[MUSIC] We'll now deal with the surface of semiconductors for an obvious reasons, contacts. That is to say, the metallization that creates a contact is always performed on the surface of the semiconductor. This is the first reason. And the second, is that the surface is by definition, an area where there is breaking of transactional symmetry of the crystal. So, the silicon covalent bonds are no longer satisfied. So, for applications, controlling the surface quality will be a major challenge. Therefore, the surface of semiconductor, such as as discussed earlier in the study of the semiconductor metal contacts, corresponded to the case, small a. That is to say the case of flat band. Flat band means that the surface is perfect, free of defects. But in practice, this is not the case. Since a bike of the crystal symmetry will therefore create defects. So, a flat band condition is not at all obvious to achieve. We will describe an example here. Considering an n-doped semiconductor, we will deal with the surface with vacuum and we assume the presence of surface states which are in this case, acceptors, as an example. More specifically, it is assumed that the acceptor levels are at a distance, delta E, relative to the edges of the conduction band as shown in the figure. So, we have an n-doped semiconductor with acceptor at the surface,as an example. So, the semiconductor electrons will naturally fill the acceptor states at the surface because they are located below EF. This will create a potential vye pi-b. Indeed, as previously mentioned, the electrons leave the first false ions for example, to let fix positive ions. Thus, creating a bond bending that will prevent the diffusion of electrons. The calculation of the charge in typical condition is given here. However, the number of states available to the surface is limited, since it depends on the number of defects present at the surface. The potential barrier increases as function of the number of electrons, filling the surface states. However, the height of these potential vye is limited by band gap. So, this means that electrons from the bulk will occupy the states, since they have a higher energy. But since pi-b is limited by Eg, it will therefore, remain uncompensated charges in the semiconductor. So, the surface states located in the band gap, NS2, could be partially filled. But since the positions of the Fermi level corresponds to the level of the states of highest occupied energy. So, if there are a lot of defects on the surface, it will not be possible to increase pi-b in such a way to populate all this state there. This means that the Fermi level will remain close to the surface states at a distance about kT. So, this effect is called in English, the pinning of the Fermi level. Of course, the semiconductor Fermi level is still determined by the bulk concentration of the bands. Because of the partially filled surface states, EF is located a few kT of this levels. So, it is band bending that fits to complete the equilibrium conditions. So, the presence of surface states can pin the Fermi level at the surface, creating band bending. We do not consider such a curvature to treat the semiconductor metal interface. In summary, the possible existence of surface states can have a very strong influence on the semiconductor interface properties, as previously discussed within the flat band assumption. We previously discussed a particular case corresponding to more surface defects that bulk electron can fill. The possible presence of band bending at the surface, can have very important consequences in applications. So, the surface passivation is a critical aspect for technological application of semiconductors. In particular, in the case of solar cell processing. In this last part of the chapter, we will look at another junction device that we'll first study as a solar cell. It is the so-called, Heterojunction. So far, we have consider homojunction. For1example, the PN-junction consist of p-doped semiconductor and n-doped semiconductor. The PN junction could be made of silicon and another semiconductor. But, we can also consider heterojunction in which a semiconductor, gallium arsenide, for instance, is associated with centers of semiconductor such gallium, aluminium arsenide. Indeed, the gallium arsenide on gallium aluminium arsenide do not have the same bond gap. Gallium, aluminium arsenide as a greater forbidden bond. Besides, these two semi-contract can be doped differently. Here, we take the case of gallium arsenide, p-dope and n-dope, aluminium, gallium arsenide. We cconsider this example because one, it's able physically and technically to go aluminium gallium arsenide on gallium arsenide, while maintaining a constant lattice parameter. That is to say, we can make a perfect interface between both semiconductors. Indeed, both gallium, aluminium arsenide on aluminium gallium arsenide semiconductors, as the same lattice parameter, which is difficult to obtain when using silicon. On the contrary, in case of same pi compounds, it is possible to combine semiconductors with different bond gaps. These hetero structure are technologically possible to achieve, fulfilling the condition of epitaxy. If the epitaxy conditions are met, we can assume the flat bond condition as shown in the figure. The work function is defined as the distance between the fermi level on vacuum, as usual. Gallium arsenide on aluminium gallium arsenide have different eloquent affinities, which will differently affect the distance between the energy bonds on vacuum. If we now put together these two semiconductors, one can apply the same reasoning as above. The electrons move from one to the other, by tunneling. So, in this case, from aluminium gallium arsenide to gallium arsenide, since fermi level is higher in n-dope aluminium gallium arsenide. At equilibrium, equal fermi level, throughout the system, will create bond bending as for metal semiconductor contacts. The heterojunction differently affects electrons on holes. Thus, the barrier for holes here has increased. It is implemented by delta EV for holes. However, it is challenged by the delta EC for electrons, because the boundaries of the condition bond at surfaces are related to electronical affinities of the two solids. Therefore, discontinuities on condition band profile is produced, providing an accumulation of electron in the interface area. It can be interpreted as the creation of a potential well as interface. This is a principle of quantum world structure. The sunspot properties are quite interesting in c-Si compounds in particular for opto-electronics applications. Crystalline silicon, amorphous silicon heterojunction will be studied later in the course. In summary, in this chapter, we treated electronic devices, different from the PN junction. Particularly, to address the contact with the metal. The rest of the course, we'll focus on solar cells, fabrication, and operations. In this context, metal semiconductor contact would be very important concept. Thank you. [MUSIC]
