Introduction

Self-assembly is a process observed in nature at different dimensions. During self-assembly, spontaneous ordering of atoms/molecules occurs over a broad length scale ranging from angstrom to centimeters in different dimensions without human intervention which makes the technique industrially relevant for the electronics and optoelectronics industry. Because of its ubiquitous occurrence in natural systems, it favors the bottom-up growth of materials. These features make it one of the most promising, low-cost, high-throughput, and practically scalable methodologies for nanofabrication [1]. Recently, there has been remarkable attention on metal oxide semiconductors (MOS) due to their optical, electrical, and thermal properties that make them unavoidable in technological frontiers like nano-electronics, solar cells, sensors, and photo catalysis. They are highly sensitive and have a very short response time when engineered in the nanoscale [2]. MOS are valence compounds with a high degree of ionic bonding. Their conduction band minimum (CBM) and valence band maximum (VBM) mainly consist of the metal (M) ns and oxygen (O) 2p orbital, respectively [3]. The interaction between the metal and oxide orbitals results in better electron transport in comparison to hole transport.

Copper oxide is an industrially important material with a wide range of applications in solar energy conversion, gas and biosensor technology, energy storage media, spintronics, optoelectronics, catalysis, biomedicine, batteries, super-capacitors, and high-temperature superconductors [4, 5]. The most common, stable, and non-toxic phases of copper oxides are cupric oxide (CuO) and cuprous oxide (Cu2O) [6]. The optical band gap of CuO ranges from 1.2 to 1.9 eV while the band gap of Cu2O ranges from 2.0 to 2.6 eV [6]. Cu2O has a high absorption coefficient in the visible region and is suitable for applications in solar cells [7]. However, Cu2O is thermally unstable and turns into CuO with heat treatment, so this latter phase is preferred for applications in solar cells operating under high-temperature conditions [8]. The CuO thin films have a p-type semiconductor property with a narrow optical bandgap and a high absorption coefficient in the visible and near-infrared (NIR) region, making the CuO thin film suitable as an absorber layer for solar cell applications [9]. The surface oxygen adsorption capacity of CuO is quite high so it can be effectively used for gas-sensing applications [10].

Very few reports on the growth of CuO thin films using wet chemical oxidative mechanisms are available in the literature that focuses on their optoelectronic applications [11]. Most of the works on wet chemistry-grown CuO are concentrated on super-capacitor application, photo-catalysis, antimicrobial applications, etc. [4, 12]. Most of the previous reports on alkaline wet chemical oxidation mechanisms are based on the use of oxidizing agents of a concentration lower than 0.5 M [11]. Aqueous KOH is alkaline in nature and dissociates to form hydroxide ions. Hence, it yields a substitution reaction when water is used as the solvent. Reports on the room temperature growth of CuO on a Cu foil using aqueous KOH are inadequate and this report is a harbinger in this area. In this work, we report on the photoconductivity of self-assembled CuO nano-structured thin films grown over Cu foil and explore its applicability for ultraviolet and visible light photo-detection.

Materials and methods

Synthesis of CuO thin films

All reagents used in this experiment require no further purification as they are of analytical (AR) grade. At first, flexible copper foil (thickness 0.25 mm, 99.98%) is cleaned using 4 M HCL solution for 15 min. After that, the foil is washed with an alkaline cleansing solution and deionized water. Cleaned copper substrates are dried in air at 60 °C for 15 min. Subsequently, 0.1 M solution of potassium hydroxide (KOH, Sigma Aldrich, 99.95%) was prepared in 100 ml distilled water. The treated copper foils were immersed into the KOH solution at room temperature and kept undisturbed for 6 days in a dark room maintained under ambient air conditions. A dark black uniform film appears over the copper foils over a period of time. At the end of the process, the copper sheets were carefully washed several times with deionized water and dried at 60 °C for 5 h. A schematic representation of thin film formation is shown in Fig. 1. Comparative studies were carried out by changing the molarity of the KOH solution to 0.1 M, 0.5 M, 1 M, 1.5 M, and 2 M with sample codes CuO0.1 M, CuO0.5 M, CuO1 M, CuO1.5 M, and CuO2 M, respectively.

Fig. 1
figure 1

Schematic image of a CuO nanostructure self-assembly and b photodetector

Full size image

Device engineering

A metal–semiconductor–metal (MSM) structure was developed for photodetector application. Silver (Ag) was used to make metallic contact over CuO surface. Ag, hence, serves as the front electrode. The base copper sheet serves as the back contact for the photodetector device engineered by us. A schematic representation of the Cu-CuO/Ag device structure is shown in Fig. 1b. Current–voltage measurements were taken under dark and visible illumination conditions. For visible light illumination, a halogen source of power density 70 W/m2 was used.

Experimental characterizations

X-ray diffraction (XRD) analysis was carried out using Bruker D8 Advance Diffractometer (Cu Ka radiation l = 1.5406 Ao) incorporated with a Lynx Eye position–sensitive detector with a step size of 0.02°. Micro-Raman spectra were recorded using a LabRAM, Horiba with a laser excitation source of wavelength 532 nm. Elemental composition confirmation was carried out using X-ray photoelectron spectroscopy with Thermo Scientific ESCALAB Xi + XPS system with Al-Kα radiation (1486.6 eV) as the source of excitation. The surface morphology was recorded using JEOL 6390LA/OXFORD XMX N. Diffused reflectance spectroscopy (DRS) studies were conducted using a JASCO V-750 double-beam spectrophotometer with a photomultiplier tube detector. Current–voltage characteristics as well as photodetector properties were studied using a Keysight B2910BL source measuring unit.

Results and discussions

Figure 2 describes the observed relationship between thin film thickness at the end of 6 days and the KOH concentration used. It is observed that the thickness of the deposited CuO thin films has a direct relationship with the KOH concentration when the number of days of self-assembly was kept fixed. The measurements show the thickness of the films is in the range of ~ 1.3–2.1 µm. Beyond the 6-day growth period, we observed that the films were beginning to peel off from the substrate.

Fig. 2
figure 2

Variation of CuO thin film thickness with KOH concentration

Full size image

XRD patterns of CuO thin films grown using different KOH molarity conditions are shown in Fig. 3. For all of the five samples, XRD peaks were observed at 2θ values of 32.56°, 35.58°, 38.76°, 48.95°, 53.39°, 58.26°, 66.46°, 68.05° and 75.20° which could be indexed to the (110), (002), (111), \((20\overline{2 }\)), (020), (202), \((31\overline{1 }\), (220), and \((22\overline{2 })\) planes respectively of the monoclinic CuO (ICCD-96-101-1195).

Fig. 3
figure 3

XRD patterns of CuO0.1 M, CuO0.5 M, CuO1 M, CuO1.5 M and CuO2 M samples

Full size image

The peaks at 2θ values of 43.40° and 50.56° could be attributed to (111) and (200) planes of cubic structured Cu (ICCD-96-901-2955). The calculated unit cell parameters are tabulated in Table 1. We observed that the intensity of Cu peaks decreases as the concentration of KOH increases which could be reasoned to the increase in thickness of the samples corresponding to the increase in the KOH concentration. For CuO1.5 M an additional peak was observed at 36.56° and for CuO2 M samples, two additional peaks at 36.56° and 42.42° were observed which corresponds to the (111) and (020) planes respectively of cubic Cu2O (96-101-0942). This observation could be recognized as the deposition of Cu2O on the copper foil due to an incomplete reaction at the film–foil interface [12]. Nuengruethai et al. reported that Cu2O impurities occur in facile deposited CuO nanostructures as the deposition time increases [12]. From XRD analysis, it is clear that when the molarity of the KOH solution is increased beyond 1 M additional Cu2O impurities occur. Among the three samples CuO0.1 M, CuO0.5 M, and CuO1 M where no Cu2O impurities were identified the CuO1 M samples show relatively higher XRD intensity for all CuO peaks. Hence this sample was identified to be the optimum in terms of its crystallization. Figure S1 (supplementary) shows the magnified peak images of 2θ values 36.56° and 42.42°.

Table 1 Summary of calculated values of unit cell parameters, crystallite size, strain, dislocation density, and crystallite number for the CuO thin film samples
Full size table

The Scherrer equation was used to determine the average crystallite size (D) of the CuO thin films [13]:

$$D=\frac{K\lambda }{\beta {\text{Cos}}(\theta )},$$
(1)

where K is the Scherrer constant (K = 0.9), which is affected by the size distribution and crystallite form, λ = 1.5418 A° is the X-ray wavelength, \(\theta\) is the Bragg angle, and β is the full width at half maximum (FWHM) of the plane.

Using the average crystallite size (D), the crystallites number (N) was determined using the relation [6]:

$$N=\frac{1}{{D}^{3}}.$$
(2)

Micro-strain (ε) in the samples were evaluated using the relation (3) which is a prominent source for the XRD peak broadening [14]:

$$\varepsilon =\frac{\beta {\text{Cot}}(\theta )}{4}.$$
(3)

Dislocation density (δ), which is defined as the number of dislocation lines per unit volume through the film surface was evaluated using relation (4) [6]:

$$\delta =\frac{1}{{D}^{2}}.$$
(4)

Table 1 summarizes the results of the calculations conducted for evaluating average crystallite size (D), micro-strain (ε), and dislocation density (δ) for the grown thin film samples.

Figure 4 shows the high-resolution TEM image and selected area electron diffraction (SAED) image of the CuO1 M sample. From Fig. 4a, b, it is evident that the deposited CuO material is nanostructured which exhibits a shape identical to that of nanopetals. Figure 4c shows the lattice under the magnified scale of 2 nm, from which the diffraction line inter-planar distance is measured as 0.25 nm and 0.27 nm which corresponds to the inter-planar spacing of (110) and (002) planes for monoclinic CuO, respectively. Figure 4d, which represents the selected-area diffraction pattern of the sample, is useful for determining the type of structural ordering in the material. The presence of concentric ring-like structures with diffuse spots indicates that the grown material is polycrystalline in nature. The planes (110), (002), (111), (020), and (220) corresponding to the monoclinic phase of CuO could be indexed from the SAED pattern. The results of the HR-TEM analysis complement the XRD analysis in identifying the crystal structure of the grown CuO films.

Fig. 4
figure 4

a, b High resolution-TEM images c lattice under 2 nm magnification scale and d selected area electron diffraction (SUED) images of CuO1 M sample

Full size image

The Raman spectra were used to analyze the nanoparticles’ structural characteristics. There are 12 vibrational modes at the zone center of CuO, including three acoustic modes (Au + 2Bu), six infrared-active modes (3Au + 3Bu), and three Raman-active modes (Ag + 2Bg)[15]. The Raman spectra of the freshly made CuO nanostructures are displayed in Fig. 5. Three Raman peaks can be recognized at 280, 330, and 620 cm−1, respectively. The Raman peaks are wider and downshifted when compared to the Raman vibrational spectra of a CuO single crystal [16]. The peak at 273 cm−1 belongs to the Ag mode, while the peaks at 321 and 606 cm−1 belongs to the Bg modes [15]. The quantum confinement effect of CuO nanorods is primarily responsible for the broadenings and downshifts of the Raman peaks [15].

Fig. 5
figure 5

Raman spectra of CuO samples

Full size image

XPS is used to characterize changes in the chemical environment of an element in a structure because of the short inelastic mean free path (IMFP) and sensitivity of the photoelectrons to their chemical surroundings [17]. Figure 6 shows the XPS results obtained from the optimal sample CuO1 M. The survey spectrum in Fig. 6a reveals the presence of the elements copper, oxygen, and carbon. Figure 6b, c shows the XPS spectra of Cu2p and O1s. In Fig. 6b two Cu peaks at 933.9 and 953.9 eV, corresponding to the Cu2p3/2 and Cu 2p1/2 peaks of Cu2+ respectively confirm the existence of CuO. The peaks at 941.72 eV and 944.20 eV are satellite peaks of Cu2p3/2. The peak at 962.51 eV corresponds to the satellite peak of Cu2p½. The O1s spectrum as shown in Fig. 6c consists of two oxygen contributions identified as O2− in CuO at 529.7 eV and oxygen adsorbed onto the surface of CuO particles at 531.2 eV [18]. XPS analysis confirms the chemical composition of the films to be CuO and complements the structural identification of the monoclinic CuO phase obtained from XRD and Raman analysis.

Fig. 6
figure 6

XPS of CuO1 M sample a survey spectrum b Cu2p3/2 spectra and c O1s spectra

Full size image

The surface pattern of CuO nano structure was studied using scanning electron microscopy (SEM). Figure 7 shows the SEM image of the optimized CuO1 M sample at different magnifications. SEM image confirms the formation of 2D self-assembled nanostructured fibrous surface morphology over Cu foil. The surface image of the CuO grown on the copper foil as shown in Fig. 7a, shows the bundled nature of the fibrous structures whereas the image shown in Fig. 7b reveals that needle like structures with an average thickness ~ 130 nm are stacked homogenously over the surface. There appears to be a globular pattern on the large scale distinct from each other composed of these needle-like structures forming a structure resembling the petals of a flower.

Fig. 7
figure 7

SEM images of the CuO1 M sample at a scale of a 10 µm and b 1 µm

Full size image

We propose the following mechanism for the growth of CuO thin films in our process: clean and new copper foils are separately submerged in 100 mL of KOH at different concentrations during the first stage of the synthesis. Copper foil, being the source of copper begins to degrade by releasing Cu2+ ions into the solutions, leaving behind electrons. The Cu2+ ions get continually discharged into the solutions at a very quick oxidation rate [19]. Since Cu2+ and O2− are the most stable oxidation states, the released ions interact with hydroxyl ions and result in the formation of Cu(OH)2 which precipitates on top of the copper foils, which then becomes CuO (solid films) after 5 h of dehydration at 60 °C. High OH− concentrations could also lead to the formation of [Cu(OH)4]−2 complex ions which is precipitated as Cu(OH)2 and then are broken down into CuO (solid films) through the dehydration process[12]. Anisotropic particles are formed at this stage as a result of the concentration of OH− ions having a net impact on the nucleation and growth processes, such that the number of nuclei, growth units, and the pace at which certain crystalline faces grow are balanced [20, 21]. The above mechanism can be illustrated using Eqs. (5)–(8).

$${\text{Cu}}+2{\text{KOH}}\to {\text{Cu}}{\left({\text{OH}}\right)}_{2}+2{\text{K}},$$
(5)
$${\text{Cu}}{({\text{OH}})}_{2}+2{({\text{OH}})}^{-}\to {[{\text{Cu}}{\left({\text{OH}}\right)}_{4}]}^{2-},$$
(6)
$$2{[{\text{Cu}}{\left({\text{OH}}\right)}_{4}]}^{2-}\to 2{\text{Cu}}{({\text{OH}})}_{2}+2{{\text{H}}}_{2}{\text{O}}+{{\text{O}}}_{2}+4{{\text{e}}}^{-},$$
(7)
$${\text{Cu}}{\left({\text{OH}}\right)}_{2}\to {\text{CuO}}+{{\text{H}}}_{2}{\text{O}}.$$
(8)

The optical properties of the samples were characterized by UV–visible diffused reflectance spectroscopy (DRS). Figure 8a shows the DRS spectra of different CuO samples. The Kubelka–Munk model is used to evaluate the optical band gap energy of the samples. The value of F(R) is calculated using the equation [12]:

$$F\left(R\right)={\frac{(1-R)}{2R}}^{2},$$
(9)

where F(R) is the Kubelka–Munk function and R is the reflectance. Figure 8b shows the values of \((F\left(R\right)hv{)}^{2}\) vs \(hv\). For obtaining the band gap energy, the straight line part of the graph is stretched out to intercept the \(hv\) axis. The intercept value corresponds to the band gap energy [10]. The calculated band gap energies of CuO0.1 M, CuO0.5 M, and CuO1 M are 1.48, 1.47 and 1.45 eV respectively, whereas the band gap of CuO1.5 M and CuO2 M are 1.6 and 1.61 eV. These large band gaps ≥ 1.6 eV can be attributed to the presence of Cu2O impurities in the samples. Generally, Cu2O has greater band gap value than CuO [18].

Fig. 8
figure 8

Graphical representation of reflectance and F(R) spectra for prepared CuO samples (a) reflectance versus wavelength and (b) F(R) versus energy

Full size image

Figure 9a–e illustrates current voltage characteristics of Ag/CuO/Cu device structure at room temperature in the voltage range − 1 V to + 1 V under dark, UV illumination, and halogen illumination conditions. Current–voltage characteristics of the device show typical diode behavior with lower reverse current and higher forward current which highlights the rectification ratio of the diode.

Fig. 9
figure 9

I–V characteristics under dark, UV and halogen illuminated conditions for samples a CuO0.1 M b CuO0.5 M c CuO1 M d CuO 1.5 M and e CuO2 M

Full size image

Under halogen and UV illumination, incident photons are absorbed, and electron–hole pairs are generated which results in the generation of photocurrent. The photosensitivity, defined as \(\left(\left(\frac{{I}_{{\text{illu}}}-{I}_{{\text{dark}}}}{{I}_{{\text{dark}}}}\right)X 100\right)\) [22], and rectification ratios of all samples are summarized in Table 2. Among all samples, CuO1 M shows a good rectification ratio and photosensitivity.

Table 2 Photosensitivity and rectification ratio values of the device structures
Full size table

As per the thermionic emission model current across the Schottky barrier can be expressed as [23]:

$$I={I}_{0}\left({\text{exp}}\left(\frac{qv}{n{K}_{{\text{B}}}T}\right)-1\right),$$
(10)

where K is the Boltzman constant, n is the ideality factor, T is the temperature in kelvin, R is the series resistance, q is the electronic charge, I0 is the reverse saturation current where I0 is defined as[24]:

$${I}_{0}=A{A}^{*}{T}^{2}{\text{exp}}\left(\frac{-q\Phi }{{K}_{{\text{B}}}T}\right),$$
(11)

Here A and A* is the effective area of the diode and Richardson constant (979.36 Acm−2 K−2 for CuO) [25] respectively and Ф is the zero bias barrier height.

The series resistance of the Schottky junction was determined using the Cheung-Cheung model [26] in the nonlinear part of the high current region. The following relation is used to calculate the series resistance

$$\frac{{\text{d}}V}{{\text{d}}l{\text{n}}I}=I{R}_{{\text{s}}}+n\left(\frac{{K}_{{\text{B}}}T}{q}\right),$$
(12)
$$H\left(I\right)=V-n\left(\frac{{K}_{{\text{B}}}T}{q}\right){\text{ln}}\left(\frac{I}{AA{T}^{2}}\right),$$
(13)
$$H\left(I\right)=I{R}_{{\text{s}}}+n\Phi .$$
(14)

From the Eq. (12), RS is calculated from the slope of dV/d(lnI) vs I curve and Y intercept gives the value of n. Series resistance can also be calculated from the slope of H(I) vs I curve. By knowing the value of n, y intercept of H(I) vs I graph determines the zero bias potential barrier Ф [26]. The series resistance, ideality factor and zero bias potential barrier was calculated for all the samples under dark, UV illumination and halogen illumination and are tabulated in Table 3. For all samples, there is no much difference in series resistance calculated from Eqs. (12) and (14). All the fabricated junctions show decrease in series resistance upon UV and halogen illumination due to the generation of photo carriers followed by photon irradiation. The devices exhibit relatively better to photo response to halogen light illumination compared to UV light illumination.

Table 3 Electrical parameters obtained for all CuO samples
Full size table

The non-linear behavior of current–voltage characteristics can be explained using the variation in diode series resistance [23]. Series resistance of CuO1 M is 1.30 Ω in dark condition, which is the lowest resistance value obtained, and results in high rectification ratio and good diode quality. Series resistances of CuO1.5 M and CuO2 M are ~ 5830 Ω and ~ 5972 Ω respectively. This high resistance value causes nonlinear bending of forward current in the current voltage characteristics and low rectification ratio[23].

The variation in ideality factor for Ag/CuO/Cu junctions from ideal one can be attributed to the contribution of interface surface defects and series resistance [26]. Even though the series resistances is small for CuO1 M and CuO0.5 M, these samples show ideality factors 4.07 and 4.02, respectively which is greater than ideal diode factor. In this case, surface defects may play the role as the series resistance is negligible. The interface surface defects and series resistance in tandem contribute to large n values (6.61 and 10.16) for CuO1.5 M and CuO2 M samples. The obtained n values are comparable with reported values [23, 27]. The zero bias potential barrier height Ф decreases with increase in the KOH molarity initially and subsequently it increases. For CuO0.1 M, CuO0.5 M, CuO1 M, CuO0.5 M and CuO2 M, the barrier heights are 0.78, 0.60, 0.53, 0.74, and 0.74 eV, respectively.

For photo detector studies, Ag/CuO/Cu structure is used as the active device. Since the photosensitivity of CuO samples towards halogen source is higher than UV source, visible light detection is studied here. The working mechanism of fabricated photodetectors could be explained as shown in Fig. 10.

Fig. 10
figure 10

Working mechanism of CuO photodetector

Full size image

The working of the photodetector is demonstrated under dark and halogen-illuminated conditions at a biasing voltage of 10 mV. In the dark condition, oxygen molecules are adsorbed to the exposed surface of CuO by capturing electrons from the conduction band.

$${{\text{O}}}_{2}+{{\text{e}}}^{-}\to {{\text{O}}}_{2}^{-} \left({\text{ad}}\right).$$
(15)

Adsorbed oxygen creates electron trap sites on the CuO surface. These trapped electron sites owe the formation of a wider less conductive depletion region. When the photodetector is illuminated with a halogen source, electron–hole pairs are generated. Photo-generated holes migrate from the valance band to the surface-trapped oxygen sites to recombine with \({{\text{O}}}_{2}^{-} ({\text{ad}})\), thereby discharging \({{\text{O}}}_{2}\) gas into the air and electrons into the conduction band.

$${{\text{O}}}_{2}^{-}\left({\text{ad}}\right)+{{\text{h}}}^{+}\to {{\text{O}}}_{2}.$$
(16)

This process increases the electron lifetime and successively increases the photocurrent. The oxygen ions are once more adsorbed to the CuO surface when the light is turned off and the current value drops as a result [14, 28].

Figure 11 shows the switching photo response of fabricated Ag/CuO/Cu photodetectors. The photocurrent time response of the CuO photodetectors was measured at 0.01 V bias under a halogen source on/off lighting with an on/off interval of 10 s. The intensity of the halogen light source was fixed at 70 W/m2. The “rise time” is defined as the time required for the photocurrent to increase from its minimum value to 63% of its maximum value and the “decay time” is the time taken for the photocurrent to drop from its maximum magnitude to 37% of its maximum value [29]. The rise and decay times of the device structures are tabulated in Table 4. The responsivity, R is another important term which is the ratio of photocurrent to incident illumination power [30]. The responsivity was calculated using the formula [30]:

$$R = \frac{I}{AE},$$
(17)

where R is the responsivity, I is the measured photocurrent, A is the photodetector device's active area, and E is the light source's irradiance intensity, which is measured by a conventional light power meter. Among the five CuO detectors, CuO1 M shows the best responsivity value which is 1.65 µA/W. Specific detectability (D*) which is a figure of merit for the device determines the weakest detectability of the optical signal and was calculated using the equation [31]:

$${D}^{*}=R{(\frac{S}{2e{I}_{{\text{dark}}}})}^\frac{1}{2} ,$$
(18)

where R, is the responsivity, S is the effective illumination area, e is the charge of the electron, and Idark is the dark current.

Fig. 11
figure 11

Switching photo-response (current vs. time) of CuO0.1 M, CuO0.5 M, CuO1 M, CuO1.5 M, and CuO2 M photodetectors

Full size image
Table 4 Photodetector parameters of all Cu/CuO/Ag detectors
Full size table

The ON–OFF ratio is defined as the ratio of photocurrent to dark current (Ip/Id)[29]. All calculated photodetector parameters are tabulated in Table 4.

The current–voltage characteristics reveal that the value of the ideality factor increases with halogen and UV illumination for all photodetectors except CuO1 M. This may be because, under dark conditions, the oxygen molecules that were adsorbed to the surface-trapped sites remain inactive. During illumination, adsorbed oxygen molecules get desorbed and the trapped sites become active for charge carrier trapping. For CuO1 M, the density of defect and trapped sites are less which increases the carrier recombination and thereby decreases the ideality factor. This is supported by the lowered diode series resistance for this device structure. This gives excellent parameters as required for a good photodetector. The inset of Fig. 11 shows the variation of responsivity and specific detectivity with KOH concentration.

Figure 12 represents the transient photo response curve for three of the samples. From Fig. 12, it is evident that the photocurrent saturates much faster for the sample CuO0.5 M samples within the illumination pulse period relative to its counterparts. The flat nature of the response indicates that no further charge carriers are being generated as a result of the illuminated condition. On the other hand, for sample CuO0.1 M, the presence of a trap state is evident from the two-level transient response curve. The photocurrent reaches a maximum when illuminated and then decays gradually even when the illumination is maintained and then abruptly decays when illumination is switched off. This temporal dependence indicates that trap states may be activated after the initial rise in photocurrent, and subsequently, the rapid depletion of charge carriers occurs which leads to the abrupt decrease in photocurrent.

Fig. 12
figure 12

Time-dependent photo-response (current vs. time) of CuO0.1 M, CuO0.5 M, and CuO1 M photodetectors

Full size image

For CuO1 M, the rise and decay of photocurrent exhibit the persistent photoconductivity nature associated with deep defect levels [32]. These results indicate that the carrier lifetime is extended in sample CuO1 M relative to all of the other samples prepared by us. The improved photo-response is a manifestation of the improved carrier dynamics which makes it the optimized sample among the set of samples engineered in this work. The persistent nature of photoconduction indicates that the charge carriers are separated as electron–hole pairs for a longer duration of time preventing their recombination. Hence, use of the appropriate electron/hole transport layers could provide an opportunity to extract the carriers for photovoltaic application.

Table 5 summarizes how the electrical parameters of the device were optimized by us compared to other similar photodetectors reported in the literature. Relative to previous reports our device shows substantial responsivity and detectivity under the application of very low bias voltage which is very encouraging. All of the previous works are based on advanced material growth techniques involving expensive processing equipment and rigorous growth conditions [33,34,35,36,37,38,39].

Table 5 Summary of photo-response parameters of previously reported CuO-based visible photodetectors
Full size table

Conclusions

Room temperature self-assembly of CuO thin films on copper foils using a facile solution-immersion method using KOH as the oxidizing agent is achieved in this work. Phase pure CuO thin films with monoclinic structure could be grown on copper foil. Raman and XPS analysis further confirmed the formation of CuO and complimented the XRD results in identifying that the CuO1 M thin films were the optimally crystallized sample. Our work shows that use of higher KOH concentration results in the formation of Cu2O impurity phases. This may be due to the decrease in oxidation rate at higher KOH concentrations. Beyond 6 days of wet oxidation, the films tend to peel off from the foil and sediment in the bath which limits the thickness of the CuO films that can be grown by our present process. To study the photo response of the grown CuO thin films, photodetectors were fabricated with the structure Cu/CuO/Ag. The device structure exhibited photo-response to both UV and visible light radiation. Temporal dependence of the photo-response of each device structure to the same illumination pulse showed that the sensitivity mechanism was different for the CuO0.5 M and CuO0.1 M samples relative to their counterparts. The current–voltage characteristics revealed the effect of series resistance and ideality factor on the photo-detecting mechanism of the engineered devices. Series resistance plays an indispensable role in the photo-detecting ability of the manufactured devices. Devices with low series resistance and ideality factor have the best photo-detecting properties. The optimized photodetector device structure Cu/CuO/Ag grown at 1 M KOH concentration is found to have a low series resistance of ~ 1.3 Ω and a high ON–OFF ratio of 69.4. The photodetector structure developed by us exhibits a responsivity of ~ 1.65 µA/W and detectivity of 22 × 105 Jones under a bias of 0.01 V. These characteristics for a wet chemically grown photodetector are competitive relative to other CuO-based photodetectors engineered using contemporary techniques.