We have used a lot of electronics products, which support our life. To
get more comfortable, useful, and friendly life, performance of electronics
products has been improved by a continuous development of various kinds
of technology. On the other hand, such development becomes more and more
difficult. One of the most familiar, used, and powerful devices in electrical
circuits is a transistor. Its performance has been increased by shrinkage
of the transistor. However, since such shrinkage gives rise to various
kinds of problems, e.g, the uncontrollable operation and high power consumption
of the transistors, a lot of efforts have been focused on the need to achieve
further improvement of the transistor, and such a trend will continue.
Our research is a development of a new type of applications using new devices
and circuits, whose mechanisms are quite different from present transistors
and electrical circuits. In order to achieve it, we focus on nanometer-scale
devices constructed by various kinds of fabrication processes of silicon
transistors.
One approach to a new application is to use one electron as one
bit of information in the circuit. To achieve it, we must control single
electrons precisely. Since conventional devices cannot control single electrons,
we use single-electron devices. Among them, a one-by-one electron transfer
device composed of transistors is useful due to its geometrical simplicity
(see another page). On the other hands, single electrons must be detected to use them as
data. The single-electron detection can be achieved by a high-charge-sensitive
sensor using nanometer-scale transistor (see another page). The combination of the one-by-one electron transfer and sensing devices
provides some applications, e.g., a multilevel memory (Electron. Lett., v. 40, p. 229, 2004) and a digital-analog converter(Appl. Phys. Lett., v. 88, p. 183101, 2006), in which one bit of information is represented by one electron.
Another application is a circuit with high flexibility. Conventional
circuits have powerful performance to give a precise answer/calculation
according to particular algorithms. However, when a question is complicated,
it takes long time to give the answer. The case that the algorithm is not
optimized to the question also provides the conventional circuit with hard
work. On the other hand, although a human brain may lack the precision
to give the answer, flexibility of the human brain allows time- and power-efficient
performance. Therefore, addition of flexibility to the circuits promises
for high-efficiency electrical circuits. In our research, the flexibility
is represented by random movement of single electrons flowing through a
transistor (Nanotechnology, v. 20, p. 175201, 2009).
