Thread Content
The band structure describes the charged single-particle states (i.e., electrons or holes). Since zinc oxide is a direct-bandgap semiconductor, that is, it has spherical extremum points where the maximum of the valence band and the minimum of the conduction band (VB and CB) coincide in the Brillouin zone, at the Г point where K = 0, it is this region that we are primarily interested in. As mentioned above, the lowest conduction band is composed of the empty 4s state of Zn2+ or the antibonding sp3 hybridized state. The compatibility table of group theory and the references therein tell us that at the conduction band bottom there is a Г1 symmetry without considering spin, and a Г1×Г7=Г7 symmetry with spin considered. The effective electron mass (more precisely, the polaron) is almost isotropic, with a value of approximately (0.28 ± 0.02)m0. If spin is ignored, the valence band originating from the 2p orbit occupied by O2- or the bound SP3 hybridized orbit is split into Г5 and Г1 states under the influence of the hexagonal crystal field. Taking spin into account, due to spin-orbit coupling, the energy bands further split into three symmetric subbands with second-order degeneracy: (Г1+Г5)×Г7=Г7+Г9+Г7. These valence bands are labeled as A, B, and C bands from high energy to low energy in all zinc blende-structured semiconductors (such as ZnS, CdS, CdSe, or GaN). In the vast majority of cases, the band ordering is AГ9, BГ7, C7, and the spin-orbit splitting is greater than the crystal field splitting. However, there has been a long-standing debate regarding ZNO: whether the order of the valence bands is the usual one, or the AГ7, BГ9, CГ7 order due to so-called \"negative spin-orbit coupling\", or an inverted order of the valence bands. In fact, spin-orbit coupling is always positive. However, due to the low nuclear charge of oxygen, the values from sulfides, selenides, and tellurides with hexagonal and cubic IIb structures can be extrapolated to give approximately 15 meV for oxides. The repulsion between this energy level (oxygen’s 2P) and the fully occupied zinc 4D energy level may easily cause a Г7 energy level to be higher than the Г9 level, resulting in an inverted valence band order; this phenomenon is also observed in the Ib-VII group semiconductor CuCl. Since the splitting between the A and B valence bands is only on the order of 5 millielectronvolts, and the selection rules are essentially the same (the transition from the two upper AГ9 and BГ7 valence bands to the Г7 conduction band is a dipole transition, while spin flips can only occur in the EC mode; from the CГ7 valence band, transitions can also only occur in the EC mode), discussing the symmetry of the A and B valence bands is an overreaction. But these types of questions tend to require a very long discussion time. The various detailed discussions proposed in recent literature in support of the normal valence band order in ZnO have not been confirmed. In Appendix A, we summarize some of the arguments and literature in support of and against inverted band order, as well as the literature on band structure calculations. Therefore, we use the following inverted order AГ7, BГ9, CГ7. The effective hole mass in zinc oxide is isotropic, and the characteristic values of the A, B, and C valence bands are as follows. Some band structure calculations (see Appendix A) predict strong mass anisotropy in the valence bands, as is the case with CdS, but the experimental data available in the literature do not indicate any such anisotropy in ZnO. The G7 symmetric band may have a small k-linear term, but the A valence band shown in Figure 7 has only k⊥c. For the conduction band, this effect can be ignored. Article source: http://www.jszfxy.cn/news/show-212.html