Enhancement in the Efficiency of Sb2Se3 Solar Cells by Triple Function of Lithium Hydroxide Modified at the Back Contact Interface

Abstract The efficiency of antimony selenide (Sb2Se3) solar cells is still limited by significant interface and deep‐level defects, in addition to carrier recombination at the back contact surface. This paper investigates the use of lithium (Li) ions as dopant for Sb2Se3 films, using lithium hydroxide (LiOH) as a dopant medium. Surprisingly, the LiOH solution not only reacts at the back surface of the Sb2Se3 film but also penetrate inside the film along the (Sb4Se6)n molecular chain. First, the Li ions modify the grain boundary's carrier type and create an electric field between p‐type grain interiors and n‐type grain boundary. Second, a gradient band structure is formed along the vertical direction with the diffusion of Li ions. Third, carrier collection and transport are improved at the surface between Sb2Se3 and the Au layer due to the reaction between the film and alkaline solution. Additionally, the diffusion of Li ions increases the crystallinity, orientation, surface evenness, and optical electricity. Ultimately, the efficiency of Sb2Se3 solar cells is improved to 7.57% due to the enhanced carrier extraction, transport, and collection, as well as the reduction of carrier recombination and deep defect density. This efficiency is also a record for CdS/Sb2Se3 solar cells fabricated by rapid thermal evaporation.


Introduction
Antimony selenide (Sb 2 Se 3 ) thin films are a promising choice for low-cost, highefficiency thin-film solar cells.Other desirable optoelectronic characteristics of Sb 2 Se 3 film include a sufficient bandgap (1.03 eV) and a high absorption coefficient (>10 5 cm −1 ).[3][4][5][6] In recent years, there has been significant progress in increasing the efficiency of Sb 2 Se 3 thin-film solar cells. [7]The highest efficiency reported so far for Sb 2 Se 3 solar cells is 10.57% for an superstrate structure, in which the Sb 2 Se 3 film was prepared using the chemical bath deposition (CBD), [8] and 10.15% for a substrate structure, in which the Sb 2 Se 3 film was prepared using the injection vapor deposition (IVD) method. [9]However, these efficiencies are still far below the theoretical limit of 32.74% for Sb 2 Se 3 solar cells. [10]Therefore, increasing efficiency remains the most critical aspect for the further development of this devices. [11]he efficiency of Sb 2 Se 3 solar cells is restricted by several factors. Another reason is the high carrier recombination at the device's interface, leading to decreased carrier transport and collection at the heterojunction interface and back surface. [14]Numerous studies have addressed these two issues, such as the in situ passivation to decrease deep-level defect density by IVD and CBD methods. [8,9]owever, these methods are unsuitable for large-scale and lowcost industrial production.Therefore, developing a simple and cost-effective deep-level defect passivation method is critical for increasing the efficiency and further development of Sb 2 Se 3 solar cells.Moreover, doping in semiconductors is essential for band energy engineering to address the issue of device interface defect passivation.[17] However, the fermi level pinning effect caused by deep-level defects in Sb 2 Se 3 film hinders effective band structure adjustment through doping, leading to insignificant improvement in the efficiency of Sb 2 Se 3 solar cells.Therefore, effective methods to modify the band alignment at the Sb 2 Se 3 /CdS interface are urgently needed to increase the open circuit voltage (V OC ) of Sb 2 Se 3 solar cells further. [18]Furthermore, carrier collection at the back surface is still limited by the back contact barrier between the Sb 2 Se 3 film and high work function Au electrode, thus, restricting the efficiency increase of Sb 2 Se 3 solar cells. [19]Constructing the p-i-n structure by selecting an appropriate hole transport layer to form a ladder band arrangement can significantly improve the device's charge carrier collection.The organic material Spiro-OMeTAD acts as the hole transport layer in the record efficiency of Sb 2 S 3 , Sb 2 Se 3 , and Sb 2 (S,Se) 3 solar cells.However, the Li-TFSI used in Spiro-OMeTAD is prone to deliquescence in the air, and the toxicity of 4-tert-butylpyridine and chlorobenzene poses a potential safety hazard. [20,21]Although many inorganic films have been used as the hole transport layer to increase the efficiency of Sb 2 Se 3 solar cells, the efficiency is still far below the recording device that used Spiro-OMeTAD as the hole transport layer.Therefore, identifying a more suitable hole transport layer or reaction etching at the back surface is crucial for the future development of Sb 2 Se 3 solar cells. [22,23]To address these issues, this paper proposes using small atomic diameter lithium (Li) ions as dopants to dope the Sb 2 Se 3 film, with lithium hydroxide (LiOH) acting as the dopant medium.Doping with Li ions can simultaneously solve the above three problems.Li-ion doping can invert the carrier type of the grain boundary to decrease the deep defect density.With the diffusion of Li ions and the reaction at the back surface, the device's carrier transport and collection are increased.
In this study, we used small atomic diameter Li ions as dopants to dope the Sb 2 Se 3 film, with LiOH acting as the dopant medium.We hypothesized that Li ions could easily enter the large gap between (Sb 4 Se 6 ) n ribbons, invert the carrier type of the grain boundary, and obtain a built-in electric field from grain interiors (p-type) to grain boundary (n-type).During the LiOH and Sb 2 Se 3 film reaction, Li ions could slightly enter the lattice of the Sb 2 Se 3 film through Li ion diffusion.By varying the Li ions concentration, we could obtain Sb 2 Se 3 films with a variable gradient band structure.The reaction between Sb 2 Se 3 and LiOH solution increased the carrier collection and transport at the back contact surface.Furthermore, we extensively investigated the structure, crystallinity, orientation, and optical and electrical properties of Sb 2 Se 3 film with a Li-ion gradient.Finally, we achieved an efficiency of 7.57% for Sb 2 Se 3 solar cells with a low deep-level defect concentration.

Results and Discussion
This study explores the formation of a Li gradient field through upper interface doping via a solution method.The detailed experiment process is shown in Figure 1.The schematic diagram of the Li gradient field is illustrated in Figure 2a.As shown in the figure, Li ions can easily enter the large gap between (Sb 4 Se 6 ) n ribbons and subsequently enter the lattice due to the corrosivity of the LiOH solution.During the diffusion process, the Liion concentration decreases as the diffusion depth increases, ultimately forming a gradient of Li ions within the Sb 2 Se 3 film and resulting in the formation of the Li gradient field.Furthermore, the J-V curves of the device with different concentrations of LiOH solution are shown in Figure S1 (Supporting Information).The figure shows that when the concentration of LiOH solution was raised to 0.01 m, the efficiency of the device increased significantly.However, with a sustained increase in concentration, especially above 0.1 m, the efficiency of Sb 2 Se 3 solar cells decreased significantly.We believe that this phenomenon is caused by the corrosion damage of high-concentration LiOH solution on the Sb 2 Se 3 film.To validate our hypothesis and understand the impact of the Li gradient field on the Sb 2 Se 3 film, detailed characterization was performed on the Sb 2 Se 3 film and device.
To investigate the influence of Li-ion gradient on Sb 2 Se 3 film, the researchers conducted a detailed analysis of the structure, crystallinity, electrical, and optical properties of the film with and without LiOH solution (AD means the Sb 2 Se 3 film without LiOH solution, Li-GS means the Sb 2 Se 3 film with 0.01 m LiOH solution).Figure 2b displays the X-ray diffraction (XRD) patterns of both Sb 2 Se 3 films.It is evident from the figure that all the peaks of both films correspond to orthorhombic Sb 2 Se 3 (JCPDS#15-0861) with a preferred orientation along the [221] and [211] directions. [24]Moreover, the intensity ratio between [221] and [211] peaks of the film with Li-ion gradient is higher than that of the film without Li-ion gradient (Figure S2, Supporting Information).This phenomenon indicates that the carrier transport of Sb 2 Se 3 film is higher in the sample with a Li-ion gradient. [25]In order to further study the effect of Li-ion alkaline solution on Sb Figure 2c,e display the X-ray photoelectron spectroscopy (XPS) results for Sb 2 Se 3 films with and without a Li gradient.The Sb 3d spectrum in Figure 2c reveals that the binding energies of Sb(III) are situated at 538 and 528.7 eV, whereas the Sb 3d 3/2 and Sb 3d 5/2 peaks are located at 538.6 and 531.1 eV, respectively, which are in agreement with literature reports. [26]However, the additional Sb─O bonds observed in Figure 2c are due to slight corrosion on the upper interface of the Sb 2 Se 3 film after adding LiOH solution.A prior study has demonstrated that an appropriate thickness of the Sb─O layer can reduce carrier recombination and enhance carrier transport in the device. [27] The Se 3d peak is presented in Figure 2d, where the Se 3d 3/2 and Se 3d 5/2 peaks are located at 54.1 and 53.3 eV, respectively.Figure 2e shows the Li 1s spectrum of the Sb 2 Se 3 film with a Li-ion gradient, where the peaks at 53.5 and 54.2 eV correspond to Li 1s, respectively. [28]The XPS data demonstrate that even a low concentration of LiOH solution can result in slight corrosion on the surface of the Sb 2 Se 3 film, and Li-ion elements can enter into the Sb 2 Se 3 film.To examine the effect of LiOH solution on the morphology of Sb 2 Se 3 film, scanning electron microscopy (SEM) was performed on Sb 2 Se 3 films with and without 0.01 m LiOH solution.As illustrated in Figure 3a, the surface morphology of the Sb 2 Se 3 film without LiOH displayed a disorderly arrangement with numerous pin-holes on the surface.However, the Sb 2 Se 3 film with LiOH solution showed a flat and orderly surface with fewer pin-holes, indicating that the LiOH solution had a corrosive effect.Figure 3c,d present cross-sectional SEM images of Sb 2 Se 3 films with and without LiOH solution, revealing that the Sb 2 Se 3 film exhibited better crystallinity with larger grain sizes when LiOH solution was added.Atomic force microscopy (AFM) images of the Sb 2 Se 3 films (Figure 3e,f) indicated that the surface roughness decreased notably when the LiOH solution was added, consistent with the SEM results.A smoother surface is generally preferred for achieving high-quality interfaces between adjacent layers.A smooth surface reduces the recombination of light-induced excitations at the absorber-metal electrode interface, thereby improving cell performance. [30]o further confirm the presence of Li ions and the formation of a concentration gradient, XPS and time-of-flight secondary ion mass spectrometry (TOF-SIMS) were conducted on Sb 2 Se 3 films treated with alkaline solutions containing Li ions. Figure 4a-c illustrates the XPS results of Sb 2 Se 3 films with and without etching to a depth of 100 nm.The results demonstrate that the Sb─O bonds disappeared when the etching depth reached 100 nm, indi-cating that the Sb─O layer only exists on the surface of the Sb 2 Se 3 film. [31]Additionally, the Li-ion signal was detected even when the etching depth had reached 100 nm.The TOF-SIMS results in Figure 4d indicate that the Li-ion signal can be detected throughout the entire Sb 2 Se 3 thin film.Unlike the accumulation of potassium and aluminium ions in the midsection of the Sb 2 Se 3 film, the concentration of Li ions gradually decreases from the surface to the bottom of the Sb 2 Se 3 film. [32]Figure 4e presents a 3D render overlay of elements Sb, Se, Li, and Au, clearly showing the position of the elements.These results more intuitively demonstrate the Li-ion concentration gradient inside the Sb 2 Se 3 film.The XPS and TOF-SIMS results confirm our hypothesis regarding the concentration gradient of Li ions.The Li gradient field can be generated using this method.
Figure 5a illustrates the electrical performance of Sb 2 Se 3 films with and without a Li gradient field.As shown in the figure, the resistivity and mobility of the Sb 2 Se 3 film increased with the addition of Li gradient, whereas the hole carrier concentration decreased.Moreover, during the Hall test, the carrier type of the Sb 2 Se 3 film with Li ions was always n-type, which resulted mainly from the inversion of the carrier type at the grain boundary.In addition, the n-type Sb 2 Se 3 film has already been reported. [33]o further verify this phenomenon and investigate the effect of the Li gradient on band energy alignment, the ultraviolet photoelectron spectroscopy (UPS) spectra and energy alignment of the films with and without a Li gradient field are presented in   , E c ) was determined from the energy gap and VBM position.The energy levels of all samples were calculated from the UPS spectra and bandgap.The UPS results revealed that the distance between the E F and VBM was calculated to be 0.62 and 0.47 eV, respectively.With the addition of a Li gradient, the Fermi level moved toward the conduction band compared to its position in the film without a Li gradient.Additionally, the results demonstrated that the CBO between the Sb 2 Se 3 film and CdS film was 0.5 eV, whereas with the addition of a Li gradient, the CBO between the Sb 2 Se 3 and CdS films was 0.21 eV, which falls within the range of moderate spike-like CBO values (0-0.4 eV). [34]The efficiency of the device can be decreased since a big CBO raises the barrier for collecting photogenerated    In the band structure calculation process, it is challenging to estimate the band gap of the gradient Sb 2 Se 3 film.Therefore, we hypothesized that the band gap of the Sb 2 Se 3 film with different etching depths is consistent with the theoretical band gap.The E C for the sample with etching depths of 0, 150, 300, and 450 nm are −4.97,−4.4,−3.93, and −3.66 eV, respectively.These results support the steady enhancement of CBM structure in the Sb 2 Se 3 films.Figure 6c shows the energy-level alignment between the CdS film and Sb 2 Se 3 film with a Li gradient.Due to the gradient CBM structure, a large barrier is transformed into many small barriers, making it easier for carriers to cross over them. [35,36]ese relevant XPS and UPS results confirm the beneficial role of the gradient Sb 2 Se 3 film in achieving high-quality heterojunctions and efficient carrier transport.
Figure 7a displays the J-V curves of Sb 2 Se 3 solar cells with and without a Li gradient field.As shown in the figure, the addition of a Li gradient field has increased all of the device parameters, especially the V oc .The Sb 2 Se 3 solar cells with LiOH solution achieved an efficiency of 7.57%, with a V oc , current density (J sc ), and fill factor (FF) of 0.41 V, 30.5 mA cm −2 and 60.51%, respectively.While the Sb 2 Se 3 solar cells without LiOH solution obtained an efficiency of 5.2%, with a V oc , J sc , and FF of 0.34 V, 28.66 mA cm −2 and 52.75%, respectively.Figure 7b displays the external quantum efficiency (EQE) results of both  devices.As seen in the figure, the device with a Li gradient field exhibited a high optical response in the long wavelength, indicating a reduction in the recombination rate and high carrier collection.Additionally, biased EQE measurements (defined as EQE (−0.5 V)/EQE(0 V)) were conducted on both devices.The results showed that the observed EQE ratio of Sb 2 Se 3 solar cells with a Li gradient field was approximately unity during the entire spectrum.However, the observed EQE ratio of Sb 2 Se 3 solar cells without a Li gradient field was strongly dependent on the bias, especially in the long-wavelength region.The biased EQE results showed that the carrier collection of the device was optimized with the addition of a Li gradient field (Figure S8, Supporting Information).Figure 7c shows the dark IV results of both devices.The results indicated that with the addition of a Li gradient field, the diode ideal factor (A) and reverse saturation current density (J 0 ) decreased significantly.The low ideal factor and reverse saturation current density can decrease carrier recombination and promote charge transport.Temperature-dependent open circuit voltage measurements were conducted from 150 to 330K (shown in Figure 7d).The obtained value of activation energy for Sb 2 Se 3 solar cells with a Li gradient field is 1.  7e,f, and the built-in voltages (V bi ) of the device with and without treatment were 0.415 and 0.38 V, respectively.The improvement in V bi can decrease the accumulation of photogenerated charge at the interfacial region, leading to an enhancement in the device V OC , as reported in previous studies. [37,38]The depletion region width, which is used to analyze orientation-related defects, was obtained from the C-V graph.For the Sb 2 Se 3 film without and with Li gradient field, the depletion region widths were calculated to be 217 and 304 nm, respectively.A larger W d value in the device with a Li gradient field could result in a greater collection of photogenerated carriers and contribute to a larger J sc , as observed in the results of EQE. [39]o further examine the impact of the Li gradient field on the profound defects of Sb 2 Se 3 film, non-radiative recombination centers, specifically electrically active defects in Sb 2 Se 3 solar cells with and without Li gradient field, were characterized using deeplevel transient spectroscopy (DLTS).DLTS is a technique that tracks the changes in the charge state of deep defect centers at various temperatures using the transient capacitance of the p-n junction. [40]In this study, traps in the samples were found using the minority carrier injection DLTS (inj-DLTS).Both electrons and holes can be injected into the depletion zone when a forward pulse voltage is used.
Three negative peaks (Li-GS-E1 at 150 K, Li-GS-E2 at 180 K, and Li-GS-E3 at 270 K) corresponding to minority-carrier (electron) traps in the p-type Sb 2 Se 3 layer were seen in the DLTS spectra for the Sb 2 Se 3 solar cell with a Li gradient field, as shown in Figure 8a.The trap levels for Li-GS-E1, Li-GS-E2, and Li-GS-E3 were 285, 331, and 400 mV below the conduction band, respectively.In contrast, a positive peak was already observed at 370K in the DLTS spectrum.Conversely, in the DLTS spectrum of Sb 2 Se 3 solar cells without a Li gradient field, no negative peaks were observed, and three positive peaks (AD-H1 at ≈210 K, AD-H2 at ≈300 K, and AD-H3 at ≈370 K) corresponding to minority-carrier (hole) traps in the p-type Sb 2 Se 3 layer were observed.As shown in Figure 8b, we calculated the trap concentration and the associated capture cross-section for each trap state using Arrhenius plots of thermal emission rates as a function of reciprocal temperature.The defect information obtained is summarized in Table 1.
Figure 8c,d illustrate the energy states and defect levels of the AD device and the Li-GS solar cells.The width of each peaks corresponding to the capture cross-section and the length corresponding to the concentration of the defects.In our rapid thermal evaporation (RTE) method, Se vapor is always in excess due to its higher vapor pressure than Sb and Sb 2 Se 3 , resulting in slightly Se-rich Sb 2 Se 3 films.The previous study reported that under Serich conditions, the dominant acceptor defects are antimony vacancy (V Sb ) and selenium antisite (Se Sb ) defects. [41]Therefore, we suspect that AD-H1 is due to the V Sb defect, while AD-H2 and H3 are tentatively attributed to Se Sb1 and Se Sb2 , respectively. As the location of Li-GS-H1 is close to the defect of AD-H3, we suspect that the Li-GS-H1 defect is due to Se Sb2.However, in the DLTS results with the addition of LiOH solution, three electrical traps appeared.We speculate that this phenomenon is due to two reasons.First, previous research has reported that the distribution of defects in trichalcogenides is dominated by antisite defects due to the similar sizes of the constituent atoms.Second, simulations have shown that Se Sb and antimony antisite (Sb Se ) are acceptors and donor defects, respectively. [42]We suspect that the Li-GS-E3 defect is due to the Se Sb defect.Second, as the location of Li-GS-E1 and Li-GS-E2 is very close to the CBM, we believe that these two defects can be attributed to Li itself.Our hypothesis is that Li ions first accumulate on the surface of the Sb 2 Se 3 film and then diffuse easily into the (Sb 4 Se 6 ) n chains and lattice of the Sb 2 Se 3 film due to the corrosive nature of LiOH solution and the large distance and small compactness at GBs. Microscale Li ions diffuse easily along the GBs and enter the gaps of the chains, resulting in n-type Li i doping.This inversion creates a local electric field between GBs and GIs, facilitating spatial separation of photogenerated electrons and holes, thereby restraining carrier recombination and enhancing carrier collection, especially for carriers generated by red and infrared photons, resulting in im-proved efficiency.Additionally, it is worth noting that the defect density of the Sb 2 Se 3 film with added LiOH solution decreased by order of magnitude.This reduction in defect density should decrease carrier recombination and increase carrier transport in the device.

Conclusion
In summary, the LiOH solution was used to modify the upper interface of the device.Due to the sensitivity of the Sb 2 Se 3 film to the alkaline solution, Li ions not only formed a back contact layer at the upper interface but also entered into the lattice of the Sb 2 Se 3 film to enhance its crystallinity, optical, and electrical properties.The UPS results also indicated that with the addition of LiOH solution, a gradient band structure was formed inside the Sb 2 Se 3 film, which should improve the carrier collection and transport of the device.The DLTS results also showed a reduction in defect density and grain boundary inversion which should suppress carrier recombination and enhance carrier collection, resulting in an improvement in PCE.Finally, an efficiency of 7.57% was achieved for Sb 2 Se 3 solar cells.This is the highest efficiency reported for Sb 2 Se 3 /CdS solar cells based on rapid thermal evaporation.We believe that this work paves the way for further improvements in the efficiency of Sb 2 Se 3 solar cells.

Experimental Section
The Fabrication of Films and Devices: In this paper, the device is fabricated based on the structure of FTO/CdS/Sb 2 Se 3 /Au.First, through single source evaporation (PTN104736, Mini SPECTROS, Kurt J. Lesker, USA), a 50 nm CdS film was evaporated onto the FTO substrate (with a sheet resistance of 6 sq −1 and transmittance 80%, Advanced Election technology Co., LTD China) using CdS powder (99.999%,China New Metal Materials Technology Co., Ltd, China).During the evaporation process, the substrate temperature was room temperature, the working pressure was 1.0×10 −6 torr, the evaporation speed was 0.65 s −1 , and the evaporation power was 35 W. The CdS film was then annealed in a tube furnace for one minute at 500 °C.Second, based on the RTE technique, Sb 2 Se 3 films with a thickness of ≈500 nm were created in a tube furnace that was maintained at a pressure of 5 m Torr by a mechanical pump.The pre-heating temperature was 310 °C for 20 min, and the depositing temperature was raised to 570 °C for 40 s.Third, the LiOH solution's concentration ranged from 0 to 0.1 m, with water serving as the solvent.Additionally, the Sb 2 Se 3 films were heated at 100 °C while spin-coating them with various LiOH concentrations for comparison.At last, Gold electrodes (0.0225 cm 2 ) with a thickness of 80 nm were prepared by evaporation.
Characterization of Films and Solar Cells: In this paper, the structure and crystallinities of all the Sb 2 Se 3 film at various LiOH concentrations were examined by the XRD based on the smart Apex II Duo system.The SEM was used to analyze the surface and cross-sections morphologies of all the samples.The AFM was used to measure the surface roughness and a UV-visible spectrophotometer (Shimadzu UV-2600) was used to measure optical performance of all the Sb 2 Se 3 films.UPS was carried out using a monochromatic He I light source (21.2 eV) and a VG Scienta R4000 analyser, while XPS was carried out using a Kratos Axis Ultra DLD system.For testing the electrical properties, Sb 2 Se 3 films with different concentrations of LiOH were deposited directly on the glass substrate and measured using a Hall measurement system with van der Pauw configuration at room temperature and a magnetic flux density between 3000-5000G (QT-50, Quatek, Germany).The metal In was used as the electrode for the hall test.Deep analysis of the elements was performed with a TOF-SMIS measurement system (IONTOF GmbH, Münster, Germany).The EQE of the solar cells was detected using a photovoltaic characteriszation system (PV

2
Se 3 film, the upper interface between Sb 2 Se 3 film and Au electrode has been treated with different concentrations of LiOH solution.The detailed results are presented in Figure S3 (Supporting Information).As shown in the figure, the Sb 2 Se 3 film was severely damaged as the concentration of LiOH solution increased, and at high concentration levels, the Sb 2 Se 3 film was almost entirely corroded away.The XRD results of these six Sb 2 Se 3 samples are displayed in Figure S4 (Supporting Information).The XRD results indicate that low concentrations of an alkaline solution containing Li-ion do not affect the structure of Sb 2 Se 3 film, whereas high concentrations of alkaline solution can severely damage the structure of Sb 2 Se 3 film.

Figure 1 .
Figure 1.Illustration of the fabrication process for Sb 2 Se 3 solar cells with Li gradient field by upper interface doping via solution method.

Figure 2 .
Figure 2. a) The schematic diagram of a Li-doped Sb 2 Se 3 film, b) XRD patterns of Sb 2 Se 3 film with and without a Li gradient concentration, and c-e) XPS patterns of Sb 2 Se 3 solar cells with and without a Li gradient concentration.

Figure
Figure 5b,c.Fitting the cut-off binding energy and long tails of the UPS spectra allowed for the determination of the Fermi level (E F ) and valence band maximum (VBM, E v ) of the Sb 2 Se 3 film.The conduction band minimum (CBM, E c) was determined from the energy gap and VBM position.The energy levels of all samples were calculated from the UPS spectra and bandgap.The UPS results revealed that the distance between the E F and VBM was calculated to be 0.62 and 0.47 eV, respectively.With the addition

Figure 3 .
Figure 3. a) Surface SEM images, c) cross-sectional SEM images, and e) AFM images of Sb 2 Se 3 film without Li gradient concentration, b) the Surface SEM images, d) cross-sectional SEM images, and f) AFM images of Sb 2 Se 3 film with Li gradient concentration.

Figure 4 .
Figure 4. a-c) XPS patterns of the Sb 2 Se 3 film with and without etching, d) TOF−SMIS patterns of the Sb 2 Se 3 film with Li gradient field, e) 3D-rendered overlays of selected elements of Sb 2 Se 3 film with Li gradient field.

Figure 5 .
Figure 5. a) Electrical properties of Sb 2 Se 3 films with and without Li gradient field, b,c) UPS cut-off edge and valence band spectra of Sb 2 Se 3 films with and without Li gradient field, d) Band structures of Sb 2 Se 3 solar cells with and without Li gradient field.

Figure 6 .
Figure 6.In-depth UPS analysis of Sb 2 Se 3 film with Li gradient: a) UPS spectra of the work function edge, b) UPS spectra of the valence-band edge, and c) Gradient energy levels at the gradient of the Sb 2 Se 3 film with Li gradient.

Figure 7 .
Figure 7. a) J−V curves, b) EQE curves, c) dark J−V curves, d) temperature-dependent open-circuit voltage measurements, e) C−V curves, and f) N C-V curves of Sb 2 Se 3 solar cells with and without a Li gradient field.

Figure 8 .
Figure 8. a) DLTS signals of Li-GS and AD Sb 2 Se 3 devices.b) Arrhenius plots obtained from DLTS signals.c) Energy states and defect levels of the AD device.d) Energy states and defect levels of the Li-GS solar cells.
2 eV, which is very close to the band gap of Sb 2 Se 3 films (1.26 eV, as shown in the Figures S9 and S10, Supporting Information), while the Sb 2 Se 3 solar cells without a Li gradient field is 0.97 eV.This result indicates that the interfacial recombination of Sb 2 Se 3 solar cells with a Li gradient field is much lower than that of Sb 2 Se 3 solar cells without a Li gradient field.The temperature-dependent open circuit voltage results indicate that the upper interface treatment of LiOH can diffuse into the entire Sb 2 Se 3 film and form a Li gradient field.The capacitance-voltage (C-V) and depletion region width (W d ) results are shown in Figure

Table 1 .
The detailed DLTS parameters of Sb 2 Se 3 solar cell with and without 0.01 m LiOH solution.