Theme I: Nanoelectronics

tunneling FET

Faculty:

  • E. Yablonovitch (lead)
  • D. Antoniadis
  • J. del Alamo
  • F. Fischer
  • J. Hoyt
  • A. Javey
  • T. Swager

The goal of the Nanoelectronics research theme is to develop a solid-state switch that can be actuated in the milli-Volt range and thus, replace the conventional transistor switch. Considerable research efforts on alternate switching mechanisms have converged around the tunnel transistor, reasoning that Moore’s Law will lead to devices so small as to require the consideration of tunneling in any case. However, state-of-the-art device results of tunnel transistors have continued to be disappointing.

Two main tunnel modulation mechanisms are in play: tunnel-distance-modulation versus density-of-states modulation, which is also called energy filtering. The latter mechanism is considered to have the most potential to meet all device specifications. Emphasizing the Center’s goal to elucidate the underlying device physics, as opposed to simple device optimization, Theme I researchers found that the preferred density-of-states mechanism demands higher interface perfection than ever previously required in solid-state electronics.

Challenges

Tunneling is an interfacial process, limited by the two-dimensional density of quantum states, which is ~1012/(cm2 eV). This desirable tunneling needs to compete with bandgap defect state density, which is a famous figure-of-merit in electronics science. In the most favorable materials systems currently used the interfacial defect density is ~1010 /(cm2 eV). Therefore, even after decades of electronic material investigations, the best interface state-density materials are far from being good enough. The Center for E3S has thus embarked on a search for new material systems with interface property, beyond what decades of research have previously accomplished.

Current Projects
Two-dimensional Chalcogenide Semiconductor Materials
 
Bottom-up Fabrication of Novel Semiconductors
 
III-V Nanowire Tunnel Field Effect Transistor
 
[Back to top]

Theme II: Nanomechanics

nanomechanics image

Faculty:

  • T.J. King Liu (lead)
  • V. Bulović
  • J. Lang
  • V. Stojanović
  • T.M. Swager
  • J. Wu
  • D. Zubia

The goal of the Nanomechanics team is to demonstrate low-voltage switching with nano-electro-mechanical (NEM) relays as ultra-low energy alternatives to the current-day transistor. A further major aim is to demonstrate feasibility of this approach in a system application. Typically, mechanical switches conduct current when two plates are in contact, and turn off current when the plates are separated. Since mechanical switches inherently have zero off-state current, they are promising solutions for the off-state leakage issue. Realizing that surface adhesion ultimately limits relay scaling, the Center has also focused on new approaches that go beyond voltage reduction through scaling and new device design. Instead, Theme II researchers pursue the concept of a tunneling relay whereby the electrical activation will occur when the two electrodes are brought into close proximity, but do not touch each other. The spacing of the electrodes can be controlled by spring-restoring forces or by folding molecular chains. The latter approach constitutes a molecular switch, or “Squitch”. Successful operation of a squitch device has recently been demonstrated by E3S researchers.

Challenges

Current and future research within Theme II aims to resolve challenges in device design and scaling, reliability, and materials processing. To achieve the ultimate goal of demonstrating reliable milli-Volt switching operation, researchers at E3S develop new fabrication processes to ensure sufficiently smooth surfaces inside the device. A major challenge is also to reduce the area of tunnel contact, while increasing the area of the electrostatic actuator. Furthermore, finding molecular building blocks with costume-designed structures and mechanical properties will be critical for optimized Squitch operation.

Current Projects
Body-Biased Switch Design
A Molecular Switch: the “Squitch”
 
Strain-induced Bandgap Modulation in Two-Dimensional Materials
 
NEM-Relay Based VLSI Circuits: Toward an Internet-of-Things Application
 
[Back to top]

Theme III: Nanophotonics

nanophotonics image

Faculty:

  • M.C. Wu (lead)
  • C. Chang-Hasnain
  • E. Fitzgerald
  • E. Yablonovitch

The goal of the Nanophotonics team is to enable optical communications between switches on a chip at unprecedented efficiency levels. In fact, Theme III researchers pursue the ultimate goal of experimentally reaching the quantum limit of photons-per-bit in a data-link. This will require reduction of photons per bit from currently used 20,000 to 20 photons. To meet this goal, the Center for E3S strives to improve energy efficiency and sensitivity of both the emitter and the photo-receiver. Central to the E3S nanophotonics research goals has been the demonstration that spontaneous emission from antenna enhanced nano-LEDs can be faster and more energy efficient than the stimulated emission of lasers, the ubiquitous light source in optical communications today. Major advances toward this goal have recently been achieved by demonstrating spontaneous emission enhancements of more 300 times, reaffirming Theme III’s strategy of introducing optical antenna enhanced spontaneous light emitters for energy-efficient short distance on-chip optical interconnects.

Challenges

Reaching the quantum limit in terms of photons-per-bit in an optical communications data-link imposes tremendous challenges on choice of materials, nanofabrication of optical components, and their on-chip integration. Ultra-efficient light sources and ultra-sensitive detectors need to be developed and miniaturized to be comparable to the size of transistors. Integration of the optical components in waveguides is also part of the challenges of nanophotonics research.

Current Projects
Antenna-coupled III-V nanoLED
 
Antenna-enhanced chalcogenide nanoLED
 
Ultra-sensitive photo-receivers
 
[Back to top]

Theme IV: Nanomagnetics

nanomagnetics image

Faculty:

  • J. Bokor (lead)
  • S. Khizroev
  • S. Salahuddin
  • V. Stojanović
  • H.-S.P. Wong

The goal of the Nanomagnetics team is to use current-driven magnetic elements for electrical communication and switching at sub-femtojoule energies, and with fast switching speeds as low as <10 picoseconds. The primary approach is to take advantage of newly discovered ultra-sensitive, current-driven switches employing actuated spin-orbit torque (Spin Hall Effect) to switch a magnet. Such a component can have current in/current out gain, as well as fanout. Since the constituents tend to be metallic, the voltage requirement is low, compatible with the goal of low dynamic power as the digital circuits switch. At the same time, these magnetic switches can operate at ultra-high speed, employing a-thermal, non-equilibrium, magnetic phase transitions. This opens the possibility to generate ultrafast switches for memory and logic gated by a single stimulus, without the need of polarization modulation. In addition, a key detractor with all currently known magnetic materials is the low on/off ratio of present magneto-resistors—a drawback for which solutions are currently explored in the Center from a circuit design and architectural perspective.

Challenges

Understanding and exploiting the fundamental physics and dynamics of the spintronic phenomena that underlie magnetic switching has been a key focus of Theme V research. In addition, development of materials platforms and device architectures that enable electrically-induced magnetic switching at high speeds remains a grand challenge.

Current Projects
Magnetic Switches Based on Spin-Orbit Torque Phenomenon
 
Spin-Transfer Torque Magnetic Tunnel Junctions
 
Circuit Architectures with Nanomagnetic Switches
 
[Back to top]