Cu/Cu2O nanocomposite films as a p-type modified layer for efficient perovskite solar cells

Cu/Cu2O films grown by ion beam sputtering were used as p-type modified layers to improve the efficiency and stability of perovskite solar cells (PSCs) with an n-i-p heterojunction structure. The ratio of Cu to Cu2O in the films can be tuned by the oxygen flow ratio (O2/(O2 + Ar)) during the sputtering of copper. Auger electron spectroscopy was performed to determine the elemental composition and chemical state of Cu in the films. Ultraviolet photoelectron spectroscopy and photoluminescence spectroscopy revealed that the valence band maximum of the p-type Cu/Cu2O matches well with the perovskite. The Cu/Cu2O film not only acts as a p-type modified layer but also plays the role of an electron blocking buffer layer. By introducing the p-type Cu/Cu2O films between the low-mobility hole transport material, spiro-OMeTAD, and the Ag electrode in the PSCs, the device durability and power conversion efficiency (PCE) were effectively improved as compared to the reference devices without the Cu/Cu2O interlayer. The enhanced PCE is mainly attributed to the high hole mobility of the p-type Cu/Cu2O film. Additionally, the Cu/Cu2O film serves as a protective layer against the penetration of humidity and Ag into the perovskite active layer.

by controlling the tilting angle of the substrate and the deposition power and period via angular rotation of the substrate during the reactive magnetron sputtering process. This angular rotation prevents bombardment damage on the perovskite, but the efficiency of the solar cell only reached 8.93% 10 .
In this work, we report, for the first time, a p-type Cu/Cu 2 O and a spiro-OMeTAD layer as combinatorial HTMs in n-i-p heterojunction PSCs to improve the hole extraction capability, device performance, and device stability. For a 20-nm-thick Cu/Cu 2 O film with an oxygen flow ratio (OFR) of 60%, a high mobility of 60.5 cm 2 /V-s was obtained, which results in a superior PCE of 17.11% for the PSC, as compared to the standard device with only spiro-OMeTAD as the HTM (PCE of 13.97%). The stability test shows that the Cu/Cu 2 O layer provides enhanced durability under ambient conditions. Hence, the p-type Cu/Cu 2 O composite film is a promising HTM modified layer for highly efficient PSCs and a good protection layer for the perovskite active layer. In addition, to meet the demand for energy level alignments in our devices, insertion of the spiro-OMeTAD layer is necessary to prevent damage of the perovskite during the high energy sputtering of the Cu/Cu 2 O composite layer.

Results
Characterizations of the Cu/Cu 2 O composite films. Figure 1 shows the X-ray diffraction (XRD) patterns of the ion-beam sputtered Cu/Cu 2 O composite films deposited on glass over a 5-min period (approximately 20-nm thick) with the OFR ranging from 0 to 60%. The diffraction from pure Cu in the absence of oxygen is shown at the bottom of Fig. 1(a). The characteristic peaks at 43.4°, 50.45°, and 74.05° can be assigned to the Cu (111), (200), and (220) planes, respectively. For the OFR increasing from 10% to 50%, the intensities of the Cu phases are weaken while the Cu 2 O phase are intensified, suggesting the increasing volume ratio of Cu 2 O in the nanocomposite films. With further increasing OFR to 60%, there is no visible Cu diffraction and the Cu 2 O (220) phase dominates. The scanning electron microscopy (SEM) images of the corresponding Cu/Cu 2 O composite films are provided in the Supporting Information Fig. S1. Owing to the difference in the contrast between Cu and Cu 2 O, a small number of bright spots were observed for the OFR ranging from 30% to 50%, which are likely associated with the existence of Cu (Supporting Information Fig. S1(c-e)). Smooth morphology with a negligible number of bright spots was observed for the Cu 2 O film with an OFR of 60%, as shown in the Supporting Information Fig. S1(f), which indicates that most of Cu was oxidized to form Cu 2 O.
For convenience, the Cu/Cu 2 O composite films grown under an OFR of X% are denoted as Cu 2 O_X in the rest of the paper. To determine the elemental composition of the Cu/Cu 2 O composite films, Auger electron spectroscopy (AES) was performed to monitor the Cu LMM transition, as seen in Fig. 1(b). The peak at 918 eV corresponds to Cu 0 bonding for an OFR of 0%. With the increase of OFR, the characteristic AES peak shifts toward lower kinetic energy (917.5 eV for Cu 2 O_40 and 917.2 eV for Cu 2 O_50), reflecting the increase of the Cu 2 O component in the composite film. For Cu 2 O_60, the characteristic peak at 916.6 eV is very close to the energy of Cu + bonding in Cu 2 O powder (916.2 eV) 33 . This shows that the films produced with high OFR exhibit spectra close to that of pure Cu 2 O.
To investigate the energy level and the relative position of the Fermi level with respect to the valence band maximum (∆E VB ) of the Cu/Cu 2 O composite films, ultraviolet photoelectron spectroscopy (UPS) and photoluminescence (PL) spectroscopy were performed, as shown in Fig. 2. Figure 2(a) shows the full UPS spectra and Fig. 2(b) shows a magnification of the spectra near the valence band region from Fig. 2(a). The Au metal rectifies the Fermi energy level position at 0 eV, and the obtained ∆E VB of the Cu 2 O_60 film is 0.55 eV. The corresponding band gap (E g ) determined from the PL spectra ( Fig. 2(c)) is 2.07 eV and the work function of Au is approximately 4.7 eV 34 . The peak values in Fig. 2(b) correspond to the energies between the valence band maximum (VBM) and the work function, and the VBM position of the Cu 2 O_60 was determined to be approximately −5.25 eV. The conduction band minimum (CBM) of the Cu 2 O_60 is −3.18 eV that was determined by the addition of E g to the VBM. The energy level of the Cu 2 O_60 film along with other commonly-used inorganic HTMs is plotted in Fig. 2  The electrical properties of the Cu/Cu 2 O composite films were analyzed by Hall effect measurements and a four-point probe method, and are summarized in Table 1. Figure S2 presents the Hall effect measurements of Cu/Cu 2 O composite films at variant OFRs. The Cu 2 O_40, Cu 2 O_50, and Cu 2 O_60 films have been determined to be p-type semiconducting materials. As evidenced from the Hall effect measurements shown in Fig. S2(a), for OFR = 30% the Cu/Cu 2 O composite film is a n-type semiconductor, which is not of interest for HTM application. The Cu 2 O_60 film is nearly insulating (sheet resistance ~83 kΩ/square) owing to the high volume ratio of Cu 2 O. The sheet resistances of the Cu 2 O_40 and Cu 2 O_50 films are 72.5 and 210.1 Ω/square, respectively, and the corresponding hole mobilities are 1.87 and 1.23 cm 2 /V-s, which are at least three orders of magnitude higher than the organic hole transport layers of spiro-OMeTAD (1.6 × 10 −4~1 .6 × 10 −3 cm 2 /V-s). Notably, the hole mobility of the Cu 2 O_60 film is as high as 60.5 cm 2 /V-s. The hole carrier concentrations of the Cu 2 O_40, Cu 2 O_50, and Cu 2 O_60 films are 2.3 × 10 22 , and 6.19 × 10 17 cm −3 , respectively.
X-ray photoelectron spectroscopy (XPS) was employed to analyze the chemical compositions of the composite Cu/Cu 2 O films grown with different OFRs. Figure 3(a-d) show the XPS spectra of Cu 2p 3/2 core level. All the films exhibit characteristic peaks at 932.3 eV and 933.1 eV that are respectively corresponding to copper(I) oxide and copper metal as reported in the literature 35,36 . By deconvoluting the signal and integrating the individual area, the ratio of Cu to Cu 2 O for Cu 2 O_30, Cu 2 O_40, Cu 2 O_50 and Cu 2 O_60 films are estimated to be 0.79, 0.50, 0.22 and 0.05, respectively. The Cu/Cu 2 O ratio as a function of oxygen flow ratios is shown in Fig. 3(e), indicating that the Cu content decreases with the oxygen flow. The tendency of Cu content in the sputtered film is consistent with the SEM results.  Devices. The device architecture is based on the n-i-p heterojunction configuration which is composed of FTO/cp-TiO 2 /mp-TiO 2 /CH 3 NH 3 PbI 3 /spiro-OMeTAD/Cu 2 O_X/Ag, as shown in Fig. 4(a). The standard (reference) device is the one without the Cu 2 O_X film interlayer. The active layer of the CH 3 NH 3 PbI 3 perovskite was prepared by solvent engineering 37 , while the Cu 2 O_X interlayer was grown by sputtering on a Cu target by tuning the OFR over a 5 min period. The thickness of the Cu 2 O_X films are approximately 20 nm, as estimated by the cross-sectional SEM images (cf. Fig. S1 in Supporting Information). For convenience and clarity, the energy level diagram of the device is depicted in Fig. 4(b). Figure 5(a) shows the current density-voltage (J-V) curves of the Cu 2 O_X based perovskite solar cells (with active area of 0.09 cm 2 ) prepared with varying OFRs, and the corresponding photovoltaic parameters are summarized in Table 2. Amongst these devices, the Cu 2 O_60 device had a short circuit current density (J sc ) of 22  TiO 2 ranges from 0.1-10 cm 2 /V-s 38,39 , while the hole mobility of Cu 2 O_X is typically between 1-60.5 cm 2 /V-s. Unbalanced ambipolar transport may be the main cause of the hysteresis observed in the prepared devices [40][41][42] . The introduction of spiro-OMeTAD not only protects the perovskite active layer during sputtering deposition but also acts as an electron blocking layer, as schematically illustrated in Fig. 4(b). Figure 5(b) shows the incident photon-to-electron conversion efficiency (IPCE) spectra of the four fabricated devices. The devices with Cu 2 O_40 and Cu 2 O_50 layers show good responses with maximum values of approximately 80% in the range of 400 to 700 nm. The IPCE is approximately 90% for the Cu 2 O_60 incorporated device over the same wavelength range. The IPCE spectra reveal that the Cu 2 O_40, Cu 2 O_50, and Cu 2 O_60 incorporated devices yield higher responses and photocurrents than the standard device. Figure S3 (Supporting Information) shows the statistical distribution of the photovoltaic parameters resulted from 15 devices for Cu 2 O_60-based PSCs as well as 15 standard devices for comparison. The results show that the Cu 2 O_60-based device exhibits better efficiency than the reference devices. We further enlarge active area of 2.06 cm 2 for Cu 2 O_60 based perovskite solar cell which exhibits a J sc of 21.66 mA/cm 2 , V oc of 1.0 V, FF of 42.47%, and PCE of 9.22% and its corresponding J-V curve is shown in Supporting Information Fig. S4. It can be found that the device with large 2.06 cm 2 active area is lower than that of the smaller 0.09 cm 2 active area ones due to the poor FF. The reducing FF is mainly resulted from the increasing sheet resistance of FTO substrate that raises the probability of the charge recombination before carrier extracted by electrode 43,44 .   Stability Test. Finally, we monitor the long-term device stability of the normalized efficiency for the encapsulated Cu 2 O_60-based devices along with the standard cell and the results are shown in Fig. 6(a). When stored in the dark, the efficiency of the standard device drops rapidly after 12 days, while the devices with the Cu/Cu 2 O composite films show long-term stability for at least 30 days. The results demonstrate that p-type Cu/Cu 2 O composite films are promising candidates for the modification of the HTM property for high efficiency PSCs, while simultaneously acting as a protecting layer. The devices with the Cu/Cu 2 O composite film were transferred to a light soaking condition (one sun illumination). The aging test indicated that approximately 67% of the initial efficiency is retained after 50 h, as shown in Fig. 6(b).

Discussion
In summary, p-type Cu/Cu 2 O composite films with a tunable Cu/Cu 2 O ratio were fabricated by controlling the oxygen flow rate during the ion beam sputtering of Cu targets. The resistance and mobility of the Cu/Cu 2 O composite films can be controlled to optimize the device performance of the PSCs with an n-i-p heterojunction configuration. In comparison to the standard device, the PCEs of the devices incorporating Cu/Cu 2 O composite films are significantly improved. Furthermore, the Cu/Cu 2 O composite films can effectively protect the perovskite active layer and enhance the lifetime of the PSCs. Characterization. The crystallographic properties of the films were determined by grazing incidence X-ray diffraction using Cu Kα radiation (λ = 1.5418 Å, D8, Bruker, Germany) at room temperature with a scanning step size of 0.005°. The surface morphology and cross-section of the samples were analyzed by field-emission scanning electron microscopy (SUPRA TM 55). The carrier type, carrier concentration, and electrical resistivity were measured using a four-terminal van der Pauw configuration at room temperature. The UPS and XPS experiment were performed at beamline 24 A of the Taiwan Light Source in the National Synchrotron Radiation Research Center (NSRRC). A microscopic PL system (MRI, Protrustech Co., Ltd., Taiwan) was used to determine the band gap of the films with a pumping wavelength of 532 nm. The J-V measurement was performed using a solar simulator (SS-F5-3A, Enlitech) with AM 1.5 G spectra and the device was connected to a source meter (Keithley 2401) for recording the J-V data. The light intensity was calibrated using reference silicon solar cells to be 100 mW/cm 2 . The scan rate was 1 V/s for the forward scan (from J sc to V oc ). A metal mask with an aperture size of 0.09 cm 2 was used to define the active area. A 300 W intensity monochromatic (Newport Cornerstone 260) xenon lamp (Newport) and a source meter (Keithley 2401) were integrated to measure the IPCE response of the devices.