 So, let us start. So, if you recap last lecture, we have looked at the physics of semiconductor, we have looked at p n junction operation and then we have looked at how p n junction works as a solar cell under when the light falls on it and then we have looked at various parameters of solar cell and in the last lecture, we looked at the design of a solar cell. So, where should be your junction, how to design anti-reflective coating, thickness, back surface field, front contact etcetera. And in the last, we were discussing about the series resistance as a one part of the design is to minimize the series resistance. So, that we will discuss briefly today, but today's two of my talks will be mainly on production of silicon and how the silicon wafers is used for manufacturing solar cell. So far, we have looked at the structure, cell structure, how it should be, but now today we should also look at how the cell, how the solar cell is manufactured using the various fabrication technology. So, the recap of the last lecture is that we have looked at the various upper limits of solar cell parameters, open circuit voltage, short circuit current, fill factor, resistive losses, sheet resistance, resistive losses we have started looking at. So, if you look at the device, the solar cell, the current flows as I said in emitter, sorry in base it flows vertical up because that is the lowest resistance path. Your emitter is very thin. So, the current in emitter flows in horizontal direction. One other reason why current is shown to flow horizontal is that the distance between the fingers is normally of the order of millimeter, while the thickness of emitter is of the order of few hundred nanometer. So, one millimeter is 1000 micrometer and 1 million nanometer. So, if you compare the different distances, the thickness of the emitter is very thin as compared to distance between the fingers. And therefore, we have to show that the current is flowing horizontal. Then the current is collected by the fingers, current is collected by the fingers and then it is given to the bus bar. Bus bar is the thicker line, basically bus bar is collecting current from all the fingers and your bus bar is the thicker line. Therefore, the contribution to the resistance comes from the base itself. So, base contributes to the resistance. The emitter contributes to the resistance. The contact between semiconductor and metal is also not a perfect contact, not perfectly omic contact. So, there is some resistance there. The metal fingers have their own resistance. The bus bar has its own resistance. The big contact has its own resistance. So, there are six component of resistance that comes in the path of the current in a solar cell and lower the resistance, better it is. Lower the resistance, better it is. And this is the series resistance. By the way, it comes in the series. So, it is a series resistance. Lower resistance is better because lower resistance means lower i square r losses. So, whatever power that is generated in solar cell by light should not be dissipated within the solar cell. Otherwise, the output power will be lower. So, the whole idea of the design is to minimize these resistances. Normally, the base resistance is very small because the cross section area is very large. So, this resistance component r 2 will be very small. The r 1 that is the back contact resistance is also very small because back contact remember it is everywhere. Back contact is everywhere. So, resistance of the back contact is also very small. And r 4 contact resistance can also be very small. So, main resistances that you have to actually minimize is the emitter resistance, the finger resistance and the bus bar resistance. Solid matter semiconductor contact resistance will also be important. So, the main resistance in the solar cell comes from the top part. Note down this point that the main resistance to the solar cell comes from the top portion. So, if I draw the solar cell diagram again, so there is a front metal contact here, then your back metal contact everywhere. So, current flows like this, then flows like this, then it flows vertically here, flows like this vertically here. So, typically this resistance of this part resistance of this part is almost 0, here r value is almost 0, here r is not equal to 0 means high value. Metal fingers have r not equal to 0 which is high value and the metal semiconductor contact will have r value not equal to 0. So, the main resistance in a solar cell comes from the top portion. Main series resistance R S contribution, main series resistance contribution come from the top. So, that is what people normally try to do that try to minimize this resistance as much as you can. You can do the analysis actually very simple analysis that current flows vertically, here it flows horizontal. So, if you know the resistance of the emitter and by the resistance of the emitter will discuss while we are discussing the solar cell itself that the resistance is first of all emitter doping is not uniform. And therefore, there is a concept called sheet resistance that we have to define. I will go directly. So, if you have the look at the parameter, if I have the spacing S, the length of the finger is L spacing between the finger is mentioned here D, but actually it is S. If I have that kind of parameter then the power loss in emitter will be given by this equation. So, J is the current density, L is the length of the finger, R is the sheet resistance of the emitter and y emitter. So, we are talking about emitter power loss, S is the spacing between the fingers. So, remember how important S is right cube power of the cube. So, F S increases slightly. So, if this space between the fingers doubles the power loss in emitter becomes 8 times. So, it is very very important design parameter of a solar cell to have the proper spacing between the fingers. It is very important design parameter to have the proper spacing between the fingers. Therefore, it should be carefully chosen. So, now if I have again, if I go back, if I go back here, let me draw again for the clarity. If I have this, if my emitter, if my fingers are here. So, this distance is S, the emitter resistance is normally given as a sheet resistance R sheet because it is a very thin layer, very thin layer and non uniform doping. And then we have seen from the expression that P depends on cube of the spacing. It also depends on R sheet. This is the power loss in emitter. Do we want power loss to happen in emitter? No, we do not want power loss to happen in emitter. We want all the power that is generated. We want all the power that is generated to come out of the solar cell and deliver to the load external not within the, within the cell. So, therefore, what is important? Spacing between fingers is important. Spacing between fingers, S is a spacing between fingers. Now, so you will say that because power loss in emitter is a function of S cube. So, why not to have very small S? Why not to have very small S? Means why not to have very small finger spacing? If you have small finger or front contact spacing, what will happen? If you have small spacing, then your fingers are like this. This is the scenario of small S, but when the scenario, when you have the small S, when your S value is small, what is, what is the problem now here? Your metal coverage of the front surface is larger. So, this your metal is shadowing the light. So, what you are doing? You are shadowing the light because of the more metal contact. So, you have said shadow loss will increase. So, you do not want high shadow loss also. So, this is a very good optimization that is required. What is the optimum spacing so that loss in the emitter is not much. So, typically you have 4, 5 percent, 3 percent loss that is allowed and according to that you can choose the finger spacing. Shadow loss, you do not want to be very high and i square r loss is in emitter also you do not want to be very high. So, therefore, this is very important. In the optimization of finger spacing is very important. Otherwise, if you have very close fingers, your shadow loss is higher, if your very fingers far from each other, then your raise to loss is higher. So, that optimization is required. So, what you can do basically in order to solve this problem which is not easy at all to solve the problem. So, instead of having larger width and less height of the finger, you should try fingers or the metal contact which is having smaller width and higher height. But in practice, this metal contacts are printed by screen printing and it is not possible to get a very thin width. This is called the aspect ratio. The ratio of W widths to height is over H is aspect ratio and people want aspect ratio as much as possible. Typically aspect ratio typically the finger width, typically the width of this finger W is in the range of about 100 micrometer. In the height will be in the range of about 20 to 30 micrometer. So, you will see the aspect ratio is will be about 0.4, 0.3. Typically, you want aspect ratio to be 1 or even higher than 1. So, that is one of the important design aspect in not only the design, implementation practice is still a challenge. You cannot actually, if you try to implement very good contact and very high aspect ratio, then your cost also increases. So, how it is going to affect? So, the series resistance affects the power loss because of internal series resistance, there is a internal losses within the solar cell. Because there is an internal losses within the solar cell, what kind of losses within the solar cell? I square losses. So, the power available to you outside the solar cell or to the load will be lower. So, when you have a high series resistance, what happens? When you have high series resistance for the same current, your voltage available will be lower. So, what I am trying to tell you is that, if I have, if my good solar cell will be like this, a less good solar cell will be like this. So, this solar cell will have high R s, this solar cell will have low R s. So, what is change, what is happening? The slope of the I V curve, the slope of the I V curve is changing, reducing towards the open circuit voltage, the slope of the I V curve reducing or you can also say that because of your R s, there will be I R s drop, voltage drop inside the solar cell because your series resistance, the V drop inside the solar cell will be I R s. So, therefore, you can say that for the same current now, the voltage available to you will be lower. For the same current, the voltage available to you will be lower. So, this is the effect of R s, series resistance. So, that you can see here, also for a solar cell is a non-linear device, but at the maximum power point, if you take the ratio of the V m over I m, you take the character, you get the value of resistance which is the characteristic resistance of the cell. And then, a series resistance at any other point is R s divided by R c h and your fill factor, your power first of all decreases, your power, maximum power that you can get under the series resistance equal to p 0, p 0 is the power when there is no series resistance and 1 minus R s, small R s is given by this. So, basically the impact of high series resistance is to reduce your power and that happens because of the reduction in the fill factor. Your fill factor also reduces, which is very clear. Fill factor is the definition of squareness of the I V curve and if your I V curve is less square, which means the fill factor is less. If your I V curve is more square, your fill factor is also higher. Similar to that, so what main difference happens is the slope of this, slope of the I V curve here towards the open circuit voltage determines the series resistance. So, there are many research papers available which will actually tell you that if you measure the slope here, you can actually measure the series resistance or at least qualitatively if you look at the slope of the I V curve and if the slope is not very high or the slope is not very high, you can say that this is the I V curve of a solar cell which is having high series resistance. I will draw you again. So, if I draw the curve like this, this can also be a solar cell curve, but it is very clear from this curve that this slope, it is very clear from this curve that this slope here is very low as compared to what otherwise is possible like this. Therefore, you can say you can say and impress people that this solar cell has very high series resistance, very high or on the other hand you can say because it is not square at all. So, you can say that fill factor for this solar cell is very low. So, this as a student of this course, now it is should be very easy for you. Similarly, shunt resistance, shunt resistance will have mainly the slope of the curve near the short circuit current will actually give the shunt resistance. Shunt resistance will actually results in the current bypassing the junction. So, bypassing the junction. So, therefore, for same open circuit voltage you are actually now getting lower shunt resistance, but let me tell you that in general the shunt resistance is not a problem for a solar cell. Shunt resistance normally very high it is the series resistance which is always the more problematic. It is the series resistance that is always the more problematic. So, that is it. So, it is shunt resistance normally is not a problem. As I said the earlier the shunt resistance should be very high we looked at the model of the solar cell. So, series resistance should be very low and shunt resistance should be very high and that is normally the case. So, this is the summary of losses. Fine. So, let me move ahead. So, this is the summary and let me give the typical parameters. So, that we have to discuss the silicon production today. So, normally we use silicon substrate because it is abundant all other people are getting clearly. So, just check Sastra. So, normally substrate for solar cell is silicon most commonly then people can use other substrate also silicon is abundance non-toxic. The cell thickness is 180 micron. Remember cell thickness is 180 micron. The doping of the base is normally tends for 15 or tends for 16 atoms per centimeter cube which is corresponding to the resistivity of 1 ohm centimeter or around that. So, typical resistivity is 0.1, 0.2, 1 ohm centimeter, 2 ohm centimeter that is typical. Emitter thickness is less than a micrometer. Emitter thickness is less than a micrometer doping is of the order of tends for 19 per centimeter cube or tends for 1918. Sheet resistance of 50 to 100 ohm per square. Optical losses can be avoided by anti-reflective coating and in normally we do the surface texturing as well as anti-reflective coating. The thick layer which is used as anti-reflective coating is silicon nitride. It has a thickness about 70 to 80 nanometer you can calculate and for the anti-reflective texturing the pyramids are formed which is normally 4 to 5 micrometer and all in the height. Grid pattern is about 20 to 200 micrometer wide fingers which are normally 2 to 5 millimetres too much actually normally 1 to 2 millimetre apart. So, not fine. So, that is the end of the design of the solar cell. If there is any question I will take one or two quick questions and then we will move on to the solar PV technologies. And mainly when you are discussing solar PV technology it is a production of silicon how to produce because we are talking about silicon silicon silicon and we are also talking you know why there is a high cost and so on. So, the production of silicon will answer why silicon is expensive. Then we will discuss the silicon wafer technology and thin film. So, one or two quick questions there is a question from Salim. Government call you Salim. Good morning sir. Sir in the solar simulator why we have to use only hydrogen lamps. In solar simulator we use halogen lamp because the spectrum of the halogen lamp is quite matching with the spectrum of the sun quite matching it is not matching completely, but it matches quite well. And therefore, the performance that you get under the halogen lamp will be close to the performance what you get in a real practice. Real practice means outside this condition. Sir we are having another question. Sir during this in the solar panel is in the dark or condition the batteries full the batteries fully charged. Now will the solar panel draw the will draw the power from the battery or not? Will it act as a load? Solar panel will not act as a load it can act as a load if the battery is fully charged, but normally in each solar panel there is what is called the blocking diode. There is a we will discuss while discussing the module that when the design of solar panel there is a blocking diode which blocks the blocks the reverse current flow. And therefore, solar panel will not act as a load there is a protection there. Any other question on the design of the solar cell ask a specific question to the design of the solar cell? Fine Gandhi Nagar. Good morning sir. Can we use ITO along with the metal fingers on top of the solar cell to get more collection? Those who do not know ITO is the indium teen oxide or the transparent one of the transparent conductive oxide used in a thin film technology for a making a contact. Normally for crystalline silicon it is not advised to use IDO along with your metal contact because first of all ITOs are not transparent they will block the light and they are not their resistivity is also higher than the emitter itself. Remember our silicon emitter is doped tends for 18 tends for 19. So, therefore, because the resistivity is higher you cannot actually you will not have much advantage of using TCO or ITO in crystalline silicon solar cell. Kakinada last question quick. Sir, I ask please research is going on in the field of conducting polymer so that by removing the metal contacts we can put conducting polymers which are transparent and get more light efficiency can be increased for the solar cell conducting polymer. Yes. So, that they collect current from the instead of metallic one where they stop the light we can use conducting polymers. Yeah conducting polymers I do not think we will have the metallic property you know because we want very low series resistance. So, that is one problem. Second problem is our solar cell is going to live for 25 years, but I do not think your any polymer actually can carry the current for 25 years. So, we are looking more for a metallic properties. So, we will get a semi conducting property, but getting metallic property from polymers will be very difficult and using it for the 25 years is even much more difficult. So, let me stop here and let me stop here and go ahead with my lecture otherwise we will run the short of time as always we do. So, now let us discuss the production of silicon production of silicon. One reason why solar cells are expensive is because of the cost of silicon itself and you might have read it if not. Let me tell you that about 50 to 60 percent cost of 60 percent even up to 60 percent cost of the solar cell is nothing but a cost of a silicon itself. So, rest everything else is very small. So, converting silicon wafer into solar cell and then converting solar cell into module and we have detailed lecture about the cells and module performance. So, we will discuss that in detail, but the important point to take home right now is that about 60 percent 50 60 sometimes even 70 percent cost is the cost of material itself. So, therefore, it becomes very important to know why it is so, why silicon is so expensive. So, that we are going to look at very quickly the details about the silicon types of silicon, silicon manufacturing and so on. So, before we go further the current status of silicon I will tell you. First of all the prices of solar cell fluctuates or actually fluctuated in the past and that is because of the prices of silicon itself. So, one there was a prices going down and suddenly in 2003 and 2004 the prices started increasing and that happened because of the shortage of silicon in the market. The history is that microelectronics industry is very old industry right, IC industry is very old industry, silicon is the main material used for IC industry and when solar cell production started in 70s and 80s whatever the material that was coming or the leftover material or you can say even the waste material that was coming from the IC industry or being utilized for the solar cell industry right. So, let us say in 2000 about 2000 only 10 percent of silicon produced in the world was utilized for solar cell only 10 percent which means we were the PV industry was completely dependent on the electronics industry microelectronics industry, but when the production and the consumption of solar cell started in 2002, 3, 4 there is suddenly shortage of silicon. The microelectronics industry was not able to supply the silicon to PV industry and because microelectronics industry can always pay the higher price right because the chips are very small 1 centimeter square half a centimeter square. So, the cost of that material is not very high, but our solar cells are very big right, solar cells are 100, 200 centimeters for each and then you have to use many solar cells. So, solar cell is a very large area device and because of that the silicon consumption in solar cell is higher and therefore, the because of the shortage solar PV industry have to pay higher prices and the module prices started increasing. In this phase about 2003, 2004 lot of people have invested money in making silicon dedicated for solar cell. Many people have invested money for producing silicon dedicated for solar cell and now in 2008, 9, 10, 11 you see that prices are again going down in this prices are going down again because of the good supply of silicon. Still manufacturing silicon is not easy at all it is not easy at all and I will show you by the slides. First of all silicon is used in many other industries, but not in that pure form. So, aluminum alloy for example, alloying aluminum people use silicon and there are other usage of silicon itself. So, actual silicon production which is not very high pure is very high 5 million 5 million metric tons and who are the main producer of silicon. I will show you some other slides. So, it is mainly Hamlock, Wacker, REC, Tokuyama, MEMC, Mitsubishi these are the main player. Now, I want to actually focus your attention to this. So, this is the slide from 2005 look at the scale of this graph. The scale of this graph is 2004, 2006, 80000. This is kind of the capacity of the various manufacturers. So, this is the scale everybody can see. So, 2005 this was the scale. The scale was 2004, 2006, 80000 of various manufacturers in the world. Also note then where is the China? China is very small silicon production. Now, let us look at the this slide. This slide is actually 2010 slide. Look at the scale now. What is the scale? One order higher 10,000, 20,000, 40,000, 60,000, 70,000. So, first of all what has happened in the last couple of years is that huge capacity of silicon production has come. One order magnitude higher than other and because of that you can see that the silicon production has really crossed too much higher level and I will show you the number. Again just this is the give it an idea that when I made this slide in 2008 and 2009 the production estimated production in 2010 was 40,000 metric tons. So, that was the title of my slide when I made in 2008, 2009 40,000 metric tons by 2010 more, but actual production in 2010 of silicon was 209,000 metric tons. This is called polysilicon. Do not worry about it. We will come to what is polysilicon referred here, but it may actually exceed 350,000 tons of silicon by 2012. So, there is a huge production of silicon is happening now dedicated for solar cell which was not the case 10 years earlier. Because of all this, because of all this production volumes are increasing for the solar cell and the cost is going down. So, this dotted line is the cost and you can see the cost of the module were somewhere like 100 dollars per watt and now you can actually get less than 2 dollars 1.5 dollars. So, that is happening this is one of the tutorial problem that I have given you can do very easy problem that if you know the solar cell efficiency and if you know the sizes then you can estimate if you want to make a 10 gigawatt modules in 2010 how much silicon is required or actually current production of silicon PV module is 20 gigawatt. So, if you are making 20 gigawatt module using silicon, how much production of silicon should happen in the world? This calculations are very easy you can do it. I have given in the tutorial. Basically, what you do if you want to estimate the silicon consumption, very quickly let me tell you that silicon consumption. So, what you do is you find out the length width and thickness length with the thickness of the module of the silicon solar cell. This will give let us say volume in centimeter cube you multiply with the density of silicon gram per centimeter cube. So, you multiply your volume of the silicon used by density of silicon. Silicon density is 2.33 gram per centimeter cube. When you multiply this actually you know the mass of your one wafer in the gram. So, one wafer mass you can find out in terms of the gram. Normally typical length you can take anything it can be like 100 10 by 10 centimeter or 12.5 by 12.5 centimeter. Thickness I told you about 200 micron or more precise 180 micron. So, you can find out what is the gram what is the mass of one wafer? Now, this one wafer let us say one wafer your standard test condition is you have 1000 watt per meter square and your efficiency is 15 percent let us say. So, this much power is available to you you multiply by the area of your cell. So, you get how much power you get and this is a watt peak power. So, you will see that some gram of material is required per watt peak. This is finally, you want to find some gram of material is required per watt peak. So, you can find out if 1 megawatt is to be produced how many grams of materials available. 10 gigawatt is produced how many grams of material is required. So, this is small, but a very interesting simple problem that you should be. So, we should be able to estimate the silicon consumption require for 1 megawatt, 5 megawatt, 10 megawatt, 10 gigawatt, 20 gigawatt, 100 gigawatt whatever is the number. So, that you should do. How to obtain silicon is the main question that we are trying to understand here. First of all silicon comes in various types. So, there is a raw silicon which is available in the form of silicon oxide. There is a metallurgical grade silicon MGS, MGS is 98 to 99 percent pure. There is a solar grade silicon which is having purity of 99.999 percent and there is a electronic grade silicon which is having impurity of 99.9999999 percent. Many 9s are there 11 9s 10 to 11 9s are there that is the purity of silicon. By the way do you know any other material which is having this purity probably not. Do you know any other material which is this pure like electronic grade silicon and produced in the quantities of 200000 tons probably not. So, silicon is by the way the most purest material and purest and largest quantity producing the largest quantity on the silicon is the only material which is so pure and produced in so high quantity on the earth. So, this is electronic grade silicon. There is one more grade called detector grade silicon which is even more pure. So, your raw silicon your MGS silicon your SOG your EGS and your detector grade. So, these are the various levels are the terminologies for silicon. Now, there are lot of impurities that are there in silicon silicon oxide which is available in the raw and that you have to you know remove. So, aluminium boron, phosphorus, titanium, tungsten, nickel, gold, iron, molybdenum, lengthenum all this impurities must be removed. How much to remove? So, that we can get purity level of this. So, that we can get the purity level of this kind of purity level. If your material is for example, only 99 percent pure or even 99 percent 99.9 percent pure you will not be able to make your solar cell. Why you will not be able to make your solar cell? If your material is only 99.9 percent pure or even 99.99 percent pure because impurities will be so high that the recombination lifetime will be higher. I am sorry much lower and therefore, diffusion length will be much lower and you will not get efficiency. So, therefore, you cannot make you cannot use raw silicon you cannot use MGS silicon also to make your solar cell. You can use solar grade silicon or you can use electronic grade silicon. So, impurity level in the solar cell or in the silicon is presented either in form of percentage either in the form percentage or it is or it is given in the parts per some parts per x atoms of silicon. So, this parts is normally given at parts per million ppm or it is given as parts per billion ppb or it is given as a parts per trillion ppt. How to get this parts of impurities per x silicon atom is you get the density of silicon atom density of silicon atom in crystalline form is 5 times transfer 22 per centimeter cube and you divide the density of impurity atoms in per centimeter cube. So, when you do this you will find out how many for how many what is the impurity level for a given number of silicon atoms. So, when you say parts per million. So, we say how in 1 million atoms of silicon how many atoms of impurity that is parts per million. When we say parts per billion in 1 billion atoms of silicon how many impurity atoms of how many atoms of impurity are parts per trillion. So, in 1 trillion atoms of silicon how many atoms of impurity. So, typically we talk about parts per million parts per billion and you can actually do the calculation again I have given tutorial problem to solve it. So, this is the way you can find out. Now, the production steps for silicon is that you actually start with the raw silicon you get the metallurgical grade silicon from metallurgical grade silicon you get the electronic grade silicon which is normally referred as a poly silicon, but you have to convert this poly silicon into the silicon wafer you to convert poly silicon into silicon wafer. So, electronic grade silicon high purity silicon containing gases first you get and then you convert that high purity gases into the solid which is of high purity, but not crystalline. What is crystalline when atoms are arranged in order then we say crystalline. When you talk about poly silicon, poly silicon is a pure silicon high quality pure silicon, but it is not crystalline in the in the form. So, therefore, what you have to do you have to convert this solid poly silicon which is pure into solid silicon which is crystalline silicon wafer. This process is called production of silicon wafer. So, you first convert poly silicon into what is called silicon ingot ingots are big rods of silicon or cubes of silicon and then you cut those ingots into the wafer. So, we look into this process quickly. First of all we have talked about three types of material metallurgical grade electronic grade and SOG solar grade metallurgical grade electronic grade and solar grade. So, look at the what kind of impurities that is allowed and this impurities are given here in terms of parts per million. So, in 1 million silicon atom how many atoms of impurity is allowed when 1 million silicon atom how many if it is metallurgical grade silicon aluminum is allowed so and so, so much 1517 atoms of 1 is allowed of aluminum in 1 million silicon. But, if it is solar grade only 0.7 atoms are allowed or if it is electronic grade 0.001 0.001 atom in 1 million which means about 1 atom in 1 billion. So, the impurity level or the purity level of silicon as far as aluminum is concerned is parts per billion. Look at the gold AU even much lower 0.0007 parts per million. So, about 0.07 parts per billion or which means 70 only 70 atoms of gold is allowed per trillion atoms of silicon. So, from this numbers you can imagine the kind of purity that is required and to purify a silicon which is high melting point. Silicon has a melting point of about 1400 80 degree centigrade. So, considering the high melting point of silicon and considering the kind of purity level that you require it is very difficult and challenging task to manufacture silicon of high quality, manufacture silicon wafer of high quality and produce silicon solar cells. For that matter so far as of today in the country nobody, nobody, nobody makes silicon. There are some efforts which are going on to manufacture silicon, but right now nobody is doing. So, let us let us look now step by step how to make a metallurgical grade silicon. So, metallurgical grade silicon before we go further as you increase the purity level of silicon your cost also increases. So, this slide here what you show what I have shown here this is a impurity contained in terms of the parts and this is the cost. So, if you look at the cost of a electronic grade silicon which is here is about 70, 80, 100 dollar per kilogram. The cost of a detector grade silicon is even 200, 300 dollar. The cost of a solar grade silicon can be about 10, 20 dollar per kilogram. The cost of the metallurgical grade silicon is about 1, 2, 2 dollar. So, as you increase your purity level your cost also increases. How increases? Look it is a log scale. So, it increases exponentially almost. So, it is a very important factor. Today's term the cost of the silicon that is used for solar cell application is available at about 40 to 50 dollar per kilogram that is that is what is available. So, silicon is available in nature you can have the sand has containing silicon oxide there are sand stones there are clays are available which is containing silicon and you can actually get a silicon from here. So, first step is to get the metallurgical grade silicon from raw silicon that is silicon oxide. In order to get the metallurgical grade silicon where the price can be the metallurgical grade silicon can have price 1, 2, 3, 4 dollar per kilogram. Very simple thing to do is you take raw material that is silicon oxide fuse with the carbon. So, you actually do this process at 2000 degree centigrade I will show you the slide here. So, you actually take put your raw material to carbon electrode supply very high voltage. So, that you know this material get heated and when the melting when the melting occurs this liquid silicon gets settled on the bottom which is about 98 to 99 percent pure and this silicon can be extracted and this silicon is referred as a metallurgical grade silicon. So, simple take silicon oxide fuse with the carbon or burn with the carbon you will get silicon and the cost can be higher. Normally this is produced in the country where electricity is cheaper because ever energy requirement remember you are melting it almost about 2000 degree centigrade 1800 to 2000 degree centigrade because that much melting requires lot of energy and therefore, if your electricity cost is cheaper than this cost of metallurgical grade silicon will be cheaper. So, not in India. So, it is produced in the countries like Brazil, Canada, USA where electricity is cheaper. So, once you get your metallurgical grade silicon what you have to do now you have to get a purified further. Now, purifying the solid metallurgical grade silicon further to a level where impurities are only in the range of parts per million and parts per billion. It is not possible to do it in the liquid phase. So, what people have done is you have to convert into the gaseous phase. So, we can we convert metallurgical grade silicon into silicon gases containing silicon and this gases are actually having very low boiling point. So, the gases which are which we convert for purification is silicon containing gases. So, we have a silane these are the silicon containing gases which are having low boiling point and can be purified by distillation. Silicon n can be purified by the distillation. So, there is silane there is a Si HCl3 tri-chlorosilane there is a Si H2Cl2 and there is also Si Cl4. So, these are the various gases which people use for purification of silicon. So, you get this gases for example, silane this is called silane and your tri-chlorosilane it is called TCS tri-chlorosilane this is called the DCS di-chlorosilane and this is tetrachlorosilane. So, normally this silane is used for purification you get a silane as a pure gas because you can do the distillation at low temperature, but what we want we do not want silane gas we want silicon in the solid form right. So, then there is a another state one thing also wanted to know tell you is when you want to make your amorphous silicon when you want to make your amorphous silicon people use silane. So, actually you can get this purified gas and directly do the deposition of silicon amorphous silicon. So, various impurities that you have to get rid of and what again you do is you have to convert the silicon into silicon containing gases. So, Si Cl4 then your Si H4 silane. So, various reaction that takes place in presence of HCl. So, you treat with HCl and you get your silicon containing gases and this gases as I told I am telling you that because of the lower boiling point can be purified by distillation and you get silicon containing. So, once you get high purity silicon containing gases what you have to do you have to get the pure solid right eventually want solid. So, you have to convert this gas pure gas into pure silicon. Once you get pure silicon, but this polysilicon is actually a not crystalline this polysilicon is not crystalline. So, then you have to convert into crystalline form. Once you convert into crystalline form then you have to cut it in the form of silicon wafer. So, then it cutting wafer. So, this is the various state. So, I will show you this I will show you the states. So, one is already that you know that it is raw silicon. Next step is to get a metallurgical grade silicon. Next step is to get a pure silicon gases or silicon containing gases next step is get a pure polysilicon. This is the polysilicon is high quality, but this is not this is not crystalline. This is not crystalline then you have to convert to crystal. So, then you have to get silicon ingot this is crystalline. What kind of crystal? What kind of crystal? Either you get a mono crystalline or you get a multi crystalline, but eventually what you want you want a wafer you want silicon wafer. So, then you have to do the dicing then you get the silicon wafer. So, this is the whole root of silicon from raw silicon to silicon wafer and then you make solar cells. I hope it is clear metallurgical grade silicon is only 98 to 99 percent pure. Pure silicon containing gases these are electronic grade silicon gases and then you polysilicon, but not crystalline then you convert into the crystalline form and then wafer. So, I hope now you are starting to realize that how difficult it is to make silicon. There are so many processors required then all these processors are high temperature processors and because this processor high temperature processors lot of energy goes into making silicon and because lot of energy goes in making silicon the cost of silicon that eventually get for making your wafer is higher and because the cost is higher the cost of solar cells is higher cost of solar module is higher and cost of electricity is higher and that is why the whole worldwide effort is actually to minimize the cost of silicon or can we make can are there are other processors where you spend less energy in making silicon is it possible that we use more impure material. So, not electronic grade silicon, but we use solar grade silicon and without compromising with the efficiency. In fact, there are some people which is who are working on I told you that metallurgical grade silicon cannot be used for making solar cell, but there are some people who are trying to upgrade metallurgical grade silicon and it is called upgraded metallurgical grade silicon. So, can we use upgraded metallurgical grade silicon to make. So, let me come to the next slides. So, these are the this is the root will quickly go through it. So, when you when you want to convert pure silicon into. So, when we want to convert pure silicon into pure silicon containing gases into solid silicon there are two ways of doing it one is called Siemens reactor and other is called fluidized bed reactor. Pure silicon gases to solid silicon there are two roots Siemens reactor and fluidized bed reactor then pure solid silicon to crystalline silicon again there are two roots one process is called CZ process and other is called FZ process. And again from crystalline silicon to silicon wafer there are two roots inner diameter sawing. So, you have to saw it and the wire sawing. So, these are the two roots each two roots for different pure silicon gas to solid silicon solid silicon to crystalline silicon crystalline silicon to silicon wafer. So, this is the Siemens reactor basically you have the rod where the you put your gases here silicon containing gases like SISCl3 is here and you again provide high temperature this gas gets broken up like here as shown here SISCl3 with hydrogen it will react and you will get a silicon. The silicon gets deposited in the form on this rod here I can see the picture here you can see the picture here that eventually this gets deposited here on the on this rod green line and actually you can get a stone of silicon like this which is pure in the form. So, one is Siemens reactor other is called fluidized bed reactor in the Siemens reactor the problem is that the silicon gets in contact with the electrode on which the deposition is taking place. And because of this contact some of the impurity of the electrode is transferred now our purity levels are so high parts per billion parts per million that even any contact with the any other surface can be a problem which is the case for Siemens reactor. So, therefore, people have come up with a fluidized bed reactor where silicon gas when you put like silane you are putting the conversion of the gas into solid silicon happens in the air it is not in contact with any other surface. So, this particles are actually floating silicon keeps get deposited on the particle and when it is heavy enough it just fall down. So, it is called fluidized bed reactor the bad is fluid it is not fixed and because now silicon is not making contact with any other surface then there are two so once you get the solid silicon you have to convert into crystalline silicon. So, definitely you have to melt it right you have to press so you have to melt it. So, there is a process called Chokralsky process it is the name given by the scientist after the scientist. So, you actually melt you put your silicon in a crucible as shown here you put your silicon in a crucible this again you melt in high temperature and you take seed silicon crystal you put in contact with the liquid and you keep rotating it so the atoms which gets deposited to the seed. So, seed is already a p type or n type and it is already monochrist line very important thing to note here is the doping level when we discussed for the solar cell the doping level is actually determined at this point. So, here in this case you put your impurities right. So, you put your boron you put your phosphorus and the desired amount right. So, if you want doping of 10. So, 14 atoms per centimeter cube or 15 or 16 you decide the doping level at this point. So, together with the pure silicon you also put the dopants in it. So, Chokralsky process is one of the most commonly used process for making silicon wafers. So, almost 70 80 percent of the material produced in the world I mean the crystalline silicon material produced in the world is coming from C Z process Chokralsky. So, this is what steps happens you put your stone here pure silicon stone it melts you put your seed crystal then you start pulling. So, slowly the ingot slowly the ingot comes up. So, here the silicon was non crystalline or you can say polycrystalline and that is why we call it as polysilicon. Here silicon is a crystalline form and typically when you have the circular shape this is the format when you get always the monochristalline silicon your monochristalline is always circular keep this point in mind because we will use this information. So, your monochristalline your monochristalline ingot monochristalline silicon ingot is circle always it is always circle it is form of ingot and cylindrical. Then you have your float zone silicon float zone silicon again the problem with the mono this kind of C Z process is that silicon comes in the contact with the other surface and impurity level can be higher. So, another way of doing it is floating the melting a particular zone in without any contact, but first of all you have to make the ingot and then you have to make another process to melt it. So, you get actually even higher purity. So, float zone silicon F Z silicon is having higher purity as compared to C Z silicon, but actually you are doing that process twice and because of that it is more expensive float zone silicon is more expensive as compared to C Z silicon, but because of its high purity many times the high efficiency records or the solar cell with the high efficiency are actually manufactured using F Z silicon, but all commercial solar cell people use C Z silicon. So, that was the case for the mono crystalline silicon for multi crystalline silicon it is done in a bit of bit of faster way. So, you actually take one crucible like this you take one crucible like this you put your silicon you melt it and then you solidify you solidify from one direction. So, in this case the whole chunk it can be 1 meter by 1 meter by 1 meter. So, whole chunk of silicon can be melted and solidified together. So, this much faster, but the problem is you do not get mono crystalline you get multi crystalline. So, there are different grain boundaries that you get here, but multi crystalline is easier to make means cheaper also. So, the multi crystalline silicon is not circle it is normally rectangular or square. So, this is how you get the multi crystalline silicon and in order to cut this silicon ingots into silicon wafer there are two techniques inner diameter and wire sawing inner diameter you have actually a thin sheet of metal which is coated with the diamond and that goes over the silicon ingot and cut it or you have the wire which is coated with the diamond. So, wire sawing. So, one of the problem is that lot of material is wasted you see almost 50 percent material is wasted in the inner diameter sawing. This loss of the material is referred as a kerf loss right when you also saw our wood whatever the wood that goes out is actually a kerf loss. Similar to that whenever you saw your silicon wafer from ingot you have to also lose some of the material because of this mechanical process and this process remember we have actually put so much of effort to make the silicon and now 50 percent of silicon is wasted because of the kerf loss and 30 percent is wasted in wire sawing because of the kerf loss. So, that is a very significant loss eventually at the end you will get this kind of wafer. So, when you get the circular wafer you should understand that it is monochrist line, but when you put circular wafer the packing density will not be very high in the module and therefore, sides are cut normally. So, we get what is called the pseudo square. So, this pseudo square you can see, but always the rounded corner then you will as soon as you see the solar cell with a rounded corner you should be able to find out that it is a monochrist line, but if you look at the multicrist line you can definitely see the different grains as shown in this photograph this is a photo of multicrist line. So, different patches are actually different crystals as shown here different patches are different crystals and normally you can see now whereas, the manufacturing is so nice that actually you cannot see the different grains. So, if you look at your solar modules try to see the experiment when you are doing one of the module will have will be of the monochrist line type and other module will be of multicrist line type. So, we have given you two modules of mono and two modules of multi try to see today when you do the experiment which modules are mono and which modules are multicrist line. So, this is the whole process as I have shown you earlier also that you start from mining get metallurgical grade silicon you get gases polysilicon then you get the ingot and you dice it then your solar cell and then wafer. So, it is a very long process and it takes lot of energy to do that and that is why silicon is expensive that look at the phases that silicon goes through it goes from solid then liquid then solid then gas then gas then solid then liquid then solid and eventually silicon wafer. So, silicon actually goes through lot of processes and because of that silicon is expensive. So, I hope that after going through this lecture you will realize that why silicon is expensive and why silicon solar cells are expensive never the less huge progress has been made and the cost of silicon based module which was 100 dollars per watt some 20, 30 years ago has brought down to less than 2 dollar per watt or almost 1 dollar per watt. So, there is lot of progress has been made both in terms of the silicon manufacturing and solar cell efficiency. My next lecture will be about how to use this silicon wafer to fabricate solar cell. Now, we are using a silicon wafer how to use silicon wafer to fabricate solar cell. So, the summary of silicon manufacturing is that you get first of all silicon of various type raw silicon metallurgical grade silicon solar grade silicon electronic grade silicon detector grade silicon and the purity levels can be different and the cost of manufacturing can also be different right the cost of manufacturing can also be very different. I have showed you how to go from raw silicon to metallurgical grade silicon metallurgical silicon to gases, gases to poly silicon which is pure silicon but not crystalline and then you go to crystalline silicon in the form of ingot and then you cut this ingot into wafer. So, that is the whole process and why it is expensive that we have discussed. Question from Vijaywada. Sir, in ingot we are having some flatness, why we are having such flatness while preparing the ingot or wafer, wafer is not round it is having some flat surfaces certain area. Why we should have some flat areas? So, in wafer, so what happens in this when you have the wafer, when you have the silicon wafer like this. So, when you are going to put in the module, in the module you put many solar cell. So, this solar cell will be actually your active area if I make solar cell like this. So, this solar cell will be your active area but the area in between like here will not be active. So, you are not utilizing your module area properly. So, therefore, what we call is this pecking density. What is the area of the solar cell? What is the percentage of cell area in the module? So, if you are using circular then you will find that the best pecking density you can get is only about 70 percent. So, your 30 percent area will not be active. So, therefore, what you do is you cut from the sides, you cut this from the sides like this. When you cut from the sides, so your wafer, you shape of the wafer becomes like this. So, then you call the pseudo square and your monocrystalline silicon is always circular. When you produce it, because of this circular motion that you have to do in CZ process, it is always circular. So, if you want good pecking density, you have to actually get what is called this kind of process, pseudo square. Sudo square shape and now when you put a pseudo square shape, you can actually make solar cell closer to each other. When you put solar cell closer to each other, your pecking density increases and that is what you can see also. Is that your question answered? Sir, one more thing is regarding monocrystalline and multicrystalline. In monocrystalline itself, on all the phases, we are not having equal mobility. Mobility varies from phase to phase. That is, in 1-1-1 plane, we will think that mobility is high. How can we think of mobility regarding multicrystalline crystals? So, in mobility, monocrystalline silicon varies from one direction to other direction. In multicrystalline silicon also, it will be kind of you have to take average mobility. So, then it is not possible to define mobility based on the direction. So, you cannot say mobility in 1-0-0-1-1-1 because there is no such thing because all the grains are very different. But then you have to, you can always say the average mobility, which will be the average of all. Nirunjali, last question, one question only. Nirunjali. Sir, which type is more efficient and why? Which type of silicon is more efficient and why? I will not answer that question. You should know that answer. You know that monocrystalline silicon is a defect free and multicrystalline silicon because of the different orientation, there are many defects, there are many grain boundaries and therefore, there are defects in multicrystalline silicon, but there are no defects in monocrystalline silicon. So, that hint should be enough to tell you that which one should having, which one should have more efficiency.