This paper presents the current state of understanding of the factors that
limit the continued scaling of Si complementary metal-oxide-semiconductor (
CMOS) technology and prov ides an analysis of the ways in which application
-related considerations enter into the determination of these limits. The p
hysical origins of these limits are primarily in the tunneling currents. wh
ich leak through the various barriers in a MOS field-effect transistor (MOS
FET) when it becomes very small. and in the thermally generated subthreshol
d currents. The dependence of these leakages on MOSFET geometry? and struct
ure is discussed along with design criteria for minimizing short-channel ef
fects and other issues related to scaling. Scaling limits due to these leak
age currents arise from application constraints related to power consumptio
n and circuit functionality. We describe her; these constraints work out fo
r some of the most important application classes: dynamic random access mem
ory (DRAM, static random access memory (SRAM), low-power portable det ices,
and moderate and high-performance CMOS logic. As a summary: we provide a t
able of our estimates of the scaling limits for various applications and de
vice types. The end result is that there is no single end point for scaling
. but that instead there are many end points, each optimally adapted to its
particular applications.