Skip to main content
NCBI home page
ACS AuthorChoice logoLink to ACS AuthorChoice
. 2023 Nov 1;15(45):53113–53121. doi: 10.1021/acsami.3c11165

Rb Diffusion and Oxide Removal at the RbF-Treated Ga2O3/Cu(In,Ga)Se2 Interface in Thin-Film Solar Cells

Elizaveta Pyatenko †,‡,*, Dirk Hauschild ‡,§,∥, Vladyslav Mikhnych ‡, Raju Edla ‡, Ralph Steininger ‡, Dimitrios Hariskos ⊥, Wolfram Witte ⊥, Michael Powalla ⊥, Clemens Heske ‡,§,∥, Lothar Weinhardt ‡,§,∥,*
PMCID: PMC10659031  PMID: 37913778

Abstract

graphic file with name am3c11165_0008.jpg

We report on the chemical structure of Cu(In,Ga)Se2 (CIGSe) thin-film solar cell absorber surfaces and their interface with a sputter-deposited Ga2O3 buffer. The CIGSe samples were exposed to a RbF postdeposition treatment and an ammonia-based rinsing step, as used in corresponding thin-film solar cells. For a detailed chemical analysis of the impact of these treatments, we employed laboratory-based X-ray photoelectron spectroscopy, X-ray-excited Auger electron spectroscopy, and synchrotron-based hard X-ray photoelectron spectroscopy. On the RbF-treated surface, we find both Rb and F, which are then partly (Rb) and completely (F) removed by the rinse. The rinse also removes Ga–F, Ga–O, and In–O surface bonds and reduces the Ga/(Ga + In) ratio at the CIGSe absorber surface. After Ga2O3 deposition, we identify the formation of In oxides and the diffusion of Rb and small amounts of F into/onto the Ga2O3 buffer layer but no indication of the formation of hydroxides.

Keywords: Cu(In,Ga)Se2 thin-film solar cells; RbF postdeposition treatment; ammonia-based rinse; photoelectron spectroscopy; HAXPES; chemical structure; surface oxides; gallium oxide buffer layer

Introduction

The development of alkali-fluoride postdeposition treatments (PDTs) has led to a significant efficiency increase for Cu(In,Ga)Se2 (CIGSe) and Cu(In,Ga)(S,Se)2 solar cells, first to 20.4% using a KF-PDT,1 then to 22.6% with RbF-PDT,2 and to 23.35% by using CsF-PDT.3 PDTs are most often followed by a rinsing step (e.g., with H2O or an NH3 solution) and/or a chemical bath deposition (CBD) step used for the subsequent deposition of a buffer layer, which removes excess alkali fluoride.4−6 In general, rinsing steps of CIGSe absorber surfaces have been employed ever since Na residue (from the soda-lime glass substrate) was observed on CIGSe surfaces,7−11 leading to a reduction or even complete removal of alkali metals12 and, in some cases, even oxides13,14 from the absorber surface.

Chemical-bath-deposited CdS (CBD-CdS) is generally used as the buffer layer in high-efficiency CIGSe-based thin-film solar cells,2,15−17 while the current world records for mini modules and lab-scale cells have already utilized sputtered Zn(O,S)18 or CBD-Zn(O,S,OH)3 buffers, respectively. Avoiding the use of CBD-CdS is motivated by reducing the environmental impact of the production process,19,20 reducing the parasitic absorption,21−23 and developing a “dry” process that does not interrupt a vacuum-based process chain.20,24

Thus, alternative buffer layer materials, including InxSy,24−28 Zn(O,S,OH),3,28,29 (Zn,Mg)O,30−32 Zn1–xSnxOy,33,34 HfOx,35 Al2O3,28,36,37 Ga2O3,23,24,38 Sn1–xGaxOy,22 (In,Ga)2O3,23 and (Al,Ga)2O3,23 have been investigated. A particularly promising candidate among the alternative buffer layers is the wide band gap transparent oxide Ga2O3, with an optical bulk band gap between 4.4 and 4.9 eV.21,24,39−43 Previously, a thin Ga2O3 layer was also applied as a passivation layer at the CdS/CIGSe interface (highest efficiencies were achieved with a 0.6 nm passivation layer). This improved the open-circuit voltage (Voc), the short-circuit current density (Jsc), and the fill factor (FF), resulting in an efficiency increase of 2.6% (absolute) as compared to the reference cells without the Ga2O3 passivation layer.21 Furthermore, amorphous (In1–xGax)2O3 buffer layers (with x ranging from 0.6 to 1) were investigated, achieving efficiencies (∼15.8%) close to those of reference cells with CdS buffers (∼17.2%) for x = 1 (i.e., pure Ga2O3).23 The highest efficiencies for these cells were achieved for buffer layer thicknesses of 60 nm.

In the first trial, ZSW used sputter-deposited gallium oxide as a buffer layer for CIGSe absorbers, reaching efficiencies up to ∼13.7% (compared to ∼17.6% with a CdS reference buffer).24 Furthermore, replacing i-ZnO in a ZnO/Al/i-ZnO/CdS/CIGSe structure by sputtered Ga2O3 resulted in cell efficiencies of 20.2% (compared to 20.4% for the reference cell).44 The absorbers were exposed to RbF-PDT as in the CdS-based ZSW process that has led to efficiencies above 22%.2 Additionally, a rinse in a 1.5 M NH3 solution to remove surplus RbF as well as a temperature optimization series for the buffer deposition were performed. Due to the wider band gap of Ga2O3, an increase in Jsc compared to reference cells with CdS was observed.24,44

In the current work, we present a detailed investigation of the chemical structure of the RbF-PDT CIGSe absorber surface, the sputter-deposited Ga2O3 buffer layer, and the Ga2O3/CIGSe interface. Particular focus is placed on the investigation of the ammonia-based rinsing step and its impact on the chemical structure of the absorber surface and the buffer/absorber interface, which is substantially more complex than a simple removal of RbF. For this purpose, samples were studied by laboratory-based X-ray photoelectron (XPS) and X-ray-excited Auger electron spectroscopy (XAES) as well as synchrotron-based hard X-ray photoelectron spectroscopy (HAXPES). Findings in this work can be correlated to electrical device parameters as reported in refs (24) and (44).

Experimental Section

The investigated samples were prepared at ZSW; the CIGSe absorbers were grown in a high-vacuum chamber by coevaporating Cu, In, Ga, and Se in an in-line multistage process onto a molybdenum/soda-lime glass substrate.45 The bulk [Ga]/([Ga] + [In]) (GGI) ratio was determined by X-ray fluorescence (XRF) measurements as 0.27, and the integral Cu content was found to be 21.3 at %.

The RbF PDT was applied in the same high-vacuum chamber without breaking the vacuum after the CIGSe process. Two different sample sets were prepared: the first set of absorbers was rinsed in 1.5 M NH3 solution for 30 s (referred to as “rinsed” in the following), while the second set was not rinsed (“non-rinsed”). The ammonia concentration for the rinsing solution, which is the same as the concentration used for the CBD CdS deposition, and the rinsing duration of 30 s were found to be sufficient for dissolving surplus RbF from the CIGSe surface.

Subsequently, Ga2O3 was deposited by radiofrequency (RF) magnetron sputtering from a ceramic target at a substrate temperature of 150 °C (details of the Ga2O3 sputter-deposition process can be found in refs (24) and (44)), generating films with thickness d of 1, 3, 10, and 100 nm. The thicknesses were estimated from sputter-deposition rates determined by optical transmittance measurements on thick Ga2O3 layers (d > 200 nm) deposited on 1 mm highly transparent quartz-glass substrates. Sister samples, processed to full solar cells (i.e., with an approximately 100 nm thick Ga2O3 buffer layer and a sputtered i-ZnO/ZnO/Al transparent front contact), showed maximum efficiencies of ∼5% for the non-rinsed and above 13% for the rinsed CIGSe absorbers, in comparison to above 16% for CdS-buffered reference devices.44 Efficiencies were measured without antireflective coating on solar cells with a total area of 0.5 cm2 with Ni/Al/Ni grids on top.

After preparation at ZSW, the samples were briefly exposed to air, sealed in a dry nitrogen atmosphere, and transported to KIT. There, the sealed samples were unpacked and mounted in an Ar-filled glovebox at the Materials for Energy (MFE) laboratory. Without any air exposure, the mounted samples were transferred directly from the glovebox into the ultrahigh vacuum (UHV) system for XPS and XAES measurements. XPS measurements were performed with a Scienta Omicron Argus CU electron analyzer, a non-monochromatized DAR 450 Mg Kα X-ray source (Scienta Omicron), and a monochromatized MECS Al Kα X-ray source (SIGMA Surface Science). The base pressure in the XPS chamber was less than 2 × 10–10 mbar. After the initial XPS and XAES measurements, the samples were transferred to the X-SPEC beamline46 at the KIT Light Source. After a brief exposure to air (less than 30 s), HAXPES spectra were measured with a Phoibos 225 electron analyzer (SPECS) and a photon energy of 2.1 keV using the Si(111) reflection of the double-crystal monochromator. The base pressure in the HAXPES analysis chamber was less than 5 × 10–10 mbar.

To calibrate the XPS/XAES measurements, the most prominent photoemission and Auger peaks of sputter-cleaned Au, Ag, and Cu foils were used.47,48 The HAXPES binding energies were calibrated using the Au 4f7/2 peak46,47 of a reference Au foil.

Results

Figure 1 shows selected HAXPES (hνexc = 2.1 keV) survey spectra of the rinsed (red) and non-rinsed (black) absorbers, as well as the corresponding 1, 3, and 100 nm Ga2O3/CIGSe interface samples. Figure S1 shows the corresponding Mg Kα XPS survey spectra. For the CIGSe absorbers, the expected core levels and XAES lines of the absorber elements (Cu, In, Ga, and Se) are detected. Also, the Rb 2p lines (Figure 1) are clearly visible (EB ∼ 1800 eV), demonstrating one strength of HAXPES to also detect such deeply bound core levels; in the case of Rb on CIGSe, uniquely identifying Rb with any other line is extremely challenging. With increasing buffer layer thickness, the Cu, Se, and In signals are increasingly attenuated by the Ga2O3 overlayer, while the Ga and O signals show a strong enhancement.

Figure 1.

Figure 1

Open in a new tab

HAXPES survey spectra of the CIGSe absorber with RbF-PDT and the 1, 3, and 100 nm Ga2O3/CIGSe samples, measured at an excitation energy of 2.1 keV. The red and black spectra correspond to the rinsed and non-rinsed sample series, respectively. Spectra were normalized to their overall integral intensity. The Ga 2p3/2 signals of the Ga2O3 buffer layers are multiplied by a factor of 0.5 for better visibility (blue dashed box). Prominent photoemission peaks and Auger signals are labeled.

Due to the higher excitation energy used for HAXPES as compared to laboratory XPS, the kinetic energy of corresponding core-level peaks increases, and thus the HAXPES measurements are less surface-sensitive than the XPS measurements. For instance, the inelastic mean free path λ for the In 3d feature is 3.0 nm for HAXPES and 1.7 nm for XPS, respectively.

In the XPS data (Figure S1), all absorber-related lines are fully attenuated for the thickest buffer layer (100 nm) samples, while the HAXPES data show very small Cu 2p, In 3d, and Se 3d signals for the rinsed and non-rinsed sample sets (enabled by the excellent signal-to-noise ratio obtainable at the X-SPEC beamline). This could be due to regions with a lower buffer layer thickness (or even a not fully closed buffer layer), a diffusion of the absorber elements into (or onto) the buffer layer, or a combination thereof. Note that a cross-section image of the Ga2O3/CIGSe interface44 suggests that regions with lower buffer layer thickness are unlikely but cannot be entirely excluded for this sample series. In both absorber spectra, small C and O 1s signals are visible. For the non-rinsed absorber, strong Rb-, F-, and Na-related signals, the latter likely due to diffusion from the soda-lime glass,9 are also detected.

A comparison of the non-rinsed and rinsed CIGSe absorbers (Figure 1) shows that the CIGSe-related lines (e.g., Cu 2p, In 3d, and Se 3d) increase after the ammonia rinse by a factor of ∼1.3. In parallel, we find a strong decrease in the intensities of the Rb 2p, F 1s, Na 1s, and O 1s lines, while the C 1s intensity increases (by a factor of 1.2). O 1s and C 1s detailed spectra are shown in Figure S2, and Na 1s spectra are presented in Figure S3. We interpret these spectral changes as follows: the rinse removes excess material of RbF-PDT (i.e., Rb and F), Na, and surface oxides (as will be discussed below) from the absorber surface. The absorber-related signals (as well as tightly bound CIGSe surface species, e.g., carbon) then increase in intensity due to the reduced attenuation by this surface layer. The larger increase in the C 1s signal for the rinsed absorber may also be due to a more “reactive” surface after the rinse that is more susceptible to the adsorption of carbon-containing species during the subsequent sample handling.

To study the differences in the chemical environment at the surfaces of the non-rinsed and rinsed absorbers, we discuss the Ga 2p3/2, In 3d5/2, Rb 2p3/2, and Se 3d spectral regions in Figure 2.

Figure 2.

Figure 2

Open in a new tab

HAXPES (hν = 2.1 keV) and monochromatized Al Kα XPS spectra of the (a) Ga 2p3/2, (b) In 3d5/2, (c) Rb 2p3/2, and (d) Se 3d regions. The black and red lines correspond to the non-rinsed and rinsed CIGSe absorbers, respectively. All spectra were normalized by the same factors as the survey spectra in Figure 1. In addition, the given multiplication factors were applied to the red spectra (rinsed) in the figure in order to maximize the contribution of the respective component. The blue lines then show the difference spectra between “non-rinsed” and “rinsed” surfaces. Colored bars show ranges of literature values of the binding energies for different compounds.4,49,50

In Figure 2a, prior to the rinse, we find a spectral component indicating Ga–F bonds (the blue box shows literature values for Ga in GaCl, GaBr3, and GaI3).50 After the rinse, this component disappears, as indicated by the difference spectrum (blue). This is in accordance with the F signal, which vanishes after the rinse (as discussed later in this paper). The Ga–F environment is much stronger in the more surface-sensitive XPS measurement, and the overall Ga 2p3/2 signal is reduced after the rinse, which we explain as follows: the RbF-PDT forms Ga–F bonds with Ga at the surface, and the rinse washes away F and (some of the) Ga, reducing the GGI ratio at the surface. This finding is supported by inductively coupled plasma–mass spectrometry, where Ga was found in the rinsing residue.25 This change in the Ga content at the surface might impact (reduce) the surface band gap and the performance of the cells.51

In Figure 2b, we see no indication of In–F bonds. However, the XPS spectrum shows a weak shoulder at ∼445.5 eV that is removed after the rinse, likely due to the reduction in In–O bonds. This observation will be discussed in conjunction with Figure 3.

Figure 3.

Figure 3

Open in a new tab

Monochromatized Al Kα XPS spectra of the In 4d/Ga 3d region. The black and red lines correspond to the non-rinsed and rinsed CIGSe absorbers, respectively. The spectra were normalized to their overall maximum. The blue line shows the difference spectrum between “non-rinsed” and “rinsed”. Colored bars show ranges of literature binding energies for different compounds.49,50

There are additional shoulders at low binding energies for the HAXPES Ga 2p3/2 and In 3d5/2 spectra of the non-rinsed samples in Figure 2a,b, respectively. In the case of Ga 2p3/2 spectra, this additional component can be assigned to the Na KL2,3L2,3 Auger transition at ∼1112 eV,49 which is removed after the rinse. The low-binding energy shoulder of the In 3d5/2 spectra at ∼443.8 eV might be attributed to In in the Rb–In–Se environment present at the surface before the rinse, as we will discuss below.

Figure 2c shows that the Rb signal decreases by more than a factor of 5 after the rinse. In addition, the difference spectrum clearly highlights that there is an additional spectral component in the non-rinsed absorber at higher binding energies. We assign it to an Rb–F environment, which, similarly to Ga–F, is removed by the rinse as well. The main spectral component can be attributed to Rb at the CIGSe surface,52,53 while the Se 3d spectra in Figure 2d indicate only a small contribution from a Rb–In–Se environment before the rinse, indicated by a weak additional spectral component at low binding energies. The binding energy corresponds to that of an alkali–In–Se environment, as reported in ref (4). We observe a narrowing of the line after the rinse, which may indicate that there are several slightly different chemical environments present on the non-rinsed sample. In ref (4), a shift of ∼0.5 eV was observed for the In 3d5/2 peak between In in a CIGSe and In in an alkali–In–Se environment. The weak low-binding energy shoulder (443.8 eV) of the In 3d5/2 spectrum in Figure 2b might be attributed to this chemical environment. In summary, a comparison between non-rinsed and rinsed absorber surfaces suggests the presence of a Rb–In–Se environment,4,52 which is subsequently reduced (if not even fully removed) by the here-applied rinse.

Figure 3 shows the In 4d/Ga 3d spectral region. Corroborating our findings above, we observe a second Ga 3d component at ∼21.5 eV for the non-rinsed CIGSe. Again, based on literature values for other Ga-halogenides (GaCl3, GaBr3, and GaI3),50 we assign this peak to Ga–F bonds at the surface. This component is not present for the rinsed sample, in agreement with the removal of F discussed above from the surface. The In 4d spectrum in Figure 3 exhibits a clear intensity reduction at ∼18 eV for the rinsed absorber as compared to the non-rinsed absorber. The blue difference spectrum (“non-rinsed” – “rinsed”) highlights this reduction, in agreement with the intensity reduction on the high-binding-energy side of the In 3d5/2 XPS spectrum in Figure 2b. We attribute these additional spectral components to the In–O bonds. Figure 2b shows that the overall In 3d5/2 intensity increases significantly after the rinse. In contrast, the overall Ga 2p3/2 intensity decreases, leading to a decrease in the surface GGI ratio as a result of the rinse.

To gain insights into the chemical composition of the Ga2O3 buffer layer and the formation of the Ga2O3/CIGSe interface, the XAES spectra of indium (M4,5N4,5N4,5, Figure 4) and gallium (L3M4,5M4,5, Figure 5) were investigated.

Figure 4.

Figure 4

Open in a new tab

Mg Kα-excited In M4,5N4,5N4,5 for the non-rinsed (left) and rinsed CIGSe (right) and Ga2O3/CIGSe samples with Ga2O3 thicknesses of 1 and 3 nm. Data points are represented as open black circles; individual species (fit components) are represented in green (In–Se bonds) and purple (In–O bonds), and the sum is presented in red. A linear background is shown in gray. Below each spectrum is given the residual. Literature values for the prominent M4N5N5 feature of different compounds49 are marked as colored bars.

Figure 5.

Figure 5

Open in a new tab

Mg Kα-excited L3M4,5M4,5 XAES spectra of gallium for the non-rinsed and rinsed CIGSe and Ga2O3/CIGSe samples with increasing Ga2O3 thickness. Data are represented as open black circles, individual species (fit components) are represented in green (selenide), purple (gallium oxide), and blue (gallium fluoride), and the sum is represented in red. Below each spectrum, the residual is shown. Literature values for the prominent L3M4,5M4,5, feature of different compounds49,50 are marked as colored bars.

While the two In MNN (Figure 4) CIGSe absorber spectra look similar to published spectra,14,47,54,55 the 1 and 3 nm thick Ga2O3 spectra show additional intensity in the “valley” at 405 eV, and the overall spectral shape is broader. To analyze these spectral changes, all In MNN spectra were fitted by using two single-species In MNN spectra as fit functions (derived from an untreated ZSW absorber sample from another batch). All In MNN fits are thus composed of two components: one with In M4N4,5N4,5 at ∼407.7 eV, attributed to In in a CIGSe environment, and the other at 405–405.5 eV, attributed to In in an oxide environment. For the 1 and 3 nm Ga2O3/CIGSe samples, an additional Gaussian broadening was applied to the oxide component to describe the presence of several slightly varying chemical environments. The oxide component is weak in the non-rinsed absorber and further decreases by about four times in the rinsed absorber. For both the rinsed and non-rinsed 1 and 3 nm Ga2O3/CIGSe samples, a relative increase in the oxide component is observed, which dominates the 3 nm spectra. This finding indicates a significant influence of the sputter deposition of the Ga2O3 layer on the absorber surface, in particular, intermixing of absorber elements and the formation of In–O bonds.

In a similar fashion, the Ga L3M4,5M4,5 region is analyzed in Figure 5. A single-species Ga L3M4,5M4,5 spectrum (derived from the same untreated ZSW absorber sample mentioned above) was used as the fit function to describe our Ga L3M4,5M4,5 absorber spectra, while for the samples with a buffer layer, the measured Ga L3M4,5M4,5 spectrum of the 100 nm rinsed Ga2O3/CIGSe sample was used. This approach is necessary due to a significant broadening and change in the spectral shape of the Ga L3M4,5M4,5 spectra for various compounds.50 While the rinsed absorber is composed of one gallium component (main peak at 1065.5 eV), attributed to Ga in a CIGSe environment, the non-rinsed absorber exhibits two additional components. The second component at 1062 eV can be attributed to Ga in an oxide environment, and the third component at 1059 eV can be attributed to Ga in a fluoride environment. This interpretation is in agreement with the results shown in Figure 2a for the Ga 2p3/2 region.

The Ga LMM spectra of all of the Ga2O3/CIGSe samples are very similar. Only a very small absorber component is present for the 1 nm Ga2O3/CIGSe samples, in addition to the main component attributed to Ga in Ga2O3. The Ga in the CIGSe environment is not visible for thicker buffer layers. To determine the chemical environment of Ga, the modified Auger parameters (α’) for gallium and oxygen were calculated by adding the binding energies of the most prominent photoemission peaks (Ga 2p3/2 and O 1s) to the kinetic energies of the most prominent Auger peaks (Ga LMM and O KVV, respectively), where α’ = Ebin (photoemission) + Ekin (Auger). They are 2180.6 ± 0.2 eV for gallium and 1040.4 ± 0.2 eV for oxygen, respectively, independent of the buffer layer thickness. These values are close to the ones reported in the literature for gallium (2180.1–2180.4 eV) and oxygen (1040.7 eV) in Ga2O3,49 suggesting that the dominant species in the buffer layer is indeed Ga2O3, independent of the rinse, and that the chemical environment does not change with increasing thicknesses. Notably, we do not find any indication of the formation of gallium hydroxides, which would be indicated by an additional component in the O 1s spectra (Figure S2).

Now, we turn to the RbF-PDT-related elements and their evolution as a function of the Ga2O3 buffer layer thickness. First, Rb is investigated, specifically the Rb 2p3/2 peak with a binding energy of ∼1804 eV in our HAXPES spectra. A clear advantage of measuring this region is that there are no overlapping peaks, in contrast to the Rb 3d/Ga 3p region (as measured with XPS).

Figure 6a shows the fit analysis of the Rb 2p3/2 peaks for the non-rinsed and rinsed samples with the Ga2O3 buffer layer, while Figure 6b shows the Rb 2p3/2 peak area as a function of the nominal buffer layer thickness.

Figure 6.

Figure 6

Open in a new tab

(a) Data and fits of the Rb 2p3/2 spectra of the non-rinsed (left) and rinsed (right) CIGSe samples with the Ga2O3 buffer layer, measured with HAXPES at an excitation energy of 2.1 keV. Data are represented as open black circles, individual species (fit components) are represented in green (Rb–F) and blue (Rb at the surface), and the sum is represented in red. Below each spectrum, the magnified (×3 or ×5) residual is colored in light gray. (b) Rb 2p3/2 peak area as a function of the nominal buffer layer thickness, normalized to the intensity of the non-rinsed absorber surface. Data for the non-rinsed and rinsed samples are shown in black and red, respectively.

Figure 6a demonstrates that a Rb 2p3/2 signal is visible on all sample surfaces. Note that these measurements are very surface-sensitive: at this excitation energy, the inelastic mean free path of the Rb 2p3/2 electrons is ∼0.8 nm; in contrast, it is ∼2.3 nm for the Rb 3d region. Combined with the complications arising from the spectral overlap with the Ga 3p lines, it is much harder to unequivocally discern the presence of Rb on the Ga2O3 buffer layer surfaces from the Rb 3d spectra. In the non-rinsed samples, Rb is mainly present in two different environments, as discussed in Figure 2c: Rb adsorbed on CIGSe and Rb–F, while in the rinsed samples, only adsorbed Rb is present.

Figure 6b shows an overall decrease in the area under the (total) Rb 2p3/2 peak with increasing buffer layer thickness. This decrease is much weaker than would be expected for (exponential) attenuation by the Ga2O3 overlayer. This discrepancy indicates a diffusion/segregation of Rb in both the rinsed and non-rinsed cases. In the rinsed case, the Rb intensity slowly but steadily decreases with buffer layer thickness. In contrast, the Rb 2p intensity in the non-rinsed series decreases sharply from the absorber to the 1 nm Ga2O3 sample, then stays approximately constant up to 10 nm Ga2O3, and then further strongly decreases from 10 to 100 nm. In addition, Figure 6a shows that the relative fraction of the Rb–F component increases with increasing buffer layer thickness in the non-rinsed sample series.

To further investigate the Rb–F diffusion/ segregation, we now turn to the fluorine signals. Figure 7 shows the In 3p and F 1s spectral regions, measured with HAXPES. Although it is possible to observe the F 1s peak with XPS, the high signal-to-noise ratio achievable with HAXPES at the X-SPEC beamline helps to clearly detect the small F 1s peak between the (strong) In 3p peaks in the rinsed sample series.

Figure 7.

Figure 7

Open in a new tab

In 3p and F 1s regions for the non-rinsed (left) and rinsed (right) CIGSe and Ga2O3/CIGSe samples with increasing Ga2O3 thickness, measured with HAXPES at an excitation energy of 2.1 keV. To highlight the F 1s signal in the rinsed sample series, it is magnified by a factor of 5 and shown in blue above each spectrum.

In the non-rinsed samples, the F 1s peak decreases with increasing buffer layer thicknesses. There is still a clear F 1s peak for the 100 nm non-rinsed sample but no In 3p peaks, indicating a clear diffusion/segregation of F to the Ga2O3 buffer layer surface. After the rinse, no (or only a very small) F 1s peak is found, indicating that the rinse removes F from the absorber surface. In contrast, the rinsed Ga2O3/CIGSe samples all show a small F 1s peak, which slightly decreases in intensity as a function of the buffer layer thickness. As a possible explanation for the presence of F on the rinsed Ga2O3/CIGSe samples, some F might diffuse into the CIGSe absorber (e.g., along grain boundaries) during RbF-PDT and then diffuse to the Ga2O3/CIGSe interface during the sputter-deposition process (assisted by the elevated process temperature).

We speculate that the large amounts of Rb–F on the non-rinsed absorber and its diffusion into the Ga2O3 buffer layer could lead to the observed significantly lower solar-cell efficiencies. Rinsing the CIGSe absorber removes F, some Rb, and associated oxides and could hence result in optimized interface properties and higher efficiencies.

Conclusions

We have presented a detailed investigation of the chemical structure of the RbF-treated CIGSe absorber surface, with and without an ammonia-based rinse, as well as their interfaces with a sputter-deposited Ga2O3 buffer layer. The rinse removes almost all F and most of the Rb in an Rb–F environment, while some remains at the CIGSe surface. The rinse removes In–O, Ga–O, and Ga–F bonds from the CIGSe surface, decreases the Ga/(Ga + In) ratio at the surface, and removes evidence of a Rb–In–Se bonding environment.

Rb and F are also found on all samples with a sputter-deposited Ga2O3 buffer layer, with and without rinse (in some cases, only trace amounts of F are observed). During sputter-deposition, a significant amount of In–O bonds is formed. The dominating chemical environment of the buffer layer is Ga2O3, independent of the buffer layer thickness and absorber rinsing.

Our findings thus indicate a rather complex chemical interface structure and diffusion/segregation behavior from the absorber to the buffer surface. This interface structure is substantially modified by applying a rinse, which then also leads to higher solar cell efficiencies. These significant changes in the chemical structure are expected to lead to changes in the electronic structure, in particular, the conduction band alignment, which are the subject of future investigations.

Acknowledgments

The authors are grateful for financial support by the German Federal Ministry for Economic Affairs and Climate Action (BMWK) in the “EFFCIS-II” project (no. 03EE1059A and 03EE1059E). D.H., L.W., and C.H. thank the Deutsche Forschungsgemeinschaft (DFG) for funding in projects GZ:INST 121384/64-1 FUGG, GZ:INST 121384/65-1 FUGG, and GZ:INST 121384/66-1.

Data Availability Statement

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Supporting Information Available

The Supporting Information is available free of charge at https://pubs.acs.org/doi/10.1021/acsami.3c11165.

  • XPS survey spectra and O 1s, C 1s, and Na 1s XPS detailed spectra for the Ga2O3/CIGSe samples (PDF)

Author Present Address

# Kvitky-Osnovyanenko 3, 61003 Kharkiv, Ukraine

The authors declare no competing financial interest.

Supplementary Material

am3c11165_si_001.pdf (181.5KB, pdf)

References

  1. Chirilă A.; Reinhard P.; Pianezzi F.; Bloesch P.; Uhl A. R.; Fella C.; Kranz L.; Keller D.; Gretener C.; Hagendorfer H.; Jaeger D.; Erni R.; Nishiwaki S.; Buecheler S.; Tiwari A. N. Potassium-Induced Surface Modification of Cu(In,Ga)Se2 Thin Films for High-Efficiency Solar Cells. Nat. Mater. 2013, 12 (12), 1107–1111. 10.1038/nmat3789. [DOI] [PubMed] [Google Scholar]
  2. Jackson P.; Wuerz R.; Hariskos D.; Lotter E.; Witte W.; Powalla M. Effects of Heavy Alkali Elements in Cu(In,Ga)Se2 Solar Cells with Efficiencies up to 22.6%. Phys. Status Solidi RRL 2016, 10 (8), 583–586. 10.1002/pssr.201600199. [DOI] [Google Scholar]
  3. Nakamura M.; Yamaguchi K.; Kimoto Y.; Yasaki Y.; Kato T.; Sugimoto H. Cd-Free Cu(In,Ga)(Se,S)2 Thin-Film Solar Cell With Record Efficiency of 23.35%. IEEE J. Photovolt. 2019, 9 (6), 1863–1867. 10.1109/JPHOTOV.2019.2937218. [DOI] [Google Scholar]
  4. Handick E.; Reinhard P.; Wilks R. G.; Pianezzi F.; Kunze T.; Kreikemeyer-Lorenzo D.; Weinhardt L.; Blum M.; Yang W.; Gorgoi M.; Ikenaga E.; Gerlach D.; Ueda S.; Yamashita Y.; Chikyow T.; Heske C.; Buecheler S.; Tiwari A. N.; Bär M. Formation of a K—In—Se Surface Species by NaF/KF Postdeposition Treatment of Cu(In,Ga)Se2 Thin-Film Solar Cell Absorbers. ACS Appl. Mater. Interfaces 2017, 9 (4), 3581–3589. 10.1021/acsami.6b11892. [DOI] [PubMed] [Google Scholar]
  5. Keller J.; Aboulfadl H.; Stolt L.; Donzel-Gargand O.; Edoff M. Rubidium Fluoride Absorber Treatment for Wide-Gap (Ag,Cu)(In,Ga)Se2 Solar Cells. Sol. RRL 2022, 6 (6), 2200044. 10.1002/solr.202200044. [DOI] [Google Scholar]
  6. Reinhard P.; Bissig B.; Pianezzi F.; Hagendorfer H.; Sozzi G.; Menozzi R.; Gretener C.; Nishiwaki S.; Buecheler S.; Tiwari A. N. Alkali-Templated Surface Nanopatterning of Chalcogenide Thin Films: A Novel Approach Toward Solar Cells with Enhanced Efficiency. Nano Lett. 2015, 15 (5), 3334–3340. 10.1021/acs.nanolett.5b00584. [DOI] [PubMed] [Google Scholar]
  7. Hedström J.; Ohlsen H.; Bodegard M.; Kylner A.; Stolt L.; Hariskos D.; Ruckh M.; Schock H.-W.. ZnO/CdS/Cu(In,Ga)Se2 Thin Film Solar Cells with Improved Performance. Conference Record of the Twenty Third IEEE Photovoltaic Specialists Conference—1993 (Cat. No.93CH3283-9), 1993; pp 364–371.
  8. Scofield J. H.; Asher S.; Albin D.; Tuttle J.; Contreras M.; Niles D.; Reedy R.; Tennant A.; Noufi R.. Sodium Diffusion, Selenization, and Microstructural Effects Associated with Various Molybdenum Back Contact Layers for CIS-Based Solar Cells. Proceedings of 1994 IEEE 1st World Conference on Photovoltaic Energy Conversion—WCPEC (A Joint Conference of PVSC, PVSEC and PSEC); IEEE: Waikoloa, HI, USA, 1994; Vol. 1, pp 164–167.
  9. Heske C.; Fink R.; Umbach E.; Riedl W.; Karg F. Na-induced Effects on the Electronic Structure and Composition of Cu(In,Ga)Se2 Thin-film Surfaces. Appl. Phys. Lett. 1996, 68 (24), 3431–3433. 10.1063/1.115783. [DOI] [Google Scholar]
  10. Niles D. W.; Al-Jassim M.; Ramanathan K. Direct Observation of Na and O Impurities at Grain Surfaces of CuInSe2 Thin Films. J. Vac. Sci. Technol. Vac. Surf. Films 1999, 17 (1), 291–296. 10.1116/1.581583. [DOI] [Google Scholar]
  11. Rockett A.; Granath K.; Asher S.; Al Jassim M. M.; Hasoon F.; Matson R.; Basol B.; Kapur V.; Britt J. S.; Gillespie T.; Marshall C. Na Incorporation in Mo and CuInSe2 from Production Processes. Sol. Energy Mater. Sol. Cells 1999, 59 (3), 255–264. 10.1016/S0927-0248(99)00026-4. [DOI] [Google Scholar]
  12. Erfurth F.Elektronenspektroskopie an Cd-Freien Pufferschichten Und Deren Grenzflächen in Cu(In,Ga)(S,Se)2. Doctoral Thesis, Universität Würzburg, 2010. [Google Scholar]
  13. Kylner A. The Chemical Bath Deposited CdS/Cu(In,Ga)Se2 Interface as Revealed by X-Ray Photoelectron Spectroscopy. J. Electrochem. Soc. 1999, 146 (5), 1816–1823. 10.1149/1.1391849. [DOI] [Google Scholar]
  14. Weinhardt L.; Fuchs O.; Groß D.; Umbach E.; Heske C.; Dhere N. G.; Kadam A. A.; Kulkarni S. S. Surface Modifications of Cu(In,Ga)S2 Thin Film Solar Cell Absorbers by KCN and H2O2/H2SO4 Treatments. J. Appl. Phys. 2006, 100 (2), 024907. 10.1063/1.2216367. [DOI] [Google Scholar]
  15. Jackson P.; Hariskos D.; Wuerz R.; Kiowski O.; Bauer A.; Friedlmeier T. M.; Powalla M. Properties of Cu(In,Ga)Se2 Solar Cells with New Record Efficiencies up to 21.7%. Phys. Status Solidi RRL 2015, 9 (1), 28–31. 10.1002/pssr.201409520. [DOI] [Google Scholar]
  16. Ramanathan K.; Contreras M. A.; Perkins C. L.; Asher S.; Hasoon F. S.; Keane J.; Young D.; Romero M.; Metzger W.; Noufi R.; Ward J.; Duda A. Properties of 19.2% Efficiency ZnO/CdS/Cu(In,Ga)Se2 Thin-Film Solar Cells. Prog. Photovolt. Res. Appl. 2003, 11 (4), 225–230. 10.1002/pip.494. [DOI] [Google Scholar]
  17. Kessler J.; Wennerberg J.; Bodegård M.; Stolt L. Highly Efficient Cu(In,Ga)Se2 Mini-Modules. Sol. Energy Mater. Sol. Cells 2003, 75 (1–2), 35–46. 10.1016/S0927-0248(02)00102-2. [DOI] [Google Scholar]
  18. New world record for CIGS efficiency. Avancis. https://www.avancis.de/en/magazine/pr-efficiency (accessed 11 03, 2022).
  19. Hynes K. M.; Newham J.. A Comparison of Window/Buffer Layer Materials for CdTe Thin Film Modules Using Environmental Risk Assessment. Conference Record of the Twenty-Eighth IEEE Photovoltaic Specialists Conference—2000 (Cat. No.00CH37036); 2000; pp 1513–1516.
  20. Hariskos D.; Spiering S.; Powalla M. Buffer Layers in Cu(In,Ga)Se2 Solar Cells and Modules. Thin Solid Films 2005, 480–481, 99–109. 10.1016/j.tsf.2004.11.118. [DOI] [Google Scholar]
  21. Garud S.; Gampa N.; Allen T. G.; Kotipalli R.; Flandre D.; Batuk M.; Hadermann J.; Meuris M.; Poortmans J.; Smets A.; Vermang B. Surface Passivation of CIGS Solar Cells Using Gallium Oxide. Phys. Status Solidi A 2018, 215 (7), 1700826. 10.1002/pssa.201700826. [DOI] [Google Scholar]
  22. Larsson F.; Keller J.; Olsson J.; Donzel-Gargand O.; Martin N. M.; Edoff M.; Törndahl T. Amorphous Tin-Gallium Oxide Buffer Layers in (Ag,Cu)(In,Ga)Se2 Solar Cells. Sol. Energy Mater. Sol. Cells 2020, 215, 110647. 10.1016/j.solmat.2020.110647. [DOI] [Google Scholar]
  23. Koida T.; Kamikawa-Shimizu Y.; Yamada A.; Shibata H.; Niki S. Cu(In,Ga)Se2 Solar Cells With Amorphous Oxide Semiconducting Buffer Layers. IEEE J. Photovolt. 2015, 5 (3), 956–961. 10.1109/JPHOTOV.2015.2396356. [DOI] [Google Scholar]
  24. Witte W.; Paetel S.; Menner R.; Bauer A.; Hariskos D. The Application of Sputtered Gallium Oxide as Buffer for Cu(In,Ga)Se2 Solar Cells. Phys. Status Solidi RRL 2021, 15 (9), 2100180. 10.1002/pssr.202100180. [DOI] [Google Scholar]
  25. Hariskos D.; Hempel W.; Menner R.; Witte W. Influence of Substrate Temperature during InxSy Sputtering on Cu(In,Ga)Se2/Buffer Interface Properties and Solar Cell Performance. Appl. Sci. 2020, 10 (3), 1052. 10.3390/app10031052. [DOI] [Google Scholar]
  26. Soni P.; Raghuwanshi M.; Wuerz R.; Berghoff B.; Knoch J.; Raabe D.; Cojocaru-Mirédin O. Sputtering as a Viable Route for In2S3 Buffer Layer Deposition in High Efficiency Cu(In,Ga)Se2 Solar Cells. Energy Sci. Eng. 2019, 7 (2), 478–487. 10.1002/ese3.295. [DOI] [Google Scholar]
  27. Hauschild D.; Mezher M.; Schnabel T.; Spiering S.; Kogler W.; Carter J.; Blum M.; Yang W.; Ahlswede E.; Heske C.; Weinhardt L. Intermixing at the InxSy/Cu2ZnSn(S,Se)4 Heterojunction and Its Impact on the Chemical and Electronic Interface Structure. ACS Appl. Energy Mater. 2019, 2 (6), 4098–4104. 10.1021/acsaem.9b00263. [DOI] [Google Scholar]
  28. Yousfi E. B.; Asikainen T.; Pietu V.; Cowache P.; Powalla M.; Lincot D. Cadmium-Free Buffer Layers Deposited by Atomic Later Epitaxy for Copper Indium Diselenide Solar Cells. Thin Solid Films 2000, 361–362, 183–186. 10.1016/S0040-6090(99)00860-3. [DOI] [Google Scholar]
  29. Mezher M.; Garris R.; Mansfield L. M.; Blum M.; Hauschild D.; Horsley K.; Duncan D. A.; Yang W.; Bär M.; Weinhardt L.; Ramanathan K.; Heske C. Soft X-Ray Spectroscopy of a Complex Heterojunction in High-Efficiency Thin-Film Photovoltaics: Intermixing and Zn Speciation at the Zn(O,S)/Cu(In,Ga)Se2 Interface. ACS Appl. Mater. Interfaces 2016, 8 (48), 33256–33263. 10.1021/acsami.6b09245. [DOI] [PubMed] [Google Scholar]
  30. Minemoto T.; Hashimoto Y.; Satoh T.; Negami T.; Takakura H.; Hamakawa Y. Cu(In,Ga)Se2 Solar Cells with Controlled Conduction Band Offset of Window/Cu(In,Ga)Se2 Layers. J. Appl. Phys. 2001, 89 (12), 8327–8330. 10.1063/1.1366655. [DOI] [Google Scholar]
  31. Nishimura T.; Chantana J.; Kawano Y.; Yamada A.; Kimoto Y.; Kato T.; Sugimoto H.; Minemoto T. Interfacial Modification Mechanism by Aging Effect for High-Performance Cd-Free and All-Dry Process Cu(In,Ga)(S,Se)2 Solar Cells. Appl. Phys. Lett. 2020, 117 (22), 223501. 10.1063/5.0031241. [DOI] [Google Scholar]
  32. Erfurth F.; Hußmann B.; Schöll A.; Reinert F.; Grimm A.; Lauermann I.; Bär M.; Niesen Th.; Palm J.; Visbeck S.; Weinhardt L.; Umbach E. Chemical Structure of the (Zn1-xMgx)O/CuIn(S,Se)2 Interface in Thin Film Solar Cells. Appl. Phys. Lett. 2009, 95 (12), 122104. 10.1063/1.3230071. [DOI] [Google Scholar]
  33. Salomé P.; Keller J.; Törndahl T.; Teixeira J. P.; Nicoara N.; Andrade R.-R.; Stroppa D. G.; González J.; Edoff M.; Leitão J.; Sadewasser S. CdS and Zn1-x SnxOy Buffer Layers for CIGS Solar Cells. Sol. Energy Mater. Sol. Cells 2017, 159, 272–281. 10.1016/j.solmat.2016.09.023. [DOI] [Google Scholar]
  34. Lindahl J.; Wätjen J. T.; Hultqvist A.; Ericson T.; Edoff M.; Törndahl T. The Effect of Zn1-x Snx Oy Buffer Layer Thickness in 18.0% Efficient Cd-Free Cu(In,Ga)Se2 Solar Cells. Prog. Photovolt. Res. Appl. 2013, 21 (8), 1588–1597. 10.1002/pip.2239. [DOI] [Google Scholar]
  35. Löckinger J.; Nishiwaki S.; Bissig B.; Degutis G.; Romanyuk Y. E.; Buecheler S.; Tiwari A. N. The Use of HfO2 in a Point Contact Concept for Front Interface Passivation of Cu(In,Ga)Se2 Solar Cells. Sol. Energy Mater. Sol. Cells 2019, 195, 213–219. 10.1016/j.solmat.2019.03.009. [DOI] [Google Scholar]
  36. Kotipalli R.; Vermang B.; Joel J.; Rajkumar R.; Edoff M.; Flandre D. Investigating the Electronic Properties of Al2O3/Cu(In,Ga)Se2 Interface. AIP Adv. 2015, 5 (10), 107101. 10.1063/1.4932512. [DOI] [Google Scholar]
  37. Werner F.; Veith-Wolf B.; Spindler C.; Barget M. R.; Babbe F.; Guillot J.; Schmidt J.; Siebentritt S. Oxidation as Key Mechanism for Efficient Interface Passivation in Cu(In,Ga)Se2 Thin-Film Solar Cells. Phys. Rev. Appl. 2020, 13 (5), 054004. 10.1103/PhysRevApplied.13.054004. [DOI] [Google Scholar]
  38. Heinemann M. D.; van Hest M. F. a. M.; Contreras M.; Perkins J. D.; Zakutayev A.; Kaufmann C. A.; Unold T.; Ginley D. S.; Berry J. J. Amorphous Oxides as Electron Transport Layers in Cu(In,Ga)Se2 Superstrate Devices. Phys. Status Solidi A 2017, 214 (5), 1600870. 10.1002/pssa.201600870. [DOI] [Google Scholar]
  39. Gallium Oxide: Materials Properties, Crystal Growth, and Devices; Springer Series in Materials Science; Higashiwaki M., Fujita S., Eds.; Springer International Publishing: Cham, 2020; Vol. 293. [Google Scholar]
  40. Onuma T.; Saito S.; Sasaki K.; Masui T.; Yamaguchi T.; Honda T.; Higashiwaki M. Valence Band Ordering in β-Ga2O3 Studied by Polarized Transmittance and Reflectance Spectroscopy. Jpn. J. Appl. Phys. 2015, 54 (11), 112601. 10.7567/JJAP.54.112601. [DOI] [Google Scholar]
  41. Tippins H. H. Optical Absorption and Photoconductivity in the Band Edge of β-Ga2O3. Phys. Rev. 1965, 140 (1A), A316–A319. 10.1103/PhysRev.140.A316. [DOI] [Google Scholar]
  42. Orita M.; Ohta H.; Hirano M.; Hosono H. Deep-Ultraviolet Transparent Conductive β-Ga2O3 Thin Films. Appl. Phys. Lett. 2000, 77 (25), 4166–4168. 10.1063/1.1330559. [DOI] [Google Scholar]
  43. He H.; Orlando R.; Blanco M. A.; Pandey R.; Amzallag E.; Baraille I.; Rérat M. First-Principles Study of the Structural, Electronic, and Optical Properties of Ga2O3 in Its Monoclinic and Hexagonal Phases. Phys. Rev. B: Condens. Matter Mater. Phys. 2006, 74 (19), 195123. 10.1103/PhysRevB.74.195123. [DOI] [Google Scholar]
  44. Witte W.; Hempel W.; Paetel S.; Menner R.; Hariskos D. Influence of Sputtered Gallium Oxide as Buffer or High-Resistive Layer on Performance of Cu(In,Ga)Se2-Based Solar Cells. J. Mater. Res. 2022, 37 (11), 1825–1834. 10.1557/s43578-022-00608-z. [DOI] [Google Scholar]
  45. Gutzler R.; Witte W.; Kanevce A.; Hariskos D.; Paetel S. VOC-Losses across the Band Gap: Insights from a High-Throughput Inline Process for CIGS Solar Cells. Prog. Photovolt. Res. Appl. 2023, 31 (10), 1023–1031. 10.1002/pip.3707. [DOI] [Google Scholar]
  46. Weinhardt L.; Steininger R.; Kreikemeyer-Lorenzo D.; Mangold S.; Hauschild D.; Batchelor D.; Spangenberg T.; Heske C. X-SPEC: A 70 eV to 15 keV Undulator Beamline for X-Ray and Electron Spectroscopies. J. Synchrotron Radiat. 2021, 28 (2), 609–617. 10.1107/S1600577520016318. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Moulder J. F.; Stickle W. F.; Sobol P. E.; Bomben K. D.. Handbook of X-Ray Photoelectron Spectroscopy: A Reference Book of Standard Spectra for Identification and Interpretation of XPS Data; Perkin Elmer Cooperation: Minnesota, USA, 1992. [Google Scholar]
  48. Seah M. P. Post-1989 Calibration Energies for X-Ray Photoelectron Spectrometers and the 1990 Josephson Constant. Surf. Interface Anal. 1989, 14 (8), 488. 10.1002/sia.740140813. [DOI] [Google Scholar]
  49. NIST X-ray Photoelectron Spectroscopy (XPS) Database Main Search Menu. https://srdata.nist.gov/xps/main_search_menu.aspx (accessed 04 07, 2022).
  50. Bourque J. L.; Biesinger M. C.; Baines K. M. Chemical State Determination of Molecular Gallium Compounds Using XPS. Dalton Trans. 2016, 45 (18), 7678–7696. 10.1039/C6DT00771F. [DOI] [PubMed] [Google Scholar]
  51. Bär M.; Repins I.; Contreras M. A.; Weinhardt L.; Noufi R.; Heske C. Chemical and Electronic Surface Structure of 20%-Efficient Cu(In,Ga)SeSe2 Thin Film Solar Cell Absorbers. Appl. Phys. Lett. 2009, 95 (5), 052106. 10.1063/1.3194153. [DOI] [Google Scholar]
  52. Bombsch J.; Avancini E.; Carron R.; Handick E.; Garcia-Diez R.; Hartmann C.; Félix R.; Ueda S.; Wilks R. G.; Bär M. NaF/RbF-Treated Cu(In,Ga)Se2 Thin-Film Solar Cell Absorbers: Distinct Surface Modifications Caused by Two Different Types of Rubidium Chemistry. ACS Appl. Mater. Interfaces 2020, 12 (31), 34941–34948. 10.1021/acsami.0c08794. [DOI] [PubMed] [Google Scholar]
  53. Maticiuc N.; Kodalle T.; Lauche J.; Wenisch R.; Bertram T.; Kaufmann C. A.; Lauermann I. In Vacuo XPS Investigation of Cu(In,Ga)Se2 Surface after RbF Post-Deposition Treatment. Thin Solid Films 2018, 665, 143–147. 10.1016/j.tsf.2018.09.026. [DOI] [Google Scholar]
  54. Kreikemeyer-Lorenzo D.; Hauschild D.; Jackson P.; Friedlmeier T. M.; Hariskos D.; Blum M.; Yang W.; Reinert F.; Powalla M.; Heske C.; Weinhardt L. Rubidium Fluoride Post-Deposition Treatment: Impact on the Chemical Structure of the Cu(In,Ga)Se2 Surface and CdS/Cu(In,Ga)Se2 Interface in Thin-Film Solar Cells. ACS Appl. Mater. Interfaces 2018, 10 (43), 37602–37608. 10.1021/acsami.8b10005. [DOI] [PubMed] [Google Scholar]
  55. van Maris V. R.; Hauschild D.; Niesen T. P.; Eraerds P.; Dalibor T.; Palm J.; Blum M.; Yang W.; Heske C.; Weinhardt L. Impact of UV-Induced Ozone and Low-Energy Ar+-Ion Cleaning on the Chemical Structure of Cu(In,Ga)(S,Se)2 Absorber Surfaces. J. Appl. Phys. 2020, 128 (15), 155301. 10.1063/5.0020253. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

am3c11165_si_001.pdf (181.5KB, pdf)

Data Availability Statement

The data that support the findings of this study are available from the corresponding authors upon reasonable request.

Articles from ACS Applied Materials & Interfaces are provided here courtesy of American Chemical Society

RESOURCES