Novel concept for full-color electronic paper Kars-Michiel H. Lenssen (SID Member) Patrick J. Baesjou (SID Member) Frank P. M. Budzelaar Marc H. W. M. van Delden Sander J. Roosendaal Leon W. G. Stofmeel Alwin R. M. Verschueren (SID Member) Jack J. van Glabbeek Johan T. M. Osenga Roland M. Schuurbiers

Abstract — Despite a steep increase in commercial devices comprising paper-like displays, a much desired feature is still missing: bright full-color electronic paper. A new reflective-display technology has been developed to solve this issue. For the first time, the principles behind this in-plane electrophoretic technology will be presented, which enables the realization of full-color reflective displays with a higher brightness than presently available e-paper technologies, without compromising paperlike properties such as viewing angle and ultra-low power consumption. An additional major advantage (e.g., for future low-cost manufacturing) is that, besides direct-drive and active-matrix configurations, a passive-matrix option with analog gray levels has been successfully developed.

Keywords — Electronic paper, reflective display, electrophoresis, passive matrix, color, digital signage. DOI # 10.1889/JSID17.4.383

1

Introduction

Electrophoretic e-paper seems well on track towards acceptance by the general public in mass applications. New products are announced1 and the number of e-readers is steadily increasing, with the Amazon Kindle being the last addition.2 Nevertheless, an essential feature is still missing: bright full color, which could truly replace (color-) printed paper. Recently, color options have been presented by several companies,3 but in our opinion further improvements are still needed to fulfill the requirements of certain applications in the future (e.g., replacing paper signage). Some proposed options are based on the use of a colorfilter array,4 but besides the costs and loss of resolution, this either limits the brightness of a reflective display to not much more than a third of the incident light, or the saturation of the colors is sacrificed (e.g., in the case of RGBW filter arrays). For a higher brightness, intrinsic colors are required, i.e., by means of colored particles,5 colored reflectors,6 or colored inks.7,8 However, if subpixelation is used, the resulting brightness of such a full-color display is still similar to technologies using a color filter. Theoretically, a significantly higher brightness could be obtained by stacking differently colored layers, but for this it is required that each layer can be switched to a highly transparent state, which is not the case for electrophoretic technologies in which the particles switch perpendicularly to the plane. Moreover, if active matrices or optical addressing would be required,8,9 this would make the stacking option costly or impractical for many applications.

A different concept of color electrophoretic technology based on microcapsules containing particles of three different colors was reported a few years ago.10 While this could be suitable for black and white plus accent color, as was demonstrated in test pixels, it is not clear how this concept could be extended to a bright full-color display in a practical manner. Moreover, realizing reproducible gray levels is notoriously difficult in perpendicular-to-the-plane switching electrophoretic cells. As a result of several years of development of a solution for bright color electronic paper, in this paper the first multi-color reflective matrix display will be presented that does not need subpixelation, stacking, or time-sequential color schemes, and which is based on in-plane electrophoretic e-paper technology.

2

In-plane electrophoresis

In-plane electrophoresis is a very favorable technological option for bright, color electronic paper because the optically active elements (i.e., colored particles) can be moved out of the viewing path completely. For that purpose, a small percentage of each pixel is covered by a black mask, where particles can be compacted and hidden. It is also very advantageous that this technology provides one extra color because the reflector color can be chosen freely or, alternatively, without a reflector, the panel can be switched to a highly transparent state, also providing options for combination with a backlight.

Extended revised version of a paper presented at Display Week 2008 (SID ‘08) held May 20–23, 2008 in Los Angeles, California. K.-M. H. Lenssen, P. J. Baesjou, F. P. M. Budzelaar, M. H. W. M. van Delden, L. W. G. Stofmeel, and A. R. M. Verschueren are with Philips Research, High Tech Campus 34, Mail Stop WB31, Eindhoven, NL-5656 AE, The Netherlands; +31-40-274-7560, fax –6330, e-mail: [email protected]. S. J. Roosendaal is with Honeywell Aerospace, Brno, Czech Republic. J. J. van Glabbeek, J. T. M. Osenga, and R. M. Schuurbiers are with MiPlaza, Philips Research, The Netherlands. © Copyright 2009 Society for Information Display 1071-0922/09/1704-0383$1.00

Journal of the SID 17/4, 2009

383

3 3.1

Passive matrix Gating

For many applications it can be a big advantage if a passivematrix configuration can be used because a passive matrix is less costly and significantly simpler to fabricate compared to an active matrix. This makes it more suitable for flexible-display applications and roll-to-roll manufacturing on a shorter term. For passive-matrix driving, a threshold is needed: it should be possible to write one row of pixels, while the other rows (that are also exposed to part of the writing voltage) remain unchanged. This threshold can be an intrinsic threshold, i.e., determined by material properties of the particles and/or substrates, but this is not easy to achieve in electrophoretic technologies and generally increases the required driving voltages significantly. We found an alternative way to create a threshold that does not depend on material properties and therefore leaves much more freedom in the choice of materials: an electric gate. The in-plane electrophoretic technology allows us to add an extra electrode between the viewing area of the pixel and the collector area (hidden by the black mask). When a repulsive voltage is applied to this electrode, it acts as a gate and no particles can move in or out the viewing area. On the other hand, if there is no repulsive voltage on the gate electrode, particles can move across freely, in a direction depending on the potential difference between collector and viewing area. The gate electrode also provides the effect of bistability: if a small repulsive voltage (e.g., 2 V) is applied to all gate electrodes in a panel and there are no other potential differences, then the panel is effectively in a hold mode. Maintaining the repulsive voltages requires only extremely small currents and therefore power consumption is ultra low in this mode (roughly 4 nW/cm2) and thus this state effectively mimics bistability.

3.2

Evolution

The optical state of a pixel is directly determined by the number of particles within the viewing area. Therefore, in order to program a pixel it is not necessary to move the particles from the collector electrode (i.e., the electrode hidden by the black mask) completely towards the other side of the pixel, but it is in first instance sufficient to move them merely from the collector across the gate. Since this distance is much smaller than the pixel size, the time needed to program all pixels in a display is significantly reduced (over a factor 10 if this distance is one fifth of the pixel size). While this programming step is done per row, the subsequent so-called evolution step (in which the particles that

Spreading

Another important ingredient of our color e-paper technology is the concept of spreading. Instead of always collecting and holding particles on electrode surfaces, in the viewing area of the pixel particles are kept essentially between electrodes. In the first place this makes it much easier to spread particles over the viewing area in a controllable manner: while a single large electrode covering the viewing area would create an equipotential plane, now a voltage gradient can be created in the viewing area between the two view electrodes. Secondly, the transparency (e.g., brightness) of the electronic paper is improved because only a small fraction of the viewing area is covered with electrode material. (This even opens up the possibility of using non-transparent electrode materials.) Last, but not least, it improves a property of passive matrix that can be an issue for some applications: the relatively long update time (as will be explained in the next section).

384

3.3

Lenssen et al. / Novel concept for full-color electronic paper

FIGURE 1 — The four phases of the passive-matrix driving scheme, respectively reset, program, evolution, hold (from top to bottom); the bars indicate applied voltages.

arrived in the viewing area are spread between the view electrodes) can be done for the entire panel at once, and thus the time involved is independent of the number of rows.

3.4

Driving scheme

The resulting driving scheme for the passive matrix consists basically of four phases, as illustrated in Fig. 1. The first phase is a reset, in which all particles are collected at the collector electrode, simultaneously for all pixels; subsequently all gate voltages are increased. In the second programming phase, the gate voltage in all pixels in a specific row is lowered and depending on the desired optical state a potential difference between the collector and the first view electrode is applied so that in certain pixels particles can cross the gate. This programming step is repeated for all rows in the display. The third phase is the evolution phase for the entire panel at once. Finally, when the desired optical image is achieved, the panel can be switched to the hold phase with a small repulsive voltage on all gate electrodes, as described above. Note that this is a simplified example for illustration purposes; of course, many optimizations are possible, e.g., multiple-row addressing, etc.

4

Experimental

All demos described in this paper were fabricated on glass substrates using optical lithography, but care was taken to make a future transition to roll-to-roll plastic manufacturing as easy as possible. The electrodes were made of indium-tinoxide (ITO) and the pixel walls of SU8 photoresist. The distance between the substrate and cover plate (cell gap) was 10–18 µm, and the pixel size 500 × 500 µm (corresponding to 50 pixels per inch). The demonstrators have been realized with a passive-matrix array of 100 × 100 pixels. The reflector consisted of standard copier paper, which was placed behind the glass panel. Demonstrators were filled with suspensions that were developed in-house using commercially available chemicals. As the dispersant Isopar-G (Exxon Mobile) or dodecane (Aldrich) was used. Suspensions consisted of either milled pigment particles with a mixture of stabilizers (surfactants) and charging agents (such as the magenta suspension used in Fig. 2), or dye-colored stabilized polymer particles with an added charging agent (such as the black suspension shown in Fig. 2). All applied voltages were 10 V or less. In a typical driving scheme, the reset phase with a 10-V potential difference applied between the collector and view2 electrode will last 120 sec. Writing the pixels with a 5-V potential difference between the collector and view1 electrode will take 20 sec per line. The evolution phase with 10 V applied between the gate and view2 electrodes will last 80 sec. In the hold phase, 2 V on the gate is sufficient to keep the particles in the collector and viewing area separated.

FIGURE 2 — Photographs of 100 × 100-pixel passive-matrix panels demonstrating gray scale.

The focus of the driving of the first lab demonstrators that are discussed in this paper was on low voltages and not at all on update speed. Much shorter update times are certainly possible, e.g., by improved suspensions, higher driving voltages, and/or smaller pixel sizes. The resulting update time will then become similar to those of other electrophoretic technologies (i.e., on the order of seconds), as was already confirmed experimentally in segmented demo panels.

5

Brightness

The demos indeed showed the expected superior brightness compared to commercially available electrophoretic paper based on perpendicularly switching particles. Optical measurements were performed with a goniometer (Autronic Melchers DMS 803) under collimated perpendicular illumination and a spectral detector at a 45° polar angle. The brightness values are integrated over the measured spectrum according to the eye sensitivity (CIELAB 1931), and consequently normalized with a lambertian reference (Labsphere Spectralon). Although the design of these first demos was conservative (resulting in an aperture of only about 75%), a high brightness of 48% was measured. It is expected that this can still be further improved by optimizing the

Journal of the SID 17/4, 2009

385

design, and also by using better suspensions or improved reflectors. The measured contrast ratio was 7:1, comparable to the electrophoretic panel of a commercial e-reader. Computer simulations indicated that a white-state reflectance as high as 57% with a contrast ratio of 10:1 should be possible.

6

Gray levels

Gray scale is proven to be a strength of our color-electronicpaper concept. Figure 2 shows that our passive-matrix approach can deliver a multitude of gray levels, both with black and magenta suspensions. In contrast to the delicate gray states in conventional electrophoretic paper (that depend on the precise meta-stable positions of particles), in our concept the gray states are well-defined and directly determined by the number of particles in the viewing area. This number of particles in its turn can be precisely controlled by the number of particles that are allowed to cross the gate (during the program step of Fig. 1). Of course, good reproducible gray scales are also a prerequisite for full-color electronic paper. Figure 3 shows the distribution of gray levels that were measured as reflectance R (at 555 nm) on the magenta panel (of Fig. 2 bottom), and subsequently scaled

FIGURE 4 — Photographs of 100 × 100-pixel passive-matrix panels demonstrating gray-scale multi-color images, without subpixelation or stacking.

into lightness values L* ≡ 116 R1/3 – 16L to better match with human perception. The obtained distribution of gray levels is not perfect, since it cannot be expected from the manually assembled demo panel to show the uniformity and zero defects of a factory-produced display. Nevertheless, the achieved L* range of gray levels, and the median separation between consecutive gray levels indicate that 32 distinguishable gray levels (5 bits) are feasible. In addition, Fig. 3 shows that at the scale of individual pixels, the gray levels are truly analog in nature (and do not result from dithering). This demonstrates that the evolution step is capable of distributing the particles evenly throughout the pixel.

7

FIGURE 3 — Distribution of gray levels as measured on a magenta panel, expressed in L*. Visible is also the truly analog nature of the gray levels.

386

Lenssen et al. / Novel concept for full-color electronic paper

Color

In-plane electrophoretic technology provides several options for color electronic paper. Obviously, a color filter could be added to a black panel (with a white reflector) for a straight forward full-color panel. The superior brightness would actually result in an improvement compared to the color electrophoretic panels demonstrated thus far, but this solution would not fulfill our goal of a bright full-color display because of the subpixelation. A much better option is to use a subtractive color scheme and stack layers with a yellow, cyan, and magenta

FIGURE 5 — Schematic illustration of a full-color concept for electronic paper that also provides an excellent black state and white state.

suspension on top of a white reflector. This is possible since every layer can be switched to a transparent state, and therefore all color combinations can be realized. Besides, in-plane electrophoresis has a unique feature that is an excellent attribute for color electronic paper: it is possible to control more than one type of particles independently in a single layer of suspension. This means that a bright multi-color reflective display can be formed in a single layer, without the need for subpixelation or stacking. This approach is actually confirmed by demonstrators that have been realized. In Fig. 4, a photo is shown of a passivematrix demonstrator based on a suspension with both cyan and orange particles. Because these differently colored particles have different charges, they can be controlled independently. As can be seen, also mixed colors of cyan and orange can be obtained without problems. The layout of this multi-color panel is similar as described earlier, except for an additional gate and collector electrode. To the authors’ knowledge, this is the first multi-color reflective matrix panel without subpixelation, stacking or color sequential driving; it really enables “any pixel any color.” In the authors’ opinion, an elegant solution for a fullcolor electronic paper is shown in Fig. 5 consisting of two layers with each two colors of particles. Excellent white is obtained by the white reflector (while both layers are switched to transparent), excellent black by the black particles, and all colors can be made (throughout each pixel!) by combinations of cyan, magenta, and yellow particles. This is the closest as an electronic-paper technology ever got to printed paper: locating cyan, magenta, yellow, and black particles on top of white paper in order to realize full-color images. This innovation truly enables the substitution of printed paper by electronic paper, not only in books but also, for example, in signage applications.

8

Conclusions and discussion

For the first time, a new reflective-display technology was reported, based on in-plane electrophoresis to control colored particles. This approach combines several advantages which enables the realization of bright full-color electronic paper. Firstly, it allows a subtractive color scheme to realize

full-color without compromising brightness or viewing angle. Secondly, it can produce accurate analog gray levels. Thirdly, it allows passive-matrix addressing which is beneficial for future low-cost manufacturing on a shorter term. Further, it consumes an ultra-low amount of power. The above advantages are demonstrated by actual working prototypes, showing 100 × 100-pixel passive-matrix addressing with over 30 gray levels with black or magenta particles. Finally, as a next step and novelty, a cyan–orange dual-particle panel was demonstrated that produces bright colors in a single layer, including orange, cyan, green, white, and all intermediate gray levels, with passive-matrix addressing.

Acknowledgments The authors would like to thank H. Regina for his contribution to the development of the electrophoretic suspensions, as well as M. H. Gökgürler and F. J. Vossen for their help with the development of the electronics.

References 1 See, e.g., E. Huitema, “Rollable displays – The start of a new mobile device generation,” Proc. 2008 Flexible Electronics & Displays Conference, presentation 4.2 (2008); B. Chan, “Microcup structural response to normal axial loading,” Proc. 2008 Flexible Electronic & Displays Conference, presentation 18.2 (2008); http://delphi.com/news/pressReleases/pr_2008_01_03_001/ and http://www.phosphorwatches.com. 2 S. Levy, “The future of reading,” Newsweek (Nov. 26, 2007) (http://www.newsweek.com/id/70983); http://www.amazon.com/kindle. 3 K.-M. H. Lenssen, M. T. Johnson, and G. Zhou, “Technologies for electronic paper displays,” Proc. Intl. Congress on Imaging Science 2006, 345–348 (2006), and references therein. 4 See, e.g., A. Bouchard, H. Doshi, B. Kalhori, and A. Oleson, “Advances in active-matrix color displays using electrophoretic ink and color filters,” SID Symposium Digest 37, 1934–1937 (2006); R. Sakurai, S. Ohno, S. Kita, and Y. Masuda, “Color and flexible electronic paper display using QR-LPD technology,” SID Symposium Digest 37, 1922–1925 (2006); “LG. Philips LCD develops world’s first flexible color A4-size E-paper,” Press release LG.Philips LCD, 13-05-2007; J-K. Lee, C.-D. Kim, I. Kang, and I.-J. Chung, “A flexible color A4 size e-paper fabricated on a thin metal foil,” Proc. IDW ‘07, 1291–1293 (2007); “Samsung Shows at SID 2007: 70-in. 120-Hz Full-HD LCD, Flexible Color E-Paper Display, and 40-in. LED-backlit, Press Release from Samsung, 19-05-2007; N.-S. Roh, W. Lee, W. S. Hong, T. H. Hwang, S. J. Kim, S. I. Kim, and P. Shin, ”Development of flexible E-Paper and its applications," Proc. IDW ‘07, 1295–1297 (2007). 5 A. Baba, S. Sunohara, and T. Kitamura, “Electrodeposition of microcapsule containing electrophoretic pigments for electrophoretic display,” Proc. IDW ‘05, 911–914 (2005). 6 T. Endo, T. Soda, S. Takagi, H. Kitayama, S. Yuasa, E. Kishi, T. Ikeda, and H. Matsuda, “Color in-plane EPD using an anisotropic scattering layer,” SID Symposium Digest 35, 674–677 (2004). 7 X. Wang, H. Zang, and P. Li, “Roll-to-roll manufacturing process for full color electrophoretic film,” SID Symposium Digest 37, 1587–1589 (2006). 8 B. J. Feenstra, R. A. Hayes, R. van Dijk, R. G. Boom, M. M. H. Wagemans, I. G. J. Camps, A. Giraldo, and B. v.d. Heijden, “Rapid switching in multiple color active matrix driven electrowetting displays,” Proc. IDW ‘05, 861–864 (2005). 9 H. Harada, M. Gomyo, Y. Okano, T. Gan, C. Urano, Y. Yamaguchi, T. Uesaka, and H. Arisawa, “Full color A6-size photo-addressable electronic paper,” Proc. IDW ‘07, 281–284 (2007). 10 S. Sunohara and T. Kitamura, “Optical characteristics of EPD utilizing electrophoretic colored particles,” Proc. IDW ‘05, 907–910 (2005); T. Yamamoto, D. Takahashi, S. Nakamura, and T. Kitamura, “Color toner display based on control of color particle movement,” Proc. IDW ‘05, 899–902 (2005).

Journal of the SID 17/4, 2009

387

Frank P. M. Budzelaar obtained his M.Sc. degree in electrical engineering from the Technical University of Eindhoven in 1988. He worked in seve r a l p r o je c ts as a h ar dware a n d software architect. In 2000, he joined the Philips Research Laboratories, Eindhoven, The Netherlands, as senior scientist. His contribution to the project is in designing the electronics for the passive-matrix driving of the electrophoretic devices.

Marc H. W. M. van Delden received his B.Sc. degree in environmental chemistry in 1986. After fulfilling his military obligations, he joined Philips Research in 1987 and has since worked on the dry-etching of semiconductors, the growth, characterization, and analysis of CZ-grown monocrystalline and wet-grown polycrystalline ferrites, LPCVD-grown thin-film ferrites, and rubbing noise in magneticrecording heads, MCM-D modules for mobile, and Q&R of a-SiNx:H thin films for passive integration. Since 2005, he has been involved in the characterization, driving, and design of electrophoretic dispersions and devices. Alwin R. M. Verschueren received his M.Sc. degree in applied physics in 1997 from Delft University of Technology, The Netherlands. Afterwards, he was a visiting scientist at the RIKEN institute in Japan. In 1998, he joined Philips Research Laboratories in Eindhoven, as a research scientist involved in liquid-crystal displays. Since 2003, his core contribution to the project has been in controlling the particle motion by electrode and waveform designs.

Jack J. van Glabbeek received his degree in analytical chemistry from the HLO in Eindhoven in 1989. He has been with Philips Research since 1986 and has been working in the field of thinfilm deposition, etching and lithography for various projects such as thin-film magnetic heads, large flat thin displays, and is presently working as a process engineer on electrophoretics.

Johan T. M. Osenga received his degree in analytical chemistry from the MLO in Eindhoven in 1990. He joined Philips in 1991 as a LCD development engineer at its FPD plant. From 1997 until 2000, he was a member of Philips PolyLED. He is currently engaged as a process engineer at Philips MiPlaza in Eindhoven.

388

Lenssen et al. / Novel concept for full-color electronic paper

Roland M. Schuurbiers received his degree in mechanical engineering from the M.T.S. (Catholic) in Eindhoven. He joined Philips Research in 1986. The largest part of his work is related to mechanical aspects for several flat displays. He is presently working as a process engineer in the DPF department (Device Processing Facilities), which is a part of the MiPlaza facility.

Sander J. Roosendaal studied experimental physics in Utrecht (The Netherlands) and Lund (Sweden). He obtained his Ph.D. in surface science from Utrecht University in 1999. He joined Philips in 2000 and has worked in LCDs for mobile applications between 2000 and 2006. From 2005 to 2007, he was involved in the research described in this paper. Currently, he works at Honeywell Aerospace as a Program Manager.

Kars-Michiel H. Lenssen is a Director and Principal Scientist at Philips Research Europe, leading the color-electronic-paper project. He received his M.Sc. degree in applied physics from Eindhoven University of Technology in 1989. In 1994, he received his Ph.D. from Delft University of Technology and joined Philips Research as a Senior Scientist. In that function, he initiated and led research projects on (giant) magnetoresistance sensors and on MRAM. From 2002 to 2003, he worked as a Philips assignee in Arizona in the framework of the MotorolaPhilips-STMicroelectronics alliance. He holds 20 granted U.S. patents and is (co-)inventor on over 35 patent applications; he (co-)authored over 25 papers in international scientific journals. He also works as an evaluator of research projects for the European Commission and is a Fellow of the Institute of Nanotechnology and a member of the International Advisory Committee of several conferences. Patrick J. Baesjou is a senior scientist at Philips Research in Eindhoven, The Netherlands. He received his M.S. degree in chemistry from Utrecht University in 1992, and obtained his Ph.D. from Leiden University in 1997 on co-ordination chemistry and homogeneous catalysis. After a postdoc at Princeton University, he joined Philips Research in 2000 and started working on plastic electronics. Since 2002, his research activities focused on non-aqueous dispersions for electrophoretic devices.

Leon W. G. Stofmeel graduated in Technical Applied Physics from the University of Professional Education in Eindhoven. In 2001, he started as a research scientist at Philips Research Laboratories writing driver software for research on active-matrix displays. In 2002, he started the characterization of transflective active-matrix displays. Since 2005, he has been involved in the optics and colorimetry of electrophoretic panels. In addition to that, designing scientific grade mockups is one of his tasks as well.