Site: Toshiba Research and Development Center
1 Komukai, Toshiba-cho
Saiwai-ku, Kawasaki 210, Japan
Tel: (81) 44-549 2318; Fax: (81) 44-520 1287
Date Visited: 24 July 1997
WTEC: H. Goronkin (report author), E. Hu, L. Jelinski, M.C. Roco, D. Shaw, C. Uyehara
Toshiba has a history over 100 years. In 1939, Shibaura Engineering Works and Tokyo Electric Company were merged to a single company named Tokyo Shibaura Electric Company. In 1978, it changed its named to Toshiba Corporation.
Toshiba has a 3-layered R&D organization with long-, medium- and short-term elements: the corporate laboratories work on new technology that may be applied to products 5 to 10 years later; the development laboratories are working on technology for deployment 3 to 5 years later; the engineering departments in the operating divisions have as their most important task the solution of problems inherent to present products.
The following sections summarize the ongoing work in various Toshiba laboratories concerned with nanostructure technologies, based on presentations made to the WTEC panel by our hosts.
"Study of Shapes Produced by Stranski-Krastanov Growth in Cold-Walled UHV-CVD"
With a silicon buffer, hut-shaped Ge dots with (001 x 010) alignment were obtained. Without a buffer, dome-shaped Ge dots were obtained. Using H-terminated Si wafers by exposure of atomic hydrogen, domes were obtained.
The H-free wafers with a prior annealing at 750oC contained the hut dots. Thermal desorption spectroscopy supports the difference in the surface: Dihydride desorbs at 415oC. Monohydride desorbs at 550oC. Annealing at 750oC removes all H. Atomic hydrogen exposure produces a monohydride surface. The goal is study of interaction of dots with 2DEG. The plan is to move to a SiO2-Si system to increase barrier height.
Working on spin electronics for high density heads, 20 Gbit in 2002 is forecast. Toshiba researchers have achieved >10% GMR ratio in layered films at room temperature. They have also achieved > 30% in nanogranular films with a coercive field of 0.1 T.
The most promising approach is the tunnel junction. It has > 25% MR ratio but drawbacks include high resistance and strong fall-off of the magnetoresistance ratio with applied voltage and pinholes in the ultrathin insulator barrier.
Dr. Inomata described two structures for possible use in future memory and logic. One new proposed structure consists of two ferromagnetic (FM) layers sandwiching a barrier containing 8 nm granules of FM material in an SiO2 matrix. The total barrier thickness is about 10 nm. Inomata and coworkers explained that the size and distribution of the FM granules must be carefully controlled. The FM contact polarization can be switched either parallel or antiparallel to the granules and to each other to provide high or low current transport through the barrier. No data were provided.
A second proposed structure places the two FM electrodes on the same surface of granular FM materials. This is a transistor structure in which lateral conduction can be controlled by the relative polarization of the contacts. No data were provided.
"Epitaxy of Ferroelectric (FE) Materials"
The goal of this project is to control FE properties by orientation and strain of the epitaxial material.
The researchers use (Sr, Ba)TiO3 with SrRuO3 electrodes, which have good metallic conduction. By reducing the thickness and sputter conditions, good performance (saturating hysteresis loop) at 1.0 V has been obtained. This material is also under investigation for DRAM charge storage capacitors. With 20% Ba content, the dielectric constant is about 900. An SiO2 equivalent thickness of 0.085 nm was obtained. The leakage current was 4 x 10-8 A/cm2 between ±1.3 V. This was said to be satisfactory for 0.12 m m DRAM. TiAlN/Pt barrier layers were used.
"Nanoparticle Phosphors Made by Thermal Plasma"
This project produces spherical particles compared to faceted particles. These particles have the particle size in the same region as those of commercially available materials. They must be heat treated in oxygen or hydrogen to restore luminescence efficiency comparable to that of the commercial phosphors. The thermal plasma also produces nanoparticle phosphors, which could be used as labeling agents for analysis.