Speaker: Leo Esaki
The Global Reach of Japanese Science
President, The University of Tsukuba
Professor Esaki received Nobel Prize for his work on electron tunneling effect. He spent so many years in this country at IBM laboratory, research institute. Professor Esaki is now President of Tsukuba University, and he will tell us probably about restructuring universities.
Distinguished Guests, Ladies and Gentlemen it is my great pleasure to give a presentation to this prominent audience. My title is the global reach of Japanese science. Of course, science is quite international, so the question is what is Japanese science? Maybe it's Japanese activities, Japanese scientific community or something like that .
First, I would like to talk about some epoch making events together with my personal history. The establishment of quantum physics around 1925-26 . I was born in Osaka, at this time though, there was no relationship. In 1947, was the invention of the transistor, and I received a BS from the University of Tokyo, but the two incidents have no relationship. After receiving my BS from the University of Tokyo, I joined a company called Kobe Kogyo as a researcher. I used to live in Kobe, where the recent earthquake hit very severely. I was very lucky to get out off Kobe a long time ago. Actually, Kobe Kogyo became bankrupt, so I had no good reason to stay in Kobe.
My area of research involved the thermionic electron emission from the cathode in a vacuum tube. Looking back at the decade of the 1950s, these were the years of the evolution from vacuum tubes to transistors. This led to the vast improvement in performance of all electronic products including consumer oriented goods, telecommunications, and data processing . It is no exaggeration to say that today's information oriented society was all made possible by this process of technological innovation. The Japanese electronic industry played an active roll in this evolution; thus, it made a significant contribution to the advancement of semiconductor science and technologies.
I feel extremely fortunate to have started my career in the middle of such a period of technological innovation. This environment stimulated me, encouraged me, and eventually lead me to my thesis work of the Esaki Tunnel Diode in 1957. I then successfully received my Ph.D.. from the University of Tokyo. Afterwards, in 1972, I received the Nobel Prize for this 1957 work; this is to me the good old days of semiconductor science. The good old days means that second rate scientists were able to write first rate papers. The bad old days means even first rate scientists only wrote mediocre papers. This is my definition of the bad old days of science.
This diode is the first quantum-mechanical electron device since electron tunneling was clearly manifested in semiconductors. The tunneling phenomenon was strictly quantum mechanical with no counterpart in Newtonian mechanics. This new diode was highly acclaimed in the United States for its novelty. As a result, I was given the opportunity to go and work in the United States. It was fortunate for me to have found a first rate research environment in a U.S. company, where we were able to pioneer a man- made semiconductor structure in low dimensions including superlattices and quantum wells.
This research was initiated through our proposal. Actually, I made the proposal to ARO (Army Research Office), which advocated to use the advanced techniques of crystal growth and micro fabrication with the view of engineering new semiconductor materials designed by applying the principles of quantum mechanics. In my mind, the root of this development was the tunnel diode, which I made in 1957. We have witnessed a remarkable evolution in this frontier of semiconductor research over the last two decades. Our original proposal and pioneering experiments apparently triggered wide spectrums of experimental and theoretical investigations of this subject.
I would like to discuss a few papers from Japan. The first is the original paper I published in the Physical review.
In appendix is a paper when I joined SONY (changed to this in 1958), at that known as Tokyo Tsuushin Kogyo. This paper was published in 1958. I moved from Kobe to Tokyo to join SONY in 1956 when it was a small company. If you join a small company, one advantage is you are promoted very quickly. I was chief scientist or something like that, with no other scientists around. The young SONY was a very ambitious company as you probably might imagine, as Morita and Ibuka were very active at that time.
Therefore, employees also became very ambitious as a result. I wrote the paper and the first presentation overseas was the one I gave at a conference in Brussels in 1958 on the heavy doped germanium and narrow p-n junction.
I will show the picture of how people changed with time . It includes me and Bill Shockley, a member of this Cosmos Club. I saw his picture in the corridor. And, at the meeting, it was mentioned that Shockley received a Nobel Prize in 1956 and he was the keynote speaker at this International Conference I just mentioned held in Brussels. This coincided with the Expo held in Brussels in 1958. This was the year Europe began to recover from the wounds of the war.
Bill Shockley was kind enough to mention my work in his keynote speech by saying "Esaki from Tokyo is giving a paper with Tunnel Diode." Because of that, I got a large audience at my talk. My English was poor, so nobody could hardly understand my presentation. Therefore, people paid more attention to my talk and the diode.
Recently, in Japanese research, man-made semiconductor structures such as the superlattices and quantum wells, is included in one of the priority areas supported by the Ministry of Education, Science, Sports and Culture. The title of this research is called Quantum Coherent Electronics Physics and Technologies. Quantum coherent electronics as I describe here contains the areas of quantum transport, quantum coherent transport, coherent electron-photon interactions, quantum coherent devices, and quantum process technologies; the organization includes fifty researchers and engineers as advisory members, three of which are participants here.
There are 83 priority areas. Now I would like to derive some cross section view of academic research in Japan through looking at the activities of these 83 priority areas. Grants-in-Aid for scientific research. I just mentioned there are 83 categories for each area, with a budget from half a million yen to six million yen per year and a period between 3 to 6 years . I'd like to focus on this particular category of grants to describe Japanese science. (You know just a Prof. Oda mentioned to the space science but...)
1) Environmental, Earth and Space Science: 8 areas and 9%. of the total. 2) Material Science, as you know, Japan is very strong in Material Science, including Nuclear and Other Materials, Chemical Materials, and New Materials, 28%. 3) Information Science and Electronics, including advanced electronics and quantum coherent electronics 6%. 4) Life Science including bio-science, oncology , neuroscience, about 40% . 5) Humanity and Social Science, 11% .
I would like to give a little more detail on each subject, what I think are important areas. In Environment, Earth and Space, the scientific approach is towards globalism, human society in harmony with the earth and a better understanding of water and energy circulation on a continental scale. It includes satellites, astrophysics with line x-rays and gamma-rays, and sounds. I just mentioned a little detail on these two subjects which were recently recognized as priority areas. The scientific research priority areas I mentioned give better understanding of water and energy circulation on a continental scale based on satellite remote sensing. This study is actually focused on the Monsoonal Asia regions which is the key area for climate system variations. This deals with global climate changes.
Another is basic research on disasters because of the big earthquake that hit the Kobe area. This is very timely Basic Research on the subject of Disaster Mitigation of Megacities to the near field earthquake ground motions.
Within Material Sciences, I would like to mention that there are so many things I am not going to say "nuclei with strangeness". This includes 30-40 scientists which Prof. Imai leads. The world of the nuclei is greatly expanded by the introduction of a new quantum number, known as strangeness.
In Chemical Materials and New Materials, we have two or three subjects. One is the New Polymers and Their Nano-Organized systems , and Molecular Science on the Specific Roles of Metal Ions in Biological Functions . Another is Innovation In Superplasticity . Polymer science is quite active in Japan.
In the Information Science and Electronics field, I will mention the Research and Development of Advanced Database Systems for Integration of Media and User Environments. This is one of the important areas in information science. Inside Advanced Electronics, I have already mentioned quantum coherent electronics. Another is a Single Electron Devices and Their High Density Integration.
Life Science, as I mentioned, includes 40% of all areas. Genome Science: New frontiers in all Biosciences, which is the genome project is stepping up from mapping to sequencing analysis, and in the near future to functional analysis of the genomes. I think, you know, certainly this is one of the fashionable research areas. another , Molecular Mechanisms of Genomic Instability and DNA Repair is again a fashionable subject with Fumio Hanaoka, Osaka University responsible for this project. Furthermore, I would like to mention System Analysis on Higher-Order Brain Functions. Brain research is certainly one of the most important research areas. Jun Tanjji, Professor of Tohoku University is one of the main researchers.
Humanity and Social Science includes 11%, I will mention only one subject. Change of Structures in Contemporary China--Interdisciplinary Studies on Present Aspect and Perspective in the 21st Century, shows that Japanese are certainly keen to know what's going on in China.
This is just a comment on Japanese science. This is a quote from Science & Engineering Indicators, 1993. This figure shows research and expenditure by country, source and performer. In Japan, industry supports nearly 80% with 20% by government.
In 1969, research on artificially structured materials was initiated when Tsu and I [ref. 1,2] proposed an engineered semiconductor superlattice with a one-dimensional periodic potential. In anticipation of advances in controlled epitaxy of ultrathin layers, two types of superlattices were envisioned: doping and compositional, as shown in.
Before arriving at the superlattice concept, we had been examining the feasibility of structural formation of potential barriers and wells that were thin enough to exhibit resonant tunneling [ref. 3]. A resonant tunnel diode [ref. 4,5] appeared to have more spectacular characteristics than the Esaki tunnel diode [ref. 6], the first quantum electron device consisting of only a single tunnel barrier. It was thought that advanced technologies with semiconductors might be ready for demonstration for de Broglie electron waves. Resonant tunneling can be compared to the transmission of an electromagnetic wave through a Fabry-Perot resonator. The equivalent of a Fabry-Perot resonant cavity is formed by the semiconductor potential well sandwiched between the two potential barriers.
The idea of the superlattice occurred to us as a natural extension of double-, triple- and multiple-barrier structures; the superlattice consists of a series of potential wells coupled by resonant tunneling. An important parameter for the observation of quantum effects in the structure is the phase-coherence length, which approximates the electron mean free path. This depends on the bulk quality as well as the interface quality of crystals, and also on the temperatures and values of the effective mass. As schematically illustrated in, if characteristic dimensions such as superlattice periods or well widths are reduced to less than the phase-coherent length, the entire electron system will enter a mesoscopic quantum regime of low dimensionality, on a scale between the macroscopic and the microscopic. Our proposal was to explore quantum effects in the mesoscopic regime.
The introduction of the one-dimensional superlattice potential perturbs the band structure of the host materials, yielding a series of narrow subbands and forbidden gaps which arise from the subdivision of the Brillouin zone into a series of minizones. Thus, the superlattice was expected to exhibit unprecedented electronic properties. At the inception of the superlattice idea, it was recognized that the long, tailor made lattice period provided a unique opportunity to exploit electric-field-induced effects. The electron dynamics in the superlattice direction were analyzed for conduction electrons in a narrow subband of a highly perturbed energy-wavevector relationship. the result led to the prediction of a negative differential resistance at a modestly high electric field, which could be a precursor of Bloch oscillations. The superlattice allows us to enter the regime of electric-field-induced quantization; the formation of Stark ladders [ref. 7,8], for example, can be proved in a (one-dimensional) superlattice [ref. 9], whereas in natural (three-dimensional) crystals the existence and nature of these localized states in a high electric field have been controversial [ref. 10,11].
This was, perhaps, the first proposal which advocated using advanced thin film growth techniques to engineer a new semiconductor material designed by applying the principles of quantum theory. The proposal was made to the US Army Research Office (ARO), a funding agency, in 1969, daringly stating, with little confidence in a successful outcome at the time, "the study of superlattices and observations of quantum mechanical effects on a new physical scale may provide a valuable area of investigation in the field of semiconductors."
Although this proposal was favorably received by ARO, the original version of the paper [ref. 1] was rejected for publication by Physical Review on the referee's unimaginative assertion that it was "too speculative" and involved "no new physics." The shortened version published in IBM Journal of Research and Development [ref. 2] was selected as a Citation Classic by the Institute for Scientific Information (ISI) in July 1987. Our 1969 proposal was cited as one of the most innovative ideas at the ARA 40th Anniversary Symposium in Durham, North Carolina, 1991.
At any rate, with the proposal we launched a program to make the "Gedanken-experiment" a reality. In some circles, the proposal was criticized as close to impossible. One of the objections was that a man-made structure with compositional variations on the order of several nanometers could not be thermodynamically stable because of interdiffusional effects. Fortunately, however, it turned out that interdiffusion was negligible at the temperatures involved.
In 1970, Chang, Tsu and I [ref. 12] studied a GaAsGaAs0.5P0.5 superlattice with a period of 20 nm synthesized by CVD (chemical vapor deposition) by Blakeslee and Aliotta [ref. 13]. Although transport measurements failed to reveal any predicted effect, the specimen probably constituted the first strained-layered superlattice having a relatively large lattice mismatch. Early efforts in our group to obtain epitaxial growth of Ge1-xSix and Cd1-xHgxTe superlattices were soon abandoned because of rather serious technical problems at that time. Instead, we focused our research efforts on compositional GaAs-Ga1-xAlxAs superlattices grown by MBE (molecular beam epitaxy). In 1972, we found a negative resistance in such superlattices [ref. 14], which was interpreted in terms of the superlattice effect.
Following the derivation of the voltage dependence of resonant tunnel currents [ref. 5], Chang, Tsu and I observed current-voltage characteristics with a negative resistance [ref. 15]. Subsequently, Chang and I measured quantum transport properties in a superlattice with a narrow bandwidth, which exhibited an oscillatory behavior [ref. 16]. Tsu et al. performed photocurrent measurements on superlattices subjected to an electric field perpendicular to the plane layers with the use of a semitransparent Schottky contact, which revealed their miniband configurations [ref. 17].
Heteroepitaxy is of great interest for the growth of compositional superlattices. Innovations and improvements in epitaxial techniques such as MBE and MOCVD (metal-organic chemical vapor deposition) have made it possible to prepare high-quality heterostructures with predesigned potential profiles and impurity distributions having dimensional control close to interatomic spacing. This great precision has cleared access to the mesoscopic quantum regime [ref. 18,19].
Since a one-dimensional potential can be introduced along with the growth direction, famous examples in the history of one-dimensional mathematical physics, including the above-mentioned resonant tunneling [ref. 3], Kronig-Penney bands [ref. 20], Tamm surface states [ref. 21], Zener band-to-band tunneling [ref. 22], and Stark ladders including Bloch oscillations [ref. 7,9], all of which had remained textbook exercises, could, for the first time, be practiced in the laboratory. Thus, do-it-yourself quantum mechanics is now possible, since its principles dictate the details of semiconductor structures [ref. 23].
Our original proposal [ref.1] and pioneering experiments have triggered a wide spectrum of experimental and theoretical investigations on superlattices and quantum wells over the last two decades. A variety of engineered structures now exhibit extraordinary transport and optical properties which do not exist in any natural crystals. This new degree of freedom offered in semiconductor research through advanced materials engineering has inspired many ingenious experiments, resulting in observations of not only predicted effects but also totally unknown phenomena. As a measure of the growth of the field, shows the number of papers related to this subject and the percentage of the total presented at the biennial International Conference on the Physics of Semiconductors. Following 1972, when the first paper [ref. 14] was presented, the field went through a short period of incubation before experiencing a phenomenal expansion in the 1980s. It appears that nearly half of all semiconductor physicists in the world are working in this area. Activity at this new frontier of semiconductor physics has in turn given immeasurable stimulus to device physics, provoking new ideas for applications. Thus, a new class of transport and opto-electronic devices has emerged.
In closing, I would like to mention that this progress clearly involves contributions of many people, where an international collaboration plays a significant role for advancement of this field.
Thank you very much for your attention.
Thank you so much Prof. Esaki. Any questions? Name and affiliation, please.References
1] L. Esaki, R. Tsu:IBM Res. Note RC-2418 (1969)
2] L. Esaki, R. Tsu: IBM J. Res. Devel. 14, 61 (1970)
3] D. Bohm: Quantum Theory (Prentice Hall, Englewood Cliffs, NJ 1951), p.283
4] L.V. Iogansen, Zh. Eksp. Teor, Fiz. 45, 207 (1963) [Sov. Phys.-JETP 18, 146 (1964)]
5] R. Tsu, L. Esaki: Appl. Phys. Lett. 22, 562 (1973)
6] L. Esaki: Phys. Rev. 109, 603 (1958)
7] H.M. James: Phys. Rev. 76, 1611 (1949)
8] G.H. Wannier: Elements of Solid State Theory (Cambridge University Press, Cambridge (1959), p.190; Phys. Rev. 117, 432 (1960)
9] W. Shockley: Phys. Rev. Lett. 28, 349 (1972)
10] J. Zak: Phys. Rev. Lett. 20, 1477 (1968); Phys. Rev. B 43, 4519 (1991)
11] A. Rabinovitch, J. Zak: Phys. Rev. B 4, 2358 (1971)
12] L. Esaki, L.L. Chang, R. Tsu: Proc. 12th Int. Conf. Low Temp. Phys., Kyoto, Japan 1970, p. 551
13] A.E. Blakeslee, C.F. Aliotta: IBM J. Res. Devel. 14, 686 (1970)
14] L. Esaki, L.L. Chang, W.E. Howard, V.L. Rideout: Proc. 11th Int. Conf. Phys. Semiconductors, Warsaw, Poland 1972, p. 431
15] L.L. Chang, L. Esaki, R. Tsu: Appl. Phys. Lett. 24, 593 (1974)
16] L. Esaki, L.L. Chang: Phys Rev. Lett. 33, 495 (1974)
17] R. Tsu, L.L. Chang , G.A. Sai - Halasz, L. Esaki: Phys. Rev. Ltt. 34, 1509 (1975)
18] L. Esaki: IEEE J Quantum Electron. QE-22, 1611 (1986)
19] L. Esaki: in Highlights in Condensed Matter Physics and Future Prospects, ed. by L. Esaki (Plenum, New York 1991), p. 55
20] R. de L. Kronig, W.G. Penney: Proc. R. Soc. London A 130, 499 (1931)
21] I. Tamm: Phys. Z. Sowjetunion 1, 733 (1932)
22] C. Zener: Proc. R. Soc. London 145, 523 (1934)
23] L. Esaki: Phys. Scr. T42, 102 (1992)