Simultaneous bioremediation of cationic copper ions and anionic methyl orange azo dye by brown marine alga Fucus vesiculosus

Textile wastewater contains large quantities of azo dyes mixed with various contaminants especially heavy metal ions. The discharge of effluents containing methyl orange (MO) dye and Cu2+ ions into water is harmful because they have severe toxic effects to humans and the aquatic ecosystem. The dried algal biomass was used as a sustainable, cost-effective and eco-friendly for the treatment of the textile wastewater. Box–Behnken design (BBD) was used to identify the most significant factors for achieving maximum biosorption of Cu2+ and MO from aqueous solutions using marine alga Fucus vesiculosus biomass. The experimental results indicated that 3 g/L of F. vesiculosus biomass was capable of removing 92.76% of copper and 50.27% of MO simultaneously from aqueous solution using MO (60 mg/L), copper (200 mg/L) at pH 7 within 60 min with agitation at 200 rpm. The dry biomass was also investigated using SEM, EDS, and FTIR before and after MO and copper biosorption. FTIR, EDS and SEM analyses revealed obvious changes in the characteristics of the algal biomass as a result of the biosorption process. The dry biomass of F. vesiculosus can eliminate MO and copper ions from aquatic effluents in a feasible and efficient method.


Results and discussion
The copper biosorption process by algae is strongly influenced by multiple factors such as initial pH, incubation time, and concentration of the copper 25 . On the other hand, decolorization of azo dye through biosorption processes is influenced by several factors including pH, algal biomass, incubation time and initial dye concentration 26,27 .

Statistical optimization of simultaneous biosorption of copper and methyl orange by Fucus vesiculosus biomass.
In the current study, the dried biomass of brown alga, Fucus vesiculosus, was used as biosorbent for simultaneous biosorption of MO and copper ions from aqueous solutions. Optimization of biosorption process independent variables was carried out using Box-Benken design to study the individual, quadratic and interaction effects between the different independent process variables and to maximize the removal percentages. Box-Benken design was also used to predict the best biosorption process conditions for maximum simultaneous biosorption of MO and copper ions from aqueous solutions. In the current experiment, the Box-Benken design of 29 experimental trials was used for optimization of the selected variables including MO conc. (X 1 ), algal biomass (X 2 ), initial pH level (X 3 ) and incubation time (X 4 ) on the biosorption of MO and copper ions from aqueous binary solutions. Table 1 presents the levels of coded and actual values of the 4 independent factors. Additionally, experimental and theoretical predicted percentages of methyl orange and copper bioremoval along with the residuals are also presented in Table 1. According to the obtained experimental results of Box-Benken design, the percentages of MO removal are ranged from 0.31 to 50.27% and the copper ions removal percentages are ranged from 51.46 to 92.76%. The low removal percentages of MO (ranged from 0.31 to 50.27%) could be explained by the simultaneous biosorption of copper ions and MO dye by Fucus vesiculosus biomass. Consequently, due to high copper concentration used in this study (200 mg/L), competition between copper ions and MO for binding sites decreases the MO biosorption.
The maximum removal percentages of both MO and copper ions were obtained in the trial no. 12 with percent of 50.27% for MO and 92.76% for copper ions when methyl orange conc. was 60 mg/L, algal biomass as Table 1. Box-Behnken design matrix of four process variables with actual factor levels, coded factor levels, predicted and experimental values of simultaneous biosorption of methyl orange dye and copper (II) ions by using Fucus vesiculosus. Adequate precision level higher than 4 is preferable and implies the model reliability. The present model used for methyl orange removal had a reasonable precision value of 33.67 revealing the model reliability. The statistically analyzed results of methyl orange removal (%) shows that the coefficient of variation % (C.V. = 8.05%) has relatively low value, indicating that the performed experiments are highly precise 29 . The PRESS value is 267.09, the standard deviation value is 2.05 and the model's mean is 25.47 ( Table 2). ANOVA of the quadratic regression model of methyl orange removal (%) indicates high significance of the quadratic model as confirmed by the very low value of probability (P-value˂ 0.0001) and the high F-value of 73.15. The non-significant lack of  www.nature.com/scientificreports/ fit (P-value = 0.5608) indicates that the present results are consistent with the model. In this study, the variables displaying P-values below 0.05 were considered to have significant impacts 30 . The interpretation of the correlation between the examined variables ( Table 2) relied on the signals of variable coefficients and P-values. Basically, the correlation between the two factors could be negative or positive. Consequently, the positive coefficient sign reveals a synergistic influence, while the negative coefficient sign reveals an antagonistic impact. It's clear from the coefficients values (Table 2) that the initial pH level (X 3 ) and incubation time (X 4 ) had positive effects on methyl orange removal (%). Whereas the negative coefficients values of both methyl orange conc. (X 1 ) and the algal biomass (X 2 ) means that they exert a negative effects on methyl orange removal (%) from aqueous solutions by Fucus vesiculosus biomass in the tested range of the examined variables. It was obvious from the P-values that the linear coefficients of X 1 , X 2 and X 4 , the interaction between X 1 X 2 , X 1 X 3 , X 1 X 4 , X 2 X 3 , X 2 X 4 , X 3 X 4 and quadratic impacts of X 3 and X 4 had significant effects. Furthermore, linear coefficients of X 3 the quadratic effects of X 1 , X 2 (P-value equals 0.4622, 0.1249; respectively) had non-significant effects on the methyl orange dye removal by F. vesiculosus. Table 3 display the fit summary for Box-Behnken design of methyl orange removal by F. vesiculosus. The fit summary applied to select the appropriate model for the experimental results (linear, 2FI or quadratic model). The appropriate model is chosen on the basis of significant model terms and non-significant lack of fit tests. The fit summary results demonstrated that appropriate model for methyl orange removal by F. vesiculosus biomass is the two factors interaction (2FI) model which is significant with a very small P-value < 0.0001. Lack of Fit Test for two factors interaction (2FI) (with F-value = 1.44 and P-value = 0.3925) and quadratic models (with F-value = 0.97 and P-value = 0.5608) of methyl orange removal percentages by F. vesiculosus biomass are nonsignificant (Table 3). Furthermore, the model summary statistics for methyl orange removal percentages by F. vesiculosus biomass quadratic model recorded the lower standard deviation of 2.05 and the highest R 2 of 0.9865, adjusted R 2 of 0.9730, but two factors interaction (2FI) model recorded the highest predicted R 2 of 0.9450.
By using the coefficients (Table 2), the 2nd-order polynomial equation describing the correlation between methyl orange removal percentages by F. vesiculosus biomass (Y) regarding MO concentrations, algal biomass, and initial pH level and incubation time as the following: where Y is the predicted value of methyl orange azo dye removal (%) and X 1 -X 4 are the coded values of methyl orange azo dye concentration, the algal biomass concentration, initial pH level, and incubation time; respectively.
Multiple regression analysis and ANOVA for copper removal. The results of Box-Behnken experimental design for removal of copper were analyzed using multiple regression analysis and the results were presented in Tables 4, 5. Table 4 shows the regression model determination coefficient (R 2 ) = 0.9934 which means that the variations in copper removal of 99.34% could be described by the model. In addition, the adjusted coefficient of determination (adj R 2 value) of 0.9867 was relatively high and validated that the model was very significant. On the other hand, the predicted value of the determination coefficient (predicted R 2 value) of 0.9772 is in an excellent agreement with the adjusted R 2 values (0.9867), which revealed a well-fit between the predicted and observed values of copper removal percentages. The model used for this experiment is therefore ideal for predicting the removal percentage of copper at the tested levels of independent parameters. The adequate precision value for the current model is 46.27, revealing the model reliability. The mean value is 69.99, PRESS value is 75.40 and the standard deviation is 1.25 (Table 4). Meanwhile, the % of the coefficient of variation (C.V. = 1.79%) is low, indicating that the performed experiment have a high level of reliability and precision 29 .
ANOVA for the regression model of copper removal (%) indicates that the model is highly significant as is apparent from a very small probability value [P-value ˂ 0.0001] with the Fisher's F test (F-value = 149.49) ( Table 4). The P-values were used to assess the significance of each coefficient. The P-values showed that linear effects of X 1 , X 2 , X 3 , and X 4 with probability values of < 0.0001 are significant. Also, the interactions effects between X 1 X 2 , X 1 X 4 , X 2 X 3 , X 2 X 4 and X 3 X 4 with probability values of < 0.0001 are also significant (Table 4). While, the interaction between X 1 X 3 (methyl orange conc. and initial pH level) had a non-significant effect (P-value˃0.05). Additionally, the probability value implied that the quadratic impact of X 2 , X 4 had a significant effects on the copper removal (P-value = 0.0250, < 0.0001 respectively), meanwhile X 1 , X 3 had an insignificant effects on the copper removal using F. vesiculosus.
The negative coefficient value reflects an antagonistic relationship between the variables and the percent removal value, whereas positive coefficient value reflects a synergism between the variables and the percent removal value. Accordingly, the negative values of coefficients means that copper removal % by the biomass of F. vesiculosus is negatively affected by the effect of linear or mutual interactions between two parameters, as well as the quadratic effects. Whereas, the positive values of coefficients means that copper removal percentages by the biomass of F. vesiculosus are increased in the evaluated levels of the selected process parameters as affected by linear effects, mutual interactions effects or quadratic effects. It can be seen from the values of coefficients ( Table 4) that X 1 , X 3 and X 4 had positive effects on copper removal % by the biomass of F. vesiculosus. However, X 2 exerted negative effect on the copper removal % by the biomass of F. vesiculosus. Table 5 indicated the results of the Fit summary of Box-Behnken experimental design of biosorption of copper by using brown alga F. vesiculosus. The fit summary results demonstrated that the quadratic model is the appropriate model for fitting copper removal by F. vesiculosus biomass with a very small P-value < 0.0004 and non-significant lack of fit (P-value = 0.9351 & F-value = 0.32). Furthermore, the model summary statistics of the quadratic model recorded the largest adjusted R 2 (0.9867), predicted R 2 (0.9772) and the lowest standard deviation (1.25).
(1) www.nature.com/scientificreports/ By using the coefficients (Table 4), the 2nd-order polynomial equation describing the correlation between copper removal percentages by F. vesiculosus biomass (Y) regarding MO concentrations, algal biomass, and initial pH level and incubation time as the following: where Y is the predicted value of copper removal (%) and X 1 -X 4 are the coded values of methyl orange azo dye concentration, the algal biomass concentration, initial pH level, and incubation time; respectively. Three dimensional (3D) surface plots for MO dye and copper removal percentages. 3D surface plots were created for determination of the optimum conditions of the bioprocess (removal of MO dye and  www.nature.com/scientificreports/ copper ions from binary aqueous solution) and to describe the relationship between the methyl orange dye and copper removal percentages by F. vesiculosus biomass and the interactions between the chosen process variables. The 3D response surface plots were created for the pairwise of the four variables (X 1 X 2 , X 1 X 3 , X 1 X 4 , X 2 X 3 , X 2 X 4 and X 3 X 4 ). All experiments were conducted using a fixed concentration of copper ions of 200 mg/L. The removal percentages of MO dye were drawn on the Z-axis versus two variables, while the remaining variables maintained fixed at their zero levels (Fig. 1). The 3D (Fig. 1A) shows the effect of MO concentrations (X 1 ) and algal biomass (X 2 ) on MO removal %, while the other two parameters (X 3 and X 4 ) are maintained fixed at their zero levels. The results have shown that the MO removal percentage was relatively high at higher MO concentrations and lower algae biomass. With the increase in the initial concentration of methyl orange dye from 20 to 57.62 mg/L, the removal percentage of methyl orange increases. The increasing percentage of removal could be mainly because of a larger surface area and the accessibility of unsaturated binding sites at the surface of F. vesiculosus biomass required to the biosorption process. Then, the removal percentage of methyl orange decreased with increasing concentration of methyl orange above 57.62 mg/L. This can be due to the saturation of binding sites on the surface of F. vesiculosus biomass. On the other hand, the removal percentage of methyl orange increases with the increase of initial concentration of F. vesiculosus biomass up to 3.2 g/L. The increased removal percentage of methyl orange could be mainly because of a larger surface area and the availability of unsaturated binding sites at the surface of F. vesiculosus biomass required to the biosorption process reaction by increased concentration of the algal biomass. The removal percentage of methyl orange then decreased with increasing the algal biomass concentration from 3.2 to 7 g/L. The agglomeration of the biomass can be a reason for the decrease in the removal efficacy. Figure 1B show the effect of MO concentration (X 1 ) and initial pH level (X 3 ) on the MO removal percentage at center levels of alga biomass (X 2 ) and incubation time (X 4 ). Figure 1B indicates that the highest percentage of MO removal (61.97 mg/L) was at MO 57.7 mg/L and pH 9.5, when alga biomass was 3 g/L and incubation time was 60 min. The increase in the initial pH resulted in the highest percentage of orange methyl removal. This could be interpreted by the enhanced access of methyl orange to the active sites of the algal biomass at alkaline pH. Figure 1C show the effect of MO concentration (X 1 ) and incubation time (X 4 ) on the percentage of MO removal while alga biomass and pH were kept at their center points. The removal percentages of methyl orange increases with the increase of both initial concentration of methyl orange dye and incubation time. In addition, the effects of brown alga F. vesiculosus biomass (X 2 ), initial pH level (X 3 ) and incubation time (X 4 ) on the MO removal percentage are also presented in Fig. 1D-F. Similarly, the three-dimensional plots (Fig. 2) represent the effects of MO concentration, brown alga F. vesiculosus biomass, initial pH level and incubation time on copper removal percentage. Figure 2A-C indicated that the higher level of MO concentration increases the percentage of copper removal by brown alga F. vesiculosus biomass in aqueous solution. The 3D plots of Fig. 2D,E indicated that lower levels of brown alga F. vesiculosus biomass causes higher removal of copper. In addition, the effects of initial pH level (X 3 ) and incubation time (X 4 ) on copper removal percentage are also presented in Fig. 2F.
Effect of initial pH value on the biosorption process. pH is a significant factor that influences the degree of ionization and the characterizations of biosorbent surface 31 . As the pH increases, the pH value leads to an increase in biosorption capacity of heavy metals ions; this may be caused by a decline of the competitiveness between positively charged metal ions and H +32 . As the pH increases, OH − and anionic dyes competing with each other and dyes uptaking were lower, the optimal pH was 6 for removal of Ni 2+ and Zn 2+ and pH was 4 for removal methyl orange 33 . In acidic conditions the electrostatic attractions take place between anionic dye (negatively charged) and biosorption sites of the adsorbent (with a positive charge), causing an increase in the biosorption capacity 34 . The current study proved that the optimum pH of Cu 2+ and methyl orange simultaneous removal was 7.
The effect of pH on MO and Cu removal can be described as an electrostatic interaction mechanism between adsorbents and alga surface; at lower pH, a competition between anionic dyes and protons found in the active sites of the biosorbent, thus the biosorption of the anion dyes is not optimal 35 . Increasing the solution pH reveals increasing the biosorption of anionic ions because the elevated of electrostatic interactions between anionic dyes and adsorbent 36 . The enhanced biosorption at higher pH values could be due to the negatively charged OHfunctional groups that are present on the surface of the biosorbent. This results in electrostatic attraction between the cationic dyes and negatively charged biosorbent surface 37 .
Effect of biomass concentration on the biosorption process. Algae biomasses are considered to be effective sustainable biosorbents for large-scale uses for metals and hazardous dyes biosorption due to their cell wall constitutes high metal removal efficiency, renewable and cost-effective 38 . The metal-binding ability of each alga was different. This can be clarified by the variation in polysaccharides and proteins' cell wall composition that provides cell surface binding sites. Schiewer and Volesky 39 claimed that the high biosorption capacity of algal surface could be attributed to the existence of polysaccharides, lipid or proteins molecules in their cell walls that containing functional groups which can act as binding sites for metals. The biosorption capacity of the algal surface is attributed to the availability of the functional groups, for example carboxylic, hydroxyl, phosphate, imidazole, sulfate, sulphuryl, phosphoryl, amino, etc 40 .
In the present study, increase in the removal percentages of Cu 2+ and methyl orange with increasing the biomass concentration could be attributed to the increase in the surface area of brown marine alga Fucus vesiculosus biomass and the availability of more active sites. Phugare et al. 41 stated that an improvement in the biosorption percentage with increasing biomass concentration is anticipated due to the increased biosorbent surface area which in turn increases the number of active sites leading to efficient biosorption. On the other hand, a reduction www.nature.com/scientificreports/ in the removal percentages at higher biomass concentrations of brown marine alga, Fucus vesiculosus, could be attributed to the agglomeration of the biomass. Karthikeyan et al. 42 stated that a reduction in the efficiency of the removal at higher concentrations of the algal biomass could be caused by a reduction in the efficiency of the biomass surface area as a result of agglomeration. Furthermore, Lata et al. 43 documented a reduction in the biosorption potential at greater algal concentrations was due to agglomerations of the biomass, which could in turn reduce the intercellular spacing which reduced the overall effective biosorption surface area, thus reducing the number of active binding sites available on the algal biomass surface. However, EL Hassouni et al. 44 concluded that the decrease in the efficiency of the biosorption process after achieving the optimal dose could be attributed to an increase in the number of unsaturated active sites on the biosorbent surface with an increase in the concentration of biomass because of the ineffective utilization of the active sites where metal ions or adsorbate particles are insufficient to bind to all available active sites. Kumar et al. 45 reported that 0.1 g of Ulva fasciata is adequate to eliminate 95% of copper from aqueous solution. The biosorption of copper was maximum by using 2 g of Callithamnion corymbosum sp. 46 . Abdulkareem & Alwared 47 reported that the increase of alginate beads derived from marine algae above 10 g/L resulted in a decrease of the biosorption processes which could be attributed to an increase in the number of unsaturated active sites on the biosorbent surface. The effective concentration of brown macro-marine algae Gelidiella acerosa as biosorbent was 0.41 g/L to remove 96.36% of copper from aqueous solution 48 . The highest removal value of copper ions (88.45%) was obtained when 4 g/L of marine brown alga Sargassum bevanom was applied as biosorbent when pH was adjusted to 6, and incubation time of 100 min 49 . The removal of copper is performed effectively by some marine algae such as Fucus vesiculosus by using 1.85 mmoL/g 21 .
The greatest removal percentage of methyl orange was 97% when using 0.4 g/L Oedogonium subplagiostomum AP1 biomass as biosorbent 50 . Five mg of the activated carbon of marine alga Gracilaria corticata has the ability for decolourization of textile dye 51 . The maximum dye decolorisation percentage (86.1%) by calcium alginate extracted from Sargassum sp. was obtained by using 40 mg/L alginate 52 .

Effect of contact time on the biosorption process.
In the present study, the simultaneous removal of cationic copper ions and anionic methyl orange azo dye by brown marine alga Fucus vesiculosus biomass from binary solution depends on the contact time. Experimental results have shown obviously that the removal percentage of cationic copper ions and anionic methyl orange azo dye increases as the contact time increase up to the optimum, which probably due to the availability of a large number of surface vacant active sites on the brown marine alga Fucus vesiculosus biomass surface and also cationic copper ions and anionic methyl orange azo dye concentrations were high. However, at higher contact time, the active sites were saturated causing no further adsorption occurs. Saturation of all active sites on the biomass surface results in a state of equilibrium 53 .
Biosorption capacity of heavy metals by brown algae increased with increasing contact times, within 60 min, the absorption of nickel and cadmium reached 95% 54 . Biosorption of Fe 3+ by Sargassum vulgare (brown alga) was elevated with increasing the time up to 50 min 55 . Uptake of methyl blue by Sargassum muticum was fast in the first 5 min and reach equilibrium within 60-90 min 56 . The elimination of dye reached 93% at 45 min. by Sargassum crassifolium 57 .

Desirability function (DF).
The key objective of the experimental design and the desirability function (DF) were used to identify the optimum predicted conditions to maximize the responses 58 . The DF values ranged between zero (undesirable) to one (desirable). The numerical optimization defines the points minimizing the desirability function. For the optimization process, the DF option in the software design expert (Version 7.0.0) was used. Figure 3 shows the optimization plot displays the optimum predicted values and the desirability function for maximum removal percentage of cationic copper ions and anionic methyl orange azo dye by brown marine alga Fucus vesiculosus.
Maximum removal percentage of cationic copper ions as mono-component by brown marine alga Fucus vesiculosus (100.34%) was obtained by using methyl orange azo dye concentration of 55.71 mg/L, the algal biomass concentration of 3.24 g/L, initial pH level of 9.75, and incubation time of 50.93 min. Whereas, the optimal predicted conditions attained for the maximum removal percentage of anionic methyl orange azo dye as monocomponent (77.91%) were methyl orange azo dye concentration of 56.12 mg/L, the algal biomass concentration of 3.26 g/L, initial pH level of 9.89, and incubation time of 87.68 min. On the other hand, the simultaneous removal percentages of cationic copper ions and anionic methyl orange azo dye in multi-component system by brown marine alga Fucus vesiculosus were obtained using methyl orange azo dye concentration of 57.62 mg/L, the algal biomass concentration of 3.20 g/L, initial pH level of 9.23, and incubation time of 88.01 min. The previous predicted conditions for simultaneous removal of copper ions and methyl orange azo dye by brown marine alga Fucus vesiculosus could be resulted in maximum removal percentages of 99.55% for copper and 76.85% for methyl orange azo dye with DF of 1.
In order to verify the removal percentages of cationic copper ions and anionic methyl orange azo dye by brown marine alga Fucus vesiculosus under the optimal predicted conditions, the experiments have been done in triplicate and compared with the predicted values. The experimental results for copper ions and methyl orange azo dye removal percentages were 97.7% and 75.49; respectively. The verification showed that the experimental results and their predicted values of a strong agreement imply that the DF effectively determines the optimal predicted conditions for the simultaneous removal of cationic copper ions and anionic methyl orange azo dye by brown marine alga Fucus vesiculosus. www.nature.com/scientificreports/ FTIR analysis. The FTIR spectrums of brown alga F. vesiculosus biomass were analyzed before and after bio-adsorption of methyl orange and copper (Fig. 4) showing differences due to the interaction of methyl orange and copper ions with active sites (functional groups) that found in cell surface in biomass. The brown algae cell walls consist of cellulose, hemicelluloses, sulphated furans and also some unique polysaccharides (alginates) that have many active groups (hydroxyl, carboxylate, amino and phosphate groups) with negative charges that can interact with cationic dye and connect the ions of heavy metal 59    Scanning electron microscopy. The SEM images demonstrated that the dry brown alga F. vesiculosus biomasses after and before the biosorption of MO and copper ions as shown in Fig. 5A,B. Figure 5B indicated the ability of F. vesiculosus to biosorb MO and copper ions. After biosorption of MO and copper ions, the surface of F. vesiculosus biomass has been more shrinking, irregular, and there are also more glossy spots as a result of accumulation of copper ions on the cell surface 75 . The morphological structure of the algae changed after biosorption of methyl orange by Oedogonium subplagiostomum 50 .  76 . In this experiment, EDS analysis was carried out to find the elements present on the surface of F. vesiculosus biomass and verified the attachment of Cu 2+ to the surface of F. vesiculosus biomass after biosorption process. The EDS-TEM analysis (Fig. 6A) indicated the existence of optical absorption peak corresponding to Cu 2+ before the biosorption process which could be due to the TEM copper grids coated with a carbon foil have been used during the analysis. The EDS spectra (Fig. 6B) reveal the presence of an optical absorption peak corresponding to Cu 2+ after biosorption process attached to F. vesiculosus biomass cell surface. The Cu 2+ weights` were 1.5 and 10.97% before and after the biosorption process; respectively that proves the role and capacity of F. vesiculosus biomass in the biosorption process of Cu 2+ from aqueous solutions. Raize et al. 77 observed that biosorption of the metallic cations to the algal cell wall components was a surface process. Biosorption by algal biomass occurs mainly through cell wall interactions 78 . Brown algae biomass cell walls contain many polymers including high concentrations of polysaccharides (sulfated polysaccharides, alginate) and proteins which involve several functional groups 77 . There are many functional groups on the algal cell wall such as carboxyl, sulphate, hydroxyl, carboxyle and amino groups, which can serve as cell surface binds for contaminants. Biosorption of metals involves several mechanisms that differ qualitatively and quantitatively according to the species used, the origin of the biomass, and its processing procedure. The principal binding mechanisms of the biosorption process by the algal biomass include ion exchange, formation of complex between contaminants cations and the ligands on the algal surface, diffusion interior the cells or surface precipitation, chelation, bioaccumulation within the cells, binding to intracellular components and proteins 19,79 (Fig. 7) and reduction reactions, accompanied by metallic precipitation on the cell wall matrix 77 . Ion-exchange is a vital concept in the biosorption process, since it reflects the fact that most brown algal biomass is either protonated (ion-exchange takes place between various ions and protons at the biomass binding sites) or contains light metal ions such as K + , Na + and Mg 2+ , which are released upon binding of a heavy metal cation with alginate 80 . Saturation of all active adsorption sites on the biomass surface results in a state of equilibrium 53 . The ion exchange capacity of the brown algae is directly related to the unique macromolecular structure of alginate that contains carboxylic groups which is the most abundant acidic functional group present in the alginate polymer polysaccharides 80 . The sulfonic acid  www.nature.com/scientificreports/ of fucoidan is the second most common functional acidic group in brown algae. Sulfonic acid groups usually play a secondary function, except when metal binding occurs at a low pH level. Hydroxyl groups are also found in all polysaccharides but they are less concentrated and charged negatively only at pH > 10. This means they play a secondary role in metal binding at low pH 80 . Jang et al. 81 acknowledged that guluronic acid-rich alginates (Na-alginate gel) display a high metal selectivity for Cu 2+ due to its higher contents of guluronic acid residues.

Materials and methods
Preparation of the biosorbent. Fucus vesiculosus was collected from Jeddah beach, Saudi Arabia in April 2019. The F. vesiculosus biomass was washed carefully by using tap water, then by distilled water to get rid of salts and sand. After cleaning, the alga biomass was subjected to dryness at 50 °C, (up to steady weight). The dried alga was crushed and then sieved using a suitable laboratory sieve with a particle size range of 1-1.2 mm. The alga biomass was then used as biosorbent for simultaneous methyl orange and copper ions removal.  Fig. 8. MO is an anionic azo organic compound that turns red in acidic medium and orange in basic medium.

Preparation of methyl orange and copper solutions.
Optimization of the copper ions and methyl orange removal by batch biosorption using Box-Behnken design. Four factors were chosen to determine the optimum conditions for maximum simultaneous removal of copper ions and MO using the Box-Behnken design 29 . Design-Expert software (version 7) for Windows was used for generating the Box-Behnken design with 3 center points and 29 different experiments for predicting the optimum levels for the significant factors and to achieve the maximum simultaneous removal of copper ions and methyl orange by Fucus vesiculosus biomass. These factors were methyl orange concentrations (X 1 ; 20, 40, 60 mg/L), algal biomass (X 2 ; 3, 5, 7 g/L), initial pH (X 1 ; 4, 7, 10) and incubation time (X 4 ; 30, 60, 90 min). The tested variables were evaluated at 3 coded levels (+ 1 for high level, 0 for middle, and − 1 for low level). The biosorption experiments were carried out by applying batch biosorption experiments in 250 mL Erlenmeyer flasks and the working solution was 100 mL. The experimental studies were made in multi component system (binary solution). All experiments were carried out using a fixed concentration of copper ions of 200 mg/L and stirring at 200 rpm at ambient temperature. The correlations between the selected variables of the biosorption process and the responses (copper ions and methyl orange removal percentages) were determined using the equation of second-degree polynomial as follows: In which Y is the predicted copper ions or methyl orange removal percentages, X i is the coded levels of the selected variables, β i (linear coefficient), β ij (interaction coefficients) , β 0 (regression coefficients) and β ii (quadratic coefficients).
Analytical methods. After the defined time, ten mL of the binary solution for each trial was centrifuged at 6000×g and the supernatants were analyzed by measuring the absorbance changes on a UV/Vis spectrophotometer at a wavelength of λ max that was 467 nm, to determine the final (residual) concentrations (C f ) of methyl orange dye. The efficiency of Fucus vesiculosus biomass for removal of methyl orange from aqueous solutions was calculated quantitatively in percentage by using the following equation: www.nature.com/scientificreports/ where: C f , C i are the final and initial concentrations of MO (mg/L); respectively. Another 10 mL of the binary solution for each trial was centrifuged at 6000×g and the supernatants were analyzed for determination of the residual concentration of Cu 2+ using Atomic absorptions (Buck scientific 2 hydrous system series Atomic Absorption (USA) by air acetylene system) at the Biotechnology Unit, Mansoura University Egypt according to "standard methods for the examination of water and wastewater 23rd edition 2017" 82 . The efficiency of F. vesiculosus biomass to get rid of Cu 2+ from hydrous solutions was detected in percentage utilizing the following equation.
where: C f , C i are the final and initial copper ions concentrations (mg/L); respectively.
All evaluations of both Cu 2+ and MO in the binary solutions were estimated in triplicate.
Statistical analysis. Design Expert version 7 (https ://www.state ase.com/softw are/desig n-exper t/) and STATISTICA version 8 (https ://www.stats oft.de/de/softw are/stati stica ) softwares have been used for the generation of the experimental design, statistical analysis and to draw the three-dimensional surface plots.
Fourier transforms infrared (FTIR) spectroscopy. The dry biomass of F. vesiculosus samples was analyzed using FTIR spectroscopy before and after methyl orange and copper ions removal. The samples of dry biomass were mixed with pellets of potassium bromide and the FTIR spectra were then detected between 400-4000 cm −1 using "Thermo Fisher Nicolete IS10, USA spectrophotometer".
Scanning electron microscopy (SEM). The samples of F. vesiculosus dry biomass were analyzed after and before methyl orange and copper removal using SEM to examine their morphology. The gold-coated dry biomass samples were detected at various magnifications using the accelerating beam voltage of 30 keV.
Electron dispersive spectroscopy (EDS). Energy dispersive spectroscopy analysis (EDS) was performed using scanning electron microscope (JEOL, JEM-2100, Japan). EDS was used to determine the contents of elements of F. vesiculosus biomass after and before the biosorption process.

Conclusions
The current study provides an interesting, harmless and environmentally-friendly approach that uses macrobrown alga Fucus vesiculosus to remove copper and methyl orange dye simultaneously from aqueous solutions. Box-Behnken design was used to optimize the experimental factors for maximum removal of both MO and copper simultaneously from aqueous solutions using marine alga, Fucus vesiculosus, biomass. The maximum removal % was obtained by using 3 g/L of F. vesiculosus biomass, MO (60 mg/L), copper (200 mg/L) at pH 7 and incubation time of 60 min with agitation at 200 rpm. F. vesiculosus dry biomass can be used as an effective and inexpensive biosorbent for the removal of MO and copper ions from wastewater effluents.