Silicon wafers to solve blue laser supply problem?

Blue lasers. The little diode is at the heart of Blu-ray and HD DVD technology, and its scarcity is the reason you might be reading this while camping out for a PlayStation 3 right now. Shimei Semiconductor Co. thinks it's found an easier way to make them by growing the gallium nitride LEDs on a silicon wafer instead of the sapphire-based process used currently. The predicted lower cost and longer lifespan of the components sounds great ...too bad these aren't expected to be available until April of 2007. Still, those waiting for a dual-format player -- or maybe a European PS3 -- might have one of these blue lasers in their future.

[Via CNet]

[james stringer @ Nov 10th 2006 5:14PM]

Might be great for the Slim PSThree when it is released! And Philly wont confirm the launch date for PAL PS3, he said there are some verysmart people working to try catchup...so maybe that means that it could be before march even?

[Alcaron @ Nov 10th 2006 5:24PM]

So, another reason to wait on buying the PS3...potentially get a rev. 2 with a longer lifespan laser?

[JS @ Nov 10th 2006 5:47PM]

blue silicon LED != blue silicon laser

[primate @ Nov 10th 2006 6:15PM]

i've seen a few attempts to grow compound semiconductor (eg. III-V materials) on silicon. All have failed -- most famous one is motorola:
http://www.eetimes.com/news/semi/showArticle.jhtml;jsessionid=MD15MF4KMKO4AQSNDLOSKHSCJUNN2JVN?articleID=10800321

others that did GaN on silicon (back in 2004) are:
http://www.azom.com/details.asp?newsID=1686
http://compoundsemiconductor.net/articles/news/8/1/34/1

we don't hear lots of stories about these guys for a reason...

Growing compound semi on silicon is inherently challenging. CTE mismatch & Growth Temp mismatch lead to some fun failures. CTE is coefficient of thermal expansion which is of course different for the two materials -- building up crazy levels of strain across the temperature range. The materials are depositied epitaxially via vapor deposition which means that the conditions have to be at the vaporization temperature of the material being deposited which is different for GaN than Si. The built in strain then causes the material to self crack.

Most of the research has been on buffer layers which allow the spread of the strain across two+ material interfaces. The result is that you have to balance:
a) the cost savings of integrating driving IC and light source on same substrate/package coupled with _Poor yield_ of this same chip
versus
b) building the better yield components discretely

more power to them if they have solved this, but this has been a longstanding problem...

[chris @ Nov 10th 2006 7:38PM]

I've heard of wafer bonding techniques for making lasers (in fact, Intel and a team at UCSB just made a hybrid silicon laser using this with InP and Si), but the process takes hours to complete- which certainly won't get a PS 3 into your hands any faster. I certainly hope they won't be trying to rush thousands of GaN lasers on Si...

[Alan Erickson @ Nov 10th 2006 6:45PM]

It's going to be interesting to see how reliable these DVD's are over time. Both MS and Sony have had drive problems in their last generation of consoles, and BlueRay seems like it's even pushing things further. What's going to happen the first time you get a few scratches on one of your disks?

[... @ Nov 10th 2006 9:43PM]

As someone that grows lasers for a living (1550nm lasers for cable tv) I can say that it will be a long wile before things start to take shape...

I can also say that that pic is faked, as all lasers (except VCSEL's) don't emit out the top of the laser like shown in the pic. They emit parallel the the wafer, which is strait into a piece of silicon in that pic. Also, you wouldn't get any lasing action because you don't have any mirrors on the ends of the cavity, so at best you would get a little blue glow... To test them you have to scribe them out of the wafer, coat them, then solder them down and power them.

I would really like to see progress being made in the violet/blue/green laser area, I can't wait until the day when you can get a laser pointer with a violet/blue/green/red (pick one) for $5 at the local bargin shop :-]....

[Gil @ Nov 11th 2006 3:03AM]

"As someone that grows lasers for a living (1550nm lasers for cable tv) I can say that it will be a long wile before things start to take shape..."

I can't help imagining someone in a whiteroom suit going into a garden watering silicon plants with blue lasers on the ends

[... @ Nov 11th 2006 3:47AM]

LOL, wow never though of it that way... If only they did grow on trees instead on inside multi-million dollar machines... Maybe (just maybe) we could finally pull off a profit x_x