[MUSIC] We will now focus on the other junction devices. Keeping the context of the operation of the shown here, I remind you that during the exposure to solar photons, some of these photons can be reflected, therefore, not used. But what we say in the first chapter, was the photon absorption mechanism due to the electron excitation from the valence bond to the condition bond. So only the energy that corresponds to the bond cap is used. The rest is lost by thermalization. So first lost was written in the second chapter. In the third chapter, they will teach you the p-n junction. We sort out this electron pairs was separated. So the carriers will be transmitted through an external circuit. So we note the losses that are related to contacts with electrodes. Indeed, the contacts between a semiconductor and metal are complex, and can result in energy losses. This is what we will see today. So we will look first at the junction metal-semiconductor equilibrium as for the PN junction. So then we will study the metal-semiconductor junction out of equilibrium. Then we will focus on the Ohmic contact. Before everything the surface of semiconductors. Seems the contacts are made on the surface of the semiconductors. Finally, we'll treat the heterojunctions. The understanding of the properties of the junction between a metal and a semiconductor is a major technological problem when the semiconductor is inserted in an external electrical circuit. In particular, the cases in which this contact is purely resisted called only contact. So then you will deal with interface between a semiconductor on a metal. So I remind you, essential difference between a semiconductor and a metal. The semiconductor is an insulator with a moderate bond gap. So the balance bond is completely filled at. On the next bond, condition bond is empty. Why in the case of a metal, on the contrary the Fermi level is located within the bond, so that the electron can change energy state very easily on both sides of the Fermi level. So for simplicity, we consider here an n-type semiconductor. We define the work function of the metal, Phi m. It is the energy required to extract an electron from the metal near the surface to the vacuum at Infinity at zero speed. Similarly, we defined the work function of a semiconductor by the distance between the Fermi level of the last occupied state in the vacuum. The problem with the semiconductor, that the Fermi level is not an intrinsic quantity since it depends on doping. So we will preferably consider, in the case of semiconductor, the so-called electron affinity. Here Chi, which is a distance between the edge of the condition bond on the vacuum. Chi is a semiconductor quantity. In general the work function is an intrinsic property of your materials. In the case of a metal, it is not the same for the silver, aluminum, or iron. And likewise, the electron affinity of silicon differs from that of Germanium ounce one. In particular, it will be different for crystalline, an amorphous silicon. Similarly to case of the p-n junction, which consists in putting in contact the p type semiconductor on n type one. We'll put the contact, semiconductor on the metal. There are two possible cases, which depend on the difference between phi m and phi s. First possible case are shown here. The metal work function is larger than that of a semiconductor. Where the two materials begins to be very close as the Fermi level of the semiconductor is greater than that of the metal, electron will pass from semiconductor to the metal by definition of the chemical potential. This process of equalization of the Fermi levels on either side of the interface will be due to the flow of electrons near the surface of the semiconductor to the metal. Therefore, the electrons leaving the semiconductor leave positive ions. This is a phenomenon similar to that of forming the p-n junctions. These positive charges are fixed ions. Therefore, there is creation of the bond building, as in the case of the p-n junction. A potential is then created as a interface between the semiconductor on metal. The value of EC minimum of the condition bond at the surface, is fixed by the electron affinity of the semiconductor. To calculate the potential at the interface we, again, use the Poisson's equation, assuming that the energy bonds are flat, far from the interface as for the p-n junction. The width of the spread charge region is obtained from solution of the Poisson equation. When a [INAUDIBLE] similar to the case of the p-n juncture is found for the metal semiconductor interface. Accordingly, when contacting a p-n junction to a metal electrode, can create another junction. In this first case, a potential is created between the metal and the semiconductor. So the electrons of the p-n junction will tend to be confined in the semiconductor. By the [INAUDIBLE] production leading to the hold back for practical applications. We solved the first case. Phi M larger than Phi S. Now we will consider the opposite case. The metal work function is smaller than that of the semiconductor. The mechanism is the same as presented before. So electrons now pass from the metal to the semiconductor since the metal has the greatest chemical potential. In order to balance the Fermi levels on both sides of the interface. This flow induces the creation of a region of positive charge as the metal surface. As before, the Chi value imposes the value of EC conduction bond edge, as semiconductor surface. In the semiconductor close to the interface, there is a zone in which the Fermi level is higher than the bond edge. Since there are many states in the condition bond there are also bond bending as in the previous case. But in the region where EC is below the Fermi level, the number of states is very large. So they will be filled fastly so that the width Wpi, in the second case, will be much smaller than the width W calculated before. In the previous case, the area of width W was a charge region, therefore resistive layer, as in the case of the p-n junction. With few carriers, because of the presence of the electric field. In this second case, on the contrary, the area will be much thinner because of the large number of states is a conduction bond, thank you. [MUSIC]
