`
`
`
`Ry
`_ Panel displays
`
`WWWILEY SID Series in Display Technology SAMSUNG Ex. 1013 - 1/30
`
`SAMSUNG EX. 1013 - 1/30
`
`
`
`
`
`Copyright © 2005
`
`John Wiley & Sons Ltd, ‘The Atrium, Southern Gate, Chichester,
`West Sussex PO19 8SQ, England
`Telephone
`(444) 1243 779777
`Email (for orders and customer service enquiries): cs-books @ wiley.co.uk
`Visit our Home Page on www.wiley.com
`
`Reprinted with corrections September 2005
`
`TK
`7382
`tb
`fFS>
`a oOo? 3S
`
`All Rights Reserved. No part of this publication may be reproduced, stored in a retrieval system or transmitted in
`any form or by any means, electronic, mechanical, photocopying, recording, scanning or otherwise, except under
`the terms of the Copyright, Designs and Patents Act 1988 or underthe terms ofa licence issued by the Copyright
`Licensing Agency Ltd, 90 Tottenham Court Road, London W1T 4LP, UK, without the permission in writing of
`the Publisher. Requests to the Publisher should be addressed to the Permissions Department, John Wiley & SonsLtd,
`The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ, England, or emailed to permreq @wiley.co.uk, or
`faxed to (+44) 1243 770620,
`
`This publication is designed to provide accurate and authoritative information in regard to the subject maiter
`covered. It is sold on the understandingthat the Publisher is not engaged in rendering professionalservices.
`If professional advice or other expert assistance is required, the services of a competent professional should
`be sought.
`
`Other Wiley Editorial Offices
`
`John Wiley & Sons Inc., !11 River Street, Hoboken, NJ 07030, USA
`
`Jossey-Bass, 989 Market Street, San Francisco, CA 94103-1741, USA
`
`Wiley-VCH Verlag GmbH,Boschstr. 12, D-69469 Weinheim, Germany
`
`John Wiley & Sons Australia Ltd, 33 Park Road, Milton, Queensland 4064, Australia
`
`John Wiley & Sons (Asia) Pte Ltd, 2 Clementi Loop #02-01, Jin Xing Distripark, Singapore 129809
`
`John Wiley & Sons Canada Ltd, 22 Worcester Road, Etobicoke, Ontario, Canada M9W ILI
`
`Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not
`be available in electronic books.
`
`Library of Congress Cataloging-in-Publication Data
`
`Flexible flat panel displays / edited by Gregory P. Crawford.
`p.
`cm.
`Includes bibliographical references and index,
`ISBN-13 978-0-470-87048-8 (alk.paper)
`ao.10 0-470-87048-6 (alk. paper)
`Information display systems. 2. Liquid crystal displays.
`F Electroluminescent display systems.
`I. Crawford, Gregory Philip.
`TK7882.16F54
`2005
`621.3815'422-de22
`
`British Library Cataloguing in Publication Data
`
`A catalogue record for this book is available from the British Library
`
`ISBN-13 978-0-470-87048-8 (HB)
`ISBN-10 0-470-87048-6 (HB)
`
`2005003238
`
`Typeset in 10/12pt Times by ThomsonPress (India) Limited, New Delhi
`Printed and bound in Great Britain by Antony Rowe Ltd., Chippenham, Wiltshire
`This book is printed on acid-free paper responsibly manufactured from sustainable forestry
`in whichat least two trees are planted for each one used for paper production.
`
`
`
`SAMSUNG Ex. 1013 - 2/30
`
`SAMSUNG EX. 1013 - 2/30
`
`
`
`
`
`15
`
`OLED Displays on Plastic
`
`DuPont Displays
`
`15.1
`
`Introduction
`
`Mark L. Hildner
`
`Organic light-emitting diode (OLED) technology has captured tremendousinterest and has
`rapidly developed since the discovery of organic electroluminescence roughly 15 years ago.
`Commercial OLED displays on glass are now available and the industry is poised for
`substantial growth in the next few years. Muchofthe attention given to OLEDsis due to the
`performance advantagesthatit has overothertypes offlat panel display (FPD) technologies,
`including the industry dominant liquid crystal displays (LCDs). Recognized advantages
`include nearly Lambertian emission, which provides wider viewing angles than LCD; fast
`Tesponse times, which facilitate grayscale and video capabilities in active matrix applica-
`tions; and low-voltage operation, which leads to low-cost components and low-profile
`packaging. Furthermore, the high efficiency of OLED materials makes OLED the lowest-
`Power emissive FPD technology and offers the potential for lower power consumption than
`backlit LCDs.
`An additional factor giving OLED technology impetus, perhaps to an extent equal to
`the performance advantages, is the perception that OLEDsare a natural choice for flexible
`displays. The very thin structure (the active layers are less than 1 jum), solid-state construc-
`Hon(there is no cell gap as in an LCD), andactive material composition of an OLED are
`:
`XIbitity.
`a
`STIs,
`v
`PCMrOlMalic
`ar
`as have many thinking that OLED is the technology path to high-performance
`Ull-color flexible displays.
`
`
`
`
`
`
`
`
`Flexible F]
`at Panel Displays Edited by G. P. Crawford
`® 2005
`John Wiley & Sons, Ltd
`
`i
`
`re
`
`SAMSUNG Ex. 1013 - 3/30
`
`SAMSUNG EX. 1013 - 3/30
`
`
`
` 286
`
`OLED DISPLAYS ON PLASTIC
`
`
`
`
`
`The two main types of OLED are based on small molecules and conjugated polymers.
`Small-molecule OLEDs (SMOLEDs) werefirst reported by Tang and VanSlyke (1987) and
`are typically thermally evaporated. Light cmission from polymer OILEDs (PLEDs) wasfirst
`reported by Burrougheser al. (1990), where a solution-processable precursor polymer was
`deposited by spin coating and then thermally converted at high temperature (> 250 °C),
`Then Braun and Heeger (1991) were able to make a light-emitting device with a polymer
`that was soluble in its conjugated form, thus climinating the need for high-temperature
`processing. Solution processing at low temperatures may revolutionize how displays are
`manufactured because it permits a number ofprocess options (spin coating, inkjet printing,
`dipping, spraying, etc.) that are lowercost than vacuum deposition; it would replace much of
`the vacuum processing usedin today’s FPD fab; it can coverlarge areas; andit is well suited
`for roll-to-roll manufacturing, which may lead to further cost reduction from the current
`batch process. That is why this chapter will focus on conjugated polymer OLEDs.
`A numberofflexible materials are being explored as OLED substrates. Thefirst flexible
`OLEDdisplay demonstration was on a transparent plastic substrate (Gustafsson etal. 1992).
`Plastic is a logical choice because its transparency allows much of the architecture of an
`OLEDonglass to be used.Plastic is rugged, more so than regularglass; able to be accurately
`cut with a laser, allowing for irregular shapes with the only downside being some
`discoloration at
`the cutting site; and is already incorporated into roll-to-roll process
`technology, both in its own manufacture and current applications.
`there are
`While flexibility may be the ultimate goal for OLED displays on plastic,
`significant opportunities that are less technologically demanding than a display that can be
`flexed or rolled up multiple times. A flat plastic OLED display is thin, lightweight, and
`rugged. These are significant attributes that may be taken advantage of in mobile applica-
`tions. Plastic displays can be easily cut into a wide variety of nonrectangular shapes, and can
`be bent into a curved, but rigid, format. These characteristics allow greater freedom in
`product design. Even for these nonflexible display manifestations,
`there are significant
`development challengesto bringing a plastic OLED display to the marketplace.
`After a brief introduction describing how a PLED display works, this chapter will present
`the challenges associated with two key technology developments that must take place. The
`first is to obtain a plastic substrate that can withstand processing and lead to a reliable and
`long-lived device. The second is to obtain an understanding of the manufacturing issues
`associated with a plastic substrate, and then to incorporate this understanding into device
`processing. The issues associated with making a passive matrix (PM) OLED will
`then
`be discussed, andfinally, there will be a review of thin film transistor technologies that are
`appropriate for plastic active matrix (AM) backplanes.
`
`15.2.1 Conjugated Polymers
`
`Conjugated polymers are characterized by alternating single and doubleor single and triple
`bonds (Heeger 2001). Overlapping of the p, orbitals from the doubleor triple bonds along
`the polymerbackboneleads to the formation ofa delocalized m-bonding system. This gives
`rise to energy bands similar to those in an inorganic semiconductor, The occupied w-band.
`analogous to the valence band,
`is comprised of hole-transport states, and the highest
`
`—rs
`4i
`
`
`
`SAMSUNG Ex. 1013 - 4/30
`
`SAMSUNG EX. 1013 - 4/30
`
`
`
`PLED BASICS
`
`287
`
`aceupied molecular orbital (HOMO)is analogousto the valence band edge. The unoccupied
` -band, analogousto the conduction band, is comprised of electron-transportstates, and the
`lowest unoccupied molecular orbital (LUMO) is analogous to the conduction band edge.
`Despite this analogy, charge transport in conjugated polymers differs in a number of ways
`from that in inorganic semiconductors(Patel et al. 2002): intrinsic and extrinsic carriers are
`generally negligible and conduction is dominated by injected carriers, the polymer chains
`distort aroundthe charge carrier so that the charged excitation is best described as a polaron
`(the charge plus the distortion); a nd the energy bands are inhomogeneously broadened due to
`the amorphous polymer structure and,
`therefore, transport is through hopping along or
`between polymer chains.
`
`15.2.2
`
`Light-Emitting Diodes
`
`A conjugated polymer can emit light because it has an energy gap. Figure 15.1 shows three
`common light-emitting polymers: poly(p-phenylenevinylene) or PPV; poly[2-methoxy,
`
`no
`rk
`fA) al {OO}
`
`HOMOand LUMO,respectively; this defines the need for a high work function anode and a
`
`5-(2'-ethyl-hexyloxy)-p-phenylencvinylene] or MEH-PPV; and polyfluorine. The basic
`polymerlight-emitting device (a diode) consists of a light-emitting polymer (LEP) film of
`~100nm sandwiched between an optically transparent anode, which sits on an optical
`quality glass or plastic substrate, and a metallic cathode. The anode is usually indium tin
`oxide (ITO), which has a high work function, whereas the cathode is typically a low work
`function metal such as Ca or Mg. Whena bias greater than the difference between the anode
`and cathode work functions (the built-in potential) is applied as illustrated im the band
`diagram of Figure 15.2(a), electrons are injected from the cathode into the nm-band, and
`holes are injected from the anode into the 7-band. The injected charges (electron and hole
`type polarons) form bound polaron-excitons,i.e. neutral bipolarons bound by their Coulomb
`attraction and their shared distortion (Heeger 2001). Electroluminescence (EL) results from
`the radiative decay (electron-hole recombination) of these excitons. The device is a diode
`because application of a reverse bias prevents charge flow (there is no light emission). The
`energy gap and thus the emission color of the diode can be tuned by changing the length of
`ithe polymer molecule, by changing the structure of the polymer repeat unit, by making
`copolymers, and/or by making polymer blends (Brauneral. 1992; Berggren et al. 1994,
`Akcelrud 2003).
`A numberoffactors influencethe efficiency of this EL process (Patel et al. 2002). Barriers
`to injection result from the mismatchesof the anode and cathode work functions with the
`
`PPV
`
`OCH,
`MEH-PPV
`
`Polyfluorene
`
`Figure 15.1 Example light-emitting conjugated polymers
`
`SAMSUNG Ex. 1013 - 5/30
`
`SAMSUNG EX. 1013 - 5/30
`
`
`
`
`OLED DISPLAYS ON PLASTIC
`
`Electrons
`
`'
`
` V-@,-9)
`
`:
`Electrons
`V—-@,- ) |
`4
`v
`wet A
`ts
`jem—
`aa z
`E,
`;
`
`| _ HOMO
`Holes
`|
`:
`Holes
`yy—a)_
`yp “
`YY rn
`“Min
`Anode HITL
`LEP
`Anode
`LEP
`To
`aie
`()
`(a)
`Illustrative band diagrams of(a) single-layer and (b) two-layer PLED diodes undera bias
`Figure 15.2
`V with LEP energy gap E,, cathode work function ec. and anode work function Dy
`
` 288
`
`Cathode
`Ca or Mg
`
`Cathode
`Ca or Mg
`
`low work function cathode. While exciton binding energies are high, not all excitons
`radiatively decay. Excitons can be in either a singlet or triplet state according to spin
`Statistics. Spin conservation only allows radiative decay from the singlet state (fluorescence).
`Radiative decaycanresult from the tripletstate (phosphorescence)if one ofthe spin states of
`the electron-hole pair were to flip, a process that can occur through interactions with
`impurities or defects: in general, however, phosphorescent emission is orders of magnitude
`smaller than fluorescent emission. If exciton formationis spin-independent, there is only one
`singlet state for every three triplet states, and the maximum efficiency obtainable is 25%.
`However, there are some polymer systems where exciton formation is spin-dependent and
`singlet fractions are as high as 60% (Wilsoneral. 2001) — the efficiency could potentially be
`
`fraction that do depends on the polymerandits purity,
`Twoother factors that can affect efficiency are preferential injection of one carrier over
`another and significant differences in the hole and electron mobilities. The hole mobility is
`typically muchgreater than the electron mobility, so in the single-layer device architecture,
`excitons will tend to form near the cathode where they nonradiatively decay due to image
`force interactions. To prevent this, it has become customary to insert a hole injection and
`transport layer (HITL) between the anode and the LEP, as shown in the band diagram of
`Figure 15.2(b)
`(Yan
`ecger
`[¢ ae ;
`Purposes:
`tacreases the likelihood of recombination in the LEP with the proper choice of HOMO and
`LUMObyblocking and localizing holes in the active layers ofthe device; (2) it can increase
`hole injectionat the anode by lowering the barrier; (3) andit can reduce leakage current and
`the prospects for Shorting by planarizing the rough, potentially spiked ITO surface (Brown
`et al. 1992; Heeger er al. 1994; Sheats e¢ al. 1996; Patel
`ey al, 2002), Polyethylene
`dioxythiophene/polystyrenesulfonate (PEDOT:PSS) and polyaniline (PANI) are two poly-
`mers Commonly used for the HITL.
`
`
`
`SAMSUNG Ex. 1013 - 6/30
`
`SAMSUNG EX. 1013 - 6/30
`
`€
`
`
`PLASTIC SUBSTRATES FOR OLED
`
`289
`
`15.2.3 OLED Display Types
`
`The pixels of a passive matrix OLED display (PMOLED)are created by patterning the
`anode into columnsandthe cathode into rows. Light is emitted from an individual pixel by
`addressing its corresponding row and columnandapplying a current-controlled forward bias
`that will provide the desired luminance. To create an image, each line is sequentially
`addressed and briefly illuminated (Sempel and Buchel 2002). The illumination must be
`extremely bright in order to obtain the desired overall display luminance and increases with
`the numberoflinesin the display (the peak luminance required is proportional to the product
`of the number of lines and the average luminance). This means that high currents and
`yoltages are needed, which has a numberof consequences. (1) The transparent anode (ITO)
`lines need to be bussed with metal to reduce significant power losses in the electrode lines.
`(2) Highly efficient polymers are needed to mitigate the other powerdissipation factors in
`the display; this requirementis an order of magnitude greater for full-color displays because,
`compared to a pixel from a monochromedisplay of the same resolution, each color subpixel
`has one-third the area and will be addressed one-third of the time. (3) High-resolution and
`large-area PM displays are impractical because the power dissipation will be too large and
`the polymer emitters will degrade faster with the increased peak current demands. Compared
`to glass displays, the power issues are more acute for plastic displays: resistive losses are
`generally greater due to a greater resistance in the anodelines, andplastic substrates have a
`lower tolerance for the Joule heating from resistive losses. For these reasons, PMOLED
`displays on plastic are limited to a display size of ~1.5 in (Innocenzo 2002).
`Active matrix OLED displays (AMOLED), in whicha thin film transistor (TFT) circuit is
`placed at each pixel, overcome many of the PMOLEDproblems. The TFT circuit provides a
`controlled current source with storage so that each pixel emits continuously. This drastically
`reduces the peak currents, which in turn significantly reduces power dissipation and the
`efficiency and lifetime demands on the polymer emitters. Current control also makes
`accurate grayscale achievable, which is difficult in PMOLED. A blanket cathode can also
`be used in an AMOLED, which eliminates many of the cathode patterning challenges
`associated with PMOLEDs.
`
`15.3.1 Substrate Requirements
`
` 15.3 Plastic Substrates for OLED
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`SAMSUNG Ex. 1013 - 7/30
`
`Theplastic substrate suitable for an OLED display mustsatisfy numerous requirements: high
`optical quality, which means few defects and high transmission (> 85%) across the visible
`Spectrum to let
`the light out; substrate smoothness in the nanometer range to prevent
`protrusions into subsequent barrier and device layers and to provide a surface that will
`promote high-quality deposition of subsequentfilms; the ability to withstand processing
`temperatures, which are expected to be at least 150°C and possibly as high as 300°C for
`AMdisplays; a good barrier to moisture and oxygen: good dimensional stability so that the
`various patterned device layers can be aligned; good resistance to any chemicals used in
`processing; and low water absorption to minimize the consequent dimensional changes and
`
`the exposure of the device to moisture.
`
`SAMSUNG EX. 1013 - 7/30
`
`
`
` 290
`
`Even for the base film, commercially available films are not adequate and require process
`and/orrecipemodificationforimproving smoothness, increasing working femperature, and/
`Orstabilizingopticaltransmission(MacDonald2004). Table 15,| listsmostofthepotentially
`Table 15.1
`Plastic film candidate Properties
`Base polymer
`PET
`PEN
`PC
`PES
`PAR
`PCO
`Pl
`Acommercial name
`Melinex
`Teonex
`Lexan
`Sumalite
`Arylite Appear Kapton
`T, (°C)
`78
`120
`ISO
`220
`340
`330
`355
`|
`400-700nm(%)
`>85
`85
`>90
`90
`90
`91.6
`orange
`|
`CTE(ppm/°c)
`15
`13
`60-70
`54
`53
`74
`30-60
`CHE (ppm/%, RH)
`100
`10
`Waterabsorption (%)
`0.14
`0.14
`0.4
`L4
`0.4
`0.03
`1.8
`yi
`Young'smodulus(GPa)
`5.3
`6.1
`L7
`2.2
`2.9
`1.9
`2.5
`f
`Tensile strength (MPa)
`225
`275
`83
`100
`50
`231
`iq
`WVTR (g/m*perday)
`40
`1.8
`62
`84
`220
`54.
`OTR(cc/m? perday
`160
`5.5
`100
`1500
`589
`388
`
`
`
`
`Chemical resistance issues_issuesgood good issues issues good
`
`
`SuitableplasticfilmsWiththeirmaterialPropertiesthatarerelevanttomakingplastic OLED
`:
`displays.Theglasstransitiontemperature, T,, ISusedasaruleofthumbforthetemperature
`q |
`extreme that
`the film can Withstand without undergoing undesirable size changes (above
`thistemperaturethepolymermoleculesstart toMoveand wil| usually rearrange themselves
`into a lower-energy Structure because many of them are not at equilibrium below T,).
`However, there are heat stabilization techniques that enable films to have a working tem-
`Taking polyethylene terephthalate (PET)as an example, with ; he ci
`200dcandidate forAMOLED He
`TIMMS Been used in agreat deal ofresearch because
`:
`oMer properties (clarity,
`low coef
`resistance, and low waterabsorption)and the fact
`that it is the most readilyavailable and
`ficient of thermal EXpansion, good chemical
`economically reasonable film. This could be a viable candidate forPMOLEDwereit not for
`the unacceptable number of defects and the surface roughness ofcommercially available
`films (the payback for addressing these issues appears smal| given the availability ofother
`films, most hotably PEN),
`Plate 15.1 shows Some images ofdefects observed on commercial grade PET and the
`consequences they have On a single diode device: surface roughness canleadto poorimage
`
`Transmission at
`
`perature above Lye
`
`
`
`OLED DISPLays ON PLASTIC
`Because there is No plastic film that meets oris likely to meet all ofthese requirements,
`there is a great deal of development work focused on obtaining a multilayer composite
`substrate that will. The 4pproachis to start with an optical quality base film meeting the
`critical Specifications — such as high transparency, high working temperature, and 200d
`dimensional Stability — and add surface treatments for smoothing; Coatings for scratch
`tesistance and, ifneeded, for chemical resistance; and barrier layers.
`15.3.2
`Plastie Base Film
`
`
`
`
`SAMSUNG Ex. 1013 - 8/30
`
`SAMSUNG EX. 1013 - 8/30
`
`
`
` sane eerie
`
`(d)
`(c)
`(b) magnification of defeet; (c)
`(a) Defects on commercially available optical grade PET;
`plate 15.1
`grainy appearance, and
`substrate having poor surface quality shows low output,
`display made with PET
`on PET substrate with improved surface quality has improved
`(d) display made
`line defect;
`characteristics
`
`quality in addition to creating a risk for device shorting. This points out that optical quality
`is not any easier to achieve than other plastic film requirements. The importance ofthe
`additional material properties will become clear in subsequent sections.
`
`15.3.3 Barrier
`OLED devices are extremely sensitive to moisture and oxygen. Emission can be easily
`quenched whenorganic light-emitting materials are exposedto water, and the highly reactive
`low work function cathodes canbe easily corroded by moisture and oxygen. This meansthat
`the substrate mustprotect the device from the ingress of these materials. Glass substrates are
`essentially impermeable to moisture and oxygen, but plastic films providelittle protection. A
`barrier structure must
`therefore be deposited onto the base film. There is currently no
`instrumental method to measure the ultra low permeation rates required for OLEDs,but an
`estimate for the waler vapor transmissionrate (WVTR) has been made by calculating the
`amount of water necded to oxidize the reactive cathode (Burrows ef al. 2001). For operating
`lifetimes in excess of 10 000 h, it was determined that the barrier structure should limit the
`WVTRto about 10~° g/m? per day. For similar lifetimes, oxygen transmission rate (OTR)
`requirements have been estimated to be somewhere between 1073 and 107° ce/m? perday.
`These transmission rate requirements have made barrier development one of the biggest
`challenges in making OLED displays on plastic because the transmission rates of bare
`plastic films are six or more orders of magnitude greater than these values, and the most
`demanding requirements outside this arena are in the packaging industry, where the
`barrier structures provide only two to three orders of magnitude improvement at best
`(Chatham 1996).
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`SAMSUNG Ex. 1013 - 9/30
`
`
`
`SAMSUNG EX. 1013 - 9/30
`
`
`
` 292
`
`
`OLED DISPLAYS ON PLASTIC
`
`The fact that the most sensitive commercially available systems that can measure WVTR
`and OTR have detection limits at 5 x 10-3 g/m? per day and 5 x 1073 cc/m* per day,
`respectively, has presented the additional challenge of developing techniques to obtain the
`needed sensitivity. A methodthatis commonly usedis the Ca test (Nisato et al, 2001). There
`are variations of this test, but they all involve observing the optical changesthat a reactive
`metal layer undergoesasit oxidizes. An effort has been made to quantify this test and water
`vapor transmission rates for barrier films have been reported to be as low as 4 x 10° 7 g/m?
`per day using this technique. The Catest, however, cannot distinguish between moisture and
`oxygen permeation. Techniques that can make this distinction and that are truly quantitative
`are currently being pursued, and if they are successful, should accelerate the understanding
`and developmentofbarrier structures (Dunkel 2004; Vogt 2004). In contrast, the Ca test has
`the benefit of being able to distinguish between bulk permeation and permeation through
`defects, which is evidenced by spots in the Ca film. Thisis important because the reduction
`in permeation through barrier films developed in the packaging industry - typically single-
`layer thin film oxides — has been shown to be limited by transport through defects such as
`pinholes, grain boundaries, and microcracks (Chatham 1996). For this reason, the current
`barrier development efforts are either directed at creating dense defect-free films (Pakbaz
`2004; Snow 2004) or multilayer films that decouple the defects and create a long circuitous
`path for the diffusing species (Burrows ef al. 2001; Rutherford 2004; Yan 2004).
`
`15.3.4 Composite Substrate
`
`
`
`Finally, the base film, appropriately chosen coatings, and barrier structure must be integrated
`into a mechanically stable composite substrate. While flexibility may ultimately place the
`strongest demands on mechanicalintegrity, the demandsarestil] great even if the substrate is
`not incorporatedinto a flexed display. This is because the substrate must be able to withstand
`the temperature, humidity, chemical, and process conditionsthatit will be exposed to during
`device fabrication and device lifetime. Adhesion ofthe various layers is of particular concern
`and can be tested for each layer after being subjected to either ambient or accelerated
`conditions that simulate the anticipated exposure (O’ Regan 2003).
`Adhesion or peel tests can be performed using a standard tensile tester. A minimum
`adhesion or bondstrength specification can be established based on an understanding ofthe
`process conditions and product environments the substrate is expected to experience.
`Samples can be tested as fabricated, after exposure to either thermal cycling or combined
`high temperature and humidity, or after outgassing in vacuum, Submersion in boiling water
`is a truly accelerated condition for testing survivability to moisture that provides quick
`m
`Ch
`_
`oe
`HOW'S
`~ peer
`tests
`to determine if a series of surface ~
`treatments on top of Teonex Q65 PEN would provide a good surface for various barrier
`layers to adhere to (Hildner 2004). Thetests are for samplesas fabricated (or received) and
`for samples after being immersed in boiling water for 2h. Each sample was cut into 4
`2.5 in x 4in strip. Two pieces of 4.5 in x 0.5 in testing tape (3M 4905/Foil) were attached
`onto the center ofthe strip and the excess length of tape was folded over to make a tab. Care
`wastaken to avoid wrinkles and excessive air pockets during tape attachment and then the
`strip was put through the heated rollers (75 °C) of a dry film laminator to remove all aif
`pockets and set the tape. The back side of the strip was mounted onto a German rotating
`wheelfixture of an Instron peel tester with 1 in double-sided tape (1 in Permacel double-
`
`
`
`SAMSUNG Ex. 1013 - 10/30
`
`SAMSUNG EX. 1013 - 10/30
`
`
`
`
`
`ure WVTR
`V per d
`per
`day,
`' obtain the
`301). There
`t a reactive
`t and water
`© 107 gfm?
`\oisture and
`‘
`: ae
`uantitati
`derstanding
`Ca test has
`.
`on through
`e reduction
`ally single-
`‘cts such as
`the current
`ns (Pakbaz,
`2 circuitous
`).
`
`2 integrated
`y place the
`substrate is
`o withstand
`:d to during
`lar concern
`accelerated
`
`. minimum
`iding of the
`experience.
`t combined
`viling water
`ides quick
`‘ of surface
`ous barrier
`ceived) and
`cut into a
`re attached
`a tab. Care
`nd then the
`
`ove all air
`ian rotating
`cel double-
`
`SUBSTRATE PROCESSING ISSUES
`
`
`293
`
`Minimum
`requirement
`
`
`
`DAsreceived
`BAtter Bw
`
`B3 B4|B1 B3 B4|B1 B3 B4]B1
`B2 B3 B4
`
`
`Barrier
`Surface treatment
`Figure 15.3 Adhesion of barrier layers on Teonex Q65 with various surface treatments; tests were
`performed on samples as received and after immersion in boiling water (BW)
`sided adhesive tape from Anderson Distributors), and the tape tab was clamped to the
`constant 90° angle peelfixture of the Instron. The tape was then peeled a minimumlength of
`1.5in at a speed of 2 in/min while the force required to peel the tape was monitored by the
`load weighting system ofthe Instron.Forthis test to be meaningful, the bondstrength of the
`tape to the sample surface, S,,, must be greater than the minimumspecification. If this
`criterion is met, then the tape will either peel off whenthe load exceedsS,, or earlier if one of
`the films or coatings in the sample has a weaker bond strength than S,,. An examination of
`the tape surface is required to see if any sample coatings peeled off, and adhesion failure is
`noted only if peel-off is observed and that it occurred al a load below specification. Figure
`{5.3 shows that the surface treatments under investigation provide the needed adhesion
`properties. Interestingly, some samples required a greater load after boiling water; this is
`probably due to a change in the surface character of the sample, which in turn changed the
`tape adhesion properties to increase Sts.
`Adhesion remains a mechanical integrity concern when the substrate is flexed or bent. An
`additional concern during bendingis the developmentof cracks in the coatings, particularly
`in the barrieras this will lead to increased permeability. Contributing to these failures are the
`internal stresses of the various layers, which will be introduced as the various materials —
`which have different thermal and mechanical properties — are processed together. One of the
`objectives in developing these composite substrates will
`therefore be to control
`these
`stresses. The critical strain — the strain above whicha film will form a crack — of individual
`layers is another factor that will determine the mechanical reliability of the composite
`substrate and the extent to which it can bend. A considerable amountof study is being
`directed at developing testing methods for understanding these issues (Nisato 2004; Bouten
`2002; Gorkhali et al. 2003).
`
`15.4 Substrate Processing Issues.
`15.4.1
`Processing Issues
`
`Conventional photolithography is used to patlern manyof the layers of PM and AM OLED
`displays on glass, and can be used on plastic as well, but the glass processes cannot be
`
`f
`
`|
`|
`|
`HI
`i
`.
`y
`y
`{
`|
`|
`i
`i]
`|
`i
`4
`|I
`
`i
`
`|
`|
`il
`
`.
`|
`
`1
`
`|
`
`|
`
`
`
`SAMSUNG Ex. 1013 - 11/30
`
`SAMSUNG EX. 1013 - 11/30
`
`
`
`
`
`294
`
`OLED DISPLAYS ON PLASTIC
`
`directly transferred to plastic. Internal stresses will be present in the deposited films, and
`they will probably be of a different character from those on glass substrates because during
`deposition, which is usually at elevated temperature, the plastic will experience significant
`dimensional change. These stresses can havea sizable influence on theflatness of a plastic
`substrate and may also lead to film adhesion issues. Furthermore,
`in the remaining
`photolithography steps — these include coating, baking, exposing, developing, and stripping
`of photoresist, and etching of deposited films — the substrate will be exposed to heat and
`solvents, which will have a significant impact on the flatness and dimensional stability of the
`substrate. Another nontrivial issue is that current flat panel display tooling is designed for
`glass and is therefore not equipped to handle substrates that flex.
`
`15.4.2
`
`Film Stress
`
`There will be stress in deposited films from the differences in thermal expansion betweenthe
`substrate and film (thermal stress) and from the microstructure of the deposited film
`(intrinsic stress) (Thornton and Hoffman 1977; Leterrier 2003). When a deposition is
`performed at elevated temperature,
`the difference between the coefficient of thermal
`expansion (CTE)of the film and the CTE of the substrate will result in a different amount
`of contraction during cooling, imposing a compressiveortensile stress in the film. On plastic
`substrates, which have higher CTEs than most films of interest,
`this thermal stress is
`typically compressive. The intrinsic stress, on the other hand,
`is determined by the
`deposition process and is thus largely independent of substrate type and film thickness.
`The physical and chemical vapor deposition processes typically used for deposition are
`nonequilibrium in nature and will lead to a quenched disordered state in the film with an
`intrinsic internal stress that can beeithertensile (resulting from attractive interactions across
`nanovoids) or compressive (as a result of high atomic density).
`Internal stresses (both thermal and intrinsic) lead to curling of the substrate, a phenom-
`enon that has been modeled extensively. According to the classic one-dimensional model of
`Stoney (1909),the radius of curvature of the coated film, X,is related to its internalstress, o;,
`by the expression
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`The internal str
`of a film. The coat
`it depends on grov
`to the surface mo;
`In addition to
`substrate will outg
`and/or controlled
`performance char:
`
`15.4.3 Dimensic
`
`The dimensional
`lithography steps
`patterned layers a
`temperature cycli
`molecular relaxati
`the temperature ay
`CTE of the subst
`
`repeated strain ex
`stresses in the sub
`size when it rett
`
`compounding effe
`quite large unless
`introduced into thr
`used to remove thi
`low shrinkage.
`Plastic substrate
`in subst