CC BY
19
Olimov Shoirbek Abduqaxxorovich, senior researcher, North China Electric Power University, Beijing Kasimahunova Anarxan Mamasadikovna, doctor, of technical sciences, professor Mamadalieva Lola Kamildjanovna, doctor of philosophy (PhD), of technical sciences Nurdinova Roziyaxon, Zokirov Sanjar Ikromjon og'li, Norbutaev Maqsud Abdurasulovich, senior researchers, Fergana Polytechnic Institute, Republic of Uzbekistan E-mail: [email protected]
DEVELOPMENT AND RESEARCH OF HETEROSTRUCTURES WITH AN INTERNAL THIN LAYER BASED ON P-TYPE SILICON
Abstract: The article presents the manufacturing technology of a heterojunction solar cell and the results of the study of its current-voltage characteristics.
Keywords: solar cells, heterojunction, technology, efficiency, characteristics.
The development of research on the development of high-efficiency solar elements based on silicon, is still gaining temp. The main goal of such research is to obtain a cost-effective and highly efficient solar cell. Among them, silicon solar cells with a more perfect heterostructure are distinguished. The best values for the conversion factor are given in [1] and [2]. In [1], analyzing the low rate of the recombination process of heterojunction solar cells made of silicon compared to direct-gap semiconductors, it was shown that at relatively low concentrations of main charge carriers (Nd~ 1015cm-3), the excess charge carriers can be comparable or higher Nd. In this case, the magnitude of the efficiency n (efficiency factor) does not depend on Nd. At higher Nd values, the n(Nd.) dependence defines two opposing trends. One of them contributes to an increase in n with increasing Nd, and the other, associated with Auger recombination, leads to the decreasing n. The authors of this work determined the optimal value Nd = 2*1016 cm3, at which the value n of such an element is maximum. It is shown that the maximum value of n on 1.5-2% higher than the value of n at 1015cm-3.
In the work [2] we can see that the creation of HIT elements (heterojunction with intrinsic then-layer solar cells) with an efficiency of 25.6% in the terms AM1.5. If we take into account the opinions of the authors [3], then the maximum theoretical value of the conversion efficiency of silicon solar cells at concentrated solar radiation can reach up to 30%. However, in practice, there are opportunities for the successful conversion of light by concentrating and achieving the highest efficiency values. In addition, research is still continuing on
methods for improving the technology of manufacturing silicon solar cells by changing its structural state and production methods. This article is devoted to finding the most efficient heterostructural solar cells made of silicon.
Formulation of the task
In connection with the presence of some unexplored sides of silicon photoconverters, it is considered effectually to investigate the possibilities of creating highly efficient solar cells based on it. The study of foreign literature shows the absence of highly efficient silicon solar cells with a base layer of p-type. This is of interest to the study of methods of manufacturing technology and the creation of a new construction of the latter. In our opinion, the most optimal methods for producing a highly efficient photoelectric converter with a base type of p-type is, firstly, the selection of the sputtering method on a vacuum-magnetron machine and, secondly, the application of fast heat treatment (RTR-Rapid Thermal Processing). In addition, we were faced with the task of finding the reasons for contesting the possibilities of obtaining highly efficient conversion of the energy of sunlight using silicon heterojunction solar cells with a base p-layer.
Production process of solar elements
A silicon semiconductor, which was packed inside a sealed foil, was taken as the starting material. Before cutting, the starting material was thoroughly washed in ethanol solution of ethanol CH3CH2OH.
Cutting the silicon wafers is made by the glass cutter without allowing the formation of any damage: cracks, lines and scratches.
Section 10. Technical sciences
The sample from which it intended to produce an effective heterojunction solar cell, after cutting, had the following geometrical dimensions: S = 4 ± 0.3 cm2 and a thickness of 420 ± 10 ^m. After the sample was prepared for sputtering, the sputtering process was carried out in a vacuum-magnetron machine (Fig. 1).
Figure 1. General view of the vacuum-magnetron machine
This operation had a duration of 3 hours. At the moment of spraying, the angle of contact of the dust-like layer with the plate surface is changed continuously. This was carried out with the help of rotating the rotary mechanism of the valve. At the end of the spraying process, after closing the target cover and turning off the heater, a break is taken to cool the sample to a temperature of ~ 250 °C. The heat treatment of the plates was carried out in a fast way. This stage involves heating the plate to 10000C for a few seconds. Heat treatment is performed parallel to the diffusion process. Moreover, in this case artificial oxidation of silicon is performed. The sequence of the method of obtaining silicon films oxide was as follows. Firstly, artificial oxidation of silicon was carried out in oxidizing stove, similar to those used
Table
for diffusion in a flow of carrier gas at a high temperature (1000-12000°C).
The basis of such stoves is, as in diffusion, a quartz tube with silicon plates heated either by high-frequency currents or by other means. An oxidizing agent was used for oxygen, and sometimes water vapor at atmospheric pressure. The working temperature in our technologies did not exceed 8000 °C. In the process of heat treatment, solid planar phosphorus sources at heating, release phosphorus pentoxide (P2O5) into the gas phase, the molecules of which diffuse to the surface of the silicon wafers and as a result of reaction
2P2O5 + 5Si 5SiO2 + 4P
form a layer of phosphorus silicate glass (FSS), from which phosphorus diffuses into the bulk of silicon. Phosphorus nitride, silicon phosphide or materials containing P2O5 in a bound form, which is released during thermal decomposition (aluminum metaphosphate, silicon pyrophosphate), is used as TPI of phosphorus. Table 1 shows the heat treatment time and the maximum operating temperatures.
Then, thick silver solutions were deposited on the surface of the sample and the sample was exposed to drying in the oven. It should be noted that the necessary action to verify the results of sputtering and diffusion is to prepare samples for grinding. For this, silicon carbide (SiC) was used. Silicon carbide is a popular abrasive because of its durability and low cost. In the manufacturing industry, due to its high hardness, it is used in abrasive processing in processes such as grinding and honing.
For grinding work, a cylindrical metal 0 50 mm and a metal sleeve for this diameter were taken. It also used heavy-duty glass with a smooth surface. The grinding process begins with the preparation of a solution of silicon carbide using water on the surface of the glass. About 50 grams of silicon carbide is taken and spilled onto the glass surface to the desired thick mass. After this, a prepared sample of 1.0 cm2 in size is placed on the inside of the cylindrical metal with the aid of a sleeve. Grinding was performed by slow rotation in the style of the eight or by the zigzag method. 1.
Step-1 Step-2 Step-3 Step-4 Step-5 Step-6
Time/temp. Time/temp. Time/temp. Time/temp. Time/temp. Time/temp.
Example-1 20s-350 30s-400 45s-450 20s-700 45s-750 15s-800
Example-2 30s-300 30s-350 45s-450 20s-600 45s-700 15s-800
Example-3 20s-250 30s-350 45s-450 20s-650 45s-750 15s-800
Example-4 20s-200 30s-300 45s-400 20s-500 45s-700 15s-800
The Results of the research
The resulting samples were consistently exposed to testing parameters. Parameters of the p and n-junction, current-voltage characteristics and finally the efficiency of the element were checked. Our results differ from the data given in foreign literature.
A general view of the volt-ampere characteristics of the measured sample is shown in Figure 4, which clearly shows the main trends in the output parameters of the current-voltage characteristic with temperature. With the increasing of the temperature, the open-circuit voltage monotonously is decreas-
ing, the short-circuit current slightly is increasing. Processing current-voltage characteristics allows to determine the nature of the changes in the main energy parameters of solar cells with temperature. The voltage ofno-load stroke varies almost linearly with temperature in a wide temperature range and is characterized by a slope from -1.84 mV / °C at low temperatures to -2.11 mV / °C at temperatures from +20 to + 60 °C. The temperature dependence of the no-load voltage for the measured samples shows that Uxx is well reproduced and at this temperature is the same for all samples with an accuracy of up to ± 5 mV.
Figure 4. Volt - ampere characteristics of a silicon photoconverter at P_ p = 0,139 V/sm2
The same can be said about the temperature coefficient of voltage, which is -1.81 ± 0.08 mV / °C in the region of negative temperatures, and -1.9 ± 0.05 mV / °C in the area near zero with a zone of 20 °C It is interesting to note that the value of the no-load voltage is somewhat higher, and the absolute value of the temperature coefficient is somewhat lower than the values given in the literature [5]. This difference, apparently, can be explained by the presence of a wide-band layer.
The voltage at which the maximum power is reached under load is 120-150 mV lower than the no-load voltage at all temperatures at which the measurements were made, and the absolute value of the temperature coefficient is approximately -2 ± 0.2 mV / °C. The larger variation of this value is due to the fact that the optimal voltage is determined with a large error than Un v. The difference Unh - Uopt for this sample is almost independent of temperature and varies from sample to sample, depending on
the shape of the current-voltage characteristic of a given solar cell. A weak temperature dependence is observed in the short circuit current and the temperature coefficient of this parameter varies from sample to sample in the range of 0.02 ± 0.2 mA / °C with an average value of 0.08 mA / °C, which corresponds to ~ 0.05% / °C. It should be noted here that this value is somewhat underestimated (by 10-20%) compared to the value of the temperature coefficient of the short-circuit current, which is obtained when the PC sample is illuminated by the atmospheric Sun, due to differences in the spectral composition illuminating the light sample. Due to the scatter of the values of the optimal current within a small area, it is not possible to show the unambiguous values of the temperature coefficient of this parameter within the specified area. The maximum power obtained from the PC sample strongly depends on temperature and the shape of this dependence is close to linear. With an accuracy of the power measurement error (± 2 ± 3%), the temperature dependence of the maximum power on the five studied samples over the entire temperature range can be approximated by a straight line with a slope of SW = 0.3 ± 0.02% / °C. Experience shows that for samples whose frontal contact is made by photolithography, this coefficient has a negative sign (SW W = -0.25% / °C).
Most samples have a tendency to increase the absolute value of SW with increasing temperature, so that in the low-temperature region (T <-20 °C) Sw -0.25% / °C, and at higher temperatures 5w = -0.39% / °C. However, due to the impossibility of stabilization of light on our experimental installation with an accuracy substantially better than we can only speak of a trend of this kind, until the accuracy of measurements is increased and the temperature interval is not increased. We note that the maximum power temperature coefficient obtained by us at a temperature of -20 + 60 °C is significantly lower 20 + 25%) given in [5]. At low temperatures (T < 20 °C) the growth ofWmax and efficiency with decreasing temperature is stronger than described in [5]. It is important that at low temperatures at the PC research we can not observe no voltage or efficiency factor and maximum power. This is apparently due to the high quality of the p-n-junction, in which even at temperatures of 100 + —140 °C we could not notice leaks.
References:
1. Sachenko A. B., Kryuchenko Yu. V., Kostylev V. P. et al. Method for optimizing the parameters of heterojunction photoelectric converters based on crystalline silicon. Semiconductor Physics and Engineering, 2016.- Vol. 50.- Issue 2.- P. 259-262.
2. Masulko K., Shigematsu M., Hashiguchi T. et ai. IEEE J. Photovolt., 4 (6), 1433 (3014).
3. Shockley W., Queisser H. J., Appl. J. Phys., 32, 510 (1961).
4. Borisov S. N., Gorodetsky S. M., Grigorieva G. M., Zvyagina K. N., Kasimahunova A. M. The effect of light intensity and temperature on the parameters of silicon photoconverters. "Science", Solar technology,- Tashkent, 1983.- No. 4.- P. 3-6.
0
200
400
600
800
