On the low-temperature mobility of holes in gated oxide Si/SiGe heterostructures

Abstract

A detailed comparison is made between theory and experiment for the low-temperature mobility of holes in gated oxide, coherently strained Si/SiGe heterostructures. We conclude that the mobility is mainly limited by interface impurities, conventional surface roughness and strain fluctuations; by contrast, we argue that alloy scattering is comparatively weak. Comments regarding possible mobility degradation due to oxide formation are also made.

Issues on the molecular-beam epitaxial growth of p-SiGe inverted-modulation-doped structures

“…The apparent universal relationship between and n s , irrespective of which interface ͑N or I͒ is involved, the cap thickness, and whether the cap surface is unpassivated or gated, strongly suggests that the 2-DHG carrier mobilities at either SiGe/Si interface are limited by similar short range scattering processes. 15 It can therefore be tentatively concluded that there are not large differences between interface roughness, interface charge, and/or alloy scattering at inverted and normal interfaces ͑all these processes playing a role at 4.2 K͒. The parameters studied merely alter the n s value and do not contribute to carrier scattering.…”

Wave function-dependent mobility and suppression of interface roughness scattering in a strained SiGe p-channel field-effect structure

“…6,13 This symmetry suggests that the interface charge might be an intrinsic effect not associated with unintentional doping of the alloy. One possibility is that it may be piezoelectric in origin.…”

“…Alloy disorder, interface roughness, and strain fluctuations are relatively less important at 4 K ͑at least for the carrier concentration range in question͒, but at 300 K they become much more significant. One can do little about alloy disorder scattering, but the predicted trend whereby the mobility due to interface roughness and strain variations increases with n s is highly desirable for SiGe pMOS applications at 300 K. 11,13,15 This trend is a direct consequence of the fact that scattering processes associated with roughness at the interface are especially sensitive to the width of the state ͑wave function͒. 11,16 As shown in Fig.…”

“…In Fig. 5 we compare the calculated interface roughness and strain fluctuation limited mobilities at 4 K for the present device configuration with an ordinary enhancement mode SiGe pMOS structure, 13 i.e., one not possessing the Si:B doping slab beneath the alloy, but otherwise having the same roughness parameters. Figure 5 does not represent a full calculation for realistic device structures intended for room temperature operation ͑with oxide layers etc.͒, but the results do suggest that the trend towards the reduced importance of the rough interface at the higher carrier concentrations is at least plausible.…”

Effective mobilities in pseudomorphic Si/SiGe/Si p-channel metal-oxide-semiconductor field-effect transistors with thin silicon capping layers

“…Simulations indicate that these points are where parasitic conduction at the Si/SiO 2 interface begins to affect the transport characteristics of the device. It should be noted that for both W01 and W09 7 Whall and Parker, 8 and Kearney and Horrell 9 have provided strong evidence that interface roughness rather than alloy scattering plays a dominant role in limiting the hole mobility in devices of this type. To determine the effect of thickness fluctuations, ⌬, in the SiO 2 insulator and Si cap, we have evaluated the associated fluctuations in the electrostatic potential 10 at the Si/SiGe interface using a self-consistent solution of Poisson's and Schröd-inger's equations.…”