You are here: Home / Seminars / Seminar Database / Beyond “Moore’s Law”--First Steps Toward New Disruptive Technologies

Beyond “Moore’s Law”--First Steps Toward New Disruptive Technologies

Main Content

Lannin Lecture by Ben Shanabrook (Naval Research Lab)
When
11 November 2007 from 4:00 PM to 5:00 PM
Where
117 Osmond Laboratory
Add event to calendar
vCal
iCal

In 1965, Intel co-founder Gordon Moore predicted that the number of transistors on a chip would double every ~2 years. His prediction, now popularly known as Moore's Law, became a reality and has fueled the worldwide technology revolution. However, as the feature sizes in parts of the transistor are approaching atomic dimensions, it is becoming increasing difficult to keep up with the predictions of Moore’s Law. These difficulties include power dissipation requirements, statistical fluctuations in feature sizes, the deleterious effect of quantum mechanics and the increasing complexity/cost of semiconductor manufacturing equipment required to make nano-scaled transistors. In this talk, I will describe the physics of two new device paradigms that may overcome limitations of conventional nano-scaled transistors. In both cases, the spin of the electron is essential for the operation of the device. In the first case, the source and drain of the transistor are replaced by ferromagnetic contacts. In a conventional transistor, changes in the current flowing between the source and drain are determined by changes in an electrostatic barrier. In contrast, in this spin transistor, current between the source and drain is determined by the relative alignment of the spin of the electron to the magnetization axis of the ferromagnetic drain contact. The gate controls the spin precession through Rashba spin-orbit coupling in the 2D channel of InAs and thereby the alignment of the electron at the drain contact. InAs is chosen because of its long spin ballistic mean free path and large g-factor. It is expected that this devices could result in a very high speed, low power consumption switch. In the second case, electrons are confined in InAs quantum dots created by the Stranski-Krastanov growth process. A quantum bit (qubit) is created by a superposition between the spin up and spin down eigenstates of the electron in the quantum dot. The goals of this research are to show that it is possible to place the electron in the quantum dot in a user-defined superposition state and to be able to entangle the spins of the electrons in adjacent quantum dots (perform a two-qubit operation). We envision controlling these qubit operations by a combination of electrical and optical fields. If it is possible to scale the number of interacting qubits to 105 to 106 (i.e. make a quantum computer), it will be possible to solve certain classes of problems that cannot be solved in a reasonable amount of time with classical computers. This long term basic research is being performed in the Nanoscience Institute of the Naval Research Laboratory by a team of staff scientists and National Research Council/NRL post doctoral fellows.

Seminars