Structure and chemical composition of the CuO nanoparticles and films
To determine the crystal structure and phase composition of the CuO NPs obtained after solvent evaporation at 120°C in a dry oven and the CuO film prepared by the spin-coating method (Fig. 2(a)), X-ray diffraction (XRD) patterns were recorded for all synthesised samples, as shown in Fig. 2(b). The XRD patterns of the CuO nanopowders were identical to that of monoclinic single-phase CuO, and the diffraction data corresponded well with that of the CuO JCPDS card (JCPDS 45–0937), with no impurity peaks present. Furthermore, the primary crystallite size (~ 20 nm) determined by the Pawley method, combined with the results of XRD analysis, confirmed the material dispersed in the suspension to be CuO NPs. Interestingly, a (220) peak, corresponding to Cu2O (JCPDS 71-3645), was observed at ~ 42° for the spin-coated films, as shown in Fig. 2(c). This peak may indicate that Cu2O is formed in the outer layers of the film owing to the partial reduction of Cu(II) to Cu(I).
X-ray photoelectron spectroscopy (XPS) was employed to investigate the chemical composition of the CuO NP powders and CuO films prepared on the ITO substrates (Supplementary Table S1). The XPS survey spectra of the CuO films showed no impurity peaks, and exhibited only the peaks corresponding to Cu, O, and C. Figures 3(a-1) and (b-1) show the high-resolution spectra of Cu 2p, which are separated into the Cu 2p3/2 and Cu 2p1/2 peaks observed at approximately 932.3 ± 0.1 eV and 952.2 ± 0.3 eV, respectively42,43. The distance between these Cu 2p main peaks was 19.9 eV, which agreed with that reported for the CuO spectrum44. Moreover, Figs. 3(a-1) and (b-1) clearly show that the Cu 2p spectra contain “shake-up” satellites, which are characteristic of the Cu2+ state43. These satellite peaks are deviated by ~ 9 eV from the main peak (the binding energies (Eb) are approximately 941 and 962 eV for Cu 2p3/2 and Cu 2p1/2, respectively). Intense peaks corresponding to Cu2+ 2p3/2 and Cu2+ 2p1/2 were observed for the CuO NPs, indicating that the prepared NP sample contained CuO (Fig. 3 (a-1)). In contrast, intense Cu+ 2p3/2 and Cu+ 2p1/2 peaks were observed for the CuO film prepared on the ITO substrate. This indicates that the film primarily contains CuO, and the outer layer contains Cu2O, in which Cu(II) partially oxidised to Cu(I)45. This was in agreement with the XRD results. The O 1s XPS profiles are presented in Figs. 3(a-2) and (b-2). A clear peak is observed at 530.3 ± 0.2 eV, which can be indexed to O2− in CuO46. Notably, three other weak O 1s peaks are also present. The peak located at 531.7 ± 0.2 eV originates due to surface hydroxyls46,47, while the peaks at 532.5 and 533.8 eV correspond to C = O and C–O, respectively48. The latter peaks can be attributed to the citric acid and PVA used in the preparation of the dispersed CuO NP suspension. The C 1s signal of the CuO films (284.8 eV) in Figs. 3(a-3) and (b-3) can be attributed to the C–C bonds of PVA and citric acid45,49.
To investigate the effect of PVA addition on the film-forming characteristics of CuO NP suspensions, field-emission scanning electron microscopy (FESEM) was used to study their morphological features. Supplementary Fig. S2 shows the surface and cross-sectional FESEM images of 1 wt.% PVA CuO films. From the cross-sectional image shown in Supplementary Fig. S2(a), the CuO films are approximately 200 nm thick and exhibit good adhesion between the ITO substrates and CuO coatings. Furthermore, the surface morphology seen in Supplementary Fig. S2(b) indicates that the film exhibits a porous structure. This porous structure could enhance the Li+ diffusion in the film and contribute to its cationic conductivity50.
Electrochemical and electrochromic properties
The electrochemical and EC properties of the CuO films deposited on the ITO substrates were analysed using a three-electrode chemical cell containing aqueous LiClO4/PC (1 mol/kg) as the electrolyte. The cyclic voltammetry (CV) measurements were conducted in the potential range from − 1.8 to + 1.3 V vs. Ag/AgCl at a sweep rate of 5 mV/s. The results are shown in Fig. 4(a) and Supplementary Fig. S3. The CuO film exhibits a complex redox reaction, and we propose the following 4 steps of electrochemical reaction mechanism. 1. The prepared thin film is in a state in which CuO and Cu2O are a mixture at the beginning of the reaction. 2. A cathodic peak51, which corresponds to Cu2O + 2CuO + 2Li+ + 2e− → 2Cu2O + Li2O52–54, appeared at approximately − 0.5 V, and the transmittance increased slightly to − 1.2 V. This is attributed to the reduction of CuO present in the thin film to Cu2O, and is consistent with the XPS results described above. 3. Furthermore, an oxidation reaction, which corresponding to 2Cu2O + 2Li2O → 4CuO + 4Li+ + 4e− 52, was initiated at approximately − 1.8 V, which caused the film color to change to dark. A peak corresponding to CuO was observed in the XPS results of the black CuO thin film (Supplementary Fig. S4), and it is considered that Cu2O55 and Li2O56 did not fully participate. Finally, 4. it is thought that the film becomes CuO state due to 4CuO + 2Li+ + 2e− → 2CuO + Cu2O + Li2O at approximately − 0.2 V. The transferred charge density (ΔQ), which indicates the electrochemical activity of the CuO film, was determined by performing chronocoulometry (CC) at constant applied potentials of − 1.8 and + 1.3 V (vs. Ag/AgCl) for 90, 120, and 120 s, respectively, to allow sufficient time for the complete redox reaction in each film (Supplementary Fig. S5). A ΔQ value of 68 mC/cm2 was observed for the CuO film.
Changes in the optical transmittance spectra of the as-deposited, transparent brown, and dark grey colour states of the CuO films were studied using ultraviolet–visible (UV–Vis) spectroscopy during CC analysis, as shown in Fig. 4 (b). The change in the transmittance of the CuO film with a thickness of 200 nm at + 1.3 and − 1.8 V was approximately 63% at a wavelength of 633 nm. However, the transmittance of the 150-nm-thick film in the coloured state increased, and the transmittance of the 300-nm-thick film in the bleached state decreased (Supplementary Fig. S6). Therefore, the film thickness of 200 nm was considered the most appropriate for application as an EC material. Moreover, the optical density (ΔOD) was used to calculate the EC colouration efficiency (CE), which is an important EC parameter that determines the electrochemical performance of the films (Supplementary Table S2). This parameter is defined in equations (3) and (4):
$$CE= \frac{\varDelta OD}{\varDelta Q}$$
3
$${\Delta }OD\left(\lambda \right)={\text{l}\text{o}\text{g}}_{10}({T}_{\lambda b}/{T}_{\lambda c})$$
4
where λ = 633 nm, and Tλb and Tλc represent the transmittance of the bleached and coloured states, respectively. The ΔOD and CE values obtained for the CuO films were 1.04 and 15.36 cm2/C, respectively.
The band gaps (Eg) of the films in the as-deposited, transparent brown, and dark grey states can be estimated using the transmittance data and estimated film thickness (d = 200 nm). Figure 4(c) shows a graphical representation of (αhv)2 vs. photon energy (hv) for these states, which can be evaluated using the following Eq. 51,57:
$$\alpha =\frac{1}{d}ln\frac{100}{T}$$
5
where α, d, and T are the absorption coefficient, film thickness, and transmittance value, respectively. The optical density can be used to obtain the band gap energy of the transparent films by plotting (αhν)1/η against hν, based on the following relation58–60:
$$\alpha h\nu =A{(h\nu -{E}_{g})}^{\eta }$$
6
where A and η are constants, and η depends on the nature of the transition. For copper-oxide-based materials, it is assumed that η = ½, which corresponds to a directly allowed electron transition mechanism45,47,49. Therefore, in the initial absorption region (hν ≈ Eg), where the plot of (αhν)2 vs. hν is linear, the intercept of the extrapolated fitted line with the hν axis gives the optical band gap energy, as shown in Fig. 4(c). Direct optical band gap energies for the Cu2O and CuO films, which have been reported to be in relatively wide ranges of approximately 2.1–2.6 eV and 1.3–1.7 eV, respectively51,58, 60–65, are dependent on the fabrication method and stoichiometry. The experimentally obtained optical band gap values were compared with the reported data, as shown in Fig. 4(d). The transmittance in the as-deposited state, the transparent brown state, and the dark grey state corresponded to the Cu2O + CuO compound film, Cu2O, and CuO + Cu films, respectively. The colour details were also evaluated based on the band gap change using the International Commission on Illumination (CIE: usually abbreviated CIE for its French name, Commission Internationale de l'Eclairage) 1976 L*a*b* colour model. The L* and b* values decreased from 83 to 32 and from 34 to 9, respectively, and the a* value increased from − 5 to − 2 as Cu2O was oxidised to CuO, which is indicative of a nearly dark grey colour (Supplementary Fig. S7). Several groups previously reported the chromaticity of dark organic EC materials prepared by different techniques. For instance, Wu et al.66 observed that the L* values of amine-based organic EC materials decreased from 92 to 6 upon changing from the bleached to the coloured state. Further, Liu et al.34 also reported a change in L* between 88 and 7 with an organic EC device composed of polyamide and viologen.
Here, we summarise the superiority of the results of this study by comparing them with those of existing studies. Although organic materials exhibit excellent EC properties, they have several disadvantages, including the use of harmful substances and complicated manufacturing processes, as well as poor durability against external stimuli (UV, temperature, etc.). In contrast, the CuO ink used in this study can be synthesised simply by mixing basic copper(II) carbonate with citric acid, which enabled coating CuO thin films. Furthermore, the CuO thin film prepared in this study showed better EC properties than any previously reported CuO thin films in Li-based electrolytes and showed low transmittance at visible wavelengths of 380–780 nm, which is comparable to that of organic EC materials. Moreover, as a nanomaterial, the CuO ink in this study has many potential applications at the cutting edge of science and technology, including catalysis67, gas sensors68, photovoltaic cells69, light-emitting diodes70, magnetic phase transitions71, and superconductors72.