Fumihiko Hirose
Mitsubishi Heavy Industries
Network
Latest external collaboration on country level. Dive into details by clicking on the dots.
Publication
Featured researches published by Fumihiko Hirose.
Meeting Abstracts | 2009
Fumihiko Hirose; Yuta Kinoshita; Suguru Shibuya; Yuzuru Narita; Hironobu Miya; Kazuhiro Hirahara; Yasuo Kimura; Michio Niwano
SiO2 ALD with precursors of tris(dimethylamino)silane (TDMAS) and ozone on Si(100) surfaces at room temperature were investigated by infrared absorption spectroscopy with a multiple internal reflection geometry. TDMAS dissociatively adsorbs on OH sites of hydroxylated Si surfaces and ozone irradiation is effective to remove the hydroaminocarbon adsorbates introduced in the course of the TDMAS adsorption. After the ozone treatment, H2O vapor treatments at substrate temperatures around 160 aC allow generation of OH sites for the TDMAS adsorption. The TDMAS adsorption and the ozone treatment at room temperature followed by the H2O treatment at 160 aC enable the cyclic deposition of SiO2. V-I measurements of SiO2 grown by the present 160 aC ALD indicated the deposited film has breakdown electric fields from 3 to 11 MV/cm. C-V measurements indicated that the present ALD is available for MOS capacitors.
Thin Solid Films | 1999
Fumihiko Hirose; Hitoshi Sakamoto; Miyako Terashi; Junko Kuge; Michio Niwano
Abstract The hydrogen adsorption and desorption processes in SiGe films epitaxially grown on Si(100), were investigated using in situ infrared absorption spectroscopy (IRAS) in the multiple internal reflection geometry. We have measured Si-H and Ge-H stretching vibration spectra to examine how the distribution of hydride species (SiH4 and GeH4 changes while exposed to atomic hydrogen at room temperature, and also to follow chemical changes of the H-exposed SiGe surface caused by thermal annealing. IRAS data demonstrate that the SiGe surface that is exposed to hydrogen after annealing at approximately 700 °, is dominantly covered with Ge hydrides. When the surface is annealed up to approximately 400 ° the Ge-H vibration peak vanished completely, and instead, the Si-H vibration peak increased its intensity. We suggest that upon thermal annealing the hydrogen adatoms migrate from Ge sites to Si sites and the Ge atoms on the outermost layer are replaced with the inner Si atoms.
Meeting Abstracts | 2008
Fumihiko Hirose; Yuta Kinoshita; Suguru Shibuya; Hironobu Miya; Kazuhiro Hirahara; Yasuo Kimura; Michio Niwano
We have studied SiO2 ALD processes with precursors of tris(dimethylamino)silane (TDMAS) and ozone on Si(100) surfaces at room temperature by infrared absorption spectroscopy with a multiple internal reflection geometry. It was found that TDMAS dissociatively adsorbs on OH sites of hydroxyrated Si surfaces and ozone irradiation is effective to remove the hydroaminocarbon adsorbates introduced in the course of the TDMAS adsorption. Compensation of OH sites by a water vapor treatment after the ozone process is effective to sustain the cyclic SiO2 deposition.
Thin Solid Films | 2000
Fumihiko Hirose
Abstract The effect of Ge on the P doping has been investigated by the in-situ observation of surface P in the P-doped Si growth using Si 2 H 6 and PH 3 . If the P-doped Si growth is performed without Ge, PH 3 adsorption reaches saturation at the higher P-doping concentrations exceeding 3×10 18 /cm 3 . With Ge on the growing surface, Ge diminishes the self-limiting behavior of the PH 3 adsorption on the Si surface and enhances the PH 3 adsorption. The enhancement of the P doping in the SiGe growth is caused by an increase of surface P coverage due to the enhancement of PH 3 adsorption.
Meeting Abstracts | 2007
Fumihiko Hirose; Naoto Fujiwara; Shigetoshi Ohshima; Hitoshi Sakamoto
Recent demands of higher integration in ultra large scale integrations (ULSIs) have accelerated shrinking of their devices to the size of nanometer scale. A major concern in nanoscale metal oxide semiconductor (MOS) devices is generation of depletion layers in poly Si gate films due to the strong electric field inherent in thin gate oxide MOS devices. Nickel monosilicide (NiSi) has been an attractive alternate since it has a lower resistivity [1,2]. So far NiSi has been formed with Ni deposition followed by a post anneal up to 350 500 . However, the conventional method does not allow the low temperature process below 300 and the concentration control of Ni in the film was difficult. One of the present coauthors, on the other hand, has developed a metal formation process of Cu by metal chloride reduction chemical vapor deposition (MCR-CVD) [3]. In the present work, we have newly developed a unique method of Ni silicidation with a Cl plasma containing NiCl based on MCR-CVD. The present technique enables the low temperature process below 300 and the concentration control of Ni in the grown film. We show a schematic of MCR-CVD system for the Ni silicidation in Fig. 1. A Ni solid source was placed at a certain distance in front of a substrate in a reaction chamber. Pure Cl2 gas diluted with a noble gas was introduced into the reaction chamber and the processed gas was evacuated through a variable conductance valve to keep the pressure at constant pressures below 1x10 Torr. The plasma was generated by radio frequency power with a frequency of 13.56 MHz and powers exceeding 1 kW by an antenna. As a substrate we prepared poly Si/SiO2 overcoated Si wafers. The thicknesses of Si and SiO2 were 150nm and 8nm, respectively. During the process, the substrate temperature was set in the rage of 280-300 . We assume the present process is explained as follows. The Cl plasma generated by the RF antenna reacts with the Ni solid source; Ni + Cl* NiCl(gas). (1) The vaporized NiCl is transported and adsorbed on the Si substrate. Cl atoms in NiCl are extracted by surface Si and SiCl desorbs from the surface, NiCl + Si Ni + SiCl(gas). (2) In the course of the process, the delivered Ni is diffused into the poly Si film. We show a cross-sectional image of the poly Si/SiO2 overcoated Si wafers processed with the Cl plasma containing NiCl in Fig.2. The thickness of NiSi was 130nm, which is smaller than the initial thickness of the poly Si. This is because the Si atoms desorb as SiCl from the film and Ni is concentrated. In our experiment, we have confirmed that the concentration of Ni is possible by extending the process time. We also show the XRD data obtained from the processed wafer in Fig.4. We can see the Ni5Si2 and Ni2Si are generated in the film. We have also confirmed the crystallinity can be changed from NiSi2 to Ni by changing the process time. The resistivity of the film was 57 ·cm, suggesting the present technique is available for the fabrication of metal gate MOSs at near RT. References [1] S. Schmidt, T. Mollenhauer, H.D.B. Gottlob, T. Wahlbrink, J.K. Efavi, L. Ottaviano, S. Cristolveanu, M.C. Lemmne, H. Kurz, Microelectronic Eng. 82(2005)497. [2] J. Kedzierski, D. Boyd, P. Ronshim, S. Zafar, J. Newbury, J. Otto, C. Canral Jr., M. Ieong, W. Haensch, IEDM Tech. Dig. (2003)315. [3] Y.Ohba, H.Sakamoto, Y.Ogura, N.Yahata, T.Nishimori and K.Hatayama, Jpn. J. Appl. Phys. 42 (2003) 6820.
Archive | 2003
Fumihiko Hirose
Archive | 2003
Fumihiko Hirose
233rd ECS Meeting (May 13-17, 2018) | 2018
Yoshiharu Mori; Takahiro Imai; Kensaku Kanomata; Masanori Miura; Bashir Ahmmad; Shigeru Kubota; Fumihiko Hirose
Archive | 2017
文彦 廣瀬; Fumihiko Hirose; 森本 洋史; Hiroshi Morimoto
231st ECS Meeting (May 28 - June 1, 2017) | 2017
Kensaku Kanomata; Makoto Ishikawa; Masanori Miura; Bashir Ahmmad; Shigeru Kubota; Kazuhiro Hirahara; Fumihiko Hirose