The first part of the chapter is devoted to a brief overview on the problem of the band alignment when a semiconductor heterostructure is formed. This is of paramount importance since it determines the sharing of the band gap discontinuity (DeltaEg) between the valence and conduction bands (DeltaEg = DeltaEv + DeltaEc), which in turns strongly influences the electrical and optical properties of the heterostructure. Considering the fact that intentionally strained heterostructures gain more and more interest in the realization of III-V based devices, attention is devoted to the effect that tetragonal distortion, induced by epitaxy, has on the band alignment in III-V semiconductor heterostructure of zinc-blende structure. This simple picture enables to simulate the Ee-->hh and Ee-->lh transitions in strained heterostructures with abrupt interfaces by solving the classical “particles-in-a-rectangular-box” problem. However, when a “real” periodic Alloy1/Alloy2 heterostructure is epitaxially grown, the Alloy1/Alloy2 and Alloy2/Alloy1 heterointerfaces are far from being perfectly planar and chemically abrupt. The absence of chemical abruptness implies that an interface layer of intermediate chemical composition is present at both Alloy1 to Alloy2 and Alloy2 to Alloy1 interfaces. The presence of these compositional interface gradients, spread over some monolayers, implies that at both interfaces unintentionally strained layers are periodically repeated along the heterostructure. As a consequence of this, it is evident that the model of a particle in a rectangular box represents an approximation of the problem: the potential profile for electrons and holes confined in real superlattices must be more complex. As a case study, the InGaAs/InP system is treated in detail, showing the presence of InAsP and InGaAsP undesired layers at the InP to InGaAs and at the InGaAs to InP interfaces respectively. The use of 4 K photoluminescence spectroscopy and high-resolution XRD in the determination of both chemical composition and width of the interface layers is discussed in detail. Conversely, high-resolution TEM technique and EXAFS, Raman, and IR (in transmission mode) spectroscopies, also useful in the study of the problem, will only be mentioned for sake of brevity.
Modelling characterization of real interfaces in III-V heterostructures
LAMBERTI, Carlo;GROPPO, Elena Clara
2003-01-01
Abstract
The first part of the chapter is devoted to a brief overview on the problem of the band alignment when a semiconductor heterostructure is formed. This is of paramount importance since it determines the sharing of the band gap discontinuity (DeltaEg) between the valence and conduction bands (DeltaEg = DeltaEv + DeltaEc), which in turns strongly influences the electrical and optical properties of the heterostructure. Considering the fact that intentionally strained heterostructures gain more and more interest in the realization of III-V based devices, attention is devoted to the effect that tetragonal distortion, induced by epitaxy, has on the band alignment in III-V semiconductor heterostructure of zinc-blende structure. This simple picture enables to simulate the Ee-->hh and Ee-->lh transitions in strained heterostructures with abrupt interfaces by solving the classical “particles-in-a-rectangular-box” problem. However, when a “real” periodic Alloy1/Alloy2 heterostructure is epitaxially grown, the Alloy1/Alloy2 and Alloy2/Alloy1 heterointerfaces are far from being perfectly planar and chemically abrupt. The absence of chemical abruptness implies that an interface layer of intermediate chemical composition is present at both Alloy1 to Alloy2 and Alloy2 to Alloy1 interfaces. The presence of these compositional interface gradients, spread over some monolayers, implies that at both interfaces unintentionally strained layers are periodically repeated along the heterostructure. As a consequence of this, it is evident that the model of a particle in a rectangular box represents an approximation of the problem: the potential profile for electrons and holes confined in real superlattices must be more complex. As a case study, the InGaAs/InP system is treated in detail, showing the presence of InAsP and InGaAsP undesired layers at the InP to InGaAs and at the InGaAs to InP interfaces respectively. The use of 4 K photoluminescence spectroscopy and high-resolution XRD in the determination of both chemical composition and width of the interface layers is discussed in detail. Conversely, high-resolution TEM technique and EXAFS, Raman, and IR (in transmission mode) spectroscopies, also useful in the study of the problem, will only be mentioned for sake of brevity.
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
