Electron Devices Laboratory
Tsukuba 305, Japan
Fax: (81) 298-56 6135
Date Visited: 23 July 1997
WTEC: H. Goronkin (report author), D. Cox, J. Mendel, H. Morishita, M.C. Roco, R.W. Siegel
Targeting: Business group R&D, from three to five years
Fundamental R&D Labs, from five to ten years
Staffing and Tsukuba: 300 (devices, materials, fundamental orientation)
Foci of NEC Kawasaki: 1,100 (computer and communication systems, Labs: software, and components)
The Goal of the Advanced Devices Research Lab is 10 nm scale lithography using e-beam. It needs a capability by 2007 for manufacturing 16 Gbit DRAMs. Previous work on short gates includes 100 nm in 1987, 45 nm in 1993 using a 50 keV field emitter beam having less than 5 nm beam diameter at 100 pA; and 40 nm in 1997.
The lab is using Tox = 3.5 nm and is working on 1 nm SiO2. Although tunneling will exist, it is believed that the small area of the gate oxide will limit gate current to a negligible fraction of the channel current.
W/L scaling is not maintained at 2:1.
The lab developed an e-beam resist (Calixarene, molecular weight 1,000) for the 10 nm project in which resolution is limited by the six benzene ring length to 6 nm; 10 nm lines show smooth edges and regular spaces. The resist was licensed to other companies for commercialization.
Starting with a Ne discharge, the neon momentum is decreased and the atoms are trapped in a laser beam. The Ne atoms are allowed to fall under gravitational attraction and they pass through a hologram plate before dropping onto a microchannel plate, where they excite an image. The large mass of Ne compared to electrons provides a large increase in resolution. According to quantum mechanics, the wave of a single Ne atom can pass through thousands of holographic channels simultaneously where it is diffracted and then recombined to form images on the microchannel plate. It is hoped that this can be used for future 1 nm lithography. It currently takes seven hours to form an image. This project began in 1995 in collaboration with the University of Tokyo.
Encouraged by Advanced Device Lab's success with e-beam lithography using Calixarene resist to obtain 10 nm features, various device structures become possible. The Lab's goal is to look for the classical-quantum crossover in MOSFET structures. There are many potential problems, including source-drain tunneling (S-D) and tunneling through the gate insulator. An EJ MOSFET (electrically variable junction MOSFET) was fabricated with gate lengths from 134 nm to 32 nm. A second gate located over the channel control gate was used to modulate the resistance of the source and drain virtual access regions. Satisfactory saturating characteristics were obtained down to 32 nm. At 32 nm, short channel effects were observed; however, no S-D tunneling was observed. At this writing, a 15 nm gate was achieved but results were not reported. It is hoped that when quantum effects are observed, useful novel devices can be made.
The advanced e-beam lithography with PMMA resist was used to fabricate a metallic single-electron transistor using Al/Al2O3/Al island/Al2O3/Al. The process involves opening a 20 nm window between source and drain. The Al is oxidided to form Al2O3 , and Al is deposited in the window to form the island region. Extremely reproducible drain current peaks as a function of gate voltage were obtained. Unlike semiconductor SETs, for example, where the peaks are of uneven height, the lack of depletion and charging of defects provides uniform heights in the metallic system. It is believed that 10 nm islands could provide room temperature operation.
NEC's Electron Devices Laboratory is one of the top labs in the world in nanoelectronics. This lab seems to be able to quickly start new programs in promising areas and to change direction when fundamental obstacles block the path to progress. The Atom Beam Holography and the 10 nm gate projects are at the leading edge. The lab appears to be clear in its understanding of how its research fits into future company needs and how the research must be directed to produce world-leading results.