The morphology of the synthesized samples was studied by means of an electronic scanning microscope (SEM) with a direct-filament tungsten cathode (FEI Company) Quanta 200i 3D. Figure 1 shows the morphology of the ZnO particles obtained at various starting NaOH concentration. An analysis of the SEM images highlights that an increase NaOH concentration in the growth solution (keeping all other synthesis parameters unchanged) leads to a decrease in the geometric parameters (diameter and length) of the synthesized ZnO particles (Table 1). In particular, at NaOH concentration of 0.14 M, large pointed elongated particles form (Figure 1A) while at a concentration of 0.27 M smaller particles of irregular spherical shape (Figure 1B) appear. At (0.4–0.7) M ZnO grows in the form of thin 2D plates (Figures 1C–E). An increase in the alkali concentration of the growth solution makes it possible to obtain thinner 2D ZnO structures, but at higher NaOH concentration, the ZnO particles tend to coalesce (see Figure 1F).

Transmission electron microscopy (TEM) images were obtained with a JEM-2100 transmission electron microscope operating at an acceleration voltage of 200 kV (Figure 2). Results of TEM analysis of ZP 15 and ZP19 samples with different morphology correspond to SEM images (Figures 1B,C).

Table 1 shows that an increase of the sodium hydroxide concentration in the growing solution up to 0.4 M leads to a gradual increase on the ZnO NP aspect ratio (AR) but with a further increase of the NaOH concentration the AR decreases. The ZP 19 sample has the highest AR.

XRD measurements were performed on an X-ray diffractometer X’pert MPD PRO (PANalitical) under the same conditions for all samples. The XRD peaks have been labelled according to the reference spectra (JCPDS Card No. 80-0075) of wurtzite ZnO structure (Figure 3).

Among all the observed diffraction peaks, the (101) reflex is particularly intense (Figure 3) that demonstrates the crystalline nature of the nanoplatelets. A high degree of ordering along (101) direction in comparison with the intensity of other peaks has a significant effect on aspect ratio of particles (Zak et al., 2013). All synthesized samples of the considered series show a slight difference in the intensity and width of the diffraction peaks.

The ZnO crystallites sizes (D) were estimated based on the XRD analysis for the most intense peak (101) by employing the Scherrer’s formula, D = k λ/βcosθ (1) here k = 0.89 is a dimensionless coefficient (Scherrer’s constant), λ = 1.54 Å is the wavelength of CuKa radiation, θ is the diffraction angle, and β is the width of the peak at half-height in radians. The values obtained in this way are consistent with the electron microscopy observation. The evaluating results of the crystal lattice parameters of the samples using Williamson—Hall method (Mote et al., 2012) are shown in Table 1. All the studied ZnO samples show a hexagonal wurtzite structure with P63mc space group (Kumaresan et al., 2017) and good crystalline quality, since all diffraction reflexes are very close to the reference sample peaks and correspondingly also the estimated lattice parameters (JCPDS Card No. 80-0075, a—3.2539 Å and c—5.2098 Å).

The optical characteristics of the materials is particularly relevant since it allows predicting the possible photocatalytic behavior of the samples under UV/Vis illumination. The optical absorption spectra of all studied samples, measured with a two-beam UV/Vis Lambda 35 spectrophotometer (PerkinElmer), are shown in Figure 4A.

All ZnO NP samples are transparent in the visible region of the electromagnetic spectrum and absorb light in the UV range. Figure 4B shows the Tauc’s graphs which can be used to identify the samples absorption edge and thus evaluate the optical band gap (E_g) using the following equation: ( αhv ) =A ( hv-Eg ) n (2) here α is the absorption coefficient; hν is the photon energy; A is a proportionality coefficient; n = 0.5, since ZnO refers to direct bandgap semiconductor material (Rusdi et al., 2011). The equation for the proportionality coefficient A has the form A = α d=−lnT (3)

Here d is the thickness of the film; T is the transmittance coefficient. The plot of value (αhν)² against energy hν allows calculating the value of band gap E_g by extrapolating the linear part of the curve to the horizontal axis. According to Figure 4B, the width of the optical band gap of the samples is in the range of 2.82–3.1 eV. The estimated band gap values for all ZnO NP samples are summarized in Table 1. In all cases, the actual gap value turned out to be smaller than the reference ZnO value of 3.37 eV (Huang et al., 2015). In particular, the smallest value was found for the ZP 26 sample (∼2.82 eV) while the largest for the ZP 20 sample (∼3.1 eV). The relationship between the E_g value and the photocatalysis efficiency was discussed in (Ong et al., 2018) and shown in (Di Mauro et al., 2017; EL-Dafrawy et al., 2021; Ramírez et al., 2021).

The photocatalytic activity of our ZnO samples was studied by verifying the degradation of a test organic molecule, namely the rhodamine-B (RhB) dye. Upon UV optical excitation of the ZnO nanoparticles, electron-hole (e⁻—h⁺) pairs are formed, which, due to the large surface/volume ratio, are easily transferred to the NP surface inducing the rhodamine-B degradation (Ruellas et al., 2019; Azfar et al., 2020; Blažeka et al., 2020). According to the literature [see e.g., (Ong et al., 2018; Khiavi et al., 2019; Nguyen and Nguyen, 2020)], the photocatalytic degradation mechanism can be summarized as follows: absorption of a UV photon excites an electron to the NP conduction band (CB); the electron, e⁻, moves to the surface of ZnO NP and it is transferred to an oxygen molecule dissolved in the rhodamine containing solution; the oxygen molecule, O₂, is transformed into a superoxide anion •O₂ ⁻, and subsequently, after capturing a proton from the solution it generates a HOO• radicals. At the same time the hole, h⁺, generated in the NP valence band (VB) moves to the NP surface reacting either with a water molecule or with a hydroxide (OH⁻) originating a reactive OH• species. Hydroxyl radicals and superoxide anions, being the most powerful oxidizing intermediates, cause decomposition of RhB molecules adsorbed on the surface of ZnO NPs. Thus, the negatively charged surface of ZnO promotes the release of holes, which decompose organic matter without an intermediate step. These intermediates are eventually converted into CO₂, H₂O and mineral acids, as summarized in Figure 5.

Figure 6 shows the degradation spectra of the aqueous RhB solution containing the different ZnO samples described in the previous sections. As it is apparent from an analysis of Figure 6 the maximum absorption intensity of the initial RhB solution with ZnO powder occurs at 554 nm.

With increasing exposure time, the intensity of the RhB absorption gradually decreases in the presence of the ZnO NPs, which indicates a decrease in the concentration of RhB dye as time proceeds. The relative concentration of the RhB dye decreases with increasing exposure time, which implies that, for all our ZnO samples, the RhB dye significantly degrades on the ZnO NP surfaces under the influence of UV illumination, already after 30 min of exposure.

It is noted that with an increase in the alkali concentration in the growth solution from 0.14 M (ZP 20, Figure 6A) up to 0.4 M (ZP 19, Figure 6C), the photocatalytic activity of the ZnO samples increases, with a further increase in the NaOH concentration, the photoactivity decreases (ZP 23, Figure 6F).

To quantitative analysis the photocatalytic activity of the different ZnO samples, we calculated the ratio R = C/C₀ at a function of exposure time (see Figure 7A), where C is the dye concentration after irradiation with UV light at maximum absorbance intensity (554 nm), while C₀ is the initial dye concentration. The value of the photodegradation rate k of the ZnO NPs was estimated using the kinetic model proposed by Langmuir–Hinshelwood (Di Mauro et al., 2017; Magalhães et al., 2017; Godin et al., 2018), according to which: ln ( C/ C 0 ) =−kt (4)

Hence k = l n R − t = l n   ( C 0 / C ) t (5)

The dependence of ln (C₀/C) as a function of the exposure time is shown in Figure 7B. The results of the photocatalytic efficiency of the synthesized samples with respect to RhB degradation, including the minimum k_min, maximum k_max after 30, 60, 90, 120, and 150 min and the average rate k_av of dye degradation are shown in Table 2. Value k_av characterizes an angle of the linear curves slope in Figure 7B. It is usualy k_max was observed after first 30 min of irradiation, minimal photodegradation rate k_min was noted after 150 min of exposure. The percentage of the degraded dye in an aqueous solution for 150 min of illumination in the presence of ZnO NPs is estimated from the ratio R* = 100 (1-R).

From an analysis of Figures 7A,B and from the data reported in Table 2, it can be noticed that the photocatalytic activity of ZP 19, ZP 22, ZP 26 samples is higher than that of ZP 15, ZP 20, ZP 23 samples. The dependence of the photocatalytic activity of the synthesized samples on the NaOH molar concentration during synthesis is nonmonotonic, as shown in Figure 8A.

The XRD and PL data have been analyzed to determine the possible reasons for this non-monotonic relationship. The width of the XRD line of the most intense reflections 100, 002, and 101 depending on the molar concentration of NaOH in the growth solution are shown in Figure 8B. Figure 9A shows the PL spectra for the obtained ZnO samples, and Figure 9B shows the dependence of the integrated PL for the same samples on the molar concentration of NaOH. It can be seen that the XRD data for the 002 reflections (Figure 8B) and the integral PL intensity (Figure 9B) have nonmonotonic dependences on the NaOH molar concentration and correlate well with the data on photocatalytic activity (Figure 8A).

The correlation of the photocatalytic activity of the samples with the XRD and PL data needs to be discussed. As can be seen from Figure 8A, the half-width of the 100 and 101 reflections increases monotonically with an increase in the NaOH molar concentration, i.e., the crystallite sizes along the corresponding crystallographic directions decrease. On the other hand, there is a correlation between the half-width of the 002 reflection (Figure 8B) and photocatalytic activity (Figure 8A), and the maximum photocatalytic activity is reached at the maximum particle size along the 002 direction.

Estimates of the particle sizes along 100 and 101 using the Scherrer formula give a value of 31 ± 2 nm at a NaOH concentration of 0.14 M, and the particle sizes decrease linearly with increasing NaOH concentration to 25 ± 2 nm at a concentration of 1 M. A similar estimate using the Scherrer formula shows that the particle size along 002 nonmonotonically depends on the concentration of NaOH, the average particle size along 002 crystallographic direction reaches a maximum of 42 ± 2 nm at concentrations of 0.4–0.7 M (Figure 8A). The data on the morphology of the samples shown in Figures 1, 2 also indicate that rods and plates of ZnO nanocrystals mainly grow at a growth solution concentration of 0.4–0.7 M NaOH.

At the same concentrations, the maximum photocatalytic activity is reached (Figure 8B) so it can be concluded that both the size of the particles and their shape, which provides a high specific surface area, have an important influence on the photocatalytic activity. Since the particle sizes along the crystallographic direction 002 reach a maximum at concentrations of 0.4–0.7 M, i.e., growth in this direction is very favorable, one can expect a low concentration of defects in the volume and on the surface of the synthesized samples.

The PL spectra also agree with this conclusion. As shown in Figure 9, PL intensity of the radiative recombination through deep defect states in samples obtained at NaOH concentrations in the growth solution of 0.4–0.7 M reaches a minimum value. These results may indicate a reduced concentration of defects in the volume and surface of ZnO nanocrystals in such samples. That is, the intensity of radiative recombination is minimal in the samples due to the low concentration of defects. These results unambiguously indicate that the concentration of defects in the volume and surface of ZnO nanocrystals is low, and, accordingly, the photogenerated free-carrier lifetime is long in such samples. It is the factor of the long lifetime of nonequilibrium carriers that can be the main reason for the high photocatalytic activity of ZnO samples. To concretize the possible photocatalytic mechanism of ZnO nanoparticles, it is necessary to carry out experiments on the analysis of the active species capture, similarly to (Xu et al., 2022). Such work is planned for the next stage of our research.

One of the advantages of photocatalysis is the reusability of the catalyst (Pirhashemi et al., 2018). In this work, experiments were carried out to verify the possibility of multiple use of the ZnO NPs as photocatalysts. To check ZnO reusability, the photocatalytic activity of the ZP 19 sample was tested five times. The irradiation time for each test was 150 min. After each rhodamine-B degradation test, the solution was centrifuged to separate the ZnO NPs, that were then added to a new fresh dye solution. The initial dye concentration was the same for all tests.

Figure 10 shows that the photocatalytic activity of ZP 19 sample remained almost unchanged after five cycles. These results confirm that ZP 19 photocatalyst has good recyclability properties.