by David Fyfe and Terry Nicklin
"Standing on the shoulders of giants," a phrase borrowed from Isaac Newton, is used with due modesty on the circumference of the British two-pound coin. But it also resonates in the world of displays, reflecting the cumulative experience and research into ink-jet printing that is being harnessed for the manufacture of polymer-OLED (P-OLED) displays.
It is a wonderful thing when technologies developed for one purpose become available just in time to be useful in another. Ink-jet printing has become ubiquitous in fields ranging from sub-$100 home desktop printers, direct-mail envelope addressing, barcoding, decoration of ceramic products, large-format banners, high-quality proofing, and much more. While these applications require different flavors of ink-jet technology, the common link is the inherent ability of ink-jet-printing systems to deposit very small quantities of highly specialized fluids reliably and accurately at high speed onto an impressive variety of substrates. Applications now on their way to production status include ink-jet printing of color filters, printing of liquid-crystal-display (LCD) spacers, printing of metal "wiring" using nanometal inks, and others.
P-OLED technology, the solution-processable type of OLED, requires just this versatility, applicable as it is to uses ranging from cell-phone screens an inch or so across to the 40-plus-in. television displays now sought after by consumers the world over. The limiting factor in the adoption of OLEDs has been the lifetime and efficiency of materials, but because the rates of progress on these factors have been rapid and accelerating, the focus has turned to the means of manufacture and the need to develop cost-effective high-yielding processes of production, including cost-effective large substrate sizes.
Whatever the end-user performance advantages of OLED displays – and these are many and well known – successful penetration into the established LCD space likely will be as much due to the potential for cost reduction, at least for the display maker, as for the undoubted attractiveness of the display itself. The headroom for this is clear: the elimination of the backlight, color filter, and liquid-crystal layer, which collectively account for roughly 50% of an LCD panel's functional materials cost.
Which brings us back to ink-jet printing. In addition to reducing process steps, ink-jet printing's inherent scalability allows the economic benefit of larger numbers of displays being produced simultaneously on larger substrates. Already, Gen 4 machines are in use for development of P-OLED technology, and a monster Gen 7+ machine with a 2.4-m-square substrate platform has been shipped by Litrex Corp., a former Cambridge Display Technology (CDT) subsidiary now owned by Japanese equipment giant Ulvac. This machine actually is destined for use in LCD manufacture, but it serves as evidence of the versatility of ink-jet printing in display applications and the capability of delivering accuracy at very large scale.
So how do the other costs associated with manufacturing stack up for ink-jet printing? Well, the large, latest generation ink-jet systems sell for in excess of $4 million for systems dedicated to the printing of a single material in production. While this may sound expensive, the total capital expenditure required for particular process steps is expected to be one-half to one-third of that of conventional equipment. This, coupled with a smaller fab footprint, elimination of physical masks, quick product changeover times, and materials-utilization rates 3–10 times better than conventional processing methods, is driving its evaluation by display makers with both quality improvement and cost reduction in mind. Ink-jet systems have been designed for production applications from Gen 4 to Gen 8, with the smaller-sized systems obviously costing somewhat less, but with the same additional production flexibility, process simplification, and cost advantages.
Fully formed pixel layouts have been produced at fairly high resolution without the need for further process steps to create the pixel pattern. Seiko-Epson raised the profile of the powerful combination of P-OLED technology fabricated by ink-jet printing when it demonstrated a 40-in. TV in 2004, following Toshiba's ink-jet-printed 17-in. display exhibited in 2002. Last year, CDT showed a 14-in. display with a WXGA+ resolution.
Even the time-honored method of spin coating may be replaced by an overall "flood" printing achieved through printing a pattern of dots that coalesce to form an even film. Existing investment in LCD production capability need not be wasted, as the ink-jet-printing station can be integrated into existing lines. Operating speeds are already fully competitive with current LCD line standards – tact times of less than 2 minutes have been demonstrated – and well ahead of the 4 minutes typical of small-molecule OLED (SM-OLED) vacuum-deposition lines, which perform the equivalent function.
So, is ink-jet printing the panacea for production of P-OLED displays? It may be, but of course, there is some progress still to be made. Present systems are capable of up to 150 pixels per inch (ppi), but for higher resolutions of up to 200 ppi, further developments in printheads, printer hardware, software, and inks are required. Borrowing from experience in the graphics and coding sectors, Dimatix has a print head that utilizes microelectro-mechanical-systems (MEMS) etched nozzle-plate technology to achieve drop sizes of 8 picoliters (pl). Xaar is developing a 1000-nozzle head where constant recirculation of fluid removes air and dirt automatically, and which, it is claimed, will be self-cleaning and self-maintaining. By the end of 2006, the company expects to see drop sizes down to an impressive 3 pl, or one-tenth of that attainablein 2000. Other players are developing their offers: Konica Minolta has a head that dispenses 7-pl drops. Other proprietary tech-niques such as applying different voltage levels to each nozzle individually ("drive per nozzle") allow nozzle-to-nozzle drop-volume variation to be controlled much more precisely.
Manufacturing high-quality displays with high yield at these tolerances requires more than precise engineering: the software driving the printers also must be somewhat more sophisticated that the bundled "printer drivers" we are accustomed to from our desktop-printer experiences. Litrex, for example, has bespoke software to control the printhead and motion stages, as well as perform nozzle inspection and drop analysis. Graphical user interfaces allow the operator to "see" the tiny individual ink drops as they fly through the air, even though they may be emerging at rates of up to 15,000 drops per second.
Equally important is the formulation of inks because the "as-printed" performance of the highly tuned P-OLED chemistry must be maintained along with the physical properties required for reliable jetting. Partly to ensure that this aspect of the supply chain is in place, CDT has formed a joint venture with Sumitomo Chemical, called Sumation™, for the development and supply of such fluids. Merck, through its acquisition of Covion, also supplies into this space.
While ink-jet printing continues to be the favored approach for P-OLED display manu-facturing scale up, other methods, including roll printing and screen printing, are getting encour-aging results, too. Seiko-Epson, Philips, DuPont, Toppan, Toshiba, Delta Optoelectronics, and other leading companies have played a part in establishing the corpus of knowledge, making P-OLED a proven solution-processable technology.
The eventual cost of P-OLED displays will depend in part on the manufacturing technique employed, and the yield, reliability, and maintenance requirement achieved in practice at commercial production output levels. Early indications are that the high robustness levels being achieved with ink-jet printing in the graphics and other sectors should ultimately be achievable in our more demanding display market.