2.1.

Ambient Contrast Ratio

White paper can be described as a close approximation of a perfectly diffuse, or Lambertian, surface,⁸ making it appear bright from any viewing angle. At the same time, black pigments absorb most of the light. The ratio between the brightest and darkest state is commonly referred to as “contrast” or CR. Although emissive displays frequently exhibit exceedingly high CRs, it is important to note that these values are typically attained under dark environmental conditions. Under bright ambient conditions these values usually greatly deteriorate for emissive displays, as can be seen in Fig. 1.

Fig. 1

Qualitative photographs of a representative emissive display under different ambient conditions. Photographs are taken in darkness (0 lux), indoor lighting (1k lux), overcast sky (5k lux), and outdoor sunlight (40k lux). Adapted with permission from Ref. 9, copyright (2021), DisplayMate Technologies.

The ambient CR (ACR), as defined in Eq. (1), offers a more accurate approach for evaluating display performance, particularly for reflective displays, as it takes into consideration the real-world lighting conditions in which these devices operate:¹⁰

$R_L$ serves as an index for evaluating the undesirable reflection of environmental light, which could compromise the legibility and visual comfort of the display interface. Using Eq. (1), with an average $R_L$ of 4% and a typical portable display luminance of 1500 lux, the ambient CR drops to below 2 in direct sunlight (40 K lux), which renders the screen illegible. Table 1 describes human readability within terms of ACR.

Table 1

Ambient CR and readability.

2.2.

White State Reflectivity

Next to ACR, white state reflectivity is important for mimicking the appearance of paper. Figure 2 compares the black and white reflectance of three different media: high-quality glossy print, standard print, and LCDs. As can be seen, glossy prints have a white state reflectivity of around 80%, which sets them apart from the average 30% reflectivity found in reflective LCDs.

Fig. 2

Reflectance spectra for white and black states of a high-quality magazine print (National Geographic), standard print, and a reflective LCD. CRs are indicated on top of the white reflectance lines. Data for Reflective LCD and National Geographic are sourced from Ref. 12.

Fig. 3

Schematic representation of different color systems. (a) Red-green-blue (RGB) color filters on top of white-black switching pixels, results in a maximum reflection of 33% and a color fraction (the effective area over which saturated color can be displayed) of 33%’ (b) red-green-blue-white (RGBW) color filters add an additional white filter, increasing maximum reflection to 50%, although at the cost of a reduced color fraction of 25%; and (c) three-layer cyan-magenta-yellow pixels with a white reflector yield an exceptional maximum reflection and a color fraction of 100%. This configuration is mostly used in print media. Adapted with permission from Ref. 5, copyright (2011), Society for Information Display.

2.3.

Color

Color reflective displays pose another layer of difficulty in obtaining bright vivid colors. Much of it boils down to the right color system choice, of which a few prominent once are displayed in Fig. 3. Emissive displays predominantly employ an RGB (red-green-blue) system or an augmented RGBW (red-green-blue-white) system. In the context of reflective displays, using an RGB system leads to a maximum white-state reflectance of 33%, as each subpixel in an RGB configuration only contributes 33% of the light. Similarly, an RGBW system can achieve a maximum white-state reflectance of 50% due to the added white subpixel, which further contributes to the overall reflectance. These values assume fully saturated color filters and are usually lower in practice.

Subtractive color systems, such as CMY (cyan-magenta-yellow) and CMYK (cyan-magenta-yellow-black), have traditionally been used in the printing industry and usually employ a stacked architecture that greatly enhances light conversion. An optimal white state and maximum color fraction, defined as the effective area over which saturated color can be displayed,⁵ can be realized through the incorporation of a tri-layered structure featuring CMY (cyan-magenta-yellow) switching layers in the display system.^13,14 At the same time, stacked displays increase costs and complexity because they require more materials, additional backplanes, and complex driving electronics. In addition, the pixel resolution must be limited to maintain enough clear aperture, which helps reduce optical losses and parallax problems. To avoid significant parallax issues, the distance between the front and rear pixels in a multi-layer stack should not be larger than the pixel size itself. This becomes challenging when dealing with very high-resolution displays. It is clear that subpixels capable of transitioning among multiple color states, surpassing the conventional binary (on/off) color limitation, offer the potential to develop single-layered display systems capable of achieving 100% color reproduction.⁵

2.4.

Bistability

Once printed on paper, no further energy is required to see the image or text. Ideally, this should be mimicked in reflective display technology. Bistability, a desirable feature for display devices, refers to the ability to maintain image persistence in two stable states without requiring continuous energy. This means that once the image or text is printed, it can be seen without the need for additional energy.

4.1.

Working Principle

In general, an electrophoretic display operates by displacing charged pigments in and out of sight using electrostatic fields. Equation (3) describes the particle velocity (with initial condition $v_0=0$). The formula is derived by equating the electric attraction forces acting on a charged particle due to an external electric field with the fluidic drag forces counteracting the particle mobility, which are given by Stoke’s law as¹⁸

As can be observed from Eq. (3), the particle velocity, and thus the speed at which a single pixel can switch states, is influenced by several factors. One of the key determinants is the particle charge. It is most often deduced from the Zeta potential, which is a measure of the electrostatic potential difference between the fluid and the outer layer of the electrical double layer surrounding the particle. This parameter provides a nuanced measure of the particle’s effective charge and its interactions with the fluid. For low-dielectric media, it is proportional to the particle charge.¹⁹ A high absolute value of the Zeta potential ensures better dispersion of the particles, reducing aggregation and enhancing the display’s long-term reliability. Separately, it is important to note the limitations imposed by the applied electric field. Research efforts have been focused on reducing the required traveling distance, reducing the electrode spacing, and improving the chemical properties of the system.

4.2.

Implementations

Microcapsule electrophoretic displays utilize electric fields to move highly scattering or absorbing microparticles vertically within a microcapsule. A multitude of these microcapsules forms one pixel, and monochrome or color systems can be developed.

4.2.1.

Monochrome system

To this date, vertical dual-particle electrophoretic-based displays have been the most commercially successful reflective display technology on the market. As can be seen in Fig. 5, depending on the applied electric field, different colored particles with either positive or negative zeta potential are selectively moved to the surface, creating the desired image or text.

Fig. 5

Microcapsule electrophoretic display. (a) Schematic representation of an electrophoretic microencapsulated black and white ink, wherein the displayed color is controlled by the polarization of the electric field and the resulting vertical position of the charged nano-pigments. (b) Microscope image of microcapsules in a working prototype. Image adapted with permission from Ref. 20, used under a Creative Commons Attribution 2.0 Generic (CC BY 2.0) license. (c) A photograph of the Amazon Kindle e-reader tablet adapted with permission from Ref. 21, used under a Creative Commons Attribution 2.0 Generic (CC BY 2.0) license.

White particles often consist of inorganic pigments, such as ${\mathrm{TiO}}_2$, ${\mathrm{ZrO}}_2$, ZnO, or ${{\mathrm{Al}}_2\mathrm{O}}_3$, and black particles are often made from Color Index black pigments, iron oxide, or carbon black.²² These particles are dispersed in a non-polar fluid, which helps reduce leakage currents within the display. Surfactants are added into the system to prevent particle agglomeration²³ and to aid particle charging.²⁴ To reduce the switching voltage, a specialized coating formulation is applied,²⁵ resulting in the microcapsules being arranged in a closely packed monolayer. This arrangement facilitates a thinner overall packaging of the microcapsules. In addition, it was found that adding an oil-soluble high molecular weight polymer to the electrophoretic fluid increased image stability to 12 days and, importantly, without increasing the fluid viscosity.²⁶ In the initial stage of electrophoretic display developments, a “flash reset” was necessary prior to rendering a new image. This process involved resetting the positions of all particles to a uniform, known state, allowing for a more predictable and reliable transition to the desired grayscale level in the subsequent image. Using an algorithm²⁷ that stores the history of each pixel’s switching for a certain amount of time allows for addressing 16 levels of grayscale without flashing. Parasitic polarization in the microcapsule polymeric binding material leads to a dwell time dependence:²⁸ the impulse required to switch a pixel to a new optical state depends on the time that the pixel has spent in its previous optical states. This issue has been greatly researched and resolved through careful selection of the materials used.^29,30

4.2.2.

Color system

Color displays for dual particle systems have been achieved through the addition of a color filter array (CFA), as visualized in Fig. 6(a), or multi-particle systems, as visualized in Fig. 6(b).

Fig. 6

Color electrophoretic reflective display technology, based on (a) superimposed color-filter arrays. Adapted with permission from Ref. 31, copyright (2006), Society for Information Display. (b) A multi-color particle system.³² (c) CFA based device “Inknote Color Plus” on the left and “Bigme Galy” device on the right with a multi-particle system offering brighter colors.³³ The image is taken at the same location and same illumination, showcasing the increased brightness of multiparticle systems. Image adapted with permission from Ref. 34, copyright (2023), Good e-reader.

Fig. 7

TIR electrophoretic display concept. (a) Schematic representation⁴³ of the working principle, showing TIR reflection and frustrated TIR when the polarity of the applied electric field draws the absorptive charged black particles into the evanescence wave field. (b) SEM image (left) of a fabricated reflection plane consisting of spheres in an index matching layer, as well as an optical microscope image showing the reflection rings and dark areas of normal incident light. Image adapted with permission from Ref. 173.

Different methods have been explored to create CFA; these include direct printing³⁵ or photolithography³⁶ of a color resist and laminated CFA films.³⁷ However, CFAs, such as RGB or RGBW, use area sharing, in which each full-color pixel consists of three or four sub-pixels, respectively. This method limits the display area for each color, leading to suboptimal brightness levels. Furthermore, the implementation of a color filter to divide monochromatic pixels into three colored pixels inevitably leads to a reduction in the overall resolution.

To overcome these limitations in filter-based color electrophoretic displays, current research is exploring multi-particle systems. The key concept behind multi-particle systems is to make use of different electrophoretic mobilities as well as charge polarity and intensity to separate the particles.³⁸ Using waveforms, these particles are pushed and pulled to mix colors and produce specific shades. The EInk Spectra system uses black, white, and colored particles (either red or yellow). For full-color displays, EInk’s Advanced Color ePaper (AceP) combines white with cyan, magenta, and yellow particles, enabling it to render up to 50,000 colors,³² akin to newspaper prints. Figure 6(c) compares two commercial displays using CFA (left) and a multi-particle system (right) under the same illumination condition. The brightness difference is clearly perceived.

4.3.

Improving Switching Speed of Microencapsulated Electrophoretic Display Device

Various studies have been carried out to improve the response time. One approach is to use low viscosity³⁹ to increase electrophoretic mobility; however, this deteriorates bistability. Other approaches focus on improving the driving waveforms⁴⁰ or on picture drawing algorithms based on the content.⁴¹ Yet another suggestion is to use the inverse electrorheological (IER) effect,⁴² essentially reducing the viscosity briefly during the switching time. It was found that such an IER effect can be induced by the rotation of LC molecules,⁴³ briefly leading to some carrying hydrodynamic motion. Adding LC molecules leads to a 2.8 times reduction of response time while cutting the applied voltage in half.

4.4.

Total Internal Reflectance Electrophoretic Displays

TIR displays use TIR for bright images and “frustrated” TIR for dark ones, achieving high contrast and fast visual changes. In TIR, light in one medium reflects internally when hitting a lower-index medium at a steep angle. Frustrated TIR happens when a third, higher-index medium is close to the interface, letting some light pass or be absorbed. This technology was first reported in 1997.⁴⁴

4.4.1.

Working principle

For TIR electrophoretic displays, the switching mechanism is achieved through the movement of charged particles in and out of the evanescent field.⁴⁵ To optimize the angular response, TIR displays are constructed using hemispheres.⁴⁶ Charged electrophoretic absorbing particles are hosted inside a liquid below the reflective hemisphere array. As depicted in Fig. 7(a), absorbing particles are pulled out or into the evanescent field, allowing the light to be reflected or absorbed, respectively.

The phenomenon of TIR occurs when light is incident upon an interface from a high refractive index material $n_1$ to a low refractive index $n_2$ at an angle of incidence $\theta$ that exceeds the critical angle ${\theta }_c$ given as

Eq. (4)

$${\theta }_c=\arcsin \left(\frac{n_2}{n_1}\right).$$

For angles of incidence less than the critical angle, light is transmitted through the interface. Conversely, for angles of incidence greater than the critical angle, the light undergoes TIR and is reflected back into the first medium. As depicted in Fig. 7(b), upon normal incidence light, these hemispheres exhibit a bright reflective ring until the light’s angle of incidence surpasses the critical angle, causing the hemisphere’s center to reflect less light. For normal incidence, the reflection is given as

Eq. (5)

$$R=1-{\left(\frac{n_2}{n_1}\right)}^2.$$

In practice, the said hemispheres are formed from high refractive index nano-composite material $(n_1\sim 1.9)$ and low-index adjacent medium, such as fluroniated hydrocarbon $(n_2\sim 1.2)$; normal incidence reflections on the order of $\sim 60\%$ can then be expected.

The size of the hemispheres can be made as small as $2\mu \mathrm{m}$ in diameter. This reduces the device thickness and thus the required travel distance for the absorbing nanoparticles, yielding a higher switching time and reducing energy consumption, as well as easing the material constraints with respect to material attenuation coefficient. Color displays are achieved incorporating CFA on top of the microsphere array, as depicted in Fig. 8.

Fig. 8

TIR electrophoretic color display. (a) Schematic of TIR display with an RGB CFA; green is in the reflective state, and red and blue are in the absorptive state. (b) A picture of a tablet implementing this technology. Adapted with permission from Ref. 48, copyright (2019), Wiley.

CLEARink released a number of working prototypes, for example, at SID 2019⁴⁸ a 9.7″ diagonal eTIR display demonstrating video and color capabilities including grayscale [see Fig. 8(b)], showcasing a very small power consumption of merely $0.029\mathrm{mJ}/{\mathrm{cm}}^2$.

4.5.

Outlook

The main challenge of microcapsule electrophoretic displays is the switching speed. The response time is usually as long as hundreds of milliseconds.⁴⁹ This limits the application to static or slow-moving images. Current challenges for TIR electrophoretic are mostly related to commercial scalability, as well as improving the color gamut, increasing video speeds, reducing ghosting, and improving the general reliability of device. A potential long-term issue for TIR-based reflective displays is that they cannot operate in transmission by design. Thus, this makes a stacked CMY(K) pixel design difficult and ultimately limits the potential for achieving high color saturation/brightness.

5.1.

Twisted-Nematic Liquid Crystal Displays

5.1.1.

Working principle

The TN-cell⁵² is the most basic LCD technology; it consists of an LC sandwiched between two crossed polarizers and a reflective layer on the backside. The first polarizer in this setup polarizes incoming light in one direction. When turned off, the LC rotates the polarization of the traversing light by 90 deg, allowing it to pass through the second polarizer and be reflected on the backside mirror. When the LC is turned on, the incoming light’s polarization is no longer rotated and is blocked on the second polarizer [see Fig. 9(a)].

Polarizer-based LC technology can attain a maximum reflectivity of 50% because a polarizer filters out half of the unpolarized ambient light.

Fig. 9

Reflective TN display: (a) schematic representation: unpolarized light first passes through polarizer A, where it is linearly polarized. In the OFF-state, the liquid crystal rotates this polarization by 90 deg, enabling it to pass through polarizer B and reflect off the back-reflector. In the ON-state, polarization is not rotated; hence it is blocked by polarizer B. (b) Early TN LCD prototype 1972. Adapted with permission from Ref. 47 under a Creative Commons Attribution-Share Alike 3.0 license.

Fig. 10

The Zenithal Bistable Display. (a) A schematic depiction of the basic structure consisting of a front polarizer, grating, liquid crystal (LC) layer, rubbed polymer, and reflector. An electric potential causes the LC layer to switch between a TN state and a HAN state, both of which are bistable. (b) Photograph illustrating a typical use case of ZBDs as a shelf label, specifically the ZBD EPOP 900 model. (Datasheet, copyright 2011 ZBD Displays Ltd.).

5.2.

Zenithal Bistable Display

Zenithal bistable displays (ZBDs), a pioneering advancement in the realm of LC display technology, derive their name from the term “zenith,” indicative of the viewer’s line of sight in relation to the display surface. This nomenclature is emblematic of the display’s unique perspective, emphasizing an overhead or top-down viewing angle. The bistability is a key feature, enabling the display to hold an image with no power consumption until a change is required.

5.3.

Cholesteric Liquid Crystal Displays

The terminology “cholesteric” was first introduced in the scientific community by Georges Friedel in 1922,⁶³ recognizing the initial identification of this particular state of matter within cholesterol esters by Friedrich Reinitzer in 1888.⁶⁴ Subsequent to this initial discovery, instances of cholesteric liquid-crystalline states have been reported in a variety of substances that bear no inherent connection to cholesterol.

5.3.1.

Working principle

Cholesteric refers to a phase of LC in which the molecules are arranged in layers with a helical stacking pattern. For normal incidence, the reflected wavelength is given as^65,66

Eq. (6)

$${\lambda }_0=n_{a}p\cos (\theta ),$$

where $n_a$ is the average refractive index, $p$ is the helical pitch, and $\theta$ is the angle between the helical axis and the direction of propagation.

As can be seen from Eq. (6), reflectivity is concentrated around a specific central wavelength. Consequently, wavelengths deviating from these central peaks contribute less to the total reflectivity, leading to an overall decrease. The bandwidth of the cholesteric reflection is given as

Eq. (7)

$$\mathrm{\Delta }\lambda =\mathrm{\Delta }np,$$

where $\mathrm{\Delta }\lambda$ is the birefringence of the LC and $p$ is the pitch. The LC molecules have an inherent electric dipole moment. By applying an electric field, these molecules experience a torque, causing the LC molecules to align along the field direction. This alignment alters the pitch $p$, thus tuning the reflected wavelength ${\lambda }_0$. The key advantage of cholesteric LC displays is their bistability, with two textures at zero electric field: the reflective planar and the non-reflective focal conic. Here, ’texture’ refers to the macroscopic molecular arrangement, as shown in Fig. 11(a).

Fig. 11

Bistable cholesteric reflective display design and examples. (a) The schematic design of a bistable cholesteric reflective display,⁶⁷ illustrating the layered composition and functional elements. (b) Exemplary reflection spectrum of said cell showing the reflection bandwidth $\mathrm{\Delta }\lambda$ and central wavelength ${\lambda }_0$⁶⁷ and (c) two-layer stacked cell with and without surface treatment (rubbing) to improve reflection. (d) 7.5-inch full-color ChLCD display. Image adapted with permission from Ref. 68, copyright (2023), AUO. (e) A flexible cholesteric display, demonstrating the technology’s adaptability and potential for diverse applications. Image adapted with permission from Ref. 69, copyright (2008), radtech.

Upon normal incidence, a cholesteric LC surface reflects the component of the light that is circularly polarized in the same handedness as the CLC, transmitting the other component, and thus it cannot reflect more than 50% of unpolarized light.⁷⁰ At oblique incidence, the reflected or transmitted light becomes elliptically polarized.

The brightness can be significantly increased using a stack of two cells with opposing helical twists. Each cell reflects light with a specific handedness of circular polarization. The first cell lets through light of one handedness, which is then reflected by the second cell, enhancing the display’s brightness. Defects in the cholesteric crystal structure lead to scattering and non-ideal reflection conditions. This can be improved using rubbed contact-surfaces. “Rubbed” refers to surfaces that have been mechanically treated by rubbing them in a specific direction to align LC molecules. This rubbing process induces an alignment of the LC molecules along the direction of the rubbing, which helps to minimize defects in the cholesteric structure. As a result, this treatment has been shown to yield an increase in reflectivity of up to 70%⁷¹ [see Fig. 11(c)].

6.1.

Interferometric Modulator Display

Each sub-pixel of an IMOD consists of a Fabry–Perot etalon composed of a movable metallic mirror and a semitransparent glass upper layer [see Fig. 12(a)]. When light falls on the glass, it is split into two beams, with one beam reflected on the glass and one on the metallic surface. The traveled distance difference between the two beams results in a phase shift that affects the interference pattern of the reflected light. Tuning the gap between the glass and metallic layer changes the phase shift and thus the wavelength at which constructive interference occurs. This sets the perceived color of the individual sub-pixels.

Fig. 12

Schematic drawing of the IMOD concept. Incoming light strikes a semi-transparent layer, where it is partially reflected and partially transmitted into a resonant cavity. Within this cavity, the light undergoes multiple reflections between the layers. Outgoing light of a specific wavelength then constructively interferes with the incoming light, leading to the color perception of that element. This process allows for precise control over the displayed color by tuning the gap $g$ between the layers.

For the most simplistic IMOD, which consists of a single layer with a reflective backplane, the gap (g) to enhance a specific wavelength $\lambda$ is given as

To create black, the gap $g$ is reduced such that the reflected wavelength maxima is moved outside the visible spectrum and absorbed within a thin film within the stack. This is accomplished through electrostatic actuation, essentially collapsing the metal membrane onto the thin film stack. To avoid failure due to stiction in the collapsed state, stiction bumps are introduced.⁸¹ In real devices, multiple layers might be used; in such cases, it is advised to refer to the transfer matrix method to compute the reflected color.

6.1.1.

Implementations

Figure 13 shows the concept and full-scale realized tablet prototype of a Mirasol Qualcomm display. Each pixel can switch between a reflective colored state and an absorptive black state, in which the initial gap of the subpixel cavity controls the reflected wavelength.

Fig. 13

The IMod architecture and full-sized display prototype. (a) Schematic drawing of a pixel element showing the on (color) and off (state);⁸² (b) picture of a prototype tablet utilizing the Qualcomm Mirasol.

As can be deduced from Eq. (8), IMODs possess by design an inherent property of incident-angle sensitivity, resulting in observable variations in perceived colors dependent upon the angle of incidence. To mitigate this issue, a diffuser can be introduced on top of the Imod pixel array,⁸³ as depicted in Fig. 14.

Fig. 14

Schematic view of a Qualcomm Mirasol display unit, including, in particular, a front-facing diffuser film to reduce angle dependence. Adapted with permission from from Ref. 84, copyright (1999), Society of Photo-Optical Instrumentation Engineers (SPIE).

As mentioned in the previous section, the initial height of the Imod sub-pixel defines the reflected wavelength. However, as in all reflective display technologies, the use of a horizontal RGB leads to suboptimal contrast due to area sharing, which reduces the white state reflectivity. Thus, an innovative single-mirror continuously tunable pixel design that uses interferometric absorption,⁸⁵ instead of reflection was proposed. Single mirror interferometric (SMI) color regulation is achieved using interferometric absorption [see Fig. 15(a)]. Incoming light, interfering with reflected light, produces standing waves with varying peaks and nulls based on the light spectrum. By strategically placing a thin absorber at a null for a specific wavelength, selective non-absorption allows for reflection of that spectral component. All other wavelengths are absorbed, resulting in a saturated color reflection. Figure 15(a) displays the reflected color as a function of the absorber-mirror gap. In practice, the mirror is mobile, and the absorber is fixed on the substrate. This essentially allows each pixel to be modulated over the entire visible spectrum, drastically increasing the reflectivity. On the other hand, SMI also suffered from angle dependence. Figure 15 shows an SMI display at different angles of incidence. One can perceive the color shift in Fig. 15(b), when observing from 10 deg or 40 deg.

Fig. 15

Interferometric absorption-based continuous color tuning: (a) schematic illustrating the principle of color regulation through strategic placement of absorbers in standing light waves. (b) Off-axis viewing examples: camera directed at 10 deg off surface normal, 20 deg, and 40 deg. Image adapted with permission from Ref. 86, copyright (2015), Optica.

In recent years, a graphene membrane-based IMOD (GIMOD) was proposed;⁸⁷ it potentially allowed for refresh speeds of up to 400 Hz and $5\mu \mathrm{m}$ pixel resolution, which translates roughly into a smartphone with a 12K display resolution, a 20 times increase in resolution compared with the typical phone resolution (QHD, $2560\times 1440\text{pixels}$). Current designs of GIMOD are fabricated using graphene-covered micro-holes, essentially creating a voltage-controlled graphene micro-drum, where the initial cavity and the subsequent deflection of the drum control the reflected wavelength. This allows for analog control over the pixel’s color, yielding continuous full-spectrum reflective-type pixels. Simulations show a CR of 1:3 utilizing 29 layers of graphene. For vibrant reflective displays, this number must be increased by at least three times. Nevertheless, the high frame rate, resolution, and full-spectrum pixel are promising avenues.

6.2.

Other MEMS-Based Reflective Display Technologies

Hereafter, two further early-stage MEMS-based potential reflective display technologies are briefly introduced. Both approaches still need substantial developments to move from early-stage demonstrators to large-scale production.

6.2.1.

Mechanical light shutter

A mechanical light shutter functions by actively introducing an object into and out of the light’s trajectory, casting a shadow on an absorptive or reflective layer. Such implementations can be observed in Jutzi et al. work.^89,90 Figure 16 demonstrates the utilization of micro-flaps to alternate between a reflective state (flaps raised, producing a white pixel) and an absorptive state (flaps lowered, resulting in a black pixel). To optimize the CR, the anchoring system of the flaps should take as little visible area as possible; thus a good balance between flap size and anchor needs to be found. Increasing the size of the flaps comes with drawbacks such as slower switching speeds, reduced pixel density, and potentially compromised reliability. Nevertheless, this design offers the possibility of excellent CRs.

Fig. 16

Micro-mechanical light shutter. (a) Schematic of the actuation method and (b) optical microscope image of the final prototype device.⁹⁰ The applied voltage was 135 V. Images adapted with permission from Ref. 89, copyright (2010), SPIE Proceedings.

6.2.2.

Mechanically tunable metasurface reflectivity modulator

Recently, an MEMS-based actively tunable metasurface was shown to be able to modulate the reflectivity in the visible spectrum.⁹¹ The working principles and a microscope image of a segmented display prototype are shown in Fig. 17. The idea is to vertically displace a membrane punctuated with a regular pattern of holes, each hosting a stationary nano-disk, which effectively tunes the surface geometry and subsequently, light absorption and reflectivity.

Fig. 17

Mechanically tunable metasurface reflectivity modulator. (a) Schematic drawing of working principle, aligned surfaces reflect light, whereas displaced surfaces absorb it. (b) An optical microscope image shows the EPFL logo on a segmented display, visible when switched from idle to actuated states. Image under open access CC-BY 4.0. from Ref. 91, copyright (2023), ACS Photonics.

7.1.

Working Principle

The Young–Lippmann’s equation⁹⁶ captures the electrowetting phenomenon in its most basic form. It describes how the contact angle of a conductive liquid changes in response to an applied voltage between a liquid and an isolated substrate, given as

In most electrowetting-based pixels, two fluids coexist. The first is usually an insulating oil ink. The second is a conductive transparent liquid electrolyte, such as DI water. At equilibrium, the colored oil film lays between a hydrophobic insulator coating of an electrode. With the application of a potential, the stack becomes no longer energetically favorable, resulting in the water displacing the oil while reducing its contact angle from ${\theta }_0$ to $\theta$ with respect to the hydrophobic layer, as can be seen in Fig. 18(a). The colored oil film can essentially be opened and closed like a curtain.

Fig. 18

Electrowetting display mechanism: (a) schematic of a unit pixel cell demonstrating the transition between expanded and contracted oil states, indicative of the “curtain-like” movement of the colored oil film modulated by potential application; (b) Actual device implementation showcasing the technology in practice. Image source from Ref. 97. Video: The Electrowetting Display IEEE Spectrum YouTube channel.

The oil displacement happens in different stages,⁹⁸ namely oil film rupture (initiation stage), oil-dewetting, and finally a slower oil drop rearrangement stage. A threshold voltage is needed for the initial oil rupture. Charges accumulate at the interface between oil and water when electricity is applied, creating pressure and causing the oil to undulate due to Rayleigh-Taylor instability.⁹⁹ As these undulations increase, the water ultimately breaks through to the dielectric substrate, resulting in a three-phase contact line. Due to the induced change in contact angle by electrowetting, the water then forces the oil to one side. The reverse mechanism or receding process is more straightforward, with the oil re-wetting the hydrophobic surface as the water retracts.

7.2.

Implementations

Electrowetting-driven reflective pixels pose several challenges that need to be addressed for reliable operations and increased optical performance. For instance, the contrast between covered and uncovered states, often referred to as “white area fraction” or “aperture,” needs to be maximized. Naturally, the colored oil cannot completely disappear from the light path. In the retracted state, there will still be some fraction of the pixel that will display the colors of the oil. This fraction depends on the pixel size and oil film thickness. Figure 19(a) depicts the relation between the pixel size and aperture for different oil thicknesses. The fraction is computed by assuming a hemispherical sphere oil droplet in the contracted state hosted in a rectangular pixel. Using volume conservation between expanded and contracted oil, the surface fractions can be easily computed.

Fig. 19

Electrowetting challenges. (a) Relationship between pixel size and aperture across different oil film thicknesses. In smaller pixels, thicker films decrease the visible white/transparent area. However, thinner films let more light through, which can lower the contrast. (b) Optical microscope images showing contact angle hysteresis.¹⁰⁰ Adapted with permission from Ref. 100 under CC BY 4.0 DEED, copyright (2021), Frontiers in Physics.

One can observe that, for pixel sizes below $40\mu \mathrm{m}$ and $5\mu \mathrm{m}$ oil thickness, the aperture drops below 50%. Thus, to increase the display resolution, thinner oil thicknesses are needed. To minimize light bleeding through, oils with higher molar extinction coefficients are thus more desirable.

Another challenge to overcome is contact angle hysteresis (CAH). CAH is a well-known physical phenomenon in electrowetting. It is defined as the difference between the advancing contact angle and the receding contact angle observed at the same voltage. Figure 19(b) illustrates an example in which there is an aperture discrepancy, albeit with the same final applied potential. Increasing the voltage from 0 to 20 V yields a smaller aperture than decreasing the voltage from 30 V to 20 V. CHA makes it more challenging to control precise gray levels in electrowetting displays. CHA has been attributed to random pinning forces, which are caused by surface heterogeneities at smaller scales.⁸⁸ These forces can be successfully reduced utilizing an increasing alternating voltage scheme,¹⁰¹ which leads to sufficient vibrational energy to overcome the pinning forces.

Finally, it is also important to account for the so-called “backflow problem.” In electrowetting systems, a dielectric material is used to separate the conducting electrodes from the liquid. When applying a potential to the system, charge trapping can occur in the dielectric layer. These remaining charges can cause the fluid to flow back, even when a constant DC voltage is applied. Charge trapping significantly impacts the stability of electrowetting-based devices. Asymmetric alternative polarity driving schemes¹⁰² are said to reduce charge trapping. In essence, any remaining trapped ions are removed by applying a reset pulse, a small voltage drop with opposite polarity at the signal termination point.

7.3.

Outlook

Electrowetting-based displays have undergone excessive research and development both in the commercial industry as well as in academic research. Despite their potential for color and video reflective display technology, this innovation never ventured beyond proof-of-concept prototypes. Manufacturing difficulties, such as proper sealing and longevity under varying conditions, might have been the cause. Moreover, with advancements in display technology, particularly in terms of resolution, electrowetting displays have become less competitive in terms of image quality because oil thickness and pixel size are tightly linked. Nevertheless, electrowetting technology remains a promising area of research with potential applications in areas such as e-readers, smart glasses, and flexible displays. Current efforts are focused on low-power sunlight-readable electronic billboards as this application is less stringent on display resolution.

8.1.

Working Principle

The core principle of electrochromic materials is their ability to change their absorption bandgap, by adding or removing electrons through redox reaction. Electrochromic materials can most often be switched between a transparent (“bleached”) state and a colored state or between two colored states. Polychromic materials may exhibit several colors.^113,114 The primary mechanisms to achieve a change in electron state are redox reactions, either directly induced or through ion intercalation. The specifics depend on the type of electrochromic material.

8.2.

Implementations

Electrochromic materials can generally be classified according to their solubility,¹¹⁵ leading to different device architectures, as depicted in Fig. 20.

Fig. 20

Electrochromic display. (a) Schematic diagrams of three types of electrochromic devices are shown: (a) type I, which operates in a solution-phase; (b) type II, a hybrid system; and (c) type III, also known as the “battery-like” ECD. ${\mathrm{K}}^{+}$ represents cations, which could be ${\mathrm{Li}}^{+}$ or ${\mathrm{H}}^{+}$, and ${\mathrm{A}}^{-}$ represents anions, for instance, ${\mathrm{ClO}}_4^{-}$, ${\mathrm{Cl}}^{-}$, ${\mathrm{BF}}_4^{-}$, or ${\mathrm{PF}}_6^{-}$. Images adapted with permission from Ref. 115 under CC BY 4.0 DEED, copyright (2020), Chemistry Europe.

Type I EC materials remain soluble in both their reduced and oxidized states. Examples include viologen and heptyl. In device architectures, these materials often have their electrochromic species dissolved directly within an electrolyte, which is sandwiched between two transparent conductive oxide substrates. Although the fabrication of such devices is relatively straightforward, a critical challenge arises in ensuring the sealing of the devices to avert potential electrolyte leakage. Furthermore, there is an inherent requirement for a continuous supply of current or voltage to uphold their redox states, which could lead to increased energy consumption.

Type II EC materials remain soluble when in their colorless redox state but transition to form a solid film upon the electrode surface when subjected to specific conditions. Devices leveraging Type II materials integrate a redox mediator in the electrolyte, acting as a counterbalance during the redox reactions. This interaction, although promising, does have its challenges. A significant one is the “loss current,” a consequence of the direct contact between the redox electrolyte and the EC layer. However, innovations, such as electronic barrier layers¹¹⁶ and catalytic ${\mathrm{IrO}}_{\mathrm{x}}$ layers,¹¹⁷ provide potential solutions to this challenge.

Type III EC materials stand apart due to their inherent solidity in both redox states. This means that they consistently form an insoluble film on the electrode surface. The gamut of Type III materials is extensive, comprising groups IV and V transition metal oxides, conductive polymers, Prussian blue, and certain metal polymers. In device configurations, these materials typically feature in thin-film “battery-like” setups. Such a device incorporates two substrates coated with transparent conductors, an electrochromic layer, and another layer designated for ion storage or as an additional electrochromic layer. The addition of an ion-storage layer with ample charge capacity ensures a complete and effective color transition, resulting in a device that boasts impressive electrochemical stability.

In practical applications, particularly when considering reflective displays, Type III often emerges as the preferred choice due to its inherent stability and enhanced optical properties.

Electrochromic materials transition from a neutral to an “excited” state during redox reactions, challenging the development of bistable electrochromic devices (ECDs) as they tend to revert to their stable, lower-energy states. This reversion diminishes the longevity of the desired optical state. To counter this, proton-coupled electron transfer¹¹⁸ and bond-coupled electron transfer¹¹⁹ have been implemented. These methods avoid the formation of high-energy intermediates, which can destabilize the material, reducing its effective life and performance. Electrode materials are fundamental in electron transfer processes. They require meticulously designed energy levels (Fermi levels) that synchronize with those of electrochromic materials. Proper alignment ensures efficient electron transfer and minimizes reverse transfers.¹²⁰ Metal dendrites, irregular tree-like structures, can form on electrode surfaces. Their presence disrupts the uniformity of the electron transfer, leading to inconsistent color transitions. To combat this, copper ions have been integrated into bismuth-based systems. The copper ions oxidize bismuth atoms, creating a smoother, spherical morphology. This morphology reduces the chance of dendrite growth, ensuring a consistent electron transfer and uniform color change.¹²¹ For enhanced device responsiveness, multivalent ions, such as ${\mathrm{Al}}^{3+}$, are incorporated into tungsten oxide ${\mathrm{WO}}_{\mathrm{x}}$ materials. These ions facilitate rapid and reversible ion insertion/extraction.¹²² The increased electrostatic forces between ions lead to an accelerated electron transfer, improving the device response time. To diversify the color palette and fasten switching, electrochromic materials are paired with metasurfaces, including Fabry–Perot cavities^123,124 and metallic nanoslits.¹²⁵ These metasurfaces induce structural colors and optimize charge-diffusion characteristics. The result is a broader color spectrum and reduced switching times.

Innovative pixel designs utilizing both lateral and vertical configurations have emerged.^126,127 Examples include Urano et al.’s three-layered CMY ECD¹²⁸ and RICOH’s full-color, flexible matrix display (see Fig. 21). These leverage vertically stacked electrochromatic elements. Such stacking allows for brighter color displays, and their flexibility enhances the device adaptability.¹²⁹

Fig. 21

Multi-layered full-color reflective electrochromic pixel. (a) Schematic representation of a unit and (b) a photograph depicting a full-color, flexible, active-matrix-based implementation of the display design. Images adapted with permission from Ref. 129 (permission pending), copyright (2013), Ricoh.

8.3.

Outlook

EC displays offer numerous benefits: they are inexpensive, low-power, insensitive to viewing angle, and capable of semi-bistability. Although electrochromism has already been commercially successful in smart window and glare-reduction applications, commercial displays are still limited to simple monochromatic segmented displays. Continued research and development in electrochromic materials and device architectures are essential to unlocking their full potential and drive advancements, particularly concerning switching speed, which currently lies around 0.8 Hz.⁷⁷ Innovations, such as the introduction of PEDOT nanotube arrays, have shown promise in achieving ultrafast switching speeds compatible with moving-image display technology, suggesting that continued innovation in material science could overcome existing limitations in switching speeds and pave the way for more versatile electrochromic displays.¹³⁰

9.1.

Working Principle

The material science behind phase change is complex. It has been shown that the switching from crystalline to amorphous states when directly applying current occurs due to rapid heat-induced dislocation nucleation, electrical wind force transport, and eventual jamming in the crystal, which leads to amorphization.¹³⁸ By contrast, the amorphous-to-crystalline transition occurs by heating the material above its crystallization temperature. Upon cooling down, the atoms reorganize into a crystalline lattice, resulting in a lower resistance state. The variations in optical properties arising from these structural modifications can be further understood by examining the corresponding changes in bonding strengths. It was found that the crystalline phase exhibits resonant bonding, whereas the amorphous phase shows covalent bonding.¹³⁸ The greater degree of electron delocalization in resonant bonding materials, as opposed to covalent bonding, can have a significant impact on the optical characteristics of the material. The higher level of interaction between delocalized electrons and incident electromagnetic waves produces specific absorption, transmission, and reflection characteristics, resulting in optical performances that differ significantly from materials having localized covalent bonding. On a broader scale, individual phase-change based pixel designs differ. However, the primary objective is to engineer a wavelength-dependent tunable resonance that is then tuned by the phase-change material, inherently altering the resonance conditions due to a change of optical properties when switching from an amorphous to a crystalline state.

9.2.

Implementations

When directly applying current to the switch state, the material undergoes a partial phase change until a low-resistance pathway is established.¹³⁹ This pathway subsequently functions as a channel for the residual current, impeding a full-phase transition. This phenomenon is referred to as the “filamentary switching issue.”¹³⁹ Utilizing nanoscale pixels can effectively reduce this issue. However, addressing such a vast number of pixels individually and simultaneously presents a challenge. As the number of pixels in a display grows, the amount of computing power and energy needed by the circuits that control each pixel also rises significantly.¹⁴⁰ Thus, alternative more homogenous switching methods need to be envisioned. To avoid the filamentary switching issue, much effort has been devoted to the design of decoupled micro-heaters as a means to uniformly fully switch the phase-change material,¹⁴⁰ as shown in Fig. 22. A typical phase-change-based pixel structure is depicted in Fig. 22(a). The SRD pixel consists of three primary layers: an electronic substrate directing drive signals to specific pixels, a microheater that converts electrical signals to uniform thermal pulses, and finally an optical layer reflecting specific colors in stable phase-change material states via strong interference effects.

Fig. 22

Microheater-based phase change display system (Fig. 21). (a) Schematic of the SRD pixel structure (b) microheater array, (c) the assembly comprising a mirror and thin-film interference stack, and (d) a microscope photograph demonstrating a $24\times 24\text{pixel}$ region being dynamically updated with a sequence of letter. Images adapted with permission from Ref. 140 (permission pending), copyright (2019), Society for Information Display.

9.3.

Outlook

Efforts to increase color saturation can be achieved by a higher refractive index and lower loss materials than GST or GeTe, such as antimox trisulphide $({\mathrm{Sb}}_2{\mathrm{S}}_3)$ and antimony triselenide $({\mathrm{Sb}}_2{\mathrm{Se}}_3)$.¹⁴¹ Although noticeable switchable color changes are possible, little has been demonstrated of producing a black state, which is needed for a fully functional display. When switching phases, the real part of the refractive index GST material usually exhibits a change of roughly 1.5 to 2, whereas the absorption coefficient can be altered by more than 3.¹⁴² Essentially, for a given pixel design, the resonance sensitivity to a variation in optical constants needs to be increased for a black state to exist. In such a state, the resonance would need to be shifted outside the visible domain. So far this remains a challenge. Furthermore, the use of microheaters severely limits the switching times. Frequencies of mere 2 Hz are reported.¹⁴⁰ With approximately $100\mathrm{mJ}/{\mathrm{cm}}^2$¹⁴² microheaters also consume more energy per frame than electrostatic-driven systems.

10.1.

Working Principle

1D photonic crystals’ colorful appearances result from interference and reflection. The wavelength that is coherently scattered is centered on $\lambda$ and is estimated by the Bragg-Snell equation:¹⁴⁶

Based on Eq. (11), the different ways to alter the reflected wavelength can be inferred, namely by modifying three variables: the diffractive plane spacing, the average refractive index, or the Bragg incidence angle.

10.2.

Implementations

Most photonic crystal-based reflective pixels are based on acting on the inter-plane distance to modulate the reflected wavelength. The key challenges are to find fast-switching methods as well as ways to respond to the inherent angle incidence dependence of such photonic crystals, which can be seen in Eq. (11).

For the most part, inter-plane distance variations are achieved in electrochemical swelling-driven color change displays¹⁴⁷ and electrokinetic-driven reflective displays.¹⁴⁸ Although crystalline structures have shown excellent color tunability, their practical application has been severely impeded by their intrinsic angle dependency. It has been demonstrated^149,150 that colloidal structures that are made up of particles with sizes similar to optical wavelengths and exhibit short-range order but lack lattice periodicity, exhibit independence of the incident angle. These systems^148,151 have therefore been seen as promising candidates for color-tunable reflective displays, albeit with lower peak reflectivity values of generally around 25% to 30% [see Fig. 23(c)].

Fig. 23

Photonic display pixel. (a) Schematic design of the photonic display pixel. (b) Photographs at varying bias voltages: 1.0 V, 2.5 V, and 4.0 V. (c) Reflection spectra recorded at increasing bias voltage levels. (d) The 165 nm shift of the photonic band position from 655 to 490 nm, governed by the applied voltage. Images adapted with permission from Ref. 148, copyright (2010), Advanced Materials.

Fig. 24

Performance map for reflective display technologies: this representation compares diverse reflective display technologies based on their essential performance parameters. Microcapsule electrophoretic displays (reflectance: 44%,¹⁵⁴ contrast: 23,¹⁵⁵ and frequency: 2.1 Hz¹⁵⁶) and their ChLCD (reflectance: 35%,⁷⁷ contrast: 10,¹⁵⁹ and frequency: 0.8 Hz⁷⁷) and electrochromic counterparts (reflectance: 47%,¹¹² contrast: 12, frequency: 1 Hz¹⁶²) find their niche in applications demanding static images given their constrained refresh rates. Conversely, emerging electrowetting displays, exhibiting up to 65% reflectance and a remarkable 62 Hz frequency,¹⁵⁵ signal promise for dynamic video renditions. Phase change displays, with a reflectance mirroring that of electrophoretic at 44%,¹⁶⁰ however, are marked by a considerable refresh energy density of $104\mathrm{mJ}/{\mathrm{cm}}^2$.¹⁶⁰ Albeit discontinued commercially, MEMS-based Qualcomm designs achieve a stellar 90% reflectance,⁸¹ contrast-ratio of $1:30$,⁸⁴ and a swift 120 Hz,⁸⁴ while being impressively efficient at $0.004\mathrm{mJ}/{\mathrm{cm}}^2$.⁸⁴ Numeric values can be found in Appendix A, Table 2.

10.3.

Outlook

Switching times of approximately 50 ms¹⁴⁸ have been documented in tunable photonic crystals, which, although not yet suitable for video applications, demonstrate potential for improvement. One possible approach to enhancing the switching speed is through the employment of lower-viscosity liquids, which may facilitate a faster response time. Furthermore, the trade-off between angle independence and color intensity presents another challenge. Although amorphous photonic structures have shown reduced angle independence, they also exhibit reduced peak reflectivity and less vibrant colors. This is due to the incoherent scattering events caused by their disordered arrangement, resulting in weaker constructive interference and diminished color intensity. Hierarchical photonic structures^152,153 have been proposed to produce both angle-independent and bright colors, but further investigations and optimization are required to address these limitations and fully realize the potential of tunable photonic crystals in display technologies. Finally, it is worth mentioning that, although continuous color can be produced, producing white without resorting back to an area-sharing subpixel layout, such as RGB, has yet to be shown. Commercially, two companies are pioneering modulation approaches using photonic crystals. Opalux focuses on color-tunable Photonic Ink, and Nanobrick works on full-spectrum tuning using ${\mathrm{SiO}}_{\mathrm{x}}$-coated nanoparticles.

11.1.

Signage

Signage applications prioritize low power consumption, long-term image stability, and moderate refresh rates. ChLCDs are particularly well suited for this domain, offering image maintenance without continuous power and a modest reflectance of around 35%. Phase change displays, still in the emerging phase, also show promise for static signage due to their low power requirements and non-volatile nature, although their current slower switching speeds make them less ideal for dynamic content.

11.2.

E-Readers

In the e-book and e-reader sector, key considerations include medium refresh rates, optional color, and high energy efficiency. Microcapsule electrophoretic displays, commonly known as E Ink, stand out in this category. They balance high readability, with reflectance up to 44%, and energy efficiency, with a refresh energy density of $0.58\mathrm{mJ}/{\mathrm{cm}}^2$, making them a preferred choice for devices used extensively over long periods.

11.3.

Smartwatches and Mobiles

For smartwatches and mobile applications, in which high refresh rates, superior pixel density, and full-color capabilities are paramount, MEMS-based reflective displays take the lead. These emerging technologies boast superior performance metrics, including high reflectance (up to 90%), impressive contrast (with a ratio of 30:1), and high-frequency operation (up to 120 Hz), making them increasingly relevant. With a pixel density of 363 PPI, these displays are especially suitable for devices that demand high-resolution visuals, ranging from advanced e-writers to premium consumer electronics. Smartwatches emerge as an especially ideal candidate for MEMS-based reflective displays due to their small display footprint (facilitating the packaging) and essential need for low-power operation as well as outdoor visibility.

11.4.

Billboards and Large-Format Displays

For billboards and large-format displays, the focus is on sunlight readability and less stringent resolution requirements. Here, electrowetting displays, an emerging technology, presents a viable option. They offer low power consumption and effective sunlight readability, making them suitable for large-scale outdoor displays such as electronic billboards, for which high resolution is not the primary concern.

Beyond core metrics, the mitigation of inadvertent internal reflections is paramount in e-paper technology. Although these displays capitalize on ambient light reflection, undesirable internal reflections—often from thin-film interfaces—can compromise performance. Advanced techniques, such as index matching and multilayer destructive interference, have been employed to counteract these challenges. Notably, antireflective coatings inspired by nocturnal moth corneal patterns¹⁶³ have achieved reductions in spurious reflectivity to a mere 0.1% within the visible spectrum.¹⁶⁴ Raut et al. provided a thorough discussion on this reflection management.¹⁶⁵