1 Introduction

Metal and metal oxide micro/nanoparticles and micro/nanostructures have a wide range of fundamental and practical application due to their distinctive properties [1,2,3,4,5,6,7,8,9,10,11]. Studies have shown that metal nanostructures have a positive effect for improving the luminous efficiency of light-emitting diodes (LEDs) [12]. In addition, metal and metal oxide micro/nanoparticles are the excellent materials for the preparation of photocatalysts [1], and they are also widely used in the field of optical instrument [5, 13]. Therefore, the researchers actively analyze different methods of the preparation of micro/nanoparticles and micro/nanostructures [14,15,16,17,18]. It has been shown that the properties of these micro/nanoparticles and micro/nanostructures strongly depend on their size [19,20,21], morphology [22, 23], composition [24], and structure [25].

The oxide of copper, cupric oxide (CuO), and cuprous oxide (Cu2O) are typical representatives of such species. Compared to macro-targets, the properties of micro/nano-sized CuO/Cu2O are quite different due to the increase of surface/volume ratio and the influence of quantum size effect. As a semiconductor material, CuO is environmentally friendly, low cost, and has numerous applications in the field of photovoltaic because of its high absorption of solar spectrum and high carrier concentration [26]. In addition, applications based on CuO/Cu2O micro/nanoparticles and micro/nanostructures, electrochemical materials [27], and super-capacitors [28] have received a great amount of attention, also. Various properties related to Cu and CuO/Cu2O micro/nanoparticles and micro/nanostructures, such as synthesis methods, synthesis mechanisms, characterization, fundamental properties, nonlinear optical characterization, and potential applications, have been discussed in [26,27,28,29,30,31,32].

Pulsed laser ablation in liquid (PLAL) can be used to prepare metals and oxide micro/nanoparticles possessing different structure and composition [33]. Application of PLAL for the synthesis of nanostructured metallic oxide and alloy has been reported in numerous studies, for example, [34,35,36]. PLAL provides a simple and efficient way to synthesize micro/nanoparticles, comparing to chemical methods. In addition, various controllable factors, such as pulse duration [17] and type of liquid [37,38,39], allow modifying the composition, morphology and structure of micro/nanoparticles. The current research has clearly shown that external experimental conditions can affect the properties of resulting oxide micro/nanoparticles through laser ablation. For example, the CuO particles from ablated bulk Cu in water using Q-switched pulsed laser (532 nm, 5 ns, 100 mJ/pulse) have been analyzed, and it was found that the increase of annealing temperature can reduce the size of particles [40]. Meanwhile, annealing also affects the bandgap and other optical properties. The ablation by 120 fs, 780 nm radiation of the Cu target in acetone showed that both Cu and CuO particles were present in the colloidal solution [41]. These particles had spherical shape with the mean diameter of 9 nm. The optical absorbance carried out immediately after ablation has shown the presence of surface plasmon resonance (SPR) at 602 nm. The green color of particles in acetone solution remained stable even after a long time. Another group of studies using Nd:YAG laser (1064 nm, 7 ns) showed that Cu target ablated in water mainly leads to the formation of Cu2O, while the products during ablation in ethanol are Cu particles [38]. However, if the ablation condition was changed to Nd:YAG laser, 532 nm, 10 ns, two kinds of stable oxides are formed in water:monoclinic CuO and cubic Cu2O [42]. In addition, the application of different solvents allows synthesizing different types of composite particles [43]. Generally speaking, many external conditions and their combinations bring uncertainty to the variations of the morphology and size of micro/nanoparticles during ablation, while providing the possibilities for more precise control of the ablation process to obtain particles with controllable morphology and size.

In this paper, we report the systematic structural analysis of the CuO and Cu2O nanostructures, which were obtained during ablation of Cu in water using various pulses (5 ns, 200 ps and 30 fs), wavelength (1064 and 355 nm radiation of 5 ns pulses, 800 nm radiation of picosecond and femtosecond pulses) and energy (high and low fluence), using the scanning electron microscopy (SEM) and spectroscopic techniques. The effects of pulse duration and wavelength on particles’ formation and aging were systematically analyzed. Particles with uniform morphology and larger sizes were synthesized. The stability of CuO/Cu2O nanostructures was studied using SEM during a few months after ablation of bulk Cu. It was found that the application of ultra-short pulses in ablation allows obtaining smaller sized particles with narrower distribution, while the pulse wavelength has a decisive effect on the morphology of the particles.

2 Experimental arrangements

These studies were carried out using focused pulsed laser to irradiate the surface of bulk Cu (ZhongNuo Advanced Material Technology Co. Ltd., China) with thickness of 3 mm, which was cleaned to remove the oxide film and stains on the surface. Cu target was kept in a quartz cell with edge length of 30 mm filled with distilled water. During ablation from the top side, the quartz cell was moved along X- and Y-directions to ensure that the ablation occurs at different places on the surface of Cu. Slow and uniform movement of the cell was maintained to ensure minimal impact on liquid level, which was 25 mm above the target surface. The laser pulse, which was focused by a focal lens with ƒ = 150 mm, irradiated from the top of cell on the surface of target. The laser beam focal length increases in water. The amount of additional focal length is directly proportional to the water height. We have corrected the position of focal lens before experiment, to ensure that the laser beam focuses on the target surface after passing through the liquid.

Experiments were accomplished using a regenerative amplifier (Spitfire Ace, Spectra Physics), which provided the 800-nm pulses at repetition rate of 1 kHz and pulse durations of either 30 fs (laser energy: 0.56 mJ, scan speed: 5 mm/s during experiments) or 200 ps (laser energy: 0.56 mJ, scan speed: 5 mm/s during experiments). The nanosecond Nd:YAG laser (Q-Smart, Coherent) provided 1064-nm pulses [as well as its third harmonic (λ = 355 nm)] of 5 ns duration at 10 Hz repetition rate (laser energy: between 10 to 100 mJ, scan speed: 5 mm/s during experiments), which were also used in these studies. The influence of pulse energy on the ablation was analyzed using three pulse durations (5 ns, 200 ps and 30 fs pulses). The duration time of ablation was maintained at 30 min during each experiment.

The size, nanostructure, distribution and morphology of particles obtained by PLAL were analyzed by SEM (S-4800, Hitachi), thus allowing determining the change of morphology during the aging process. Chemical composition and microscopic structure of particles were analyzed by X-ray diffraction (XRD, D8 Discover, Bruker AXS) and energy-dispersive spectroscopy (EDS, S-4800, Hitachi). The absorption spectroscopy (Cary Series, Agilent Technologies) was carried out to determine the bandgaps and SPRs of particles in the water colloids.

The experimental parameters are shown in Table 1. The information presented in this table allows determining the role of different pulses in the ablation process, similar to earlier presented data for ablation and formation of nanoparticles using pulses of different duration [44].

Table 1 The parameters of experiments
Full size table

The purity of the Cu target used in this study was 99.99%, in accordance with the data provided by the manufacturer. The materials of such purity were also used in previous studies [45,46,47]. According to the quality analyze report provided by the manufacturer, the Cu bulk contained some impurities. The types and proportion of elements are shown in Table 2.

Table 2 Certificate of analysis of Cu plate
Full size table

The types of impurity are given in the test report. Most of the impurities did not show in EDS because of their content was extremely small, but some elements, such as Cl and K, showed very weak EDS peaks. In addition, there was a strong peak of Si because of EDS was detected using the silicon wafers to prepare the ablated samples. Moreover, there were three main sources of oxygen in EDS. First, oxygen is the most abundant impurity in Cu target according to the analysis report. Second, to improve the surface smoothness of silicon wafer, a very thin oxide film is deposited on the surface of silicon wafer, which was shown up in EDS tests with silicon. Third, there were oxygen components in copper oxide and cuprous oxide formed by oxidation of copper particles in air and liquid. The oxygen element presented in the EDS is a collective representation of many sources superimposed.

For some elements (especially impurity elements), only weak single peak appeared during EDS measurements. The content of impurity elements in the sample was extremely low. When the EDS analysis was conducted on different positions of the sample, we found that the individual impurity elements appeared occasionally. The presence of the element with only one peak during EDS measurements is also a common case in previous studies [48, 49].

The EDS supported our expected results. In [50, 51], the authors have presented the EDS of CuO and Cu2O nanostructures prepared with chemical methods which coincided with our results. The amount of the elements was measured using EDS of the CuO films deposited on the surfaces of a silicon substrate. At the meantime, we analyzed EDS of the pure silicon substrate.

3 Result and discussion

3.1 Spectroscopy of colloidal suspensions

The initial colloidal solution containing particles prepared by ablation using different pulse duration (nanosecond (ns), picosecond (ps), femtosecond (fs)) showed green color. However, the color of solution obtained by ps pulses was the deepest, while that of ns and fs was relatively light. The depth of color of colloidal solution depends mainly on the concentration of particles. It is worth noting that in the case of ns pulses with 10 Hz repetition rate, the number of pulses acting on the target surface during the same time (30 min) was 100 times less compared with the case of ps pulses. In addition, the water level has a strong effect on the laser power at the 1064 nm wavelength [52]. About 5.5% of the laser power will be lost during propagation of 1064-nm radiation through a 2-mm-thick water layer. We used a 25-mm-thick water layer above the surface of the target, which weakens the energy of the laser pulses (1064 nm, 5 ns) on the target surface. The total amount of particles produced by ns pulses during the same ablation time was significantly lower than that in the case of ps pulses. During ablation by fs pulses, strong white light was emitted when these pulses propagate through the liquid. The white light generation is a threshold effect, which depends on the power of the focused beam. This effect led to a decrease of pulse intensity on the target surface, resulting in a lower efficiency of nanoparticle formation [53]. As a result, the concentration of particles in that case was significantly smaller than in the case of ablation using ps pulses. Therefore, the efficiency of the synthesis of particles at the condition of ablation by ns and fs pulses during these experiments was significantly lower than that of ps pulses, which explained the lighter color of suspensions in the former cases. After a few hours from ablation, the colloidal solutions of particles obtained by ns and ps pulses showed brown color, while the change of the color of solution obtained by fs pulses was insignificant. Since then, colloidal solutions became extremely stable in color.

The absorption spectra of suspensions perfectly reflected the temporal change of color of nanoparticle-contained suspensions. The absorption spectra of the newly prepared colloidal solutions of particles were measured immediately after ablation (Fig. 1a). We compared the absorption spectra of the solution obtained by PLAL using different pulse durations. Although the duration time of ablation was the same, the absorption of the sample obtained by ps pulses was significantly stronger than that of the other two (ns and fs). The difference in absorption spectra of particles obtained by various wavelengths (1064 and 355 nm of 5 ns pulses) correlated with the difference in the morphology and size of these particles, which was confirmed by SEM measurements (see the following subsection).

Fig. 1
figure 1

Absorption spectra of the suspensions containing CuO/Cu2O nanostructures synthesized by ablation in distilled water. a Spectra of suspensions obtained by various pulse durations (5 ns, 200 ps and 30 fs) and wavelengths (1064 and 355 nm in the case of 5 ns pulses). b Spectra of suspensions obtained at different energies of 1064 nm, 5-ns pulses. c Spectra of aged colloidal solutions obtained by ps pulses. d Spectra of aged colloidal solutions obtained by fs pulses

Full size image

We observed the absorption peak at λ = 640 nm (E = 1.94 eV) in three samples [5 ns (1064 nm), 200 ps (800 nm) and 30 fs (800 nm)], which showed the highest intensity in the case of ps pulses. One can assume that this absorption peak is related to the SPR of non-spherical Cu nanoparticles. The position of peak was similar to those reported in previous studies [40, 54]. It is worth mentioning that there are many different reports about the location of the absorption peaks of Cu particles, which were mainly attributed to the difference in the ablation conditions and the physical properties of nanoparticles. Particularly, in the case of ablation in water, the absorption resonance peak appeared at 650 nm [42], while in the case of ablation in vacuum, the absorption curves of deposited Cu particles showed the absorption bands at shorter wavelength region (λ = 590 nm, [55]).

A dependence of the ablation threshold on the pulse duration for bulk copper was demonstrated in [56]. It was shown that the highest ablation threshold corresponded to the nanosecond ablation regime. In the case of nanosecond ablation, we used higher energy of the ablating pulses at the same conditions of the studies. During laser ablation of the copper, we observed the change of the color of the suspension due to the formation of the nanoparticles of the copper. The color of the suspension became the green, which indicates stronger linear absorption in the near-infrared range. The transparency and loss of the ablation pulses energy were negligible at the wavelength of 800-nm and 1064-nm pulses (Fig. 1). In the case of ablation with 355 nm, the strong absorption of the ablation pulses energy led to the decrease concentration of the ablated copper nanoparticles (see Fig. 1a, solid purple). As was mentioned, in the case of the femtosecond ablation the energy of ablating pulses was lost due to the white light generation starting from the surfaces of the cell and continuing inside the water. The conversion of the energy of ablating femtosecond pulses to white light generation led to the decrease of the energy of pulses on the surfaces of the bulk copper immersed in the distilled water for ablation of the copper nanoparticles.

In the case of 5-ns pulses (λ = 1064 nm), ablation was carried out using two different energies (E = 80 and 45 mJ) (Fig. 1b). Spectral measurements showed that the absorption of colloidal solution obtained at higher energy (E = 80 mJ) is stronger. The intensity of the absorption spectrum reflects the amount of the particles contained in the colloidal solution, that is, the volume of particle formation. The ablation efficiency of particles depends on the type of material used, heat of vaporization, and laser power used. Higher pulse energy corresponds to higher laser power (at the same repetition rate, 10 Hz), resulting in a larger nanoparticle generation rate. Therefore, more nanoparticles can be produced at higher ablation energy (80 mJ), while the shapes of absorption curves in these two cases are similar to each other. The same conclusion can be obtained in the case of ps and fs pulses with different ablation energies. Similar results were reported in [41]. Besides a broad absorption band located at λ = 640 nm, there is a very strong absorption peak that appeared at λ = 230 nm (E = 5.64 eV) in the case of ablation by 5-ns and 1064-nm pulses. This peak (λ = 230 nm) is related to the copper electrons band-to-band transition [57, 58]. The oxidation process under these conditions is extremely rapid. The interaction time between single ns pulse and target surface is relatively long. The less stable particles formed at these conditions easily interact with the surrounding environment, which leads to faster aging.

In the case of ps ablation, the UV spectrum of initial colloidal solution of particles showed a wide “shoulder” at λ = 640 nm. However, the spectra significantly changed during the subsequent measurements (Fig. 1c). The absorption peak at 640 nm disappeared, while a very strong absorption peak appeared near λ = 300 nm (E = 4.17 eV), which is caused by the Brillouin transitions of Cu2O [59, 60]. This observation pointed out the formation of Cu2O during aging. The spectra measured at the 2nd, 15th, and 30th days were similar to each other. Meanwhile, the color of colloidal solution was changed from green to brown. The species obtained at the beginning of ablation were mainly Cu nanoparticles, which led to the green color of colloidal solution. However, during aging, the particles reacted with oxygen in the solution to form CuO/Cu2O shell during aging process [25], which led to the change of color from green to brown. This led to the change of absorption peak position. Such core–shell structures also appear during the oxidation process in other metals [24, 61, 62].

The absorption peak located at λ = 640 nm in the case of fs ablation was smaller than the one observed using ps pulses mainly due to the lower rate of particles synthesized by fs pulses. Similar to the sample obtained by ps pulses, the common feature of the sample synthesized by fs pulses after several days of aging was a strong resonance peak at λ = 300 nm. It is noteworthy that, compared with the former (ns and ps) regimes of ablation, the aging process in the case of fs ablation was relatively slow. On the second day after ablation, the color of the colloidal solutions obtained by ns or ps pulses changed due to oxidation, and the absorption spectra also showed the corresponding changes. However, the absorption spectrum of the solution obtained by ablation using fs pulses did not show significant difference during first few days after ablation due to slower oxidation process.

As it has been shown in the case of longer pulses (ns and ps) that the plasma plume produced during ablation forms the nanoparticles [63]. The longer interaction time between laser radiation and target surface caused a longer heating. The internal structure of particles was determined by cooling and aggregation of the plume interacting with external environment for a longer period. On the other hand, the particles synthesized under ultra-short-pulse (fs) ablation did not participate in a similar heating/cooling process, which made those particles more stable. This probably was a reason why the color and spectral curves of colloidal solution changed in the case of ns and ps pulses, while the aging of solution under fs pulse ablation was relatively slow and more stable. In addition, from the analysis of these spectra, we inferred a negligible absorption of the long-wavelength part of the spectrum, which means that in case of 800 and 1064 nm relatively long ablating pulses, the absorption in suspension was insignificant.

3.2 Morphology of particles

SEM was carried out to characterize the morphology, size and distribution of particles obtained at different pulse durations and wavelengths. Taking into account the fact that the initial particles became oxidized during a very short time (few hours), which led to unstable morphology, the analysis was mainly carried using the samples, where the nanoparticles completed the oxidation process and remained at stable conditions. The SEM images of particles obtained under different pulse durations are shown in Fig. 2. These studies demonstrated that pulse duration during ablation of Cu target in water plays a crucial role in the process of particles’ formation, which resulted in the remarkable difference of the morphology and particle sizes of synthesized species.

Fig. 2
figure 2

SEM images of CuO/Cu2O nanostructures obtained in distilled water using the pulses of different duration. a, b Ablation using 5-ns, 1064-nm pulses; c, d ablation using 200-ps, 800-nm pulses; e, f ablation using 30-fs, 800-nm pulses

Full size image

The particles obtained at some conditions, particularly during ablation by 800-nm, 200-ps pulses, showed the same morphology, and the particle sizes had a very good homogeneity. SEM images showed that the particles obtained by 200-ps pulses were smaller compared with the 5-ns case. The conclusion that the size of particles became smaller under the condition of shorter pulses ablation was further confirmed in the case of 30-fs pulses. This peculiarity has also been underlined in [64]. In addition, the synthesized particles demonstrated a common feature. Their surface did not show a smooth and holistic structure, but rather contained the aggregated filamentary structures. A detailed description of growth characteristics at these conditions is presented in [65].

To analyze the effect of wavelength on the morphology of particles, the ablation of Cu target was carried out using fundamental (1064 nm) and third harmonic (355 nm) pulses of Nd:YAG laser (t = 5 ns). The images of particles synthesized at these conditions are shown in Fig. 3. In the case of 355-nm radiation, the morphology of particles always presented a regular spherical structure. Overall, the distribution of particle sizes was extremely broad (from a few nanometers to 100 nm, Fig. 3a). It was found that there are no significant morphology differences between initial and aging particles. The ablation using 1064-nm, 5-ns pulses showed a completely different pattern. The particle size became larger, while the size distribution range was narrower compared with the former case (Fig. 3b). In addition, we analyzed the ablation at different energies (45 and 80 mJ) of 5-ns, 1064-nm pulses, which showed no significant difference in the size and morphology of synthesized particles in these two cases.

Fig. 3
figure 3

SEM images of CuO/Cu2O nanostructures obtained in distilled water using 5-ns pulses. a Ablation using 5-ns, 355-nm radiation. b Ablation using 5-ns, 1064-nm radiation

Full size image

The temporal variation/stabilization of particles during aging process was analyzed using the sample obtained by 200-ps, 800-nm pulses. The SEM images at different periods from ablation are shown in Fig. 4. As was mentioned earlier, the particles in the colloidal solution obtained by PLAL rapidly oxidize during a short period (few hours), which changes the color of the solution from green to brown. After that, the color of solution became stable. The brown solution was analyzed by SEM, which showed that there was no obvious difference between the measurements at different periods from ablation. Thus, the morphology and particle size of particles in brown solution maintained excellent stability.

Fig. 4
figure 4

Dynamics of CuO/Cu2O particles’ morphology variations after laser ablation of bulk Cu using 200-ps, 800-nm pulses. a Within 24 h from ablation. b One month from ablation. c Four months from ablation

Full size image

The influence of pulse duration on different metal targets is different. At the beginning of these experiments, it was not clear which pulse duration could ablate the copper target more efficiently. The results show that the morphology and particle size of nanoparticles obtained with various pulse durations vary during the ablation process. In the case of 5 ns, the difference of the wavelength of heating pulses (1064 and 355 nm) showed a significant effect on the morphology. Similarly, the effect of pulse duration (200 ps and 30 fs) in ablation process was studied at 800 nm wavelength.

We did not perform the wavelength dependence in the case of picosecond ablation due to the low conversion efficiency towards the second harmonic of 800-nm, 200-ps pulses. Due to the low energy of 400-nm pulses, we did not perform the comparative analysis of the ablation of copper and formation of nanoparticles in water using two wavelengths (800 and 400 nm). On the other hand, we demonstrated the inefficiency in application of the 355-nm, 5-ns pulses for ablation and formation of copper nanoparticles due to the strong linear absorption of the light at the wavelength of 355 nm.

The ablation of copper using 532-nm, 4.5-ns pulses was performed in [66]. The larger ablation threshold of copper can be related to its larger thermal depth as compared to aluminum and titanium. The large thermal depth of copper results in a lesser localization of the excitation energy. Thereby, it causes the difficulties during the melting and evaporation of the copper surface. The difference between the ablation rates of metals in the visible and infrared ranges originates from the different optical and thermal proprieties of materials at these wavelengths.

To analyze the composition of particles, EDS of evaporated suspension was carried out. Apart from Cu and silicon, on which the samples were dried, there were some impurities. Among them, the presence of potassium (K) and chlorine (Cl) was detected. The EDS in Fig. 5a shows all the elements in the studied sample. The sources of these elements are very clear. Silicon comes from the base used to measure EDS, while potassium and chlorine may come from the impurities in the Cu target (see Table 2). However, we pay more attention to the proportion between the basic elements of our sample (oxygen and copper). The two main peaks of copper were located at 0.94 and 8.08 keV, respectively, while the peak attributed to oxygen appeared at 0.53 keV. The table embedded in Fig. 5a presents the parameters of two (O and Cu) elements. After removing impurities, only the relative atomic and weight parts of oxygen and copper are shown in the table. The comparative ratio of the atoms presented in the sample should be 1:1 for CuO while our measurements show 49.91:50.09, which is almost the same.

Fig. 5
figure 5

a EDS and b XRD of the particles produced during ablation of bulk Cu using 200-ps, 800-nm pulses

Full size image

Structural analysis of particles obtained by 200-ps, 800-nm pulses was carried out by XRD. The results are shown in Fig. 5b. All diffraction peaks were analyzed according to the standard monoclinic structures of Cu (JCPDS no. 00-001-1241), Cu2O (JCPDS no. 00-077-0199) and CuO (JCPDS no. 00-048-1548). There were three sharp peaks located at 2θ = 31.75°, 35.64° and 38.54°, which corresponded to diffraction from the (110), (002, 11-1) and (111, 200) of CuO (JCPDS no. 00-048-1548), respectively [67]. Meanwhile, a very strong peak appeared at 2θ = 61.71°, which corresponded to (220) of Cu2O (JCPDS no. 00-077-0199) [61]. These results fitted with the existing reports [68, 69]. CuO and Cu2O were also detected in the other samples, which were obtained by 5-ns (1064 and 355 nm) and 30-fs pulses. Notice that XRD image shows that the copper oxide formed after aging exhibited a crystalline state. Additionally, Cu2O has a self-oxidation process oriented to CuO [70]. Table 3 summarizes the particle sizes and composition obtained in present study and provides a comparison with previous reports.

Table 3 Observed sizes and compositions of synthesized particles and comparison with previous studies
Full size table

4 Conclusions

The laser ablation of Cu target in distilled water was carried out using different durations, wavelengths and energies of heating pulses. The experimental results show that colloidal solutions obtained at different conditions comprise CuO and Cu2O. It was found that the colloidal solution obtained by ablation will be oxidized during a very short time, which changes the color of the solution from green to brown. We have shown that, under ultra-short-pulse ablation conditions, the oxidation rate of suspension is slower and the particles are more stable.

We have demonstrated that the pulse duration has a decisive influence on the morphology of synthesized particles, and there is a corresponding relationship between them. Ultra-short pulses allow the synthesis of smaller sized particles, which means that the decrease of pulse duration directly promotes the decrease of produced particle sizes. In a small range of pulse energy variations, the change of pulse intensity has no significant effect on the morphology and size of the particles. When the same pulse duration was used, the morphology of particles depends on the wavelength of used radiation.

This study has demonstrated that PLAL has broad prospects as a simple, environmental-friendly and low-cost method to synthesize particles, particularly Cu nanourchins. Moreover, it is feasible to selectively synthesize particles and particle oxides with specific morphology, size and components by adjusting the laser–matter interaction conditions.