Introduction

Several industries extensively use synthetic dyes and generate a large volume of dye effluent. This dye effluent contains enormous salt and organic content and they are less biodegradable. Discharge of these dyes into natural water system without the treatment results in ecological imbalance [1,2,3,4]. Synthetic dyes are resistance to the physicochemical reaction and have thermal and optical stability due to the complex aromatic structure [5, 6]. By conventional methods, it is challenging to treat the wastewater which contains dyes [7, 8]. Azo dyes account for 60–70% of total dye consumption in many industries [7]. Congo red (CR) is the synthetic anionic azo dye [9] and the effluent had to be adequately treated and discharged into the environment, otherwise it can metabolize into benzidine, which is a human carcinogen [10]. CR dye contains complex aromatic structure which complicates the treatment process. It has thermal and optical stability, and resistance to biodegradation and photodegradation [11]. CR dye is used for staining purpose and as a pH indicator in the laboratory. It is also used for detecting bacteria, amyloidosis diagnosis and to treat protein folding disorders, dermatological disorders, neurodegenerative diseases, etc. However, CR dye causes allergic reactions and it is a cytotoxic, genotoxic, hematotoxic, neurotoxic, as well as mutagenic substance [12, 13]. Generally, chemical oxidation, ozonisation, membrane filtration, ion exchange or electrochemical techniques are the different methods used to treat dye-contaminated wastewater. Adsorption is proved to be the best method compared to all other technologies due to its less capital outlay. Detailed literature review shows that various adsorbents were used for Congo red dye removal from wastewater, which includes polypyrrole–polyaniline nanofibers [14], modified zeolites [2], sugarcane bagasse [15], aniline propyl silica xerogel [16], tamarind fruit shell [17], acrylamide–sodium dodecyl sulfate layered double hydroxides [18], mesoporous ZrO2 fibers [19], ammonium aluminum carbonate hydroxide-nickel hydroxide composite [20], zinc oxide–alumina composite [21], nickel cobaltite [22] and Funalia trogii [23]. However, some of these are not effective for the anionic dye. The biosorbent (Chrysanthemum indicum) belongs to Asteraceae family and are known as Indian chrysanthemum which are widely used for decoration purposes [24, 25]. Further, the literature review indicates that Chrysanthemum indicum flowers (CIF) which are used in this study as an adsorbent for CR dye removal have not yet been studied previously. In India, CIF is generally used for the ornamental purpose and the regular return of CIF is 16.7 tons/hectare that is reported in our previous study [26]. The main aim deals with CIF ability as an adsorbent to treat synthetic CR dye wastewater. CCD is widely adapted to observe the effect of individual process factors and their interactive effects.

Materials and methods

Microparticles

CIF was obtained from the native marketplaces of Vellore district, Tamil Nadu, India and meticulously washed to remove dust and dirt particles. Petals were collected and dried and powdered using a domestic mixer. This powder was sieved. Particles of size less than 100 microns were further used as an adsorbent. Physical characteristics of the microparticles are presented elsewhere [26].

Dye

Congo red (CR) dye (C32H22N6Na2O6S2; MW- 696.665; λmax = 497 nm) working derivative solutions made appropriately by diluting the stock solution (1 mg mL−1) with double distilled water.

Adsorption experiment

Experimental trials were carried based on the procedure mentioned in our previous studies [26]. 250-mL Erlenmeyer flasks were used with 100 mL of CR dye solution based on the experimental run generated as mentioned in Table 1. Changes in the reduction were analyzed in the residual supernatant obtained after centrifugation (8000 rpm for 10 min) by UV–Vis spectrophotometer (λ—497 nm). Percentage dye reduction was calculated using the following equation:

$$ {\text{Percentage}}\;{\text{ reduction }} = \, \left[ {\left( {C_{\text{i}} {-} \, C_{\text{f}} } \right) \, / \, C_{\text{i}} } \right] \, \times \, 100, $$

where Ci, Cf are the initial and final dye concentrations (mg L−1), respectively.

Table 1 Level of independent variables and experimental range
Full size table

Experimental design

In tradition method of adsorption, variable response effect is studied by varying one variable at a time, ending up in large experimental trails associated with high time consumption process without representing the interaction effect between factors. Therefore, RSM is considered as one of the most useful methods for optimizing the process and portray the influence by several parameters [27, 28]. The experimental trails are carried out based on the quadratic equation (Eq. 1) between the independent variable and coefficients of central composite design (CCD), as expressed below:

$$ y \, = \, \mu_{0} + \, \mu_{1} F_{1} + \, \mu_{2} F_{2} + \, \cdots \, + \mu_{n} F_{n} \pm \, c, $$
(1)

where y is the response (yield), µ is the response function, c is the error, and F1, F2, F3 … Fn are the independent process parameters.

Results and discussion

Central composite design

Based on CCD principle, the design of six variables consists of 53 experiments that are calculated by the following expression:

$$ N = 2^{M - 1} + 2M + f_{0} , $$
(2)

where N is an experimental trail, M is the independent variables, and f0 is the central points.

Table 1 shows the independent variables and their experimental range. Table 2 shows the 26 fractional factorial design CCD matrix.

Table 2 Fractional factorial CCD matrix with code and real variables
Full size table

Equation 2 is expanded to Eq 3 for a six-variable design:

$$ \begin{aligned} y & =\, \mu_{0} + \mu_{1} F_{1} + \mu_{2} F_{2} + \mu_{3} F_{3} + \mu_{4} F_{4} + \mu_{5} F_{6} + \mu_{11} F_{12} + \mu_{22} F_{22} + \mu_{33} F_{32} + \mu_{44} F_{42} \\ & \quad + \mu_{55} F_{52} + \mu_{66} F_{62} + \mu_{12} F_{1} F_{2} + \mu_{13} F_{1} F_{3} + \mu_{14} F_{1} F_{4} + \mu_{15} F_{1} F_{5} + \mu_{16} F_{1} F_{6} + \mu_{23} F_{2} F_{3} \\ & \quad + \mu_{24} F_{2} F_{4} + \mu_{25} F_{2} F_{5} + \mu_{26} F_{2} F_{6} + \mu_{34} F_{3} F_{4} + \mu_{35} F_{3} F_{5} + \mu_{36} F_{3} F_{6} + \mu_{45} F_{4} F_{5} + \mu_{46} F_{4} F_{6} \\ & \quad + \mu_{56} F_{5} F_{6} , \\ \end{aligned} $$
(3)

where y is the dye removal (%); F1, F2, F3, F4, F5 are the independent variables; µ0 is the regression coefficient; µ1, µ2, µ3, µ4, µ5, µ6 are the linear coefficients; µ11, µ22, µ33, µ44, µ55, µ66 are the quadratic coefficients; and µ12, µ13, µ14, µ15, µ16, µ23, µ24, µ25, µ26, µ34, µ35, µ36, µ45, µ46, µ56 are the interaction coefficients.

With the coefficients in Eq 3:

$$ \begin{aligned} y & \quad = \, 59.3528 \, {-} \, 3.8176F_{1} + \, 5.9990F_{2} {-} \, 3.1209F_{3} + \, 0.7326F_{4} + \, 2.1713F_{5} {-} \, 0.4646F_{6} \\ & \quad + 2.0057F_{12} {-} \, 0.4492F_{22} + \, 0.7676F_{32} {-} \, 0.6081F_{42} - \, 1.1907F_{52} + \, 1.3164F_{62} {-} \, 0.0364F_{1} F_{2} \\ & \quad + .9034F_{1} F_{3} {-} \, 0.2190F_{1} F_{4} + \, 4.3763F_{1} F_{5} {-} \, 1.1797F_{1} F_{6} + \, 1.0092F_{2} F_{3} {-} \, 0.5795F_{2} F_{4} \\ & \quad + 3.4821F_{2} F_{5} + \, 0.3418F_{2} F_{6} + \, 2.7554F_{3} F_{4} {-} \, 1.0721F_{3} F_{5} + \, 1.4783F_{3} F_{6} + \, 0.4062F_{4} F_{5} \\ & \quad - 2.2752F_{4} F_{6} + \, 0.5504F_{5} F_{6} . \\ \end{aligned} $$
(4)

Microparticle characteristics

The morphology of pre- and post-treated adsorbents surface was studied using scanning electron microscopic technique and presented in Fig. 1. It can be observed from the figure that the surface of the adsorbent was irregular and multifarious in nature. In addition, the particles are asymmetric and the uneven surface increases the physical adsorption. FTIR spectrum depicts the functional groups present in the adsorbent [29]. Figure 2 shows the FTIR spectra ranging between 400 and 4000 cm−1 of adsorbent pre- and post-experimental trails. In Fig. 2, the peaks at wavenumbers 1737.8 and 1600.9 cm−1 indicate the presence of C=O group. Broad absorption near 3400–2400 cm−1 confirms the occurrence of O–H group. Strong intensity absorptions near 1300–1000 cm−1 show the presence of esters(C–O group). Some functional group’s peaks in figure are shifted after adsorption. The downshift of the wavenumber from 1600 to 1641 cm−1 indicates the bending vibrations of N–H group in the adsorption process. The intensity of peaks at the regions 1737 cm−1, 1641 cm−1, 1429 cm−1, 1365 cm−1, 1228 cm−1, 1149 cm−1 significantly shifted due to the adsorbent contact with CR, indicating the complexation and attachment to the surface of the dye

Fig. 1
figure 1

SEM images of a raw and b CR dye-loaded adsorbent

Full size image
Fig. 2
figure 2

FTIR spectra of a raw and b CR dye-loaded adsorbent

Full size image

Point of zero charge (PZC)

PZC is to determine the pH influence on CR dye removal. 100 mL of double distilled water has taken in conical flasks. Adjust the pH of all flasks from 2 to 10 using 0.1 N NaOH and 0.1 N HCl and take as initial pH. Add 0.1 g of adsorbent to all flasks and agitated at 120 rpm for 24 h. Measure the pH after equilibrium which has been considered as final pH. Plot the graph of initial pH vs difference in initial and final pH values. The pH at which it is zero is considered as PZC value. From Fig. 3, it is noted that PZC value of CI is 6.6. This shows that at pH < 6.6, CI adsorbent carries positive surface charge, whereas at pH > 6.6 adsorbent carries negative surface charge.

Fig. 3
figure 3

Point of zero charge

Full size image

Response surface methodology

Table 1 shows the independent variables and their experimental range. Table 2 shows the 26 fractional factorial design CCD matrix, whereas Table 3 depicts the experimental results and predicted. Analysis of variance (ANOVA) which is used in evaluating adequacy and significance of the model is shown in Table 4. Values from the ANOVA table indicate that all variables and their interactions are significant. Degree of significance of each variable is represented by p value and variation of the data about its mean is described by F-value [30]. R2 (0.9569) value shows the reliability of the model. 3D surface plots were obtained from the validated model. The obtained F value (12.72 > Fcritical 1.94) is 95% significant endorses the model is suitable [31]. R2 (0.9569) which is near to 1 designates the fit of regression line with the data. From Table 5, the coefficients µ1, µ2, µ3 are significant single effects, µ11, µ55, µ66 are the significant square effects and µ13, µ15, µ16, µ25, µ34, µ36, µ46 are the significant interactive effects as p < 0.05.

Table 3 Fractional factorial CCD matrix and the output response for dye removal
Full size table
Table 4 ANOVA for fit of dye removal (%) from central composite design
Full size table
Table 5 Estimated regression coefficients for removal of dye (%)
Full size table

Interpretation of 3D surface plots

Effect of pH

Adsorbent dose As pH increases dye adsorption decreases this is because at higher pH level, OH− ions compete with dye anions for active adsorption sites (Fig. 4a). Removal efficiency rises with increase in adsorbent dose due to unsaturated adsorption sites availability. The maximum dye removal is at low pH and increasing adsorbent dose which arguably would increase the positive terminal site at low pH to favor adsorption.

Fig. 4
figure 4

Interactive effect of a pH and adsorbent dose, b pH and initial dye concentration, c pH and RPM, d pH and contact time, e pH and temperature and f adsorbent dose and initial dye concentration, on removal of CR dye

Full size image

Initial dye concentration Maximum percentage dye removal is obtained at low pH and low dye concentration (Fig. 4b) as dye ions are less than a number of adsorption sites, indicating overlapping of sites (based on PZC value).

RPM Percentage removal of CR dye decreases as pH increases with given RPM (Fig. 4c). This is based on the competition of OH− ions with negatively charged dye ions. RPM alone does not have much effect on dye removal. Maximum removal efficiency is obtained at low pH and high RPM.

Contact time As pH increases dye removal decreases (Fig. 4d) by electrostatic repulsion of dye ions and negatively charged adsorbent according to PZC value of the adsorbent. At high pH as contact time increases removal efficiency of dye molecules increases as sufficient time is available to reach equilibrium.

Temperature At high temperature because of swelling of adsorbent, large dye molecules will penetrate into the adsorbent which intend increases the dye adsorption (Fig. 4e). Same observation is reported in the literature [32]. Maximum dye removal is seen at low pH and high temperature. At high pH region with low-temperature condition removal efficiency increases as mobility of dye molecules decreases which facilitates the adsorption.

Effect of adsorbent dose

Initial dye concentration The vital factor which limits the sorbent–sorbate equilibrium in the system is adsorbent [33]. Figure 4f depicts an increase in the removal of dyes with an increase in adsorbent dose and is maximum at a high dosage. This is attributed as the convenience of sufficient amount of unsaturated adsorbent spots for dye molecules.

RPM From Fig. 5a, it can be noted that RPM alone does not have much effect on percentage dye removal. Maximum adsorption of dye molecules is obtained at high adsorbent dose with given RPM. Since adsorbent dose increases the availability of active adsorption sites also increases which intend increases the adsorption.

Fig. 5
figure 5

Interactive effect of a adsorbent dose and RPM, b adsorbent dose and contact time, c adsorbent dose and temperature, d RPM and initial dye concentration, e initial dye concentration and contact time and f initial dye concentration and temperature, on removal of CR dye

Full size image

Contact time Figure 5b shows at a low dose when contact time increases removal efficiency decreases due to the occurrence of desorption. At high dose, with an increase in contact time adsorption also increases and is maximum at a high dose and high contact time. This is because at high dose, a maximum number of unsaturated adsorption sites are available and sufficient time is available to reach equilibrium.

Temperature Figure 5c shows that as adsorbent dose increases dye adsorption also increases due to an upsurge in adsorption sites. Dye removal decreases first as temperature increases then increase again. This is because at high-temperature viscosity in the medium decreases which increases the diffusion frequency of the dye molecules into adsorbent pores [34]. Moreover, at low temperature, the mobility of the dye molecules decreases which increases the adsorption process.

Interactions

RPM and initial dye concentration Removal efficiency declines with an increase in the concentration of dye at lesser RPM (Fig. 5d). This is because of non-availability of enough adsorption sites to attain equilibrium. At high RPM with an increase in dye concentration, percentage dye removal also increases. This is attributed to increased concentration gradient (driving force to overcome mass transfer).

Initial dye concentration and contact time Maximum dye removal efficiency is obtained at low concentration and high contact time (Fig. 5e). This is due to the availability of huge amount of surface capacity for adsorption of dye and enough time for interaction of adsorbent and adsorbate. Removal efficiency decreased at higher adsorbate concentration.

Initial dye concentration and temperature Interactive effect of initial concentration of adsorbate (dye) and the temperature is shown in Fig. 5f. At low temperature, increasing dye concentration decreases the removal rate due to deficiency of adsorption surfaces. As temperature increases surface area of adsorbent increases due to swelling effect which intends increases the adsorption of dye molecules.

RPM and contact time Percentage dye removal increased first with an increase in RPM and then decreases (Fig. 6a). This is because, as RPM increase desorption takes place first and then once again dye molecules adsorbs on the adsorbent. As time increases removal efficiency also increases irrespective of RPM.

Fig. 6
figure 6

Interactive effect of a RPM and contact time, b RPM and temperature, and c temperature and contact time, on removal of CR dye

Full size image

RPM and temperature From Fig. 6b it can be noted that the maximum removal is seen at less RPM and high temperature due to increase in pore size of the adsorbent. At high temperature, removal efficiency decreases w.r.t RPM. However, at a lesser temperature, percentage removal increases as RPM increases.

Temperature and contact time At higher temperature, when time improves adsorption of dye molecules also increases and is maximum at high time and temperature. This is because of increase in pore size of adsorbent due swelling effect and availability of sufficient time to reach equilibrium. At low temperature, increasing or decreasing of contact time from mid value has obvious decrease in adsorption of dye molecules (Fig. 6c).

Table 6 presents the comparison of CR dye removal in the present study with the different adsorbents. It can be observed from Table 6 that, the CR dye removal from aqueous solution by Chrysanthemum indicum microparticles is noteworthy and equivalent with other adsorbents reported in the literature.

Table 6 Comparison of CR dye removal from aqueous solution by various adsorbents
Full size table

Conclusion

Chrysanthemum indicum microparticles showed good ability as an adsorbent to reduce the pollutant CR. This study aims to optimize the process parameters of Congo red dye adsorption from chrysanthemum indicum using a CCD which is a RSM module. The optimum condition for maximum dye removal (84.1%) was obtained at pH1, adsorbent dose 0.3 g/100 ml, dye concentration 150 ppm, RPM 150, contact time 75 min and temperature 30 °C. FTIR analysis indicated the formation of new peaks, suggesting the complexation of dye on the surface of the adsorbent. Higher dosage of adsorbent-exhibited better removal of CR from the aqueous solution, suggesting that synthesized microparticles can be used as an efficient adsorbent.