Two-step spin-coating of vacancy-ordered double perovskites enables growth of thin films for electronic devices

Abstract

Vacancy-ordered double perovskites (VODPs), such as Cs₂TeX₆ (X = Cl, Br, I), are lead-free alternatives to conventional metal-halide perovskites (MHPs). One limitation of VODPs is the lack of processes to form thin films relevant for physical characterization and electronic devices. A two-step spin-coating method was developed for synthesizing high-quality films of Cs₂TeBr₆. Independently depositing CsBr and TeBr₄ enables high precursor concentrations and control over crystallization kinetics. By optimizing the spin-coating parameters, conversion of precursors to phase pure films was observed using structural and surface characterization methods. The growth of mixed-halide systems was investigated using alternative salts including CsCl and CsI. The formation of halide alloys was found to depend on the existence of routes to byproducts. Lastly, single carrier diodes of Cs₂TeBr₆ were designed following valence band characterization with photoelectron spectroscopy. Temperature-dependent space-charge-limited current measurements revealed that transport occurs by hopping and the hole mobility is 3.2 × 10⁻⁵ cm² V⁻¹ s⁻¹ near room temperature. The insights from the 2-step procedure provide a pathway towards making semiconducting devices from VODPs.

---

Introduction

The pressing demand for low-temperature, solution-processable semiconductors has driven a revolution in research on metal-halide perovskites (MHPs).^1,2 Three-dimensional (3D) MHPs have remarkable carrier lifetimes and diffusion lengths, resulting in high-performing semiconducting devices like solar cells, light-emitting diodes (LEDs), and sensors.^3–6 These materials have the formula ABX₃, where A is a monovalent organic cation, B is a divalent metal cation, and X is a halogen anion. Despite having impressive performance, many 3D MHPs exhibit poor stability.⁷ Because of this, many studies have been conducted on lower-dimensional MHP systems (2D, 1D), where reduced corner-sharing of the BX₆ octahedra has been shown to reduce ion mobility and sensitivity to water thereby improving their stability relative to 3D MHPs.^8–12 The B-site metal cation, however, remains prone to oxidation since it is divalent. Vacancy-ordered double perovskites (VODPs) of the form A₂BX₆ remedy this issue because the B-site cation is tetravalent and less prone to oxidation, or in some cases, in the maximum oxidation state, granting the material greater resistance against degradation in the ambient.^13,14 VODPs are considered zero-dimensional because the octahedra experience no corner-sharing and are relatively isolated from one another. Despite this structure, the relatively soft heavier halide anions can still allow for disperse electronic bands. The unique geometry of these compounds, while advantageous for stability, welcomes a new suite of optical, electronic, and structural properties due to the molecule-like isolated octahedra.

A notable consequence of a vacancy-ordered lattice is enhanced carrier- and exciton–phonon coupling. The lack of connectivity between octahedra enables greater susceptibility to deformation upon formation of charge carriers or excitons, thereby leading to polaron formation after excitation.¹⁵ Prior studies have shown that this strong interaction induces localized carriers and excitons that consequently affect emission pathways and carrier transport.^16,17 Experimentally observed broad optical emission has been attributed to the potential wells created by localized charges predicted by computation.^15,17–20 Several contradicting results on the electronic properties of various VODPs have also been published recently, complicating characterization and device design. For example, reported values of the bandgap for Cs₂TeBr₆ based on density functional theory calculations have differed by over 1.5 eV.^21,22 In the case of carrier mobility measurements, we are not aware of space-charge-limited current (SCLC) measurements for VODPs. SCLC measurements of diodes can reveal transport behavior in thin films relevant for devices, providing an understanding of charge carrier dynamics.^23,24 However, effective diode fabrication requires high-quality thin films and careful choice of transport materials and electrodes. The limited co-solubility of all-inorganic precursors for VODP formulations has hindered film synthesis. These limitations and discrepancies motivated us to study new growth methods for VODPs.

The most common method for casting films of MHPs is one-step spin-coating, where precursors for a given perovskite are co-dissolved and spin-coated onto a substrate followed by a thermal annealing treatment. This simple method typically produces uniform films, but it offers little control over crystallization kinetics. Additionally, the precursors for desirable MHP systems are often not sufficiently co-soluble to make one-step deposition methods viable.^25,26 Researchers have recently been investigating two-step deposition methods as a viable synthetic alternative for MHPs.^27,28 Here, the B-site metal precursor is independently spin-coated to create a film which is then either immersed in a solution of the A-site metal precursor or spin-coated again with this solution. Two-step methods have been shown to reduce defects in both hybrid and all-inorganic MHP films, contributing to competitive device performance among other synthetic routes.²⁹

In this work, we explore new routes to grow the Cs₂TeX₆ (X = Cl, Br, I) class of VODPs as thin films. These materials have a tetravalent B-site cation that reduces degradation of the perovskite due to oxidation. Herein, we report a two-step spin-coating method for fabricating high-quality films of Cs₂TeBr₆. The rapid crystallization of this material from precursors CsBr and TeBr₄ enables dynamic two-step spin-coating with strong repeatability and precise control over film properties. We show that the quality of the resulting thin film is highly dependent on the conditions of both spin-coating steps and careful use of structural characterization techniques is essential to verify full conversion of TeBr₄ in the reaction to Cs₂TeBr₆. This method is also adapted to investigate the formation of mixed halide phases with CsCl and CsI precursors. Lastly, we uncover valence band energetics and fabricate single-carrier devices for measuring the carrier mobility using SCLC measurements.

Methods

Chemicals

Cesium bromide (99% metals basis) was purchased from Alfa Aesar. Tellurium(IV) bromide (99% metals basis), acetone (certified ACS grade), and isopropyl alcohol (99.5%) were purchased from Thermo Fisher Scientific. Methanol (≥ 99.9%) was purchased from Sigma-Aldrich. Ethylene glycol (≥ 99%) was purchased from VWR Chemicals BDH.

Precursor solutions

CsBr solution was made by dissolving purified CsBr in ethylene glycol at a target concentration of 1 M. TeBr₄ solution was made by dissolving TeBr₄ in methanol at a target concentration of 1 M. Solutions were heated at 50 °C and stirred for 2 hours to ensure complete dissolution.

Two-step sequential spin-coating

Z-cut quartz and indium tin oxide (ITO)/quartz substrates were purchased from University Wafer. Substrates were cleaned with sonication in water, acetone, and isopropyl alcohol for 10 minutes each, followed by an air plasma treatment for 10 minutes. In an N₂ glovebox, 50 μL of TeBr₄ solution was spin-coated at 500 rpm for 10 seconds and then ramped up to a speed between 2000 and 8000 rpm for 30 seconds, depending on desired thickness. The resulting film was annealed at 100 °C for 5 minutes. After the first annealing step, the film was re-spun at 2000 rpm for 30 seconds. Once ramping to 2000 rpm was complete, 50 μL of the CsBr solution was deposited on top of the TeBr₄ film to form Cs₂TeBr₆. The final film was annealed again at 100 °C for 5 minutes.

Powder X-ray diffraction

Powder X-ray diffraction patterns were collected with a panalytical Empyrean powder diffractometer in reflection mode. Cu Kα₁ was used as the X-ray radiation source with an accelerating voltage of 45 kV and beam current of 40 mA. Scans were performed from 2θ = 3° to 2θ = 60° with a step size of 0.04° and step time of 40 ms.

UV-Vis

A Shimadzu UV3600 UV-Vis-NIR spectrometer was used to collect diffuse reflectance spectra. Scans were performed from 300 to 900 nm with a step size of 0.5 nm. Reflectance data was converted to absorbance with the Kubelka–Munk transform. Tauc plots for indirect band gap materials were generated to determine the band gap with a linear fit of the absorption edge.

Photoluminescence

Emission spectra was collected in reflectance mode with a 430 nm long pass filter, spectrometer, and visible CCD detector. Samples were excited at 405 nm with a continuous wave laser diode.

Grazing incidence wide-angle X-ray scattering (GIWAXS)

GIWAXS experiments were performed at Stanford Synchrotron Radiation Lightsource (SSRL) on beam line 11–3, which has a fixed energy of 12.7 keV and is equipped with a two-dimensional Rayonix MX225 CCD area detector. Lanthanum hexaboride (LaB₆) was used to refine the beam center and sample-to-detector distance. Data was collected with an incidence angle of 3° to access a large range in q. Geometric corrections to the raw images were made using Nika. Data was azimuthally integrated with WAXStools.

Scanning electron microscopy (SEM)

SEM images were collected with a Thermo Scientific Apreo C LoVac SEM. Films were mounted onto SEM stubs with double-sided copper tape and imaged with accelerating voltages of 5.00 kV and beam currents of 0.40 nA.

Ultraviolet and X-ray photoelectron spectroscopy

Ultraviolet photoelectron spectroscopy (UPS) and X-ray photoelectron spectroscopy (XPS) experiments were performed with a Thermo Scientific ESCALAB Xi⁺ XPS Microprobe. For UPS experiments, before film casting, 20 nm of chromium followed by 90 nm of gold were deposited by thermal evaporation onto quartz substrates so that films did not experience charging. Cs₂TeBr₆ films were cast with an initial TeBr₄ layer spun at 8000 rpm so that the film was sufficiently thin to prevent charging issues. Nickel tape was adhered to the surface of the film and wrapped around the edge of the substrate to further assist with charge dissipation. A helium I radiation source was used, performing 5 scans with a pass energy of 1.5 eV, dwell time of 150 ms, and energy step size of 0.05 eV. Additional low energy scans were conducted with electron charge compensation to ensure that sample charging didn't alter the onset, aligning with data collected without compensation. A corrected helium I satellite line background was subtracted using CasaXPS. For XPS experiments, X-rays were generated with a monochromated aluminum anode (1486.7 eV). Scans were performed with a pass energy of 100 eV, dwell time of 20 ms, and energy step size of 0.5 eV. Charge compensation was applied, and charge shift was accounted for by calibrating to trace oxygen photoemission at 531 eV. The Thermo Scientific Avantage data system was used to fit peaks with a smart background and to quantify atomic percentages.

Device fabrication and testing

The device architecture of SCLC diodes is pictured in Fig. 5a. Cs₂TeBr₆ was deposited onto indium tin oxide (ITO) coated glass substrates through two-step spin-coating. TeBr₄ was spun at 4000 rpm, followed by 1.5 M CsBr at 2000 rpm. 20 nm of molybdenum(VI) oxide (MoO₃) followed by 100 nm of gold were thermally evaporated in high vacuum and deposited onto the Cs₂TeBr₆ film through a mask to form mm² contacts. Devices were tested with a LakeShore Cryogenic Vacuum Probe Station and Keithley 2400 SourceMeter. Current–voltage scans were performed with a step size of 0.05 V.

Results

Synthesis and film casting

The reported growth procedures for thin films of VODPs have been limited in comparison to higher-dimensional MHPs. In the case of Cs₂TeX₆ (Fig. 1a), its bulk powder dissolves with a polar aprotic solvent into Cs⁺ cations and [TeX₆]²⁻ anions, producing solution that is highly stable. Drop-casting and pad printing have been demonstrated to form films of Cs₂TeX₆.³⁰ Drop cast films frequently suffer from defects, such as pinholes or variations in thickness, that form as the solvent evaporates upon casting.³¹ In contrast, spin-coating can alleviate inhomogeneity and provide greater control over grain size, thickness, and other film properties.³² Producing high quality films via spin-coating depends on numerous factors, including the spin speed, concentration and the viscosity of precursor solution, and annealing procedure. The relatively low solubility of Cs₂TeX₆ (0.125 M) in polar aprotic solvents, like dimethylformamide (DMF) and dimethyl sulfoxide (DMSO), makes it a challenge to spin-coat uniform and continuous thin films. Our attempts at spin-coating films of Cs₂TeX₆ with DMSO and DMF solutions resulted in virtually no material being retained on the substrate due to the low concentration of the solution. Typically, minimum concentrations of ≈30 wt% are desirable for MHP films.³³ This difficulty suggested the need to examine other routes to form thin films of VODPs.

Fig. 1 (a) Crystal structure of Cs₂TeBr₆, depicting characteristic isolated octahedra of VODPs. (b) Schematic of two-step spin-coating procedure for Cs₂TeBr₆ films and corresponding photographs of the resulting films. The films shown are on 15 mm × 15 mm quartz substrates.

The simple, rapid reaction between a cesium halide salt (CsX) and a tellurium tetrahalide (TeX₄) (eqn (1)) encouraged us to investigate two-step spin-coating as a potential route for producing high-quality continuous films.

2CsX + TeX₄ → Cs₂TeX₆ (1)

Casting a perovskite film with two deposition steps adds complexity to the overall procedure relative to using a single precursor. The spin-coating parameters must be carefully optimized such that the two reactants completely convert into the desired perovskite phase while still producing a film that is free of morphological defects. Most importantly, the first layer must not be soluble in the solvent of the second precursor solution or else the entire film will wash off the substrate. Solvent choice and sequence of casting therefore play a critical role in developing a successful procedure. Two-step studies with lead iodide perovskites cast the PbI₂ layer first because it does not degrade upon annealing and readily crystallizes to form a film.²⁷ We took a similar approach and deposited the metal halide TeX₄ in the first layer (Fig. 1b). In the case of the bromine derivative, methanol is an effective solvent choice for TeBr₄ because of its solubility at room temperature. Ethylene glycol was found to rapidly dissolve CsBr at concentrations as high as 2 M, but not TeBr₄, determining the order of the two-step process. We rationalize this slow dissolution with the viscosity of ethylene glycol. The conversion of TeBr₄ to Cs₂TeBr₆ is proposed to proceed by the dissociation of (TeBr₄)₄ clusters into [L₂TeBr₃]⁺Br⁻ solvent-coordinated complexes followed by reaction with the metal halide salt (see ESI† for the reaction steps).^34–36 Thus, we deduce that ethylene glycol can successfully coordinate with TeBr₄ to form these intermediate complexes when CsBr solution is deposited, but the higher viscosity relative to other solvents slows down the dissolution considerably. As a result, the TeBr₄ film is not completely dissolved and washed away during deposition.

We examined different deposition processes to investigate the two-step growth process. In solution-based two-step procedures, growth of the final material is initiated by either immersing the first layer in a precursor solution or by depositing a precursor solution onto the first layer before, or during, spinning. These are referred to as immersion, static spin-coating, and dynamic spin-coating, respectively.²⁷ In our case, adding CsBr via static spin-coating or immersion resulted in patchy, irregular films with residual CsBr observed by powder X-ray diffraction (PXRD). These methods expose the TeBr₄ film to a significant excess of CsBr which is then incorporated into the sample. In contrast, dynamically depositing the second layer limits the amount of CsBr available to react because the volume of liquid on top of the substrate is determined by the spin speed. Depositing CsBr solution while spinning therefore prevents residual CsBr in the final film. This stoichiometric control is enabled by the rapid kinetics of the reaction between CsBr and TeBr₄. Films are annealed after each deposition step to promote crystallization and remove solvent, but the reaction between CsBr and TeBr₄ commences immediately upon deposition, evident by a stark color change.

The fabrication method significantly influences the morphology, phase purity, and thickness of Cs₂TeBr₆ films. Scanning electron microscopy (SEM) images reveal morphological differences between films made via two-step spin-coating and drop-casting (Fig. 2a). The initial TeBr₄ film, spun at 2000 rpm, has large cracks that separate highly porous domains. The porosity likely prevents the formation of a capping layer when CsBr is deposited because the solution can completely permeate the TeBr₄ layer and react. The resulting films have mostly spherical grains that are smaller than the octahedral crystallites found in drop-cast films. PXRD experiments indicate that two-step spin-coating produces Cs₂TeBr₆ films with higher phase purity than drop-cast films (Fig. 2b). Diffraction patterns were compared to ICSD #24151 for Cs₂TeBr₆ in the Inorganic Crystal Structure Database (ICSD) to confirm the identity and purity of samples. PRXD spectra for drop-cast samples show excess CsBr present in the films given the (110) reflection at q = 2.07 Å⁻¹, which is the strongest reflection in CsBr reference spectra (ICSD #236387). Conversely, the PXRD spectra for spin-coated films indicate phase-pure Cs₂TeBr₆ with no missing or additional reflections when compared to the ICSD reference. Changing the spin speed of the TeBr₄ layer dramatically alters the final thickness of the Cs₂TeBr₆ film. By varying the TeBr₄ spin speed between 2000 and 8000 rpm and depositing CsBr solution at 2000 rpm, the final film thickness can be tuned between roughly 2 and 5 microns (Fig. 2c and Fig. S1, ESI†). This is in stark contrast to drop-cast films, which are limited to substantial thicknesses of about 30 microns. Thickness control is one of the key advantages of adopting spin-coating methods for MHP films.

Fig. 2 (a) SEM images of TeBr₄ and Cs₂TeBr₆ films made via two-step spin-coating procedure (top) and drop-casting (bottom). The drop-cast film was made by dropping 200 μL of 0.125 M Cs₂TeBr₆ solution onto a quartz substrate at 170 °C. 50 μL of toluene was deposited in tandem as an anti-solvent. The two-step spin-coated film was made by spinning 50 μL of 1 M TeBr₄ at 2000 rpm followed by 50 μL of 1 M CsBr deposited dynamically at 2000 rpm. (b) PXRD patterns for the spin-coated Cs₂TeBr₆ film (top), the drop-cast Cs₂TeBr₆ film (middle), and the ICSD #24151 Cs₂TeBr₆ reference pattern (bottom). (c) Profilometer thickness data of films made via drop-casting and spin-coating. The spin-coating speed refers to the procedure for the initial TeBr₄ layer, which primarily influences final film thickness.

The two-step spin-coating procedure introduces many process variables that impact film quality and phase purity. Conditions like the spin speed of both layers and concentration of precursor solution interact in complex ways, making optimization challenging. In addition, bulk characterization techniques may not fully capture the nuances of the growth of Cs₂TeBr₆ within these films. To follow the transformation, in situ photoluminescence experiments were conducted. The photoluminescence quantum yield (PQLY) of Cs₂TeBr₆ is sufficiently low that no emission is observed when exciting at 405 nm at a reasonable acquisition time, so the two-step procedure was amended to grow Cs₂TeCl₆ on glass substrates (see Mixed halide compositions section for more details). Cs₂TeCl₆ has a higher PLQY due to the greater ionicity of Te–Cl bonds causing flatter band dispersion.³⁵ In addition, the 3D Pb-containing CsPbBr₃ films were also synthesized in two steps by dissolving PbBr₂ in DMF to compare VODP growth to that of 3D structures. In situ photoluminescence data collected during Cs₂TeCl₆ and CsPbBr₃ film casting at room temperature highlights the rapid reaction of precursors in two-step spin-coating. Characteristic Cs₂TeCl₆ emission centered around 2.05 eV was observed immediately upon deposition of CsCl onto TeCl₄ (Fig. S2, ESI†).³⁷ CsPbBr₃ emission was initially centered around 2.40 eV but gradually red shifted to 2.37 eV during spin-coating. This behavior has been attributed to quantum confinement and is possibly not observed in Cs₂TeCl₆ because of the molecular nature of VODPs.³⁸ The described observations hint at different growth mechanism but the degree of TeX₄ conversion as well as crystal phase evolution are not revealed with optical characterization. We therefore turned to grazing-incidence wide-angle X-ray scattering (GIWAXS) to understand how different processing parameters influence the composition of the resulting films.

The detailed influence of the concentration of CsBr solution required to form Cs₂TeBr₆ was evaluated using GIWAXS. While laboratory PXRD provides a fingerprint of crystalline phases in the bulk of a film, GIWAXS provides access to a wider region of reciprocal space and can selectively probe different sample depths by varying the grazing-incidence angle.³⁹ The higher brilliance of synchrotron X-ray sources also enables detection of weakly scattering phases that are not easily visible with PXRD. Films made with different concentrations of CsBr were characterized with GIWAXS to determine ideal synthetic conditions for phase-pure samples (Fig. 3a). None of the Cs₂TeBr₆ films analyzed with GIWAXS exhibit significant texturing through the depth (Fig. S3, ESI†), as the 2D diffraction patterns exclusively display rings without any discernible peaks indicating orientation. Phase information was therefore compared to powder reference patterns by azimuthally integrating the diffraction patterns to generate 1D scattering patterns. Unlike PXRD patterns, residual TeBr₄ is detectable with GIWAXS given the presence of reflections that match a collected GIWAXS pattern for a TeBr₄ film. The lack of sensitivity to TeBr₄ in PXRD patterns is due to residual TeBr₄ primarily populating the surface of the film, which is more easily revealed with GIWAXS. This distribution is made clear by the increasing intensity of TeBr₄ reflections relative to Cs₂TeBr₆ reflections as the grazing-incidence angle decreases (Fig. S4, ESI†) and is further supported by X-ray photoelectron spectroscopy (XPS) that shows excess tellurium on the surface of films (Fig. S5, S6 and Tables S2, S3, ESI†). In situ photoluminescence data demonstrated that when TeX₄ reacts with CsX, Cs₂TeX₆ rapidly precipitates, leaving unreacted TeBr₄ to be coordinated with ethylene glycol in the liquid layer. Solvent is then evaporated during annealing and residual TeBr₄ is left on the surface of the Cs₂TeBr₆ film. Increasing the concentration of CsBr solution deposited onto TeBr₄ results in diminished presence of TeBr₄ reflections in GIWAXS and increased intensity of Cs₂TeBr₆ reflections, highlighting greater conversion of TeBr₄ to the VODP. The most prominent reflection attributed to TeBr₄ in GIWAXS is at q = 2.22 Å⁻¹ (highlighted with pink shading in Fig. 3a) and its diminishing presence in Cs₂TeBr₆ patterns best illustrates this concentration dependence. In addition, no reflections expected from CsBr are observed (Fig. S7, ESI†).

Fig. 3 (a) Azimuthally integrated GIWAXS data collected at a 3° incidence angle of Cs₂TeBr₆ films made by depositing 0.5 M, 1 M, and 1.5 M CsBr solution (top) in two-step spin-coating procedure. The TeBr₄ pattern (middle) and ICSD 24151 reference pattern (bottom) are included for peak assignment. The pink shaded region marks the most prominent TeBr₄ reflection. (b) SEM images of Cs₂TeBr₆ films made with varying CsBr deposition spin-speeds.

The spin speed during deposition of CsBr was optimized for growth of phase pure films and to minimize any residual TeBr₄ on the surface. After successfully producing films at 2000 rpm, higher spin speeds were tested to vary the film thickness. To our surprise, films made by spin-coating both layers at higher spin speeds had an overwhelming residual TeBr₄ presence in integrated GIWAXS patterns despite a thinner TeBr₄ layer (Fig. S8, ESI†). This suggests that the spin speed of the second layer is a crucial factor in the conversion to Cs₂TeBr₆, which was further studied with SEM experiments (Fig. 3b). Films with the TeBr₄ layer cast at 8000 rpm and subsequent CsBr depositions at 2000 rpm, 4000 rpm, 6000 rpm, and 8000 rpm were made and imaged with SEM. Higher spin-speed films show unreacted TeBr₄ crystallites that dominate the substrate surface. As the second-layer speed decreases, the surface is further populated with Cs₂TeBr₆ crystallites in tandem with the disappearance of the large TeBr₄ domains. The same behavior is observed in films made at lower TeBr₄ spin speeds (Fig. S9, ESI†). The CsBr liquid layer on top of TeBr₄ is made thinner at higher spin speeds, limiting the amount of CsBr available to react and form perovskite. This result indicates that the thickness of the final Cs₂TeBr₆ must be controlled dominantly by varying the speed of the initial TeBr₄ layer. Thickness variation was achieved by spin-coating TeBr₄ at 2000 rpm, 4000 rpm, and 8000 rpm followed by CsBr at 2000 rpm for all samples.

Using this data, we can estimate the thickness of the CsBr liquid layer which, along with concentration, largely determines the extent of conversion to the perovskite. Given that GIWAXS shows very little remaining TeBr₄ when depositing 1.5 M CsBr solution onto a TeBr₄ film spun at 2000 rpm, we can assume 100% yield at this concentration to estimate the thickness of the second liquid layer when spinning at 2000 rpm. We estimate a thickness of ≈40 μm with these process parameters, informing the conditions required to induce full conversion in the film (see ESI†).

Mixed-halide compositions

After successfully producing Cs₂TeBr₆ films via the two-step method, the procedure was adapted to determine if Cs₂TeCl₆ and Cs₂TeI₆ could also be grown. To form these materials, CsCl/CsI and TeCl₄/TeI₄ were used in place of CsBr and TeBr₄, respectively. TeCl₄ was similarly dissolved in methanol at 1 M, while TeI₄ was dissolved in dimethyl formamide (DMF) due to poor solubility in methanol. CsCl and CsI were both dissolved in ethylene glycol at 1 M. While the two-step method still formed the target phase on quartz substrates (Fig. S10, ESI†), the coverage was poor and most of the substrate was left bare. Unlike TeBr₄, TeCl₄ and TeI₄ are susceptible to sublimation and degradation, respectively, during the first annealing step. TeCl₄ has been shown to have an appreciable vapor pressure at 100 °C and, given the thin films here, sublimation is likely. TeI₄ decomposes to Te(s) and I₂(g) at even lower temperatures limiting the annealing temperature.^40,41 In contrast, TeBr₄ is more stable and decomposes to TeBr₂(g) and Br₂(g) at temperatures above 400 °C.⁴² Annealing TeCl₄ and TeI₄ at lower temperatures gave better results but the samples were still not continuous films. If annealing the first layer is skipped entirely, the desired product forms on the substrate but most washes off after depositing the cesium salt. While the inability to form pure phases with chloride and iodide limit the method, the stability of TeBr₄ suggested a route to form mixed halide compositions.

Multi-halide alloys provide means to tune the electronic properties of a given material system and we targeted compositions that would show clear shifts in optical properties.^22,43,44 Our two-step spin-coating procedure was adapted by depositing CsCl and CsI at concentrations of 0.5 M, 1 M, and 1.5 M in the second step to react with TeBr₄ films (Fig. 4a and b). Tauc plots of UV-vis reflectance data for these films reveal that the optical absorbance can be modified by the choice of halide salt.^45,46 Films made with CsI show a shift in the absorbance onset and optical band gap to lower energy as the concentration of deposited CsI solution increases. However, films made with CsCl show no difference in UV-vis spectra and are all relatively identical to the spectra of Cs₂TeBr₆ (Fig. 4c and d).

Fig. 4 (a) and (b) Pictures of films made by depositing (a) CsCl and (b) CsI on top of initial TeBr₄ layer. (c) and (d) Tauc plots generated from UV-Vis reflectance data for films made by depositing (c) CsCl and (d) CsI solutions of varying concentrations with calculated bandgaps. (e) Azimuthally integrated GIWAXS data of films made with increasing CsCl concentration and ICSD 24151, ICSD 29031, and TeBr₄ patterns for reference. (f) Azimuthally integrated GIWAXS data of films made with increasing CsI concentration and the ICSD 24151, ICSD 38105, and TeBr₄ patterns for reference. (g) Photoluminescence of films made with increasing CsI concentration, excited at 405 nm.

These films were analyzed with GIWAXS to determine the structure and composition after reacting with CsCl and CsI. As was done with Cs₂TeBr₆, GIWAXS data was azimuthally integrated to compare with reference powder patterns (Fig. 4e, f and Fig. S5, ESI†). Diffraction patterns of films made with CsCl reveal that Cs₂TeBr₆ is the only compound formed, with no deviation in reflections from the Cs₂TeBr₆ reference pattern. This is consistent with the optical data which appears identical to that of Cs₂TeBr₆. Additionally, XPS scans show that no Cl ions are apparent on the surface of films (Fig. S11, ESI†). The puzzling reaction of Cs⁺ ions with TeBr₄ and the absence of any Cl suggests a new reaction pathway. We hypothesize that TeCl₄ is formed as a byproduct based on DFT formation energy calculations of the overall reaction (see ESI†). While true Cs₂TeCl₃Br₃ mixed-halide single crystals have been synthesized in solution via antisolvent vapor, instead the rapid reaction while spin-coating prefers Cs₂TeBr₆ and TeCl₄.²² Annealing the films then sublimes the TeCl₄ leaving no chloride compounds to be detected in the film by GIWAXS. Like films made with CsBr, an increase in the concentration of CsCl results in increased TeBr₄ conversion. In contrast, films made with CsI show new features in diffraction patterns, indicating the presence of additional phases. While Cs₂TeBr₆ reflections are still dominant, weak scattering intensity is observed that mirrors the known (222) and (400) Cs₂TeBr₆ reflections at q = 1.99 Å⁻¹ and q = 2.30 Å⁻¹, respectively, except at slightly lower q. As the CsI concentration increases, these features shift to lower q and approach the position of the (222) and (400) reflections in the Cs₂TeI₆ reference powder pattern. Cs₂TeI₆ reflections are more evident in samples where a lower concentration TeBr₄ solution was used to cast the initial layer (Fig. S12, ESI†). This observation supports the existence of mixed halide octahedra in the perovskite, the quantity of which increases with increasing CsI concentration. As more iodine is incorporated, the lattice parameter of the average unit cell increases, and these reflections shift to lower q. However, this change is not conclusive because the peak is in a crowded section of the diffraction pattern and therefore is hard to identify and definitively attribute to mixed halide octahedra. Additionally, this observation can be interpreted in two different ways: either the increased CsI concentration induces the formation of more octahedra with the same I : Br ratio, or an increased I : Br ratio in octahedra. Photoluminescence (PL) experiments were performed to confirm the identity of mixed halide phases. PL data shows a broad emissive peak that is characteristic of VODPs and is largely attributed to the presence of self-trapped excitons.^47,48 This peak shifts to lower energy with increased CsI concentration, supporting the presence of octahedra with an increased I : Br ratio (Fig. 4g). The lower electronegativity of iodine than bromine reduces the optical transition energy of TeX₆²⁻ octahedra. The conduction band of tellurium-based VODPs is derived from the anti-bonding hybridization of the Te 5s and halogen np orbitals, so a less electronegative X-site halogen leads to a higher energy p orbital and consequently lower energy conduction band minimum.^21,49 The bandgap therefore shrinks, causing a redshift in the PL emission. Br and I p orbitals both contribute to the conduction band in mixed octahedra, and the bandgap is effectively tuned between both extremes. These results demonstrate the potential for two-step spin-coating methods to be adapted for multi-halide alloy film growth.

Single-carrier device characteristics

Our 2-step route allows us to make thin Cs₂TeBr₆ films making it easier to examine the valence band structure with photoemission. Ultraviolet photoelectron spectroscopy (UPS) was conducted to extract band energy values (Fig. S13, ESI†). We defined the onset of the VBM from a logarithmic plot because of the shape of the observed onset; we note that taking the onset from the plot on linear scale results in a value shifted deeper that depends on the range used for the extrapolation to the edge. The Fermi level (E_F) was found to be −4.99 eV, and the valence band maximum (VBM) was found to be −5.29 eV from the logarithmic plot with reference to the vacuum level. Prior estimates on the VBM position of Cs₂TeBr₆ at −5.97 eV have been made with XPS.⁵⁰ X-rays excite deeper core-level electrons, making XPS less sensitive compared to UPS, which primarily excites electrons from the valence band. We find the Fermi level to be closer to the VBM, suggesting Cs₂TeBr₆ films are p-type as cast.

Enabled by thin film synthesis and VBM measurements, hole-only devices were fabricated to measure space-charge-limited hole currents in Cs₂TeBr₆ films. In contrast to Hall effect measurements, SCLC measurements reveal the behavior under conditions where injection, defect states and grain boundaries can significantly affect transport. Indium tin oxide (ITO) was chosen as the hole injecting bottom contact and gold was chosen as the top contact (Fig. 5a). Both of these have work functions that align well with the VBM of Cs₂TeBr₆, as determined with UPS.^51,52 XPS experiments confirmed that Cs₂TeBr₆ films grown on ITO-coated substrates were phase-pure (Fig. S14 and Table S5, ESI†). The addition of a layer of molybdenum oxide (MoO₃) prevented diffusion of gold through the active layer and it is known to act as a hole transport layer.^53–55J–V curves showed low magnitude of the current density consistent with the isolated nature of octahedra in VODPs (Fig. 5b). Minimal hysteresis was observed, and injection did not strongly depend on bias direction. This agrees with the energetically well-aligned contacts and hence supports our estimate for the VBM. After correcting the data by subtracting shunt leakage current, we fit the J–V data with a space charge limited current model with a hole mobility of 3.2 × 10⁻⁵ cm² V⁻¹ s⁻¹ (Fig. S15, ESI†). The low magnitude of the observed mobility when compared to all-inorganic 3D MHPs suggests hopping transport caused by the localization of charge carriers in VODPs.¹⁶ The relatively flat bands in calculated band structures in literature agree with this assessment.^21,22 We expected an Arrhenius temperature dependence of the mobility if hopping was the primary charge transport mechanism. J–V curves were collected at several temperatures between 280 K and 220 K to further observe this relation (Fig. S16, ESI†). An Arrhenius plot of hole mobilities (Fig. 5c) reveals, an activation energy of 0.24 eV (see ESI†) which is reasonable for polaron hopping. The pre-factor, μ₀ = 0.35 cm² V⁻¹ s⁻¹, can be interpreted as the high temperature limit for hole mobility in Cs₂TeBr₆. While we cannot fully rule out the influence of grain boundaries on the extracted mobility, the high temperature limit is still relatively low and is again consistent with typical values for hopping transport. To our knowledge, this is the first study to successfully fabricate a single carrier diode with a VODP active layer and experimentally determine carrier mobility.

Fig. 5 (a) Schematic depicting space-charge-limited current (SCLC) diode device architecture. (b) J–V plot for a Cs₂TeBr₆ device at 280 K. Arrows depict the direction of voltage sweep. (c) Arrhenius plot of the hole mobility from J–V curves acquired at different temperatures.

Conclusion

In this work, we demonstrated the successful fabrication of high-quality Cs₂TeBr₆ films via a two-step spin-coating process, offering a reliable method to produce lead-free vacancy-ordered double perovskites. By optimizing spin-coating parameters, we determined the conditions required for complete conversion of TeBr₄ to Cs₂TeBr₆ with PXRD and GIWAXS analysis. The adaptability of this method was further explored with by mixing the halides, revealing the selective formation of mixed-halide phases as confirmed with GIWAXS and PL. Lastly, the valence band energetics of Cs₂TeBr₆ were characterized by UPS and hole-only devices were fabricated that enable the observation of space-charge-limited current. These findings lay a strong foundation for the future design of multi-layer devices that leverage the unique properties of Cs₂TeBr₆ and related MHPs, offering promising alternatives for sustainable optoelectronic applications.

Author contributions

The manuscript was written through contributions of all authors. O. K. synthesized the materials and fabricated films and devices with assistance from A. B. and L. T. and characterized them. A. S. M. performed DFT calculations. O. K. analyzed the data. C. M. S.-F. assisted with experiment design for in situ optical characterization at The Molecular Foundry. T. K. developed Python script for analyzing and plotting in situ PL data. M. L. C. and L. T. assisted with device testing and analysis. O. K., A. B., M. L. C., and R. S. designed the project scope. All authors have given approval to the final version of the manuscript.

Data availability

Data for this article, including X-ray scattering, spectroscopic data, and electronic transport data, are available at the Dryad Repository at: https://doi.org/10.5061/dryad.qnk98sfv0.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

DE-SC0024422, DE-AC02-05CH11231, NSF DMR-2308708, DoD ARO DURIP 66886LSRIP, CNS-1725797. This work was supported by the U. S. Department of Energy, Office of Science, Basic Energy Sciences, under DE-SC0024422. The investigation made use of shared facilities of the National Science Foundation (NSF) Materials Research Science and Engineering Center (MRSEC) (NSF DMR-2308708) and the Optical Characterization Facility (DoD ARO DURIP 66886LSRIP) at UC Santa Barbara. Use was made of computational facilities purchased with support from the National Science Foundation (CNS-1725797) and administered by the Center for Scientific Computing (CSC). The CSC is supported by the California NanoSystems Institute and the Materials Research Science and Engineering Center (NSF DMR-2308708) at UC Santa Barbara. Use of the Stanford Synchrotron Radiation Lightsource, SLAC National Accelerator Laboratory, is supported by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Contract No. DE-AC02-76SF00515. Work at the Molecular Foundry was supported by the Office of Science, Office of Basic Energy Sciences, of the U.S. Department of Energy under Contract No. DE-AC02-05CH11231.

References

1. S. Wu, Z. Chen, H.-L. Yip and A. K.-Y. Jen, The Evolution and Future of Metal Halide Perovskite-Based Optoelectronic Devices, Matter, 2021, 4(12), 3814–3834,  DOI:10.1016/j.matt.2021.10.026.
2. H. Dong, C. Ran, W. Gao, M. Li, Y. Xia and W. Huang, Metal Halide Perovskite for Next-Generation Optoelectronics: Progresses and Prospects, eLight, 2023, 3(1), 3,  DOI:10.1186/s43593-022-00033-z.
3. Y. Chen, H. T. Yi, X. Wu, R. Haroldson, Y. N. Gartstein, Y. I. Rodionov, K. S. Tikhonov, A. Zakhidov, X.-Y. Zhu and V. Podzorov, Extended Carrier Lifetimes and Diffusion in Hybrid Perovskites Revealed by Hall Effect and Photoconductivity Measurements, Nat. Commun., 2016, 7(1), 12253,  DOI:10.1038/ncomms12253.
4. S. Liu, V. P. Biju, Y. Qi, W. Chen and Z. Liu, Recent Progress in the Development of High-Efficiency Inverted Perovskite Solar Cells, NPG Asia Mater., 2023, 15(1), 27,  DOI:10.1038/s41427-023-00474-z.
5. D. Yang, B. Zhao, T. Yang, R. Lai, D. Lan, R. H. Friend and D. Di, Toward Stable and Efficient Perovskite Light-Emitting Diodes, Adv. Funct. Mater., 2022, 32(9), 2109495,  DOI:10.1002/adfm.202109495.
6. T. H. Bui and J. H. Shin, Perovskite Materials for Sensing Applications: Recent Advances and Challenges, Microchem. J., 2023, 191, 108924,  DOI:10.1016/j.microc.2023.108924.
7. M. Xie and J. Tian, Operational Stability Issues and Challenges in Metal Halide Perovskite Light-Emitting Diodes, J. Phys. Chem. Lett., 2022, 13(8), 1962–1971,  DOI:10.1021/acs.jpclett.1c04210.
8. L. Mao, C. C. Stoumpos and M. G. Kanatzidis, Two-Dimensional Hybrid Halide Perovskites: Principles and Promises, J. Am. Chem. Soc., 2019, 141(3), 1171–1190,  DOI:10.1021/jacs.8b10851.
9. D. Duan, C. Ge, M. Z. Rahaman, C.-H. Lin, Y. Shi, H. Lin, H. Hu and T. Wu, Recent Progress with One-Dimensional Metal Halide Perovskites: From Rational Synthesis to Optoelectronic Applications, NPG Asia Mater., 2023, 15(1), 8,  DOI:10.1038/s41427-023-00465-0.
10. Z. Huang, A. H. Proppe, H. Tan, M. I. Saidaminov, F. Tan, A. Mei, C.-S. Tan, M. Wei, Y. Hou, H. Han, S. O. Kelley and E. H. Sargent, Suppressed Ion Migration in Reduced-Dimensional Perovskites Improves Operating Stability, ACS Energy Lett., 2019, 4(7), 1521–1527,  DOI:10.1021/acsenergylett.9b00892.
11. J. Cho, J. T. DuBose, A. N. T. Le and P. V. Kamat, Suppressed Halide Ion Migration in 2D Lead Halide Perovskites, ACS Mater. Lett., 2020, 2(6), 565–570,  DOI:10.1021/acsmaterialslett.0c00124.
12. H. Lai, B. Kan, T. Liu, N. Zheng, Z. Xie, T. Zhou, X. Wan, X. Zhang, Y. Liu and Y. Chen, Two-Dimensional Ruddlesden–Popper Perovskite with Nanorod-like Morphology for Solar Cells with Efficiency Exceeding 15%, J. Am. Chem. Soc., 2018, 140(37), 11639–11646,  DOI:10.1021/jacs.8b04604.
13. A. E. Maughan, A. M. Ganose, D. O. Scanlon and J. R. Neilson, Perspectives and Design Principles of Vacancy-Ordered Double Perovskite Halide Semiconductors, Chem. Mater., 2019, 31(4), 1184–1195,  DOI:10.1021/acs.chemmater.8b05036.
14. Y. Cai, W. Xie, H. Ding, Y. Chen, T. Krishnamoorthy, L. H. Wong, N. Mathews, S. G. Mhaisalkar, M. Sherburne and M. Asta, Computational Study of Halide Perovskite-Derived A₂BX₆ Inorganic Compounds: Chemical Trends in Electronic Structure and Structural Stability, Chem. Mater., 2017, 29(18), 7740–7749,  DOI:10.1021/acs.chemmater.7b02013.
15. A. E. Maughan, A. M. Ganose, A. M. Candia, J. T. Granger, D. O. Scanlon and J. R. Neilson, Anharmonicity and Octahedral Tilting in Hybrid Vacancy-Ordered Double Perovskites, Chem. Mater., 2018, 30(2), 472–483,  DOI:10.1021/acs.chemmater.7b04516.
16. F. Zhang, W. Gao, G. J. Cruz, Y. Sun, P. Zhang and J. Zhao, Giant Excitonic Effects in Vacancy-Ordered Double Perovskites, Phys. Rev. B, 2023, 107(23), 235119,  DOI:10.1103/PhysRevB.107.235119.
17. Z. Xu, X. Jiang, H. Cai, K. Chen, X. Yao and Y. Feng, Toward a General Understanding of Exciton Self-Trapping in Metal Halide Perovskites, J. Phys. Chem. Lett., 2021, 12(43), 10472–10478,  DOI:10.1021/acs.jpclett.1c02291.
18. M. Adhikari, N. Shrivastava, S. T. McClain, C. M. Adhikari, B. Guzelturk, R. Khanal, B. Gautam and Z. Luo, Luminescence from Self-Trapped Excitons and Energy Transfers in Vacancy-Ordered Hexagonal Halide Perovskite Cs₂HfF₆ Doped with Rare Earths for Radiation Detection, Adv. Opt. Mater., 2022, 10(19), 2201374,  DOI:10.1002/adom.202201374.
19. A. E. Maughan, A. M. Ganose, M. A. Almaker, D. O. Scanlon and J. R. Neilson, Tolerance Factor and Cooperative Tilting Effects in Vacancy-Ordered Double Perovskite Halides, Chem. Mater., 2018, 30(11), 3909–3919,  DOI:10.1021/acs.chemmater.8b01549.
20. P. Bhumla, M. Jain, S. Sheoran and S. Bhattacharya, Vacancy-Ordered Double Perovskites Cs₂BI₆ (B = Pt, Pd, Te, Sn): An Emerging Class of Thermoelectric Materials, J. Phys. Chem. Lett., 2022, 13(50), 11655–11662,  DOI:10.1021/acs.jpclett.2c02852.
21. B. Cucco, C. Katan, J. Even, M. Kepenekian and G. Volonakis, Fine Structure of Excitons in Vacancy-Ordered Halide Double Perovskites, ACS Mater. Lett., 2023, 5(1), 52–59,  DOI:10.1021/acsmaterialslett.2c01010.
22. M. C. Folgueras, J. Jin, M. Gao, L. N. Quan, J. A. Steele, S. Srivastava, M. B. Ross, R. Zhang, F. Seeler, K. Schierle-Arndt, M. Asta and P. Yang, Lattice Dynamics and Optoelectronic Properties of Vacancy-Ordered Double Perovskite Cs₂TeX₆ (X = Cl⁻, Br⁻, I⁻) Single Crystals, J. Phys. Chem. C, 2021, 125(45), 25126–25139,  DOI:10.1021/acs.jpcc.1c08332.
23. M. Sajedi Alvar, P. W. M. Blom and G.-J. A. H. Wetzelaer, Space-Charge-Limited Electron and Hole Currents in Hybrid Organic-Inorganic Perovskites, Nat. Commun., 2020, 11(1), 4023,  DOI:10.1038/s41467-020-17868-0.
24. V. M. Le Corre, E. A. Duijnstee, O. El Tambouli, J. M. Ball, H. J. Snaith, J. Lim and L. J. A. Koster, Revealing Charge Carrier Mobility and Defect Densities in Metal Halide Perovskites via Space-Charge-Limited Current Measurements, ACS Energy Lett., 2021, 6(3), 1087–1094,  DOI:10.1021/acsenergylett.0c02599.
25. X. Cao, G. Zhang, L. Hao, X. Ding, T. Dong, X. Li, Q. Zeng, X. He, Y. Jia and J. Wei, Achieving One-Step Solution Deposition of High Quality CsPbBr3 Films for Efficient Solar Cells through Halide Ion Exchange, J. Alloys Compd., 2022, 919, 165722,  DOI:10.1016/j.jallcom.2022.165722.
26. N. Sun, S. Fu, Y. Li, L. Chen, J. Chung, M. M. Saeed, K. Dolia, A. Rahimi, C. Li, Z. Song and Y. Yan, Tailoring Crystallization Dynamics of CsPbI₃ for Scalable Production of Efficient Inorganic Perovskite Solar Cells, Adv. Funct. Mater., 2024, 34(6), 2309894,  DOI:10.1002/adfm.202309894.
27. Y. Han, H. Xie, E. L. Lim and D. Bi, Review of Two-Step Method for Lead Halide Perovskite Solar Cells, Sol. RRL, 2022, 6(6), 2101007,  DOI:10.1002/solr.202101007.
28. M. Chauhan, Y. Zhong, K. Schötz, B. Tripathi, A. Köhler, S. Huettner and F. Panzer, Investigating Two-Step MAPbI 3 Thin Film Formation during Spin Coating by Simultaneous in Situ Absorption and Photoluminescence Spectroscopy, J. Mater. Chem. A, 2020, 8(10), 5086–5094,  10.1039/C9TA12409H.
29. M. F. Mohamad Noh, N. A. Arzaee, I. N. Nawas Mumthas, A. Aadenan, F. H. Saifuddin, A. D. A.-G. Syakirin, M. Abd Mutalib, M. Q. Lokman, M. A. Ibrahim and M. A. Mat Teridi, Superiority of Two-Step Deposition over One-Step Deposition for Perovskite Solar Cells Processed in High Humidity Atmosphere, Opt. Mater., 2021, 118, 111288,  DOI:10.1016/j.optmat.2021.111288.
30. M. C. Folgueras, S. Louisia, J. Jin, M. Gao, A. Du, S. C. Fakra, R. Zhang, F. Seeler, K. Schierle-Arndt and P. Yang, Ligand-Free Processable Perovskite Semiconductor Ink, Nano Lett., 2021, 21(20), 8856–8862,  DOI:10.1021/acs.nanolett.1c03308.
31. S. L. Hamukwaya, H. Hao, Z. Zhao, J. Dong, T. Zhong, J. Xing, L. Hao and M. M. Mashingaidze, A Review of Recent Developments in Preparation Methods for Large-Area Perovskite Solar Cells, Coatings, 2022, 12(2), 252,  DOI:10.3390/coatings12020252.
32. D. Zheng, F. Raffin, P. Volovitch and T. Pauporté, Control of Perovskite Film Crystallization and Growth Direction to Target Homogeneous Monolithic Structures, Nat. Commun., 2022, 13(1), 6655,  DOI:10.1038/s41467-022-34332-3.
33. S. C. Ezike, A. D. Ahmed, R. M. Obodo and M. A. Salawu, Perovskite Precursor Concentration for Enhanced Recombination Suppression in Perovskite Solar Cells, Hybrid Adv., 2022, 1, 100006,  DOI:10.1016/j.hybadv.2022.100006.
34. N. N. Greenwood, B. P. Straughan and A. E. Wilson, Behaviour of Tellurium(IV) Chloride, Bromide, and Iodide in Organic Solvents and the Structures of the Species Present, J. Chem. Soc. A, 1968, 2209,  10.1039/j19680002209.
35. A. Brumberg, O. Kuklinski, G. T. Kent, E. E. Morgan, A. A. Mikhailovsky, T. A. Strom, M. L. Chabinyc and R. Seshadri, Tuning the Optical Absorption Edge of Vacancy-Ordered Double Perovskites through Metal Precursor and Solvent Selection, Chem. Mater., 2024, 36(19), 9625–9635,  DOI:10.1021/acs.chemmater.4c01701.
36. B. Buss and B. Krebs, Crystal Structure of Tellurium Tetrachloride, Inorg. Chem., 1971, 10(12), 2795–2800,  DOI:10.1021/ic50106a035.
37. J. Jiang, G. Niu, L. Sui, X. Wang, X. Zeng, Y. Zhang, L. Che, G. Wu, K. Yuan and X. Yang, Six-Photon Excited Self-Trapped Excitons Photoluminescence in Lead-Free Halide Perovskite, Adv. Opt. Mater., 2023, 11(5), 2202634,  DOI:10.1002/adom.202202634.
38. J. Butkus, P. Vashishtha, K. Chen, J. K. Gallaher, S. K. K. Prasad, D. Z. Metin, G. Laufersky, N. Gaston, J. E. Halpert and J. M. Hodgkiss, The Evolution of Quantum Confinement in CsPbBr₃ Perovskite Nanocrystals, Chem. Mater., 2017, 29(8), 3644–3652,  DOI:10.1021/acs.chemmater.7b00478.
39. J. A. Steele, E. Solano, D. Hardy, D. Dayton, D. Ladd, K. White, P. Chen, J. Hou, H. Huang, R. A. Saha, L. Wang, F. Gao, J. Hofkens, M. B. J. Roeffaers, D. Chernyshov and M. F. Toney, How to GIWAXS: Grazing Incidence Wide Angle X-Ray Scattering Applied to Metal Halide Perovskite Thin Films, Adv. Energy Mater., 2023, 2300760,  DOI:10.1002/aenm.202300760.
40. L. D’Alessio, D. Ferro and V. Piacente, Sublimation Study of Tellurium Tetrachloride and Tetraiodide from Their Vapour Pressure Measurements, J. Alloys Compd., 1994, 209(1–2), 207–212,  DOI:10.1016/0925-8388(94)91099-5.
41. S. A. Shlykov, N. I. Giricheva, A. V. Titov, M. Szwak, D. Lentz and G. V. Girichev, The Structures of Tellurium(IV) Halides in the Gas Phase and as Solvated Molecules, Dalton Trans., 2010, 39(13), 3245,  10.1039/b922474b.
42. S. A. Shlykov, H. Oberhammer, A. V. Titov, N. I. Giricheva and G. V. Girichev, A Combined Gas-Phase Electron Diffraction/Mass Spectrometric Study of the Sublimation Processes of TeBr₄ and TeI₄: The Molecular Structure of Tellurium Dibromide and Tellurium Diiodide, Eur. J. Inorg. Chem., 2008,(33), 5220–5227,  DOI:10.1002/ejic.200800700.
43. M. Karlsson, Z. Yi, S. Reichert, X. Luo, W. Lin, Z. Zhang, C. Bao, R. Zhang, S. Bai, G. Zheng, P. Teng, L. Duan, Y. Lu, K. Zheng, T. Pullerits, C. Deibel, W. Xu, R. Friend and F. Gao, Mixed Halide Perovskites for Spectrally Stable and High-Efficiency Blue Light-Emitting Diodes, Nat. Commun., 2021, 12(1), 361,  DOI:10.1038/s41467-020-20582-6.
44. Q. Wei, M. Ghasemi, R. Wang, C. Wang, J. Wang, W. Zhou, B. Jia, Y. Yang and X. Wen, Metal Halide Perovskite Alloy: Fundamental, Optoelectronic Properties and Applications, Adv. Photonics Res., 2023, 4(2), 2200236,  DOI:10.1002/adpr.202200236.
45. J. Tauc, R. Grigorovici and A. Vancu, Optical Properties and Electronic Structure of Amorphous Germanium, Phys. Status Solidi B, 1966, 15(2), 627–637,  DOI:10.1002/pssb.19660150224.
46. A. Dolgonos, T. O. Mason and K. R. Poeppelmeier, Direct Optical Band Gap Measurement in Polycrystalline Semiconductors: A Critical Look at the Tauc Method, J. Solid State Chem., 2016, 240, 43–48,  DOI:10.1016/j.jssc.2016.05.010.
47. Q. Guo, X. Zhao, B. Song, J. Luo and J. Tang, Light Emission of Self-Trapped Excitons in Inorganic Metal Halides for Optoelectronic Applications, Adv. Mater., 2022, 34(52), 2201008,  DOI:10.1002/adma.202201008.
48. S. Li, J. Luo, J. Liu and J. Tang, Self-Trapped Excitons in All-Inorganic Halide Perovskites: Fundamentals, Status, and Potential Applications, J. Phys. Chem. Lett., 2019, 10(8), 1999–2007,  DOI:10.1021/acs.jpclett.8b03604.
49. B. Pal, A. J. Kale, M. Sharma, K. C. Bhamu, S. G. Kang, V. K. Singh and A. Dixit, Inorganic Cs2TeX6 (X = Cl, Br, I) Lead-Free Vacancy-Ordered Double-Perovskite Absorber-Based Single-Junction Solar Cells with a Higher Efficiency of ∼24%: Theoretical Insights, Energy Fuels, 2024, acs.energyfuels.3c03030,  DOI:10.1021/acs.energyfuels.3c03030.
50. J. Jin, C. Chen, P. W. Smith, M. C. Folgueras, S. Yu, Y. Zhang, P.-C. Chen, F. Seeler, B. Schaefer, C. Lizandara-Pueyo, R. Zhang, K. Schierle-Arndt and P. Yang, Benzyl Alcohol Photo-Oxidation Based on Molecular Electronic Transitions in Metal Halide Perovskites, ACS Photonics, 2023, 10(3), 772–779,  DOI:10.1021/acsphotonics.3c00042.
51. W. M. H. Sachtler, G. J. H. Dorgelo and A. A. Holscher, The Work Function of Gold, Surf. Sci., 1966, 5(2), 221–229,  DOI:10.1016/0039-6028(66)90083-5.
52. Y. Park, V. Choong, Y. Gao, B. R. Hsieh and C. W. Tang, Work Function of Indium Tin Oxide Transparent Conductor Measured by Photoelectron Spectroscopy, Appl. Phys. Lett., 1996, 68(19), 2699–2701,  DOI:10.1063/1.116313.
53. E. Avigad and L. Etgar, Studying the Effect of MoO₃ in Hole-Conductor-Free Perovskite Solar Cells, ACS Energy Lett., 2018, 3(9), 2240–2245,  DOI:10.1021/acsenergylett.8b01169.
54. B. Dasgupta, W. P. Goh, Z. E. Ooi, L. M. Wong, C. Y. Jiang, Y. Ren, E. S. Tok, J. Pan, J. Zhang and S. Y. Chiam, Enhanced Extraction Rates through Gap States of Molybdenum Oxide Anode Buffer, J. Phys. Chem. C, 2013, 117(18), 9206–9211,  DOI:10.1021/jp3114013.
55. S. Sun, M. Xu, Y. Zhang, R. Liu, X. Wang, L. Zhang, Y. Fang and P. Wang, Study of Molybdenum Oxide Optimized Hole Carrier Transport in Perovskite Solar Cells, Org. Electron., 2023, 113, 106697,  DOI:10.1016/j.orgel.2022.106697.

---

Footnote

† Electronic supplementary information (ESI) available: In situ optical characterization, GIWAXS 2D diffraction patterns and grazing angle dependence, spin-coating layer thickness estimate, further X-ray diffraction phase identification, DFT methods, UPS band energy analysis, SCLC leakage current correction and Mott–Gurney fits. See DOI: https://doi.org/10.1039/d5tc00502g

---