Silicon quantum-well model

Figure 3 shows the Eg values of 1-D silicon clusters ls a function of size. The calculated Eg value decreases with increase of the number of silicon atoms, that is, the cluster size. Figure 3 also indicates the Eg values evaluated by Matsumoto s quantum-well (QW) modell l in which the wavefunctions are confined in the potential well and asolute Eg values were determined by interpolating the experimental Eg values of disilane (6.5eV) and polysilane (4eV) by the size dependance of the energy levels of the wavefunctions. 

Figures 3, 4 and 5 also show the Eg values when silicon 3d orbitals are added to the basis sets. All of the calculated Eg values without Si 3d orbitals are about leV higher than those of the QW model. However, when Si 3d orbitals are taken into account, the calculated Eg values decrease by about leV and good agreement with the values given by the QW model is obtciined. These results imply that the energy levels around HOMO and LUMO of silicon clusters are well modeled by the quantum confinement of delocalized wavefunc-...

We calculated the electronic structure of 1-D, 2-D, and 3-D silicon clusters using (DV)-Xa MO method. The calculated energy gap Eg) between HOMO and LUMO decreases with increase of the cluster size and the number of dimensions of the cluster. It is found that including silicon 3d orbitals as basis sets results in lowering the Eg values by about leV. The results also show that the components of silicon 3d orbitals in the unoccupied levels near LUMO axe over 50%. In the case of silicon clusters, the size effect on the Eg value is well described by the quantum-well(QW) model in which the delocalized wavefunc-tions are confined in the potential well. The calculated results predict that Eg... 

Since the discovery of the intense red photoluminescence of porous silicon [1,2], much work has been devoted to this particular nanostructured material [4, 5] and, in the meantime, also to silicon nanoparticles [6, 7]. An important issue of current studies is the influence of the surface passivation on the photoluminescence properties. It has already been said that, in the quantum confinement model, it is essential that the surface is well passivated to avoid any dangling bonds [8]. Being middle-gap defects, these dangling bonds will quench the PL. On the other hand, the surface itself may lead to surface states that can be the origin of another kind of photoluminescence [9,10]. 

In 1990, Canham observed intense visible photoluminescence (PL) from PSi at room temperature. Visible luminescence ranging from green to red in color was soon reported for other PSi samples and ascribed to quantum size effects in wires of width 3 nm (Ossicini et al, 2003). Several models of the origin of PL have been developed, from which we chose two. In the first (the defect model), the luminescence originates from carriers localized at extrinsic centers that are defects in the silicon or silicon oxide that covers the surface (Prokes, 1993). In the second model (Koch et al., 1996), absorption occurs in quantum-confined structures, but radiative recombination involves localized surface states. Either the electron, the hole, both or neither can be localized. Hence, a hierarchy of transitions is possible that explains the various emission bands of PSi. The energy difference between absorption and emission peaks is explained well in this model, because photoexcited carriers relax into surface states. The dependence of the luminescence on external factors or on the variation of the PSi chemistry is naturally accounted for by surface state changes. 

---