Absolute Accuracy Evaluation of Si Single-Electron Transfer Devices

 

Gento Yamahata, Katsuhiko Nishiguchi, and Akira Fujiwara
Physical Science Laboratory

 Single-electron (SE) transfer is expected to be used in ultralow power consumption circuits and current standards in metrology. Accuracy of the SE transfer is crucial for determining bit errors or uncertainty in the current standards. A sufficiently low error rate, typically an error per transferred electron of about 1 part in 108, is necessary for the current standards. Here we demonstrate an absolute accuracy evaluation of the SE transfer [1] in high-speed transfer devices constructed by using Si nanowire metal-oxide-semiconductor field-effect transistors (MOSFETs) [2].
 Figures 1 and 2 show a schematic diagram and a scanning electron microscope image of the device, respectively. The SE transfer device comprises three Si nanowire MOSFETs (FET1, 2, and 3) on a silicon-on-insulator wafer. A high-sensitivity charge sensor is composed of another Si nanowire MOSFET (S-FET). For the absolute evaluation of accuracy of the SE transfer, a node is first formed by turning FET3 off. Then, pulse voltages are applied to FET1, FET2, and the source (S) to shuttle an SE between the source and the node. Finally, we evaluate the transfer error by detecting the number of electrons in the node during the SE shuttle transfer. The merit of this scheme is that the error of the high-speed transfer is evaluated by detecting the average number of electrons in the node with a low-speed sensor. As a first step, we carried out single-shot measurement and thereby separated SE captures into the node and SE ejections from the node to investigate the transfer mechanism.
 Figure 3 shows a typical result of the detection during the SE shuttle transfer at a temperature of 17 K. The abrupt changes of the detection current (IS-D) indicate the capture and ejection of an SE. Here, two capture errors are observed. From longer measurements, the best error rates of the capture and ejection are estimated to be about 2.1 % and 0.25 %, respectively. The former can be explained by thermal fluctuations, indicating the possibility of achieving high-accuracy transfer by reducing temperature or scaling down the device size. The latter can not be explained by thermal fluctuations. From the dependence of the pulse voltage amplitude, we attribute it to transfer mediated by a trap level. Since the trap-mediated transfer shows low error, it could be possible to achieve high-accuracy SE transfer.

[1] G. Yamahata et al., Appl. Phys. Lett. 98 (2011) 222104.
[2] A. Fujiwara et al., Appl. Phys. Lett. 92 (2008) 042102.
 

Fig. 1. Schematic of the SE transfer device
for accuracy evaluation.
Fig. 2. Scanning electron microscope
image of the device.
Fig. 3. Charge detection during SE
shuttle transfer.

[back] [Top] [Next]