<%BANNER%>

Efficient Near-Infrared and Ultraviolet Organic Light Emitting Devices

Permanent Link: http://ufdc.ufl.edu/UFE0043752/00001

Material Information

Title: Efficient Near-Infrared and Ultraviolet Organic Light Emitting Devices
Physical Description: 1 online resource (208 p.)
Language: english
Creator: Yang, Yixing
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: eletroluminescence -- infrared -- oled -- ultraviolet
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: There has been a growing interest in the development of extreme light sources including near-infrared (near-IR) and ultraviolet (UV) emitting organic light-emitting devices (OLEDs) due to their potential applications. Existing near-IR and UV OLEDs generally have low external quantum efficiencies and limited choice of emitter materials. So, new materials and devices need to be developed to achieve OLEDs with tunable emission and high efficiency in near-IR and UV ranges. In the first part of this dissertation, we achieve very nice coverage of the near-IR emission range from 700 nm to over 1000 nm, by using different novel near-IR emitting materials. Fluorescent near-IR OLEDs based on two donor-acceptor-donor oligomers, BEDOT-TQMe2 and BEDOT-BBT, are first demonstrated with maximum external quantum efficiency (EQE) up to 1.6% and emission peak wavelengths over 800 nm. The efficiencies of these fluorescent OLEDs are further increased by two to three times by using sensitized fluorescence structure to funnel the triplet excitons formed on the host molecules to the fluorescent emitters, which are not utilized in fluorescent devices. Even higher efficiencies and longer emission wavelengths are achieved by phosphorescent near-IR OLEDs based on extended conjugation Platinum (II) porphyrins, with record high EQE of 9.2%. Furthermore, we do realize that the PL efficiencies in solution do not necessarily carry over to the EL efficiencies of devices. In the second part, towards the short wavelength end of the visible light range, a blue-violet emitting device has been constructed featuring a highly fluorescent donor-acceptor purine in the emissive layer, with a maximum EQE of 3.1% and peak emission at 430 nm. With different donor groups connected to the purine heterocycle ring, another purine derivative with further blue-shifted emission is also employed for UV emitting OLEDs, achieving a maximum EQE of 1.5% and emission peak below 400 nm, at wavelength of 393 nm. We realize that, due to the high energy band gap of UV emitting molecules, it is very important to appropriately align the energy levels of different layers in multilayer UV OLEDs to efficiently extract the desired light emission.
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 Yixing Yang.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Xue, Jiangeng.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043752:00001

Permanent Link: http://ufdc.ufl.edu/UFE0043752/00001

Material Information

Title: Efficient Near-Infrared and Ultraviolet Organic Light Emitting Devices
Physical Description: 1 online resource (208 p.)
Language: english
Creator: Yang, Yixing
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2011

Subjects

Subjects / Keywords: eletroluminescence -- infrared -- oled -- ultraviolet
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: There has been a growing interest in the development of extreme light sources including near-infrared (near-IR) and ultraviolet (UV) emitting organic light-emitting devices (OLEDs) due to their potential applications. Existing near-IR and UV OLEDs generally have low external quantum efficiencies and limited choice of emitter materials. So, new materials and devices need to be developed to achieve OLEDs with tunable emission and high efficiency in near-IR and UV ranges. In the first part of this dissertation, we achieve very nice coverage of the near-IR emission range from 700 nm to over 1000 nm, by using different novel near-IR emitting materials. Fluorescent near-IR OLEDs based on two donor-acceptor-donor oligomers, BEDOT-TQMe2 and BEDOT-BBT, are first demonstrated with maximum external quantum efficiency (EQE) up to 1.6% and emission peak wavelengths over 800 nm. The efficiencies of these fluorescent OLEDs are further increased by two to three times by using sensitized fluorescence structure to funnel the triplet excitons formed on the host molecules to the fluorescent emitters, which are not utilized in fluorescent devices. Even higher efficiencies and longer emission wavelengths are achieved by phosphorescent near-IR OLEDs based on extended conjugation Platinum (II) porphyrins, with record high EQE of 9.2%. Furthermore, we do realize that the PL efficiencies in solution do not necessarily carry over to the EL efficiencies of devices. In the second part, towards the short wavelength end of the visible light range, a blue-violet emitting device has been constructed featuring a highly fluorescent donor-acceptor purine in the emissive layer, with a maximum EQE of 3.1% and peak emission at 430 nm. With different donor groups connected to the purine heterocycle ring, another purine derivative with further blue-shifted emission is also employed for UV emitting OLEDs, achieving a maximum EQE of 1.5% and emission peak below 400 nm, at wavelength of 393 nm. We realize that, due to the high energy band gap of UV emitting molecules, it is very important to appropriately align the energy levels of different layers in multilayer UV OLEDs to efficiently extract the desired light emission.
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 Yixing Yang.
Thesis: Thesis (Ph.D.)--University of Florida, 2011.
Local: Adviser: Xue, Jiangeng.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2013-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2011
System ID: UFE0043752:00001


This item has the following downloads:


Full Text

PAGE 1

1 EFFICIENT NEAR INFRARED AND ULTRAVIOLET ORGANIC LIGHT EMITTING DEVICES By YIXING YANG 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 2011

PAGE 2

2 2011 Yixing Yang

PAGE 3

3 T o my m om

PAGE 4

4 ACKNOWLEDGMENTS I would like to express my sincere gratitude to my advisor, Dr. Jiangeng Xue, for his guidance and support during my four years graduate study at University of Florida. His substandial knowledge in the field and high standards on academic research has sharpened my mind and helped me a lot to accomplish my goals. I would also like to thank my supervisory committee, Dr. Franky So, Dr. Elliot Douglas, Dr. Jennifer Andrew, and Dr. Kirk Schanze, for their advice and support on my research. Thanks Dr. John Mecholsky for attending my final defense exam. I owe many people for their active collaboration on accomplishing this thesis. Among them I wo uld like to thank the collaborators in the Department of Chemistry at the University of Florida: Jonathan Sommer for Pt porphyrin material synthesis, Timothy Steckler and Dan Patel for donor acceptor donor oligomers synthesis, Pamela Cohn for purine synthe sis, Richard Farley, Kenneth Graham, and Aubrey Dyer for their help in the material photophysical charaterizations, and Dr. John Reynolds, Dr. Kirk Schanze, and Dr. Ronald Castellano for their advice in finishing the research projects. I would also like to thank Dr. Sang Hyun Eom for the extensive discussion on the device design and fabrication and Renjia Zhou for the synthesis and characterization of CdSe nanorods for me. I am grateful to the fellow graduate students and post docs in my group that I work with during my graduate study. I would like to thank Dr. Ying Zheng, Dr. Wei Zhao, Dr. William Hammond, Dr. Teng kuan Tseng Dr. Jason Myers, Weiran Cao, Edward Wrzesniewski, John Mudrick, Matthew Rippe, Nathan Shewmon, Shuang Zhao, and Zhifeng Li for the joyful discussions with them about the research projects and beyond. I also need to acknowledge the financial support from U.S. Defense Advanced Research Projects Agency (DARPA) and the U.S. Army Aviation and Missile

PAGE 5

5 Research, Development, and Engineering Center (AMRDEC), the Florida Energy Systems Consortium (FESC), and Sestar Technologies, LLC I also need to thank Dr. Wei Qiu for his kind help for my job searching, and thank my classmates in Department of Materials Science and Engineering, Rui Qing, Chan ghua Liu, Bo Xu, Jiaqing Zhou, and Song Chen for their support outside the lab. I would also like to thank all my other friends in Gainesville especially Long Yu, for sharing the joy with me and giving me help. Finally, this thesis would not have been com plete d without the support and understanding from my mother. I show my deepest gratitude and appreciation to her for her love, trust and encouragement throughout my college study.

PAGE 6

6 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 9 LIST OF FIGURES ................................ ................................ ................................ ........ 10 LIST OF ABBREVIATIONS ................................ ................................ ........................... 15 ABSTRACT ................................ ................................ ................................ ................... 21 CHAPTER 1 INTRODUCTION TO ORGANIC LIGHT EMITTING DEVICES .............................. 23 1.1 Lighting Emitting Devices (LEDs) ................................ ................................ ..... 23 1.2 Advantages of Organic Light Emitting Devices ................................ ................. 25 1.3 Applications of Organic Light Emitting Devices ................................ ................ 27 1.3.1 Flat Panel Display ................................ ................................ .................... 28 1.3.2 White OLEDs Lighting ................................ ................................ ............. 29 1.4 High Efficie ncy Blue and White PHOLEDs ................................ ....................... 30 1.5 Near Infrared and Ultra Violet OLEDs ................................ .............................. 33 1.6 Outline of Dissertation ................................ ................................ ....................... 37 2 PHYSICS OF ORGANIC SEMICONDUCTORS ................................ ..................... 38 2.1 Introduction to Organic Semiconductors ................................ ........................... 38 2.2 Electronic St ructures of Organic Semiconductors ................................ ............. 40 2.3 Charge Transport ................................ ................................ .............................. 44 2.3.1 Hopping Transport ................................ ................................ ................... 45 2.3.2 Band Transport ................................ ................................ ........................ 46 2.3.3 Transport vs. Optical Band Gaps ................................ ............................ 46 2.4 Excitons ................................ ................................ ................................ ............ 47 2.4.1 Classification of Excitons ................................ ................................ ......... 47 2.4.2 Multiplicity of Excitons ................................ ................................ ............. 48 2.4.3 Metal Ligand Charge Transfer Excit on ................................ .................... 49 2.5 Intra Molecular Energy Transfer ................................ ................................ ....... 50 2.5.1 Absorption ................................ ................................ ............................... 52 2.5.2 Vi brational Relaxation and Internal Conversion ................................ ....... 52 2.5.3 Intersystem C rossing ................................ ................................ ............... 52 2.5.4 Fluorescence ................................ ................................ ........................... 53 2.5.5 Phosphorescence ................................ ................................ .................... 53 2.5.6 Frank Condon Shift ................................ ................................ ................. 54 2.6 Inter Molecular Energy Transfer ................................ ................................ ....... 55 2.6.1 Frster Energy Transfer ................................ ................................ .......... 55

PAGE 7

7 2.6.2 Dexter Energy Transfer ................................ ................................ ........... 56 2.6.3 Exciton Dissociat ion ................................ ................................ ................ 57 3 FUNDAMENTALS OF ORGA NIC LIGHT EMITTING DEVICES ............................ 58 3.1 Device Structure of OLEDs ................................ ................................ ............... 58 3.2 Fabrication of OLEDs ................................ ................................ ........................ 60 3.3 Principle of OLEDs Operation ................................ ................................ ........... 62 3.4 Electromagnetic Spectrum ................................ ................................ ................ 64 3.4.1 Photometry vs. Radiometry ................................ ................................ ..... 65 3.4.2 Photopic vs. Scotopic Response ................................ ............................. 66 3.4.3 Plane vs. S olid Angles ................................ ................................ ............. 67 3.4.4 Lambertian Emission S ource ................................ ................................ ... 68 3.4.5 CIE C olor Space ................................ ................................ ...................... 69 3.5 Characteristics of OLEDs ................................ ................................ .................. 72 3.5.1 Radiance ................................ ................................ ................................ 74 3.5.2 Luminance ................................ ................................ ............................... 74 3 .5.3 Luminance Efficiency ................................ ................................ ............... 75 3.5.4 Power Efficiency ................................ ................................ ...................... 75 3.5.5 Luminous Power Efficiency ................................ ................................ ..... 75 3.5.6 External Quantum Efficiency ................................ ................................ ... 76 3.6 Efficiency of OLEDs ................................ ................................ .......................... 77 4 NEAR INFRARED ORGANIC LIGHT EMITTING DEVICES BASED ON FLUORESCENT DONOR ACCEPTOR DONOR OLIGOMERS ............................. 80 4.1 Introduction to the Near IR OLEDs ................................ ................................ ... 80 4.2 Donor Acceptor Donor Oligomers ................................ ................................ .... 81 4.3 Experimental Details ................................ ................................ ......................... 84 4.4 Near I nfrared OLEDs based on DAD oligomers ................................ ............... 87 4 .4. 1. Optical P roperties of the DAD O ligomers ................................ ............... 87 4.4. 2. Fluorescent NIR OLEDs based on the DAD O ligomers .......................... 88 4.4. 3. Sensitized F luoresc ent NIR OLEDs based on the DAD O ligomers ........ 95 4.5 Aggregation of Donor Acceptor Donor Moleclues ................................ .......... 101 4.6 Summary ................................ ................................ ................................ ........ 108 5 NEAR INFRARED ORGANIC LIGHT EMITTING DEVICES BASED ON PHOSPHORESCENT PLATINUM (II) PORPHYRINS ................................ .......... 111 5.1 Introduction ................................ ................................ ................................ ..... 111 5.2 Experimental Details ................................ ................................ ....................... 115 5.3 Extended Conjugation Platinum (II) Porphyrins ................................ .............. 119 5.3.1 Near IR OLEDs ba sed on Pt TPTNP ................................ .................... 121 5.3.2 Near IR OLEDs based on Pt Ar 4 TAP ................................ .................... 125 5.4 Effect of Structure on EL Efficiency in Platinum (II) Tetrabenzopor phyrins ..... 131 5.5 Electroluminescence Transient Lifetime of Devices ................................ ........ 139 5.5.1 Theory ................................ ................................ ................................ ... 140

PAGE 8

8 5.5.2 Transient Analysis of NIR OLEDs based on Pt TBPs ............................ 143 5.6 Aggregation Effect of Pt TBPs on EL Efficiency ................................ ............. 146 5.7 Su mmary ................................ ................................ ................................ ........ 148 6 ULTRA VIOLET ORGANIC LIGHT EMITTING DEVICES ................................ .... 150 6.1 Introduction to UV OLEDs ................................ ................................ ............... 150 6.2 Donor Acceptor Purines ................................ ................................ .............. 151 6.3 Experimental Details ................................ ................................ ....................... 154 6.4 Blue Violet OLEDs based on Purine 1 ................................ ............................ 157 6.5 Violet UV Emitting OLEDs based on Purine 2 ................................ ................ 166 6.6 Summary ................................ ................................ ................................ ........ 174 7 CONCLUSIONS AND FUTURE WORKS ................................ ............................. 177 7.1 Conclusions ................................ ................................ ................................ .... 177 7.1.1 Near IR OLEDs based on Fluorescent Donor Acceptor Donor Oligomers ................................ ................................ ................................ .... 177 7.1.2 Near IR OLEDs based on Phosphorescent Platinum (II) Porphyrins .... 178 7.1.3 Ultra Violet Organic Light Emitting Devices ................................ ........... 179 7.2 Future Works ................................ ................................ ................................ .. 180 7.2.1 Near IR OLEDs of Longer Emission and Higher Efficiency ................... 180 7.2.2 Application of Near IR Emitters in OPVs ................................ ............... 181 7.2.3 Ultra Violet OLEDs of Shorter Emisison ................................ ................ 185 APPENDIX A LIST OF ORGANIC MOLECULAR STRUCTURES ................................ .............. 189 B LIST OF PUBLICATIONS AND CONFERENCES ................................ ................ 193 LIST OF REFERENCES ................................ ................................ ............................. 195 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 208

PAGE 9

9 LIST OF TABLES Table page 1 1 Recent research progress of blue and white OLEDs. ................................ ......... 33 4 1 List of so me existing near IR OLEDs based on different near IR emitters. ........ 81 5 1 Photophysical data for Pt TPTBP, Pt TPTNP and Pt Ar 4 TAP in toluene. ......... 12 1 5 2 Photophysical data in toluene and device characteristics for Pt TPTBP, Pt Ar 4 TBP, Pt DPTBP, Pt Ar 2 TAP, Pt TAr 2 TBP, and Pt Ar 2 OPrTBP. ................... 133 6 1 List of some existing blue (B), blue viole t (B V) and ultraviolet (UV) OLEDs ... 151 6 2 Photophysical data for purine 1 and 2 measured in CH 2 Cl 2 ............................ 154

PAGE 10

10 LIST OF FIGURES Figure page 1 1 (color) Examples of white LEDs for lighting. ................................ ....................... 24 1 2 (color) Performance and current progress of several light sources including inorganic and orga nic LEDs. ................................ ................................ ............... 25 1 3 (color) CIE color coordinates of various phosphorescent cyclometalate platinum complexes. ................................ ................................ ........................... 26 1 4 (color) Examples of flexible OLEDs applications ................................ ............... 27 1 5 (color) Various OLEDs display applications ................................ ....................... 28 1 6 (color) Examples of OLEDs lighting applications ................................ ............... 30 1 7 (color) Various near IR light source applications ................................ ............... 34 1 8 (color) Various ultra violet light source applications ................................ ........... 36 2 1 (color) Classification of organic semiconductors with increasing structural complexity. ................................ ................................ ................................ .......... 39 2 2 Examples of several organic semiconductors in electronic and optoelectronic applications. ................................ ................................ ................................ ........ 40 2 3 Electronic configurations of a carbon atom in the ground state and sp n hybridized states ( n = 1, 2, 3). ................................ ................................ ............ 41 2 4 Molecular orbitals of two adjacent carbon atoms with sp 2 hybridization ............ 42 2 5 Delocalized bond of benzene and extended conjugation system. ................... 43 2 6 Energy level diagrams of molecules in different phases. ................................ .... 44 2 7 Schematic illustration of three types of excitons in a solid ................................ 47 2 8 (color) Schematic illustration of fluorescence and phosphorescence ................ 49 2 9 (color) Two types of metal organic phosphors with heavy transition metals ...... 50 2 10 (color) Jablonski energy diagram. ................................ ................................ ....... 51 2 11 Frank Condon principle energy diagram. ................................ ........................... 54 2 12 (color) Schematic illustration of two non radiative inter molecular energy transfer processes between host and dopant ................................ .................... 56

PAGE 11

11 3 1 (color) Typical structures of OLEDs ................................ ................................ ... 58 3 2 Chemical structures of PEDOT and PSS ................................ .......................... 59 3 3 (color) Schematic illustration of organic thin film growth processes ................... 61 3 4 (color) Schematic energy band diagrams of OLEDs operation. .......................... 63 3 5 (color) The electromagnetic spectrum as a function of wavelength (nm). ........... 64 3 6 Comparison between Photometry and Radiometry. ................................ ........... 65 3 7 Photopic and scotopic responses of human eye. ................................ ............... 67 3 8 Plane and solid angles ................................ ................................ ...................... 67 3 9 (color) Schematic illustration of a Lambertian emission ................................ ..... 68 3 10 (color) Color matching function curves ................................ .............................. 70 3 11 (color) CIE chromaticity diagram represented by ( x,y ) coordinates. ................... 71 3 12 (color) Three opti cal modes in bottom emitting OLEDs ................................ ..... 78 4 1 Molecular structures and PL spectra of two near IR emitting DAD oligomers in CH 2 Cl 2 solutions. ................................ ................................ ............................ 85 4 2 (color) Absorption spectra of BEDOT TQMe 2 and BEDOT BBT in CH 2 Cl 2 solutions. ................................ ................................ ................................ ............ 87 4 3 (color) Device characteristics and energy level diagram of OLEDs with various BEDOT TQMe 2 doping con centrations. ................................ ................. 89 4 4 (color) EQE s of OLEDs based on BEDOT TQMe 2 with various ETL thickness. ................................ ................................ ................................ ........... 90 4 5 (color) EQE s of OLEDs based on BEDOT BBT with various ETL thickness. ..... 91 4 6 Device characteristics of optimal NF OLEDs based on two DAD oligomers. ...... 92 4 7 (color) Device e fficiencies of NF and SF OLEDs based on two oligomers. ........ 95 4 8 (color) Energy transfer mechanisms in NF and SF systems. .............................. 96 4 9 RJ V cha racteristics for SF OLEDs based on two oligomers ............................... 98 4 10 (color) EL spectra for SF OLEDs based on two oligomers. ................................ 99

PAGE 12

12 4 11 (color) Dev ice characteristics for SF OLEDs with different doping concentration of near IR fluorescent emitter. ................................ .................... 100 4 12 (color) Device characteristics for SF OLEDs with different doping concentration of th e phosphorescent sensitizer ................................ ............... 101 4 13 (color) The spectral over laps between absorption of BEDOT BBT and emission s of Al q 3 and CBP ................................ ................................ .............. 102 4 14 Molecular structures of bulky donor acceptor donor oligomers. ....................... 103 4 15 (color) Device characteristics of NIR OLEDs based on BEDOT BBT using CBP and Al q 3 as host materia ls. ................................ ................................ ....... 105 4 16 (color) EQEs of OLEDs based on three DAD oligomers with different doping concentrations. ................................ ................................ ................................ 106 4 17 (color) EL spectra of NIR OLEDs based on thr ee DAD oligomers ................... 108 5 1 conjugation series of Pt porphyrins ....... 119 5 2 PL spectra of Pt TPTBP, Pt TPTNP and Pt Ar 4 TAP in toluene solutions. ........ 120 5 3 (color) Device characteristics for devices based on Pt TPTNP with various doping concentrations and ETL thickness. ................................ ....................... 122 5 4 EL spectra for NIR OLEDs based on Pt TPTBP and Pt TPTNP ...................... 124 5 5 (color) Schemati c energy level diagram of OLEDs based on Pt Ar 4 TAP. ......... 125 5 6 (color) Device characteristics for Pt Ar 4 TAP based OLEDs with different preparation recipes of the active layer. ................................ ............................. 127 5 7 (color) Device characteristics for Pt Ar 4 TAP based OLEDs with different ETLs and solution concentration s ................................ ................................ .... 128 5 8 Device characteristics for the optimal NIR OLEDs based on Pt Ar 4 TAP. ......... 130 5 9 Molecular structures of Pt TBPs. ................................ ................................ ...... 132 5 10 ( c olor) Device characteristics of the NIR OLEDs based on Pt TBPs. ............... 135 5 11 (color) Device efficiencies of the NIR OLEDs based on Pt TBPs. .................... 136 5 12 (color) Device characteristics of the NIR OLEDs based on Pt TAr 2 TBP and Pt Ar 2 OPrTBP ................................ ................................ ................................ .. 138 5 13 (color) Transient EL decay of Pt Ar 4 TBP, Pt DPTBP and Pt Ar 2 TBP .............. 144

PAGE 13

13 5 14 (color) Transient EL decay of Pt TAr 2 TBP and Pt Ar 2 OPrTBP ........................ 145 5 15 EQEs of OLEDs based on Pt TPTBP a nd Pt Ar 4 TBP as a function of doping concentration ................................ ................................ ................................ ... 147 6 1 Generic structure of the donor acceptor purines ................................ ............. 152 6 2 PL spectra of pur ine 1 in solid state film and in CH 2 Cl 2 solution ...................... 158 6 3 (color) Schematic en ergy level diagram of OLEDs based on purine 1 with different ETLs. ................................ ................................ ................................ .. 159 6 4 (color) Device characteristics o f purine 1 based OLEDs with different ETLs. ... 160 6 5 (color) Device efficiencies for purine 1 b ased devices with various OXD 7 layer thicknesses. ................................ ................................ ............................. 162 6 6 Device efficiencies of OLEDs with different doping concentration of purine 1 162 6 7 Device characteristics of the optimal p urine 1 based OLED ............................ 163 6 8 EL spectrum for the OLED based on purine 1 and d evice picture ................... 164 6 9 (color) EL intensity decay of optim al OLEDs based on purine 1 ...................... 165 6 10 PL spectra of purine 2 in solid state films ................................ ........................ 166 6 11 (color) EL spectrum for the OLEDs based on neat purine 2 as EML. ............... 167 6 12 EL spectra for purine 2 based OLED and control device. ................................ 168 6 13 (color) EL spectra for the OLEDs based on purine 2 with different ETLs ......... 169 6 14 (color) EL spectra for the OLEDs based on purine 2 with different HTLs ........ 170 6 15 (color) Sche ma tic energy level diagram of OLEDs based on purine 2 with different HTLs ................................ ................................ ................................ .. 171 6 16 EL spectra for optimal OLEDs based on purine 2 ................................ ............ 172 6 17 (color) Device characteristics for purine 2 based OLEDs and control device. .. 173 7 1 (color) J V characteristics of OPV based on BEDOT TQMe 2 .......................... 182 7 2 (color) EQEs of OPV based on BEDOT TQMe 2 ................................ .............. 182 7 3 (color) J V characteristics of OPV based on BEDOT BBT .............................. 183 7 4 (color) EQEs of OPV based on BEDOT BBT. ................................ .................. 184

PAGE 14

14 7 5 (color) Absorption and PL spectra of t Bu BP4M in solution and solid state film ................................ ................................ ................................ ................... 185 7 6 (color) EL spectr a for OLEDs based on t BuBP4M with different doping concentrations ................................ ................................ ................................ 186 7 7 (color) EL spectra for OLEDs based on t BuBP4M with different device structures. ................................ ................................ ................................ ......... 187

PAGE 15

15 LIST OF ABBREVIATION S AFM Atomic force microscopy Ag S ilver Al A luminum Al q 3 T ris(8 hydroxyquinoline) aluminum AMOLED A ctive matrix organic light emitting device BCP B athocuproine BEDOT TQMe 2 4,9 bis(2,3 dihydrothieno[3,4 b ][1,4]dioxin 5 yl) 6,7 dimethyl [1,2,5]thiadiazolo[3,4 g ]quinoxaline BEDOT BBT 4,8 bis(2,3 dihydrothieno [3,4 b ][1,4]dioxin 5 yl)benzo[1,2 c ;4,5 c ]bis [1,2,5]thiadiazole BOLED Bottom emitting OLED BPhen B atho phenanthroline CB C onduction band CBP 4,4' bis(carbazol 9 y l)biphenyl CCD C harge coupled device CCT C orrelated color temperature CIE Commission CRI C olor rendering index CRT C athode ray tube Cs Cesium CT Charge transfer CV Cyclic v oltammetry DAD Donor acceptor donor DCM Dichlorometha ne DPV Differential p ulse v oltammetry

PAGE 16

16 EBL Electron blocking layer EIL E lectron injection layer EL Electroluminescen ce EM Electromagnetic EML E missive layer ETL E lectron transporting layer F Fluorescence HBL Hole blocking layer HIL H ole injection layer HOMO H ighest occupied molecular orbital HTL H ole transporting layer IC I nternal conversion Ir Iridium IR Infrared Ir(ppy) 3 T ri s ( 2 phenylpyridine) iridium (III) ISC I ntersystem crossing ITO I ndium tin oxide LCD L iquid crystal display LUMO L owest uno ccupied mole cular orbital mCP dicarbazolyl 3,5 benzene MLCT M etal ligand charge transfer m MTDATA 4,4 ,4 tris(N 3 methylphenyl N phenyl amino)triphenylamine MO Molecular orbital NIR Near infrared NPD B is[N (1 naphthyl) N phenyl amino] biphenyl

PAGE 17

17 OLED O rganic li ght emitting device OMBD Organic molecular beam deposition Os Osmium OVJP Organic vapor jet printing OVPD Organic vapor phase deposition OXD 7 1,3 bis[2 (4 tert butylphenyl) 1,3,4 oxadiazo 5 yl]benzene PBD 2 (4 tert Butylphenyl) 5 (4 biphenylyl) 1,3,4 oxad iazole PDP P lasma display panel PEDOT Poly(3,4 ethylenedioxythiophene) PHOLED P hosphorescent organic light emitting device PL P hotoluminescen ce PLED P olymer based OLED P LQY P hotoluminescen ce quantum yield PQIr B is(2 phenylquinoline)(acetylacetonate)iridium (III) PSS Poly(styrenesulfonate) Pt Platinum Pt Ar 4 TAP Platinum (II) tetraaryl tetra anthro porphyrin Pt Ar 2 TBP Platinum (II) diaryl tetrabenzoporphyrin Pt Ar 4 TBP Platinum (II) tetraaryl tetrabenzoporphyrin Pt DPTBP Platinum (II) diphenyl tetrabenzoporphyrin Pt TBPs Platinum (II) tetrabenzoporphyrin s Pt TPTBP Platinum (II) tetraphenyl tetrabenzoporphyrin Pt TPTNP Platinum (II) tetraphenyl tetranaphthoporphyrin PVK P oly(9 vinylcarbazole) QCM Quartz crystal monitor

PAGE 18

18 RPM Revolutions per minute Ru Ruthenium SF Sen sitized fluorescence SMOLED S mall molecule based OLED SSL S olid state lighting TAPC 1,1 bis (di 4 tolylaminophenyl)cyclohexane TAZ 3 (4 biphenylyl) 4 phenyl 5 tert butylphenyl 1,2,4 triazole TBAPF 6 Tetrabutylammonium hexafluorophosphate T c T a 4,4 ,4 tris(c arbazol 9 yl)triphenylamine TIR Total internal reflection TOLED Top emitting OLED TPBi 2,2 ,2 (1,3,5 benzinetriyl) tris(1 phenyl 1 H benzimidazole) UGH2 p bis(triphenylsilyly)benzene UV Ultra Violet VB V alence band VDW V an der Waals VTE V acuum thermal eva poration WOLED W hite organic light emitting device 2 AP 2 Aminopurine 3TPYMB T ris[3 (3 pyridyl)mesityl]borane A i Normalized amplitude of i th decay C onversion factor E ex Exciton binding energy E F Fermi level E opt Optical band gap

PAGE 19

19 E tr Transport band gap f G eometric factor G( ) P hotopic response G ( ) S cotopic response I det P hotocurrent I D D evice current J D Device c u rrent density k nr Non radiative decay rate constant k r Radiative decay rate constant k Total decay rate constant L L uminance S olid angle em PL quantum yield R Radiant emittance R v Radiance S 0 G round state S S inglet S( ) S pectrum EL EL lifetime em PL lifetime t avg Average EL lifetime t i Time constant of i th decay T T riplet M obility V V oltage

PAGE 20

20 L C urrent efficiency (or lumin ance efficiency) L P L uminous power efficiency P P ower efficiency IQE I nternal quantum efficiency EQE E xternal quantum efficiency out O utcoupling efficiency

PAGE 21

21 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy EFFICIENT NEAR INFRARED AND ULTRAVIOLET ORGANIC LIGHT EMITTING DEVICES By Yixing Yang December 2011 Chair: Jiangeng Xue Major: Materials Science and Engineering There has been a growing interest in the development of extreme light sources including near infrared ( near IR) and ultraviolet (UV) emitting organic light emitting device s (OLEDs) due to their potential applications. Existing near IR and UV OLEDs general ly have low external quantum efficiencies and limited choice of emitter materials. So new materials and device s need to be developed to achieve OLEDs with tunable emission and high efficiency in near IR and UV ranges. In the first part of this dissertatio n, we achieve very nice coverage of the near IR emission range from 700 nm to over 1000 nm, by using different novel near IR emitting materials. Fluorescent near IR OLEDs based on two donor acceptor donor oligomers, BEDOT TQMe 2 and BEDOT BBT, are first dem onstrated with maximum external quantum efficiency (EQE) up to 1.6% and emission peak wavelengths over 800 nm The efficiencies of these fluorescent OLEDs are further increased by two to three times by using sensitized fluorescence structure to funnel the triplet excitons formed on the host molecules to the flu orescent emitters, which are not utilized in fluorescent devices. Even higher efficiencies and longer emission wavelengths are achieved by phosphorescent near IR OLEDs based on extended conjugation Pl atinum (II)

PAGE 22

22 porphyrins with record high EQE of 9.2% Furthermore, we do realize that the PL efficiencies in solution do not necessarily carry over to the EL efficiencies of devices. In the second part, t owards the short wavelength end of the visible light range a blue violet emitting device has been constructed featuring a highly fluorescent donor acceptor purine in the emissive layer with a maximum EQE of 3.1% and peak emission at 430 nm With different donor groups connected to the purine heterocyc le r ing, another purine derivative with further blue shifted emission is also employed for UV emitting OLEDs achieving a maximum EQE of 1.5 % and emission peak below 400 nm, at wavelength of 393 nm. We realize that, d ue to the high energy band gap of UV emitti ng molecule s it is very important to appropriately align the energy l evels of different layers in multilayer UV OLEDs to efficiently extract the desired light emission.

PAGE 23

23 C HAPTER 1 INTRODUCTION TO ORGANIC LIGHT EMITTING DEVICES 1. 1 Lighting E mitting D evic es (LEDs) There has been an enormous increase in the global demand for energy in recent years as a result of industrial development and population growth. Nowadays, 85% of renewable energy including oil, coal and natural gas, which will be used up in a very close future due to huge demand of energy. 1 6 To balan ce the energy demand and supply, it is imperative to develop new sources of energy such as wind, solar and nucl ear energy. On the other hand, the introduction of efficient electronic devices is also necessary to reduce the consumption of electric power, and therefore mitigate the energy crisis. Currently, research into high efficiency and energy saving solutions is widely studied and gained top priority. Amony all the energy consumed, lighting in buildings in the United States consumes ~765 TeraWatt hours (TWh) of electricity a year, which accounts for approximately 22% of the total electricity, with 40% of that amo unt consumed by inefficient incandescent lamps. 7 The incandescent light usually has the power conversion efficiency as low as 5%, which means 95% of the supplied energy is wasted as heat. A nother common light source, fluorescent tube, has better power conversion efficiency of 20%. However, fluorescent tubes contain hazardous mercury which is not environmentally friendly. 8 Given these f igures, it is easy to see that if the efficiency of lighting can be increased to more than 20%, it would bring tremendous energy savings. One alternative solution is to use the solid state lighting. Rather than electrical filaments or plasma, solid state lighting (SSL) uses semiconductor materials based light emitting devices (LEDs) as sources of illumination with high effciency Conventional

PAGE 24

24 LEDs are made of inorganic semiconductors, mostly g roup III nitride which have direct band gaps and can convert th e electric energy directly into the light emission with less energy loss. Although almost 100% internal quantum efficiency could be achieved by using inorganic LEDs 9 there are still a few is sues such as high co st of materials and processing low color rendering index (CRI) for white light sources, and difficulty for device scalability. Nevertheless, the inorganic LEDs market has been expanded to more and more fields in the last few years. 10 Figure 1 1 shows several examples of white LEDs for application of lighting. Figure 1 1. (color) Example s of white LE Ds for lighting. A) A white LED from Cree. B ) Recessed downlight. C ) Audi LED headlight D ) Portable desk/task lighting. As the name indicates, o rganic light emitting devices (OLEDs) a new generation of LEDs, use the organic materials including polymer s as the active electroluminescent materials in response of the input electric power. Since the OLEDs were first demonstrated by Tang and Vanslyke in 1987 at Eastman Kodak, 11 extensive research ha s been conducted in academia and industry to achieve comparable high efficiency

PAGE 25

25 and pract ical OLEDs. Figure 1 2 schematically compares the performance and progress of different light sources including inorganic and organic LEDs 12 It clearly shows that the performance of OLEDs has developed very fast and now has surpassed the conventional low efficiency incandescent light and can be competitive with the LEDs and fluorescent light. Other than that, OLEDs have many advantages over the other light sources, such as low cost, low temperature processing, compatible with flexible and large area substrates, tunable material properties via structure modification. Details will be discussed in the following section. Figure 1 2 (color) P erformance and current progress of several light sources including inorganic and organic LEDs. 12 The performance is presented by power eff iciency (lm/W). 1 2 Advantages of Organic Light E mitting Devices There are many advantages for organic semiconductors used for organic light emitting devices. To date, about two million organic compounds have been made and this constitutes nearly 90% of a ll known materials in the world. So, organic materials provide a huge amount of choices for specific device fabrication requirements. And their

PAGE 26

26 electronic and optical properties can be easily tailored by chemical structure modification. For example, the em ission spectra of the phosphorescent cyclometalated platinum complexes developed for organic light emitting devices can be tuned throughout the visible spectrum by simply changing the functional cycl ometalating ligand conjugation, 13 as shown in Figure 1 3. Most organic materials are inexpensive, although currently the cost for some materials may be substant ial due to the low yield of chemistry synthesis or difficulties in the purification. Figure 1 3 (color) C ommission (CIE) color coordinates of various phosphorescent cyclometalate platinum complexes. 13 Another advantage of organic materials is the compatibility with low cost and large area manufacturing processes. Vacuum the rmal evaporation (VTE) 14 or organic vapor phase deposition (OVPD) 1 5,16 are usually used to deposit small molecular organic thin films. Polymer semiconductors can be processed from solution by casting, spin coating and ink jet printing. 17 22 These processes only require at or nea r room

PAGE 27

27 temperature processing, which is significantly lower than required for most inorganic materials. Some of these deposition techniques are compatible with large area, flexible 23 and roll to roll processing, 24 30 leading to potentially hig h throughput manufacturing of organic electronic devices. Figure 1 4 shows some flexible OLEDs (FOLEDs) applications available in the market. Figure 1 4 (color) Examples of flexible OLEDs (FOLED) applications : A ) 4.5 inch f lexible AM OLED display by Sam sung Mobile Display (SMD ); B ) F lexible OLEDs on garments OLED displays also have many merit s such as fast response time, wide viewing angle, high contrast, and low powe r consumption, compared to the main competitor, liquid crystal display s (LCD s ). 1 3 Application s of Organic Light E mitting Devices There are mainly two application s of OLEDs especially the white light OLEDs (WOLED) T he first is the flat panel display device where OLEDs are considered as the next generation display technique and will co mpetes with the conventional LCD s T he other application is to be used as the next generation solid stae lighting source to replace the conventional low efficiency incandescent light bulbs and even fluorescent tubes There are already many commercialized p roducts in the market for both applications of OLEDs.

PAGE 28

28 1 3.1 F lat P anel D isplay LCDs have replaced cathode ray tube (CRT) displays in more and more applications and been due to its advantages of wide variety o f display sizes, energy efficient and environment safe. Compared to LCD s OLEDs are capable of providing markedly better performance features, including thinner and lighter design, faster response times, wider viewing angles, higher contrast ratios and bri ghter, more saturated colors. OLEDs can also be more energy efficient than LCDs with operating lifetimes now in the tens of thousands of hours. The fabrication of OLED diplays also has the potential to be cost effective with those low cost and high through put manufacturing methods like roll to roll processing. There have been many commercialized products in the market, as shown in Figure 1 5. Figure 1 5 (color) Variou s OLEDs display applications: A ) 11 OLED display (Sony ) ; B ) 4 active matrix OLED d isplay for Galaxy S smartphone (Samsung ) ; C ) 31 OLED TV ( LG ) ; D ) 40 OLED TV (S amsung )

PAGE 29

29 1, was released by Sony in 2007 as shown in Fig ure 1 5 A with only 3 mm thick. Nowadays, active matrix OLED ( AMOL E D) has been widely used for cell pho ne screen display. Figure 1 5B shows the world largest flexible AMOLED with a WVGA 800x480 pixel resolution by Samsung Since the relatively small size of OLED displays have been widely introduced to the market l arge size prototype OLED TVs ( and above ) have been also continuously reported as shown in Fig ure 1 5 C and D Several companies including DuPont and LG had stated that they can produce the even larger size OLED TVs up to 50 inch in minutes Moreover it is expected that the OLEDs would be more compatible in realizing the 3 D images due to their fast response 1. 3.2 White OLEDs L ighting The primary requirements for a general illumination light source lie in three aspects: high power efficiency at high brightness, high CRI and high stability at high brightness. The conventional incandescent lamps are widely used but very inefficient (~15 lm/W). This has generated increased interest in the use of W OLEDs, owing to their potential for significantly improved efficiency and highe r CRI over incandescent sources combin ing with low cost, high throughput manufacturability. Currently, the efficiencies of WOLEDs can be achieved by 4 0 60 lm/W 31 33 and this value can be increased to more than 100 l m/W if light outcoupling enhancement techniques are employed. Meanwhile, lifetime s longer than 10,000 hours for WOLED are also demonstrated. In terms of efficiency and stability, WOLED lighting ha s definitely surpassed the incandescent lamps and can be eve n competitive with the fluorescent light tubes. Except for the merit s of high efficiency and stability WOLED is also a surface light emitting source which allows great design freedom in producing new concept lightings.

PAGE 30

30 They can be fabricated on any subst rates including the flexible ones and are able to make lighting devices in any shapes and designs. Figure 1 6 sh ows several existing white OLED lighting products. Figure 1 6 (color) Examples of OLEDs lighting applications : A) 30 cm lighting panels; B ) GE ; C ) Flexible OLED general lighting by GE and Konica Minolta 1.4 High Efficiency Blue and White PH OLEDs The OLED applications such as flat panel display and solid state lighting require integrating individual pixels of blue ( B), green (G), and red (R) emission to realize the white OLEDs. While green emitting phosphorescent OLEDs (PHOLEDs) with nearly 100% internal quantum efficiency have been demonstrated, 34,35 improving the efficiency and stability of blue emitting OLEDs is, however, one of the remaining challenges in OLEDs for full color display and lighting source applications. Blue phosphorescent emitting materials usually have higher triplet excited energy levels as the blue emissio n goes to shorter wavelengths. So it is much more difficult to effectively confine the

PAGE 31

31 exctions in the blue emitters by using the typical host and charge transporting materials which work well for g reen and red emitters in OLEDs. Most of the works on blue OLEDs have been focused on phosphorescent metal organic complexes such as FIrpic 36 an d FIr6. 37,38 FIr6 based devices show deep blue emission, 37,39 while FIrpic exhibits greenish blue color, which is not sufficient for practical full color displays and lightin g applications. 40 However, FIr6 has a higher triplet energy (T 1 ) of 2.72 eV, 41 compared to 2.65 eV for FIrpic. 36 Holmes et al. used a wide gap material, UGH2 with high triplet energy of 3.5 eV 41 and inserted an mCP layer (T 1 = 2.9 eV) 36 between the hole transporting la yer and the emissive layer to serve as the efficient electron/exciton blocking layer. 37 A maximum external quantum eff iciency ( EQE ) of 12% and power efficiency ( P ) of 14 lm/W at low luminances were demonstrated by this device structure. The power efficiency can be further improved by employing the p i n dual emissive layer (D EML) structures. A low turn on voltage of 3.2 V and a maximum power efficiency of 25 lm/W were achieved. 42 By carefully choosing the charge transporting layers with high triplet energies to better confine the excitons, maximum EQE of 20% and peak P of 36 lm/W were achieved maintaining CIE coordinates of (0.16, 0.28). 43 Some existing efficient blue OLEDs are listed in Table 1 1. Nonetheless, alternative materials and device structures are still needed to further enhance the performance of blue OL EDs, in order for display and lighting applications. Since the first demonstration of white organic light emitting devices (WOLEDs) by Kido et al., 44 47 there have been many different approaches, trying to enhance the device efficiency so that WOLEDs can be used for solid state lighting and displays to replace the conventional products. These devices have now surpassed incandescent

PAGE 32

32 light bulbs (~15 lm/W) in terms of efficiency and lifetime. Active research has been focused on further improving their device efficiency to beyond that of fluorescent light (60 90 lm/W) and achieving high color rendering indices (CRI). Su et al. has reported high efficiency white PHOLEDs with a peak power efficiency of 44 lm/W using two p hosphorescent dopants of a greenish blue emitter, FIrpic, and a red emitter, PQ2Ir. 32 Due to the limited spectral coverage of the two emitters this device possessed a low CRI of only 68 in spite of the high efficiency. To improve the spectral coverage of WOLEDs, a blue fluorescent emitter in combination with green and red phosphor dopants has been exploited in the EML with a high CRI of 85, and maximum P of 22 lm/W. 48 WOLEDs based on all phosphorescent dyes of deep blue emitter, FIr6, green emitter, Ir(ppy) 3 and red emitter, PQIr have also been demonstrated with maximum EQE of 12% and CRI of 80. 49 The efficiencies can be further improved with enhanced charge and exciton confinement in the EML by charge transporting layers as reported by Xue et al., achieving maximum EQE of 19% and peak power efficiency of 40 lm/W. 31 Only recently, Universal Display Corpo ration (UDC ) has announced all phosphorescent WOLED lighting panels with record high power efficiency of 58 lm/W CRI of 83, and operating lifetime of 30,000 hours (to 70% of initial luminance) Some recent research progress of efficient WOLEDs is liste d in Table 1 1. Moreover, the planar OLEDs structure will typically suffer from low light extraction efficiency of approximately 20%. The light output from emissive layer in OLEDs will be reduced by absorption losses and wave guiding within the device and its substrate due to total internal reflection (TIR) at multiple interfaces. 34,50 Numerous works have been reported to further improve the light extraction efficiencies in OLEDs, 51 54 and recent

PAGE 33

33 reports from UDC show the peak power efficiency higher than 100 lm/W has been demonstrated with incorporation of light outcoupling enhancement techniques, which is already very competitive with the conventional fluorescent light source. Table 1 1 Recent research p rogress of blue (B) and white (W) OLEDs. The maximum external quantum ( EQE ) and power ( P ) efficiencies t he CIE ( x,y ) coordinates and CRI are shown for comparison. OLED architecture EQE (%) ( max ) P ( lm/W ) ( ma x ) CIE (x, y) CRI B1 FIr6 as blue emitter and UGH2 as host with mCP exciton blocking layer 37 12 % 14 lm/W (0. 16 ,0.26) B2 p i n D EML with FIr6 as blue emitter 42 17% 25 lm/W (0.16,0.28) B3 p i n D EML with FIr6 as blue emitter and 3TPYMB as ETL 43 20% 36 lm/W (0.16,0.28) W1 WOLED with two phosphorescent dopants of FIrpic and PQ2Ir 32 25 % 44 lm/W (0.34,0.40) 68 W2 Fluorescent blue emitter with green and red phosphor dopants 48 11% 22 lm/W (0.38,0.40) 85 W3 Phosphorescent triple doped EML 49 12% 26 lm/W (0.43,0.45) 80 W4 Phosphorescent triple doped D EML with p i n 31 19% 40 lm/W (0.37,0.40) 79 W5 All phosphorescent WOLED lighting panel ( UDC ) 58 lm/W 83 1.5 Near Infrared and Ultra Violet OLEDs Since the first report of near infrared ( NIR) emission from an organic light emitting device (OLED) based on a lanthanide complex, 55 there has been considerable interest in the development of materials and devices that exhibit wavelength tunable and efficient light emission in the NIR wavele ngth region. Such work is of interest for a variety of applications in security and defense, telecommunications, and biomedical sectors 56 63 as shown in Figure 1 7. For example, NIR OLEDs could serve as a new clas s of illuminators for night vision, providing advantages such as light weight, low thermal signature, low power consumption, and compatibility with large area and flexible

PAGE 34

34 substrates. Much of the existing work on NIR OLEDs has focused on the use of lanthan ide containing organometallic complexes based on Yb, Nd, and Er. 55,56,64 73 The metal phthalocyanines (Pc) with emissions at the near IR range are also used for the device demonstration. 74,75 However, because of the inherent low efficiency of emission from these complexes, the most efficient OLEDs based on these materials possess external quantum efficiencies (EQE) less than 0. 3 %. Phosphorescent transition metal organic com plexes provide attractive candidates for use in PLEDs and OLEDs due, in part, to their high luminescence quantum yields and nearly 100% internal quantum efficiencies due to their ability to efficiently emit from the spin orbit mixed triplet states. 34,35 Figure 1 7 (color) Various near IR light source applications: A ) Night vision; B ) Heating; C ) Thermal bandage healing; D ) Telecommunication. A recent report of an electrophosphorescent device that used a cyclomet alated [ (pyrenyl quinolyl) 2 Ir(acac)] complex as the phosphor gave max = 720 nm and an EQE of 0.1%. 76 More specifically, a subclass of ph osphorescent metal organic complexes, metalloporphyrins, has shown intense absorption and emission in the red to NIR region

PAGE 35

35 of the spectrum. 77,78 There are a number of reports of OLEDs fabricated with PtOEP, PtTPP, or analogues of these compounds as phosphorescent emitters, with emission maxima between 630 and 650 nm. 35,79 86 Porphyrin chromophores with fused aromatic moieties at the pyrrole positions, for example, tetrabenz oporphyrin (TBP), exhibit a bathochromic shift (relative to unsubstituted porphyrin) of the absorption and emission energy, owing to the expansion of the electronic system of the porphyrin core. 87 The addition of bulky gro ups to the meso positions of the porphyrin macrocycles with substituted pyrroles leads to the formation of nonplanar porphyrins, and further re d shifts the absorption spectra Based on these considerations extended Pt porphyrin complexes such as Pt tetrapheny ltetrabenzoporphyrin (Pt TPTBP ) have been studied fo r NIR OLEDs with a reported maximum EQE of 8.5% and a peak wavelength of 7 70 nm. 88,89 Despite the high efficiencies, there still remains the need for the development of alternative materials and devices that exhibit wavelength tunable and highly efficient emission at > 800 nm. Another interest has recently turned to organic systems that emit at the other end of the visible range, the shorter wavelengths from the deep blue, through the violet, to the ultraviolet regions. Ultraviolet (UV) to deep blue organic light e mitting diodes (OLEDs) have found, or are sought for, applications in biological and chemical sensing, 90,91 sterilization, high density information storage devices, 92 and full color light emitting displays, 93,94 as shown in Figure 1 8. Recently, there have been several reports 95 104 of high efficiency fluorescent blue to violet OLEDs with a peak emission wavelength in the range of 40 0 4 5 0 nm that possess maximum EQE values up to 3 4 %, although the shorter wavelength emission generally leads to lower device efficiencies.

PAGE 36

36 Despite a number of examples of deep blue to UV emitting OLEDs based on organ ic fluorescent emitters including smal l molecules 101 108 and polymers, 99,100,109,110 only a few 106,108 realize the device emission peak wavelengths below 400 nm, which c an be considered as the UV emission, with external quantum efficiencies (EQE ) greater than 1%. However, these so called UV OLEDs still have relatively broad emission peaks, with emission tail extending as far as to the longer wavelength ranges of 450 550 n m. Figure 1 8 (color) Various ultra violet light source applications: A ) Protein in jellyfish that glowed bright green under UV light (Nobel Prize 2008); B ) A collection of mineral samples brilliantly fluorescing at various wavelengths as seen while being irradiated by UV light; C ) Blu ray disc using blue violet laser to read the information; D ) Sterilization of food and tools. The challenges of fabricating UV OLEDs lie in two parts: one is the molecular design of materials capable of efficient UV emission, 106 and the other is appropriately designed device structure, 108,111,112 where both hole and electron carriers are injected into a wide energy band gap UV emitting material and non UV em issions from longer wavelength emitting materials need to be prevented. So, h ighly sought are organic m aterials that can achieve emissive color tunability and provide high efficiency and

PAGE 37

37 brightness in the UV to violet region and device structures with app ropriate energy levels alignment. 1. 6 Outline of D issertation T his dissertation endeavors to deliver the fundamental background of OLEDs as well as material study and device design for high efficiency near IR and UV OLEDs. First, Chapter 2 will explain the optical, electrical, and physical properties of organic semiconductors. Chapter 3 will help understanding more details about OLEDs from the organic thin film growth to OLED measurements. The important terminologies and definitions for understanding the OL ED operation and evaluation will be described in Chapter 3 as well. In Chapter 4 and 5, high efficiency fluorescence and phosphorescence near IR OLED s will be demonstrated by employing donor acceptor donor oligomers and Platinum (II) porphyrins, respective ly. The effect of molecular structure for both types of NIR emitters on the device performance will also be studied. In Chapter 6, efficient blue violet and violet UV emitting OLEDs will be demonstrated based on the highly fluorescent donor acceptor purine s The challenges of fabricating UV OLEDs will be realized and then the effect of ener g y level alignment in the multilayer devices on acquiring d esired device emissions will be investigated. F inally, this dissertation will make conclusions and leave some f uture topics for further OLEDs especially UV emitting OLEDs as well as for organic photovoltaic (OPV) device research in Chapter 7

PAGE 38

38 CHAPTER 2 PHYSICS OF ORGANIC SEMICONDUCTORS 2. 1 Introduction to Organic Semiconductors Organic semiconductors differ sig nificantly in their optical, mechanical, and electronic properties from the conventional covalent solid semiconductors such as Si, Ge, II VI and III V semiconductors, which have been used as the active materials in electronic and optoelectronic devices. Th is can be mainly attributed to the different bonding interactions between the constituent atoms or molecules for the organic and conventional semiconductor materials. The majority of organic materials belong to the category of molecular solids, and the int ermolecular interaction forces that hold the molecules together is primarily of the van der Waals (VDW) force, which is considerably weaker than that of covalent or ionic bonds in the conventional semiconductors. 113,114 Organic semiconductors have a number of advantages over their inorganic counterp arts. To date, about two million organic compounds have been made and this constitutes nearly 90% of all known materials in the world. So, organic materials provide a huge amount of choices for a particular application. And their electronic and optical pro perties can be easily tailored by chemical structure modification. 13 Most organic materials are inexpen sive, although the cost for some materials may be substantial due to the low yield of chemistry synthesis or difficulties in the purification. Another advantage of organic materials is the compatibility with low cost and large area manufacturing processes. Vacuum thermal evaporation (VTE) 14 or organic vapor phase deposition (OVPD) 15,16 are usually used to deposit small molecular organic thin films. Polymer semiconductors can be processed from solution by casting, spin coating and ink jet printing. 17 22 These processes require at o r near room temperature

PAGE 39

39 processing, which is significantly lower than required for most inorganic materials. Some of these deposition techniques are compatible with large area, flexible and roll to roll processing, 24 30 leading to potentially high throughput manufacturing of devices. Despite the above advantages, organic semiconductors face some challenges for use in electronic and optoelectronic devices. Due to the weak intermolecular interactions of van der Waals f orce, organic materials generally have much lower carrier motilities than inorganic materials. The purification of organic materials is also problematic, as no available purification method can achieve organic materials with purities reaching that of, for example, silicon. Finally, most organic materials are susceptible to degradation when exposed to water vapor, oxygen and other contaminations. So, their reliability and stability in electronic and optoelectronic devices are in question. Figure 2 1 (colo r) Classification of organic semiconductors with increasing structural complexity. 115 The molecules shown from left to right are: an Al quinoline complex small molecule (left), a polymer PPV (middle), and a protein complex (right). Depend ing on the structural complexity and molecular weight, organic semiconductors can be furthe r divided into three types: small molecules, polymers, and biological molecules, as shown in Figure 2 1. 115 Small molecules usually have molecular weight less than 1000 and have well defined molecular structures. Polymers

PAGE 40

40 are long chain molecules consisting of an undefined number of repeating units. Their molecular weight can b e anywhere from a few thousand to a million. The molecules having the highest structural complexity are of biological origin, such as DNA and protein. Small molecules and polymers have been extensively used as the active materials in electronic and optoele ctronic devices, while the use of biological molecules in electronic applications has not been clearly demonstrated. 2.2 Electronic Structures of Organic Semiconductors Figure 2 2 shows several examples of typical organic semiconductors used in electro nic and optoelectronic devices. The framework of the organic molecules is usually composed of the carbon atoms connected by different bonds such as single and double bonds. Other atoms, for example, sulfur or nitrogen atoms, sometimes are also incorporated For the carbon atom, the electronic configuration in the ground state is 1 s 2 2 s 2 2 p 2 as shown in Figure 2 3, with two paired electrons in 2 s orbital and two unpaired electrons in two 2 p orbitals. Figure 2 2. Examples of several organic semiconductors i ncluding small molecules and polymers used in electronic and optoelectronic applications.

PAGE 41

41 Due to the Pauli Exclusion Principle, only the unpaired 2 p electrons in carbon atom may be shared with other atoms to form covalent bonds. The paired 2 s orbital, howe ver, may mix or hybridize with one or more 2 p orbitals, leading to the hybridized orbitals of sp sp 2 and sp 3 with single electron occupancy in each orbital, as shown in Figure 2 3. The hybridized orbital configurations are more favorable than the ground state for the carbon atoms to form covalent bonds with other atoms, as all four valence electrons are now unpaired and available for bonding. Figure 2 3. Electronic configurations of a carbon atom in the ground state and sp n hybridized states ( n = 1, 2, 3). When two carbon atoms are bonded together, the inter atomic interaction leads to the formation of bonding and anti bonding molecular orbitals (MOs). Figure 2 4 shows the formation of the molecular orbitals of two adjacent carbon atoms with sp 2 hybridi sp 2 hybridized orbitals, one from each carbon atoms. Similarly, the remaining two sp 2

PAGE 42

42 bonding MOs with s p or sp n formed between the two p z opposite tren d. So the four electrons will occupy the low lying energy levels first, and a ormed, as shown in Figure 2 4A with the two sp 2 hybridized orbitals connected in the head on geometry, whereas the two p z orbi as illustrated in Figure 2 4B The Figure 2 4. Molecular orbitals (MOs) of two adja cent carbon atoms with sp 2 hybridization: A bonding and anti bonding MOs; B ) Schematic illustration of the bonding between the two adjacent carbon atoms. bonding molecular orbital s are the highest occupied and lowest unoccupied molecular orbital s (HOMO and LUMO, respectively) for this molecule. Since anti bonding MOs play an important role in determining various electronic properties of the molecule.

PAGE 43

43 For a molecule containing several atoms, the electronic structure of this molecule can be treated in the same way as the situation of two atoms, althoug h the results may be much more complicated. Figure 2 5A shows a simple example of the electronic structure of benzene (C 6 H 6 ) molecule. All the six carbon atoms have the sp 2 hybridization and are connected with alternating single and double bonds. A delocal ized super p z orbitals, leading to all six carbon atoms and the bonds between them being equivalent. Hence, rather than using alternating single and double bonds, the structure of benzene with a circle inside the carbon hexagon skeleton, as sh own in Figure 2 5A Figure 2 5. A ) Resonance molecular structure of a benzene molecule (C 6 H 6 ) and the p orbitals. B ) Extended conjugation system. The e xtension of the conjugation will lead to a general delocalization of the electrons across all of the adjacent parallel aligned p orbitals of the atoms, which will lower the overall energy of the molecule and enable the charge transport within the

PAGE 44

44 organic m olecule, as shown in Figure 2 5B Additionally, the band gaps of the organic molecules are strongly affected by the degree of conjugation, i.e. more conjugated molecules have smaller band gaps and thereby longer absorption maximum. 2.3 Charge Transport In general, the dynamics of charge carrier motion in solids has two extreme types: one is highly delocalized motion in the form of a Block wave, and the other is highly localized as a result of interactions with the local surroundings. Actual systems fall bet ween these two extremes due to the competition between energies associated with the delocalization and localization processes. Figure 2 6 E nergy level diagrams of a single molecule in gas phase, ionized electron and hole pairs in the solid crystal and a disordered Gaussian density of states in an amorphous solid. 116 Localizing a charge carrier in wave like motion to form a wave packet requires certain energy. Such a localization energy is a direct result of the uncertainty principle. Narrow bands are formed in organic materials with weak intermolecular interactions. The large density of states in a narrow band leads to a relatively small localization energy, and the tendency towards carrier localization is more pronounced in organic

PAGE 45

45 materials. 113,114 On the other hand, the interaction between a charge carrier and its electronic and nuclear surroundings causes local polarization, which lowers the total energy of the system, as shown in Figure 2 6. 116 The change in the total energy is called the polarization energy. For the amorp hous solids, locally different polarization energies create a Gaussian distributed density of states for transporting sites. The mobility for delocalized charge carriers is usually high, as they transport in wide energy bands. For strongly localized carriers, they move via stochasticall y uncorrelated hopping motion from site to site, and the mobility is generally low and is thermally activated. Organic semiconductors usually have several orders of magnitude lower charge transport capabilities compared to their inorganic counterparts A strong covalent bond is formed in inorganic semiconductors creating the delocalized states and leading to the continuous transport energy band w hereas, a weak VDW interaction creates discontinuous localized states in organic semiconductors resulting to the less efficient charge hopping transport. 2.3.1 Hopping T ransport Most organic thin films are in an amorphous solid state, and the weak VDW interactions throughout the amorphous structure creates discontinuous localized states in the organic molecules, instead of a continuous transporting energy band So an intermolecular hopping transport mechanism dominates in the amorphous solid state. The typical charge carrier mobility ( ) of amorphous organic semiconductors is in the range of 10 3 10 10 cm 2 /V s. T he mobility in amorphous organic semiconductors can be expressed as a function of electric field and temperature : 117

PAGE 46

46 w here F is the electric field, T is the temperature, is the activation energy for interm olecular hopping, k is Boltzmann constant, and is a constant. 2.3.2 Band T ransport Band transport has been observed in high quality organic molecular crystals such as pentacene anthracene and naphthalene, with a slightly delocalized transport energy band and room temperature mobility of RT 2 /V s. 114,118,119 The mobility for crystalline organic semiconductor has a temperature dependence : 120 However, in most cases, the weak intermolecular interactions and the strong polarizability of organic molecules lead to localization of charge carriers in organic materials, and a thermally activated mobility with RT < 1 cm 2 /V s is obtained. 114 2. 3.3 T ransport vs. O ptical Band G aps Typically for inorganic semiconductors, ther e is a negligible difference between the optical band gap ( E opt ) and transport ban d gap ( E tr ) due to the small exciton binding energy (~10 meV) in delocalized energy states. However for organic semiconductors, the E opt and E tr should be differentiated due to the s trong exciton binding energy (~1 eV) in the localized organic molecule s 114 The accurate relationship of E opt and E tr for organic semiconductors can be expressed as : E opt = (E gap E p ) E ex = E tr E ex w here E gap is the energy level difference between the HOMO and LUMO in a single moelcule, E p is the energy loss due to polarization, and E ex i s the exciton binding energy. From this relationship, it can be found that the optical b and gap corresponds to exciton gap and is obtained from optical absorption, whereas transport band gap refers to the

PAGE 47

47 gap separating the electron and hole transport level. Optical band gap is smaller than the transport band gap. 2. 4 Excitons 2. 4 .1 Classific ation of E xcitons Due to the strong tendency of localization of charge carriers, light absorption or charge injection in organic semiconductors leads to the creation of excitons, or bound electron hole pairs, instead of free electrons and holes as in inorg anic semiconductors. The excitons are the excited states of the molecules and can be treated as chargeless particle capable of diffusion. Figure 2 7 Schematic illustration of three types of excitons in a solid: 121 A ) Frenkel exciton, localized on single molecule; B ) Charge transfer exciton, slightly delocalized over two o r several adjacent molecules; C ) Wannier Mott exciton, highly delocalized with a radius much greater th an the lattice constant. There are three types of excitons in solids based on different extent of charge delocalization: the Frenkel exciton, charge transfer exciton, and Wannier Mott exciton, as shown in Figure 2 7. 121 The Frenkel exciton has the electron and hole localizing on the same molecule for the excited state. And the radius of the exciton is less than or close to the lattice constant. With the increasing intermolecular interactions, the excited

PAGE 48

48 state may delocalize over two or several adjacent molecules, forming the charge transfer (CT) exciton. Frenkel and CT excitons are typically present in organic semiconductors. For the conventional inorganic semiconductors, the hig hly delocalized hydrogen like Wannier Mott excitons with a radius much greater than the lattice constant are commonly found. Due to the Coulombic interaction between the constituent electron and hole, the Frenkel or CT excitons are tightly bound, with a b inding energy ranging from 0.1 to 2 eV, 122 124 while that for Wannier Mott excitons in inorganic semiconductors is only a few meV. 125 Similar to the tran sport of single carriers, exciton migration or diffusion among the same species of molecules can occur through band transport or hopping process. The weak intermolecular interactions also lead to a limited exciton mobility in organic semiconductors. 2. 4 .2 Multiplicity of Excitons Based on the spin statistics, the symmetric and anti symmetric wave functions ( s a nd a respectively ) of the excited state for a two electron system can be expressed as : s = 1 2 s = 1 2 symmetric states ( S = 1, triplet ) s = 1 2 1 2 a = 1 2 1 2 anti symmetric state ( S = 0, singlet ) w here n (n=1,2) is the spin function of electron and ( ) or ( ) rep resent the possible spin states of elec tron. For the states of symmetric s the spin states of the two electrons are parallel and the total spin S = 1, while the two electrons have the opposite spin states for the state of anti symmetric a with the total spin S = 0. So if the formation

PAGE 49

49 probability of singlet and triplet excited state is identical, then the ratio of singlet and triplet excitons can be expected to be 1:3, which means the singlet only comprise 25% of the excitons. Figure 2 8 shows the r adiat ive relaxation processes of fluorescence and phosphorescence from singlet and triplet excitons respectively Triplet excited state has lower energy because it is spatially symmetric under exchange of electrons. As shown in Figure 2 8, fluorescence decays from singlet excited state to the ground state with conserved symmetry and is usually a very fast (~10 9 s) process. Disallowed by the symmetry, phosphorescence decays from the triplet excited state to the ground state slowly (1 ms to 1 s). Figure 2 8 (color) Schematic illustration of f luorescence (decay from the singlet exciton left) and phosphorescence (decay from the triplet exciton right). (Courtesy of V. Bulovic) 2. 4 3 Metal Ligand Charge Transfer Exciton The singlet and triplet excited states n eed to be mixed to make both singlet and triplet decay allowed, so that all the excitons can contribute to the radiative emission. This can be achieved by utilizing metal organic complexes with heavy transition metals

PAGE 50

50 such as iridium (Ir) platinum (Pt) o sm ium (Os) and ruthenium (Ru), due to very strong spin orbit coupling which is proportional to the atomic number (Z 4 ) of heavy metals Figure 2 9 (color) Two types of metal organic phosphors with heavy transition metal s : 126 A ) Type I: exciton localized on organic ligand; B ) Type II: metal ligand charge transfer exciton (MLCT). There are two types of metal organic phosphors based on the different extent of spin orbit coupling, a s shown in Fig ure 2 9. 126 For the Type I phosphor such as Pt complex, excitons are localized within the organic ligands with less singlet and triplet states mixing. H eavier transiti on metals such as Ir can effectively enhance the state mix ing generating emissive metal ligand charge transfer (MLCT) excitons which delocalize among the metal atom and organic ligands This Type II phosphor has much lower triplet lifetime (~1 s) than t he less mixing Type I phosphor (~100 s) due to the spin orbit coupling. Therefore, it is theoretically possible to convert all the singlet and triplet excitons to phosphorescence, leading to almost 100% photon conversion efficiency by using these organome tallic compounds as emitters. 34,35 2. 5 Intra M olecular E nergy T ransfer After the excitons are formed, there are several routes for the excitons to relax back to the ground state (S 0 ) within the same species of mol ecule. Figure 2 10 shows the electronic states of a molecule and the possible transition process between them.

PAGE 51

51 This energy diagram is also called Jablonski diagram. In this diagram, the electronic states are arranged vertically by energy and grouped horizo ntally by spin multiplicity. The vibrational ground states of each electronic state are indicated with thick lines, the higher vibrational states with thinner lines. Non radiative transitions are indicated by squiggly arrows and radiative transitions by st raight arrows. The time scale for each transition is also indicated. The Jablonski diagram shows what sorts of transitions that can possibly happen in a particular molecule. Each of this possibility is dependent on the time scale of each transition. The fa ster the transition, the more likely it is to happen as determined by selection rules. Figure 2 10 (color) Jablonski energy diagram. 127 The electronic states are arranged vertically by energy and grouped horizontally by spin multiplicity (singlet: S, triplet: T). The vibrational ground states of each electronic state are indicated with thick lines, the higher vibrational states with thinner lines. Non radiative transitions are indicated by squiggly arrows and radiative transitions by straight arrows. The time s cale for each transition is also indicated. IC and ISC are referred to internal conversion and intersystem crossing, respectively.

PAGE 52

52 2. 5 1 Absorption The absorption process occu r s with excitation energy of photon ( ) larger than the band gap and there is a series of wavelength s that can lead to a transition from the ground state to the excited states (S 0 S 1,2 ). Thus, electronic absorption spectra generally show the broad bands, rather than the single lines. Abso rption transition is a very fast process, on the order of 10 15 s. 2.5.2 Vibrational Relaxation and Internal Conversion Once the electron on the ground state is excited to the higher energy excited states by the absorption of photon, the exciton is generat ed. There are several ways that the energy may be dissipated. The first is through vibrational relaxation, which is a non radiative process. This relaxation occurs between different vibrational states within the same electronic state. The dissipated energy through the vibrational relaxation may stay within the same molecule, or it may be transferred to other molecules around the excited molecule, depending on the phase of the probed sample. This process is also very fast (10 14 10 10 s), so it is very likel y to occur immediately following absorption. A transition may also happen from a vibrational state in one electronic state to another vibrational state in a lower electronic state, which is called internal conversion (IC). Internal conversion occurs becaus e of the overlap of vibrational and electronic states and is mechanistically identical to vibrational relaxation. The molecular spin state remains the same for both vibrational relaxation and internal conversion transitions. 2. 5 3 Intersystem C rossing The radiationless transition between different vibrational states with different spin multiplicity is another path for the dissipation of energy, called intersystem crossing

PAGE 53

53 (ISC). In this process, a singlet excited state non radiatively transfers to a triplet excited state, or the corresponding converse process from a triplet excited state to a singlet excited state. The probability of this transition occurring is more favorable when the vibrational levels of the two excited states overlap, since little or no energy will be gained or lost in the transition. Intersystem crossing is most common in heavy atom containing molecules due to the spin orbit coupling effect. 2. 5 4 Fluorescence Once excitons are generated, they quickly relax to the lowest vibrational lev el of an ex c ited singlet state via vibrational relaxation or internal conversion processes. Then excitons can radiatively decay from the singlet excited state to the ground state ( S 1 S 0 ) with conserved symmetry, called fluorescence. Fluorescence can only exploit the singlet excitons ( about 25% of the total excitons ) and has a short radiative lifetime of approximately 10 9 to 10 6 s. Delayed fluorescence may happen as the triplet excited state transfer back to the singlet excited state via intersystem cros sing, leading to the emissive transition to the ground state. 2. 5 5 Phosphorescence In the intersystem crossing transition, the singlet excited state non radiatively transfers to the triplet excited state. The triplet excited state then undergoes the int ernal conversion and falls to the lowest vibrational level of the triplet electronic state. The radiative transition from the lowest vibrational level of the triplet electronic state to the ground state (T 1 S 0 ) is possible with strong spin orbit coupling This transition is called phosphorescence and is a very slow process, on the order of 10 3 to 1 s The molecules are therefore able to emit slowly after the excitation.

PAGE 54

54 2. 5 6 Frank Condon S hift Vibronic transitions are the simultaneous changes in elect ronic and vibrational energy levels of a molecule due to the absorption or emission of a photon of the appropriate energy. During an electronic transition, a change from one vibrational energy level to another will be more likely to happen if the two vibra tional wave functions overlap more significantly. Figure 2 1 1 Frank Condon principle energy diagram. The vibronic transitions in a molecule with Morse like potential energy functions in both the ground and excited electronic states are shown. (Courtesy of J. Xue) Figure 2 11 show the energy diagram of Frank Condon principle. 114 The absorption is dominated by transition from the zero order vibrational mode of ground state to the higher order vibrational mode of the excited state (S a 0 S a2 ). The excited electrons then undergo a fast vibrational relaxation by releasing heat to the zero order mode (S a 2 S a 0 ). Light emission occurs from transition of the zero order vibrational mode of the excited state to various vibrational mode s of the ground state (S a 0 S c1 ).

PAGE 55

55 So there exists difference between the absorption excitation energy and emission photon energy, which leads to the red shifted emission peak compared to the absorption peak, 128 called Frank Condon shift. 2. 6 Inter M olecular E nergy T ransfer Inter Molecular energy transfer between a donor molecule (initial exciton site) and an acceptor molecule (final exciton site) may occur through photon re absorption, Frster energy trans fer or Dexter energy transfer. For the first process, the photon emitted by recombination of the exciton on the donor molecule is re absorbed by the acceptor molecule. This process is a radiative transfer and may occur at long distances, typically more tha n 100 An overlap in the emission spectrum of the donor molecule and the absorption spectrum of the acceptor molecule is required. The latter two processes are non radiative energy transfers. The mechanism s of Frster and Dexter energy transfers are illu strated in Figure 2 1 2 2. 6 1 Frster E nergy Transfer Frster energy transfer 129 also depends upon the spectral overlap, however, no photon is actually emitted. Rather, the dipole dipole interaction between the donor and acceptor molecules induces a resonant transition of the donor molecule to the ground state and the acceptor molecule to the excited state, as shown in Figure 2 12. Fr ster energy transfer happens very fast (< 10 9 s) typically in singlet singlet transitions The distance over which Frster energy transfer occurs may be up to 100 shorter than that for the photon re absorption process due to the strong distance depende nce of the dipole dipole interaction. The Frster energy transfer rate constant is given by: 114,130

PAGE 56

56 w here K is an orientation factor, n is the refractive index of the medium, D is the radiative lifetime of the donor, r is the distance (c m) between donor and acceptor, and J is the spectral overlap (in coherent unit s: cm 6 mol 1 ) between the absorption spectrum of the acceptor and the fluorescence spectrum of the donor. Figure 2 1 2 (color) Schematic illustration of two non radiative i nter molecular energy transfer processes between host and dopant: ( left ) Frster e nergy transfer with long range interaction (~ 30 100 ) ; ( right ) Dexter energy transfer with short range interaction (~ 6 2 0 ) The typical transition types are also indicated. (Courtesy of J. Xue) 2. 6 .2 Dexter E nergy T ransfer Dexter energy transfer involve s bilaterally elect ron exchange between molecules, as shown in Figure 2 12. The exchange mechanism typically occurs within a very short range of ~10 and allows for the singlet singlet and triplet triplet transitions. Unlike the six power distance depende nce of Frster energy transfer, the transfer rate constant of Dexter energy transfer exponentially decays as the distance between the two molecules increases. The transfer rate constant, k ET is given by: 130

PAGE 57

57 w here r is the distance between donor and acceptor, L and P are constants not easily related to experimentally determinable quantities and J is the spectral overlap integral. For this mechanism the spin conservation rule s are obeyed. 2.6.3 Exciton Dissociation An exciton may also dissociate into a free electron and hole when it is subject to an electric field or at heterointerfaces. Due to the large exciton binding energy, an intense electric field of 10 6 V/cm is needed t o cause only a small fraction of Frenkel excitons to dissociate. 131 Alternatively, heterojunctions between organic materials with different electron affinities and ionization potentials, and structural and che mical defects present in these materials, may provide efficient exciton dissociation sites for rapid charge transfer. For example, exciton dissociation at an organic heterojunction between a donor with a low ionization potential and an acceptor with a high electron affinity may be energetically favorable, leading to electrons in the LUMO of the acceptor and holes in the HOMO of the donor. 131,132 Such a donor acceptor heterojunction structure is used in organic photovoltaic cells.

PAGE 58

58 CHAPTER 3 FUNDAMENTALS OF ORGA NIC LIGHT EMITTING DEVICES 3.1 D evice S truct ure of OLEDs OLED architecture can be divided into two groups depending on the light emitting direction, the bottom emitting OLED (BOLED) and the top emitting OLED (TOLED) G eneral device structure s of both types of OLEDs are sh own in Fig ure 3 1 In t he co nventional B OLED structure light i s directed out from the emissive layer (EML) to the air through the semi transparent anode and transparent glass substrate (bottom direction), as opposed to the TOLED, in which the light is directed out through the semi t ransparent cathode to the air (top direction). The light hit the opaque electrodes (that is cathode for BOLED, and anode for TOLED) will be reflected back to the opposite directions. Figure 3 1 (color) Typical structures of OLEDs : ( Left ) the bottom emit ting OLED (BOLED) and ( Right ) the top emitting OLED (TOLED) H ere HTL = hole transporting layer, ETL = electron transporting layer, and EML = emissive layer. Indium tin oxide (ITO) is generally used as an anode with high transmitivity (~90%) to visible lig ht and low sheet resistance (~ patterned on the glass substrate for

PAGE 59

59 typical BOLED ITO has high treatment sensitive work function, which promotes injection of holes into the HOMO level of the successive organic layer. Usually in the OLEDs based on polymer EMLs (PLEDs), a layer o f polythiophene derivative doped with polystyrene sulfonic acid, PEDOT:PSS, is also deposited after ITO to enhance the hole injection. 133 PEDOT:PSS is the p type conducting polymer with HOMO level of 4.8~ 5.2 eV, and has a high lateral conductivity and also transparent over the visible spectrum. The chemical structures of PEDOT:PSS are shown in Figure 3 2. Figure 3 2. Chemical structures of Poly(3,4 ethylenedioxythiophene) (PEDOT) and Poly(styrenesulfonat e) (PSS). A h ole transporting layer (HTL) facilitates hole injection from the anode to the emissive layer (EML) and also functions as an electron blocking layer (EBL) and helps confine excited states to the EML Similarly, the electron transporting layer (ETL) helps better electron injection from the cathode to the EML, typically with a thin electron injection layer (EIL) between the ETL and the cathode. Injected holes and electrons from the anode and cathode, respectively, recombine and create excitons in the EML, and emit the light depending on the band gaps of emitters, via the radiative relaxation process. Highly reflective metals such as silver (Ag) and aluminum (Al) can be typically used as the cathode materials in BOLED due to their high reflectivity and low work

PAGE 60

60 functions (4.1~ 4.2 eV). In this dissertation, the BOLED structure will be mainly employed and discussed. 3.2 F abrication of OLEDs BOLEDs are typically fabricated on glass substrates commercially pre coated with an indium tin oxide (ITO) anod e ( ) A typical way to clean t he substrates are, place the substrates in ultrasonic baths of deionized water, acetone, and isopropanol consecutively sonicating for 15 minutes each, and are then exposed to the ultraviolet ozone amb ient for 15 minutes immediately before the next processing step D epend ing on the molecular weight of the organic molecules which are used for the thin film growth OLED s can be divided into two types; small molecule OLEDs (SMOLEDs) and polymer OLEDs (PLED s) Small molecule is defined by a mole cular weight less than 1000 g/mol whereas polymer has long and complicated structures with the molecular weight approximately more than 1000 g/mol There have been many different methods to grow organic thin films in cluding small molecules thin films deposition methods such as vacuum thermal evaporation (VTE ) 14 organic vapor phase deposition (OVPD) 15,16 organic vapor jet printing (OVJP), 134,135 and organic molecular beam deposition (OMBD), 136,137 and polymer thin films deposition methods such as spin coating, spraying metho d, 138,139 inkjet printing, 17 and even the roll to roll process ing 24 Among those metho ds, for research in the laboratory, VTE is commonly used to fabricate SMOLEDs, whilst the spin coating is most widely used for deposition of PLEDs. Figure 3 3 A illustrates the VTE system in which the main vacuum chamber is pumped to the typical pressure un der 10 6 ~ 10 7 Torr 14,140 T he source boat placed in the bottom of the vacuum chamber is electrically heat ed up and organic materials in the source boat are sublimed in a vacuum environment. The ballistic molecular thin film

PAGE 61

61 deposition is possible with tens of centimeter sc ale mean free path under the ultra high vacuum in the chamber, larger than the chamber size and therefore avoiding the collision of molecules when traveling from source boat to substrate. A quartz crystal monitor (QCM, coated with a thin gold layer) provide s film thickness control with accuracy up to 0.1 /s. The patterning of organic and metal deposition can be realized by using shadow masking. The VTE method provides several advantages of the o rganic thin films deposition for OLEDs fabrication. 140 It has very clean environment with low impurity incorporated, due to the ultra high vacuum during the deposition process. It is also capable of depositing multi layers, mixed or doped layers, and metal layers. But the VTE method is only applied to relatively small area OLEDs, due to the poor uniformity of large area film thick ness, considerable waste of source materials and challenging patterning for large area films. The shadow masks and chamber wall need to be routinely cleaned. Figure 3 3 (color) Schematic illustration of organic thin film growth processes : 140 A ) V acuum thermal evaporation (VTE) for small molecules thin film deposition and B ) S pin coating for polymer thin film deposition.

PAGE 62

62 Figure 3 3 B sc hematically shows the spi n coating process where polymers are dissolved in a specific solvent and dropped onto the substrate which rotates at an angular velocity. 140 Centrifugal forces spread the solution over the surface of substrate, and a thin layer of polymer film will be formed as the solvent evaporates. A thermal annealing process is typically required to completely evaporate the solvent residue and further densify the coating after spin coating. The thickness of a polymer thin film can be accurately controlled by adjusting the spin speed and the polymer concentration in the solution Benefits of spin coating include fast process time and high uniformity of thin films. Similar to the VTE method, spin coating can only be used for relatively small area PLEDs and the waste of solution also cannot be neglected. Spin coating method cannot deal with the metal deposition, so other methods such as VTE method need to be used to finalize the device fabrication. Finally, the encapsulation of device is not required for OLEDs research in the laboratory, but will definitely increase the stability of devices for longer duration. 3.3 P rinciple of OLED s O peration Fig ure 3 4 illustrate the energy band diagram s for the operation processes of the OLED The OLED is simplified as a single organic layer structure sandwiched by the anode and the cathode in each side for electrical contacts Fermi level s ( E F ) of organic layers are not aligned before the electrical contact with anode and cathode as shown in Figure 3 4 A After the electrical contact, there is a thermal equilibrium state with aligned E F However, charge injection into the organi c layer is still not preferable due to the built in potential barrier shown in Fig ure 3 4B When the voltage bias is applied to the electrodes, charge carriers are ready to be injected into the organic layer at V = V bi as shown in Figure 3 4C When applie d voltage

PAGE 63

63 bias, V > V bi charge carriers are then injected into the organic layer, as shown in Figure 3 4 D Holes move upward through the HOMO level and electrons transport downward through the LUMO level of the organic layer, respectively. Excitons are fo rmed once both types of charge carriers meet with each other in the organic layer, and energetically bounded together (Coulombic interaction ). T he radiative recombination of excitons in the organic layer will generate light emission The wavelength of the light emission depends on the band gap of the material in the organic layer. Figure 3 4 (color) Schematic energy band diagrams of OLEDs operation. A ) b efore the electrical contact, B ) a fter the electrical contact but no voltage bias applied C ) w ith vol tage bias applied through the OLED and charge carriers ready to be injected at V = V bi and D ) charge carriers injected in to the organic layer at V > V bi and light emitted by the recombination of excitons

PAGE 64

64 3.4 Electromagnetic Spectrum The electromagnet ic (EM) spectrum is the range of all possible wavelengths (frequencies) of electromagnetic radiation, as shown in Figure 3 5. The EM spectrum extends from low frequencies used for modern radio communication to gamma radiation at the high frequency end. Lig ht is defined as visible radiation of electromagnetic wave s which can be observed by the human eyes, and located in the wavelength ranges between 380 nm and 7 7 0nm The visible spectrum can be further subdivided according to different colors, with red at th e long wavelength end and violet at the short wavelength end, as illustrated in the inset of Figure 3 5. White light is a combination of lights of different wavelengths in the visible spectrum. Figure 3 5. (color) The electromagnetic spectrum as a functi on of wavelength (nm) The visible spectrum is further specified. Next in frequency comes ultraviolet (UV) radiation. The wavelength of UV rays is shorter than the violet end of the visible spectrum but longer than the X ray. UV rays are very energetic due to the short wavelengths. In this dissertation, OLEDs with emission spectra covering from blue, violet in visible spectrum to UV will be studied.

PAGE 65

65 The infrared (IR) part of the EM spectrum comes after the red part of the visible spectrum in wavelengths. I t covers the range from roughly 750 nm to 1 mm, and can be divided into three ranges: far infrared, mid infrared, and near infrared. This dissertation will focus on the OLEDs emitting near infrared radiation with spectrum range from 750 nm to 2500 nm. F igure 3 6 Comparison between P hotometry and R adiometry. T he terminologies and their corresponding SI units for P hotometry and R adiometry are listed. 3.4.1 Photometry vs. R adiometry P hotometry is defined as the science of the measurement of light in visib le spectrum wavelengths ( = 380 7 7 0 nm), whereas R adiometry is the science of measurement of radiant energy in terms of absolute power, and can be applied to the entire spectrum of electromagnetic radiation. 141 T he terminologies and their corresponding units for photometry and radiometry are su mmarized in Figure 3 6 A

PAGE 66

66 primary distinction between Photometry and Radiometry is, radiant flux uses [Watt] as the unit, but luminous flux uses [Lumen] as the unit. In photometric quantities, every wavelength is weighted according to how sensitive the human eye is to it, while radiometric quantities use unweighted absolute power. Since in this dissertation, we will focus on OLEDs with near infrared and ultraviolet emission, therefore, the more general concept Radiometry will be used 3.4.2 Photopic vs. Scotopic Response The illuminance at daytime is about 100,000 lux (lx), as opposed to about 0.0003 lx at nig ht. In order to detect such a wide range of illuminance, the size of the pupil and the responsivity of the retina of the human eye change s according to the brightness of the environment. The human eye has different responses as a function of wavelength wh en it is adapted to light conditions and dark conditions. In a relatively bright environment (> 3 cd/m 2 ) the human eye detect s the light based on the photopic response, while scotopic response is dominant in the dark conditions (< 0.0003 cd/m 2 ) Mesopic r esponse occurs in intermediate lighting conditions and is effectively a combination of scotopic and photopic vision A s sho wn in Figure 3 7, t he photopic response, G( ) has the peak luminous power efficiency of 0 = 683 lm/W at = 555 nm, whereas the sco topic response, G ( ), has the blue shifted peak value of 0 = 1700 lm/W at = 507 nm and insensitive to the wavelengths longer than approximately 640 nm (red) So, the ph otopic and scotopic responses can be expressed as:

PAGE 67

67 w here g( ) and g ( ) are the normalized photopic and scotopic responses, respectively. Figure 3 7 P hotopic and scotopic responses of human eye The vertical axis is plotted as luminous efficacy in lm/W. 3.4.3 Plane vs. Solid Angle s For the application of photometry and radiometry, the concept of solid angle, instead of the widely known plane angle, is usually used. Figure 3 8 P lane and solid angles : A ) a plane angle ( ) in 2 D and B ) a solid angle ( ) in 3 D. Here, r is the radius of the circle or s phere; L is the length of arc in the circle; and A is an arbitary spherical surface area. (Courtesy of S. H. Eom) 142

PAGE 68

68 As shown in Figure 3 8, a plane angle ( ) is defined as a ratio of the length of the arc to the radius of the circle in two dimensional space defined as a ratio of the spherical surface area to the square of the radius of the sphere in three dimensional space The SI units for the plane an gle and solid angle are radian (rad) and steradian (sr), respectively. In spherical coordinates, the solid angle can be expressed as: 3.4.4 Lambertian Emission S ource An isotropic light emitting with equal radiance ( luminance ) into a ny solid angle s is defined as a Lambertian source Figure 3 9 (color) Schematic illustration of a Lambertian emission with equal ra diance into any solid angles. 143 A ) Emission from an area element ( d A) in a normal and angle direction; B ) Observing from normal and angle direction. An important property for the Lambertian source is the cosine law that, 143 the radiance (luminance) is exactly the same when a Lambertian source is viewed from any angle. This is because altho ugh the emitted power from a given area element is reduced by the cosine of the emission angle, the apparent size (solid angle) of the

PAGE 69

69 observed area, as seen by the viewer, is also decreased by a corresponding amount. Therefore, the radiances are the same. This can be further quantitatively demonstrated as illustrated in Figure 3 9. Suppose the radiance along the normal of the circle is I (energy per unit solid angle per projected source area). So, the radiance emitted into the vertical wedge and wedge at angle both with solid angle of d is, I d d A and I cos( ) d d A, respect ively, as shown in Figure 3 9A From the view of observer, the observer directly above the area element ( d A) will see the scene through an aperture of area d A 0 and the area element will subt end a solid angle of d 0 as shown in Figure 3 9B So this normal observer will then record the radiance as: Similarly, the observer at angle to the normal will be seeing the scene through the same aperture ( d A 0 ) and the area eleme nt will subtend a solid angle of d 0 cos( ). Then the observer will record the radiance as: which is the same as what the normal observer has seen. 3.4.5 CIE C olor S pace The color space wa s created in 1931, 144 and can be defined by the tristimulus values of X Y and Z The human eye has photoreceptor (called cone cells) with sensitivity peaks in short ( S 420 440 nm), middle ( M 530 540 nm), and long ( L 560 580 nm) wavelengths. The tristimulus values of a color are the amount of three primary colors needed to match that test color. The

PAGE 70

70 CIE color space has defined a set of three color matching functions, call x ( ), y ( ), and z ( ), as shown in Figure 3 10. The tristimulus values ( X,Y,Z ) for color with a spectral power distribution S( ) are given in terms of the color matching functions by: Figure 3 10 (color) Color matching function curves: x ( ), y ( ), and z ( ). 142 The t ristimulus values of ( X,Y,Z ) can be obtained by integrating the color matching function curves Since the human eye has three types of color sensors tha t respond to different ranges of wavelengths, a full plot of all visible colors is a three dimensional figure. However, the CIE color space was deliberately designed so that the Y parameter was a measure of the brightness of color, while the chromaticity o f a color was then specified by the two derived parameters, x and y as the functions of three tristimulus values X, Y, and Z :

PAGE 71

71 This derived color space specified by x y and Y is known as the CIE xyY color space and leads to the wid ely used two dimensional CIE chromaticity diagram, as shown in Figure 3 11. Figure 3 1 1 (color) CIE chromaticity diagram represented by ( x,y ) coordinates Planckian locus (black body radiation) is also illustrated with lines of constant correlated colo r temperature ( T C ) The diagram represents all of the chromaticities visible to the average person. The horseshoe shaped region shown in colors is called the gamut of human vision. The outer curved boundary is the spectral locus and corresponds to monochro matic light

PAGE 72

72 (each point representing a pure hue of a single wavelength), with wavelengths shown in nanometers. The lower part on the border of the gamut, without counterpart in monochromatic light, is called line of purples. The chromaticities of black bod y light sources of various temperatures and lines of constant correlated color temperature ( T C ) are also shown in Figure 3 11. Note that the chromaticity diagram is a tool to specify how the human eye will experience light with a given spectrum. It cannot specify colors of objects since the chromaticity observed while looking at an object depends on the light source as well. 3.5 Characteristics of O LED s The accurate and consistent measurement of OLED characteristics is important for comparison among differe nt research groups and different devices A measurement method for the OLED has been suggested previously The measurement method, definitions, and calculations in this dissertation are based on the previous suggested report and equipment in one specific r esearch laboratory. Although the available equipment and measurement set up might be somewhat different among different research groups, it can provide a consistent device result comparison by following the similar understanding of fundamental concepts. Ra diant emittance ( R ) or luminance ( L ) current density ( J ) voltage ( V ) characteristics of the OLEDs were measured in ambient using an Agilent 4155C semiconductor parameter analyzer and a calibrated silicon detector ( Newport 818 UV) The luminance was cal ibrated using a Konica Minolta LS 100 luminance meter (with No. 110 close up lens) The Lambertian emission source is assumed for both radiant emittance and luminance calibration Electroluminescent (EL) spectra were taken using the corresponding spectrome ter s for devices with different range emission

PAGE 73

73 The spectrum with absolute power intensity can be obtained by the EL spectra measurement, written by the following expression: (3 1) where I ( ) is the spectrum with absolute power intensity, [W]; S ( ) is the normalized spectrum with amplitude constant of I 0 The semiconductor parameter analyzer will record the photo current by using the sili con photo detector. The photo current, I det can be written as: (3 2) where I det ( ) is the photo current at a single wavelength of [A]. Th en the spectrum with absolute power intensity can be re written as: (3 3) where f is the geometry factor, which is the fraction of photons reaching the photo detector comparing to the total photons emitted by the OLEDs; R ( ) is the responsivity of the photo detector, [A/W]. By replacing the term of I det ( ) in Equation 3 2, the photo current can be then re written as: (3 4) where (3 5) By replacing I 0 in Equation 3 1, I ( ) can be further re written as: (3 6)

PAGE 74

74 Both Equation 3 4 and 3 6 show the relation between the measured spectrum and photo current. Based on the data sets from the R J V or L J V characteristics and EL spectrum measurement all the properties of OLEDs can be derived as follows. 3.5.1 Radiance Radiance ( R v ) is defined as the radiant flux per unit area per unit solid angle (or radiant intensity per unit area). The unit for radiance is [W/sr m 2 ] and it characterizes the total emission power emitted by the sources. Another term, radiant emittance ( R ), does not consider the solid angle in a specified direction emitted from the sources, and the unit is [W/m 2 ]. Based on the definition, the radiant emittance can be expressed as: (3 7) where A is the device area. Substituted the I ( ) with Equation 3 6: (3 8) In this dissertation, the term radiant emittance ( R ) and unit of [W/m 2 ] will be used. 3.5. 2 Luminance L uminance ( L ) is defined as the luminous flux per unit area and per unit solid angle (or lum inous intensity per unit area) and t he unit for luminance is [ cd/m 2 ] (= [ lm/sr m 2 ] or [ nit ] ) Different from the radiance which only considers the total power of emission, lumi nance needs to incorporate the photopic response of human eyes. So the expression of luminance can be written as: (3 9)

PAGE 75

75 where (3 10) Equation 3 9 and 3 10 are also used for the calibration of geometr y factor, f 3.5. 3 Luminance E fficiency Luminance efficiency ( L ) is also called current efficiency and it is defined as a ratio of luminance to the injected current density to the OLED J D as expressed in Equation 3 11. The unit of L is [ cd/A ] (3 11) 3.5.4 Power Efficiency Power Efficiency ( P ) is defined as the ratio of output optical power to the input power driving the OLEDs and the unit of P is [W/W]. Power efficiency can be calculated by the following equation: (3 12) where V is the applied voltage of OLEDs, [V]. 3.5. 5 Luminous P ower E fficiency Luminous power efficiency ( L P ) is defined as the ratio of output luminous flux to the input power driving the OLEDs and the unit of L P is [lm/W]. Similar to the luminance, the photopic response of human eyes needs to be included here and the expression of L P can be written as: (3 13)

PAGE 76

76 Also, luminous power efficiency of the OLED has the simple conversion with the luminance efficiency ( L ) as shown in Equation 3 14: (3 14) 3.5. 6 External Q uantum E fficiency The external quantum efficiency ( EQE ) physically means a ratio of the number of photons emitted from the OLED to the number of charge carriers injected into the OLED and can be expressed as: (3 15) w here q is the electric charge h is Planck s constant and c is speed of light. Substituted I ( ) with Equation 3 6: (3 16) By comparing Equation 3 12 and 3 16, the conversion between power efficiency and external quantum efficiency can be derived as: (3 17) Similarly, the conversion between luminous power efficiency and external quantum effici ency can be derived as: (3 18)

PAGE 77

77 3.6 E fficiency of O LED s There are basically four steps for the OLED operation including: 1) injection of electrons and holes; 2) capture to from excitons; 3) radiative decay; 4) light emission. So the efficiency of OLEDs will be composed of the efficiency of each step, as shown by the following expression of quantum efficiency of OLEDs: (3 19) Here, R is the charge balance factor indicating the ratio of excitons to charge injected. This efficiency can be easily achieved to unity for OLEDs by balancing the charge s inject ed from the anode and cathode, respectively. Balanced charge injection can maximiz e the exciton generation in the middle zone of the EML and also minimiz e the possible non radiative exciton polaron (excited charge carrier) quenching s is the spin statistics factor indicating the ratio of excitons which can radiatively decay. The fluo rescent emission can convert only singlet (25%) exciton s into photon s whereas the phosphorescent emission can convert singlet (25%) as well as triplet (75%) excitons into photon s leading to theoretical ly 100% photon conversion efficiency. PL is the PL efficiency of organic dye and it depends on the property of that material. The PL efficiency of organic dyes can be as high as over 90% in the solution matrix, but the PL efficiency of solid state thin film reduces significantly due to strong intermolecula r interaction s So, higher PL efficiency can ultimately provide higher device efficiency but also the higher electroluminescent quantum yield of the emissive layer in the OLED should be optimized by alternatively choosing appropriate host guest systems an d /or adjusting the do ping concentration of emissive guest These three efficiencies can also be summarized as the internal quantum efficiency of OLED, I QE which is defined as

PAGE 78

78 the ratio of total number of photons generated in the OLED to the total number of electrons injected into the O LED: (3 20) For the phosphorescent organic dye with 100% PL efficiency, the internal quantum efficiency of 100% can be achieved. However, the external quantum efficiency of BOLED is typically limited to ~20% due to the limitation of light outcoupling efficiency, out The light output from emissive layer in OLEDs will be reduced by absorption losses and wave guiding within the device and its substrate due to total internal reflection (TIR ) at multiple interfaces 34,50 The OLED has a multi layered structure usually consisted of a metal reflector, organic layers, ITO, and a glass substrate. Hence, the isotropic dipole s generated in the emissive o rganic layer must go through multiple interfaces in order to finally com e out to the air Figure 3 1 2 (color) T hree optical modes in bottom emitting OLEDs (BOLEDs) : i ) a ir mode (~20%) ; ii ) glass substrate mode (~30%) ; and iii ) l ocalized plasmonic a nd organic/ITO wave guiding mode (~50%) The values in parentheses are the percentage of photons in that mode. (Courtesy of S. H. Eom) 142 There are three optical modes depend ing on the different optical confinement mechanisms as shown in Figure 3 12 The first mode is the organic/ITO wave guiding

PAGE 79

79 modes where almost 50 60% of the generated photons are confined near the metal cathode and organic/ITO layers due to the strong localized electric field on the metal cathode and TIR mainly at the interface between ITO and g las s substrate. 72,73 The second mode is glass substrate m ode, causing a pproximately 20 30% loss of generated photons. In this mode, photons are trapped and lost in the thick glass substrate due to the TIR at the glass/air interface. Therefore, it is believ ed that only ~ 20% of photons in the third air mode can contribute to the out in a planar BOLED system.

PAGE 80

80 CHAPTER 4 NEAR INFRARED ORGANIC LIG HT EMITTING DEVICES BASED ON FLUORESCENT DONOR ACCEPTOR DONOR OLIGOMERS 4 .1 Introduction to the Near IR OLEDs Since the first report of near infrared (NIR) emission from an organic light emitting device (OLED) based on a lanthanide complex, 55 there has been considerable interest in the development of materials and devices that exhibit wavelength tunable and efficient light emission in the NIR wavelength region. Such work is of inter est for a variety of applications in security and defense, telecommunications, and biomedical sectors. 56 63 For example, NIR OLEDs could serve as a new class of illuminators for night vision, providing advantages su ch as light weight, low thermal signature, low power consumption, and compatibility with large area and flexible substrates. Much of the existing work on NIR OLEDs has focused on the use of lanthanide containing organometallic complexes based on Yb, Nd, an d Er. 55,56,64 73 The metal phthalocyanines (Pc) with emissions at the near IR range are also use d for the device demonstration. 74,75 However, because of the inherent low eff iciency of emission from these complexes, the most efficient OLEDs based on these materials possess external quantum efficiencies (EQE) less than 0. 3 %. More recently, Thompson and co workers reported the development of a phosphorescent Pt porphyrin comple x that phosphoresce with high quantum yield at a peak wavelength of 770 nm, and achieved EQE up to 8.5% in the optimized OLEDs. 88,89 However, the emission of this device is still not as deep as into the near IR emission range. So, there still remains the need for the development of alternative materials and devices that exhibit wavelength tunable and highly efficient emission at > 800 nm. Table 4 1 summarizes some of the existing near IR OLEDs in the literature

PAGE 81

81 Table 4 1. List of some existing near IR OLEDs based on different near IR emitters. The maximum radiant emittances ( R ) and external quantum efficiencies ( EQE ) are compared for different devices. Emitters max em (nm) R 2 ) EQE (%) Ref Ln(TPP)TP Ln (TPP)L (OEt) 977 1570 <1 <0.1% 65,67,68 PPyrPyrPV 800 0.06 (17V) Low 145 Ethyne briged Porphyrin 720, 820 0.3%, 0.1% 146 Ir Complexes 720 319 (20V) 0.1 % 76 ClInPc 880 74 PdPc, PtPc 1025, 966 0.03%, 0.3% 75 Pt TPTBP 772 1200 8.5% 88,89 Os Complexes 814 718 65 (17V) 172 (14V) 1.5% 2.7% 147 4 .2 Donor Acceptor Donor Oligomers c onjugated d onor a cceptor d onor (DAD) oligomers and polymers have been extensively investigated for their controlled optoelectronic properties that include line ar and nonlinear optical effects 148,149 When the electron rich donor and electron deficient acceptor species are covalently bonded with one another in the DAD molecule s the ene rgy of the highest occupied molecular orbital ( HOMO ) is controlled by the donor portion, while the lowest unoccupied molecular orbital ( LUMO ) energy is controlled by the acceptor and, as such, the HOMO LUMO gap can be easily controlled by carefully selecti ng the appropriate structures of the donor and acceptor units By combining strong donors with strong acceptors, a set of DAD oligomers and polymers with narrow HOMO LUMO gaps has been achieved. In the area of non linear optics DAD molecules can provide a large two photon cross section suggesting possible applications in three dimensional patterning via photopolymerization and optical limiting. 150 Reynolds and co workers have exploited oxidative polymerization for t he synthesis of narrow gap polymers, as exemplified by our recent report of electrochemically prepared polymer

PAGE 82

82 films deposited on single walled carbon nanotube films exhibiting band gaps as low as 0.5 eV 151 and demonstrated NIR light emission (albeit weak) with solution prepared narrow gap polymers. 145 Considering the spectral prop erties of a set of DAD oligomers prepared in the course of their polymer work we note d that the oligomers have especially long wavelength optical absorption and efficient photoluminescence (PL), making them excellent candidates as the light emitting speci es for NIR OLEDs prepared via vacuum deposition methods 152,153 In this chapter, we report NIR OLEDs based on two DAD conjugated oligomers, 4,8 bis(2,3 dihydrothieno [3,4 b ][1,4]dioxin 5 yl)benzo[1,2 c ;4,5 c ]bis [1 ,2,5]thiadiazole (BEDOT BBT) and 4,9 bis(2,3 dihydrothieno[3,4 b ][1,4]dioxin 5 yl) 6,7 dimethyl [1,2,5]thiadiazolo[3,4 g ]quinoxaline ( BEDOT TQMe 2 ) as illustrated by the structures in Figure 4 1, wh ich have the same donor component but different acceptor ones Using tris(8 hydroxyquinoline) aluminum (Al q 3 ) as the host in the emissive layer (EML), fluorescent OLEDs based on these two NIR emitters were fabricated by doping B E DOT TQMe 2 or BEDOT BBT into the host matrix By varying the doping concentration of the NIR emitter and the thickness of the electron transp orting layer, a maximum EQE of EQE = 1.6% and a maximum power efficiency of P = 7.0 mW/W are achieved in the optimized devices based on BEDOT TQMe 2 in which the electroluminescence (EL) peak s at 692 nm but e xtends to well above 800 nm. BEDOT BBT based OLEDs show even longer wavelength emission wit h a maximum at 815 nm and exten ding to as far as 950 nm due to the stronger acceptor ( BBT) used in this molecule ( which correspond s to a smaller HOMO LUMO gap than that of BEDOT TQMe 2 ). T he maximum efficiencies of the optimized BEDOT BBT devices were reduced to EQ E = 0.51% and P = 2.1 mW/W

PAGE 83

83 due to the significantly lower fluorescent quantum yield of the corresponding NIR emitter. The efficiencies of these optimized fluorescent OLEDs were further increased by two to three times by incorporating a phosphorescent sen sitizer in the emissive layer to funnel the triplet excitons formed on the host molecules to the fluorescent emitters, which are not harvested in fluorescent devices Using this sensitized fluorescence (SF) device architecture 154 we achieved maximum efficiencies of EQE = 3.1% and P = 12 mW/W for BEDOT TQMe 2 based devices, an d EQE = 1.5% and P = 4.0 mW/W for BEDOT BBT based devices. These OLEDs emit with the highest efficiency reported to date for near IR emitting devices that are based on fluorescent chromophores. 146,152,153 The ag gregate effect of the donor acceptor donor oligomers in the devices is finally studied by utilizing different host matrix and changing the stacking manners of doping molecules. Changing from the previous host of Al q 3 to CBP, which has more spectral overlap with the longer emission DAD NIR emitter s the concentration of NIR emitters needed to allow complete energy transfer for excitons on the host molecules will be decreased due to increased F rster radius, leading to less significant aggregate quenching eff ect. The NIR OLEDs based on BEDOT BBT doped into CBP host matrix has a maximum EQE of EQE = ( 0.92 0.07 )%, which is almost double that of the devices with Al q 3 as host. The power efficiency is also higher for the CBP device than the device with Al q 3 host with a maximum of P = ( 3.6 0. 3 ) mW/W for the former device The effect of molecular structure on aggregation in devices is then studied by adding bulky substituted groups to the two ends of DAD oligomers. However, t he NIR OLEDs based on two bulky DAD oligomers show similar aggregate effect, compared to

PAGE 84

84 BBT based devices. The efficiencies of these bulky DAD oligomers based devices show no enhancement and the emissions are red shifted with peak wavelength s up to 905 nm. This chapter is organ ized as follows. The experimental details on material synthesis, device fabrication and characterization are described in section 4 3 In section 4 4 we present the experimental results and discussion on the optical properties of the DAD oligomers and per formance of the fluorescent and sensitized fluorescent devices. We also study the aggregation effect of normal and bulky DAD oligomers in the device in section 4 5 Finally, we conclude in section 4 6 4 .3 Experimental Details The molecular structures of B EDOT TQMe 2 and BEDOT BBT are shown in Fig ure 4 1. The synthesis of these two DAD oligomers follows a similar process to tha t reported previously. 151 Briefly the synthesis started with the bromination of 2,1,3 benzothiadiazole followed by nitration. A Stille coupling between the nitrated product and trimethyltin ethylenedioxythiophene yielded a three ring precursor which w as reduced to the diamine with iron in acetic acid A final ring closure was done with N thionylaniline in pyridine or 2,3 butanedione in acetic acid to yield BEDOT BBT and BEDOT TQMe 2 respectively. The synthesis of the bulky DAD oligomers follows the sam e way. The materials were synthesized by Timothy Steckler and Dan Patel in Dr. John Reynolds group in the Department of Chemistry at the University of Florida. UV Vis absorption spectra were measured in CH 2 Cl 2 solutions using a PerkinElmer Lambda 25 UV vi s spectrometer. The PL spectra were obtained by excitation at the absorption maxima and recorded with an ISA SPEX Triax 180 spectrograph coupled to a Spectrum 1 liquid nitrogen cooled charge coupled device

PAGE 85

85 ( CCD ) detector. This part of work was done by Rich ard Farley in Dr. Kirk Schanze group in the Department of Chemistry at the University of Florida. Figure 4 1. Molecular structures and photoluminescence spectra of two near IR emitting donor acceptor donor oligomers, BEDOT TQMe 2 and BEDOT BBT in CH 2 Cl 2 solutions. (courtesy of Richard Farley) NIR OLED devices were fabricated via vacuum thermal evaporation on glass substrates commercially pre coated with an indium tin oxide (ITO) anode ( sheet ) The substrates were cleaned in ultrason ic baths of deionized water, acetone, and isopropanol consecutively for 15 minutes each, and were then exposed to an ultraviolet ozone ambient for 15 minutes immediately before loading into a high vacuum chamber (base pressure ~ 10 7 Torr). To fabricate a fluorescent OLED, a 40 nm thick hole transporting layer (HTL) of bis[N (1 naphthyl) N phenyl amino] biphenyl ( NPD) a 20 nm thick emissive layer (EML) consisting of an Al q 3 host doped with either NIR emitter, an electron transport layer (ETL) of bathocuproine (BCP) and a 1 nm thick LiF layer followed an Al cathode layer (50 nm thick) were successively

PAGE 86

86 deposited on the substrates. 155 The doping concentration of the NIR emitters in the EML was varied from 0% to 5% by weight, whereas the thickness of the BCP ETL was varied from 40 nm to 110 nm. For the sensitized fluorescent 154 d evices, 4,4 bis(carbazol 9 yl)biphenyl (CBP) was used as the host material in the EML, which also incorporates a phosphorescent sensitizer in addition to the NIR emitters. t ris(2 p henylpyridine) iridium (III) (Ir(ppy) 3 ) 154,156 and bis(2 phenylquinoline)(acetylacetonate)iridium(III) (PQIr) 49,157 were used as the phosphorescent sensitizers, both with 10 wt.% doping concentration, for BEDOT TQMe 2 and BEDOT BBT based dev ices, respectively. The deposition rates for the organic layers were 0.1 0.2 nm/s, as monitored by quartz crystal microbalances. The active device area was 4 mm 2 with the electrodes arranged in a cross bar geometry, and each substrate featured four indepen dently addressable device pixels. Ir(ppy) 3 and PQIr were purchased from Luminescence Technology Corp., Al q 3 and BCP from TCI America, NPD from e Ray, and CBP from Springchem & Jadetextile Group L td. All organic materials were used as obtained except for NPD and Al q 3 which were purified in house using vacuum gradient sublimation method for one to two cycles. 14 Radiant emittance ( R ) current density ( J ) voltage ( V ) characteristics of the NIR OLEDs were measured in ambient using an Agil ent 4155C semiconductor parameter analyzer and a calibrated silicon detector ( Newport 818 UV) EL spectra were taken with the ISA SPEX Triax 180 spectrograph with the device s driven at a constant current using a Keithley 2400 source meter The radiant emit tance was calibrated assuming Lambertian emission and t he EQE ( EQE wall plug power efficienc y ( P ) were

PAGE 87

87 derived based on the recommended methods. 158 The organic layer thickness non uniformity across the samples and run to run variations typica lly lead to 5 10% error in device efficiencies. 4 .4 Near I nfrared OLEDs based on DAD oligomers 4 .4. 1. Optical P roperties of the DAD O ligomers The absorption spectra of the two DAD molecules in CH 2 Cl 2 solutions are shown in Fig ure 4 2. BEDOT TQMe 2 features a strong absorption peak at 370 nm and also a longer wavelength absorption peak at = 531 nm The absorption of BEDOT BBT in the 350 400 nm range is similar to that of BEDOT TQMe 2 ; however, the longer wavelength absorption band is red shifted by 120 nm, to = 650 nm due to the stronger electron acceptor component in BEDOT BBT, which l eads to a lower lying LUMO level and thus a lower HOMO LUMO energy band gap. The PL spectrum of Al q 3 host is also shown in Figure 4 2 for comparison. Figure 4 2 (color) Absorption spectra of BEDOT TQMe 2 (red) and BEDOT BBT (blue) in CH 2 Cl 2 solutions. The photoluminescence spectrum of Al q 3 is also shown by green dashed line.

PAGE 88

88 Accordingly, compared to the PL band of BEDOT TQMe 2 which has a maximum wavelength at = 698 nm, the PL of BEDOT BBT is also red shifted by more than 100 nm, to a maximum wavelength at = 805 nm, as shown in Figure 4 1 The PL quantum yield for BEDOT TQMe 2 is measured to be 21%, whereas that for BEDOT BBT is three times lower, only 7.6%. The lower quantum yield of the latter molecule is at least in part due to the increased non radiative decay rate in this lower band gap molecule. From the absorption and PL spectra, the HOMO LUMO gaps of BEDOT TQMe 2 and BEDOT BBT are determined to be appro ximately 2.0 and 1.6 eV, respectively. 4 .4. 2. Fluorescent NIR OLEDs based on the DAD O ligomers The electroluminescence spectra of OLEDs with various doping concentrations of BEDOT TQMe 2 in the Al q 3 host matrix and t he schematic energy level diagram of thes e devices are shown in Fig ure 4 3 A The electrode work functions and the HOMO/LUMO energ NPD Al q 3 and BCP are taken from the literature 159,160 whereas the HOMO/LUMO energies for the NIR emitters are est imated from a combination of solution cyclic voltammetry, differential pulse voltammetry, and spectroscopic measurements. The BCP thickness in these devices is maintained at 50 nm. In Figure 4 3 A t he undoped device (0% doping) shows an emission peak at = 518 nm, which is solely from the Al q 3 host molecules. With BEDOT TQMe 2 doped into Al q 3 the Al q 3 host emission is quenched and emission from the dopant molecules is observed. The peak emission wavelengths of the doped devices are at = 690 698 nm, sligh tly blue shifted compared to the PL peak of BEDOT TQMe 2 in solution but red shifted significantly from the emission of the Al q 3 host Efficient F rster energy transfer 114 occurs from Al q 3 to BEDOT TQMe 2 molecules, as the emission spectrum of Al q 3 almost completely overlaps with the absorption spectrum of BEDOT TQMe 2 as

PAGE 89

89 Figure 4 3. (color) A) Normalized el ectroluminescence (EL) spectra of OLEDs with various BEDOT TQMe 2 doping concentration in the emissive layer: 0% (green solid lines), 2% (red dash dotted lines) and 5% (blue dashed lines) The BCP layer thickness is 50 nm. Inset: T he s chematic energy lev el diagram of the OLEDs. The energies (in eV) are measured from the vacuum level. The upper dashed line and the dash dotted line in the Al q 3 layer correspond to the LUMOs for BEDOT TQMe 2 and BEDOT BBT, respectively, whereas the lower dashed lines correspon d to the HOMO of both molecules. B ) External quantum efficiency, EQE and power efficiency, P as functions of current density, J for these OLEDs shown in Fig ure 4 2. The intensity of the original Al q 3 emission is less than 1% of that of the NIR peak for the 2% doped device. The Al q 3 emission completely disappears when the BEDOT TQMe 2 doping concentration is increased to 5%; however, this leads to a considerable decrease of the EQE, due to aggregate quenching, 161 from a maximum of

PAGE 90

90 0.91% for the 2% doped device to 0.74% for the 5 % doped device, as shown in Fig ure 4 3 B Similarly, the power efficiency is also decreased from a maximum of 5.4 mW/W for the 2% doped device to 3.9 mW/W for the 5% doped device. Therefore in subsequent work, we employed a 3.5% BEDOT TQMe 2 doping concentration to achieve a proper balance between the maximum energy transfer and minimum concentration quenching. Figure 4 4. (color) A ) External quantum e fficiency, EQE of OLEDs with various BCP ETL thickness ( with 3.5% BEDOT TQMe 2 doping concentration in the emissive layer), as functions of current density, J B ) Maximum EQE (solid) and EQE at J = 1 mA/cm 2 (open) of these OLEDs as functions of the BC P layer thickness, t BCP With a constant 3.5% BEDOT TQMe 2 doping concentration in Al q 3 the thickness of the BCP ETL, t BCP was varied from 40 nm to 100 nm. Figure 4 4A shows t he EQE s of these devices change as the varied BCP ETL thicknes. The maximum EQE as well as the efficiencies at J = 1 mA/cm 2 are summarized in Figure 4 4B Both efficiency curves show the same trend, i e the efficiencies increase with t BCP for 40 nm < t BCP < 80 nm, and then decrease with further increase in t BCP to 100 nm. With an 80 nm thick BCP ETL, a maximum EQE of EQE = (1.6 0.2)% is achieved.

PAGE 91

91 The optimal ETL thickness here is much greater than that usually used in OLEDs emitting visible light. This can be attributed to the weak microcavity effect 52,131 existing in the OLED device structure, as the emission zone needs to be located approximately at a distance of / 4 n away from the reflecting electrode (cathode) to maximize the coupling of the emission into the external mode s Here, is the emission wavelength, and n ( 1.5 2.0) is the refractive index of the organic materials. Therefore, the longer wavelengths of NIR emission, compared to visible emission, require a thicker ETL to achieve the optimized performance The thicker ETL, howev er, does lead to higher operating voltages for the OLEDs O verall, when t BCP is increased from 40 nm to 80 nm, both the maximum EQE and EQE at J = 1 mA/cm 2 are increased by a factor of 2.5 whereas the maximum P is only increased by 80% and the enhancement in P at J = 1 mA/cm 2 is even smaller, only 30%, due to the significant increase in operating voltage. Figure 4 5 (color ) External quantum efficiency, EQE as a function of current density, J for OLEDs ( with 3.5% BEDOT BBT doping concentration in the emissive layer) with different BCP ETL layer thickness: 50 nm ( solid green squares), 80 nm ( open black circles), 100 nm ( ope n red triangles), and 110 nm ( solid blue diamonds).

PAGE 92

92 We have also studied the effect of doping concentration and ETL thickness on the efficiencies of BEDOT BBT based NIR OLEDs The optimal do ping concentration of BEDOT BBT is found to be 3.5%, the same as B EDOT TQMe 2 Figure 4 5 shows the EQE for BEDOT BBT based NIR OLEDs with different BCP layer thickness, which was varied from 50 nm to 110 nm. By comparing the EQE of devices with different BCP thickness, the optimal BCP thickness is found to be 100 nm, sli ghtly thicker than that in the optimal BEDOT TQMe 2 based devices, consistent with the longer wavelength emission of the BEDOT BBT (Fig ure 4 1 ) according to the microcavity effect. Figure 4 6 A ) Normalized electroluminescence (EL) spectra for optimal fl uorescent (NF) OLEDs based on BEDOT TQMe 2 ( dashed line) and BEDOT BBT ( solid line). B) Current density J and the radiant emittance in the forward viewing direction R as functions of the voltage V for these two optimal OLEDs.

PAGE 93

93 As shown in Fig ure 4 6 A with 3.5% BEDOT BBT doped into Al q 3 the host emission is barely apparent with an intensity of approximately 3% of that of the NIR peak at 815 nm, compared to the nearly complete quenching of host emission in the BEDOT TQMe 2 based device with the same doping co ncentration. I ncreasing the BEDOT BBT doping concentration to over 5% could completely remove the Al q 3 emission as well; however, this also leads to a decrease in the EQE by more than 25%, again due to aggregate quenching. 161 The higher doping concentration of BEDOT BBT needed to completely quench the host emission, compared to BEDOT TQMe 2 is consistent with the extent of overlap between the emission spectrum of the Al q 3 host ( F ig ure 4 2 ) and the absorption spectra of the tw o NIR emitters. As shown in Fig ure 4 2, BEDOT TQMe 2 has an absorption peak at = 531 nm which overlaps with the emission from Al q 3 very significantly. However, for the BEDOT BBT molecule, the absorption peak is red shifted to = 650 nm leading to a much less overlap with the Al q 3 emission spectrum. As the F rster energy transfer radius from the host to guest is determined by such spectral overlap, 114 this means that the F rster radius is smaller for energy transfer from Al q 3 to BEDOT BBT compared to that for Al q 3 to BEDOT TQMe 2 and therefore a higher concentr ation of BEDOT BBT is needed to allow complete energy transfer for excitons on Al q 3 host molecules. Based on the methods previously reported, 129,162,163 the F rster radii for energy transfer from Al q 3 to BEDOT TQMe 2 and BEDOT BBT are 28 and 22 respectively, wh ich are consistent with what we expect. Another host material with more spectral overlap could then be used to replace Al q 3 to improve the F rster energy transfer in the BEDOT BBT device s Alternatively, cas caded energy transfer 164 from the host to an intermediate molecule and then to the

PAGE 94

94 NIR emitter could also be used. This will be important when NIR emitters with even longer wavelength emissions are to be used. The R J V characteristics for the BEDOT TQMe 2 and BEDOT BBT based fluorescent NIR OLEDs with 3.5 wt% doping concentrations in the EML and the optimal ETL thicknesses are shown in Fig ure 4 6B The BEDOT BBT based device has slightly higher turn on voltage and lower current density than the BEDOT TQMe 2 based device, due to the slightly thicker BCP ETL of in the former device (100 nm vs. 80 nm). Maximum radiant emittances of R = 6 and 0.5 mW/cm 2 are achieved for BEDOT TQMe 2 and BEDOT BBT based fluorescent devices, respectively, at V = 20 V. Figure 4 7 shows the dependenc i e s of EQE and P on the current density for BEDOT TQMe 2 and BEDOT BBT based fluorescent OLEDs (labeled as N F ) with 3.5 wt% doping concentrations in the EML and the optimal ETL thicknesses The BEDOT TQMe 2 based device has a maximum EQE of EQE = (1.6 0.2)% achieved at J 1 0 2 mA/cm 2 slowly decreasing to EQE = 0.7% at J = 400 mA/cm 2 The EQE for the BEDOT BBT based device is lower, with a maximum of EQE = (0.51 0.05)% achieved at 10 3 mA/cm 2 < J < 10 2 mA/cm 2 although its roll off at higher current densities is less si gnificant and it remains above 0.35%, or 70% of the maximum value, for J up to 100 mA/cm 2 The maximum EQE of these two devices are approximately proportional to the corresponding fluorescent quantum yields of the DAD oligomers The power efficiency is als o higher for the BEDOT TQMe 2 based device than the BEDOT BBT based device, with a maximum of P = (7.0 0.7) mW/W for the former device and (2.1 0.2) mW/W for the latter, both achieved at low current densities ( J < 10 3 mA/cm 2 ). The lower power efficiency in the latter device is mainly due to the lower

PAGE 95

95 EQE (0.5% vs. 1.6%), but is also affected by the slightly higher operating voltage (thicker BCP ETL) and the longer emission wavelength (thus lower photon energy) for BEDOT BBT compared with BEDOT TQMe 2 (Figure 4 1 ). Figure 4 7 (color) External quantum efficiency, EQE and power efficiency, P as functions of current density, J for fluorescent (NF) and sensitized fluorescent (SF) OLEDs based on: A) BEDOT TQMe 2 and B) BEDOT BBT 4 .4. 3 Sensitized F luorescent NIR OLEDs based on the DAD O ligomers The energy transfer mechanism of the above fluo rescent NIR OLEDs can be summarized by the schematics shown in Fig ure 4 8 A Both singlet (S) and triplet (T) excitons are formed in the host molecules with a ratio of 1:3 according to spin

PAGE 96

96 statistics, 165 But only the singlet excitons in the host molecules are transferred to the NIR emitting molecules through the F rster energy transfer process, from which fluorescent emissions are achieved. However, the triplet excitons formed on the host molecules, which are three times more in number than the singlet excitons, are non radiatively recombined and not harvested at all, due to the requirement of sp in symmetry conservation for these fluorescent molecules. Figure 4 8 (color) Energy transfer mechanisms in : A) normal fluorescence ( N F) and B) sensitized fluorescence (SF) systems. F rster energy transfers are represented by solid black lines and Dext er energy transfers by dashed black lines. Intersystem crossing (ISC) is represented by solid green line. Dashed red lines indicate that certain processes have no contribution to the device emission and result in a loss in efficiency. To harvest both singl et and triplet excitons in OLEDs with fluorescent emitters, the so called sensitized fluorescence device architecture has been demonstrated, in which a phosphorescent dye is incorporated in the emissive layer together with the host and fluorescent molecu les to funnel the triplet excitons formed on the host molecules to the fluorescent emitters, or sensitize the fluorescent molecules. 154 Theoretically, all the excitons now can be converted to singlet excitons on fluorescent dyes. This mechanism of sensitized fluorescence is s hown in Fig ure 4 8 B So ideally, by applying the

PAGE 97

97 se nsitized fluorescence to the devices, the efficiencies can be increased by three times compared to the normal fluorescent devices. But in fact, the real enhancement factors will usually be lower due to several loss mechanisms involved in the sensitized flu orescence system, 154 shown by the dashed red lines in Fig ure 4 8 B There i s a probable source of loss of direct transfer from the triplet states of host molecules into the triplet states of the fluorescent dye by a Dexter transfer, which is very similar to what happens in normal fluorescence system. Furthermore, singlet excition s in the phosphorescent sensitizer are subject to intersystem crossing (ISC) and transfer to the triplet states, from which triplet excitons may Dexter energy transfer to the triplet states of fluorescent dye. This is another loss mechanism. Finally, direc t formation of triplets on the fluorescent dye is an additional path to loss (not shown in the figure). Overall, all these loss paths will result in a less enhancement than that expected by ideal situation in the sensitized fluorescent devices. Here, we u se two molecules, Ir(ppy) 3 and PQIr, 49,154,156,157 as the phosphorescent sensitizers for BEDOT TQMe 2 and BEDOT BBT, respectively, and CBP is used as the host instead of Al q 3 The doping concentration of both phospho rescent sensitizers is 10% by weight. BEDOT TQMe 2 doping concentration is slight ly decreased from 3.5% to 3%, whereas that of BEDOT BBT is maintained at 3.5%. The BCP ETL thickness es are the same as in the optimized fluorescent devices (80 nm for BEDOT TQM e 2 and 100 nm for BEDOT BBT). The J V characteristics of these two sensitized fluorescent NIR OLEDs are shown in Fig ure 4 9 Compared to the fluorescent devices exhibit ing turn on voltages of 2.7 V, the two sensitized fluorescent devices have slightly hig her turn on voltages of 3.4 V,

PAGE 98

98 probably due to the charge trapping behavior of the phosphorescent sensitizers. 156 At high voltages ( V > 5 V for BEDOT TQMe 2 and V > 11 V for BEDOT BBT), the sensitized fluoresce nt devices show higher current densities than the corresponding fluorescent devices. This could be attributed to the higher electron mobility of the CBP host in the sensitized fluorescent devices than that of the Al q 3 host in the fluorescent devices. 166,167 The radiant emittances in the forward viewing direction of the d evices are also shown in Fig ure 4 9 The devices with sensitized fluorescence, as expected, have higher radiant emittances than their corresponding f luorescent devices. The maximum radiant emittances for BEDOT TQMe 2 and BEDOT BBT based sensitized fluorescent devices are R = 19 and 2 mW/cm 2 respectively, achieved at V = 20 V, which are more than three times higher than those of the corresponding fluore scent devices. Figure 4 9 Current density, J and the radiant emittance in the forward viewing directions, R as functions of the voltage, V for sensitized fluorescent (SF) OLEDs based on BEDOT TQMe 2 and BEDOT BBT The EQE and power efficiencies of the sensitized fluorescent devices (labeled as SF ) as compared to those of the optimized fluor escent devices are shown in Fig ure 4 7 The EQE of the BEDOT TQMe 2 based sensitized fluorescent device reaches a

PAGE 99

99 maximum of EQE = ( 3 1 0. 3 )% at J 10 2 mA/cm 2 and slowly decreas es with increasing current density, but still achieves EQE = 1 6 % at J 1 00 mA/cm 2 For the BEDOT BBT based sensitized fluorescent device, the maximum EQE is EQE = ( 1 5 0. 2 )% also achieved at J 10 2 mA/cm 2 and its slow roll off at higher current densities is similar to the corresponding fluorescent device. Hence, using the sensitized fluorescent device structure, we have achieved a two and three times higher EQE for the BEDOT TQMe 2 and BEDOT BBT based NIR OLEDs, respectively. These enhancement, which are lower than the theoretical value in the ideal situation result from several possible loss mechanisms involved in the sensitized fluorescence system, as we discussed above ( Fig ure 4 8 ). The m aximum power efficiency is P = ( 12 2 ) mW/W for the BEDOT TQMe 2 based sensitized fluorescent device, and is ( 4.0 0. 4 ) mW/W for the BEDOT BBT based device, both approximately twice of that achieved in the normal fluorescent devices. Figure 4 1 0 (col or) Normalized electroluminescence (EL) spectra for sensitized fluorescent (SF) OLEDs based on BEDOT TQMe 2 and BEDOT BBT. The EL spectra for the two fluorescent (NF) OLEDs are also shown for comparison.

PAGE 100

100 Figure 4 1 0 shows the EL spectra of the sensitized f luorescent NIR OLEDs based on BEDOT TQMe 2 and BEDOT BBT. For comparison, the EL spectra of the two fluorescent devices are also shown. The peak emission wavelength for BEDOT TQMe 2 based sensitized fluorescent devices is at = 692 nm whereas that for BEDOT BBT based sensitized fluorescent devices is at = 815 nm both of which are nearly identical to the corresponding fluorescent devices However, unlike the fluorescent devices in which the host emission is almost complete ly quenched, the two sensitized fluorescent devices still show appreciable emissions (10 15%) from the phosphorescent sensitizers, with the emission peak at around 510 nm from Ir(ppy) 3 and that at 583 nm from PQIr 49 ,154,156,157 This suggests an incomplete energy transfer of the triplet excitons from the phosphorescent sensitizers to the NIR fluorescent emitters, similar to that has been observed previously. 154 Figure 4 1 1 (color) A ) Normalized electroluminescence spectra for BEDOT TQMe 2 based sensitized fluorescence OLEDs with differen t doping concentration of near IR fluorescent emitter B ) External quantum efficiency, EQE as a function of the current density, J for these devices. More complete energy transfer could be induced with h igher concentrations of the fluorescent emitter i n the emissive layer. As shown in the Fig ure 4 1 1 A, as the BEDOT

PAGE 101

101 TQMe 2 concentration increases from 3% to 5%, the intensity of the Ir(ppy) 3 emission is nearly completely suppressed. The EQE of this SF device, however, is also reduced as shown in Figure 4 1 1 B. While a maximum EQE of EQE = 3 1 % is achieved for a 3% BEDOT TQMe 2 doping concentration, a maximum EQE of only 1.6% is obtained for the device with a 5% BEDOT TQMe 2 doping concentration, which again is attributed to aggregation induced fluorescence quenching. 161 Figure 4 1 2 (color) A ) Normalized electroluminescence spectra for BEDOT BBT based sensitized fluorescence OLEDs with different doping concentration of the phos phorescent sensitizer, PQIr. B ) External quantum efficiency, EQE as a function of the current density, J for these devices. The concentration of the phosphorescent sensitizer also affects the energy transfer, although the effect is less prominent. As shown in Figure 4 12 A when th e PQIr concentration in the BEDOT BBT based SF device is increased from 10% to 15%, the PQIr emission is slightly reduced. Accordingly, the maximum EQE of the device is also slightly reduced from EQE = 1.5% to 1.4%, as shown in Figure 4 12 B 4 .5 Aggregat ion of Donor Acceptor Donor Moleclues Previously, we have studied the NIR OLEDs with different doping concentration based on BEDOT TQMe 2 and BEDOT BBT (Fig ure 4 3). It is found that when low

PAGE 102

102 doping concentration (2%) is used in the emissive layer with Al q 3 host molecules, the Al q 3 emission cannot be completely quenched for both BEDOT TQMe 2 and BEDOT BBT based devices, due to the incomplete energy transfer between the host and guest molecules. Further increasing the doping concentration will help solv e this problem. As shown in Fig ure 4 3, the Al q 3 emission completely disappears when the BEDOT TQMe 2 doping concentration is increased to 5%; however, this leads to a considerable decrease of the EQE and power efficiency, due to aggregate quenching. 161 Figure 4 13. (color) The spectral overlaps between absorption spectra of BEDOT BBT (blue solid) in CH 2 Cl 2 solutions and emission spectrum of Al q 3 (green dashed) and CBP (red dashed) For BEDOT BBT, the doping concentration needs to be increase to over 5% to completely quench the emission from Al q 3 because the absorption of BEDOT BBT has a much less overlap with the Al q 3 emission spectrum compared to BEDOT TQMe 2 (Figure 4 2) As the F rster energy transfer radius from the host to guest is determined by such spectral overlap, 114 this mean s that the F rster radius is smaller for energy transfer from Al q 3 to BEDOT BBT compared to that for Al q 3 to BEDOT TQMe 2 and therefore a higher concentration of BEDOT BBT is needed to allow complete energy

PAGE 103

103 transfer for excitons on the Al q 3 host molecules. So in that case, the aggregate quenching issue will be more significant for BEDOT BBT. In previous work, we employed a 3.5% doping concentration for both BEDOT TQMe 2 and BEDOT BBT based near IR OLEDs to achieve a compromise between the maximum energy tr ansfer and minimum aggregate quenching, although the near IR OLEDs based on both molecules still show small amount intensity of Al q 3 emission. Figure 4 1 4 Molecular structures of two near infrared emitting bulky donor acceptor donor oligomers. The mole cular structure of BEDOT BBT is also shown for comparison. But, as the NIR emitters with even longer emission wavelength were used, methods have to be found out to better address these issues. One way is to use another host material to replace Al q 3 to incr ease the spectral overlap with the NIR emitters, by which the concentration of NIR emitters needed to allow complete energy transfer for excitons on the host molecules will be decreased due to increased F rster radius. Then the aggregate quenching issue wi ll be less significant. Another way is to

PAGE 104

104 change the stacking manner of emitter molecules and thus manipulate the inter molecular interaction and aggregation. Here, we replace the host material of Al q 3 with CBP, which has more spectral overlap 168 with the longer emission near IR emitter, BEDOT BBT, as shown in Figure 4 13. Instead of very limited overlapping with the absorption peak at = 6 50 nm for Al q 3 host, CBP host will have more spectral overlap with the shorter absorption peak of BEDOT BBT at = 350 nm. Furthermore, t o decrease the aggregation between molecules, we add two different types of bulky end groups to the two sides of t he DAD oligomers, called bulky DAD oligomers. The molecular structures of these two bulky DAD oligomers, Bu 3 Si Th BBT and EtHx Th BBT, are shown in Fig ure 4 1 4 Compared to the structure of BEDOT BBT which is also shown in Fig ure 4 1 4 the two types of bu lky end groups in the bulky DAD oligomers are designed to reduce the aggregation between molecules. Figure 4 1 5A shows the EL spectra comparison of the NIR OLEDs based on BEDOT BBT with different host materials. The doping concentration of BEDOT BBT is 3.5 % and BCP thickness is 100 nm for both devices. The host emission, which is supposed to be peaked at 420 nm, is nearly completely quenched in the devices using CBP as host, while the devices with Al q 3 as host still show approximately 3% of the intensity of the NIR peak at 815 nm. This is mainly due to the larger spectral overlap of CBP with BEDOT BBT than Al q 3 which leads to more complete F rster energy transfer between host and guest. The two devices show similar turn on voltages but the devices with CBP show higher current densities than the devices with Al q 3 as shown in Fig ure 4 1 5B This could be attributed to the higher electron mobility of the CBP host than that of

PAGE 105

105 Figure 4 1 5 (color) A ) Normalized electroluminescence (EL) spectra of NIR OLEDs bas ed on BEDOT BBT using CBP (black dashed lines) and Al q 3 (red solid lines) as host material (with 3.5%. doping concentration of BEDOT BBT). B ) Current density, J and radiant emittance in the forward viewing directions, R as functions of voltage, V fo r the two OLEDs with different hosts : CBP (black squares) and Al q 3 ( red triangles). C ) External quantum efficiency, EQE and power efficiency, P as functions of current density, J for the two OLEDs with different hosts : CBP (black squares) and Al q 3 (re d triangles).

PAGE 106

106 the Al q 3 host. 166,167 The radiant emittances in the forward viewing direction of these two devices are also compared in Fig ure 4 1 5B The devices with CBP host, as expected, have higher radiant emittan ces than the devices with Al q 3 host. The maximum radiant emittance for the devices with CBP host is R = 5.5 mW/cm 2 achieved at V = 15 V, which is more than ten times higher than that of the devices with Al q 3 host. Figure 4 1 5 C shows the dependenc i e s of EQE and P on the current density for these two NIR OLEDs The device with CBP host has a maximum EQE of EQE = ( 0.92 0.07 )% achieved at J 10 1 mA/cm 2 which is almost double that of the devices with Al q 3 as host. The power efficiency is also higher f or the CBP device than the device with Al q 3 host with a maximum of P = ( 3.6 0. 3 ) mW/W for the former device, achieved at current densities of J 10 2 mA/cm 2 Figure 4 1 6 (color) External quantum efficiency, EQE of OLEDs based on three DAD oligomer s (BEDOT BBT: black squares, Bu 3 Si Th BBT: red triangles, EtHx Th BBT: blue circles) as a function of doping concentration, x ( x = 2%, 4%, 7% and 10% ), of these NIR e mitters. S olid and open symbols are for maximum EQE and EQE at a current density of 10 mA/ cm 2 respectively.

PAGE 107

107 We apply the same device structure which uses CBP as host material to the NIR OLEDs based on Bu 3 Si Th BBT and EtHx Th BBT. To study the aggregation of the DAD oligomers in the devices, the doping concentration of the DAD oligomers in the OLEDs was varied from 2% to 10%. The maximum EQE as well as the efficiency at J = 10 mA/cm 2 are summarized in Fig ure 4 1 6 The corresponding device results of NIR OLEDs based on BEDOT BBT are also shown for comparison. All six efficiency curves show the same trend that the EQE of these OLEDs gradually decreases with the increasing doping concentration of DAD oligomers, which can be attributed to the aggregate quenching as we have mentioned above. The decrease rate, which can be indicated by the slope of t he efficiency curve, represents the extent of aggregation of the DAD oligomers in the devices. The NIR OLEDs based on EtHx Th BBT have similar efficiency decrease rate as the BEDOT BBT based OLEDs, while the decrease rate for the Bu 3 Si Th BBT based OLEDs i s slightly higher. This means Bu 3 Si Th BBT and EtHx Th BBT have similar aggregation in the OLEDs as the BEDOT BBT based OLEDs have. This result is surprise to us as the bulky groups in the two sides of the DAD oligomers actually did not reduce the aggregat ion between molecules as we have expected. The EQE of the NIR OLEDs based on EtHx Th BBT is much lower than that of the BEDOT BBT based OLEDs at all doping concentrations, while the Bu 3 Si Th BBT based OLEDs shows close EQE to the BEDOT BBT based OLEDs at 4 % and 7% doping concentration, and 30% lower EQE when doping concentration increases to 10%. The EL emission of NIR OLEDs based on Bu 3 Si Th BBT with 4% doping concentration is moved to longer wavelength with peak at 865 nm as shown in Fig ure 4

PAGE 108

108 1 7 while th e efficiency still keeps the same, compare to the BEDOT BBT devices. NIR OLEDs based on EtHx Th BBT show even longer emission peaked at 905 nm, although the EQE is much lower than that of BEDOT BBT based OLEDs (Fig ure 4 1 6 ). The OLEDs based on Bu 3 Si Th BBT and EtHx Th BBT both show small amount intensity of host emission at wavelength of 420 nm, due to incomplete energy transfer. Different from BEDOT BBT based device which only show one single peak in the near IR range, the OLEDs based on these two bulky DA D oligomers also exhibit a similar secondary peak at = 730 740 nm with an intensity of ap proximately 25% of that of the near IR peaks. This can be probably attributed to the decomposition of the two bulky DAD oligomers that the chemical bonds connecting the bulky end groups and DAD macrocycle rings break down. Figure 4 1 7 (color) Normalized electroluminescence (EL) spectra of NIR OLEDs based on three DAD oligomers: BEDOT BBT (black dashed lines), Bu 3 Si Th BBT (red solid lines), EtHx Th BBT (blue dotted lines). The doping concentration is 4% for all devices with CBP as host in the emissive layer. 4 .6 Summary In this chapter, we demonstrate here that low gap DAD oligomers are good candidate material s for use in OLEDs to achieve efficient NIR emission. A m aximum

PAGE 109

109 external quantum efficiency of 1.6% and a maximum power efficiency of 7.0 mW/W were achieved in devices based on BEDOT TQMe 2 whose emission peak s at 692 nm, but extend s to well above 800 nm. With a stronger acceptor and thus a reduced energy gap, l onger wavelength NIR emissions peaked at 815 nm were achieved in BEDOT BBT based devices, although the efficiencies were approximately three times lower than the BEDOT TQMe 2 based devices, due to the significant ly lower fluorescent quantum yield of BEDOT B BT. Using the sensitized fluorescence device structure, the efficiencies were further increased by two to three times, and we achieved the maximum efficienc ies of EQE = 3.1% and P = 12 mW/W for BEDOT TQMe 2 based devices, and EQE = 1.5% and P = 4.0 mW/W for BEDOT BBT based devices. The aggregate effect of the donor acceptor donor oligomers in the devices is also studied by utilizing different host matrix and changing the stacking manners of doping molecules. Changing from the previous host of Al q 3 to CBP which has more spectral overlap with the DAD NIR emitters, the concentration of NIR emitters needed to allow complete energy transfer for excitons on the host molecules will be decreased due to increased F rster radius, leading to less significant aggreg ate quenching effect. The NIR OLEDs based on BEDOT BBT doped into CBP host matrix has a maximum EQE of EQE = ( 0.92 0.07 )%, which is almost double that of the devices with Al q 3 as host. The power efficiency is also higher for the CBP device than the devi ce with Al q 3 host with a maximum of P = ( 3.6 0. 3 ) mW/W for the former device The effect of molecular structure on aggregation in devices is then studied by adding bulky substituted groups to the two ends of DAD oligomers. However, the NIR OLEDs based on two bulky DAD BBT based

PAGE 110

110 devices. The efficiencies of these bulky DAD oligomers based devices show no enhancement and the emissions are red shifted with peak emission wavelengths up to 905 nm.

PAGE 111

111 CHAPTER 5 NEAR INFRARED ORGANIC LIG HT EMITTING DEVICES BASED ON PHOSPHORESCENT PLATINUM ( II) PORPHYRINS 5.1 Introduction In last chapter, the fluorescent d onor acceptor donor oligomers featur ing tunable emission from 700 900 nm have been integrated into OLEDs with device efficiencies in the 1 % to 3% range for devices emitting between 700 860 nm, but dropping below 0.5% for devices emitting above 900 nm A major shortcoming of the oligomers and other fluorescent materials is that they only emit from the singlet state, which limits the electroluminescent internal quantum efficiencies achievable to approximately 25%. The triplet excitons, which are three times more in number than the singlet excitons, are not harvested at all, due to the requirement of spin symmetry conservation for these fluorescent molecules. To harvest both singlet and triplet excitons in OLEDs with fluorescent emitters, the so called sensitized fluorescence device architecture has been used, in which a phosphorescent dye is incorp orated in the emissive layer together with the host and fluorescent molecules to funnel the triplet excitons formed on the host molecules to the fluorescent emitters, or sensitize the fluorescent molecules. 154 But there still exists several loss mechanisms of triplet excitoins, leading to less enhancement than that expected b y ideal situation. Additionally, the solution quantum yields of near IR emitting donor acceptor donor oligomers are usually low ( em < 0.25) which also results in the low efficiencies for the OLEDs. So it is necessary to develop the real phosphorescent ne ar IR emitters by which 100% internal quantum efficiency can be achieved, for the higher efficiencies of the near IR OLEDs. Phosphorescent metal organic complexes provide attractive candidates for use in PLEDs and OLEDs due, in part, to their high lumines cence quantum yields and nearly

PAGE 112

112 100% internal quantum efficiencies due to their ability to efficiently emit from the spin orbit mixed triplet states. 34,35 Previously, two classes of phosphorescent metal organic comp lexes have been employed as dopants in NIR OLEDs, The first utilized trivalent lanthanide cations (Ln 3+ ) as the emitting centers, for example, Er 3+ or Nd 3+ chelated with chromophoric ligands to sensitize excitation energy transfer to the lanthanide ion. 169 Schanze e t al. have reported a near IR OLED utilizing Ln 3+ in conjugation with a porphyrin/polystyrene matrix, with EQE ra nging from 8.0 10 4 to 2.0 10 4 % a t approximately 1 mA/cm 2 68 Similarly, a Nd(phenalenone) 3 based NIR OLED had an EQE of 0.007% at max =1065 nm. 70 The second class of NIR OLEDs is transition metal complexes, similar to those used in the visible region. A recent report of an electrophosphorescent device that used a cyclometalated [ (pyrenyl qui nolyl) 2 Ir(acac)] complex as the phosphor gave max = 720 nm and an EQE of 0.1%. 76 More specifically, a subclass of phosphorescent metal o rganic complexes, metalloporphyrins, has shown intense absorption and emission in the red to NIR region of the spectrum. 77,78 There are a number of reports of OLEDs fabricated with PtOEP, PtTPP (OEP = 2,3,7,8,12,13, 17, 18 octaethylporphyrin, TPP = 5,10,15,20 tetra phenylporphyrin), or analogues of these compounds as phosphorescent emitters, with emission maxima between 630 and 650 nm. 35,79 86 Porphyrin chromophores with fused aromatic moieties at the pyrrole positions, for example, tetrabenzoporphyrin (TBP), exhibit a bathochromic shift (relative to unsubstituted porphyrin) of the absorption and emission energy, owing to the expansion of the electronic system of the porphyrin core. 87 The addition of bulky groups to the meso positions of the porphyrin macrocycles with substituted pyrroles leads to the formation of nonplanar porphyrins, and further red shifts the absorption

PAGE 113

113 spectra. 170 Coordination of a heavy metal atom increases the rate of the intersystem crossing between sing let and triplet states of the metalloporphyrins, thereby enhancing the rate of radiative decay from the triple t states. Considering all these reasons, extended Pt porphyrin complexes such as Pt tetrapheny ltetrabenzoporphyrin (Pt TPTBP ) have been studied for NIR OLEDs with a reported maximum EQE of 8.5% at a peak wavelength of 7 70 nm. 88,89 Analysis of the Pt TPTBP complex by X ray crystallography reveals a nonplanar molecular structure with a saddle type distortion similar to that found in other TPTBP derivatives. 171 These Pt porphyrins show relatively narrow phosphorescence emission bandwidths and high quantum yield in solution. But so f ar, to the best of our knowledge all of the NIR emitting Pt porphyrins that have been incorporated into PLEDs or OLEDs have p hosphorescence peaks below 900 nm. In this chapter, based on the NIR emitting phosphorescent Pt TPTBP, we further extend conjugation system by replacing the benzo groups in the pyrrole positions with naphtha and an th ro groups, which are Pt tetr aphenyl tetranaphthoporphyrin (Pt TPTNP) and Pt tetraaryl tetra anthro porphyrin (Pt Ar 4 TAP), to achieve emission s with even longer wavelengths into NIR region. The NIR OLEDs based on Pt TPTNP have t he emission wavelength peaked at ~ 900 nm and, when incorpor ated into OLEDs, a maximum EQE of 3.8 % has been obtained 172 while the more conjugated Pt Ar 4 TAP based NIR OLEDs have further longer emission wavelength peaked at ~1000 nm and EQE up to 0.2%. This is the longest wa velength demonstrated to date for a triplet phosphor in an electroluminescent device. The photophysics of various ex tend ed conjugation porphyrins have been previously investigated, and it has been demonstrated that substituents play a major

PAGE 114

114 role in determi ning porphyrin photophysical properties. 84,173 176 For example, Vinogradov et al. demonstrated that Pd 5,15 diphenyltetrabenzoporphyrin has a phosphorescence quantum yield of 0.15, whereas Pd 5,10,15,20 tetraphenyl tetrabenzoporphyrin has a quantum yield of only 0.08. 173 Furthermore, Beeby, et al. have demonstrated an approximate doubling of solution em from 0.11 to 0.21 for 5,10,15,20 tetrasubstituted free base porphyrins when monophenyl substituents are substituted for more bulky fluorene or terphenyl substituents. 173,174 Drawing from th ese previous work, a family of P t tetrabenz oporphyrins (Pt TBPs ) incorporating 5,15 diaryl and 5,10,15,20 tetraaryl derivatives with varying substituent groups has been synthesized, and the diaryl derivatives display a 50% enhancement in solution quantum yie ld as compared to the tetraaryl derivativ es. 177 But when this family of Pt TBPs is incorporated into NIR OLEDs, the device efficiencies show inconsistent trend of enhancement with that in solution, that the NIR OLEDs based on 5,15 diaryl derivat ives show no enhancement of the device efficiencies compared to the 5,10,15,20 tetraaryl derivatives based devices. Then the devices based on this family of Pt TBPs are further characterized through solid state electroluminescence transient lifetime measur ements, and a strong correlation is observed between the EL lifetimes and device efficiencies. It was also found that the addition of 3,5 di tert butylphenyl groups in place of phenyl groups on the benzoporphyrin ring periphery results in increased device efficiencies A record high maximum EQE of 9.2% is obtained for NIR OLEDs based on Pt TBP with 5,10,15,20 tetraaryl groups, to which 3,5 di tert butylphenyl groups attach (Pt Ar 4 TBP). Finally, the aggregate effect of Pt TBPs on EL efficiency is studied bas ed on the two Pt porphyrin molecules: Pt Ar 4 TBP, which has bulky end groups added in place of

PAGE 115

115 the meso aryl groups on the benzoporphyrin ring periphery and the non substituted counterpart, Pt TPTBP. The extent of aggregation for the two molecules in devic es is compared through the device efficiency decay rate with the increasing doping concentration of dopants. This chapter is organized as follows. The experimental details on optical characterization, device fabrication and characterization are described i n section 5 2 In section 5 3 we present the experimental results and discussion on the optical properties of the extended conjugation Platinum (II) porphyrins and performance of the NIR OLEDs. In section 5 4 we show the experimental results and discussi on on the effect of molecular structure on EL efficiency in Platinum (II) tetrabenzoporphyrins (Pt TBPs). Beginning with a brief introduction of triplet excited states decay and quenching models, w e also study the solid state electroluminescence transient lifetime of the NIR OLEDs based on these Pt TBP emitters in section 5 5 The aggregate effect of Pt TBPs on EL efficiency is studied in section 5 6. Finally, we conclude in section 5 7 5.2 Experimental Details All the P t (II) po rphyrin molecules were synth esized and purified as reported 177 by Jonathan Sommer in Dr. Kirk Schanze group in the Department of Chemistry at the University of Florida He also measured the following solution photophysical data. Ab sorption spectra were measured using a PerkinElmer Lambda 25 UV vis spectrometer. The PL spectra of Pt TBPs and Pt TPTNP were obtained by excitation at the absorption maxima and recorded with an ISA SPEX Triax 180 spectrograph coupled to a Spectrum 1 liqui d nitrogen cooled charge coupled device ( CCD ) detector This spectrometer has a relatively flat spectral response to 900 nm, although there is some loss in efficiency due to the grating, which is blazed in the visible regi on. For Pt Ar 4 TAP,

PAGE 116

116 a Spex Fluorolo g II equipped with an InGaAs near IR photomultiplier tube detector was used. The s olution phosphorescence quantum yield was calculated relative to ZnTPP in CH 2 Cl 2 ( = 0.033) according to a previously described method. The sample and actinometer solutions have matched optical density at the excitation wavelength, and the emission spectra were corrected for the spectrometer response prior to being used to compute the quantum yield. Solution lifetimes were determined from transient absorbance measurements. Th e transient absorption spectra were collected by using previously described laser systems for the visible and near IR regions. NIR OLED devices were fabricated on glass substrates commercially pre coated with an indium tin oxide (ITO) anode ( sheet resista nce ~ 20 ) The cleaning procedure for substrates is already described in Chapter 4 All the layers (except those in the Pt Ar 4 TAP based device), including the cathode, were deposited by vacuum thermal evaporation at a base pressure of 1 x 10 7 Torr following pr eviously published procedures. 39 To fabricate the OLED, a 40 nm thick hole transporting layer (HTL) of NPD an emissive layer (EML) consisting of host matrix doped with the NIR Pt porphyrin emitters, an electron transport layer (ETL) of bathophenanthroline (BP hen ) and a 1 nm thick LiF layer followed an Al cathode layer (50 nm thick) were successively dep osited on the substrates. The emissive layer of all Pt TBPs based OLEDs is Al q 3 with 4% Pt TBPs and has the thickness of 25 nm, while the EML of Pt TPTNP bassed OLEDs is 20 nm thick and cons ists of the CBP host doped with Pt TPTNP. The thickness of the BPh en ETL was 80 nm for Pt TBPs and varied from 100 nm to 120 nm for Pt TPTNP based devices, respectively. More specifically, in one device of Pt TPTNP, the entire BPhen ETL is nominally undoped (specified as the undoped device ). In another device

PAGE 117

117 (specifie d as the n doped device ), Cs was used to n dope BPhen (BPhen:Cs = 1:0.2, molar ratio) except in the 15 nm layer adjacent to the EML. Cs has been shown to serve as an effective n type dopant in BPhen to increase the conductivity of the ETL and improve the efficiency of elect ron injection from the cathode. 178 Pt Ar 4 TAP could not be thermally evaporated due to its high molecular weight; the refore, Pt Ar 4 TAP multi layer O LEDs were constructed by spin casting a layer of Pt Ar 4 TAP in PVK:PBD blend onto the PEDOT:PSS coated ITO /glass substrate followed by thermal evaporation of a 40 nm thick electron transporting layer of BPhen or tris[3 (3 pyr idyl)mesityl]borane ( 3TPYMB ) a 1 nm thick LiF and an Al cathode layer (100 nm thick ) The film of PEDOT:PSS (Clevios P VP Al 4083) was deposited by spin coating at 4000 RPM, and annealed on a hotplate in an argon glovebox for 20 minutes a t 130 C (~40 nm thick). As th e OLEDs were characterized in air without device encapsulation, a 200 nm thick MoO x overlayer is thermally evaporated to the surface of Pt Ar 4 TAP devices, to improve the OLED device stability as they are exposed to the laboratory ambient for t he short time period of testing and characterization. Individual solutions of poly(9 vinylcarbazole) (PVK, 20 mg/mL), 2 (4 tert Butylphenyl) 5 (4 biphenylyl) 1,3,4 oxadiazole ( PBD 20 mg/mL), Pt Ar 4 TAP (3.5 mg/mL) were prepared in anhydrous and deoxygenate d chlorobenzene and stirred overnight. Solutions were combined to obtain PVK:PBD ratios varied from 6:4 to 7:3, and the doping concentration of Pt Ar 4 TAP varied from 2% to 4% The (Whatman puradisc) filters and using a clean glass syringe they were directly deposited onto the PEDOT:PSS coated ITO /glass substrates The solution concentrations were 12 or 9 mg/mL and then spi n cast at 1000 or 2000 RPM for 60 s both two parameters

PAGE 118

118 changed for slightly different active layer thickness. The active layer thickness for 12 mg/mL solution spun coating at 1000 RPM was ~110 nm as determined by AFM. These solution processed film deposi tions and characterizations are done by Kenneth Graham in Dr. John Reynolds group in the Department of Chemistry at the University of Florida. The deposition rates for the organic layers were 0.1 0.2 nm/s, as monitored by quartz crystal microbalances. The active device area was 4 mm 2 with the electrodes arranged in a cross bar geometry, and each ITO/glass substrate featured four independen tly addressable device pixels. Al q 3 and BPhen were purchased from TCI America, 3TPYMB from Lumtec, NPD from e Ray, an d CBP from Springchem & Jadetextile Group L td. All organic materials were used as obtained except for NPD and BPhen, which were purified in house using vacuum gradient sublimation method for one to two cycles. 14 Radiant emittance ( R ) current density ( J ) voltage ( V ) characteristics of the NIR OLEDs were measured in ambient using an Agilent 4155C semicondu ctor parameter analyzer and a calibrated silicon detector ( Newport 818 UV) EL spectra of Pt TBPs and Pt TPTNP based devices were taken with the ISA SPEX Triax 180 spectrograph coupled to a Spectrum 1 liquid nitrogen cooled silicon CCD detector with the de vice s driven at a constant current using a Keithley 2400 source meter For Pt Ar 4 TAP based OLEDs, a Spex Fluorolog II equipped with an InGaAs near IR photomultiplier tube detector was used. But for the visible emission of this device, the CCD was still use d. The radiant emittance was calibrated assuming Lambertian emission and t he EQE ( EQE wall plug power efficienc y ( P ) were derived based on the recommended methods. 158 The organic layer thickness non uniformity across the samples and run to

PAGE 119

119 run variations typically lead to 5 10% error in device efficiencies. All measurements are averages over at least ten pixels and are presented along with standard deviations. Electroluminescence transient lifetimes were measured by applying a 1 ms voltage pulse to the device. The voltages were created by a Tektronix AFG3101 function generator (100 MHz) and the magnitude of voltages was chosen to keep the current density at 10 mA/cm 2 The decay sig nals were captured by a Newport 818 UV photodetector which was connected with a Tektronix DPO3054 digital phosphor oscilloscope (500 MHz) for data acquisition. The electroluminescence decay was fitted to a bi exponential decay function and the average life time was calculated by a standard method. 179 181 5.3 Extended Conjugation Platinum (II) Porphyrins The molecular structures of the series of increasing conjugation phosphorescent Pt porphyrins: Pt TPTBP, Pt TPTNP and Pt Ar 4 TAP, are shown in Fig ure 5 1. Figure 5 1. Molecular structures of increasing conjugation series of phosphorescent Pt porphyrins: Pt TPTBP, Pt TPTNP and Pt Ar 4 TAP. The PL spectra of these Pt porphyrins in tol uene solutions are shown in Fig ure 5 2. system is extended across the series of Pt TPTBP, Pt TPTNP, and Pt Ar 4 TAP the solution emission wavelength maxima shift from 773 to 891 to 1022 nm, respectively as expected. A ccompanying this wavelength shift is a decreasing

PAGE 120

120 phosphorescence quantum yield ( em ) and decreasing lifetime ( em ) as predicted by the energy gap law 182 184 The PL quantum yield of Pt TPTNP is measured to be 0.20, two times lower than Pt TPTBP, whose PL quantum yield is 0.46. And Pt Ar 4 TAP shows even lower em of 0.11. The solution PL lifetimes follow the trend as the PL quantum yield, in which the lifetime of Pt TPTBP is 29.9 s, and then decreases to 12.7 and 3.2 s for Pt TPTNP and Pt Ar 4 TAP, respectively. All these data including PL emission maxima max em PL quantum yields em and lifetimes em in the toluene solution are listed in Table 5 1 for comparison. Figure 5 2 The photoluminescence spectra of Pt TPTBP (solid lines), Pt TPTNP (dashed lines) and Pt Ar 4 TAP (dash dotted lines) in toluene solutions. The trend of decreased quantum yields and lifetimes is partially due to the increased non radiative decay rate in th ese lower band gap molecules. To better prove this conclusion, the radiative and nonradiative decay rate constant, k r and k nr are calculated by the following equations: em = k r em ( 5 1) k = 1/ em = k r + k nr ( 5 2)

PAGE 121

121 where k is the total decay rate constant The results of the calculated k r and k nr are also listed in Ta ble 5 1. The non radiative decay rate of Pt Ar 4 TAP is more than ten times higher than Pt TPTBP, and four times higher than Pt TPTNP, while the radiative decay rates of all three Pt porphyrins are within the same order of magnitude. Table 5 1. Photophysica l data ( PL emission maxima max em PL quantum yields em lifetimes em and radiative and non radiative decay rate constant, k r and k nr ) for Pt TPTBP, Pt TPTNP and Pt Ar 4 TAP in deoxygenated toluene. Pt porphyrins max em (nm) em em k r (10 2 1 ) k nr (10 2 1 ) Pt TPTBP 773 0.46 29.9 1.5 1.8 Pt TPTNP 891 0.20 12.7 1.6 6.3 Pt Ar 4 TAP 1022 0.11 3.2 3.4 28 The solution em value generally provides a good figure of merit to predict the efficiency of a series of structu rally related chromophores in the OLED s ; therefore, a similar trend of decreasing OLED efficienc ies across the series of Pt TPTBP, Pt TPTNP, and Pt Ar 4 TAP is expected as the emission is shifted into the near IR region 5.3.1 Near IR OLEDs based on Pt TPTNP NIR OLEDs based on Pt TPT BP was first fabricated to compare with the already reported results by Thompson and co workers. The EL emission wavelength peaked at 773 nm and EQE can be achieved up to 8.0% in our devices. These results are very similar to what Thompson and co workers h ave reported, considering the allowable deviation between different scientific labs probably due to the different film thickness calibrations of vacuum thermal evaporation method 88,89 OLEDs based on the Pt TPTNP we re fabricated via vacuum thermal evaporation. The emissive layer (EML) consisting of CBP doped with Pt TPTNP, was sandwiched between a hole transport layer (HTL) of NPD and an electron transport layer (ETL) of

PAGE 122

122 BPhen. 172 The R J V and EQE characteristics and of Pt TPTNP based NIR OLEDs with various doping concentrations and ETL thickness are shown in Figure 5 3. Figure 5 3. (color) A) Current density, J and the radiant emittance in the forward viewing directions, R as functions of the voltage, V for the devices based on Pt TPTNP with various doping concentrations and ETL thickness Cs was also used to n dope the 100 nm ET L layer ( n doped device). B) External quantum efficiency, EQE of these devices as functions of J Inset: Power efficiency, P of the undoped and n doped devices with 100 nm thick ETL and 8% doping concentration, as functions of J As mentioned in Chapter 4, the ETL thickness will affect the device efficiencie s due to the microcavity effect in OLEDs By using the 10% doping concentration of Pt TPTNP, the EQEs of devices achieve the maximum value as the ETL thickness is reduced from 120 nm to 1 00 nm, as shown in Figure 5 3B Then, by using this optimal

PAGE 123

123 ETL thick ness of 100 nm, the doping concentration of Pt TPTNP is varied to achieve a balance between the maximum energy transfer and minimum aggregate quenching. The doping concentration with maximum efficiency is found to be 8%, as shown in Figure 5 3 B by compari ng the devices with doping concentration from 8% to 10%. All these devices show close current densities and radiant emit tance, as shown in Figure 5 3A T o increase the conductivity and improve the efficiency of electron injection from the cathode, Cs was also used to n dope the 100 nm BPhen ETL layer ( n doped device) As shown in Figure 5 3A a t V > 2.5 V, the current density was approximately two orders of magnitude higher in the n doped device, due to the significantly enhanced conductivity in the n dope d ETL compared with that in the nominally undoped 100 nm thick ETL The undoped device show s a low turn on voltage of approximately 2.2 V even though the BPhen ETL is rather thick (100 nm) which is further reduced to 2.0 V for the n doped device M aximum radiant emittance s of 1. 8 mW/cm 2 are obtained ( at V = 17 V in the undoped device and V = 1 2 V in the n doped device), which are similar to those obtained from our previously reported fluorescent NIR OLEDs in Chapter 4 that emit at shorter wavelengths (pea k emission s from 700 nm to 815 nm). 185,186 For the undoped device with optimal doping concentration of 8% and ETL thickness of 100 nm t he EQE is relatively constant at low current densities and reaches a maximum o f EQE = ( 3.8 0. 3 )% at J 0.1 mA/cm 2 ; however, at J > 1 mA/cm 2 it decreases significantly with the increase of the current density to EQE = 2.0% at J = 1 0 mA/cm 2 and EQE = 0. 6 % at J = 1 00 mA/cm 2 The significant roll off at higher current densities is likely due to the triplet triplet exciton annihilation process that commonly occurs in phosphorescent OLEDs. 187 The maximum power efficiency of the undoped

PAGE 124

124 device is P = ( 19 3 ) mW/W achieved at low current densities ( J 10 3 mA/cm 2 ) as shown in the inset of Figure 5 3 B Such efficiency is approximately three to ten times higher than the maximum P of the NIR fluorescent OLEDs we mentioned earlier. 185,186 These efficiencies are lower compared to Pt TPTBP based devices, but are approximately proportional to the corresponding PL quantum yi elds of these two molecules ( Table 5 1). Compared with the undoped device, t he n doped de vice ha s slightly lower quantum efficiencies with a maximum of EQE = 3.3%. The maximum P of the n doped device is P = 1 7 mW/W, also slightly lower than the undoped device; however the lower drive voltage resulted from the increased conductivity of the n doped ETL leads to higher power efficiencies at J > 1 0 2 mA/c m 2 for the n doped device. For example, at J = 1 mA/cm 2 P = 1 2 mW/W for the n doped device, more than 40% higher than that of the undoped device ( P = 8.4 mW/W). Figure 5 4. Normalized electroluminescence (EL) spectra for NIR OLEDs based on Pt TPTBP (solid line) and Pt TPTNP (dashed line). Figure 5 4 shows the EL spectra of the NIR OLEDs based on Pt TPTNP as well as the Pt TPTBP for comparison. The peak emission wavelengths for Pt TPTBP and Pt

PAGE 125

125 TPTNP based devices are 773 nm and 891 nm, nearly identica l to the corresponding PL peaks in solution (Table 5 1). It is noticed that the full width half maxima (FWHM) of the EL peaks of Pt porphyrin based devices are in the range of 50 60 nm, much narrower than those of the fluorescent DAD oligomers based NIR OL EDs, which are usually in the range of 100 200 nm. 5.3.2 Near IR OLEDs based on Pt Ar 4 TAP The NIR OLEDs based on Pt Ar 4 TAP were fabricated with the device structure of ITO/PEDO:PSS/PVK:PBD:Pt Ar 4 TAP/BPhen or 3TPYMB/LiF/Al. Figure 5 5 schematically shows th e energy level diagram of the NIR OLED based on Pt Ar 4 TAP, in which the electrode work functions and the HOMO/LUMO energies for PVK, PBD, BPhen and 3TPYMB are taken from the literature. 43,188 PEDOT:PSS is served as the hole injection layer (HIL) while PBD and BPhen or 3TPYMB is served as the ETLs Figure 5 5. (color) Schematic energy level diagram of OLEDs based on Pt Ar 4 TAP The energies (in eV) are measured from the vacuum level. The red and blue lines correspond to the HOMO/LUMO of PVK and PBD, respectively. The energy levels of Pt Ar 4 TAP are not indicated here. The blend ratio of PVK and PBD, doping concentration of Pt Ar 4 TAP, and spin rate for the active layer, were studied to optimize the efficiency of device s. BPhen is used as

PAGE 126

126 the ETL and the concentration of solution for active layer deposition is 12 mg/mL for all these devices. Figure 5 6 A shows the J V characteristics of these devices with different preparation recipes of the active layer s All the devices with active layers spun coated with 2000 RPM (Device 2, 3, 6, and 7) show lower turn on voltages and higher injection currents than the corresponding devices with 1000 RPM spin rate (Device 1, 4, 5, and 8) which is as we predicted due to the reduced devic e thickness. Furthermore, by increasing the doping concentration of Pt Ar 4 TAP from 2% to 4% while keeping all the other parameters the same, the injection current densities of devices as well as the turn on voltages are also decreased (for examples, Device 2 vs 3, and Device 5 vs 8), suggesting that the charge trapping by the dopant molecules cannot be neglected. Finally, the blend ratios of PVK and PBD also play a role for the device performance that decreasing the portion of PBD in the blend active laye r from 6:4 to 7:3 will enhance the current injection and then decrease the turn on voltages. Experiment has also been done to further decrease the PBD concentration in PVK:PBD blend to 8:2 in the devices, but no significant change has been observed. Overal l, the highest current density and lowest turn on voltage can be achieved by using spin rate of 2000 RPM, 2% doping concentration of Pt Ar 4 TAP and PVK:PBD ratio of 7:3 ( Device 6 ), as shown in Fig ure 5 6 A The turn on voltage of Device 6 is approximately 8 V, still much higher than the normal devices, mainly due to the thick active layer (~100 nm). So only 40 nm thick ETL layer is used here to avoid further increase of turn on voltages. The EQE of these devices are also compared in Fig ure 5 6 B Device 6, whi ch has the maximum current injection, also shows the highest EQE among all these devices, indicating the charge balance in Device 6 is also optimized.

PAGE 127

127 Figure 5 6. (color) A ) Current density, J as functions of voltage, V for Pt Ar 4 TAP based OLEDs with d ifferent preparation recipes of the active layer. B ) External quantum efficiency, EQE as functions of current density, J for these OLEDs. The recipes for each device are list ed in the table. The solution concentration is 12 mg/mL and ETL is BPhen for al l the devices. We also studied the effect of using different ETL layers by replacing BPhen with 3TPYMB, while maintaining the recipe of Device 6 whi ch leads to the optimal current injection and device efficiency. As shown in Fig ure 5 5, 3TPYMB has a lower lying HOMO level than BPhen which can combine the holes within the active layer more effectively. 3TPYMB also has higher triplet energy than BPhen, 43 which is beneficial to the exciton blocking. The J V characteristics of NIR OLEDs based on Pt Ar 4 TAP with different ETLs are shown in Fig ure 5 7 A The concentrations of the solution for

PAGE 128

128 depositing active layers are also varied from 12 to 9 mg/mL to reducing the device thickness. As predicted, the devices spun coated with 9 mg/mL solution (Device 10 and 11) show higher current d ensities and lower turn on voltages than the devices prepared with 12 mg/mL solution (Device 6 and 9) Figure 5 7. (col or) A ) Current density, J as functions of voltage, V for Pt Ar 4 TAP based OLEDs with different ETLs of BPhen or 3TPYM B and solution co ncentration. B ) External quantum efficiency, EQE as functions of current density, J for these OLEDs. The recipes for each device are list ed in the table. The PVK:PBD ratio is 7:3, the Pt Ar4TAP is 2%, and spin rate is 2000 RPM for all these devices. By comparing the effect of using different ETLs, it is found that the devices with BPhen as ETL have higher current density than 3TPYMB, while the turn on voltages are almost the same. This can be attributed to the lower conductivity of 3TPYMB and higher inje ction barrier of electrons due to the lower lying LUMO energy l evel of 3TPYMB. 43 As shown in Fig ure 5 7 A the turn on voltage can be further reduced to 5.5 6.0 V by using 9

PAGE 129

129 mg/mL solution for active layer deposition and either BPhen or 3TPYMB as the ETL. The EQE of t hese dev ices are shown in Fig ure 5 7 B It is found that the difference of solution concentration does not affect the device efficiencies appreciably. The devices with 3TPYMB as ETL show higher maximum EQE than the BPhen devices, but this is achieved at low current density o f J 1 mA/cm 2 and its roll off at higher current densities is more significant than the BPhen devices. So at high current density ranges ( J > 10 mA/cm 2 ), the devices with BPhen as ETL have higher EQE than the 3TPYMB devices as shown in F ig ure 5 7 B The power efficiencies can be predicted to have the same trend as the EQE comparison. The device characterization of Device 10 with the optimal device structure (PVK:PBD ratio is 7:3, the doping concentration of Pt Ar 4 TAP is 2%, solution concentration is 9 mg/mL, s pin rate for active layer is 2000 RPM, and BPh en as ETL) is summarized in Fig ure 5 8. 189 This device exhibits turn on voltage of approximately 5.7 V. Maximum radiant emittance of 1.3 mW/cm 2 is achieved for this device at V 13 V, as shown in Fig ure 5 8 A The EQE and power efficiency of this optimal device are shown in Fig ure 5 8 B This device has a maximum EQE of EQE = ( 0.24 0. 03 )% achieved at J 10 mA/cm 2 slowly decreasing to EQE = 0. 14 % at J = 5 00 mA/cm 2 The maxim um power efficiency is P = ( 0.47 0. 03 ) mW/W also achieved at J 10 mA/cm 2 These efficiencies of EQE and power efficiency are lower than those expected from the quantum yield, which may originate from degradation or impurities. It should be noted that the Pt Ar 4 TAP device does show some visible emission as shown in Fig ure 5 8 C and it is thought that this could be due to degradation or impurities since the material is light sensitive and was also difficult to purify. The visible

PAGE 130

130 Figure 5 8. A ) Curr ent density, J and radiant emittance in the forward viewing directions, R as functions of voltage, V for the optimal NIR OLEDs based on Pt Ar 4 TAP. B ) External quantum efficiency, EQE and power efficiency, P as functions of current density, J for this device. C ) Normalized electroluminescence (EL) spectra of this device. emission is relatively broad with peak emission wavelength of ~440 nm, and the intensity is about 20% of t hat of the NIR peak at 1005 nm from Pt Ar 4 TAP. The peak emission wavelength is slightly blue shifted compared to the corresponding PL peak i n

PAGE 131

131 solution. For the radiant emittance and EQE measurement and calculation, the visible emission from the Pt Ar 4 TAP d evice was not removed which may result in the deviation to some extent. The measurement can be improved by using an 800 nm long pass filter to remove the visible emission from the devices. Through t he extension of the system from Pt TPTBP to Pt TPTNP and Pt Ar 4 TAP, the electroluminescence was shifted up to 1005 nm, which is the longest wavelength demonstrated to date for a triplet phosphor in an electroluminescent device. These efficiencies of EQEs and power efficienc y, and radiant emittance values achieved by NIR OLEDs based on this series of extended conjugation Pt porphyrins are also some of the highest reported for OLED s emitting in 750 1050 nm range. 5.4 Effect of Structure on EL Efficiency in Platinum (II) Tetr abenzoporphyrins The photophysics of various expanded conjugation porphyrins have been previously investigated, and it has been demonstrated that substituents can play a major role in determining porphyrin photophysical properties. 84,173 176 Here, a family of P t tetrabenzoporphyrins (Pt TBPs ) incorporating 5,15 diaryl and 5,10,15,20 tetraaryl derivatives with varying s ubstituent groups has been synth esized to provide insight into how structural variation of the exp anded conjugation Pt porphyrins impact their solution and solid state photophysics, and how these properties relate to the OLED performance. 189 The molecular structures of the series of Pt porphyrins are shown in Fig ure 5 9. The photophysical properties including solution PL quantum yields em and lifetimes em of th e se Pt TBPs are listed in Table 5 2 Radiative and nonradiative decay rate constant, k r and k nr are also calculated by equation 5 1 and 5 2 sh own above.

PAGE 132

132 Figure 5 9. Molecular structures of Pt porphyrins: Pt TPTBP, Pt Ar 4 TBP, Pt DPTBP and Pt Ar 2 TBP, Pt TAr 2 TBP, and Pt Ar 2 OPrTBP. As shown in Table 5 2, t he solution em and em values are 50 60% larger for the di substituted 5,15 diaryl porphyrin derivatives relative (except Pt DPTBP) to the 5,10,15,20 tetraaryl substituted porphyrins. The tetra substitute d Pt porphyrins have a non planar molecular structure with a large degree of out of plane (saddle type) distortion, while the less substituted diaryl porphyrins still keep the planar structure in spite of in plane distortion due to the insertion of meso ar yl groups perpendicular to the molecular plane. 173,176,190 192 The nonplanar deformation will dramatically enhance the nonradiative decay of triplet states of Pt porphyrins, similar to the Pd porphyrins reported ea rlier. 173 This is consistent with the calculated results of k r and k nr values listed in Table 5 2. The nonradiative decay rates of di substituted Pt porphyrins are approximately 2 3 times lower than the tetra substituted Pt porphyrins, while the radiative decay rates are within 20% difference for all the Pt porphyrins.

PAGE 133

133 Furthermore, meso aryl substituted Pt porphyrins have been reported to be rather poor red light em itting dyes in OLEDs, which is mainly ascribed to some conjugative effect of the meso aryl substituents that stems from their somewhat free rotational movement. It is believed that such a conjugative interaction of meso aryl substituents will also accelera te the nonradiative decay of the triplet excited state of the Pt porphyrins. 193,194 The di substituted Pt porphyrins now can decrease one pair of meso aryl substituents and also their free rotational movement, by w hich the nonradiative decay rate will also be reduced. So both the lack of out of plane distortion and less rotation of the meso aryl groups for di substituted Pt porphyrins can suppress the nonradiative decay rate while the radiative rate does not change, leading to the much higher solution quantum yields and lifetimes of di substituted Pt porphyrins compared to the tetra substituted ones. Table 5 2. Photophysical data ( PL quantum yields em lifetimes em radiative and nonradiative decay rate constant k r and k nr ) measured in toluene and device characteristics (maximum EQEs, EQE ,max and EL lifetimes, EL ) for Pt TPTBP, Pt Ar 4 TBP, Pt DPTBP, Pt Ar 2 TAP, Pt TAr 2 TBP, and Pt Ar 2 OPrTBP. The EL lifetimes are measured at 10 mA/cm 2 Pt porphyrins Pt TPTBP Pt A r 4 TBP Pt DPTBP Pt Ar 2 TBP Pt TAr 2 TBP Pt Ar 2 OPrTBP em 0.44 0.44 0.40 0.65 0.58 0.60 em ( s) 29.9 32.0 28.0 53.0 51.7 51.8 k r (10 2 1 ) 1.5 1.4 1.4 1.2 1.1 1.2 k nr (10 2 1 ) 1.8 1.7 2.2 0.7 0.8 0.7 EL EL ( s) 27.0 24.0 26.5 19.2 23.6 EQE max (%) 8.0 0.5 9.2 0.7 5.0 0.3 7.8 0.5 3.2 0.2 7.0 0.5 T o further reduce the effect of rotational freedom o f meso aryl substituents, steric bulky end groups are added in place of these meso aryl groups on the benzoporphyrin ring periphery, to s everely restrict the rotational freedom of meso aryl substituents. These steric bulky end groups can also prevent the stacking interactions and then aggregation when in the solid state films. This is beneficial for application in devices

PAGE 134

134 because the decreased aggregation of molecules will reduce the triplet triplet annihilation processes in the solid state. We first a dd 3,5 di tert butylphenyl groups to each of meso aryl substituents in the Pt TBPs for both di and tetra substituted Pt TBPs. With these bulky end groups, Pt Ar 4 TBP and Pt Ar 2 TBP both have higher quantum yields and lifetimes in the solution compared to th eir non bulky substituted counterpart. Especially for Pt Ar 2 TBP, the lifetime in the solution is almost doubled of that of Pt DPTBP. Molecules with even bulkier end groups providing even significant restriction effect are also synthesized based on the di s ubstituted Pt TBPs, which are Pt TAr 2 TBP and Pt Ar 2 OPrTBP. But the solution quantum yields and lifetimes are not obviously increased compared to Pt Ar 2 TBP, which may indicate the 3,5 di tert butylphenyl groups already have enough size to effectively restri ct the rota tion of meso aryl substituents. Based on the solution behavior of the above Pt TBPs, we predict that the di substituted Pt TBPs with bulky end groups to the meso aryl substituents would give rise to more efficient NIR OLEDs compared to the exist ing OLEDs based on Pt TPTBP. Figure 5 10 A shows the EL spectra of the NIR OLEDs based on Pt TPTBP, Pt Ar 4 TBP, Pt DPTBP and Pt Ar 2 TBP. The structural modifications of these Pt TBPs do not shift the emission peaks of the corresponding devices, all at wavelen gths of 773 777 nm. The two di substituted Pt TBPs based devices have very narrow emission peaks with FWHM of only 40 nm, 50% narrower than the tetra substituted Pt TBPs based devices. This may be attributed to the out of plane distortion of tetra substitu ted Pt TBPs. The host emission of Al q 3 is completely quenched for all the four devices and only emission from the dopant molecules is observed, as shown in Fig ure 5 10 A

PAGE 135

135 Figure 5 10. ( c olor) A ) Normalized electroluminescence (EL) spectra of the NIR OLED s based on Pt TPTBP (solid red squares), Pt Ar 4 TBP (solid blue triangles), Pt DPTBP (open black circles) and Pt Ar 2 TBP (open green diamonds). B ) Current density, J and radiant emittance in the forward viewing directions, R as functions of voltage, V fo r these devices. The J V characteristics of these OLEDs are shown in Fig ure 5 10 B All the four devices exhibit very close turn on voltage of approximately 2.3 V. The tetra substituted Pt TBPs based OLEDs show higher current densities than the di substitut ed Pt TBPs devices, while the two devices within the same category do not show too much difference of the current injection. Accordingly, the tetra substituted Pt TBPs based OLEDs also exhibit higher radiant emittances than the di substituted Pt TBPs devic es, especially at voltages from 3 7 V. The maximum radiant emittances of 3.4 and 4.4

PAGE 136

136 mW/cm 2 are achieved for the Pt TPTBP and Pt Ar 4 TBP based devices, respectively, while the Pt DPTBP and Pt Ar 2 TBP based devices show the values of 2.1 and 3.0 mW/cm 2 all a t V = 15 V. Figure 5 11 shows the dependencies of EQE and P on the current density for NIR OLEDs based on these four Pt TBPs. NIR OLEDs based on Pt Ar 4 TBP exhibit the highest EQE and power efficiencies while the efficiencies for Pt TPTBP based devices are slightly lower. This result is consistent with t he solution photophysical data of these two Pt porphyrins. Pt DPTBP based devices have the lowest efficiencies which are also in agreement with its low est solution quantum yield and lifetime. Figure 5 11. (color) External quantum efficiency, EQE and p ower efficiency, P of the NIR OLEDs based on Pt TPTBP (solid red squares), Pt Ar 4 TBP (solid blue triangles), Pt DPTBP (open black circles) and Pt Ar 2 TBP (open green diamonds) as functions of J To our surprise, not like the solution photophysical data ha s shown for Pt Ar 2 TBP that the solution quantum yield and lifetime are much higher than the other three Pt

PAGE 137

137 TBPs, the OLEDs based on Pt Ar 2 TBP have maximum EQE efficienc y approximately 15% lower than Pt Ar 4 TBP based devices, and only close to the Pt TPTBP b ased devices. It indicates that the trends observed when comparing the em and em values for the series of Pt TBPs in solution do not carry over to the corresponding efficiencies of devices, in which these Pt TBP molecules are in solid state films mixed w ith host matrix Thus, while the solution photophysical data gives rise to the prediction that the di substituted porphyrins will give more efficient electroluminescence in the devices, the device results contradict this, suggesting that the photophysical behaviors are different while in the solution and in the solid state A maximum EQE of EQE = ( 9.2 0.7) % is achieved at J 2 mA/cm 2 slowly decreasing to EQE = 4.0% for J up to 10 mA/cm 2 which is still over 40% of the maximum value. The OLEDs based on Pt TPTBP and Pt Ar 2 TBP have slightly lower EQEs of EQE = (8.0 0.5) % and (7.8 0 .5) % respectively, while Pt DPTBP based devices show significantly lower efficiencies than the other three OLEDs, achieved of EQE = (5.0 0.3) % The bulkier di substituted Pt porphyrin s Pt TAr 2 TBP and Pt Ar 2 OPrTBP were also used for NIR OLEDs and comp ared with the Pt Ar 2 TBP derivative as shown in Fig ure 5 12. These two molecules have similar solution photophysical behavior with Pt Ar 2 TBP as shown in Table 5 2. However, the bulkier substituents to the benzoporphyrin ring periphery may lead to further s uppression of triplet triplet annihilation due to decreased porphyrin/porphyrin interactions. The EL spectrum of Pt TAr 2 TBP based devices shows the same emission peak of 777 nm with the Pt Ar 2 TBP based devices but the emission is broader, as shown in Fig ur e 5 12 A The emission peak wavelength red shifts to 792 nm for the NIR OLEDs based on Pt Ar 2 OPrTBP As shown in Fig ure 5 12 B these two

PAGE 138

138 Figure 5 12. (color) A ) Normalized electroluminescence (EL) spectra of the NIR OLEDs based on Pt TAr 2 TBP (red) and Pt Ar 2 OPrTBP (blue). The EL spectrum of NIR OLEDs based on Pt Ar 2 TBP (black) is also shown for comparison. B ) Current density, J and radiant emittance in the forward viewing directions, R as functions of voltage, V for these two devices. C ) External qua ntum efficiency, EQE and power efficiency, P as functions of current density, J for these two devices.

PAGE 139

139 devices both have close turn on voltage of approximately 2.3 V to that of Pt Ar 2 TBP based devices and the maximum radiant emittances of 1.9 and 3.2 are achieved at V = 15 V for Pt TAr 2 TBP and Pt Ar 2 OPrTBP based devices, respectively, both lower than the Pt Ar 2 TBP based devices. Figure 5 12 C shows the EQE and power efficiencies of these two devices. The Pt Ar 2 OPrTBP based device has a maximum EQE of E QE = ( 7 0 0. 4 )% achieved at J 10 2 mA/cm 2 while t he EQE for Pt TAr 2 TBP based device is much lower, with a maximum of EQE = ( 3.2 0. 3 )% at J 10 2 mA/cm 2 although its roll off at higher current densities is less significant. The maximum power effic iency is P = ( 37 3 ) mW/W for the Pt Ar 2 OPrTBP based device, and is ( 16 2 ) mW/W for the Pt TAr 2 TBP based device, both achieved at low current densities ( 10 3 < J < 10 2 mA/cm 2 ). Not like the similar solution behaviors for these two molecules with Pt Ar 2 TBP, these efficiencies for the devices based on these two molecules are all lower than the Pt Ar 2 TBP based devices especially for the Pt TAr 2 TBP devices, which only has less than 50% of the maximum EQE of Pt Ar 2 TBP based devices. This is not the first time the disagreement between solution photophysical data and device effciencies has been found, as we have previously found that Pt Ar 2 TBP based devices exhibit a lower EQE than the solution photophysical data would indicate. 5.5 Electro luminescence Trans ient Lifetime of Devices The disagreement between solution photophysical data and device effciencies has been found previou sly. T he trends observed when comparing the em and em values for the series of Pt TBPs in solution do not carry over to the corresp onding efficiencies of devices based on these molecules in which these Pt TBP molecules are in solid state

PAGE 140

140 films mixed with host matrix To explain these results and better evaluate the device performance in the future, the photophysical data such as quan tum yield and lifetime measured in solid state instead of in solution should be used, so that they have the most similar situation as that in devices. Therefore in this section, we did the transient analysis of organic electrophosphorescence for the NIR OL EDs based on the series of Pt TBPs shown in Fig ure 5 9. 189 5.5.1 T heory First, we consider the simplest situation that the triplet excited states in phosphor, [ 3 T *], decay naturally through radiative and non radiative ways without any possible quenching paths .The concentration of triplet excitons is determined by the fo llowing equation: (5 3) where k T is the total triplet exciton decay rate constant and has the following relation with phosphorescent recombinat ion lifetim e : (5 4) Assuming that the power intensity ( I ) is linearly proportional to the concentration of excited states, then the phospho rescent emission intensity can be determined by solving the E quation 5 3: (5 5) It is a mono exponential decay for the situation that the triplet excited states only naturally decay, without any other ways to quench the triplet excitons.

PAGE 141

141 Usually, one pronounced characteristics of electrophosphorescence is a roll off efficiency at high current densities. 35,79 ,156 It has been noted that the onset of this roll off occurs at increasing current densities as the transient phosphorescent lifetime is decreased. 156 Hence the phosphor Ir(ppy) 3 with a ~500 ns excited state lifetime, 156 has a significantly higher quantum efficiency at typical operating current densities of J 1 mA/cm 2 than does PtOEP, with a lifetime of ~30 s. 35 A possible explanation is that long transient lifetimes increase the likelihood for saturation of phosphorescent sites. Thus, devices incorporating a condu ctive organic host materials doped with Ir(ppy) 3 saturate at higher current densities than similar PtOEP doped devices. However, the onset of the efficiency roll off (~1 mA/cm 2 for PtOEP and ~100 A/cm 2 for Ir(ppy) 3 ) occurs at much lower current density tha n is required to fully saturate phosphorescent sites with a density of ~10 19 cm 3 156 We would also expect that saturation of phosphorescent sites would lead to an efficiency roll off proportional to 1/ J where J is the current density. But at the current densities of interest, the roll off is much more gradual. Thus, saturation alone cannot explain the observed behavior. Rather, the observations are consistent with triplet triplet ( T T ) annihilation dominating electrophosphorescence until relatively high current densities. For simplicity of analysis we assume that only guest triplets participate in T T annihilation, which will be another pathway of triplet loss. So E quation 5 3 then can be re written as: 114 (5 6) where k TT is triplet triplet annihilation rate constant. By solving the E quation 5 6, the phosphorescent emission intensity can be written as:

PAGE 142

142 (5 7) where [ 3 T *(0)] is the ini tial concentration of triplet excitons. Since only the triplet triplet annihilation of guest triplets is considered in E quation 5 6, it means, the host triplet states should be much higher than that of guest, so that the triplets can be well confined on gu est. But if there is only small energy difference between the host and guest triplet states, T T annihilation will be present in both the host and guest, and that energy transfer between the two species is responsible for the multiple lifetimes needed to m odel the transient decays. 187 An alternative model for quenching is tr iplet polaron annihilation. 114,195 The concentration of triplet excitons can be determined by the following equation: 114,195 (5 8) where k TP is triplet polaron annihilation rate constant, [ h ] is the density of trapped holes. By solving the E quation 5 8, the phosphorescent emission intensity can be written as: (5 9) where N is the total number of guest molecules. Equation 5 9 can be re written as: (5 10) where: (5 11) It is a bi exponential decay that one component of the transient lifetimes is the natural decay lifetime, 1 and the other component 2 obviously smaller than 1 originating

PAGE 143

143 from the quenching of triplet excitons by guest molecule cations. The average transient lifetime, avg can be calculated as follows: 179 181 (5 12) 5.5.2 Transient Analysis of NIR OLEDs based on Pt TBPs Here, e lectroluminescence transient lifetimes of near IR OLEDs based on the series of Pt TBPs shown in Fig ure 5 9 were measured by applying a very short ( 1 ms ) voltage pulse to th e device s based on these Pt TBPs and t he phosp horescence decay signals of triplet excited states then can be captured to extract the EL lifetimes of the Pt TBPs in devices. In this measurement, the Pt TBP molecules are in the solid state films mixed with host matrix, and excitons are formed by boundin g of electrically injected holes and electrons. Therefore, it may provide the best estimation for the behaviors happened in devices. Figure 5 13 shows the phosphorescence decay of Pt Ar 4 TBP, Pt DPTBP and Pt Ar 2 TBP in their corresponding NIR OLEDs. The devi ces have exactly the same structures as mentioned above, and are operated in ambient at the constant current density of 10 mA/cm 2 The decay curves are fitted to a bi exponential function as recommended by literature 89 and the average EL transient lifetim e s can be calculated by E quation 5 12. The fitting results are also shown in the inset of Figure 5 13. The bi exponential function has a very good fitting agreement with the measured curves. The EL lifetimes of Pt Ar 4 TBP, Pt Ar 2 TBP and Pt DPTBP are measure d to be 27.0, 26.5 and 24.0 s, respectively. Note that the EL lifetime of Pt Ar 2 TBP in devices decreases by 50% of the value in the solution while the EL lifetimes for the other two molecules are only slightly decreased. Pt Ar 2 TBP does not display a longer EL lifetime than Pt Ar 4 TBP

PAGE 144

144 as was observed in the solution PL lifetimes, which is in good correlation with the trend for OLEDs efficiencies that Pt Ar 2 TBP based devices did not show higher efficiencies than the Pt Ar 4 TBP devices. Figure 5 13. (color) Phosphorescence decay of Pt Ar 4 TBP, Pt DPTBP and Pt Ar 2 TBP in their corresponding NIR OLEDs operated at current density of 10 mA/cm 2 The experimental measured curves are shown in red lines and the fitting curves are shown in blue lines. The fitting results are also shown in the inse t. The unit for 1 2 and avg are s, and A 1 and A 2 are unitless.

PAGE 145

145 Figure 5 14. (color) Phosphorescence decay of Pt TAr 2 TBP and Pt Ar 2 OPrTBP in their corresponding NIR OLEDs operated at current density of 10 mA/cm 2 The experimental measured curves are shown in red lines and the fitting curves are shown in blue lines. The fitting results are also shown in the inset. The unit for 1 2 and avg are s, and A 1 and A 2 are unitless. The disagreement between solution photophysical data and device effciencies for Pt TAr 2 TBP and Pt Ar 2 OPrTBP can also be explained by the EL lifetime measurement results As shown in Fig ure 5 14, Pt TAr 2 TBP exhibit a much lower EL lifetime of 1 9. 2 s compared to the Pt Ar 2 TBP, consistent with the much lower device efficiencies for Pt TAr 2 TBP based OLEDs. The reason for the significant decrease in EQE from Pt Ar 2 TBP to Pt TAr 2 TBP is not known; it is possible that these compounds were decomposed to some extent in the thermal vapor deposition process (e.g., perhaps by loss of one or m ore of the aryl substituent groups). Pt Ar 2 OPrTBP has the EL lifetime of

PAGE 146

146 23.6 s also lower than Pt Ar 2 TBP The comparison of the EL lifetimes well matches the difference between the OLEDs based on these two molecules. The EL lifetimes measured at current density of 10 mA/cm 2 and maximum EQE for the NIR OLEDs based on the series of Pt TBPs are summarized in Table 5 2. 5.6 Aggregation Effect of Pt TBPs on EL Efficiency Previously in Chapter 4, we have studied the NIR OLEDs with different doping concentration based on donor acceptor donor oligomers. The doping concentrations of those emitters need to be appropriately adjusted to achieve the balance between the maximum energy transfer and minimum aggregate quenching. One way to reduce the aggregation of dopants is to change the stacking and interaction of dopant molecules. For some mole cules in the series of Pt TBPs, for example, Pt Ar 4 TBP, bulky end groups are added in place of the meso aryl groups on the benzoporphyrin ring periphery compared to the non substituted Pt TPTBP These bulky end groups can prevent the stacking interact ions of molecules and therefore molecular aggregation when in the solid state films and devices The aggregate effect of Pt TPTBP and Pt Ar 4 TBP based NIR OLEDs is studied by varying the doping concentration of these two Pt TBPs in devices from 2% to 12%. Except for the doping concentration of NIR emitters, the device structure for both molecules is exactly the same as that used in previous sections of this chapter. The maximum EQE for the devices with different doping concentrations are shown in Figure 5 1 5. The curves for both Pt TPTBP and Pt Ar 4 TBP based devices exhibit the same trend that the EQE of devices decrease with the increasing doping concentration of doped Pt TBPs, which is obviously due to the aggregate quenching as we discussed.

PAGE 147

147 Figure 5 15 External quantum efficiency, EQE of OLEDs based on t wo Pt TBPs: Pt TPTBP ( black squares ) and Pt Ar 4 TBP ( red triangles) as a function of doping concentration, x ( x = 2%, 4%, 8 % and 1 2 % ), of these NIR e mitters. But the decay rates, which can be represented by the slopes of the efficien cy curves, are different between the Pt TPTBP and Pt Ar 4 TBP based OLEDs, indicating the different extent of aggregation effect between these two molecules in devices. For example, as the doping concentration is increased from 2% to 12%, the EQE s of Pt Ar 4 T BP based devices drops from 9.3% to 7.9%, approximately 15% decrease, but Pt TPTBP based devices show 35% decrease of EQE s from 8.2% to 5.3%, more than two times higher than that of Pt Ar 4 TBP based devices. Especially, when the doping concentration is incr eased from 8% to 12%, the EQE for Pt Ar 4 TBP based devices almost stay the same with less than 5% efficiency drop off while Pt TPTBP based devices still show significant decrease of EQE by approximately 20%. This means, by adding the bulky end groups in pla ce of the meso aryl groups on the benzoporphyrin ring periphery the bulkier Pt Ar 4 TBP exhibits much less significant aggregate than Pt TPTBP in the devices.

PAGE 148

148 5. 7 Summary In this chapter, extended Pt porphyrins, including the novel Pt Ar 4 TAP derivative, into NIR OLEDs. Through the system to Pt Ar 4 TAP the electroluminescence was shifted to 1005 nm, which is the long est wavelength demonstrated to date for a triplet phosphor in an electrolumines cent device. The photophysic al extended Pt porphyrins were compared, revealing decreasing solution em and device system i s extended and the emission is bathochromically shift ed from 770 nm to 1005 nm. The photophysics of a series of variously substituted Pt tetrabenzoporphyrins were characterized both in solution and solid state and compared with their performance in OLEDs Although relatively large differences in em and em between di and tetra substituted Pt TBPs were observed in solution, the difference in em was barely found in host matri x of devices. The results of this study clearly demonstrate that the large dif ferences observed for the solution em do not directly correlate with the performance of the chromophores in devices. It was also found that the addition of 3,5 di tert butylphenyl groups in place of phenyl groups on the benzoporphyrin ring periphery resu lts in increased device efficiency; however, further increasing the size of the substitutents to bulkier groups (such as terphenyl ) does not improve the device performance. Although the efficiency improvements obtained with the di substituted Pt benzoporp hyrins were not as high as predicted by their solution em values, the NIR OLEDs based on the series of di substituted Pt TBPs still achieve very high efficiencies with maximum EQE up to 7 8% in the near IR emission range. A record high EQEs

PAGE 149

149 were finally obtained by OLEDs based on Pt Ar 4 TBP emitting at the peak wavelength of 773 nm with EQE of 9.2%. Finally, the aggregate effect of Pt TBPs on EL efficiency is studied based on the two Pt porphyrin molecules: Pt Ar 4 TBP, which has bulky end groups added in pl ace of the meso aryl groups on the benzoporphyrin ring periphery and the non substituted counterpart, Pt TPTBP. By comparing the efficiency decay rate with the increasing doping concentration of the two emitters, we find that the bulkier Pt Ar 4 TBP exhibit s much less significant aggregate than Pt TPTBP in the devices.

PAGE 150

150 CHAPTER 6 ULTRA VIOLET ORGANIC LIGHT EMITTING DEVICES 6.1 Introduction to UV OLEDs While research efforts remain strong to develop organic materials that emit visible light, we have shown the promising results of efficient OLEDs with near IR emi ssion based on fluorescent donor acceptor donor oligomers and phosphorescent Pt porphyrins with record high EQE up to 9.2% and emission peak wavelength over 1 m. Another interest has recently turned t o organic systems that emit at the other end of the visible range, the shorter wavelengths from the deep blue, through the violet, to the ultraviolet regions. Ultraviolet (UV) to deep blue organic light emitting diodes (OLEDs) have found, or are sought f or, applications in biological and chemical sensing, 90,91 sterilization, high density information storage devices, 92 and full color light emitting displays. 93,94 Recently, there have been several reports 95 104 of high efficiency fluorescent blue to violet OLEDs with a peak emission wavelength in the range of 4 0 0 4 8 0 nm that possess maximum EQE values up to 3 6 %, although the shorter wavelength emission generally leads to lower device efficiencies. Despite a number of examples of deep blue to UV emitting OLEDs based on organ ic fluorescent emitters including small molecules 101 108 and polymers, 99,100,109,110 only a few 106,108 realize the device emission peak wavelength s below 400 nm, which can be considered as the UV emission, with external quantum efficiencies (EQ E ) greater than 1%. However, these so called UV OLEDs still have relatively broad emission peaks, with emission tail extending as far as to the longer wavelength ranges of 450 550 nm. Table 6 1 lists some of the existing blue violet UV OLEDs in the litera ture. The challenges of fabricating UV OLEDs lie in two parts: one is the molecular design of organic materials

PAGE 151

151 capable of efficient UV emission, 106 and the other is appropriately designed device structure, 108,111,112 where both hole and electron carriers are injected into a wide energy band gap UV emitting material and non UV emissions from longer wavelength emitting materials need to be prevented. So here in this dissertation, h ighly sought are organic m aterials that can exploit the donor acceptor concept 105,107 to achieve emissive color tunability and provide high efficiency and brightness in the UV to violet region and device structures with appropriate energy levels alignment. Table 6 1. List of some existing blue (B), blue violet (B V) and ultraviolet (U V) OLEDs based on different emit ters including small molecules and polymers. Emitter s max em (nm) Claimed Emission QY Efficiency (%) Ref. DO PPP 425 B 0.85 2.0% 99 PPP 424 B V 1.2% 100 TPD p BPD 404 415 B V 0.68 0.78 1.4 % 1.2% 101 p TTA 435 B V 2.8% 102 F 2 PA PFFA 405 422 B V 0.56 0.76 2.0% 1.1% 103 Bifluorene 374,392 UV 0.70 3.6% 106 2SBFN 381 UV 0.70 (film) 2.0% 108 indoles 443 B 0.59 0.87% 104 6.2 D onor Acceptor Purines 2 Aminopurine (2 AP), an isomer of adenine, remains perhaps the simplest functional nucleobase variant that is also fluorescent; a high quantum yield (68% in water for the riboside 196 ) underlies its widespread usage as an optical probe 197 and continued mechanistic and structural study. 198 200 Even so, molecules based on this platform have sparsely filtered into traditional organic materials and related sensing applications 201,202 and in surprisingly few cases has their optimization and br oader photophysical evaluation toward this goal been performed. 203

PAGE 152

152 Figure 6 1. Gen eric structure of the donor acceptor purines (D = donor, A = acceptor). Atoms have been numbered in the conventional way around the purine core. The organic framewo rk considered here is that of the donor acceptor purine, a heterocycle central to DNA/RNA structure that can be made highly fluorescent upon the judicious placement of electron donor an d acceptor groups on its rings 203,204 as shown in Fig ure 6 1 Sign ificant photophysical changes occur upon introduction of acceptor substituents to the purine C(8) position that complement typical donor groups at C(2) and C(6) positions Recently prepared derivatives have displayed fluorescence quantum yields over 80% an d emission wavelengths that can be tuned from 350 to 450 nm in solution, 203,204 suitable starting parameters for blue violet OLED construction. The interest in using purines for electronic device development is fu rther derived from their rich chemical structure that underlies molecular recognition and charge transport 205 surface tends to facilitate aromatic stacking and can allow access to different hydrogen bonding patterns along their heteroatom lined edges. 206 These structural features have already been shown important to the ordering of simple nucleobases in the solid state 207 where they are relevant to de vice performance. 208,209 Hardly considered to date have been emissive purine analogs in solid state devices, such as OLEDs. As a first example, efficient blue violet emitting OLEDs based on one type of donor accept or purine, methyl 9 benzyl 2 N,N dimethylamino 9 H purine 8 carboxylate

PAGE 153

153 1 ( Figure 6 1) are first demonstrated in this chapte r. 210 A maximum EQE of 3.1% is achieved in device s with the purine emitter dispersed in a N,N dicarbazolyl 3,5 benzene (mCP) host matrix as the emissive layer. The electroluminescence (EL) of the OLED has a peak emission wavelength of = 430 nm Another donor acceptor purine derivative, methyl 2 amino 9 benzly 6 (dimethylamino) 9 H purine 8 carboxylate 2 (Fig ure 6 1), which has even shorter PL emission peak wavelength below 400 nm, is also used for OLEDs device demonstration. Different from the visible light and near IR emitting materials, the blue violet UV emitting materials usually have v ery wide energy band gap (>2.6 eV), so i t is very important to find out the appropriate organic materials for multilayer in devices including hole transporting layers (HTLs) or electron blocking layers (EBL), emissive layers (EMLs) and electron transportin g layers (ETLs) or hole blocking layers (HBLs), to well align the energy levels between different layers in the devices when the emitters are incorporated. 108 Emission from single layers such as HTLs or exciplex emissions from interfaces between organic layers may be observed if the electrons or holes, respectively, are not effectively injected and confined within the EMLs. These issues indicate the importance of choosing the appropriate organic materials for a multilayer OLEDs, in order to obtain the desired EL emission and high efficiencies. Based on the UV emitter, Purine 2 we study the effect of device structures by utilizing different HTL s ETL s, EMLs and insertion of thin buffer layer to the device s for the extraction of violet to UV emission from OLEDs. By using the optimized device structure with appropriate energy level alignment among multi layer s a maximum EQE of 1.5 % is achieved in the devices with a peak emission wavelength below 400 nm, at = 393 nm

PAGE 154

154 This chapter is organized as follows. The experimental details on optical characterization, electrochemistry device fabrication and characterization are described in section 6 3 We present the experimental results and discussion on the opti cal properties and performance of the OLEDs for Purine 1 in section 6 4 In section 6 5 we study the effect of device structure by utilizing different HTL s ETL s EML s and insertion of thin buffer layer, for achieving the short wavelength emitting OLEDs b ased on purine 2 Finally, we conclude in section 6 6 Table 6 2 Photophysical data (absorption and emission maxima, max abs and max em and quantum yields, em ) for purine 1 and 2 measured in CH 2 Cl 2 Purine max abs (nm) lo (M 1 cm 1 ) em max em (nm) 1 362 4.1 0.81 433 2 330 4.2 >0.95 393 6.3 Experimental Details The synthesis of purine 1 and 2 has been reported pre viously. 203 The purine molecules used in this work are synthesized by Pamela Cohn in Dr. Ronald Castellano group in the Department of Chemistry at the University of Florida. The photophysical data including the absorption and emission maxima and quantum yield s in solution, for these two purines are listed in Table 6 2 Solution steady state fluorescence emission spectra were recorded on a SPEX Fluoromax spectrophotometer. The optical density was less than 0.1, and the sample concentration ranged from 5 S olid state films were prepared by vacuum thermal evaporation on quartz substrates. Photoluminescence spectra for the films were obtained choosing certain excitation wavelengths that usually corresponded to the absorption maxima of test materials as recor ded with a JASCO FP 6500 spectrofluorometer.

PAGE 155

155 Cyclic Voltammetry (CV) and Differential Pulse Voltammetry (DPV) measurements were performed in an Ar atmosphere glovebox maintained with < 4 ppm trace oxygen and moisture Tetrabutylammonium hexafluorophosphate (TBAPF 6 ) was purchased from Aldrich and used as received. Dichloromethane (DCM) was passed through a MBraun solvent purification system that renders the solvent dry on retrieval. The purine solid was dissolved to a concentration of 4.5 mM in a 0.1 M TBA PF 6 /DCM electrolyte. The working reference, and counter electrodes were a Pt button (0.02 cm 2 ), a silver wire pseudoreference calibrated with the ferrocene/ferrocenium (Fc/Fc + ) redox couple, and a platinum flag, respectively. The above PL spectra of purin e molecules in solution and CV/DPV characterization are measured by Aubrey Dyer in Dr. John Reynolds group in the Department of Chemistry at the University of Florida. The multilayer OLEDs were fabricated on glass substrates commercially precoated with an indium tin oxide (ITO) anode with a sheet resistance of The substrates were cleaned and processed by the standard procedure described in previous chapters. All the layers, including the cathode, were deposited using vacuum thermal evaporation following procedures published previously 39 T he organic layers in the devi ce structure usually consists of hole transporting layers (HTLs), emissive layer and electron transporting layers (ETLs) in sequent The HTLs used in this chapter include 1,1 bis[(di 4 tolylamino)phenyl]cyclohexane (TAPC) 4,4 ,4 tris(N 3 methylphenyl N p henyl amino)triphenylamine (m MTDATA), 4,4 ,4 tris(carbazol 9 yl)triphenylamine (TcTa), and NPD. mCP is used as the host material. For some deivces, neat emitter layers are directly used, sandwiched by HTL and ETL layers,

PAGE 156

156 without mixing into the host ma trix. The ETLs used include 1,3 bis[2 (4 tert butylphenyl) 1,3,4 oxadiazo 5 yl]benzene (OXD 7) 2,2 ,2 (1,3,5 benzinetriyl) tris(1 phenyl 1 H benzimidazole) (TPBi), 3 (4 biphenylyl) 4 phenyl 5 tert butylphenyl 1,2,4 triazole (TAZ), BPhen, and 3TPYMB. The thickness of the HTL and ETL were both 40 nm, whereas the EML was 20 nm consisting of either the mCP host doped with emitters at different doping concentrations or the neat emitter layers. A 5 nm thick p bis(triphenylsilyly)benzene (UGH2) 37 is inserted between host and ETLs in some devices as the buffer layer. A 1 nm thick layer of LiF followed by a 50 nm Al layer was the n deposited as the cathode. The deposit i on rates for the organic layers were 0.1 0.2 nm/s, as monitored by quartz crystal microbalances. The active device area was 4 mm 2 with the electrodes arranged in a cross bar geometry, and each substrate featured f our independently addressable device pixels. TAPC, m MTDATA, TcTa, mCP OXD 7 TPBi, TAZ 3TPYMB and UGH2 were all purchased from Luminescence Technology Corp BPhen from TCI America, and NPD from e Ray. All organic materials were used as obtained excep t for NPD and BPhen, which were purified in house using vacuum gradient sublimation method for one to two cycles. 14 The HOMO/LUMO energ ies for all the materials above are taken from the literature 39,43,108,211,212 whereas the HOMO/LUMO energies for the purines are determined experimentally using CV and DPV Radiant emittance ( R ) current density ( J ) voltage ( V ) characteristics of the OLEDs were measured under ambient conditions using an Agilent 4155C semiconductor parameter analyzer and a calibrated silicon detector ( Newport 818 UV) EL spectra and the Commission Int ernationale de L Eclairage (CIE) coordinates were

PAGE 157

157 taken using an OceanOptics HR400 0 spectrometer with the device s driven at a constant current using a Keithley 2400 source meter. The radiant emittance was calibrated assuming Lambertian emission, and the E QE ( EQE wall plug power efficiency ( P ) were derived based on the recommended methods 158 The organic layer thickness nonuniformity across the samples and run to run variations typi cally lead to 5 10% error in device efficiencies. 6.4 Blue Violet OLEDs based on Purine 1 Figure 6 2 shows the photoluminescence (PL) spectra of 1 in solid state film with 5% doping concentration in mCP. The films are prepared by vacuum thermal evaporation on quartz substrates and the film thickness is 20 nm. Independent PL spectra for the films were obtained using excitation wavelengths that correspond to the absorption maxima of mCP ( ex = 292 nm) and purine 1 ( ex = 362 nm). Compared to the PL spectr a of neat mCP film excited at ex = 292 nm shown in Figure 6 2 the purine 1 doped mCP film shows the red shifted emission peak at em = 432 nm when also excited at 292 nm, which can be at tributed to the emission from purine 1 This result indicates that Frster energy transfe r 114 of excitons occurs from the mCP host to the purine dopant molecules It can be further proved by the PL spectr a of the same doped film excited at ex = 362 nm a wavelength at which purine 1 has maximum but mCP has negligible absorption The PL emission of doped film excited at the absorption maximum of purine 1 in spite of the lower intensity due to low composition of purine 1 in the film, is iden tical to that excited at the absorption maximum of mCP as shown in the inset of Figure 6 2. Additionally, t he PL spectra of purine 1 in solution and the solid state show identical emission maxima ( em = 432 nm) suggesting that the p urine molecules are

PAGE 158

158 wel l dispersed in the mCP host matrix and do not exhibit aggregati on induced red shifted emission Fig ure 6 2. P hotoluminescence spectra of purine 1 in solid state film (5 wt% in mCP) when excited at ex = 292 nm (solid line) and 362 nm (dashed line). PL s pectrum of mCP film (excited at ex = 292 nm ) is also shown for comparison. Inset: The molecular structure of purine 1 and normalized PL spectra of purine in solid state films (5 wt% in mCP) when excited at ex = 292 nm (solid line) and 362 nm (dashed line ). PL spectrum of purine 1 in CH 2 Cl 2 solution (dashed dotted line ; ex = 320 nm) is also shown for comparison. Most encouraging is that while t he PL quantum yield for purine 1 in organic solution (81% in CH 2 Cl 2 Table 1 ) is high relative to violet emitter s reported previously 101,104,106 it is not the highest reported for donor acceptor purines. 203 Likewise, it is known that small changes to the donor and acceptor groups of purine 1 can tune its emission across the blue UV region 203 potentially suitable for a variety of OLED applications The energies of the highest occupied molecular orbital (HOMO) and lowest occupied molecular orbital (LUMO) for purine 1 were determined experimentally using

PAGE 159

159 cyclic voltammetry (CV) and differential pulse voltammetry (DPV) This knowledge has guided the choice of charge transport and injection layers for proper energy level alignment with the active material to maximize OLED efficiency. The E 1/2 for oxidation and reduction (vs. Fc/Fc + ) were measured as 0.97 V and 2.24 V, respectively. The HOMO and LUMO energy values were then calcul ated from the electrochemical data as 6.07 and 2.86 eV, respectively, considering that Fc/Fc + = 5.1 eV relative to vacuum. 213 The corresponding HOMO LUMO gap en ergy (3.2 eV) lies between those reported previously 203 from optical absorption data (3.0 eV) and electronic structure calculations (3.88 eV at the B3LYP/6 311++G** level). Figu re 6 3. (color) Schematic energy level diagram of the OLEDs using purine 1 as the emitter with different ETLs. The energies (in eV) for the highest occupied and lowest unoccupied molecu lar orbitals (HOMO and LUMO, respectively) of the organic materials and the Fermi levels of the electrodes are referred from the vacuum level. The dashed lines in the mCP layer correspond to the LUMO and HOMO levels of purine 1 The schematic energy level diagram of the multilayer OLED structure based on purine 1 is shown in Fig ure 6 3 We first study the effect of different ETLs o n the OLEDs based on purine 1 With a constant 3% doping concentration in mCP, the ETLs with the thickness of 40 nm are varied from OXD 7, 3TPYMB to BPhen 43,214

PAGE 160

160 Figure 6 4. (color ) A ) Normalized electroluminescence (EL) spectra of the purine 1 based OLEDs with OXD 7 (red), 3TPYMB (blue), and BPhen (green) as ETLs. B ) Current density J and radiant emittance in the forward viewing directions, R as functions of voltage, V for t hese three devices. C ) External quantum efficiency, EQE and power efficiency, P as functions of current density, J for these three devices.

PAGE 161

161 The EL spectra of these three devices show almost the same emission peaked at approximately 430 nm, as shown in Fig ure 6 4 A consistent with the PL emission in both solution and solid state films. The current density and radiant emittance of these devices are shown in Fig ure 6 4 B The current densities of devices decrease from BPhen to 3TPYMB and to OXD 7, 43,214 which can be attributed to the decreasing conductivity of these three ETL mate rials. Accordingly, the radiant emittances of these three devices follow the same trend as current densities. But a maximum radiant emittance is achieved by OLEDs with the least conductive OXD 7, due to the higher saturation voltage of the OXD 7 device com pared to the other two devices. Here, the actual optical power density of the OLED emission is reported instead of the more commonly used luminance, as the latter incorporates the human eye sensitivity and is intended for display and lighting applications. Moreover, the luminance value of the OLED is dominated by the longer wavelength (blue and green) components in the emission spectrum, and does not clearly present the contributions from the shorter wavelength (UV to violet) emissions, the focus of this wo rk. The OXD 7 device also shows the highest EQE among these three devices as shown in Fig ure 6 4 C This is probably due to the lower electron injection barrier for host/ETL interface (Fig ure 6 3) and optimal charge balance within the device when using OXD 7 as the ETL. 215 Therefore, in th e following work we will use OXD 7 as ETL for further device performance optimization. The OXD 7 ETL layer thickness is then optimized considering the further charge balance optimization and microcavity effect as mentioned in the previous chapters. As the thickness of OXD 7 increases from 30 to 40 nm, both the EQE and power

PAGE 162

162 efficiencies are improved, while further increas e of the thickness to 50 nm will lead to the decrease of device efficiencies as shown in Fig ure 6 5. Therefore, the optimal OXD 7 thickne ss in the devices is determined to be 40 nm. Figure 6 5. (color) External quantum efficiency, EQE and power efficiency, P as functions of current density, J for the purine 1 based devices with various OXD 7 ETL layer thicknesses Figure 6 6. Maxim um e xternal quantum efficiency, EQE and power efficiency, P of OLEDs ( with 40 nm thickness of OXD 7 ETL ) as functions of the doping concentration x of purine 1

PAGE 163

163 The maximum EQE and power efficiencies of OLEDs with various doping concentraions of puri ne 1 in the mCP host matrix are shown in Fig ure 6 6. The OXD 7 thickness in these devices is maintained at 40 nm. Both the EQE and power efficiency curves show the same trend; i.e., the efficiencies increase with doping concentration of purine 1 for 1%< x <4 %, and then decrease with further increase of x to 6%. So the doping concentration of purine 1 in OLEDs is optimized as 4%. Figure 6 7. External quantum efficiency, EQE of the optimal purine 1 based OLED as a function of the current density, J Inset: Current density, J and the radiant emittance, R as functions of the voltage, V for this device. Current density voltage ( J V ) and radiant emittance voltage ( R V ) c haracteristics of an OLED based on purine 1 with optimized purine 1 doping concentration of 4% and OXD 7 ETL thick ness of 40 nm, are shown in the set of Figure 6 7 The turn on voltage of this device, which is defined as the operating voltage that yields d etectable light emission (in this case it means R > 10 5 mW/cm 2 ), is approximately V T = 2.9 0.1 V. Note that the photon energy correspon em = 432 nm) is 2.9 eV. This suggests that there is negligible energy loss to induce the EL from purine in

PAGE 164

164 this OLED device architecture. The light output increases drastically with the operating voltage at V > V T and a radiant emittance in the range of 10 to 18 mW/cm 2 is achieved at V > 15 V Figure 6 7 shows the dependency of EQE on the current density. The device has a maximum EQE of (3.1 0.3)%, a value that approaches the highest efficiency OLEDs with p eak emission appro aching 440 nm 16 18 The power efficiency for the device reaches a maximum of (23 3) mW/W at low current densities (10 3 < J < 10 2 mA/cm 2 ), and is reduced to (13 2) mW/W at J = 1 mA/cm 2 mostly due to the increase in operating voltage. Figure 6 8. Nor malized EL spectrum for the OLED based on purine 1 at J = 1 mA/cm 2 Inset: Device picture with four pixels in one substrate all lighten ed up. As shown in Figure 6 8, the EL spectrum of the device measured at J = 1 mA/cm 2 identical to the PL emission obtained in solid state films. The Commission Internationale The emission p eak of = 430 nm falls into the violet spectral range (380 450 nm); however, we do recognize that the emission from these OLEDs does contain a significant blue component (440 490 nm) and even longer wavelengths. As a result, we define the electroluminesce nce

PAGE 165

165 from the purine based OLEDs as blue violet emission. Considering that the human eye has a much higher photopic response to blue (and even higher to green) than to violet light, overall the emission of these purine based OLEDs appears to be blue to our eyes, as shown by the device picture in the in set of Fig ure 6 8. Figure 6 9. (color) Electroluminescence intensity decay of optimal purine based OLEDs driven at a constant current density of 1 mA/cm 2 in air. The operational voltage change with the consta nt current density is also shown. The operational stability of the purine based OLEDs is also tested in air by measuring the electroluminescence intensity decay of the OLEDs driven at a constant current density of 1 m A/cm 2 As shown in Fig ure 6 9, t he devi ce shows a half intensity life time of about 500 seconds, but the decay appears to slow down after an initial burn in period. For example, from 200 seconds to 1200 seconds (an interval of 1000 seconds), the EL intensity is reduce d by approximately 50%. Th e relatively low device lifetime may be due to the chemical stability of the purine, and we also believe that potentially useful device lifetimes should be achievable if we employ good encapsulation for these devices and/or optimization of device structure as has been accomplished in the commercial sector for visible light emitting OLEDs. The operational

PAGE 166

166 voltages are also gradually increased with the device operation time, from the initial voltage of 7.2 V to 7.6 V after 20 min continuous operation, probabl y due to the degradation of devices. 6.5 Violet UV Emitting OLEDs based on Purine 2 With different donor groups connected to the heterocycle ring, another donor acceptor purine 2 exhibits further shorter emission wavelength peaked at 393 nm and very high q uantum yield close to unity in the solution (Table 6 1 ). Figure 6 10 shows the photoluminescence (PL) spectra of 2 in solid state films (as a dispersion in mCP or neat film ) Doped s olid state films were prepared by vacuum thermal evaporation and consist ed of a 20 nm thick mCP layer doped with 5% (by weight) purine 2 Independent PL spectra for the films were obtained using excitation wavelengths that correspond to the absorption maxima of mCP ( ex = 292 nm) and purine 2 ( ex = 3 30 nm). Figure 6 10. P h otoluminescence spectra of purine 2 in solid state films with 5 wt% in mCP when excited at ex = 292 nm (solid line) and 330 nm (dashed line), and neat purine 2 film excited at ex = 330 nm (dashed dotted line). Inset: The molecular structure of purine 2

PAGE 167

167 The PL spectra for these doped films show identical emission maxima with tha t in the solution ( em = 393 nm) suggesting that the p urine molecules are well dispersed in the mCP host matrix and do not exhibit aggregation induced red shifted emission. Addit ionally, the two doped solid state film shows the same PL emission em = 393 nm) whether excited at the absorption maxim um of mCP ( ex = 292 nm) or the absorption maxim um of purine 2 ( ex = 3 30 nm, a wavelength at which mCP has barely absorption). This re sult indicates that Frster energy transfe r 114 of excitons occurs from the mCP host to the purine dopant m olecules But for the solid state film with neat purine 2 a significant red shift over 70 nm relative to the solution PL emission peak has been observed, which indicates that the purine 2 have very strong inter molecular interaction, leading to aggregatio n induced red shifted emission. Figure 6 11. (color) EL spectr um for the OLEDs based on neat purine 2 as EML, measured at various current densities from J = 1 0 mA/cm 2 to 375 mA/cm 2 The OLEDs using a 20 nm thick neat purine 2 as the EML show similar emis sion with solid state films, both with largely red shifted emission peak wavelengths at 470

PAGE 168

168 nm, as shown in Fig ure 6 11. The devices have TAPC and BPhen as HTL and ETL, respectively, both with thickness of 40 nm. The emission of this device obviously falls into the range of blue emission instead of violet as predicted by the solution PL data. As suggested by PL emission of both doped and neat solid state films, the purine 2 must be dispersed into host matrix in the devices to prevent the aggregation induce d red shifted emission We still use mCP as host material doped with 5% purine 2 molecules as EML for the OLEDs, as the previous film PL emission results have shown efficient Frster energy transfe r of excitons from the mCP host to the purine dopant molecu les ( Figure 6 10) Figure 6 12. Normalized EL spectra for the OLEDs based on purine 2 doped EML (doped device, solid ) and neat mCP EML (control, dashed ). HTL and ETL for these devices are TAPC and BPhen, respectively. The combinations of a series of HTL / ETL materials including TAPC/BPhen, TAPC/TAZ, TAPC/OXD 7, TAPC/3TPYMB, NPB/BPhen, m MTDATA/BPhen and PEDOT:PSS/OXD 7 are used for the OLEDs with doped EML. Unfortunately, no emission of purine 2 has been observed for all those devices. F or example, the E L spectra of the OLEDs with TAPC and BPhen as HTL and ETL, respectively, are shown

PAGE 169

169 in Fig ure 6 12. The control device, which contains only neat mCP in the EML without purine is also made for comparison. T h e doped EML device shows almost the same spectra a s the control device both with emission peak wavelengths of 410 nm suggesting there is no contribution of emission from purine 2 molecules for the device with purine 2 doped EML. This same emission from both doped and control devices can be attributed to the exciplex emission 216 218 at the interfaces of EML and ETLs. The exciplex emissions at the interfaces are also observed when other ETL materials are used, with peak wavelength s of approximately 400 410 nm as sh own in Figure 6 13. The energy differences at mCP/ETL interfaces, which are the energy gaps between the HOMO level of mCP and LUMO levels of ETLs, can be calculated of approximately 2.9 eV, 3.1 eV, and 3.2 eV for BPhen, OXD 7 and TAZ based devices, respect ively ( Figure 6 3). So the exciplex emissions at the interfaces will then gradually blue shi ft from BPhen to OXD 7, and TAZ based devices, which is in good agreement with the trend shown in Fi gure 6 13. Figure 6 13. (color) Normalized EL spectra for the OLEDs based on purine 2 doped EML with different ETLs (ETLs = BPhen, OXD 7 and TAZ). TAPC is used as the HTL for these devices The spectra are only shown for the range of 330 530 nm for better comparison of peak shift.

PAGE 170

170 Figure 6 14. (color) Normalized EL spectra for the OLEDs based on purine 2 doped EML with different HTLs (HTLs = NPD and m MTDATA). BPhen is used as the ETL for these devices. Another possible emission path is the exciton emission of HTL layers, as shown in Figure 6 14, for OLEDs with doped EML and BPhen as ETL. The similar emissions peaked at 430 440 nm for both NPD and m MTDATA based devices are due to the emission from the corresponding HTLs, 219,220 while TAPC based devices do not show this emission The layer of NPD has 0.4 eV low er LUMO level than TAPC as shown in Figure 6 15, and there is almost no energy barrier for the electron injection from mCP host to NPD. So the relatively large amount of injected electrons in NPD recombines with holes in NPD, giving out the emission of NPD. For m MTDATA based devices, the hole injection barrier from m MTDATA to mCP host is much larger than the TAPC based devices, while the electron injections from host to HTL are almost the same for both devices (Figure 6 15). So the holes are continuously injected and accumulate in the layer of m MTDATA due to the high energy barrier at HTL/host interface. These accumulated holes then have greater chances to recombine with the injected electrons in m MTDA TA layer and lead to the emission of m MTDATA. The tail

PAGE 171

171 emission extended to as far as 600 nm in both NPD and m MTDATA based devices can be probably attributed to the exciplex emission at the HTL/host interface. Figu re 6 15 (color) Schematic energy level diagram of the OLEDs using purine 2 as the emitter with different HTLs (HTLs = TAPC, NPD and m M TDATA). The HOMO and LUMO of purine 2 are not indicated here The exciton formation is usually more favorable when the electron s and holes binded with lower energy In the above purine 2 based devices, due to its high energy band gap which corresponds to t he emission peak of 393 nm, the formations of exciplex at host/ETL interfaces (corresponding to the emission peak over 400 nm) and excitons of HTLs (corresponding to the emission peak at 430 440 nm) are then more favorable compared to the formation of exci tons of purine 2 for the variable ETLs and HTLs devices, respectively. As a result, the devices will emit through the recombination of these favorable states instead of the purine 2 excited states. These results above suggest the importance of appropriate energy level alignment in the multilayer OLEDs with large band gap molecules as emitter All the functional layers, including hole and electron transporting layers and host matrix, need to be carefully chosen to facilitate the formation of stable excited states of emitter molecules, and at the same time avoid other possible competitive emission paths.

PAGE 172

172 Figure 6 1 6 (color) Normalized EL spectra for the OLEDs based on purine 2 doped EML (doped device, red solid line ) and neat mCP EML (control, black solid line ) with the insertion of thin layer of UGH2. HTL and ETL for these devices are TAPC and OXD 7 respectively. The PL emission of purine 2 doped mCP film (black dashed line) is also shown for comparison. Inset: the s chema tic energy level diagram of the purine 2 doped device The HOMO and LUMO of purine 2 are not indicated Based on the above discussion, TAPC is eligible for the hole transporting layer since no HTL emission is observed for all TAPC based devices (Figure 6 13). To prevent the exciplex emis sion at the EML/ETLs interfaces, a thin layer of UGH2 (5 nm) with very wide band gap is inserted between the host and ETLs (OXD 7). mCP is still used as the host material doped with 5% purine 2 Figure 6 16 shows the EL spectrum of this device as well as t he control device with only neat mCP in the EML and the PL emission of purine 2 doped mCP film for comparison. After the insertion of UGH2 thin layer, the

PAGE 173

173 OLEDs based on purine 2 show very similar emissions with the PL emission of purine 2 doped film both with emission peak at 393 nm. This result strongly indicates that the purine 2 emission is achieved by the device with UGH2 thin layer. Comparatively, the control device shows relatively blue shifted and very noisy emission and an additional peak at appro ximately 580 nm with an intensity of 15% of that of the primary peak. Figure 6 17 (color) A ) Current density, J and the radiant emittance, R as functions of the voltage, V for the OLEDs based on purine 2 doped EML (doped device, red triangles ) and ne at mCP EML (control, blue squares ) with the insertion of thin layer of UGH2 B ) External quantum efficiency, EQE of the se two OLED s as a function of the current density, J One more evidences for successfully obtaining the purine 2 emission in OLEDs with UGH2 thin layer is the comparison of radiant emittance ( R ) current density ( J ) voltage ( V ) characteristics a nd device efficiencies between the doped and control

PAGE 174

174 devices, as shown in Figure 6 17. These two devices show completely different R J V characteristics including the different turn on voltages, and different orders of magnitude of current densities and ra diant emittances. The EQE of doped device is also two times higher than the control device. All these differences suggest that the doped and control devices have completely different emission manners within the devices, therefore leading to the different E L emission spectra as shown in Figure 6 16. The small amount of doped purine 2 (5% by weight) may only slightly affect the current densities and radiant emittances if these two devices have the same emission mechanism, and the EQE should be not significant ly different between the two devices, which are obvious ly not the case as shown in Figure 6 17. The largely enhanced current density, radiant emittance and device efficiency indicate the doped device has a robust emission mechanism, which is greatly likely to be the Frster energy transfe r from the host to purine molecules The purine 2 based OLEDs, which employ the device structure of ITO/TAPC/mCP:5% Purine 2 /UGH2/OXD 7/LiF/Al, have a maximum EQE of EQE = (1.5 0.2)% achieved at J 1 mA /cm 2 and a maxim um radiant emittance of 5 mW/cm 2 at V = 17 V, as shown in Figure 6 17. The efficiency of purine 2 based OLEDs is two times lower than the purine 1 devices but the former devices shift the emission further towards the short wavelength range with the emissi o n peak below 400 nm, at = 393 nm 6. 6 Summary In this chapter, firstly, a blue violet electroluminescent device has been constructed featuring a highly fluorescent donor acceptor purine 1 in the emissive layer. Its defining parameters a maximum external quantum efficiency of 3.1%, a turn on

PAGE 175

175 voltage of 2.9 V, and peak emission at 430 nm are achieved by optimizing the doping concentration of purine in host matrix, and materials and thickness of electron transporting layer. These results are comparable to state of the art UV to violet emitting OLEDs and approach the highest performing deep blue OLEDs based on fluorescent organic compounds. With different donor groups connected to the purine heterocycle ring, another donor acceptor purine derivative 2 which exhibits further blue shifted emission with peak wavelength below 400 nm and very high quantum yield close to unity in the solution is also employed for violet UV emitting OLEDs. Due to the high energy band gap of purine 2 molecules, it is very important to appropriately alig n the energy levels of different layers in the multilayer devices to efficiently extract the desired light emission. So we study the effect of device structures by utilizing different HTL s ETL s, EMLs and insertion of thin buffer layer for purine 2 based O LEDs. By using the optimized device structure with appropriate energy level alignment among multilayer, the purine 2 based devices have a maximum external quantum efficiency (EQE) of 1.5 % two times lower than the purine 1 based devices, but the purine 2 b ased devices further shift the emission towards the short wavelength range with emission peak wavelength below 400 nm, very close to the real UV emission. Central to the purine based emissive component is a rich (and highly modifiable) chemical structure t hat is linked to broad functions in biology and emerging app lications in materials science. 221,222 Given that extension of the donor acceptor concept to the purine heterocycle has allowed highly efficient and tunab le fluorescence emission across the blue UV spectrum in solution, the results reported here bode well for

PAGE 176

176 continued OLED development using the molecules to design devices with tailored electroluminescence. The extent to which the molecular recognition func tionality of the purine (e.g., hydrogen bonding sites) can be used to modulate the EML and interface structure toward improved device performance will be a focal point of future studies. As the emission of OLEDs goes to the even shorter wavelength range, i t is more and more difficult to find out the appropriate organic materials for different functional layers such as HTLs, ETLs, EMLs and buffer layers to achieve energy level alignment for multilayer, since so far, most of available organic materials are de signed for the use of OLEDs emitting visible light, which corresponds to the relatively small energy band gaps compared to the UV emitters. So organic materials as use of the functional layers in violet UV OLEDs with specific requirements, such as larger e nergy band gap, reasonable HOMO/LUMO levels, and good electrical properties, need to be designed and synthesized in the future.

PAGE 177

177 CHAPTER 7 CONCLUSIONS AND FUTU RE WORKS 7.1 Conclusions 7.1.1 Near IR OLEDs based on F luorescent D onor A cceptor D onor O ligomers In C hapter 4 we demonstrate that low gap DAD oligomers are good candidate material s for use in OLEDs to achieve efficient NIR emission. A maximum external quantum efficiency of 1.6% and a maximum power efficiency of 7.0 mW/W were achieved in devices base d on BEDOT TQMe 2 whose emission peak s at 692 nm, but extend s to well above 800 nm. With a stronger acceptor and thus a reduced energy gap, longer wavelength NIR emissions peaked at 815 nm were achieved in BEDOT BBT based devices, although the efficiencies were approximately three times lower than the BEDOT TQMe 2 based devices, due to the significant ly lower fluorescent quantum yield of BEDOT BBT. 185 Using the sensitized fluorescence device structure, 154 the efficiencies were further increased by two to three times, and we achieved the maximum efficienc ies of EQE = 3.1% and P = 12 mW /W for BEDOT TQMe 2 based devices, and EQE = 1.5% and P = 4.0 mW/W for BEDOT BBT based devices. 186 The aggregate effect of the donor acceptor donor oligomers in the devices is also studied by utilizing different ho st matrix and changing the stacking manners of doping molecules. Changing from the previous host of Al q 3 to CBP, which has more spectral overlap with the DAD NIR emitters, the concentration of NIR emitters needed to allow complete energy transfer for excit ons on the host molecules will be decreased due to increased F rster radius, leading to less significant aggregate quenching effect. The NIR OLEDs based on BEDOT BBT doped into CBP host matrix has a maximum EQE of EQE = ( 0.92 0.07 )%, which is almost double that of the devices with Al q 3 as host. The

PAGE 178

178 power efficiency is also higher for the CBP device than the device with Al q 3 host with a maximum of P = ( 3.6 0. 3 ) mW/W for the former device The effect of molecul ar structure on aggregation in devices is then studied by adding bulky substituted groups to the two ends of DAD oligomers. However, the NIR OLEDs based on two bulky DAD BBT based devices. T he efficiencies of these bulky DAD oligomers based devices show no enhancement and the emissions are red shifted with peak wavelengths up to 905 nm. 7.1.2 Near IR OLEDs based on Phosphorescent Platinum (II) Porphyrins In C hapter 5 we have demonstrated the extended Pt porphyrins, including the novel Pt Ar 4 TAP derivative, into NIR OLEDs. 172,177,189 system to Pt Ar 4 TAP the electroluminescence was shifted to 1005 nm, which is the longest wavelength demonstrated to date for a triplet phosphor in an electrolumines cent device. The photophysic al and OLED data on this extended Pt porphyrins were compared, revealing decreasing solution em and device e system is extended and the emission is bathochromically shifted from ~770 nm to ~1005 nm. The photophysics of a series of variously substituted Pt tetrabenzoporphyrins were characterized both in solution and solid state and compared wi th their performance in OLEDs. Although relatively large differences in em and em between di and tetra substituted Pt TBPs were observed in solution, the difference in em was barely found in host matri x of devices. The results of this study clearly d emonstrate that the large differences observed for the solution em do not directly correlate with the performance of the chromophores in devices. It was also found that the addition of 3,5 di tert

PAGE 179

179 butylphenyl groups in place of phenyl groups on the benzo porphyrin ring periphery results in increased device efficiency; however, further increasing the size of the substitutents to bulkier groups (such as terphenyl ) does not improve the device performance. Although the efficiency improvements obtained with th e di substituted Pt benzoporphyrins were not as high as predicted by their solution em values, the NIR OLEDs based on the series of di substituted Pt TBPs still achieve very high efficiencies with maximum EQE up to 7 8% in the near IR emission range. A r ecord high EQEs were finally obtained by OLEDs based on Pt Ar 4 TBP emitting at the peak wavelength of 773 nm with EQE of 9.2%. 189 Finally, the aggregate effect of Pt TBPs on EL efficiency is studied based on the two Pt porphyrin molecules: Pt Ar 4 TBP, which has bulky end groups added in place of the meso aryl groups on the benzoporphyrin ring periphery and the non substituted counterpart, Pt TPTBP. By comparing the efficiency decay rate with the increasing doping concentration of the two emitters, we find that the bulkier Pt Ar 4 TBP exhibits much less significant aggregate than Pt TPTBP in the devices. 7.1.3 Ultra Violet Organic Light Emitting Devices In Chapter 6, a blue violet electroluminescent device has been constructed featuring a highly fluorescent donor acceptor purine in the emissive layer. 210 Its defining parameters a maximum EQE of 3.1%, a turn on voltage of 2.9 V, and peak emission at 430 nm are comparable to state of the art UV to violet emitting OLEDs and approach the highest performing deep blue OLEDs based on fluores cent organic compounds. With different donor groups connected to the purine heterocycle ring, another donor acceptor purine derivative 2 which exhibits further blue shifted emission with

PAGE 180

180 peak wavelength below 400 nm and very high quantum yield close to u nity in the solution is also employed for violet UV emitting OLEDs. Due to the high energy band gap of purine 2 molecules, it is very important to appropriately align the energy levels of different layers in the multilayer devices to efficiently extract t he desired light emission. So we study the effect of device structures by utilizing different HTL s ETL s, EMLs and insertion of thin buffer layer for purine 2 based OLEDs. By using the optimized device structure with appropriate energy level alignment amon g multilayer, the purine 2 based devices have a maximum external quantum efficiency (EQE) of 1.5 % two times lower than the purine 1 based devices, but the purine 2 based devices further shift the emission towards the short wavelength range with emission p eak wavelength below 400 nm, very close to the real UV emission. Central to the purine based emissive component is a rich (and highly modifiable) chemical structure that is linked to broad functions in biology and emerging applications in materials science Given that extension of the donor acceptor concept to the purine heterocycle has allowed highly efficient and tunable fluorescence emission across the blue UV spectrum in solution, the results reported here bode well for continued OLED development using the molecules to design devices with tailored electroluminescence. The extent to which the molecular recognition functionality of the purine (e.g., hydrogen bonding sites) can be used to modulate the EML and interface structure toward improved device perfo rmance will be a focal point of future studies. 7.2 Future Works 7.2.1 Near IR OLEDs of Longer Emission and Higher Efficiency Already seen in this dissertation, a s the energy band gaps get lower for near IR emitting materials (including both DAD oligomers and Pt porphyrins) the quantum

PAGE 181

181 yields and lifetimes will decrease, partially due to the increased non radiative decay rate in the lower gap molecules. This fact will, to some extent, prevent the further developmetn of high efficiency near IR OLEDs with ev en longer emission wavelength (> 1 m) Novel near IR emitters with long er emission wavelengths but still exhibiting high quantum yield may be found out. On the other hand, another limitation for the relatively low efficiencies of these NIR OLEDs is the low light outcoupling efficiency, whi ch is usually considered to be less than 20%. So there is still large improvement space for light outcoupling. Many methods such as the application of microlen s arrays have been reported to improve the light coupling for visible light emitting OLEDs 51 54 Those same methods may be also used for NIR OLEDs. Simulation need to be done first to realize the estimated enhancement. 7.2.2 Application of Near IR Emitters in OPVs These low band gap near IR emitting materials h ave the absorption in relatively long er wavelength range, which will have perfect spectra coverage with the solar spectrum. 223 225 So it is possible to use these materials in the organic photovoltaic (OPV) devices a nd tandem OPVs Standard OPV devices have been tried based on the two donor acceptor donor oligomers, BEDOT TQMe 2 and BEDOT BBT. 226 Figure 7 1 shows the J V characteristics in the dark and under simulated AM1.5 solar illumination for a bilayer cell with the structure of ITO/BEDOT TQMe 2 (200 )/C 60 (40 0 )/BCP(80 )/Al(800 ). The output parameters of this cell under illumination are summarized in the inset. The low short circuit current density ( J sc ) less than 0.01 mA/cm 2 and fill factor ( FF ) of 0.20, lead to the poor device power conversion efficiency ( PCE ) of 0.001%, probably due to the low conductivity of BEDOT TQMe 2 molecules.

PAGE 182

182 Figure 7 1. (color) Current density vs voltage ( J V ) characteristics of an organic photovoltaic cell with the structure of ITO/BEDOT TQMe 2 (200 )/C 60 (400 )/BCP(80 )/Al(800 ) in the dark (black) and under simulated AM 1.5G solar illumination 1 sun (red). The device output parameters including short circuit current density ( J sc ), open circuit voltage ( V oc ), fill factor ( FF ), power conversion efficiency ( PCE ) are listed in t he inset. Figure 7 2. (color) External quantum efficiency, EQE as a function of the wavelength for BEDOT TQMe 2 :C 60 bilayer cell. Inset: the schematic energy level diagram of this cell. The external quantum efficiency of this cell, as shown in Figure 7 2, is very low, less than 1% for the whole spectrum, and shows most of the contributions from C 60

PAGE 183

183 corresponding to the peaks at about 350 nm and 440 nm. The BEDOT TQMe 2 peak, which should appear at ~530 nm (Figure 4 2), can be barely seen, indicating t here is very little or probably no contribution of electrical output from BEDOT TQMe 2 Figure 7 3. (color) Current density vs voltage ( J V ) characteristics of an organic photovoltaic cell with the structure of ITO/BEDOT BBT(200 )/C 60 (400 )/BCP(80 )/Al (800 ) in the dark (black) and under simulated AM 1.5G solar illumination (red). The device output parameters including short circuit current density ( J sc ), open circuit voltage ( V oc ), fill factor ( FF ), power conversion efficiency ( PCE ) are listed in the inset. The DAD oligomer, BEDOT BBT, with smaller energy band gap and longer absorption wavelengths, is also used for the OPV devices. As shown in Figure 7 3, the cells based on BEDOT BBT exhibit improved device output than the BEDOT TQMe 2 with J sc and PCE greatly enhanced to 0.19 mA/cm 2 and 0.02%, respectively, and V oc and FF stay the same. The low FF for both oligomers based OPV devices suggests the charge collection in these two devices is not efficient. Since BEDOT BBT and BEDOT TQMe 2 have the similar conductivities (Figure 4 9), there should be some other reasons leading to the difference of the J sc for these two devices. One possible explanation is that the BEDOT BBT molecules have higher absorption coefficient and also more absorption overlap with so lar spectrum.

PAGE 184

184 Figure 7 4. (color) External quantum efficiency, EQE as a function of the wavelength, for BEDOT BBT:C 60 bilayer cell. Inset: the schematic energy level diagram of this cell. The EQE of BEDOT BBT based cells is higher than the BEDOT TQMe 2 achieving 1.5% at the maximum peak but still at very low level, as shown in Figure 7 4. Except for the contribution from C 60 at peaks of 350 nm and 440 nm, we can clearly find the BEDOT BBT peak at about 650 nm (Figure 4 2) with peak value of 0.7%. The a bove preliminary results of OPV devices based on the two DAD oligomers indicate that the molecular structures of DAD oligomers need to be carefully designed to have certain properties such as high conductivity, high absorption coefficient, appropriate ener gy band gap s and HOMO/LUMO levels, for the application of high efficiency OPV devices. And those properties can be easily tuned by connecting different donor and acceptor units for the DAD oligomers Furthermore, the highly phosphorescence Pt porphyrins in this dissertation can be dissolved in solvent or be thermal evaporated, 189 so both vapor thermal evaporation and solution processing methods can be used for OPV device fabrication. The conductivity

PAGE 185

185 might be an issue before applying to the OPV devices The cost will be another concern since some of the phosphorescent nea r IR emitters contain the expensive heavy metals. 7.2. 3 Ultra Violet OLEDs of Shorter Emisison We already realize the importance of appropriate energy level alignment for the multilayer devices, in order to extract the desiring light emission from the UV e mitters. In Chapter 6, we already successfully achieve the violet UV emitting OLEDs with peak emission wavelength below 400 nm, at = 393 nm However, UV emitting OLEDs with even shorter emission wavelengths are still needed. Figure 7 5 (color) Absorption (dashed red line) and PL spectra of t BuBP4M in the solution (dashed blue line) and solid state film (solid blue line) when ex cited at ex = 2 70. Inset: molecular structure of UV emitting t BuBP4M. Based on a new UV emitter, t BuBP4M, with PL emission peak wavelength at ~332 nm in the solution, as shown in Figure 7 5, we have tried to fabricate the UV emitting OLEDs. The PL emis sion measured in solid state film is also shown in Figure 7 5, in which ~30 nm red shift of PL emission peaks is observed, indicating that the t BuBP4M have very strong inter molecular interaction So the t BuBP4M needs to be dispered into host matrix to a void the aggregate induced red shifted emission.

PAGE 186

186 Figure 7 6. (color) Normalized EL spectra for the OLEDs with t BuBP4M doped in UGH2 host matrix with different doping concentrations (1%, 5% and 10%). TAPC and BCP are used as the HTL and ETL, respectively for these devices. We first use very wide band gap UGH2 (~4.4 eV) as the host matrix doped with t BuBP4M of different doping concentrations. TAPC and BCP are used as the HTL and ETL, respectively, in these devices. As shown in Figure 7 6, all the devices show no UV emission as indicated by the PL spectra of t BuBP4M (Figure 7 5). The longer emissions peaked at approximately 450 nm and 650 nm are probably due to the exciplex emission at the interfaces. OLEDs with other device structures are then fabricated as shown in Figure 7 7, to better align the energy levels of different layers. 106,108 However, none of these devices shows the similar emission as the solution PL spectra or even the solid state film PL spectra of t BuBP4M. For example, the neat t BuBP4M film is used as the EML layer with TcTa and TPBi as the HTL and ETL, respectively. The device only shows a noisy and unstable peak at wavelength of ~380 nm. The host material used in previous purine based devies, mC P, also does not work for the t BuBP4M based devices. The insertion

PAGE 187

187 of thin buffer layer of UGH2 between the host and ETL layers does not help, either. These devices show no UV emission from t BuBP4M and similar to the EL spectrum of control device with on ly neat mCP layer sandwiched by TcTa and TPBi layers. Figure 7 7. (color) Normalized EL spectra for the OLEDs based on t BuBP4M (UV) with different device structures. The existing organic materials used in OLEDs are mostly for the application of visibl e light emitting OLEDs, in which charges can be easily injected and then effectively co nfined, and energy transfer can successfully occur. But for UV OLEDs, as mentioned in the above of the t BuBP4M based devices, it is not the case. Due to the wide band g ap of the UV emitters ( E g > 3.3 eV ) the host materials should have even wider gap than UV emitters, so that the energy transfer effectively occurs. The HTL layers should have shallower LUMO levels to effectively confine the injected electrons. The HOMO le vels of HTL also need to be deeper, to reduce the injection barrier of holes from ano de. A hole injection layer (HIL ) may also help. Similar requirement including effectively

PAGE 188

188 injecting electrons and confining holes also apply to the ETL layer. At the same time, the emission of exciplex occurred at the interface of organic layers also need to be eliminated by changing the interface properties or energy level alignment. So, the materials of HTL, host and ETL need to be carefully designed and synthesized speci fically for UV OLEDs fabrication However, even the energy level alignment is finally achieved; there are still lots of other issues that will affect the device performance such as the device stability and efficiency. So it is still very challenging but pr omising to achi eve the OLEDs emitting UV light in the future

PAGE 189

189 APPENDIX A LIST OF ORGANIC MOLECULAR STRUCTURES This table lists the molecular structures for all the organic materials used in this dissertation. Please refer to LIST OF ABBREVIATIONS for full name of each molecule. Al q 3 BCP BEDOT TQMe 2 BEDOT BBT BPhen Bu 3 Si Th BBT CBP EtHx Th BBT

PAGE 190

190 Ir(ppy) 3 mCP m MTDATA NPD OXD 7 PBD PQIr Pt Ar 2 OPrTBP

PAGE 191

191 Pt Ar 4 TAP Pt Ar 2 TBP Pt Ar 4 TBP Pt DPTBP Pt TAr 2 TBP Pt TPTBP Pt TPTNP Purine 1

PAGE 192

192 Purine 2 PVK TAPC TAZ TcTa TPBi UGH2 3 TPYMB

PAGE 193

193 APPENDIX B LIST OF PUBLICATIONS AND CONFERENCES 1. R. Zhou, Y. Zheng, D. Xie, W. Cao, Y. Yang R. Stalder, M. Plaisant, K. S. Schanze, P. H. Holloway, J R. Reynolds, and J. Xue, High effciency solution processed hybrid polymer:colloidal nanocrystal photovoltaic cells by interface engineering Nature Mater. 2011 ( in preparation ). 2. R. Zhou, Y. Zheng, L. Qian, Y. Yang P. H. Holloway, and J. Xue, Solution processed nanostructured hybrid organic inorganic solar cells with broad spectral sensitivity Nano Research 2011 ( in preparation ). 3. K. R. Graham, Y. Yang J. R. Sommer, A. H. Shelton, K. S. Schanze, J. Xue and J. R. Reynolds, Extended conjugation platin um(II) porphyrins for use in near infrared emitting organic light emitting diodes Chem. Mater. 2011 ( accepted ). 4. W. Zhao, J. P. Mudrick, Y. Zheng, W. T. Hammond, Y. Yang and J. Xue, Enhancing photovoltaic response of organic solar cells using a crystalli ne molecular template Org. Electron. 2011 ( accepted ). 5. D. G. Patel, Y. Y. Ohnishi, Y. Yang S. H. Eom, R. T. Farley, K. R. Graham, J. Xue, S. Hirata, K. S. Schanze and J. R. Reynolds, Conjugated polym ers for pure UV light emission: poly(meta phenylenes ) J. Polym Sci Part B: Polym Phys 49 557 (2011). 6. Y. Yang J. R. Sommer, K. R. Graham, S. H. Eom, J. R. Reynolds, K. S. Schanze and J. Xue, High Efficiency near infrared phosphorescent OLEDs MRS Fall Conference Poster Presentation Boston (2010). 7. Y. Yang P. Cohn, A. L. Dyer, S. H. Eom, J. R. Reynolds, R. K. Castellano and J. Xue, Blue violet electroluminescence from a highly fluorescent purine Chem.

PAGE 194

194 Mater. 22 3580 (2010). 8. Y. Yang R. T. Farley, T. T. Steckler, S. H. Eom, J. R. Reynolds, K. S. Schanze and J. Xue, Efficient near infrared organic light emitting devices based on low gap fluorescent oligomers J. Appl. Phys. 106 044509 (2009). 9. Y. Yang R. T. Farley, T. T. Steckler, J. R. Sommer, S. H. Eom, K. R. Graham, J. R. Reynolds, K. S. Schanze and J. Xue, Near infrared fluorescent and phosphorescent organic light emitting devices MRS Symp. Proc. 1154 B05 98 (2009). 10. Y. Yang R. T. Farley, T. T. Steckler, J. R. Sommer, S. H. Eom, K. R. Graham, J. R. Reynolds, K. S. Schanze and J. Xue, Near infrared fluorescent and phosphorescent organic light emitting devices MRS Spring Conference Poster Presentation San Francisco (2009). 11. J. R. Sommer, R. T. Farley, K. R. Graham, Y. Yang J .R. Reynolds, J .Xue and K. S. Schanze, Efficient near in frared polymer and organic light emitting diodes based on electrophosphorescence from (Tetraphenyltetranaphtho[2,3]porphyrin) P latinum(II) ACS Appl. Mater. Interfaces 1 274 (2009). 12. Y. Yang R. T. Farley, T. T. Steckler, S. H. Eom, J. R. Reynolds, K. S. Schanze and J. Xue, Near infrared organic light emitting devices based on donor acceptor donor oligomers Appl. Phys. Lett. 93 163305 (2008).

PAGE 195

195 LIST OF REFERENCES 1 U. S. Department of Energy, National Lighting Inventory and Energ y Consumption Estimate Report, Vol 1 (2002). 2 M. Lenzen, Energies 3, 462 (2010). 3 S. S. Dihrab and K. Sopian, Renew Energy 35, 1303 (2010). 4 J. Brouwer, Curr Appl Phys 10, S9 (2010). 5 J. Paska, P. Biczel, and M. Klos, Renew Energy 34, 2414 (2009) 6 C. J. Lawn, Proc Inst Mech Eng C 223, 2717 (2009). 7 U.S. Department of Energy, US Lighting Market Characterization (2001). 8 K. T. Kamtekar, A. P. Monkman, and M. R. Bryce, Adv Mater 22, 572 (2010). 9 E. Dupont, H. C. Liu, M. Buchanan, S. Chiu and M. Gao, Appl. Phys. Lett. 76, 4 (2000). 10 C. J. Humphreys, MRS Bull 33, 459 (2008). 11 C. W. Tang and S. A. Vanslyke, Appl. Phys. Lett. 51, 913 (1987). 12 J. R. Sheats, H. Antoniadis, M. Hueschen, W. Leonard, J. Miller, R. Moon, D. Roitman, and A. Stocking, Science 273, 884 (1996). 13 J. Brooks, Y. Babayan, S. Lamansky, P. I. Djurovich, I. Tsyba, R. Bau, and M. E. Thompson, Inorg. Chem. 41, 3055 (2002). 14 S. R. Forrest, Chem. Rev. 97, 1793 (1997). 15 P. E. Burrows, S. R. Forrest, L. S. Sapochak, J. Schwartz, P. Fenter, T. Buma, V. S. Ban, and J. L. Forrest, J Cryst. Growth 156, 91 (1995). 16 M. Shtein, H. F. Gossenberger, J. B. Benziger, and S. R. Forrest, J. Appl. Phys. 89, 1470 (2001). 17 J. H. Burroughes, D. D. C. Bradley, A. R. Brown, R. N. Mar ks, K. Mackay, R. H. Friend, P. L. Burns, and A. B. Holmes, Nature 347, 539 (1990). 18 C. J. Drury, C. M. J. Mutsaers, C. M. Hart, M. Matters, and D. M. de Leeuw, Appl. Phys. Lett. 73, 108 (1998). 19 T. R. Hebner, C. C. Wu, D. Marcy, M. H. Lu, and J. C. St urm, Appl. Phys. Lett. 72, 519 (1998).

PAGE 196

196 20 D. Braun and A. J. Heeger, Appl. Phys. Lett. 58, 1982 (1991). 21 M. Singh, H. M. Haverinen, P. Dhagat, and G. E. Jabbour, Adv Mater 22, 673 (2010). 22 P. Andersson, R. Forchheimer, P. Tehrani, and M. Berggren, Ad v Funct Mater 17, 3074 (2007). 23 T. Sekitani and T. Someya, Adv Mater 22, 2228 (2010). 24 F. C. Krebs, Org Electron 10, 761 (2009). 25 F. C. Krebs, Sol Energy Mater Sol Cells 93, 465 (2009). 26 F. C. Krebs, Sol Energy Mater Sol Cells 93, 1636 (2009). 27 F. C. Krebs, S. A. Gevorgyan, and J. Alstrup, J Mater Chem 19, 5442 (2009). 28 F. C. Krebs, T. D. Nielsen, J. Fyenbo, M. Wadstrom, and M. S. Pedersen, Energy Environ Sci 3, 512 (2010). 29 F. C. Krebs and K. Norrman, ACS Appl. Mater. Interf aces 2, 877 (2010). 30 F. C. Krebs, T. Tromholt, and M. Jorgensen, Nanoscale 2, 873 (2010). 31 S. H. Eom, Y. Zheng, E. Wrzesniewski, J. Lee, N. Chopra, F. So, and J. G. Xue, Appl. Phys. Lett. 94, 153303 (2009). 32 S. J. Su, E. Gonmori, H. Sasabe, and J. Ki do, Adv Mater 20, 4189 (2008). 33 H. Sasabe, J. Takamatsu, T. Motoyama, S. Watanabe, G. Wagenblast, N. Langer, O. Molt, E. Fuchs, C. Lennartz, and J. Kido, Adv Mater 22, 5003 (2010). 34 C. Adachi, M. A. Baldo, M. E. Thompson, and S. R. Forrest, J. Appl Phys. 90, 5048 (2001). 35 M. A. Baldo, D. F. O'Brien, Y. You, A. Shoustikov, S. Sibley, M. E. Thompson, and S. R. Forrest, Nature 395, 151 (1998). 36 R. J. Holmes, S. R. Forrest, Y. J. Tung, R. C. Kwong, J. J. Brown, S. Garon, and M. E. Thompson, Appl. P hys. Lett. 82, 2422 (2003). 37 R. J. Holmes, B. W. D'Andrade, S. R. Forrest, X. Ren, J. Li, and M. E. Thompson, Appl. Phys. Lett. 83, 3818 (2003). 38 J. Li, P. I. Djurovich, B. D. Alleyne, M. Yousufuddin, N. N. Ho, J. C. Thomas, J. C. Peters, R. Bau, and M E. Thompson, Inorg. Chem. 44, 1713 (2005).

PAGE 197

197 39 Y. Zheng, S. H. Eom, N. Chopra, J. W. Lee, F. So, and J. G. Xue, Appl. Phys. Lett. 92, 223301 (2008). 40 S. Tokito, T. Iijima, Y. Suzuri, H. Kita, T. Tsuzuki, and F. Sato, Appl. Phys. Lett. 83, 569 (2003). 41 X. F. Ren, J. Li, R. J. Holmes, P. I. Djurovich, S. R. Forrest, and M. E. Thompson, Chem. Mat er 16, 4743 (2004). 42 S. H. Eom, Y. Zheng, N. Chopra, J. Lee, F. So, and J. G. Xue, Appl. Phys. Lett. 93, 133309 (2008). 43 S. H. Eom, Y. Zheng, E. Wrzesniewski J. Lee, N. Chopra, F. So, and J. G. Xue, Org Electron 10, 686 (2009). 44 J. Kido, K. Hongawa, K. Okuyama, and K. Nagai, Appl. Phys. Lett. 64, 815 (1994). 45 J. Kido, W. Ikeda, M. Kimura, and K. Nagai, J pn. J Appl Phys ., Part 2 35, L394 (1996). 46 J. Kido, M. Kimura, and K. Nagai, Science 267, 1332 (1995). 47 J. Kido, H. Shionoya, and K. Nagai, Appl. Phys. Lett. 67, 2281 (1995). 48 Y. R. Sun, N. C. Giebink, H. Kanno, B. W. Ma, M. E. Thompson, and S. R. Forrest, Nature 440, 908 (2006). 49 B. W. D'Andrad e, R. J. Holmes, and S. R. Forrest, Adv Mater 16, 624 (2004). 50 S. Y. Kim and J. J. Kim, Org Electron 11, 1010 (2010). 51 T. Yamasaki, K. Sumioka, and T. Tsutsui, Appl. Phys. Lett. 76, 1243 (2000). 52 V. Bulovic, V. B. Khalfin, G. Gu, P. E. Burrows, D Z. Garbuzov, and S. R. Forrest, Phys Rev B 58, 3730 (1998). 53 T. Tsutsui, M. Yahiro, H. Yokogawa, and K. Kawano, Adv Mater 13, 1149 (2001). 54 S. Reineke, F. Lindner, G. Schwartz, N. Seidler, K. Walzer, B. Lussem, and K. Leo, Nature 459, 234 (2009). 55 R. G. Sun, Y. Z. Wang, Q. B. Zheng, H. J. Zhang, and A. J. Epstein, J. Appl. Phys. 87, 7589 (2000). 56 L. H. Slooff, A. Polman, F. Cacialli, R. H. Friend, G. A. Hebbink, F. van Veggel, and D. N. Reinhoudt, Appl. Phys. Lett. 78, 2122 (2001). 57 R. J. Cu rry and W. P. Gillin, Appl. Phys. Lett. 75, 1380 (1999).

PAGE 198

198 58 N. Tessler, V. Medvedev, M. Kazes, S. H. Kan, and U. Banin, Science 295, 1506 (2002). 59 S. A. Priola, A. Raines, and W. S. Caughey, Science 287, 1503 (2000). 60 H. T. Whelan, R. L. Smits, E. V. B uchman, N. T. Whelan, S. G. Turner, D. A. Margolis, V. Cevenini, H. Stinson, R. Ignatius, T. Martin, J. Cwiklinski, A. F. Philippi, W. R. Graf, B. Hodgson, L. Gould, M. Kane, G. Chen, and J. Caviness, J. Clin. Laser Med. Sur. 19, 305 (2001). 61 T. Karu, Th e Science of Low Power Laser Theraphy (Gordon and Breach Scientific, New York, 1998). 62 R. Raghavachari, Near Infrared Applications in Biotechnology (CRC, Boca Raton, FL, 2001). 63 E. Desurvire, Erbium Doped Fiber Amplifiers: Principles and Applications ( Wiley Interscience, New York, 1994). 64 B. S. Harrison, T. J. Foley, M. Bouguettaya, J. M. Boncella, J. R. Reynolds, K. S. Schanze, J. Shim, P. H. Holloway, G. Padmanaban, and S. Ramakrishnan, Appl. Phys. Lett. 79, 3770 (2001). 65 B. S. Harrison, T. J. Fol ey, A. S. Knefely, J. K. Mwaura, G. B. Cunningham, T. S. Kang, M. Bouguettaya, J. M. Boncella, J. R. Reynolds, and K. S. Schanze, Chem. Mat er 16, 2938 (2004). 66 A. de Bettencourt Dias, Dalton Trans. 2229 (2007). 67 T. S. Kang, B. S. Harrison, T. J. Fole y, A. S. Knefely, J. M. Boncella, J. R. Reynolds, and K. S. Schanze, Adv Mater 15, 1093 (2003). 68 K. S. Schanze, J. R. Reynolds, J. M. Boncella, B. S. Harrison, T. J. Foley, M. Bouguettaya, and T. S. Kang, Synth. Met. 137, 1013 (2003). 69 F. X. Zang, Z. R. Hong, W. L. Li, M. T. Li, and X. Y. Sun, Appl. Phys. Lett. 84, 2679 (2004). 70 A. O'Riordan, E. O'Connor, S. Moynihan, P. Nockemann, P. Fias, R. Van Deun, D. Cupertino, P. Mackie, and G. Redmond, Thin Solid Films 497, 299 (2006). 71 A. O'Riordan, R. Va n Deun, E. Mairiaux, S. Moynihan, P. Fias, P. Nockemann, K. Binnemans, and G. Redmond, Thin Solid Films 516, 5098 (2008). 72 Z. F. Li, J. B. Yu, L. Zhou, H. J. Zhang, R. P. Deng, and Z. Y. Guo, Org Electron 9, 487 (2008).

PAGE 199

199 73 Z. F. Li, J. B. Yu, L. Zhou, R. P. Deng, and H. J. Zhang, Inorg Chem Commun 12, 151 (2009). 74 S. Yu, C. Ma, C. Cheng, X. Wang, D. Ji, Z. Fan, D. Xia, W. He, Y. Chang, and G. Du, Dyes Pigments 76, 492 (2008). 75 T. C. Rosenow, K. Walzer, and K. Leo, J. Appl. Phys. 103 (2008). 76 E. L. Williams, J. Li, and G. E. Jabbour, Appl. Phys. Lett. 89, 083506 (2006). 77 K. Kalyanasundaram, Photochemistry of Polypyridine and Porphyrin Complexes (Academic Press, London, 1992). 78 T. V. Duncan, K. Susumu, L. E. Sinks, and M. J. Therien, J Am Ch em Soc 128, 9000 (2006). 79 D. F. O'Brien, M. A. Baldo, M. E. Thompson, and S. R. Forrest, Appl. Phys. Lett. 74, 442 (1999). 80 M. Colle, C. Garditz, and M. Braun, J. Appl. Phys. 96, 6133 (2004). 81 H. D. Burrows, M. Fernandes, J. S. de Melo, A. P. Monkm an, and S. Navaratnam, J Am Chem Soc 125, 15310 (2003). 82 Y. Wang, Appl. Phys. Lett. 85, 4848 (2004). 83 V. A. Montes, C. Perez Bolivar, N. Agarwal, J. Shinar, and P. Anzenbacher, J Am Chem Soc 128, 12436 (2006). 84 M. Ikai, F. Ishikawa, N. Aratan i, A. Osuka, S. Kawabata, T. Kajioka, H. Takeuchi, H. Fujikawa, and Y. Taga, Adv Funct Mater 16, 515 (2006). 85 Q. Hou, Y. Zhang, F. Y. Li, J. B. Peng, and Y. Cao, Organometallics 24, 4509 (2005). 86 J. Kalinowski, W. Stampor, J. Szmytkowski, M. Cocchi, D. Virgili, V. Fattori, and P. Di Marco, J. Chem. Phys. 122, 16 (2005). 87 V. V. Rozhkov, M. Khajehpour, and S. A. Vinogradov, Inorg. Chem. 42, 4253 (2003). 88 C. Borek, K. Hanson, P. I. Djurovich, M. E. Thompson, K. Aznavour, R. Bau, Y. R. Sun, S. R. For rest, J. Brooks, L. Michalski, and J. Brown, Angew Chem Int Ed 46, 1109 (2007). 89 Y. Sun, C. Borek, K. Hanson, P. I. Djurovich, M. E. Thompson, J. Brooks, J. J. Brown, and S. R. Forrest, Appl. Phys. Lett. 90, 213503 (2007). 90 J. Shinar and R. Shinar, J. Phys. D Appl. Phys. 41, 26 (2008).

PAGE 200

200 91 S. Landgraf, J. Biochem. Biophys. Methods 61, 125 (2004). 92 H. van Santen and J. H. M. Neijzen, Jpn. J. Appl. Phys. Part 1 42, 1110 (2003). 93 Y. Fukuda, T. Watanabe, T. Wakimoto, S. Miyaguchi, and M. Tsuchida, S ynth. Met. 111, 1 (2000). 94 A. Haldi, J. B. Kim, B. Domercq, A. P. Kulkarni, S. Barlow, A. P. Gifford, S. A. Jenekhe, S. R. Marder, and B. Kippelen, J. Disp. Technol. 5, 120 (2009). 95 C. J. Tonzola, A. P. Kulkarni, A. P. Gifford, W. Kaminsky, and S. A. J enekhe, Adv Funct Mater 17, 863 (2007). 96 S. L. Lin, L. H. Chan, R. H. Lee, M. Y. Yen, W. J. Kuo, C. T. Chen, and R. J. Jeng, Adv Mater 20, 3947 (2008). 97 N. Matsumoto, T. Miyazaki, M. Nishiyama, and C. Adachi, J. Phys. Chem. C 113, 6261 (2009). 98 T. Y. Zhang, J. Wang, T. Li, M. Liu, W. F. Xie, S. Y. Liu, D. L. Liu, C. L. Wu, and C. T. Chen, J. Phys. Chem. C 114, 4186 (2010). 99 Y. Yang, Q. Pei, and A. J. Heeger, J. Appl. Phys. 79, 934 (1996). 100 J. S. Huang, H. F. Zhang, W. J. Tian, J. Y. Hou, Y. G. Ma, J. C. Shen, and S. Y. Liu, Synth. Met. 87, 105 (1997). 101 K. Okumoto and Y. Shirota, Appl. Phys. Lett. 79, 1231 (2001). 102 M. Kinoshita, H. Kita, and Y. Shirota, Adv Funct Mater 12, 780 (2002). 103 K. Okumoto and Y. Shirota, Chem. Mat er 15, 69 9 (2003). 104 J. R. Hwu, Y. C. Hsu, T. Josephrajan, and S. C. Tsay, J Mater Chem 19, 3084 (2009). 105 K. R. Wee, H. C. Ahn, H. J. Son, W. S. Han, J. E. Kim, D. W. Cho, and S. O. Kang, J. Org. Chem. 74, 8472 (2009). 106 T. C. Chao, Y. T. Lin, C. Y. Yang, T. S. Hung, H. C. Chou, C. C. Wu, and K. T. Wong, Adv Mater 17, 992 (2005). 107 C. T. Chen, Y. Wei, J. S. Lin, M. Moturu, W. S. Chao, Y. T. Tao, and C. H. Chien, J Am Chem Soc 128, 10992 (2006). 108 H. Etori, T. Yasuda, X. L. Jin, K. Fujita, S. Mata ka, and T. Tsutsui, Jpn J Appl Phys ., Part 1 46, 5071 (2007). 109 S. Hoshino, K. Ebata, and K. Furukawa, J. Appl. Phys. 87, 1968 (2000).

PAGE 201

201 110 J. Kido, K. Hongawa, K. Okuyama, and K. Nagai, Appl. Phys. Lett. 63, 2627 (1993). 111 S. Hoshino, K. Furukawa, K Ebata, C. H. Yuan, and H. Suzuki, J. Appl. Phys. 88, 2892 (2000). 112 H. Q. Zhang, B. Yang, Y. Zheng, G. D. Yang, L. Ye, Y. G. Ma, X. F. Chen, G. Cheng, and S. Y. Liu, J Phys Chem B 108, 9571 (2004). 113 E. A. Silinsh and V. Capek, Organic Molecular C rystals: Interaction, Localization, and Transport Phenomena (AIP, New York, 1994). 114 M. Pope and C. E. Swenberg, Electronic Processes in Organic Crystals and Polymers 2nd ed. (Oxford University Press, New York, 1999). 115 S. R. Forrest, Nature 428, 911 (2004). 116 W. Brutting, Physics of Organic Semiconductors (Wiley VCH, 2005). 117 J. D. Wright, Molecular Crystals 2nd ed. (Cambridge University Press, 1995). 118 L. B. Schein, W. Warta, A. R. McGhie, and N. Karl, Chem Phys Lett 75, 267 (1980). 119 S. R. Forrest, M. L. Kaplan, and P. H. Schmidt, J. Appl. Phys. 56, 543 (1984). 120 N. Karl and J. Marktanner, Mol Cryst Liq Cryst 355, 149 (2001). 121 J. Xue, Ph. D Thesis, Princeton University, 2005. 122 I. G. Hill, A. Kahn, Z. G. Soos, and R. A. Pascal, Chem Phys Lett 327, 181 (2000). 123 M. Knupfer, H. Peisert, and T. Schwieger, Phys Rev B 65 (2002). 124 S. Barth and H. Bassler, Phys Rev Lett 79, 4445 (1997). 125 N. W. Ashcroft and N. D. Mermin, Solid State Physics (Saunders College, Philadelphi a, 1976). 126 M. A. Baldo, Ph. D Thesis, Princeton University, 2001. 127 S. Pawlizak, Diploma Thesis, University of Leipzig, 2009. 128 V. Bulovic, P. E. Burrows, S. R. Forrest, J. A. Cronin, and M. E. Thompson, Chem. Phys. 210, 1 (1996). 129 T. Forster, Di scuss Faraday Soc 7 (1959).

PAGE 202

202 130 A. D. McNaught and A. Wilkinson, IUPAC Compendium of Chemical Terminology (Royal Society of Chemistry, Cambridge, UK). 131 P. Peumans, A. Yakimov, and S. R. Forrest, J. Appl. Phys. 93, 3693 (2003). 132 C. W. Tang, Appl. P hys. Lett. 48, 183 (1986). 133 W. H. Kim, A. J. Makinen, N. Nikolov, R. Shashidhar, H. Kim, and Z. H. Kafafi, Appl. Phys. Lett. 80, 3844 (2002). 134 Y. R. Sun, M. Shtein, and S. R. Forrest, Appl. Phys. Lett. 86 (2005). 135 M. S. Arnold, G. J. McGraw, S. R. Forrest, and R. R. Lunt, Appl. Phys. Lett. 92 (2008). 136 K. Hamano, T. Kurata, S. Kubota, and H. Koezuka, J pn. J Appl Phys ., Part 2 33, L1031 (1994). 137 F. Schreiber, Phys Status Solidi A 201, 1037 (2004). 138 R. Green, A. Morfa, A. J. Ferguson, N. K opidakis, G. Rumbles, and S. E. Shaheen, Appl. Phys. Lett. 92 (2008). 139 K. X. Steirer, M. O. Reese, B. L. Rupert, N. Kopidakis, D. C. Olson, R. T. Collins, and D. S. Ginley, Sol Energy Mater Sol Cells 93, 447 (2009). 140 M. Shtein, Ph. D Thesis, Princ eton University, 2004. 141 Meyerare.Jr, Appl. Optics 7, 2081 (1968). 142 S. H. Eom, Ph. D Thesis, University of Florida, 2010. 143 W. H. Carter and E. Wolf, J Opt Soc Am 65, 1067 (1975). 144 CIE (1932), Commission Internationale de L'Eclairage Proceedi ngs (Cambridge University Press, Cambridge, 1931). 145 B. C. Thompson, L. G. Madrigal, M. R. Pinto, T. S. Kang, K. S. Schanze, and J. R. Reynolds, J. Polym. Sci. Pol. Chem. 43, 1417 (2005). 146 J. C. Ostrowski, K. Susumu, M. R. Robinson, M. J. Therien, and G. C. Bazan, Adv Mater 15, 1296 (2003). 147 T. C. Lee, J. Y. Hung, Y. Chi, Y. M. Cheng, G. H. Lee, P. T. Chou, C. C. Chen, C. H. Chang, and C. C. Wu, Adv Funct Mater 19, 2639 (2009). 148 K. Mullen and G. Wegner, Electronic Materials: The Oligomer App roach (Wiley, New York, 1998).

PAGE 203

203 149 T. A. Skotheim and J. R. Reynolds, Handbook of Conducting Polymers (CRC, New York, 2007). 150 M. Albota, D. Beljonne, J. L. Bredas, J. E. Ehrlich, J. Y. Fu, A. A. Heikal, S. E. Hess, T. Kogej, M. D. Levin, S. R. Marder, D McCord Maughon, J. W. Perry, H. Rockel, M. Rumi, C. Subramaniam, W. W. Webb, X. L. Wu, and C. Xu, Science 281, 1653 (1998). 151 T. T. Steckler, K. A. Abboud, M. Craps, A. G. Rinzler, and J. R. Reynolds, Chem. Commun. 4904 (2007). 152 G. Qian, B. Dai, M. Luo, D. B. Yu, J. Zhan, Z. Q. Zhang, D. G. Ma, and Z. Y. Wang, Chem. Mat er 20, 6208 (2008). 153 G. Qian, Z. Zhong, M. Luo, D. B. Yu, Z. Q. Zhang, Z. Y. Wang, and D. G. Ma, Adv Mater 21, 111 (2009). 154 M. A. Baldo, M. E. Thompson, and S. R. Forrest, Na ture 403, 750 (2000). 155 M. G. Mason, C. W. Tang, L. S. Hung, P. Raychaudhuri, J. Madathil, D. J. Giesen, L. Yan, Q. T. Le, Y. Gao, S. T. Lee, L. S. Liao, L. F. Cheng, W. R. Salaneck, D. A. dos Santos, and J. L. Bredas, J. Appl. Phys. 89, 2756 (2001). 156 M. A. Baldo, S. Lamansky, P. E. Burrows, M. E. Thompson, and S. R. Forrest, Appl. Phys. Lett. 75, 4 (1999). 157 S. Lamansky, P. Djurovich, D. Murphy, F. Abdel Razzaq, R. Kwong, I. Tsyba, M. Bortz, B. Mui, R. Bau, and M. E. Thompson, Inorg. Chem. 40, 1704 (2001). 158 S. R. Forrest, D. D. C. Bradley, and M. E. Thompson, Adv Mater 15, 1043 (2003). 159 J. G. Xue and S. R. Forrest, J. Appl. Phys. 95, 1869 (2004). 160 I. G. Hill, D. Milliron, J. Schwartz, and A. Kahn, Appl. Surf. Sci. 166, 354 (2000). 161 C. W Tang, S. A. Vanslyke, and C. H. Chen, J. Appl. Phys. 65, 3610 (1989). 162 A. Shoustikov, Y. J. You, P. E. Burrows, M. E. Thompson, and S. R. Forrest, Synth. Met. 91, 217 (1997). 163 J. Morgado, F. Cacialli, R. Iqbal, S. C. Moratti, A. B. Holmes, G. Yahio glu, L. R. Milgrom, and R. H. Friend, J Mater Chem 11, 278 (2001). 164 M. Berggren, A. Dodabalapur, R. E. Slusher, and Z. Bao, Nature 389, 466 (1997). 165 M. A. Baldo, D. F. O'Brien, M. E. Thompson, and S. R. Forrest, Phys Rev B 60, 14422 (1999). 166 S. C. Tse, H. H. Fong, and S. K. So, J. Appl. Phys. 94, 2033 (2003).

PAGE 204

204 167 J. W. Kang, S. H. Lee, H. D. Park, W. I. Jeong, K. M. Yoo, Y. S. Park, and J. J. Kim, Appl. Phys. Lett. 90, 223508 (2007). 168 V. Adamovich, J. Brooks, A. Tamayo, A. M. Alexander, P. I. Djurovich, B. W. D'Andrade, C. Adachi, S. R. Forrest, and M. E. Thompson, New J. Chem. 26, 1171 (2002). 169 J. Kido and Y. Okamoto, Chem. Rev. 102, 2357 (2002). 170 O. S. Finikova, S. E. Aleshchenkov, R. P. Brinas, A. V. Cheprakov, P. J. Carroll, and S. A. Vinogradov, J. Org. Chem. 70, 4617 (2005). 171 R. J. Cheng, Y. R. Chen, S. L. Wang, and C. Y. Cheng, Polyhedron 12, 1353 (1993). 172 J. R. Sommer, R. T. Farley, K. R. Graham, Y. Yang, J. R. Reynolds, J. Xue, and K. S. Schanze, ACS Appl. Mater. Interfac es 1, 274 (2009). 173 A. Y. Lebedev, M. A. Filatov, A. V. Cheprakov, and S. A. Vinogradov, J Phys Chem A 112, 7723 (2008). 174 C. A. Barker, X. S. Zeng, S. Bettington, A. S. Batsanov, M. R. Bryce, and A. Beeby, Chem. Eur. J. 13, 6710 (2007). 175 C. M. D rain, C. Kirmaier, C. J. Medforth, J. Nurco, K. M. Smith, and D. Holten, J. Phys. Chem. 100, 11984 (1996). 176 B. Roder, M. Buchner, I. Ruckmann, and M. O. Senge, Photochem Photobiol Sci 9, 1152 (2010). 177 J. R. Sommer, A. H. Shelton, A. Parthasarathy, I. Ghiviriga, J. R. Reynolds, and K. S. Schanze, Chem. Mat er 2011 ( accepted ) 178 K. Walzer, B. Maennig, M. Pfeiffer, and K. Leo, Chem. Rev. 107, 1233 (2007). 179 A. Mishra, N. Periasamy, M. P. Patankar, and K. L. Narasimhan, Dyes Pigments 66, 89 (2005) 180 K. Zhang, Z. Chen, C. L. Yang, Y. Zou, S. L. Gong, J. G. Qin, and Y. Cao, J. Phys. Chem. C 112, 3907 (2008). 181 V. Kishore, K. L. Narasimhan, and N. Periasamy, Phys. Chem. Chem. Phys. 5, 1386 (2003). 182 W. Siebrand, J. Chem. Phys. 44, 4055 (1966). 183 W. Siebrand, J. Chem. Phys. 46, 440 (1967).

PAGE 205

205 184 J. V. Caspar, E. M. Kober, B. P. Sullivan, and T. J. Meyer, J Am Chem Soc 104, 630 (1982). 185 Y. Yang, R. T. Farley, T. T. Steckler, S. H. Eom, J. R. Reynolds, K. S. Schanze, and J. Xue, Appl. Phys. Lett. 93, 163305 (2008). 186 Y. Yang, R. T. Farley, T. T. Steckler, S. H. Eom, J. R. Reynolds, K. S. Schanze, and J. Xue, J. Appl. Phys. 106, 7 (2009). 187 M. A. Baldo, C. Adachi, and S. R. Forrest, Phys Rev B 62, 10967 (2000). 188 K. Zhang, Z. Chen, C. L. Yang, Y. Zou, S. L. Gong, Y. T. Tao, J. G. Qin, and Y. Cao, J Mater Chem 18, 3366 (2008). 189 K. R. Graham, Y. Yang, J. R. Sommer, A. H. Shelton, K. S. Schanze, J. Xue, and J. R. Reynolds, Chem. Mat er 2011 ( accepted ) 190 J. E. Rogers, K. A. Nguyen D. C. Hufnagle, D. G. McLean, W. J. Su, K. M. Gossett, A. R. Burke, S. A. Vinogradov, R. Pachter, and P. A. Fleitz, J Phys Chem A 107, 11331 (2003). 191 A. Rosa, G. Ricciardi, and E. J. Baerends, J Phys Chem A 110, 5180 (2006). 192 V. N. Knyukshto, E. I. Sagun, A. M. Shul'ga, S. M. Bachilo, D. A. Starukhin, and E. I. Zen'kevich, Opt Spectros 90, 67 (2001). 193 R. C. Kwong, S. Sibley, T. Dubovoy, M. Baldo, S. R. Forrest, and M. E. Thompson, Chem. Mat er 11, 3709 (1999). 194 J. M. Lupton, I. D. W. S amuel, M. J. Frampton, R. Beavington, and P. L. Burn, Adv Funct Mater 11, 287 (2001). 195 R. J. Holmes, S. R. Forrest, T. Sajoto, A. Tamayo, P. I. Djurovich, and M. E. Thompson, Org Electron 7, 163 (2006). 196 D. C. Ward, E. Reich, and L. Stryer, J B iol Chem 244, 1228 (1969). 197 J. D. Ballin, S. Bharill, E. J. Fialcowitz White, I. Gryczynski, Z. Gryczynski, and G. M. Wilson, Biochemistry 46, 13948 (2007). 198 G. Kodali, K. A. Kistler, S. Matsika, and R. J. Stanley, J Phys Chem B 112, 1789 (2008) 199 S. Bharill, P. Sarkar, J. D. Ballin, I. Gryczynski, G. M. Wilson, and Z. Gryczynski, Anal Biochem 377, 141 (2008). 200 R. K. Neely, S. W. Magennis, S. Parsons, and A. C. Jones, Chem PhysC hem 8, 1095 (2007).

PAGE 206

206 201 S. J. Kim and E. T. Kool, J Am Chem Soc 128, 6164 (2006). 202 F. D'Souza, S. Gadde, D. M. S. Islam, S. C. Pang, A. L. Schumacher, M. E. Zandler, R. Horie, Y. Araki, and O. Ito, Chem. Commun. 480 (2007). 203 R. S. Butler, P. Cohn, P. Tenzel, K. A. Abboud, and R. K. Castellano, J Am Chem Soc 131, 623 (2009). 204 R. S. Butler, A. K. Myers, P. Bellarmine, K. A. Abboud, and R. K. Castellano, J Mater Chem 17, 1863 (2007). 205 J. Vura Weis, M. R. Wasielewski, A. K. Thazhathveetil, and F. D. Lewis, J Am Chem Soc 131, 9722 (2009). 206 S. Sivakova and S. J. Rowan, Chem. Soc. Rev. 34, 9 (2005). 207 S. Lena, G. Brancolini, G. Gottarelli, P. Mariani, S. Masiero, A. Venturini, V. Palermo, O. Pandoli, S. Pieraccini, P. Samori, and G. P. Spada, Chem. Eur. J. 13, 3757 (2007). 208 R. Rinaldi, G. Maruccio, A. Biasco, V. Arima, R. Cingolani, T. Giorgi, S. Masiero, G. P. Spada, and G. Gottarelli, Nanotechnology 13, 398 (2002). 209 S. D'Amico, G. Maruccio, P. Visconti, E. D'Amone, R. Cingolani, R. Rinaldi, S. Masiero, G. P. Spada, and G. Gottarelli, Microelectron. J. 34, 961 (2003). 210 Y. Yang, P. Cohn, A. L. Dyer, S. H. Eom, J. R. Reynolds, R. K. Castellano, and J. G. Xue, Chem. Mat er 22, 3580 (2011). 211 M. Ichikawa, K. Kobayashi, T. Koyama, and Y. Taniguchi, Thin Solid Films 515, 3932 (2007). 212 M. H. Tsai, H. W. Lin, H. C. Su, T. H. Ke, C. C. Wu, F. C. Fang, Y. L. Liao, K. T. Wong, and C. I. Wu, Adv Mater 18, 1216 (2006). 213 A. J. Bard and L. R. Faulkner, Electrochemical Methods: Fundamentals and Applications 2nd ed. (Wiley, New York, 2001). 214 D. Tanaka, Y. Agata, T. Takeda, S. Watanabe, and J. Kido, Jpn. J. Appl. Phys. Part 2 46, L117 (2007). 215 N. Chopra, J. Lee, Y. Zheng, S. H. Eom, J. E. Xue, and F. So, ACS Appl. Mater. Interfaces 1, 1169 (2009). 216 B. J. Chen, X. H. Zhang, X. Q. Lin H. L. Kwong, N. B. Wong, C. S. Lee, W. A. Gambling, and S. T. Lee, Synth. Met. 118, 193 (2001). 217 D. Kolosov, V. Adamovich, P. Djurovich, M. E. Thompson, and C. Adachi, J Am Chem Soc 124, 9945 (2002).

PAGE 207

207 218 J. S. Kim, B. W. Seo, and H. B. Gu, Synth. Met. 132, 285 (2003). 219 S. F. Chen, Z. J. Wu, Y. Zhao, C. N. Li, J. Y. Hou, and S. Y. Liu, Org Electron 6, 111 (2005). 220 D. W. Zhao, Z. Xu, F. J. Zhang, S. F. Song, S. L. Zhao, Y. Wang, G. C. Yuan, Y. F. Zhang, and H. H. Xu, Appl. Surf. Sci. 253, 402 5 (2007). 221 A. M. Martin, R. S. Butler, I. Ghiviriga, R. E. Giessert, K. A. Abboud, and R. K. Castellano, Chem. Commun. 4413 (2006). 222 H. A. Wagenknecht, Nat. Prod. Rep. 23, 973 (2006). 223 J. Peet, J. Y. Kim, N. E. Coates, W. L. Ma, D. Moses, A. J. H eeger, and G. C. Bazan, Nature Mater 6, 497 (2007). 224 E. Zhou, M. Nakamura, T. Nishizawa, Y. Zhang, Q. Wei, K. Tajima, C. Yang, and K. Hashimoto, Macromolecules 41, 8302 (2008). 225 J. Hou, H. Y. Chen, S. Zhang, G. Li, and Y. Yang, J Am Chem Soc 130 16144 (2008). 226 B. C. Thompson, Y. G. Kim, T. D. McCarley, and J. R. Reynolds, J Am Chem Soc 128, 12714 (2006).

PAGE 208

208 BIOGRAPHICAL SKETCH Yixing Yang was born in Jinhua, China in 1985. He grew up in the same city with good academic education and f inished his middle school there in 2000 with top 3 ranking in the whole school Then he graduated from Jinhua No.1 High School in 2003. In the same year he was admitted to one of the best colleges in China, University of Science and Technology of China (U STC), for higher level education. In 2007, he successfully graduated from USTC with a b achelor degree of Polymer Science and Engineering. Because of the outstanding academic performance there he was honored with the Presidential Guo Moruo Scholarship. A t fall of 2007 he traveled abroad to United States and started to pursue his D octorate degree in the Department of Materials Science and Engineering at University of Florida. Working on the research fields of organic light emitting devices and organic pho tovoltaic devices, he became a well qualified and dedicated researcher in organic semiconductor thin film deposition, device fabrication and characterization He published more than 10 peer reviewed papers in scientific journals and attended several intern ational conferences Under the supervision of his advisor, Dr. Jiangeng Xue, in December 2011, he received his Doctor of Philosophy degree in the Department of Materials Science and Engineering at University of Florida.