1977 - 2003

1977 - 1979 Electron correlation effects on Anderson localization (Theory)

From 1977 to 1979, we were engaged in theoretically studying the interplay of randomness and electron correlation in the Anderson localized states in doped semiconductors.


As the donor concentration ND is increased in donor-doped semiconductors, the system becomes metallic due to the Mott-Anderson transition at a certain critical donor concentration Nc. In the intermediate concentration region ND < Nc, where the electrons are involved in the Anderson localization, the spin susceptibility and electronic specific heat was observed to exhibit anomalous phenomena.


In order to explain the anomaly, we took into account the effects of electron correlation on the Anderson localized states. We then proposed a new type of Hamiltonian for this system and solved the statistical mechanics based on the Hamiltonian. As a result, this theoretical model explained the anomaly on the spin susceptibility and electronic specific heat well. (Click here to see related papers) (Click here to see related conferences)

(PageTOP)

1979 - 1984 Low dimensional electron systems in III-V semiconductors (Experiment)

From 1979 to 1984, we were engaged in experimentally studying III-V semiconductor MISFETs (Metal-Insulator-Semiconductor Field Effect Transistors) .


III-V semiconductor MISFETs were expected to have higher mobility than Si MISFET's due to lighter effective mass. However, no insulator suitable for III-V semiconductor MISFET's which exhibited the interface inducing almost no interface trap states had ever been found so far. We thus tried to find out the insulator which gave minimized interface trap states for InAs and InP. We succeeded in growing PAsxNy and BN insulator on InAs and InP by CVD method for the first time. We also developed a new method for measuring the energy distribution of interface trap states on the basis of ICTS method.


Experimental studies for five years finally allowed us to make InAs and InP MISFETs on which two dimensional transport properties could be clearly observed. We analyzed the subband structures of two dimensional electrons on InAs and InP MISFET's, and, for the first time, proved that anomalous characteristics on mobility found so far was due to the inter-subband scattering effects. (Click here to see related papers) (Click here to see related conferences)

(PageTOP)

1984 - 1991 Electronic structures of III-V semiconductors (Theory)

From 1985 to 1990, we were engaged in theoretically studying the electronic structures for the interface, superlattices and deep levels in III-V semiconductors.


It had experimentally been shown that each III-V semiconductor exhibits its own specific characteristics for the energy distribution of the interface trap states. For instance, the interface between GaAs and any insulator always induces the empty interface states above the Fermi level, which make it impossible to make GaAs MISFETs. What is the origin of such specific characteristics for the interface states?


To clarify the origin, we developed the theory for obtaining the electronic structures of the system consisting of the junction of two different materials, based on the Green's function technique and the superlattice technique. As a result, we found that the specific characteristics of the interface states in III-V semiconductors are produced by the fact that the anti-bonding states or bonding states at the junction of two different materials destroyed the band structures near the junction.  We then, for the first time, found that the characteristic energy distribution of the interface states for III-V semiconductors can be controlled by inserting a certain monolayer of III-V between the two different materials.


We used the theory for obtaining the band structure of thin film superlattices as a function of film thickness. As a result, we, for the first time, explained the experimental results for the G-X crossover phenomena for the GaAs-AlAs superlattices.


We applied the Green's function theory to deep levels, especially deep donor levels, in III-V semiconductors. It had been found since 1960's that donor impurities produce deep levels in AlxGa1-xAs, being named DX centers. The deep donor levels had then been thought to be due to large lattice relaxations, but its microscopic origin had been unknown. As a result of the present theoretical investigation, we found that the anti-bonding electronic states for the junction of host AlxGa1-xAs and donors could produce deep levels. This result suggested that the origin of the DX centers might not be large lattice relaxations but be the anti-bonding states.


Later, an American group claimed that, based on the first principle calculation, a certain configuration of large lattice relaxation exhibited the energetically most stable states, producing the DX centers as doubly occupied electronic states. In order to confirm if their calculation was correct or not, we performed the first principle calculation with more precision, and found that the configuration of small lattice relaxation exhibited the most stable states and that the configuration as claimed by the American group never exhibited the most stable states. (Click here to see related papers) (Click here to see related conferences)

(PageTOP)

1991 - 1998 Optical properties of III-V semiconductors (Experiment & Theory)

Later, the deep donor levels due to the anti-bonding states at the junction of donor atoms and host atoms, predicted by us, was experimentally found by European group. They reported that this deep donor level was different from the DX center. From 1991 to 1996, in order to investigate the relationship between this anti-bonding states and the DX center, we were engaged in experimentally studying the optical properties of deep donor levels in III-V semiconductors.


If our theory is correct, band-edge photoluminescence should be observed by the below-bandgap photoexcitations in AlxGa1-xAs. Namely, thermal barrier should exist between the conduction band and the deep donor levels. To confirm this hypothesis, we performed the experiment of this anti-Stokes photoluminescence effect for Si doped AlxGa1-xAs. As a result, we succeeded in observing the band-edge photoluminescence (2.0 eV) by below-gap photoexcitation (1.8 eV) . The experimental results for the excitation spectra, the power and temperature dependence of the luminescence amplitude justified my theory well.


From 1996 to 1998, we theoretically investigated the electronic structures and deep level properties of nitride semiconductors. The Green's function calculation revealed that the deep donor level, which appears in the bad-gap in AlxGa1-xAs, never appears in the band-gap. Therefore, it is predicted that the insulator AlN can be used as electronic barrier layer of GaN. It suggests that GaN/AlN heterostructures are promising materials for electronic devices of the next generation.


Furthermore, we calculated the deep levels produced by nitrogen vacancy in nitride semiconductors, using the Green's function technique. An anti-bonding s-like state produced by nitrogen vacancy is predicted to appear at 0.3 eV below the conduction-band (CB) edge for GaN, which gets shallower and then resonant with the CB for InxGa1-xN with In content (x) . The theoretical results can provide a new and consistent model for explaining anomalous optical properties in nitride semiconductors. To confirm this prediction, we also performed the first principle calculation by using a 64-super-cell system, and obtained the results that the deep level produced by nitrogen vacancy is already resonant with the conduction band in GaN. (Click here to see related papers) (Click here to see related conferences)

(PageTOP)

1989 - 1998 Anormalous nuclear effects in deuterated solids (Experiment)

In March 1989, two chemists in the United States (Martin Fleischmann, Professor of Southampton Univ.,UK, and Stanley Pons, Professor of Univ. of Utah) reported that anomalous amount of excess enthalpy was produced by the electrolysis of heavy water with Pd as cathode, and claimed the possibility of "cold fusion" of deuterons in solids. However, the observation of nuclear product is really necessary to prove the occurrence of nuclear reactions. We invented a new method of observing this anomalous phenomenon in vacuo, and performed the experiment from 1989 to 1998 to confirm the possibility of anomalous excess enthalpy effect in deuterated solids.


As a result of this new type of experiment, we succeeded in observing the production of He-4 at the same time of anomalous heat effects. We found the experimental conditions to obtain 100 % reproducibility for the excess enthalpy. (Click here to see related papers) (Click here to see related conferences)

(PageTOP)

1998 - 2003 Science & technology public policy (Social science)

In 1998, we started the public policy study for R & D (research and development) at the 21st Century Public Policy Institute of Keidanren (Japan Federation of Economy) . We investigated the historical background of R & D in Japan and necessary condition for R & D policy of the Japanese government.


We surveyed the historical path of Japan's innovation by holding interviews with a number of brilliant scientists / engineers. As a result, we found the specific limitations of Japan's R & D methodology for the innovations, and proposed a new methodology to enhance the innovation.


We further collected the data of various "states" constituting the Japan's society as a function of time and space. Here, the elements of "states" are varied from the populations with various attributes (sex, age, labor status, daily transport, etc.) , corporations with various categories (working field, scale, benefit status, etc.) , land prices, budgets of local governments, medical/care status, election status, criminal status, etc. The unit cell for space is the third mesh or the city/town/village. Based on the constructed database, we clarified the state structures of Japan by making the multi-dimensional topography. (Click here to see related papers) (Click here to see related conferences)

(PageTOP)

2003 - 2013

2003 - 2013 Innovation and technology management (Multidisciplinary)

Investigating an error embedded in the discussion of 'disruptive innovation' by Clayton Christensen (which we call 'performance disruptive innovation' hereafter) , we discovered a new type of innovation, 'paradigm disruptive innovation'. This paradigm disruptive innovation is found to be successful only when the team which co-owns the tacit knowledge, while, on the other hand, the performance disruptive innovation is successful only when the team overcomes the value network trap.

Applying this theory to transistors (bipolar transistors, MOSFETs, HEMTs) and blue LEDs, we analyzed the mechanisms for the reason why small startups can only be innovators for creating new industries. From this analysis, we clarified the detailed pictures of the 'fields of resonance' model. It is found to be essential for triggering chain reactions of innovation processes that there are a number of fields of resonance, fields in which people embodying the knowledge can acquire the tacit knowledge from people creating the knowledge, beyond the framework of companies.

To investigate this 'fields of resonance' model in more detail, we performed three approaches: (1) economical physics approach for analyzing macroscopic structures and dynamics of 'fields of resonance' network beyond intra-companies, (2) sociological approach for studying social characteristics of organizational scientists / engineers, i3j cultural-anthropological approach for analyzing existential desire of organizational scientists / engineers who have been key persons of innovation processes in fact.

To perform the first approach (economical physics) , we focused on patents as a key quantity representing innovation processes, and build up the patent network database which include more than 100,000 Japanese companies. We found as follows: (1) The network has stratified structures. (2) The nearness between companies gives rise to a positive effect on collaborative R&D. (3) Typical network growth model called Preference Attachment can reproduce this patent network, but cannot reproduce geological structures. The patent network can be reproduced just by adding a condition that the nearest node can be selected as an end point of the link.

To perform the second approach (sociology) , we focused on the mobility of Japan's scientists / engineers from a view point of social structures. Specifically, we analyzed the transient process of privatizing top-class national research institutes. Furthermore, we performed various surveys for Kyoto high-tech companies and extracted the characteristics of these Kyoto companies and discovered their fine structures.

To perform the third approach (cultural anthropology) , we performed long-term sustained field survey for key engineers and technicians of science-based venture companies in Taiwan-Hsinchu, and found essential aspects of Taiwan Innovation System. We also performed interview survey for key scientists / engineers in Silicon Valley. Furthermore, as a typical example of paradigm disruptive innovation, we targeted the innovation process of CCD and performed detailed interview to key persons of invention, development and marketing of CCD. (Click here to see related papers) (Click here to see related conferences)

(PageTOP)

2003 - 2013 Development of nitride semiconductor devices and its industrialization (Multidisciplinary)

In 2002, we founded Powdec K.K., and started to make a large scale GaN HEMT wafers as a commercial base. We measured Shubnikov de Haas effect at low temperatures under strong magnetic fields and found the Powdec's HEMT had an excellent 2DEG properties, where electron mobility was 1400cm2/Vs at room temperature and more than 10000cm2/Vs at 4.2 K.

In 2005, we founded ALGAN K.K., and started to develop ultra-violet (UV) photodetector on GaN, InN, AlN, and their alloys. We have succeeded in sensing vacuum ultra-violet (VUV) light and in making a sensor for UV with specific wavelength. Furthermore, we are also successful in the developping sensors of charged particle with high energies. (Click here to see related papers) (Click here to see related conferences)

(PageTOP)