Meta: Why did we choose SiC?
On March 6, Meta (formerly Facebook) published an article on its official website, describing the process and advantages of choosing silicon carbide as the core material when developing AR glasses waveguide technology.
The Meta team has not only solved key bottlenecks such as field of view, weight and optical artifacts of AR glasses through silicon carbide waveguide technology, but also sees it as a "game changer" in the AR industry, which may become a mainstream material in the future:
Meta Orion team explains: Why choose SiC technology
In 2019, the Orion team prepared Meta founder and CEO Mark Zuckerberg for a pivotal demonstration of the potential waveguide technology for augmented reality glasses - the moment when theoretical calculations on paper became reality for the first time and revolutionized the trajectory of subsequent development.
Meta released AR glasses -Orion
Pascual Rivera, a Meta-optics scientist, recalls: "When wearing glasses with glass base waveguides and multiple laminated panels, it felt like being in a disco - there were rainbow spots everywhere, and the interference was so strong that it was impossible to see the AR content. But when you put on the prototype glasses with silicon carbide waveguides, it is instantly like being in a symphony Hall listening to a quiet classical movement, and your attention is always focused on the complete experience we have built. It's a total game changer."
However, while the choice of silicon carbide as a substrate may seem obvious today, it was far from a given when the Meta Orion team embarked on the development of AR glasses a decade ago:
Pascual Rivera explained that silicon carbide is often heavily doped with nitrogen, which makes it appear green, or even black if it is thick enough. Such a material simply cannot be used to make optical lens-it is essentially electronic, and its color is closely related to its electronic properties.
Giuseppe Calafiore, head of waveguide technology at Meta AR, adds that silicon carbide has a long history as an applied material, mainly in high-power electronics. Take electric cars for example: All electric cars require a chip that can withstand extremely high power to drive the wheels and complete vehicle systems. Traditional silicon substrates cannot meet this demand, and only materials such as silicon carbide that allow high current and high power through can be competent.
Before the renewable energy issue heated up in recent years, the market for such high-power chips was far smaller than for consumer electronics chips. In addition, the long-term price of silicon carbide is high, but due to the small amount of substrate for automotive chips, the cost is still acceptable, and manufacturers lack the motivation to reduce prices.
But it turns out that silicon carbide also has key properties required for waveguides and optics, and the parameter the Meta Orion team is most focused on is the refractive index. The high refractive index of silicon carbide means that it can conduct and output huge amounts of optical data - an analogy to Internet bandwidth: the greater the bandwidth, the more data can be transmitted within the channel. Optics follow the same logic: the higher the refractive index of a material, the greater its optical expansion, and the greater the amount of optical data transmitted through that channel.
Calafiore further explained that in our application scenario, the channel is the waveguide, and the greater optical expansion directly translates into a wider field of view. The higher the refractive index of the material, the larger the field of view that the display can support.
SiC refractive index up to 2.7: far more than glass, lithium niobate and other materials
When Calafiore first joined Oculus Research (Meta's research and development lab) in 2016, the highest refractive index glass they had was only 1.8 - multiple layers of glass had to be stacked to achieve the target field of view. Optical artifacts aside, the assembly process is extremely complex: the first two waveguides must be perfectly aligned, and then the entire stack must be perfectly matched to the third waveguide.
"Not only is this expensive, but it's also obvious that there's no way you can fit three pieces of glass into each lens." Calafiore recalled, "They were too heavy, and the thickness was far beyond the limit of aesthetics - no one would buy such products. So we went back to square one: trying to increase the refractive index of the substrate material, thereby reducing the number of glass plates required."
In the beginning, the research team first focused on lithium niobate, which has a refractive index of about 2.3, significantly higher than glass's 1.8.
Calafiore said that we realized that we could just stack two boards, or maybe even cover the field of view with one board. At the same time, we began to explore other materials - which is why we found excellent transparency in high-purity silicon carbide in our work with suppliers in 2019. More importantly, the refractive index of silicon carbide is as high as 2.7, setting a record for optical applications.
For the research team, this value means that the refractive index of silicon carbide is 17.4% higher than that of lithium niobate and 50% higher than that of glass. Calafiore explained, "It is possible to prepare transparent silicon carbide with only a small modification of existing industrial equipment. So we adjusted the process to strictly control the parameters - no longer optimizing for electronic properties, but focusing on optical properties: core metrics such as transmittance and refractive index uniformity."
Solving problems such as ghosting and rainbow effect: SiC technology finally stands out
At the time, the Reality Labs team was the first to attempt to convert opaque silicon carbide wafers into transparent substrates. Because silicon carbide is one of the hardest materials known, its cutting and polishing must rely on diamond tools, which leads to extremely high costs of non-repetitive engineering and ultimately expensive substrates.
Although there are more cost-effective alternatives to silicon carbide substrates, there are advantages and disadvantages to any technology, and Meta ultimately decided to go with silicon carbide. Silverstein, scientific director of Meta Research, explained that finding the ideal solution for wide-field AR displays is essentially a game of performance versus cost, which may be compressed, but if performance is not up to par, the cost advantage is meaningless.
At the same time, the Meta Orion's field of view is up to 70 degrees, and new problems such as gghost and rainbow effect begin to appear: gghost is a repeat image of the main image projected on the display, and rainbow effect is a dynamic color pattern formed by the reflection of ambient light on the waveguide surface.
For example, Silverstein explains, if you're driving at night and the headlights move around you like rainbow stripes, or playing volleyball on a sunny beach, the dynamic rainbow effect can cause you to miss your shot. One of the magical properties of silicon carbide is that it can completely eliminate these disturbances. Another unique advantage of silicon carbide is its thermal conductivity. Plastics are poor insulators, as are glass and lithium niobate, but silicon carbide is both transparent as glass and efficient at conducting heat - defying conventional wisdom.
Therefore, in July 2020, the Meta Orion team selected silicon carbide based on three core factors:
First, shape optimization: single-layer substrate and smaller support structure greatly reduce the volume of equipment;
Second, optical advantages: high refractive index and anti-rainbow effect improve the display quality;
The third is lightweight: compared with the double glass scheme, the weight is significantly reduced.
Meta solves the problem of slope etching: We hope that more enterprises will participate in the research and development of optical grade SiC
After the material was identified, the next hurdle turned to the fabrication of waveguides - specifically, an unconventional grating technique called bevel etching.
Calafiore explained, "The grating is the nanostructure responsible for coupling light into and out of the lens, and for the silicon carbide to work, the grating must be etched with a bevel. The etched lines are not arranged vertically, but are distributed at an oblique Angle.
Nihal Mohanty, research manager at Meta, added that they are the first team in the world to achieve slope etching directly on the device, and the entire industry has relied on nanoimprint technology in the past, but this cannot be applied to high refractive index substrates. For this reason, no one had considered the silicon carbide option before.
In 2019, Nihar Mohanty and his team partners jointly built an exclusive production line, before which, because the slope etching technology is not mature, most semiconductor chip suppliers and foundries lack relevant equipment. Therefore, at that time, there was no facility in the world that could produce etched silicon carbide waveguides, and it was impossible to verify the technical feasibility outside the laboratory.
Nihal Mohanty further revealed that it was a major investment and they built the complete production chain. The processing equipment was customized by the partners and the process was developed by Meta itself - initially the equipment was only up to research grade standards because there was no manufacturing grade system at the time, so they then worked with the manufacturing partners to develop the production grade bevel etching equipment and process.
Now that the potential of silicon carbide has been proven, the Meta team is looking forward to the rest of the industry starting to develop their own devices, because the more companies invest in optical grade silicon carbide research and development and equipment development, the more robust the industry ecosystem for consumer AR glasses will be.
SiC cost reduction and efficiency path is clear: it will shine in the field of AR glasses
While the Meta team is still exploring alternatives, a strong consensus has emerged: in the right market window, the right people are working together to drive the silicon carbon-based AR glasses revolution.
Silverstein and Giuseppe Calafiore said that prior to this, all silicon carbide manufacturers had significantly expanded production in preparation for the expected electric vehicle boom, and the current overcapacity situation did not exist when Orion was in development. Now, because of oversupply, the cost of the substrate has begun to fall.
The Orion project proved the viability of silicon carbide in AR glasses, and there is now strong interest from supply chains across three continents, with suppliers excited about new opportunities to manufacture optically grade silicon carbide. After all, compared to electronic chips, each waveguide lens consumes a larger amount of material, and their existing technical capabilities can be seamlessly transferred to this field, they are betting on this opportunity, silicon carbide will eventually win.
In addition, there are already manufacturers shifting from 6-inch to 8-inch substrates, and there are pioneer companies developing cutting-edge technologies for 12-inch substrates - which will make AR glasses production capacity increase exponentially. In the future, these developments will continue to drive costs downward, and while the industry is still in its early stages, the picture of the future is becoming clearer.
Calafiore believes that at the beginning of any new technological revolution, people will always try multiple paths, and television technology is an example: from the cathode ray tube to the LED plasma screen, and now MicroLED, we have gone through multiple iterations of the technology architecture. Most of the paths in the exploration are eventually falsified, but there are always a few options that are repeatedly chosen because of their great potential. We have not yet reached the end, nor can we fight alone, but silicon carbide is undoubtedly a miracle material worthy of heavy investment.
Silverstein concluded that they have successfully demonstrated the crossover potential of silicon carbide in electronics and photonics, and its future may shine in areas such as quantum computing. At the same time, the possibility of significantly reducing the cost of silicon carbide has appeared, although there are still many challenges, but its revolutionary energy is immeasurable.
ZMSH SIC Wafer 4H-N & Semi Type:
* Please contact us for any copyright concerns, and we will promptly address them.
