Statistical analysis on the removal of malachite green dye using active carbons of Achyranthes aspera and Allamanda blanchetii plants

Malachite Green Dye (MGD), a basic dye classified in the tri phenyl methane family, has widely been used in dyeing of leather, jute, silk, paper and wool and also in distilleries. Further, the dye is widely probed as a parasiticide, bactericide and fungicide and in many aquacultural industries globally (Zhang et al. 2008). The effluents from these industries, if not completely free from the dye before discharging, enter into water bodies (Uma et al. 2013). As the dye is non-degradable, it becomes persistent in the water steams and poses an impending environmental problem. It is highly lethal to freshwater fishes, in both acute and chronic exposures. Moreover, due to its mutagenic, carcinogenic, genotoxic effects, it causes serious public health hazards to aquatic fauna, flora and human beings (Srivastava et al. 2004). The clinical observations described so far reveal MGD as a ‘multi-organ toxin’. It decreases the intake of food, affects the progress of growth and fertility; and also cause severe harm to heart, kidney, spleen, liver. Further, the dye imposes lesions on the lungs, skin, and also in eyes; and poses erotogenic effects (Yonar & Yonar 2010). The colored waters, besides possessing non-aesthetic nature, obstructs with the passage of sunlight into the water bodies and thereby decreases the natural photosynthetic action and disturbs the eco-systems (Ashtoukhy 2009).

Various conventional methods based on membrane separation, biodegradation, electro-coagulation, reverse osmosis, chemical oxidation, adsorption and photo degradation (Sharma et al. 2010) are reported in literature for the removal of dyes form polluted waters and these methods suffer from one or the other disadvantage such as being non-economical, involving tedious procedures etc. (Yadav et al. 2013; Sujitha & Ravindhranath 2018a). The adsorbents derived from bio-materials are attracting the attention of the researchers in controlling various dyes. In fact, our research group made some progress in this aspect and found interesting results and they have been reported to the literature (Sujitha & Ravindhranath 2017a, 2017b, 2018b; Naga Babu et al. 2018).

On perusal of the literature, it is inferred that there are reports on the removal of MGD using bio-adsorbents, activated carbon of Durian seed (Mohd Azmier Ahmad et al. 2014), Coconut fronds (Mohammad et al. 2017), Avena sativa hull (Banerjee et al. 2016), Artocarpus altilis fruit skin (Lim et al. 2016), brown marine macro algae (Jerold & Sivasubramanian 2016), Lindley seed biomass (Aksakal et al. 2009), Biopolymer adsorbent (Sekhar et al. 2009), neem leaves (Odoemelam et al. 2018), potato plant waste (Gupta et al. 2011), fruit shell of tamarind (Saha et al. 2010b), CO₂-activated carbon derived from cattail biomass (Yu et al. 2017), activated carbons of epicarp of Ricinus communis (Santhi et al. 2010), cashew nut bark (Parthasarathy et al. 2011), neem sawdust (Khattri & Singh 2009) and organically modified clay (Arrellano-Cárdenas et al. 2013) have been employed as bio-adsorbents for the removal of MGD.

In the present investigation, we examined various plant materials as adsorbents for the removal of MGD and observed that the active carbons developed from the stems of Allamanda blanchetii and Achyranthes aspera have affinity towards MGD. In the present work investigations are made in this regard and are presented comprehensively.

Malachite green dye (MGD) is a cationic dye; classified as tri aryl methane dye and is widely used in pigment industries. It exists as a chloride salt, the molecular formula of MGD is [C₆H₅C (C₆H₄N (CH₃)₂)₂] Cl. The dye shows a strong absorption band at 618 nm and hence the intense green color results. The structural formula of MGD is shown in Figure 1.

Achyranthes aspera [Figure 2(a)] is a herb of the Amaranthaceae family and it grows to a height of nearly 15 cm in all parts of India. Allamanda blanchetii [Figure 2(b)] is a species of flowering plants, belongs to the family Apocynaceae. Allamanda plants are evergreen, erect or weakly climbing shrubs growing up to 3 meters tall. Both the plants are distributed universally all around the world especially in the southern parts of India.

All chemicals used in this work are of Analytical Reagent Grade and were purchased from Merck Pvt. Ltd.

(a) Achyranthes aspera active carbon preparation:

The cut pieces of the stems of Achyranthes aspera and Allamanda blanchetii plants are washed with distilled water; dried under sunlight and carbonized at 300 °C for 120 minutes in a muffle furnace. The carbons obtained were ground and sieved with 75 µm ASTM mesh and the sieved carbons were distilled water washed, filtered and oven dried at 110 °C. The carbons were activated by heating the carbon powders with 1N HNO₃ at 800C^o for 3 h. Thus activated carbons were filtered and washed thoroughly with de-ionized water till the washings were neutral, oven dried (110 °C) for 4 hrs and stored in air tight bottles. Thus obtained nitric acid activated carbons prepared from the stem parts of Achyranthes aspera and Allamanda blanchetii plants are named as NACSAA and NACSAB respectively.

The surface morphologies of the adsorbents: NACSAA and NACSAB (before and after MGD loading) were studied to detect the possible changes on the adsorbents' surface. The Field emission scanning electron microscope (FESEM) analytical studies were made using a JEOL JSM-7600F model instrument. The SEM monographs were taken with varying resolutions at 10.0 kV.

The Fourier transform-infrared (FT-IR) spectra of the adsorbents: NACSAA and NACSAB (before and after MGD loading) and malachite green dye were noted by a FT-IR spectrophotometer of model: BRUKER ALFA in the frequency range 4,000–500 cm⁻¹.

Further, the adsorbents were characterized for various textual properties using standard procedures described elsewhere (Bureau of Indian Standards 1989; Namasivayam & Kadirvelu 1997; ASTM D4607-94 2014). The BET surface area for NACSAA and NACSAB (before and after adsorption of MGD) were measured using a nitrogen gas adsorption analyzer (computer-controlled), at 78 K (Brunauer et al. 1938).

Batch type adsorption studies were executed as described in previous literature (Sujitha & Ravindhranath 2018c). The dye concentration in the solution was assayed by UV-visible spectrophotometer (λ_max: 618 nm) as described elsewhere (American Public Health Association 1998).

The influence of the combined effects of parameters such as pH, initial dye concentration, adsorbent dose and contact time for the two adsorbents NACSAA and NACSAB were investigated and the observations are presented in Figures 5(a)–5(f) and 6(a)–6(f).

The influence of temperature on the extraction of MGD was investigated and the results are depicted in Figure 7(a) and 7(b) and Table 2. The influence of a five times excess of co-ions exist naturally in water was investigated and the results are depicted in Figure 7(c) and 7(d).

The adsorptive nature of the adsorbents NACSAA and NACSAB was investigated by using different isotherms of adsorption and models of kinetics. The results are presented in Figure 8(a)–8(h) and Tables 3 and 4.

A preliminary sensitivity analysis is carried out on the physicochemical parameters namely, pH (A), sorbent dosage (B), agitation time (C), and initial MGD concentration (D), which affect the adsorption capabilities of adsorbents. However, to analyze the combined effect of the said parameters on the removal efficiency of the adsorbents, a statistical optimization modelling was carried out using a Response Surface Method (RSM) based Central Composite Design (CCD) package of the statistical software package Design-Expert, Minneapolis, USA (Box & Hunter 1957; Sen et al. 2014). The empirical model is developed in terms of percentage of MGB adsorption (Y) using the CCD package, as it employs a minimum possible number of samples in optimizing the identified physicochemical parameters to achieve the maximum possible removal efficiencies. Further, an ANalysis Of VAriance (ANOVA) model was adopted to investigate the reliability of the developed empirical model.

The adsorbents NACSAA and NACSAB were characterized for various textual properties and the obtained results are depicted in Table 1. BET surface area for NACSAA and NACSAB (before after adsorption of MGD) was measured and the obtained values are noted in Table 1.

The SEM monographs were taken for the adsorbents: NACSAA and NACSAB (before and after MGD loading) and are presented in Figure 3(a)–3(d). These SEM monographs reveal the porosity and surface texture of the samples. Figure 3 shows the SEM monographs of NACSAA and NACSAB before and after MGD loading. It is clearly seen from Figure 3(a) and 3(b) that the surfaces of the adsorbents are rough and uneven in nature and have considerably porous textures, which are requisite for the effective adsorption of MGD. The surfaces of the adsorbents (Figure 3(c) and 3(d)) are smooth and homogeneous in nature and the pores are completely clogged after adsorption of MGD. These results confirm the sorption of MG dye onto the surface of both the adsorbents.

Figure 3(i) and 3(ii) represent the FT-IR spectra of MGD, NACSAA and NACSAB before and after MGD loading.

On perusal of the FT-IR spectrum of MGD, the ten significant peaks, such as two peaks at 710 cm⁻¹ and 818 cm⁻¹ pertaining to bending vibrations of the C–H group; the peak at 1,040 cm⁻¹ corresponding to the -C-O group; the band at 1,222 cm⁻¹ corresponding to C-N stretching vibrations; the band at 1,435 cm⁻¹ pertaining to –C-H- deformations in alkanes; the peaks at 1,525 cm⁻¹ and 1,595 cm⁻¹ corresponding to C = C stretching of the benzene ring; the band at 1,736 cm⁻¹ pertaining to the –C = O group and two bands at 3,401 cm⁻¹ and 3,420 cm⁻¹ pertaining to stretching vibrations of the N–H group in primary amines are noticed.

On comparing the spectra of NACSAA before and MGD after adsorption, it is observed that the band at 1,238 cm⁻¹ (before adsorption) pertaining to C-N stretching vibrations is shifted to 1,375 cm⁻¹ (after) and the intensity of this band is decreased. The peak at 1,583 cm⁻¹corresponding to C = C stretching of the benzene ring appeared in the spectrum after adsorption of MGD. Moreover, the intensity of the peak representing the stretching vibration of the O–H group is decreased and the absorption peak is shifted from 3,429 cm⁻¹ to 3,332 cm⁻¹ after MGD loading. The shifting and appearance of new peaks may be attributed to the fact that there are interactions between the adsorbate MGD and the adsorbent.

On perusal of spectra of NACSAB (before and after MGD loading), it is noted that the band pertaining to C–H bending vibrations at 670 cm⁻¹ (before) is shifted to 650 cm⁻¹ (after); 768 cm⁻¹ (before) shifted to 788 cm⁻¹ (after); a band at 1,050 cm⁻¹ (before) pertaining to the –C-O- group is shifted to 1,045 cm⁻¹ (after); a peak at 1,420 cm⁻¹ (before) corresponding to –C-H- deformations in the alkanes is shifted to 1,445 cm⁻¹. A new band at 1,598 cm⁻¹ pertaining to C = C stretching of the benzene ring appeared in the spectrum after adsorption of MGD. The peak at 1,698 cm⁻¹ (before) pertaining to the –C = O group is shifted to 1,693 cm⁻¹ (after); Further, the strength of the peak representing the stretching vibration of the O–H group is decreased and the absorption peak is shifted from 3,394 cm⁻¹ to 3,378 cm⁻¹ after adsorption of MGD. These changes indicate that MGD adsorption has taken place via surface interactions and molecular adsorption into the pores of the adsorbents by capillary forces.

The analysis of the combined effect of physicochemical parameters for the extraction of MGD on to the adsorbents NACSAA and NACSAB was carried out and the findings are presented in Figures 4(a), 4(b), 5(a)–5(f) and 6(a)–6(f).

Figures 5(a), 5(b), 6(a) and 6(b) represent the combined influence of pH with sorbent dosage and contact time respectively on extraction of MGD ions for the adsorbents NACSAA and NACSAB. The influence of pH on the removal ability of the adsorbents was examined by changing the pH from 2 to 10, keeping constant the other parameters of extraction. It is seen from the figures that the percentage of extraction of MGD increases with increase in pH and the extraction is most favorable at basic pH 8–9 with 91% and 100% extraction of MGD onto NACSAA and NACSAB respectively. This can be explained from the pHzpc values which are found to be 5.0 for NACSAA and 5.53 for NACSAB as described in Figure 4(a) and 4(b).

At pH < pHzpc, the % removal of dye is less because of the protonation of the adsorbents' surface, which obstructs the extraction of MGD ions. But when the solution pH > pHzpc, the surface of the adsorbents becomes deprotonated and at high pH values, the surface acquires negative charge. Thus the acquired negative charge imparts a kind of a thrust on the inter-surface of the adsorbent and solution for positive ions. Hence, the MGD being a cation below pH: ∼10.5 (pK–10.3), shows more adsorption towards the adsorbents in the pH range 8–9 because the ideal conditions for adsorption are that the dye is in the cationic form and the surface must have a negative charge; these conditions are satisfied in the said pH range. But when the solution pH is further increased, the dye species are hydrolyzed in basic conditions which leads to the loss of the positive surface charge and so the percentage of extraction is less favored at high pH values.

Figures 5(c) and 6(c) represent the combined influence of initial concentration with pH for the extraction of MGD ions on to the adsorbents NACSAA and NACSAB. The influence of the initial MGD concentration on the rate of adsorption was investigated by changing the initial dye concentration in the range of 100–500 mg/L. From the figures it is observed that, with rise in the initial MGD concentration, the % removal of MGD decreased from 91% to 48% with the adsorbent NACSAA and 100% to 53% with NACSAB. This is attributed to the fact that, at low MGD concentrations, the ratio of the number of adsorbate ions to the available sorption sites of the adsorbents is fewer. Hence, the adsorption is independent of the initial concentration of dye resulting in more % removal. But as the initial concentration increases, the available sites become far less and so the extraction of MGD is dependent on the initial dye concentration (Saha et al. 2010a) and hence, the % removal is less.

But, the adsorption ability (q_e) is enhanced from 15.16 to 40.0 mg/g for NACSAA and with NACSAB it is increased from 20.0 to 53.0 mg/g. This is because, with a rise in the initial dye concentration, the dye molecules are induced to correlate with the active sorption sites of the adsorbents and hence, the uptake of dye by the adsorbents is increased.

Figures 5(d), 5(e), 6(d) and 6(e) represent the combined influence of adsorbent dosage with contact time and initial concentration respectively on the adsorption of MGD ions onto the adsorbents NACSAA and NACSAB respectively. The influence of sorbent dosage on the removal efficiency of the adsorbents was investigated by changing the adsorbents' dosage from 0.2 g/100 mL to 1.0 g/100 mL while keeping constant the other parameters of adsorption. It is noted from the figure that with increase in adsorbent dosage, the % removal is enhanced and 91% (maximum) of extraction is noted with 0.6 g/100 mL of NACSAA and 100% extraction of MGD is observed with 0.5 g/100 mL of NACSAB. Thereafter, the % removal decreased slightly with increase in the sorbent dosage, which is attributed to the fact of aggregation of active sorption sites. This results in a decrease in the existing surface area to MGD for adsorption and hence, less percentage removal.

Figures 5(f) and 6(f) represent the combined influence of contact time and initial concentration on extraction of MGD onto the adsorbents NACSAA and NACSAB respectively. The influence of contact time on the extraction ability of the adsorbents was studied by changing the contact time from 15 minutes to 75 minutes while keeping constant the other parameters of adsorption. It is noted from the figure that the adsorption of MGD is enhanced with the rise in contact time and reached the maximum percentage of extraction of 91% with NACSAA and 100% with NACSAB after 50 minutes and 40 minutes of agitation respectively and further increase of time of contact did not enhance the removal efficiency of MGD as all the active sites are utilized with time till a steady state is reached. The adsorption process occurred rapidly in the initial stages because of the large obtainability of sorption sites of the adsorbents, and in the advanced stages, the number of active sites were scarcely available. Hence, the extraction process becomes attachment-controlled and the process of extraction slows down.

Moreover, the plots of actual values vs. predicted values for the extraction of MGD onto the adsorbents NACSAA and NACSAB respectively are shown in Figure S1. It is observed from the figures that, the actual values and the expected values are in agreement with one another, as has been revealed from the good correlation coefficient values (R²) for the adsorbents NACSAA (0.9872) and NACSAB (0.9683). The statistical values obtained from the ANOVA model for the adsorption of MGD are presented in Figures S2 and S3.

The optimum conditions of extraction for the adsorbent NACSAA are found to be: pH: 8–9; sorbent dosage: 0.6 g/100 mL; contact time: 45 minutes; initial MGD concentration: 50 mg/L and temperature: 30 °C ± 1 °C.

% removal (Y) with NACSAA: − 9.55000 + 18.70000 * A + 42.00000 * B + 0.36333 * C − 0.22633 * D − 2.42187 * AB + 0.051042 * AC + 3.12500E-004 * AD + 0.18958 * BC + 6.87500E-003 * BD − 9.08333E-004 * CD −1.07266 * A² −19.76563 * B² − 4.62500E-003 * C² + 5.28750E-004 * D² = 91.0%

The optimum conditions of extraction for the adsorbent NACSAB were found to be: pH: 8–9; sorbent dosage: 0.5 g/100 mL; contact time: 30 minutes; initial MGD concentration: 50 mg/L and temperature: 30 °C ± 1 °C.

% removal (Y) with NACSAB = + 78.00 + 14.05 * A + 0.31 * B + 1.89 * C − 5.43 * D − 1.21 * AB − 1.09 * AC − 1.22* AD − 0.54 * BC − 0.025 * BD + 0.35 * CD − 2.77 * A² − 2.10 * B² − 0.10 * C² + 2.77 * D² = 100.0%.

Thermodynamic parameters (ΔH, ΔS and ΔG) were assessed by changing the temperature from 303 k to 333 k by keeping constant the other adsorption parameters. The observations are as shown in Figure 7(a) and 7(b). For this study, lnK_d was plotted against 1/T as per the Van't Hoff Equation: ln K_d = ΔS/R − ΔH/RT, where K_d = q_e/C_e and from which ΔH and ΔS are calculated; the ΔG value was calculated from the relation ΔG = ΔH – TΔS, where K_d (L/mg) is the distribution coefficient for the adsorption, q_e is the amount of dye adsorbed on the adsorbent at equilibrium, C_e is the equilibrium concentration of the dye in the solution, T is the absolute temperature, and R is the gas constant (Romero-Gonzalez et al. 2005). K_d values in L/mg were converted into L/g by multiplication with a factor of 1,000 and the values acquired were multiplied with the m. wt. (molecular weight) of the MGD to convert the units of K_d to L/mol (Lima et al. 2015).

The values of different thermodynamic parameters are tabulated in Table 2. It is noted from the table that the positive value of ΔH reflects the endothermic nature of the adsorption process and that fact is supported by the increase in K_d values with rise in temperature. ΔS values are positive and they demonstrate the affinity between the adsorbents and MDG. Moreover, the ΔG values are negative, indicating the spontaneity of adsorption.

The influence of co-anions, namely NO₃⁻, Cl⁻, CO₃²⁻, SO₄²⁻ and PO₄³⁻ and co-cations; that is, Na⁺, Zn²⁺, Mg²⁺, Ca²⁺, and Cu²⁺ are naturally present in water and these ions strive with the MGD molecules for the active vacant sites of adsorption and subsequent decrease in the removal efficiencies of adsorbents results. The influence of competing ions on the extraction ability of dye was investigated using 500 mg/L concentrations of co-ions and an initial dye concentration of 100 mg/L, keeping constant the other adsorption parameters: pH: 8; agitation time: 50 min for NACSAA, 40 min for NACSAB; sorbent concentration: 0.6 g/100 mL for NACSAA, 0.5 g/100 mL for NACSAB and temperature 30 °C ± 1 °C. The data obtained is depicted in Figure 7(c) and 7(d). It is observed from the figure that the percentage extractability of MGD is slightly influenced by the anions but cations have some effect. The interference of cations on the % of extraction of MGD is in the order: Zn²⁺ > Ca²⁺ > Mg²⁺ > Cu²⁺ > Na⁺.

R_L is a dimensionless separation factor obtained from the Langmuir model, R_L is described using R_L=1/(1 + a_LC_i). The R_L values for the adsorbents NACSAA and NACSAB are 0.00986 and 0.00988 respectively, as the R_L values are less than 1, the adsorption process is favorable (Atkins 1999; Monika et al. 2009).

The R² values from the Table 3 indicate that the process of extraction is explained in the following decreasing order: Freundlich model > Langmuir > Temkin > Dubinin-Radushkevich for both the adsorbents. The well fitted model is Freundlich representing the heterogeneity of the adsorbents surface and multi-layered process of adsorption.

R² values and the rate constants of the described kinetic models are measured and depicted in Tables 3 and 4 respectively. It is revealed from the table that the R² value is high for the pseudo-second order model for both the adsorbents: 0.994 for NACSAA and 0.995 for NACSAB. This infers the pseudo-second-ordered adsorption kinetics for both adsorbents.

Desorption experiments were made using the procedure as described in previous literature (Sujitha & Ravindhranath 2018c). It is observed that 0.1 M HCl was effective. Figure 9 shows the performance of regenerated adsorbents in the removal of MGD as % removal vs. No. of cycles. It is observed from the figure that the removal efficiency is more than 80.0% for every cycle and it did not come down below 80.0% even after five regenerations for both the adsorbents. Thus, it can be deduced that the adsorbents can be used repeatedly in the treatment of wastewater containing dyes.

The developed NACSAA and NACSAB-based methodologies were applied to the effluent samples obtained from different dyeing industries in A.P. (Andhra Pradesh) and the observations are noted in Table 5. Not less than 80.0% extraction of MGD was observed with these effluent samples.

The NACSAA and NACSAB-based methods developed in this investigation were compared with the reported literature (Zhang et al. 2008; Arivoli et al. 2009; Khattri & Singh 2009; Sekhar et al. 2009; Saha et al. 2010b; Tsai & Chen 2010; Parthasarathy et al. 2011; Santhi et al. 2011; Chieng et al. 2014; Gautam et al. 2015; Mirzajani & Ahmadi 2015) with respect to q_e (mg/g) and pH (Table 6). It is inferred from the table that NACSAA (40.0 mg/g) and NACSAB (53.0 mg/g) have shown higher adsorption capacity than hitherto reported adsorbents in the literature.

Two adsorbents were developed based on HNO₃ activated carbons from stems of Achyranthes aspera and Allamanda blanchetii plants; namely, NACSAA and NACSAB. These bio-adsorbents were probed for the extraction of cationic malachite green dye from contaminated water by varying the extraction conditions for the maximum possible extraction.

The adsorbents were characterized by various surface morphological techniques including FESEM and FT-IR. The adsorption abilities were found to be 40.0 mg/g for NACSAA and 53.0 mg/g for NACSAB respectively, and these figures are higher than many sorbents reported in literature.

The percentage extractability of MGD was slightly affected by the anions but cations had some effect. The interference of cations on the % of extraction of MGD was in the order: Zn²⁺ > Ca²⁺ > Mg²⁺ > Cu²⁺ > Na⁺.

The mechanism of adsorption was assessed by different adsorption isotherm models and is explained in the following decreasing order: Freundlich model > Langmuir > Temkin > Dubinin-Radushkevich for both the adsorbents. The best fit model is Freundlich, representing the heterogeneous nature of the surfaces of sorbents and a multi-layer process of extraction.

Of the different kinetic models investigated, the adsorption is well described by pseudo-second ordered kinetics. Thermodynamics parameters (ΔG, ΔS and ΔH) reveal the endothermic and spontaneous process of sorption for both the adsorbents. The fall in ΔG values with the rise in temperature demonstrates the favorable extraction conditions at high temperatures.

Substantial amounts of dye (not less than 80.0%) are noted even with five time regenerated spent adsorbents, NACSAA and NACSAB. The procedures developed in this investigation were successfully applied to water samples procured from different dyeing industries.

The authors thank the Koneru Lakshmaiah Education foundation for providing the needed facilities to conduct the present work. The authors thank the Particulate Materials laboratory, MEMS Department, IIT Bombay, for facilitating the instruments for the characterization of samples in this work.