Is the FED Finally Unchained?


by Daniel den Engelsen

At EuroDisplay '96, Canon presented the principle of the surface-conduction electron emitter (SCE) for field-emission displays (FEDs). This concept was revolutionary because the paradigm for cold emission in FEDs at that time was an array of microtips. For a long time after that, nothing was heard of these SCEs and their suitability for FEDs. Beginning in 1998, Canon and Toshiba cooperated on the development of a surface-conduction electron-emitter display (SED) and that was the bulk of the information given to the outside world.

The silence on SED was undone at the big electronics show CEATEC Japan 2004, held in Tokyo, October 5–9, 2004. Toshiba and Canon showed a 36-in. W SED. At the end of 2004, SED, Inc., the daughter company of Canon and Toshiba, announced that 50-in. SEDs would be introduced to the market in 2006. The company has started pilot activities in Hiratsuka, Japan, to enable manufacturing.

At SID '05 in Boston, SED, Inc., presented two interesting papers on the technical capabilities of the SED. Although the anode voltage is said to be 10 kV, the lumen efficiency of the screen is only 5 lm/W, leading to a power consumption of 110 W (at 20% APL) for the 36-in. W panel. This figure compares favorably to that of power-hungry PDPs and LCDs. Nonetheless, some power is lost because of the rather low aperture ratio of 31% for the anode pixels and because of ohmic losses in the cathode plate.

The contrast ratio of the SED at low ambient illumination is about 10,000:1, leaving LCDs far behind. The large black-matrix area and Toshiba's color-filter technology yield a daylight contrast that is better than that of CRTs, but not as good as the daylight contrast of an LCD.

Spacers are always an issue with FEDs. In the case of the SED sample shown at CEATEC, the spacers were placed on the scan wires of the cathode plate and were virtually invisible.

Because of the fast-decaying CRT phosphors used, the response time for switching off is about 1 msec, which enables good video performance without any motion artifacts.

The uniformity of the emission from pixel to pixel is the hottest issue in FEDs besides spacer technology. Because the emission efficiency of SCEs is 3%, most of the current is flowing in the cathode plate. The fairly good luminance uniformity of the SED implies that the tunneling currents of the individual pixels are identical within 2%. This is achieved by applying a carbon nano-gap to a palladium oxide (PdO) thin film as part of the process of making a rupture in the PdO film and depositing carbon on top of it. Driving schemes for matrix displays such as PDPs, LCDs, and FEDs are designed not only to generate gray scales, but also to improve luminance uniformity. The SED applies pulse-width modulation; however, no details were disclosed.

The carbon source is an organic vapor such as acetone, and the nature of the deposited carbon is graphite. This is one of the key processes for generating uniform emission characteristics in a SED all over the screen. At this moment, the peak luminance of an SED is limited to 400 cd/m2. It was said that the peak luminance will be increased to 500 cd/m2 in the near future.

The SCE current was presented as being very stable during (accelerated) life test; however, since the phosphor loading at 10 kV is 3–4 times larger than in CRTs, the aging of the phosphor will probably determine the end of life.

The audience got a good impression of SED at the Boston presentations, but it is still too early to conclude that FEDs are finally unchained. SED, Inc., has not started manufacturing yet, which implies that SED, Inc., has no information on yield or real cost.

Motorola and Sony/Candescent were also able to show nice FED prototypes in the past, but either failed in manufacturing or decided not to start manufacturing at all. The pioneering FED company PixTech went bankrupt in 2002. From this recent history, it can be concluded that the manufacture of FEDs is difficult. SED, Inc., may need deep pockets to climb the manufacturing learning curve.

Another question is how will SED, Inc., position its 50-in. W SED – the company's first intended product – in the market? SED is certainly a low-power alternative to PDPs and large-format LCD TVs, but there are many other comparative issues.

Of course, these questions were not addressed in Boston. We will learn the answers in the next 2 years. I wish SED, Inc., a lot of success in challenging its two big brothers.


Daniel den Engelsen, Fellow of the SID since 2002, developed CRTs and FEDs before his retirement from Philips Display Components, LG.Philips Displays Netherlands B.V., P.O. Box 807, 5600 AV Eindhoven, The Netherlands; telephone +31-(0)-40-2304-642, fax -160, e-mail: daan.den.engelsen@ He is now a Visiting Professor at Southeast University in Nanjing, China.