Stanford Breakthrough Could Make Better Chips Cheaper 56
angry tapir writes: Researchers at Stanford University have come up with a new way to make chips and solar panels using gallium arsenide, a semiconductor that beats silicon in several important areas but is typically too expensive for widespread use. "[I]t can cost about $5,000 to make a wafer of gallium arsenide 8 inches in diameter, versus $5 for a silicon wafer, according to Aneesh Nainani, who teaches semiconductor manufacturing at Stanford. The new Stanford process (abstract) seeks to lessen this thousand-to-one cost differential by reusing that $5,000 wafer. Today the working electronic circuits in a gallium arsenide device are grown on top of this wafer. Manufacturers make this circuitry layer by flowing gaseous gallium arsenide and other materials across the wafer surface. This material condenses into thin layer of circuitry atop the wafer. In this scenario, the wafer is only a backing. The thin layer of circuitry on top of this costly platter contains all of the electronics."
Re: (Score:1)
only used ones.
Watched the YouTube video but left wondering... (Score:1)
...if they can deposit a layer of GaAs on top of the sacrificial layer and make circuits out of that, then why do they need the bottom wafer at all? Why not add the sacrificial layers on something less expensive and then deposit the GaAs circuit layer on top of that?
Re: (Score:1)
Yeah I thought the same thing?????
Re: Watched the YouTube video but left wondering.. (Score:3, Informative)
To get the single-crystal purity of the surface layers, they need to perfectly match the crystal dimensions of the substrate. Making it out of the same thing is aa easy way to achieve that.
Re: (Score:3)
That makes sense if you're building devices directly on the wafer, but wouldn't the three sacrificial layers interrupt that?
=Smidge=
Re: (Score:1)
Nope.
I believe the two materials have different thermal expansion - and that causes cracks in the top layers.
Single crystal needed (Score:2)
...if they can deposit a layer of GaAs on top of the sacrificial layer and make circuits out of that, then why do they need the bottom wafer at all? Why not add the sacrificial layers on something less expensive and then deposit the GaAs circuit layer on top of that?
Because the chips need to be made on single-crystal material, which needs to be grown on a single crystal substrate.
This is, by the way, not particularly new in the solar cell research community. Photovoltaics researchers have been developing technologies like this for a long time-- it's called "epitaxial lift-off" or "monolithic metamorphic" in the most recent versions (with "metamorphic" indicating a change in lattice constant), but older variants were called "cleft" and "peeled film technology".
The Wankel Engine of the Semiconductor Industry (Score:5, Interesting)
One of the very first papers I read for a VLSI design course had one of the weirdest final sentences I have ever heard, from a geeky see-my-smarts cross between physics and car geeks. As I recall, it was something like this:
"And then, of course, there is the problem of gallium arsenide, which is the Wankel Engine of the semiconductor industry."
To which the class (a bunch of undergrads wading into the delightful bliss and head-scratching geekery of academic journals for the first time) collectively and perplexedly said "WTF?"
Re: (Score:1)
BadAnalogyGuy must have been a co-author.
Re: (Score:1)
Good guess, but it's due to the sealing issues: https://en.wikipedia.org/wiki/Wankel_engine#Disadvantages
Re: (Score:3)
Typical otto-cycle piston engines are popular be
Re: (Score:3)
Well that's laughably wrong.
First, most electronics are already treated hazardous waste because they often contain lead, mercury, cadmium and other toxic materials. Adding a spec of gallium arsenide sealed in a plasticf IC package that will outlast the human race is not going to make any of it more hazardous.
Second, LEDs. Red, yellow, orange and infrared LEDs use gallium arsenide. Seen any hazmat warning stickers on your TV remote lately? Of course not.
Having gallium arsenide in a chip does cause it to
Re:The Wankel Engine of the Semiconductor Industry (Score:4, Insightful)
Correct. As a compound GaAs is not toxic to deal with. Sodium is nasty, as is Chlorine, but combined together you can eat the stuff.
Re: The Wankel Engine of the Semiconductor Industr (Score:3)
Not to mention pretty much everyone has a GaAs amplifier chip in their cell phone. Also CD and DVD drives use GaAs-based lasers.
Silicon has many distinct advantages over GaAs for logic. To many to go in to here.
Re:The Wankel Engine of the Semiconductor Industry (Score:5, Interesting)
I guess that would be because like Wankel engines, gallium arsenide is a better solution to the problem at hand, but won't ever get a break because the existing design is entrenched...
No, entrenched designs are not what is holding back either GaAs or Wankel engines. Although both are, in theory, elegant solutions, in practice, both have major flaws, and just don't work very well. Both have been subjected to decades of research that have not found solutions to the problems. I don't expect this research to change things much. GaAs will just go from ridiculously impractical, to very impractical. I expect to have a graphene CPU before I have a GaAs CPU.
Re: (Score:3)
readers can find out what a Wankel engine is with a few seconds search and a couple minutes of reading.
It sounds like a great nickname for a motorized Fleshlight.
Re: (Score:2)
Unfortunately, most people (including geeks) still have an alarming lack of curiosity and will be perfectly content to say "WTF?" and never even attempt to discover what an unfamiliar term refers to.
Re: (Score:1)
Boost -> Engine -> Apex Seals
The engine is a good design where the ability to withstand abuse (not racing abuse, but bad gas, bad maintenance [no lifter adjustment, no head gaskets, no camshaft timing, no timing belt], etc) matters. It is also a much lighter design and inherently has low vibration. It's also a cheaper design, if anyone actually wanted enough of them to make it worthwhile. The efficiency, however, is low, and the engine is dirty (burns oil by design). And repairs will come sooner t
What makes it so expensive? (Score:3)
Are gallium, arsenic, or both markedly more difficult to purify enough to serve as reliable semiconductors? Is growing sufficiently flawless crystals large enough to be cut into wafers too error prone to get good yields? Some other unpleasant aspect of processing or handling the material?
Re: (Score:1)
Well, they re-use the same wafer 50-100 times, but I presume the additional processing steps add some additional per-re-use cost.
It's still $5 versus $50, but given that wafer processing itself can cost $5000 to $10000 per wafer, the wafer cost is now insignificant - especially if GaAS processing is cheaper in any way than silicon wafer processing.
Re:What makes it so expensive? (Score:5, Informative)
From what I understand of it (which is very little) it's relatively easy to coax a crucible of pure, molten Si to grow into a single crystal - those long grey sausage-like boules are a single crystal of silicon, so are incredibly pure with a consistent crystalline structure. It's a lot harder to get gallium arsenide to do the same thing.
Re:What makes it so expensive? (Score:4, Informative)
Re: (Score:2)
Well, one look at a silicon PV cell should tell you how much material is used - practically the entire wafer. I mean, the big panels with the blue squares are basically single wafers of silicon. The cheaper ones use cut up wafers which is why they're a lot more irregularly shaped (the wafers are circular for processing, and then sides are lopped off to square them up. Those sides a
Second ones? (Score:3)
No cheaper, just recyclable. (Score:1)
The new manufacturing method won't make the wafer any cheaper, but it does allow it to be reused roughly 50 to 100 times, dramatically reducing the per-chip cost and opening up gallium arsenide for wider use.
unless they are going to start buying back CPUs, this development means very little.
Re:No cheaper, just recyclable. (Score:4, Interesting)
No, they are making a wafer, building chips on top, remove thin top layer to sell the chips, and reuse the bottom part.
Why does the wafer need to be GaAs? (Score:1)
...when it isn't part of the finished product?
Re: (Score:3, Informative)
You need a wafer with the same crystal structure and lattice constant. If there is a mismatch
between the inter-atomic distance (aka lattice parameter or lattice constant), the atoms
deposited on this wafer will try to adjust to this lattice.
If the layer is thin, the deposited crystal will in effect be compressed or expanded. While this is OK from a mechanical and
crystal point of view, the electronic properties of the grown semiconductor will change. E.g. the bandgap (energy distance
between filled and empty e
the presentation is BS (Score:5, Interesting)
The article follows the youtube presentation and the summary is, for once, accurate (i.e. does not introduce new errors).
The trouble is that the presentation is utter BS. The GaAs devices are NEVER made out of a solid GaAs wafer; the process starts with a plain silicon wafer, on which GaAs is grown epitaxially. The secret sauce is, and always has been, how to minimize the defect density at the Si/GaAs interface.
Such a wafer is more expensive than the plain Si one, but not 1000x more! Oh, and every purchaser would kill to get $5 8" wafers...
Since the Stanford guys are no dummies, I guess that the announcement was deliberately made to sound ridiculous. For what purpose? Time will tell.
Re: (Score:3)
I have been designing GaAs MMIC's and RFIC's for 14 years, and none of them were on a silicon wafer. GaAs makes a nearly lossless substrate that makes microwave circuits much better than if they were over a conductive silicon wafer.
lets do the math (Score:1)
Now we need to make a few more assumptions for the rest. Assuming ~50% circuit density and similar cost, the remaining substrate would cost around $50. That's pretty significant, especially considering that many chips will be significantly smaller than a square inch.
What is also significant is the additional weight savings.
The future (Score:3)
Gallium arsenide, the semiconductor of the future, and always will be.
Turns out to be true afterall.
Flexible silicon curls up (Score:3)
I recall an AMD engineer presenting at MICRO in 2012 telling us that one of the problems with making wafers too thin is that they tend to curl up. I'm not sure whether the warping is inherent in the silicon or doesn't occur until after all the circuit layers are put on top. Regarding the article, the wafer doesn't start out thin. The circuits are formed, and then chips are (in a manner of speaking) shaved off the surface, exposing fresh GaAs to make another set of chips.