Tiny displays look poised for a bright future

18 Jun 2004

Microdisplays are vital for a number of applications, from next-generation TV to light-beam modulators. Daniel Krueerke provides an overview of the devices available.

From Opto & Laser Europe June 2004

Microdisplays are miniature displays which are so small that magnifying optics are needed to exploit them. Typically measuring 1 cm or so across, they are at the heart of many up-and-coming consumer items. Examples of popular equipment containing microdisplays are front projectors, rear-projection television (RPTV), home-theatre projectors, digital-cinema projectors and head-mounted-systems, as well as viewfinders in digital camcorders and cameras. Microdisplays also have numerous non-display applications as light modulators for tasks such as optical correlation, printing, micro-machining, pulse-shaping, holography and optical networking.

The family of microdisplay technologies can be classified as follows:

* transmissive liquid crystal;

* microelectromechanical systems (or MEMS), often known as digital light processors (DLP);

* liquid-crystal-on-silicon (LCOS);

* organic light-emitting diode (OLED) and organic light-emitting diode on silicon (OLEDoS).

Depending on the technology, a microdisplay is transmissive (light passes through the image-forming element), reflective (light reflects off the element) or emissive (the element generates light itself) mode. In the latter case no external light source is required.

Of these different technologies, DLP and transmissive liquid-crystal (LC) solutions are the most mature and established in the market, while reflective LC and emissive OLED microdisplays are just emerging.

Before you purchase a microdisplay, it is essential to be clear about its purpose so you can select the most appropriate technology. The following questions should help.

Selection criteria * Do you need the microdisplay to be in transmissive, reflective or emissive form?

* What is the required resolution?

* Would you like it for amplitude modulation (for display applications) or phase modulation (for use as a diffractive element)?

* What is the desired switching speed, or the frame rate?

* What colour depth do you need?

* Do you require high brightness?

* Do you prefer the addressing to be analogue or digital?

* Do you need a large viewing angle (e.g. on-axis, or off-axis application)?

* What is the operating temperature range?

* Do you have to consider harsh environmental aspects?

Although the microdisplays market began with transmissive LC panels, in recent years reflective microdisplay technology has led to a remarkable improvement in image quality, contrast and light efficiency.

While both transmissive and reflective versions are often used in near-to-eye applications (mainly viewfinders and simulation), this sector is now moving towards emissive solutions. Systems using emissive microdisplays have the big advantage of being very compact as they do not require an external light source or any illumination optics.

Today, the resolution of commercially available microdisplays ranges from QVGA (320 x 240 pixels), which is typically used in viewfinders and mobile phones, to QXGA (2048 x 1536), which targets next-generation digital cinema and high-definition television (HDTV).

For many display applications, 16 million colours (24-bit colour depth) are desired, and the colour gamut (range) should match or exceed that of a standard colour cathode-ray tube (CRT). In order to show high-quality video with a performance that matches the 60 Hz frame rate of a CRT, the rise and fall times for switching pixels on and off must be less than a few milliseconds.

Transmissive microdisplays Transmissive microdisplays are always based on LC technology and usually high-temperature polysilicon (HTPS) thin-film transistor (TFT) technology is used to drive (address) each pixel. Several types of microdisplays based on different liquid-crystal phases are available. The nematic liquid-crystal phase is used in most microdisplays and offers a switching time in the millisecond region. The most common types are twisted nematic (TN), super-twisted nematic (STN) and double-layer super-twisted (DSTN). All have slightly different characteristics in terms of driving voltages and switching times.

The LC-film is placed between two transparent electrodes in order to create a sandwich structure which is placed between an orthagonally aligned input polarizer and output polarizer. One electrode is pixelated and contains the TFT array for driving each pixel, while the opposite electrode is common (e.g. earth). When no voltage is applied to the pixel, light transmission is blocked. However, when a voltage is applied to a pixel it causes the polarization of the light to rotate as it passes through the liquid crystal, and thus pass through the output polarizer.

A couple of other commercially available microdisplays use the ferroelectric smectic C* liquid-crystal (FLC) phase. FLC microdisplays offer very fast switching, typically around 10-100 µs, or about three orders of magnitude faster than the ordinary switching times of most other liquid crystals. Additionally, the FLC's in-plane switching provides excellent viewing-angle characteristics.

An electro-optic effect in nematic liquid crystals, well known since the early 1970s as deformation of aligned phase (DAP), is now enjoying a great comeback in so-called vertically aligned nematic (VAN) displays. This mode exhibits very high contrast ratios and a large viewing angle because of its good black state and in-plane bright state. It can be used in transmission as well as reflection.

Other devices based on electroclinic and anti-ferroelectric liquid-crystal materials have also been developed in laboratories.

Reflective microdisplays Reflective microdisplays are based on LC technology or on MEMS. Both feature a silicon backplane, consisting of an open silicon memory chip, with a highly reflective top metal layer, usually made of aluminium. This top layer is subdivided to create an array of tiny mirrors (pixels or stripes) which are individually addressable. The addressing adjusts either the tilt of the pixels (in the case of MEMS technology) or the voltage applied to the pixels (LC technology).

The MEMS technology is widely known as digital light processing (DLP) and has been patented by Texas Instruments. The image is generated by light reflected from pixels that are switched either on or off, depending on their tilt angle. This technology is often found in the digital projectors that project the image from a laptop computer onto a large screen.

In contrast, the LC technology is generally described as liquid crystal on silicon (LCOS) and works by controlling the polarization of the light reflected from the pixels. The configuration of an LCOS microdisplay is similar to a transmissive display. The main difference is that the latter's pixelated transparent electrode is replaced with a reflective backplane, which incorporate compact, integrated electronics. In effect, an LCOS microdisplay is an on-chip solution, which can perform image processing such as gamma correction or image decompression.

Polarizers are required to achieve a contrast between the ON and OFF pixels, unless the device operates in a special mode called binary phase modulation. In this mode the display is designed so that the light reflected from different pixels interferes either constructively or destructively. Binary phase modulation is used in a wide range of non-display applications where the microdisplay is a programmable diffractive optical element.

LCOS microdisplays are often an attractive solution because they are easy to build and suit a wide range of applications. They also provide important price and performance advantages compared with HTPS and MEMS technology. For example, LCOS devices require less power than transmissive HTPS devices and the pixels on LCOS panels can be made smaller without compromising picture quality or manufacturability.

As a result, LCOS displays can be scaled up to a high resolution without increasing the size and cost of the panel and other optical components in the final system. With this advantage and their excellent optical properties, LCOS microdisplays pose a strong challenge to their established TFT and MEMS light engines in projection applications.

However, the requirement for polarizers and the transmission properties of the liquid-crystal film may limit LCOS technology for certain uses, especially applications involving light in the ultraviolet wavelength region. MEMS technology is typically more robust than LCOS when it comes to thermal stability and a wide range of wavelengths and high intensities of the incident light.

MEMS microdisplays also have the advantage of higher light throughput (less optical loss) because they do not need polarizers. They also have a broader modulation bandwidth because of their fast switching, which can be as short as 2 µs but is typically 10 µs.

On the other hand, MEMS are much more expensive than LCOS devices, because of their complex, multi-step semiconductor manufacturing process.

Emissive microdisplays Emissive microdisplays are the latest development on the scene and many aspects of their performance, such as lifetime and colour quality, are still the subject of intensive R&D. Based on organic light-emitting diode (OLED) technology, emissive microdisplays rely on the electroluminescence and semiconductor characteristics of organic molecules (polymers). When a silicon backplane is integrated with the organic material the technology is known as organic light-emitting diode-on-silicon (OLEDoS). In both cases, applying a voltage to a pixel made from OLED material causes it to emit light at a wavelength that is characteristic of the material. The intensity of the light increases as the voltage rises.

Once this technology is established, it could revolutionize the entire direct-view display market. OLEDoS technology does not require an external light source or polarizers and as a result can create displays that are lightweight and cheap. Furthermore, OLED devices that do not use a silicon backplane can produce robust and bendable plastic displays. With still some way to go technically, the availability of OLEDoS microdisplays is likely to mature in the future.

Colour-image generation Two important markets for microdisplays are direct view and projection. For direct-view applications, a microdisplay should meet or exceed typical CRT characteristics, while for projection applications the key factors are light throughput and resolution. In the case of projection applications, the frame rate can be less than 60 Hz (30 Hz is typical).

It can be difficult to get both fast switching and good full-colour performance in a single nematic liquid-crystal microdisplay. There are two reasons for this. Firstly, the switching time in nematic liquid-crystal devices is generally in the region of several milliseconds. Secondly, the polarization-conversion efficiency of the liquid crystal varies across the visible wavelength spectrum.

A way around this is to use a three-channel system that superimposes red (R), green (G) and blue (B) images generated from three separate microdisplays. Even microdisplays that are fast enough to perform single- or dual-channel projection (like ferroelectric LCOS, digital VAN-LCOS and DLP) are often implemented in three-channel light engines in order to increase the brightness of high-end projectors.

The demand for smaller, lighter, cheaper, high-performance colour products is a key driver for manufacturers; single-channel systems are an ideal solution. To generate full colour in a single-channel system a "time sequential" mode is often employed. Here the information for generating the R, G and B images is passed onto the same display sequentially. Above a certain frequency the human visual system fuses the short flashing colour sub-images into a full-colour image without noticing the temporal nature of the image generation. In this mode, the microdisplay is illuminated with LEDs or white light from a bulb that is passed through a spinning colour wheel.

An alternative approach to generating single-channel full colour is to cover R,G or B sub-pixels by colour filters or a diffraction element which separates the R,G and B parts of the illuminating white light onto the designated sub-pixel.

All imaging systems incorporating microdisplays based on LC technology use polarized light. This causes an efficiency loss of at least 50%, because only one state of polarization can be used for the display. Fortunately, methods have been developed to decrease this efficiency loss by sufficient polarization-conversion techniques.

An overview of various microdisplay characteristics is given in the table. If you want to use a microdisplay as a spatial phase modulator, you should focus on the liquid-crystal technology. Only vertically switching MEMS are suitable as phase modulators, and they are rare and expensive. One particular MEMS solution for phase modulation could be the grating light-valve (GLV) approach.

Many microdisplays can be bought together with driving electronics - and, in the case of direct-view applications, with an illumination source, illumination optics and viewing optics. The driving electronics often enable you to use the DVI, VGA or even S-Video and RGB signals out of your computer, video recorder or other common source.

* For more information about microdisplays visit The Microdisplay Page at www.elis.ugent.be.