Introduction

Nowadays, effluents emanating from industrial actions are one of the biggest problems in the world [1]. Various process industries, such as textile, pulp, and paper, due to the consumption of thousands of colored chemicals in their production activities are significant sources of dye pollutants in the environment [2]. The global consumption of organic dyes is estimated to be over 700,000 tons per year, and textile consumption is 75% [3]. Water consumption for production in the textile industry is estimated between 100 and 200 L/kg, which result in very high volumes of wastewater containing acids, alkali, toxic compounds as well as colored materials [3, 4]. Colorants are typical of artificial origin with a complex molecular structure derived from coal tar, which contains hydrocarbons such as benzene, naphthalene, anthracene, toluene, and glycine [5].

Reactive dyes are soluble anionic dyes. After the azo dyes group, they are the most widely used colorants in the textile industry [6]. These dyes have been used to replace direct, azo, and Watts dyes because of their desirable properties such as transparency and low energy consumption [7].

The reactive dye, with its characteristic chemical structure, is normally resistant to chemical attack and has a small stabilization effect due to the competition between the formation of its reactive state (vinyl sulfone) and hydrolysis reactions [8]. They can pose a negative impact on the environment due to the reduction in light penetration and impairment of photosynthesis process [9, 10]. Reactive Blue 19 (RB19) also known as Remazol Brilliant blue is an anthraquinone dye used by the textile industries [11]. Therefore, the removal of the dye from wastewaters is of great necessity. Removal of color is possible through various physical, chemical, biological methods or a combination of them [12].

Several treatment methods have been proposed for the removal of RB19 from contaminated waters, which include photodecomposition [13], electro-coagulation [3] adsorption [14], dissolved air flotation [15], biodegradation [16], and other processes. The afore-mentioned methods cause problems such as their control, injection of chemicals and production of high volumes of sludge with attendant problems of filtration and sludge disposal [17]. Advanced oxidation process (AOP) is highly effective in the removal of organic compounds because of its high efficiency and high oxidation potential [18, 19]. The AOP generally refers to a process in which a catalyst such as magnesium oxide, and a strong oxidizing agent such as hydrogen peroxide and ozone are used in the presence or absence of an ultraviolet radiation source [18, 20]. This process results in the production of free and active radicals (OH). The hydroxyl ion is very significant due to its high oxidation power and the reduction in environmental pollution caused by organic matter [21, 22]. This method is applicable to the degradation of resistant compounds in comparison to other methods and, in view of its high oxidizing power; it converts them into mineral forms under special conditions [23]. In addition, recent advances on the removal of dyes using Fe-based metallic glasses/AOPs are also attractive [24,25,26]. Among different oxidation processes, hydrogen peroxide (H2O2)/nanoparticles oxidation process is a novel technology in which toxic materials are degraded to harmless compounds [17]. H2O2 is a common oxidizing agent that generates ·OH radicals and oxidizes organic and inorganic materials with high oxidizing capacity [27]. Magnesium oxide nanoparticles (MgO NPs) were used as a catalyst in the purification of dangerous, anti-bacterial and resistive materials [28].

The significant characteristics of MgO NPs include availability, low cost, non-volatility, non-toxicity, stability, high adsorption capacity and high reactivity [29]. For this reason, it is an appropriate option used for remediation processes.

The purpose of this study is to optimize the removal of RB19 dye from aqueous solutions via AOP using MgO NPs. The central composite design (CCD) based on response surface method (RSM) was used to design the experiments for RB19 decolorization. It is an important branch of empirical design and basic tool in developing new processes, optimizing their performance and improving the design and formulation of new products [30, 31]. RSM was also applied to investigate the individual effects of the process variables and their relationship or interactions to maximize the efficiency of RB19 decolorization as well as determine the optimal conditions. The test system used is a discontinuous type and the variables studied include the solution pH, H2O2/MgO NPs molar ratio, concentration of RB19, and time of reaction.

Materials and methods

Materials

NaOH (98%), HCl (37%) and H2O2 (30%) used in the study were of analytical grade (supplied by Sigma-Aldrich Co, Germany). RB19 is an anionic dye with a molecular weight of 626.54 g/mol and maximum absorption (λmax) of 594 nm. The RB19 (C22H16N2Na2O11S3) used in this work was of analytical grade (supplied by Sigma-Aldrich—US). The magnesium oxide nanoparticles (MgO NPs) of 98% percentage purity and 50 nm size were purchased from Sigma-Aldrich (US).

Characterization of MgO NPs

Fourier-transform infrared spectroscopy (FTIR) was done on a JASCO 640 plus machine (4000–400 cm−1) at room temperature to examine the functional groups present in the MgO NPs, which took part in the AOP for RB19 decolorization. Value-stream mapping (VSM) was also performed on the MgO NPs. Vibrating magnetometry (VSM) was analyzed using a Kavir Precise Magnetic instrument (MDKFT, Iran). X-ray diffraction (XRD) was determined using an X-ray diffract meter with XRD SMART Lab to evaluate the crystalline properties of MgO NPs.

Batch experiments

Magnesium oxide nanoparticles (MgO NPs) were used as the catalyst while hydrogen peroxide (H2O2) was used as the oxidizer for the process. The effect of the oxidation process parameters, pH (3–7), molar ratio of H2O2/MgO NPs (1–3), initial concentration of RB19 (20–80 mg/L), and contact time (30–90 min) on the process were investigated. All stock solutions of RB19 were prepared with double-distilled water. Batch experiments were conducted in 250 mL Erlenmeyer flasks to study the removal of RB19 by AOP. First, the RB19 stock solution was prepared and then the samples with different concentrations (20, 40, 60 and 80 mg/L) were prepared from the stock. The pH of the solution was adjusted by adding 0.1 N HCl or NaOH solutions. The pH was measured using an MIT65 pH meter. A specific amount of catalyst (MgO NPs) and oxidizer (H2O2) were added to the color solution with a specific initial RB19 concentration and placed on the shaker and blended. All solutions underwent constant mixing at the 180 rpm for 2 h. After the desired time of mixing, the mixture was filtered using the Whatman filter paper (size: 40 µm). Then, the filtrate was analyzed for the residual RB19 concentration. The initial and final RB19 concentrations remaining in solutions were analyzed by a UV–visible recording spectrophotometer (Shimadzu Model: CE-1021-UK) at a wavelength of maximum absorbance, λmax = 595 nm. Percentage of decolorization was calculated as follows [32, 33]:

$${\text{Decolorization}}\,(\% ) = \frac{{(C_{0} - C_{\text{f}} )}}{{C_{0} }} \times 100,$$
(1)

where C0 and Cf are the initial and final RB19 dye concentration, respectively.

Design of experiments and statistical analysis

The response surface methodology (RSM) is an effective method for the optimization of processes. The RSM method can be employed on the basis of different designs including CCD, Box–Behnken design (BBD), one-factor design, d-optimal design, etc. [34]. In the study, the CCD was applied to evaluate the impact of the independent variables [pH, H2O2/MgO NPs ratio, time and initial concentration of RB19 on the dependent variable (RB19 decolorization, %)]. The CCD was applied using Design Expert software (Stat-Ease, 7.1 trial version). Four factors at three levels of full factorial CCD based on RSM was used which gave a total of 30 experiments (including 8 axial points, 16 factorial points and 6 replicates at the center points) to optimize the chosen variables. The experimental range of the contributory factors and selected levels of the independent variables used in the study are presented in Table 1.

Table 1 Experimental range and levels of independent variables tested
Full size table

The fitting on the efficiency of eliminating color was implemented using the coefficient of determination (R2) and the adjusted R2. Furthermore, analysis of variance (ANOVA) was applied as a statistical method to analyze the responses and to understand better the effects of variables. The reason for using this statistical model is due to the high expense of nanoparticles, sample shrinking size, and presentation of linear equation. The experimental settings, which were designed using CCD for the assessment of the effect of several factors and optimization of process parameters, are presented in Table 2. The response, Y, can be represented by a mathematical equation that correlates the response surface. The response, Y, can be expressed as a polynomial model based on the following quadratic equation [35, 36]:

$$Y = b_{0} + b_{1} A + b_{2} B + b_{3} C + b_{4} D + b_{11} A^{2} + b_{22} B^{2} + b_{33} C^{2} + b_{44} D^{2} + b_{12} AB + b_{13} AC + b_{14} AD + b_{23} BC + b_{24} BD + b_{34} CD$$
(2)

where Y represents the predicted response (RB19 percentage decolorization); A, B, C and D are the coded values of the independent process factors: initial pH of solution, initial RB19 concentration, reaction time and H2O2/MgO NPs ratio, respectively. b0, b1, b2, b3, b4, b11, b12, b13, b14, b22, b23, b24, b33, b34, and b44 are the constant regression coefficients.

Table 2 Experimental design matrix with the experimental and predicted values for RB19 discoloration efficiency (%) using AOP
Full size table

ANOVA was used to evaluate the significance of the quadratic regression model. Also, the model terms were assessed using the p value with 95% confidence level. The coefficient parameters were assessed by response surface regression analysis using the software Design Expert (version 7). It was also applied to obtain the residuals, 3-D surface and 2-D contour plots of the response models.

Results and discussion

Magnetization characterization of catalyst

Value-stream mapping is a lean management technique for examining the present state and designing an impending state for the series of events that take a product or service from its beginning through to the customer. Magnetization measurements of the catalyst, magnesium oxide nanoparticles (MgO NPs) were done using the VSM technique at fields of − 8000 to 8000 emu/g (before and after calcination, respectively) operated at room temperature (see Fig. 1). It was observed that MgO NPs produced a higher value of saturation magnetization (0.02 emu/g) with the smallest particle size.

Fig. 1
figure 1

VSM hysteresis loop of magnetic MgO NPs

Full size image

Fourier-transform infrared spectroscopy (FTIR) spectroscopy

Figure 2 shows the FTIR spectrum (%transmittance versus wave number) for MgO NPs. The functional groups present in MgO NPs (which took part in RB19 decolorization) were identified. The peak of 1615.92 cm−1 [C–C stretching (in-ring) is assigned to aromatics, 1384.58 cm−1 (N=O bending] shows the presence of nitro compounds, 1087.17 cm−1 (C–N stretching) is assigned to aliphatic amines, 3700.56 cm−1 (C–N stretch) indicates aliphatic amines. Peaks 3419.46 cm−1 (O–H stretch, H–bonded) assigned to alcohols and phenols are very strong and broadband which also took part actively in the decolorization of RB19.

Fig. 2
figure 2

FTIR spectrum of the MgO NPs

Full size image

X-ray diffraction study

X-ray diffraction was done to determine the diffractive pattern of MgO NPs. Figure 3 shows the XRD pattern of MgO NPs. Broad peaks were observed on the XRD image which indicates the presence of crystalline constituents in the NPs. The XRD image showed that the maximum peak is about 43° with extremely high intensity.

Fig. 3
figure 3

XRD patterns of MgO NPs

Full size image

Model fitting and statistical analysis of RB19 decolorization using RSM

The first step in this study was to identify the variables that influenced the decolorization process and their range. Thus, initially, the capacity of H2O2/MgO NPs ratio was tested in the decolorization of RB19 at pH 3, 5 and 7 while keeping other process variables constant. It was found that the H2O2/MgO NPs ratio was only able to decolorize the dye RB19 at pH 3 and the decolorization obtained was 95% at 60 min. Then, the experimental runs were performed according to the operating parametric values given in Table 2. The experimental results were analyzed using RSM to obtain an empirical model for the best response. The actual and the predicted dye decolorization percentages are shown in Table 2. The actual values were found to be close to the predicted values evaluated from the model for a specific run.

In order to develop a statistically model, the significance of the coefficients of regression was determined in relation to the p values. From Table 3 (the ANOVA results), the quadratic model developed from the RSM was statistically significant for RB19% decolorization (Y). A p value less than 0.05 indicate that the effect of a term is significant. The model F value of 9.37 and Prob > F value less than 0.0500 implies that the model is significant for RB19 decolorization. Low value of Prob > F (less than 0.05) indicates the randomness of the result and the significant effect of the model terms to the response (Table 3). The “Lack of Fit F value” of 625,049.41 implies that the Lack of Fit is significant. There is only a 0.01% chance that a “Lack of Fit F value” this large could occur due to noise. Adequate precision measures the signal to noise ratio. A ratio larger than 4 is desirable. Adequate precision ratio of 12.349 shows an adequate signal. This model can be used to navigate the design space. The quadratic model was used to explain the mathematical relationship between the independent variables and the dependent response. The mathematical expression for the relationship between RB19 decolorization with the independent variables: pH, time, concentration, and H2O2/MgO NPs molar ratio is shown in terms of coded factors in Eq. (3):

$$Y = 79.51 - 13.34A + 6.40B - 41.28C + 1.20D - 11.53AD - 13.07BC + 1.81BD - 3.90CD + 5.35A^{2} - 16.09B^{2} - 91.49C^{2} + 8.07D^{2} .$$
(3)
Table 3 ANOVA, lack-of-fit (LOF) test, and regression coefficients and the significance of the response surface quadratic model of AOP decolorization of the dye solution
Full size table

“Prob > F” less than 0.0500 indicates that the model terms are significant. Values greater than 0.1000 indicate that the model terms are not significant. Statistical analysis showed that the coefficients; C, BC, B2, and C2 were statistically significant. F value shown in Table 3 indicates that time (F value: 9.27) has the greatest effect on the AOP for RB19 decolorization followed by concentration and pH. Also, the interaction between concentration and time showed a high F value of 7.54 than the other interactions. The high value of the coefficient of determination (R2 = 0.912) being a measure of the goodness of fit to the model indicates a high degree of correlation between the predicted response and the experimental responses. The adjusted R2 (0.805) also confirmed the high correlation between the observed values and the theoretical values. From Fig. 4, it was also confirmed that the experimental (actual) values were very close to the predicted values. The coefficient of a factor reflects the effect of that particular factor while the coefficients of two combined factors show the interaction between them, whereas the second-order term indicates quadratic effect. The ANOVA results showed that the quadratic model (Eq. 3) is adequate to predict RB19 dye decolorization at the variables’ studied range. A residual is defined as the change between an observed value, Y and it is fitted, Ŷ [37]. Normal probability plot is used to check the normality distribution of the residuals [38]. Great deviation from normality was not observed in the normal probability plots of the residuals (Fig. 5).

Fig. 4
figure 4

Plot of the predicted values versus the observed values of RB19 decolorization

Full size image
Fig. 5
figure 5

The studentized residuals and normal % of probability residuals for RB19 decolorization

Full size image

Response surface plots

The interactive effects of the process variables on the RB19 decolorization using advanced process dye were studied further by plotting three-dimensional (3D) surface curves and two-dimensional (2D) figures against any two independent variables while keeping the other variables at their central (0) level. The 3D surface and 2D contours plots of the response (percentage decolorization) from the interactions between the variables are shown in Figs. 6, 7, 8, 9, 10 and 11. The pH is a very important parameter that actively affects the AOP. The effect of pH on RB19 decolorization is related to pH of the solution and the functional groups present in the MgO NPs, which affects its surface charge. Figure 6 shows that 90.5% decolorization was achieved at pH of 3.8 and concentration of 53.6 mg/L. 89% efficiency was observed at pH 4.3 and time of 67.8 min (Fig. 6). The percentage of decolorization was increased by decreasing the pH of the solution. However, it was observed that at higher pH, the decolorization of RB19 was reduced rapidly (Figs. 6, 7, 8). The solution pH is one of the parameters affecting the adsorbent level and ionization of pollutants [26, 34]. The impact of pH also depends on the zero point charge of the catalyst and acidity constant (pKa). RB19 is an anionic dye and the value of pHZPC for MgO NPs was 12.4 [39]. Thus, MgO NPs at pH lowest of 12.4 had a positive charge. At pH higher than these values, MgO NPs had a negative charge. But at pH above 12.4 and close to this value, the RB19 and nanoparticles had a neutral charge and leads to reduced efficiency. The cause of color decomposition in acidic pH is due to the instability of color rings with the presence of MgO NPs at the pH and the presence of H+ on MgO NPs which resulted in better color reduction [30, 40]. Also, the decolorization of RB19 being favorable in the acidic environment is due to the electrostatic attractions between the negatively charged functional groups present on the anionic dye and the positively charged MgO NPs surface. A study that investigated the effect of nanoparticles on the treatment of dairy wastewater resulted in an optimum pH of 4 [41].

Fig. 6
figure 6

3D surface and 2D contour plot of the interactive effect of pH and RB19 concentration on RB19 removal efficiency using advanced H2O2/MgO oxidation process at constant H2O2/MgO NPs ratio and time

Full size image
Fig. 7
figure 7

3D surface and 2D contour plot of the interactive effect of pH and time on RB19 removal efficiency using advanced H2O2/MgO oxidation process at constant H2O2/MgO NPs ratio and RB19 concentration

Full size image
Fig. 8
figure 8

3D surface and 2D contour plot of the interactive effect of pH and H2O2/MgO NPs ratio on RB19 removal efficiency using advanced H2O2/MgO oxidation process at constant RB19 concentration and time

Full size image
Fig. 9
figure 9

3D surface and contour plot of the interactive effect of time and RB19 concentration on RB19 removal efficiency using advanced H2O2/MgO oxidation process at constant pH and H2O2/MgO NPs ratio

Full size image
Fig. 10
figure 10

3D surface and 2D contour plot of the interactive effect of H2O2/MgO NPs ratio and RB19 concentration on RB19 removal efficiency using advanced H2O2/MgO oxidation process at constant pH and time

Full size image
Fig. 11
figure 11

3D surface and 2D contour plot of the interactive effect of H2O2/MgO NPs ratio and time on RB19 removal efficiency using advanced H2O2/MgO oxidation process at constant pH and RB19 concentration

Full size image

The contact time is also important in all experimental processes. Figures 6, 9, and 11 show that higher amount of RB19 was decolorization at increased contact time. This is due to the high availability of the active sites with time. This is also due to an increase in the time leading to the production of radical hydroxyl needed for the reaction and these hydroxyl radicals will have enough time to react with organic compounds [42].

From Fig. 9, it can be seen that the RB19 concentration with time has a negative on the process. The removal efficiency was increased rapidly from contact time of 30–65 min. Previous studies have shown that the efficiency of removal is reduced at high concentrations due to saturation of the catalyst coating surface with the reacting compounds [43]. From the results obtained, in order to increase the efficiency at higher concentrations an increase in MgO NPs and ·OH in the reaction medium is required [44]. A higher rate of decolorization was observed at an H2O2/NPs ratio of 3. Studies have shown that the application of high molar ratios does not increase the efficiency of removal because higher molar ratios imply excess amounts of H2O2 react with ·OH radicals produced in the process and result in the production of weaker radicals that are less active than ·OH. Hydroxyl radical in the reaction zone with H2O2 is combined in the environment and form scavenger radicals (HO ·2 ) that are weaker and contribute to the reduction of the removal efficiency [45, 46].

Optimization of RB19 decolorization

Optimization was done using the Design expert software (Stat-Ease, 7.1 trial version) to determine the optimum conditions for decolorization of RB19. Optimum conditions of pH 3, RB19 concentration of 80 mg/L, time of 60 min and H2O2/MgO NPs ratio of 3 were obtained, and the optimum decolorization efficiency at this optimum condition was predicted to be 93.77%. Experiment was carried out at these optimum conditions to validate the predicted optimum values. An experimental value of 92.93% was obtained, which was found to be in close agreement with the predicted percentage decolorization (93.77%). The desirability of a model close to unity and with low error depicts the applicability of the model [47]. A desirability of 0.973 (Fig. 12) confirms the applicability of the model and the predicted responses. Also, Fig. 13 shows the desirability effect of the individual process variables. Their desirability values are close to unity, which implies that each of the operating parameter satisfies the model. The values furthermore indicate how well each variable satisfies the model.

Fig. 12
figure 12

The desirability effect for RB19 decolorization

Full size image
Fig. 13
figure 13

Desirability effect of the individual operating parameter

Full size image

Conclusions

The advanced H2O2/MgO oxidation process (AOP) removal of RB19 dye from aqueous solution was optimized using CCD in RSM. The effect of oxidation process parameters, pH (3–7), molar H2O2/MgO NPs ratio (1–3), initial concentration of RB19 (20–80 mg/L), and contact time (30–90 min) was investigated. The CCD of RSM was applied to study the interactive effects of the process parameters and the optimum conditions. A quadratic model equation relating the independent variables to the dependent variable was also generated from the design. One-way ANOVA was applied for statistical data analysis. The observed values were found to be very close to the predicted (theoretical) response values. Optimum conditions of pH 3, contact time of 60 min, RB19 concentration of 80 mg/L and H2O2/MgO NPs molar ratio of 3 were obtained, which gave RB19 removal efficiency of 93.77%. The value of the coefficient of determination (R2 = 0.912%) and adjusted R2 (0.805%) showed that the removal of RB19 dye using AOP can be described by the RSM. The ANOVA results showed that the quadratic model developed from the RSM was statistically significant for RB19% decolorization. From the study, it was deduced that the RSM can be used for optimization of the process parameters for RB19 removal from aqueous solution and the advanced H2O2/MgO oxidation process.