We have presented in the preview sequence, mythological silicon production on purification. All of these processes represents a substantial energy consumption. We will look in these accounts how to produce crystalline silicon wafers. The physical principle which leads to the crystallization is easy. From some of dynamic point of view, crystalline solid phase is more stable than amorphous one. Solidifying the liquid silicon thus lead to the formation of a crystalline material. So wafering process is based on metallurgical silicon mentioned above. The liquid silicon may also contain [inaudible] phosphorus or born. This method is also carried out at a high-temperature liquid silicon. As will be seen later, the monocrystalline silicon wafers are grown from a crystalline seed that can then propagate the crystalline order. We present the various forms of silicon. Starting from metallurgical silicon that can be liquified in order to obtain mono or multi-crystalline silicon, as discussed later. The multi crystalline silicon can be produced in massive quantities, 700 kilos or more, resulting in substantial cost reductions. This silicon is ultimately used to produced solar cells, mono or multi-crystalline, as seen here. The mechanism for obtaining mono-crystalline silicon is illustrated in this animation. This is called the Czochralski method. The entire process is performed in clean room environment in order to prevent impurity in corporation in the silicon. On RF heating system combined with graphite susceptor is used to remove any metallic contact with silicon. At high temperature, such contacts may induce contamination in silicon. Their satin material is purified metallurgical silicon that is melted at 1,500 degrees. The process is induce from a small seed that is itself mono-crystalline. The seed crystal is soaked in liquid silicon, so liquid silicon will cool in contact with the solid gem and thus, crystalized. This solidification will lead to the propagation of the crystalline order carried by the gem. By pulling of watts out of the RF heating, it leads to the growth of mono-crystalline ingot, proceeding slowly. The symmetry of the system is naturally cylindrical. The size of the ingot is limited to 200 or 300 millimeter in order to avoid thermal gradients. The various feature of the Czochralski method are shown right here. This is the method devoted to microelectronics, giving the best performances in terms of energy efficiency of photovoltaic cells on modules as shown here. The maximum size of mono-crystalline wafer is currently a diameter of 300 millimeter. In the absence of mono-crystalline seed, one obtains multi-crystalline silicon as illustrated in the second animation. As previously, the cooling liquid, when it leaves the heating zone induces the crystallization of silicon. In the absence of seed, the crystallized appear randomly during cooling. Then, when the crystallized grow at the expense of the liquid during the movement, you don't want to get in touch. There is this formation or great boundaries between the crystallized. So, advantage of this technique is to be spread over large area, about meter square. Departing from a circular geometry, poorly suited for double takes applications. In industrial application, the methods of manufacture is continuous. Metallurgical silicon is introduced continuously during the pulling of the multi-crystals. This continuous process allows to obtain ingots of several hundred kilos. This is followed by the sowing of ingots or multi-crystalline. As shown here, to finally get solar cells, multi-crystalline here, thickness of about 200-300 microns. Yields on cells on modules are slightly lower than for mono-crystalline silicon. The various characteristics of the solar multi-crystalline silicon are shown here. The road map of the evolution of crystalline silicon ingots are shown in this slide. The size of the ingots continues to rise in order to reduce the cost of the cells. We thus passed the milestone of 800 kilos for multi-crystalline ingots and 200 kilos for single crystal. These trends will continue in the coming years. Let us mention the ways to reduce the cost of silicon that's totally declined in recent years, so one can relax the requirement of purity. As we have seen, the level of impurities, ten to the minus eight in microelectronics. However, two order of the magnitude, lower for silicon solar cells. There has been in recent year, at a very fast output goes are thin here. You see very clearly by comparing pinks on red histogram that the main application of crystalline silicon is photovoltaics, one order of magnitude higher than for micro-electronics. It says here on the right scale, the growth rate. This growth rate has varied much in recent years. In 2012, there was a crises in the supply of silicon. Then, the market regained in growth rate about 25% per year. Currently, the thickness of the wafers that are used in manufacture of solar cell is [inaudible] 200 microns, which is about seven grams of silicon per watt peak. The peak watt is a watt at EM 1.5. The weight of silicon per watt use was divided by a factor two in ten years. We reach the final step. The preparation of the p-n junction. The silicon used in solar cells is often p-doped. Also, its strengths appear to be reversed in recent year. We can mention some arguments in favor of p-doped silicon. First, the residual impurities in crystal and silicon are generally acceptor. Under diffusion lengths of minority chaos in the p-regions, electrons is higher. This argument is particularly relevant for thick cells. For the p-n junction, if the bold material is p-doped, one will then diffuse phosphorus at high temperature on the surface region of silicon, as shown in this figure. It can be optionally use, gaseous precursor, phosphorus PH3 phosphine. It operates similarly to the n-doped silicon. After this high-temperature process, silicon is covered with a thin layer of oxide. With this [inaudible] in this silicons, the manufacture of the solar cells. I present here, a mono-crystalline silicon wafer polished surface. We will turn to the description of the solar cell made from crystalline silicon. Thank you.
