1 Introduction

Water quality, throughout history, has been an essential factor in ensuring human well-being. Nowadays, one of the most severe environmental problems is the pollution of the hydrosphere, which is to a great extent, threatened by human activity (Streimikiene 2015). The quality of this vital element is deteriorating due to industrial wastes, especially by heavy metals, which are a characteristic feature of these industries and become more and more worrying. The protection of the environment requires the early elimination of these metals, considering the levels allowed by international chemical standards discharged into the wild (Abdellaoui et al. 2019).

The carcinogenic and teratogenic properties on humans of most heavy metals have considered them a severe danger to human beings (Elouahli et al. 2018; Wu et al. 2008). However, chromium can be released into the environment by various industries, including the metal finishing industry, the iron and steel industry, and the production of inorganic chemicals.

The toxicity of chromium depends not only on its concentration but also on its degree of oxidation. Indeed, it is commonly accepted that chromium (VI) is much more toxic than chromium (III). Chromium (VI) ions are a kind of toxic inorganic pollutant that poses a severe problem to ecosystems, whether its effluents are not appropriately treated (Huang et al. 2013).

Among the developed viable technologies for the removal of heavy metals from wastewater effluents is the sorption process as a promising and highly effective method (Cimá-Mukul et al. 2019). Unfortunately, most conventional adsorption systems are broadly applied the activated carbon to remove a variety of contaminants despite its disadvantages relating to the cost of production, separation difficulties, and regeneration (Abatal et al. 2018). Therefore, the researches towards treatment processes using less costly and widely available natural materials are encouraged. Indeed, the performance and effectiveness of this technique depend in a preponderant way of the support nature used as an adsorbent, its cost, abundance, and regeneration.

Recently, research has focused on the study of chitinous products and their derivatives, especially chitosan, owing to its intrinsic properties, abundance, and biodegradability that constitute significant assets for its use for a respective environmental clean-up (Pal and Pal 2019; Sessarego et al. 2019). Moreover, natural minerals showed to be a promising sorbent of heavy metals owing to their excellent environmental compatibility and removal capability, for instance, clay minerals and iron ore slimes (Panda et al. 2011; Uddin 2017). Among these minerals, phosphate rocks exist widely in nature under several mineralogical classes. However, the apatite is considered as the most abundant phosphate on earth (Hughes and Rakovan 2002). Fluorapatite (Fa) with the formula [Ca5(PO4)3F] is one of the most common apatite phases in nature. Natural phosphate showed great potential to be one of the efficient adsorbents (Aklil et al. 2004). Previous works have evaluated the maximum sorption capacity (qm) of natural phosphate of divalent heavy metals; the qm was to be 89.29, 32.15, and 23.70 mg/g for Pb (II), Cu (II), and Zn (II), respectively.

Further work demonstrated that modification of natural phosphate with hydroxyl function improves the adsorption capability of heavy metals ions (Mobasherpour et al. 2011; Sun et al. 2018). Furthermore, such as functional groups that strongly react with heavy metals, the importance of amino groups could not be ignored. Therefore, the presence of both amino and hydroxyl groups on chitosan surface made it one of the most prevalent modifier agents, and it has a high affinity for chromium ions and thus could effectively improve the adsorption of Cr (VI) (Aydin and Aksoy 2009; Rojas et al. 2005; Sampaio et al. 2015).

In this context, the work is devoted to the synthesis, characterization of chitosan-fluorapatite (CS-Fa) composite; somehow, it could control and promote the active sites of the CS surface, and then condition the properties of the suitable adsorbent for chromium (VI) ions. Furthermore, an investigation of the different parameters on the sorption of chromium ions using response surface methodology (RSM) and regeneration study of CS-Fa will also be carried out.

2 Materials and Methods

2.1 Materials

Potassium dichromate (K2Cr2O7, Sigma-Aldrich, ≥ 99.5%), sodium hydroxide (NaOH, Sigma-Aldrich, ≥ 99%), 1,5-diphenylcarbazide (C13H14N4O, Sigma-Aldrich, ≥ 98%), nitric acid (HNO3, Sigma-Aldrich, 65%), hydrochloric acid (HCl, Sigma-Aldrich, 37%), sulfuric acid (H2SO4, Sigma-Aldrich, 37%), ethanol (C2H5OH, Sigma-Aldrich, ≥ 95%) are applied. The CAS registry number, mass fraction purity, and supplies’ name of all chemicals are listed in Table 1.

Table 1 CAS registry number and mass fraction purity
Full size table

2.2 Adsorbent Preparation

2.2.1 Preparation of Chitosan

Crude chitin was extracted from shrimp shells, which was then processed to obtain chitosan (CS). CS was obtained using chemical deacetylation process where the chitin was brought in contact with a solution of sodium hydroxide NaOH (48 wt%) at a temperature of 100 °C, for 6 h with a solid-to-liquid ratio of 1:20 to remove some or all the acetyl groups of the chitin. The product obtained was then filtered and washed several times with distilled water until neutral pH was reached, the substrate was dried at T = 50 °C. The obtained chitosan was labeled as CS.

2.2.2 Preparation of Fluorapatite

The natural phosphate used in the present work comes from an ore located in Khouribga (Morocco). A determined mass of the natural phosphate (125 μm) was transferred into a 500-mL conical flask; then, the dissolution reaction of the ore was carried out by adding a volume of 250 mL of the distilled water at 75 °C for 2 h. The precipitate was filtered under vacuum, washed with distilled water, and then dried in an oven at 100 °C. The dried phosphate was subjected to two treatments, firstly sodium hydroxide (1 M) and then with nitric acid (1 M) for 3 h. Filtered, washed, and dried, the solid was calcined at 900 °C.

2.2.3 Preparation of CS-Fa Composites

The cross-linked chitosan was prepared by dissolving 1 g of chitosan in 100 mL of 1% (v/v) acetic acid solution, and then a determined mass of Fa was added to the chitosan solution. The mixture is then stirred for 24 h. Finally, the resulting composite was washed with distilled water and dried at 50 °C overnight.

2.3 Characterizations

X-ray diffraction analysis was carried out using a Bruker D8 diffractometer operating at 45 kV/100 mA, using CuKα radiation with Ni filter. FTIR spectrometer was obtained using a thermo-scientific spectrometer in the mid-infrared region between 400 and 4000 cm−1 with a resolution of 4 cm−1. The surface morphology of the samples was obtained from a scanning electron microscope (SEM) Philips XL 30 ESEM (Acc spot Magn 20.00 kV). Thermogravimetric analysis was performed using a Discovery TGA from TA instruments at a heating rate of 10 °C/min under nitrogen atmosphere.

2.4 Adsorption Studies

The adsorption experiments of Cr (VI) and CV onto CS and CS composite were carried out in a batch system. The adsorption tests were conducted in glass beakers (150 mL) containing 50 mL of CV/MB solutions at a consistent stirring rate. The effects of experimental parameters such as pH, adsorbent mass, contact time, initial chromium concentration, and temperature on the adsorption process were examined. The initial pH of the solution was adjusted by the addition of 0.1 M HCl or 0.1 M NaOH. The concentration of Cr (VI) in the solution was determined using a UV-Vis spectrophotometer (UV-Vis 1200) at 540 nm. The adsorption capacity qt (mg/g) and adsorption percentage (%Removal), at a specified contact time, were calculated using the following equations:

$$ {\mathrm{q}}_{\mathrm{t}}=\frac{\left({C}_o-{C}_t\right)V}{m} $$
(1)
$$ \%\mathrm{Removal}=\frac{\left({C}_o-{C}_t\right)}{C_o}\times 100 $$
(2)

where Co and Ct are the amounts of initial and retained Cr (VI) in the solution at time t (mg/L), respectively. V is the solution volume (L) and m is the mass of adsorbent (g).

2.5 Data Analysis

In order to describe the mechanism involved in the adsorption process, the experimental kinetic data were analyzed by application of the pseudo-first-order (PFO), the pseudo-second-order (PSO), and intraparticle diffusion (IPD) kinetic models, while Langmuir and Freundlich models were applied to describe the obtained isotherms to propose the sorption mechanism involved (Table 2) (Ho and Mckay 1998; Lagergren 1898; Langmuir 1916; U. 1906; Weber and Morris 1963). Thermodynamic parameters including the Gibbs free energy (ΔG), enthalpy (ΔH), and entropy (ΔS) were also analyzed to predict the feasibility and determine the nature of the adsorption process in the temperature range of 25–45 °C (Hameed et al. 2007; Tsai and Chang 1995).

Table 2 Kinetic and isotherm models
Full size table

2.6 Regeneration Study

The regeneration of the CS and CS-Fa was carried out using NaOH (0.5 N) solution as eluent for desorption of chromium ions (Cr (VI)). Desorption experiments were conducted by mixing 0.1 g of chromium loaded CS or CS-Fa with 50 mL of eluent at 25 °C for 2 h. The regenerated adsorbent was separated, then thoroughly washed with distilled water and then dried. The recovered materials were further subjected to adsorption experiments. This regeneration procedure was repeated up to six successive cycles. The loss of adsorbent mass after each regeneration cycle was considered to keep the same adsorbent/solution ratio in further use for adsorption.

3 Results and Discussion

3.1 Characterizations

3.1.1 X-ray Diffraction

X-ray patterns of CS, Fa, and CS-Fa composite were studied respectively in Fig. 1. The XRD pattern of biopolymer CS is depicted in Fig. 1. The diffractogram of prepared chitosan showed to 2θ, 10, and 20 corresponding the semi-crystalline patterns of chitosan (Sethy and Sahoo 2019), which has been confirmed by the JCPDS file no. 039-1894. Besides, Fig. 1 indicates the existence of a single fluorapatite phase with no characteristic peaks of impurities (e.g., CaO or tricalcium phosphate). As can be seen from Fig. 1, all diffraction peaks of Fa are in agreement with the standard pattern of fluorapatite (JCPDS file no.15-0876) with a hexagonal structure (Essamlali et al. 2019). The graphs of pure CS and Fa were taken as a reference to explain the data from the X-ray patterns of CS-Fa composite. The XRD patterns of CS-Fa is given in Fig. 3c showed impressive results since the major peaks and patterns in XRD data correspond to the composites. However, the characteristic peaks of Fa remain the same in the case of CS-Fa; a noticeable trend is that the peaks of CS become less defined, shifted in their initial position, and intensity also decreased. These results strongly show that the synthesis of CS-Fa composite was successfully realized under the chosen conditions.

Fig. 1
figure 1

X-ray patterns of chitosan (CS), fluorapatite (Fa), and CS-Fa composite

Full size image

3.1.2 FTIR Analysis

The characterization of chitosan by infrared spectroscopy is illustrated in Fig. 2; all bands that we found in this spectrum of chitosan are in good agreement with reported studies (Haffad et al. 2019; Song et al. 2013). The bands’ attribution appearing in the spectrum are grouped in Table 3 (Nan et al. 2019).

Fig. 2
figure 2

FTIR spectra of chitosan (CS), fluorapatite (Fa), and CS-Fa composite

Full size image
Table 3 FTIR bands’ attribution of CS, Fa, and CS-Fa
Full size table

The FTIR spectra of Fa impregnated in the CS biopolymer matrix confirm that some characteristic peaks of CS were shifted and others not. The broad peak between 3000 and 3600 cm−1 corresponding to the vibration of N–H and O–H changed its initial position and became more intense; the same remark was observed for the peak allocated to C–H symmetric stretching at 2930 cm−1. The peak at 1661 cm−1 becomes very intense, while the peak at 800 cm−1 in the spectrum of Fa is almost disappearing, which confirms that residual N-acetyl (C=O stretching of amide I) reacted with the carbonate substitution of FA and formed a robust structure of CS-FA composite.

3.1.3 Scanning Electronic Microscopy-Energy-Dispersive X-ray Spectroscopy

The SEM image of obtained chitosan is presented in Fig. 3. Irregular structure and cracks can be clearly seen in SEM image of prepared CS. These characteristics (surface morphology with several cracks) give the sheet microstructure surface of CS. In total, surface changes were observed when CS-Fa was prepared. In fact, the surface becomes very rigorous and the particles of Fa are well deposited on the sheet surface of CS in a fairly homogeneous way, which was confirmed by EDS analysis throughout the homogeneous distribution of Ca and P on CS-Fa.

Fig. 3
figure 3

SEM images and EDS analyses of chitosan (CS) and CS-Fa composite

Full size image

3.1.4 Thermogravimetric Analysis

Next, the thermal stability of CS and CS-Fa was compared by TGA. According to Fig. 4, CS and CS-Fa showed a slight weight loss of 9% and 5%, respectively, at 100 °C that attributed to the evaporation of free water (non-bond) and hydrogen-bonded water (Abdel-Fattah et al. 2015; Dessì et al. 2013). More weight losses at 250 °C and 400 °C corresponded to the decomposition of volatile organic matter and released volatile substance of the thermal decomposition of the inorganic compound for CS and CS-Fa, respectively (Sivakumar et al. 2002). Additionally, it was observed that the CS decomposition was slowed when incorporated by FAP, suggesting good thermal stability to be applied as an adsorbent. The total weight losses were found to be 67% and 56% for CS and CS-Fa, respectively (Ali et al. 2018).

Fig. 4
figure 4

TGA graphs of CS and CS-Fa

Full size image

3.2 Adsorption Studies

3.2.1 The Effect of pH

The influence of pH on Cr (VI) adsorption onto CS and CS-Fa was studied in a pH range from 1.0 to 7.0, using 0.1 g of CS or CS-Fa in a 50 mL of Cr (VI) solution (10 mg/L) for 1 h at room temperature. The pH was adjusted using 0.1 mol/L NaOH and 0.1 mol/L HCl. The results are shown in Fig. 5a. At this pH range, the highest levels of Cr (VI) removal on CS (66.2%) and CS-Fa (92.0%) were observed at pH 3.0, and they dropped for CS to 19.8% and 21.22% at pH values of 1.0 and 7.0, respectively, while they decreased to 69.0% and 48.0% at pH values of 1.0 and 7.0, respectively for Cs-Fa. As shown in Fig. 5a, the Cr (VI) removal diminished as the solution pH increases from 3.0 to 7.0. Cr (VI) speciation in solution is recognized to be highly pH-dependent (Dinker and Kulkarni 2015). Chromic acid (H2CrO4) appears when the pH is less than 2.0. Also, when the pH ranged from 1.0 to the neutral pH 7.0, the HCrO4− ions occur, while, above the neutral pH, only CrO42− ions are present in the solution (Fig. S2) (Elouahli et al. 2018). As disclosed in Fig. 5a, the removal proficiency at lower pH values is higher because of the predominance of Cr (VI) species mainly exists in monovalent HCrO4− form. As the pH increased, the CS and CS-Fa surface develops the negative surface charges increasingly, leading to a drop in the removal efficiency of Cr (VI). Consequently, the removal percentage of Cr (VI) at lower pH values (pH < pHPZC) is higher compared to higher pH values (Fig. S1). This result demonstrates that the pH is a significant factor affecting the removal efficiency of Cr (VI) on CS and CS-Fa. Later studies of adsorption were carried out at pH 3.0.

Fig. 5
figure 5

a Effect of pH solution and b adsorbents mass on adsorption of Cr (VI)

Full size image

3.2.2 The Effect of Adsorbent Mass

The effect of CS and CS-Fa dosage on Cr (VI) removal displayed in Fig. 5b shows that the percentage removal of Cr (VI) increases with increasing the amount of the adsorbent from 20 to 100 mg, and then it remains stable. This could be attributed to the availability of active binding sites (Abdellaoui et al. 2018). As Fig. 5b shows, 100 mg Cs-Fa was enough to remove over 91% of Cr (VI) (Ci = 100 mg/L), while 100 mg of Cs uptake was limited in 66.29%. In general, the removal of Cr (VI) by Cs-Fa was comparatively higher and faster than Cs materials, as will be discussed in the next section.

As the removal efficiency was to be the highest when we use 100 mg of CS and CS-Fa, it will be considered for further experiments.

3.2.3 Adsorption Kinetics

Figure 6 illustrates the adsorption of metal ions of Cr (VI) on CS and CS-Fa with different contact times (0 to 120 min). The adsorption capacity increase with increasing contact time up to 20 min and 15 min, respectively, for CS and CS-Fa, after which it is more constant. The higher adsorption of CS-Fa at the initial contact time compared to CS could be related to the driving force of Cr (VI) ions into the surfaces of CS-Fa and the abundance of active sites on the CS-Fa compared to CS (Li et al. 2010; Wu et al. 2008) (Table 4).

Fig. 6
figure 6

a Kinetic studies by PFO, PSO, and IPD models. b Isotherm studies by Langmuir and Freundlich models

Full size image
Table 4 Kinetics adsorption parameters of Cr(VI) on CS and CS-Fa
Full size table

To examine the adsorption kinetics of Cr (VI) metal ion onto CS and CS-Fa and the mechanism and rate-controlling step in the whole adsorption process, the nonlinear forms of pseudo-first-order (PFO), pseudo-second-order (PSO), and intraparticle diffusion (IPD) models are used (Table 2). Table 2 summarizes the analogous parameters for Cr (VI) metal ion adsorption on CS and CS-Fa. The results showed that the R2 of PSO and IPD exhibit a poor connection of Cr (VI) metal ion onto CS and CS-Fa as opposed to parameters of the PFO. In addition, PFO model show well-coordinating correlations between calculated Qcal values (CS 34.65 mg/g and CS-Fa 46.12 mg/g) and experimental qe values (CS 34.15 mg/g and CS-Fa 45.53 mg/g).

3.2.4 Adsorption Isotherms

Langmuir and Freundlich isotherms (Table 2) were useful to define the adsorption capacity of CS and CS-Fa for Cr (VI) metal ion (Elouahli et al. 2018). The fitting plots of Langmuir and Freundlich isotherm models for Cr (VI) adsorption on CS and CS-Fa are shown in Fig. 6. The calculated isotherm parameters, along with correlation coefficients (R2), are set in Table 5. Rendering to R2 values (R2 > 0.997), the Langmuir model tailored better than the Freundlich model to experimental data for the adsorption of Cr (VI) metal ion. Owing to the Langmuir model is established on the homogenous adsorption assumption, it can be said that the adsorption of Cr (VI) metal ion onto CS and CS-Fa surface is homogenous. By comparing the obtained qmax, the qmax values for CS-Fa (100.92 mg/g) is higher than CS (81.34 mg/g), which demonstrates the efficiency of the addition of Fa.

Table 5 Equilibrium adsorption parameters of Cr(VI) on CS and CS-Fa
Full size table

However, the higher sorption capacity of the CS-Fa composite compared to CS could be attributed to the presence of more reactive amino groups in chitosan as well as active binding sites already possess fluorapatite in the surface. Therefore, under the studied condition, the chromium fraction is negatively charged (mainly HCrO4−), which promotes strong electrostatic attraction with surface groups of CS-Fa (Kousalya et al. 2010). Consequently, it is confirmed that the use of composite possesses an enhanced sorption capacity, which demonstrates its selectivity towards Cr (VI) ions. Compared to some chitosan-based adsorbents reported in the literature, CS-Fa could be a promising and effective adsorbent for the removal of chromium ions from aqueous medium (Table 6).

Table 6 Adsorption capacity comparison of various chitosan-based adsorbents
Full size table

3.2.5 Thermodynamic Study

One of the indispensable tools in the prediction of the adsorption mechanism, whether it is a physisorption or chemisorption process are the thermodynamic parameters. The thermodynamic parameters could be determined from the thermodynamic laws as it is presented in Table 1 (Fig. 7). The plots of thermodynamic parameters are listed in Table 7. The free energies of both adsorbents, CS and CS-Fa, exhibit negative values in the studied temperature range, indicating that the adsorptive of chromium has occurred favorably and spontaneously with low adsorption-energy requirements. The negative ΔH° quantities (− 32.549 and − 20.095 J/mol) suggest that the adsorption of Cr (VI) is exothermic onto CS and CS-Fa, respectively. Moreover, the decrease of ΔG values with increasing the temperature from 298 to 333 K asserted the exothermicity of the adsorption process, which was also confirmed by the negative value of ΔH. In addition, the ΔS° < 0 reflects a decrease in the degree of freedom of Cr (VI) ions on the solid-solution interface of the adsorbents (Aydin and Aksoy 2009; Mobasherpour et al. 2011; Pal and Pal 2019).

Fig. 7
figure 7

Plot of ln(Kd) versus 1000/T for Cr(VI) adsorption onto CS and CS-Fa

Full size image
Table 7 Thermodynamic parameters of Cr(VI) using CS and CS-Fa
Full size table

3.3 Regeneration

The evaluation of the reusability and recovery of the adsorbents is one of the crucial factors in the adsorption processes. The CS and CS-Fa loaded Cr (VI) after adsorption were recovered using NaOH solution. From the results shown in Fig. 8 CS-Fa showed a high regeneration capacity compared to CS; however, the uptake ability of Cr (VI) using CS-Fa composite slightly decreased after four cycles from 89.32 to 79.7% in the sixth cycle. Whereas uptake percent was decreased from 66.8 in the initial three cycles to 52.5 in the fifth cycle. Accordingly, the synergistic use of the adsorbents (chitosan and fluorapatite) as a composite has improved the regeneration capacity of the adsorbent. Consequently, CS-Fa could be reused for the fifth time in the adsorption process of Cr (VI).

Fig. 8
figure 8

Recycling study of CS and CS-Fa adsorbents

Full size image

3.4 Modeling Using CCD-RSM

The adsorption of Cr onto CS and CS-Fa was optimized by using a multivariate analysis consisting of response surface methodology (RSM) coupled with central composite design (CCD) (Ait Ahsaine et al. 2017, 2018; Anfar et al. 2017). Therefore, the variables were assessed at five coded levels (− α, − 1, 0, + 1, and + α) (Table S1). Appropriate parameters affecting the adsorption of Cr were found by using the data of the first step in this work. Therefore, pH, time, temperature, and mass were considered as the design variables. Ordinary equations and parameters of the second-order polynomial model have presented details elsewhere (Anfar et al. 2018; Zbair et al. 2018, 2019) (Table S1).

The obtained data were fed into a computer for statistical study. The results of statistical treatment are as follows:

  • The relationship between the response (removal %) and variables was confirmed by ANOVA and shows a strong significance of variables at a 95% confidence level (P value < 0.05) for both cases (Cr adsorption onto CS and CS-Fa) (Anfar et al. 2018).

  • Based on Table S2, the coefficients of determination values indicate a good fit between the experimental and calculated results by the model in the case of Cr adsorption onto CS-Fa and the opposite for Cr adsorption onto CS (Anfar et al. 2019b).

  • The significance of the coefficients was examined using the P value (Table S2). Indeed, all coefficients are significant at 5% or 1% in both cases, except the pH effect, the interactions pH × temperature, temperature × time, and pH × mass (Anfar et al. 2019c).

The results of CCD matrix were used to elaborate a quadratic polynomial equation for Cr adsorption onto CS-Fa (Eq. 1).

$$ {\displaystyle \begin{array}{l} Adsorption\ \left(\%\right)=88.033+0.265B+2.398C+1.141D\hbox{--} 2.014{A}^2+0.807{B}^2\\ {} CD-1.396{C}^2-0.495{D}^2\hbox{--} 1.193 AC\hbox{--} 0.296 BD-1.247\end{array}} $$
(1)

All reported considerable interactions among different parameters in the CCD were taking into account to determine most significant effects by using RSM presentations (Fig. S3) (Anfar et al. 2019a; Elouahli et al. 2018; Haffad et al. 2019; Ouasfi et al. 2019). Based on these three-dimensional response surface plots, the influence related to time and temperature depicted in Fig. S3-a indicates that high removal of Cr was found at this regain. The suggested optimized value by RSM for pH was found to be 2.54 for pH, 25.75 °C, 36.63 min, and by using of 86.72 mg of CS-Fa adsorbent. The estimated adsorption removal efficiency by the model, at these values, was 91.28%. Under some optimal conditions, the adsorption of Cr onto CS-Fa was carried out and led to an achievement of 91.8%. These results show an indication of proper accordance between estimated and experimental results and confirm the suitability of the model (Table S2).

4 Conclusion

In this study, it has been demonstrated that synergistic use of the adsorbents (chitosan and fluorapatite) as a composite has improved the sorption capacity of Cr (VI) and the regeneration capacity of the adsorbent as well. The chromium uptake was slightly influenced by the pH range from 2 to 6. The Langmuir isotherm model could be applied to the sorption of Cr (VI) ions, both sorbent CS and CS-Fa, with maximum capacities of 81.34 and 100.92 mg/g, respectively. The rate of sorption followed the pseudo-first-order kinetic model, where the process was spontaneous and exothermic in nature.