Scaling of buried-channel MOSFETs for submicron applications

The evolution of CMOS technology has led to the development of two types of bulk MOS devices; surface channel (SC), where channel conduction begins and remains at the Si/SiO2 interface, and buried-channel (BC) where conduction begins below the interface. It is shown in this work that so-called buried-channel MOSFETs may in fact be either surface or buried-channel in operation. A more complete classification of bulk MOS architectures as surface-channel inversion (SCI), buried-channel accumulation (BCA) and surface-channel accumulation (SCA) devices is used which distinguishes physical behavior.;As CMOS technology scaling has progressed, the industry has increasingly favored SCI devices for both n-channel and p-channel MOSFETs, both for manufacturing and device performance reasons. It is shown here that the use of alternative gate materials allows the design of high-performance, scalable SCA/BCA or accumulation MOSFETs. These devices offer higher gm in both the triode and saturation regions and generally lower capacitances than their SCI counterparts. Proper gate engineering allows the use of complementary accumulation structures in a CMOS technology. Of particular interest is the special case of the device design just at the boundary between SCA and BCA operation (LF-SCA) which provides a transverse surface field of nearly zero at threshold. It is shown that this device promises the best SCE control for the best performance for short-channel digital SCA/BCA applications.;In addition to digital CMOS applications, accumulation devices show promise for analog and RF power technologies. For the first time, LDMOS-like low-field LF-SCA devices are investigated for use in low-power RF power amplifiers. The devices provide excellent power added efficiency of 70% with good high frequency performance and subthreshold behavior. Simulations and measured data of polysilicon gate SCA/BCA designs, and simulation studies of metal-gate designs demonstrate the advantages and disadvantages of these devices for use in both digital and RF power applications.