Effect of Pocket Doping and Annealing Schemes on the Source-Pocket Tunnel Field-Effect Transistor

Abstract

Low operating power is an important concern for sub-45-nm CMOS integrated circuits. Scaling of devices to below 45 nm leads to an increase in active power dissipation (CV <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup> .f) and subthreshold power (I <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">OFF</sub> .V <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">DD</sub> )Hence, new device innovations are being explored to address these problems. In this paper, we simulate and experimentally investigate the source-pocket tunnel field-effect transistor (TFET), which is based on the principle of band-to-band tunneling, p-i-n and source-pocket TFETs are fabricated with different pocket conditions to observe the effect of the source-side pocket on device performance. Different annealing schemes (spike and conventional rapid thermal annealing) are used to study the effect of annealing conditions on TFET performance. The source-pocket TFET shows a higher I <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">ON</sub> (~10 times) and steeper subthreshold swing as compared to a p-i-n TFET. The ambipolar conduction is also reduced by using a low-doped drain extension. Low-temperature measurements of the source-pocket TFET were performed, and the subthreshold swing of the source-pocket TFET shows very little temperature dependence, which confirms the dominant source injection mechanism to be band-to-band tunneling.