The saturated bond in semiconductors

The origin of this hybridization gap in tetrahedral semiconductors can best be understood by taking the four sp3 hybrid orbitals as our starting basis rather than the four free atomic orbitals s, p, p , and pz. As is well known (see, for example, McWeeny (1979)), the former are linear 

These two sharp bonding and antibonding levels are then broadened into a band of states by coupling between different hybrids on the same atom, since 

Since we are neglecting overlap between hybrids on different atoms unless they point along the same bond, only one contribution remains in eqn (7.87), so that 

The energy of the bottom of the valence band El may now be found, since it corresponds to the most bonding state illustrated in the top panel of Fig. 7.18, in which each bond orbital is in phase with the six neighbouring bond orbitals. Hence, its energy is given by 

---

FREQUENCIES OF THE LOCAL MODES OF VIBRATION AT 5 K OF HYDROGEN SATURATING A DANGLING BOND IN A VACANCY IN VARIOUS SEMICONDUCTORS. 

Consequently, the bond is fully saturated for A sp = 0 with a bond order of 1, but it is only partially saturated by the time the gap closes for AEap/2 h = 1 (cf eqn (7.92)) when the bond order equals 0.76. This simple second moment model has been extended to include the compound semiconductors. The resultant values of the bond order are given in Table 7.2. We see that the bonds in tetrahedral carbon and silicon are almost fully saturated, but those in zinc selenide and cadmium telluride are only about 75% saturated due partly to the mismatch in the sp orbitals between chemically distinct atoms. 

Surface states can arise simply because the atomic bonding at a semiconductor surface is necessarily different from that in the bulk. For example, in a Si lattice, the bonds at the Si surface are not ftilly coordinatively saturated. To relieve this unsaturation, either a surface reconstruction can occur and/or bonds to the metallic material can be formed. This distinct type of surface bonding results in a localized electronic structure for the surface which is different from that in the bulk. The energies of these localized surface orbitals are not restricted to reside in the bands of the bulk material, and can often be located at energies that are inside the band gap of the semiconductor. Orbitals that reside in this forbidden gap region are particularly important, because they will require modifications of our ideal model of charge equilibration at semiconductor/metal interfaces. ... 

---