# Fileset
[Valenta_Adv_Mater_Interf2024.pdf](https://mdr.nims.go.jp/filesets/d50468ea-98e7-4c4d-b39f-7b9bddba83d5/download)
## Creator
[Donald Valenta](https://orcid.org/0009-0003-8080-0285), [Hasan Arif Yetkin](https://orcid.org/0000-0002-1401-5866), [Tim Kodalle](https://orcid.org/0000-0002-8792-9669), [Jakob Bombsch](https://orcid.org/0000-0002-0820-162X), [Raul Garcia‐Diez](https://orcid.org/0009-0000-9374-1083), [Claudia Hartmann](https://orcid.org/0000-0002-8017-8161), [Shigenori Ueda](https://orcid.org/0000-0001-9425-0614), Roberto Félix, [Johannes Frisch](https://orcid.org/0000-0002-7895-7529), [Lucas Bodenstein‐Dresler](https://orcid.org/0000-0001-7602-2773), [Regan G. Wilks](https://orcid.org/0000-0001-5822-8399), Christian A. Kaufmann, [Marcus Bär](https://orcid.org/0000-0001-8581-0691)
## Rights
[Creative Commons BY Attribution 4.0 International](https://creativecommons.org/licenses/by/4.0/)
## Other metadata
[The Energy Level Alignment at the Buffer/Cu(In,Ga)Se2 Thin‐Film Solar Cell Interface for CdS and GaOx](https://mdr.nims.go.jp/datasets/f07de4cc-5880-46e6-9d8f-af3747f6687d)
## Fulltext
The Energy Level Alignment at the Buffer/Cu(In,Ga)Se2 Thin‐Film Solar Cell Interface for CdS and GaOxRESEARCH ARTICLEEditor’s Choice www.advmatinterfaces.deThe Energy Level Alignment at the Buffer/Cu(In,Ga)Se2Thin-Film Solar Cell Interface for CdS and GaOxDonald Valenta,* Hasan Arif Yetkin, Tim Kodalle, Jakob Bombsch, Raul Garcia-Diez,Claudia Hartmann, Shigenori Ueda, Roberto Félix, Johannes Frisch,Lucas Bodenstein-Dresler, Regan G. Wilks,* Christian A. Kaufmann, and Marcus Bär*Sputter-deposited GaOx (i.e., oxygen-deficient gallium oxide) films areevaluated as a potential replacement for the standard CdS buffer layers inCu(In,Ga)Se2 (CIGSe) based thin-film photovoltaics. The energy levelalignment at the GaOx/CIGSe and CdS/CIGSe interfaces are compared bymeans of direct and inverse photoemission. For the GaOx/CIGSe a(0.04 ± 0.07) eV (i.e., a small spike-like) conduction band offset (CBO) and a(−3.21 ± 0.19) eV (i.e., a large cliff-like) valence band offset (VBO) are found,which suggests a nearly ideal charge-selective contact. The derived GaOxband gap of (4.80 ± 0.25) eV confirms its utility as a highly transparent bufferlayer. However, the GaOx (with x derived to be 1.1 ± 0.1) exhibits considerable(presumably) defect-related occupied states above the valence bandmaximum. It is proposed that these states may increase charge carrierrecombination and decrease open circuit voltage in respective devices; alsoexplaining why solar cells with standard CdS buffer outperform devices withGaOx buffer, despite less ideal electronic interface properties (CBO:(−0.18 ± 0.07) eV, VBO: (−0.98 ± 0.15) eV) and the smaller CdS band gap of(2.35 ± 0.22) eV.D. Valenta, J. Bombsch, R. Garcia-Diez, C. Hartmann, R. Félix, J. Frisch,L. Bodenstein-Dresler, R. G. Wilks, M. BärInterface DesignHelmholtz-Zentrum Berlin für Materialien und Energie GmbH (HZB)Berlin GermanyE-mail: donald.valenta@helmholtz-berlin.de;regan.wilks@helmholtz-berlin.de; marcus.baer@helmholtz-berlin.deH. A. Yetkin, T. Kodalle, C. A. KaufmannPVcomBHZBBerlin GermanyS. UedaSynchrotron X-ray Station at SPring-8National Institute for Materials Science (NIMS)Kouto, Sayo, Hyogo JapanThe ORCID identification number(s) for the author(s) of this articlecan be found under https://doi.org/10.1002/admi.202301110© 2024 The Authors. Advanced Materials Interfaces published byWiley-VCH GmbH. This is an open access article under the terms of theCreative Commons Attribution License, which permits use, distributionand reproduction in any medium, provided the original work is properlycited.DOI: 10.1002/admi.2023011101. IntroductionThin-film solar cells based onCu(In,Ga)Se2 (CIGSe) chalcopyritelight absorbers are a stable, low band gapbottom device technology for tandemsolar cells. They are potentially flexibleand lightweight and low material andenergy consumption promises lower pro-duction costs. CIGSe devices can reachpower conversion efficiencies exceeding>23%.[1,2] For these high efficiencies,cadmium sulfide (CdS) is commonlyemployed as a buffer layer between theCIGSe absorber and the emitter. Theuse of a heavy metal, Cd, and parasiticabsorption due to the short-wavelengthrange are strong motivations for replac-ing it with a nontoxic, more transparentalternative buffer. To be considered aviable alternative to CdS in CIGSe-basedsolar cells, certain requirements mustS. UedaResearch Center for Electronic and Optical MaterialsNIMSTsukuba, Ibaraki JapanR. Félix, J. Frisch, R. G. Wilks, M. BärEnergy Materials In-Situ Laboratory Berlin (EMIL)HZBBerlin GermanyM. BärHelmholtz-Institute Erlangen-Nürnberg for Renewable Energy (HIERN)Berlin GermanyM. BärDepartment of Chemistry and PharmacyFriedrich-Alexander-Universität Erlangen-Nürnberg (FAU)Erlangen GermanyAdv. Mater. Interfaces 2024, 11, 2301110 2301110 (1 of 7) © 2024 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbHhttp://www.advmatinterfaces.demailto:donald.valenta@helmholtz-berlin.demailto:regan.wilks@helmholtz-berlin.demailto:marcus.baer@helmholtz-berlin.dehttps://doi.org/10.1002/admi.202301110http://creativecommons.org/licenses/by/4.0/http://crossmark.crossref.org/dialog/?doi=10.1002%2Fadmi.202301110&domain=pdf&date_stamp=2024-03-08www.advancedsciencenews.com www.advmatinterfaces.debe fulfilled.[3] The conduction band offset (CBO) is one of themost crucial parameters for unimpeded electron transport acrossthe buffer/absorber.[4] An aligned conduction band (i.e., a CBOof zero) is ideal for electron transport, however, theoreticalanalysis[5–7] shows that high performance can also be obtainedwith a positive CBO of up to +0.2 to +0.4 eV and with a neg-ative CBO of up to −0.1 to −0.4 eV depending on whether ornot a Cu-poor CIGSe surface layer (often called OVC—orderedvacancy compound)[8,9] is present and how in the underlyingmodels the offset is distributed between the CdS/OVC and theOVC/CIGSe interfaces. For CdS/chalcopyrite stacks resulting inhigh-efficiency solar cells, aligned[10,11] or slightly positive[11,12]CBO values have, indeed, been reported in the past.Gallium oxide is a wide band gap semiconductor prominentfor field effect passivation[13,14] with minimal related environ-mental and health concerns and thus is a possible CdS bufferlayer alternative candidate material. It displays n-type conductiv-ity when O-deficient compared to its Ga2O3 stoichiometry (wewill refer to such material as GaOx below).[3,15–18] This n-typeconductivity is attributed to oxygen vacancy-derived defect states;consequently, the conductivity of GaOx films is generally con-nected to an oxygen-deficient environment during the prepara-tion process,[9,11] which may open a convenient route for delib-erately tuning the electronic structure of GaOx when applied asbuffer layer in CIGSe devices. GaOx films may also fulfill anotherhighly desirable requirement, namely producing an interface thatis less prone to temperature-induced interdiffusion (comparedto layer stacks with CdS as interface partners). This is a crucialrequirement to benefit from high-temperature deposition of thetransparent conductive oxide (used as front contact) or for theapplication of CIGSe cells in tandem or multijunction devicesto avoid negative effects of subsequent high-temperature processstages (e.g., during the deposition of the top cell).[18]In this work, we compare the electronic structure (and herespecifically the energy level alignment at the interface) of thestandard buffer/absorber interface CdS/CIGSe with that of thepotential alternative GaOx/CIGSe. In order to do so, we com-bined several photoelectron spectroscopy techniques such assynchrotron-based hard X-ray photoelectron spectroscopy (HAX-PES), lab-based X-ray photoelectron spectroscopy (XPS), lab-based ultraviolet photoelectron spectroscopy (UPS) and inversephotoelectron spectroscopy (IPES) to thoroughly study and di-rectly determine the electronic structure at the buffer/absorberinterfaces. The derived energy level alignment will provide a de-tailed insight into the charge carrier transport mechanism acrossthe buffer/absorber interface and how it may affect device effi-ciency.2. Experimental Section2.1. Sample PreparationIn this study, buffer/CIGSe samples were made at the Com-petence Centre for Photovoltaics Berlin (PVcomB) at HZB em-ploying their standard deposition process.[19] On a cleaned50 × 50 × 2 mm3 soda-lime glass (SLG) substrate, the molybde-num back contact of 800 nm thickness was deposited by DC sput-tering, followed by the 2.1-μm-thick CIGSe thin-film absorberprepared by a three-stage physical vapor deposition (PVD) pro-cess. To complete the buffer/CIGSe/Mo/SLG stack, buffer lay-ers of CdS or GaOx were deposited (a thorough characteriza-tion of the physical properties of similarly prepared buffer lay-ers and related devices has previously been reported[19]). For CdSdeposition the CIGSe was first rinsed in a 2.5% aqueous am-monia solution, followed by chemical bath deposition (CBD) us-ing anhydrous cadmium acetate (Cd(CH3COOH)2) and thiourea(S═C(NH2)2) in an aqueous ammonia-containing solution. Toobtain a nominal layer thickness of 50 nm, the CIGSe/Mo/SLGstack was immersed into 1800 mL CBD solution for 16 min dur-ing which the temperature of the CBD solution increased from 25to ≈50 °C. Thinner layers were produced by interrupting the CBDearlier; a 10 nm nominal thickness was obtained after a CdS CBDof 4 min and 36 s. The 5- and 50-nm-thick GaOx buffer layers weredeposited via radio frequency (RF) magnetron sputtering usinga 5 N stoichiometric Ga2O3 target. Deposition was carried outwith 70 W power in an Ar2 atmosphere with a sputtering pres-sure of 5 μbar without additional oxygen. During the sputteringprocess, the substrate was not externally heated. After the deposi-tion processes, in both cases, the buffer/absorber samples werevacuum-sealed to minimize air exposure and surface contami-nation. Solar cells prepared using identical processes resulted inefficiencies of 13.0% for devices with GaOx buffer and 16.2% forCdS buffered cells.[18]2.2. Photoelectron SpectroscopyHAXPES measurements were performed using an excitation en-ergy of linearly polarized 5.95 keV X-rays (referred to as 6 keVin this manuscript) at BL15XU of the SPring-8 electron storagering,[20] equipped with a hemispherical electron analyzer (VGScienta R4000) oriented perpendicular to the X-ray propagationaccess and with the analyzer axis and X-ray polarization vectoraligned. Using the high-resolution Si 333 channel-cut monochro-mator and an electron analyzer pass energy of 200 eV, a total en-ergy resolution of 0.25 eV was achieved for all HAXPES measure-ments. Calibration of the binding energy was done by referencingthe Au 4f7/2 peak to 84.00 eV. The base pressure of the analysischamber during measurements was <6 × 10−9 mbar. The take-off-angle of photoelectrons was set to ≈88° with respect to thesample surface for all measurements.Laboratory-based XPS and UPS measurements were per-formed in the Energy Materials In Situ Laboratory Berlin(EMIL) at HZB. XPS spectra were obtained by using non-monochromatized Al K𝛼(1486.58 eV) and Mg K𝛼(1253.56 eV)excitation from a SPECS XR 50 twin-anode X-ray source. Thesource and analyzer are arranged in “magic-angle” geometry(≈54.7° between X-ray propagation and analyzer axes). For theUPS measurements, He I (21.2 eV) and He II (40.8 eV) exci-tation from a Prevac UVS 40A2 gas-discharge lamp was used.In both cases the photoelectrons were detected by a ScientaOmi-cron Argus CU electron analyzer operating at a base pressure of<5 × 10−9 mbar. By setting the electron analyzer pass energy to20 eV, total energy resolutions of 1.2 and 0.9 eV for excitationenergies of Al K𝛼and Mg K𝛼, respectively, were achieved. The to-tal energy resolution for the He I and He II UPS measurementswas 100 meV. Binding energy calibration for XPS was carried byreferencing the Au 4f7/2 peak to 84.00 eV; for UPS the Au FermiAdv. Mater. Interfaces 2024, 11, 2301110 2301110 (2 of 7) © 2024 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 21967350, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202301110 by Cochrane Japan, Wiley Online Library on [07/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmatinterfaces.dewww.advancedsciencenews.com www.advmatinterfaces.deedge (EF) was set to 0.00 eV. Samples were oriented with samplenormal colinear with the analyzer entrance.For fitting HAXPES and XPS spectra, Voigt functions wereused to model individual spectral species; for a given photoe-mission line and experimental apparatus, the Gaussian andLorentzian contributions were linked and optimized during si-multaneous fitting of all measurements. Spin-orbit doublet sep-arations were linked and optimized similarly, while doublet ratioswere assigned based on the multiplicity of the core level. Further-more, in cases where multiple contributions (i.e., from distinctchemical/electronic environments) to a photoemission line werepresent, the energy separation between contributions was linkedand optimized during the fit procedure. This fit approach was in-tended to minimize the uncertainty in relative peak positions byconstraining the fits to conform to a common physical model;the uncertainty in peak shifts was therefore primarily influencedby instability of the measurement apparatus, which was consid-ered to be negligible compared to the more subjective errors indetermining the VBM and CBM deriving spectral onsets of UPSand IPES spectra (discussed in detail below). The peak intensi-ties were used to estimate stoichiometry of the measured sam-ples according to the procedures described in the Supporting In-formation; relevant values are contained in (Table S1, SupportingInformation).The IPES measurements were performed in the same systemas the lab-based XPS and UPS studies. A Kimball Physics inc.EGPS-1022E electron gun with a BaO cathode firing unit wasused for excitation, and an OmniVac IPES1000 solid-state detec-tor was used to record photon intensity. The base pressure duringmeasurements was <5 × 10−9 mbar. The energy was calibratedby setting the Au EF to 0.00 eV. The total energy resolution of thissolid-state IPES setup is ≈1 eV.Since both UPS and IPES are very surface sensitive techniquesthat can easily be affected by surface contaminations, a gentlecleaning procedure using subsequent Ar+ ion treatment cycleswith ion energies of 50 eV (i.e., energies significantly below anysputter threshold) was performed on the samples. To make surethat surface contaminations were being efficiently removed, Ar+ion treatment cycles of 20–30 min duration were performed withsubsequent XPS measurements monitoring changes in C 1s (andfor the CdS/CIGSe and CIGSe sample also the O 1s) spectra in-tensity. The Ar+ ion treatment cycles were stopped when changesof the C 1s (O 1s) spectrum were no longer visible. For the bareCIGSe absorber this was the case after an accumulated Ar+ iontreatment time of 120 min; for the absorbers with depositedbuffer layers, i.e., CdS and GaOx, an Ar+ ion treatment of 200 minwas needed. Exemplarily, the XPS survey spectra before and af-ter 120 min Ar+ treatment are shown for the bare CIGSe and theGaOx/CIGSe sample in Figure S1 (Supporting Information).The positions of the valence band maximum (VBM) and con-duction band minimum (CBM) were derived by linear approx-imation of the leading edges of the UPS and IPES spectra, re-spectively. This inherently caused some uncertainty, in particu-lar for the significantly less energy resolved IPES data. In thiscase, in an attempt to obtain the highest possible degree of accu-racy, the determination of the relative changes of the IPES onsetswere focused—and thence the CBOs. The following steps wereperformed: 1) A linear background obtained by a least-squaresfit of the region below the calibrated EF position (i.e., where noFigure 1. UPS (He I or He II excitation as indicated) and IPES spectraof the studied samples: bare CIGSe absorber (after 120 min of mild Ar+ion treatment) and CIGSe with nominal 50-nm-thick CdS and GaOx bufferlayer (after 200 min of mild Ar+ ion treatment), respectively. The linear ex-trapolation of the leading edge positions of the UPS (±0.10 eV for the bareCIGSe and the CdS/CIGSe sample and ±0.15 eV for the GaOx/CIGSe sam-ple in absolute position) and IPES (±0.20 eV in absolute position) spec-tra (and the computed surface band gap obtained from their differences(±0.22 eV for the bare CIGSe and the CdS/CIGSe sample and ±0.25 eVfor the GaOx/CIGSe sample)) are indicated. The IPES spectra have hada linear background subtracted; the background was obtained by a fit ofthe 0–2 eV region. The spectra before subtraction are plotted in light gray.Vertical offsets are added for clarity.unoccupied states are possible) was subtracted from all spectra;the spectra before subtraction are shown in light-gray points inFigure 1. 2) The spectra were shifted and scaled such that theirleading edges overlap over the largest possible energy range (seeFigure S2, Supporting Information), and a common linear ex-trapolation (also shown in Figure S2, Supporting Information)was used to determine the intersection of the spectral leadingedge with the background. This method attempted to eliminatethe variations in the somewhat subjective process of determiningthe spectral leading edge—any suggested change to the line po-sition would need to be applied to all spectra equally, leading to auniform change in the absolute value of the obtained onsets withno change in their relative values. It is due to this argumentationthat values derived from relative positions were assigned—i.e.,the CBOs—with relatively low experimental uncertainty.Adv. Mater. Interfaces 2024, 11, 2301110 2301110 (3 of 7) © 2024 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 21967350, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202301110 by Cochrane Japan, Wiley Online Library on [07/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmatinterfaces.dewww.advancedsciencenews.com www.advmatinterfaces.de3. Results and DiscussionThe HAXPES survey spectrum (see Figure S3, Supporting Infor-mation) of the bare CIGSe sample shows all expected peaks re-lated to Cu, In, Ga, and Se. These CIGSe-related peaks are com-pletely attenuated in the survey spectra of the samples with thick(nominally 50 nm) CdS and GaOx buffer layers; and only respec-tive buffer-related signals are observed, indicating complete cov-erage of the CIGSe by both CdS and GaOx layers, with a bufferthickness exceeding the HAXPES information depth, which inthis case is estimated to be between 20 and 30 nm.[21]Additional signals related to Na, O, and C can be identifiedin the survey spectrum of the bare CIGSe sample. The pres-ence of Na at the CIGSe surface has repeatedly been observedbefore[22–24] and is ascribed to thermally induced Na diffusionfrom the SLG through (and accumulation on top of) the CIGSeduring absorber growth at elevated temperatures. The presenceof O and C can be explained by surface contamination caused bythe short exposure to air during sample mounting and introduc-tion into the load lock of the HAXPES setup and/or incorpora-tion of O and C into the CIGSe during absorber during prepara-tion. To some extent Na, O, and C signals can also be observed inthe survey spectrum of the CdS/CIGSe sample, which can be ex-plained by the same reasoning and thus points to a Na diffusionalso into the CdS buffer. However, neither Na nor C is observed inthe HAXPES survey spectrum of the GaOx/CIGSe sample, sug-gesting that GaOx prevents/reduces Na diffusion and that the Cobserved in the spectra of the (CdS/)CIGSe samples does indicatea (partial) incorporation into the buffer and absorber.In this work, we use HAXPES data to derive the surface com-position of the studied samples, as it is less affected by surfaceeffects. Based on the shallow core levels Ga 3d, In 4d, and Cu3p (see Figures S4,S5, Table S1, Supporting Information), we de-rived the CGI = [Cu]/([Ga]+[In]) and GGI = [Ga]/([Ga]+[In]) ra-tios for the bare CIGSe absorber surface to be 0.6 and 0.4 (±0.1),respectively (for a detailed description of the HAXPES data quan-tification see description in conjunction with Table S1, Support-ing Information). Hence, the CIGSe surface is significantly Cudeficient compared to its expected bulk composition, in agree-ment with previous reports.[25–27] As one explanation for this find-ing the formation of an OVC with Cu:(In+Ga):Se compositionsof 1:3:5 or 1:5:8 at the CIGSe surface has been suggested in thepast.[8,9,28]Next, the composition of the buffer layers is addressed. Forthe nominal 50-nm-thick CBD CdS buffer, the [Cd]/[S] ratio wasderived based on the Cd 4d and S 2p measurements. To deter-mine the GaOx composition, we used the O 1s and the Ga 3dcore levels of the nominal 50-nm-thick GaOx buffer. As thesephotoemission lines have significantly different binding ener-gies, the respective line intensities (derived by correspondingfits, see Figure S4 and S5, Supporting Information) had to becorrected by the corresponding photoionization cross-sections,the inelastic mean free paths (IMFP) of the photoelectrons, andthe transmission function of the analyzer (for details see discus-sion in conjunction with Table S1, Supporting Information). Forthe CdS buffer, we find a slightly Cd-rich [Cd]/[S] ratio of 1.1(±0.1); in agreement (within the experimental uncertainty) withthe expected bulk composition ratio.[18] For the GaOx, we derivex to be 1.1 (±0.1), which matches well with the oxygen-deficient(compared to stoichiometric Ga2O3) composition previously re-ported for amorphous GaOx layers deposited in Ar2 atmosphere(via pulsed laser deposition)[29] or via oxidation of metallic Ga in1 × 10−6 mbar partial pressure of O2.[30]The UPS and IPES spectra of the CIGSe absorber with andwithout nominal 50-nm-thick CdS and GaOx buffers, respec-tively, are plotted in Figure 1 on a common energy scale relative tothe EF level. The leading edge of the UPS and IPES spectra wereobtained by linear extrapolation as discussed in the experimentalsection; they represent the VBM and CBM positions with respectto EF, respectively. As the VBM (CBM) is usually below (above) EF,the UPS leading edge = – VBM (IPES leading edge = – CBM), byconvention. The UPS spectra were obtained using different exci-tation energies, i.e., He I (21.2 eV) for the CdS and GaOx bufferedsamples, and He II (40.8 eV) for the bare CIGSe samples. Thelatter was specifically done to increase the photoionization crosssection of the Cu 3d derived states that dominate the CIGSe VBM,outweighing the intensity decrease when going from He I to HeII excitation.While the extrapolation of the leading VB edges is relativelystraightforward in the case of CIGSe (as the leading edge isformed by hybridized Se 4p—Cu 3d derived states[31,32]) and CdS(as the leading edge is formed by S 3s—S 3p derived states[31]),it is more complicated in the case of GaOx. The dominatingVB feature leading to the sloping onset at ≈ 4 eV is assignedto hybridized O 2p—Ga 4s derived states;[3] however, there issignificant spectral intensity visible between this edge and EF,which has previously been ascribed to defect states related to oxy-gen vacancies.[33,34] A contribution from band tails related to theamorphous structure of the deposited GaOx is also likely. Hence,we will treat these states as being distinct from the “true” valenceband and use the dominant O 2p—Ga 4s derived edge to deter-mine the VBM. Figure 1 shows two intersections of the leadingUPS edge with linear extrapolations of different backgrounds. Incase (i), the above-edge defect states are treated as the backgroundsignal resulting in an onset position of 3.75 eV. This procedureapproximates the leading edge that would result from subtrac-tion of the defect-related states and linear extrapolation of the re-mainder. In case (ii), we extrapolate the main edge to the spectralnoise floor observed around EF, ignoring these states completelyand resulting in a leading UPS edge position of 3.55 eV. This un-certainty dominates the error in deriving the VBM position of theGaOx; thus, in Table 1 and associated calculations, we use the av-erage of these two boundary values: (3.65 ± 0.15) eV. Togetherwith the derived CBM position ((1.15 ± 0.20) eV), this resultsin a GaOx surface band gap (EgSurf = CBM-VBM) of (4.8 ± 0.30)eV. This agrees well with the optically derived bulk band gap of4.71 eV reported for similarly prepared GaOx thin films.[18] TheVBM (±0.10 eV), CBM (±0.20 eV), and EgSurf (±0.25 eV) valuesfor the bare CIGSe absorber (the 50-nm-thick CdS/CIGSe sam-ple) are −0.40 (−1.55), 1.15 (0.80), and 1.55 (2.35) eV, respec-tively, as depicted in Figure 1. The values are in line with previ-ous UPS/IPES measurements for CIGSe absorbers having simi-lar surface CGI and GGI ratios (resulting in EgSurf values between1.40 and 1.61 eV[25–27]) and for CdS buffers (for which EgSurf val-ues between 2.40 and 2.65 eV have been reported[25,26]), in par-ticular when considering the experimental uncertainty. As previ-ously observed, the Cu-deficient CIGSe surface has a wider bandgap (EgSurf = (1.55 ± 0.25) eV) than the absorber bulk.[9,35]Adv. Mater. Interfaces 2024, 11, 2301110 2301110 (4 of 7) © 2024 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 21967350, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202301110 by Cochrane Japan, Wiley Online Library on [07/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmatinterfaces.dewww.advancedsciencenews.com www.advmatinterfaces.deTable 1. Core level binding energies (uncertainty ≈±0.02 eV) and UPS/IPES leading edge (-VBM/-CBM) positions (see main text for uncertainties) usedto calculate VBO and CBO at the CdS/CIGSe (top table) and GaOx/CIGSe (bottom table) interfaces.CdS/CIGSe interface (bare ref.) CIGSe (thick ref.)CdS/CIGSe(thin) CdS/CIGSe BE shift (thin – ref.)In 4d5/2 (Mg K𝛼) 18.04 eV – 18.17 eV 0.13 eVSe 3d5/2 (Mg K𝛼) 54.36 eV – 54.50 eV 0.14 eVS 2p3/2 (Mg K𝛼) – 161.92 eV 161.86 eV −0.06 eVCd 3d3/2 (Mg K𝛼) – 412.42 eV 412.41 eV −0.01 eVUPS leading edge 0.40 eV 1.55 eV – –IPES leading edge −1.15 eV −0.80 eV – –EgSurf 1.55 eV 2.35 eV – –VBO – – −0.98 eV –CBO – – −0.18 eV –GaOx/CIGSe interface (bare ref.)CIGSe(thick ref.)GaOx/CIGSe(thin)GaOx/CIGSeBE shift(thin – ref.)In 4d5/2 (6 keV) 17.72 eV – 17.84 eV 0.12 eVSe 3d5/2 (6 keV) 54.16 eV – 54.26 eV 0.10 eVGa 3d5/2 (6 keV) – 20.62 eV 20.69 eV 0.07 eVUPS leading edge 0.40 eV 3.65 eV – –IPES leading edge −1.15 eV −1.15 eV – –EgSurf 1.55 eV 4.80 eV – –VBO – – −3.21 eV –CBO – – 0.04 eV –To determine the VBO and CBO at the buffer/CIGSe interface,Kraut’s method is applied to determine the alignment at the in-tact interface.[36] In this context, a sample with the “thinnest pos-sible” closed buffer layer that is assumed to well-represent theinterface in the regular layer stack and a surface-sensitive pho-toemission measurement are needed in order to isolate the corelevel binding energies at the interface (required to determine theinterface-induced band bending) without “diluting” this informa-tion by integrating into the bulk of the sample.[37–39] Unfortu-nately, for the GaOx/CIGSe sample set it was not possible to ob-tain (closed) capping GaOx layers thin enough to allow probingof the covered CIGSe absorber by surface-sensitive photoemis-sion measurements. In this case, the interface could only be ac-cessed by using HAXPES on samples with intermediate (nomi-nally 5 nm) capping layer thicknesses; therefore, the contributionfrom the interface would be small, and any shift derived fromthese measurements should be considered a lower bound. Theinformation is therefore considered to be incomplete, and wetherefore increase the error budget of the derived energy levelalignment in this case.In both cases (CdS/CIGSe with XPS and GaOx/CIGSe withHAXPES), we detect a shift of the In 4d and Se 3d peaks rel-ative to the respective line positions at the surface of the bare(i.e., uncovered) CIGSe absorber. (NB: Unlike the measurementof the capping layer, the largest contribution of the buried CIGSelayer to the measured photoemission signal will originate fromnear the interface. However, when using bulk-sensitive HAX-PES, the larger IMFP still means that the interface contributionis smaller, compared to when probed by more surface sensitiveXPS measurements). The values needed to determine VBO andCBO via Kraut’s method are depicted in Table 1 (a more detaileddescription of the methodology and data are found in SI, section“Derivation of the valence (VBO) and conducting (CBO) band off-sets” including Figures S4–S7, Supporting Information). For thepurpose of determining peak positions and shifts representingsimilar sample regions with high confidence, peaks with rela-tively low natural line widths and similar binding energies wereselected.By combining the UPS and IPES leading edge positions withthe derived core-level shift values at the interface, the band offsetscan be calculated according to the following equations:VBO =(UPSa + shif ta)− (UPSb + shif tb) (1)CBO =(IPESa + shif ta)− (IPESb + shif tb) (2)where “a” and “b” denote the absorber and buffer, respectively.For shifta and shiftb the average of the derived respective val-ues for each core level was used. The resulting energy align-ment for the CdS/CIGSe interface is shown in Figure 2a) and forthe GaOx/CIGSe interface, it is depicted in Figure 2b). The val-ues for the VBM and CBM positions in the figures are correctedfor the interface induced band bending. The CBOs between theCIGSe absorber and the buffer layers are crucial, because theygovern the transport (and thus collection) of the electrons pho-togenerated in the absorber into the emitter. At the CdS/CIGSeinterface, the formation of a “cliff-like” (i.e., negative) offset of(−0.18 ± 0.15) eV is derived. This is in contrast to thealigned[10,11] or slightly positive[11,12] CBO values reported forCdS/chalcopyrite stacks resulting in high-efficiency solar cells.Adv. Mater. Interfaces 2024, 11, 2301110 2301110 (5 of 7) © 2024 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 21967350, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202301110 by Cochrane Japan, Wiley Online Library on [07/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmatinterfaces.dewww.advancedsciencenews.com www.advmatinterfaces.deFigure 2. Energy level alignment at the interface between the CIGSe ab-sorber layer and a) the CdS and b) the GaOx buffer layer. Each stated valueof VBM and CBM is corrected for interface-induced band bending effects.Please see text for the experimental uncertainties of the derived VBO andCBO values.If a Cu-poor CIGSe surface layer is present, a negative CBObetween −0.1 and −0.4 eV (depending on how in the underly-ing models the offset is distributed between the CdS/OVC andthe OVC/CIGSe interfaces) would be in agreement with reach-ing high efficiencies.[5–7] Thus, our finding might represent aninteresting borderline case with respect to whether the derivedCBO already limits the device efficiency. However, a conclusionon whether in the here studied case we have a rather small CBOat the OVC/CIGSe interface, which would mean that (for a high-efficiency devices) the acceptable “cliff-like” buffer/absorber CBOis rather small (i.e., −0.1 eV7) or whether the OVC/CIGSe CBO issimilar to that at the CdS/OVC interface and thus the acceptable“cliff-like” CBO can be as large as −0.4 eV6 is not possible.Subject to the caveats regarding the uncertainty in the corelevel binding energy shifts derived from the HAXPES measure-ments described above, for the GaOx/CIGSe interface, we findnearly perfect alignment, with a small “spike-like” (i.e., pos-itive) CBO of (0.04 ± 0.07) eV, which is in agreement witha buffer/absorber interface that principally could reach highefficiencies.[5–7]At both buffer/absorber interfaces, we find pronounced neg-ative VB offsets ((−0.98 ± 0.15) eV for the CdS/CIGSe and(−3.21 ± 0.19) eV for the GaOx/CIGSe interface) that are princi-pally sufficient to block photo-generated holes preventing a high-rate recombination path across the buffer/CIGSe interface. How-ever, despite having the seemingly better energy level alignment,solar cells with GaOx buffer are clearly inferior to their counter-parts with standard CdS buffer.[18] This indicates that the spectralintensity above the VBM of GaOx (see Figure 1) can indeed be as-cribed to defect states that act as charge carrier recombinationcenters, undermining the principally (almost) ideal electronic in-terface properties.4. ConclusionWe directly measured VBM and CBM values, as well as corelevel shifts, for CdS/CIGSe GaOx/CIGSe interfaces. While forthe CdS/CIGSe interface a “cliff-like” CBO (−0.18 ± 0.07 eV) isfound, for the GaOx/CIGSe interface a small “spike-like” offsetof (0.04 ± 0.07 eV) is derived, which is seemingly the more idealelectronic interface structure in view of potentially reaching highsolar cell efficiencies. However, this was found to be in contradic-tion to device results,[18] showing that devices with standard CdSbuffers outperform their counterparts with GaOx buffer. This wastentatively being explained by the presence of significant densityof occupied states above the GaOx VBM, extending almost to EF.These states (ascribed to band tails or to oxygen vacancy relateddefects) present in the significantly oxygen deficient amorphousGaOx, with x = 1.1 (±0.1) derived by HAXPES, likely serve ascharge carrier recombination centers in the working device andmay account for the observed reduced open circuit voltage (VOC)in the respective solar cells compared to CdS/CIGSe referencedevices.Supporting InformationSupporting Information is available from the Wiley Online Library or fromthe author.AcknowledgementsOpen access funding enabled and organized by Projekt DEAL.Conflict of InterestThe authors declare no conflict of interest.Data Availability StatementThe data that support the findings of this study are available in the sup-plementary material of this article.KeywordsCdS buffer, chalcopyrite thin-film solar cell, energy level alignment, GaOxbuffer, interface propertiesReceived: December 30, 2023Revised: February 21, 2024Published online: March 8, 2024[1] M. Nakamura, K. Yamaguchi, Y. Kimoto, Y. Yasaki, T. Kato, H.Sugimoto, IEEE J. Photovoltaics 2019, 6, 1863.[2] M. A. Green, E. D. Dunlop, M. Yoshita, N. Kopidakis, K. Bothe, G.Siefer, X. Hao, Prog. Photovolt. 2023, 31, 651.Adv. Mater. Interfaces 2024, 11, 2301110 2301110 (6 of 7) © 2024 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 21967350, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202301110 by Cochrane Japan, Wiley Online Library on [07/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmatinterfaces.dewww.advancedsciencenews.com www.advmatinterfaces.de[3] M. D. Heinemann, M. F. A. M. van Hest, M. Contreras, J. D. Perkins,A. Zakutayev, C. A. Kaufmann, T. Unold, D. S. Ginley, J. J. Berry, Phys.Status Solidi A 2017, 214, 1600870.[4] T. Minemoto, Y. Hasimoto, Shams-Kolahi W., T. Satoh, T. Negami, H.Takakura, Y. Hamakawa, Sol. Energy Mater. Sol. Cells 2003, 75, 121.[5] T. Minemoto, T. Matsui, H. Takakura, Y. Hamakawa, T. Negami, Y.Hashimoto, T. Uenoyama, M. Kitagawa, Sol. Energy Mater. Sol. Cells2001, 67, 83.[6] X. Liu, J. R. Sites, AIP Conf. Proc. 1996, 353, 444.[7] F. Obereigner, N. Barreau, W. Witte, R. Scheer, J. Appl. Phys. 2015,117, 155704.[8] J. R. Tuttle, D. S. Albin, R. Noufi, Solar Cells 1991, 30, 21.[9] D. Schmid, M. Ruckh, F. Grunwald, H. W. Schock, J. Appl. Phys. 1993,73, 2902.[10] M. Morkel, L. Weinhardt, B. Lohmüller, C. Heske, E. Umbach, W.Riedl, S. Zweigart, F. Karg, Appl. Phys. Lett. 2001, 79, 4482.[11] S. Teshima, H. Kashiwabara, K. Masamoto, K. Kikunaga, K. Takeshita,T. Okuda, K. Sakurai, S. Ishizuka, A. Yamada, K. Matsubara, S. Niki,MRS Online Proceedings Library 2007, 1012, 319.[12] S. Ishizuka, J. Nishinaga, K. Beppu, T. Maeda, F. Aoyagi, T. Wada, A.Yamada, J. Chantana, T. Nishimura, T. Minemoto, M. M. Islam, T.Sakurai, N. Terada, Phys. Chem. Chem. Phys. 2022, 3, 1262.[13] F. Zang, H. Li, Y. T. Cui, G. L. Li, Q. Guo, AIP Adv. 2018, 8, 045112.[14] S. Garud, N. Gampa, T. G. Allen, R. Kotipalli, D. Flandre, M. Batuk, J.Hadermann, M. Meuris, J. Poortmans, A. Smets, B. Vermang, Phys.Status Solidi A 2018, 215, 1700826.[15] M. D. Heinemann, J. Berry, G. Teeter, T. Unold, D. Ginley, Appl. Phys.Lett. 2016, 108, 022107.[16] S. I. Stepanov, V. I. Nikolaev, V. E. Bougrov, A. E. Romanov, Adv.Mater. Sci. 2016, 44, 63.[17] J. Kim, T. Sekiya, N. Miyokawa, N. Watanabe, K. Kimoto, K. Ide, Y.Toda, S. Ueda, N. Ohashi, H. Hiramatsu, H. Hosono, T. Kamiya, NPGAsia Mater. 2017, 9, e359.[18] H. A. Yetkin, T. Kodalle, T. Bertram, Villanueva Tovar A., R. Klenk,M. Rusu, J. Ibaceta-Jaña, F. Ruske, I. Simsek, R. Muydinov, B.Szyszka, R. Schlatmann, C. A. Kaufmann, IEEE J. Photovolt. 2021, 11,648.[19] M. D. Heinemann, R. Mainz, F. Österle, H. Rodriguez-Alvarez, D.Greiner, C. A. Kaufmann, T. Unold, Sci. Rep. 2017, 7, 45463.[20] S. Ueda, Y. Katsuya, M. Tanaka, H. Yoshikawa, Y. Yamashita, S.Ishimaru, K. Kobayashi, AIP Conf. Proc. 2010, 1234, 403.[21] QUASES-IMFP-TPP2M Ver. 3.0, Inelastic electron mean free pathscalculated from the TPP-2M formula based on S. Tanuma, C. J. Pow-ell, D. R. Penn, Surf. Interf. Anal. 1994, 21, 165.[22] M. D. Heinemann, D. Greiner, T. Unold, R. Klenk, H. W. Schock, R.Schlatmann, C. A. Kaufmann, IEEE J. Photovolt. 2014, 5, 378.[23] S. Puttnins, S. Levcenco, K. Schwarzburg, G. Benndorf, F. Daume, A.Rahm, A. Braun, M. Grundmann, Sol. Energy Mater. Sol. Cells 2013,119, 281.[24] E. S. Mungan, X. Wang, M. A. Alam, IEEE J. Photovolt. 2012, 3, 451.[25] K. Horsley, S. Pookpanratana, S. Krause, T. Hofmann, M. Blum, L.Weinhardt, M. Bär, K. George, J. Van Duren, D. Jackrel, C. Heske,presented at 37th IEEE Photovoltaic Specialists Conf., Seattle, WA,USA, June, 2011.[26] N. Terada, R. T. Widodo, K. Itoh, S.-H. Kong, H. Kashiwabara, T.Okuda, K. Obara, S. Niki, K. Sakurai, A. Yamada, S. Ishizuka, ThinSolid Films 2005, 480–481, 183.[27] E. Handick, P. Reinhard, J. H. Alsmeier, L. Köhler, F. Pianezzi, S.Krause, M. Gorgoi, E. Ikenaga, N. Koch, R. G. Wilks, S. Buecheler,A. N. Tiwari, M. Bär, ACS Appl. Mater. Interfaces 2015, 7, 27414.[28] H. Mönig, C. H. Fischer, R. Caballero, C. A. Kaufmann, N. Allsop,M. Gorgoi, R. Klenk, H. W. Schock, S. Lehmann, Lux-Steiner M. C., I.Lauermann, Acta Mater. 2009, 57, 3645.[29] L. Nagarajan, R. A. De Souza, D. Samuelis, I. Valov, A. Börger, J. Janek,K. D. Becker, P. C. Schmidt, M. Martin, Nat. Mater. 2008, 7, 391.[30] T. E. Hsieh, J. Frisch, R. G. Wilks, M. Bär, ACS Appl. Mater. Interfaces2023, 15, 47725.[31] T. V. Kuznetsova, V. I. Grebennikov, H. Zhao, C. Derks, C. Taubitz, M.Neumann, C. Persson, M. V. Kuznetsov, I. V. Bodnar, R. W. Martin,M. V. Yakushev, Appl. Phys. Lett. 2012, 101, 111607.[32] S. Siebentritt, M. Igalson, C. Persson, S. Lany, Prog. Photovolt. 2010,18, 390.[33] L. Dong, R. Jia, B. Xin, B. Peng, Y. Zhang, Sci. Rep. 2017, 7, 40160.[34] Z. Hajnal, J. Miró, G. Kiss, F. Réti, P. Deák, R. C. Herndon, J. M.Kuperberg, J. Appl. Phys. 1999, 86, 3792.[35] C. Rincón, R. Márquez, J. Phys. Chem. Solids 1999, 60, 1865.[36] E. A. Kraut, R. W. Grant, J. R. Waldrop, S. P. Kowalczyk, Phys. Rev. Lett.1980, 44, 1620.[37] D. Wippler, R. G. Wilks, B. E. Pieters, van Albada S. J., D. Gerlach, J.Hüpkes, M. Bär, U. Rau, ACS Appl. Mater. Interfaces 2016, 8, 17685.[38] O. Romanyuk, A. Paszuk, I. Bartoš, R. G. Wilks, M. Nandy, J.Bombsch, C. Hartmann, R. Félix, S. Ueda, I. Gordeev, J. Houdkova, P.Kleinschmidt, P. Machek, M. Bär, P. Jǐrírček, T. Hannappel, Appl. Surf.Sci. 2021, 565, 150514.[39] O. Romanyuk, O. Supplie, A. Paszuk, J. P. Stoeckmann, R. G. Wilks, J.Bombsch, C. Hartman, Garcia-Diez R., S. Ueda, I. Bartoš, I. Gordeev,J. Houdkova, P. Kleinschmidt, M. Bär, Surf. Interface Anal. 2020, 52,933.Adv. Mater. Interfaces 2024, 11, 2301110 2301110 (7 of 7) © 2024 The Authors. Advanced Materials Interfaces published by Wiley-VCH GmbH 21967350, 2024, 13, Downloaded from https://onlinelibrary.wiley.com/doi/10.1002/admi.202301110 by Cochrane Japan, Wiley Online Library on [07/05/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons Licensehttp://www.advancedsciencenews.comhttp://www.advmatinterfaces.de The Energy Level Alignment at the Buffer/Cu(In,Ga)Se2 Thin-Film Solar Cell Interface for CdS and GaOx 1. Introduction 2. Experimental Section 2.1. Sample Preparation 2.2. Photoelectron Spectroscopy 3. Results and Discussion 4. Conclusion Supporting Information Acknowledgements Conflict of Interest Data Availability Statement Keywords