Emissive Materials Generate Excitement at the Show
MicroLEDs created the most buzz at Display Week 2018, but quantum dots and OLEDs sparked a lot of interest too.
by Ken Werner
THE emissive materials that matter to displays are micro light-emitting diodes (µLEDs), organic light-emitting diodes (OLEDs), quantum dots (QDs), and phosphors. While phosphors remain a very mature technology with little new innovation, the other three material types are evolving rapidly and there were many new developments available to see on the show floor at Display Week 2018 in Los Angeles.
MicroLED displays – displays consisting of micron-scale inorganic LED chips that are not individually packaged – have been much discussed but seldom seen. At Display Week, though, several µLED technology demonstrations could be seen on the show floor, and at least two others were shown behind closed doors. And there was extensive coverage in the technical symposium.
In his Monday seminar, “MicroLEDs: Recent Advances and Applications,” Jongseung Yoon of University of Southern California observed that some of microLED’s advantages are infinite contrast ratio (similar to OLED’s), a response time in the nanoseconds (compared with microseconds for OLED and milliseconds for LCD), long lifetime, low energy consumption, high viewing angle, and extremely high pixel density.
The Canadian company VueReal, which did not appear to be represented at Display Week this year (the company presented an invited paper in 2017), reported 6,000 ppi with microLEDs last year. That’s impressive and useful for microdisplays and projection, but for direct-view applications such as TV sets, this level of pixel density isn’t needed and cannot easily be supported by the rest of the system. So, technologies for selectively picking up microLEDs from the wafer on which they were made and depositing them on a substrate with much greater pixel spacing to make a “sparse matrix” are crucial if µLEDs are to be a major display technology for consumer electronics.
In their paper, “Status and Prospects of microLED Displays,” Eric Virey (Yole Development) and Nicolas Baron (KnowMade) wrote, “Assuming traditional pick and place equipment could manipulate such small devices, those tools typically deliver processing speeds of around 25,000 units per hour (UPH). At this pace, it would take more than 1 month to assemble a single display!” The authors continue on to say that in order to be cost-compatible with most consumer applications, microLED chip transfer must reach rates of about 50 to 100 million per hour. The authors mentioned X-Celeprint’s polymer stamp transfer process, Apple/Luxvue’s electromagnetic and electrostatic MEMS approaches, and eLux’s fluidic assembly. Virey also commented that µLED chips are less efficient than LED chips of normal size, “but it just has to be better than OLED.” (Virey wrote an article on microLED technology for the May/June 2018 issue of this magazine.)
In the invited paper, “MicroLED Displays: Key Manufacturing Challenges and Solutions,” Ajit Paranjpe and colleagues from Veeco said that meeting cost targets for µLED displays will be challenging. Veeco believes: “A two-step mass transfer approach using a dense interposer substrate or cartridge provides the benefit of maintaining low overall transfer costs while increasing epitaxy wafer usage and yield through the use of smaller transfer fields.” In other words, instead of using microtransfer printing (µTP) to directly transfer µLED chips from the wafer to the final large display substrate, Veeco suggests it would involve less cost and produce greater chip consistency if chips were extracted from a relatively small area of the wafer and deposited on a relatively small interposer substrate. Then the interposer substrates will be transferred onto the final display substrate. The paper makes a convincing argument for this approach.
Looking at µLED Hardware
Behind closed doors, X-Celeprint showed a technology demonstration of its 5.1-in. µLED display. Until I saw the AUO display (see Fig. 1), I believed this to be the largest microLED display yet constructed. The display has 70 pixels per inch, active-matrix switching using micro-ICs (not TFTs), pixel-level compensation, subjectively very high contrast, and highly saturated colors including a very red red.
X-Celeprint has accumulated an extensive IP portfolio on µTP, including the transfer of µLED chips and the transfer of microcircuits and other small objects. The company intends to spin off a new company devoted exclusively to displays within the next 12 months. (Disclosure: The author is a member of an X-Celeprint advisory committee, for which he is paid modestly – very modestly.)
In a private conversation, Optovate’s Graham Woodgate and Paul May updated me on their company. Founded by former Sharp and CDT people some years ago, the company developed significant IP only to find that nobody was interested in µLED displays at the time. Now interest is high, and the company is being re-energized. Optovate’s IP focuses on two areas. The first is removing the LED chips from their wafer via patterned laser lift-off (p-LLO) rather than using a sacrificial layer process that separates all the LEDs from the wafer at once, with chip selection relegated to a separate step. The second major focus is sheets of catadioptric optics for controlling the angular emission of the LEDs. One application involves thin miniLED backlights for LCDs that rival OLEDs in thinness and functionality. MiniLED backlit displays with local dimming (some using Optovate technology) appeared in several booths on the show floor, and Paul May said Optovate was receiving “lots of interest on the optics part now.”
AUO won a Best in Show Award in the medium exhibit category at Display Week, in part for its 8-in., full-color µLED with 1,280 × 480 pixels (Fig. 1).
Fig. 1: AUO’s 8-in. µLED display is the largest shown publicly to date. Photo: AUO
AUO’s Norio Sugiura said each LED measured less than 30µ, and that the display was driven with an LTPS backplane. The display uses color conversion technology, which means it employs only blue µLEDs and presumably obtains red and green with phosphors or quantum dots. I asked if the display uses a mass-transfer technique (such as µTP) or whether it was assembled from relatively large sections of two or more LED wafers. (The latter approach would be expensive and totally impractical for volume production, but could be a way of getting a demo to the show floor if you had not yet mastered an appropriate mass-transfer technology.) Sugiura would only say, “That is a very good question.”
Hong Kong Beida Jade Bird Display, an I-Zone honoree at Display Week this year, showed a green µLED display with 3 million – yes, million – nits (Fig. 2). This first-generation, proof-of-concept display measured 0.65 inches on the diagonal, had a pixel pitch of 20 µm, and pixel dimensions of 640 × 480. A different chip had 2,560 × 1,920 pixels. Jade Bird, which was founded in 2015, is targeting its displays for projection.
Taiwanese company PlayNitride was also an I-Zone honoree, for “utilizing its PixeLED display technology to build a transparent display with an innovative and unique process to transfer RGB microLEDs onto a pixel.” The fabless company, established in 2014, focuses on gallium-nitride µLEDs, which the company says it can make at pixel densities up to 1,500 ppi. In April 2017 there were rumors that Samsung Display was interested in acquiring PlayNitride. In April 2018, several online publications reported that PlayNitride was discussing a cooperative arrangement with Apple. True or not, an Apple investment rumor tends to make investors more patient.
There was lots of µLED activity at Display Week, and lots of work remains to be done. Companies such as Jade Bird and New York start-up Lumiode, which are focusing on microdisplays and don’t have to tackle microtransfer issues, have a shorter path to commercialization. But they are also leaving the largest markets on the table.
Fig. 2: Jade Bird’s 0.65-in. green display has a luminance of 3 million nits! Photo: Ken Werner
The 800-Pound Gorilla in the Quantum-Dot Room
If you buy a quantum-dot or QD-enhanced LCD TV, it will have quantum dots made by or licensed by Nanosys. There are other quantum-dot companies operating in this industry – but for some reason those dots don’t appear in products any of us buy.
So what do you do if you have a virtual monopoly on the display market for quantum dots? First, because quantum dots are currently used only in premium sets, you realize you don’t want to limit your customers to those whose names begin with “S.” Also, you want to develop QD products and architectures that will improve TV performance and lower cost. And you don’t want to limit applications only to LCDs.
At Display Week, Nanosys’s Jeff Yurek said, “Quantum dots are the technology platform for all future displays. We don’t care where the photons come from.” In other words, the color conversion performed by QDs can be just as useful for downconverting the light from blue µLEDs as from the blue LED backlight in an LCD TV.
Yurek added later that AUO was showing “a fantastic QD gaming monitor” and that Tianma was showing a demo with 90 percent BT.2020 based on what Tianma called “less-Cd [cadmium] Dots.”
In its booth, Nanosys showed technology demos that included one for ink-jetted QD matrices to replace the matrix color filters used in current LCDs. Another showed early-stage electroluminescent QDs. These demos were shown privately at CES. At Display Week they were available for all to see.
Also shown in the booth, and new to me, was the Vizio P Series Quantum 65-in. TV (Fig. 3).
Fig. 3: The Vizio P Series Quantum TV with Nanosys’s Hyperion QD film and 192 full-array local dimming zones delivered an impressive image. Photo: Ken Werner
The set uses Nanosys’s Hyperion low-cadmium QD film and has a full-array backlight with 192 local-dimming zones. Peak brightness is greater than 2,000 nits, according to a data card next to the exhibit, with DCI-P3 coverage of greater than 98 percent and BT.2020 coverage of greater than 80 percent. This impressive-looking set was scheduled to be available in a few months at an MSRP of $2,200, according to the card. This price compares with an MSRP of $3,500 for Samsung’s stunning 65-in. Q9F with 500 local dimming zones. It would be interesting to compare these sets side by side.
Also in the booth was a side-by-side comparison of the Samsung Q9 and an unidentified OLED TV. We reported on this comparison at CES, where the Q9 competed with the OLED very effectively. What was unexpected here was the serious burn-in of a logo in the OLED set (Fig. 4). Yurek said the set had only been running for about 60 hours and offered to show me the dated sales receipt. The burn-in was a complete surprise to everyone who saw it, and it’s hard to believe this is typical of modern OLED TV sets.
Fig. 4: The Familytime logo (left) appeared almost constantly on both the OLED set and the Samsung Q9 set in the Nanosys booth. The logo burned into the OLED set (right) after about 60 hours, according to Nanosys’s Jeff Yurek. There was no burn-in on the quantum-dot-enhanced LCD TV. Photos: Ken Werner
TADF Players on the Show Floor Ranged from C to K
A blue OLED emitter with long life, high efficiency, and “deep-blue” color coordinates is the holy grail for OLED materials developers. Everybody is looking for it and nobody has found it. Thermally activated delayed fluorescence (TADF) is a clever quantum-mechanical trick that permits us to make use of the three quarters of quantum states in fluorescent OLED emitters that are usually unavailable, thus increasing the internal quantum efficiency from 25 percent to 100 percent. Universal Display Corporation (UDC) already does this with phosphorescent OLED emitters, which appear in commercial OLED displays ranging from the smartphone displays made by Samsung Display to the TV displays made by LG Display.
If you are hopelessly cynical, you might say that the main reason to pursue TADF is to find a way around UDC’s impressive patent portfolio. If you have a more benevolent nature, you might observe that although UDC has very nice red, green, and yellow phosphorescent materials, the company has not yet been able to develop an efficient, long-lived, deep blue that would make an all-phosphorescent OLED display possible. Current displays use UDC’s phosphorescent red and green (or, in LG Display’s case, yellow) and then fill out the spectrum with a less efficient fluorescent blue.
There may be uses for TADF greens, reds, and yellows, but once you have acknowledged the creative synthetic chemistry and quantum physics involved, these materials are not going to change the OLED display game in any fundamental way. What the industry wants is a “deep blue,” as opposed to a “sky blue,” which may have applications in lighting but not in displays, at least not in any straightforward way. (UDC has demonstrated a four-subpixel configuration that uses both sky blue and deep blue to reduce the aging of the short-lived deep-blue emitter, but this has not appeared in any commercial display as far as I know.)
In the past few years, the general attitude toward TADFs has evolved from skeptical curiosity to hopeful respect. In his Sunday seminar, “OLEDs: Recent Progress and Applications,” Jian Li of Arizona State University said, “I used to be a doubter.” Although he no longer doubts, Li said, “[It is] difficult to push TADF design forward.” However, people are developing new device architectures in their search for “a new route to harvest triplets.”
If OLED Generation 1 is fluorescence, Gen 2 is phosphorescence, and Gen 3 is TADF, then a more recent approach, TADF-assisted fluorescence (TAF), could be considered Gen 3.5. The advantage of TAF, said Li, is that it expands the molecular design possibilities beyond TADF, thus increasing the likelihood of being able to synthesize molecules with the desired color coordinates (such as “deep blue”), lifetime, and efficiency.
On the show floor, the range of TADF developers ran from C to K; that is, from Cynora to Kyulux. Kyulux, the Japanese company founded in 2015 on the basis of technology licensed from Kyushu University, showed green, yellow, and sky-blue TADF OLED emitters. Daniel Tsang told Information Display that the green and yellow are in customer development, while the sky blue is still at the in-house development stage (Fig. 5). A deep blue is possible, he said, but lifetime still needs improvement.
Fig. 5: Kyulux’s sky-blue TADF OLED emitter appeared in the company’s Display Week booth. Photo: Ken Werner
In their paper, “Progress of Highly Efficient Blue TADF Emitter Materials Toward Mass Production,” Thomas Baumann and Matthias Budzinsky of the German company Cynora said that one problem with deep-blue emitters is that they have too broad an emission spectrum. However, by using molecular design principles, Cynora increased the percentage
of narrow emitters from about 15 percent to nearly 50 percent over the first three quarters of 2017. The team’s most recent results “of 14 percent EQE [external quantum efficiency] at a CIEy coordinate of 0.15 [deep blue] with a lifetime LT97 [lifetime measured to 97 percent of initial luminance] at 700 nits of ~10 hours show that blue TADF emitters are not far from mass production specifications.” The specifications for volume production “are typically considered to be around 0.15 CIEy, 15 percent EQE and >100h LT97 at 700 nits. We therefore expect that our blue TADF emitters can be found in first products by 2019 after all the necessary processing tests will have been done in 2018.”
In the company’s booth, Cynora CEO Gildas Sorin gave Information Display more recent results for TADF blue: 20 to 26 percent EQE and LT97 (700 nits) of 20 hours. Early last year that lifetime was only 1 minute, Sorin said. This compares to an EQE for fluorescent-blue OLEDs of 7 percent and an LT97 of 150 hours.
Since the company has solved two of the three key problems, said Sorin, Cynora can now focus on lifetime exclusively, a much easier task than having to work on color coordinates, efficiency, and lifetime all at once, which is the challenge facing Cynora’s competitors, according to Sorin. However, Sorin was more conservative than his technical colleagues in predicting commercial introduction. He said he hopes that Cynora will achieve the lifetime goal in the middle of 2019, with commercialization at the end of 2020.
There were also lots of LCDs discussed in the technical sessions and shown at the exhibition at Display Week, but the excitement seemed to be generated by emissive displays and the materials and processes that enable them. •
Ken Werner is the principal of Nutmeg Consultants, specializing in the display industry, manufacturing, technology, and applications, including mobile devices, automotive, and television. He consults for attorneys, investment analysts, and companies re-positioning themselves within the display industry or using displays in their products. He is the 2017 recipient of the Society for Information Display’s Lewis and Beatrice Winner Award. You can reach him at firstname.lastname@example.org.