If you have heard of Corning Incorporated, it might have been for CorningWare, Corelle or Pyrex dishes in the past; perhaps for the glass on that big screen TV; or even as the ones behind Verizon’s FTTH (Fiber to the Home) initiative; or possibly you knew it was behind the catalytic converter in your car; or maybe you know about Gorilla Glass in so many phones and tablets.
But you probably don’t think of Corning as a laser company. Yet one of the big ‘cool tech’ stories from last year, the Microvision SHOWWX projector that was going to offer the ability to simply hook up to your iPhone and do a corporate grade production, was fueled by Corning’s ‘synthetic’ green laser.
That makes it even bigger news that Corning recently ‘shelved’ the Synthetic Green Laser program. Why? Read on and see!
So what is a ‘synthetic’ laser compared to a ‘native’ laser?
This might get a bit esoteric, so I forgive you in advance if you nod off … and I promise not to be deriving Maxwell’s Equations anywhere in the article!
For me, the best way to describe the problem is to look at the image at the top. You see (unless you are color blind) that red light comes from a small diode laser at the upper left, and blue light comes from a small laser diode lower and to the right. Then you notice that the green light comes from a HUGE box in the middle of the two diodes! What gives?
Green light is notoriously hard to produce from a direct laser perspective. Way back in the late 80’s I worked at a research instrumentation company and we had a product that used an early ‘solid state’ green laser that was about the size of a very large loaf of bread. The system used lamps to excite a Neodynium-doped YAG (Yttrium Aluminum Garnet) crystal, which output light at 1064nm. Then it is sent through a BBO (Beta barium borate) nonlinear crystal that propogates a nonlinear polarization wave that is twice the frequency of the original light but also phase matched to the original wave, allowing energy transfer from the original wavelength to the doubled frequency.
But what does that MEAN? It means that you had to use a bunch of energy to generate a moderate amount of light (normal lasing process), and then eat us most of that light trying to generate light at a different wavelength. And if you care about mode control, frequency stability, and so on … you lose even more of the light you want.
Contrast this with red and blue lasers: each of these lasers can be directly created using a compound semiconductor (GaAs for red and GaN for blue) with layers of atoms forming quantum wells that allow for the light amplification needed for lasing to occur. You get a direct output with the power and stability, and merely need to efficiently direct the light output.
In the case of the PicoProjector technology of Microvision, Corning needed to squeeze that same technology into an ultra-small package. The image above shows the ‘first generation’ of such a device, which is what Microvision built their devices around.
So is it as simple as doing the same thing as I did back in the 80’s … but smaller? Not really – you start with the same basic concept as before but really need to change the entire way of accomplishing things. You start with a solid state infrared laser with 1064nm output, and put it through a non-linear doubling crystal to get the green output as before.
But in this case you really need to concern yourself with wavelength and mode control, meaning there are more optics and electronics involved as well as tighter tolerances. Also, since infrared light basically means heat, and also you have electronics which also mean heat … and since the alignment of elements as well as their shape and performance are all impacted by heat … you have a control system dream/nightmare on a very small scale.
But since a microprojector is a consumer device, and therefore you need to deliver consistent performance in a power-conscious package. The first generation of PicoProjectors were nifty gadgets, but pretty much useless in a normal setting. Gone are the days when you would completely darken a room for a presentation – but that is what these devices required.
Earlier this year Corning introduced the G-2000 laser, which was aimed at the next generation of PicoProjectors. These 20 Lumen projectors would actually be useful for real people to do real presentations!
In fact, financial analysts said that “PICOPROJECTORS WILL DRIVE THE GREEN LASER MARKET”, which led to a very bullish feeling feeling that soon people would be carrying a smartphone and projector with them and not much else.
So what was the problem? Why would Corning kill their green laser program in such an environment? Simple – all of this stuff looked great on paper, but unfortunately the ramps in orders and production that Microvision kept predicting simply wasn’t happening. Part of it is the price – the ShowWX costs $550, more than an unlocked cell phone! So you have two laser suppliers (Corning and Osram) building lasers based on a large predicted ramp … and therefore an unsustainable backlog. And there is also something else I’ll get to soon …
Earlier in the year Osram had announced that they were suspending further green laser development, but were going to continue to supply lasers for the Microvision PicoP module.
So why is everyone suddenly abandoning the green laser for pico projector boat just as analysts are predicting it as the ‘future of the green laser market’?
Simple – native green lasers.
Native green lasers would replace the complex box at the top with a simple small laser diode just like the ones producing red and blue light. Then you would have a lower power requirement, less heat dissipation, lower cost, and so on. It is the ultimate goal …
… and it is almost here!
Over the last couple of years researchers at Samsung, Osram, Corning, and other places have shown the direct generation of green light using InGaN (indium gallium nitride) semiconductor. The wavelengths have ranged from 510nm – 531nm (green is usually defined as 515-535nm, and the ‘standard’ established by frequency doubled lasers is 532nm), but the problem has been the power output.
To put it into context, in a 2009 Forbes article, Corning said “can make a 1-milliwatt native green laser, he says. (Its synthetic green laser pumps out 60 milliwatts of light.) Industry would like 100-milliwatt devices.” That has been the issue – for a true room-capable projector you need 100W, but the best sources are well short of that goal.
But what Corning and Osram realize (which makes sense, since they are leading researchers) is that the power gains for native green lasers have been coming more rapidly than anticipated. So instead of being 5+ years away from a 100mW 532nm source, we are now only perhaps 2 or 3 years away!
To me that is technically amazing, knowing how difficult it is to accomplish this and how many year folks have been struggling with the solutions and implementations, to finally see it coming to fruition!
But think of the business models – you have a dead-end technology (synthetic green lasers) with a slower-than-expected adoption rate accompanied by nothing that builds confidence of an incipient ramp anytime soon, and on top of that you have an accelerating technical ability to deliver on the solution everyone wanted in the first place! It is definitely not the place a company wants to be investing, and that is why the major synthetic green laser makers have ‘cut the cord’.
But as I said at the start, the death of the old green laser is something to rejoice, not mourn! Because it comes along with the birth of a new and better laser technology that will enable not just pico projectors and other commercial products, but also enable embedded spectroscopy and other research devices that benefit greatly from green light but have been held back by the bulk and complexity of current solutions.