The Cu₂O/CuO/SnO₂ transparent pn junction film device towards photovoltaic enhancement with Cu²⁺ self-oxidation transition layer

Abstract

The Cu₂O/CuO/SnO₂ transparent pn junction film device with Cu²⁺ self-oxidation transition layer is prepared via a simple thermal oxidation-sputtering method. There, the Cu₂O is prepared via a simple sputtering method, the CuO transition layer is prepared via a thermal self-oxidation method on the surface of Cu₂O film, and subsequently the SnO₂ film is deposited via the sputtering method on the surface of CuO transition layer. As revealed, the as-prepared transparent photovoltaic device exhibits highly transparency of about ~ 75%, obvious photovoltaic conversion enhancement of about ~ 450 folds than unmodified device, decent stability during 10000 s’ cycle, which can be mainly ascribed to the Cu²⁺ self-oxidation transition layer; there, the appropriate Fermi level and nicer lattice matching can provide a decent channel for charge carrier transport, and the visible light response can improve photo-generated carrier excitation and injection. Additionally, the structure design of Cu₂O/CuO/SnO₂ can prevent the oxidation of Cu₂O to increase photovoltaic stability.

Graphical abstract

Introduction

Due to the rapid consumption and non-renewability of fossil fuels, the energy crisis is becoming a more and more serious problem. At the same time, with the rapid industrial development, the greenhouse effect caused by fossil fuel combustion also brings more and more challenges for environment [1, 2]. So, the exploration of efficient alternatives for fossil fuels would be extremely urgent. There, the renewable energy is regarded as an efficient candidate for solving this problem [3,4,5,6], including water energy, wind energy, solar energy, biomass energy or tidal energy, etc. [7, 8]. Among these, the solar cells or photovoltaic devices [9,10,11,12], with the convenience, environmental friendly and can directly convert solar energy into electric energy, have attracted a great of attentions.

For traditional solar cells, most of them mainly focus on light response [13, 14], and the dark semiconductors would be primary candidates, so that which encounters prominent limitations in practical applications, including building walls or windows. Herein, the transparent solar cells or photovoltaic devices, with the balanced transmittance and photovoltaic conversion [15,16,17,18,19], are regarded as ideal solutions, such as ZnO-based device, TiO₂-based device and so on.

Among these, the SnO₂-based device, with a wide band gap (3.6 eV) and decent intrinsic highly conductivity [20,21,22], is reported as a hot topic, such as Cheema et al. have prepared transparent SnO₂-based dye-sensitized solar cells with remarkable photon conversion performance [23], Jia et al. have reported decent perovskite solar cells with SnO₂ transport layers [24], etc. In addition, the stale physical–chemical performances of SnO₂ would be additional advantages [25, 26].

However, the transparent device is still a difficult topic, because that, the opportune and matched p-type semiconductor would be a great challenge for the fabrication of pn junction [27, 28], especially the low intrinsic mobility and carrier concentration would be the most significant issues [29,30,31], for example, SnO, CuAlO₂, etc. Herein, as a typical p-type direct band gap semiconductor, the Cu₂O [32], with easy preparation, low cost, non-toxic and decent transmittance (2.0–2.6 eV), is reported as an outstanding candidate [33, 34], such as Yun and Kim groups have reported the Cu₂O/ZnO/AZO photovoltaic device for transparent applications [35], Osorio-Rivera et al. have prepared the Cd₂SnO₄/CdS/Cu₂O/Ag solar cell with remarkable photovoltaic performance [36], Wong groups have used the metal oxide frameworks for transparent photovoltaic devices [37], etc. Additionally, the Cu vacancy and interstitial oxygen can induce the formation of excess holes charge carriers [38, 39], which is regarded as another advantage for improving the photovoltaic efficiency.

For all this, the pn junction device is still hard to be applied in windows, because that the low intrinsic charge carriers caused by p-type semiconductor and great energy band alignment from different materials would restrict the photovoltaic conversion significantly [40,41,42]. That the efficient hole carrier donor or band gap adjustment is regarded as effective approach [43,44,45], such as element doping and plasma injection. There, the transition layer modification [46], with easy preparation and sufficient candidates, is reported as an efficient method. Especially that, the appropriate transition layer [47, 48], can not only act as a donor to inject sufficient charge carriers, but also provide a decent channel for charge carriers transport, for instance, Qin et al. have prepared the Cu₂O/SnO₂ pn junction solar cell with MgO transition layer [49], Azizi groups have fabricated the Cu₂O/ZnO pn junctions with ZnS transition layers [50], etc. For this Cu₂O/SnO₂ pn junction, the CuO [51], self-oxidation by the Cu₂O, would obtain a unique advantage, because that, compared with other transition layers, due to the homogeneous oxidation, the CuO formed at the surface of Cu₂O would cause minimal lattice distortion [52,53,54,55], which can improve the pn junction crystalline quality to increase the photovoltaic conversion [56, 57]. In addition, CuO, obtains a narrower band gap of 1.3–2.0 eV [58], can increase light response for increasing the charge carriers injection further, and the good physical–chemical stability would be another important advantage [59]. What’s more, the ultrathin thickness can shorten the transport time for restrain the photo-generated charge carrier recombination, which is also an important reason [60, 61].

In this work, we have fabricated the Cu₂O/CuO/SnO₂ transparent pn junction film device with Cu²⁺ self-oxidation transition layer via a simple thermal oxidation-sputtering method. The results indicate that the as-prepared Cu₂O/CuO/SnO₂ transparent pn junction film device obtains a highly transparency of about ~ 75% in visible light, an efficient photovoltaic conversion enhancement of about ~ 450 folds than unmodified device and a decent stability during 10000 s’ cycle. Further, the mechanism of photovoltaic enhancement of Cu₂O/CuO/SnO₂ transparent pn junction device is investigated via series of electrochemical characterizations.

Experiment

The Cu₂O/CuO/SnO₂ transparent pn junction film device with Cu²⁺ self-oxidation transition layer was prepared via a simple thermal oxidation-sputtering method. There, the Cu₂O film was prepared via a sputtering method, and then the CuO ultrathin layer was prepared by thermal self-oxidation of Cu₂O film surface in air at 180 °C. Subsequently, the SnO₂ film was prepared via a sputtering method on surface of CuO film. The process is displayed as Fig. 1. Herein, the transparent pn junction devices with different oxidation time were labelled as Cu₂O/SnO₂-CuO-0 (0 min), Cu₂O/SnO₂-CuO-1 (15 min), Cu₂O/SnO₂-CuO-2 (30 min), Cu₂O/SnO₂-CuO-3 (45 min) and Cu₂O/SnO₂-CuO-4 (60 min), respectively. The detail of preparation and characterization is shown in ESI.

Figure 1

The process diagram of Cu₂O/CuO/SnO₂ transparent pn junction film device with Cu²⁺ self-oxidation transition layer

Results and discussions

Figure 2 is the XRD of Cu₂O/CuO/SnO₂ transparent devices with different ratios of CuO transition layers. As displayed, due to the ultrathin thickness and lower crystallinity, the peaks of Cu₂O and SnO₂ could be hardly observed and would be characterized by the HRTEM and HRXPS. However, it’s interesting that, with the increased oxidation time, the weak diffraction peaks at 31.66° and 35.21° exhibit an increasing and are ascribed to the (110) and (002) planes of CuO(PDF#48-1548) [51, 52], respectively. Further, no other diffraction peaks could be observed, which manifest that the thermal oxidation process would not introduce other impurity.

Figure 2

The XRD of different Cu₂O/CuO/SnO₂ transparent pn junction film devices

Figure 3 is the SEM of Cu₂O/CuO/SnO₂ pn junction film device, including the cross section. As revealed, the Cu₂O (Fig. 3a) prepared by sputtering obtains a smooth and uniform surface. Then, with the formation of CuO transition layer (Fig. 3b) by thermal oxidation, there is no obvious change could be observed, which can be ascribed to that the ultrathin layer and decent lattice matching of CuO. Figure 3c is the surface of SnO₂ film, as shown, which is smooth and uniform and is beneficial for the fabrication of pn junction film device. Figure 3d is the cross section of as-prepared ternary pn junction film device. As demonstrated, the Cu₂O/CuO/SnO₂ device obtains a decent interfaces, such high-quality pn junction is beneficial for charge carrier transport. It’s significant that, restricted by the ultra-micro structure and distinguishability, the CuO transition layer could be hardly observed in SEM, and would be characterized by HRTEM. By calculation, the thicknesses of SnO₂ and Cu₂O films are about ~ 200 nm and ~ 100 nm, respectively.

Figure 3

The SEM of Cu₂O/CuO/SnO₂ pn junction film device(Cu₂O/SnO₂-CuO-3) at different stage, a the surface of Cu₂O film, b the surface of CuO transition layer, c the surface of SnO₂ film and d the cross section of Cu₂O/SnO₂-CuO-3

Figure 4a is the TEM of Cu₂O/CuO/SnO₂ pn junction film device (Cu₂O/SnO₂-CuO-3) interface. As revealed, such pn junction obtains a clear and high-quality interface, which corresponds to SEM. As demonstrated in HRTEM, the lattice spaces of 0.264 nm (Fig. 4b) and 0.248 nm (Fig. 4d) are ascribed to the (101) plane of SnO₂ [29] and (111) plane of Cu₂O [33], respectively. It’s significant that, the CuO transition layer can be observed at the interface between Cu₂O and SnO₂ layers, by fitting, the corresponding lattice space of 0.254 nm (Fig. 4c) is ascribed to the (002) plane of CuO [51, 52], which is an reasonable evidence for the presence of CuO transition layer.

Figure 4

The TEM of Cu₂O/CuO/SnO₂ pn junction film device (Cu₂O/SnO₂-CuO-3), a the interface of Cu₂O/SnO₂-CuO-3, b the HRTEM of SnO₂, c the HRTEM of CuO and d the HRTEM of Cu₂O

HRXPS reveals the different chemical valences of each component, especially the presence of CuO. As revealed in Fig. 5a, the peaks at 933.09 eV and 952.96 eV are ascribed to Cu 2p_3/2 and Cu 2p_1/2 of Cu⁺, and the peaks at 934.66 eV and 954.59 eV are ascribed to Cu 2p_3/2 and Cu 2p_1/2 of Cu²⁺ [29, 51, 52], such results indicate that the successful deposition of Cu₂O film and formation of CuO transition layer. Figure 5b is the HRXPS of Sn 3d; there, the peaks at 486.92 eV and 494.92 eV are ascribed to Sn⁴⁺ [29], which manifest that the successful deposition of SnO₂ film. All above indicate that the Cu₂O/CuO/SnO₂ pn junction film device with Cu²⁺ self-oxidation transition layer has been fabricated successfully.

Figure 5

The HRXPS of Cu₂O/CuO/SnO₂ pn junction film device (Cu₂O/SnO₂-CuO-3), a Cu 2p, b Sn 3d

As revealed, due to the wide band gap of Cu₂O and SnO₂, the as-prepared Cu₂O/SnO₂ pn junction exhibits a highly transparency at visible light (Fig. 6), and because the band gap of Cu₂O is narrower than SnO₂, the absorption edge mainly depends on Cu₂O at about ~ 550 nm [29, 51, 52]. Further, with the formation of CuO self-oxidation transition layer, the transparency of modified pn junction device exhibits a certain decrease, which is ascribed to the narrow band gap of CuO [46, 47]. By calculation (ESI Fig. S1), the corresponding band gaps are 2.274 eV (Cu₂O), 1.673 eV (CuO) and 3.626 eV (SnO₂), respectively. However, most of Cu₂O/CuO/SnO₂ pn junction film devices with CuO transition layer still remain a highly transparency of about ~ 75% in visible light, which would not influence its application in windows and can be supported by the clear and transparent photo in insert.

Figure 6

The optical transmittance spectra of different Cu₂O/CuO/SnO₂ pn junction film device inset the photo of transparent device

Figure S2 (ESI) is the I–V curve of Cu₂O/SnO₂-CuO-3 under illumination, as shown, which manifest that the as-prepared transparent device is a typical pn junction [62]. Additionally, all of as-prepared Cu₂O/CuO/SnO₂ pn junction film devices exhibit similar open-circuit voltage of about ~ 150 mV, which is another important factor for photovoltaic conversion.

Herein, the photovoltaic conversion performances of different Cu₂O/CuO/SnO₂ pn junction film devices are demonstrated in Fig. 7. As displayed, the as-prepared pn junction device without CuO transition layer (Cu₂O/SnO₂-CuO-0) exhibits a weak intrinsic photocurrents of about ~ 6.62 × 10^–7 A. Subsequently, with the formation of CuO transition layer, the photocurrents exhibit obvious enhancement to about ~ 4.40 × 10^–5 A (Cu₂O/SnO₂-CuO-1), 1.16 × 10^–4 A (Cu₂O/SnO₂-CuO-2), 2.92 × 10^–4 A (Cu₂O/SnO₂-CuO-3) and 1.95 × 10^–4 A (Cu₂O/SnO₂-CuO-4), respectively. By calculation, the Cu₂O/CuO/SnO₂ pn junction film device (Cu₂O/SnO₂-CuO-3) exhibits significant enhancement of about ~ 450 folds than unmodified device, and the PEC (photovoltaic conversion efficiency) is about 1.17%, that outclasses the increased absorption and can be regarded as a decent result in transparent device, which manifests that the charge carrier transport would play a more important role than absorption. It’s interesting that, the excess CuO would lead a decrease in photovoltaic conversion, because the CuO itself obtains weak conductivity and blocked up transition layer would lead an interface scattering for charge carrier, so that the photocurrent of Cu₂O/SnO₂-CuO-4 exhibits a decreasing. The response time can be fitted by the formulas [62] of $\Delta\sigma =\Delta {\sigma }_{s}(1-{e}^{-\frac{t}{\tau }})$ and $\Delta\sigma =\Delta {\sigma }_{s}{e}^{-\frac{t}{\tau }}$; there, the rise time is 68 ms, and decay time is 63 ms, respectively.

Figure 7

The photocurrents of different Cu₂O/CuO/SnO₂ pn junction film devices

Subsequently, the photovoltaic stability is evaluated by the recycle I-T curve in Fig. 8. As revealed, during the 10000 s’ cycle, the as-prepared Cu₂O/CuO/SnO₂ pn junction film device (Cu₂O/SnO₂-CuO-3) exhibits a decent stability, including the photocurrent and response time, which indicate that this transparent pn junction device is promising to be applied in windows.

Figure 8

The recycle I-T curve of Cu₂O/SnO₂-CuO-3

Additionally, the illumination depending photovoltaic response is another important factor for practical application and is revealed in Fig. 9. It’s interesting that with the decreased illumination of 0.1 solar, the photocurrent exhibits a nonlinear descent, by calculation, at the intensity of 0.1 solar, the transparent pn junction (Cu₂O/SnO₂-CuO-3) can still remain a higher photovoltaic conversion about 30% of standard value, which is regarded as a decent result.

Figure 9

The illumination depending photovoltaic response of Cu₂O/SnO₂-CuO-3

All above indicate that the formation of self-oxidation CuO transition layer can improve the photovoltaic conversion efficiency efficiently, so the mechanism of photovoltaic enhancement would be an interesting issue.

As revealed in Fig. 10a, the decreased arcs indicate that, with the formation of CuO transition layer, the interface charge carrier transferring exhibits an obvious improving and obtains the optimal value at Cu₂O/SnO₂-CuO-3, which corresponds the photovoltaic performance and indicate the CuO transition layer can accelerate the charge carrier transport [42]. As demonstrated in Fig. 10b, fitted by the formula [62] of $\tau_{n} = \frac{1}{{2\pi f_{m} }}$, the charge carrier lifetime exhibits prolonging with the formation of CuO transition layer and corresponds to the EIS (optimal value at Cu₂O/SnO₂-CuO-3). Such results also can be supported by the PL (ESI Fig. S3). As displayed, with the introducing of CuO transition layer, the fluorescence obtains significant decrease, which indicates that the interface charge carriers recombination is optimized and corresponds to photovoltaic performance. All above manifest that the interface charge carrier efficiency obtains an improving with the formation of CuO transition layer, including the transport, recombination and lifetime.

Figure 10

a The EIS Nyquist plots and b the phase angle Bode plots of different Cu₂O/CuO/SnO₂ pn junction film devices

The charge carrier transport route would be another significant factor for photovoltaic conversion, which depends on the potential and can be evaluated by the Mott–Schottky plot via the formula [42, 62, 63] of

$$\frac{{1}}{{C^{{2}} }} = \pm \frac{{2}}{{{\text{e}}\varepsilon \varepsilon_{{0}} N_{{\text{d}}} }}\left( {E - E_{{{\text{FB}}}} - \frac{{K_{{\text{b}}} T}}{{\text{e}}}} \right)$$

where the x-axis intercept is the E_fb (band potential, corrected by AgCl, vs. 0.197 eV), and the slope denotes the type of semiconductor. As demonstrated, the negative slopes in Fig. 11a and b manifest the Cu₂O and CuO are typical p-type semiconductors, and the positive slope in Fig. 11c manifests the SnO₂ is typical n-type semiconductor. As displayed, the flat band potentials of different films are 0.723 eV (VB of Cu₂O, vs. Ag/AgCl), 0.629 eV (VB of CuO, vs. Ag/AgCl) and -0.083 eV (CB of SnO₂, vs. Ag/AgCl), respectively. By correction, the corresponding Fermi levels are 0.920 eV (VB of Cu₂O, vs. NHE), 0.826 eV (VB of CuO, vs. NHE) and 0.114 eV (CB of SnO₂, vs. NHE). Incorporation with the intrinsic band gap (ESI Fig. S1), the CB of the Cu₂O and CuO is -1.354 eV (vs. NHE) and -0.847 eV (vs. NHE), and the VB of SnO₂ is 3.740 eV (vs. NHE). It’s obviously, such Fermi levels are hard to form a decent channel for charge carrier transferring, so the Mott–Schottky plot of as-prepared transparent pn junction device (Cu₂O/SnO₂-CuO-3) is revealed in Fig. 11d. By correction, the corresponding Fermi levels of each component are 0.295 eV (VB of Cu₂O, vs. NHE), 0.486 eV (VB of CuO, vs. NHE) and -0.045 eV (CB of SnO₂, vs. NHE). Correspondingly, the CB of Cu₂O and CuO is -1.979 eV (vs. NHE) and -1.187 eV (vs. NHE), and the VB of SnO₂ is 3.581 eV (vs. NHE). It’s evident that the E_CB-Cu2O < E_CB-CuO < E_CB-SnO2 and E_VB-Cu2O < E_VB-CuO < E_VB-SnO2. It’s obvious that, with the presence of Cu²⁺ self-oxidation transition layer, the appropriate potential of CuO would make the photo-generated electron transport channel from Cu₂O to CuO to SnO₂ being achievable [42]. Correspondingly, the photo-generated hole would transfer along the channel of SnO₂ to CuO to Cu₂O [29, 62]. Above results manifest that the Cu²⁺ self-oxidation CuO transition layer with an appropriate potential can provide a decent channel for accelerating photo-generated carrier transport, which is beneficial for photovoltaic conversion.

Figure 11

The Mott–Schottky plots of a Cu₂O, b CuO, c SnO₂ and d Cu₂O/SnO₂-CuO-3

Thus, the photovoltaic enhancement mechanism of the Cu₂O/CuO/SnO₂ transparent pn junction film device can be proposed as Fig. 12, which is mainly ascribed to that the appropriate potential of Cu²⁺ self-oxidation transition layer and excess visible light response of CuO.

Figure 12

The photovoltaic enhancement mechanism of Cu₂O/CuO/SnO₂ transparent pn junction film device

As shown, the Cu₂O/SnO₂-CuO-3 obtains a significant photovoltaic enhancement (~ 450 folds than unmodified device), but a few decreased transparency (15%), which indicates that the charge carrier transport improved via Cu²⁺ self-oxidation transition layer would play a more important role than the visible light response during the photovoltaic conversion [42]. There, with the formation of Cu₂O/SnO₂ pn junction, the intrinsic potential structure is hard for photo-generated carriers transferring. That, with the formation of Cu²⁺ self-oxidation transition layer, the appropriate potential of CuO would fabricate a decent gradient [29, 51, 52], which would provide a favourable channel for charge carriers transferring, such process can improving the photo-generated carrier separation/recombination, transport and prolong the lifetime efficiently for improving photovoltaic conversion, and can be supported by the EIS, PL, Bode and Mott–Schottky plots. There, as depicted in Fig. 12, driven by self-photovoltage, the photo-generated electron would transfer along the route of Cu₂O to CuO to SnO₂; correspondingly, the photo-generated hole would transfer to Cu₂O, this process is the core issue for this transparent pn junction device [51, 52, 62, 63]. In addition, the Cu₂O and SnO₂ with wider band gap exhibit weaker response in visible light, but the CuO with narrower band gap can absorption visible light to excite excess photo-generated charge carriers and inject into pn junction, which is also an important advantage [23,24,25, 51, 52]. Further, with the decent adjustment of thickness, such CuO transition layer would not influence the transparency for transparent device application. On the other hand, the decent lattice matching of Cu²⁺ self-oxidation transition layer would provide a decent channel for charge carrier transport, though the excess transition layer would lead an interface scattering [29, 51, 52], so that the Cu₂O/SnO₂-CuO-4 exhibits a decrease. What’s more, the structure design of Cu₂O/CuO/SnO₂ can prevent the oxidation of Cu₂O film efficiently, which is an important reason for the decent photovoltaic stability.

Conclusion

We have prepared the Cu₂O/CuO/SnO₂ transparent pn junction film device with Cu²⁺ self-oxidation transition layer via a simple thermal oxidation-sputtering method. The as-prepared Cu₂O/SnO₂-CuO-3 obtains highly transparency of about ~ 75%, obvious photovoltaic conversion enhancement of about ~ 450 folds than unmodified device, decent stability during 10000 s’ cycle. There, the Cu²⁺ self-oxidation transition layer with appropriate Fermi level and decent lattice matching can provide a decent channel for charge carrier transport, and the visible light response can improve excess photo-generated carrier injection are regarded as the main reasons. In addition, the structure design of Cu₂O/CuO/SnO₂ can prevent the oxidation of Cu₂O film to increase the photovoltaic stability.

Thus, this Cu₂O/CuO/SnO₂ transparent pn junction film device with balanced transparency and photovoltaic performance is expected to be a promising candidate in windows.