Heterojunction - a building block of heterostructures The interface arising by joining two different semi-infinite semiconductors is called a heterojunction.

The forbidden energy gaps of the constituent layers of the semiconductors are usually different, and the materials electronic properties abruptly change at the interface plane. It is desirable for the difference of lattice constant between the constituent semiconductors to be about (0.01) 1 per cent or less of the lattice parameter. Fortunately, many semiconductor materials crystallize in the diamond or zinc-blende symmetry and satisfy the lattice matching requirement. The electronic properties depend largely on the band offsets and on the bulk forbidden gaps of the constituents. The band alignment determines the properties of heterojunctions. The topology of the band alignments is classified according to the relative ordering of the band-edge energies as shown in the Figure 1.1.

The pattern of alignment in Figure 1.1(a) is called straddling configuration. The remaining types, given in Figures 1.1(b) and (c), are known as staggered and broken gap. From the superlattice engineering point of view, straddling is known as Type-I and staggered and broken gap as Type-II. Heterostructures quantum well Heterostructures are formed from multiple heterojunctions. The double junction structures are known as quantum wells (QW) and the multi-junction structures as superlattices (SL) or multiple quantum wells (MQW). In Type-I quantum wells electrons and holes are confined in the same layers; however, in Type-II quantum wells electrons and holes are not spatially confined in the same region. The recombination times of electrons and holes are long in Type-II heterostructures. The electron (solid circle) and hole (open circle) confinement in Type-I and Type-II single quantum wells are shown in Figure 1.2.

Superlattice and multiple quantum well A superlattice is a set of quantum wells which are sufficiently closely spaced for the wavefunction, representing the electrons, to tunnel between the wells. The model of the superlattices' electronic structure is based on the assumption that the separation between adjacent wells is small enough for the wave function overlap to be finite. However, the overlap may be regarded as a small correction if the wave function is highly localized in the well. A microstructure in such situation is called a multiple quantum well (MQW). The wavefunction becomes very small between the wells in MQWs. When the alloy layers (barriers of SL) are sufficiently thick (>100 ), the tails of the wave functions associated with the states confined in the individual wells do not penetrate far enough into the barrier for them to overlap; as a result, the system will not be able to form a superlattice, but a MQW. The potential profile of the superlattices is analogous to the Kronig-Penney model. QWs have applications in lasers, LEDs, etc; while superlattices have a variety of applications in resonant tunneling structures and far-infrared detectors; likewise, MQW structures have applications in various electronic devices, including quantumcascade lasers and solar cells. 2.4 Basic approaches for heterostructures 2.4.1 Confinement potentials In a QW, where different materials are combined to form a heterostructure, the shift in the band structure induces a step like potential profile in the growth direction as shown in Figure 2.2. The height of the discontinuity (in the so called abrupt junction approximation) is called band offsets. The conduction band offset confines the electrons whereas the valence band offset confines the holes. At present it is possible to realize different shapes of confining potentials due to advancements in growth techniques. The simplest and most commonly employed configurations are the single or multiple square quantum wells. The degree of confinement as well as the degree of overlap between confined electronic states can be controlled by the width and depth of the confining potential well, and the width of the barrier layer. Although the square well configuration in the form of single, multiple or SL structure remains the most commonly studied, techniques such as MBE permit realization of other functional forms of the potential through compositional grading of the barrier and well layers. Some of those potential forms of quantum well are parabolic, triangular, trapezoidal, V-shaped etc. Any such potential shape may create the new electronic quantum systems which may be exploited for novel devices. 2.4.2 Quantum well and confined states When the Fermi energies through three layers of material 1/material 2 multilayer structure equalize, then the schematic of quantum well will be as shown in the Figure 2.2. Material 1 and material 2 represent well and barrier respectively. Experimentally, QW are formed by sandwiching a narrower band gap crystal or semiconductor (the well) with the wider band gap crystals as adjacent layers (the barriers). The relative heights of the discontinuities in the conduction band (CB) and valence band (VB) edges are intrinsic properties of the two materials involved. Since the narrower band gap material forms a potential well in the CB and VB, the electrons and holes are bound states in the growth direction (z-direction), while their motion in x-y plane is unrestricted. This implies the carriers are quantized in one-dimension, i.e. in the direction perpendicular to the interfaces. This is known as quantum confinement. Thus, a one-dimensional confinement effect occurs in QW which is used to engineer the electronic

properties of the materials. The chemical composition, dimension of the layers, and the band gap energies vary the properties of QW. The confinement effect may be increased by nanostructuring as quantum wire and quantum dot.

Figure 2.2 Schematic of quantum well as a function of distance in the growth direction z and energies of the sub-bands within the well The width of the quantum well determines whether the confinement effect will be important. Usually, the carriers experience significant quantum confinement effects for well widths of order 100 Angstrom or less. It is reasonable to consider the well width be thick enough (L > 20 Angstrom) so that the well layer material has the properties characteristic of macroscopic crystal and thin enough (L < 400 Angstrom) so that the width L is shorter than the mean free path ( = v; = m* /ne2, is resistivity), i.e. an electron is likely to traverse the well without undergoing a collision of any kind.