2013-06-14
A Biased Look at the Prospects for Sapphire
The substrate provider Rubicon Technology has released a sponsored white paper that paints a rosy picture of sapphire's prospects, while slighting the competitive technologies.
The paper (PDF) was authored by industry analyst Jamie Fox of IMS Research (a subsidiary of IHS Electronics and Media). It gives an overview of the role of sapphire in the LED market, where 2-inch and 4-inch wafers form the substrates on which the majority of LEDs are grown. (About 7 percent of industry capacity, by area, is devoted to 6-inch and 8-inch wafers.)
The white paper doesn't estimate what proportion of the LEDs being produced use something other than sapphire as a substrate.
The report defines the total market for LEDs as being made up of backlighting for cellphones, tablets, laptops, and TVs; general lighting; automotive; and display/signage. The analyst notes that most LED producers do not distinguish among these applications in terms of the sapphire they order, and therefore, sapphire producers that offer only one substrate regardless of the end use will have advantages of simplicity and economies of scale over those that use different substrates for different applications. We can guess which camp the report's sponsor falls into.
General lighting will be the largest application for LEDs by 2014, as backlighting will be either saturated or beginning to decline, this report confirms, echoing other industry analyses. The report claims strong advantages for sapphire in the lighting arena, where high reliability, long life, and resistance to high temperatures are particularly important in the substrate.
Slighting the competition
This report provides a data point based on interviews with industry insiders, but its point of view is decidedly favorable to that of the company sponsoring the research. Here is what the report says about alternatives to sapphire as a substrate for LED epitaxy:
LED manufacturers have been on the lookout for alternatives to sapphire as a substrate. GaN and ZnO can be used as substrates, but the costs are high. Silicon is still in the developmental stage as there are issues with LED die quality and thermal expansion coefficients that require the additional cost and complexity of a buffer layer. There are no other likely alternate candidate materials. It is expected that either sapphire will continue to dominate in the future, silicon will take over, or there will be a mixture. Many people view the mid- to long-term prospects for silicon as an LED substrate as good.
Silicon substrates will fight for supremacy with sapphire on the battlefield of 6-inch and 8-inch wafers, we are told. Bridgelux, backed by Toshiba, will be in volume production of GaN-on-Si in 2013, and "by 2015 silicon may have a notable share of the market for the first time," according to the report.
Startup Soraa is not mentioned. This company has pioneered GaN-on-GaN technology, and its products arguably demonstrate some of the advantages Soraa claims for GaN as a substrate material: radically fewer defects, leading to higher brightness and improved spectral characteristics.
And Cree, which bases its LEDs on silicon carbide, is mentioned only in passing; the company uses SiC only for "historical reasons," apparently.
There is good industry data in this report, but read it with eyes open to the biases likely to inhere to any corporate-sponsored industry white paper of this kind.
Related links
— Keith Dawson 
, Editor-in-Chief, All LED Lighting