Reflective perovskite solar cells for efficient tandem applications

Abstract

Tandem solar cells combining a wide bandgap, efficient perovskite absorber with a low bandgap photovoltaic module, such as a c-Si cell, can potentially achieve a high theoretical efficiency of over 30%. Instead of using the conventional parallel stacking tandem, we report here a reflective tandem configuration, with the perovskite solar cell acting as the spectral filter that absorbs high energy photons, while sub-bandgap photons are reflected to a Si sub-cell using a highly reflective back metal electrode. The perovskite solar cell exhibits a high reflectance of over 60% in the near infrared spectral region, which allows the subsequent silicon cell to absorb photons in this region, resulting in a high current density of 13.03 mA cm⁻². In such a tandem configuration, we achieved a combined efficiency of 23.1% using a four-terminal measurement. This result demonstrates the promise of employing perovskite solar cells in a reflective tandem for a high efficiency solar energy conversion system, with an efficiency of up to 30%.

---

Introduction

Metal halide perovskite solar cells are emerging as a competitive candidate for energy conversion in the photovoltaic industry due to their inexpensive material costs, solution processability and high levels of performance.^1–9 Their power conversion efficiency (PCE) has improved tremendously from 3.8% to a certified 22.1% in the last six years.^3,10 Moreover, the bandgap of perovskites can be tuned from 1.55 to 2.3 eV, making them highly attractive for use in tandem cells in combination with other commercial photovoltaics such as silicon (Si) cells.^11–18 In perovskite/Si tandem cells, short-wavelength photons are converted in the perovskite sub-cell at a relatively high voltage, which reduces thermalization losses, while long-wavelength photons are absorbed by the small bandgap Si sub-cell, to cover a wide spectral range. This approach could realistically surpass the Shockley–Queisser efficiency limit for a single junction cell and potentially achieve an efficiency of over 30% in a cost-effective way.^19–24

A regular perovskite solar cell consists of multiple layers, including a transparent conducting electrode as the optical window, an electron transport layer, a perovskite layer as the light absorber, a hole transport layer, and a metal back electrode made from a material such as gold, silver or aluminum.^25,26 To ensure good conductivity, the thickness of the back metal electrode is typically above 100 nm, which makes the perovskite cell opaque. However, this is not favorable for tandem applications because the perovskite top cell has to be highly transparent in the near-infrared (NIR) spectral region to transmit light to the bottom cell. Several groups have reported semi-transparent perovskite cells through replacing the metal electrode with silver nanowires, ultra-thin Au film and sputtered transparent conducting oxides (TCO).^18,28–31 Since the back electrode has to be deposited on top of the complete cell stack, sputtering can damage the sensitive perovskite layer and charge transport layer.^31,32 A thin buffer layer made of metal oxides has to be introduced prior to the sputtering of TCO to protect the perovskite cell.^17,30,31 Despite the high transparency obtained from TCO or silver nanowires, semi-transparent perovskite cells suffer a substantial efficiency drop, mainly due to the low conductivities of these transparent electrodes compared to thick metal films.^27–34 Furthermore, the sophisticated deposition process for a transparent back electrode increases the fabrication cost of perovskite devices.

An alternative approach to the conventional stacked tandem is to construct a spectrum splitting system, where incident light is selectively split to different sub-cells.^35–44 The opaque perovskite solar cell itself can serve as a spectral filter, where the perovskite layer absorbs the high energy photons and the remaining NIR photons are reflected from the metal electrode. The reflective property of the perovskite cell is beneficial for the construction of a reflective tandem that does not require a transparent back electrode.^35–37 Recently, Grant et al. analyzed the optical loss in a perovskite/Si reflective tandem cell and showed that the parasitic absorption in the NIR region was mainly caused by fluorine doped tin oxide (FTO) and the hole transporting material (Spiro-OMeTAD).⁴³ This issue can be overcome through the use of a perovskite/Si reflective tandem solar cell based on an inverted planar perovskite device with high NIR reflection. In this work, we demonstrate the feasibility of such a design. By optimizing the perovskite composition, we fabricated an inverted perovskite cell with a light conversion efficiency of over 16% and high reflectance in the NIR region. This enabled us to build a perovskite/Si reflective tandem with a total efficiency of up to 23.1%, a significant gain compared to each sub-cell.

Results and discussion

The reflective perovskite/Si tandem architecture is illustrated in Fig. 1a. The perovskite cell is positioned at 45° to the incident light, and the Si cell is oriented at 45° to the perovskite cell. The 45° angle between the perovskite cell and the Si cell leads to normal incidence of the reflected light on the Si cell, which makes the tandem system more efficient, with a minimized Si module area compared with other angles (detailed discussion can be found in the ESI†). The perovskite cell was prepared on indium tin oxide (ITO) glass with a p–i–n multi-layer structure, as depicted in Fig. 1b. A thin hole transport layer (poly-TPD, 30 nm) was first deposited on the ITO and a perovskite layer was then spin-coated on the poly-TPD. After annealing the perovskite film, an electron extraction layer (PCBM, 30 nm) was spin-coated and the cell was completed through the evaporation of bathocuproine (BCP, 6 nm), followed by an Ag electrode (Fig. 1c). The inverted perovskite cell has a planar configuration, hence mirror-like reflection can be achieved compared to a normal configuration, where incident light is scattered using a mesoporous TiO₂ layer.⁷ Here, a mixed-cation mixed-halide perovskite (FA_0.9Cs_0.1Pb(I_1−xBr_x)₃) was used due to its high thermal and environmental stability and bandgap tunability.^17,45–47

Fig. 1 Schematic diagram of the perovskite/Si reflective tandem solar cell (a), illustration of the inverted perovskite solar cell structure (b), and the corresponding layers in a cross-section scanning electron morphological image (c).

For application in such a configuration, the optical properties of the cells are very important. Both the transmittance and reflectance of the perovskite solar cell were investigated. The transmittance is recorded without a back electrode at normal incidence, and the reflectance is measured at 45° incidence (Fig. 2). The perovskite layer has a high absorption coefficient leading to low transmittance at photon energies above its band gap. The average transmittance reached 67% in the 800 to 1200 nm region due to parasitic absorption in the perovskite layer as well as the charge transport layers. When the device is fabricated with a highly reflective Ag electrode, all the transmitted light can be reflected back. However, the reflected light in the NIR region was found to be 53%, lower than the transmittance, possibly due to re-absorption of reflected light within the perovskite cell. Moreover, the perovskite cell shows an average reflection of ∼10% in the visible range (400 to 800 nm). As this portion of light is reflected from the front of the perovskite cell, it causes major optical loss in transmittance mode.^23,24 However, they can be further utilized using a Si sub-cell in a reflection tandem, where the two sub-cells form a light-trapping configuration. When coupled with AM 1.5G solar radiation, the reflectance of the perovskite solar cell corresponds to a photon current of 14.8 mA cm⁻², slightly higher than the 14.6 mA cm⁻² obtained in transmittance mode. In a stacked tandem, the semi-transparent perovskite cell is completed with a TCO electrode which typically reduces the NIR transmittance by 5 to 10%.^20,48 By taking this parasitic loss into consideration, the total reflectance of a perovskite cell could generate a greater photon current than when in transmission configuration. These results show that a reflective tandem has the potential for a higher solar energy conversion efficiency with greater ease of fabrication.

Fig. 2 Transmittance of the perovskite cell without a back electrode, and reflectance of the cell with a silver back electrode.

The performances of a series of perovskite solar cells with different bandgaps ranging from 1.53 eV to 1.7 eV were examined. The variation in bandgap is achieved through composition variation, i.e., varying the ratio of Br to I (FA_0.9Cs_0.1Pb(I_1−xBr_x)₃). Their current density–voltage (J–V) characteristics are shown in Fig. 3 and their corresponding short circuit current density (J_sc), open circuit voltage (V_oc), fill factor (FF) and power conversion efficiency (PCE) values are summarized in Table 1. Note that all the device properties are measured at a 45° angle to the incident light with an aperture to define the illumination area. The PCE obtainable in such a configuration is slightly lower than with direct irradiation at normal incidence, due to a reduced V_oc and a slightly lower J_sc (Fig. S4, ESI†). From the EQE curves at two incident angles, it is highly probable that the comparable J_sc from the two measurements is due to an elongated light path inside the perovskite layer due to tilting compensating the reflection loss at 45° compared to normal incidence.

Fig. 3 J–V characteristics of perovskite solar cells with different compositions.

Table 1 J–V parameters of perovskite solar cells with different compositions

Composition	J sc (mA cm^−2)	V oc (V)	FF	PCE (%)
(FA0.9Cs0.1PbI3)	23.2	0.94	0.67	14.7
Br0.03	22.5	1.03	0.71	16.5
Br0.1	21.8	1.07	0.72	16.4
Br0.2	20	1.14	0.71	15.4
Br0.3	18.1	1.12	0.72	14.5

---

From the J–V characteristics in Table 1, we see that a high PCE of 14.7% can be achieved using a pure iodine perovskite absorber (FA_0.9Cs_0.1PbI₃) with a J_sc of 23.2 mA cm⁻². By adding a small amount of bromine (x = 0.03, Br_0.03), the PCE increased to 16.6%. Although the slightly higher bandgap resulted in a reduced J_sc, the V_oc of the mixed halide device improved to 1.025 V compared with 0.94 V for FA_0.9Cs_0.1PbI₃. The effect of a small amount of bromine improving the efficiency is not fully understood with recorded high efficiency obtained from mixed halides of Br and I.^49,50 With more bromine added into the perovskite absorber, the total efficiency dropped due to a narrower absorption range with a larger band gap. However, the wide band gap perovskites generating a higher V_oc with smaller thermodynamic loss may perform better when used in tandem with a Si solar cell.¹⁷

The corresponding external quantum efficiency (EQE) spectra for the various compositions are shown in Fig. 4a. All the perovskite solar cells show high single wavelength conversion efficiencies above their band gaps. The photon current calculated through integrating the EQE spectra matches well with the value obtained from the J–V curves (Table 1). While the EQE curves exhibit a sharp drop when approaching their band gaps, the cell reflection increases correspondingly, and reaches >50% reflection in the wide NIR region (Fig. 4b).

Fig. 4 EQE spectra for the perovskite solar cells (a), and their corresponding reflectance (b).

The high performance of the perovskite solar cells and their good reflectance make them highly attractive for use in reflective tandems with a smaller bandgap Si solar cell. We evaluated the potential of reflective perovskite for use in tandem cells with an efficient crystalline Si solar cell in a four-terminal configuration. The perovskite cell with a composition of FA_0.9Cs_0.1Pb(I_0.9Br_0.1)₃(Br_0.1) was first used here because of the high efficiency achieved. As shown in Fig. 5a, most of the visible light was converted using the perovskite cell, resulting in a J_sc of 21.2 mA cm⁻² while the Si sub-cell received the reflected NIR photons and yielded an EQE of ∼55%, contributing an additional 12.7 mA cm⁻². Interestingly, the Si sub-cell also collected the reflected light from the front of the perovskite cell, which is usually lost in a transmission tandem. As a result, the reflective tandem achieved an EQE of ∼90% from 400 nm to 800 nm. Unfortunately a large portion of NIR photons were lost due to an insufficient reflection efficiency.

Fig. 5 EQE spectra of a perovskite cell and a filtered Si cell (a), and the J–V characteristics of a perovskite solar cell, a Si cell (unfiltered) and a Si cell (filtered) (b).

The J–V curves measured under 1-sun illumination are shown in Fig. 5b and the extracted characteristic parameters are summarized in Table 2. The stand-alone Si cell had a PCE of 21.3% with a J_sc of 39.58 mA cm⁻², a V_oc of 0.665 V and a FF of 0.809. After filtering, the Si sub-cell exhibits a J_sc of 13.03 mA cm⁻², close to the photon current calculated from the EQE curve. The V_oc is reduced to 0.631 V, due to a lower J_sc. With a FF of 0.811, the Si sub-cell reaches a PCE of 6.67%, leading to a cumulative tandem efficiency of 23.1% in combination with the reflective perovskite sub-cell. This is higher than both sub-cells working alone and one of the highest efficiencies reported for a perovskite/Si tandem. The perovskite sub-cell used here possesses a lower PCE compared with the recorded highest PCE of 22.1%.¹⁰ As the perovskite sub-cell efficiency is not compromised in this reflective tandem configuration, we can expect a total efficiency of 28.5% if the highest efficiency perovskite cell is applied even with the modest Si cell we used here. Through further optimizing the parasitic optical loss and reflection light management, a high PCE of 30% could be attained. Besides, the Si module area used in this reflective tandem is equal to the direct illumination area. When a bifacial Si cell is employed, the Si cell can receive the reflected light from both sides. This can further reduce the Si module area by 50%, which provide the most cost-effective way to construct a photovoltaic system.

Table 2 J–V parameters of Si cells, both unfiltered and filtered using different perovskite cells

Si cell	J sc (mA cm^−2)	V oc (V)	FF (%)	PCE (%)	Tandem PCE (%)
Unfiltered	39.58	0.665	80.9	21.3
Filtered by Br0.1	13.03	0.631	81.1	6.67	23.1
Filtered by Br0.2	14.43	0.633	81.3	7.43	22.8
Filtered by Br0.3	14.83	0.638	80.8	7.66	22.2

---

Since the Si bandgap is 1.1 eV, the “top cell” requires a wide bandgap in the range of 1.6–1.8 eV in order to obtain maximum tandem efficiency.¹⁹ We investigated Si cells in tandem with a range of wider bandgap perovskite materials, and their parameters are summarized in Table 2. There is a trade-off between the two sub-cell efficiencies because the incident light is split between them. The wider bandgap perovskite cells possess a lower efficiency because they absorb a narrower range and reflect more of the longer wavelength light which gives the Si sub-cells higher efficiencies. However, their tandem efficiencies dropped compared with smaller bandgap perovskites due to the substantial optical loss.

Conclusions

In summary, we reported an efficient perovskite/Si tandem in a reflective configuration that is easily integrated with each other. Inverted perovskite solar cells with efficiencies of up to 16% have been fabricated. The high reflection properties of the perovskite cell act as a spectral filter, utilizing the high energy photons while reflecting the sub-band photons to the Si cell. We achieved a total efficiency of 23.1% using four-terminal measurements. These results show the promise of employing perovskite solar cells in a reflective tandem to obtain a high efficiency of up to 30%. Furthermore, this reflective tandem could dramatically reduce material costs through employing a bifacial Si cell while retaining a high efficiency.

Experimental

Materials and synthesis

Poly(N,N′-bis(4-butylphenyl)-N,N′-bis(phenyl)-benzidine) (poly-TPD) and bathocuproine (BCP) were purchased from American Dye Source and Lumtec, respectively. PbI₂ and PbBr₂ were purchased from TCI (Tokyo Chemical Industry). PCBM (60) was purchased from Nano-C. Formamidinium iodide (FAI) and methylammonium iodide (MAI) were synthesized according to the published method. The crude material was dissolved in ethanol and recrystallized at −3 °C overnight. A perovskite precursor solution was prepared through dissolving FAI, MAI, PbI₂ and PbBr₂ at respective stoichiometric ratios in DMF/DMSO (4 : 1 v/v).

Device fabrication

Perovskite solar cells were prepared on a pre-cleaned patterned ITO substrate. Poly-TPD was spin-coated from solution (6 mg mL⁻¹ in chlorobenzene) at 4000 rpm for 40 s, followed by baking at 120 °C for 20 min. Prior to perovskite coating, the poly-TPD surface was treated in a UV-ozone cleaner for 10 s. The perovskite precursor solution was spin-coated onto poly-TPD at 1000 rpm for 10 s and 4000 rpm for 30 s; 180 μL of chlorobenzene was dripped on the sample surface after 15 s of the second spin-coating step. Subsequently, the perovskite layer was annealed at 80 °C for 5 min, 120 °C for 5 min and 180 °C for 30 min. The PCBM solution was coated from a chlorobenzene solution (20 mg mL⁻¹) at 1000 rpm for 60 s. The devices were completed through evaporating BCP (6 nm) and Ag (120 nm) sequentially under a high vacuum (1 × 10⁻⁶ mbar). The active area was 7 mm², as defined by the overlap between the back electrode and ITO.

Characterization

The absorption, transmittance and reflection were measured using an Agilent Cary 5000 UV-Vis-NIR Spectrometer. The J–V characteristics were measured using a Keithley 2400 sourcemeter under the illumination of a solar simulator under AM 1.5G conditions. The light intensity was calibrated with a standard Si photodiode. The EQE spectra were collected through combining a monochromated 450 W xenon lamp with an SR830 Lock-in amplifier. For tandem measurements, a passivated emitter rear contact (PERC) Si cell was used with an exposed area of 2 × 2 cm². The J–V measurements for each cell were carried out separately, while the illumination area was defined using an aperture.

Acknowledgements

Y. M. L. acknowledges financial support from an MOE AcRF Tier 1 grant (RG99/14). This work was supported by the National Natural Science Foundation of China (51272033, 51572037 and 51335002), the Priority Academic Program Development of Jiangsu Higher Education Institutions, and the Natural Science Foundation of the Jiangsu Higher Education Institutions of China (Grant No. 14KJA430001 and EEKJA48000).

Notes and references

1. W. S. Yang, J. H. Noh, N. J. Jeon, Y. C. Kim, S. Ryu, J. Seo and S. I. Seok, Science, 2015, 348, 1234–1237.
2. W. Chen, Y. Wu, Y. Yue, J. Liu, W. Zhang, X. Yang, H. Chen, E. Bi, I. Ashraful and M. Grätzel, Science, 2015, 350, 944–948.
3. A. Kojima, K. Teshima, Y. Shirai and T. Miyasaka, J. Am. Chem. Soc., 2009, 131, 6050–6051.
4. T. Salim, S. Sun, Y. Abe, A. Krishna, A. C. Grimsdale and Y. M. Lam, J. Mater. Chem. A, 2015, 3, 8943–8969.
5. H. Zhou, Q. Chen, G. Li, S. Luo, T.-b. Song, H.-S. Duan, Z. Hong, J. You, Y. Liu and Y. Yang, Science, 2014, 345, 542–546.
6. J. M. Ball, M. M. Lee, A. Hey and H. J. Snaith, Energy Environ. Sci., 2013, 6, 1739–1743.
7. J. Burschka, N. Pellet, S.-J. Moon, R. Humphry-Baker, P. Gao, M. K. Nazeeruddin and M. Grätzel, Nature, 2013, 499, 316–319.
8. M. Liu, M. B. Johnston and H. J. Snaith, Nature, 2013, 501, 395–398.
9. A. Mei, X. Li, L. Liu, Z. Ku, T. Liu, Y. Rong, M. Xu, M. Hu, J. Chen and Y. Yang, Science, 2014, 345, 295–298.
10. NREL record solar cell efficiency table, http://www.nrel.gov/ncpv/images/efficiency_chart.jpg, accessed: August, 2016.
11. E. Edri, S. Kirmayer, D. Cahen and G. Hodes, J. Phys. Chem. Lett., 2013, 4, 897–902.
12. J. H. Noh, S. H. Im, J. H. Heo, T. N. Mandal and S. I. Seok, Nano Lett., 2013, 13, 1764–1769.
13. Y. Ogomi, A. Morita, S. Tsukamoto, T. Saitho, N. Fujikawa, Q. Shen, T. Toyoda, K. Yoshino, S. S. Pandey and T. Ma, J. Phys. Chem. Lett., 2014, 5, 1004–1011.
14. N. K. Noel, S. D. Stranks, A. Abate, C. Wehrenfennig, S. Guarnera, A.-A. Haghighirad, A. Sadhanala, G. E. Eperon, S. K. Pathak and M. B. Johnston, Energy Environ. Sci., 2014, 7, 3061–3068.
15. G. E. Eperon, S. D. Stranks, C. Menelaou, M. B. Johnston, L. M. Herz and H. J. Snaith, Energy Environ. Sci., 2014, 7, 982–988.
16. F. C. Hanusch, E. Wiesenmayer, E. Mankel, A. Binek, P. Angloher, C. Fraunhofer, N. Giesbrecht, J. M. Feckl, W. Jaegermann and D. Johrendt, J. Phys. Chem. Lett., 2014, 5, 2791–2795.
17. D. P. McMeekin, G. Sadoughi, W. Rehman, G. E. Eperon, M. Saliba, M. T. Hörantner, A. Haghighirad, N. Sakai, L. Korte and B. Rech, Science, 2016, 351, 151–155.
18. C. D. Bailie, M. G. Christoforo, J. P. Mailoa, A. R. Bowring, E. L. Unger, W. H. Nguyen, J. Burschka, N. Pellet, J. Z. Lee and M. Grätzel, Energy Environ. Sci., 2015, 8, 956–963.
19. S. Essig, M. A. Steiner, C. Allebé, J. F. Geisz, B. Paviet-Salomon, S. Ward, A. Descoeudres, V. LaSalvia, L. Barraud and N. Badel, IEEE J. Photovolt., 2016, 6, 1012–1019.
20. C.-W. Chen, S.-Y. Hsiao, C.-Y. Chen, H.-W. Kang, Z.-Y. Huang and H.-W. Lin, J. Mater. Chem. A, 2015, 3, 9152–9159.
21. P. Löper, B. Niesen, S.-J. Moon, S. M. De Nicolas, J. Holovsky, Z. Remes, M. Ledinsky, F.-J. Haug, J.-H. Yum and S. De Wolf, IEEE J. Photovolt., 2014, 4, 1545–1551.
22. T. P. White, N. N. Lal and K. R. Catchpole, IEEE J. Photovolt., 2014, 4, 208–214.
23. Y. Jiang, I. Almansouri, S. Huang, T. Young, Y. Li, Y. Peng, Q. Hou, L. Spiccia, U. Bach and Y.-B. Cheng, J. Mater. Chem. C, 2016, 4, 5679–5689.
24. S. Albrecht, M. Saliba, J.-P. Correa-Baena, K. Jäger, L. Korte, A. Hagfeldt, M. Grätzel and B. Rech, J. Opt., 2016, 18, 064012.
25. S. Sun, T. Salim, N. Mathews, M. Duchamp, C. Boothroyd, G. Xing, T. C. Sum and Y. M. Lam, Energy Environ. Sci., 2014, 7, 399–407.
26. J. You, Z. Hong, Y. M. Yang, Q. Chen, M. Cai, T.-B. Song, C.-C. Chen, S. Lu, Y. Liu, H. Zhou and Y. Yang, ACS Nano, 2014, 8, 1674–1680.
27. B. Chen, Y. Bai, Z. Yu, T. Li, X. Zheng, Q. Dong, L. Shen, M. Boccard, A. Gruverman and Z. Holman, Adv. Energy Mater., 2016, 6, 1601128.
28. F. Fu, T. Feurer, T. Jäger, E. Avancini, B. Bissig, S. Yoon, S. Buecheler and A. N. Tiwari, Nat. Commun., 2015, 6, 8932.
29. S. Albrecht, M. Saliba, J. P. C. Baena, F. Lang, L. Kegelmann, M. Mews, L. Steier, A. Abate, J. Rappich and L. Korte, Energy Environ. Sci., 2016, 9, 81–88.
30. J. Werner, L. Barraud, A. Walter, M. Bräuninger, F. Sahli, D. Sacchetto, N. Tétreault, B. Paviet-Salomon, S.-J. Moon and C. Allebé, ACS Energy Lett., 2016, 1(2), 474–480.
31. T. Duong, N. Lal, D. Grant, D. Jacobs, P. Zheng, S. Rahman, H. Shen, M. Stocks, A. Blakers and K. Weber, IEEE J. Photovolt., 2016, 6, 679–687.
32. J. Werner, G. Dubuis, A. Walter, P. Löper, S.-J. Moon, S. Nicolay, M. Morales-Masis, S. De Wolf, B. Niesen and C. Ballif, Sol. Energy Mater. Sol. Cells, 2015, 141, 407–413.
33. B. W. Schneider, N. N. Lal, S. Baker-Finch and T. P. White, Opt. Express, 2014, 22, A1422–A1430.
34. J. Werner, J. Geissbühler, A. Dabirian, S. Nicolay, M. Morales-Masis, S. D. Wolf, B. Niesen and C. Ballif, ACS Appl. Mater. Interfaces, 2016, 8, 17260–17267.
35. B. V. Andersson, N.-K. Persson and O. Inganäs, J. Appl. Phys., 2008, 104, 124508.
36. K. Tvingstedt, V. Andersson, F. Zhang and O. Inganäs, Appl. Phys. Lett., 2007, 91, 123514.
37. V. Andersson, K. Tvingstedt and O. Inganäs, J. Appl. Phys., 2008, 103, 094520.
38. M. Peters, J. C. Goldschmidt, T. Kirchartz and B. Bläsi, Sol. Energy Mater. Sol. Cells, 2009, 93, 1721–1727.
39. H. Uzu, M. Ichikawa, M. Hino, K. Nakano, T. Meguro, J. L. Hernández, H.-S. Kim, N.-G. Park and K. Yamamoto, Appl. Phys. Lett., 2015, 106, 013506.
40. R. Sheng, A. W. Ho-Baillie, S. Huang, M. Keevers, X. Hao, L. Jiang, Y.-B. Cheng and M. A. Green, J. Phys. Lett., 2015, 6, 3931–3934.
41. A. Goetzberger, J. Goldschmidt, M. Peters and P. Löper, Sol. Energy Mater. Sol. Cells, 2008, 92, 1570–1578.
42. M. E. Ellion, High efficiency photovoltaic assembly, World Pat., 8701512, 1987.
43. D. Grant, K. Weber, M. Stocks and T. P. White, in Light, Energy and the Environment 2015, OSA Technical Digest (online) (Optical Society of America, 2015), paper PTh3B.3.
44. A. Imenes and D. Mills, Sol. Energy Mater. Sol. Cells, 2004, 84, 19–69.
45. C. Yi, J. Luo, S. Meloni, A. Boziki, N. Ashari-Astani, C. Grätzel, S. M. Zakeeruddin, U. Röthlisberger and M. Grätzel, Energy Environ. Sci., 2016, 9, 656–662.
46. Z. Li, M. Yang, J.-S. Park, S.-H. Wei, J. J. Berry and K. Zhu, Chem. Mater., 2015, 28, 284–292.
47. J. W. Lee, D. H. Kim, H. S. Kim, S. W. Seo, S. M. Cho and N. G. Park, Adv. Energy Mater., 2015, 5, 1501310.
48. M. Filipič, P. Löper, B. Niesen, S. De Wolf, J. Krč, C. Ballif and M. Topič, Opt. Express, 2015, 23, A263–A278.
49. N. J. Jeon, J. H. Noh, W. S. Yang, Y. C. Kim, S. Ryu, J. Seo and S. I. Seok, Nature, 2015, 517, 476–480.
50. D. Bi, W. Tress, M. I. Dar, P. Gao, J. Luo, C. Renevier, K. Schenk, A. Abate, F. Giordano and J.-P. C. Baena, Sci. Adv., 2016, 2, e1501170.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6tc04510c

‡ These authors contributed equally to this work.

---