Background

The dye effluents are considered to be highly toxic to the aquatic species and are found to affect the symbiotic process by disturbing the natural equilibrium by reducing photosynthetic activity and primary production due to the colorization of the water[1]. Removal of these dyes from effluents in an economic way remains as a major problem for textile industries[2, 3]. The most commonly used methods for color removal are biological and chemical precipitation. However, these processes are effective and economic only in cases where solute concentrations are relatively high[4]. Even though many methods were explored to address this environmental and economic concern, there are advantages and disadvantages in each and every method used for dye removal from wastewaters[5]. Many physicochemical methods have been tested, but adsorption is considered to be superior to other techniques. This is attributed to its low cost, easy availability, simplicity of design, high efficiency, easy operation, biodegradability, and ability to treat dyes in more concentrated forms[6, 7]. The adsorption technique has been proven to be an excellent way to treat effluents, offering advantages over conventional processes, especially from the environmental point of view[2]. Weber and Morris[2] have identified many advantages of adsorption over several other conventional treatment methods. The adsorption of dyes on various types of materials has been studied in detail. Those include the following: activated carbon[8], peat[9], chitin[10], silica[11], hardwood[12], hardwood sawdust[13], bagasse pith[14], fly ash[15, 16], mixture of fly ash and coal[17], chitosan fiber[18], paddy straw[19], rice husk[20], slag[21], chitosan[22], acid-treated spent bleaching earth[23], palm fruit bunch[24], and bone char[25]. Adsorption with activated carbon has been used widely for the removal of dyes from wastewater[26–31]. Activated carbon was being used as a potential adsorbent because of its high efficiency[32]. Commercially available activated carbons are usually derived from natural materials such as wood or coal which are expensive[33, 34]. Increase in the price of activated carbon results in increased treatment cost for the developing countries like India. Hence, no-cost or low-cost, alternate adsorbents with equivalent capabilities of activated carbon are the current thirst in this area of research. Consequently, low-cost activated carbons from agricultural solid wastes are being investigated for a long time.

The conventional method of studying a process by changing one variable at a time and keeping the other variables at a constant level does not depict the combined effect of all the factors involved. Moreover, this method is time-consuming and also requires a large number of experiments to determine optimum levels, which may or may not be reliable. These limitations of the conventional method can be eliminated by simultaneously varying all the affecting parameters by using a statistical experimental design such as the response surface methodology (RSM)[35]. RSM, initially described by Box and Wilson[36], is an experimental approach to identify the optimum conditions for a multivariable system. RSM is a collection of statistical and mathematical techniques useful for developing, improving, and optimizing processes that can be used to evaluate the relative significance of all the factors involved in the process even in the presence of complex interactions. The main objective of RSM is to determine the optimum operational conditions for the system or the process[37]. The application of statistical experimental design techniques in adsorption process development can result in improved product yields, reduced process variability, and reduced development time and overall costs[38].

The main objectives of the present investigation are the utilization of the low-cost, agricultural waste adsorbent, cashew nut shell (CNS), and the determination of the optimum conditions for Congo red (CR) dye removal from an aqueous solution using response surface central composite design methodology. In this study, the central composite design was selected to determine the effect of experimental parameters and their interactions over the removal of CR. The interactions between factors that influence the percentage of CR removal were established. The optimum value of the parameters was also determined for removal of CR from the aqueous solution using RSM.

Methods

Cashew nut shells collected from Pudukkottai District, Tamilnadu, India were used as an adsorbent. This natural waste was thoroughly rinsed with water to remove dust and the soluble material and was allowed to dry at room temperature. The dried natural waste was crushed into powder and sieved through the sieves. CNS particles with a mesh size of 200 to 30 were taken for the adsorption studies. The proximate and ultimate analysis of cashew nut shells are shown in Table 1. The specific surface area and the pore structure of the CNS were determined by using a surface area and pore size analyzer (Autosorb-I, Quantachrome Instruments, Boynton Beach, FL, USA) on nitrogen adsorption at 77 K. The specific surface area was calculated using the BET equation[39]. It was found that the BET surface area, pore volume, average pore diameter, and bulk density of the CNS were 395 m2/g, 0.4732 cm3/g, 5.89 nm, and 0.415 g/cm3, respectively.

Table 1 Properties of the cashew nut shells
Full size table

Congo red (CI = 22120) was supplied by Merck (Mumbai, India). A stock solution of CR dye was prepared (100 mg/L) by dissolving a required amount of dye powder in deionized water. The stock solution was diluted with deionized water to obtain the desired concentration ranging from 20 to 100 mg/L.

The concentration of CR in the experimental solution was determined from the calibration curve prepared by measuring the absorbance of different known concentrations of CR solutions at λmax = 497 nm using a UV–vis spectrophotometer (Shimadzu, Kyoto, Japan). The pH of the solution was measured with a Hanna pH meter using a combined glass electrode (model HI 9025C, Hanna Instruments, Singapore).

Adsorption experiments were conducted as per the design developed with response surface central composite design methodology. The experiments were carried out in 250-mL Erlenmeyer flasks with the working volume of 100 mL of reaction mixture. The initial pH of the solution was adjusted to the desired value by adding 0.1 M NaOH or HCl. The flasks were shaken for the specified time period in a rotary shaker. The flasks were withdrawn from the shaker after the desired time of reaction. The residual dye concentration in the reaction mixture was analyzed by centrifuging (5,000 rpm, R-24 REMI Centrifuge, Mumbai, India) the reaction mixture and then measuring the absorbance of the supernatant at the wavelength that corresponds to the maximum absorbance of the sample. Dye concentration in the reaction mixture was calculated from the calibration curve. The λmax values of the wastewater samples varied by ±10 nm from the λmax values of pure samples. Each determination is repeated three times, and the results given were their average value. The percentage of CR removal was taken as a response (Y) of the experimental design and calculated as follows:

% CR dye removal = C i − C f C i × 100 ,
(1)

where Ci is the initial concentration of dye in the solution and Cf is the final concentration of dye in the solution.

The pH, adsorbent dose, initial dye concentration, time and temperature are chosen as the independent input variables and the efficiency of dye removal as dependent response variable. Independent variables, experimental range and levels for congo red dye removal are given in Table 2. A 25 full factorial experimental design[36], with ten replicates at the center point, 10 experiments at axial points and thus a total of 52 experiments are employed in this study. The alpha value is chosen as face centered (α = 1). When α = 1, the axial points are placed on the ‘cube’ portion of the design. This is an appropriate choice when the cube points of the design are at the operational limits. The center point replicates are chosen to verify any change in the estimation procedure, as a measure of precision property. The experimental plan showing the coded value and uncoded value of the variables together with the dye removal efficiency for CR is given in Table 3. For statistical calculations, the variables X i are coded as x i according to the following relationship:

x i = X i − X 0 δ X ,
(2)

where x i is the dimensionless coded value of the variable X i , X0 is the value of the X i at the center point, and δ X is the step change.

Table 2 Experimental range and levels of independent process variables
Full size table
Table 3 Central composite design matrix with experimental and predicted values
Full size table

The behavior of the system is explained by the following quadratic equation:

Y = β 0 + ∑ β i x i + ∑ β ii x i 2 + ∑ β ij x i x j
(3)

where β0 is the offset term, β i is the linear effect, β ii is the squared effect, β ij is the interaction effect, and x i is the dimensionless coded value of the variable X i .

The results of the experimental design are studied and interpreted by MINITAB 14 (State College, PA, USA) statistical software to estimate the response of the dependent variable.

Results and discussion

The experiments were performed at the specified combinations of the physical parameters using statistically designed experiments in order to study the combined effects of these factors. The main effects of all the variables taken into consideration in this work are presented in Figure 1. From the figure, it is clear that the variables considered for this investigation play an important role in the adsorption studies of CR dye removal since each variable has a significant contribution on the percentage of dye removal. A main effect occurs when the mean response changes across the levels of a factor. The main effect plots were used to plot data means when multiple factors are involved in the system/process and also to compare magnitudes of marginal means. The points in the plot were the means of the response variables at the various levels of each factor, with a reference line drawn at the grand mean of the response data. Grand mean is the mean of all observations, as opposed to the mean of individual variables. The line drawn across the plot (center line) represents the grand mean. From Figure 1, it is clear that the grand mean of the response falls at around 50% dye removal. Out of the variable considered, adsorbent dose and time are the dominant factors for the percentage of dye removal when compared to other variables.

Figure 1
figure 1

Main effect plot of parameters on CR dye removal.

Full size image

The percentage of CR dye removal decreases with an increase in initial pH of the system due to its influence on the surface properties of the CNS and ionization/dissociation of the CR molecule. At higher pH, the grand mean drops down to 30%. The maximum removal of dye can be obtained at a low pH. Low pH leads to an increase in H+ ion concentration in the system, and the surface of the CNS acquires a positive charge by absorbing H+ ions. As the CNS surface is positively charged at low pH, a significantly strong electrostatic attraction appears between the positively charged CNS surface and anionic dye molecule, leading to maximum adsorption of dye. As the pH of the system increases, the number of negatively charged sites increases and the number of positively charged sites decreases. A negatively charged surface site on the CNS does not favor the adsorption of anionic dye molecules due to the electrostatic repulsion. Furthermore, lower adsorption of the CR dye in alkaline medium is also due the competition from excess OH− ions with the anionic dye molecule for the adsorption sites.

The percentage of CR dye removal is increased with an increase in adsorbent dose. At higher CNS-to-CR concentration ratios, a very fast superficial sorption onto the CNS surface occurs which produces a lower CR concentration in the solution than when the CNS-to-CR concentration ratio is lower. This is because a fixed dose of CNS can only adsorb a certain amount of dye. Therefore, with higher adsorbent dosage, a larger volume of effluent can be purified with a fixed dosage of CNS. The decrease in the amount of CR adsorbed on to an adsorbent qe (mg/g) with increasing CNS dose is due to the split in the flux or the concentration gradient between the CR concentration in the solution and the CR concentration in the surface of the CNS. Thus, with increasing CNS dose, the amount of dye adsorbed onto the unit weight of CNS gets reduced, thereby causing a decrease in the qe value.

The effect of initial dye concentration on adsorption shows that the percentage of CR dye removal decreases with an increase in the initial concentration of CR dye. The initial dye concentration provides the necessary driving force to overcome the resistances to the mass transfer of CR between the aqueous phase and the solid phase. The increase in an initial dye concentration also enhances the interaction between CR and CNS. Therefore, an increase in an initial concentration of CR enhances the adsorption uptake of CR. This is due to an increase in the driving force of the concentration gradient, as an increase in the initial dye concentration.

The effect of contact time on the percentage of CR dye removal increases rapidly with time. This indicates that a longer contact time favors the reaction towards the equilibrium between the dye and adsorbent. It may be due to the fact that the available CR dye molecules are preferentially adsorbed to the most active sites of the CNS. The short contact time between the CNS and CR dye molecules gives the less percentage of CR dye removal and equilibrium adsorption capacity and vice versa. The adsorption of CR dye onto the CNS was investigated as a function of temperature, and the maximum removal of CR dye was obtained at lower temperature. This is mainly due to the decrease in the surface activity, suggesting that adsorption between CR and CNS was an exothermic process.

Response surface methodology has been successfully applied for optimizing conditions for CR dye removal. The experimental results are analyzed through RSM to obtain an empirical model for the best response. The uncoded regression model equation (second-order polynomial) relating the removal efficiency and process parameters is developed and given in Equation 4. The mathematical expressions of the relationship between the independent variables and dependent response are given in terms of uncoded factors. Apart from the linear effect of the parameter for the dye removal, the RSM also gives an insight into the quadratic and interaction effect of the parameters. These analyses are done by means of Fisher's F test and Student's t test. Student's t test was used to determine the significance of the regression coefficients of the parameters. The P values are used as a tool to check the significance of each of the interaction among the variables, which in turn may indicate the patterns of the interactions among the variables. In general, the larger the magnitude of t and smaller the value of P, the more significant is the corresponding coefficient term[40]. The regression coefficient and the t and P values for all the linear, quadratic, and interaction effects of the parameters are given in Table 4. From the high t values and very small P values, it is observed that the coefficients for the linear effect and quadratic effect of all the factors are highly significant. The high P values of 0.593, 0.224, and 0.752 for the interaction effect of pH and adsorbent dose, adsorbent dose and temperature, and initial concentration and temperature did not seem to be significant, respectively. The response surface plots to estimate the removal efficiency over independent variables are shown in Figure 2. The peak of the surface plot shows an optimum value of the percentage of dye removal for the relative effects of variable pH and adsorbent dose (Figure 2a), adsorbent dose and temperature (Figure 2b), initial concentration and time (Figure 2c), and initial concentration and temperature (Figure 2d) by keeping the other three variables at the midpoint values. From Figure 2, it is clear that the interaction effects of adsorbent dose-pH, adsorbent dose-temperature, and initial concentration-temperature are not significant towards the percentage of dye removal. The response surfaces of mutual interactions between the variables other than those discussed above are found to be elliptical which explains the significance of those variables. The significance of these interaction effects between the variables would have been lost if the experiments are carried out by conventional methods. The experimental results and the results of theoretically predicted responses (using the model equation) are shown in Table 3.

η = ( 2.9226 ) + ( 0.9618 X 1 ) + ( 0.7943 X 2 ) − ( 0.3165 X 3 ) + ( 2.1283 X 4 ) − ( 0.1324 X 5 ) − ( 0.0850 X 1 2 ) − ( 0.0228 X 2 2 ) + ( 0.0027 X 3 2 ) − ( 0.0132 X 4 2 ) + ( 0.0013 X 5 2 ) + ( 0.0005 X 1 X 2 ) − ( 0.0005 X 1 X 3 ) − ( 0.0096 X 1 X 4 ) + ( 0.0017 X 1 X 5 ) − ( 0.0003 X 2 X 3 ) + ( 0.012 X 2 X 4 ) + ( 0.00032 X 2 X 5 ) − ( 0.0015 X 3 X 4 ) − ( 0.00003 X 3 X 5 ) − ( 0.0012 X 4 X 5 ) ,
(4)

where X1 is the pH, X2 is the adsorbent dose (g/L), X3 is the initial concentration of adsorbate (mg/L), X4 is the time (min), and X5 is the temperature (°C).

Table 4 Estimated regression coefficients and corresponding t and P values
Full size table
Figure 2
figure 2

Response surface plot of CR dye removal. (a) pH and adsorbent dose, (b) adsorbent dose and temperature, (c) initial concentration and time, and (d) initial concentration and temperature.

Full size image

The statistical significance of the ratio of the mean square variation due to the regression and mean square residual error is tested using analysis of variance (ANOVA). ANOVA is a statistical technique that subdivides the total variation in a set of data into component parts associated with specific sources of variation for the purpose of testing hypothesis on the parameters of the model[41]. According to the ANOVA (Table 5), the F statistic values for all regressions are higher. The large value of F indicates that most of the variation in the response can be explained by the regression equation. The associated values of P lower than 0.01 indicate that the model is statistically significant[42]. The ANOVA table also shows a term for the residual error, which measures the amount of variation in the response data left unexplained by the model. The form of the model chosen to explain the relationship between the factors and the response is correct and has very good agreement with the experimental value. The high value of R2 and R2adj explains that the model is significant.

Table 5 ANOVA results for the quadratic model
Full size table

The FStatistics value of 48,028 is greater than the tabulated F20,31 which indicates that the fitted model exhibits a lack of fit of 0.000 at the 99% confidence level. The ANOVA indicates that the second-order polynomial model Equation 4 is highly significant and adequate to represent the actual relationship between the response (percent removal efficiency) and variables, with P values (0.0000) and a high value of the coefficient of determination (99.7%). This implies that 99.7% of the sample variation is explained by the independent variables. Even though the interaction terms pH and adsorbent dose, adsorbent dose and temperature, and initial concentration and temperature are found to be insignificant, the oval model holds good agreement with the experimental data.

The model Equation 4 is optimized using a multistage Monte-Carlo optimization technique[43]. The optimum values of the process parameters are obtained in uncoded units. The optimum values of the process variables for the maximum removal efficiency are 3.2, 24.76 g/L, 20 mg/L, 67 min, and 30°C for the variables pH, adsorbent dose, initial dye concentration, time, and temperature, respectively. The percentage of dye removal at the optimum conditions simulated from the model is 100%. The experiments were done at optimum conditions, and the percentage of dye removal thus obtained is 99.99% which has a very close agreement with the model. These results closely agree with the optimum values obtained from the response surface methodology as 3.2557, 24.7634 g/L, 20 mg/L, 67.1815 min, and 30°C for the variables pH, adsorbent dose, initial dye concentration, time, and temperature, respectively. These results confirm that the RSM could be effectively used to optimize the process parameters in complex processes using the statistical design of experiments.

Experimental

Adsorption experiments were conducted as per the design developed with response surface central composite design methodology. All the experiments were done in duplicate to minimize the experimental error. The results given in Table 3 are the average experimental values. The detailed experimental procedure was discussed in ‘Methods’ section.

Conclusion

A novel, low-cost adsorbent prepared from agricultural waste, cashew nut shell, is successfully applied for the removal of CR from its aqueous solution. The specific surface area, pore volume, and average pore diameter calculated by the BET equation states the effectiveness of the CNS adsorbent for the removal of CR from aqueous solution. The effect of pH, adsorbent dose, initial dye concentration, time, and temperature on the adsorption was studied with the 52 experiments designed by the 25 full factorial central composite designs. This study clearly shows that the response surface methodology is one of the suitable methods to optimize the best operating conditions to maximize the dye removal. The statistical analysis results proved the significance of the model developed from experimental data to optimize the parameters. A significant interaction was observed in CR removal with the combination effects of pH, adsorbent dose, initial dye concentration, time, and temperature. The optimum values of pH, adsorbent dose, initial dye concentration, time, and temperature were found to be 3.2, 24.76 g/L, 20 mg/L, 67 min, and 30°C for the complete removal of CR dye, respectively. The experimental values were in good agreement with the model predicted values.

Authors’ information

RS is working in the Department of Chemical Engineering, University of Louisiana at Lafayette, Lafayette, USA. His areas of research are environmental engineering, bio processing, renewable energy, and process modeling and simulation. SP is working in the Department of Chemical Engineering, SSN College of Engineering Chennai, India. Currently, he is doing research in the fields of waste water treatment and chemical reaction engineering.