Luminescent europium-doped titania for efficiency and UV-stability enhancement of planar perovskite solar cells

Journal Pre-proof Luminescent europium-doped titania for efficiency and UV-stability enhancement of planar perovskite solar cells P. Chen, Z. Wang, S. Wang, M. Lyu, M. Hao, M. Ghasemi, M. Xiao, J.-H. Yun, Y. Bai, L. Wang PII:

S2211-2855(19)31106-1

DOI:

https://doi.org/10.1016/j.nanoen.2019.104392

Reference:

NANOEN 104392

To appear in:

Nano Energy

Received Date: 17 September 2019 Revised Date:

30 November 2019

Accepted Date: 8 December 2019

Please cite this article as: P. Chen, Z. Wang, S. Wang, M. Lyu, M. Hao, M. Ghasemi, M. Xiao, J.-H. Yun, Y. Bai, L. Wang, Luminescent europium-doped titania for efficiency and UV-stability enhancement of planar perovskite solar cells, Nano Energy, https://doi.org/10.1016/j.nanoen.2019.104392. This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version will undergo additional copyediting, typesetting and review before it is published in its final form, but we are providing this version to give early visibility of the article. Please note that, during the production process, errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain. © 2019 Published by Elsevier Ltd.

Luminescent europium-doped titania for efficiency and UV-stability enhancement of planar perovskite solar cells Peng Chen, Zhiliang Wang, Songcan Wang, Miaoqiang Lyu, Mengmeng Hao, Mehri Ghasemi, Mu Xiao, Jung-Ho Yun, Yang Bai*, Lianzhou Wang* P. Chen, Dr. Z. Wang, Dr. S. Wang, Dr. M. Lyu, M. Hao, M. Ghasemi, Dr. M. Xiao, Dr. J.-H. Yun, Dr. Y. Bai*, Prof. L. Wang* Nanomaterials Centre, School of Chemical Engineering and Australian Institute for Bioengineering and Nanotechnology, The University of Queensland, Brisbane, QLD 4072, Australia Corresponding Authors: L. Wang ([email protected]); Y. Bai ([email protected])

Graphical Abstract

Luminescent europium-doped titania (Eu-TiO2) thin films are fabricated via a facile and scalable chemical-bath deposition at low-temperature (70 oC). The use of Eu-TiO2 in planar perovskite solar cells (PSCs) enables effective down-shifting of damaging UV light to extra visible luminescence and also leads to a more optimal band alignment, resulting in an enhanced power conversion efficiency of 21.4% and significantly improved device stability under UV illumination for 500 hours.

Abstract Perovskite solar cells (PSCs) have demonstrated high power conversion efficiencies (PCEs) but poor stability against ultraviolet (UV) irradiation. Here, we report a one-pot synthesized luminescent europium-doped titania (Eu-TiO2) via chemical-bath deposition at lowtemperature (70 oC) for planar PSCs, which enables simultaneous efficiency and UV-stability enhancement. We show that the Eu-TiO2 could effectively convert damaging UV photons into 1

useful visible luminescence for additional light harvesting. A more optimal energy band alignment at the Eu-TiO2/perovskite interface leads to facilitated charge extraction and suppressed non-radiative recombination. The use of Eu-TiO2 in PSCs results in increased photocurrent and open-circuit voltage, yielding an enhanced PCE of 21.40% with relative to pristine TiO2 device (19.22%). More importantly, the Eu-TiO2 devices exhibit remarkably improved UV stability, retaining 75% of the initial PCE after exposing to UV illumination for 500 hours, while the devices with pristine TiO2 lost the majority of their original PCEs in 150 hours. As a proof-of-concept, we further demonstrate the scalability of our method by fabricating a large-area Eu-TiO2 film of 64 cm2 showing excellent uniformity.

Keywords: down-shifting, europium-doped titanium dioxide, UV stability, band alignment, planar perovskite solar cells

1. Introduction In recent years, the emerging organic-inorganic metal halide perovskite solar cells (PSCs) have attracted tremendous attention due to their unprecedented progress in power conversion efficiency (PCE) increasing from 3.8% to certified 25.2%.[1, 2] In addition to pursuing an even higher PCE, the long-term stability of PSCs against various external stimuli remains a critical hurdle for their further practical applications. While considerable research efforts have been devoted to prolonging the device lifespan in humid environment, there have been very few study for addressing the negative effect of ultraviolet (UV) irradiation on the photovoltaic performance of PSCs. The high-energy photons from the UV portion of solar spectrum can induce the dissociation of perovskite materials into metal halide and halogen via a superoxidemediated pathway.[3-5] Furthermore, titanium dioxide (TiO2), which is widely used as electron transport materials (ETMs) in high-performing PSCs, would trigger interfacial

2

photocatalytic reaction of perovskite films under UV light.[6, 7] Although alternative widebandgap ETMs (>3.2 eV) such as SnO2, ZnO, and BaSnO3, [8-10] or inserting an interfacial buffer layers [11, 12] have been developed to suppress the interfacial degradation of perovskite films, the long-term stability of polycrystalline perovskite films when coupling with the penetrated UV light remains an unresolved problem.

The chemical doping strategy has been recognized as a facile approach to modify the electronic property of TiO2, including Nb5+, In3+, Al3+, Li+, K+, etc.[7, 13-17] Among them, trivalent rare earth (RE) ions, such as Eu3+, Er3+, Sm3+, represent one kind of effective dopants to enhance the electron mobility of TiO2 with reduced surface trap states and shifted Fermi energy level.[18-20] In addition to the facilitated charge transport ability, the introduction of RE ion dopants with well-shielded 4f-orbital configurations would functionalize TiO2 as down-shifting phosphors that can convert a single high-energy photon to multiple low-energy photons.[21-30] For example, down-shifting phosphor composed of trivalent europium ion (Eu3+) emitting species and TiO2 host matrix can exhibit sharp red luminescence via intra-4f transition under UV excitation.[22, 24, 28] Thus, the use of luminescent TiO2 as a dualfunctional layer in PSCs may open a promising avenue for the simultaneous enhancement of UV stability and light harvesting of PSCs by converting damaging UV light to useful visible luminescence. However, no previous study demonstrate high-efficiency PSCs employing down-shifting phosphors,[29-35] and the positive effect of down-shifting Eu-doped TiO2 (EuTiO2) on the UV stability of PSCs has not been well investigated.

Here, we report a one-pot synthesized luminescent Eu-TiO2 thin films via chemical-bath deposition at low-temperature (70 oC), which not only function as efficient charge transport layers but also effectively convert damaging UV light to extra visible luminescence. The controllable doping of Eu3+ also leads to a more optimized energy band alignment and 3

therefore facilitates the electron extraction of Eu-TiO2-based PSCs. As a result, the use of EuTiO2 as ETM in planar PSCs leads to 5% increase in photocurrent and yields a high PCE of 21.40%, which is the highest compared to previously reported PSCs employing down-shifting phosphors and also among the best in conventional PSCs based on low-temperature processed ETMs (< 120 oC). More Encouragingly, PSCs based on Eu-TiO2 exhibit remarkably improved long-term UV stability, retaining 75% of the peak PCE under continuous UV light illumination (5 mW cm-2) after 500 hours. We further demonstrate that our method can be easily scaled up to module size and a large-area Eu-TiO2 film of 64 cm2 is fabricated as a proof-of-concept showing excellent uniformity.

2. Results and Discussion To prepare high-quality and uniform TiO2 films, a low-temperature chemical-bath deposition method was employed,[36] simply immersing cleaned FTO glasses in a diluted TiCl4 aqueous solution at 70 oC for one hour with subsequent heat treatment at 100 oC for 30 min (details in experimental section). With the incorporation of different molar ratios of europium acetylacetonate in the chemical-bath precursor solutions, Eu-TiO2 nanoparticles with different doping ratios of 2.5%, 5%, and 8% can be grown on FTO substrates. Unless stated otherwise, all the characterization results of Eu-TiO2 films were based on 5% Eu doping. In the system of Eu-TiO2, TiO2 can not only serve as an efficient ETL in PSCs, but also be investigated as an ideal host matrix to sensitize the luminescence of rare earth ion under UV illumination.[22, 24, 27, 28] As illustrated in Figure 1a, high-energy UV photons were firstly absorbed by the TiO2 host matrix, and then relaxed to the Eu3+ activator, subsequently emitting efficient visible luminescence via the intra-4f transitions between the Russell-Saunders multiplets 5D0 to 7Fj (j = 0, 1, 2, 3, 4).[22, 24, 27, 37] To verify the down-shifting effect of Eu-TiO2, the steady-state photoluminescence (PL) spectra of TiO2 and Eu-TiO2 films were measured upon 360 nm UV 4

light excitation (Figure 1b). The Eu-TiO2 film showed several featured PL emission peaks with a dominant red emission peak at 615 nm associated with the electric-dipole 5D0-7F2 transition, while that of the TiO2 film did not show any peaks in the visible light region.

In addition to the down-shifting effect, the doping of europium ion has influence on the optoelectronic properties of the TiO2 host matrixes. Generally, the lower edge of conduction band (CB) of TiO2 is mainly made up of Ti4+ 3d bands, while its upper edge of valence band (VB) consists of O2- 2p bands.[17] In this case, we assume that the presence of Eu3+ in TiO2 lattices would probably introduce some sub energy bands below the CB edge of TiO2 and therefore affect its energy bandgap. To verify this hypothesis, we performed the ultravioletvisible (UV-vis) absorption spectroscopy and ultraviolet photoelectron spectroscopy (UPS) characterizations on both TiO2 and 5% Eu-TiO2 samples. As shown in Figure 1c, the 5% EuTiO2 sample exhibited a narrower energy bandgap of 3.13 eV than that (3.17 eV) of pristine TiO2. Meanwhile, the UPS results of TiO2 and 5% Eu-TiO2 films showed similar VB edges positioned at 7.44 and 7.45 eV below the vacuum level, respectively (Figure S1). Thus, we determine that the 5% Eu-TiO2 film has a lower CB edge of 4.32 eV than that (4.27 eV) of pristine TiO2 film, which confirmed our hypothesis on the energy band structure of 5% EuTiO2 film before and after Eu doping. Then, we measured the UV-vis absorption spectra of 2.5% and 8% Eu-TiO2 samples, showing energy bandgaps of 3.16 and 3.11 eV, respectively. Assuming Eu-doping would not affect the VB edge of TiO2, the CB edges of 2.5% Eu-TiO2 and 8% Eu-TiO2 samples can be estimated to be 4.29 and 4.34 eV below the vacuum level, respectively. As Eu-doping may affect the photocatalytic activity of Eu-TiO2 film, we further evaluated the photoelectrochemical performance of TiO2 and 5% Eu-TiO2 films by employing them as photoanodes in a three-electrode system. As shown in Figure S2, the 5% Eu-TiO2 film shows a photocurrent density of 0.176 mA cm-2 at 1.23 V versus reversible hydrogen electrode (RHE), which is much lower than that (0.317 mA cm-2) of the pristine TiO2 film. 5

Such deteriorated photocatalytic activity of 5% Eu-TiO2 film under light may be due to the excessive doping of Eu3+ ions, compared to the commonly used low Eu-doping ratios (<1%) in reported photocatalysts based on Eu-TiO2 materials.[38-42]

As shown in the X-ray diffraction (XRD) pattern of the pristine TiO2 sample (Figure 1d), all the featured peaks can be indexed to the typical rutile phase (JCPDS 01-089-0552). Except for the identified peaks of rutile TiO2, there is no additional characteristic peak of europiumbased compounds in the XRD pattern of Eu-TiO2 sample, which indicates the preserved TiO2 crystal structure and the incorporation of Eu3+ in lattices.[43] The reduced peak intensities of the XRD pattern of Eu-TiO2 sample was related with the relatively lower crystallinity, which was induced by the crystal structural distortion due to the discrepancy in valency and ionic radii of Eu3+ (0.95 Å) and Ti4+ (0.68 Å).[44] Compared to its crystallinity, the energy band alignment between Eu-TiO2 film and the perovskite layer may play a more determinant role in efficient electron transport, as various amorphous TiOx films were also reported as efficient ETMs in PSCs.[45-49] Although the XRD results indicated lower crystallinity in the Eu-TiO2 film, there is no significant difference in the surface morphology of TiO2 and Eu-TiO2 films. As shown in the SEM images of bare FTO, and TiO2 and Eu-TiO2 layers coated on FTO (Figure 1e-g and Figure S3), interconnected TiO2/Eu-TiO2 nanoparticles with a grain size of 15-20 nm were uniformly grown and fully covered all the exposed facets of FTO grains, manifesting their great potential as effective electron transport and hole blocking layers for PSCs. To investigate the distribution of Eu element, we performed the energy dispersive spectroscopy (EDS) mapping and element analysis on Eu-doped TiO2 films. As shown in Figure S4a-c, Eu element (red dots) and Ti element (green dots) homogeneously distributed across the sample and the Eu molar ratio was determined to be 4.64%, which was consistent with the ratio (5%) of europium source in the precursor solution used in the chemical-bath deposition. We further carried out the X-ray photoelectron spectroscopy (XPS) measurement 6

to study the surface chemical bonding of TiO2 and Eu-TiO2 films (Figure S5). The presence of Eu3+ ions in the Eu-TiO2 film can be confirmed by the emerged peaks at 1163.9, 1134.6, 141.9, and 136.1 eV, which can be assigned to Eu3+ 3d3/2, Eu3+ 3d5/2, Eu3+ 4d3/2, and Eu3+ 4d5/2, respectively,[39, 50] while these peaks are absent in the Eu 3d and Eu 4d XPS spectra of the pristine TiO2 film. In addition, typical Ti4+ 2p1/2 (464.5 eV) and Ti4+ 2p3/2 peaks (458.8 eV) were identified in the Ti 2p XPS spectra of TiO2 and Eu-TiO2 films, and the featured Ti-O bonding (530.1 eV) were observed in their O 1s XPS spectra.[51]

Then, such deposited Eu-TiO2 layers were used as a dual-functional layer to fabricate PSCs with a device configuration of FTO/ Eu-TiO2/ perovskite/ Spiro-OMeTAD/ Au, as illustrated in Figure 2a. On one hand, the generated electrons from the perovskite layer can be transported by the Eu-TiO2 layer. On the other hand, the Eu-TiO2 layer can convert harmful UV light to visible luminescence via its down-shifting transition, which boosts the light harvesting of the adjacent perovskite layer. In the completed devices, perovskite layers with a composition of (Cs0.05FA0.80MA0.15)Pb(I0.85Br0.15)3 were prepared on TiO2/Eu-TiO2 substrates by a one-step antisolvent-assisted method,[52, 53] resulting in compact and pinhole-free lightabsorbing layers with an average grain size of ~200-300 nm and a thickness of ~600 nm, as shown in the surface and cross-section SEM images (Figure S4d-e). The deposited perovskite film exhibited an optical bandgap of 1.62 eV (Figure S4f), and its VB edge (5.92 eV) and CB edge (4.30 eV) were determined by UPS measurement (Figure S6). Consequently, the energy band diagram of different functional layers in PSCs is summarized in Figure 2b. Since TiO2 and 2.5% Eu-TiO2 exhibit higher CB positions above that of perovskite layer, it may build an energy barrier for the photogenerated electrons injecting from the perovskite layer to ETM and therefore lead to undesirable consequences such as charge accumulation and nonradiative recombination at the interface.[46, 49] Compared to that (40 meV) of 8% Eu-TiO2, 5% Eu-TiO2 possesses a more favorable smaller band offset of 20 meV below that of the 7

perovskite layer. A more optimal energy band alignment between 5% Eu-TiO2 and the perovskite layer would probably allow more efficient electron extraction.[54, 55]

To examine the electron transport property and down-shifting effect, we fabricated solar cells using the Eu-TiO2 films with different Eu doping ratios as ETMs and the pristine TiO2 film was used as a reference. Figure 2c shows the current density-voltage (J-V) curves of the bestperforming devices based on Eu-TiO2 and TiO2 films, and their photovoltaic parameters with standard deviation obtained from 28 devices are summarized in Table 1. The control devices based on the pristine TiO2 films delivered an average PCE of 18.58% with a short-circuit current density (Jsc) of 22.20 mA cm-2, an open-circuit voltage (Voc) of 1.09 V, and a fill factor (FF) of 0.76. When TiO2 was doped with 2.5% Eu, the average PCE was enhanced to 19.24% as a result of the increased Jsc. By increasing the Eu-doping ratio to 5%, the average PCE was further boosted to 20.13% with a significantly improved average Jsc of 23.32 mA cm-2 and a Voc of 1.12 V. Further increasing the doping ratio to 8% deteriorated the Jsc and PCE, which may be due to the structural distortion in TiO2 crystal caused by excessive Eudoping.[27, 28] Figure 2d shows the statistics of the PCE distributions, which demonstrates the reliability and repeatability of the PCE enhancement by Eu-doping. The champion device based on Eu-TiO2 (5%) exhibited a Jsc of 23.85 mA cm-2, a Voc of 1.13 V and a FF of 0.79, yielding a PCE of 21.40%. To the best of our knowledge, this is the highest PCE compared to previously reported PSCs employing down-shifting materials and also among the best in the reported n-i-p planar PSCs based on low-temperature processed ETMs (< 120 oC) (Table S1 and S2). No obvious photocurrent hysteresis was observed in the device made with Eu-TiO2 by changing the sweeping direction (Figure 2e and Table S3). The steady-state photocurrent and efficiency measured at the maximum power point of 0.92 V are shown in Figure S7a, which further confirms the device performance parameters extracted from the J-V curve. We observed that the increase of Jsc is remarkable for Eu-TiO2-based devices (Figure S7b), which 8

could be attributed to both the down-shifting induced light harvesting improvement and the facilitated charge extraction resulting from better energy band alignment in Eu-TiO2-based devices.

To quantify the photocurrent increased by the down-shifting effect, we performed the external quantum efficiency (EQE) measurement, as shown in Figure 2f. The integrated photocurrent from the PSCs based on TiO2 and Eu-TiO2 films reached 21.35 and 22.15 mA cm-2, respectively, which are comparable with the Jsc extracted from J-V curves and thus confirm the validity and calibration of our AM 1.5 light source.[35] Notably, the device based on EuTiO2 films displayed much higher EQE values with an enhanced integrated photocurrent by ~0.4 mA cm-2 in the UV region (< 400 nm), revealing the photocurrent contribution from the down-shifting luminescence. To further confirm the impact of the down-shifting effect on photocurrent, we deliberately exposed PSCs based on both TiO2 (8 devices) and Eu-TiO2 films (8 devices) under pristine UV light illumination with a high intensity of 10 mW cm-2, and measured their J-V curves. As shown in Figure S9a, devices based on Eu-TiO2 films delivered an average Jsc of 1.67 mA cm-2, which is over two times higher than that of TiO2 films (0.52 mA cm-2), highlighting the photocurrent contribution from the luminescence converted from UV light by the Eu-TiO2 layer.

To understand how Eu-doping impact the charge transfer dynamics, we firstly performed the steady-state PL and time-resolved PL (TRPL) measurements on various perovskite films deposited on glass, TiO2, and Eu-TiO2 films (Figure 3a-b). Under light source excitation of 450 nm, the sample with a perovskite layer coated on a glass substrate exhibited a strong steady-state emission peak at 760 nm, which is consistent with the measured optical bandgap of the perovskite materials (1.62 eV). The PL emissions dramatically quenched when contacting with TiO2 and Eu-TiO2 layers (Figure 3a), indicating the efficient electron 9

extraction by ETM layers. Compared to that of TiO2 films, a relatively lower intensity of emission from the sample of perovskite on Eu-TiO2 films was observed, which can be ascribed to the facilitated electron extraction from a more optimized energy band alignment.[56] Then, the charge recombination dynamics were studied by the TRPL decay measurements, in which average recombination lifetimes of different samples can be estimated by fitting with a bi-exponential function of time. After fitting, the estimated average lifetimes of the samples with perovskite on TiO2 and Eu-TiO2 films were 29.93 and 23.34 ns, respectively (Figure 3b). The shorter PL lifetime in the sample of perovskite on Eu-TiO2 films indicated a faster electron extraction process, which was in accordance with the stead-state PL results. All the fitted TRPL parameters for different samples are summarized in Table S4. Furthermore, impedance spectroscopy (IS) measurements were performed on the completed PSCs based on TiO2 and Eu-TiO2 films in the dark under different applied potentials (0.1-0.9 V), and a simplified circuit was used to fit the Nyquist plots (Figure S8a). Apart from a series resistance (RS) which represents the resistance of conductive glass and metal electrode, the resistance (RSC) extracted from the high-frequency arc is assigned to the selective contacts and their interfaces with the perovskite layer, while the resistance (Rrec) derived from the lowfrequency arc refers to the charge carrier recombination of the perovskite layer.[57, 58] As the hole transport layers are identical in this study, the difference in RSC can be attributed to the charge transfer at the ETM/perovskite interface. All the parameters derived from the fitted Nyquist spectra were summarized in Table S5. Compared to those of pristine TiO2 films , the lower RSC values of PSCs based on Eu-TiO2 films under different applied potentials (0.1, 0.3, 0.5, 0.7, and 0.9 V) indicates a more facilitated electron transfer process at the EuTiO2/perovskite interface, which suppresses the undesirable interfacial non-radiative recombination and thus leads to improved Jsc and FF. To further investigate the charge transfer dynamics, we estimated the built-in potential at the ETM/perovskite heterojunction interface by performing Mott-Schottky plot measurement on both devices under dark 10

condition. As shown in Figure 3c-d, the Eu-TiO2-based device delivered a larger built-in potential of 0.89 V than that of the control device (0.86 V). The higher built-in potential would provide larger driving force for improving the photoexcited carrier extraction and thus reducing the interfacial trap-states due to charge accumulation, which leads to suppressed carrier recombination at the ETM/perovskite interface.[59, 60] To confirm the reduction of trap-states, we fabricated the electron-only devices with a structure of FTO/ TiO2 (Eu-TiO2)/ Perovskite/ PCBM/ Au for space charge limited current (SCLC) measurement. As presented in Figure 3e-f, the fitted trap-filled limit voltage (VTFL) of the electron-only devices based on TiO2 and Eu-TiO2 films are 0.48 and 0.35 V, respectively. The trap state density (Ntrap) is determined by the VTFL using the following equation 1: N trap =

2ε 0ε VTFL eL2

(1)

where ε0 is the vacuum permittivity, ε is the relative dielectric constant of FA-based perovskite (ε = 62.23),[61] e is the electron charge, and L is the film thickness. After calculation, the estimated electron trap density of the perovskite films coated on TiO2 and EuTiO2 films are 1.32 × 1016 and 9.64 × 1015 cm-3, respectively. Apart from photocurrent loss, the non-radiative recombination induced by the electron trap states at the ETM/perovskite interface would also contribute to the discrepancy between the radiative Voc limit (Voc,rad) and the obtained Voc, as indicated in the following equation 2:[62-64] VOC ,non− rad = VOC , rad − VOC = −

kT ln ( EQEEL ) q

(2)

Where q is the elementary charge, k is Boltzmann’s constant, T is sample temperature, and EQEEL is the quantum yield of electroluminescence. In this case, the reduced trap-states would lead to suppressed non-radiative recombination at the Eu-TiO2/perovskite interface and thus facilitate electron extraction process and reduce the interfacial voltage loss, which

11

corresponds to the improvement in all photovoltaic parameters in the 5% Eu-TiO2-based devices (Table 1).

To investigate their UV tolerance of the fabricated PSCs, we placed the encapsulated devices under continuous UV illumination (365 nm, 5 mW cm-2) in ambient air (relative humidity level of 60±10%, RH 60±10%) as depicted in Figure 4a and monitor their photovoltaic performance evolution with time. The control device with pristine TiO2 lost the majority of the initial PCE after aging for 150 hours. In a sharp contrast, PSCs based on Eu-TiO2 films retained about 75% of the original PCE even after aging for 500 hours under UV light (Figure 4b). The original J-V curves showing the degradation can be found in Figure S9b-c for both types of perovskite devices. The variation of key photovoltaic parameters versus illumination time is summarized in Figure S9d-f. We note that the Jsc of Eu-TiO2 devices maintained ca. 80% of the initial value after UV light aging for 500 hours, while the control devices experienced a fast decline in Jsc under the same conditions. Such rapid photocurrent degradation in control devices can be ascribed to the gradual dissociation of bulk perovskite films as well as the interfacial degradation accelerated by the photocatalytic TiO2 under UV illumination.[6] With down-shifting Eu-TiO2 films, the corrosive UV photons that drives the decomposition of perovskite films can be effectively converted to visible luminescence. Consequently, the perovskite films showed improved robustness against UV illumination and the corresponding device lifespan was prolonged. In addition to the UV light stability, we have investigated the operational stability of PSCs based on both Eu-TiO2 and TiO2 films under continuous full-spectrum light illumination (100 mW cm-2) with an UV intensity of 4.6 mW cm-2 in the air (RH 40±10%).[11] The encapsulated devices were held at a static load of 680 Ω to ensure they remained near their maximum power points, and the temperature of the aging devices (ca. 60 oC) was measured by an infrared thermometer.[65] As shown in Figure S10a, PSCs based on Eu-TiO2 films maintained more than 80% of the original efficiency after 12

exposure under one-sun full-spectrum illumination for 100 hours, while the control devices lost more than 50% of their original PCEs. The performance loss in control devices mainly results from the rapid degradation of Jsc and FF (Figure S 10b-d). The Jsc decay is mainly due to the UV-induced decomposition of perovskite films, and the reduced FF can be attributed the accumulated electrons at the interface of TiO2 and perovskite layers which leads to increased non-radiative recombination. For Eu-TiO2-based devices, the UV-induced decomposition and interfacial photocatalytic degradation of perovskite films were largely mitigated by down-shifting the damaging UV photons to visible light. Additionally, the more optimized band alignment would facilitate charge extraction and thus suppress the undesirable charge accumulation at interfaces, leading to significantly improved operational stability.

As a proof-of-concept, we demonstrated the scalability of our deposition method by fabricating a large-area Eu-TiO2 substrate of 64 cm2, as shown in Figure 4c. It is noted that any substrate size should be feasible with the chemical-bath deposition method. To check the uniformity, this large-area Eu-TiO2 substrate was cut into 16 pieces (2 × 2 cm2) along the dotted lines for device fabrication. The corresponding subcells exhibited highly reproducible photovoltaic performance with an average PCE of 19.46% (Figure 4d-e). All the photovoltaic parameters of these 16 subcells were summarized in Table S5. Unlike the commonly used compact and mesoporous TiO2 layer prepared by spray pyrolysis or sol-gel solution method, the cost-effective Eu-doped TiO2 films developed in this work does not require hightemperature annealing (500 oC) treatment. Integrating with scalable techniques for perovskite layers and hole transport materials,[66-68] it is very promising to fabricate efficient and UVstable perovskite solar modules based on such facile, cost-effective, and scalable Eu-TiO2 films.

3. Conclusion 13

In summary, we report a luminescent Eu-TiO2 ETM for efficient and UV-stable n-i-p planar PSCs. Benefiting from the enhanced light harvesting by down-shifting effect and a more optimal energy band alignment, PSCs based on Eu-TiO2 films deliver a boosted PCE of 21.40%. More encouragingly, these Eu-TiO2 devices exhibit significantly improved UV stability, retaining ca. 75% of the original PCE after aging under continuous UV illumination for 500 hours in ambient condition (RH 60±10%). We further demonstrate the scalability of our method by fabricating a highly uniform and large-area Eu-TiO2 film of 64 cm2, which can be easily applied for industrial manufacturing of perovskite solar modules. Our findings offer an effective way to simultaneously improve the long-term UV-stability and efficiency of PSCs, and this facile strategy is also applicable to other perovskite-based optoelectronic devices beyond photovoltaics.

4. Experimental Section Materials and Precursor Preparation: lead iodide (99.999%), lead bromide (99.999%), cesium iodide (99.999%), titanium (IV) chloride (99.9%), europium acetylacetonate hydrate (Eu(acac)3), dimethyl sulfoxide (DMSO), N,N-dimethylformamide (DMF), chlorobenzene (CB), acetonitrile (ACN), Bis(trifluoromethane)sulfonamide lithium (LiTFSI) and 4-tertbutylpyridine (TBP) were purchased from Sigma-Aldrich. Formamidinium iodide (FAI) and methylammonium bromide (MABr) were bought from Greatcell Solar, while Spiro-OMeTAD was purchased from Borun Chemicals. All the purchased chemicals were used as received unless stated otherwise. The perovskite precursor solution was prepared by dissolving FAI (1.10 M), PbI2 (1.25 M), MABr (0.22 M) and PbBr2 (0.25 M) in a mixed solvent of DMF and DMSO (4:1 v/v), followed by adding CsI stock solution (1.5 M in DMSO) to achieve the desired triple-cation composition of (Cs0.05FA0.80MA0.15)Pb(I0.85Br0.15)3. The precursor solution for hole transport layer was prepared by dissolving 72.3 mg of Spiro-OMeTAD in 1

14

mL of CB with additives of 17.5 µL of LiTFSI solution (520 mg mL-1 in ACN) and 28.8 µL of TBP. The chemical bath deposition of TiO2 and Eu-TiO2 films: For the growth of TiO2 film, 200 mM TiCl4 aqueous solution was prepared by dropwise adding 4.5 mL of TiCl4 in 200 g ice water. As for the precursor solution of Eu-TiO2 film, 2.5, 5, and 8 mol% of Eu(acac)3 was further added in 200 mM TiCl4 aqueous solution, forming a yellowish and clean solution. After ultrasonic washing with acetone and isopropanol, cleaned FTO glass (FTO22-7, Yingkou OPV) substrates were dried and treated with UV-ozone for 20 min. Kapton polyimide tape was used to cover the unwanted deposition areas. Subsequently, FTO substrates with different sizes (2 × 2 cm2 and 8 × 8 cm2) were fully immersed into the prepared precursors in a sealed glass container, followed by storing in an oven at 70 oC for 1 hour. After washing by water and ethanol, the obtained TiO2 or Eu-TiO2 coated FTO substrates were dried at 100 oC for 30 min on a hotplate. Device fabrication: The perovskite layer was deposited via a one-step antisolvent-assisted method. In detail, 80 µL of perovskite precursor was drop cast on the TiO2 or Eu-TiO2 coated FTO substrates (2 × 2 cm2), followed by spin coating at 2000 rpm for 10 s and 6000 rpm for 30 s, in which 200 µL of CB was dropped on the spinning substrate 15 s prior to the end of second step. After a further annealing step at 100 oC for 60 min, the crystalline perovskite layer was obtained. Then, 70 µL of Spiro-OMeTAD precursor solution was spin-coated on top of the perovskite layers at 4000 rpm for 30 s. In the end, the devices were completed by depositing 80 nm of gold electrodes using thermal evaporation with an active area of 0.10 cm2. The practical active area of the devices were defined by a metal shadow mask with an aperture area of 0.0625 cm2 to avoid the edge effects. Characterizations and measurements: XRD was characterized by a Bruker Advanced X-ray diffractometer (40 kV and 30 mA) with Cu Kα radiation. UV-Vis absorption spectra of different films were measured by a spectrophotometer (V-650, Jasco), and UPS was 15

characterized by an X-ray photoelectron spectroscopy (Kratos Axis Supra). The surface chemical bonding of films were measured by XPS (Krato Ultra) with Al Kα radiation source, and all the binding energies were calibrated by using C 1s (284.8 eV) as a reference. The surface and cross-section SEM images were recorded using a field-emission scanning electron microscope (FE-SEM, JEOL 7001). The steady-state and time-resolved PL spectra were measured at room temperature on a fluorescence spectrophotometer (FLSP-900, Edinburgh Instruments). The steady-state PL emission spectra of TiO2 and perovskite films were measured at 360 nm and 450 nm light source excitation using monochromatized Xe lamp, respectively. The time-resolved PL spectra were carried out with a 377 nm pulsed diode laser excitation source. Nyquist plots of the completed devices were measured under different applied potentials of 0.1-0.9 V with an AC perturbation signal of 10 mV in the frequency range of 1 Hz to 100 kHz in the dark. The J-V curves were measured with a source meter (Keithley 2420) and a solar simulator (Newport, Oriel Class AAA, 94063A) at 100 mW cm-2 illumination (AM 1.5 G). The light intensity was calibrated by a silicon reference cell and meter (Newport, 91150V) certificated by NREL. The J-V curves were measured in reverse scan (from 1.2 V to -0.1 V) or forward scan (from -0.1 V to 1.2 V) modes at a scan speed of 100 mV s-1. The incident photo-to-current conversion efficiency (IPCE) was measured using a 300 W xenon light source (Newport, 66902) with an Oriel Cornerstone monochromator and a multimeter in DC mode. A standard Si reference cell was used for calibration before the IPCE measurements. For the stability testing, the devices were encapsulated by UV curable resin and glasses. The UV stability was monitored by exposing the sealed devices under UV lamp (Black-Ray B100, 365 nm, 5 mW cm-2) illumination in ambient conditions (RH 60±10%), and the full-spectrum light stability was carried out under a solar simulator (Newport, Oriel Class ABB, 94061A) at 100 mW cm-2 illumination (AM 1.5 G). A static resistor of 680 Ω was connected to devices to ensure they remained near their maximum power points.

16

Acknowledgements The financial support from the Australian Research Council Discovery Projects (ARC DPs) and Laureate Fellowship are highly appreciated. The authors acknowledge the facilities and the scientific supports from the Queensland node of the Australian National Fabrication Facility (ANFF), Central Analytical Research Facility (CARF) and Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy and Microanalysis (CMM), The University of Queensland. Y. B. acknowledges the support from ARC DECRA Fellowship. P. Chen thanks the support from Australian Government Research Training Program (RTP) and UQ Centennial Scholarships.

Appendix A. Supporting Information Supplementary data associated with this article can be found in the online version at

References:

[1]

NREL

Chart,

https://www.nrel.gov/pv/assets/pdfs/best-research-cell-

efficiencies.20190802.pdf, (Accessed August 2019). [2] A. Kojima, K. Teshima, Y. Shirai, T. Miyasaka, J. Am. Chem. Soc. 131 (2009) 6050-6051. [3] N. Aristidou, I. Sanchez-Molina, T. Chotchuangchutchaval, M. Brown, L. Martinez, T. Rath, S.A. Haque, Angew. Chem. Int. Ed. 54 (2015) 8208-8212. [4] N. Aristidou, C. Eames, I. Sanchez-Molina, X. Bu, J. Kosco, M.S. Islam, S.A. Haque, Nat. Commun. 8 (2017) 15218-15228. [5] S.-W. Lee, S. Kim, S. Bae, K. Cho, T. Chung, L.E. Mundt, S. Lee, S. Park, H. Park, M.C. Schubert, S.W. Glunz, Y. Ko, Y. Jun, Y. Kang, H.-S. Lee, D. Kim, Sci. Rep. 6 (2016) 3815038160. [6] T. Leijtens, G.E. Eperon, S. Pathak, A. Abate, M.M. Lee, H.J. Snaith, Nat. Commun. 4 (2013) 2885-2893. [7] S.K. Pathak, A. Abate, P. Ruckdeschel, B. Roose, K.C. Gödel, Y. Vaynzof, A. Santhala, S.-I. Watanabe, D.J. Hollman, N. Noel, A. Sepe, U. Wiesner, R. Friend, H.J. Snaith, U. Steiner, Adv. Funct. Mater. 24 (2014) 6046-6055. [8] S.S. Shin, E.J. Yeom, W.S. Yang, S. Hur, M.G. Kim, J. Im, J. Seo, J.H. Noh, S.I. Seok, Science 356 (2017) 167-171. [9] Q. Jiang, Z. Chu, P. Wang, X. Yang, H. Liu, Y. Wang, Z. Yin, J. Wu, X. Zhang, J. You, Adv. Mater. 29 (2017) 1703852. 17

[10] J. Song, L. Liu, X.-F. Wang, G. Chen, W. Tian, T. Miyasaka, J. Mater. Chem. A 5 (2017) 13439-13447. [11] W. Li, W. Zhang, S. Van Reenen, R.J. Sutton, J. Fan, A.A. Haghighirad, M.B. Johnston, L. Wang, H.J. Snaith, Energy Environ. Sci. 9 (2016) 490-498. [12] M.M. Tavakoli, F. Giordano, S.M. Zakeeruddin, M. Gratzel, Nano Lett. 18 (2018) 24282434. [13] J. Peng, T. Duong, X. Zhou, H. Shen, Y. Wu, H.K. Mulmudi, Y. Wan, D. Zhong, J. Li, T. Tsuzuki, K.J. Weber, K.R. Catchpole, T.P. White, Adv. Energy Mater. (2016) 1601768. [14] H. Zhou, Q. Chen, G. Li, S. Luo, T.B. Song, H.S. Duan, Z. Hong, J. You, Y. Liu, Y. Yang, Science 345 (2014) 542-546. [15] W. Chen, Y.Z. Wu, Y.F. Yue, J. Liu, W.J. Zhang, X.D. Yang, H. Chen, E.B. Bi, I. Ashraful, M. Gratzel, L.Y. Han, Science 350 (2015) 944-948. [16] T. Singh, S. Oz, A. Sasinska, R. Frohnhoven, S. Mathur, T. Miyasaka, Adv. Funct. Mater. 28 (2018) 1706287. [17] B. Roose, S. Pathak, U. Steiner, Chem. Soc. Rev. 44 (2015) 8326-8349. [18] Y. Xiang, Z. Ma, J. Zhuang, H.L. Lu, C.Y. Jia, J.S. Luo, H.M. Li, X.W. Cheng, J. Phys. Chem. C 121 (2017) 20150-20157. [19] Z. Ren, J. Wu, N. Wang, X. Li, J. Mater. Chem. A 6 (2018) 15348-15358. [20] Z. Xu, J.H. Wu, T.Y. Wu, Q.L. Bao, X. He, Z. Lan, J.M. Lin, M.L. Huang, Y.F. Huang, L.Q. Fan, Energy Technol. 5 (2017) 1820-1826. [21] S. Gai, C. Li, P. Yang, J. Lin, Chem. Rev. 114 (2014) 2343-2389. [22] A. Conde-Gallardo, M. Garcı́a-Rocha, I. Hernández-Calderón, R. Palomino-Merino, Appl. Phys. Lett. 78 (2001) 3436-3438. [23] W. Strek, P. Deren, A. Bednarkiewicz, J. Lumin. 87–89 (2000) 999-1001. [24] H. Zhang, Y. Sheng, K. Zheng, X. Zhou, Z. Shi, X. Xu, H. Zou, Eur. J. Inorg. Chem. 2014 (2014) 3305-3311. [25] R.S. Fontenot, W.A. Hollerman, K.N. Bhat, M.D. Aggarwal, J. Lumin. 132 (2012) 18121818. [26] J. Wu, J. Wang, J. Lin, Y. Xiao, G. Yue, M. Huang, Z. Lan, Y. Huang, L. Fan, S. Yin, T. Sato, Sci. Rep. 3 (2013) 2058. [27] J. Yin, L. Xiang, X. Zhao, Appl. Phys. Lett. 90 (2007) 113112. [28] K.L. Frindell, M.H. Bartl, A. Popitsch, G.D. Stucky, Angew. Chem. Int. Ed. 41 (2002) 959-962.

18

[29] W. Chen, Q. Luo, C. Zhang, J. Shi, X. Deng, L. Yue, Z. Wang, X. Chen, S. Huang, J. Mater. Sci.: Mater. Electron. 28 (2017) 11346-11357. [30] J. Jia, J. Dong, J. Lin, Z. Lan, L. Fan, J. Wu, J. Mater. Chem. C 7 (2019) 937-942. [31] L. Jiang, W. Chen, J. Zheng, L. Zhu, L.e. Mo, Z. Li, L. Hu, T. Hayat, A. Alsaedi, C. Zhang, S. Dai, ACS Appl. Mater. Inter. 9 (2017) 26958-26964. [32] L. Jiang, J. Zheng, W. Chen, Y. Huang, L. Hu, T. Hayat, A. Alsaedi, C. Zhang, S. Dai, ACS Appl. Energy Mater. 1 (2018) 93-102. [33] C. Chen, H. Li, J.J. Jin, X. Chen, Y. Cheng, Y. Zheng, D.L. Liu, L. Xu, H.W. Song, Q.L. Dai, Adv. Energy Mater. 7 (2017) 1700758. [34] Q. Wang, X.S. Zhang, Z.W. Jin, J.R. Zhang, Z.F. Gao, Y.F. Li, S.Z.F. Liu, ACS Energy Lett. 2 (2017) 1479-1486. [35] F. Bella, G. Griffini, J.P. Correa-Baena, G. Saracco, M. Gratzel, A. Hagfeldt, S. Turri, C. Gerbaldi, Science 354 (2016) 203-206. [36] A. Yella, L.P. Heiniger, P. Gao, M.K. Nazeeruddin, M. Gratzel, Nano Lett. 14 (2014) 2591-2596. [37] K.L. Frindell, M.H. Bartl, M.R. Robinson, G.C. Bazan, A. Popitsch, G.D. Stucky, J. Solid State Chem. 172 (2003) 81-88. [38] Z. Xu, Q. Yang, C. Xie, W. Yan, Y. Du, Z. Gao, J. Zhang, J. Mater. Sci. 40 (2005) 15391541. [39] Y. Ma, J. Zhang, B. Tian, F. Chen, S. Bao, M. Anpo, Res. Chem. Intermed. 38 (2012) 1947-1960. [40] G.V. Khade, N.L. Gavade, M.B. Suwarnkar, M.J. Dhanavade, K.D. Sonawane, K.M. Garadkar, J. Mater. Sci.: Mater. Electron. 28 (2017) 11002-11011. [41] Y. Zhang, H. Zhang, Y. Xu, Y. Wang, J. Mater. Chem. 13 (2003) 2261-2265. [42] Z.M. El-Bahy, A.A. Ismail, R.M. Mohamed, J. Hazard. Mater. 166 (2009) 138-143. [43] M. Pal, U. Pal, J.M.G.Y. Jiménez, F. Pérez-Rodríguez, Nanoscale Res. Lett. 7 (2012) 1. [44] C. Leostean, M. Stefan, O. Pana, A.I. Cadis, R.C. Suciu, T.D. Silipas, E. Gautron, J. Alloys Compd. 575 (2013) 29-39. [45] D. Yang, R. Yang, J. Zhang, Z. Yang, S. Liu, C. Li, Energy Environ. Sci. 8 (2015) 32083214. [46] A. Kogo, Y. Sanehira, Y. Numata, M. Ikegami, T. Miyasaka, ACS Appl. Mater. Inter. 10 (2018) 2224-2229. [47] K.E. Roelofs, V.L. Pool, D.A. Bobb-Semple, A.F. Palmstrom, P.K. Santra, D.G. Van Campen, M.F. Toney, S.F. Bent, Adv. Mater. Interfaces 3 (2016) 1600580. 19

[48] X. Deng, G.C. Wilkes, A.Z. Chen, N.S. Prasad, M.C. Gupta, J.J. Choi, J. Phys. Chem. Lett. 8 (2017) 3206-3210. [49] J.P. Correa Baena, L. Steier, W. Tress, M. Saliba, S. Neutzner, T. Matsui, F. Giordano, T.J. Jacobsson, A.R. Srimath Kandada, S.M. Zakeeruddin, A. Petrozza, A. Abate, M.K. Nazeeruddin, M. Gratzel, A. Hagfeldt, Energy Environ. Sci. 8 (2015) 2928-2934. [50] X. Tan, Q. Fan, X. Wang, B. Grambow, Environ. Sci. Technol. 43 (2009) 3115-3121. [51] H. Chen, G. Liu, L. Wang, Sci. Rep. 5 (2015) 10852. [52] N.J. Jeon, J.H. Noh, Y.C. Kim, W.S. Yang, S. Ryu, S.I. Seok, Nat. Mater. 13 (2014) 897903. [53] M. Saliba, T. Matsui, K. Domanski, J.Y. Seo, A. Ummadisingu, S.M. Zakeeruddin, J.P. Correa-Baena, W.R. Tress, A. Abate, A. Hagfeldt, M. Gratzel, Science 354 (2016) 206-209. [54] P. Chen, Y. Bai, S.C. Wang, M.Q. Lyu, J.H. Yun, L.Z. Wang, Adv. Funct. Mater. 28 (2018) 1706923. [55] W. Tress, J. Phys. Chem. Lett. (2017) 3106-3114. [56] C. Xie, C. Zhou, Y. Bin, L. Shen, L. Ke, L. Ding, Y. Yuan, Appl. Phys. Express 12 (2019) 064006. [57] P. Yadav, M.I. Dar, N. Arora, E.A. Alharbi, F. Giordano, S.M. Zakeeruddin, M. Gratzel, Adv. Mater. 29 (2017) 1701077. [58] Y. Jiang, E.J. Juarez-Perez, Q. Ge, S. Wang, M.R. Leyden, L.K. Ono, S.R. Raga, J. Hu, Y. Qi, Mater. Horiz. 3 (2016) 548-555. [59] X. Gong, Q. Sun, S. Liu, P. Liao, Y. Shen, C. Gratzel, S.M. Zakeeruddin, M. Gratzel, M. Wang, Nano Lett. 18 (2018) 3969-3977. [60] Y. Lin, B. Chen, Y. Fang, J. Zhao, C. Bao, Z. Yu, Y. Deng, P.N. Rudd, Y. Yan, Y. Yuan, J. Huang, Nat. Commun. 9 (2018) 4981. [61] D. Yang, R. Yang, K. Wang, C. Wu, X. Zhu, J. Feng, X. Ren, G. Fang, S. Priya, S. Liu, Nat. Commun. 9 (2018) 3239. [62] J. Yao, T. Kirchartz, M.S. Vezie, M.A. Faist, W. Gong, Z. He, H. Wu, J. Troughton, T. Watson, D. Bryant, J. Nelson, Phys. Rev. Appl. 4 (2015) 014020. [63] A. Fakharuddin, L. Schmidt-Mende, G. Garcia-Belmonte, R. Jose, I. Mora-Sero, Adv. Energy Mater. 7 (2017) 1700623. [64] W. Tress, Adv. Energy Mater. 7 (2017) 1602358. [65] J.A. Christians, P. Schulz, J.S. Tinkham, T.H. Schloemer, S.P. Harvey, B.J. Tremolet de Villers, A. Sellinger, J.J. Berry, J.M. Luther, Nat. Energy 3 (2018) 68-74.

20

[66] Y.H. Deng, X.P. Zheng, Y. Bai, Q. Wang, J.J. Zhao, J.S. Huang, Nat. Energy 3 (2018) 560-566. [67] M. Yang, Z. Li, M.O. Reese, O.G. Reid, D.H. Kim, S. Siol, T.R. Klein, Y. Yan, J.J. Berry, M.F.A.M. van Hest, K. Zhu, Nat. Energy 2 (2017) 17038. [68] H. Chen, F. Ye, W. Tang, J. He, M. Yin, Y. Wang, F. Xie, E. Bi, X. Yang, M. Gratzel, L. Han, Nature 550 (2017) 92-95.

21

Figure 1. (a) Illustrated mechanism of down-shifting transition from UV photons to visible luminescence; (b) the steady-state PL spectra of TiO2 and Eu-TiO2 films on glass; (c) the UVvis absorption spectra of TiO2 and Eu-TiO2 samples with different doping ratios; (d) the XRD patterns of TiO2 and Eu-TiO2 samples; the SEM images of (e) bare FTO, (f) Eu-TiO2 film coated on FTO, and (g) Eu-TiO2 film coated on FTO with a higher magnification of the dotted box area.

22

Figure 2. (a) Schematic illustrations of the device configuration of n-i-p planar PSCs employing Eu-TiO2, (b) energy band alignment diagram of different layers of PSCs, (c) the JV curves of PSCs based on TiO2, 2.5% Eu-TiO2, 5% Eu-TiO2, and 8% Eu-TiO2 films, (d) statistics of the PCE distributions of PSCs based on TiO2 and 5% Eu-TiO2 films, (e) the J-V curves of the champion device based on 5% Eu-TiO2 film under forward and reverse scans, (f) IPCE measurement and the integrated Jsc of PSCs based on TiO2 and 5% Eu-TiO2 films. Note that 2.5%, 5%, and 8% are the molar ratios of Eu-doping.

23

Figure 3. (a) The steady-state PL spectra of perovskite layer on glass, TiO2/FTO and EuTiO2/FTO substrates, (b) the time-resolved PL decay curves of perovskite layer on TiO2/FTO and Eu-TiO2/FTO substrates; the Mott-Schottky plots of PSCs based on (c) TiO2 and (d) EuTiO2 films under dark; the SCLC curves of electron-only devices based on (e) TiO2, and (f) Eu-TiO2 films.

24

Figure 4. (a) The digital photo of encapsulated PSCs under UV light illumination in ambient conditions, (b) the stability testing of the encapsulated PSCs based on TiO2 and Eu-TiO2 films exposing under continuous UV illumination in ambient conditions (RH 60 ± 10%), (c) the digital photo of the fabricated large-area FTO substrates (8 × 8 cm2) coated with Eu-TiO2 films, (d) the J-V curves of 16 subcells (2 × 2 cm2) based on the labelled areas of the largearea FTO substrates, (e) the statistic of the PCE distribution of the fabricated subcells.

25

Table 1. Summary of the best photovoltaic performance of the fabricated PSCs based on TiO2 and Eu-TiO2 films with different Eu doping ratios. ETM TiO2 2.5% Eu-TiO2 5% Eu-TiO2 8% Eu-TiO2 a b

PCE (%) a,b

Jsc (mA cm-2) a,b

Voc (V) a,b

FF a,b

18.58 ± 0.74

22.20 ± 0.95

1.09 ± 0.03

0.76 ±0.02

(19.22)

(22.28)

(1.12)

(0.77)

19.24 ± 0.96

22.79 ± 0.77

1.10 ±0.02

0.76 ± 0.02

(20.20)

(23.56)

(1.10)

(0.78)

20.13 ± 1.27

23.32 ± 0.68

1.12 ± 0.03

0.76 ± 0.03

(21.40)

(23.85)

(1.13)

(0.79)

17.97 ± 1.08

22.07 ± 0.99

1.10 ± 0.02

0.74 ± 0.05

(19.05)

(21.95)

(1.10)

(0.79)

Average values with standard deviation are obtained from 28 devices; The values in brackets are the photovoltaic parameters of the champion devices.

Author Information

Peng Chen is a current Ph.D. student under the supervision of Prof. Lianzhou Wang at School of Chemical Engineering, The University of Queensland, Australia. He received his B. Eng. (2010) and M. Eng. (2013) from Shanghai University, China. His research interests mainly focus on the development of efficient and stable perovskite solar cells.

26

Zhiliang Wang is currently a postdoctoral research fellow in the School of Chemical Engineering, the University of Queensland. He received his Ph.D. degree from Dalian Institute of Chemical Physics, University of Chinese Academy of Sciences in 2017. His research interests focus on photoelectrochemical cells and photocatalysis.

Songcan Wang is currently a professor at Institute of Flexible Electronics (IFE), Northwestern Polytechnical University (NPU), China. He received his B.Eng. (2011) and M.Eng. (2014) from Central South University (CSU), China, and Ph.D. degree from The University of Queensland (UQ), Australia in 2018. Before joining NPU, he has been working as a postdoctoral research fellow in Professor Lianzhou Wang’s group at UQ for about 1.5 years. His research interests focus on the synthesis of semiconductor nanomaterials for solar energy conversion and storage, including photoelectrochemical cells, photocatalysis, and rechargeable batteries.

Miaoqiang Lyu is a current post-doctoral researcher from School of Chemical Engineering, University of Queensland. He received his M.Sc. from Xiamen University in 2013, and Ph.D. degree from The University of Queensland (UQ), Australia in 2017. His research mainly focuses on metal halide perovskites for optoelectronic applications. 27

Mengmeng Hao is a current Ph.D. student under the supervision of Prof. Lianzhou Wang at School of Chemical Engineering, The University of Queensland, Australia. He received his B. Eng. (2012) and M. Eng. (2015) from Central South University, China. His research interests are mainly focused on the development of novel quantum dot solar cells.

Mehri Ghasemi is a current Ph.D. student under the supervision of Prof. Lianzhou Wang at School of Chemical Engineering, The University of Queensland, Australia. She received his B. Eng. (2011) from Shahid Chamran University of Ahvaz and M. Eng. (2013) from Shiraz University of Technology, Iran. Her research interests are mainly focused on the development of low-toxic bismuth-based semiconducting light-absorbing materials for photovoltaics.

Mu Xiao obtained her Ph.D. degree from the University of Queensland in 2019 under the supervisor of Prof. Lianzhou Wang and Dr. Bin Luo. She currently works as a post-doctoral fellow at the University of Queensland, Australia. Her research interest is developing functional nanomaterials for solar energy conversion and storage. 28

Jung-Ho Yun is an ARC DECRA Fellow, who has held an appointment at Nanomaterails Centre (Director: Prof. Lianzhou Wang), School of Chemical Engineering, the University of Queensland, Australia after his Ph.D. at the University of New South Wales, Australia (2013). His researches focus on photoelectrochemistry and high efficiency perovskite solar cells, including the fundamental study on defects in halide perovskite single crystals for optoelectronic applications.

Yang Bai is a current discovery early career research fellow, Austraian Research Council, working with Professor Lianzhou Wang in Nanomateirals Centre, The University of Queensland (UQ), Australia. He received his Ph.D. degree from School of Chemical Engineering, UQ in 2014. Before moving back to UQ in 2017, he has worked in Prof. Jinsong Huang's group at the University of Nebraska-Lincoln in the United States for two years. His research interests focus on the development of low-cost, efficient and robust thin film photovoltaics including perovskite solar cells and quantum dot solar cells.

29

Lianzhou Wang is a Professor in School of Chemical Engineering and Director of Nanomaterials Centre, The University of Queensland (UQ), Australia. He received his Ph.D. degree from Shanghai Institute of Ceramics, Chinese Academy of Sciences in 1999. Before joining UQ in 2004, he has worked at two national institutes (NIMS and AIST) of Japan for five years. His research interests include the design and application of semiconducting nanomaterials in renewable energy conversion/storage systems, including photocatalysts, low cost solar cells and rechargeable batteries.

30

Highlights Down-shifting Eu-TiO2 can emit efficient visible luminescence under UV excitation.

The use of Eu-TiO2 in perovskite solar cells prolongs device lifespan under UV light.

Perovskite solar cells based on dual-functional Eu-TiO2 show a high PCE of 21.40%.

The low-temperature chemical-bath deposition of Eu-TiO2 can be scaled up to 64 cm2.