1 Introduction

Development of nano-materials is an attractive issue among scientists and industrialists, since the materials offer unique properties which can be of outstanding value in various fields like catalysis, photonics, electronics, and ceramics [1,2,3,4]. As a consequence, the number of investigations on preparing metal carbonates, as an important class of materials has been greatly boosted. The compounds are widely used due to their natural abundance and widespread applications in areas such as rubber, plastic, paint, and paper industries. Carbonate salts of metal are also used as precursors for preparing corresponding oxides [5,6,7]. CuCO3, as a common basic salt, has widespread applications in insecticides and fungicides, pigments, industrial catalysts, coloring of brass black, as poisoning antidote for phosphorus toxins, crude desulfurization, and wood preservatives [8, 9]. CuO nanoparticles (NPs), on the other hand, are used as pigments in ceramics leading to create red, blue, and green (or occasionally black gray, or pink) glazes. CuO is further used in magnetic storage systems, in the production of p-type semiconductors with fine band-gaps, catalysis and photocatalysis, photothermal applications, polishing tools, dry cell batteries and solar cells, sensors, super-capacitors and photo-detectors [10,11,12,13].

The photocatalytic application of semiconductors like CuO is usually in the degrading organic pollutants. Such photocatalytic materials are usually used in light induced self-sterilizing or self-cleaning of surfaces, water and air purification, odor control, bacterial deactivation and the like [14,15,16,17,18].

Direct precipitation of the particles of an insoluble salt through mixing solutions of its components is a routine chemical procedure, the critical step of which is the manipulation of the operating conditions affecting the properties of the product (i.e. its shape, size, composition, and substructure) [19, 20] and hence requires optimization.

It is known that optimization based on fractional factorial design of experiments considerably limits the number of experiments required for optimizing the operating parameters of the process, while as opposed to the sequential approaches, it also allows to acquire more information from the limited number of tests. Accordingly, the application of Taguchi robust experiment design for optimizing chemical procedures has been very common [21,22,23,24].

Given all of the above, the present study was initially focused on optimizing the direct precipitation of CuCO3 NPs through the Taguchi robust design, followed by the preparation of CuO NPs by thermal treatment of the carbonate salt as the precursor. In the light of the reports in the literature, indicating the concentrations of the cation and anion solutions, reactor temperature and the flow-rate of adding one solution to the other as the major factors influencing the dimensions of the product of a precipitation reaction [25], an orthogonal array was designed to optimize the effects of these parameters on the product. The final stage of the study was focused on evaluating the photocatalytic effects of the optimal carbonate and oxide particles in the UV-induced degradation of the methyl orange (MO).

2 Experimental

2.1 Materials and synthesis CuCO3 NPs

Lab-grade Na2CO3, and CuCl2 were obtained from Merck Co. and used as received. The CuCO3 particles were prepared via the direct precipitation reaction between aqueous solutions of Cu2+ and CO32− ions. This involved feeding the cation solution to that of the anion in the reactor under vigorous stirring. The produced CuCO3 NPs was centrifuged, repeatedly washed with distilled water and absolute ethanol, before drying at 80 °C for 120 min. The factors influencing the results of the precipitation were, as mentioned above, optimized by a Taguchi-based orthogonal array (L9). These variables were the Cu2+ and CO32− concentrations (CCu, CCO3), reactor temperature (T) and flow rate of the copper solution (F), the values of which are presented in Table 1.

Table 1 Designed trials for the process optimization and average particle size of occasioned copper carbonate
Full size table

2.2 Synthesis of CuO NPs

CuO was obtained by the thermal treatment of optimally prepared CuCO3, at 350 °C in a furnace under a static flow of air for 120 min. The typical experiments involved transferring about 0.2 g of the precursor to a 20 mm × 40 mm alumina crucible, sealing the crucible with a thin aluminum foil, and placing it in a furnace.

2.3 Characterization of the CuCO3 and CuO NPs

Field emission scanning electron microscopy (FE-SEM) studies intended to characterize the samples, were performed using a ZEISS Sigma/up instrument. To perform the analyses the samples were loaded onto the system after coating with a layer gold. The coating of the samples with gold was performed using a SCD005, BAL-TEC sputter coater. X-ray powder diffraction (XRD) analyses were performed using a Rigaku D/max 2500 V diffractometer and the Fourier transform infra-red spectroscopy (FT-IR) analyses were carried out on a Perkin-Elmer instrument in a wavenumber window of 4000–500 cm−1. The FT-IR samples were loaded into the instrument through the KBr pellet approach. The thermogravimetric data were acquired using a Perkin-Elmer STA 6000 instrument and an Avantes Spectrometer Avaspec-2048-TEC (Electron Engineering Pishro Pajuhesh, Iran) was used to collect the UV–Vis diffuse reflectance spectroscopy (DRS) data of the NPs at ambient temperature and the MO content of the water samples before and after photocatalytic treatments.

2.4 UV–Vis DRS studies

To determine the optical band gap energy (OBGE) of the CuCO3 and CuO NPs, UV–Vis DRS analyses were performed. OBGE is defined as the lowest energy a photon should have to transfer an electron from the valance band of a semiconductor to its conduction band. The relation between energy and the absorption edge of a semiconductor is described by the following equation also known as Tauc’s equation [26, 27]:

$$\alpha h\upsilon ={\text{ A}}{\left( {h\upsilon - {E_g}} \right)^\eta }$$
(1)

where h, υ, α, Eg, A are Planck’s constant, frequency of the light, absorption coefficient, OBGE and a constant. η is another constant and can be 1/2 or 2 depending on the nature of the electron transition. In this case η was considered 1/2 for both NPs due to the direct nature of their electron transition mechanisms.

2.5 Photocatalytic activity

UV-induced photocatalytic behaviors of the CuCO3 and CuO NPs in the presence of MO were monitored in a Pyrex cylindrical double-pipe air-lift photoreactor equipped with a high pressure mercury lamp (250 W, λ > 280 nm) as the UV source.

The reaction suspensions were prepared through mixing 0.05 g of one of the NPs with 500 mL of a 5 mg/L MO solution and stirring the suspension in the dark for 20 min to allow the adsorption phenomena to reach their equilibrium conditions. Next the suspension was subjected to UV light from the lamp under constant aeration. In addition to the initial MO concentration of the reaction mixture, the MO content of the mixture was repeatedly measured at 10 min intervals during the course of the reaction. This involved taking small aliquots of the sample and their UV–Vis analysis. The reaction temperature was kept constant (25 °C) using a cooling system.

The quantitative analyses were performed using the Lambert–Beer equation [28]:

$${\text{A}}=\varepsilon {\text{bC}}$$
(2)

where A, ε, b and C reflect the absorbance of light, molar absorptivity, path length of light through the sample solution, and concentration of the analyte. Further considering C0 and A0 as the initial MO concentration and corresponding absorbance and Ct and At, as these values at any time t, the following equations were also derived from Eq. 2:

$$\frac{{\text{A}}}{{{{\text{A}}_0}}}=\frac{{\text{C}}}{{{{\text{C}}_0}}}$$
(3)
$${\text{Degradation~efficiency}}~\left( \% \right)=\frac{{{{\text{A}}_0} - {{\text{A}}_t}}}{{{{\text{A}}_0}}} \times 100$$
(4)

The kinetics of the photocatalytic degradation reaction was studied based on the Langmuir–Hinshelwood model, expressed as below [29,30,31]:

$$- \frac{{{\text{dC}}}}{{{\text{dt}}}}={{\text{k}}_{{\text{app}}}}{\text{C}}$$
(5)

where C, kapp, t and \(- dC{\text{/}}dt\) represent the concentration of the degraded species (i.e. MO), reaction rate constant, degradation time and reaction rate.

3 Results and discussion

3.1 Optimization of the procedure for preparing CuCO3

Controlling the size and morphology of salts prepared through the direct precipitation reaction between the solutions of their ingredients can be rather complex [32]. Optimizing the parameters influencing the outcome of the reaction requires performing a large number of experiments. As an alternative, statistical procedures can be used to minimize the number of the required experiments [33, 34].

Accordingly, as summerized in Table 1, the effects of the concentrations of Cu2+ and CO32− solutions (CCu and CCO3), flow rate of adding the cation solution to that of the anion (F) and reaction temperature on the dimensions of the CuCO3 were evaluated at three levels based on an orthogonal array (OA9) proposed by Taguchi [35, 36]. The CuCO3 under the conditions in Table 1, were evaluated through FE-SEM and some of the results have been presented in Fig. 1. The FE-SEM results indicated that the size of the CuCO3 particles is a function of these conditions. The average sizes of the CuCO3 particles obtained at each experiment, given in the last column of Table 1, were used as inputs to calculate the mean size of the CuCO3 particles attributed to the effect of each examined parameters at any level [37, 38], and the results are presented as bar graphs in Fig. 2. The effects of the three levels of CCu and CCO3 (i.e., 0.01, 0.05 and 0.1 mol/L) on the size of the CuCO3 NPs is illustrated in Fig. 2a, b, which indicate 0.01 M as the optimal value for CCu, and 0.1 M is that of CCO3.

Fig. 1
figure 1

FE-SEM images of copper carbonate NPs fabricated at diverse trials of Table 1: a trial 1; b trial 3; c trial 5; and d trial 8

Full size image
Fig. 2
figure 2

Middling effects of explored parameters at triple levels on the copper carbonate particle size: a concentration of copper ion solution, b concentration of carbonate solution, c reactor temperature and d feeding flow rate

Full size image

Figure 2c illustrates the effect of the influence of the reactor temperature at the three levels of 0, 25 and 50 °C, while Fig. 2d indicates that of the flow rate at which copper ion solution is added to the carbonate solution (i.e. 2.5, 10 and 40 mL/min). Clearly, changing the levels of these parameters has slight effects on the size of the CuCO3 NPs.

The role of these parameters on the size of the NPs were further evaluated by ANOVA and the results are shown in Table 2. These indicate that at 90% confidence interval, and ignoring the possible interactions of the parameters, CCu and CCO3 are key factors in determining the size of the NPs, while T and F do not have significant effects in this regard. Based on the results, the optimal values for CCu and CCO3 are 0.01 and 0.1 mol/L, respectively. Also according to the Taguchi process, the size of the NPs prepared under the optimal conditions can be given by the equation below [39, 40]:

$$Y_{{opt}} = \frac{T}{N} + \left( {\mathop {\text{C}}\nolimits_{{{\text{Cu}}}} - \frac{T}{N}} \right) + \left( {{\text{C}}_{{{\text{CO3}}}} \, - \frac{T}{N}} \right)$$

where T/N is the overall average of the size of CuCO3 NPs, prepared through all experiments in Table 1. Yopt is the optimal size of the NPs and CCu and CCO3 represent the mean size of the CuCO3 NPs (Fig. 2) at optimal Cu2+ and CO32− concentrations.

Table 2 ANOVA data on the size of copper carbonate particles produced by inspected precipitation reaction
Full size table

The confidence interval (CI) of the resulting values is calculated using the following equation [39, 40]:

$$CI= \pm \sqrt {\frac{{{F_\alpha }({f_1},{f_2}).{V_e}}}{{{n_e}}}}$$

in which V e is the error variance, F α (f 1 ,f 2 ) is the critical value of F at the significance level α (here 90%), f 1 and f 2 are the degrees of freedom (DOFs), (f 1  = DOF of the mean (always 1) and f 2  = DOF of the error in Table 2). The effective number of replications (n e ) is also given by:

$${n_e}=\frac{{Number\,of\,experiments}}{{DOF\,of\,mean\,\left( {always\,1} \right)\,+\,DOF\,of\,all\,factors\,at\,optimum\,conditions}}$$

The results proved that the optimal size of the CuCO3 NPs is 17 ± 2 nm, which complies with the results of the third trial in Table 1 (CCu = 0.01 mol/L and CCO3 = 0.1 mol/L). Figure 1b shows the FE-SEM image of the CuCO3 NPs under these conditions. According to this figure and the data in Table 1, the size of the NPs is 16 nm, which agrees with the calculated value of 17 ± 2 nm. Once the optimal CuCO3 NPs were prepared, their chemical, thermal and optical properties were evaluated using FT-IR, thermogravimetry (TG) analysis and DRS.

3.2 Characterization of CuCO3 NPs

The results of the FT-IR analysis of the optimal CuCO3 NPs are given Fig. 3. As seen in this figure, the spectrum contains bands associated with the vibrations of CO32− ion in the range of 400–1600 cm−1. The strong absorption around 1395 cm−1 is corresponds to the asymmetric stretching vibration signals, and that at 1508 cm−1 was attributed to the υ3 mode of CO32− ions. The stretching modes of carbonate ion are reflected by the peaks at 1073, 847, 740 and 693 cm−1 [41, 42]. The side peak at 3370 cm−1 reflects the bending and stretching of the hydroxide groups of the residual surface-absorbed water molecules [43].

Fig. 3
figure 3

FT-IR spectrum of the copper carbonate NPs gained at optimal conditions

Full size image

TG experiments are usually performed to monitor the thermal behavior of inorganic compounds [44], and hence, the optimal CuCO3 NPs were subjected to TG studies and the results are given in Fig. 4. The curve in this figure indicates that the CuCO3 NPs undergone two mass loss steps. The first step occurred from 40 to 200 °C and accounted for a 2.2% loss of the total mass of the sample. This was attributed to the evaporation of the surface-adsorbed water. The second step was observed between 200 and 350 °C and led to a 28.4% mass loss, which could be due to the decomposition of the carbonate to the oxide salt through losing CO2 and CO species. This was very evident between 200 and 300 °C, while over 350 °C no further mass loss was observed, indicating the completion of the decomposition reaction. The total mass loss of the sample through the TG experiments from 40 to 350 °C came to around 30.6%. Based on these the calcination of the CuCO3 NPs upto 350 °C was concluded to lead to the complete decomposition of the carbonate salt to CuO, and hence 350 °C was chosen and applied for production of the CuO NPs.

Fig. 4
figure 4

TG curve of copper carbonate NPs; sample mass 3.0 mg; heating rate 10 °C/min; nitrogen atmosphere

Full size image

3.3 Preparation of CuO NPs

The copper oxide NPs were formed through calcinating optimal CuCO3 NPs at 350 °C for 120 min. The SEM image of the product (Fig. 5) indicated the particles having a spherical morphology and an average size of around 33 nm.

Fig. 5
figure 5

SEM image of copper oxide NPs fabricated by calcination of carbonate precursor

Full size image

The results of XRD and FT-IR analyses are presented in Figs. 6 and 7. Comparing the XRD patterns of the oxide salt (Fig. 6) with JCPDS 01-080-1916 revealed that all diffraction peaks of the sample comply with a monoclinic phase (cc space group, a = 4.6927, b = 3.4283 and c = 5.1370 Å). The pattern further indicate the high crystallinity and phase purity of the product. Using the Debye–Scherrer equation:

$${\text{D}}=\frac{{0.9{{{\uplambda}}}}}{{{{{\upbeta}~Cos{\uptheta}}}}}$$

where λ or the X-ray radiation wavelength was 0.154059 nm, β is corrected band broadening, and θ represents the Bragg angle [45], the average crystal size of the product was determined to be nearly 29 nm.

Fig. 6
figure 6

XRD pattern of copper oxide prepared by calcination of carbonate precursor

Full size image
Fig. 7
figure 7

FT-IR spectrum of copper oxide NPs obtained by calcination of carbonate precursor

Full size image

The FT-IR spectrum (Fig. 7) clearly shows that the calcination of the precursor at 350 °C fully decomposed the carbonate ions and all peaks could only be ascribed to CuO as reported in the literature [46]. The bands at 591 and 542 cm−1, in this figure, reflect the vibrational modes of Cu–O bond, while the high-frequency modes at 518 cm−1 could be attributed to the stretching vibrations of the Cu–O bonds.

3.4 Photocatalytic properties of the NPs

Figure 8a, b depicts the DRS spectra of the optimal CuCO3 and CuO NPs and the corresponding Tauc’s plots are further presented in Fig. 9a, b. Based on the Tauc’s plots CuCO3 NPs have an OBGE of around 3.85 eV, while that of the oxide salt is about 3.6 eV. The corresponding band gaps can be observed at 322 and 344 nm respectively. The calculated OBGE for CuCO3 NPs is different from the value reported for bulk CuO [47], yet it fully matches the value reported elsewhere [48].

Fig. 8
figure 8

Diffuse reflectance UV–Vis spectra for the fabricated NPs: a copper carbonate, and b copper oxide

Full size image
Fig. 9
figure 9

Tauc’s plots of fabricated NPs: a copper carbonate and b copper oxide

Full size image

Figure 10 illustrates the photocatalytic behavior of the optimal CuCO3 and CuO NPs in the UV-induced degradation of MO in water. Further, Fig. 11 shows the C/C0 and degradation efficiency behaviors against the illumination time. According to these results the concentration of MO is effectively reduced in the presence of CuCO3 and CuO NPs and the respective maximum yields reached after 120 min of UV light irradiation, were 64 and 90%.

Fig. 10
figure 10

UV–Vis absorbance spectra of MO after irradiation during diverse interval times in the presence of 0.1 g/L photocatalyst: a copper carbonate and b copper oxide NPs

Full size image
Fig. 11
figure 11

Degradation of MO solution during UV irradiation in the presence of photocatalyst: a copper carbonate and b copper oxide NPs

Full size image

The pseudo first-order kinetic plots, obtained for the reaction in the presence of the two particles can be seen in Fig. 12. The reaction rate constants were determined using the slopes of the linear regressions of the two plots. All data are summerized in Table 3, which can be held as proof that the two NPs are suitable photocatalysts for the light-induced elimination of organic pollutants like MO.

Fig. 12
figure 12

Plots of Pseudo first order kinetics of MO degradation in the presence of: a copper carbonate and b copper oxide NPs

Full size image
Table 3 Calculated rate constants for pseudo first order reactions and degradation efficacy of photocatalysts after 120 min UV irradiation
Full size table

4 Conclusion

CuCO3 NPs with definite size and morphology were fabricated successfully in this study without using non-aqueous solvents or applying severe operating conditions via the facile and scale up able direct precipitation technique. Also, CuO NPs were prepared via single step solid state thermal decomposition of the optimal carbonate precursor NPs. The optimization was performed through the Taguchi robust design and the results showed that only the concentrations of the cation and anion significantly influenced the dimensions of the product particles. The optimal conditions led to the production of copper carbonate with an average size of 17 nm. Results of TG analyses revealed that the CuCO3 NPs undergone a decomposition reaction to form CuO NPs of 33 nm in average size. The advantages of the optimized processes include simplicity, low-cost, and good potentials for upscaling. The CuCO3 and CuO NPs were further evaluated as photocatalysts in the UV-induced degradation of MO and the respective maximum degradation efficiencies were found to be 64 and 90% after 120 min.