Low Damage Scalable Pulsed Laser Deposition of SnO2 for p–i–n Perovskite Solar Cells

Pulsed laser deposition (PLD) has already been adopted as a low damage deposition technique of transparent conducting oxides on top of sensitive organic charge transport layers in optoelectronic devices. Herein, SnO2 deposition is demonstrated as buffer layer in p–i–n perovskite solar cells (PSCs) via wafer‐scale (4 inch) PLD at room temperature. The PLD SnO2 properties, its interface with perovskite/C60, and device performance are compared to atomic layer deposited (ALD) SnO2, i.e., state‐of‐the‐art buffer layer in perovskite‐based single junction and tandem photovoltaics. The PLD SnO2‐based solar cells exhibit on par efficiencies (17.8%) with that of SnO2 fabricated using ALD. The solvent‐free room temperature processing and wafer‐scale approach of PLD open up possibilities for buffer layer formation with increased deposition rates while mitigating potential thermal or physical damage to the top organic layers. This is a promising outlook for fully physical vapor‐processed halide PSCs and optoelectronic devices requiring low thermal budget.


Introduction
SnO 2 has been a key enabler of the record efficiencies of halide perovskite-based single junction (26.0%) and Si/perovskite tandem solar cells (33.7%). [1]SnO 2 owns superior properties such as high spectral transmittance (E g = 3.6-4.24] Moreover, it is earth abundant, can be processed via several methods, and shows excellent mechanical, thermal, chemical, and UV-light stability [5][6][7][8][9][10] The reported perovskite/Si and perovskite/CIGS tandems have integrated SnO 2 in both n-i-p and p-i-n solar cell configurations.[13] In p-i-n structure, on the other hand, depositing SnO 2 requires methods to prevent direct damage to the organic transport layer underneath and the halide perovskite absorber.[15] This is crucial because the high kinetic energy of the sputtered species could damage the underlying organic and perovskite layers by directly breaking the chemical bonds or by inducing stress due to the mismatch in the thermal expansion coefficients. [13,16]nO 2 has been fabricated using various techniques ranging from solution-based methods such as sol-gel and chemical bath deposition to vapor-based methods such as e-beam evaporation and sputtering. [6,7,12,17,18]Among all, atomic layer deposition (ALD) has been most widely used due to its capability to deposit uniform and compact films on textured surfaces and over large areas due to self-limiting surface chemistry. [19]While ALD SnO 2 has been the reference material in high-efficiency perovskite solar cells, challenges such as limited growth rates are currently addressed by processing SnO 2 in batch and spatial ALD systems.Moreover, recent literature studies have addressed the chemical interaction between ALD precursors and perovskite chemistry. [20,21]ere, we propose the use of 4 inch wafer-scale pulsed laser deposition (PLD) as a physical vapor deposition method to deposit SnO 2 in p-i-n configuration (Figure 1).PLD is a dry, vapor phase deposition method in which the film is formed by laser-ablated material from a target.Wafer-scale PLD of SnO 2 has previously been reported by Zanoni et al. who integrated SnO 2 in all vapor-deposited n-i-p perovskite solar cells DOI: 10.1002/solr.202300616Pulsed laser deposition (PLD) has already been adopted as a low damage deposition technique of transparent conducting oxides on top of sensitive organic charge transport layers in optoelectronic devices.Herein, SnO 2 deposition is demonstrated as buffer layer in p-i-n perovskite solar cells (PSCs) via wafer-scale (4 inch) PLD at room temperature.The PLD SnO 2 properties, its interface with perovskite/C 60 , and device performance are compared to atomic layer deposited (ALD) SnO 2 , i.e., state-of-the-art buffer layer in perovskite-based single junction and tandem photovoltaics.The PLD SnO 2 -based solar cells exhibit on par efficiencies (17.8%) with that of SnO 2 fabricated using ALD.The solvent-free room temperature processing and wafer-scale approach of PLD open up possibilities for buffer layer formation with increased deposition rates while mitigating potential thermal or physical damage to the top organic layers.This is a promising outlook for fully physical vapor-processed halide PSCs and optoelectronic devices requiring low thermal budget.
demonstrating efficiencies of 18.2%. [22]Chen et al. used PLD to deposit SnO 2 on a flexible substrate realizing a perovskite solar cell (PSC) with 14.0% efficiency. [23]In addition, low damage deposition of tin-and zirconium-doped indium oxides (ITO & IZrO) by PLD as the top electrodes in p-i-n PSCs has been demonstrated in a number of studies. [24,25]This stems from the possibility to use a broad range of deposition pressures and laser energy densities to mitigate the kinetic energy of the ablated species arriving to the substrate.
In this work, we first present a comparison between PLD SnO 2 and ALD SnO 2 films in terms of their optical properties and solar cell performance within the same batch of opaque and semitransparent PSCs.The p-i-n PSCs are composed of the following structure: ITO-coated glass/2-PACz/Cs 0.15 FA 0.85 PbI 2.76 Br 0.24 /MgF 2 /C 60 /SnO 2 /Ag, where 2-PACz (([2-(9H-carbazol-9-yl) ethyl]phosphonic acid) is a self-assembled monolayer performing as a hole-transport layer (HTL), and FA stands for formamidinium.The unencapsulated PLD SnO 2 -and ALD SnO 2 -based cells exhibited power conversion efficiencies above 17.8% and retain a shelf life time of 86% and 87.8%, respectively, after 70 d of storage inside the N 2 -filled glove box.Therefore, both PLD and ALD SnO 2 can serve as efficient and stable buffer layers in PSCs.Next, we discuss the effects of PLD parameters such as O 2 pressure and the laser fluence on the optical properties of PLD SnO 2 and on the performance of full PSCs.

PLD and ALD SnO 2 : A Comparative Study
PLD of SnO 2 was performed on a wafer-scale system as depicted in Figure 1a.The 4 inch wafer size holder is scanned in x-y plane over the laser-generated plasma plume to form a uniform layer over the mounted substrates.The thickness homogeneity of the deposited film is above 98.5% over the full 100 cm 2 area. [22]The deposition is carried out at room temperature and at working pressure of 5.0 Â 10 À3 mbar under 100% O 2 atmosphere.ALD process used is reported elsewhere. [21]To check the crystalline phase of the grown layers, X-ray diffraction (XRD) measurements were carried out for the SnO 2 films grown on glass as given in Figure S1a, Supporting Information.No diffraction peak was observed for ALD-and PLD-grown SnO 2 layers, indicating amorphous nature of the grown SnO 2 films.The SnO 2 processes were tested on p-i-n devices with the stack presented in Figure 1b.
The optical properties of the PLD and ALD SnO 2 layers grown on glass were measured by UV-vis-NIR spectroscopy and spectroscopic ellipsometry as shown in Figure 2a,b.The associated Tauc plot is also calculated and shown in Figure S1b, Supporting Information.The estimated bandgap is 3.3 and 3.1 eV for ALD and PLD SnO 2 layers, respectively.PLD SnO 2 has a higher parasitic absorptance compared to the case of ALD in UV to visible spectrum.This is correlated to O:Sn atomic ratio in the SnO 2 layers.To quantify the atomic concentrations of O and Sn, X-ray photoelectron spectroscopy (XPS) surface measurements, summarized in Figure S2 and S3, Supporting Information, were carried out following a gentle surface cleaning with a 2s Ar þ sputtering step to remove adsorbed species caused by air exposure during sample transfer.The measurements confirmed the presence of Sn in the Sn 4þ oxidation state, whose 3d 5/2 peaks were detected at 486.9 AE 0.1 and 486.7 AE 0.1 eV, for PLD and ALD SnO 2 , respectively. [22]Also, the results indicate that the PLD layer has an O:Sn ratio of 1.8, with no C or N being detected.It is important to note that throughout the article the notation of PLD SnO 2 is used despite the off-stoichiometric Sn:O ratio (1:1.8)measured by XPS.The ALD layer instead presents 6.5% and 2.5% of C and N contamination, respectively.The inclusion of C and N species is the result of incomplete precursor ligand elimination during the low-temperature thermal ALD process. [26]An O:Sn ratio of 2.0 is calculated for ALD SnO 2 , if hydroxyl species are not included.If instead those are included, a ratio of 2.3 is obtained.Such a high O:Sn ratio can be explained by the inclusion of unreacted OH groups caused by the steric hindrance of the chemisorbed Sn-precursor molecules. [27]The lower substoichiometric ratio of O:Sn implies higher oxygen vacancies (V Ö ) in the case of the PLD SnO 2 layer.It has previously been shown that high oxygen vacancies (V Ö ) in PVD-grown SnO 2 result in lower optical transmittance in UV-visible region of the spectrum. [22,28]However, this is not an issue for p-i-n stacks as SnO 2 is at the rear side of the solar cell.This is confirmed by the similar external quantum efficiency (EQE) and reflectivity spectra measured for the full solar cell stack as shown in Figure 3a.The slight difference in wavelength region of 650-800 nm is due to the different measured thicknesses of the ALD and PLD SnO 2 on c-Si, that is, 24 and 19 nm, respectively.On the other hand, the refractive index (n) of PLD SnO 2 is significantly higher than that of ALD SnO 2 , e.g., n PLD = 2.17 at a wavelength (λ) of 633 nm while n ALD = 1.85 at the same wavelength.This indicates a denser SnO 2 layer when grown by PLD.
The lower density of ALD SnO 2 is attributed to the detected C and no N contaminations compared to PLD SnO 2 , according to XPS depth profile analysis performed on PLD and ALD SnO 2 on c-Si samples shown in Figure S4a,b, Supporting Information.That is a result of incomplete precursor ligands elimination during the low-temperature ALD process. [26]While a denser SnO 2 layer could serve as a stronger barrier to the ingress of environmental molecules such as H 2 O, it might be beneficial to reduce the required thickness of the layer needed to protect against ITO sputtering damage.In parallel, this would also reduce the parasitic absorbance of the PLD SnO 2 , thus minimizing optical losses.In regard to morphology, atomic force microscopy (AFM) images show compact and smooth layers for both ALD-and PLD-grown SnO 2 on glass (root mean square (RMS) in pm range), C 60 (RMS of 1.2 nm), and on perovskite/ C 60 substrates (RMS of 8.6 and 12.0 nm), as shown in Figure S5, Supporting Information.To confirm the nondamaging SnO 2 deposition processes, XPS depth profiles were carried out and the corresponding atomic concentration trends, reported as relative at%, are given in Figure 2c,e.The transition between the Ag top contact, the SnO 2 layers, and the underlying C 60 electron transport layer (ETL) is all well defined for both ALD and PLD SnO 2 .This indicates that, at a first approximation, there is no diffusion/growth of SnO 2 into C 60 , and that the homogeneity and material density of ALD and PLD SnO 2 are sufficient to prevent the diffusion of Ag toward the ETL.This is further supported by cross-sectional scanning electron microscope (SEM) analysis, as shown in Figure 2d,f.While no difference in the length of the sputtering process for the two layers is detected, different at% of Sn is detected for PLD and ALD SnO 2 .This difference stems from the combined effect of both density and thickness of the two SnO 2 layers.As mentioned earlier, ALD SnO 2 is 5 nm thicker than its counterpart but at the same time has a lower density following its lower refractive index.Therefore, while less Sn is detected in PLD SnO 2 , the sputtering process to go through the layer has the same duration as the case with ALD SnO 2 .
The performance of the ALD and PLD SnO 2 layers was compared in p-i-n PSCs in the same batch.As control solar cells, an ETL stack of C 60 /bathocuproine (BCP) was used to monitor the impact of the experimental processes.The EQE and J-V characteristic curves for the resultant PSCs are shown in Figure 3a,b, respectively.The statistical data are given in Figure S6, Supporting Information.While all the cells exhibit very close J sc values, there is a slight hysteresis in the J-V curves. [29]For instance, the best performing cell with ALD SnO 2 exhibited a hysteresis index (HI) of 6.2% versus 0.2% of the best performing PLD SnO 2 cell.The ALD SnO 2 -based cells showed somewhat higher V oc (1069 mV) than PLD SnO 2 (1054 mV).To determine whether the loss is due to higher nonradiative recombination caused by any process damage to the C 60 layer in the case of PLD SnO 2 , we measured photoluminescence quantum yield (PLQY) and the subsequent implied V oc (iV oc ) considering the quasi-Fermi-level splitting in the absorber for both ALD and PLD SnO 2 -based cells. [30]The calculation is via: iV oc = V oc,rad þ KT/q ln(PLQY), where V oc,rad indicates the radiative limit of V oc at the emission wavelength of the absorber.The results are given in Figure 3d.Considering the emission of the Cs 0.15 FA 0.85 PbI 2.76 Br 0.24 layer at λ = 775 nm, the iV oc for PLD SnO 2 (1139 mV) was 4 mV lower than ALD SnO 2 (1143 mV).The similar iV oc for ALD and PLD SnO 2 -based solar cells implies negligible process-induced differences at the SnO 2 / C 60 interfaces during ALD and PLD.On the other hand, a discrepancy between the measured and implied V oc is observed in both the ALD and PLD cases, with the iV oc higher in both cases.This is a well-understood phenomenon associated with the difference in the illuminated area of the cells during J-V (1.00 mm 2 ) and PLQY (3.46 mm 2 ) measurement. [31]An example of measured V oc for a device fabricated with ALD SnO 2 and with different illumination areas is shown in Figure S7, Supporting Information.Nevertheless, we can conclude that the proof-ofconcept wafer-scale PLD SnO 2 layer in p-i-n PSCs performs virtually on par with ALD SnO 2 .To confirm the rigidity of 20 nm SnO 2 grown by PLD and ALD against subsequent TCO deposition relevant for tandem structures, semitransparent (ST) cells were fabricated with ITO as the top electrode instead of Ag.The deposition of ITO was carried out through PLD. [25]The nominal EQE and J-V curves for the cells are given in Figure S8, Supporting Information.While the EQE spectra for both PLD and ALD SnO 2 ST-PSCs are similar, ALD SnO 2 -based cells exhibited slightly higher FF (78.3% vs 75.6%) and thereby higher efficiency than PLD (17.1% vs 16.5%).It should be noted that the effect of SnO 2 thickness variation on optical and electrical properties of SnO 2 layer will be addressed in future.It is important to note that the high-quality SnO 2 grown by PLD is deposited at a rate of 0.64-0.70nm min À1 while the SnO 2 ALD is performed at 0.44-0.48nm min À1 .The higher deposition rate and room temperature growth in the case of PLD is useful for higher throughput production and growth on flexible substrates relevant for industrial use. [32]The unencapsulated solar cells with both PLD and ALD SnO 2 exhibited stable performance as measured by maximum power point tracking and shown in Figure 3c.They also retained 86.0% (PLD SnO 2 ) and 87.8% (ALD SnO 2 ) of their initial efficiency when kept under N 2 gas environment for 70 d as shown in Figure S9, Supporting Information.The control BCPbased cell, on the other hand, dropped to 81.7% of its initial efficiency.

Effect of PLD Parameters on the SnO 2 Properties and Cell Performance
While the analysis above was performed on an optimized PLD SnO 2 (room temperature, 1.0 J cm À2 , 100% O 2 , 5.0 Â 10 À3 mbar working pressure), it is important to note that the PLD process parameters have a strong influence in the final device performance.To demonstrate this, the effect of deposition pressure was first investigated while the laser fluence (F) was kept constant at 1.5 J cm À2 .Note that for all deposition pressures, the introduced gas is 100% O 2 .UV-vis-NIR measurements revealed that increasing the deposition (O 2 ) pressure from 2.5 Â 10 À3 to 1.0 Â 10 À2 mbar enhances the optical transparency with a slight shift of the band edge to higher energies, as shown in Figure 4a.This is attributed to the decreased oxygen vacancies within SnO 2 at higher O 2 pressure. [22]The performance of the SnO 2 films depending on the O 2 pressure was tested in full solar cell stacks with results shown in Figure S10a, Supporting Information and the statistical data given in Figure S10b, Supporting Information.It is important to note that as the deposition pressure was in the range of 10 À3 mbar, the deposition was preceded with a thin (%2.5 nm) SnO 2 intermediate layer (IL) deposited at 0.1 mbar, which was used previously to prevent possible damage to the sublayers. [25]Overall, even though the cells with 5.0 Â 10 À3 mbar of pressure showed the highest average J sc and V oc values, the J-V curves exhibited an S-shape which was exacerbated with increased O 2 pressure to 10 À2 mbar regime.Additionally, this pressure regime caused a lateral electrical short between the top Ag fingers following the abnormally high J sc values in J-V curves as shown in Figure S10a, Supporting Information.To elucidate the reason for such behavior, Hall-effect measurements were carried out for 20 nm SnO 2 layers on glass deposited under an O 2 pressure of 10 À3 and 10 À2 mbar with results given in Table S1, Supporting Information.While a pressure regime of 10 À3 mbar results in highly resistive SnO 2 , a higher pressure of 10 À2 mbar resulted in conductive SnO 2 (conductivity of 165.8 S cm À1 ) corresponding to a high carrier concentration of 7.7 Â 10 19 cm À3 and a relatively high mobility of 13.5 cm 2 V À1 s À1 .This type of conductive SnO 2 has been previously reported by a number of studies. [28,33]The high carrier concentration may further compromise the hole blocking property of the SnO 2 layer by increasing the charge carrier recombination. [34,35]Thus, we evaluated whether the 2.5 nm thick SnO 2 IL is beneficial as it is deposited at an even higher O 2 pressure of 0.1 mbar.To do this, PSCs including SnO 2 layers with and without the IL were fabricated at a pressure of 5.0 Â 10 À3 mbar (Figure S11, Supporting Information).It was found that the SnO 2 IL deposited at high O 2 pressure exacerbates the S-shape in the characteristic J-V curves of the PSCs as shown in Figure S11b, Supporting Information.There was a net 4% boost in the average efficiency of the cells without the IL which is due to significant FF and V oc improvement.In a parallel set of samples, keeping a constant O 2 pressure of 5.0 Â 10 À3 mbar, the effect of the laser fluence (F) was investigated.Hence, three F values of 2.0, 1.5, and 1.0 J cm À2 were tested on the performance of the PLD SnO 2 in PSCs.The results are shown in Figure S12, Supporting Information.It was found that there is a clear uptrend in V oc and FF once F decreases from 2.0 to 1.0 J cm À2 .Lower fluence is associated with less kinetic energy of the ablated material, which could also lead to lower damage to the underlying C 60 sublayer.Following the above conclusions, PLD SnO 2 was made without the IL and with low fluences with the results shown in Figure 4b and S13, Supporting Information.It was interesting to see that a further decrease in the laser fluence to 0.3 J cm À2 did not yield any meaningful performance boost of the solar cells with somewhat lower average FF values.The best performing PSC with an average efficiency of 16.6% (champion: 17.3%) belonged to the PLD SnO 2 layer deposited with F = 1.0 J cm À2 at a pressure of 5 Â 10 À3 mbar.This was mainly due to its higher average FF of 75.5% compared to 71.8% measured for the case of F = 0.3 J cm À2 .

Conclusion
PLD SnO 2 for p-i-n devices was demonstrated as a room temperature, buffer layer alternative to ALD SnO 2 .The PLD SnO 2 layers are dense, smooth, and resulted in nonhysteretic PSCs with efficiencies of 17.8% comparable to that of state-of-the art ALD SnO 2 (18.3%).The p-i-n PSCs based on PLD retained 86.0% of their performance after 1680 h.The wafer scale capability of the PLD and the demonstrated capability for dry and room temperature deposition of metal oxides broaden the application of PLD for low damage contact layer deposition onto chemically or physically sensitive devices.

Experimental Section
PLD SnO 2 : PLD SnO 2 was grown by pulsed laser deposition (waferbased system from Solmates BV, now LAM research) from a SnO 2 target at room temperature.All experiments were performed using a KrF (λ = 248 nm) Compex-pro laser (COHERENT) with a pulse repetition rate of 10 Hz.To conduct PLD, the substrates were transferred via a loadlock to another chamber evacuated to 2 Â 10 À7 mbar.The wafer holder with the mounted substrates was scanned on the laser plasma plume to form a uniform layer on 100 cm 2 area (adaptable to 225 cm 2 ).The depositions were carried out under pure O 2 pressure.
ALD SnO 2 : ALD SnO 2 films were carried out in an Oxford Instruments FlexAL thermal ALD reactor at 10 À5 mbar and 100 °C substrate temperature.
The growth process employed tetrakis(dimethylamido)-Sn(IV) (TDMA-Sn), 99.9% from STREM Chemicals, as the metal-organic precursor and water as the coreactant.The precursor was kept at 50 °C and supplied to the ALD chamber in vapor-drawn mode.An ALD cycle consisted of a 500 ms TDMA-Sn dose, followed by a purge step of 5 s, then a 50 ms H 2 O vapor dose, followed by a purge step of 10 s. [21] Solar Cell Fabrication: The prepatterned ITO-coated glass substrates (Ossila, 20 Ω sq À1 ) were sonicated for 10 min in 2 vol% Hellmanex III (Sigma-Aldrich) aqueous solution, acetone, and ethanol, subsequently.Later, the substrates were treated by UV-ozone cleaner for 15 min and transferred to N 2 -filled glove box (Kiyon).As HTL, 3.6 mmol of 2-PACz ([2-(9H-carbazol-9-yl)ethyl]phosphonic acid purchased from TCI-Chemicals) was dissolved in 1 mL of dry ethanol (Acros Organics) and spin-coated at 5000 rpm for 30 s onto the substrates before annealing at 100 °C for 10 min.Later, the 2-PACz-coated substrates were washed 2 times by ethanol and annealed for 5 min at 100 °C.To fabricate the perovskite layer, a 1.33 M solution of Cs 0.15 FA 0.85 PbI 2.76 Br 0.24 was prepared using CsI (99.999%, abcr), FAI (99.9%,Greatcell Solar Materials), FABr (99.9%,Greatcell Solar Materials), PbI 2 (99.999%,Sigma-Aldrich), and PbBr 2 (99.9%,TCI) in DMF:DMSO (9:1 vol.ratio).Subsequently, 100 μL of the perovskite precursor solution was spin-coated at 2000 rpm for 10 s followed by 5000 rpm for 30 s while 240 μL of trifluorotoluene was dripped as antisolvent 20 s before the spin coating ended.Next, the substrates were annealed at 100 °C for 10 min.The perovskitecoated substrates were later transferred into a glove box adapted evaporation chamber (NANOVAK R&D) and evacuated to a base pressure of 2 Â 10 À7 mbar to deposit 0.5 nm of MgF 2 as the passivation layer followed by 25 nm of C 60 vapor deposition. [36]Then, the substrates were transferred to dedicated vacuum chambers to deposit SnO 2 using PLD or ALD as explained above.As reference cells, C 60 -terminated substrates were coated with a 7 nm of BCP.Later, the samples were transferred to the glove box adapted vacuum chamber for depositing 100 nm Ag as the top electrode.
Device Characterization: The J-V measurement of the devices was carried out under a Wavelabs LED-based Sinus 70 sun simulator (class AAA) with A.M1.5G spectrum (100 mW cm À2 ) and scan rate of 0.11 V s À1 and a voltage step size of 15 mV.The devices were illuminated using a shadow mask with 1 mm 2 area.The devices are unencapsulated and the measurement was carried out at room temperature in the air (humidity of 36%).
Scanning Electron Microscopy: SEM images were acquired with an acceleration voltage of 1.4 kV using an in-lens detector within Zeiss MERLIN HR-SEM tool.
Ellipsometry: The thickness of the ALD and PLD SnO 2 layer is measured by means of spectroscopic ellipsometry, using a J.A. Woollam Inc. M2000 UV ellipsometer and the measured growth per cycle is 0.11 (AE0.01)nm.The SnO 2 layers are fitted using a Cauchy model with an Urbach tail absorption considering a bandgap of 3.6 eV.
XPS: The XPS depth profile measurements were performed on a KA1066 spectrometer with a monochromatic Al K-α X-ray source (hν = 1486.6eV) from Thermo Fisher Scientific and analyzed with the Thermo Scientific Avantage software.The surface scans were carried out after a 2s Ar þ sputtering step with an ion energy of 200 eV.The depth profiles were performed with an ion energy of 500 eV and an exposure of 5s.Elemental atomic concentrations, given as at%, were performed.Note that due to preferential sputtering the depth profile results were qualitative and not indicative of materials' stoichiometries.
Optical Properties: The optical properties of the glass/SnO 2 films were measured on a UV-vis-NIR spectrophotometer PerkinElmer Lambda 950S using an integrating sphere.
Hall Effect: The sheet resistance, Hall mobility, and carrier concentration of glass/SnO 2 samples were measured on a Hall effect measurement setup in the van der Pauw configuration, using 1.5 Â 1.5 cm 2 prediced samples.
AFM: AFM images of the samples were obtained using a Bruker ICON Dimension microscope in tapping mode.
Absolute PL: Photoluminescence was measured using a 520 nm excitation laser set to a photon flux of 1 sun.The luminescence was collected with a StellarNet Blue-wave spectrometer.To calibrate the system and measure the absolute photon flux, an Avantes calibrated halogen light source was used.This calibrated light source was equipped with a cosine corrector to provide a Lambertian emission of light, and subsequently aligned to the surface of the sample to perform the calibration.
XRD: Measurements were done in a symmetric configuration using a PANalytical X'Pert PRO with a Cu anode kα (λ = 1.54 Å) X-ray source.

Figure 1 .
Figure 1.a) Schematics of the PLD system for the 4 00 size sample holder with 12 sample slots.b) PSC structure used throughout the study.In the case of control devices, BCP layer is used instead of SnO 2 .

Figure 2 .
Figure 2. a) The optical transmittance, reflectance, and absorptance measurement for glass/SnO 2 deposited by optimized PLD and ALD.b) Refractive index measurements for PLD and ALD SnO 2 grown on Si substrate, XPS depth profiles from the full PSC stack with SnO 2 grown by c) ALD (24 nm) and e) PLD (19 nm).Cross-sectional SEM images of perovskite solar cells with d) ALD SnO 2 and f ) PLD SnO 2 .

Figure 3 .
Figure 3. a) EQE and 1-R measurements for the full solar cell stacks with ALD and PLD SnO 2 as the ETLs, b) J-V curves measured for best performing PSCs based on optimized PLD SnO 2 , ALD SnO 2 , and BCP as the control cells (the inset table shows the performance summary and the HI), c) maximum power point tracking for PSCs based on PLD and ALD SnO 2 over 300 s under continuous 1 sun illumination of 100 mW cm À2 , and d) photoluminescence and the correlated implied V oc (iV oc ) of full solar cells based on optimized PLD SnO 2 and ALD SnO 2 .

Figure 4 .
Figure 4. a) Optical properties of PLD SnO 2 (T, R, A) deposited under different deposition pressures.b) J-V curves measured for PLD SnO 2 -based PSCs where two laser fluence (F) values of 0.3, 1.0, and 1.5 J cm À2 were used to grow SnO 2 .The SnO 2 layer is 35 nm thick in this experiment.The inset indicates a table of the performance parameters of the cells.