University of California at Santa Cruz Jack Baskin School of Engineering Electrical Engineering Department EE-145L: Properties of Materials Laboratory Lab 6: Solar Cells Fall 2004 Dawn Hettelsater, Yan Zhang and Ali Shakouri, 05/09/2002 1.0 References Sections 6.1-6.3 of S.O. Kasap. Principles of Electrical Engineering Materials and Devices. 2 nd Ed. 2002. The Semiconductor applet Service List of Simulation Applets http://jas.eng.buffalo.edu/ #4 PN Junction Diode at Equilibrium #5 PN Junction Diode under Applied Bias Voltage Includes interactive applets on: Semiconductor statistics PN junction diode (In equilibrium, Applied bias, Fabrication) Also includes parameters, mathematical analysis, and detailed explanations. Solar Electricity How Solar Cells Work http://www.soton.ac.uk/~solar/intro/tech0.htm Includes solar cell simulations and general theory Institute of Electrical Power Engineering - Renewable Energy Sect. http://emsolar.ee.tu-berlin.de/~ilse/solar/index_e.html Includes theory as well as interactive components Center for Photovoltaic Engineering, University of New South Wales, Australia http://www.pv.unsw.edu.au/research/usefulinfo.asp Useful tables, current research in photovoltaics. 2.0 Theory 2.1 Formation of the pn Junction A pn junction is formed when a p-type semiconductor material is brought into contact with an n-type material. The p side has an excess number of holes, whereas the n side has an excess number of electrons. When the two materials come into contact a concentration gradient is formed on each side due to the excess charges. The holes begin to diffuse toward areas of low concentration in the n side, and the electrons diffuse to the p side. Eventually, a region in the middle becomes devoid of any electrons or holes 1
because they have diffused to the opposite sides. Within the middle region are the ionized acceptor atoms in the p side and donors in the n type. This forms a negative charge on one side and a positive charge on the other, creating an electric field between. The built in equilibrium potential is given by V 0. Figure 1 below illustrates the pn junction. Figure 1 The middle region containing the electric field is referred to as the space charge region, or the depletion region. Due to the electric field, electrons and holes experience a drift current. Equilibrium is reached when the drift current exactly opposed the diffusion current. The net current flow is therefore zero. The total current density for either electrons or holes is given by: J = J + J (1) ntotol ptotol ndrift J = J + The current densities are therefore given by: J J p n pdrift J ndiff pdiff dn( x) = qnµ ne + qdn (2) dx dn( x) = qpµ pe qd p dx Where n and p are the electron and hole concentrations respectively, µ is the drift mobility, E is the electric field, and D n,p are the diffusion coefficients. 2.2 Forward Bias When a positive voltage is applied to the p-type material there is a net current flow in the positive direction. Figure 2 below demonstrates the pn junction under forward bias: 2
Figure 2 The applied voltage, V, results in an electric field that opposes the internal field across the space charge region. This effectively lowers the potential barrier by the amount V-V 0. In equilibrium, some of the majority carriers in the conduction band have a high enough energy to surmount the barrier. Under forward bias, the barrier is lowered, increasing the probability for majority charge carriers to diffuse across by a factor of e (qv/kt). As the majority charge carriers cross the junction they are injected into the other type of material, becoming minority charge carriers. This is referred to as minority carrier injection. Unlike the diffusion current, the drift current remains fairly insensitive to the applied bias. This current is caused by minority carriers wandering toward the barrier and then being swept away by the field. Comparatively, there are very few minority carries compared to the doped regions, therefore very few charges contributing to drift current. The total current is given by adding the diffusion and drift currents, with the diffusion current dominating. During equilibrium diffusion current is equal in magnitude to the absolute value of the drift current. Under forward bias, the diffusion current can be given by: qv kt I diff = I drift e (3) The total current is then the diffusion current minus the absolute value of the drift current, since the drift current is in the opposite direction. The current is therefore given by: 3
qv kt I = I 0 e 1 (4) Where I 0 is the absolute value of the drift current. This equation is referred to as the ideal diode equation. Using equation 2 and solving for the minority injection currents leads to a more accurate diode equation given by: D qv p D n kt I = qa p + 1 n n p e (5) L p Ln Where L p,n are the diffusion lengths of the holes and electrons, p n and n p are the minority charge concentrations, V is the applied voltage, and A is the area of the device. 2.3 Solar Cells 2.3.1 Generation of photocurrent With semiconductors, it is possible to convert the energy of the solar radiation into electric energy. Theoretically, each absorbed light quantum (photon) could generate an electron-hole pair. If the energy of the photon surpasses the band gap, then such a generation takes place. The surplus energy is converted into heat. In order to generate more than one electron-hole pair(ehp), the photon must provide a multiple of the energy of the band gap. Figure 3 Generation of electron-hole pairs 4
Figure 4 Generation of photocurrent in a diode If the minority carrier created from the EHP diffuses toward the junction, it will be swept to the opposite side by the internal field. Due to the larger number of charges crossing the junction, the drift current increases. The charges then begin to build up on the other side, holes on the p side and electrons on the n-side, making the p-side positive and the n-side negative. This resembles the forward bias and creates the same diode current from equations 4 and 5. The excess majority charge carriers begin diffusing away from the junction. This creates diffusion current, called the photogeneration current (I PH ) that opposes the diode current. 2.4 Important characteristics of solar cells Two important characteristics of a solar cell are its open circuit voltage (V OC ) and short circuit current (I SC ). Under short circuit conditions, the charges are free to travel through the circuit and no build up bias is produced. The photogeneration current assumes a maximum since there is no opposing diode current. Because the minority carriers are produced from the EHP s, the photogeneration current is directly proportional to the intensity of the sunlight. Under direct sun conditions, the I SC is at its maximum. Under open circuit conditions there is a charge buildup on each side creating a diode current. The device reaches equilibrium when the diode current is equal and opposite to the photogeneration current. In order for the diode current, otherwise referred to as the internal or injection current, the internal potential barrier is lowered. This further demonstrates the forward bias characteristics. Figure 5 below demonstrates the short circuit and open circuit conditions. 5
Figure 5 Short Circuit Current Figure 6 Open Circuit Voltage Under illumination, the total current is given by the diode current minus the photogeneration current: I qv kt ( e ) I Ph = I 1 0 (6) Where I Ph is the photogeneration current. The typical current for a solar cell is around 1mA and the typical voltage for a silicon cell is about 0.5V. Figure 6 below shows the IV curve for a cell. 6
Figure 7 Berlin University of Technology, Institute of Electrical Power engineering, Renewable Energy Section The power output of the cell is the product of the current and voltage. A measure of the quality of a cell is given by the fill factor (ff). The fill factor is defined as the ratio of the maximum power to the product of the open circuit voltage and the short circuit current: I V M M ff = (7) I SCV OC Typical values of the fill factor range from 0.75-0.85. If the IV curve were in the shape of a rectangle, the fill factor would be 1. The efficiency is a measure of the maximum power over the input power. P out η = (8) Pin Where P out is the maximum power point and P in is the power due to the photons incident on the cell, given by: ( hc λ) dλ E = N (9) ph The value of E with the sun perpendicular to a solar cell s surface is referred to as the solar constant. Its average value is approximately 1kW/m 2. The efficiency can then be rewritten as: ff I SCV E A OC η = (10) 7
2.5 Factors Affecting Efficiency Typical values for solar cell efficiencies are 10-15% for thin film cells, 15-20% for crystalline silicon cells, and 30% or more for concentrating systems (focus sunlight onto small area sun, up to 1000 sun concentration). The best theoretical values for efficiencies are 20-28% for normal cells. The reason for this low value is simply that not all of the energy reaching a solar cell from sunlight can be converted into electricity. About 25% of incoming photons have energies below the bandgap energy and cannot produce an EHP. About 30% of the photons will have too much energy and will either be re-emitted or wasted as heat. This accounts for a total of 55% of the energy that can t be used. Of the ~75% of absorbed photons, about 43% of the energy from absorbed photons is lost as heat. In addition to this, electrons can be lost due to recombination with in the semiconductor material. The extent to which this is a problem depends on the type and purity of the material. Without treating the cell, about 30% of incoming photons can be reflected off the surface. For this reason the surface is texturized in the shape of pyramids, maximizing the chance that a photon is reflected back into the cell. Antireflection (AR) coatings are also applied. The combination of both can result in reflection losses of less than 1%. Another problem is the natural resistance to electron flow. Large metal contacts on the surface of the cell can minimize this, but will block the incoming light. The contacts are therefore designed as a grid with conducting fingers. Research is also being applied to creating back only contacts as well as transparent contacts. Temperature can also greatly affect a solar cell s efficiency. The warmer the cell, the less it behaves as a semiconductor and the efficiency falls. 2.6 Types of Solar Cells (Reference Reading) Types of solar cells fall into three categories. Table 1 below shows the different types of cells. Crystalline Thin Film Other Crystalline Silicon (c-si) Gallium Arsenide (GaAs) and Alloys Amorphous Silicon (a-si) Thin film Silicon Copper Indium Diselenide (CIS) Cadmium Telluride (CdTe) Quantum dot solar cells Dye sensitized photochemical cells Polymer cells Photoelectrochemical Cells Crystalline silicon currently makes up about 86% of the photovoltaic market. The reason for this dominance is that the material, technology, and equipment come right out the electronics industry. Whatever is wasted is used in the PV industry. The 8
development of c-si cells is very energy intensive. It is therefore very expensive to process these cells and the technology is leaning toward the production of polycrystalline Si cells and thin film technologies. Polycrystalline silicon cells use less energy to produce. Molten Si is allowed to solidify under specific conditions. The solidified Si is then sliced into rectangles and then individual square cells. This process eliminates the time and energy intensive step of growing a single ingot and then slicing wafers. The end product leaves small crystalline silicon areas separated by grain boundaries. The grain boundaries decrease the efficiency of the cell. However, the benefit of lower energy consumption and cost make up for this loss. Gallium arsenide can be alloyed with indium, phosphorous, and aluminum to produce multijunction cells with very high efficiencies. In forming multiple junctions with decreasing bandgap energies, the incoming photons can be sifted through with the longer wavelength photons being absorbed at the bottom. Currently two junction devices are used for spacecraft with GaInP as the top layer and GaAs as the bottom. Research is being conducted to make a four-junction device boosting its efficiency to more than 40%. Thin film semiconductors are only a few microns thick and therefore use much less material than their crystalline counterparts. These materials are cheaper to manufacture and likely to lead solar energy into a competitive market. Thin films are made by depositing the semiconductor material directly onto a low cost substrate. Amorphous silicon makes up most of the remaining 14% of the PV market. Stable modules have efficiencies of 6-9%. The minimal material used and therefore the inexpensive price of modules account for this low efficiency. The p and n regions are made very thin with a thicker intrinsic layer between in order to lengthen the space charge region. To maximize light absorption and minimize recombination, the layers need to be thinner than that needed to absorb the light. Several layers are therefore stacked on top of each other. Germanium is added to each successive layer in order to decrease the band gap energy and therefore absorb wavelengths previously unabsorbed. Cadmium telluride is a newer thin film technology with immature manufacturing steps. With time it is thought to be the most promising thin film to meet the cost goals needed for PV to be a competitive market. Laboratory cell efficiencies are around 16% with module efficiencies between 6-9%. Some benefits to CdTe are its high absorption coefficient, minimal amount of material, only 1µm, and the 12 or more manufacturing steps that can be used to make the modules. CIS and its alloys is also a promising thin film material with laboratory efficiencies of 18% and module efficiencies greater than 11%. This product is currently on the market and boasts 20+ years of research and development. Some problems include immature manufacturing step, slow vacuum steps, and a more complex structure than the other thin films. Another type of thin film is thin film crystalline silicon in which the inexpensive amorphous silicon is combined with the more efficient crystalline silicon. This is a new technology that is in the experimental stages. Among the other category include quantum dot solar cells in which a nanocyrstalline CdSe semiconductor is embedded in the conductive polymer/c60 composite. This has the potential for low-cost, large-area production. Dye-sensitized photochemical cells have a dye sensitizer that absorbs light and generates EHP s in a 9
nanocrystalline titanium dioxide semiconductor layer. Only certain wavelengths can be absorbed but because the device is clear, research is being conducted to create a clear window that will absorb and convert UV light into energy. 3.0 Experimental Procedure 3.1 Overview In this experiment we will plot the IV curve for a solar cell, using the open circuit voltage and short circuit current for the limits. Data will be taken along the IV curve using a variable resistor. Next perform a linear regression line and determine the maximum power point. 3.2 Questions to answer before starting the lab: Read through the first three references above. Next go to the last reference from the Institute of Electrical Power Engineering, Berlin. Run through the simulations on series and parallel resistance, and temperature effects. What are your observations? Will the current be positive or negative? What values of current and voltage should you expect to measure? What values will you use for the photocurrent and the voltage from the diode equation? How will you determine the fill factor and the efficiency? What are the factors to affect Isc and Voc, how about temperature, is it a factor? 3.3 Equipment Solar cell Multimeters Variable resister Light source Light intensity meter 3.4 Schematic setup Figure 8 below gives the schematic setup for measuring the IV curve for a solar cell. Figure 8 10
3.5 Procedures 1. Set up the circuit from figure 8. 2. Set up the light source directly above the circuit. 3. Set the resistance to zero in order to measure the short circuit current. Record the current and the voltage. 4. Turn off the light and observe the effect of decreased illumination on the short circuit current. 5. Increase resistance until current is very close to zero. This corresponds to the open circuit voltage. Record the current and voltage. 6. Turn off the light and observe the effect of decreased illumination on the open circuit voltage. 7. Vary the resistance and take a few current and voltage measurements along the curve. a. To get a good fit of the I-V curve, you need to measure most of your points near the elbow of the I-V curve; b. The strategy here is to start with the short circuit condition, and increase resistance until the current falls a few percent, take a measurement, then repeat. c. When the current drops by 30%, then it passed the elbow of the I-V curve and only need a few more measurements 3.6 Questions to answer after completing the lab 1. Calculate the fill factor. 2. Calculate the efficiency of the cell. Does this seem reasonable? Why would your value be incorrect? (What is the E value used?) 3. Why is the voltage not equal to zero when measuring I SC? 11