INTRODUCTION:

Energy requirements and climate change are two major threads for the sustainability of the human community. Photovoltaics (PV) — the conversion of sunlight into electricity using solar cells — has progressed more rapidly as one of the most promising technology that address both issues. Importantly, CO₂ and other pollutants are not involved in the PV conversion rendering it a totally green process. Although playing an increasingly important role in energy market, PV is limited by its low energy efficiency, which is the power release per unit volume. Of all the semiconductors (amorphous or crystalline), Si is the least expensive, the most abundant (26% of Earth’s crust), and uniformly distributed element on earth [1]. The Si PV industry controls almost 90% of the global market [2].

There have been many improvements and developments since the first Si based solar cell was made in 1941 [3]. The improvement in Si solar cell performance came with time and involves improvement in material quality, device design advancement in cell passivation technology [4] and light trapping structures [5]. In addition to this, emitters are an important aspect to improve the efficiency of Si solar cells because a highly doped and its type of region introduces crystalline defect states that increase the electron-hole recombination thereby decreasing the blue response (i.e., the short-circuit current Isc) as well as the open-circuit voltage (Voc). Another unwanted effect is the increase in photon absorption near the front surface due to band-gap narrowing [6]. On the other hand, a light dopant concentration and its type near the cell's front surface reduce the surface recombination and thus improving the Vocand Isc[7, 8].

One way to improve the solar energy conversion efficiency is to increase the open circuit voltage by increasing the concentration of dopants. However, this also leads to an increase of recombination of electron-hole pairs in the region of high doping. Thus, reducing the high doping area due to the doping depth is important for improving the conversion efficiency [9]. Reiter et al [10] demonstrated that the short circuit current density loss is higher in n type Si than the p type, which helps to optimize the front junction. The studied are reported for the large thickness of above 20nm. Romer et al, [11] reported that open circuit voltages are 732 mV for n type poly- Si and 711mV for p type poly-Si. The standard processing sequence for industrial screen printed p-type Cz silicon solar cells is well optimized, but it suffers from a rather low bulk lifetime which is caused by the boron and oxygen related light induced degradation [12]. This limitation can be overcome by using n-type silicon, which does not contain any boron and is more tolerant for metal impurities [13]. Applying the standard processing sequence to n-type silicon leads to a rear emitter solar cell with a phosphorous front surface field [14]. Since, the electron-hole pairs are mainly generated close to the front surface of the solar cell, therefore, the type of Si either n or p may significantly affect the solar cell efficiency.

Over the huge development over 66 years, still the major limitation of today’s industrial solar cells is recombination at the front surface hence; it is desirable to utilize the excellent front surface selection of Si junctions. In addition to this, solar technology is approaching to thin film which is the most emerging topic of interest for the researchers over world. Hence, it is better to optimize the front surface for future thin film technologies. Therefore, the purpose of this work is therefore to investigate the p-n junction properties of Si layers, and to determine constraints regarding thickness for an acceptable optical absorption and hence, efficient solar cells. In present study; numerical modeling of Si solar cell has been carried out by PC1D numerical simulator to optimize the thickness of front surface with its suitable type of material i.e., n or p type.

PC1D Simulation Method:

Numerical modeling is widely used to gain better understanding into the details of the physical operation of solar cells. For numerical modeling of PV devices excellent, intuitive, open-source simulation programs such as PC1D [15, 16], AforsHet [17, 18], AMPS [19, 20], SCAPS [21], and ASA [22] have been developed over the years. Numerical simulation is routinely performed in developing and studying crystalline silicon solar cells. They are extremely convenient in most cases and shed light on numerous phenomena occurring in solar cells, especially those based on Si. Amongst the various programs in use, PC1D has been chosen as a simulation tool for this research regarding its user-friendly system. PC1D is the most common and perhaps simplest simulation software. The process parameters can be adjusted by choosing the appropriate layers or contacts in the schematic diagram of the device. PC1D is usually used for solving the one-dimensional semiconductor equations based on Shockley–Read Hall recombination statistics [23]. Based on these considerations, PC1D [24] was chosen to describe bifacial solar cell in this study. PC1D was originally developed by Basore and coworkers at Sandia National Labs and was further developed at UNSW, Australia. It allows simulating the behaviour of PV structures based on semiconductor by respecting to axial symmetry. The PC1D contains the files of the solar spectrum AM1.5 [25]. In present study, we have chosen “one-sun.exc” and 'scan-qe.exc' in excitation that gives us the characteristics I-V and quantum efficiency of cells solar, respectively. These files contain standard parameters to be used during the simulation of solar cells and provide the respective output parameters as:

The “one-sun.exc” file gives us the following results:

1 The value of short circuit current.

2 The value of the open circuit voltage.

3 The value of the maximum power.

The file “scan-qe.exc” gives us the following results:

1 The value of short circuit current.

2 The value of the maximum power.

Figure 1a

Figure 1b

Figure 1 PC1D-simulated a) Characteristics I-V and b) external quantum efficiency (EQE) for 10µm layer thickness of both p and n type materials as front/emitter surface of solar cell.

RESULTS AND DISCUSSION:

In the solar cell simulations the doping concentration plays an important role while optimizing the other parameters. Therefore, in present study, the doping concentration of p and n are 1×10¹⁹cm^-3 and 1×10¹⁶cm^-3 respectively considered as suggested by Belarbi et al [26]. We selected a total thickness of 100 μm and a surface of 100 cm². To look at the front/emitter surface material effect on cell performance, 10µm layer of both p and n types are considered as an example. Figure 1a clearly shows the significant current enhancement for p type in comparison to n on front/emitter surface. Since the electron-hole pairs are mainly generated close to the front surface of the solar cell hence, to look into the nature of this difference in n or p type emitter, we have calculated the external quantum efficiency (EQE), which is defined as the ratio of number of charge carriers collected by the solar cell to the number of photons of a given energy shining on the solar cell from outside under the AM 1.5G illumination. Figure 2 depicts that while keeping p as front surface of solar cells clearly improve EQE upto 90% in the short wavelength to visible region i.e. 300nm to 700nm. But in case of n type as emitter, EQE can be achieved upto maximum of 70% in the near infrared region with relatively narrow band width in comparison to p. Hence, energy photons are collected for the p type emitter in comparison to n type which results in comparatively large current density. Area in lower wavelength limit of figure 1b (for external quantum efficiency) refers about losses caused by surface recombination velocity, in middle region due to surface reflection and diffusion length of carriers and in longer wavelength limit refers to about recombination of the rear side of the cell [27]. Therefore, looses due to surface recombination are minimum in case of p type as emitters. In overall, p type allows the broad wavelengths of the illumination to be collected which in turn contribute to more electron–hole pair (EHP) generation. Therefore, the values for open circuit voltage (V_OC) and short-circuit current (I_SC) may be increased.

Now to probe the Voc and Isc in details, the layer thickness have been varied from 0.5 to 10 µm. Figure 2a depicts the variation of open circuit voltage (V_oc) as a function of emitter layer thickness for both p and n type. Ostensibly, V_oc decreases from 0.723V to 0.657V with increase in the n type layer thickness from 0.5 to 10µm. This decrease in V_oc is about 9% which is considerably significant. This considerable decrease may be due to the reduced EQE with thickness as shown in figure 2b. Clearly figure 2b pictured that the lower EQE in short wavelength region is due to increased surface recombination on the front surface increased reflectance from the surface in visible regime. These losses ultimately reduce the Voc and hence, efficiency of solar cells. Hence, the parameters Voc calculations suggest keeping its thickness as small as possible to get efficient cells. On the other hand, about 2% increase in Voc for p type layer thickness increase is also seen in figure 2a. This increase may be due to the reduced reflectance from the front surface of the solar cell and can be easily understood from figure 2c. Figure 2c clearly indicates the increase in EQE with thickness in the visible region i.e. the losses due to reflectance can be improved by increase in layer thickness of p type but not much significant as we believe.

Figure 2a

Figure 2b

Figure 2c

Figure 2 a) Variation of open circuit voltage with emitter/front layer thickness for both p and n type. Simulated EQE of the solar cells with different emitter layer thickness as b) n type and c) p type.

Figure 3a

Figure 3b

Figure 3 a) Short circuit current b) maximum power plotted against the p and n type layer thickness.

Figure 3 shows the variation of I_scand P_max with thickness of emitter/front layer. The calculated maximum of I_sc is 3.74A for 0.5µm thickness of n type front surface (see figure 3a). This I_sc is about 21% more than its value for n type of 10µm thickness. This decrease can again be easily understood with the help of EQE variation with layer thickness as explained above (figure 2b). It is also clear from figure 3a that these parametric I_sc trends are in reverse fashion while keeping p side towards the irradiated sun beam. It is pointed out that although Icc is about 1.7 times higher for p type as front in comparison to n at layer thickness of 10µm but at the thickness of 0.5µm fashion is reversed and it becomes 0.5 times. This reverse in fashion occurs at the layer thickness of about 6µm. Therefore, in future thin film solar cells techniques using n type as the front/emitter medium may be beneficial. Similar trends for power maximum (P_max) are also pictured when plotted against the layer thickness (see figure 3b). Maximums of P_max 2.72 Watts and 1.67 Watts are obtained for 0.5µm n type layer thickness and for 10 µm p type layer thickness, respectively. At thickness of 0.5µm, P_max for n type is almost two times larger than the p type materials as front medium.

Figure 4a

Figure 4b

Figure 4 a) fill factor (FF) and b) power conversion efficiency (PCE) as a function of emitter/front surface thickness.

Actual silicon solar cells also suffer from ohmic losses due to parasitic resistance R_s (series) and R_sh (shunt). R_sis mainly attributed to the sheet resistance of doped regions in the case of two dimensional current flows, bulk resistance of silicon substrate, metallic resistance of gridline and specific contact resistance between silicon and metal. Increasing the shunt resistance (R_sh) and decreasing the series resistance (R_s) lead to a higher fill factor (FF), thus resulting in greater efficiency, and bringing the cell's output power closer to its theoretical maximum [28]. Hence, the factor FF which is directly affected by the value of the cell's series and shunt resistances is also calculated. Interestingly, maximum FF of 84.1% is obtained for n type layer thickness of 0.5µm and it decreases with increase in layer thickness (see figure 4a). In case of p type as emitter, maximum FF of 84.2% is seen for 6µm layer thickness. Now, to fully probe the performance of solar cells, we have also calculated the power conversion efficiency (PCE). Figure 4b represents the variation of PCE with the layer thickness of p type and n type front layer thickness. The maximum value of PCE of 25.24% is found for 0.5um thickness of n type. As the layer thickness increased from 0.5µm to 10µm, the PCE decreases from 25.24% to 10.64%. In case of p type as front surface, as the layer thickness increased from 0.5 to 10µm, PCE increased from 12.32% to 18.55%. Fascinatingly, the decrease in PCE with layer thickness is much steeper than the increase. In general, it is obtained from the study that the solar cell parameters show the layer thickness dependent reverse fashion as we change the emitter surface from p to n type and vice versa.

In conclusion, using PC1D, we have studied the solar cells parameters for both n type and p type considered as front/emitter materials. From cell parameters like open circuit voltage, I_sc, P_max, FF and PCE, we have seen that n type as front surface in solar cells make them more efficient if the layer thickness is below 6µm. If the layer thickness above 6µm is considered then the solar performance is better by considering p type as front/emitter surface materials. The presented simulations results may help to design efficient thin film Si solar cells in future.

REFERENCES:

1. Ali, A.; Cheow, S. L.; Azhari, A. W.; Sopian, K.; Zaidi, S. H. Results in Physics. 2017, 7, 225–232

2. Asim, N.; Sopian, K.; Ahmadi, S.; Saeedfar, K.; Alghoul, M. A.; Saadatian, O.; Zaidi, S. H. Renewable Sustainable Energy Rev. 2012, 16, 5834–5847.

3. Green, M. A. Physica E: Low dimensional Systems and Nanostructures. 2002, 14, 11-17.

4. Aberle, A. G. Prog. Photovolt Res Appl. 2000, 8, 473-487.

5. Green, M. A.; Zhao, J.; Wang, A.; Wenham, S. R. IEEE Trans Electron Devices. 1999, 46, 1940-1947.

6. Brammer, T.; Zastrow, U.; Stiebig, H. Proc. 17^th European Photovoltaic Solar Energy Conf., Munich, Germany 2001, 22-26.

7. Shetty, K. D.; Boreland, M. B.; Shanmugam, B.; Cunnusamy, J.; Wu, C.; Iggo, S.; Antoniadis, H. Energy Procedia, 2013, 33, 70-75.

8. Mikeska, K. R.; Carroll, A. F.; Hanna, L. G.; Lewittes, M. E.; Li, Z.; Liang, L.; Meisel, A.; Scardera, G.; Subramoney, S.; VerNooy, P. D. Proc. 27^th European Photovoltaic Solar Energy Conf., Messe Frankfurt, Germany, 2012, 1717-1721.

9. Alexander, V.; Korovina, José Alvareza; Jean-Paul, K. Energy Procedia, 2016, 92, 103 – 108.

10. Reiter, S.; Koper, N.; Reineke-Koch, R.; Larionova, Y.; Turcu, M.; Krügener, J.; Tetzlaff, D.; Wietler, T.; Höhne, U.; Kähler, J.; Brendel, R.; Peibst, R. Energy Procedia. 2016, 199-204.

11. Römer, U.; Peibst, R.; Ohrdes, T.; Lim, B.; Krügener, J.; Bugiel, E.; Wietler, T.; Brendel, R. Solar energy materials and solar cells. 2014, 131, 85-91.

12. Glunz, S. W.; Rein, S.; Lee, J. Y.; Warta, W. J Appl Phys. 2001, 90, 2397-404.

13. Macdonald D.; Geerligs, L. J. Appl Phys Lett. 2004, 85, 4061-3.

14. Meier, D. L.; Davis, H. P.; Garcia, R. A.; Salami, J.; Rohatgi, A.; Ebong, A. Solar Energy Materials and Solar Cells. 2001, 65, 621-7.

15. R. Brendel et al, Proceedings of the 31^st European Photovoltaic Solar Energy Conference and Exhibition. 2015, 264.

16. Singh, H. K.; Mondal, S.; Arunachalam, B.; Soman, A.; Sharma, P.; Solanki, C. S. Plasmonics 2017, DOI: 10.1007/s11468-017-0563-8

17. Clugston, D. A.; Basore, P. A. IEEE 26^thPhotovolt. Spec. Conf., 1997, 207–210.

18. Stangl, R.; Kriegel, M.; Schmidt, M. IEEE 4^th World Conference on Photovoltaic Energy Conference. 2006, 2, 1350–1353.

19. Analysis of Microelectronic and Photonic Structures (AMPS) software was developed at Pennsylvania State University under the direction of Fonash, S.J., with funding from the Electric Power Research Institute. Also, see. http://www.psu.edu/dept/AMPS/.

20. Hernández-Como, N.; Morales-Acevedo, A. Solar Energy Materials and Solar Cells. 2010, 94, 62−67.

21. Burgelman, M.; Nollet, P.; Degrave, S. Thin Solid Films, 2000, 361–362, 527–532. Also, see http:// www.elis.ugent.be/ELISgroups/solar/projects/scaps.html.

22. Schropp, R. E. I.; Zeman, M. Kluwer Academic Publishers, Boston, 1998.

23. Shui-Yang Lien; Dong-Sing Wuu. Prog. Photovolt: Res. Appl. 2009, DOI: 10.1002/pip.

24. https://www.engineering.unsw.edu.au/energy-engineering/research/software-data-links/pc1d-software-for-modelling-a-solar-cell

25. http://rredc.nrel.gov/Solar/Spectra/am1.5/

26. Belarbi, M.; Benyoucef, A.; Benyoucef, B. Advanced Energy: An International Journal. 2014, 1, 1-10.

27. Meier, D.; Good, E. A.; Garcia, R. A.; Bingham, B. L.; Yamanaka, S.; Chandrsekaran, V.; Bucher, C. IEEE 4^th World Conference on Photovoltaic Energy Conference. 2006, DOI: 10.1109/WCPEC.2006.279656.

28. Green, M. A., Solid State Electronics 24, 788, (1981)

Received on 20.08.2017 Accepted on 21.10.2017

Research J. Engineering and Tech. 2017; 8(4): 414-418.

DOI: 10.5958/2321-581X.2017.00072.1