Intel, IBM Announce Chip Breakthrough 112
Intel announced a major breakthrough in microprocessor design Friday that will allow it to keep on the curve of Moore's Law a while longer. IBM, working with AMD, rushed out a press release announcing essentially equivalent advances. Both companies said they will be using alloys of hafnium as insulating layers, replacing the silicon dioxide that has been used for more than 40 years. The New York Times story (and coverage from the AP and others) features he-said, she-said commentary from dueling analysts. If there is a consensus, it's that Intel is 6 or more months ahead for the next generation. IBM vigorously disputes this, saying that they and AMD are simply working in a different part of the processor market — concentrating on the high-end server space, as opposed to the portable, low-power end.
printer/ad free version (Score:3, Informative)
Re:Please tag article... (Score:3, Informative)
Re:RFI? Electromigration? (Score:4, Informative)
Re:Not news (Score:3, Informative)
The alternative would have been just to shrink the devices, gain less on performance and use circuit techniques to battle parasitic power consumption. That is what most companies in cost sensitive markets are going to do.
Re:Rename? (Score:2, Informative)
"Most zirconium minerals contain 1 to 3% hafnium. Hafnium is a ductile metal with a brilliant silver lustre. Its properties are influenced considerably by the impurities of zirconium present. Of all the elements, zirconium and hafnium are two of the most difficult to separate. Hafnium is a Group 4 transition element.
Because hafnium has a good absorption cross section for thermal neutrons (almost 600 times that of zirconium), has excellent mechanical properties, and is extremely corrosion resistant, it is used for nuclear reactor control rods.
Hafnium carbide is the most refractory binary composition known, and the nitride is the most refractory metal nitride (m.p. 3310C)."
Intel is not going to be burned by thermal problems again, and you can also hide behind your CPU if "the big one" goes off in the neighborhood. OK, several CPUs and a water tank. But still.
Most efficient.
Last price I could find is $150/pound.
Re:This is a big deal (Score:2, Informative)
> http://hardware.slashdot.org/comments.pl?sid=2191
HfO2/Hf Silicates is mature technology (Obviously, else they wouldn't be in production this year) - however, I disagree with it having been mature for more than 10 years. There were all sorts of compatibility problems with respect to the new layer of "foreign materials" killing the mobility of the electrons responsible for the transistor action in the absence of the kind of relatively perfect interface that Si/SiO2 had. Finding new metals for the right band alignment (different for both the PMOS vs NMOS) was an added absolutely non-trivial challenge. The amount of research activity that this problem has generated is insane. For example, a prominent research review article that was published back in 2001 "High-kappa gate dielectrics: Current status and materials properties considerations" by Wilk et. al, (J. App. Phys. 89, 5243-5275 2001) has been cited 1429 times since then when I checked today and it's still growing...
Ten years ago, scaling down of SiO2 had not really hit the wall, it was coming; so they began this work back then; It's only the last couple of years that frequency scaling has not been going upwards... The real fear was if high-k technology would miss the "45 nm technology node" - in which case it might have had to wait till the 38 nm or whatever node that came next. It was thus a question of timing, and frankly it is impressive that Intel/IBM has managed to converge upon a set of solutions which have overcome all the new problems** that the new manufacturing technology (they use ALD, atomic layer deposition) and new materials and their interfaces bring about. Hopefully these chips won't start exhibiting flaky behavior when the overclockers get their hands on these chips....
** Some of the tough problems they had to solve include (sorry for the karma whoring, check the wikipedia high-k article for the links http://en.wikipedia.org/wiki/High-k_Dielectric [wikipedia.org])
* Permittivity
* Band gap
* Band alignment to silicon - sufficiently large band offsets are needed to keep the leakage current low and protect the film from hot carrier injection.
* Thermodynamic stability
* Minimization of electric fields due to phonons in the dielectric to reduce scattering in the Si substrate so as to achieve high mobility of charge carriers in the MOSFET channel
* Minimization of the concentration of electrically charged and/or electrically active defects in the film
* Film morphology - Amorphous or epitaxial films seem to be the promising candidates - polycrystalline materials are generally ruled out.
* Interface quality
* Compatibility with the current or expected materials to be used in processing for CMOS devices
* Process compatibility - for one, the film must survive sufficiently high temperatures such as a Rapid thermal anneal to 1000 C for say, 10 s (as dictated by the CMOS technological process)
* Reliability
* Stability against degradation by the electric field and injected carriers.
* Precursor availability
* Precursor and process costs