Cobalt ferrite nanoparticles (CoFe2O4 NPs) are used as an efficient adsorbent to remove cobalt (II) phthalocyanine (CoPc) dye from aqueous solutions. The characterization of adsorbent is investigated by field emission scanning electron microscopy (FE-SEM), X-ray diffraction (XRD), Fourier transform infrared (FT-IR) spectroscopy, energy dispersive X-ray spectroscopy (EDX), and the vibrating sample magnetometer (VSM) technique. To optimize the effective factors, response surface methodology (RSM) through using Box–Behnken design (BBD) is applied. By proper running of the Desirability function option in MINITAB software, the optimum conditions were found as pH 3.2, adsorbent mass (m) 11 mg, contact time of nine minutes (t), and initial dye concentration (Cd) of 30 mg L−1. Isotherm studies of the adsorption process are carried out where the Langmuir isotherm shows the maximum monolayer capacity (qmax) is 431 mg g−1. The kinetic studies including pseudo-first-order, pseudo-second-order and intra-particle diffusion models indicate that the pseudo-second-order kinetic model describes better the adsorption kinetic behavior. This study shows that CoFe2O4 NPs have excellent potential for the removal of CoPc dye from an aqueous solution.

  • According to literature review, the removal of cobalt (II) phthalocyanine dye by adsorption process has not been reported yet.

  • In this work, cobalt ferrite nanoparticles are used as highly efficient adsorbent for this purpose with maximum uptake capacity 431 mg g−1.

  • Optimum conditions are found by applying BBD.

Graphical Abstract

Graphical Abstract
Graphical Abstract

In recent years, the contamination of water resources has been a major global environmental concern. Among all, dyes are considered as the most serious water pollutants. The widespread usage of dyes in various industries causes critical problems in the environment, where some dyes have carcinogenic and mutagenic effects (Forgacs et al. 2004; Aksu 2005; Crini 2006; Gupta & Suhas 2009). Phthalocyanine (Pc) dyes are used in various industries, including printing, computers, textiles, pharmaceuticals, coatings, and paint. Since they exhibit resistance against biological aerobic degradation in water, these high-risk colored wastewaters need to be purified before being released in nature (Moser & Thomas 1964; Matthews et al. 2009; Silva et al. 2012; Kaušpėdienė et al. 2013). Phthalocyanine reactive dyes are metal complexes and are used to produce shades of a blue and blue-green color. Most of the phthalocyanine dyes are complexed with copper. Due to the presence of metals such as copper, nickel, and cobalt, they are a major source of toxicity. Co(II) phthalocyanine is among this class of dyes and has a blueish pigment stain and there are no reports about its removal from water resources. To reduce their harmful effects on the environment, removing them from industrial effluents and wastewater is an important subject (Ambashta & Sillanpää 2010; Gupta et al. 2012). Among various separation methods, adsorptive removal is one of the most effective processes due to its specific characteristics such as affordability, non-toxicity, high-uptake capacity and high removal rate for physical water purification which can be applied to produce high-quality water. In the adsorption process, introducing new sorbents and developing their efficiency is a major objective (Khajeh et al. 2013; Kyzas & Matis 2015; Alalwan et al. 2020). In recent years, usage of metal nanocomposite has promoted the role of nanomaterials in the adsorption removal process due to their unique physical and chemical properties in comparison with conventional materials. Moreover, magnetic nanocomposite provides a new simple and facile separation step in adsorption removal procedures (Khajeh et al. 2013; Kyzas & Matis 2015). For this purpose magnetic oxides such as FeO, Fe2O3, and Fe3O4 are widely applied as a component of nanomagnetic composite adsorbent (Absalan et al. 2011; Panneerselvam et al. 2011; Khajeh et al. 2013; Kyzas & Matis 2015; Bagtash et al. 2016; Zolgharnein et al. 2016a; Zolgharnein & Feshki 2017). Since some of other transition metal oxides such as cobalt, copper and zinc have magnetic behavior, their role as components of nanomagnetic composite adsorbents has become increasingly interesting for researchers. So far, MFe2O4 magnetic nanoparticles such as CoFe2O4, MgFe2O4, MnFe2O4, and NiFe2O4 have been used as magnetic adsorbents (Zhao et al. 2010; Glover et al. 2012; Ghaemi et al. 2014; Singh et al. 2014; Ayazi et al. 2016; Berger et al. 2017; Mahmoodi & Abdi 2019). Cobalt ferrite (CoFe2O4) has been promoted as an efficient and powerful magnetic absorbent due to its chemical and electrochemical stability, relatively high permeability, and catalytic behavior (Nyokong 1995; Chinnasamy et al. 2002; Maya et al. 2006; Glover et al. 2012; Ghaemi et al. 2014).

In this work, cobalt ferrite nanoparticles (CoFe2O4 NPs) are synthesized in an alkaline media through a co-precipitation method (Chinnasamy et al. 2002; Ghaemi et al. 2014). The capability of using cobalt ferrite nanoparticles as an adsorbent for removal of CoPc is investigated. Since the multivariate optimization method has advantages over one at a time, the response surface approach (RSM) has been applied reaching the maximum removal percentage (R%) (Zolgharnein et al. 2016a; Zolgharnein & Feshki 2017). It has also led to a mathematical model which relates the removal response to their effective factors and finding optimization conditions (Kazemi et al. 2010; Zolgharnein et al. 2013, 2016a; Zolgharnein & Feshki 2017). Box–Behnken design (BBD) is applied for optimization of adsorption of CoPc on CoFe2O4 NPs (Kazemi et al. 2010; Zolgharnein et al. 2013). Adsorption characterization by various surface analyses is conducted. Finally, adsorption kinetics of CoPc on CoFe2O4 NPs and isotherm behavior are studied.

Materials and apparatus

The commercial reactive cobalt (II) phthalocyanine was used in this study (Figure 1). Chemicals of cobalt (II) chloride hexahydrate (CoCl2.6H2O), sodium hydroxide and hydrochloric acid were all supplied by Scharlab (Spain). Ferric chloride hexahydrate (FeCl3·6H2O) was purchased from Alfa Aesar (Germany). Double-distilled water was used for each run. A pH meter (Metrohm, model 744, Herisau, Switzerland), UV/Vis spectrophotometer (Analytikjena double beam SPECORD250, Germany), Fourier transform infrared (FT-IR) spectroscopy (SHIMADZU series FT-IR 8400S, Japan), and X-ray diffraction (XRD; Philips Co., The Netherlands) were used. The scanning electron microscopy (SEM) measurements were carried out using TESCAN (VEGA3, Czech). VSM measurements were performed by using a vibrating sample magnetometer (Meghnatis Daghigh Kavir Co. Kashan, Iran).

Figure 1

Chemical structure of cobalt (II) phthalocyanine.

Figure 1

Chemical structure of cobalt (II) phthalocyanine.

Synthesis of cobalt ferrite NPs

Magnetic cobalt ferrite nanoparticles were synthesized with a stoichiometric ratio of 1:2 of Co2+ and Fe3+ using the chemical coprecipitation method in an alkaline medium (Chinnasamy et al. 2002; Ghaemi et al. 2014). Briefly, 2.564 g of CoCl2 and 5.862 g FeCl3 were dissolved in 20 mL distilled water. Then, the prepared solution was kept at 60 °C. This mixture was added into a 250 mL sodium hydroxide solution (1.5 M) dropwise at a constant temperature of 80 °C with constant stirring. During the precipitation process, the reaction solution was stirred vigorously and nitrogen gas was bubbled to degas it for 50 minutes. The obtained precipitate was washed with distilled water at least three times until the solution pH became neutral.

Adsorption studies

To investigate the adsorption CoPc capability by prepared cobalt ferrite nanoparticles, adsorption examinations were accomplished in V = 25 mL of Co(II) phthalocyanine solution with primary concentrations of C0 (mg L−1). The pH was adjusted using NaOH and HCl 0.1 M by a pH meter (Metrohm, 744). M (g) of magnetic cobalt ferrite nanoparticles was added to the solution of CoPc and adsorptive removal was carried out by stirring the solutions by a mechanical mixer. After enough time, an external magnet was placed at the bottom of the beaker to separate the adsorbent from the solution. The residual concentration of CoPc, Ce (m L−1), was measured by a spectrophotometer (Analytikjena SPECORD 250) at λmax = 659 nm. The least square equation of calibration plot is y = 0.0355x + 0.0274 (R2 = 0.9992) with a dynamic range from 1 to 50 mgL−1 (Zolgharnein & Feshki 2017). The CoPc removal percentage (R%) was determined according to Equation (1):
formula
(1)
To investigate the adsorption amount, the adsorption capacity q (mg·g−1) was determined at a specified time according to Equation (2):
formula
(2)
where C0 (mg L−1) is the initial dye concentration, Ce (mg L−1) is the concentration of the dye at equilibrium, V (L) denotes the solution volume and M (g) is the mass of sorbent.

Multivariate optimization using Box–Behnken design

As mentioned earlier, multivariate optimization has many advantages over conventional optimization methods. It is due to this fact that the adsorption process is influenced by several effective factors (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008; Kazemi et al. 2010; Zolgharnein et al. 2013, 2014). For this purpose, the response surface methodology (RSM) was applied throughout the running Box–Behnken design (BBD) (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008; Kazemi et al. 2010; Zolgharnein et al. 2013, 2014). RSM is a method that can determine the impact of factors and their interactions on a response value by modeling, specifically here for removal percentage (R%) of CoPc (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008; Kazemi et al. 2010; Zolgharnein et al. 2013, 2014). Box–Behnken is of three-level factors (−1, 0, +1) and is a spherical and rotatable design (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008; Kazemi et al. 2010; Zolgharnein et al. 2013, 2014). Its total required runs are equal to N = 2k(K − 1) + Cp, where N is the number of experiments, K is the number of factors and Cp is the number of center points (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008; Kazemi et al. 2010; Zolgharnein et al. 2013, 2014). Box–Behnken design gives a quadratic model which relates the adsorption responses (R% and q) with effective factors and their interactions (Verma et al. 2008; Ghaemi et al. 2014; Zolgharnein et al. 2014):
formula
(3)
where x1, x2,. . ., xk are effective factors, β0 is a constant, is the slope or linear effect of the input factor xi, βii is the quadratic effect of the input factor xi (i = 1,2,. . ., k), βij is an interaction effect between the input factors xi and ε is a random error. The coefficients should be determined by the least square method (Ferreira et al. 2007a, 2007b; Zolgharnein et al. 2014). MINITAB 18 statistical software was used for multiple regression analysis of the design matrix of the obtained experimental data.

Cobalt ferrite nanoparticle adsorbent was characterized by various techniques.

Characterization of adsorbent

To identify involved functional groups of adsorbents, the FT-IR spectra of (a) cobalt ferrite nanoparticles, (b) CoPc and (c) the loaded adsorbent in the range of 400–4,000 cm−1 are shown in Figure 2. In the case of CoFe2O4 NPs (a), the 485 cm−1 and 421 cm−1 bands are attributed to the stretching vibrations of the M–O bond (Absalan et al. 2011; Panneerselvam et al. 2011; Berger et al. 2017; Mahmoodi & Abdi 2019). The band at 3,403 cm−1 is due to surface OH groups of adsorbent (Zolgharnein et al. 2016a, 2020; Zolgharnein & Feshki 2017). In Figure 2(b), the banding peaks at 1,605 cm−1 are related to the C=C macrocyclic ring deformation. The bending of C=N stretching is observed at 1,521 cm−1. The presence of C–H in-plane bending is indicated by the peak at 1,120 cm−1. The band at 1,030 cm−1 attributed to the C‒N stretching modes in pyrrole vibration is also present in the FT-IR spectrum (Verma et al. 2008). The 633 and 599 cm−1 bands in the spectrum of Figure 2(b) which are due to Co–N bonds in the CoPc chemical structure show the presence of bonded Co(II) in this compound. The distortion and deformation of M–O and OH bands in the pristine spectrum in comparison with the Figure 2(c) spectrum clearly exhibit the involving of M–O (M; Co, Fe) and OH groups of the adsorbent surface with CoPc, and the changes of fingerprint region of these two spectra also confirm the loading of CoPc onto CoFe2O4 NPs (Zolgharnein et al. 2016a; Berger et al. 2017; Zolgharnein & Feshki 2017).

Figure 2

FT-IR spectra of (a) pristine CoFe2O4 NPs, (b) cobalt (II) phthalocyanine dye (CoPc), and (c) loaded CoFe2O4 NPs.

Figure 2

FT-IR spectra of (a) pristine CoFe2O4 NPs, (b) cobalt (II) phthalocyanine dye (CoPc), and (c) loaded CoFe2O4 NPs.

The morphology of CoFe2O4 NPs was studied by the field emission scanning electron microscopy (FE-SEM) of TESCAN (MIRA3). Figure 3(a) and 3(b) illustrate images of these particles with different sizes of 28.7–36.89 nm (Panneerselvam et al. 2011; Ghaemi et al. 2014). CoFe2O4 NPs elemental analysis was performed by EDX analysis. The presence of iron, oxygen, and cobalt can be observed clearly in the EDX spectrum (Figure 3(c)). Also, the wide-angle XRD patterns of CoFe2O4 NPs obtained by the co-precipitation method at room temperature were recorded across a range of 2θ: 10° to 80° by the PW Philips instrument 3710 (Figure 3(d)) (Zolgharnein et al. 2016a; Berger et al. 2017; Zolgharnein & Feshki 2017; Mahmoodi & Abdi 2019). These XRD patterns correspond to the inverse spinel crystalline structure with cubic symmetry (space group Fd3 m), matching with JCPDS 79-0418 and JCPDS 22-1086, respectively (Berger et al. 2017).

Figure 3

(a), (b) FE-SEM-evaluated morphology, (c) EDX analysis, and (d) X-ray diffraction pattern of the CoFe2O4 NPs.

Figure 3

(a), (b) FE-SEM-evaluated morphology, (c) EDX analysis, and (d) X-ray diffraction pattern of the CoFe2O4 NPs.

Figure 4 shows the hysteresis loop of the superparamagnetic behavior of nano-CoFe2O4 adsorbent. The saturation magnetization value of CoFe2O4 NPs at high fields from −10,000 to +10,000 Oe is about 45 emu g−1. So, this superparamagnetic property of CoFe2O4 NPs causes their facile separation from the sample matrix by an external magnet (Chinnasamy et al. 2002; Bagtash et al. 2016; Zolgharnein & Feshki 2017).

Figure 4

VSM magnetization curves for CoFe2O4 NPs.

Figure 4

VSM magnetization curves for CoFe2O4 NPs.

Figure 5

Normal probability plot of residuals for CoPc removal efficiency.

Figure 5

Normal probability plot of residuals for CoPc removal efficiency.

Box–Behnken design as optimization approach

As mentioned in the previous section, to get the maximum amount of removal percentage (R%), the effective factors such as solution pH, adsorbent mass, primary dye concentration, and contact time were evaluated using the Box–Behnken design approach (BBD). In this design, for k = 4 (effective factors), Cp = 3 (replicate of center point), the total required runs N are 27 experiments (N = 2k(K − 1) + Cp) (Kazemi et al. 2010; Zolgharnein et al. 2013). The appropriate ranges for the CoPc adsorption factors are illustrated in Table 1. The matrix design obtained for running BBD which includes experimental response (R%) is shown in Table 2. The multiple linear regression analysis (MLR) of the obtained BBD matrix design leads to the following second-order polynomial model in terms of coded factors for CoPc removal percentage (Table 3) (Ferreira et al. 2007a, 2007b; Zolgharnein et al. 2014):
formula
(4)
Table 1

Factors and their levels used in the Box–Behnken design

Independent variable unitsCoded levels
−10+ 1
pH (A) ‒ 3.2 6.1 9.0 
Initial concentration of dye (Cd) (B) mgL−1 30 165 300 
Adsorbent mass (m) (C) mg 11 20 
Contact time (t) (D) min 16 
Independent variable unitsCoded levels
−10+ 1
pH (A) ‒ 3.2 6.1 9.0 
Initial concentration of dye (Cd) (B) mgL−1 30 165 300 
Adsorbent mass (m) (C) mg 11 20 
Contact time (t) (D) min 16 
Table 2

Box–Behnken design matrix and obtained response

Run orderpHmCdtR%
3.2 (−1) 2 (−1) 165 (0) 9 (0) 69.0 
9 (+1) 2 (−1) 165 (0) 9 (0) 0.0 
3.2 (−1) 20 (+1) 165 (0) 9 (0) 86.0 
9 (+1) 20 (+1) 165 (0) 9 (0) 0.0 
6.1 (0) 11 (0) 30 (−1) 2 (−1) 21.1 
6.1 (0) 11 (0) 300 (+1) 2 (−1) 19.8 
6.1 (0) 11 (0) 30 (−1) 16 (+1) 29.0 
6.1 (0) 11 (0) 300 (+1) 16 (+1) 25.4 
3.2 (−1) 11 (0) 165 (0) 2 (−1) 74.0 
10 9 (+1) 11 (0) 165 (0) 2 (−1) 0.0 
11 3.2 (−1) 11 (0) 165 (0) 16 (+1) 84.5 
12 9 (+1) 11 (0) 165 (0) 16 (+1) 3.0 
13 6.1 (0) 2 (−1) 30 (−1) 9 (0) 18.5 
14 6.1 (0) 20 (+1) 30 (−1) 9 (0) 28.0 
15 6.1 (0) 2 (−1) 300 (+1) 9 (0) 17.0 
16 6.1 (0) 20 (+1) 300 (+1) 9 (0) 24.6 
17 3.2 (−1) 11 (0) 30 (−1) 9 (0) 87.0 
18 9 (+1) 11 (0) 30 (−1) 9 (0) 8.0 
19 3.2 (−1) 11 (0) 300 (+1) 9 (0) 83.0 
20 9 (+1) 11 (0) 300 (+1) 9 (0) 6.5 
21 6.1 (0) 2 (−1) 165 (0) 2 (−1) 8.2 
22 6.1 (0) 20 (+1) 165 (0) 2 (−1) 17.7 
23 6.1 (0) 2 (−1) 165 (0) 16 (+1) 16.0 
24 6.1 (0) 20 (+1) 165 (0) 16 (+1) 23.2 
25 6.1 (0) 11 (0) 165 (0) 9 (0) 29.7 
26 6.1 (0) 11 (0) 165 (0) 9 (0) 29.5 
27 6.1 (0) 11 (0) 165 (0) 9 (0) 29.6 
Run orderpHmCdtR%
3.2 (−1) 2 (−1) 165 (0) 9 (0) 69.0 
9 (+1) 2 (−1) 165 (0) 9 (0) 0.0 
3.2 (−1) 20 (+1) 165 (0) 9 (0) 86.0 
9 (+1) 20 (+1) 165 (0) 9 (0) 0.0 
6.1 (0) 11 (0) 30 (−1) 2 (−1) 21.1 
6.1 (0) 11 (0) 300 (+1) 2 (−1) 19.8 
6.1 (0) 11 (0) 30 (−1) 16 (+1) 29.0 
6.1 (0) 11 (0) 300 (+1) 16 (+1) 25.4 
3.2 (−1) 11 (0) 165 (0) 2 (−1) 74.0 
10 9 (+1) 11 (0) 165 (0) 2 (−1) 0.0 
11 3.2 (−1) 11 (0) 165 (0) 16 (+1) 84.5 
12 9 (+1) 11 (0) 165 (0) 16 (+1) 3.0 
13 6.1 (0) 2 (−1) 30 (−1) 9 (0) 18.5 
14 6.1 (0) 20 (+1) 30 (−1) 9 (0) 28.0 
15 6.1 (0) 2 (−1) 300 (+1) 9 (0) 17.0 
16 6.1 (0) 20 (+1) 300 (+1) 9 (0) 24.6 
17 3.2 (−1) 11 (0) 30 (−1) 9 (0) 87.0 
18 9 (+1) 11 (0) 30 (−1) 9 (0) 8.0 
19 3.2 (−1) 11 (0) 300 (+1) 9 (0) 83.0 
20 9 (+1) 11 (0) 300 (+1) 9 (0) 6.5 
21 6.1 (0) 2 (−1) 165 (0) 2 (−1) 8.2 
22 6.1 (0) 20 (+1) 165 (0) 2 (−1) 17.7 
23 6.1 (0) 2 (−1) 165 (0) 16 (+1) 16.0 
24 6.1 (0) 20 (+1) 165 (0) 16 (+1) 23.2 
25 6.1 (0) 11 (0) 165 (0) 9 (0) 29.7 
26 6.1 (0) 11 (0) 165 (0) 9 (0) 29.5 
27 6.1 (0) 11 (0) 165 (0) 9 (0) 29.6 

Cd, initial concentration of dye; m, adsorbent mass; t, contact time.

Table 3

Estimated effects and regression coefficients and statistical parameters of the suggested model for R% (coded values)

TermCoef.SE coef.TP
Constant 29.6000 0.0728 406.55 0.0001 
pH −38.8333 0.0364 −1,066.74 0.0001 
m 4.2333 0.0364 116.29 0.0001 
Cd −1.2750 0.0364 −35.02 0.0001 
t 3.3583 0.0364 92.25 0.0001 
pH2 16.5958 0.0546 303.92 0.0001 
m2 −7.5042 0.0546 −137.42 0.0001 
t2 −5.7917 0.0546 −106.06 0.0001 
pH × m −4.2500 0.0631 −67.40 0.0001 
pH × Cd 0.6250 0.0631 9.91 0.0001 
pH × t −1.8750 0.0631 −29.74 0.0001 
m × Cd −0.4750 0.0631 −7.53 0.0001 
m × t −0.5750 0.0631 −9.12 0.0001 
Cd × t −0.5750 0.0631 −9.12 0.0001 
TermCoef.SE coef.TP
Constant 29.6000 0.0728 406.55 0.0001 
pH −38.8333 0.0364 −1,066.74 0.0001 
m 4.2333 0.0364 116.29 0.0001 
Cd −1.2750 0.0364 −35.02 0.0001 
t 3.3583 0.0364 92.25 0.0001 
pH2 16.5958 0.0546 303.92 0.0001 
m2 −7.5042 0.0546 −137.42 0.0001 
t2 −5.7917 0.0546 −106.06 0.0001 
pH × m −4.2500 0.0631 −67.40 0.0001 
pH × Cd 0.6250 0.0631 9.91 0.0001 
pH × t −1.8750 0.0631 −29.74 0.0001 
m × Cd −0.4750 0.0631 −7.53 0.0001 
m × t −0.5750 0.0631 −9.12 0.0001 
Cd × t −0.5750 0.0631 −9.12 0.0001 

T = coef./SE coef.; p-value: significance level (p < 0.05).

R2 = 0.9999, R2predi = 0.9988.

The validation and adequacy of the obtained model were investigated by various statistical analysis tests such as analysis of variance (ANOVA), analysis of residuals and coefficient of determination result (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008; Kazemi et al. 2010; Zolgharnein et al. 2013). ANOVA results for the model found for CoPc removal percentage are presented in Table 4. In this Table, the first parameters seen are the F and P values of the model (F = 96415, P<0.05) that indicate the significance of the quadratic model (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008; Kazemi et al. 2010; Zolgharnein et al. 2013, 2014). All linear terms of pH, m (adsorbent mass), t (contact time) and Cd (initial concentration of CoPc) and quadratic terms are significant, except for the interaction term Cd × Cd. Moreover, the lack of fit of the model is not significant (P = 0.425>0.05) which is a good indication of the adequacy of the model (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008). Analysis of residuals according to the normal probability plot (Figure 5) illustrates the residuals (R%experimentalR%model) with a random distribution and justifies the validity of the model. This is also confirmed by the values calculated for coefficient of determination R2 = 0.9999, R2predi = 0.9988 (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008; Kazemi et al. 2010; Chen et al. 2011; Zolgharnein et al. 2013, 2014). As a conclusion here, all three tests prove the model's adequacy and validity.

Table 4

Analysis of variance (ANOVA) for suggested model for adsorption responses R% of CoPc using the CoFe2O4 NPs

SourcedfaSeq SSbContributionAdj SSAdj MSF-valueP-value
Model 14 21,465.7 100.00% 21,465.7 1,533.3 96,415.04 0.000 
 Linear 18,466.2 86.03% 18,466.2 4,616.6 290,298.89 0.000 
  pH 18,096.3 84.30% 18,096.3 18,096.3 1,137,935.37 0.000 
  m 215.1 1.00% 215.1 215.1 13,523.00 0.000 
  Cd 19.5 0.09% 19.5 19.5 1,226.67 0.000 
  t 135.3 0.63% 135.3 135.3 8,510.52 0.000 
 Square 2,908.1 13.55% 2,908.1 727.0 45,716.53 0.000 
  pH × pH 2,473.8 11.52% 1,468.9 1,468.9 92,368.49 0.000 
  m × m 233.9 1.09% 300.3 300.3 18,885.60 0.000 
  Cd × Cd 21.4 0.10% 0.0 0.0 0.58 0.460 
  t × t 178.9 0.83% 178.9 178.9 11,249.49 0.000 
 2-Way Interaction 91.4 0.43% 91.4 15.2 958.14 0.000 
  pH × m 72.2 0.34% 72.3 72.3 4,543.23 0.000 
  pH × Cd 1.6 0.01% 1.6 1.6 98.25 0.000 
  pH × t 14.1 0.07% 14.1 14.1 884.28 0.000 
  m × Cd 0.9 0.00% 0.9 0.9 56.75 0.000 
  m × t 1.3 0.01% 1.3 1.3 83.16 0.000 
  Cd × t 1.3 0.01% 1.3 1.3 83.16 0.000 
Error 12 0.2 0.00% 0.2 0.0   
 Lack-of-fit 10 0.2 0.00% 0.2 0.0 1.71 0.425* 
 Pure error 0.0 0.00% 0.0 0.0   
Total 26 21,465.9 100.00%     
SourcedfaSeq SSbContributionAdj SSAdj MSF-valueP-value
Model 14 21,465.7 100.00% 21,465.7 1,533.3 96,415.04 0.000 
 Linear 18,466.2 86.03% 18,466.2 4,616.6 290,298.89 0.000 
  pH 18,096.3 84.30% 18,096.3 18,096.3 1,137,935.37 0.000 
  m 215.1 1.00% 215.1 215.1 13,523.00 0.000 
  Cd 19.5 0.09% 19.5 19.5 1,226.67 0.000 
  t 135.3 0.63% 135.3 135.3 8,510.52 0.000 
 Square 2,908.1 13.55% 2,908.1 727.0 45,716.53 0.000 
  pH × pH 2,473.8 11.52% 1,468.9 1,468.9 92,368.49 0.000 
  m × m 233.9 1.09% 300.3 300.3 18,885.60 0.000 
  Cd × Cd 21.4 0.10% 0.0 0.0 0.58 0.460 
  t × t 178.9 0.83% 178.9 178.9 11,249.49 0.000 
 2-Way Interaction 91.4 0.43% 91.4 15.2 958.14 0.000 
  pH × m 72.2 0.34% 72.3 72.3 4,543.23 0.000 
  pH × Cd 1.6 0.01% 1.6 1.6 98.25 0.000 
  pH × t 14.1 0.07% 14.1 14.1 884.28 0.000 
  m × Cd 0.9 0.00% 0.9 0.9 56.75 0.000 
  m × t 1.3 0.01% 1.3 1.3 83.16 0.000 
  Cd × t 1.3 0.01% 1.3 1.3 83.16 0.000 
Error 12 0.2 0.00% 0.2 0.0   
 Lack-of-fit 10 0.2 0.00% 0.2 0.0 1.71 0.425* 
 Pure error 0.0 0.00% 0.0 0.0   
Total 26 21,465.9 100.00%     

aDegree of freedom. *Not significant (p>0.05).

bSum of square.

Figure 6

Response surface plots showing effective factors and their mutual effects on removal percentage (R%) of CoPc while the other variables are at middle level: (a) the effect of pH and amount of adsorbent (m); (b) the effect of pH and dye concentration amount (Cd) ; (C) the effect of pH and contact time (t); (d) the effect of amount of adsorbent (m) and dye concentration amount (Cd); (e) dye concentration amount (Cd) and contact time (t); (f) the effect of contact time (t) and amount of adsorbent (m).

Figure 6

Response surface plots showing effective factors and their mutual effects on removal percentage (R%) of CoPc while the other variables are at middle level: (a) the effect of pH and amount of adsorbent (m); (b) the effect of pH and dye concentration amount (Cd) ; (C) the effect of pH and contact time (t); (d) the effect of amount of adsorbent (m) and dye concentration amount (Cd); (e) dye concentration amount (Cd) and contact time (t); (f) the effect of contact time (t) and amount of adsorbent (m).

The 3D response surface plots (Figure 6) visualize the relation of removal percentage R% with two effective factors (pH, m, t, Cd) while the others are kept constant at the middle level. These plots show the geometry of the response area of the adsorption process and all factors' interactions, especially pH with m, Cd and t. It is seen that CoPc removal performance could be strongly decreased for pH rising above 3.2. By proper running of the Desirability function option in MINITAB software, the optimum conditions were found as pH 3.2, adsorbent mass (m) 11 mg, contact time of nine minutes (t), and initial dye concentration (Cd) of 30 mg L−1 (Ferreira et al. 2007a, 2007b; Bezerra et al. 2008; Kazemi et al. 2010; Zolgharnein et al. 2013, 2014). These are in good concordance with primary optimized results before running the experimental design runs.

Isotherm studies

Isotherm models for the adsorption process of CoPc onto CoFe2O4 NPs can be used to investigate their equilibrium behavior and find the uptake capacity of the adsorbent. So, the Langmuir, Freundlich and Dubinin—Radushkevich (D–R) isotherms were applied to describe the equilibrium performance of the adsorption process and their results are illustrated in Figure 7(a)–7(c) (Langmuir 1918; Dubinin & Radushkevich 1947; Argun et al. 2007; Liu & Liu 2008; Foo & Hameed 2010; Chen et al. 2011; Rangabhashiyam et al. 2014; Zolgharnein et al. 2014; Rastgordani et al. 2020). Their equations and the parameter values are introduced and listed in Table 5. Two statistical tests, coefficient of determination (R2) and chi-square () are applied for evaluating the goodness of fit of these isotherms with experimental data (Zolgharnein et al. 2016a, 2020; Zolgharnein & Feshki 2017; Rastgordani et al. 2020). This study begins with the Langmuir isotherm, which is based on homogeneous sites of adsorbent and monolayer within adsorption (Langmuir 1918; Liu & Liu 2008; Foo & Hameed 2010). The maximum uptake capacity (qmax) and b values (Table 5) are found from both plots of Ce/qe versus Ce (Figure 7(a)) and nonlinear regression analysis (Figure 7(d)) (Argun et al. 2007; Chen et al. 2011; Rangabhashiyam et al. 2014). The estimated values for qmax and b from both methods are equal to 431.03 and 434.33 mg·g−1, 0.072 and 0.076 Lmg−1, respectively. In the second step, the Freundlich isotherm as an experimental model is considered (Argun et al. 2007; Liu & Liu 2008; Foo & Hameed 2010; Chen et al. 2011; Rangabhashiyam et al. 2014; Zolgharnein et al. 2014). The Freundlich isotherm is applied for adsorption onto heterogeneous surfaces with a non-uniform energy distribution of adsorption CoPc onto the target surface leading to multilayer adsorption (Langmuir 1918; Argun et al. 2007; Liu & Liu 2008; Verma et al. 2008; Foo & Hameed 2010; Absalan et al. 2011; Chen et al. 2011; Rangabhashiyam et al. 2014; Zolgharnein et al. 2014, 2016a; Zolgharnein & Feshki 2017). Table 5 indicates its equation along with parameters such as KF and 1/nF found from the linear plot and nonlinear regression. The values of KF and 1/nF are 44.89, 73.90 ((mg·g−1)(Lmg−1)1/n), 0.493 and 0.357, respectively (Argun et al. 2007; Liu & Liu 2008; Foo & Hameed 2010; Absalan et al. 2011; Chen et al. 2011; Rangabhashiyam et al. 2014; Zolgharnein et al. 2016a; Zolgharnein & Feshki 2017).

Table 5

The equations and parameters of various Isotherms considered for the adsorption of CoPc by CoFe2O4 NPs

Parameters
IsothermsFormEquation (mgg−1)  (Lmg−1)
Langmuir Linear  431 0.072 0.9981 – 
Non-linear  434 0.076 0.9908 9.673 
((mgg−1) (Lmg−1)1/n)
1/nF
Freundlich Linear 1/nF 44.89 0.493 0.9135 – 
 Non-linear  73.90 0.357 0.9057 96.86 
(mgg−1)
β (mol²(kJ²)−1)
E (kJmol−1)
D-R Linear  259.4 4.0 × 10−9 11.32 0.9398 – 
Parameters
IsothermsFormEquation (mgg−1)  (Lmg−1)
Langmuir Linear  431 0.072 0.9981 – 
Non-linear  434 0.076 0.9908 9.673 
((mgg−1) (Lmg−1)1/n)
1/nF
Freundlich Linear 1/nF 44.89 0.493 0.9135 – 
 Non-linear  73.90 0.357 0.9057 96.86 
(mgg−1)
β (mol²(kJ²)−1)
E (kJmol−1)
D-R Linear  259.4 4.0 × 10−9 11.32 0.9398 – 

qmax: maximum monolayer qe; b: Langmuir constant; KF: Freundlich adsorption constant; 1/nF is the heterogeneity factor; E: mean free energy; = ; R: gas constant (8.314 J (mol K)−1); T: absolute temperature in Kelvin; Ce: equilibrium concentration (mol·L−1).

Figure 7

(a) Linear Langmuir, (b) linear Freundlich and (c) D-R isotherm, and (d) both nonlinear Langmuir and Freundlich isotherms for adsorption of cobalt (II) phthalocyanine by CoFe2O4 NPs (conditions: 30 mL solutions containing 10 mg CoFe2O4 NPs adsorbent, pH = 3.0 and 10-minute contact times with different initial concentrations (10–300 mg. L−1)).

Figure 7

(a) Linear Langmuir, (b) linear Freundlich and (c) D-R isotherm, and (d) both nonlinear Langmuir and Freundlich isotherms for adsorption of cobalt (II) phthalocyanine by CoFe2O4 NPs (conditions: 30 mL solutions containing 10 mg CoFe2O4 NPs adsorbent, pH = 3.0 and 10-minute contact times with different initial concentrations (10–300 mg. L−1)).

The Dubinin–Radushkevich isotherm (D-R) is also investigated to estimate the porosity characteristics and apparent free energy of adsorption (Dubinin & Radushkevich 1947; Argun et al. 2007; Liu & Liu 2008; Foo & Hameed 2010; Absalan et al. 2011; Chen et al. 2011; Rangabhashiyam et al. 2014; Zolgharnein et al. 2016a; Zolgharnein & Feshki 2017). The D-R equation is given in Table 5 and its components are given in the footnote; qmax (mg g−1) and β (mol2(J2)−1) are obtained from the plot of ln qe versus ɛ2, and β is a coefficient related to the energy of the adsorption and it may be an estimate of the adsorption mechanism (Table 5 and Figure 7(c)) (Langmuir 1918; Dubinin & Radushkevich 1947; Argun et al. 2007; Liu & Liu 2008; Foo & Hameed 2010; Chen et al. 2011; Rangabhashiyam et al. 2014; Rastgordani et al. 2020; Zolgharnein et al. 2020). The mean free energy () states information about adsorption mechanisms whether the interactions are physical or chemical (Langmuir 1918; Dubinin & Radushkevich 1947; Argun et al. 2007; Liu & Liu 2008; Foo & Hameed 2010; Chen et al. 2011; Rangabhashiyam et al. 2014; Zolgharnein et al. 2020). According to the literature, if E < 8 kJ·mol−1, the predominant adsorption process is physical and if E is between 8 and 16 kJ·mol−1, a chemical ion-exchange takes place, and if E > 16 kJ·mol−1, chemisorption is the dominant interaction (Langmuir 1918; Dubinin & Radushkevich 1947; Argun et al. 2007; Liu & Liu 2008; Foo & Hameed 2010; Chen et al. 2011; Rangabhashiyam et al. 2014; Zolgharnein et al. 2020). In this work, E was found to be 11.32 kJ·mol−1, which indicates that one of the probable mechanisms for the adsorption of CoPc onto nano-CoFe2O4 may be a chemical ion-exchange (Langmuir 1918; Dubinin & Radushkevich 1947; Argun et al. 2007; Liu & Liu 2008; Foo & Hameed 2010; Absalan et al. 2011; Chen et al. 2011; Rangabhashiyam et al. 2014; Zolgharnein et al. 2016a; Zolgharnein & Feshki 2017). According to the results listed in Table 5, comparing R2 (0.9981, 0.9135) and (9.673, 96.86) values of the examined isotherms, the Langmuir isotherm successfully describes the equilibrium behavior of CoPc by CoFe2O4 (Dubinin & Radushkevich 1947; Zolgharnein et al. 2020).

Kinetic studies

Kinetic studies provide valuable information about the rate and mechanism of the adsorption process. The type of kinetic model and adsorption mechanism depends on various factors such as the physical and chemical properties of adsorbents, as well as the mass transfer process (Dubinin & Radushkevich 1947; Argun et al. 2007; Liu & Liu 2008; Foo & Hameed 2010; Absalan et al. 2011; Chen et al. 2011; Zolgharnein et al. 2016a; Zolgharnein & Feshki 2017). In this study, the adsorption kinetics of CoPc dye onto nano-CoFe2O4 was examined by two kinetic models of pseudo-first order and pseudo-second order, at the previously obtained optimum conditions (Figure 8). Their equations and the related parameter values are listed in Table 6 (Dubinin & Radushkevich 1947; Zolgharnein et al. 2020). In the pseudo-first-order and pseudo-second-order models the qe, k1 and k2 values were calculated from the plots of ln (qeqt) vs t and t/qt vs t and their values for initial concentration of dye equal to 100 mg·L −1 were found to be qe, = 129, qexp = 222.22 mg·g−1, k1 = 0.8309, k2 = 0.0405min−1, respectively (Figure 8(a) and 8(b)) (Dubinin & Radushkevich 1947; Absalan et al. 2011; Zolgharnein et al. 2014, 2016a, 2020; Zolgharnein & Feshki 2017). The comparison of the coefficient of determination for both models (R12 = 0.7593 and R22 = 0.9999 respectively) indicates that the pseudo-second-order model is perfectly fitted with the experimental data (clearly shown in Figure 8(b)). Further consideration shows experimental uptake capacity (qe,expe = 221.46 mg·g−1) is very close to qe (222.22 mg g−1) obtained by pseudo-second-order model which confirms the predominant behavior of the pseudo-second-order kinetic model on the adsorption process of CoPc NPs (Dubinin & Radushkevich 1947; Zolgharnein et al. 2020). It is interesting to see that in pseudo-second-order with increasing the initial CoPc concentration (20, 100, 150 mg L−1), the rate constant k2 decreases (0.1159, 0.0605, 0.0300 g mg−1min−1), which is reasonable and may be due to some interaction such as ion exchange, chelation, and physical adsorption (Dubinin & Radushkevich 1947; Ho & McKay 1998; Zolgharnein et al. 2016b, 2020). Moreover, the rate and diffusion manner of CoPc adsorption onto nano-CoFe2O4 is considered with the intra-particle diffusion model (Zolgharnein et al. 2016b, 2020; Rastgordani et al. 2020). This model is based on Weber and Morris theory () in which the plots of qt versus t1/2 are nonlinear throughout the adsorption and usually divided into two linear segments (Liu & Liu 2008; Foo & Hameed 2010; Zolgharnein et al. 2016b; Rastgordani et al. 2020). The diffusion of dye onto the adsorbent exhibits three steps: the first step is the mass transfer of the dye from the bulk of the solution onto the surface of the adsorbent, which is a fast stage and not recognizable (Boudechiche et al. 2017). The linear plots are due to the second and third steps (Figure 9). The second step is internal diffusion of dye into the cell wall and then reaching equilibrium in the pore wall of the adsorbent, which is illustrated by the second segment of the linear plots in Figure 9 (Liu & Liu 2008; Foo & Hameed 2010; Zolgharnein et al. 2016b, 2020; Rastgordani et al. 2020). Figure 9 indicates that the first linear segment of the plots does not pass through the origin. Weber and Morris's theory shows that if the plot of qt versus t1/2 is linear and passes through the origin, then intra-particle diffusion is the only rate-limiting process (Liu & Liu 2008; Foo & Hameed 2010; Zolgharnein et al. 2016b, 2020; Rastgordani et al. 2020). Increasing the intra-particle diffusion rate constant kid with increasing initial dye concentration is due to the rise of the driving force of the diffusion pattern and means that the rate of dye reaching the sites of adsorbent increases (for three initial dye concentrations (20, 100, 150 mgL−1) with intercept C boundary layer thickness (C = 34.113, 189.33, 294.12) and kid (4.7641, 12.069, 15.587 mgg−1 min−0.5)). So, the intra-particle diffusion is not the only rate-limiting step in the kinetic process of the adsorption of cobalt (II) phthalocyanine (Figure 9) (Liu & Liu 2008; Foo & Hameed 2010; Zolgharnein et al. 2016b; Rastgordani et al. 2020). All kinetic results are given in Table 6. In overall and as a conclusion of the isotherm and kinetic results, it seems that the type of interaction involved in adsorption of CoPc with CoFe2O3 NPs may be a chemisorption process (Argun et al. 2007; Absalan et al. 2011; Chen et al. 2011; Rangabhashiyam et al. 2014; Zolgharnein et al. 2016a; Zolgharnein & Feshki 2017).

Table 6

The equations and parameters of kinetic models of pseudo-first order, pseudo-second order, and intra-particle diffusion of CoPc by CoFe2O4 NPs

Kinetic modellinear equation/plotParameterCoPc concentration (mgL−1)
20100150
Pseudo-first order  vs t  (min−10.7036 0.8309 0.7671 
(mgg−128.06 129.05 152.12 
(mgg−144.41 221.46 334.78 
 0.7806 0.7593 0.7835 
Pseudo-second order  vs t  (gmg−1 min−10.1159 0.0405 0.0300 
(mgg−145.87 222.22 333.33 
(mgg−144.41 221.46 334.78 
 0.9999 0.9999 0.9999 
Intra-particle diffusion  vs   (mgg−1 min−0.55.363 12.069 15.587 
C 33.45 189.33 294.12 
 0.9595 0.9942 0.9710 
Kinetic modellinear equation/plotParameterCoPc concentration (mgL−1)
20100150
Pseudo-first order  vs t  (min−10.7036 0.8309 0.7671 
(mgg−128.06 129.05 152.12 
(mgg−144.41 221.46 334.78 
 0.7806 0.7593 0.7835 
Pseudo-second order  vs t  (gmg−1 min−10.1159 0.0405 0.0300 
(mgg−145.87 222.22 333.33 
(mgg−144.41 221.46 334.78 
 0.9999 0.9999 0.9999 
Intra-particle diffusion  vs   (mgg−1 min−0.55.363 12.069 15.587 
C 33.45 189.33 294.12 
 0.9595 0.9942 0.9710 

: pseudo-first-order rate constant; : pseudo-second-order rate constant; t: contact time; kid: intra-particle diffusion rate constant; C: boundary layer thickness: .

Figure 8

(a) Pseudo-first-order and (b) pseudo-second-order kinetics plots for the adsorption process (conditions: pH = 3, m = 0.04 g, Cd: 20, 100, and 150 mgL−1).

Figure 8

(a) Pseudo-first-order and (b) pseudo-second-order kinetics plots for the adsorption process (conditions: pH = 3, m = 0.04 g, Cd: 20, 100, and 150 mgL−1).

Figure 9

Intra-particle diffusion plot of cobalt (II) phthalocyanine adsorption by CoFe2O4 NPs (conditions: pH = 3, m = 0.04 g, Cd: 20, 100, and 150 mgL−1).

Figure 9

Intra-particle diffusion plot of cobalt (II) phthalocyanine adsorption by CoFe2O4 NPs (conditions: pH = 3, m = 0.04 g, Cd: 20, 100, and 150 mgL−1).

Desorption and regeneration studies

Desorption and reusability are important characteristics and advantages of an adsorbent. According to the nature of the organic dyes, proper solvents should be chosen. Organic solvents and their mixture along with salt solution are proper for investigation of dye desorption. This helps the dye to be released from the surface of the adsorbent. In this work, an aliquot (40 mg L−1) of CoPc dye solution at optimum conditions was treated with several organic solvents accompanied by NaCl solution.

All the following cases are examined: (a) methanol, (b) a mixture of methanol and acetic acid (1:1 ratio), (c) 1 M of NaCl Solution, (d) a mixture of acetone and NaCl (1 M) solution (1:1 ratio). The 2 mL of the latter solution was the best eluent and desorbs 95% of CoPc from the surface of CoFe2O4 NPs in a short time. The consequence of effective desorption of CoPc dye is the successful regeneration of CoFe2O4 NPs. It was used five times without significant reduction in the removal percentage of dye (Zolgharnein & Feshki 2017).

Finally, a comparison of the uptake capacity (qmax) of CoPc by other adsorbents is shown in Table 7. As mentioned in the introduction, reports of the removal of CoPc were not found (Forgacs et al. 2004; Aksu 2005; Crini 2006; Gupta & Suhas 2009), so this comparison was made with CuPc. These results showed that the CoFe2O3 NPs pose excellent potential (431 mg g−1) for removal of CoPc dye from an aqueous solution (Anbia & Mohammadi 2008; Kaušpėdienė et al. 2013; Boudechiche et al. 2017).

Table 7

Maximum uptake capacity (qmax) of phetalocyanine dyes with some different adsorbents

AdsorbentqmaxReference
Nano-CoFe2O4 431 mgg−1 Present study 
MCM-48a 300.5 mgg−1 Anbia & Mohammadi (2008)  
MCM-41b 285.5 mgg−1 Anbia & Mohammadi (2008)  
Crataegus azarolusc 24.02 mgg−1 Boudechiche et al. (2017)  
Rolite AQ 500PSd 86.9 μmolg−1 Kaušpėdienė et al. (2013)  
Macronet MN 200f 2.3 μmolg−1 Kaušpėdienė et al. (2013)  
AdsorbentqmaxReference
Nano-CoFe2O4 431 mgg−1 Present study 
MCM-48a 300.5 mgg−1 Anbia & Mohammadi (2008)  
MCM-41b 285.5 mgg−1 Anbia & Mohammadi (2008)  
Crataegus azarolusc 24.02 mgg−1 Boudechiche et al. (2017)  
Rolite AQ 500PSd 86.9 μmolg−1 Kaušpėdienė et al. (2013)  
Macronet MN 200f 2.3 μmolg−1 Kaušpėdienė et al. (2013)  

a,bRemoval of copper-phthalocyanine from aqueous solution by cationically templated MCM-41 and MCM-48 nanoporous adsorbents.

cValorization of Crataegus azarolus stones for the removal of Direct Blue 86 (copper-phthalocyanine) as a textile anionic dye.

d,fRemoval of the copper-phthalocyanine dye from acidic solutions using resins with the polystyrene divinylbenzene matrix.

The synthesis of CoFe2O4 NPs was performed by a chemical co-precipitation procedure in alkaline media. The prepared cobalt ferrite nanocomposite was characterized by FT-IR, FE-SEM, EDX, XRD and VSM analyses. The effects of pH, adsorbent dose, contact time and initial dye concentration on the CoPc removal percentage were investigated using BBD. A dye removal percentage of 87% was obtained under the optimum condition for PC removal. The pseudo-second-order model displayed the highest agreement with the experimental data. Examination of Langmuir, Freundlich and Dubinin–Radushkevich isotherm models indicates that the adsorption procedure corresponds better with the Langmuir model, and the maximum adsorption capacity is 431 mg g−1. For adsorption of cobalt (II) phthalocyanine from aqueous solutions, the produced magnetic nanoparticles can be used as an appropriate adsorbent.

The authors gratefully appreciate Professor Saeid Amani for his valuable guides.

All relevant data are included in the paper or its Supplementary Information.

Alalwan
H. A.
Kadhom
M. A.
Alminshid
A. H.
2020
Removal of heavy metals from wastewater using agricultural byproducts
.
J. Water Supply: Res. Technol.
69
(
2
),
99
112
.
Ambashta
R. D.
Sillanpää
M.
2010
Water purification using magnetic assistance : a review
.
J. Hazard. Mater.
180
,
38
49
.
Argun
M. E.
Dursun
S.
Ozdemir
C.
Karatas
M.
2007
Heavy metal adsorption by modified oak sawdust: thermodynamics and kinetics
.
J. Hazard. Mater.
141
,
77
85
.
Bagtash
M.
Yamini
Y.
Tahmasebi
E.
Zolgharnein
J.
Dalirnasab
Z.
2016
Magnetite nanoparticles coated with tannic acid as a viable sorbent for solid-phase extraction of Cd2+, Co2+ and Cr3+
.
Microchim. Acta.
183
(
1
),
449
456
.
Berger
D.
Georgescu
D.
Bajenaru
L.
Zanfir
A.
Stănică
N.
Matei
C.
2017
Properties of mesostructured silica coated CoFe2O4 versus Fe3O4-silica composites
.
J. Alloys Compd.
708
,
278
284
.
Bezerra
M. A.
Santelli
R. E.
Oliveira
E. P.
Villa
L. S.
Escaleira
L. A.
2008
Response surface methodology (RSM) as a tool for optimization in analytical chemistry
.
Talanta
76
,
965
977
.
Chinnasamy
C. N.
Jeyadevan
B.
Perales-Perez
O.
Shinoda
K.
Tohji
K.
Kasuya
A.
2002
Growth dominant co-precipitation process to achieve high coercivity at room temperature in CoFe2O4 nanoparticles
.
IEEE Trans. Magn.
38
,
2640
2642
.
Dubinin
M. M.
Radushkevich
L. V.
1947
The equation of the characteristic curve of activated charcoal
.
Proc. Acad. Sci. Phys. Chem. Sect.
55
,
331
.
Ferreira
S. L. C.
Bruns
R. E.
da Silva
E. G. P.
dos Santos
W. N. L.
Quintella
C. M.
David
J. M.
de Andrade
J. B.
Breitkreitz
M. C.
Jardim
I. C. S. F.
Neto
B. B.
2007a
Statistical designs and response surface techniques for the optimization of chromatographic systems
.
J. Chromatogr. A
1158
,
2
14
.
Ferreira
S. L. C.
Bruns
R. E.
Ferreira
H. S.
Matos
G. D.
David
J. M.
Brandão
G. C.
da Silva
E. G. P.
Portugal
L. A.
dos Reis
P. S.
Souza
A. S.
dos Santos
W. N. L.
2007b
Box–Behnken design: an alternative for the optimization of analytical methods
.
Analyt. Chim. Acta.
597
(
2
),
179
186
.
Forgacs
E.
Cserháti
T.
Oros
G.
2004
Removal of synthetic dyes from wastewaters: a review
.
Environ. Int.
30
,
953
971
.
Ghaemi
M.
Absalan
G.
Sheikhian
L.
2014
Adsorption characteristics of Titan yellow and Congo red on CoFe2O4 magnetic nanoparticles
.
J. Iran Chem. Soc.
11
(
6
),
1759
1766
.
Gupta
V. K.
Suhas
2009
Application of low-cost adsorbents for dye removal – a review
.
J. Environ. Manage.
90
(
8
),
2313
2342
.
Gupta
V. K.
Ali
I.
Saleh
T. A.
Nayak
A.
Agarwal
S.
2012
Chemical treatment technologies for waste-water recycling – an overview
.
RSC Adv.
2
,
6380
6388
.
Ho
Y. S.
McKay
G.
1998
Sorption of dye from aqueous solution by peat
.
Chem. Eng. J.
70
,
115
124
.
Kaušpėdienė
D.
Gefenienė
A.
Kazlauskienė
E.
Ragauskas
R.
Selskienė
A.
2013
Simultaneous removal of azo and phthalocyanine dyes from aqueous solutions using weak base anion exchange resin
.
Water Air Soil Pollut.
224
,
1769
.
Kyzas
G. Z.
Matis
K. A.
2015
Nanoadsorbents for pollutants removal: a review
.
J. Mol. Liq.
203
,
159
168
.
Liu
Y.
Liu
Y.
2008
Biosorption isotherms, kinetics and thermodynamics
.
Sep. Purif. Technol.
61
,
229
242
.
Mahmoodi
N. M.
Abdi
J.
2019
Surface modified cobalt ferrite nanoparticles with cationic surfactant: synthesis, multicomponent dye removal modeling and selectivity analysis
.
Prog. Color Color. Coat.
12
,
163
177
.
Matthews
R. D.
Bottomley
L. A.
Pavlostathis
S. G.
2009
Palladium-catalyzed hydrogen reduction and decolorization of reactive phthalocyanine dyes
.
Desalination
248
,
816
825
.
Maya
E. M.
de la Torre
G.
Lozano
A. E.
Torres
T.
de la Campa
J. G.
de Abajo
J.
2006
Novel cobalt (II) phthalocyanine-containing polyimides: synthesis, characterization, thermal and optical properties
.
Macromol. Rapid Commun.
27
(
21
),
1852
1858
.
Moser
F. H.
Thomas
A. L.
1964
Phthalocyanine compounds
.
J. Chem. Educ.
41
(
5
),
245
249
.
Rangabhashiyam
S.
Anu
N.
Giri Nandagopal
M. S.
Selvaraju
N.
2014
Relevance of isotherm models in biosorption of pollutants by agricultural byproducts
.
J. Environ. Chem. Eng.
2
(
1
),
398
414
.
Silva
M. C.
Corrêa
A. D.
Amorim
M. T. S. P.
Parpot
P.
Torres
J. A.
Chagas
P. M. B.
2012
Decolorization of the phthalocyanine dye reactive blue 21 by turnip peroxidase and assessment of its oxidation products
.
J. Mol. Catal. B Enzym.
77
,
9
14
.
Verma
D.
Dash
R.
Katti
K. S.
Schulz
D. L.
Caruso
A. N.
2008
Role of coordinated metal ions on the orientation of phthalocyanine based coatings
.
Spectrochim. Acta Part A
70
,
1180
1186
.
Zhao
L.
Li
X.
Zhao
Q.
Qu
Z.
Yuan
D.
Liu
S.
Hu
X.
Chen
G.
2010
Synthesis, characterization and adsorptive performance of MgFe2O4 nanospheres for SO2 removal
.
J. Hazard. Mater.
184
(
1–3
),
704
709
.
Zolgharnein
J.
Choghaei
Z.
Bagtash
M.
Feshki
S.
Rastgordani
M.
Zolgharnein
P.
2016a
Nano-Fe3O4 and corn cover composite for removal of Alizarin Red S from aqueous solution: characterization and optimization investigations
.
Desalin. Water Treat.
57
,
27672
27685
.
Zolgharnein
J.
Shahmoradi
A.
Zolgharnein
P.
Amani
S.
2016b
Multivariate optimization and adsorption characterization of As(III) by using Fraxinus tree leaves
.
Chem. Eng. Commun.
203
,
210
223
.