Plasmonic GaInP solar cells with improved energy conversion efficiency

Abstract

We report here a new method to improve the performance of GaInP solar cells by fabricating Ag nanoparticles (Ag NPs) atop the devices through a simple wet-chemical method. The size and density of Ag NPs can be easily optimized by changing the concentration of AgNO₃ solution and/or the reaction time. The Ag NPs support surface plasmonic resonances (SPRs) and thus greatly increase light absorbance and photocurrent responses of the solar cells by the scattering and re-reflection of incident light. We find that, for a Ag NPs density of ∼3.37 × 10¹⁰ cm⁻², the short-circuit current density (J_sc) increases by 19.5% from 14.9 to 17.8 mA cm⁻², and the power conversion efficiency (PCE) increases by 20.4% from 15.2% to 18.3%.

---

Introduction

Thin-film III–V compound solar cells have tremendous potential in space and terrestrial applications due to their long minority carrier lifetimes and superior radiation resistance.^1–3 Although III–V semiconductor materials are usually more expensive, the adoption of thin-film solar cells greatly reduces material consumption and hence lowers production costs.⁴ However, decreasing the active layer thickness also diminishes light absorption by the thin-film solar cells.⁵ In addition, the incident light is also subjected to Fresnel reflection.^6,7 For a planar GaInP solar cell with AlInP as the window layer, one can calculate that a large proportion of solar light up to 28.5% will be reflected back to the air in the full solar spectrum range. As a result, a variety of techniques have been developed to improve light absorption efficiency of thin-film solar cells. However, these techniques usually have some drawbacks. For example, surface texturing often involves fabricating pyramidal structures of several micrometers that are too large to be used in thin-film solar cells.^8–10 Although substrate texturing can partially solve this problem, it cannot circumvent Fresnel loss and may increase the recombination rate.^11,12 Antireflection coating (ARC) can only be non-reflective at some specific wavelengths.^13–15

Recently, it was theoretically predicted that metal nanoparticles (NPs) on photovoltaic devices can increase their energy conversion efficiency.¹⁶ In particular, noble metal NPs, such as Ag, Au, and Pd, support surface plasmon resonances (SPRs) (collective oscillations of the conduction electrons) in the solar spectrum range, therefore enabling confinement of incident solar energy in a small volume and mediation of the photovoltaic processes.^17,18 Note that the optical properties of SPRs depend on the nanoparticle's size, shape and the surrounding dielectric environment.¹⁹ To date, some techniques have been developed to fabricate plasmonic thin-film solar cells. For example, Yu et al. discovered that colloidal Au NPs drop coated on the surface of p–n, n–p, and p–i–n Si solar cells can increase the power conversion efficiency (PCE) by 8.8%.^20–22 Lare et al. found that periodic Ag nanoparticles (NPs) arrays fabricated on Si solar cells using substrate conformal imprint lithography can increase the photocurrent by 10% due to second- and third-order grating coupling.²³ In addition to conventional Si solar cells, Pryce et al. found that Ag NPs arrays fabricated on the GaN/InGaN/GaN quantum well solar cells using anodic alumina template as a mask can increase the photocurrent by 6%.²⁴ Liu et al. fabricated Ag NPs on GaAs solar cells by thermal evaporation of a thin layer of Ag followed by annealing, and they demonstrated a 14.2% increase of the short-circuit current density (J_sc).²⁵

In this work, we report a one-step facile galvanic displacement process to fabricate Ag NPs on GaInP solar cells by in situ reduction of dilute AgNO₃ solution on the n-AlInP window layer. The size and density of the Ag NPs were optimized by adjusting the concentration of the AgNO₃ solution and/or the reaction time. An engineered enhancement was observed in J_sc and the PCE.

Results and discussion

Fig. 1 shows the schematic diagram of the GaInP solar cells used in this investigation. The single-junction n-on-p GaInP solar cells were grown by metal organic chemical vapor deposition (MOCVD) on a p-GaAs substrate. It consists of a p-GaAs substrate, a 100 nm-thick p-AlGaInP buffer layer, a 30 nm-thick p-GaInP back-side field (BSF), a 420 nm-thick p-GaInP base, a 20 nm-thick n-GaInP emitter, and a 30 nm-thick n-AlInP window layer. The BSF and window layers create energy barriers to reduce the recombination rate of minority carriers at the interfaces.^26,27 A 500 nm-thick n-GaAs finger contact layer was grown on top of the n-AlInP layer to form ohmic contact. Au finger electrodes and bottom electrodes were deposited on the n-GaAs cap layer and p-GaAs substrate, respectively.

Fig. 1 Schematic diagram of the GaInP single junction solar cells with Ag NPs deposited on the n-AlInP window layer.

The deposition of Ag NPs on the AlInP window layer follows the spontaneous galvanic displacement process. The electrons in the conduction band of n-type AlInP can transfer directly into the Ag⁺/Ag couple, resulting in the formation of Ag NPs on the AlInP surface. Fig. 2 shows SEM images of the samples before (a) and after deposition of Ag NPs for 0.5 (b, S1), 1 (c, S2) and 1.5 min (d, S3), respectively. The blank sample surface is smooth, flat and featureless (Fig. 2a). After deposition, Ag NPs appear on the sample surfaces. The Ag NPs are randomly distributed and spherical in shape, most of which remain isolated when the reaction times are 0.5 and 1 min. Increasing the reaction time to 1.5 min results in the agglomeration of some Ag NPs, as shown in Fig. 2d. Both the particle densities and diameters increase gradually upon elongation of the reaction time. The particle densities for S1, S2, and S3 are ∼1.9 × 10⁹, ∼6.7 × 10⁹ and ∼3.4 × 10¹⁰ cm⁻², respectively. The particle diameters increase from ∼30 (S1), to ∼40 (S2) to ∼60 nm (S3).

Fig. 2 SEM images of GaInP solar cells samples without (a) and with (b–d) Ag NPs after galavnic displacement reactions for 0.5 min (b), 1 min (c), and 1.5 min (d).

Fig. 3 shows the photocurrent density versus voltage (J–V) curves of the GaInP solar cells before and after deposition of Ag NPs under AM 1.5 irradiation. The results are summarized in Table 1. It is evident that the presence of Ag NPs significantly increases the J_sc from 14.9 to 16.8, 17.6 and 17.8 mA cm⁻², for S1, S2, S3, respectively. In comparison, the open-circuit voltage (V_oc) and the fill factor (FF) show only a slight increase. We calculated the PCE (η) of the solar cells by

where P_m is the maximum power output, P_in is the incident solar power, V_oc is the open-circuit voltage, I_sc is the short-circuit current, and FF is the fill factor.²⁸ We observed that the PCE increased from 15.2% to 17.1%, 18.1% and 18.3% for S1, S2, and S3, respectively. The PCE shows a maximum increase of 20.8% for Ag NPs decorated samples.

Fig. 3 The J–V characteristics of the GaInP solar cells under AM 1.5 irradiation. The black curve represents the blank sample without Ag NPs; the red, magenta, and green curves represent samples reacted for 0.5, 1, and 1.5 min, respectively.

Table 1 The short-circuit current density (J_sc), open-circuit voltage (V_oc), fill factor (FF), and power conversion efficiency (PCE) of the samples with and without Ag NPs under AM 1.5 irradiation

	Without Ag	Ag 0.5 min	Ag 1 min	Ag 1.5 min
Jsc (mA cm^−2)	14.9	16.8	17.6	17.8
Voc (V)	1.30	1.31	1.31	1.31
Fill factor (%)	78.3	78.4	78.8	79.0
PCE (η, %)	15.2	17.1	18.1	18.3

---

To understand the underlying physical principles, we measured the photocurrents under laser irradiation of different wavelengths. Fig. 4 shows the photocurrents for S3 as a function of laser wavelengths. Table 2 shows the values with standard deviation of the J_sc and the corresponding photocurrent enhancement ratio. Apparently, the photocurrent enhancement depends strongly on the laser wavelength. The enhancement ratios were 74.0%, 26.3%, 23.5% and 4.3% for 405, 445, 473, and 532 nm lasers, respectively. The maximum photocurrent enhancement occurs at 405 nm around the localized surface plasmon resonances (LSPRs) of Ag NPs. In contrast, when the laser wavelength is 650 nm, around the absorption edges of the GaInP, the J_sc shows no difference between samples with and without Ag NPs.

Fig. 4 Photocurrent responses of the GaInP solar cells under irradiation of different wavelengths for the blank sample (black) and S3 (red) and the photocurrent enhancement ratio (blue).

Table 2 The short-circuit current density (J_sc) and the corresponding photocurrent enhancement ratio (%) for the blank sample and S3

	405 nm	445 nm	473 nm	532 nm	650 nm
Jsc without Ag NPs (mA cm^−2)	15.7 ± 1.2	29.3 ± 1.3	39.7 ± 2.0	21.5 ± 1.0	0
Jsc with Ag NPs (mA cm^−2)	27.4 ± 1.1	37.0 ± 1.9	49.0 ± 2.0	22.4 ± 1.1	0
Photocurrent enhancement ratio (%)	74.0 ± 5.5	26.3 ± 6.1	23.5 ± 6.3	4.3 ± 2.1	0

---

We discuss the abovementioned results based on LSPRs of Ag NPs on GaInP solar cells. A schematic diagram of the possible physical principles is shown in Fig. 5. It is well-known that the efficiency of thin-film solar cells is limited by their poor absorption efficiency of incident solar energy. LSPRs can greatly enhance the local electromagnetic field, increase forward scattering of incident light, and increase absorption at the resonance frequencies.^29–32 In this study, we chose Ag NPs because of their low absorption loss, compared with other metals.¹⁸ Light that interacts with Ag NPs creates LSPRs, which enhance the electromagnetic fields in the vicinity of the particles and greatly increase the scattering cross section several times larger than the geometric cross section. As a result, they can greatly increase the forward scattering of incident light into the active layer of the GaInP solar cells.^33–35 In addition, light that is not absorbed may be captured by the LSPRs of Ag NPs, which can deflect it again back into the active region for absorption. As a result, the optical path of the incident light can be greatly increased such that more effective photoelectric conversion can be expected.

Fig. 5 A schematic of the solar cell's surface decorated with Ag NPs, which can greatly increase forward scattering of incident light and re-reflection of the backscattered light.

Experimental

Prior to the deposition of Ag NPs, the samples are cleaned in acetone, ethanol, and HF solution with sonication to remove possible organic contaminations and oxides. After refreshing with deionized water and drying in N₂, 10 mM AgNO₃ (Alfa Aesar) solution was dropped onto the sample's surface to undergo galvanic displacement reactions for 0.5, 1, and 1.5 min, respectively. The samples are hereafter named S1, S2, and S3. After reactions, the samples were washed with deionized water and blow-dried with N₂.

The morphologic features of the samples were studied by a scanning electron microscope (SEM, Hitachi S-4800, Japan) operated at 5 kV under high vacuum. The illuminated current density versus voltage (J–V) characteristics was measured by a SolarIV-150A solar simulator at room temperature under Air Mass 1.5 (AM 1.5) irradiation. The photocurrent densities under single wavelength irradiation were also measured by using semiconductor lasers of 405, 445, 473, 532 and 650 nm, respectively. The power density was adjusted by using a linear variable density filter and monitored by an optical power meter (Ophir Optronics, PD300-UV-193). During the tests, the power densities were kept constant at 100 mW cm⁻² for all the laser beams. The laser beams illuminated uniformly and vertically on the sample's surface.

Conclusions

In conclusion, we have demonstrated that Ag NPs deposited on GaInP solar cells can greatly increase the J_sc and the PCE, whereas they had little influence on the V_oc. We attributed the enhancement to LSPRs of Ag NPs that can greatly increase the forward scattering and re-reflection of the incident light. After parameter optimization, we demonstrated an increase of the J_sc by 19.5% from 14.9 to 17.8 mA cm⁻², and an increase of the PCE by 20.4% from 15.2% to 18.3% when the Ag NPs have an average diameter of ∼60 nm and a concentration of ∼3.37 × 10¹⁰ cm⁻². This method is general and could be extended to other III–V optoelectronic devices for improved performance.

Acknowledgements

The authors thank the National Science Foundation of China (NSFC) (Grant nos 91233122, 91123007, 51472143), the Innovation Projects of Shandong University (Grant nos 2014JC032, 2014YQ003), the Young and Middle-Aged Scientists Research Awards of Shandong Province (Grant no. BS2013DX007), and the National Basic Research Program of China (973 Program) (Grant no. 2009CB930503) for financial support.

Notes and references

1. A. W. Bett, F. Dimroth, G. Stollwerck and O. V. Sulima, Appl. Phys. A: Mater. Sci. Process., 1999, 69, 119–129.
2. R. W. Miles, Vacuum, 2006, 80, 1090–1097.
3. R. W. Miles, G. Zoppi and I. Forbes, Mater. Today, 2007, 10, 20–27.
4. B. Parida, S. Iniyan and R. Goic, Renewable Sustainable Energy Rev., 2011, 15, 1625–1636.
5. G. J. Bauhuis, J. J. Schermer, P. Mulder, M. M. A. J. Voncken and P. K. Larsen, Sol. Energy Mater. Sol. Cells, 2004, 83, 81–90.
6. M.-L. Kuo, D. J. Poxson, Y. S. Kim, F. W. Mont, J. K. Kim, E. F. Schubert and S.-Y. Lin, Opt. Lett., 2008, 33, 2527–2529.
7. T. Knüttel and S. Bergfeld, Laser Tech. J., 2013, 10, 21–24.
8. P. Campbell and M. A. Green, J. Appl. Phys., 1987, 62(1), 243–249.
9. J. Kim, D. Inns, K. Fogel and D. K. Sadana, Sol. Energy Mater. Sol. Cells, 2010, 94, 2091–2093.
10. L. A. Dobrzański and A. Drygala, J. Achieve. Mater. Manuf. Eng., 2008, 31, 77–82.
11. J. Müller, B. Rech, J. Springer and M. Vanecek, Sol. Energy, 2004, 77, 917–930.
12. Y. Wang, X. Zhang, L. Bai, Q. Huang, C. Wei and Y. Zhao, Appl. Phys. Lett., 2012, 100, 263508.
13. Y. F. Makableh, R. Vasan, J. C. Sarker, A. I. Nusir, S. Seal and M. O. Manasreh, Sol. Energy Mater. Sol. Cells, 2014, 123, 178–182.
14. Z. N. Adamian, A. P. Hakhoyan, V. M. Aroutiounian, R. S. Barseghian and K. Touryan, Sol. Energy Mater. Sol. Cells, 2000, 64, 347–351.
15. S.-Y. Lien, D. S. Wuu, W. C. Yeh and J.-C. Liu, Sol. Energy Mater. Sol. Cells, 2006, 90, 2710–2719.
16. K. R. Catchpole and A. Polman, Opt. Express, 2008, 16, 21793–21800.
17. A. Polman, Science, 2008, 322, 868–869.
18. S. Pillai, K. R. Catchpole, T. Trupke and M. A. Green, J. Appl. Phys., 2007, 101, 093105.
19. S. Linic, P. Christopher and D. B. Ingram, Nat. Mater., 2011, 10, 911–921.
20. D. M. Schaadt, B. Feng and E. T. Yu, Appl. Phys. Lett., 2005, 86, 063106.
21. P. Matheu, S. H. Lim, D. Derkacs, C. McPheeters and E. T. Yu, Appl. Phys. Lett., 2008, 93, 113108.
22. D. Derkacs, S. H. Lim, P. Matheu, W. Mar and E. T. Yu, Appl. Phys. Lett., 2006, 89, 093103.
23. M. V. Lare, F. Lenzmann, M. A. Verschuuren and A. Polman, Appl. Phys. Lett., 2012, 101, 221110.
24. I. M. Pryce, D. D. Koleske, A. J. Fischer and H. A. Atwater, Appl. Phys. Lett., 2010, 96, 153501.
25. W. Liu, X. Wang, Y. Li, Z. Geng, F. Yang and J. Li, Sol. Energy Mater. Sol. Cells, 2011, 95, 396–698.
26. P. Dai, S. L. Lu, Y. Q. Zhu, L. Ji, W. He, M. Tan, H. Yang, M. Arimochi, H. Yoshida, S. Uchida and M. Ikeda, J. Cryst. Growth, 2013, 378, 604–606.
27. T. Hayakawa, T. Ohta, Y. Nakashima, K. Koyama, K. Ohdaira and H. Matsumura, Jpn. J. Appl. Phys., 2012, 51, 101301.
28. C. Hu and R. M. White, Solar cells from basics to advanced systems, McGraw-Hill, New York, 1st edn, 1983, ch. 3, pp. 57–60.
29. S. H. Lim, W. Mar, P. Matheu, D. Derkacs and E. T. Yu, J. Appl. Phys., 2007, 101, 104309.
30. E. Hutter and J. H. Fendler, Adv. Mater., 2004, 16, 1685–1706.
31. Y. A. Akimov, K. Ostrikov and E. P. Li, Plasmonics, 2009, 4, 107–113.
32. E. T. Yu, D. Derkacs, S. H. Lim, P. Matheu and D. M. Schaadt, Proc. SPIE, 2008, 7033, 73001V.
33. X. Chen, B. Jia, J. K. Saha, B. Cai, N. Stokes, Q. Qiao, Y. Wang, Z. Shi and M. Gu, Nano Lett., 2012, 12, 2187–2192.
34. T. L. Temple, G. D. K. Mahanama, H. S. Reehal and D. M. Bagnall, Sol. Energy Mater. Sol. Cells, 2009, 93, 1978–1985.
35. S. Zhou, Q. Li, S. Townsend, Y. M. Xie and X. Huang, 10^th world congress on structural and multidisciplinary optimization, Orlando, Florida, USA, 2013.

---