ZnO Heterointerfaces - ACS

Jul 25, 2016 - Potentiostatic and galvanostatic two-step electrodeposition of semiconductor Cu 2 O films and its photovoltaic application. Tonghui Yan...
1 downloads 5 Views 1MB Size
Subscriber access provided by CORNELL UNIVERSITY LIBRARY

Article 2

Band Alignment Engineering at CuO/ZnO Heterointerfaces Sebastian Siol, Jan C. Hellmann, S. David Tilley, Michael Grätzel, Jan Morasch, Jonas Deuermeier, Wolfram Jaegermann, and Andreas Klein ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.6b07325 • Publication Date (Web): 25 Jul 2016 Downloaded from http://pubs.acs.org on July 26, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Band Alignment Engineering at Cu2O/ZnO Heterointerfaces Sebastian Siol1*+, Jan C. Hellmann1, S. David Tilley2&, Michael Graetzel2, Jan Morasch1, Jonas Deuermeier3, Wolfram Jaegermann1, and Andreas Klein1* 1

Technische Universität Darmstadt, Institute of Materials Science, Surface Science Division, Petersenstrasse 32, 64287 Darmstadt, Germany 2

Ecole Polytechnique Fédérale de Lausanne (EPFL), Institut des Sciences et Ingénierie Chimiques, Laboratory of Photonics and Interfaces, Station 6, CH-1015 Lausanne, Switzerland 3

i3N/CENIMAT, Universidade NOVA de Lisboa and CEMOP/UNINOVA, Department of Materials Science, Faculty of Science and Technology, Campus de Caparica, 2829-516 Caparica, Portugal & Present Address: University of Zurich, Department of Chemistry, Winterthurerstrasse 190, 8057 Zurich, Switzerland + Present Address: National Renewable Energy Laboratory, 15013 Denver West Pkwy, 80401 Golden, CO, USA

*E-mail: [email protected], [email protected] Keywords: Band alignment, Cu2O, ZnO, XPS, Interface experiment, Fermi level pinning, Band Offset

Abstract Energy band alignments at heterointerfaces play a crucial role in defining the functionality of semiconductor devices, yet the search for material combinations with suitable band alignments remains a challenge for numerous applications. In this work we demonstrate how changes in deposition conditions can dramatically influence the functional properties of an interface, even within the same material system. The energy band alignment at the heterointerface between Cu2O and ZnO was studied using photoelectron spectroscopy with stepwise deposition of ZnO onto Cu2O and vice versa. A large variation of energy band alignment depending on the deposition conditions of the substrate and the film is observed, with valence band offsets ranging from ΔEVB = 1.45 – 2.7 eV. The variation of band alignment is accompanied by the occurrence or absence of band bending in either material. It can therefore be ascribed to a pinning of the Fermi level in ZnO and Cu2O, which can be traced back to oxygen vacancies in ZnO and to metallic precipitates in Cu2O, respectively. The intrinsic valence band offset for the interface, which is not modified by Fermi level pinning, is derived as ΔEVB ≈ 1.5 eV, being favorable for solar cell applications.

 

1   ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 18

Introduction The energy band alignment between two semiconductors is characterized by the discontinuities in the valence band maximum EVBM and the conduction band minimum energies ECBM at the interface1. These discontinuities are the basis for a large number of modern opto-electronic devices as they enable controlled quantum confinement of charge carriers and generation of interfacial two dimensional electron gases for high mobility transistors2. A popular example for the relevance of energy band alignments are thin film compound semiconductor solar cells such as Cu(In,Ga)Se2 or CdTe3, where suitable energy band discontinuities at the interfaces are required in order to enable transfer of photo generated charge carriers4–6. In recent years the term Band Alignment Engineering has been used to describe the new research paradigm of purposefully altering the band alignment at heterointerfaces to enable desired functionality7–10. Even though significant progress has been made in the prediction of energy band alignments, a generally applicable model for quantitative predictions has yet to be developed11,12. Energy band alignment of covalently bonded semiconductors like Si and III-V compounds are governed by induced interface dipoles, which result in an alignment of charge neutrality levels 1,13–15. For semiconductors with more ionic type of bonding, the density of induced interface states is expected to be lower 13,16. A large variation of Schottky barrier heights with both high and low work function oxide contact materials (RuO2, ϕ = 6.1 eV; In2O3:Sn (ITO), ϕ = 4.5 eV) has been explicitly demonstrated, for example, for a number of semiconducting oxides like (Ba,Sr)TiO3 and Pb(Zr,Ti)O3 17,18. Experimental determinations of valence band discontinuities ΔEVB of a number of semiconducting oxides using photoelectron spectroscopy indicate weak Fermi level pinning for defect free interfaces. In these cases the energy band alignment can be understood in terms of the orbital contributions to the valence bands19–23. However, defects are often formed during interface formation, which can lead to a considerable modification of energy band alignment and/or Schottky barrier heights22,24–29. This is also true for Cu2O, where the presence of metallic Cu or CuO can lead to a modification of band alignment30–32. No strong dependence of band alignment on the details of the interface structure is expected, if the alignment is governed by charge neutrality levels or orbital contributions to the valence bands, since both are bulk properties. But a variation may be expected in dependence on atomic reconstruction for interfaces between materials with different polarity1,33,34. However, except for a few examples (see e.g. 35), a significant variation of energy band alignment, which can be related to intrinsic interface properties, has not yet been demonstrated. Materials availability and toxicity issues motivated the search for alternative materials for photovoltaic solar energy conversion36–38. Among them, semiconducting oxides have recently been considered 39,40. Currently, the most successful oxide solar cells are based on heterostructures of Cu2O. In 2011 reports of Cu2O/ZnO solar cells with efficiencies of up to 3.8% led to a renewed  

2   ACS Paragon Plus Environment

Page 3 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

interest in the material for photovoltaic applications41,42. In recent years the conversion efficiencies Cu2O based solar cells could be improved significantly. Utilization of alternative buffer layers led to much higher efficiencies of 5.3% with Ga2O343,44 and 8.1% with Zn1−xGexO 45, respectively. Here the main improvement is due to a reduction of the conduction band offset, which leads to an increase in open circuit voltage45. Cu2O is one of the oldest known semiconducting materials 42,46. It crystallizes as cubic cuprite with a lattice constant of a = 4.27 Å and has a direct band gap of 2.17 eV47. Its common p-type conductivity originates from copper vacancies with an ionization energy of 0.28 eV48. ZnO has a direct band gap of 3.3 eV and crystallizes in hexagonal wurtzite structure with lattice constants of a = 3.25Å and c = 5.20 Å47. ZnO is intrinsically doped n-type, though it is still unsettled which defect causes this doping 49–51. Different values for the valence band offset at Cu2O/ZnO have been reported 52–54. Several groups have reported on successfully altering the band alignment of Cu2O/ZnO heterointerfaces55–57. However, the mechanisms, leading to the different band alignments, have not yet been completely identified. In this work the origin of this phenomenon is systematically investigated. A strong dependence of the energy band alignment at the oxide heterointerface between Cu2O and ZnO on the preparation of the layers is reported. The observed variation can be directly related to tiny differences in the chemical composition of the layers, which lead to various concentration of metallic precipitates for Cu2O or oxygen vacancies for ZnO. A dependence of band alignment on defect properties has been shown previously for various oxide/metal 25,29,58 and oxide/semiconductor interfaces 37,59–62. In the present case, the assignment becomes evident from the different magnitude of surface potential changes in the layers, which are prohibited in the case of larger defect concentrations. The results presented in this contribution therefore do not only explain the variation of ΔEVB at the Cu2O/ZnO interfaces reported in this work and in the literature, they also provide a general strategy to control the energy band alignment at oxide heterointerfaces. Experiment Band alignment and band bending (BB) during interface formation are obtained from Xray photoelectron spectroscopy (XPS) performed in the Darmstadt Integrated System for Materials Research (DAISY-MAT)63, which combines a Physical Electronics PHI 5700 multi technique surface analysis system with several thin film deposition chambers by a vacuum transfer system. XPS measurements were performed using monochromatic Al Kα radiation with an energy of 1486.6 eV and an overall energy resolution of less than 0.4 eV as derived from the Gaussian broadening of the Fermi edge of a sputter cleaned Ag foil, which is also used for binding energy (BE) calibration. For ultraviolet photoelectron spectroscopy (UPS) measurements a He(I) discharge lamp (hν = 21.2 eV) is used as excitation source. The interface properties were studied using increasing film thickness of ZnO on Cu2O and vice versa. Interface experiments with films deposited by magnetron sputtering were  

3   ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 18

performed in-situ in the DAISY-MAT without breaking vacuum between individual deposition and analysis steps using a single sample. Atomic layer deposition (ALD) and electrodeposition (ED) growth were performed ex-situ with mail transfer of samples in ambient atmosphere. In this case, each film thickness corresponds to a separate sample. A summary of the performed experiments is given in Table S1. Due to a non-negligible photo voltage upon illumination no work functions and hence no values for the interface dipole could be obtained by UPS measurements. A small potential drop over the interface due to x-ray induced photo voltage might occur during regular XPS measurements. Performing calibration measurements on bulk samples, this photo voltage could be identified to be smaller than VPh < 0.10 eV. Overall this adds to the uncertainty of the measurements but does not change the overall conclusion of this study. ZnO layers were deposited either by magnetron sputtering (MS) or ALD, and Cu2O films by reactive magnetron sputtering or ED, respectively. MS-Cu2O was deposited at room temperature. The stoichiometry of the films can be controlled by changing the argon- to oxygen fluxes, which were adjusted using mass flow controllers. The ZnO layers were likewise deposited via RF-magnetron sputtering using a ceramic ZnO target at room temperature or 300°C. A change of defect concentration was further achieved by adding up to 3 % oxygen to the process gas. The atomic ratios have been calculated via XPS measurements by evaluating the area under curve values for the Cu 2p3/2, Zn 2p3/2 and O 1s spectra taking into account the respective sensitivity factors. Additional experiments were performed using electrodeposited Cu2O. The ED of Cu2O was performed at EPFL as described in 64, and interfaces with ALD ZnO and ZnO:Al were investigated by preparation of several samples with different ZnO or ZnO:Al thickness. The experimental details for the ALD of ZnO and ZnO:Al are given in the supporting information. Results In this contribution the band alignments for several Cu2O/ZnO heterointerfaces are presented (see Table 1). To determine the band alignments of the respective Cu2O/ZnO interfaces, XPS measurements were carried out while successively depositing ZnO on a thick layer of Cu2O or vice versa. The depositions were carried out as described in the experimental section with the deposition parameters supplied in Table S1. For each experiment survey spectra were taken for each deposition step to check the sample for contamination. Special attention was given to the stoichiometry of the Cu2O layers, which was verified by checking the Cu 2p3/2 and Cu LMM spectra, respectively. In the Cu 2p region characteristic shake up satellites would indicate the presence of Cu2+ (CuO, as well as Cu(OH)2)20. In the copper Auger spectra a shoulder at binding energies of 568 eV would be visible if elemental copper is embedded in the film. While no metallic Cu or CuO could be identified for the in-situ samples, ex-situ samples prepared by ED showed trace amounts of surface oxidation. The following discussion describes in an exemplary manner how the band alignment is  

4   ACS Paragon Plus Environment

Page 5 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

determined from the photoelectron (PE) spectra, following a commonly practiced procedure as described in65. First XPS measurements are carried out while successively depositing the Film material on a thick layer of the Substrate material (see Table 1). XPS measurements are performed after every deposition step to determine the evolution of binding energies with increasing film thickness. Since the evolution of the valence band maximum cannot be precisely determined for steps of intermediate coverage, core level binding energies were measured of Zn 2p3/2 and O 1s as well as Cu 2p3/2 and CuLMM, respectively (a exemplary set of photo emission spectra is shown in Figure S1). For the determination of the band alignment the peak positions of substrate- as well as filmspecific photoemission lines have to be considered. The binding energies of the core level spectra were determined by fitting using Wavemetrics IGOR Pro software. The position of the valence band maximum was determined by a linear extrapolation of the low binding energy edge of the valence band emission.

 

5   ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 18

From the evolution of core level binding energies over film thickness the band diagram can be extracted. A display of this evolution of core level BE for all discussed interface experiments is given in Figure 1.

Figure 1: Evolution of the core level BE as well as the valence band maximum (VBM) with increasing film thickness for the individual interface experiments discussed in this study. The core level binding energies are normalized to the VBM in the bulk of the respective material. Interface experiments A-D were carried out in-situ. Experiments E-G were prepared ex-situ.

 

Based on the evolution of the core level BE for each material the surface potential changes can be determined, respectively. The surface potential changes in the substrate ΔEBE-sub add to any surface band bending, which might be present prior to deposition. To determine the band bending in the growing film (BBfilm), the deposition induced  

6   ACS Paragon Plus Environment

Page 7 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

surface potential change (ΔEBE-sub) has to be subtracted from the shift of the film-specific core level binding energies (ΔEBE-film), since it has to be taken into account that a change in the substrate’s electric potential causes a shift of all core level binding energies: BBfilm = ΔEBE-film – ΔEBE-sub.

(1)

The valence band offset at the interface is calculated using the BE difference of the Cu 2p3/2 and Zn 2p3/2 peaks at intermediate coverage (ΔBE) as well as the BE of the Cu 2p3/2 (BECu2p3/2-VBM) and Zn 2p3/2 (BEZn2p3/2-VBM) core levels with respect to the valence band maximum. ΔEVB = BECu2p3/2-VBM – BEZn2p3/2-VBM + ΔBE

(2)

Using ΔEVB and the band gaps from literature (Eg = 2.1 eV for Cu2O and Eg = 3.3 eV for ZnO) the conduction band offset can be determined as ΔECB = 2.1 eV + ΔEVB - 3.3 eV

(3)

The band offsets as well as the magnitude of the substrate surface potential changes ΔEBE-sub and the band bending in the film for all interface experiments are listed in Table 1. The typical uncertainty in the determination of these values is on the order of ± 0.1 eV. Table 1: Investigated interfaces. Shown are the substrate (Sub) and the film material, as well as the respective deposition method (MS: RF-magnetron sputtering, ED: electrodeposition, ALD: atomic layer deposition). Depending on the EVBM, different substrate surface potential changes ΔEBE-sub and film band bending BBfilm can be observed. The magnitude of the BB is related to the variation of the valence band offsets ∆EVB. #

Sub

EVBM

ΔEBE-sub

Film

EVBM

BBfilm

∆EVB

A

MS-Cu2O

0.25

0.5

MS-ZnO

2.8

0.1

1.9

B

MS-ZnO

2.75

0.55

MS-Cu2O

0.21

0.35

1.6

C

MS-Cu2O

0.45

0.0

MS-ZnO

2.6

0.6

1.6

D

MS-ZnO

2.44

0.5

MS-Cu2O

0.48

0.1

1.45

E

ED-Cu2O

0.26

0.55

MS-ZnO

2.86

-0.15

2.15

F

ED-Cu2O

0.41

0.0

ALD-AZO

3.4

0.3

2.7

G

ED-Cu2O

0.30

0.3

ALD-ZnO

3.3

0.25

2.5

Besides the deposition method of the respective layers (MS: RF-magnetron sputtering, ED: electrodeposition, ALD: atomic layer deposition) the values for valence band offset ΔEVB, substrate surface potential changes ΔEBE-sub and film band bending BBfilm are listed for every experiment. Additionally the energy values for the valence band maxima EVBM for the substrate as well as the thick film are given.

 

7   ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 18

  Figure 2: Energy band diagrams for the 7 interface experiments carried out in this study. Shown are the respective Fermi level (EF), the band gap (Eg), substrate surface potential change (ΔEBE-sub), film band bending (BBfilm), as well as the valence band offset (∆EVB) and conduction band offset for the interface (∆ECB). Experiments A-D were carried out in-situ. Interface experiments E-G were prepared ex-situ. Surface band bending in the substrate and the film are neglected. With these values the band diagram for the ZnO/Cu2O-interface can be drawn. Figure 2 shows the band diagrams for all seven interface experiments that were carried out for this study. Discussion For this work a total of seven interface experiments was carried out. The results are given in the previous section. Looking at the values in Table1 a broad range of valence band offsets (VBO) can be observed. Dependence of energy band alignment on sample preparation has been observed at a few hetero interfaces, in particular if oxide semiconductors are involved20,54,59,62. In the case of the Cu2O/ZnO heterointerfaces investigated in this study, the development of the valence band discontinuity depends strongly on the valence band maximum in the bulk of both substrate and film as well as on the occurrence of band bending in the respective materials. For layers deposited via RF-magnetron sputtering the magnitude of  

8   ACS Paragon Plus Environment

Page 9 of 18

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

band bending can be controlled by changing the oxygen partial pressure during deposition. Cu2O layers that have been deposited using 4.0% oxygen (Exp. A-B) show a band bending of up to 0.5 eV whereas for 3.7% oxygen (Exp. C-D) no band bending can be observed. Taking into account the differences in intrinsic doping, which arise from the alternating deposition parameters for layers prepared under more oxidizing conditions, Fermi levels in the range of 0.2 eV-0.8 eV can be achieved at the interface whereas for layers prepared under a more reducing atmosphere the Fermi level seems to be pinned at around 0.5 eV. Aside from the intrinsic doping the deposition conditions directly affect the materials stoichiometry (see Table S1). The Fermi level pinning effect observed for CuxO layers with atomic Cu/O ratios of x > 2 (Exp. C & D) can tentatively be explained by a large density of Cu2O/Cu-Schottky barriers at internal interfaces spread throughout the material, which are commonly known to promote Fermi level pinning66. This way the Fermi level in Cu2O adopts a value of EF - EVB = (0.5 ± 0.1) eV. The barrier height could be confirmed by performing Cu/Cu2O in-situ interface experiments58. A schematic illustration of this pinning mechanism is displayed in Figure 3.a.

Figure  3:  a)  Schematic  illustration  of  the  Fermi  level  pinning  mechanism  in  Cu-­‐rich   Cu2O  layers.  Shown  are  the  work  functions  (Φ)  and  band  gaps  (Eg)  of  Cu  and  Cu2O,   respectively.  If  the  distance  d  between  two  interfaces  becomes  significantly  smaller   than  the  space  charge  region  W  of  the  interface,  the  EVBM  will not reach the bulk value in the areas between the contacts. b) Correlation between the (bulk) EVBM  of  the  Cu2O   layers  discussed  in  this  study  and  their  respective  band  bending  at  the  interface   with  ZnO.

 

Figure 3.b displays the correlation between intrinsic doping and observed band bending. For CuxO-layers with atomic Cu/O ratios of x ≤ 2 (EVBM < 0.4 eV) band bending of up to 0.55 eV can be observed. For these layers a limitation of the Fermi level through formation of point defects can be observed. Here the Fermi levels are found to be in the range of 0.2 - 0.8 eV. The lower limit corresponds well to the charging of copper vacancy defect states whereas the upper limit is in accordance with a discontinuity of formation energy for copper interstitials67. For the ex-situ prepared layers of experiments E-G, this stoichiometry is not as easily accessible since adsorbates on the sample’s surface falsify the values for the oxygen content. Nevertheless the big discrepancy in band bending  

9   ACS Paragon Plus Environment

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 18

between the experiments can be explained by the different methods for the ZnOdeposition. The highest band bending can be observed for the case of magnetron sputtered ZnO (Exp. E) whereas no band bending is obtained for the ALD preparation of ZnO:Al (Exp. F) on top of the electrodeposited Cu2O. It has been shown in the literature how, depending on the choice of precursors, the substrates surface can be reduced during ALD-deposition62,31. As opposed to the ZnO deposition via magnetron sputtering this may result in an increased density of Cu-precipitates at the junction. For ZnO the restriction of the Fermi level shift within the band gap due to contact formation can be traced back to oxygen vacancies and is most pronounced for ZnxO layers that have been deposited in pure argon atmosphere and exhibit a stoichiometry of x ≥ 1.2 (Exp. A & E). For these layers, that naturally show an abundance of oxygen vacancies, the Fermi level seems to be pinned to values of EF > 2.5 eV, which corresponds well to the charging point for oxygen vacancies as reported in the literature50,27. For the case of a pinned Fermi level in ZnO, taking into account the observed restrictions for the Fermi level in Cu2O of 0.2