Pages

July 17, 2007

Silicon spintronics closer for cheap continuation of Moore's law

From EEtimes, Naval Research Laboratory (NRL) scientists will next month describe a technique that would allow spintronics to be inserted into the standard silicon CMOS processes using ferromagnetic materials similar to those already used for magnetic random access memory.

Our demonstration showed a 30 percent polarization of the injected electrons, which is not bad considering that polarizaiton of electrons in magnetic metals is about 45 percent," said lead scientist Berend Jonker. "Now we want to build an electronic detector, rather than use an LED, as the next step toward silicon spintronics."

The NRL scientists claim to have injected electrons through a ferromagnetic film over a high-k aluminum oxide dielectric. Specifically, they formed an iron/aluminum oxide tunnel barrier contact over a light-emitting diode built from layers of n-doped silicon over a silicon-oxide insulator over p-doped silicon.

Despite the weak electroluminescence of a silicon LED, the resulting circularly polarized light confirmed the injection of spin-polarized electrons.

OTHER READING from EETimes.com:
Intermolecular Inc. (San Jose, Calif.) says its High-Productivity Combinatorial (HPC) platform of "fab in a lab" technologies will facilitate R&D of IC materials, processes and device structures

Intermolecular is addressing what he called the "perfect storm" in semiconductor R&D. "R&D spending is running out of control," he said. "Our mission is to make R&D more productive." While it remains to be seen whether Intermolecular's strategy will work, there is a crying need in the chip industry for feasible new R&D models. In 1978, total semiconductor R&D was $600 million, according to IC Insights. Intermolecular claims that figure had grown to $45 billion by 2006 and is expected to hit a whopping $100 billion in 2012.

The F30 fab-in-a-lab tool provides cleaning, electroless deposition, self-assembly and other functions in the same unit. The modular tool taps 28 separate screening models.

Intermolecular insists the tool will not replace a production machine but says it can be used by chip makers to develop a process "10 to 100 times faster" than conventional methods.

For example, to develop its "molecular masking layer" IP technology, Intermolecular ran 7,635 experiments with 60 base molecular types in the F30. In a short time, the tool discovered two "hits," or matches, according to the company.


In a move that will raise some eyebrows — and could upset others — International Sematech Monday (July 16) launched a new and dedicated 450-mm fab R&D program. Sematech will now have two separate and parallel 450-mm efforts. The new program is bound to upset most chip-equipment companies, which lack the R&D dollars to move to the next-generation and costly wafer size. 450-mm fabs are not expected to emerge until 2012 or later. Only a few companies can afford to build these fabs, namely Intel, Samsung, and perhaps TSMC.

0 comments: