however, high performance (over 25 lm/W) of WOLED achieved by D Adrade [12] in 2004 have opened the opportunity of WOLEDs in the applications for general lighting field. There are many intrinsic and unique advantages for WOLEDs to be an alternative lighting source.

PAGE 21

21 One is that power efficiencies of recent OLEDs achieved or are approaching to 100 lm/W at 1000 cd/m 2 which is the critical requirement to be an alternative source of the general white light illumination. In second, the WOLED is environmental friendly comparing to other white light source s such as mercury containing fluorescent lamps and many toxic chemicals used inorganic LED s in their fabrication process. In third, the intrinsic softness of organic materials gives a big potential for flexible applications In contrast to point or line li ght sources like bulbs, LED, and tube lamp the WOLED is the first planar light source and it is easy to scale down and up (from microns to centimeters). The use of flexible substrate also makes bendable or foldable light sources possible. Transparent WOLE D can be applied on the building and car windows. T he WOLED doesn t require ugly luminaries such metal box, aluminum tube, baffle, and sockets. T he WOLED is the only pure and honest light source [13] In fourth, the WOLE D is easily tunable. CIE and CRI of the WOLED can be controllable by the combination of light emitting materials which will broaden its applicability from specialty to commodity lighting In fifth, the WOLED ha s big potential to their low cost manufacturi ng process. Without use of high vacuum chamber, it is possible to fabricate the WOLED using inkjet printing or roll to roll process. 1.1.3 The Status of E fficient W OLEDs There are several ways to make an efficient high quality white light source with CIE c oordinates similar to that of a blackbody radiator with a correlated color temperature between 2500 K and 6000 K, and a color rendering index above 80. Three particularly sophisticated WOLED structures are shown in Fig. 1 2. Typically several emissive mate rials are combined in organic material sandwiched between two electrodes. The device in Fig. 1 2(a) uses multiple dopants into emissive layer. After

PAGE 22

22 D Andrade demonstrated 26 lm/W WOLED in 2004 [12] there were several groups to use this approach. The device in Fig. 1 2(b) uses multiple emissive layers. This approach is nowad ays widely used in lighting industry such as Kodac, Philips and Universal Display Corporation The white light emitting device in Fig. 1 2(c) uses down conversion phosphor s on the top of blue emitting OLED. General electric and O SRAM have done with this a pproach [14, 15] Here, I summarize the external quantum efficiency power efficiency, CIE and CRI of significant WOLEDs in the late 2000s in Table 1 1. 4 different types of WOLED s are categorized here. All phosphorescen t WOLED s including multiple doping and multiple emissive layers are shown first In r ecent all phosphorescent WOLEDs show the highest power efficiency (more or less than 50 lm/W) depending on CRI, which corresponds to whether two or three color of phosphorescent dopants ar e used. Even though all phosphorescent WOLED has the highest efficiency, it lifetime is not enough for the commercialization due to the unstable blue phosphor until now. The hybrid WOLED with fluorescent blue and phosphorescent green red was introduced as an alternative choice of long lasting device by sacrificing its efficiency. All fluorescen ce doped WOLED have been studied by the benefit of its very long lifetime for display application. Down conversion WOLED is not studied much since efficient blue OLED didn t exist until very recent days In this dissertation we used this approach and power efficiency of 68 lm/W is achieved, which is 2.7 times higher than previous record. All these efficiency data are without additional light extraction scheme. A progre ss of WOLED efficiency in Table 1 1 is re plotted in Fig. 1 3. well summarized the status of the WOLED before 200 4 [8]

PAGE 23

23 1.2 Physics of O rganic L ight E mitting D iodes There are two fundamental aspects in physics of OLEDs. One is electrical properties, which decided by intrinsic band gap, mobi lity, interface, and carrier density of organic materials. The other is optical properties, which determined by intrinsic refractive index, absorption, reflection, and interference in organic devices. When excitons generates by electrons and holes it s un clear to separate two properties In this section, electric and optical properties of organic materials will be discussed. 1.2.1. OLED Basics: Electrical P roperties 1.2.1.1 Organic vs. I norganic S emiconductor M aterials The fundamental difference between or ganic and inorganic semiconductor materials is the nature of bonding. Table 1 2 summarizes the major difference between organic and inorganic semiconductor materials. Organic semiconductor materials are van der Waals bonded solids, which its intermolecula r bonding is considerably weaker than covalently bonded inorganic semiconductor materials. The most important consequence of different bonding nature is much stronger localization of charge carriers and excitons in organic semiconductor materials As a res ult of strong localization tendency of organic semiconductors, their carrier mobilities are much lower than inorganic materials. It is also difficult to generate free charges in organic semiconductors by intentionally doping with impurity atoms or molecule s due to localized and stable organic molecules. The weak intermolecular bonding of organic semiconducting materials also weakens mechanical strength and lowers thermal melting or glass transition temperature of organic materials. This softness nature of o rganic materials gives flexible features of organic electronic devices comparing to

PAGE 24

24 inorganic semiconductor materials Organic materials are also easy to modify their electrical and optical properties by tailoring molecule structure. Organic semiconductor materials have superior advantages in their process over their inorganic semiconductor materials. Organic materials are compatible with low cost and large area manufacturing processes. Vacuum thermal evaporation (VTE) is usually used to deposit small mole cular organic thin films. Polymer semiconductors can be processed by wet procession without high vacuum environment such as roll to roll, casting, spin coating, and ink jet printing. These processes don t require high temperature like most inorganic materi als. Some of these deposition techniques (such as VTE and ink jet printing) are compatible with large scale applications, giving the potential to low cost industrial manufacturing of organic electronic devices. Moreover, the soft nature of the organic mat erials makes them intrinsically compatible with flexible substrates. Hence, various low cost substrates, such as glass, plastic, and stainless steel foils, can thus be used for organic devices. 1.2.1.2 Charge I njection 1.2.1.2.1. Metal/organic I nterface I n the most of organic semiconductor device, metal contacts are necessary for charge injection. Therefore, the understanding of metal semiconductor interfaces is crucial to the design of electronic devices. Depending on the work function of the metal and th e HOMO and LUMO levels of the semiconductor, the interface may be either ohmic or rectifying (also called Schottky contact). The potential barrier that charge carriers have to overcome at the metal/semiconductor interface is determined by the relative heig ht of the Fermi level of the two materials. Figure 1 4 (a) shows the energy

PAGE 25

25 band diagram of a metal and an n type semiconductor before contact. m is the ionization potential of the metal and is the electron affinity of the semiconductor. When put in co ntact, electrons will diffuse from the semiconductor into the metal until the Fermi levels of both materials align ( Figure 1 4 (b)). Electrons in the metal now face a barrier q b which is q b = q ( m ) (1.1) On the other hand, the barrier encounte red by electrons in the semiconductor, the so called built in voltage V bi is equal to: qV bi = q ( m s ) (1.2) where m is the work function of the metal and s the work function of the semiconductor. In an ohmic contact, these barriers are negligib le and current can flow across the interface in both directions with very little resistance. In a Schottky contact, these barriers are not negligible and the carriers have to overcome them (e.g. through thermionic emission or tunneling). In the previous d escription, an ideal metal semiconductor interface was assumed. However, in practice, several factors arise: chemical reactions between the two materials, physical strains, interface states. Particularly, atoms on the surface of the semiconductor only have neighbors on one side with which their valence electrons could form covalent bonds. Therefore, a net charge in the surface states appears. These charges, together with their image charges on the metal surface, constitute a dipole layer between the metal a nd the semiconductor. The dipole layer introduces a misalignment of vacuum levels ( Figure 1 4(c)) and alters the barrier height. The previously stated model has to be modified to:

PAGE 26

26 Q b = q ( m (1.3) where is the dipole barrier. Its height depends on several factors such as the permittivity of the dipole layer, its thickness and the density of surface states. The formation of a dipole layer at a metal semiconductor interface and t herefore the injection efficiency of a device depend largely on sample preparation conditions. A number of possible factors that drive the dipole formation were proposed, among them chemical reactions, ion formation, mirror forces or surface electronic rea rrangement as well as the presence of permanent dipoles at the interface in the case of some organic materials. 1.2.1.3 Charge T ransport Once the charges have injected into organic materials from electrodes, the charge carriers can transport (move) throug h the organic semiconductor s Carrier transport at a certain electric field is described through the conductivity of the material, defined as: = E (1.4) where J is the current density ( = / with I the current and A the surface) and E the electric field. If only electrons are responsible for the current, the current density can be expressed by: = (1.5) where e is the elementary charge, n e the electron density and the drift velocity of ele ctrons. The quantity linking the drift velocity of electrons to the applied electric field is called the electron mobility e : = E (1.6)

PAGE 27

27 Respectively, the hole mobility h is defined as: = E (1.7) where is the drift velocity of holes. While in inorganic materials or highly pure organic single crystals, carriers undergo band transport, in most organic materials, electrons and holes are transported by a hopping mechanism. In band transport, carrie rs move as quasi free particles with an effective mass and their wavefunction is modulated by the periodic potential of the crystal. In this model, transport is mainly limited by scattering of carriers with modes of atomic vibrations, called phonons. Exper imentally, in the case of band transport, carrier mobility decreases with temperature within a large temperature range [16] However, in the most organic materials, the band transport model is not valid. Mobility values typically increase with temperature and are smaller by several orders of magnitude compared to inorganic semiconductors In these materials, transport occurs by hopping of carriers between localized states. Experimentally, it has been shown through time of flight (TOF) measurements over large ranges of electric fields E, that in many materials, the carrier mobility exhibit s following behavior: exp ( ) Several models have been proposed to explain this dependence, amongst which the Bssler model (or Gaussian Disorder Model (GDM)), which will be outlined here. The Bssler model assumes that the energetic distribution of states (or density of states, DOS) is a Gaussian distribution with a width [17] The ho pping rate depends on the difference in temperature T Monte Carlo simulations of charge transport considering these hypotheses lead to the following expression for mob ility, confirming the dependancy:

PAGE 28

28 = 0 { 2 3 2 } { 2 2 } (1.8) where k C is a constant. 1.2.1.4. Fluorescence and P hosphorescence Exciton s ha ve spin. An e xciton may be considered as two electron systems: one electron is excited into an unfilled orbital of a given molecule, while the second remains in a partially filled ground state. The total spin of a two electron system is either S=0, or S=1. The S = 0 s tate or singlet state ( S ) is anti symmetric under particle exchange and is given by: S = 1 2 { | 1 2 > 1 2 > ( 1.9 ) state ( T ) contains three states, all symmetric under particle exchange: T = 1 2 { | 1 2 > + 1 2 > (1.10) T = | 1 2 > ( 1.11) T = | 1 2 > (1.12) The degeneracy of each state is reflected in its title: singlet (S=0) and triplet (S=1). This relative degeneracy of the singlet and triplet states determines a 1:3 singlet:triplet ratio. By convention, the ground electronic singlet state is represented by S 0 and electronically excited singlet states are labeled S 1 S 2 etc, where the subscript refers to the first and second singlet energy level manifolds above the ground singlet energy, respectively. Similarly T 1 T 2 are the first and second triplet energies above the ground energy state ( Figure 1 5 ). Radiative singlet decay is known as fluorescen ce and r adiative triplet decay is known as phosphorescence In nature, phosphorescence is very rare except for some

PAGE 29

29 materials, which has strong singlet triplet mixing with large spin orbit coupling due to the presence of heavy metal atoms such as Pt or Ir [18, 19] Exciton transfer from singlet state to triplet state by spin orbit coupling is termed Intersystem crossing ( Figure 1 5) 1.2. 1.5 Excitons Injected and transported, electrons and holes from electrodes generate e xcitons before emit photons in the emitting zone There are t hree types of excitons found in crystalline solids as shown in Fig. 1 6 A Frenkel exciton corresponds that the electron hole pair locat es on t he same molecule. The radius (R) of Frenkel exciton is similar to the lattice constant a (< 5 A ) A charge transfer (CT) exciton is slightly delocalized over two or several adjacent molecules w ith increasing intermolecular interactions In organic semiconductor, Frenkel and CT excitons are commonly observed. Wannier Mott excitons ( R >> a ) are easily found in highly delocalized R inorganic semiconductor crystalline (40 100 A ). Due to the Coulombic interaction between the constituent electron and hole the Frenkel or CT excitons are tightly bound, with a binding energy ranging from 0.1 eV to 2 eV while that for Wannier Mott excitons in inorganic semiconductors is only a few meV [20, 21] Frenkel excitons are the most commonly o bserved excitons in organic materials, and especially in amorphous organic materials since these excitons are tightly bound and highly localized. More delocalized c harge transfer excitons are often found in polycrystalline or even crystalline organic mate rials. Materials may often exhibit CT states in addition to Frenkel excitonic transitions [21, 22] 1.2. 1. 6 Excitonic E nergy T ransfer There are three main excitonic energy transfer from one molecule to another: cascade (trivial) Frster, and Dexter energ y transfer mechanisms [21] Cascade energy

PAGE 30

30 transfer (or trivial energy transfer) happens by photons from emitted by donor to absorbed by acceptor. The acceptor and donor materials doesn t influence directly. Photon is a medium of e nergy transfer. This process can be very long range. The rate of energy transfer depends on the emission efficiency of the donor and the absorption efficiency of the acceptor. Frster energy transfer (resonance energy transfer) needs the overlap between th e absorption spectrum of the acceptor and emission spectrum of the donor. The dipole dipole interaction (dipole dipole coupling) between donor and acceptor is the way of energy transfer. Photon is not the medium of energy transfer. Dipole dipole interactio n between molecules is R 3 (where R is the mean spatial separation between donor and acceptor). However, the lifetimes of excited donor and acceptor states are too short to make coherent dipole dipole interaction. The rate of energy transfer is also R 6 by considering coherency of dipoles [21] Dexter energy transfer occurs when electrons are physically exchanged between molecules. Electron clouds must significantly overlap with much shorter separation between donor and acce ptor molecules. It doesn t need the overlap of the emission of the donor and the emission of the absorption. Dexter energy can happen in triplet exciton energy transfer independent of the oscillator strengths of the donor and acceptor transitions. 1.2.1. 7 The D efinition of B rightness and E fficiency There are two ways to define the light intensity in light emitting devices. One is based on counting the number of photons (radiometric units) and the other is based on the light power of a source as perceived b y the human eye (photometric units) [23] UV and IR beyond the range of the perception of human eye need to be measured with radiometric units such as the photon energy and optical power (Watt).

PAGE 31

31 Quantum efficiency is given for expressing how efficient the light emitting device is. In other case, especially in this dissertation, photometric units with the informati on of what human actually feel the brightness are more relevant for the range of visible spectrum. Photometric units will be discussed in section 1.2.2.4 in more detail. Here, the most important radiometric efficiency, external quantum efficiency (EQE), is defined. The external quantum efficiency ( EQE ) is a measure of how many photons are emitted into the forward viewing direction for each injected electron by means of photo energy or electric energy : = = ( 1.1 3 ) where is the charge balance factor, and is equal to unity for balanced injection of electrons and holes into the EML, denotes the fraction of radiative excited states given spin selection rules ( ~ 0.25 for fluorescence and ~ 1 fo r phosphor escence ), and is the fraction of emitted photons that escape int o the forward viewing direction the light extraction efficiency ). From Eq. 1.1 3 we de fi ne the internal quantum e ffi ciency ( IQE ) as the total number of photons int ernally emitted in the device T he intrinsic photoluminescence (PL) efficiency ( ) of the emissive layer is defined : = + ( 1 .1 4 ) where k R and k NR are the radiative and non radiative decay rates. Non radiative decay ind icates the energy transition from t he excited state to the ground state without the emission of the photon due to the relaxation through intermolecular excited state

PAGE 32

32 quenching. Measurement of the PL efficiency is quite useful to characterize the performanc e of emitting materials. Before we understand the current efficiency and power efficiency, the most important photometric efficiency units, we need to know the units, the candela (cd) and the lumen (lm). The definition of candela is a monochromatic light source emitting an optical power of (1/683) Watt at 555nm into the solid angle of 1 steradian (sr) has a luminous intensity of 1 candela (cd) Originally 1 standardized candle emi ts a luminous intensity of 1.0 cd. Lumen is defined a monochromatic light source emitting an optical power of (1/683) Watt at 555nm has a luminous flux of 1 lumen (lm) The relationship between cd and lm is, = ( 1 15 ) 1 = 4 = 12 57 ( an isotropically emitting light source) ( 1 16 ) The luminance (cd/m 2 ) is the luminous flux incident per unit area. The current efficiency (cd/A) is the luminance by applied current density, = = 2 / 2 = / ( 1 17 ) The power efficiency (lm/W) in lumens per electrical w att, is a measure of perceived optical power per electrical p ower Therefore, for a given emission wavelength, the power efficiency is maximized by maximizing the quantum efficiency and minimizing the operating voltage. 1.2.2. OLED Basics : Optical P roperties emitting devices. Their performances a re greatly affected by the optical design of OLEDs. Optical properties of OLEDs greatly affects on spectra and

PAGE 33

33 efficiencies of devices. In this section, the fundamental optics of OLEDs will be presented Generated light from organic phosphors encounters ma ny surfaces and the light can be transmitted, absorbed, reflected, refracted or diffused (or some combination) by these surfaces and the material s Each of these properties is discussed in this section. Furthermore, Fabry Perot interference, lambertian emi ssion, light extraction efficiency in OLEDs is also discussed for optical applications. The concept of CIE and CRI follows in the end of this section. 1.2.2.1. Ref lection, R ef raction, and A bsorption in O rganic M aterials Reflection There are three general types of reflection: specular spread and diffuse as shown in Figure 2 1. A specular reflection, such as what you see in a mirror or a polished surface, occurs when light is reflected away from the surface at the same e. Specular reflections simply follows the law of reflection, which states that the angle between the incident ray and a line that is normal (perpendicular) to the surface is equal to the angle between the reflected ray and the normal. A spread reflectio n occurs when an uneven surface reflects light at more than one angle, but the reflected angles are all more or less the same as the incident angle. A diffuse reflection, sometimes called l ambertian scattering or diffusion, occurs when a rough or matte s urface reflects the light at many different angles. Some of the light extraction scheme uses these spread or diffuse reflection [24] Refractive I ndex (n) The refractive index, n, of a medium is defined as the ratio of the velocity, c, of a wave phenomenon such as light or sound in a vacuum the phase velocity, v p :

PAGE 34

34 ( 1. 18 ) In other media, the refractive index has electric and magnetic properties of materials: ( 1. 19 ) r r is its relative permeability. For most r is very close to 1 at optical frequencies, therefore n is approximately In real materials, the polarization does not respond instantaneously to an applied field. This causes dielectric loss, which can be expressed by a permittivity that is both complex and frequency dependent. Real materials are not perfect insulators either, i.e. they have non zero direct current conductivity. Taking both aspect s into consideration, we can define a complex index of refraction: ( 1. 20 ) Here, n is the refractive index called the extinction coefficient, which indicates the amount of absorption loss when the the frequency (wavelength). Not e that the sign of the complex part is a matter of convention, which is important due to possible confusion between loss and gain. Figure 1 8 shows wavelength dependent n and k of some important materials in OLEDs. Most of organic materials have n of 1 7 in contrast to n of inorganic materials (i.e. n of Si is ~ 4) In Figure 1 8 (b), k of these materials is close to 0, which indicates negligible loss. High conductive organic materials like PEDOT:PSS has a small loss. Snell s law

PAGE 35

35 on shows the relationship between the incident angle and the refractive index: n i i = n t t ( 1.21 ) Where n i = the refractive index of medium i, n t = the refractive index of medium t, i = the incident angle of the light ray (with respect to th e normal) t = the refracted angle (with respect to the normal) Figure 1 9 when light encounters glass from air to glass In OLEDs, these combinations of refraction and reflection always happens at every bound ary of organic/ITO, ITO/glass, and glass/air. The direction and intensity of refracted and reflected light follows this Fresnel R eflection (Fresnel equation) When light moves from a medium of a given refractive index n 1 into a second m edium with refractive index n 2 both reflection and refraction of the light may occur. Fresnel equation is generalized form of snell s law. The calculations of R and T depend on polari z ation of the incident ray. If the light is polarized with the electric field of the light perpendicular to the plane of the diagram above ( s polarized ), the reflection coefficient is given by: ( 1. 22) t i by Snell's law and is simplified using trigonometric identities. If the incident light is pol arised in the plane of the diagram ( p polarised), the R is given by:

PAGE 36

36 ( 1.2 3 ) The transmission coefficient in each case is given by T s = 1 R s and T p = 1 R p If the incident light is unpolarised (containing an equal mix of s and p polarisations), the r eflection coefficient is R = ( R s + R p )/2. In common case, reflection is 4% from air to glass at the normal incidence. This reflection at the normal direction derived from the Fresnel equation is called as the Fresnel reflection. This Fresnel reflection is fundamental reason of weak microcavity effects on OLED. 1.2.2.2. Fabry Perot I nterference The interference phenomena always exist in OLEDs due to its intrinsic Fresnel reflection Fabry Perot interference is the simplest form of optical cavity effect con sisting of two coplanar mirrors separated by a distance L cav in Figure 1 10 Constructive interference occurs if the transmitted beams are in phase, and this corresponds to a high transmission peak. If the transmitted beams are out of phase, destructive in terference occurs and this corresponds to a transmission minimum. Very small cavities, with typical dimensions of L cav will be denoted as microcavities. Plane waves propagating inside the cavity can interfere constructively and destructively resulting in stable (allowed) optical modes and attenuated (disallowed) optical modes, respectively ( Figure 1 10 (b) ) The tran smittance through a Fabry Perot cavity can be expressed in terms of a geometric series. The transmitted light intensity is then given by T = T 1 T 2 1 + R 1 R 2 2 R 1 R 2 cos 2 ( 1.2 4 )

PAGE 37

37 Where is the phase change of the optical wave for a sin gle pass between the two reflectors. Phase changes at the reflectors are neglected. The maxima of the transmittance occur if the condition of constructive interference is fulfilled, i.e. if 2 =0, 2 1.28 ) yields the transmittance maxima as T max = T 1 T 2 1 R 1 R 2 2 ( 1.2 5 ) 1.2.2.3 Lambertian emission The index contrast between the light emitting material and the surrounding material leads to a non isotropic emission pattern. Consider a light ray e mitted from the source at an angle with respect to the surface normal. The light ray is refracted at the semiconductor air interface and the refracted light ray has an angle with respect to the surface normal. The light intensity in air can be derived to [23] = 4 2 = 2 2 ( 1. 26 ) The l lambertian emission pattern is shown schematically in Figure 1 11 Several other surface shapes are also shown in Fig. 1 11 The non planar surfaces exhibit various emission patterns. An isotropic emission pattern is obtained from hemispherically shaped LEDs, which h ave the light emitting region in the center of the sphere. A strongly directed emission pattern can be obtained in LEDs with parabolically shaped surfaces. However, both hemispherical as well as parabolic surfaces are difficult to fabricate.

PAGE 38

38 1.2.2. 4 Light O ut coupling E fficiency (Light extraction efficiency) A l arge amount of generated l ight doesn t escape out to air due to its TIR at the organic semiconductor air interface. Light extraction efficiency is one of big issue to increase device efficiency. In t his section, the light extraction efficiency of OLED will be discussed. Light escape cone Assume that the angle of incidence in the semiconductor at the semiconductor air interface is given by = ( 1. 27 ) w here n s and n air are the refractive indices of the semiconductor and air, respectively. The critical angle for TIR Figure 1 12 = / ( 1. 28 ) The angle of total internal reflection defines the light escape cone. Light emitted into the cone can escape from the semiconductor, whereas light emit ted outside the cone is subject to TIR. Next, we calculate the surface area of the spherical cone with radius r in order to determine the total fraction of light that is emitted into the light escape cone. The surface area of the calotte shaped surface sho wn in Figs. 1 12 (b) and (c) is given by the integral = = 2 2 = 0 = 2 2 ( 1 cos ) ( 1. 29 ) Let us assume that light is emitted from a point like source in the semiconductor with a total power of P source Then the power that can escape from the semicondu ctor is given by

PAGE 39

39 = 2 2 ( 1 cos ) 4 2 ( 1.3 0 ) Where 4 2 is the entire surface area of the sphere with radius r. The calculation indicates that only a fraction of the light emitted inside a semic onductor can escape from the semiconductor. This fraction is given by below for only high index materials (inorganic semiconductor materials). = 1 2 ( 1 cos ) 1 2 1 1 1 2 2 1 2 1 2 2 1 4 2 2 ( 1.3 1 ) Organic vs. Inorganic semiconductor materials Maximum of light extraction efficiency Based on above light escape cone calculation the light extraction efficiency ( ext ) of organic and inorganic materi als (Si) is shown in Table 1 3. The most of OLEDs has metal cathode mirror like Al or Ag. Cathode reflects back and increases the light extraction efficiency to almost double ( 1/2n 2 ) ideally Based on this calculation, the light extraction efficiency is k nown about 17.3% at maximum for OLEDs [25, 26] There is one more important interference effect to be considered. The recombination zone in an OLED is g enerally located from ~50nm away from cathode mirror. In this distance, reflected light waves ha ve in phase constructi ve alignment Therefore, the maximum light extraction efficiency can go up to double 1/ n 2 Therefore, the maximum light extraction efficiency of OLEDs is 34.6% if it has perfect reflective mirror. Use of aluminum (reflectivity of ~ 85%) as a cathode mirror will lower down the limit of light extraction efficiency (~30%) Extracted, substrate guided, and ITO/organic modes The light emitted by an OLED device can be divided into three modes: the extracted mode, which is the light escaped from the substrate to air; the substrate

PAGE 40

40 mode, which is the light trapped in the substrate; and the ITO (Indium Tin Oxide)/organic mode, which is the light trapped in the ITO and the organic layers in Figure 1 13. Because of the mismatch of the refractive indices between air an d glass, and between glass and ITO, a fraction of the generated light is trapped in the glass and the ITO/organic layer due to total internal reflection (TIR), depending on the angle of incidence of the light ray undergoing TIR at different interfaces [27] The external quantum efficiency, is given by = where is the internal quantum efficiency and extraction schemes have been used to enhance the light extraction efficiency of O LEDs. These methods can be divided into two approaches. The first approach is to extract the substrate guided mode in the glass substrate to enhance light extraction. The second approach is to extract the ITO/organic mode. Some approach use the combination of these two approaches. Table 1 4 is the summary of these approaches Until now, macrolens [9, 28] or microlens array [29, 30] is one of best opt ion to extract the substrate guided mode. Photonic crystal [31] high refractive index subs trate [9] and silica ae ro gels [32] are used for out coupling the ITO/organic mo de. 2.4 times enhancement of power efficiency is reported by combining two approaches [9] 1.2.2. 5 P hotopic and S cotopic R esponse When human eyes see ( perceive ) the light (photons), all photons aren t counted equivalently in the eye Detectors in the eye have different weight ing factors for specific wavelength s of photon (specific color). The eye s weight ing factors correspond to photopic and s cotopic response. There are two main classes of photodetectors in the eye: rods and cones. Rods are extremely sensitive to light regardles s wavelength and

PAGE 41

41 are responsible for scotopic or night vision. The cones ha ve the ability to discriminate color and determine the photopic or day vision. Scotopic vision is active below 0.01 cd/m 2 and photopic vision requires luminance levels to be greate r than 3 cd/m 2 [23, 33] Figure 1 14 shows the photopic and scotopic responsivities normalized with respect to peak height. The units of b rightness, lumen (lm) and candela (cd), which is explained in the section 1.2.1.6, contain the photopic information. Therefore, the luminance (cd/m 2 ) has strong dependence on the wavelength of spectrum. 1.2.2. 6 The CIE The CIE chromaticity coordinate syste m is method to indicate the specific color, originally recommended in 1931 by the (CIE). The CIE is define d by giving the amounts X, Y, Z of three imaginary primary colors in Figure 2 3. These amounts are calculated as a summation of the spectral compositions of the radiant power of the source times the spectral tristimulus values or color matching functions for an equal power source [23] : = ( 1.32 ) = ( 1.33 ) = ( 1.3 4 ) Here, S( ) is the spectral irradiance of the source, and and are the spectral tristimulus values plotted in Fig. 2 3. Chromaticity coordinates ( x, y, z ) are then calculated as follows: = + + ( 1.35) = + + ( 1.36 ) = + + (1.37 )

PAGE 42

42 S ince 1 = x + y + z, the CIE chromaticity is expressed in terms of x and y (x, y). The conventiona l CIE chromaticity coordinate system is shown in Fig. 2 4. 1.2.2. 7 The CRI ant CCT The standards of the CRI and CCT are come from the solar spectrum, which is described by Planck s blackbody spectrum. These two concepts are very useful to characterize the quality of white illumination. The Color Rendering Index (CRI) varies from 0 (monochromatic spectrum) to 100 (Blackbody like spectrum). For example, two white light radiation can have the identical CIE coordinate between white light consisted of monochrom atic red, green, and blue, and spectrum like blackbody covering whole visible spectrum range. However, the CRI would be very different close to from 0 to 100. If the white illumination is required for the purpose of the lighting of the colorful Figure s or photos, white light with high CRI (~90) is necessary. There are several kinds of white lights. White light emitted by the Sun varies from day light in the clear sky (Cool white) to sunset (Warm white) due to scattering and absorption of atmosphere. An i nc andescent bulb emits the warm light and a fluorescent tube emits the cool white. The Correlated Color Temperature (CCT) is used for describing the color difference of white light based on the sun s black body radiation. The color temperature of a blackbody is defined as its temperature for a particular perceived color. For example, a blackbody radiator with a color temperature of 10,000 K appears blue. T he correlated color temperature (CCT) is the temperature of a blackbody radiator which has a color that m ost closely resembles that of the light source. Table 1 5 shows the representative white light sources and their CRI and CCT.

PAGE 43

43 1.3 Disserta t ion O utline The final goal of this dissertation is to show the route s to achieve efficient white OLED with down con version. Down conversion and microcavity needs a lot of electrical and optical understanding of OLEDs. These basic electrical and optical properties of OLEDs are given in Chapter 1. To achieve efficient down conversion, efficient blue emitting device is es sential. The most important part of down conversion is an efficient blue OLED. Two key factors of efficient blue PHOLED are explained in the Chapter 2. Chapter 3 describe s the weak microcavity effects on OLEDs. Based on understanding of weak microcavity ef fects, Chapter 4 elucidate s the strong microcavity effects on blue PHOLEDs. Efficiency enhancement and light extraction efficiency enhancement of strong microcavities are basic momentum for efficient down conversion. Chapter 5 describe s how to integrate th e down conversion white OLEDs and their detail analysis. We conclude in Chapter 6 with a brief out look at future research challenges in white OLEDs

PAGE 44

44 Table 1 1. Review of recent w hite OLEDs Architecture Ext (%) p (lm/W) CIE CRI Refs. Year All phospho rescen ce 12 19 26 24.1 26 40 55 49.3 (0.43, 0.45) (0.37, 0.40) (0.34, 0.39) (0.49, 0.41) 80 79 68 62 [12] [34] [35] [36] 2004 2009 2008 2008 Fluorescence and phosp horescence hybrid 33 18.7 14 37.6 (0.38, 0.44) (0.40, 0.41) 82 85 [37] [38] 2006 2006 All fluorescen ce 19.2 c d/A 10.6 cd/A 16 9.5 (0.36, 0.43) (0.32, 0.36) <70 <70 [39] [40] 2007 2008 Down conversion 1.3 39 cd/A 69 cd/A 3.7 25 67.7 4130K (0.26, 0.40) (0.43, 0.46) 93 <60 83 [15] [41] This work 2002 2006 2009 Ext (%) : Maximum reported forward viewing external quantum efficiency. Sometime current efficiency is given instead of external quantum efficiency p (lm/W) : Maximum reported forward viewing external power efficiency All efficiency data is without light o ut coupling scheme, although down conversion white OLEDs already extracts some of wave guided mode out. Three tandem structure

PAGE 45

45 Table 1 2. Organic semiconductor vs. inorganic semiconductor materials. Modified from R ef. [42] Organic semiconductor materials (e.g., pentacene type crystal) Inorganic semiconductor materials (e.g., silicon type crystal) Properties 1. Weak Van der Waals type of intera ction (characteristic interaction energies E VDW =0.001~0.01 eV) 2. Tendency of charge carrier and exciton localization 3. Charge carriers and excitons as polaron type quasiparticles 4. 2 /Vs) and small mean free path (l=a 0 = latt ice constant) at room temperatures 5. Hopping type charge carrier transport dominant 6. Excitons as molecular Frenkel type quasi particles 7. Low melting and sublimation temperatures 8. L ow mechanic strength, high compressibility 9. Easy to modify their electric al and optical properties by tailor molecule structure 1. Strong covalent type interaction (characteristic interaction energies Ecov=2 4ev) 2. Pronounced charge carrier delocalization 3. Charge carriers as free holes and electrons 4. cm 2 /Vs) and large mean free path [l=(100~1000)a 0 ] 5. Band type charge carrier transport dominant 6. Excitons as Wannier type quasi particles 7. High melting and sublimation temperatures, 8. H igh mechanical strength low compre ssibility 9. Difficult to modify their electr ical and optical properties Processes 1. Compatible with low cost and large area manufacturing processes 2. Less restrictive on the substrate and growth conditions 3. Very difficult to make high purity due to impurity and molecular defect 1. Compatible with high performance and small area manufacturing processes 2. Lattice matched substrate required 3. High purity crystal is possib le

PAGE 46

46 Table 1 3 Extraction efficiencies for emitter in a bare situation (top) or modified by a mirror. For an emitter lying close to a mirror (bottom), there is a negligible phase change for rays emitted between normal incidence and critical angle, unlike for Ref. [43]

PAGE 47

47 Table 1 4 Schemes of enhancing the light extraction efficiency Outcoupling scheme Factor Extracting Substrate guided mode Extracting ITO/organic mode Refs. High refractive index substrate + High index macrolens 2.4 O O [9] Macrolens 1.7 O X [9, 28] Microlens array 1.7~1.8 O X [29, 44] Photonic crystal pattern 1 .5 O X [31] Low index grid 1.3 X O [44] Scattering layer 1.4 O X [45] Silica aerogels 1.8 X O [32]

PAGE 48

48 Table 1 5 CIE, CCT, CRI and power efficiency of the representative w hite light sources from [8, 46] Light source CIE (x, y) CCT CRI lm/W Daylight (Sun) 0.313, 0.329 6500K 90 Incandescent bulb 0.448, 0.408 2854K 100 15 20 lm/W Fluorescent, cool white 0.3 16 0.3 36 6500 K 8 5 50 8 5 lm/W Fluorescent, warm white 0.440, 0.408 2940K 72 50 8 5 lm/W Tungsten Halogen 0.448, 0.407 2856K 100 25 lm/W High pressure sodium vapor light bulb 0.519, 0.417 2100K 24 100 140 lm/W Down conversion white OLED (this work) 0.43, 0.46 ~2800K 83 88 lm/W A t 100 nits

PAGE 49

49 Figure 1 1. LED display by Cambridge Display Technology (b). Figure 1 2 Three main approaches for generating white light in OLEDs. (a) Multiple doping in emissive layer, (b) Multiple emissive layers, (c) down conversion with blue OLED

PAGE 50

50 Figure 1 3 Recent progres s es of white OLEDs. All power efficiency is without light out coupling scheme.

PAGE 51

51 Figure 1 4 Metal semiconductor interface (a) before and (b) after contact for ideal case. Metal semiconductor interface (c) with interface dipole layer Figure 1 5 Energy level diagram illustrating the energy level splitting between singlet and triplet states arising from the exchange interaction K.

PAGE 52

52 Figure 1 6 Schematic representations of three types of excitons in a solid: (a) a Frenkel exciton, localized on a single molecule; (b) a charge transfer exciton, slightly delocalized over two or several adjacent molecules; and (c) a Wannier Mott exciton, which is highly delocalized with a radius much greater than the lattice constant Adapted from Ref. [20]

PAGE 53

53 Figure 1 7 Specular, spread, and diffuse reflections from a surface.

PAGE 54

54 Figure 1 8 (a) Refractive index (n) of ITO, NPD, Alq 3 PEDOT:PSS, and SiO 2 Adapted from the data base of Setfos software [47] Figure 1 8 (b) Extinction coefficient (k) of ITO, NPD, Alq3, PEDOT:PSS, and SiO 2. Adapted from data base of Setfos software [47]

PAGE 55

55 Figure 1 s refracted at the boundary between two mediums.

PAGE 56

56 Figure 1 10 (a) Transmission of a light wave with electric field amplitude E 0 through a Fabry Perot cavity Figure 1 10 (b) Schematic illustration of allowed and disallowed optical modes in a Fabry P erot cavity consisting of two coplanar reflectors.

PAGE 57

57 Figure 1 11 Light emtting diodes with (a) planar, (b) hemispherical, and (c) parabolic surfaces. (d) Far = 60, the lambertian emission pattern decreases to 50% of its maximum The three emission patterns are normalized to unity Adapted from [23]

PAGE 58

58 Figure 1 12 (a) Definition of the escape cone by the critical angle c (b) Area element dA. (c) Area of calotte shaped section of the sphere defined by radius r and angle c Adapted from [23] Figure 1 13 3 optical modes in OLEDs. Extracted, substrate guided, and ITO/organic modes

PAGE 59

59 Figure 1 14 photopic vision regime. Also shown is the eye sensitivity function for the Adapted from [23]

PAGE 60

60 Figure 1 15 CIE (1931) and CIE (1978) xyz color matching functions (CMFs). The y CMF is identical to the eye sensitivit CMF is the currently valid official standard. Adapted from [23]

PAGE 61

6 1 F igure 1 16 CIE 1931 (x, y) chromaticity diagram. Mono chromatic colors are located on the perimeter. Color saturation decreases towards the center of the diagram. White light is located in the center. Also shown are the regions of distinct colors. The equ al energy point is loc a ted at the center and has the coordinates (x, y) =(1/3, 1/3). Adapted from [23]

PAGE 62

62 CHAPT ER 2 HIGH LY EFFICIENT BLUE PHOSP HORESCENT OLEDS Light emission in OLEDs depends mainly on two opto electronic processes: fluorescence and phosphorescence. In organometallic complexes, the presence of a heavy metal ion leads to very efficient intersystem cr ossing, and hence these materials exhibit strong phosphorescence even at room temperature. With efficient mixing of triplet and singlet states, achievement of 100% internal quantum efficiency (IQE) is theoretically possible. Using phosphorescent emitters v ery high efficiency green [48] and red [49] OLEDs have been demonstrated. However, the efficiency of blue devices, which is an essential component f or achieving high efficiency white OLEDs, was still lagging behind until very recently [50] [51] In this Chapter we will discuss some of the most importance device parameters affecting the efficiency of these blue phosphorescent OLEDs and how to maximize the efficiency of these devices. 2 .1 H ighly E fficient B lue P hosphorescent OLEDS The OLED external quantum efficiency is defined in the section 1.2.1.7 : = = ( 1.13 ) oc is the light out coupling efficiency and is the in ternal quantum efficiency of the device. Light extraction efficiency is dependent on the device architecture [52] Most of the l ight emitted from the OLED is lost in the glass substrate and the ITO and organic layers due to total internal reflection (TIR) Many efforts have been focused on improving the light extraction efficiency of OLED devices by surface roughening [53] use of micro lenses [54] [55] low index gratings [56] or microcavity structures [57] In the present study we use microcavity to enhance the light output from the device which will be discussed later in detail.

PAGE 63

63 The internal quantum efficiency depends on the following parameters [48] : which is the fraction of total excitons formed which result in radiative transitions (~1 for phosphorescent materials), the charge balance factor and the intrinsic quantum efficiency for radiative decay. He nce, to maximize the efficiency of these devices, all these factors need to be optimized. In order to maximize for phosphorescent devices, triplet exciton confinement in the device is very important [58] For triplet exciton confinement, the triplet energy of the host material [58] as well as the nearby charge transport la yers [59] should be greater than that of the phosphorescent dopant. This prevents any back wards energy transfer from dopant to host or transport materials. We will illustrate the effect of the triplet energy of the transport layers as well that of the host material on the performance of FIrpic based blue phosphorescent OLEDs. The other factor which is very important for achieving high efficiency is balance of electron and hole carriers in the device [48] Imbalance in charge transport can lead to accumulation of charges in an OLED heterostructure thereby resulting in loss in efficiency [60] and lifetime [61] Conventionally, this factor is assumed to be close to one [48] However, in most OLED devices charge balance factor is far from its ideal value [62] We will also demonstrate the effect of charge balance on the location of the recombination zone of the device and how that in turns greatly affects the device performance. We demonstrate the effect of all these factors on blue PHOLED device performance. By optimization of all the above mentioned factors, we were able to achie ve high efficiency blue phosphorescent OLEDs with a high efficiency of 60cd/A (about 50 lm/W) without any out coupling enhancement which is one of the highest

PAGE 64

64 value reported for such devices [62] 2.1.1 Effect of Trip let E nergy of T ransport L ayers for T riplet E xciton C onfinement To demonstrate the effect of triplet state energy (T 1 ) of transport layer we fabricated blue phosphorescent devices with different hole transporting materials as well as electron transporting m aterials. Using TAPC as the hole transporting layer (HTL), we L = (19.8 1) cd/A, external quantum EQE P ) of (11 0.6) lm/W, which is approximately 35% highe r than that of previously reported devices with NPD as HTL EQE = 7.5 0.8%). We attribute this improvement to the enhanced electron and exciton confinement in the devices as TAPC possesses a smaller lowest unoccupied molecular orbital (LUMO) ene rgy and larger triplet energy than those in NPD. The device structure used in this study is shown in Figure 2 1. mCP was selected as a host because of its triplet energy and its high photoluminescence efficiency [63] when it is doped with FIrpic. A 20 nm thick f ilm of PEDOT:PSS as a hole injection layer was spin coated over ITO substrate and baked at 180C for 15 minutes. To complete the device fabrication, a 20 nm thick HTL (NPD, TPD, or TAPC), a 20 nm thick 3% FIrpic doped mCP emitting layer, a 20 nm thick BCP as a hole blocking layer ( HBL ) a 20 nm thick Alq 3 electron injection layer, and a lithium fluoride/aluminum (1 nm/100 nm thick) cathode were deposited sequentially. The device efficiencies with different HTLs are shown in Figure 2 2. The maximum current efficiencies of devices with different HTLs are 14.3 cd/A, 17.5 cd/A, and 19.8 cd/A for NPD, TPD, and TAPC, respectively. The triplet energies of the HTLs

PAGE 65

65 used in this study are 2.29 eV, 2.34 eV and 2.87 eV for NPD, TPD, and TAPC respectively. Our results indicate that TAPC confines the triplet excitons within the emitting layer most effectively. In devices with TPD and NPD, exciton energy readily transfers from FIrpic to the HTL, resulting in lower device efficiency. It should be noted that current efficie ncy of TPD device is slightly higher than that of the NPD device, even though T 1 of TPD is almost similar to T 1 [64] also found similar results in their green phosphorescent devices where they observed a significant difference in efficiency despite the small difference in the triplet energies between TPD and NPD. Given that the triplet energies of both TPD and NPD are lower than that of FIrpic, the emission zone should be kept away from the HTL interface. Since TPD has a higher hole mobility ( TPD = ~1x10 3 cm 2 /Vs at ~10 5 V/cm) than NPD ( NPD = ~5 x 10 4 cm 2 /Vs at ~10 5 V/cm) [65] it is expected that the TPD device should have an emission zone fu rther away from the HTL interface compared with the NPD device, resulting in higher efficiency in the TPD device. To study the effect of triplet energy of electron transport layer (ETL)/ HBL we fabricated devices with device structure shown in Figure 2 3. Three different electron transport materials were used for comparison: BCP, BPhen and 3TPYMB. 3TPYMB was chosen since its electron mobility (~10 5 cm 2 /Vs) [66] is about an order of magnitude higher than that of BCP and it has one of the highest triplet energy (T 1 =2.98 eV) [50] amongst all electron transport materials (ETMs) used in OLEDs. BPhen (T 1 =2.5 eV) [67] has triplet energy similar to that of BCP while its electron mobility (10 4 cm 2 /Vs) is the highest amongst all three ETMs. HOMO (highest occupied molecular orbital) and LUMO (lowest unoccupied molec ular orbital) energy levels, triplet energy and mobility

PAGE 66

66 for all the HTLs and ETMs have been summarized in Table 2 1. Figure 2 4 shows the device efficiencies for all three devices. The BCP device has the lowest efficiency of all which is in agreement with its lowest mobility and low triplet energy. Although the electron mobility of BPhen is substantially higher than those of the other two ET L s, due to its low triplet energy, the device efficiency is only slightly higher than that of the BCP device. Finally the 3TPYMB device shows substantially higher efficiency (60 cd/A) compared to the other two devices. Our optimized 3TPYMB device shows a peak luminous power efficiency of 50 lm/W. This clearly demonstrates that the device efficiency also depends on the t riplet energy of electron transport layer. This effect will be discussed in more detail when we talk about charge balance in these devices in section 2.1.3 2.1.2. Host M aterials As discussed previously, selection of host material for a blue phosphorescent OLED is a very important factor [58] [68] The efficiency of the blue PHOLEDs is generally low compared to that of their green and red counterparts due to the fact that it is more challenging to design wide band gap host and charge transport materials. Additionally, these wide gap materials pose problems for charge injection and transport, resulting in lower device efficiency. Hence, it is desirable to have wide gap materials with good charge injection and transp ort properties. As discussed, triplet energy of the host [68] as well as charge transport materials [59] [64] used in PHOLEDs, is one of the most important factors determining the performance of PHOLEDs. In PHOLEDs triplet excitons have much longer lifetimes than their singlet counterparts with exciton diffusion length>1000 [69] [70] [49] therefore we need host and transport materials with higher triplet energy as compared

PAGE 67

67 to the dopant material in order to prevent any backward energy transfer. This is especially challenging to blue PHOLEDs where very large band ga p is required. Initially, CBP was used as a host for FIrpic OLEDs [68] With CBP as host endothermic energy transfer from guest to host was observed, as the triplet energy of FIrpic (T 1 = 2.7 eV) is higher than that of CBP (T 1 = 2.6 eV) [68] Thereafter, higher triplet energy hosts were introduced for FIrpic such as mCP [58] with triplet energy of 2.9 eV or CDBP [71] with triplet energy of 3.0 eV which show exothermic host guest energy transfer. We used a very wide gap host for FIrpic OLEDs; UGH2 for FIrpic devices in difluorophenylpyridinato)tetrakis (1 pyrazolyl) borate) based blue PHOLEDs [72] In FIr6 OLEDs with UGH2 electrophosphorescence was observed due to guest charge trapping, i.e. luminescence is due to direct charge injection into energy levels of FIr6. Also, UGH2 has very high triplet energy of 3.5 eV [73] which can effectively confine the excitons on the dopant molecule. Hence, we studied FIrpic UGH2 guest host system and compared it to aforementioned host systems to investigate the effect of a wide band gap host and study how the triplet energy of the host material affects the efficiency of the device. For this study we have compared four different host materials: CBP, mCP, CDBP and UGH2. As mentioned CBP, mCP, CDBP behave very differently as compared to UGH2 in terms of route for electro phosphorescence i.e. charge trapping or energy transfer (exothermic or endothermic). Hence, these materials were selected for comparison with UGH2. Devices were made with device structure shown in Figure 2. 3. In the device with UGH2 as the host, 10% FIrpic was used in EML. A higher doping conce ntration is necessary as the transport is occurring through the dopant molecules.

PAGE 68

68 All the other hosts had only 3% FIrpic doping. Figure 2 5 shows the efficiency plots for the different host materials. The lowest device efficiency is seen for the devices with CBP host (14.5 cd/A) which is consistent with the fact that these devices show endothermic energy transfer from CBP to FIrpic. The peak device efficiencies for mCP (20.7 cd/A) and CDBP (19.9 cd/A) host materials are very similar and these results are expected as these materials have almost equal triplet energies. The efficiencies for these devices differ in roll off which is probably due to the difference in their charge transporting properties. The efficiency was found to be the highest for UGH2 host (31.7 cd/A) which has the highest triplet energy as well as the highest energy band gap which helps in efficient charge carrier confinement. Shape of efficiency curve for UGH2 devices is very different from others and has a steeper efficiency rise and roll off. This might be due to the difference in charge transport and balance in these devices. In fact we have seen that by modifying the device architecture and tuning the charge balance the efficiency for these devices can be further enhan ced and roll off c an be reduced [74] Though, higher triplet energy of the host might be the dominating factor responsible for 1.5X enhancement seen in device efficiency, other factors might also play an important role which will be discussed towards the end of this section. 2. 1.3. Effect of C harge B alance Charge balance is a very important factor in the performance of an OLED. It has already been established that internal quantum efficiency of the device is directly dependent on the balance of electron and hole carriers in the device [48] For maximizing the efficiency of blue PHOLEDs we need to confine the recombination zone in the bulk of EML. Charge imbalance can cause the recombination zone to move from

PAGE 69

69 the bulk of emitting lay er to the interface. Conventionally used transport materials such NPD, T 1 =2.29 eV [64] BCP (T 1 =2.5 eV) [18] and Alq 3 (T 1 = 2.0 eV) [18] have lower triplet energy than that of FIrpic, T 1 =2.7 eV [58] Hence, if the recombination zone is at the interface then the excitons can be quenched by lower triplet energy charge transporting layer as discussed in section III. Also, charge balance strongly affect s the efficiency roll off in a phosphorescent device [75] If the recombination zone is just a narrow zone at the interface it can lead to increased triplet exciton density and hence enhanced triplet triplet annihilation. Hence, the charge balance plays a vital role in achieving high efficiency for PHOLEDs. Charge balance is investigated in FIrpic based OLEDs by fabricating single carrier devices. Electron only and hole only devices were fabricated and their current voltage characteristics were compared. The devices were found to be largely hole dominant and hence it is expected that there is quenching from lower triplet energy BCP. The location of the emitting zone was further verified by probing the r ecombination zone in the device. The device structure used for this study was same as that shown in Figure 2 3 with mCP as the host. The hole mobility of TAPC (~ 1.0X10 2 cm 2 /Vs) [76] is four orders of magnitude higher than the electron mobil ity of BCP (5.5X10 6 cm 2 /Vs) [77] Hence, it is expected that the recombination zone will be located at the interface of EML and ETL. In addition, the triplet energy of BCP is lower than that of FIrpic. Hence, if the recombination zone is located near the BCP interface it could lead to luminescence quenching. To verify this single carrier devices were made corresponding to OLED dev ice structure. The hole only device has the following structure: ITO/ PEDOT (25 nm)/

PAGE 70

70 HTL (200 nm)/ EML (20 nm)/Gold (Au) and electron only device has the following structure: ITO/EML (20 nm)/ ETL (200 nm)/ LiF /Al. The device structure for these devices is shown in Figure 2 6 Current voltage (IV) measurements were performed on these devices. Figure 2 7 shows the current density vs. voltage plot for these devices. From this IV data it can be clearly seen that hole current density is orders of magnitude hig her than the electron current density, indicating that the device is very hole dominant. As the devices are found to be deficient in electron transport, it is expected that the recombination zone of the device is located at the EML/ETL interface. To verify the location of recombination zone and charge balance two modifications of our control dual carrier device were fabricated which have different location of FIrpic doping in the emitting layer. The left doped device (doped at HTL/EML interface only) as sh own in Figure 2 8 (b ) has the following structure: ITO/PEDOT/TAPC/mCP:FIrpic (10 nm) /mCP (10 nm)/BCP/LiF/Al The right doped (doped at HTL/EML interface only) device as shown in Figure 2 8 (c ) has the following structure: ITO/PEDOT/TAPC/mCP (10 nm)/mCP:F Irpic (10 nm) /BCP/LiF/Al. Since in mCP host based FIrpic devices, the charge carriers are transported through mCP host [58] the location of doping should not affect the location of recombination zone which is also reflected in almost identical IV characteristics for three devices as shown in F igure 2 9 ( a). However, as the dopant is located away from the recombination zone in EML, the luminance for the device doped on TAPC/mCP interface is expected to be lower which is illustrated in the LIV data in Figure 2 9 ( a). Hence, we expect to see lower efficiency for the device doped on TAPC/mCP interface. The efficiencies of these two partially doped devices and the control device are shown in Figure 2 9 ( b). The device which was doped at the

PAGE 71

71 TAPC/mCP interface has very low efficiency compared to the o ther two devices. Whereas for the devices doped on mCP/BCP interface, the efficiency is almost similar to that of control device. This clearly shows that recombination zone is located on mCP/BCP interface, which can lead to quenching due to BCP. 2.1. 4 Dem onstration of H igh ly E fficient B lue PHOLEDs Based on the above data it was concluded that we have two issues hindering the efficiency of blue PHOLED devices: 1) deficiency of the number of injected electrons and 2) lower triplet energy of ETL. Hence, if BC P is replaced with electron transport materials with higher electron mobility and higher triplet energy, charge balance will be improved and exciton quenching will be reduced resulting in higher overall device efficiency. Now let s revisit the data discuss ed in Figure 2 4 where we show the effect of triplet energy confinement of different electron transport layers. The efficiency enhancement that we saw was a result of both the high mobility as well as high triplet energy of 3TPYMB. 3TPYMB has a high elect ron mobility of (~10 5 cm 2 V 1 s 1 ) (Refer to Table 2 1 ) which is about an order of magnitude higher than that of BCP and as we already noted it has high triplet energy (T 1 =2.98 eV) to provide exciton confinement for FIrpic. Hence, if we look at the data in Figure 2 4 considering both charge balance and triplet energy effect, we note that BCP device has the much lower efficiency of all which is in agreement with its lowest mobility and low triplet energy. Although the electron mobility of BPhen is substanti ally higher than those of the other two ETMs, due to its low triplet energy, the device efficiency is only slightly higher than that of the BCP device. 3TPYMB device shows substantially higher efficiency (60 cd/A). The results demonstrate that both high mo bility and high triplet energy electron transport materials are required for high device efficiency.

PAGE 72

72 Finally, t o demonstrate the highly efficient blue PHOLED based on understanding of these two key factors, the triplet energy and the charge balance, we fab ricated another set of devices with mCP as host. We made a set of two devices with following device structure: ITO/ PEDOT:PSS(25nm) / TAPC ( 2 0 nm)/ mCP ( 2 0 nm) layer doped with 10 wt% FIrpic (20 nm) as an EML/ 3TPYMB (40 nm) as a ETL / LiF or CsCO 3 / Al catho de. In order to increase the electron injection efficiency and to achieve charge balance, CsCO 3 also used here as a electron injection layer in Figure 2 10. Figure 2 1 0 (a) shows the current efficiency and power efficiency of two blue PHOLEDs LIV charac teristics are shown in Figure 2 10 (b). Better electron injection efficiency in CsCO 3 devic e lowers down turn on voltage (< 3.0eV) and increases current density. This enhanced electron injection greatly affects on efficiency and its roll off. Because of th e effect of lowered turn on voltage on CsCO3 device, the maximum power efficiency increases from 50 (LiF) to 60 lm/W (CsCO 3 ) at very low current density (~ 0.05 mA/cm 2 ). However, as current density increases, efficiency drop (efficiency roll off) is signif icant for CsCO 3 device. This significant roll off difference might be due to triplet triplet annihilation since enhanced electron injection generated much larger number of triplet exciton in emitting zone at high current density ( > 1 mA/cm 2 ). From this ex ample understanding of charge balance on OLEDs gives a perception not only to achieve high efficiency OLEDs but also to control the roll off of efficiency. 2. 2 Conclusion In this Chapter we focus on the factors affecting device performance of blue pho sphorescent organic light emitting diodes (PHOLEDs). Effect of key factors such as triplet exciton confinement, host materials and charge balance is illustrated for FIrpic based blue PHOLEDs. All these factors affect the device performance and hence, it i s

PAGE 73

73 necessary to optimize all these factors in tandem to achieve high efficiency blue emitting PHOLEDs. With the optimization of all the above mentioned factors, a high efficiency blue PHOLED with 6 0 cd/A and 50 lm/W was demonstrated.

PAGE 74

74 Table 2 1 Energy lev els, triplet energy and mobility parameters for different electron transport materials used in this study Mobility is the hole mobilities for NPD, TPD, and TAPC and the electron mobilities for BCP, BPhen, and 3TPYMB. Charge transporting layers HOMO (eV) L UMO (eV) T 1 (eV) Mobility (cm 2 V 1 s 1 ) NPD [64] 5. 4 2.4 2.29 10 4 TPD [64] 5. 6 2.5 2.34 10 3 TAPC [64] 5.5 1.8 2.87 10 2 BCP [77, 78] 6.5 3.0 2.5 10 6 BPhen [67, 79] 6.4 3.2 2.5 10 4 3TPYMB [66] 6.77 3.3 2.98 10 5

PAGE 75

75 Figure 2 1 Device structures of blue PHOLEDs for studying effect of triplet energy confinement from HTL Figure 2 2 Current efficiency and power efficiency of blue PHOLEDs by changing HTLs

PAGE 76

76 Figure 2 3 Device structure used for studying effect of different electron transport layers (ET L s). Figure 2 4 Current efficienc ies of blue PHOLEDs with dif ferent electron transport layers.

PAGE 77

77 Figure 2 5 Current efficiency of blue PHOLEDs fabricated with different host materials.

PAGE 78

78 Figure 2 6 Device structures for single carrier devices used in this study ( a) Hole only device structure ( b) El ectron only device structure. Figure 2 7 Current density voltage (J V) characteristics for single carrier devices shown in Figure 2 6

PAGE 79

79 Figure 2 8 Devices fabricated for probing the recombination zone. a) Device doped on the entire emitting layer, b) Device doped only on interface of HTL/EML (left doped), c ) Device d oped only on interface of EML/ETL (right doped).

PAGE 80

80 (a) (b) Figure 2 9 ( a) LIV characteristics for devices in Figure 2 8 and contro l device. The filled symbols indicate the current density whereas open symbols indicate luminance of the corresponding device. ( b) Current efficiency for these devices.

PAGE 81

81 (a) ( b ) Figure 2 1 0 (a) the current efficiency and luminous efficacy for highly efficient blue PHOLEDs with LiF or CsCO 3 layer as a electron injection layer (b) LIV characteristics these two devices. Turn on voltage is reduced significantly.

PAGE 82

82 CHAPTER 3 WEAK MICROCA VIT Y EFFECTS ON OLEDS 3.1 Introduction Most of OLEDs inherently uses met allic electrode as a anode or cathode or both. This intrinsic metal mirror creates weak microcavity effects in OLEDs due to Fresnel reflection at the ITO/glass and glass/air interface, see section 1.1.1.1. Weak microcavity effects were ignored in many case s in contrast to strong microcavity effect. However, recently it was found that even weak microcavity effects affect on the emission spectra and efficiencies of OLED [80 82] In this Chapter we will show that weak microcavity effects are significant on not only emission spectra but light extraction efficiency and lambertian emission pattern d epending on optical cavity length of OLEDs. 3.2. Experiment The devices used in this study have the following structure ( Figure 3 1 ) : a 85nm thick indium tin oxide (ITO) as the anode, a 10 nm thick CuPc as the hole injection layer, a 60 nm thick NPD as t he hole transporting layer, a 20 nm thick Alq 3 as the emitting layer, a 20 nm thick BCP as the hole blocking layer, a second Alq 3 layer as a buffer to tune the cavity length, and a 1 nm thick lithium fluoride (LiF) and 100 nm thick aluminum as the cathode. The thicknesses of all layers were kept constant except for the Alq 3 buffer layer, which was varied from 20 nm to 103 nm. The emitting zone was confined within the 20 nm thick Alq 3 emitting layer by a BCP hole blocking layer and the NPD electron blocking (hole transport) layer. The substrate guided mode was measured with an index matching gel (n ~ 1.52 Norland Product, NOA65 ) coated silicon photodiode in direct contact with the substrate glass. Under these conditions, the air gap between the silicon diode and the glass substrate is absent, and the light trapped in

PAGE 83

83 the glass substrate will be directly coupled into the photodiode. Hence the amount of substrate guided mode is simply the difference between the measured light intensity with (I gel ) and without ( I ext ) the index matching gel. In addition, the light intensity as a function of viewing angle was measured to confirm the l ambertian shape of the OLED emitters. The polar plot of the emitted light was calculated using the UniMCO 4.0 OLED simulation tool by UniCAD [83] The radiative emission from the recombining excitons is modeled by oscillati ng dipoles in front of a mirror. The method used in UniMCO is based on the transfer matrix formalism including additional source terms for electric field. The use of UniMCO for OLED simulation has previously been reported [81] 3.3. Results and Discussions 3.3.1. Weak M icrocavity E ffects o n t he S pontaneous E mission of OLEDs Measured emission spectra and simulated emission spectra by changing the thickness of Alq 3 buffer layer is shown in Figure 3 2. Weak microcavity effects affect on light intensity and spectra shifting. At the Alq 3 thickness of 20nm, the optical cavity length (L cav ) creates constructive interference, which has the strongest light intensity. As i ncreasing Alq 3 thickness from 20 to 103nm, increased optical cavity length has no longer constructive interference In the Alq 3 thickness of 93~103nm strong destructive interference forms in the OLED and it decreases light intensity in Figure 3 2. (a b). I n normalized spectra, peak wavelength shifts from 525nm at 20nm thick Alq 3 to 600nm at 93nm thick alq 3 by resonant cavity effects. After reaching the optical cavity length of the maximum of the destructive interference ( about 93nm thick Alq 3 ), another opti cal cavity mode is generated at the 480nm with 103nm thick Alq3 in Figure 3 2. (c d).

PAGE 84

84 3.3.2. Weak M icrocavity E ffects on the L ight E xtraction E fficiency of OLEDs The light emitted by an OLED device can be divided into three modes: the extracted mode, whic h is the light escaped from the substrate to air; the substrate mode, which is the light trapped in the substrate; and the ITO (Indium Tin Oxide)/organic mode, which is the light trapped in the ITO and the organic layers. Because of the mismatch of the ref ractive indices between air and glass, and between glass and ITO, a fraction of the generated light is trapped in the glass and the ITO/organic layer due to total internal reflection (TIR), depending on the angle of incidence of the light ray undergoing TI R at different interfaces [27] Based on classical ray optics, the light extraction efficiency ( ext ) is generally assumed to be 1/2n 2 where n is the index of refraction of the substrate [25] without consideration of weak mic rocavity effect For glass substrates, n = 1.52, and therefore ext is estimated to be about 20%. With this assumption [82] it is estimated that the extracted mode, the substrate guided mode and the ITO/organic mode should have a ratio of 1:1.8:2.9 which means 17.5%, 31.5%, and 51%, respective ly, and the dependence of device architecture has been neglected [26] This classical calculation has been used for the estimation of the internal quantum efficiency of OLED [48] Various studies of improving light out coupling in OLEDs have also based on this calculation. It is assumed that the ratio of three modes is constant independent of the device architecture [12, 26, 30 32] One of the deficiencies of the classical ray optics model is that the cavity effect has been ignored. More accurate quantum mechanical calculation was introduced [80, 82] and shows that these three modes are strongly dependent on the device geometry and the thickness of ITO [82]

PAGE 85

85 In this session we will show that due to cavity effects, the substrate extraction efficiency, ext sub_ext = I ext /(I ext + I sub ), where I ext is the extracted light intensity and I sub is the substrate guided mode intensity, is a strong function of the device geometry and it varies from 22% to 55% depending on the location of the recombination zone. T hat means the amount of light trapped in the glass substrate and the amount of light extracted are not constant, and hence the device geometry needs to be taken into consideration in order to determine the light extraction efficiency of an OLED. We will also show that OLEDs are not Lambertian emitters when the optical distance between the cathode and the emitting zone exceeds the quarter wavelength value. In addition, our simulation results indicate that as the optical cavity length exceeds two quarter wavelengths, more light is trapped in the substrate than that in the ITO/organic layers Figure 3 3 show s the ratio I gel / I ext as a function of the Alq 3 bu ffer layer thickness. T he ratio varies from 1.8 for the device with a 20 nm thick Alq 3 buffer layer to 4.45 for the device with a 103 nm thick buffer layer. This means the substrate guided mode is strongly influenced by the optical thickness of device. Fig ure 3 3 also shows the substrate extraction efficiency decreases from 55% for the device with a 20 nm thick buffer layer to 22% for the device with 103 nm thick buffer layer. These results clearly indicate that the light extraction efficiency is a strong function of the device geometry. 3.3.3. Weak M icrocavity E ffects on A ngle D ependent E mission P attern of OLEDs The angular dependence of the light intensity versus the Alq 3 buffer layer thickness is shown in Figure 3 4 The emission profiles have a Lamber tian shape for devices with a 20~40 nm thick Alq 3 buffer layer. As the buffer layer becomes thicker, the intensity maximum is found at angles off from the normal angle, indicating that the

PAGE 86

86 emitters are no longer Lambertian. For devices with 20 40 nm thick buffer layers, the distance of the recombination zone from the cathode (L rec ) is about a quarter wavelength, thus maximum light output at normal angle is achieved. Increasing the buffer layer thickness increases the optical cavity length of the device as w ell as L rec When L rec exceeds the quarter wavelength value, the light intensity at normal angle starts to decrease due to destructive interference. As L rec and the device optical path increase further, the constructive interference of higher order mode wi ll be reached at angles off the normal axis, resulting in non Lambertian emission with devices having thick (60 103 nm) buffer layers. Figure 3 5 shows the simulated electroluminescence (EL) emission pattern (the optical power per unit solid angle) integ rating all wavelengths for different Alq 3 buffer layer thicknesses The angle of 0 corresponds to the direction normal to the substrate surface while the angle of 90 corresponds to the direction parallel to the substrate. Here, 1 (= 35.3) is the critical angle at the glass substrate and air interface, and 2 (= 60.2) is the critical angle at the ITO and glass substrate interface. For comparison, the EL emission pattern for a Lambertian emitter is also shown. For a Lambertian emitter, about half of the light is extracted and half of the light is trapped in the glass substrate. For devices with different Alq 3 buffer layer thicknesses, the distribution of the EL emission pattern is shifted to larger solid angles with increasing b uffer layer thickness, and more light is trapped in the substrate and organic/ITO layers due to TIR. Polar plots of the EL emission patterns for three wavelengths, 450nm, 550nm and 650nm, are shown in Figure 3 6 It is apparent that the distribution of the EL emission pattern is shifted to larger solid angles as the wavelength decreases, resulting in more short

PAGE 87

87 wavelength light trapped in the device substrate The simulation results show that, given the present device geometry, the long wavelength light (re d and green) has higher extraction efficiency compared with short wavelength (blue). This phenomenon can be explained by optical cavity effect. For the device with a 20 nm thick buffer layer, assuming the recombination zone is within 10 nm from the Alq 3 / NPD interface, L rec is about 50 60 nm which roughly corresponds to the quarter wavelength of red and green light, and that results in maximum light output. However, for blue light, the value of L rec exceeds the quarter wavelength, and it results in lower light extraction efficiency and more light being trapped in the glass substrate. These results are consistent with the fact that the edge emitted light in Alq 3 devices tend to be bluish in color. 3.4. Conclusion In this Chapter weak microcavity effects o n OLEDs characteristics are studied. W e demonstrated that the substrate guided mode of a n OLED structure strongly depends on the optical cavity length of the device due to the weak cavity effect, and the substrate extraction efficiency varies from 22% to 5 5% depending on the device geometry Our results indicate that estimates of internal quantum efficiencies by assuming a light extraction efficiency of 20% based on classical ray optics is not constant. It changes by the optical cavity thickness of OLEDs. W e also showed that some OLEDs might not be Lambertian emitters, depending on the device structure. The results from our experimental data and simulation indicate that due to cavity effect more light is trapped in the substrate and organic/ITO layers if the distance between the recombination zone and the cathode exceeds the quarter wave value

PAGE 88

88 Figure 3 1 Device structure and measurement of substrate mode. The distance of the recombination zone from the cathode, L rec is also shown.

PAGE 89

89 Figure 3 2 (a) Measured spectra by changing Alq3 thickness, (b) Simulation spectra by changing Alq3 thickness, (c) normalized spectra of (a), (d) normalized spectra of (b)

PAGE 90

90 Figure 3 3 The ratio of the substrate guided mode to the sum of substrate and external modes ( I gel / I ext ) and substrate extraction efficiency versus Alq 3 buffer thickness.

PAGE 91

91 Figure 3 4 EL intensity as a function of angle for different Alq 3 buffer layer thicknesses

PAGE 92

92 Figure 3 5 Simulated polar plot of OLED emission patterns for devices wit h different Alq 3 1 2 are critical angles at the air/glass and glass/ITO interfaces, respectively. The emission pattern for a Lambertian emitter is also shown for comparison. Figure 3 6 Simulated polar plots of OLED emission patterns for three different wavelengths and two different Alq 3 1 2 are critical angles of air/glass, glass/ITO, respectively.

PAGE 93

93 CHAPTER 4 STRONG MICROCAVITY E FFECTS ON OLEDS The ways to enhance the luminous efficiency fall into two categories. On e increases the internal quantum efficiency such as harvesting phosphorescence, which is theoretically 4 times higher efficient than fluorescent OLEDs [18, 19] The highly efficient blue OLEDs are demonstrated by confining triplet energy and balancing char ges in the emitting zone [51, 58, 64, 84, 85] The other modifies optical structures in OLEDs to enhance the light out coupling, which enhances the light extraction efficiency by the substrate surface texturing, using microlens arrays, and scatter ing layers [26, 29 31, 45, 86] Micro cavity structure is the one of ways to increase the light extraction efficiency [87] by extracting the wave guided mode, which is trapped in organic films. due to total internal refl ection by the refractive index mismatching between air/glass substrate and glass substrate/ITO layer. 4.1. I ntroduction Micro cavity OLED was introduced as one of the ways to increase external quantum efficiency (EQE) by extracting trapped light in OLED [88 90] Micro cavity OLED has been known as its several merits such as brightness enhancement, color tenability, spe ctral narrowing and top emitting structure. In the design of active matrix OLED (AMOLED), a top emitting structure which has intrinsically microcavity structure is preferred because the thin film transistors can be buried underneath of the organic light em itting diode. Thus, complicated pixel circuits can be fabricated without diminishing the aperture ratio in a top emitting organic light emitting diode (TOLED) [91] The use of microcavities to increase the out coupling efficiency and EQE of inorganic LED has been studied [43, 92, 93] Using a planar microcavity to

PAGE 94

94 redirect spontaneous emission toward the surface, the constructive interferences can bring 15% the out coupling efficiency strongly depends on spectral width and reflectivity of cavity [43] Even without strong cavity effects, we already showed in the previous section that week cavity effects with simple bottom emitting OLED mo dified the distribution of spectral and spatial emission characteristics and changed light out coupling efficiency [94, 95] There are 3 modes in the conventional the bottom emitting OLED which are extracted mode to air, substrate guided mode trapped in glass substrate, and ITO/organic mode trapped in ITO/organic layers due to total internal reflection (TIR) by the intrinsically refractive index mismatching between air/glass substrate and glass substrate/ITO layer. The me thods for enhancing out coupling efficiency can be distinguished by which mode between substrate guided mode and ITO/organic mode is extracted. The first approach harvests the substrate guided mode out to the air by the substrate surface texturing, using m icrolens arrays, and scattering layers [26, 29 31, 45, 86] The second approach focus on extracting ITO/organic mode out to the air such as using low refractive index grids, aerogel layer and photonic crystal patterns at the glass/ITO i nterface [32, 44, 96] Micro cavity structures using metal/metal and dielectric/metal mirrors are demonstrated as the one of ways to increase the out coupling efficiency [87 90, 97] Microcavity structure redistributes the spectral and spat ial emission characteristics of OLED s Dodabalapur et al., briefly explained that the density of states such that only certain wavelengths, which correspond to allowed c avity

PAGE 95

95 [97] Peng et al., and Agrawal et al., explained that efficiency enhancement of microcavity OLED is due to increasing out coupling efficiency by reducing wave guided mode in cavity OLED [87, 89] misunderstand that high directionality of microcavity OLED is the main reas on of out coupling efficiency enhancement. Spatial change of emission pattern by directional emission to the normal axis seems to decrease the amount of substrate guided mode. In this study, we found that the directional emission to the normal angle t enhance the out coupling efficiency by extracting glass substrate guided mode but do enhance the out coupling efficiency mainly suppressing or extracting ITO/organic mode. We will also show that the relationship between the amount of trapped modes and th e change of cavity thickness and cavity strength (reflectivity of dielectric mirrors). In order to support this argument, the substrate guided mode and angular dependent spectrum are measured. It will give the insight of appropriate use of microcavity stru cture for efficient OLED design. In this Chapter we will show the microcavity OLEDs improves light intensity at the normal angle as well as all integrated angle which means EQE improvement due to enhancing the light extraction efficiency. Analysis of opt ical cavity modes in strong cavity OLEDs is also discussed here. 4.2. Experiment Dielectric mirrors made of alternate low index/high index layers, denoted (LH) n are often the only low loss or simply feasible mirror solution. Two different reflectivities o f micro cavity OLEDs with dielectric/metal cavity were fabricated to compare the strength of cavity effects on the light out coupling efficiency in Figure 4 1. 2 and 4 layers of SiO 2 /TiO 2 quarter wave stacks (2 and 4 QWS) was deposited by RF magnetron sput tering on the glass substrate and 50nm thick of Indium Tin Oxide (ITO) was

PAGE 96

96 followed with the QWSs. FIrpic doped blue phosphorescent device has a fairly broad free space emission spectrum covering the range 460~600nm. The reflectivities (R) of 2 and 4 QWSs designed as 39 % and 7 0% at 475nm wavelength to achieve the maximum efficiency enhancement at all integrated angle in Figure 4 2. In order to obtain a high value of reflectance in a multilayer film, a stack of alternate layers of high index, n H and low ind ex, n L materials is used, the thickness of each layer being ( quarter ) wavelength. The reflectance of multi QWSs is approximately given by equation 4.1 1 as follow [98] : = 2 1 2 + 1 2 (4.1) where n H n L and N are refractive index of high index material, refractive index of low index material, and the number of high/low index layer pair. Microcavity of dielectric/metal mirror using QWSs was used instead of metal/metal microcavity str ucture, since dielectric mirrors has the lower absorption and better operating stability than metal mirror such as commonly used silver and aluminum. In Figure 4 1, blue microcavity PHOLED devices with 2 and 4 layers of QWSs have been fabricated. Previousl y Tokito et al., demonstrated brightness enhancement only at normal angle using 6 QWSs [99] Strong microcavities using more than 5 QWSs have much more saturated color purity and light intens ity enhancement at normal angle, but light intensity at all integrated angle decreases. Thickness of each layer is cautiously decided to maximize the micro cavity effect at the Iridium(III)bis [(4,6 di fluorophenyl) pyridinato N,C2] picolinate (FIrpic) em ission of 475nm by UniCAD simulation [83] i.e. (glass substrate (1mm)/SiO 2 (79nm) /TiO 2 (48nm) or SiO 2 (7 9nm)/TiO 2 (48nm)/SiO 2 (79nm)/TiO 2 (48nm)/ ITO (50nm)/ PEDOT:PSS ( 25nm) / TAPC ( 50nm) / FIrpic doped

PAGE 97

97 mCP ( 20nm)/ BCP ( 40nm)/Lithium Fluoride (1nm)/Aluminum (100nm)). FIrpic 3% is doped in the emitting layer of mCP host. 4.3. Results and D iscussion 4.3.1 Str ong M icrocavity E ffects on E fficiency of OLEDs The luminous power efficiency and EQE of 3 devices are compared in Figure 4 3 ( a ). Summary of efficiency results of cavity devices is in Table 4 1. The maximum power efficiency and EQE of noncavity, 2QWS and 4QWS devices are 10.8, 20.4, 19.1 lm/W and 10.1 11.2, 14.3 %, respectively. The maximum power efficiency enhancement is X1.88 (2QWS) and X1.72 (4QWS) comparing to that of noncavity device The CIE coordinates of noncavity, 2QWS, and 4QWS devices are (0.16, 0.34), (0.122, 0.260), and (0.114, 0.242), respectively. As the spectrum of 4QWS device has more blue shifted color, its luminance decreases even the flux remains the same. The reason of lower current and power efficiency with the 4QWS device than that of 2QWS device is blue shifted spectrum of 4QWS device which has lower photopic response than 2QWS device. There is the trade off relationship between saturated blue color optimization and luminance (current and power efficiency ) due to the human eye s photo pic response Measured spectrum of noncavity, 2QWS, and 4QWS devices at the fixed current density are compared in Figure 4 3 ( b ). Spectrum of 4QWS device has more than 3 times higher peak intensity than that of noncavity device. Spectrum of 2QWS device als o has intensity enhanced spectrum. The spectrum of 4QWS device has a narrower Full Width at the Half Maximum ( FWHM ) and blue shifted spectrum than those of the 2QWS and noncavity devices Angular dependent light intensities (polar plots) of noncavity, 2Q WS and 4QWS devices are compared in Figure 4 4 2QWS and 4QWS devices have more directional

PAGE 98

98 emission to the normal direction than lambertian shape of noncavity device. All angle integrated light intensities of 2QWS and 4QWS are 11% and 41% more enhancement than that of noncavity device which means EQE increases approximately 11% and 41%, even though current efficiency and power efficiency of 2QWS device are slightly higher than that of 4QWS. EQE of 4QWS device is higher than that of 2QWS device, but its lum inous efficiency is lower than that of 2QWS device due to lower photopic response by bluer spectrum. The reason that the emission of 2QWS device is more directional than emission of 4QWS device is not that 2QWS device has stronger cavity effect but resonan t wavelength of 2QWS device locates more blue shifted than that of 4QWS due to a little thinner thickness of 2QWS. Previously we showed high EQE blue PH OLED [51] of 23% using high triplet energy of electron transporting layer, 3TPYMB, and wide bandgap host layer, UGH2. High ly efficient blue microcavity OLEDs are demonstrated based on this device architecture and the same QWSs from Figure 4 1. The micro cav ity OLEDs have the following structures: glass substrate ( 1mm )/ 2QWSs or 4QWSs /ITO (50 nm) /TAPC( 6 0 nm)/mCP(10 nm)/20% FIrpic doped UGH2(20 nm)/3TPYMB(40 nm)/LiF (1 nm) /Aluminum (100 nm)). 60nm thick TAPC used to maximize the cavity effect instead of prev ious optimized 50nm thickness here [51] The current efficiencies and EL spectra of the noncavity, 2QWS, and 4QWS devices are compared in Figure 4 5 Eve n there is a slight blue shift in the EL spectrum due to cavity effects, both the 2QWS and the 4QWS devices have higher current efficiency (2QWS : 63 cd/A, 4QWS : 53 cd/A) than that of the noncavity device (46 cd/A) at the maximum. Luminous power efficienc y of noncavity, 2QWS, and 4QWS devices are 24, 32, and 29 lm/W, respectively.

PAGE 99

99 Comparing to previous mCP host devices, luminous power efficiency enhancement is not comparable (2QWS : X1.33, 4QWS : X1.1). It is due to weaker cavity effect of the UGH2 host mi crocavity devices than the mCP host microcavity devices. The difference between the refractive index of PEDOT:PSS and TAPC decreases the reflectivity of QWSs. Comparing the light intensity between Figure 4 3 (b) and Figure 4 5 ( b ), the light intensity enh ancement in the UGH2 host devices from noncavity to 2QWS and 4QWS devices is not as high as that in mCP host devices. The FWHM of 2QWS and 4QWS devices in UGH2 host devices are wider than that of mCP host devices. It indicates that cavity strength of micro cavity with UGH2 is not as strong as that with device structure in mCP host. 4.3.2 Strong M icrocavity E ffects on E mission P rofile of OLEDs Redistribution of cavity modes is controlled by the optical cavity length. The dependence of cavity effects on the cavity length is more apparent, as cavity strength increases in microcavity device. In Figure 4 6 and 4 7 the thickness of HTL in 2 and 4 QWS devices is intentionally changed from 40 to 60nm. Figure 4 6 (a) is measured angular light intensity for 3 dif ferent HTL thicknesses from 45, 50 to 60nm. Lambertian emission pattern and emission pattern of noncavity device are also shown in Figure 4 6 (a) for the comparison. As the cavity length decreases from 60 to 45nm, angular l ambertian like phenomena was also reported as a week microcavity effects in Alq 3 OLED in the previous Chapter [94] Measured spectra for 3 different thicknesses of 4QWS devices at the normal angle and noncavity device are shown in Figure 4 7 (b). 4QWS device with thick HTL (60nm) has longer resonant cavity length and it shifts spectrum to greenish

PAGE 100

100 (513nm). T he changes of optical cavity length alter the spectral and spatial free space emission of extracted mode. More detail explanation is in the section 4 3.4 4.3.3 Strong M icrocavity E ffects on L ight E xtraction E fficiency of OLEDs In Figure 4 8 measured I ge l /I air of noncavity, 2QWS, and 4QWS OLED was shown by changing the thickness of TAPC layer (hole transporting layer) using the introduced method in above. As the thickness of HTL change from 30 to 55nm, resonant peak wavelength shifts from 455 to 501 nm wa velength. Inset of Figure 4 8 shows measured free space spect ra of 4QWS devices by changing the thickness of HTL from 30 to 55nm. The increased resonant cavity length by changing the thickness of TAPC layer shifts resonant peak wavelength to longer wavele ngth. With 30nm thick HTL, the peak wavelength of resonant cavity mode locates less than 470nm (should be the intrinsic FIrpic emission spectrum doesn t exist at lower wavelength than 470nm. It i s noticeable that the spectrum of 4QWS device with 30nm thick HTL is even broader than that with 37.5nm thick HTL due to the week emission at 500nm This will be discussed later as a small portion of 2 nd cavity mode in the section 4.3.4 The I gel /I air rati o of noncavity device (HTL thickness) It means that the amount of substrate guided mode is about 70 80 % of extracted mode ying the cavity length. 2QWS devices also have quite similar I gel /I air ratio to noncavity device. The cavity effect of 2QWS devices is not strong enough to modify the distribution of substrate guide mode in OLED. 4QWS devices show high dependence of I gel /I air ratio on the resonant cavity thickness. Reflectivity (70%) of 4QWS generates strong cavity effect to redistribute the substrate guided mode in the OLED. I gel /I air ratio varies from 2.56 to 1.74

PAGE 101

101 by changing the thickness of HTL from 30 to 55nm. The amou nt of substrate guided mode in 4QWS device varies from 156 to 74% comparing to the extracted mode by changing the optical cavity length. We noticed that efficiency of 4QWS devices also increases by changing the HTL thickness from 30 to 55nm. It is attribut ed to the increased amount of the substrate guided mode lowering the efficiency of microcavity device With 40~60nm thick HTL, EQE of 4QWS device increased even with similar or larger amount of the substrate guided mode. This means that efficiency enhancem ent of microcavity structure is not due to out coupling of the substrate guided mode. It would be due to suppressing the ITO/organic mode by redistributing spatial cavity mode and generating the destructive cavity mode in ITO/organic mode. 4.3.4 Analysis of C avity M odes Cavity strength defined by reflectivity of QWS affects the generation and distribution of cavity modes in microcavity OLED. As increasing the reflectivity of QWS, 2 nd and 3 rd cavity mode are generated in microcavity OLEDs. Schematic diagram of cavity mode distribution by changing the cavity strength is shown in Figure 4 9 Cavity modes of noncavity, 4QWS and 6QWS devices are compared for the comparison of distribution of cavity modes. Among the extracted, substrate guided, and ITO/organic mo des, only extracted mode is out coupled to the air, the escap ed emission patterns ( blue colored circle or ellipse) marked by orange circle in Figure 4 9 (a). Because of the mismatch of the refractive indices between air (n air =1) and glass (n glass =1.52), an d between glass and ITO (n ITO =1.9), a fraction of the generated light is trapped in the glass substrate and the ITO/organic layer (n organic = 1.7) due to total internal reflection (TIR), depending on the angle of outgoing of the generated light at the emit ting zone 1 2 (60.2) is the critical angles of

PAGE 102

102 law, shown in Figure 4 9 (a). As the cavity strength increases from n oncavity to 4QWS and 6QWS devices, patterns of the free space emission changes from lambertian to strong directional emission (upper diagrams of Figure 4 9 ). In contrast to noncavity device, 2 nd cavity mode is generated in 4QWS device and 3 rd cavity mode might be created in 6QWS device. In Figure 4 9 (b) the peak of 2 nd cavity mode is located in the 1 2 which means that some amount of substrate guided mode is created by the 2 nd cavity mode. In Figure 4 9 (c) 2 nd and 3 rd cavity modes are created in the substrate guided mode and ITO/organic mode. It also increases the amount of trapped in the substrate guided mode and ITO/organic mode, leads to drop the EQE of any EQE enhancement in Alq 3 OLED, although emission to the normal direct ion is enhanced [100] In Figure 4 8 4QWS device shows the large fraction of the substrat e guided mode. This is the consistent case with schematic diagram in Figure 4 9 (b). Schematic diagrams of each 4QWS device with 40, 50 and 60nm HTL thickness are shown in Figure 4 10 (a c) to explain how cavity length to affect spatial emission and the o ut coupling efficiency of microcavity OLED in Figure 4 10 (a). With 60nm thick HTL, cavity length of 4QWS device is longer than optimized cavity length in Figure 4 10 (a). Maximum peak of 1 st cavity mode located off from the normal axis (0) makes the pola Batwing This Batwing shaped emission pattern is shown in Figure 4 7 (a). For the maximum efficiency of microcavity devices, cavity condition in Figure 4 10 (a) is desirable, since this redistribution of cavity mode reduces the substrate guided mode and maximizes the extracted mode. Spectrum shifting to

PAGE 103

103 green is drawback to consider. With 50nm thick HTL, maximum peak of 1 st cavity mode positions on the exact normal axis. It is noticeable that 2 nd cavity mode is begin to shif t toward the normal axis and forms some amount of substrate guided mode In Figure 4 10 (c), cavity length is too thinner than the optimized thickness. Even at the normal direction, the first cavity mode doesn t make the maximum constructive interference a nd it loose some efficiencies. 2 nd and even 3 rd cavity mode are clearly created at substrate guided mode and ITO/organic mode. In this optical cavity condition, the large amount of the substrate guided mode and ITO/organic mode is generated and some amount of extracted mode is suppressed. It has more directionality than 50nm thick 4QWS device, but total light intensity escaping to the air decreases. This is why the amount of substrate guided mode in 4QWS device with thin HTL increases in Figure 4 8 In ord er to confirm the presence of 2 nd cavity mode, angular dependent spectrum of 4QWS device is measured in Figure 4 11 Angular emission intensity of 4QWS device strongly depends on the viewing angle due to strong cavity effect on altering optical cavity leng th. Normalized spectrum is shown in Figure 4 11 (b). 510nm peak of greenish spectrum is measured at the glance viewing angle. This is the direct evidence of 2 nd cavity mode rise. Most of 2 nd cavity mode is trapped in the substrate guided mode but a small p ortion of 2 nd mode can be measured at only around the first critical angle 1 ), which is almost 90 to the normal angle from glass surface. In schematic diagram of Figure 4 11 (b) most of free space emission disappears by destructive interference and the very small amount of 1 st and 2 nd cavity modes co exist abo 1 Angular dependent color spectrum is shown in Figure 4 11 (b). Spectrum of 2 nd cavity mode is

PAGE 104

104 more greenish spectrum not blue spectrum. It is explained that the spectrum of substrate guided mode trapped by 2 nd cavity mode is more greenish color Op tical simulation to confirm the 2 nd cavity mode is performed using UniCAD 4.0 in Figure 4 12 Figure 4 12 (a) shows comparison of only extracted mode between noncavity and 4QWS devices. Once again, 4QWS has much more directional emission to normal viewing angle. Figure 4 12 (b) is simulated inside of the substrate. Angular light intensity in Figure 4 12 (b) includes both of the extracted and substrate guided mode. 4QWS device has much higher substrate guided mode, which is generated by 2 nd cavity mode. I TO/organic mode is not shown here due to the limit ation of Unicad simulation. 4.4 Conclusion The blue microcavity OLEDs with 2 and 4QWS were fabricated using SiO2/TiO2 dielectric mirrors. The strong cavity effects on the out coupling efficiency of each 3 devices were explained using measurement of the amount of substrate guided modes and its schematic diagrams. It is shown that the amount of substrate guided mode of microcavity devices is similar or larger than that of noncavity device. In addition to cav ity effects on the mode distribution, spectral and spatial changes of the free space emission in the microcavity devices are discussed here. Angular dependent emission bat l cavity thickness. It will give the insight of appropriate use of microcavity structure for efficient OLED design. Efficiency enhancement of blue microcavity devices is also shown. The maximum current efficiency and luminous power efficiency of microcavit y OLEDs with 2 and 4 QWS devices are 32.3, 30.9 cd/A and 20.4, 19.1 lm/W, which are 88% (2QWS) and 72% (4QWS) higher than that of noncavity device. External quantum

PAGE 105

105 efficiency (EQE) enhancement of 11% (2QWS) and 41% (4QWS) was achieved. Finally highly effi cient blue microcavity OLED of 63 cd/A current efficiency is demonstrated even with more saturated blue spectrum than the noncavity OLED. This is attributed to extracting ITO/organic mode not substrate guided mode.

PAGE 106

106 Table 4 1. Summary of the maximum curre nt efficiency, luminous power efficiency, external quantum efficiency and CIE coordinate of Noncavity, 2QWS, and 4QWS devices. Current efficiency (cd/A) Power efficiency (lm/W) EQE (%) CIE (x, y) Noncavity 23.7 10.8 10.1 (0.16, 0.34) 2QWS 32.3 (X1.36) 20.4 (X1.88) 11.2 (X1.11) (0.122, 0.260) 4QWS 30.9 (X1.3) 19.1 (X1.72) 14.3 (X1.42) (0.114, 0.242)

PAGE 107

107 Figure 4 1 Structure of blue Microcavity OLED and its 2, 4 quarter wave stacks

PAGE 108

108 Figure 4 2 Measured r eflect ivities of 2, 4 layers of qua r ter wave stacks with 50nm thick ITO and aluminum. Most of FIrpic emission is from 470 to 500nm.

PAGE 109

109 Figure 4 3 (a) The power effic i ency and external quantum efficiency of noncavity, 2QWS, and 4QWS blue OLEDs with mCP host. EQE of 2QWS an d 4QWS are calculated from measured angular dependent light intensity. Figure 4 3 (b) Measured spectrum of noncavity, 2QWS, and 4QWS blue OLEDs at the 0.4 mA/cm 2

PAGE 110

110 Figure 4 4 Measured angular dependent light intensity (the number of photons) of non cavity, 2QWS, and 4QWS blue OLEDs

PAGE 111

111 Figure 4 5 (a) Current and power effic i ency of noncavity, 2QWS, and 4QWS blue OLEDs with UGH2 host. Figure 4 5 (b) Measured spectr a of noncavity, 2QWS, and 4 QWS blue OLEDs at the 0.4 mA/cm 2

PAGE 112

112 Figure 4 6 (a) Measured angular dependent light intensity and (b) those measured spectrum at the normal direction (b) by changing the thickness of HTL from 40 to 45, 50nm on 2QWS PHOLED. Emission patterns of noncavity device and lambertian are shown t ogether for the comparison. Figure 4 7 (a) Measured angular dependent light intensity and (b) those measured spectrum at the normal direction (b) by changing the thickness of HTL from 45 to 50, 60nm on 4QWS PHOLED. Emission patterns of noncavity devic e and lambertian are shown together for the comparison.

PAGE 113

113 Figure 4 8 The ratio of the substrate guided mode to the sum of substrate and extracted modes (I gel /I air ) versus HTL (TAPC) thickness for each noncavity, 2QWS, and 4QWS devices. Peak of the 1 st re sonant cavity wavelength is also shown in bottom axis of Figure Inset of Figure is measured and normalized spectrum of 4QWS with 4 different HTL thicknesses from 32.5 to 56nm. Normalized spectrum of 4QWS is also shown in the inset for comparison.

PAGE 114

114 Figu re 4 9 Schematic diagram of mode distribution of (a) noncavity, (b) 4QWS, and (c) 1 2 are critical angles of TIR for substrate guided mode and ITO/organic mode. 1 st 2 nd and 3 rd cavity mode generated by strong cavity effect of 6QWS device is shown in Figure (c). Free space emission pattern escaping to air (extract ed mode) for each 3 devices are shown in upper diagram. Color changing in Figure (b) is to illustrate color spectrum dependence on the optical cavity length, which is correspondent to the outgoing angle of emitted light.

PAGE 115

115 Figure 4 10 Schematic diagra m of cavity mode distribution in 4QWS device with 3 different HTL thicknesses of 60 (a), 50 (b), and 45nm (c). Figure 4 11 Measured angular dependent spectra (a) and its normalized spectra (b) of 4QWS device. A small fraction of the 2 nd cavity mode i s observed at the glancing angle.

PAGE 116

116 Figure 4 12 Unicad simulation of angular intensity at free space (a) and at the oranic medium (b) for noncavity and 4QWS devices.

PAGE 117

117 CHAPTER 5 DOWN CONVERSION WHIT E OLED White organic light emitting diodes are of int erest and potential importance for use as solid state lighting sources. In very recent decade, tremendous progress has been made in white OLEDs, since the discovery of phosphorescent OLEDs has opened the serious use of OLEDs for the illumination white ligh t by its intrinsic high internal quantum efficiency [8, 9, 12, 14, 34, 38, 41, 101] Reineke et al. demonstrated 90lm/W white OLED at 1000cd/m 2 using multiple phosphorescent emitting layers with 2.4 times enhancement by the light extraction schemes of high refractive index substrates and macrolens [9] Universal Display Corporation announced 102 lm/W with color rendering index of 70 on white OLEDs at 1000cd/m 2 using light extraction enhancement [102] These recent promising W OLEDs has been fueling the developing flat large area lighting illumination, which can be an alternative light source of the fluorescent tube. 5.1 Introduction Several approaches ha ve been introduced for WOLED s. One approach uses multiple doping within the same emitting layer using the same host materials such as RGB (red, green and blue) or BO (blue and orange) dopants in Figure 1.2 (a) [12, 34] In this process, device architecture is identical to monochromatic OLED except multiple doped emitting layer, but it needs extremely cautiously controlled doping concentrati ons of RGB dopants in the emitting layer. E lectroluminance spectra of triple or double doped white OLED are greatly affected by applied current density (brightness). Second approach uses multiple ( separate ) RGB emitting layers within devices in Figure 1.2 (b) [9, 38] In this approach, since electroluminance spectra are greatly affected by charge balance in 3 emitting layers. It needs very careful control for each thicknesses of

PAGE 118

118 RGB emitting layers and the concentration of RGB dopants. In the previous two approaches, it is also very difficult to have operational color stability due to differential aging of the RGB dopants. Recently, the hybrid emitting lay ers using fluorescent blue and phosphorescent green and red have been introduced to overcome the color shifting by the operating age [38] Third approach uses a single color emitting OLED in the combination with a down conversion layer in Figure 1.2 (c) [15, 41] After Schlotter et al. in troduced white light emitting devices based on an inorganic blue light emitting LED with down conversion phosphor [103] down conversion technique is very commonly used in cost effective white LED. Duggal et al. implemented down conversion concept to the field of OLEDs [15] In 2006, Krummacker et al. published 25 lm/W white OLED based on blue phosphorescent polymer device [41] The generation of white light is achieved by the combination of non absorbed blue wavelength photons from blue emitting OLED and re emitted yellow or r ed wavelength photons by the down converting phosphor layers Tuning color is easily obtained by the selection of phosphors (the combination of yellow and red phosphors) and phosphor film thickness. The spectrum of under various a pplied current density (brightness) and device aging controls of multiple doping or emitting layer is also very appropriate for cost effective applications of large scale lighting In the Chapter 2, we have shown highly efficient blue phosphorescent OLEDs, which have the power efficiency of 50lm/W at the maximum Based on this highly efficient blue PHOLEDs, here we have fabricated the efficient blue microcavity OLEDs of 41 lm/W wi th more saturated blue color using reflective metal cathode and dielectric

PAGE 119

119 mirror s made with quarter wave stack on the glass substrate. Then, down conversion phosphors are applied on the top of glass substrate of blue microcavity OLEDs. We obtained the hig h power efficiency of 73, 61 lm/W with yellow phosphor and 67.7, 57 lm/W with red:yellow phosphor mixture at 20 and 100 cd/m 2 on the blue microcavity OLED with 4 layers of QWS mirrors. Finally, macrolens is used for enhancement of light out coupling effici ency. Obtained power efficiency of down conversion white OLED are 107, 96 lm/W (CRI : 59, CIE : 0.36, 0.50) with yellow phosphor and 99.3, 8 7 lm/W (CRI : 83, CIE : 0.43, 0.46) with mixture of red:yellow phosphors at 3 0 and 100 cd/m 2 on the blue microcavit y OLED with 4 layers of QWS mirrors. 5.2 Concept of D own C onversion W hite L ight I llumination with B lue M icrocavity OLEDs High ly efficient blue emitting OLEDs are essential for efficient down conversion W OLEDs since device efficiency of down conversion WO LED is determined by the combination of non absorbed blue photons by blue device and re emitted yellow or red photons by phosphor films. The absorption spectrum of the yellow phosphor has the peak at the range of 450~500 nm. Quantum Yield (QY) of this yell ow phosphor decreases below 80% beyond 510nm. However, the intrinsic emission spectrum of FIrpic device is quite broad from 450 to 600 nm in Figure 5 1 (a). It leads to lower the down conversion efficiency. Use of micro cavity structure makes sky blue emis sion of FIrpic doped PHOLED shift to more saturated blue emission. Microcavity structure redistributes the photon density of state from green photons (over 500nm) to blue photons (460 ~ 500nm) by optical resonant cavity effects. It leads to high down conve rsion efficiency in Figure 5 1 (b). Finally efficient white light is generated by incorporating blue microcavity OLEDs with down conversion films in Figure 5 1 (c). In

PAGE 120

120 Figure 5 2 QY of yellow phosphor and normalized spectra of noncavity, 2QWS, and 4QWS mi crocavity PHOLED are plotted together. Calculated QYs are 56.3, 63.7, and 64.3% for noncavity, 2QWS, 4QWS blue microcavity PHOLED with taking into account the entire emission spectra of blue emission region (from 450 to 600nm). More saturated blue emission of 4QWS microcavity structure increases QY of yellow phosphor about 14% comparing to noncavity device (from 56.3 to 64.3%). In the previous Chapter blue microcavity enhances the EQE of 41% at the maximum with more saturated blue emission by extracting ITO /organic mode Strong blue emission of microcavity device is appropriate for efficient down conversion WOLED. The amount of the substrate guided mode in the strong microcavity device (4QWS) is much larger than noncavity blue device. Down conversion layer i s known to increase the out coupling efficiency by the volumetric scattering of phosphor particles Phosphor particles in down con version films scatter the incoming blue photons and reduce TIR at the air/substrate interface [45] Th e detail light extraction analysis of phosphor film will be dis cussed in the section 5.3.5. 5. 3. Experiment and discussion 5.3.1. Down C onversion P hosphor F ilms Yellow and red phosphors from commercial suppliers under a nondisclosure agreement were used for down conversion of blue to white light T herefore the charac teristics that can be reported are limited. Photoluminescence (PL) and photoluminescence excitation (PLE) spectra, CIE coordinates QYs of phosphors and phosphor thin film were determined using a JASCO FP6500 spectrophotometer equipped with an integrating sphere. The down converted white light was also characterized using a SpectraScan 650 Photo Research camera. Absorption and

PAGE 121

121 emission spectra of y ellow emitting and red emitting phosphors were shown in Figure 5 3 In Figure 5 3 (a) yellow phosphor has the maximum absorption from 450 to 470nm wavelength The PL peak from the yellow emitting phosphor was centered at 560nm, and the QY for thin film samples was 89 % at the 460nm and the CIE coordinates are in the yellow at (0.446, 0.541). In Figure 5 3 (b) red phosphor has the maximum absorption from 450 to 520nm. The emission from the red emitting phosphor was centered at 610 nm and t he QY of the red phosphor was 84% at 500nm. The CIE coordinates are (0.60, 0.39) When a mixture of the red and yellow phosp hors was used in the silicone, the CRI of the down converted white light increased from 72 to 83 Photoluminescent QYs of yellow, red, and yellow:red mixture phosphor films are plotted in Figure 5 4. Phosphor thin films were prepared using two different host materials: (1) polymethyl methacrylate (PMMA; Molecular Weight = 950,000; solution of 4 wt% in chlorobenzene), and (2) silicone. The refractive indices of the PMMA and silicone were 1.40 and 1.44, respectively. The silicone matrix exhibited superior p roperties and therefore the data reported below are from silicone matrix based devices, unless noted otherwise. Phosphor powders were thoroughly mixed into the transparent silicone (Nusil Silicone Technology, LS 3140P) and manually doctor bladed onto eithe r microslide glass (Corning Cover glass, 25 mm x 25 mm, ~0.2 mm thick), or directly onto the glass substrate of the OLED sample. The type and weight of phosphor and weight of silicone used for different down conversion thin films are listed in the Table 5 1. After the dispersed phosphor layer was added, the coated glass or coated OLED substrates were baked at 70C for 2 h ours in air. When a phosphor coated glass slide

PAGE 122

122 was used for down conversion, an index matching gel was used between the phosphor coated m icroslide and the glass supporting the OLED. Either 30mg of yellow emitting or 20/10mg of yellow/red emitting phosphor was diluted to either 60, 150 or 330mg of silicone, as listed in Table 5 1. The effects of the loading of phosphor on down conversion wil l be discussed in the device integration section below. Stereoscopic photomicroscopy (Olympus BX60) and scanning electron microscopy (SEM, JEOL JSM 6400) were used to characterize the phosphor dispersion and to measure the film layer thickness on glass and OLED substrates. A SEM cross section micrograph of a down conversion film is shown in Figure 5 5 (a) for sample Y180 High resolution micrographs of the dispersed phosphor are shown in Figure 5 5 (b) & 5 5 (c). The uniformity of the thickness of the fil ms was measured using a profilometer (Tencor P2 long scan), and the thicknesses reported in Table 5 1 are averages from at least four different areas across the sample. 5.3.2. Fabrication of D own C onversion W hite OLEDs First of all, m icro cavity blue PHOL EDs were fabricated on the two different dielectric mirrors: two layer quarter wave stacks with reflectivity of 0.39 (R 2QWS = 0.39), and 4 layer quarter wave stacks with reflectivity of 0.7 at 475nm (R 4QWS = 0.7). The thickness of each dielectric layer is designed to maximize the micro cavity effect at 475nm by the optical OLED simulation [83] The devices hav e the following structure: glass substrate (1mm)/SiO 2 (79nm)/TiO 2 (48nm) or SiO 2 (79nm)/TiO 2 (48nm)/SiO 2 (79nm)/TiO 2 (48nm)/ITO (50nm)/ PEDOT:PSS (60nm)/ TAPC (25nm)/ mCP (25nm) with 20 wt % FIrpic/ 3TPYMB (40nm)/ CsCO 3 (0.8nm)/ Aluminum (100 nm) in Figure 5 6

PAGE 123

123 Thickness of PEDOT:PSS layer cautiously was optimized to achieve the maximum resonant cavity effect. Current efficiency and power efficiency without down conversion phosphor films are shown in Figure 5 7 (a). Spectra of blue micro cavity PHOLED are s hown in Figure 5 7 (b). The maximum current and luminous power efficiencies of noncavity, 2QWS, and 4QWS OLEDs was 44.3, 45.7, 43.8 cd/A and 39.7, 39.9, 41.1 lm/W. In Figure 5 7 (b), spectra of 2QWS and 4QWS devices become narrow and more bluish. Power effi ciency more blue shift spectra of microcavity devices has ~20% stronger integrated light intensity than spectrum of noncavity device due to reduced luminance of blue shif ted of noncavity, 2QWS, and 4QWS devices are (0.15, 0.32), (0.14, 0.23), and (0.15, 0.22), respectively. Three different thicknesses of yellow and yellow:red phosphor mi xture f ilms are prepared on the separate glass slide (0.25 mm thick), which are described in the previous section. Phosphor films are applied to blue micro cavity PHOLED with refractive index matching gel to minimize the total internal reflection (TIR) in Figure 5 6 Finally, macrolens with index matching gel is applied on the top of down conversion phosphor film to increase light out coupling efficiency with index matching gel 5.3.3. High E fficienc ies of D own C onversion W hite OLEDs Current efficiency and power efficiency of integrated down conversion W OLEDs are shown in Figure 5 8 ~ 5 10 Current efficiencies, p ower efficienc ies, CIE coordinates and CRI with/without phosphor films and macrolens are compared between noncavity and 4QWS microcavity devices in T able 5 2. Y180 and YR180 phosphor films are used

PAGE 124

124 here (See Table 5 1 for phosphor film information). Efficiency enhancement by applying the macrolens are 46 and 31% for yellow and yellow:red mixture phosphor films Power efficiencies of down conversion WO LED with noncavity device are 37lm/W with Y180 phosphor and 32lm/W with YR180 phosphor at 100 cd/m 2 which means no enhancement of power efficiency from pristine noncavity devices (38lm/W at 100 cd/m 2 ). In contrast to noncavity device, p ower efficienc ies o f down conversion WOLED with 4QWS microcavity shows 61lm/W with Y180 phosphor and 57 lm/W with YR180 phosphor at 100 cd/m 2 which is 67 ~79 % enhancement from pristine blue microcavity device (34lm/W) at 100 cd/m 2 This efficiency enhancement only by 4QWS m icrocavity device is attributed to 3 reasons. One is that blue shifted spectrum of microcavity device has more efficient down conversion than noncavity device. As discussed i n the section 5.3.1, calculated QYs over all integrated wavelength of 4QWS microca vity device (64.3%) used down conversion was 14% higher than the QY of noncavity device used down conversion (56.3%). A s econd reason is that the number of blue photons created by blue microcavity is much larger than noncavity device. In Chapter 4, we show ed that EQE of blue microcavity device is 1 0~40% higher than noncavity blue device. Blue microcavity PHOLED used here might have about 20% higher EQE than noncavity blue PHOLED. It leads high efficiency of down conversion WOLEDs. A t hird reason is the diff eren t amount of the substrate guided mode between noncavity and 4QWS devices In the previous Chapter it is explained that the redistribution of optical cavity mode by the microcavity structure increases the amount of trapped substrate guided mode for 4QW S device This trapped substrate guided mode is utilized and harvested by the integration

PAGE 125

125 of light scattering phosphor layer on the substrate. It is known that phosphor films increases the light extraction efficiency from 4 0~10% depending on the concentrat ion and thickness of phosphor films [45] T he detail of the light extraction efficiency and the substrate guided mode of down conversion WOLED will be discussed in the section 5.3.5. 5.3.4. The CIE and CRI of D own C onversion WOLED Normalized spectra of down conversion from with blue micro cavity PHOLEDs are shown in Figure 5 1 1 As changing yellow phosphor films from Y90 (95um) to Y360 (450um) blue spectra from 460~500nm begin to decrease by phosphor absorption and yellow spectrum around 560nm increases. For the yellow:red phosphor mixture, the two emissions of yellow (560nm) and red (610nm) emission spectrum are overlapped. Down conversion spectra with 2QWS and 4QWS microcavity devices still possess the large amount of blue photons which is not absorbed by phosphor particles due to the large num ber of blue photons generated by strong microcavity effects CIE coordinates of down conversion devices were measured using JASCO system and shown in Figure 5 1 2 CIE coordinate of noncavity device with yellow phosphor films shifts from (0.15, 0.32) to (0. 27, 0.46), (0.34, 0.52), (0.38, 0.54) by changing yellow phosphor films from Y 90 to Y 180, Y 360 and CIE of noncavity device with yellow:red phosphor mixture films shifts to (0.36, 0.46), (0.42, 0.47), (0.49, 0.47) by changing yellow:red phosphor mixture fil ms from YR 90 to YR 180, YR 360 in Figure 5 12 (a). In the same ways, CIE coordinates of 4QWS device with yellow phosphor films shifts from (0.15, 0.22) to (0.28, 0.42), (0.36, 0.50), (0.50, 0.46) and CIE of 4QWS with yellow:red phosphor mixture films shifts from (0.15, 0.22) to (0.38, 0.44), (0.43, 0.46), (0.50, 0.46) in Figure 5 1 2 (b). This blue shifted spectrum of microcavity devices move

PAGE 126

126 down y value of the CIE coordinate. In order to achieve pure white light (0.33, 0.33), deeper blue device (~460nm peak wavelength ) or increased concentration of red phosphor are required. 5.3.5. L ight E xtraction E fficiency on D own C onversion W OLED In previous Chapter we discussed that how much blue shifted spectrum and increased the number of blue photons in microcavity device help to achieve an efficient down conversion WOLED. The purpose of this section is the analy sis of the difference of the light extraction efficiency between noncavity and microcavity down conversion WOLEDs, which is the third factor of efficient dow n conversion WOLED with microcavity device. As mentioned before, mac rolens was used for extracting the substrate guided mode in phosphor film s. Efficiency enhancements by macrolens 42% for yellow phosphor and 31% for yellow:red mixture phosphor. Efficiency enhancement by macrolens is known as 70~80% for an OLED device [9] therefore the discrepancy between efficiency enhancement of 70~80% without phosphor films and 40~30% with phosphor films is attributed to the present out coupling by the volumetric light scattering of phosphor particles [41, 45, 104] In order to confirm the influence of volumetric light scattering to efficiency of down conversion WOLED, the red emitting PQIr doped device is additionally fabricated to distinguish the each contribution by absorption/reemission and by the volumetric light scatter ing of the phosphor films Figure 5 1 3 shows QY of yellow phosphor and normalized spectrum of PQIr device. There is clearly no absorption of PQ Ir emission by yellow phosphor, therefore PQIr emission doesn t experience down conversion process (absorption a n reemission). The measured photocurrent of photodiode is shown in Figure 5 1 4 The photocurrents without phosphor films are set

PAGE 127

127 as a reference 1 As applying yellow phosphor films from Y90 to Y180, Y360, measured photocurrent decreases 8, 12, 27% in red e mitting PQIr device. It indicates that the light extraction enhancement by the volumetric scattering of phosphor particles is negligible or some amount of photons is trapped or lost in phosphor films used here. Photocurrent drops of blue noncavity and 2QWS device are much bigger than that of PQIr device. It might be due to the photo n loss in the absorption/reemission process of phosphor particles. In the section 5.3.1, QYs of down conversion with noncavity and 2QWS devices are only 56.3 and 64.3%. Blue phot ons of 44~36% encountering down conversion process will be lost. Only 4QWS device shows the photocurrent enhancement by applying phosphor films. Comparing to 2QWS device, this photocurrent enhancement is mainly due to the out coupling of the substrate guid ed mode trapped in the 4QWS device. In order to confirm the contribution of the substrate guided mode in 4QWS device, the substrate guided mode is measured by means of direct photodiode contact with index matching gel, which is already explained in the Cha pter 3 and 4 in Figure 5 15 The I gel /I air ratio of 4 devices with yellow phosphor films are shown in Figure 5 16. The measured I gel /I air ratio of 4QWS device without phosphor films is close to 2.6, which means that the substrate guided mode of 4QWS device is 160% comparing to the extracted mode (I air ) This amount of the substrate guided mode of 4QWS device is 2 times larger than that of other 3 devices; the measured I gel /I air ratio of PQIr, noncavity, 2QWS devices is 1.8~1.9, which means that the substrat e guided modes of these devices are 80~90% comparing to the extracted mode. After the integration with phosphor films, all devices have similar amount of the substrate guided mode; I gel /I air ratio varies between 1.6 and 1.3. It indicates that some amount o f trapped

PAGE 128

128 substrate guided mode in the 4QWS device is out coupled by v olumetric scattering of phosphor films. 5.4 Conclusion High ly efficient d own conversion phosphors are demonstrated with blue microcavity OLEDs We obtained the high power efficiency o f 73, 61 lm/W with yellow phosphor and 67.7, 57 lm/W with red:yellow phosphor mixture at 3 0 and 100 cd/m 2 on the blue microcavity OLED with 4 QWS microcavity device Finally, macrolens is used for enhancement of light out coupling efficiency. Obtained powe r efficiency of down conversion white OLED are 107, 96 lm/W (CRI : 59, CIE : 0.36, 0.50) with yellow phosphor and 99.3, 8 7 lm/W (CRI : 83, CIE : 0.43, 0.46) with the mixture of red:yellow phosphors at 3 0 and 100 cd/m 2 on the blue microcavity OLED Blue m icrocavity PHOLED with 4QWS dielectric mirror is required for efficient down conversion WOLEDs. Three advantages of blue microcavity PHOLED for down conversion are discussed here. One is that blue shifted spectrum of microcavity device compromises high QY of phosphor. The second is that the number of blue photons created by blue microcavity increases efficiency of down conversion WOLEDs. The third is that phosphor films extract out the more trapped substrate guided mode in 4QWS device.

PAGE 129

129 Table 5 1 Weight o f components in different down conversion thin film samples Label Silicone A (mg) Silicone B (mg) Yellow Phosphor (mg) Red Phosphor (mg) Total Phosphor Wt (mg) Film Thickness Y90 30 30 30 -90 95 YR90 30 30 20 10 90 Y180 75 75 30 -180 220 YR180 75 75 20 10 180 Y360 165 165 30 -360 450 YR360 165 165 20 10 360

PAGE 130

130 Table 5 2 Summary of current and power efficiencies for down conversion of nonc avity and 4QWS microcavity OLEDs at 30, 100, and 1000 cd/m 2 CIE and CRI are also given here.

PAGE 131

131 Figure 5 1 Schematic diagram of down conversion W OLED with microcavity structure.

PAGE 132

132 Figure 5 2 Quantum yield of yellow phos phor and normalized spectra of noncavity, 2QWS, 4QWS microcavity PHOLED. Figure 5 3 (a) Photoluminescence and photoluminescence excitation spectra from yellow phosphor and (b) red phosphor

PAGE 133

133 Figure 5 4 Photo luminescent quantum yield for yellow, red and yellow:red mixture phosphor films.

PAGE 134

134 Figure 5 5 (a) SEM micrograph of cross section of Y30S150 thin film sample (scale bar resolution SEM micrographs of phosphor particl es incorporated in the silicone

PAGE 135

135 Figure 5 6 Device structure of blue microcavity PHOLED with down conversion phosphor. The p hosphor film and m acrolens were attached with index matching gel.

PAGE 136

136 Figure 5 7 (a) Current efficien cy and power efficiency of noncavity, 2QWS, and 4QWS PHOLEDs before applying phosphor films Figure 5 7 (b) Normalized spectra of noncavity, 2QWS, and 4QWS PHOLEDs before applying phosphor films

PAGE 137

137 Figure 5 8 Current and po wer efficiency of blue noncavity PHOLED and its down converted efficiencies. Efficiency enhancement with macrolens is also shown here. Open symbol s are power efficiency.

PAGE 138

138 Figure 5 9 Current and power efficiency of blue 2QWS PHOLED and its down converted efficiencies. Efficiency enhancement with macrolens is also shown here. Open symbol s are power efficiency.

PAGE 139

139 Figure 5 10 Current and power efficiency of blue 4QWS PHOLED and its down converted efficiencies. Efficiency enhancement with macrolens is also shown here. Open symbol s are power efficiency.

PAGE 140

140 Figure 5 1 1 Normalized spectra of down conversion white OLEDs. 3 different thicknesses of yellow phosphor films and yellow:red phosphor mixture films are applied (a) noncavity with phosphor films, (b) 2QW S with phosphor films, (c) 4QWS with phosphor films

PAGE 141

141 Figure 5 1 2 CIE coordinate of noncavity device with incorporating to phosphor films (a), 4QWS device with incorporating to phosphor films (b).

PAGE 142

142 Figure 5 1 3 Quantum yield of yellow phosphor and normalized spectrum of red emitting PQIr device

PAGE 143

143 Figure 5 1 4 Measured r elative photocurrent of photodiode for PQIr, noncavity, 2QWs, and 4QWS devices. Photocurrent without phosphor film is set as 1 for comparison.

PAGE 144

144 Figure 5 1 5 Measurement of the substrate guided mode. See Chapter 3 and 4. Figure 5 1 6 I gel /I air ratios for PQIr, blue noncavity, 2QWS, and 4QWS devices.

PAGE 145

145 CHAPTER 6 C ONCLUSION S AND FUTURE WORK 6.1 Conclusion s Several approaches have been introduced for th e generation of white light emitting OLEDs (WOLED) One approach uses multiple doping (RGB) within the same emitting layer using the same host material The other uses separate doped RGB multiple emitting layers Recently Reineke et al. demonstrated 90 lm/ W WOLED at 1000 cd/m 2 based on multiple emitting layer approach [9] However, this process requires extremely cautiously controlled doping concentrations of RGB dopants and emissive layer thicknesses E lectrolumin escent spectra of these approaches are greatly affected by charge balance, applied current density (brightness). Down conversion WOLED using yellow and red phosphor is introduced as a alternative approach to generate color stable and cost effective white lighting. After Duggal et al. implemented down conversion concept to the field of OLEDs [15] the previous highest efficiency of down conversion WOLED was done by Krummacker et al. in 2006. In this dissertation, we achieved 9 9.3 and 8 7 lm/W of down conversion WOLED at 3 0 and 100 cd/m 2 with CRI 83 using blue m icrocavity OLED which is 4 times efficient than previously reported down conversion WOLED [41] High ly efficient deep blue OLED is the key to achieve efficient down conversion. First of all, we achieve d 50 lm/W blue PHOLED by understanding of triplet exciton confinement and charge balance. Microcavity with dielectric/metal mirrors was used here for accomplishing deep blue color and further efficiency enhancement. Redistribution of optical cavity mode by microcavity design allows extracting wave guided mode out to air. It leads to increase external quantum efficiency (40%

PAGE 146

146 enhancement at maximum ). Emission spectrum is also easily tuned by optical cavity thickness. CIE coordinate of sky blue FIrpic device ( 0.16, 0.34) is shifted to deeper blue color (0.114, 0.242). The generation of white light is achieved by the combination of non absorbed blue photons from blue emitting OLED and re emitted yellow or red photons by down convert ed phosphor s Color temperatu re and CRI are easily controlled by the selection of phosphors and the change of phosphor film thickness. The spectrum of white light is stable under operation and changing brightness. The simple process of down fine controls of multiple doping or emitting layer is very appropriate for cost effective and large scale lighting applications D own conversion phosphors are applied on the blue microcavity OLEDs We obtained the high power efficiency of 73, 61 lm/W with yellow phosphor and 67.7, 57 lm/W with red:yellow phosphor mixture at 3 0 and 100 cd/m 2 on the blue microcavity OLED with 4 layers of QWS mirrors. Finally, macrolens is used for further enhancement of the out coupling efficiency. Obtained power efficiency o f down conversion white OLED are 107, 96 lm/W (CRI : 59, CIE : 0.36, 0.50) with yellow phosphor and 99.3, 8 7 lm/W (CRI : 83, CIE : 0.43, 0.46) with mixture of red:yellow phosphors at 3 0 and 100 cd/m 2 on the blue microcavity OLED with 4 layers of QWS mirro rs. Photo summaries are shown in Figure 6 1. Efficient blue PHOLED is fabricated first in Figure 6 1 (a) then blue microcavity PHOLED is successfully demonstrated in Figure 6 1 (b). Finally, down conversion WOLED is presented in Figure 6 1 (c).

PAGE 147

147 6.2 F utu re W ork 6.2.1 Challenges for White OLEDs The future of WOLED is very promising. A WOLED has strong potential for both display and lighting applications. For the display application, a WOLED with out backlight and liquid crystal can be used for each self lig ht emitting pixel with the color filter [105] It gives a lot of cost reduction in the fabri cation process. For the lighting application, a WOLED will be a n alternative light source instead of light bulb and fluorescent lamp. A planar, flexible, and efficient WOLED has potential to be totally a new type of light source, we never experienced befor e. T here are few challenges left for the commercialization of WOLED. One is the discovering stable (long lasting) phosphorescent blue materials. Stable green and red phosphorescent materials are already developed now (Lifetime, T 50 is over 100,000 hours at 1000 cd/m 2 ) To commercialize WOLED for the lighting application, phosphorescent blue materials is required. Until now reported in the market the longest lifetime of sky blue phosphorescent material is about 20,000 hour Another challenge is finding th e light extraction scheme possible for the mass production for a WOLED. Many approaches of enhancing the out coupling efficiency are introduced for several researchers. Most of approaches are not relevant for the large scale light ing application. At least 2 time enhancement of the out coupling efficiency is required for achievement of 100 lm/W power efficiency, which is necessary condition for the commodity lighting. Use of bight enhancement film (BEF) and high refractive index substrate might be possible c andidate s in the mass productive solutions. Nowadays, even though WOLEDs over 100 lm/W is demonstrated by several groups by use of all phosphorescent RGB materials and light extraction scheme, its

PAGE 148

148 lifetime doesn t still match with the need of customers du e to unstable blue phosphorescent materials. A WOLED with 50 60 lm/W and 20,000 hour lifetime (T 50 ) over 1000 cd/m 2 will be possible with help of the light extraction based on the combination of fluorescent blue and phosphorescent green red hybrid structur e in 1~2 years future With this high performance WOLED will begin to penetrate into the lighting market for the quality lighting applications. 6.2.2 Future work Down conversion W OLED in this dissertation has the several obstacles to overcome for the indu strial fabrication. One is the short lifetime and steep efficiency roll off of the blue phosphorescent OLED. The lifetime of highly efficient blue OLED in Chapter 2 is lower than 100 hours at 500 cd/m 2 High triplet energy materials of TAPC, mCP, and 3TPYM we used here are the very initial stage materials for phosphorescent devices. Bulk degradation of these materials looks quite dramatic under device operation based on my experience. It will take 3 5 years to find robust phosphorescent materials from the h elp of chemist. In a meantime, blue fluorescent materials can be applied instead of phosphorescent materials. Current available blue fluorescent materials have 8% EQE with ~15,000 hour lifetime at 1000 cd/m 2 [106] Down conversion white fluorescent OLE D with ~ 30 lm/W power efficient is expected by the combination of blue fluorescent OLED, microcavit y and down conversion phosphor. T he roll off efficiency is quite severe in down conversion WOLED (from 9 9 lm/W at 3 0 cd/m 2 to 4 7 lm/W at 1000 cd/m 2 ) due to the steep roll off of blue PHOLED. It is known that triplet triplet quenching and poor charge balance is the sources of the steep roll off. Recently several research groups found the way to maintain the high efficiency even at several thousand luminance (~ 10 000 cd/m 2 ) by tuning charge balance and

PAGE 149

149 controlling the location of the recombination zone (dual emitting layers, bipolar host, and mixed host layer) [34, 74, 107, 108] Use of charge balancing technique s over 80lm/W down conversion WOLED at 1000 cd/m 2 might be possible. The finding light extraction scheme for down conversion phosphor WOLED is also one of re quired futre work In the Chapter 5, we showed that the macrolens extracts the substrate guided mode out about 46% more. Microcavity structure is used for extracting ITO/organic mode out. However, enhancement of EQE enhancement by the microcavities is ~ 40 % EQE at the maximum and the macrolens is not option choice for the industrial use. Therefore, it needs to develop efficient out coupling method for down conversion architecture For example, high index substrate can be applied instead of microcavity struc ture for extracting ITO/organic mode. Surface texture treatment (roughing patterning, stamping and implanted microlens array structure) on the phosphor film surface might be one of considerations. There are several applications of microcavity structure on an OLED. One is the modification of cavity mode distribution. In this dissertation, blue microcavity OLED has directional and elliptical emission shape in Figure 4 4. As shown in Figure 4 7, batwing shaped emission pattern is also possible by controllin g the optical cavity thickness. Microcavity OLED with this batwing shaped emission profile has the smallest amount of the substrate guided mode, even lower than noncavity device. This microcavity design might give the highe st efficiency enhancement. For the lighting application, batwing shape emission pattern will have more advantages than lambertian emission pattern depending on the usage. Without optical luminaries like focusing or colliminating lens and extra convex mirrors, varieties of emission p attern from directional to batwing

PAGE 150

150 will be great advantages of WOLED. Without strong microcavity structure, controlling the emission profile is also possible with weak microcavity effects. The understanding of optical field distribution in the OLED can b e applied to modeling of the organic photovoltaic or detector. Microcavity can be implemented as an optical band pass filter, which open or close the specific spectrum windows for desired Infrared or UV wavelength. Field distribution inside organic photovo ltaic needs to be considered. The maximum (constructive) electromagnetic field should be formed at the photon absorbing materials and the minimum (destructive) field should be at the electrodes or reflective interfaces to increase the light harvesting effi ciency.

PAGE 151

151 (a) (b) (c) Figure 6 .1 Photo taken of (a) blue FIrpic doped PHOLED, (b) blue microcavity PHOLED with 4QWS, and (c) down conversion white OLED.

PAGE 152

152 APPENDIX A MOLECULAR STRUCTURE OF ORGANIC MATERIALS USED THIS DISSERTATI ON NPD Alq3 T PD TAPC CuPC BCP PEDOT:PSS BPhen

PAGE 153

153 CBP mCP CDBP UGH2 FIrpic 3TPYMB PQIr

PAGE 154

154 LIST OF REFERENCES [1] M. Pope, H. P. Kallmann, and P. Magnante, "Electroluminescence in organic crystals", J. Chem. Phys. vol. 38, pp. 2042, 1963. [2] R. H. Partridge, "Electroluminescence from polyvinylcarbazole films. Pt. 1: carbazole cations", Polymer vol. 24, pp. 733 738, 1983. [3] C. W. Tang and S. A. VanSlyke, "Organic electroluminescent diodes", App. Phys. Lett. vol. 51 pp. 913 915, 1987. [4] Samsung Mobile Display, www.samsungsmd.com Dec. 2009 [5] General Electric Global Research Center, http://www.grcblog.com/?cat=19 Dec. 2009 [ 6] S. Bruner, D. Xu, and C. Phillips, "54.3: Drop Landing Accuracy Improvements in Inkjet Printed OLED Displays," SID 2007. [7] F. So, J. Kido, and P. Burrows, "Organic Light Emitting Devices for Solid State Lighting", MRS BULLETIN vol. 33, 2008. [8] B. W. D'Andrade and S. R. Forrest, "White Organic Light Emitting Devices for Solid State Lighting", Adv. Mater. vol. 16, pp. 1585 1595, 2004. [9] S. Reineke, F. Lindner, G. Schwartz, N. Seidler, K. Walzer, B. Lussem, and K. Leo, "White organic light emitti ng diodes with fluorescent tube efficiency", Nature vol. 459, pp. 234 238, 2009. [10] I. Navigant Consulting, Washington, "U.S. Department of Energy," 2001. [11] www.energy.ca.gov California Energy Commission, "L ighting Efficiency Standards," 2008. [12] B. W. D'Andrade, R. J. Holmes, and S. R. Forrest, "Efficient Organic Electrophosphorescent White Light Emitting Device with a Triple Doped Emissive Layer", Adv. Mater. vol. 16, pp. 624 628, 2004. [13] P. Y. Nga i, "Mainstreaming OLEDs for the Future of OLED Lighting", Workshop, Department of Energy 2009. [14] F. So, B. Krummacher, M. K. Mathai, D. Poplavskyy, S. A. Choulis, and V. E. Choong, "Recent progress in solution processable organic light emitting device s", J. App. Phys. vol. 102, pp. 091101, 2007.

PAGE 155

155 [15] A. R. Duggal, J. J. Shiang, C. M. Heller, and D. F. Foust, "Organic light emitting devices for illumination quality white light", App. Phys. Lett. vol. 80, pp. 3470 3472, 2002. [16] C. Kittel and P. Mc Euen, Introduction to Solid State Physics : Wiley, 2004. [17] H. Bassler, "Charge transport in disordered organic photoconductors: A Monte Carlo simulation study", Phys. Stat. Sol. B vol. 175, pp. 15 56, 1993. [18] M. A. Baldo and S. R. Forrest, "Transie nt analysis of organic electrophosphorescence: I. Transient analysis of triplet energy transfer", Physical Review B vol. 62, pp. 10958, 2000. [19] M. A. Baldo, C. Adachi, and S. R. Forrest, "Transient analysis of organic electrophosphorescence. II. Trans ient analysis of triplet triplet annihilation", Phys. Rev. B vol. 62, pp. 10967, 2000. [20] J. Xue, "Pure and Mixed Organic Thin Films and Heterojunctions for Optoelectronic Applications," in Electrical Engineering vol. PhD: Princeton University, 2005. [21] Martin Pope and C. E. Swenberg, Electronic Processes in Organic Crystals and Polymers Second ed: Oxford Science Publications, 1999. [22] S. R. Forrest, "Ultrathin Organic Films Grown by Organic Molecular Beam Deposition and Related Techniques", Che m. Rev. vol. 97, pp. 1793 1896, 1997. [23] E. F. Schubert, Light Emitting Diodes : Cambridge, 2003. [24] N. Toshitaka, T. Naoto, J. Noriyuki, and F. Hironaka, "Improvement of coupling out efficiency in organic electroluminescent devices by addition of a diffusive layer", J. App. Phys. vol. 96, pp. 6016 6022, 2004. [25] N. C. Greenham, R. H. Friend, and D. C. Bradley, "Angular Dependence of the Emission from a Conjugated Polymer Light Emitting Diode: Implications for efficiency calculations", Adv. Mater. vol. 6, pp. 491 494, 1994. [26] G. Gu, D. Z. Garbuzov, P. E. Burrows, S. Venkatesh, S. R. Forrest, and M. E. Thompson, "High external quantum efficiency organic light emitting devices", Opt. Lett. vol. 22, pp. 396 398, 1997. [27] J. S. Kim, P. K. H. H o, N. C. Greenham, and R. H. Friend, "Electroluminescence emission pattern of organic light emitting diodes: Implications for device efficiency calculations", J. Appl. Phys. vol. 88, pp. 1073 1081, 2000.

PAGE 156

156 [28] S. H. Eom, Y. Zheng, E. Wrzesniewski, J. Lee, N. Chopra, F. So, and J. Xue, "Effect of electron injection and transport materials on efficiency of deep blue phosphorescent organic light emitting devices", Org. Elec. vol. 10, pp. 686 691, 2009. [29] Y. Sun and S. R. Forrest, "Organic light emitting devices with enhanced outcoupling via microlenses fabricated by imprint lithography", J. App. Phys. vol. 100, pp. 073106, 2006. [30] S. Moller and S. R. Forrest, "Improved light out coupling in organic light emitting diodes employing ordered microlens ar rays", J. Appl. Phys. vol. 91, pp. 3324 3327, 2002. [31] Y. J. Lee, S. H. Kim, J. Huh, G. H. Kim, Y. H. Lee, S. H. Cho, Y. C. Kim, and Y. R. Do, "A high extraction efficiency nanopatterned organic light emitting diode", App. Phys. Lett. vol. 82, pp. 377 9 3781, 2003. [32] T. Tsutsui, M. Yahiro, H. Yokogawa, K. Kawano, and M. Yokoyama, "Doubling Coupling Out Efficiency in Organic Light Emitting Devices Using a Thin Silica Aerogel Layer", Adv. Mater. vol. 13, pp. 1149 1152, 2001. [33] B. W. D'Andrade, "W hite Organic Light Emitting Devices," in Electrical Engineering vol. PhD: Princeton University, 2004. [34] S. H. Eom, Y. Zheng, E. Wrzesniewski, J. Lee, N. Chopra, F. So, and J. Xue, "White phosphorescent organic light emitting devices with dual triple d oped emissive layers", App. Phys. Lett. vol. 94, pp. 153303, 2009. [35] S. Shi Jian, G. Eisuke, S. Hisahiro, and K. Junji, "Highly Efficient Organic Blue and White Light Emitting Devices Having a Carrier and Exciton Confining Structure for Reduced Effic iency Roll Off", Adv. Mater. vol. 20, pp. 4189 4194, 2008. [36] G. Schwartz, S. Reineke, K. Walzer, and K. Leo, "Reduced efficiency roll off in high efficiency hybrid white organic light emitting diodes", App. Phys. Lett. vol. 92, pp. 053311 053313, 200 8. [37] H. Kanno, N. C. Giebink, Y. Sun, and S. R. Forrest, "Stacked white organic light emitting devices based on a combination of fluorescent and phosphorescent emitters", App. Phys. Lett. vol. 89, pp. 023503 023503, 2006. [38] Y. Sun, N. C. Giebink, H. Kanno, B. Ma, M. E. Thompson, and S. R. Forrest, "Management of singlet and triplet excitons for efficient white organic light emitting devices", Nature vol. 440, pp. 908 912, 2006.

PAGE 157

157 [39] J. H. Jou, M. H. Wu, C. P. Wang, Y. S. Chiu, P. H. Chiang, H. C. Hu, and R. Y. Wang, "Efficient fluorescent white organic light emitting diodes using co host/emitter dual role possessed di(triphenyl amine) 1,4 divinyl naphthalene", Org. Elec. vol. 8, pp. 735 742, 2007. [40] J. Jwo Huei, W. Chun Jan, L. Yu Pu, C. Yu C hiao, C. Po Hsuan, W. Ming Hsuan, W. Chung Pei, L. Chun Liang, and C. Champion, "Color stable, efficient fluorescent pure white organic light emitting diodes with device architecture preventing excessive exciton formation on guest", App. Phys. Lett. vol. 92, pp. 223504, 2008. [41] B. C. Krummacher, V. E. Choong, M. K. Mathai, S. A. Choulis, F. So, F. Jermann, T. Fiedler, and M. Zachau, "Highly efficient white organic light emitting diode", App. Phy. Lett. vol. 88, pp. 113506, 2006. [42] E. A. Silinsh an d V. Capek, organic molecular crystals : American Institute of Physics, 1997. [43] H. Benisty, H. De Neve, and C. Weisbuch, "Impact of planar microcavity effects on light extraction Part I:basic concepts and analytical trends", IEEE Journal of Quantum Elec tronics vol. 34, pp. 1612 1631, 1998. [44] Y. Sun and S. R. Forrest, "Enhanced light out coupling of organic light emitting devices using embedded low index grids", Nat. Phot. vol. 2, pp. 483 487, 2008. [45] J. J. Shiang, T. J. Faircloth, and A. R. Dug gal, "Experimental demonstration of increased organic light emitting device output via volumetric light scattering", J. Appl. Phys. vol. 95, pp. 2889, 2004. [46] Philips, "Philips Product Catalog," 2009. [47] Fluxim simulation tools, www.fluxim.com Dec. 2009 [48] C. Adachi, M. A. Baldo, M. E. Thompson, and S. R. Forrest, "Nearly 100% internal phosphorescence efficiency in an organic light emitting device", J. App. Phy. vol. 90, pp. 5048 5051, 2001. [49] D. F. O. Br ien, M. A. Baldo, M. E. Thompson, and S. R. Forrest, "Improved energy transfer in electrophosphorescent devices", App. Phy. Lett. vol. 74, pp. 442 444, 1999. [50] D. Tanaka, Y. Agata, T. Takeda, S. Watanabe, and J. Kido, "High luminous efficiency blue or ganic light emitting devices using high triplet excited energy materials", Japanese J. App. Phy. Part 2 Letters & Express Letters vol. 46, pp. L117 L119, 2007.

PAGE 158

158 [51] N. Chopra, J. Lee, Y. Zheng, S. H. Eom, J. Xue, and F. So, "High efficiency blue phosphor escent organic light emitting device", Appl. Phys. Lett. vol. 93, pp. 143307 143303, 2008. [52] J. Lee, N. Chopra, and F. So, "Cavity effects on light extraction in organic light emitting devices", App. Phy. Lett. vol. 92, pp. 033303, 2008. [53] R. Win disch, C. Rooman, S. Meinlschmidt, P. Kiesel, D. Zipperer, G. H. Dohler, B. Dutta, M. Kuijk, G. Borghs, and P. Heremans, "Impact of texture enhanced transmission on high efficiency surface textured light emitting diodes", App. Phy. Lett. vol. 79, pp. 2315 2317, 2001. [54] N. Toshitaka, T. Naoto, J. Noriyuki, and F. Hironaka, "Thin film waveguiding mode light extraction in organic electroluminescent device using high refractive index substrate", J. App. Phy. vol. 97, pp. 054505, 2005. [55] Y. Sun and S. R. Forrest, "Organic light emitting devices with enhanced outcoupling via microlenses fabricated by imprint lithography", J. App. Phy. vol. 100, pp. 073106, 2006. [56] Y. Sun and S. R. Forrest, "Enhanced light out coupling of organic light emitting devic es using embedded low index grids", Nat Photon vol. 2, pp. 483 487, 2008. [57] C. L. Lin, H. W. Lin, and C. C. Wu, "Examining microcavity organic light emitting devices having two metal mirrors", App. Phy. Lett. vol. 87, pp. 021101, 2005. [58] R. J. Ho lmes, S. R. Forrest, Y. J. Tung, R. C. Kwong, J. J. Brown, S. Garon, and M. E. Thompson, "Blue organic electrophosphorescence using exothermic host -guest energy transfer", App. Phy. Lett. vol. 82, pp. 2422 2424, 2003. [59] J. Lee, N. Chopra, S. H. Eom, Y. Zheng, J. Xue, F. So, and J. Shi, "Effects of triplet energies and transporting properties of carrier transporting materials on blue phosphorescent organic light emitting devices", App. Phy. Lett. vol. 93, pp. 123306, 2008. [60] J. H. Lee, T. C. Lin, C. C. Liao, and F. H. Yang, "Study on organic light emitting device with more balanced charge transport," presented at SPIE, 2005. [61] V. E. Choong, S. Shi, J. Curless, C. L. Shieh, H. C. Lee, F. So, J. Shen, and J. Yang, "Organic light emitting diodes w ith a bipolar transport layer", App. Phy. Lett. vol. 75, pp. 172 174, 1999. [62] N. Chopra, J. Lee, Y. Zheng, S. H. Eom, J. Xue, and F. So, "Effect of charge balance on high efficiency blue phosphorescent organic light emitting diodes", Accepted for publ ication, ACS Applied Materials & Interfaces 2009.

PAGE 159

15 9 [63] Y. Kawamura, K. Goushi, J. Brooks, and J. J. Brown, "100% phosphorescence quantum efficiency of Ir(III) complexes in organic semiconductor films", App. Phy. Lett. vol. 86, pp. 071104, 2005. [64] K. Goushi, R. Kwong, J. J. Brown, H. Sasabe, and C. Adachi, "Triplet exciton confinement and unconfinement by adjacent hole transport layers", J. App. Phys. vol. 95, pp. 7798 7802, 2004. [65] S. Naka, H. Okada, H. Onnagawa, Y. Yamaguchi, and T. Tsutsui, "C arrier transport properties of organic materials for EL device operation", Synthetic Metals vol. 111 112, pp. 331 333, 2000. [66] D. Tanaka, T. Takeda, T. Chiba, S. Watanabe, and J. Kido, "Novel Electron transport Material Containing Boron Atom with a Hi gh Triplet Excited Energy Level", Chemistry Letters vol. 36, pp. 262 263, 2007. [67] Q. Xin, W. L. Li, W. M. Su, T. L. Li, Z. S. Su, B. Chu, and B. Li, "Emission mechanism in organic light emitting devices comprising a europium complex as emitter and an electron transporting material as host", J. App. Phy. vol. 101, pp. 044512, 2007. [68] C. Adachi, R. Kwong, C, P. Djurovich, V. Adamovich, M. Baldo, A M. Thompson, E, and S. Forrest, R "Endothermic energy transfer: A mechanism for generating very effic ient high energy phosphorescent emission in organic materials," vol. 79: AIP, 2001, pp. 2082 2084. [69] C. W. Tang, S. A. VanSlyke, and C. H. Chen, "Electroluminescence of doped organic thin films", J. App. Phy. vol. 65, pp. 3610 3616, 1989. [70] M. A. Baldo, D. F. O'Brien, Y. You, A. Shoustikov, S. Sibley, M. E. Thompson, and S. R. Forrest, "Highly efficient phosphorescent emission from organic electroluminescent devices", Nature vol. 395, pp. 151 154, 1998. [71] S. Tokito, T. Iijima, Y. Suzuri, H. Ki ta, T. Tsuzuki, and F. Sato, "Confinement of triplet energy on phosphorescent molecules for highly efficient organic blue light emitting devices", App. Phy. Lett. vol. 83, pp. 569 571, 2003. [72] R. J. Holmes, B. W. D. Andrade, S. R. Forrest, X. Ren, J. Li, and M. E. Thompson, "Efficient, deep blue organic electrophosphorescence by guest charge trapping", App. Phy. Lett. vol. 83, pp. 3818 3820, 2003. [73] X. Ren, J. Li, R. J. Holmes, P. I. Djurovich, S. R. Forrest, and M. E. Thompson, "Ultrahigh Energy Gap Hosts in Deep Blue Organic Electrophosphorescent Devices", Chem. Mater. vol. 16, pp. 4743 4747, 2004.

PAGE 160

160 [74] B. P. Asanga, K. K. Philip, C. Lelia, P. Evgueni, S. S. James, C. Neetu, S. Franky, S. S. Linda, and J. G. Daniel, "Tuning charge balance in PH OLEDs with ambipolar host materials to achieve high efficiency," 2009. [75] N. C. Giebink and S. R. Forrest, "Quantum efficiency roll off at high brightness in fluorescent and phosphorescent organic light emitting diodes", Physical Review B (Condensed Mat ter and Materials Physics) vol. 77, pp. 235215, 2008. [76] P. Strohriegl and J. V. Grazulevicius, "Charge Transporting Molecular Glasses", Adv. Mater. vol. 14, pp. 1439 1452, 2002. [77] Y. Q. Li, M. K. Fung, Z. Xie, S. T. Lee, L. S. Hung, and J. Shi, An Efficient Pure Blue Organic Light Emitting Device with Low Driving Voltages", Adv. Mater. vol. 14, pp. 1317 1321, 2002. [78] M. Ikai, S. Tokito, Y. Sakamoto, T. Suzuki, and Y. Taga, "Highly efficient phosphorescence from organic light emitting devices with an exciton block layer", App. Phy. Lett. vol. 79, pp. 156 158, 2001. [79] G. T. Lei, L. D. Wang, L. Duan, J. H. Wang, and Y. Qiu, "Highly efficient blue electrophosphorescent devices with a novel host material", Synthetic Metals vol. 144, pp. 249 252, 2004. [80] V. Bulovic, V. B. Khalfin, G. Gu, P. E. Burrows, D. Z. Garbuzov, and S. R. Forrest, "Weak microcavity effects in organic light emitting devices", Phys. Rev. B vol. 58, pp. 3730, 1998. [81] B. Krummacher, M. K. Mathai, V. E. Choong, S. A. Choulis, F. So, and A. Winnacker, "Influence of charge balance and microcavity effects on resultant efficiency of organic light emitting devices", Org. Elec. vol. 7, pp. 313 318, 2006. [82] M. H. Lu and J. C. Sturm, "Optimization of external coupling an d light emission in organic light emitting devices: modeling and experiment", J. App. Phys. vol. 91, pp. 595 604, 2002. [83] Unicad simulation tool, www.unicadinc.com Dec. 2009 [84] L. Jaewon, C. Neetu, E. Sang Hy un, Z. Ying, X. Jiangeng, S. Franky, and S. Jianmin, "Effects of triplet energies and transporting properties of carrier transporting materials on blue phosphorescent organic light emitting devices", Appl. Phys. Lett. vol. 93, pp. 123306, 2008. [85] Y. K awamura, K. Goushi, J. Brooks, J. J. Brown, H. Sasabe, and C. Adachi, "100% phosphorescence quantum efficiency of Ir(III) complexes in organic semiconductor films", Appl. Phys. Lett. vol. 86, pp. 071104 071103, 2005.

PAGE 161

161 [86] B. C. Krummacher, M. K. Mathai, V. Choong, S. A. Choulis, F. So, and A. Winnacker, "General method to evaluate substrate surface modification techniques for light extraction enhancement of organic light emitting diodes", J. Appl. Phys. vol. 100, pp. 054702 054706, 2006. [87] M. Agrawal Y. Sun, S. R. Forrest, and P. Peumans, "Enhanced outcoupling from organic light emitting diodes using aperiodic dielectric mirrors", App. Phy. Lett. vol. 90, pp. 241112 241113, 2007. [88] R. H. Jordan, A. Dodabalapur, and R. E. Slusher, "Efficiency enh ancement of microcavity organic light emitting diodes", App. Phy. Lett. vol. 69, pp. 1997, 1996. [89] H. J. Peng, X. L. Zhu, J. X. Sun, X. M. Yu, M. Wong, and H. S. Kwok, "Efficiency improvement of phosphorescent organic light emitting diodes using semit ransparent Ag as anode", App. Phy. Lett. vol. 88, pp. 033509, 2006. [90] C. J. Yang, S. H. Liu, H. H. Hsieh, C. C. Liu, T. Y. Cho, and C. C. Wu, "Microcavity top emitting organic light emitting devices integrated with microlens arrays: Simultaneous enhan cement of quantum efficiency, cd/A efficiency, color performances, and image resolution", App. Phy. Lett. vol. 91, pp. 253508, 2007. [91] S. Tatsuya, S. Mitsunobu, Y. Akira, Y. Jiro, H. Takashi, I. Yuichi, Y. Takao, I. Tadashi, M. Takao, A. Mitsuru, T. S hinichiro, and U. Tetsuo, "24.4L: Late News Paper: A 13.0 inch AM OLED Display with Top Emitting Structure and Adaptive Current Mode Programmed Pixel Circuit (TAC)", SID Symposium Digest of Technical Papers vol. 32, pp. 384 387, 2001. [92] E. F. Schubert Y. H. Wang, A. Y. Cho, L. W. Tu, and G. J. Zydzik, "Resonant cavity light emitting diode", Appl. Phys. Lett. vol. 60, pp. 921, 1992. [93] D. L. Huffaker, C. C. Lin, J. Shin, and D. G. Deppe, "Resonant cavity light emitting diode with an Al[sub x]O[sub y]/GaAs reflector", Appl. Phys. Lett. vol. 66, pp. 3096 3098, 1995. [94] J. Lee, N. Chopra, and F. So, "Cavity effects on light extraction in organic light emitting devices", App. Phy. Lett. vol. 92, pp. 033303, 2008. [95] T. Nakamura, N. Tsutsumi, N. Juni, and H. Fujii, "Improvement of coupling out efficiency in organic electroluminescent devices by addition of a diffusive layer", J. App. Phy. vol. 96, pp. 6016, 2004. [96] Y. J. Lee, S. H. Kim, J. Huh, G. H. Kim, Y. H. Lee, S. H. Cho, Y. C. Kim, and Y. R. Do, "A high extraction efficiency nanopatterned organic light emitting diode", App. Phy. Lett. vol. 82, pp. 3779, 2003.

PAGE 162

162 [97] A. Dodabalapur, L. J. Rothberg, R. H. Jordan, T. M. Miller, R. E. Slusher, and J. M. Phillips, "Physics and applications of organic microcavity light emitting diodes", J. App. Phy. vol. 80, pp. 6954, 1996. [98] G. R. Fowles, "Introduction To Modern Optics", Dover Publications, INC., New York [99] S. Tokito, T. Tsutsui, and Y. Taga, "Microcavity organic light emitting diode s for strongly directed pure red, green, and blue emissions", J. Appl. Phys. vol. 86, pp. 2407 2411, 1999. [100] S. Tokito, T. Tsutsui, and Y. Taga, "Microcavity organic light emitting diodes for strongly directed pure red, green, and blue emissions", J. App. Phy. vol. 86, pp. 2407, 1999. [101] H. Kanno, N. C. Giebink, Y. Sun, and S. R. Forrest, "Stacked white organic light emitting devices based on a combination of fluorescent and phosphorescent emitters", App. Phy. Lett. vol. 89, pp. 023503, 2006. [ 102] B. W. D'Andrade and J. Esler, "Realizing white phosphorescent OLED effi ciency limits", SPIE optics+Photonics vol. 7051, pp. 23, 2008. [103] P. Schlotter, R. Schmidt, and J. Schneider, "Luminescence conversion of blue light emitting diodes", Applied Physics A: Materials Science & Processing vol. 64, pp. 417 418, 1997. [104] B. Krummacher, M. Klein, N. von Malm, and A. Winnacker, "Optical analysis of down conversion OLEDs," 2008. [105] Y. Xiong, W. Xu, C. Li, B. Liang, L. Zhao, J. Peng, Y. Cao, and J. Wang, "Utilizing white OLED for full color reproduction in flat panel display", Org. Elec. vol. 9, pp. 533 538, 2008. [106] L. Jiun Haw, H. Yu Hsuan, L. Tien Chun, and W. Chia Fang, "High Efficiency Fluorescent Blue Organic Light Emitting Device with Balanced Carrier Transport", J. Electrochem. Soc. vol. 154, pp. J226 J228, 2007. [107] K. Jae Wook, L. Se Hyung, P. Hyung Dol, J. Won Ik, Y. Kyung Mo, P. Young Seo, and K. Jang Joo, "Low roll off of efficiency at high current density in phosphorescent o rganic light emitting diodes", App. Phys. Lett. vol. 90, pp. 223508, 2007. [108] Y. C. Tsai and J. H. Jou, "Long lifetime, high efficiency white organic light emitting diodes with mixed host composing double emission layers", App. Phys. Lett. vol. 89, p p. 243521, 2006.

PAGE 163

163 BIOGRAPHICAL SKETCH Jaewon Lee was born in Busan, South Korea on April 8 19 7 7. He received the B.S. degree in physics from the Chung Ang University at Seoul, South Korea, in 2000 the M.S. degree in fiber optics from the Gwangju Institu te Science and Technology (GIST) in 2003 and Ph.D. degrees in materials science and engineering from the University of Florida at Gainesville, Florida, i n 2009 After graduation with the M.S he joined Institute of Information Technology Assessment (IITA) as a research er planning the Korean government strategy of IT human resource development for supply and demand, of academy and industry. His Ph.D. dissertation was completed in the organic electronic materials and devices laboratory especially down conve rsion white OLED with blue microcavity phosphorescent OLED.