Application of Response Surface Methodology for Optimization of Nanosized Zinc Oxide Synthesis Conditions by Electrospinning Technique

Zinc oxide (ZnO) is a well-known semiconductor material due to its excellent electrical, mechanical, and unique optical properties. ZnO nanoparticles are widely used for the industrial-scale manufacture of microelectronic and optoelectronic devices, including metal oxide semiconductor (MOS) gas sensors, light-emitting diodes, transistors, capacitors, and solar cells. This study proposes optimization of synthesis parameters of nanosized ZnO by the electrospinning technique. A Box–Behnken design (BB) has been applied using response surface methodology (RSM) to optimize the selected electrospinning and sintering conditions. The effects of the applied voltage, tip-to-collector distance, and annealing temperature on the size of ZnO particles were successfully investigated. Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images confirm the formation of polyvinylpyrrolidone-zinc acetate (PVP-ZnAc) fibers and nanostructured ZnO after annealing. X-ray diffraction (XRD) patterns indicate a pure phase of the hexagonal structure of ZnO with high crystallinity. Minimal-sized ZnO nanoparticles were synthesized at a constant applied potential of 16 kV, with a distance between collector and nozzle of 12 cm, flow rate of 1 mL/h, and calcination temperature of 600 °C. The results suggest that nanosized ZnO with precise control of size and morphology can be fabricated by varying electrospinning conditions, precursor solution concentration, and sintering temperature.


Introduction
Zinc oxide (ZnO) is an eminent semiconductor material due to its excellent electrical, mechanical, and unique optical properties. ZnO has a wide direct band gap width (3.37 eV), a vast excitation binding energy (60 meV), and ultraviolet (UV) absorption ability at room temperature, as well as various distinctive characteristics, including high electron mobility and excellent transparency [1][2][3]. It is well known that particle size reduction significantly affects the fundamental properties of semiconductor materials, especially nanosized metal oxides, which are typically 1-100 nm in size [4]. Nanosized metal oxides and semiconductors are of particular interest because they exhibit physical and chemical features and benefits [5,6], where their structural properties, lattice symmetry, and cell parameters as well as electrical and optical characteristics vary depending on their particle size [7,8].
In particular, ZnO nanoparticles are commonly used for the industrial manufacturing of microelectronic and optoelectronic devices, including metal oxide semiconductor were found to be the most effective factor in determining of the fiber diameter [35]. Nowadays, RSM has been frequently utilized in the evaluation of the electrospinning conditions for fiber and particle synthesis [36][37][38], where statistical software Design Expert can be used for designing these experiments. So far, there has been no literature found on the optimization of ZnO nanoparticle-obtaining conditions by RSM in the electrospinning process using PVP and ZnAc. In this study, ZnO nanoparticles were synthesized from a PVP/ZnAc precursor solution using the electrospinning method and sintering, and their synthesis parameters were optimized by utilizing the RSM method. The use of RSM to optimize the conditions for electrospinning and sintering of ZnO nanoparticles facilitates the procedure and allows for the consideration of interactions between the selected parameters as well as the advancement of the process. Furthermore, guiding the experimental approach to save time and lower the cost assists in the control of morphology and diameter by conducting a smaller number of experimental runs. In addition, RSM helps evaluate the impact of each factor on the resulting ZnO diameter through a high-quality process [39,40].
PVP/ZnAc fibers were previously electrospun with fixed parameters such as precursor solution, annealing temperature, and spinning time to obtain a uniform diameter [41]. In this study, an efficient investigation of ZnO nanoparticle synthesis was carried out by optimizing and varying the parameters such as applied voltage and distance between collector and nozzle, sintering temperature, taking into account their interaction and effect of each condition. The modeling was conducted using Design Expert Software and the RSM method. Ideal conditions were set, under which pure and well-shaped ZnO nanoparticles with higher crystallinity and an average minimal diameter of 43 nm were synthesized by electrospinning and sintering processes. In our study, a power of 16 kV, a distance of 12 cm from the nozzle to the collector, and a sintering temperature of 600 • C were found to be optimal conditions for fabricating the smallest particles.

ZnAc-PVP Nanofiber Preparation and ZnO Synthesis
PVP solution in ethanol (8 wt%) was prepared and stirred for 5 h. Then, ZnAc (1 g) was added to 10 mL of PVP solution and stirred for an additional 5 h. The spinning solution was injected through a stainless steel needle connected to a high-voltage DC power supply of 12-16 kV and a distance of 8-12 cm and electrospun. Nanofibers were accumulated on the Al foil, and the obtained PVP-ZnAc composite nanofibers were annealed in air at 600-800 • C (at a heating rate of 5 • C/min) for 2 h.

Experimental Design, Statistical Analysis, and Optimization by RSM
In the electrospinning and sintering processes, the most effective synthesis parameters and their ranges that affect ZnO particle diameter and properties were determined by performing a preliminary experiment and reviewing the literature [42]. These parameters were the applied potential (X 1 : 12-16 kV), the distance between the collector and the nozzle (X 2 : 8-12 cm), and the calcination temperature (X 3 : 600-800 • C). Fifteen experiments were conducted based on the BB design and optimized by the RSM method. Each experiment was run twice, and the average response was taken into consideration. The ranges of minimum and maximum coded values of process parameters in the present study were fixed according to the initial trial runs (presented in Table 1). A second-order polynomial model with the coded independent variables (X i , j ) was used to obtain minimized-size ZnO particles (Y) as shown in the equation below (Equation (1)): Here, Y is the response variable to be modeled (ZnO size), X i and X j define the independent variables, b 0 is the constant coefficient; b i is the coefficient of linear effect, b ij is the coefficient of interaction effect, b ii is the coefficient of quadratic effect, and n is the number of variables. To specify the significance of the model, an ANOVA (analysis of variance) was conducted. The response surface and contour plots of the model-predicted responses were applied to specify the interactive relations between the significant variables. Design Expert, v. 8.0.7.1 (Stat-Ease Inc., Minneapolis, MN, USA), was used for designing the tests as well as regression and graphical analysis of the obtained data.

Characterization
The microstructure and morphology of the fibers and ZnO nanoparticles were observed by scanning electron microscopy (SEM, EDX ZEISS Crossbeam 540, Zeiss, Germany). Samples for SEM analysis were coated with 5 nm gold by an automatic sputter coater (Q150T, Tokyo, Japan) to reduce charging. The structure of the obtained fibers and ZnO nanoparticles were observed by transmission electron microscopy (TEM, JEOL JEM-1400 Plus, Peabody, MA, USA). The accelerated voltage of the TEM was 120 kV. The structural properties of the ZnO samples were characterized by X-ray diffraction (XRD; Rigaku Smart-Lab, Tokyo, Japan). The optical properties of the obtained samples were studied using an Evolution 300 UV-Visible Spectrophotometer.

Response Surface Model
This study used an efficient RSM modeling approach based on a three-level BB design with three variables to reveal the influence of the chosen spinning and sintering parameters on the size of the electrospun and sintered ZnO nanoparticles. Optimization is highly desired to see the effect of the various electrospinning conditions on the structural integrity of fibers. To achieve optimal conditions, a set of measurable investigative factors such as applied potential, distance between nozzle and collector, sintering temperature, and the observed response of the specified conditions were selected. Among the chosen parameters, the most influential one is the applied potential because it directly causes elongation of a polymer fluid drop retained at a needle tip by surface tension. At some threshold potential difference, the charge repulsion overcomes surface tension, and viscous forces within the fluid drop. A complex force balance governs the ejection of a fluid jet and the subsequent creation of nanofibers [20]. It has been noted that the formation of electrical arcs between electrodes is responsible for the discontinuous withdrawal of fluid jet beyond the maximum potential difference and the smallest separation distance set [43]. In the current research, the potential difference was divided into three levels, ranging from 12 kV to 16 kV. The spinning of the fibers started at 12 kV, and an accumulation of polymer solution mass near the needle tip, combined with the short residence time of the droplets, can explain the erratic behavior of the solution. The fibers obtained at this voltage in the observed microscopic fields had larger diameters. Bead-free weaved threads were electrospun under a constant feed rate (1 mL/h) and separation distance value (12 cm) with an applied potential difference of 12 kV (Figure 1a), where the average size of the obtained fibers was 193 nm; 14 kV ( Figure 1b), where the average size of the obtained fibers was 125 nm; and 16 kV (Figure 1c), where the average size of the obtained fibers was 91 nm. Drop formation at the needle tip should have a longer residence time, and the charged ejected jet should have a longer "time-of-flight" under these ideal conditions, which should boost solvent evaporation and the formation of fiber from the solution [44].  Three variables of the spinning-calcination temperature (X1), tip-to-collector distance (X2), and applied potential (X3)-were selected for the optimization process to form ZnO particles. The interaction effect of the chosen parameters on the response observed in the experimental runs can be explained using the analysis of variance (ANOVA). Furthermore, the adequacy of the model was examined using diagnostic diagrams, and the model should be validated by evaluating the optimum experimental conditions, as previously explained [45]. The optimization results show the parameter interaction effect, which consists of 15 experiments, as is presented in Table 2. For each experiment, the ZnO diameter (the response) was measured; a two-set average is noted in Table 3, column Y. When one decides if the overall results are significant, the F statistic must be used in combination with the p-value. The p-value indicates the degree to which the data is consistent with the null hypothesis. The successive p-value of <0.0001 and F value of 99.22 indicate significant model terms. In this scenario, there were key model terms: A, B, C, A 2 , B 2 , and C 2 . Only the quadratic terms for the two electrospinning variables made it into the refined model, which was an interesting finding [43]. Coefficient of variance (CV) indicates the reproducibility of the model, for which a value of less than 10% is desirable [46]. According to the results, the CV value is 3.38%, and the model was statistically valid. The predicted residual error sum of squares (PRESS) is cross validation used in data analysis to offer a statistical summary of model performance [47]; the obtained PRESS of the model is 58.71. The adjusted R 2 of 0.9546 in the improved model agreed reasonably well with the expected R 2 of 0.9424. Figure 2 depicts the relationship between the measured diameter of the ZnO particles and the models' anticipated diameter. The linear correlation coefficient indicates reasonable agreement between the experimental and model values across the factor space. Three variables of the spinning-calcination temperature (X 1 ), tip-to-collector distance (X 2 ), and applied potential (X 3 )-were selected for the optimization process to form ZnO particles. The interaction effect of the chosen parameters on the response observed in the experimental runs can be explained using the analysis of variance (ANOVA). Furthermore, the adequacy of the model was examined using diagnostic diagrams, and the model should be validated by evaluating the optimum experimental conditions, as previously explained [45]. The optimization results show the parameter interaction effect, which consists of 15 experiments, as is presented in Table 2. For each experiment, the ZnO diameter (the response) was measured; a two-set average is noted in Table 3, column Y. When one decides if the overall results are significant, the F statistic must be used in combination with the p-value. The p-value indicates the degree to which the data is consistent with the null hypothesis. The successive p-value of <0.0001 and F value of 99.22 indicate significant model terms. In this scenario, there were key model terms: A, B, C, A 2 , B 2 , and C 2 . Only the quadratic terms for the two electrospinning variables made it into the refined model, which was an interesting finding [43]. Coefficient of variance (CV) indicates the reproducibility of the model, for which a value of less than 10% is desirable [46]. According to the results, the CV value is 3.38%, and the model was statistically valid. The predicted residual error sum of squares (PRESS) is cross validation used in data analysis to offer a statistical summary of model performance [47]; the obtained PRESS of the model is 58.71. The adjusted R 2 of 0.9546 in the improved model agreed reasonably well with the expected R 2 of 0.9424. Figure 2 depicts the relationship between the measured diameter of the ZnO particles and the models' anticipated diameter. The linear correlation coefficient indicates reasonable agreement between the experimental and model values across the factor space.

Response Surface Plots
The ZnO nanoparticles' diameter and the response variables are shown in three-dimensional surface plots in Figure 3 versus two factors at a time (with the third variable Nanomaterials 2022, 12, 1733 7 of 13 kept constant at the center value of (0)). As shown in Figure 3b, at 16 kV applied potential and 12 cm distance (feed rate of 1 mL/h constant), nanoparticles with the smallest diameter were formed. However, when the voltage and distance were reduced to the absolute minimum of 12 kV and 8 cm, the largest average diameter fibers (193 nm) were generated. These plots revealed a few generalized conclusions about electrospun ZnO nanoparticles: (a) the particle diameter changed in an inverse relationship to the applied voltage, (b) increasing the distance of separation led to a decrease in fiber diameter, and (c) fiber diameter was smaller for the minimal value of calcination temperature of 600 • C.

Response Surface Plots
The ZnO nanoparticles' diameter and the response variables are shown three-dimensional surface plots in Figure 3 versus two factors at a time (with the th variable kept constant at the center value of (0)). As shown in Figure 3b, at 16 kV appli potential and 12 cm distance (feed rate of 1 mL/h constant), nanoparticles with t smallest diameter were formed. However, when the voltage and distance were reduc to the absolute minimum of 12 kV and 8 cm, the largest average diameter fibers (193 n were generated. These plots revealed a few generalized conclusions about electrosp ZnO nanoparticles: (a) the particle diameter changed in an inverse relationship to t applied voltage, (b) increasing the distance of separation led to a decrease in fiber dia eter, and (c) fiber diameter was smaller for the minimal value of calcination temperatu of 600 °C. The Box-Behnken (BB) design [48] is a part of the standard response surface a proach and can be used to establish, analyze, and identify the quantitative correlati among electrospinning variables and mean particle diameter. When compared to oth symmetrical second-order experimental designs such as Doehlert, central composi and three-level factorial designs in terms of efficiency and characteristics, the use of design is common in industrial research due to the economic advantages that requ only three levels for each factor, with the settings of 1, 0, +1 [49]. In this study, the re tionship between the following variables was investigated using the BB desi three-level, three-factor model. In combination with RSM, this design is frequently us to optimize a variety of physical, chemical, and biological processes [50,51]. Fiber dia eter accord of different polymeric materials has been studied using various electrosp ning parameters [52]. Numerous studies on electrospinning support these findings fiber diameter variation in response to parametric manipulation. Changing the distan between the tip and the collector which has a tremendous effect on flight time and fie strength significantly impacts the fiber morphology. Separation distance and fiber ameter are opposite to each other, and in many cases, the droplet formation was due The Box-Behnken (BB) design [48] is a part of the standard response surface approach and can be used to establish, analyze, and identify the quantitative correlation among electrospinning variables and mean particle diameter. When compared to other symmetrical second-order experimental designs such as Doehlert, central composite, and three-level factorial designs in terms of efficiency and characteristics, the use of BB design is common in industrial research due to the economic advantages that require only three levels for each factor, with the settings of 1, 0, +1 [49]. In this study, the relationship between the following variables was investigated using the BB design three-level, three-factor model. In combination with RSM, this design is frequently used to optimize a variety of physical, chemical, and biological processes [50,51]. Fiber diameter accord of different polymeric materials has been studied using various electrospinning parameters [52]. Numerous studies on electrospinning support these findings of fiber diameter variation in response to parametric manipulation. Changing the distance between the tip and the collector which has a tremendous effect on flight time and field strength significantly impacts the fiber morphology. Separation distance and fiber diameter are opposite to each other, and in many cases, the droplet formation was due to the inability of the jet to maintain sufficient distance between the tip and the collector as a result of the increased field strength [53]. To stretch the solution before it deposits on the collector, one can increase the separation distance, leading to a decreased fiber diameter [54]. An increase in fiber diameter is associated with decreased field strength when separated by a greater distance [55]. However, fibers are not deposited if the separation distance is too large [54]. Because of this, it appears that voltage and the resulting electric field have a significant impact on the jets' stretching, acceleration, and diameter. A higher voltage causes the solution to be stretched out more in the jet due to stronger columbic forces, which reduces the fiber size [56,57]. It is known that the formation of beads cannot be completely avoided at higher voltages. However, here, no beads were observed in the selected range of applied potential, and it was excellent for the fabrication of nanosized fibers. Below 12 and above 16 kV, the possibility of the formation of beads is raised. The density of the beads increases with the rise of voltage, and the beads unite to form a thicker diameter fiber [50,51,58].

Response Surface Plotting and Characterization of ZnO Nanoparticles at Optimized Conditions
As demonstrated in Figure 3, response surface plots illustrate the effect of interaction between sintering temperature, the distance between nozzle and collector, and the applied voltage on the size of ZnO nanoparticles. The nanoparticle size became bigger as the calcination temperature increased from 600 to 800 • C. The fundamental reason for this is that an increase in temperature influences the growth of crystal. The particle size grew rapidly as the temperature rose from 600 to 800 • C. These findings could imply that the increase in size was mainly due to crystal development, as evidenced by multiple previous investigations [59,60]. When these two parameters (voltage and calcination temperatures) are combined, they have a more significant impact on particle size, comparable to the effects seen in Figure 3a,b. At a constant calcination temperature of 600 • C, the particle size of ZnO changes dramatically as the voltage is reduced from 16 to 12 kV. The electrospinning process and calcination temperature highly influenced the diameter of ZnO nanoparticles. Previous optimization by employing response surface methodology was used to form smooth and homogeneous nanofiber architectures [59]. SEM images shown in Figure 1 represent fibers before annealing, whilst Figure 4 shows those after calcination. Figure 5 provides EDS mapping of the obtained ZnO nanoparticles. By optimizing the electrospinning parameters, uniform and bead-free fibers were obtained. The average diameter of electrospun and sintered ZnO nanoparticles was determined using the Image J program. The experimental design parameters were evaluated by varying processing settings with a fixed collector distance of 12 cm and applied potential in the range from 12 kV to 16 kV. A significant decrease in diameter is due to the fact that a higher electrical voltage and distance contribute to a greater stretching of the polymer, resulting in a decrease in the diameter of the electrospun fiber and particles. A constant flow rate of 1 L/min was maintained throughout the experiment. As shown in Figure 4a-c, the formation of well-shaped ZnO particles can be observed after annealing of the obtained fibers at various temperatures from 600 • C to 800 • C. a result of the increased field strength [53]. To stretch the solution before it deposit the collector, one can increase the separation distance, leading to a decreased fiber ameter [54]. An increase in fiber diameter is associated with decreased field stren when separated by a greater distance [55]. However, fibers are not deposited if the aration distance is too large [54]. Because of this, it appears that voltage and the resul electric field have a significant impact on the jets' stretching, acceleration, and diam A higher voltage causes the solution to be stretched out more in the jet due to stron columbic forces, which reduces the fiber size [56,57]. It is known that the formatio beads cannot be completely avoided at higher voltages. However, here, no beads w observed in the selected range of applied potential, and it was excellent for the fabr tion of nanosized fibers. Below 12 and above 16 kV, the possibility of the formatio beads is raised. The density of the beads increases with the rise of voltage, and the be unite to form a thicker diameter fiber [50,51,58].

Response Surface Plotting and Characterization of ZnO Nanoparticles at Optimized Conditio
As demonstrated in Figure 3, response surface plots illustrate the effect of inte tion between sintering temperature, the distance between nozzle and collector, and applied voltage on the size of ZnO nanoparticles. The nanoparticle size became bigge the calcination temperature increased from 600 to 800 °C. The fundamental reason this is that an increase in temperature influences the growth of crystal. The particle grew rapidly as the temperature rose from 600 to 800 °C. These findings could im that the increase in size was mainly due to crystal development, as evidenced by m ple previous investigations [59,60]. When these two parameters (voltage and calcina temperatures) are combined, they have a more significant impact on particle size, c parable to the effects seen in Figure 3a,b. At a constant calcination temperature of 600 the particle size of ZnO changes dramatically as the voltage is reduced from 16 to 12 The electrospinning process and calcination temperature highly influenced the diam of ZnO nanoparticles. Previous optimization by employing response surface metho ogy was used to form smooth and homogeneous nanofiber architectures [59]. SEM ages shown in Figure 1 represent fibers before annealing, whilst Figure 4 shows th after calcination. Figure 5 provides EDS mapping of the obtained ZnO nanoparticles optimizing the electrospinning parameters, uniform and bead-free fibers were obtai The average diameter of electrospun and sintered ZnO nanoparticles was determi using the Image J program. The experimental design parameters were evaluated by ying processing settings with a fixed collector distance of 12 cm and applied potenti the range from 12 kV to 16 kV. A significant decrease in diameter is due to the fact th higher electrical voltage and distance contribute to a greater stretching of the polym resulting in a decrease in the diameter of the electrospun fiber and particles. A cons flow rate of 1 L/min was maintained throughout the experiment. As shown in Figure  c, the formation of well-shaped ZnO particles can be observed after annealing of the tained fibers at various temperatures from 600 °C to 800 °C.   We performed an analysis of SEM/EDS micrographs of all electrospun ZnO nanoparticles shown in Figure 5. From the EDS report, the weight percentage and atomicity of Zn and O were 88.2 and 11.8, respectively, which is close to the bulk ZnO weight percentage. Figure 6 displays the UV-Vis absorption spectra of ZnO nanoparticles produced at optimized conditions using the electrospinning technique; absorbance was measured in the range of 300-600 nm wavelengths. With the decrease of ZnO particle size, the energy gap widens due to their quantum size effect, caused by the confinement of electrons within particles of dimensions smaller than the bulk counterpart [61]. The energy band gap of a semiconductor becomes more pronounced when the size of the nano-crystallites is smaller than the Bohr radius of the bulk excitation [62]. In nanoscale materials, columbic interactions between holes and electrons are critical and charge carriers can be quantum bound, which changes the semiconductor's valence and conduction bands [63]. We performed an analysis of SEM/EDS micrographs of all electrospun ZnO nanoparticles shown in Figure 5. From the EDS report, the weight percentage and atomicity of Zn and O were 88.2 and 11.8, respectively, which is close to the bulk ZnO weight percentage. Figure 6 displays the UV-Vis absorption spectra of ZnO nanoparticles produced at optimized conditions using the electrospinning technique; absorbance was measured in the range of 300-600 nm wavelengths. With the decrease of ZnO particle size, the energy gap widens due to their quantum size effect, caused by the confinement of electrons within particles of dimensions smaller than the bulk counterpart [61]. The energy band gap of a semiconductor becomes more pronounced when the size of the nano-crystallites is smaller than the Bohr radius of the bulk excitation [62]. In nanoscale materials, columbic interactions between holes and electrons are critical and charge carriers can be quantum bound, which changes the semiconductor's valence and conduction bands [63]. We performed an analysis of SEM/EDS micrographs of all electrospun ZnO nanoparticles shown in Figure 5. From the EDS report, the weight percentage and atomicity of Zn and O were 88.2 and 11.8, respectively, which is close to the bulk ZnO weight percentage. Figure 6 displays the UV-Vis absorption spectra of ZnO nanoparticles produced at optimized conditions using the electrospinning technique; absorbance was measured in the range of 300-600 nm wavelengths. With the decrease of ZnO particle size, the energy gap widens due to their quantum size effect, caused by the confinement of electrons within particles of dimensions smaller than the bulk counterpart [61]. The energy band gap of a semiconductor becomes more pronounced when the size of the nano-crystallites is smaller than the Bohr radius of the bulk excitation [62]. In nanoscale materials, columbic interactions between holes and electrons are critical and charge carriers can be quantum bound, which changes the semiconductor's valence and conduction bands [63]. The UV-Vis absorption spectra of synthesized ZnO nanoparticles showed an absorption peak at 320 nm. The Tauc plot of the samples showed that the band gap of the ZnO nanoparticles prepared at a calcination temperature of 600 • C was 3.72 eV, which is comparable to the previously reported value [64]. Equation (2) was used to calculate the direct band gap energy: Characterization of ZnO nanoparticles using UV-Vis spectrum analysis is usually used to study the size and shape of the particles [65]. The rate and width of the surface plasma on the absorbent depend on the size and shape of the nanoparticles. It is worth noting that the obtained E g value was different from the band gap of bulk ZnO (3.37 eV). Bang gap can be attributed to the optical confinement effect, corresponding to the size and length of nanoparticles [66].

XRD Patterns and Transmission Electron Microscopy (TEM)
TEM examination was conducted to ascertain the size of the nanoparticles (given in Figure 7a). The ZnO nanoparticles fabricated under optimized conditions reveal that the particles are hexagonal with slight variation in thickness, which corroborates with the SEM results. The majority of ZnO nanoparticles are sphere-shaped and have an average particle size of 43 nm. The XRD patterns of ZnO nanoparticles fabricated by electrospinning at calcination temperatures of 600, 700, and 800 • C are shown in Figure 7b. All of the peaks in the XRD patterns correspond to the structure of ZnO wurtzite. The intensity of the peaks and the average crystallite size increased with the increasing calcination temperature, indicating the formation of ZnO nanoparticles with a larger size and an increase in crystallinity [67]. The diffraction peaks were assigned to the ZnO (100), (002),  . UV-Visible spectra and energy band gap of ZnO nanoparticles synthesized at a constant applied potential of 16 kV, a distance between collector and nozzle of 12 cm, a flow rate of 1 mL/h, and a calcination temperature of 600 °C.
The UV-Vis absorption spectra of synthesized ZnO nanoparticles showed an absorption peak at 320 nm. The Tauc plot of the samples showed that the band gap of the ZnO nanoparticles prepared at a calcination temperature of 600 °C was 3.72 eV, which is comparable to the previously reported value [64]. Equation (2) was used to calculate the direct band gap energy: Characterization of ZnO nanoparticles using UV-Vis spectrum analysis is usually used to study the size and shape of the particles [65]. The rate and width of the surface plasma on the absorbent depend on the size and shape of the nanoparticles. It is worth noting that the obtained Eg value was different from the band gap of bulk ZnO (3.37 eV). Bang gap can be attributed to the optical confinement effect, corresponding to the size and length of nanoparticles [66].

XRD Patterns and Transmission Electron Microscopy (TEM)
TEM examination was conducted to ascertain the size of the nanoparticles (given in Figure 7a). The ZnO nanoparticles fabricated under optimized conditions reveal that the particles are hexagonal with slight variation in thickness, which corroborates with the SEM results. The majority of ZnO nanoparticles are sphere-shaped and have an average particle size of 43 nm. The XRD patterns of ZnO nanoparticles fabricated by electrospinning at calcination temperatures of 600, 700, and 800 °C are shown in Figure 7b. All of the peaks in the XRD patterns correspond to the structure of ZnO wurtzite. The intensity of the peaks and the average crystallite size increased with the increasing calcination temperature, indicating the formation of ZnO nanoparticles with a larger size and an increase in crystallinity [67]. The diffraction peaks were assigned to the ZnO (100), (002),

Conclusions
The present work vouches for the successful application of a BB design and RSM method to predict the diameter of electrospun and sintered ZnO nanoparticles. The se-

Conclusions
The present work vouches for the successful application of a BB design and RSM method to predict the diameter of electrospun and sintered ZnO nanoparticles. The selected method and the applied experimental design effectively determined the optimal parameters of the three chosen variables to fabricate ZnO nanoparticles with minimal diameter by the electrospinning technique. The selected variables (applied potential, distance between nozzle and collector, and calcination temperature) had a significant effect on the ZnO nanoparticles' size, where the results of a second-order polynomial regression model