Slashdot Mirror
Chicken-Feather Chips
gtaylor writes "The Washington Post reports that University of Delaware chemical engineer proposes to replace silicon with chicken feather composites -- since the feathers apparently make the electrons fly. (Unlike turkeys.)"
since the feathers apparently make the electrons fly. (Unlike turkeys.)
I hate to disappoint you, but...Turkeys can fly. In the wild turkeys actually roost in trees.
hen traveling near solids, however, the movement tends to kick up opposing positive charges. These charges can distract the signal from completing its appointed rounds.
So what are they saying? Air offers no electrical resistance? Last I heard, air was one of the best insulators around. Or did they perchance confuse resistance with the dielectric value?
Yes, they confused resistance with dilectric value. The phenomenon described is the the slowing of propogation of signals in a wire surrounded by a material of high dilectric constant.
What puzzles me is the description of this material as a replacement for silicon. The point of the silicon is that it is a suitable material for the fabrication of transistors. The article talks as if the transistors were painted on and the wiring was in the silicon, rather than the other way around. While chips sometimes have a layer of polysilicon wiring for interconnecting slow signals, the bulk of the wiring is successive layers of metal separated by glass above the chip.
Now maybe if they laid layers of this stuff on top of the wafer and built the wiring in it, or etched away the silicon around the active components and filled it with this stuff, it would be useful.
And once the transistors are again discrete components fabricated by nanotech, perhaps something like this might make a suitable microscopic "circuit board". But the techniques to fab nanotransistors in bulk may also provide a way to construct a low-dilectric-constant matrix to contain them and their interconnecting wiring.
Bantam Dominique roosters crow a four-note song. Once you've heard it as "Happy BIRTHday" you can't NOT hear it that way
The reason silicon is used for integrated circuit chips is because it's a semiconductor - a material that can conduct or insulate depending on the electrical conditions around it. Chicken feathers do not semiconduct.
As for electric signals travelling best through air... would you rather be standing ten feet from a power line, or reaching out with a metal fishing rod to touch it?
As far as I can tell, the discussion seems to be a garbled description of using organic fibers/composites as a dielectric (insulating) material instead of oxides or nitrides. Much research has been done over the past several years looking for "low-K dielectrics". The "K" parameter is a measure of how an insulator interacts with an electric field imposed on it. A high-K material has more capacitance when you put a voltage across it; low-k materials for bulk insulators reduce the capacitance between wires (and between wires and the substrate). This reduces wire delays.
An attractive area of research has been to put voids (bubbles or pores) into the dielectric material. Because gases tend to have low dielectric constants, introducing gas-filled voids in the dielectric will reduce the capacitance that two wires insulated by the dielectric will feel. This is what the "microbubbles" comment in the article refers to.
I guess this guy wants to grind up chicken feathers and paste them on to a wafer instead of growing an oxide. Among other things, he'll need to remove all particles larger than a few tens of nanometres for this to not introduce defects in the chip. Good luck.
"WITH many companies struggling to integrate low-k dielectrics, STMicroelectronics researchers claim to have succeeded in reducing the dielectric constant (k) of silicon dioxide, the traditional CMOS dielectric, using air holes. Low k values translate into higher signal propagation speeds. Using air as part of the dielectric is not a new idea, but the difficulty has been producing an interlayer dielectric with sufficient mechanical strength, reliability and uniformity.
"The ST team integrated an SiO[subscript2] air-gap material into a dual-damascene copper process. The dielectric constant was measured at less than 1.7. The researchers say there are no issues with current leakage, electrical resistance or electromigration of the copper. The team is looking for implementation at sub-130nm. Fig.A (above left) is a cross-section of a three-level copper/air-gap architecture with a 640nm line spacing. Fig.B (above right) shows that the deposited oxide at the top of the image is planar to within 60nm above the voids."
Air-gap dielectric
"Dielectric constant is not an easy property to measure or to specify, because it depends not only on the intrinsic properties of the material itself, but also on the test method, the test frequency and the conditioning of samples before and during the test. Dielectric constant tends to shift with temperature.
What is Dielectric Constant
Taking stuff apart since 1969 (TM)