Nitrogen-doped Cu₄O₃ thin films as high-performance counter electrodes for quantum dot-sensitized solar cells

Abstract

P-type metal oxide semiconductors are critical components in the development of next-generation optoelectronic and photovoltaic devices. While n-type materials such as SnO₂, ZnO, and ITO are well-established, the lack of stable, high-performance p-type transparent oxides with suitable bandgaps remains a key limitation. Among copper oxides, Cu₄O₃—a mixed-valence oxide—offers promising electronic and optical tunability, yet has been underexplored for device integration. In this work, nitrogen-doped Cu₄O₃ thin films were synthesized via DC magnetron sputtering under an (Ar + 10% O₂)/N₂ atmosphere. The incorporation of nitrogen effectively modulated the electronic structure, enhanced chemical stability, and improved electrical transport properties. The optimal film, denoted as Cu₄O₃-30 (30% N₂), exhibited a direct optical bandgap of 2.18 eV, resistivity of 4.19 Ω cm, hole concentration of 3.33 × 10¹⁷ cm⁻³, and hole mobility of 4.48 cm² V⁻¹ s⁻¹. When implemented as a counter electrode in TiO₂/CdS/CdSe:Cu/ZnS quantum dot-sensitized solar cells (QDSSCs), the device achieved a power conversion efficiency of 7.29%, exceeding the performance of its Cu₂S-based counterparts. These results highlight the potential of N-doped Cu₄O₃ as a scalable, chemically stable, and electrically efficient p-type oxide for emerging optoelectronic and photovoltaic technologies.

---

1. Introduction

Metal oxide thin films have long been integral to the advancement of photovoltaic technologies, owing to their favorable optical and electronic properties. Among these, wide bandgap n-type semiconductors such as SnO₂, ZnO, and indium tin oxide (ITO) are widely employed. These materials serve as transparent conductive layers, electron transport layers, or buffer layers in diverse device architectures, enabling efficient charge extraction and minimizing optical losses. However, next-generation optoelectronic devices—including solar cells, transparent transistors, and light-emitting diodes—demand not only efficient n-type materials but also transparent p-type layers with high hole conductivity and suitable band alignment. Such layers play a critical role in enhancing charge separation and transport under visible-light irradiation, thereby improving device performance.

Achieving stable p-type conductivity in metal oxides, however, remains a significant challenge. This challenge originates from intrinsic factors such as the deep nature of acceptor levels, self-compensation effects, and the strong localization of holes in O 2p orbitals, which collectively limit carrier mobility. In many cases, the difficulty in incorporating aliovalent dopants to create acceptor states is compounded by poor dopant solubility and low substitution efficiency—an issue noted in studies on p-type copper oxide films synthesized by ALD and sputtering methods.¹ Despite this, some metal oxides naturally exhibit p-type behavior, including SnO, NiO, and various copper oxides. Of these, SnO and NiO possess wide bandgaps (>3 eV), restricting their optical absorption in the visible range. In contrast, copper oxides, particularly those with narrower bandgaps and inherent p-type characteristics, offer strong potential for visible-light-driven optoelectronic applications.

Recent studies have drawn increasing attention to Cu₄O₃, a mixed-valence copper oxide comprising both Cu(I) and Cu(II) species. This compound inherits key features from Cu₂O (direct bandgap ∼1.82–2.56 eV)^2–7 and CuO (indirect bandgap ∼1.42–2.07 eV),^8–12 resulting in tunable optical and electronic properties. Notably, Cu₄O₃ itself exhibits a direct bandgap ranging from 1.34 to 2.34 eV,^13–16 rendering it suitable for photoactive and photovoltaic applications. In addition to its optical versatility, the mixed-valence nature of Cu₄O₃ provides opportunities for defect engineering, particularly via aliovalent doping, to manipulate carrier concentration, mobility, and catalytic activity.

Copper-based oxides have been extensively explored for applications such as photocatalysis,⁵ supercapacitors,¹⁰ water splitting,¹⁷ and light sensing,^9,18 with solar energy conversion receiving significant focus. Among these, Cu₂O has emerged as a leading candidate due to its optimal bandgap and high absorption coefficient (∼10⁵ cm⁻¹).^{2–4,6,7,19–27} By contrast, Cu₄O₃ remains relatively underexplored. While several studies have demonstrated the synthesis of Cu₄O₃ films via DC^13,14,16 and RF sputtering,¹⁵ and efforts have been made to tune its properties through thickness and phase engineering,²⁸ studies on dopant-induced electronic modification—particularly nitrogen doping—are rare. Nitrogen, with its comparable ionic radius to oxygen but distinct electronegativity, offers a potential route to substitute oxygen sites, introduce acceptor-like states, and tailor the electronic environment; however, such investigations for Cu₄O₃ are virtually absent. To date, only one report has addressed nitrogen incorporation in Cu₄O₃via oxygen substitution.¹⁵ Recent progress in QDSSC research has focused on interfacial engineering, advanced counter electrode design, and multifunctional quantum dot integration to enhance efficiency and stability.^29–31 Furthermore, no previous studies have evaluated Cu₄O₃ as a hole-conductive material in photovoltaic systems, particularly in quantum dot-sensitized solar cells (QdSSCs). This gap is critical because existing hole transport materials, such as Cu₂S, suffer from severe photocorrosion and instability, limiting their long-term device performance. This represents a critical knowledge gap. In this work, we address this void by synthesizing nitrogen-doped Cu₄O₃ thin films using DC magnetron sputtering under a mixed (Ar + 10% O₂)/N₂ atmosphere. We systematically investigate their structural, optical, and electrical properties and, for the first time, demonstrate their use as hole-transport layers in QdSSCs, replacing Cu₂S in the traditional FTO/TiO₂/CdS/CdSe:Cu/ZnS/Cu₂S/FTO configuration.³²

The N-doped Cu₄O₃ films display improved chemical stability, tunable electrical behavior, and enhanced catalytic robustness. These properties directly address the limitations of Cu₂S-based counter electrodes, which suffer from photocorrosion, chemical instability, and poor long-term performance.^33,34 By bridging the knowledge gap between fundamental defect engineering and device-level application, this study highlights the untapped potential of N-doped Cu₄O₃ as a scalable, stable, and efficient p-type layer for next-generation optoelectronic platforms.

2. Experiments

2.1. Fabrication of Cu₄O₃-x thin films

Nitrogen-doped copper oxide (Cu₄O₃-x) thin films were fabricated via direct current (DC) magnetron sputtering onto quartz and fluorine-doped tin oxide (FTO) substrates. Reactive sputtering was performed at a working pressure of 1 × 10⁻³ torr in a mixed Ar + O₂ + N₂ atmosphere. The total gas flow rate was maintained at 50 sccm. The O₂ concentration was fixed at 10% of the total flow (i.e., 5 sccm O₂), while the N₂ content was systematically varied from 0% to 40% of the total gas flow in 10% increments (i.e., 0 to 20 sccm N₂). The balance of the flow (i.e., 45–25 sccm) was Ar. For example, the 10% N₂ condition included 5 sccm O₂, 5 sccm N₂, and 40 sccm Ar; the 40% N₂ condition included 5 sccm O₂, 20 sccm N₂, and 25 sccm Ar. All gases (purity: 99.9999%) were introduced through high-precision mass flow controllers to ensure accurate control of the reactive gas composition. Prior to deposition, substrates were preheated to 150 °C to promote film adhesion and ensure uniform nucleation. High-purity (99.999%) copper targets with a 3-inch diameter and 0.25-inch thickness were utilized as the sputtering source. Film deposition was carried out in a Leybold Univex 450 high-vacuum system (Germany), where the chamber base pressure was maintained at 1 × 10⁻⁶ torr. The sputtering power was fixed at 80 W, with the target-to-substrate distance set at 10 cm. Prior to deposition, both quartz and FTO substrates underwent sequential cleaning using deionized water, acetone, and a dilute NaOH solution to remove organic and ionic contaminants.

Fluorine-doped tin oxide (FTO) glass substrates (resistivity ∼7 Ω sq⁻¹), titanium dioxide (TiO₂) nanoparticle paste (typically anatase phase, particle size ∼20 nm), cadmium nitrate tetrahydrate (Cd(NO₃)₂·4H₂O, ≥99%), sodium sulfide nonahydrate (Na₂S·9H₂O, ≥98%), selenium powder (Se, 99.99%), sodium borohydride (NaBH₄, ≥98%), copper(II) acetate monohydrate (Cu(CH₃COO)₂·H₂O, ≥98%), and zinc acetate dihydrate (Zn(CH₃COO)₂·2H₂O, ≥99%) were used. All reagents were purchased from Sigma-Aldrich, Alfa Aesar, or equivalent suppliers and used as received without further purification. Ethanol, methanol, and deionized (DI) water were used as solvents throughout the process.

2.2. Fabrication of FTO/TiO₂/CdS/CdSe:Cu(0.3)/ZnS electrodes

The creation of the FTO/TiO₂/CdS/CdSe:Cu(0.3)/ZnS photoanode began with the formation of the FTO/TiO₂ film using the doctor blade technique and annealing at 500 °C for 1 h. Following this, the CdS layer is formed through the SILAR method, a process that involves placing the FTO/TiO₂ substrate in a 0.1 M cadmium ion (Cd²⁺) solution for six minutes, rinsing with ethanol, and subsequently placing it in a 0.1 M sulfide ion (S²⁻) solution for five minutes, and rinsing with methanol. This sequence was repeated thrice to ensure the creation of a consistent CdS layer. Subsequently, a CdSe:Cu(Cd:Cu = 0.3) layer was formed using the SILAR method. Initially, the Se precursor solution was prepared by dissolving 3.7812 g of sodium sulfite (Na₂SO₄) and 1.135 g of selenium powder in 50 mL of distilled water, followed by adding 2.5 milliliters of 1 mM sodium hydroxide (NaOH) to improve selenium solubility and agitating the mixture in a 70 °C oil bath for 16 h to produce a stable Se²⁺ solution. A cadmium–copper solution was prepared by dissolving 0.0025572 g of CuCl and cadmium acetate in a 1 : 1 ethanol and water mixture. The FTO/TiO₂/CdS substrate was then immersed in this solution for five minutes, rinsed with ethanol, and dried in air. Next, the substrate was placed in a 70 °C Se²⁺ solution for five minutes, cleaned with distilled water, and air-dried at room temperature. This process was performed three times to achieve controlled formation of the CdSe:Cu(0.3) layer. To enhance photostability and suppress surface recombination, a ZnS passivation layer was deposited via 2 SILAR cycles using 0.1 M zinc acetate and 0.1 M sodium sulfide aqueous solutions.

2.3. Fabrication of QDSSCs with different counter electrodes

Counter electrodes were prepared using previously synthesized Cu₄O₃-x coated FTO substrates. The photoanode and counter electrode were assembled into a sandwich-type cell with a polysulfide electrolyte (0.5 M Na₂S, 2 M S, 0.2 M KCl in DI water/methanol, 3 : 7 v/v) introduced between the electrodes. The assembled QDSSCs were sealed and stored under ambient conditions for performance characterization.

2.4. Characterization

The optical characteristics of Cu₄O₃-x thin films were examined utilizing UV-vis spectroscopy (JASCO V-730) within the wavelength range of 190–1100 nm. Electrical conductivity measurements were conducted at 28 °C utilizing Hall effect measurement equipment (HL5500PC; Nanometrics Inc., USA). The elemental composition, comprising copper (Cu), oxygen (O), and nitrogen (N), was examined using X-ray spectroscopy through the energy-dispersive mechanism (EDX; FE-SEM; JEOL JSM-IT800). The films' morphology, including both the surface and cross-section, was examined utilizing FE-SEM. The binding energies of the elements were analyzed via X-ray photoelectron spectroscopy (XPS) using an AXIS Ultra DLD spectrometer (Kratos–Shimadzu) equipped with a monochromatic Al Kα X-ray source (hν = 1486.6 eV). The chamber pressure was maintained below 1 × 10⁻⁸ torr during analysis. Survey spectra were collected over the binding energy range of 0–1350 eV with a pass energy of 160 eV and a step size of 1.0 eV to identify elemental composition. High-resolution spectra for core levels (e.g., Cu 2p, O 1s, and N 1s) were acquired with a pass energy of 20 eV and a step size of 0.1 eV. The binding energy scale was calibrated using the adventitious C 1s peak at 284.8 eV. A Shirley background was subtracted, and the peak shapes were fitted using a Voigt function, which represents a convolution of Gaussian and Lorentzian components. The fitting parameters were carefully adjusted to minimize the residual error while maintaining physical relevance. Structural investigation was conducted via X-ray diffraction (XRD) utilizing a Shimadzu XRD-6100 instrument with Cu-Kα radiation. The rough surface with microscale features was assessed using microscopy based on the atomic force interaction mechanism (AFM; Bruker Dimension Edge). The electrochemical characteristics of the Cu₄O₃-x/FTO counter electrode were examined by spectroscopic impedance analysis and cyclic voltage measurement, employing a CHI 650E potentiostat/galvanostat. The redox electrolyte utilized in EIS studies consisted of 2 M Na₂S, 2 M elemental sulfur (S), and 0.2 M KCl. The counter electrode possessed an active area of 0.25 cm², with the inter-electrode distance between the counter and working electrodes regulated by using a set spacer. The current–voltage (I–V) properties of quantum dot-sensitized solar cells (QDSSCs) were assessed under light and dark conditions utilizing a solar simulator and a Keithley 2450 source meter. The light intensity was calibrated to 100 mW cm⁻² (AM 1.5G equivalent) using a silicon photodetector-based radiometer.

3. Results and discussion

3.1. Analysis of atomic composition via EDX

Nitrogen-doped Cu₄O₃ (denoted as Cu₄O₃-x) thin films were deposited on silicon substrates using a mixed gas containing varying ratios of N₂, in order to eliminate interference from oxygen originating from the glass substrate. This approach enables a more precise analysis of the elemental composition of Cu, O, and N within the films, providing a solid foundation for future research. The energy-dispersive X-ray spectroscopy (EDX) spectra of the Cu₄O₃-x films, shown in Fig. 1, exhibit characteristic K_α peaks for N and O at 0.39 keV and 0.52 keV, respectively, while the Cu L_α peak appears at 0.93 keV. Additionally, the Si K_α signal at 1.74 keV is observed, attributed to the Si substrate. The atomic compositions, summarized in Table 1, indicate an increase in N content with increasing x, which is correlated with a concurrent reduction in oxygen content. Here, the parameter x represents the increasing N fraction in the sputtering gas mixture. This trend suggests that N is substituting for O at lattice sites. Moreover, the atomic ratio r = Cu/(N + O) is observed to decrease with increasing x, implying the formation of copper vacancies (V_Cu) within the lattice. The incorporation of N in place of O appears to promote the generation of these vacancies. The r value further supports the stoichiometric baseline of undoped copper oxide as Cu₄O₃. A declining r ratio with increasing N incorporation also suggests an evolution in the chemical environment, possibly indicating a shift in copper oxidation states or local bonding configurations. Moreover, the mapping images show the homogeneous distribution of all elements over the whole Cu₄O₃ films as seen in Fig. 2.

Fig. 1 EDX spectra of Cu₄O₃-x films deposited with varying N₂ contents in a mixed sputtering gas.

Table 1 The composition of Cu₄O₃-x films determined by EDX spectra

Samples	% at. Cu	% at. O	% at. N
Cu4O3-0	59.57	40.03	0.4	≈1.47
Cu4O3-10	58.54	38.91	2.55	≈1.41
Cu4O3-20	57.71	39.01	3.28	≈1.36
Cu4O3-30	56.92	38.65	4.43	≈1.32
Cu4O3-40	55.85	37.83	6.32	≈1.27

---

Fig. 2 EDX mapping images of Cu₄O₃-x films deposited with varying N₂ contents in a mixed sputtering gas.

3.2. Analysis of atomic composition via XPS

XPS was utilized to provide a comprehensive understanding of the elemental composition and chemical states of N, O, and Cu in the Cu₄O₃-x thin films, using the C 1s peak at 284.8 eV as a reference. The corresponding spectra are shown in Fig. 3, with Fig. 3A–C presenting the core-level binding energy profiles of N 1s, O 1s, and Cu 2p, respectively. As illustrated in Fig. 3A, the N 1s peak appears within the binding energy range of 393.3–398.8 eV, depending on the x value. A noticeable shift toward lower binding energies is observed with increasing x, which is attributed to enhanced electronic screening effects as N becomes increasingly incorporated into the lattice. Fig. 3C displays the Cu 2p spectrum, characterized by a well-defined spin–orbit doublet corresponding to the 2p_3/2 and 2p_1/2 transitions, ranging from 932.7 to 952.7 eV. Notably, no significant shift in the Cu binding energies is detected as x increases, suggesting that competing effects—namely electron withdrawal by N substituting for O or the reduction in oxygen vacancies, and electron donation associated with Cu vacancies effectively compensate for each other. This compensation maintains a relatively constant electron density around the Cu ions. To elucidate the oxidation states of Cu, the Cu 2p_3/2 peak was deconvoluted into two distinct components corresponding to Cu¹⁺ and Cu²⁺, located at 932.6 eV and 934.0 eV, respectively as reported in ref. 2, 3, 5 and 22. Fig. 3B shows the O 1s spectrum, which exhibits an asymmetric peak centered near 529.8 eV, indicating the presence of multiple oxygen species. Deconvolution reveals four components: (1) a peak at 529.8 eV assigned to lattice oxygen bonded to Cu²⁺ (O_Cu2), (2) a peak at 530.4 eV corresponding to oxygen bound to Cu¹⁺ (O_Cu1), (3) a peak at 531.0 eV attributed to V_O, and (4) a peak at 532.5 eV related to adsorbed oxygen species. These spectral features collectively demonstrate the evolution of chemical states with varying composition. The atomic percentage of each element (detailed in Table 2) was calculated using the integrated peak areas of the high-resolution XPS spectra and the corresponding atomic sensitivity factors (ASF) mentioned in ref. 35 (S_Cu2p3/2 = 16.73; S_O1s = 2.93; S_N1s = 1.8). The quantification follows the standard formula , where i and j represent Cu, N, and O, respectively. I_[i,j] signifies the integrated intensity for element [i, j], whereas S[i, j] denotes the sensitivity factor. As x increases, the N content increases while the O and V_O contents decline, confirming that N atoms substitute for lattice O sites. Additionally, an increase in Cu²⁺ content is evident with increasing x, as shown by the growing Cu²⁺/Cu⁺ ratio, which is consistent with the increase in the O_Cu2 component and the reduction in both O_Cu1 and V_O contributions. Furthermore, Table 2 indicates that the atomic ratio of total Cu (Cu⁺ + Cu²⁺) to the combined O and N content decreases from 1.47 to 1.28 with increasing x. This trend aligns with the nominal stoichiometry of Cu₄O₃ and reflects the overall reduction in Cu content relative to anion content due to N incorporation into the host lattice. Lastly, a comparison between XPS and EDX analyses reveals minor discrepancies in elemental ratios; however, the overall agreement supports the conclusion that the elemental distribution is homogeneous throughout thin films. In summary, XPS analysis confirms the substitution of O with N in Cu₄O₃, which intensifies with increasing x and is linked to a decrease in V_O and an increase in V_Cu, as seen by the increase in Cu²⁺. The obtained results indicate the following expression for the Kröger–Vink doping reaction (eqn (1)):Therein, – represents an oxygen vacancy that is doubly positively charged; denotes a neutral oxygen atom situated on an oxygen site; – signifies a neutral copper atom located on a copper site; – indicates a copper atom on a copper site with a positive effective charge (Cu²⁺ or hole); – represents a negatively charged copper vacancy; – refers to a nitrogen atom substituting an oxygen site with a single negative charge; – represents a hole; – denotes N₂ gas.

Fig. 3 XPS spectra of Cu₄O₃-x films deposited with varying N₂ contents in a mixed sputtering gas: (A) N 1s spectra, (B) O 1s spectra, and (C) Cu 2p spectra.

Table 2 The composition of Cu₄O₃-x films determined by XPS spectra

Samples	Cu4O3-0	Cu4O3-10	Cu4O3-20	Cu4O3-30	Cu4O3-40
% at. Cu^2+	28.03	28.68	28.99	29.18	29.19
% at. Cu^1+	31.42	29.56	28.56	27.7	26.96
% at. Cu(Cu^1+ + Cu^2+)	59.45	58.24	57.55	56.88	56.15
% at. OCu2	23.76	24.45	24.84	25.15	24.59
% at. OCu1	13.37	12.79	12.76	12.61	12.35
% at. VO	2.96	2.11	1.81	1.21	0.9
% at. O(O1 + O2 + VO)	40.09	39.35	39.41	38.97	37.84
% at. N	0.46	2.41	3.04	4.15	6.01
	≈0.89	≈0.97	≈1.02	≈1.05	≈1.08
	≈1.78	≈1.91	≈1.95	≈1.99	≈1.99
	≈1.47	≈1.40	≈1.36	≈1.32	≈1.28

---

3.3. Analysis of structural properties with XRD patterns

The structural properties of Cu₄O₃-x thin films were confirmed using X-ray diffraction (XRD) analysis, with reference to the standard JCPDS card no. 03-0879. For films deposited with N content up to x ≤ 20, the diffraction pattern prominently exhibits the (004) lattice plane, accompanied by a weaker (220) peak (Fig. 4), indicating a preferential orientation along these planes. However, as the N incorporation increases to x = 30 and 40, the (004) reflection diminishes and is completely suppressed, while a new (202) peak emerges, suggesting a reorientation of the crystalline structure induced by higher N content. Additionally, the intensity of the (220) peak shows a consistent decline with increasing x, further evidencing structural modifications due to N doping. XRD peak positions for all films display a noticeable shift toward lower angles, as summarized in Table 3. This peak shift is indicative of lattice expansion, which strongly supports the successful substitution of O atoms by N within the Cu₄O₃ crystal structure. Since Cu₄O₃ is an ionic crystal, the ionic radii of N³⁻ (146 pm) and O²⁻ (140 pm) should be considered. The substitution of smaller O²⁻ ions by larger N³⁻ ions leads to an increase in interplanar spacing, resulting in the observed diffraction peak shifts. Moreover, the average crystallite size, estimated using the Scherrer equation, decreases progressively with increasing N content. Despite this size reduction, the overall crystal quality of the films remains preserved, suggesting that N incorporation does not significantly deteriorate the crystal structure but instead alters the microstructure in a controlled manner. Correspondingly, the calculated lattice parameters a and c exhibit a monotonic increase with increasing x, further supporting the substitution of O²⁻ by N³⁻ ions in the Cu₄O₃ lattice and confirming the formation of an N-doped phase.

Fig. 4 XRD patterns of Cu₄O₃-x films deposited with varying N₂ contents in a mixed sputtering gas.

Table 3 Structural parameters of Cu₄O₃-x films deposited with varying N₂ contents in a mixed sputtering gas

Samples	Lattice reflections of Cu4O3	Diffractive angle 2θ (degree)	Full width at half-maximum (degree)	Crystallite size D (nm)	Weight percent of the phase	Average crystallite size D (nm)	Lattice parameters (Å)
Cu4O3-0	(004)	36.16	0.30	27.87	0.7	28.07	c = 9.9332
	(220)	43.78	0.30	28.55	0.3		a = 5.8467
Cu4O3-10	(004)	36.10	0.30	27.86	0.77	26.78	c = 9.9491
	(220)	43.74	0.37	23.15	0.23		a = 5.8518
Cu4O3-20	(004)	36.04	0.39	21.43	0.47	21.71	c = 9.9652
	(220)	43.68	0.39	21.95	0.53		a = 5.8594
Cu4O3-30	(202)	35.30	0.41	20.34	0.98	20.26	c = 10.1298
	(220)	43.54	0.53	16.15	0.02		a = 5.8773
Cu4O3-40	(202)	34.90	0.49	17.00	1	17.00	—

---

3.4. Analysis of morphology via AFM

The microscale surface morphology of Cu₄O₃-x films, as observed by atomic force microscopy (AFM) in Fig. 5, indicates that the root mean square (RMS) surface roughness tends to decrease with increasing N content in the sputtering gas. However, this reduction in RMS does not correspond to better surface quality in all cases. For films deposited with N₂ concentrations up to 30%, both the RMS roughness and the surface grain distribution are favorable—smooth and uniform—making these films well-suited for interface-related applications. In contrast, the film produced at 40% N₂ exhibits inhomogeneous surface grain structures, despite having a lower RMS value. This inconsistency arises from N atoms substituting for O sites in the Cu₄O₃ lattice, which disturbs the microstructural development and leads to irregular surface grain growth. Therefore, among the tested conditions, the film deposited with 30% N₂ offers the best combination of surface smoothness and surface grain uniformity, making it the most appropriate choice for optoelectronic devices involving sensitive interfaces.

Fig. 5 AFM images of Cu₄O₃-x films deposited with varying N₂ contents in a mixed sputtering gas.

3.5. Analysis of morphology via FESEM

To comprehensively investigate the nanoscale morphology of the Cu₄O₃-x thin films, FESEM analysis was performed, capturing both top-view and cross-sectional images as shown in Fig. 6. These images reveal a uniform grain distribution across the surface and through the cross-section of the films. The surface morphology of Cu₄O₃-x films fabricated with N doping levels from x = 0 to 10 displays a characteristic tetrahedral grain structure, marked by sharp angles and well-defined edges. As the N₂ concentration in the sputtering mixed gas increases (x ≥ 20), there is a slight reduction in the sharpness of these angular features and edges; however, the overall grain shapes remain distinct and clearly defined, indicating that the crystalline integrity is largely preserved despite minor morphological smoothing. The cross-sectional FESEM images further reveal a well-developed columnar microstructure extending throughout the film thickness, mirroring the surface texture. The Cu₄O₃-x films were deposited with varying N₂ concentrations while maintaining a similar total thickness (∼280–300 nm) by adjusting the deposition time. The deposition rate decreased with increasing N₂ content in the reactive atmosphere, from approximately 0.83 nm s⁻¹ at 0% N₂ to 0.27 nm s⁻¹ at 40% N₂ (Fig. 7). This reduction is attributed to target poisoning effects and changes in plasma chemistry under higher N₂ flow conditions. Additionally, a progressive decrease in grain size with increasing N content is observed, aligning with the crystallite size trends obtained from XRD analysis discussed in Section 3.3. Overall, the structural characterization by XRD, AFM, and FESEM collectively indicates that the Cu₄O₃-x film fabricated at x = 30 exhibits an optimal combination of surface uniformity, low RMS, fine grain structure, and ordered columnar growth. This composition is particularly advantageous for applications requiring well-defined and stable interfacial features, such as in the integration of electrical or optoelectronic devices.

Fig. 6 FESEM images of Cu₄O₃-x films deposited with the varying N₂ contents in a mixed sputtering gas.

Fig. 7 Deposition rate of Cu₄O₃-x with varying N₂ percent (x) in a mixed sputtering gas.

3.6. Analysis of optical characteristics utilizing UV-vis spectroscopy

In addition to the structural properties, it is crucial to investigate the optical behavior of Cu₄O₃-x thin films in order to elucidate the influence of N incorporation on the electronic structure of the Cu₄O₃ lattice. UV-vis spectroscopy results, as presented in Fig. 8A, reveal that the films exhibit high optical transmittance, exceeding 70% across the visible spectrum. Notably, a red shift in the optical absorption edge is observed with increasing x, which signifies a progressive substitution of O atoms by N within the Cu₄O₃ lattice. This spectral shift is indicative of modifications in the electronic band structure, specifically due to the hybridization of N 2p orbitals with O 2p orbitals in the valence band. Such orbital mixing leads to an expansion of the valence bandwidth, thereby reducing the energy required for electronic transitions across the bandgap. This phenomenon is analogous to the band-edge shift mechanism previously reported for Cu₂O systems, as described in ref. 36 and 37. To quantitatively evaluate the effect of N doping on the band structure, the optical bandgap values were determined using the Tauc method, assuming a direct allowed transition model in accordance with prior studies on Cu₄O₃ (ref. 38–41). The Tauc plots, shown in Fig. 8B, clearly illustrate a systematic narrowing of the bandgap with increasing N content (increasing x). This reduction in bandgap occurs because the N atoms interact more strongly with the O atoms in the material. Their electron energy levels mix together, which increases the energy of the valence band and makes it easier for electrons to jump to the conduction band, effectively lowering the bandgap.

Fig. 8 (A) UV-vis transmittance spectra and (B) band gap of Cu₄O₃-x films deposited with varying N₂ contents in a mixed sputtering gas.

3.7. Analysis of electrical characteristics by Hall measurements

As previously discussed, Cu₄O₃-x thin films synthesized with varying N contents exhibit clear evidence of N atoms substituting for O within the lattice framework. This substitution leads to the formation of N-substituted O sites, which, in conjunction with intrinsic Cu vacancies, contribute to an increased hole concentration, as detailed in Table 4. However, despite the observed increase in hole concentration with increasing x, the overall electrical resistivity of the films does not continuously decrease. Instead, the minimum resistivity of 4.19 Ω cm is observed at x = 30, beyond which resistivity increases. This non-monotonic behavior is primarily attributed to a pronounced reduction in hole mobility in the film with x = 40 (2.04 cm² V⁻¹ s⁻¹) relative to that with x = 30 (4.48 cm² V⁻¹ s⁻¹). The decline in mobility at higher N content may result from increased lattice disorder or scattering centers introduced by excessive N incorporation, which impedes efficient charge carrier transport.

Table 4 The results of the Hall effect measurement of Cu₄O₃-x films deposited with varying N₂ contents in a mixed sputtering gas

Samples	Electrical resistivity (ρ) (Ω cm)	Hall mobility (μ) (cm^2 V^−1 s^−1)	Carrier concentration (n) (cm^−3)	Type
Cu4O3-0	64.07	7.26	1.34 × 10^16	p
Cu4O3-10	9.3	6.83	9.84 × 10^16	p
Cu4O3-20	6.63	4.55	2.07 × 10^17	p
Cu4O3-30	4.19	4.48	3.33 × 10^17	p
Cu4O3-40	5.68	2.04	5.39 × 10^17	p

---

3.8. Examination of the electrochemical properties of Cu₄O₃-x in a Na₂S polysulfide electrolyte

To evaluate the charge carrier exchange dynamics between Cu₄O₃-x thin films and the Sn²⁻/S²⁻ redox couple, cyclic voltammetry (CV) measurements were performed over a potential range of −1.6 V to 0.8 V with a scan rate of 10 mV s⁻¹. As shown in Fig. 9A, all CV curves exhibit distinct anodic and cathodic peaks, confirming the occurrence of reversible charge transfer processes between the films and the electrolyte. The key electrochemical parameters extracted from the CV plots—including anodic and cathodic peak currents, anodic and cathodic peak voltages, and the peak-to-peak voltage separation (ΔV_PP)—are summarized in Table 5. For films with x values from 0 to 30, both the anodic and cathodic currents, as well as the peak voltages, increase with increasing N content, while ΔV_PP exhibits a decreasing trend, suggesting enhanced electrochemical kinetics. Moreover, the film fabricated at x = 30 demonstrates the most favorable electrochemical performance, exhibiting the highest peak currents and the smallest ΔV_PP, indicative of the most efficient charge transfer characteristics. Consistently, Tafel polarization measurements further corroborate this observation: the film with x = 30 achieves the highest limited current density (J_lim) of 29 × 10⁻³ mA cm⁻² and exchange current density (J₀) of 6.9 × 10⁻³ mA cm⁻² as seen in Fig. 9C. These results imply that the Cu₄O₃-30 film possesses superior catalytic activity toward polysulfide redox reactions. Furthermore, it facilitates the most efficient diffusion of polysulfide species at the counter electrode (CE)/electrolyte interface, underscoring its potential as an effective CE material for polysulfide-based electrochemical systems. This is confirmed further by the CV plot (Fig. 9B) of Cu₄O₃-30 with 50 cycles which shows a weak current decreasing trend of 0.015 mA cm⁻².

Fig. 9 (A) Cyclic voltammetry; (B) Tafel polarization of Cu₄O₃-x films deposited with varying N₂ concentrations in a mixed sputtering gas; and (C) cyclic voltammetry of Cu₄O₃-30 film subjected to 50 cycles.

Table 5 Parameters of cyclic voltammetry and Tafel electrochemical characteristics of Cu₄O₃-x/FTO CE films with varying N₂ concentrations in a mixed sputtering gas

Counter electrodes	J o(×10^−3) (mA cm^−2)	J lim(×10^−3) (mA cm^−2)	J PC (mA cm^−2)	J PA (mA cm^−2)	ΔVPP (V) versus Ag/AgCl
Cu4O3-0/FTO	2.3	6	−11.07	24.38	0.90
Cu4O3-10/FTO	4.3	13	−20.71	35.25	0.84
Cu4O3-20/FTO	5.1	18	−25.74	49.44	0.76
Cu4O3-30/FTO	6.9	29	−31.58	66.15	0.59
Cu4O3-40/FTO	5.6	23	−28.18	56.10	0.69

---

3.9. EIS spectra and J–V characteristics of CdS/CdSe:Cu/ZnS QDSSCs with Cu₄O₃-x CEs

A schematic of the quantum dot-sensitized solar cell (QDSSC) structure, designed specifically to investigate electrochemical impedance spectroscopy (EIS) and current–voltage (J–V) characteristics, is illustrated in Fig. 10A. These devices are configured as TiO₂/CdS/CdSe:Cu/ZnS/QDSSCs employing Cu₄O₃-x counter electrodes (CEs). The fabrication methodology for the TiO₂/CdS/CdSe:Cu/ZnS photoanodes closely follows the procedures detailed in ref. 32. The Nyquist plots obtained from EIS measurements, presented in Fig. 10B, reveal two distinct semicircles for all tested devices. The appearance of two semicircles is attributed to two separate interfacial processes: the smaller high-frequency semicircle corresponds to the charge transfer resistance at the CE/electrolyte interface, while the larger mid-frequency semicircle originates from charge recombination processes at the photoanode. Based on these observations, appropriate equivalent circuit models were developed and are shown in Fig. 10C. Key electrochemical parameters were extracted by fitting the EIS data using Autolab software, with the results summarized in Table 6. In this analysis, the resistance R_ct2 represents the charge transfer resistance at the CE/electrolyte interface, directly reflecting the catalytic activity of the counter electrode. In contrast, R_ct1 corresponds to the resistance associated with electron recombination at the TiO₂/quantum dot (QD) interface. The principal emphasis of this study is the efficacy of counter electrodes, with particular attention devoted to the fluctuation of R_ct2 values. The results indicate that the Cu₄O₃-30 counter electrode exhibits the lowest R_ct2 value among all samples, signifying the most efficient and reversible electrochemical charge transfer process. Interestingly, a similar trend is observed for R_ct1, despite the identical fabrication conditions for all QD photoanodes, suggesting a synergistic improvement possibly related to the overall interfacial properties of the device. Overall, the Cu₄O₃-30 counter electrode's better electrochemical performance, as evidenced by cyclic voltammetry (CV), Tafel polarization, and Nyquist analyses, directly correlates with the highest photovoltaic efficiency of 7.29% observed for the solar cells utilizing the Cu₄O₃-30 CE, as summarized in Table 6 and shown in Fig. 10D. This confirms the critical role of optimized Cu₄O₃-30 counter electrodes in enhancing the performance of polysulfide-based QDSSCs. The optimized Cu₄O₃-30 CE in this work delivers a PCE of 7.29%, surpassing most reported Cu₂S/Cu_xS-based counterparts (1.68–5.59%) in comparable device architectures as seen in Table 7, owing to its superior electrocatalytic activity and lower charge-transfer resistance. We attribute the enhanced performance to the unique morphology and optoelectronic characteristics of the Cu₄O₃ films, including improved light absorption and interfacial charge transfer.

Fig. 10 (A) Diagram of solar cells; (B) Nyquist plots (C) the equivalent circuit; (D) J–V curves.

Table 6 Parameters of EIS electrochemical characteristics and solar cell efficiency based on Cu₄O₃-x/FTO CE films with different N₂ contents in a mixed sputtering gas

Sample	Open voltage (Voc, V)	Short current density (Jsc, mA cm^−2)	Fill factor (FF)	Efficiency (η, %)	R s (Ω)	R ct1 (Ω)	R ct2 (Ω)
Cu4O3-0	0.50	8.05	0.52	∼2.09	38.16	284.7	70.04
Cu4O3-10	0.50	13.57	0.53	∼3.60	38.58	248.9	58.12
Cu4O3-20	0.51	15.98	0.53	∼4.32	38.77	150.3	48.23
Cu4O3-30	0.51	25.51	0.56	∼7.29	38.90	34.9	15.14
Cu4O3-40	0.51	19.88	0.51	∼5.17	38.86	99.4	36.42

---

Table 7 Photovoltaic performance of the Cu₄O₃-30 counter electrode compared with various Cu₂S/Cu_xS-based CEs in QDSSCs

CE/source	Photoanode structure	J sc (mA cm^−2)	V oc (V)	FF (%)	η (%)	References
Cu4O3	FTO/TiO2/CdS/CdSe/ZnS	25.51	0.51	56.0	7.29	This work
Cu2S (electrodeposited)	FTO/TiO2/CdS/CdSe	19.60	0.45	48.62	4.24	42
Cu2S (bare)	FTO/TiO2/CdS/CdSe	15.84	0.56	50.07	4.35	42
Cu2S/PbS (3 SILAR cycles)	FTO/TiO2/CdS/CdSe/PbS	18.08	0.55	53.55	5.28	43
ATO/CuS	ATO/TiO2/CdS/CdSe	20.22	0.57	41.49	4.79	44
ATO/PbSe	ATO/TiO2/CdS/CdSe/PbSe	17.18	0.59	55.06	5.59	45
Cu2S/Cu2Se composite	FTO/TiO2/CdS/CdSe	19.18	—	—	4.60	46
CuInS2–Cu2S nanocomposite	FTO/TiO2/CdS/CdSe	4.74	0.73	58.0	2.01	47
Se-doped CuS	FTO/TiO2/CdS/CdSe	6.74	0.37	67.10	1.68	48

---

4. Conclusion

Cu₄O₃-x thin films were successfully deposited under varying nitrogen (N₂) concentrations in the sputtering gas mixture to systematically optimize their structural, optical, and electrical properties. Among the fabricated samples, the Cu₄O₃-30 film, synthesized with 30% N₂ in the sputtering atmosphere, exhibited the most favorable characteristics. This film demonstrated an optimal balance between crystallinity and morphology, with an average crystallite size of approximately 20 nm, a surface roughness (RMS) of 2.06 nm, and a grain size of about 50 nm as observed by FESEM. In terms of optical properties, the Cu₄O₃-30 film possessed a direct optical bandgap of 2.18 eV. Regarding electrical performance, it achieved a resistivity of 4.19 Ω cm, a hole concentration of 3.33 × 10¹⁷ cm⁻³, and a hole mobility of 4.48 cm² V⁻¹ s⁻¹. Leveraging these superior properties, the Cu₄O₃-30 film was employed as a CE material in CdS/CdSe:Cu/ZnS/TiO₂ quantum dot-sensitized solar cells (QDSSCs). The resulting devices exhibited a remarkable power conversion efficiency (PCE) of 7.29%, primarily ascribed to the superior catalytic performance and effective charge transfer at the CE/electrolyte interaction. Notably, this efficiency surpasses that of comparable QDSSCs utilizing conventional Cu₂S counter electrodes, highlighting the significant potential of optimized Cu₄O₃-x films for high-performance photovoltaic applications.

Conflicts of interest

There are no conflicts to declare.

Data availability

All data relevant for the reproduction of the results presented in this work are included within the article.

Acknowledgements

This research is funded by Vietnam National University, Ho Chi Minh City (VNU-HCM) under grant number C2025-18-08.

References

1. W. Maeng, S.-H. Lee, J.-D. Kwon, J. Park and J.-S. Park, Atomic layer deposited p-type copper oxide thin films and the associated thin film transistor properties, Ceram. Int., 2016, 42, 5517–5522.
2. A. Lakshmanan, Z. C. Alex and S. R. Meher, Cu₂O thin films grown by magnetron sputtering as solar cell absorber layers, Mater. Sci. Semicond. Process., 2022, 148, 106818.
3. F. Yang, W. Peng, Y. Zhou, R. Li, G. Xiang, J. Z. YueLiu, J. Zhang, Y. Zhao and H. Wang, Thermal optimization of defected Cu₂O photon-absorbing layer and the steady p-Cu₂O/n-Si photovoltaic application, Vacuum, 2022, 198, 110876.
4. J.-S. Yoon, J.-W. Lee and Y.-M. Sung, Enhanced photoelectrochemical properties of Z-scheme ZnO/p–n Cu₂O PV-PEC cells, J. Alloys Compd., 2019, 771, 869–876.
5. M. R. Dustgeer, S. T. Asma, A. Jilani, K. Raza, S. Z. Hussain, M. B. Shakoor, J. Iqbal, M. Sh. Abdel-wahab and R. Darwesh, Synthesis and characterization of a novel single-phase sputtered Cu₂O thin films: Structural, antibacterial activity and photocatalytic degradation of methylene blue, Inorg. Chem. Commun., 2021, 128, 108606.
6. X. Wu, J. Liu, P. Huang, Z. Huang, F. Lai, G. Chen, L. Lin, P. Lv, W. Zheng and Y. Qu, Engineering crystal orientation of p-Cu₂O on heterojunction solar cells, Surf. Eng., 2017, 33, 542–547.
7. S. Aseena, N. Abraham, G. Sahaya Dennish Babu, S. Kathiresan and V. Suresh Babu, Solution-Synthesized Cu₂O As a Hole Transport Layer for a ZnO-Based Planar Heterojunction Perovskite Solar Cell Fabricated at Room Temperature, J. Electron. Mater., 2022, 51, 1692–1699.
8. D. Prasanth, K. P. Sibin and H. C. Barshilia, Optical properties of sputter deposited nanocrystalline CuO thin films, Thin Solid Films, 2019, 673, 78–85.
9. H. Güney, D. İskenderoğlu, M. E. Güldüren, K. Ç. Demi̇r and S. M. Karadeniz, An investigation on CuO thin films grown by ultrasonic spray pyrolysis at different substrate temperatures: Structural, optical and supercapacitor electrode characterizations, Opt. Mater., 2022, 132, 112869.
10. P. Venkateswari, P. Thirunavukkarasu, M. Ramamurthy, M. Balaji and J. Chandrasekaran, Optimization and characterization of CuO thin films for P–N junction diode application by JNSP technique, Optik, 2017, 140, 476–484.
11. A. H. Shukor, H. A. Alhattab and I. Takano, Electrical and optical properties of copper oxide thin films prepared by DC magnetron sputtering, J. Vac. Sci. Technol., B: Nanotechnol. Microelectron.: Mater., Process., Meas., Phenom., 2020, 38, 012803.
12. Y. Akaltun, Effect of thickness on the structural and optical properties of CuO thin films grown by successive ionic layer adsorption and reaction, Thin Solid Films, 2015, 594, 30–34.
13. L. Radjehi, L. Aissani, A. Djelloul, S. Lamri, K. Nomenyo, S. Achache, G. Lerondel and F. Sanchette, Effect of vacuum annealing on the structural and optical properties of sputtered Cu₄O₃ thin films, Surf. Eng., 2021, 37, 422–428.
14. M. A. M. Patwary, C. Y. Ho, K. Saito, Q. Guo, K. M. Yu, W. Walukiewicz and T. Tanaka, Effect of oxygen flow rate on properties of Cu₄O₃ thin films fabricated by radio frequency magnetron sputtering, J. Appl. Phys., 2020, 127, 085302.
15. M. A. M. Patwary, K. Saito, Q. Guo, T. Tanaka, K. Man Yu and W. Walukiewicz, Nitrogen Doping Effect in Cu₄O₃ Thin Films Fabricated by Radio Frequency Magnetron Sputtering, Phys. Status Solidi B, 2020, 257, 1900363.
16. D. S. Murali and A. Subrahmanyam, Synthesis of low resistive p type Cu₄O₃ thin films by DC reactive magnetron sputtering and conversion of Cu₄O₃ into CuO by laser irradiation, J. Phys. D: Appl. Phys., 2016, 49, 375102.
17. L. Yang, R. Wang, D. Chu, Z. Chen, F. Zhong, X. Xu, C. Deng, H. Yu and J. Lv, BiVO₄ photoelectrodes for unbiased solar water splitting devices enabled by electrodepositing of Cu₂O simultaneously as photoanode and photocathode, J. Alloys Compd., 2023, 945, 169336.
18. Y. Wang, N. Yu and Y. Wu, Cu₂O/GaN ultraviolet detector synthesized using ZnO nanorod arrays as template by aqueous method, Nano-Struct. Nano-Objects, 2020, 23, 100494.
19. T. Minami, H. Tanaka, T. Shimakawa, T. Miyata and H. Sato, High-Efficiency Oxide Heterojunction Solar Cells Using Cu₂O Sheets, Jpn. J. Appl. Phys., 2004, 43, L917.
20. T. Minami, Y. Nishi and T. Miyata, Efficiency enhancement using a Zn_1−xGe_x-O thin film as an n-type window layer in Cu₂O -based heterojunction solar cells, Appl. Phys. Express, 2016, 9, 052301.
21. D.-C. Perng, M.-H. Hong, K.-H. Chen and K.-H. Chen, Enhancement of short-circuit current density in Cu₂O/ZnO heterojunction solar cells, J. Alloys Compd., 2017, 695, 549–554.
22. Y. Liu, J. Zhu, L. Cai, Z. Yao, C. Duan, Z. Zhao, C. Zhao and W. Mai, Solution-Processed High-Quality Cu₂O Thin Films as Hole Transport Layers for Pushing the Conversion Efficiency Limit of Cu₂O/Si Heterojunction Solar Cells, Sol. RRL, 2020, 4, 1900339.
23. K. Eom, D. Lee, S. Kim and H. Seo, Modified band alignment effect in ZnO/Cu₂O heterojunction solar cells via Cs₂O buffer insertion, J. Phys. D: Appl. Phys., 2018, 51, 055101.
24. V. S. Nguyen, A. Sekkat, D. Bellet, G. Chichignoud, A. Kaminski-Cachopo, D. Muñoz-Rojas and W. Favre, Open-air, low-temperature deposition of phase pure Cu₂O thin films as efficient hole-transporting layers for silicon heterojunction solar cells, J. Mater. Chem. A, 2021, 9, 15968–15974.
25. C. Qin, Y. Wang, Z. Lou, S. Yue, W. Niu and L. Zhu, Surface modification and stoichiometry control of Cu₂O/SnO₂ heterojunction solar cell by an ultrathin MgO tunneling layer, J. Alloys Compd., 2019, 779, 387–393.
26. T. Minami, Y. Nishi and T. Miyata, Cu₂O-based solar cells using oxide semiconductors, J. Semicond., 2016, 37, 014002.
27. T. Minami, Y. Nishi and T. Miyata, Heterojunction solar cell with 6% efficiency based on an n-type aluminum–gallium–oxide thin film and p-type sodium-doped Cu₂O sheet, Appl. Phys. Express, 2015, 8, 022301.
28. K. Sahu, A. Bisht, S. A. Khan, I. Sulania, R. Singhal, A. Pandey and S. Mohapatra, Thickness dependent optical, structural, morphological, photocatalytic and catalytic properties of radio frequency magnetron sputtered nanostructured Cu₂O–CuO thin films, Ceram. Int., 2020, 46, 14902–14912.
29. H.-H. Zeng, A. S. Rasal, M. Alemayehu Abate, A. Vithal Ghule and J.-Y. Chang, Surface and interfacial engineering for aqueous-processed quantum dots-sensitized solar cell with efficiency approaching 11%, Chem. Eng. J., 2024, 495, 153702.
30. L. G. Tartuci, E. Raphael, J. A. Carvalho Junior, W. S. Machado and M. A. Schiavon, Counter Electrodes Based on Polyaniline/carbon Nanotube/Cu_xS Nanocomposites for CdS Quantum Dots-Sensitized Solar Cells, Energ. Tech., 2025, 2402251.
31. Z. Luo, W. Yin, J. Wang, Y. Hua, Z. Zhou, W. Zhang, S. Chen, X. Zhang and W. Zheng, Quantum Dots Enable Perovskite Solar Cells Performance: Interactions, Mechanisms, Progresses, and Future Perspectives, Adv. Funct. Mater., 2025, 35, 2419268.
32. D. H. Phuc and H. T. Tung, Quantum dot sensitized solar cell based on the different photoelectrodes for the enhanced performance, Sol. Energy Mater. Sol. Cells, 2019, 196, 78–83.
33. K. Zhao, H. Yu, H. Zhang and X. Zhong, Electroplating Cuprous Sulfide Counter Electrode for High-Efficiency Long-Term Stability Quantum Dot Sensitized Solar Cells, J. Phys. Chem. C, 2014, 118, 5683–5690.
34. E. J. Radich, R. Dwyer and P. V. Kamat, Cu₂S Reduced Graphene Oxide Composite for High-Efficiency Quantum Dot Solar Cells. Overcoming the Redox Limitations of S^2–/S_n^2– at the Counter Electrode, J. Phys. Chem. Lett., 2011, 2, 2453–2460.
35. B. V. Crist, Handbooks of Monochromatic XPS Spectra – the Elements and Native Oxides, XPS Inter-national LLC, Mountain View (CA, USA), 1999, vol. 1.
36. S. Hassaballa, A. Aljabri, S. H. Mohamed, A. M. Bakry, A. M. A. El-Rahman and M. A. Awad, Photocatalysis, wettability and optical properties of N-doped Cu₂O/CuO thin films for smart coating applications, Phys. Scr., 2024, 99, 115974.
37. M. Li, J.-Y. Zhang, Y. Zhang and T.-M. Wang, Oxygen vacancy in N-doped Cu₂O crystals: A density functional theory study, Chin. Phys. B, 2012, 21, 087301.
38. A. Thobor and J. F. Pierson, Properties and air annealing of paramelaconite thin films, Mater. Lett., 2003, 57, 3676–3680.
39. J. F. Pierson, D. Wiederkehr and A. Billard, Reactive magnetron sputtering of copper, silver, and gold, Thin Solid Films, 2005, 478, 196–205.
40. J. F. Pierson, A. Thobor-Keck and A. Billard, Cuprite, paramelaconite and tenorite films deposited by reactive magnetron sputtering, Appl. Surf. Sci., 2003, 210, 359–367.
41. J. F. Pierson, E. Duverger and O. Banakh, Experimental and theoretical contributions to the determination of optical properties of synthetic paramelaconite, J. Solid State Chem., 2007, 180, 968–973.
42. J. Wang, M. M. Rahman, C. Ge and J.-J. Lee, Electrodeposition of Cu₂S nanoparticles on fluorine-doped tin oxide for efficient counter electrode of quantum-dot-sensitized solar cells, J. Ind. Eng. Chem., 2018, 62, 185–191.
43. I.-R. Jo, J. A. Rajesh, Y.-H. Lee, J.-H. Park and K.-S. Ahn, Enhanced electrocatalytic activity and electrochemical stability of Cu₂S/PbS counter electrode for quantum-dot-sensitized solar cells, Appl. Surf. Sci., 2020, 525, 146643.
44. L. Zhou, H. L. Ren, C. Q. Yang, Y. X. Wu and B. B. Jin, ATO/CuS composite counter electrodes enhanced the photovoltaic performance of quantum dot sensitized solar cells, Inorg. Chem. Commun., 2022, 138, 109273.
45. B. B. Jin, H. S. Huang, S. Y. Kong, G. Q. Zhang, B. Yang, C. X. Jiang, Y. Zhou, D. J. Wang and J. H. Zeng, Antimony tin oxide/lead selenide composite as efficient counter electrode material for quantum dot-sensitized solar cells, J. Colloid Interface Sci., 2021, 598, 492–499.
46. Y. Chen, D. Wang, Y. Lin, X. Zou and T. Xie, An in situ inward etching strategy for constructing a p-p heterojunction Cu₂S/Cu_2-xSe material based on brass as an effective counter electrode for quantum dot sensitized solar cells, J. Power Sources, 2019, 442, 227222.
47. M. Mousavi-Kamazani, Z. Salehi and K. Motevalli, Enhancement of quantum dot-sensitized solar cells performance using CuInS₂–Cu₂S nanocomposite synthesized by a green method, Appl. Phys. A, 2017, 123, 691.
48. G. Vinoth, S. Abinaya and M. R. Kadiresan, Hydrothermal Synthesis of Se-Doped CuS Quantum Dots with Respect to the DW/EN Solvent Ratio and Application as a Counter Electrode for Quantum Dot-Sensitized Solar Cells, Braz. J. Phys., 2023, 53, 1.

---