Vertical Divider

Playnitride Targeting End of 2017 to Start Production of Micro LED Displays
September 11, 2017

Charles Li, chairman and CEO of Taiwan-based Micro LED maker PlayNitride, which earlier in the year was rumored to be a takeover candidate by Samsung, claims it takes about 10 seconds to mass transfer 200,000 Micro LED chips in laboratory and it will take 10 minutes to produce a 5-inch Micro LED smartphone panel. Li believes the key factor affecting Micro LED panel production has shifted from how to break through the technological bottlenecks to how to decrease production cost. PlayNitride plans to start trial production of Micro LED in the second half of 2017, Li said. While the process of mass transferring Micro LEDs has often been cited as a major volume production barrier, PlayNitride has attained lab yield rates of over 99% in mass transferring and placing Micro LED chips, Li noted. He says a Micro LED smartphone panel costs about US$300, much higher than AMOLED's US$70-80 and LCD's US$15. Therefore, it is difficult to use Micro LED panels in smartphones for the time being, Li noted. For commercializing Micro LED technology, it is necessary to seek feasible application in terms of cost, such as smartwatches, automotive transparent displays and VR (virtual reality)/AR (augmented reality) devices, Li indicated.
Micro LED chip sizes are about 1% the size of standard LED chips and their structural arrays allow pixel pitches to be reduced to micrometer level, facilitating resolutions over 1,500ppi, much higher than 400ppi for Retina LCD displays. Note OLED smartphone displays typically reach 700-800 ppi. For use as displays, Micro LED chips must be mass transfer repeatedly. There are several technologies for such repeated mass transfer. US-based LuxVue Technology owns a technology patent for mass transfer based on static absorption. US-based lux, of which Foxconn Electronics and Sharp are shareholders, applies fluid dynamics to mass transfer. Taiwan government-sponsored ITRI has developed mass transfer technology based on electromagnetic absorption.
The current status of PlayNitride seems quite promising, but there is quite a bit if work remaining.

• The transfer rate yields needs to be more like 10-4 than 10-2, a 2 or 3 order of magnitude improvement
• If Micro LEDs were to target smartphones, the most likely choice would be the high-end, where the margins are substantial. High-end smartphones have resolutions of 2960 x 1440, which translates to 12.8m sub-pixels or more depending on the sub-pixel layout.
• At 200K pixels/10minutes, total transfer time would more like 27 minutes minutes than Li’s 10 minutes
• But, on average a Gen 6 OLED Fab produces one 5.7” panel every .65 seconds, unyielded
• To achieve the equivalent production capacity of a Gen 6, this version of Micro LEDs would require 2,500 transfer tools
• Given that the OLED deposition and encapsulation equipment for a 15K Gen 6 fab with 15K substrate/month capacity cost around US$2.0B, the average tool transfer tool would need to cost ~ US$800,000 to have a comparable cost base to OLED deposition and encapsulation.

However, there are additional costs for Micro LEDs, which must also be encapsulated and require some form of redundancy to account for any Micro LED failures, which are likely to happen. UC Santa Barbara claimed that minimum redundancy of 2X or possibly 3X would be necessary depending on the failure rate of the LEDs.  Mitigating the higher volume could be some designs that share Micro LEDs on a 2-pixel basis, which might cut the number of sub-pixels down by 1/3.
This reality should be enough to scare panel makers away from the smartphone market, right now, where rigid 5.5” OLEDs sell for US$12.50 now and will likely drop by 2020, when more production kicks in.  The use of a Micro LED for smartwatches as suggested by Lee may make sense but the average OLED panel cost is <US$2.50.  Typical resolution is 320x320, so a panel would take 15 seconds to fully load without redundancy.
There is, obviously, more work to be done as moving from the test lab to production usually involves a doubling or tripling of the costs until very high volume kicks in. Moreover, targeting the high-end smartphones or for that matter smartwatches would likely require a flexible backplane, not a showstopper for Micro LEDs, but certainly more difficult and more expensive. Another issue would be the price of the Micro LED chips, which for smartphones, should not be more than 40% of the total cost. If 12m Micro LED were needed, the cost/chip would be US$0.000002, not much revenue for an LED chip maker used to getting US$0.01 per LED chip. Nonetheless, the semiconductor industry has been able to reduce costs significantly over time so long term this may not be an issue.