October 6, 2014
(Press release material)
Nippon Telegraph and Telephone Corp. (NTT) (Head Office, Chiyoda-ku, Tokyo; Hiroo Unoura, President and CEO) has succeeded for the first time in speeding up one-by-one transfer of electrons via a charge-confining trap level*1 in silicon transistors*2.
Since a flow of electrons accurately transferred in one direction leads to a high-accuracy current flow, the technique is expected to lead to high-accuracy current sources (current standards*3), which will contribute to the recently proposed redefinition of the ampere (the base unit of electric current). In addition, if we realize a device with much higher accuracy than traditional electrical standards, we can contribute to the electrical standards field and the measurement instrument industry.
This work will be reported in the UK science journal “Nature Communications” on the 6th of October, 2014.
This work was partly supported by the Funding Program for Next Generation World-Leading Researchers of the Japan Society for the Promotion of Science (GR 103).
NTT Basic Research Laboratories have been developing single-electron devices that can manipulate and detect an individual electron, which is the smallest constituent of electric current. Such devices have attracted much interest for ultralow-power-consumption information processing and ultrahigh-sensitivity sensors because they can operate with a small number of charges and have a high charge sensitivity. We have so far demonstrated the operation of single-electron transfer devices and single-electron detectors using stable and reproducible silicon transistors.
The redefinition of the ampere, the base unit of electric current defined in the International System of Units (SI)*4, was proposed in 2011 together with the abolition of the international prototype kilogram*5, and it has been attracting much attention. In the new SI units, the value of the elementary charge e, which has so far been determined from measurements, is fixed and the ampere is set from the fixed value. It is therefore desirable to realize the current standard using single-electron transfer because it directly connects e to the ampere (Figure 1 ).
Most single-electron transfer devices control the transfer by capturing electrons in a single-electron island*6, an artificially fabricated fine region. To increase the transfer accuracy, we have to scale down the single-electron island to increase the energy (electron addition energy*7) required to capture an electron (Figure 2 ). However, the limitation on the downscaling of semiconductor devices has been an obstacle to increasing the transfer accuracy.
We have been studying single-electron manipulation to apply it to the electrical standards field, where high accuracy and reliability are required.
We have succeeded for the first time in high-speed single-electron transfer via a charge trap level in silicon transistors (Figures 3 , 4 ) at frequencies of up to 3.5 gigahertz and a measurement temperature of 17 kelvin*8 (Figure 5 ). In the high-speed operation, the error rate*9 is below the level that can be measured using a current meter (~ 10-3). The error rate obtained at 3.5 gigahertz is much lower than that of other single-electron transfer devices operating at a similar speed. In addition, from a theoretical estimation, the error rate is likely to be below 10-8 (the target for the current standards) at 10 to 20 K and 1 GHz, indicating the possibility of high-accuracy device operation.
Owing to our wafer-level fabrication techniques for nanometer-scale silicon transistors having a double-layer gate structure, many devices can be fabricated with high yield. As a result, optimal devices can be efficiently selected. In addition, our knowledge of the single-electron manipulation, which has been acquired from many studies of silicon single-electron devices, has enabled us to clarify the operating mechanism of the transfer via the trap level and thereby to achieve the high-speed operation.
Instead of the single-electron island, using a naturally existing trap level having an extremely fine confinement area of less than 10 nm, which is difficult to artificially fabricate, can lead to high-accuracy operation owing to its large electron addition energy (Figure 2 ). We investigated the electron capture/emission to/from the trap level in detail to clarify the mechanism of the transfer and found that the capture and emission speeds can be widely modulated by controlling applied voltages. On the basis of the findings, we achieved the high-speed operation by increasing speed of the capture and emission. In addition, by investigating temperature dependence of the transfer current, we theoretically anticipated the transfer error rates.
We aim to experimentally prove that the single-electron transfer has the high accuracy necessary for the current standards. Furthermore, we also aim to contribute to the electrical standards field and the measurement instrument industry by developing devices with accuracy that exceeds traditional electrical standards. In addition, since the findings in this work are applicable not only to trap levels but also to other localized states, we expect to be able to improve device yields by using intentionally introducible impurity levels. By using such a technique, high-yield current standards are expected to be accomplished. Furthermore, in the long term, we aim to achieve ultralow-power-consumption devices based on single-electron manipulation and thereby contribute to the realization of a low-energy society.
G. Yamahata, K. Nishiguchi, A. Fujiwara
“Gigahertz single-trap electron pumps in Si”
Nature Communications (2014).
Nippon Telegraph and Telephone Corporation
Science and Core Technology Laboratory Group, Public Relations
NTT Has Instituted a Logo to Represent R&D Activities.
Information is current as of the date of issue of the individual press release.
Please be advised that information may be outdated after that point.