The vast majority of semiconductor integrated circuits in use today rely on silicon
Metal-Oxide-Semiconductor Field Effect Transistors (MOSFETs), whose electronic advantages and
operational device physics are summarized in [37,40]. Given the extreme usefulness
and success of MOSFET-based electronics in VLSI silicon, it is naturally desirable to
implement high-performance inversion channel MOSFETs in SiC. Like silicon, SiC forms a thermal
oxide when it is sufficiently heated in an oxygen environment. While this enables SiC MOS
technology to somewhat follow the highly successful path of silicon MOS technology, there are
nevertheless important differences in insulator quality and device processing that are
presently preventing SiC MOSFETs from realizing their full potential. While the following
discourse attempts to highlight key issues facing SiC MOSFET development, more detailed
insights can be found in [94,95]. In highlighting the difficulties facing SiC
MOSFET development, it is important to keep in mind that early silicon MOSFETs faced similar
developmental challenges that took many years of dedicated research efforts to successfully
overcome.
From a purely electrical point of view, there are two prime operational
deficiencies of SiC oxides and MOSFETs compared to silicon MOSFETs. First, effective inversion
channel mobilities in most SiC MOSFETs are much lower (typically well under 100 cm/V-s for
inversion electrons) than one would expect based on silicon inversion channel MOSFET carrier
mobilities. This seriously reduces the transistor gain and current-carrying capability of SiC
MOSFETs, so that SiC MOSFETs are not nearly as advantageous as theoretically
predicted. Second, SiC oxides have not proven as reliable and immutable as silicon oxides, in
that SiC MOSFETs are more prone to threshold voltage shifts, gate leakage, and oxide failures
than comparably-biased silicon MOSFETs. SiC MOSFET oxide electrical performance deficiencies
appear mostly attributable to differences between silicon and SiC thermal oxide quality and
interface structure. SiC oxides to exhibit higher levels of interface state densities
(
eV cm), fixed oxide charges (
cm), charge trapping, carrier oxide tunneling, and roughness-related scattering of
inversion channel carriers.
One of the most obvious differences between thermal
oxidation of silicon and SiC to form SiO is the presence of C in SiC. While most of the C
in SiC converts to gaseous CO and CO and escapes the oxide layer during thermal oxidation,
leftover C species residing near the SiC-SiO interface nevertheless appear to have a
detrimental impact on SiO electrical quality [94,95]. Cleaning treatments
and oxidation/anneal recipes aimed at reducing interfacial C appear to improve SiC oxide
quality. Another procedure employed to minimize detrimental carbon effects has been to form
gate oxides by thermally oxidizing layers of silicon deposited on top of
SiC [96]. Likewise, deposited insulators also show promise toward improving SiC MOSFET
characteristics. Recently improved SiC inversion channel carrier mobilities (
cm/Vs) using thick deposited gate-insulators has been reported [97].
SiC surfaces are well known to be much rougher than silicon surfaces, due to off-angle
polishing needed to support SiC homoepitaxy as well as step-bunching (particularly pronounced
in 4H-SiC) that occurs during SiC homoepilayer growth
(subsection 2.3.3.2) [52]. The impact of surface morphology on the inversion
channel mobility is highlighted by the work of [88], in which improved mobility
( cm/Vs) was obtained by specifically orienting SiC MOSFETs in a direction so that
the current flowed parallel to the surface step texture. The interface roughness of SiC may also be a
factor in poor oxide reliability by assisting unwanted injection of carriers that damage and
degrade the oxide.
The wide bandgap of SiC reduces the potential barrier impeding
tunneling of hot carriers through SiC thermal oxides, so that perfectly-grown oxides on
atomically smooth SiC would not be as reliable as silicon thermal
oxides [98]. Therefore, it is highly probable that alternative gate-insulators
will have to be developed for optimized implementation of inversion channel SiC FETs for the
most demanding high-power and high-temperature electronic applications.