Electrowetting Technology Aims to Improve on the Performance of LCDs for Mobile Applications

In a world where liquid-crystal displays have overcome seemingly every shortcoming they have been criticized for, is there room for another disruptive display technology? This article will examine the properties of a technology that has demonstrated a promising future since its inception in 2003 and can be manufactured using existing infrastructure.

by B. Johan Feenstra

WHY would anyone consider developing another novel display technology? After all, liquid-crystal displays (LCDs) are ubiquitous and customers seem quite happy with their performance. However, there are always elements that can be improved. For example, consumers have indicated that the function of their cell phones that they would most like to see improved is battery lifetime, which is markedly impacted by the display of the cell phone.1 Outdoor readability and improved color gamut are two other areas where LCDs used in mobile phones can be improved.

Figure 1 shows a reproduction of a slide that was presented during the SID 2005 Keynote Address given by Harold Hoskens (who at that time was head of Philips Mobile Display Systems). The slide illustrates his view on the seven most important parameters for displays in mobile handsets. The fill on the bars indicates the current status, while the arrows indicate the amount of improvement required to achieve a performance that is equal to the performance of a television. Obviously, all of the parameters require improvement, but none more than the optical parameters in outdoor conditions (indicated by the sun symbol). Also, the power consumption of current state-of-the-art displays is inadequate for the future. The trend toward "TV on Mobile" and more display-oriented services enhances this deficiency as the display becomes an ever larger percentage of the cell phone's power budget.

Johan Feenstra is the Director of Technology Platforms for Liquavista, High Tech Campus 48, Rm. 2-164, 5656 AE Eindhoven, The Netherlands; telephone +31-40-2742116, fax +31-40-2743695, e-mail: johan.feenstra@ liquavista.com

Fig. 1: Improvements required in future mobile handsets for the seven key display properties. Courtesy of H. Hoskens (TPO).

Figure 1 re-emphasizes why there is such a strong effort in the display industry to find a technology that can improve on these deficiences. Without a doubt, the performance of LCDs will continue to improve in the coming years. However, with a mature technology such as LCD technology, the improvements will be small rather than revolutionary as required.

There is room for a more radical approach, which led us to the development of electro-wetting technology. This well-patented display technology has progressed at a remarkable pace since its first appearance in the literature in 2003,2 advancing from single pixels demonstrating the principle at the time of public disclosure to the full-color 2.5-in. displays shown at the 2006 SID exhibition. This article will explore the properties that make electrowetting technology unique and will compare it to existing and emerging technologies.


In electrowetting, a voltage is used to modify the wetting properties of a material system between a hydrophobic and a hydrophilic state. Rapid progress in the performance of electrowetting has been achieved in the last 20 years due to improvements in materials and processing. As a result, in the past decade, electrowetting has been utilized for an increasing number of applications, including pixelated optical filters,3 fiber optics4, adaptive lenses,5,6 lab-on-a-chip,7 and curtain coating.8

In Fig. 2, the principle of a (reflective) electrowetting display is shown. Figure 2(a) shows the optical stack, comprising a (reflecting) electrode, a hydrophobic insulator, a colored oil layer, and water. In a display, these layers are sandwiched between glass or polymeric substrates. In equilibrium, the colored oil naturally forms a continuous film between the water and the hydrophobic insulator [Fig. 2(a)] due to the fact that this is the lowest energy state of the system. At the typical length scales used in displays (pixel sizes around or below 200 μm), the surface-tension force is more than 10,000 times stronger that the gravitational force. As a result, the oil film is stable in all orientations.


Fig. 2: Electrowetting-display principle.

When a voltage difference is applied across the hydrophobic insulator, the stacked state is no longer energetically favorable. The system can lower its energy by moving the water into contact with the insulator, thereby displacing the oil [Fig. 2(b)] and exposing the underlying surface.

The balance between electrostatic and surface tension forces determines how far the oil moves to the side. In this way, the optical properties of the stack when viewed from above can be continuously tuned between a colored off-state and a transparent or reflecting on-state, provided the pixel is sufficiently small so that the human eye averages the optical response.

The photographs show a typical oil retraction obtained for a group of pixels with a size of 160 x 160 μm2, confirming the 80% white area required for a 70% in-pixel color reflectivity. Part of the electrode is omitted in the upper-right-hand corner of each pixel to control the oil motion.9 The control of oil motion greatly improves the pixel-to-pixel homogeneity and, hence, display uniformity.

Mechanical Display Properties

Size: From a manufacturing point of view, the ability to scale electrowetting technology has been demonstrated by the rapid scale-up of the demonstrator size from the single-pixel level to the current 2.5-in. full-color displays. A further increase in display size is anticipated, in particular when the manufacturing process is successfully transferred from laboratory-sized substrates to large industrial substrate sizes.

Power: One of the most important advantages of electrowetting displays is their very low power consumption at high color brightness. In Fig. 3, the power consumption of existing and emerging video-capable technologies for full-color video content are summarized. For LCDs and OLEDs, the typical power performance of contemporary displays are indicated.

The transmissive/emissive technologies have high power consumption as generating light costs much in terms of energy consumed. For a typical LCD, the backlight consumes about 85–90% of the display power budget, while the power consumption of present OLEDs can be even higher.

Making use of ambient light therefore results in a reduction in display power consumption by a factor of 8–10. Analog gray scaling, which is possible with electrowetting displays, results in a much lower power consumption than displays that generate similar optical performance using pulse-width modulation (PWM), such as Qualcomm's iMoD technology.

Robustness: A major drawback of LCDs is their cell-gap dependence. Not only do customers notice this when touching the display, it also significantly complicates the manufacture of the displays. For electrowetting displays, the electro-optic curve shows no dependence on cell gap.In addition, the average cell gap is tens of microns, reducing the sensitivity during manufacturing even further. From a module point of view, this allows for the elimination of the frame that is now used for LCDs to fixate the cell gap, resulting in significantly thinner display modules.

Display Performance

Contrast Ratio: The contrast ratio of an electrowetting display, defined as the reflectivity in the on-state divided by the reflectivity in the off-state, is mainly determined by the optical absorption in the off-state. At present, the contrast ratio for a reflective architecture is typically between 6 and 10 and can be improved further by introducing a black mask at the non-active areas. Such a contrast ratio resembles that of paper, which is about 10–15. Because the display performance scales with the ambient lighting conditions, similar to paper, this contrast ratio delivers a much improved viewer experience compared to that of transmissive displays.

Resolution: The resolution of the current 2.5-in. QVGA electrowetting displays is about 160 ppi. Further improvement in the resolution is very beneficial to the effect, as the interfacial tension – the force that is controlled with electrowetting – becomes even more dominant with reduced pixel size. In addition, the response times of the pixel (currently about 5 msec on and off) will become even shorter for increased resolution.

Brightness: The basic optical switch, as illustrated in Fig. 2, has a high optical efficiency. The absence of polarizers provides an intrinsic advantage of a factor of 2 in optical efficiency over that of typical LCDs (without reflective pre-polarizers) in all possible modes (transmissive, reflective, and transflective). The viewing angle of electrowetting displays is naturally very broad (having little or no contrast-ratio or brightness reduction for angles between 0 and 70¼) when a diffuse Lambertian reflector is used. This can be used to achieve a very-wide-angle high-contrast viewing cone using either ambient lighting or lighting solutions provided by the system.


Fig. 3: Power consumption for a variety of video displays scaled per square inch (full-color video content).a

Color saturation: In most display technologies, including transmissive LCDs, the color-gamut performance can be altered markedly by narrowing or widening the spectral transmission of the color filters, resulting in an engineering tradeoff between light efficiency and saturation of the primary colors. For an electrowetting display, one can also exchange the brightness advantage mentioned above for a broadening of the color gamut when desired. Moreover, in the reflective mode, this color gamut is achieved in almost all lighting conditions (provided the ambient light is sufficiently "white"). This can be contrasted with the performance of the typical transflective LCD employed in contem-porary mobile handsets, where color gamut is often greatly reduced in outdoor conditions.

Product Platforms

ColorMatch: ColorMatch displays exploit the fact that by changing the color of the dye, a wide range of colors can be created. An example showing some of the available colors can be seen in Fig. 4. By mixing the different dyes, one can achieve intermediate shades. The color tunability provides a strong design freedom, e.g., making the display part of the packaging, as shown in Fig. 4. Also, the color can be "matched" to a desired colorpoint, e.g., company colors. The high reflectivity in the on-state combined with an unlimited viewing angle provides excellent readability, also at a greater distance.

The ColorMatch platform is expected to be launched in the first half of 2007 in simple seg-mented displays. At a later stage, this product line will be extended to active-matrix displays.

ColorBright: The ColorBright platform adds a color filter to the ColorMatch architecture and uses a black dye as the optical switch. The result is a display that has a greatly improved optical performance compared to that of an LCD. The similarity to an LCD in the build-up of a ColorBright display enables us to manufacture them on standard LCD production lines. Not only are the manufacturing processes nearly identical, but the process flow is also the same for both front- and back-end processing. Finally, we make use of standard components, including active-matrix backplanes, driver ICs, and color filters. As the technology continues to mature, more amplitude-modulated gray scales will be introduced into this product line, thereby further increasing its application scope.

ColorFull: The next step on the Liquavista product roadmap is an architecture that resembles the approach of color printing.10ColorFull displays have significant advantages in optical performance because it enables the creation of any color in any area. Clearly, this approach is more complicated from the module-manufacturing point of view, but most of the challenge lies in the engineering of the full module, not the basic principle.


Electrowetting displays have very favorable optical properties, combining a paper-like performance with video-speed switching speed. In addition, they can be manufactured using existing infrastructure. This implies that electrowetting displays present a disruptive technology from the user-experience point of view, while exploiting the existing LCD value chain.


The author would like to thank his Liquavista colleagues, the many students, and PRE technicians who helped pave the way for electro-wetting-display technology.


aThe value for LCDs was taken by using a typical power consumption for a 2.5-in. display of 250 mW. For the AMOLED, we used the power-consumption numbers from a 2.2-in. Kodak Nuvue display. Because active- matrix driving is still not often used in the OLED industry, not much data is available. In principle, OLED displays provide a power advantage over LCDs because of selective lighting. However, to date, OLED displays still consume significantly more power than LCDs of the same size. Furthermore, an emissive technology will always require light to be generated, and therefore requires significant power. For the iMoD technology developed by Qualcomm, the power consumption was calculated by ab-initio calculations, assuming a 6-bit gray scale by dithering and a higher frame rate. For electrowetting displays, we used the same calculations with amplitude modulation and our present swing voltage of 20 Vdc. Because we expect the driving voltage to be significantly reduced in the near future, the power consumption will also be significantly reduced as well.

1Based on results from a TNS Survey of 6800 users across 15 countries (2005).

2R. A. Hayes and B. J. Feenstra, Nature 425, 383 (2003).

3M. W. J. Prins, W. J. J. Welters, and J. W. Weekamp, Science 291, 277 (2001).

4P. Mach, et al., Appl. Phys. Lett. 81, 202 (2002).

5S. Kuiper and B. H. W. Hendriks, Appl. Phys. Lett. 85, 1128 (2004).

6B. Berge and J. Peseux, Eur. Phys. JE3, 159 (2000).

7M. G. Pollack, R. B. Fair, and A. D. Shenderov, Appl. Phys. Lett. 77, 1725 (2000).

8T. D. Blake, A. Clarke, and E. H. Stattersfield, Langmuir 16, 2928 (2000).

9B. J. Feenstra, et al.J. Soc. Info. Display 12, 293 (2004).

10B. J. Feenstra, et al.Proc. Intl. Display Workshop (2005). •


Fig. 4: Sample displays of a variety of colors illustrating the ColorMatch concept.