Flexible Substrates and Packaging for Organic Displays and Electronics
Despite the performance challenges that flexible-display technology faces, the first commercial products are scheduled to reach the marketplace in the 2006–2007 time frame.
by Nicole Rutherford
AS ELECTRONIC DEVICES become ever-more integral to our everyday lives, displays are becoming an increasingly important component, providing quick and easy communications. Large companies – such as Philips, Sony, and General Electric – want to bring even more displays into the home and personal environment, but the rigid, square, glass-based display is not the ideal shape or form factor for integration into the human environment. Most consumers prefer curved shapes, a foldable or rollable form, and electronic accessories that can withstand multiple accidental "drop tests." The fact that the glass displays in today's personal digital assistants (PDAs) and notebook computers often do not survive these "drop tests" is seen as a cost of portability and mobility. But paper and clothing, for example, are flexible, eminently portable, and function well even after mechanical abuse. As designers look to new and different ways to set their products apart, electronic devices that are thinner, lighter, unbreakable, and not square in shape are considered distinctive and highly marketable.
Organic displays and electronics are among the technologies being developed to meet these needs. Substrates and packaging that can provide protection from the environment while providing the required robustness and flexibility are critical limiting factors.
Organic vs. Inorganic Materials
Materials that have been used to make conventional displays and in microelectronics are primarily, though not exclusively, inorganic. These inorganic materials are brittle, but are relatively mechanically and chemically stable at high temperatures and tolerate high levels of moisture, oxygen, and light.
Organic materials, especially polymers and plastics, have lower moduli of rigidity and high strength-to-weight ratios and are therefore more resistant to defects caused by mechanical bending or flexing. But organic materials have much lower heat resistance and less tolerance to UV light, particularly in the presence of oxygen or moisture.
To meet the demanding requirements for reliability, robustness, and lifetime, inorganic materials must be protected against oxygen and moisture. However, organic optoelectronic and electronic devices will require even more protection, although lifetime requirements might be shorter for disposable products. If organic components are to compete effectively in the marketplace, they must be protected against degradation from moisture and oxygen, and this requirement will become even more important with time.
The Drive to Lower Temperature
To fabricate a flexible display, the same processes developed for rigid displays can be used, and the glass substrate can be simply replaced by a flexible substrate, such as plastic. But current device-fabrication processes consist of many high-temperature steps usually related to photolithographic patterning, annealing of deposited layers, or deposition at high temperatures, which are used to achieve good electrical properties.
Initially, this conflict between process temperature and material properties led to demands for higher-temperature or higher-glass-transition-temperature plastics. A num-ber of suppliers responded by providing plastic films having impressive characteristics, but most of these failed because of undesirable chemical and mechanical properties, a lack of manufacturing maturity, or excessive cost. Market studies indicate that flexible displays will not support a premium over glass displays – at least not for any significant length of time.
To lower cost, one option is to use a low-temperature plastic film such as PET. Another advantage of lowering process temperatures is to reduce the effects of any thermal mismatch of thin-film structures that may contain both organic and inorganic materials. The larger thermal-expansion coefficients of many organic materials can cause stress to build as temperatures change, leading to cracking and delamination.
Although it is possible to account for reversible coefficient-of-thermal-expansion (CTE) effects, small but irreversible shrinkage can also occur with each high-temperature cycle, making pattern alignment a significant challenge. Another driver for the use of lower temperatures is process cost. It would simply be cheaper if more processing could be performed at or near room temperature.
Requirements for Flexible Substrates and Packaging
To realize flexible organic electronics and displays, the backplanes and frontplanes must have the required flexibility and must provide adequate protection against moisture and oxygen. In some cases, a flexible transparent conductor is also required. The options under consideration are thin glass, polymers (plastic films), and stainless-steel foil.
Plastic films with a barrier coating or inherent barrier properties are available for food and pharmaceutical packaging. The properties of these films are not targeted for display or elec-tronic applications, but might be adequate when moisture sensitivity, transparency, and surface quality are not critical. Typically, these films are not available with a transparent conductor.
Also available are films that are not targeted for barrier applications, but which have gained entry to the FPD market as polarizers, brightness-enhancement films, and touch screens. Manufacturers of these films know the industry and are more sensitive to the industry's optical and processing requirements.
For the most critical applications, the limitations of commercially available plastic films are insufficient barrier properties, dimensional instability with temperature and humidity, high coefficient of thermal expansion, solvent solubility, surface quality (roughness), and optical clarity. Some films score high in one area, but not in others. There is no one "winner" – as glass is. The suitability of each film depends on the type of display and the targeted application.
Thin glass (0.03–1.1 mm) is an extension of the trend to continuously reduce thickness and weight in rigid displays. The thin glass, coated with polymer so it can better tolerate handling and separation into individual displays, should have an advantage for conformable (if not flexible) displays. But even with a polymer coating, microcracks are prone to form from the edges, and the glass is easily broken. Thus far, polymer-coated glass has not been widely adopted as a first step toward flexible displays.
Stainless-steel foil is used in making flexible thin-film solar cells, using processes that have much in common with the high-temperature processes used to fabricate displays and thin-film transistors (TFTs) on glass. But stainless steel's opacity limits it to certain applications. Polishing and/or smoothing layers are required to reduce surface roughness, and additional layers are required to achieve electrical insulation between the device and the substrate and to form a barrier against contaminants from the steel.
Metal foil can go beyond the conformability possible with thin glass, but it has limited flexibility compared to plastic. Finally, as with thin glass, adding polymer-based layers to im-prove the stainless steel's properties limits its compatibility with high-temperature processing.
Flexible Emissive Displays
The current full-color high-resolution FPD technologies – TFT-LCD and plasma – are not naturally amenable to making flexible displays. Of the newer technologies, only organic light-emitting diode (OLED) technology is a potential competitor that could also conceivably be made flexible.
The shift of most high-information-content displays to active-matrix addressing has raised the hurdle for flexible technologies. To be fully flexible, compatible processes for both the transistor and the display must be developed. Active-matrix flexible OLED-display demonstrators have been developed. Seiko-Epson replaced low-temperature polysilicon (LTPS) with plastic in 2003; Universal Display Corp. (UDC) and Samsung SDI fabricated LTPS on foil in 2004 and 2005, respectively (Fig. 1); and NHK fabricated organic thin-film transistors (OTFTs) on plastic in 2005, but actual products could be 10 years away. It is more likely that the first successful flexible displays will be low-information-content displays targeting unconventional markets.
Samsung SDI Samsung SDI Universal Display Corp.
Fig. 1: A 4.1-in. 100 x 246 top-emission AMOLED on steel-foil substrate with Barix™ thin-film encapsulation was shown by Samsung SDI at SID 2005 (left and center). UDC's PHOLED™ displays on 100-μm flexible foil substrates prepared by Xerox PARC are packaged using Barix™ encapsulation (right).
A relatively small contingent of companies is addressing these unconventional markets. They are focusing on near-term flexible emissive displays, albeit for low-resolution shorter-lifetime applications. And, while not exactly flexible, ultra-thin OLEDs for the simpler backlight applications are also considering some of the same substrates as flexible.
OLEDs. In terms of packaging, the requirements of even simpler shorter-lifetime OLEDs remain the most challenging of all the display technologies. The substrate must have a high barrier to moisture and oxygen, a high-quality surface, good transparency, and a flexible transparent conductor with a sheet resistance 10–30 Ω/. The device could be encapsulated by laminating a second substrate (backplane) over the cathode, but a thin-film encapsulation should be used to avoid moisture permeation through organic sealants.
A thin – even thinner than the substrate – flexible plastic film can be laminated over the encapsulation for mechanical protection. Barrier requirements for the encapsulation are the same as for the substrate. Transparency is not required for bottom-emission devices.
Electrophoretic Displays. Another option where impressive flexibility has been demonstrated is the electrophoretic display (EPD) (Fig. 2). But even for low-information-content applications, EPDs require an active-matrix backplane. The lower-transistor-performance requirements (compared to OLEDs), however, provide a path for organic transistors. Because EPDs use reflected light, the backplane need not be transparent. This makes stainless steel and colored plastics an option.
Moisture permeation rates should be in the 10-1–10-3 g/m2/day range, and advanced electronic inks are more sensitive to moisture and oxygen, thus requiring permeation at the lower end of the scale. Organic TFT materials will also require protection from moisture and oxygen in order to pass reliability and lifetime tests.
For plastic substrates, the challenge will be the dimensional-stability requirements of the TFT-fabrication process. For the frontplane, the same moisture-permeation requirements will apply. Dimensional-stability requirements are relaxed, but the frontplane must be transparent. The requirements for a transparent conductor are moderate; they are even within the range of possibility for transparent organic conductors such as PEDOT or carbon nanotubes.
Bridgestone
Fig. 2: This Bridgestone electrophoretic flexible display was shown at FPD International 2004.
Eastman Kodak Co.
Fig. 3: Eastman Kodak's flexible cholesteric-LCD as seen at SID 2004.
Cholesteric Displays. Although the technology has been around for a long time, cholesteric liquid-crystal-display (Ch-LCD) technology continues to advance, targeting electronic-paper and label applications (Fig. 3). Active-matrix addressing is not required, making flexibility easier to achieve. Barrier requirements are moderate, but some type of barrier is required to prevent air bubbles from forming in the display. The substrate surface quality (smoothness) and sheet resistance of the transparent conductor are within the realm of what is available for low-temperature-film substrates. Since Ch-LCD technology is one of the few liquid-crystal technologies that do not require polarizers, the low-cost birefringent plastics, such as polyesters, can be used. Nematic LC and polymer-dispersed liquid-crystal (PDLC) technologies are two other liquid-crystal technologies with similar requirements that could be used for flexible displays.
Electrochromic Displays. The flexible-display technology that is first to the marketplace could be the electrochromic display. These small, simple displays are targeted for smart labels, packaging, and smart cards. Because of the high-volume low-cost requirements of the applications, manufacturers are focused on using existing printing technologies to fabricate the displays.
In terms of substrate and packaging requirements, the focus is on a low-cost thin package that is rather robust and particularly flexible. The backplane can be opaque and should include a conductor with relatively high sheet resistance. The frontplane should be transparent and provide mechanical protection for the display. Coated paper, plastic film, and metal foil are possible candidates.
Materials under Development
Sumitomo Bakelite and Teijin DuPont Films have demonstrated sustained efforts to supply plastic film suitable for flexible-display substrates, although General Electric, Teijin, Nitto Denko, and Promerus have all supplied film to researchers for evaluation.
For the base plastic film, the focus has been on improving surface quality and achieving low and predictable dimensional change as a function of temperature and humidity. For example, Teonex® Q65A polyethylene napthalate (PEN) from Teijin DuPont Films is inherently resistant to moisture absorption, manufactured with special controls to optimize surface smoothness and heat-stabilized to limit shrinkage to less than 0.1% when exposed to temperatures up to 180°C for 30 min.
Sumitomo Bakelite supplies not only base films, but also PES film with a gas barrier layer and ITO targeted for plastic LCDs. This company is a member of TRADIM in Japan – a consortium for researching advanced display materials for plastic electronics and displays – and has published at least one paper on an improved PES barrier film being developed within the project. The improved film has a low coefficient of thermal expansion (CTE) (14 ppm/°C, compared to 55 ppm/°C for standard PES) and maintains a high light transmission of 89%. The moisture permeation of the SiO2 barrier is in the 10-3 g/m2-day range.
Nitto Denko and Toyota CRDL have reported on the fabrication of OLEDs on epoxy substrates with an SiNx(barrier)/CNx:H (buffer) multilayer barrier [Akedo, Proc. IDW '04 (2004)]. The focus of this work was on the optimization of the surface smoothness to reduce leakage and bright spots in the display.
Vitex Systems, Inc., has a long history of developing high-barrier ITO-coated plastic substrates for flexible displays (Flexible Glass™engineered substrate), as well as high-barrier thin-film encapsulation materials and equipment for OLED displays (Barix™ Resin and Guardian™ Encapsulation Systems). The barrier technology is based on a multilayer organic–inorganic structure whose feasibility was first demonstrated for OLEDs in 2000 by Graff and co-workers at Pacific Northwest National Laboratory.
The first long-lived OLED made on a substrate utilizing laboratory samples of the material had an operating device lifetime (to halfluminance) of 3800 hours [M. Weaver, Appl. Phys. Lett. 81 (2002)]. Although this lifetime was not adequate for commercial displays, it demonstrated an order-of-magnitude increase in lifetime over previous barrier approaches.
Vitex Systems, Inc.
Fig. 4: These OLED pixels encapsulated with the Barix™ multilayer barrier structure easily meet telecommunications display specifications. The barriers for the different pixels have different numbers of organic–inorganic layer pairs, called dyads. Shown are N dyads at 1000 hours of testing (left), N-2 dyads at 1000 hours (center), and N-3 dyads at 500 hours (right). The sample shown at 500 hours is still in test.
Current OLEDs on glass encapsulated with the optimized Barix™ multilayer barrier have lifetimes that typically exceed the telecommunications display specification of 500 hours at 60°C and 90% relative humidity (Figs. 4 and 5), and commercial processing equipment is now available.
For flexible displays, the same barrier deposited on plastic film has viably packaged calcium metal for up to 1000 hours at 60°C and 90% relative humidity (Fig. 6). The combined frontplane and backplane permeation rate for calcium packaged with the multilayer barrier is calculated to be less than 10-6 g/m2-day.
The fact that such results are possible from a fast low-temperature process, with high yield, is explained by a model in which the barrier performance is dependent not on the series addition of each barrier layer but on the lag time through the barrier [Graff, J. Appl. Phys. 96(4) (2004)]. The model predicts that a 5-dyad barrier with a typical defect size and defect density can more than meet the requirement for OLED moisture permeation.
Recently, the General Electric Research Center presented results on graded barrier coatings of SiOxCy (organic zone)/SiOxNy(inorganic zone), and Dow Corning presented results on SiC alloy. For the General Electric barrier, OLED pixels reached 144 hours at 60°C and 85% relative humidity with no black spots.
Although most developers have made the transition to multilayer structures to improve yield and manufacturability, Symmorphix has demonstrated a single-layer aluminosilicate barrier on PC and PEN film. The most recent data shows that a permeation rate in the 10-5–10-6 range was achieved for a 150–200-nm barrier on PEN film. For a bending radius of 12 mm, no cracking was observed.
A more recent competitor to thin glass is thin stainless steel. Before it is usable for a display application, steel foil must be planarized and a dielectric buffer layer applied. The USDC has recently funded a contract to develop a production-capable design and process for flexible stainless-steel substrates.
New Manufacturing Concepts
In order for flexible organic displays to become ubiquitous, they must be inexpensive to produce. Some radical changes to current manufacturing processes are being considered, most notably printing and roll-to-roll (R2R) processes. No one believes that complex displays will be made solely by R2R or printing techniques, but significant progress is being made to make part of the display by leveraging these processes.
For flexible substrates, R2R processes are already being used to manufacture and coat plastic films. Further development of these processes to meet the cleanliness and defect levels required by the display industry is occurring. Although innovations will be required, the historical experience in this manufacturing technique should help to advance the technology.
Neither printing nor R2R techniques have any history with display manufacturers and makers of TFTs. This implies a more difficult transition. Yet, the R2R manufacture of thin-film solar cells on stainless-steel foil is already a reality. Ink-jet printing of polymer OLED materials has opened the way for the development of ink-jet printers for the fab that will specialize in printing electronic materials. Many other printing techniques are also being tried, such as screen printing, flexo, and gravure.
Of course, O-TFT developers see high-volume low-cost printing as the way to succeed in new applications where current inorganic TFTs are too expensive. The drive to use printing as a technique for manufacturing organic electronic products is causing governments to invest money into research consortia at a rapid rate. Consortia for organic or flexible electronics funded by government entities exist in Japan (TRADIM), Taiwan, Korea, the U.S. (Flexible Display Center at Arizona State University), and Europe (PolyApply).
Philips
Fig. 5: Philips made this monochrome 96 x 64 passive-matrix display on glass with Barix™ encapsulation.
One accelerant for the advancement of new, cheaper manufacturing techniques for electronics is the drive for radio-frequency identification (RFID) tags, helped along by large retailers and government security concerns. Although the challenges are significant, there is money and brainpower behind this technology.
The "First Wave"
Printed electronic products have now entered the marketplace in the form of flexible circuits made from conductive inks. There are a wide variety of applications, including interactive learning games and interactive in-store audio advertising from T-Ink; intelligent packages with passive RFID, microprocessors, and sensors from Cypak; electronic transdermal drug delivery patches from Travanti Pharm; and printable flexible batteries from Power Paper. Simple thermochromic displays also exist that change color upon application of heat and are typically used as indicators.
The flexible display that goes beyond a color indicator has not yet arrived. But significant progress in printable transistors and organic display technology has been made, laying the ground work for the first products to emerge in the 2006–2007 time frame. The task now is to manufacture them cheaply and reliably. The first products to incorporate thin, flexible displays will likely be conformable, electronically updatable labels and signage for advertising in retail stores and mass-transit systems. Fujitsu and Dai Nippon Printing are two large manufacturers who have targeted 2006–2007 as the release date for their first products. •
Fig. 6: For flexible displays, the same barrier used on the glass display shown in Fig. 5 has been deposited on plastic film and used to viably package calcium metal for up to 1000 hours at 60°C and 90% relative humidity.