Introduction

The discharge of the heavy metals into the environment is a critical pollution problem. Unlike organic pollutants, the heavy metals are not biodegradable. They keep accumulating in the organisms and to incorporate within the food chains through multiple pathways, which causes severe harm to human health (Lu et al. 2009; Amzal et al. 2009; Nagajyot et al. 2010; Ghnainia et al. 2016). Consequently, toxic heavy metals including zinc, nickel, copper, lead, chromium, cadmium, and mercury are of special concern in the treatment of industrial wastewater.

The toxicity of chromium ions to the mammals and the aquatic organism is approved. For mammal, the toxicity manifests also by other heavy metal ions, such as Cr(III), generally, due to her lower solubility and her little mobility in the ecosystem compartments of heavy metals. The chromium ions are mentioned by the International Agency for Research on Cancer (IARC 2012) as strong carcinogenic agent, which changes the DNA transcription process (Nandy et al. 1990; Wielinga et al. 2001; Wang et al. 2012). At excessive concentrations, Nickel (Ni(II)) induces lungs, nose, and bone cancers. For example, the dermatitis is the major consequence of the exposure to nickel, likely costume jewelry and coins. Moreover, Ni carbonyl [Ni (CO)] has been considered as lethal to humans at atmospheric exposures of 30 ppm for 30 min (Beliles 1979; Natasha and Vernon 2006). The acute poisoning of nickel causes dizziness, headache, fast respiration, nausea and vomiting, dry cough and breath shortness, chest pain and tightness, cyanosis, and extreme weakness (ATSDR 2003).

Nowadays, many treatment technologies have been developed to remove the heavy metal ions from water and wastewater such us chemical precipitation (Villa-Gomez et al. 2011; Mbamba et al. 2015), ion exchange processes (Dabrowski et al. 2004; Kang et al. 2004; Figoli et al. 2010; Ahmad et al. 2011), coagulation (Kurniawan et al. 2006; Pang et al. 2011), and adsorption. Among the best techniques for the removal of heavy metals from wastewater is adsorption, because the adsorption is a reversible process and the adsorbent can be regenerated by simple desorption process for another use (Pan et al. 2009). Further, the adsorption mechanism is characterized by low maintenance cost, high capacity of elimination, and ease of operation.

In addition, the adsorption process has come to the forefront as one of the popular techniques for heavy metal removal from water/wastewater. Many studies developed a new adsorbent, such as activated phosphate rock (Elouar et al. 2008; Boujelben et al. 2008), activated carbon (Baccar et al. 2009; Omri et al. 2016), biomaterials (Wahaba et al. 2011; Ghrab et al. 2017a), and especially clay minerals (Chaari et al. 2008; Hamdi and Srasra 2012; Eloussaief and Benzina 2010; Ghrab et al. 2013; Eloussaief et al. 2014; Sdiri et al. 2016), characterized by a low-cost, large specific area, chemical and mechanical stability, layered structure, and high cation exchange capacity. These important characteristics could remove significantly the undesirable metals from wastewater.

In this study, a Tunisian smectite sample (R5) was characterized using X-ray diffraction (XRD), chemical analysis, Fourier transform infrared spectroscopy (FT-IR), thermal behavior, and scanning electron microscopy (SEM). Later, batch adsorption experiments were performed to evaluate the adsorption properties of the (R5) toward nickel(II) and chromium(III). In order to understand the adsorption mechanism, kinetic and thermodynamic model studies were also examined.

Materials and methods

Adsorbent

The smectite sample (R5) was collected from Aleg Formation in Jebel Romana. It is early Coniacian Age. Jebel Romana is located in the N-E termination of the anticline Zemlet El Bidha. (R5) was kept in an oven at 60 °C.

Characterization of smectite sample

The mineralogical analysis was determined out using a Philips® X-Pert diffractometer with Cu Kα radiation. The diffraction data results were analyzed with the X powder® computer program (Martín-Ramos 2004). For minerals quantification, the data results obtained by the classical method (area measurement of the peaks and relative power) were corrected according to the chemical composition of rock, following López-Galindo et al. (1996). The relative error was mentioned by 5%.

The chemical composition of the major elements of the total rock (R5) was obtained by XRF and flame-photometric methods, using a spectrometer of the type BRUKER S4 Pioneer X-ray Fluorescence, associated in the anode with X-ray RH (60 Kv, 150 my). Quantification was determined by the fundamental method of parameters using the software related to the equipment (Spectra Plus).

The FT-IR spectrum was obtained with a Nicole Impact 410 FT-IR spectrophotometer. The sample is depressed in KBr pellet between 400 and 4000 Cm−1.

Thermogravimetric analyses were obtained using 1 g of sample, analyzed with TGA-50 SHIMADZU equipment operating in an air atmosphere, with a heating rate of 20 °C/min. TGA/DTA curves were obtained between 0 and 950 °C range.

A detailed study of the morphology and texture of the selected clay sample (R5) was carried out by the Scanning Electron Microscope AURIGA (FIB-FESEM) of Carl Zeiss SMT with high resolution (FESEM) and with the tension of variable acceleration.

The specific surface area was determined by Brunauer-Emmet-Teller (BET) method (Micromrtrics ASAP 2020 V3.04 H). The N2 adsorption experiments were performed after a suitable thermal treatment (150 °C) under vacuum (10−4 Pa) for 24 h.

The cation exchange capacities (CEC) were estimated by washing thoroughly the sample with deionized water to eliminate superficial cations, and then, 1 g of (R5) powder was dispersed in 25 ml (1 M) aqueous solution of tetramethylammonium bromide to displace the constituent cations. The dispersion was shaken overnight at 50 rpm in water bath with 25 ± 1 °C. The content of the dispersion (Na+, Ca2+, and Mg2+) in solution was determined by atomic absorption spectroscopy (PerkinElmer Spectrometer (5100 mod)), and the CEC was calculated as the sum of exchangeable cations, expressed in meq/100 g of (R5).

Adsorbate

All chemical reagents used were obtained from Fluka (purity 99%). A stock solution of each metal—Ni(II) (1041.09 ppm) and Cr(III) (1026.82 ppm)—was prepared by dissolving NiSO4 7(H2O) and CrCl3 6(H2O) in distilled water.

Batch adsorption experiments

Batch adsorption was selected as an appropriate technique in the current study. The experiments were carried out by mixing of 0.5 g of (R5) sample with 50 ml of solutions containing heavy metals of the desired concentration. After equilibrium, suspensions were filtered and analyzed by the atomic absorption spectrophotometer (HITACHI model Z-6100).

To investigate the kinetic adsorption, a set of Erlenmeyer was prepared as described above, but then shaken for 2, 5, 10, 20, 30, 60, and 120 min. The initial concentration of 20 mg/L was the same in all of the kinetic study, as was the temperature (25 °C).

The thermodynamic studies of the metals adsorption were investigated by varying the concentration of each metals from 1 to 50 mg/L. The experiment isotherms were realized at 10 and 40 °C.

Results and discussion

Characterization of the smectitic clay adsorbent

The X-ray diffraction analysis of total rock (Fig. 1) and oriented aggregates (Fig. 2) indicates that the adsorbent clay R5 is mostly composed of phyllosilicates (76%) and associated minerals (24%). The mineralogical composition of phyllosilicates is fairly variable; it is composed of smectite (64%), with a weak percentage of kaolinite (8%) and illite (4%). Quartz (8%), calcite (8%), gypsum (7%), and potassium feldspar (1%) are associated minerals in this adsorbent smectitic clay (Table 1).

Fig. 1
figure 1

Diffractogram XRD of total rock of adsorbent smectite clay R5

Full size image
Fig. 2
figure 2

Diffractogram XRD of oriented aggregates of adsorbent smectite clay R5

Full size image
Table 1 Mineralogical composition of adsorbent smectite clay R5 (weight%/weight)
Full size table

The adsorbent smectitic clay of Aleg Formation is characterized by a significant rate of Si (43.52%) and Al (15.28%) (Table 2). The enrichment of (R5) with aluminosilicate is confirmed by the significant rates of silica and aluminum. The (R5) sample is rich in iron oxide, CaO, and MgO, which verifies the high loss of ignition (LOI).

Table 2 Major element content of adsorbent smectite clay R5 (%)
Full size table

The FT-IR spectrum (Fig. 3) is correlated with the results obtained following the mineralogical characterization and chemical analysis. This confirms the presence of smectite (915 cm−1) and quartz (780 cm−1). Furthermore, these analyses prove that mineral surface clay is negatively charged with SiO− (470, 530, and 1040 cm−1) and AlO− (3620 cm−1) (Ghrab et al. 2017b).

Fig. 3
figure 3

FT-IR spectra of adsorbent smectite clay R5

Full size image

Results of thermal study (Fig. 4) show multiple endothermic peaks: The first stage weight loss (11.05 and 0.7%) with two endothermic peaks, respectively, at 113.68 and 267.89 °C corresponds to the loosely bound water molecules (Baran et al. 2001). The second stage loss (3.79%) with an endothermic peak at 519.56 °C is due to the deshydroxylation of the octahedral sheet (Brigatti et al. 2005). The third stage loss (2.36%) with an endothermic peak at 715.31 °C is contributed to the decomposition of carbonates. Exothermic peak appeared at 989.56 °C due to the crystallization of new phases (Gillot 1987). The total weight loss is about 18.26%.

Fig. 4
figure 4

TGA-DTA curves of R5 clay sample

Full size image

As for the identification of the morphology by SEM (Fig. 5), the sample R5 presents the layers of alumunosilicates with a honeycombs form having corrugated edges. This structure is characteristic of smectite clay (Azizi et al. 2013).

Fig. 5
figure 5

SEM microphotographs of the adsorbent smectite clay R5 a (1 µm) b (200 µm) c (100 nm) d (200 nm)

Full size image

The BET method of N2 adsorption-desorption was shown a BET surface area of R5 sample equal to 74.163 m2/g. The result (Fig. 6) indicates that N2 adsorption isotherm of R5 exhibits a sorption behavior of type II according to the classification of Brunauer, Deming, Deming and Teller (BDDT) (Tuccimei et al. 2015) with the appearance of hysteresis. The large uptake N2 can be noticed close to the saturation pressure. This apparent step in adsorption branch following by a sharp decline in the desorption branch confirms the presence of mesopore (Park et al. 2013).

Fig. 6
figure 6

N2 adsorption/desorption isotherms of R5 sample of clay

Full size image

CEC of the R5 sample are found to be 121.011 meq/100 mg. The total surface of the adsorbent smectitic clay is high which fosters the attachment of large amounts of metal ions. The main exchangeable cation is Ca2+ (95.437 meq/100 g), allowing this sample to be classified the calcium clay (Table 3).

Table 3 Exchangeable cation of the studied sample (meq/100 g)
Full size table

Kinetics adsorption

Figure 7a presents the curve for Ni(II) and Cr(III) adsorption kinetics on (R5). The adsorption of heavy metal ions into R5 is found to occur rapidly in the first 10 min in the process, remaining virtually at equilibrium over time. At the beginning of adsorption, this rapidity is due to the active adsorption sites on the (R5) surface, which are more available to easily interact with metal ions. Afterwards, adsorption continues at a slower rate and finally reaches the equilibrium by the saturation of adsorption sites on (R5). The equilibrium is reached within 30 min for the both metal ions Ni(II) and Cr(III). However, having the highest affinity interaction, the adsorption system Cr/R5 is slower than the Ni/R5 one to reach the equilibrium state of the adsorption.

Fig. 7
figure 7

Kinetic curve for nickel and chromium adsorption onto R5. a Adsorbed amount at equilibrium. b Pseudo-first-order kinetic model. c Pseudo-second-order kinetic model. d Intraparticle kinetic model

Full size image

The adsorption kinetics was adjusted with the pseudo-first-order kinetic model (Ho 2004; Febrianto et al. 2009) (Fig. 7b), the pseudo-second-order kinetic model (Ho and McKay 1999; Ho et al. 2000) (Fig. 7c) and the intraparticle diffusion kinetic model (Karthika et al. 2010) (Fig. 7d) which are described in Table 4. The linear coefficients and the constants of kinetic models are given in Table 5.

Table 4 Kinetic adsorption models used in this work and their parameters
Full size table
Table 5 Ni and Cr adsorption rate coefficients for pseudo-first-order model, pseudo-second-order model, and intraparticle diffusion model onto R5
Full size table

The correlation coefficient for the pseudo second order kinetic model was higher than the other models (R2 = 1, R2 = 0.999) for Ni and Cr ions, respectively, attesting that the adsorption perfectly complies with the pseudo-second-order kinetic model and the adsorption process is controlled by the pseudo-second-order model. K2 was employed to describe the chemisorption involving valency forces through the sharing or exchange of electrons between (R5) and heavy metal ions like covalent forces and ions exchanges (Ho 2006; Assameur and Boufatit 2012; Ghrab et al. 2018). According to Table 5, K2 constant confirms that the adsorption rate of Ni(II) onto (R5) is more rapid than the adsorption of Cr(III) onto (R5). In addition, the theoretical qe values determined from the pseudo-second-order kinetic model are in agreement with the experimental qe values.

Moreover, intraparticle diffusion model was less suitable for the experimental values if compared with the second-order kinetic model. The curve does not pass through the origin (Fig. 7). This marked that the pseudo-second-order kinetics adsorption mechanism is not limited by intraparticle diffusion of the metal ions within (R5).

Thermodynamic study

The thermodynamic parameters ΔG, ΔH, and ΔG are presented in Table 6 and were determined using the following equations (Eqs. 6 and 7) and calculated values obtained slope and intercept Van’t Hoff (Eq. 8) (Fig. 8).

Table 6 Thermodynamic parameters for Cr and Ni adsorption onto R5
Full size table
Fig. 8
figure 8

Graphical representation of the Van’t Hoff equation

Full size image

The spontaneity of reactions is mentioned by the negative Gibbs free energy with a significant contribution of the positive entropy. These results confirmed the interaction of metal ions (Ni or Cr) with the reactive sites on the surface R5. The complexation of the heavy metal ions on the (R5) surface promotes the adsorption of new free ions. Consequently, heavy metal ions lead to an entropy increase.

The decrease in the Gibbs free energy with the increase of the temperature, identically for Ni and Cr, presents that the adsorption mechanism is favorable at higher temperature. In addition, the values of enthalpy of adsorption onto (R5) are positive; it indicates that the adsorption mechanism of Ni(II) and Cr(III) is an endothermic process, controlled by physical and chemical mechanisms of adsorption (Malkoc and Nuhoglu 2005; Huang et al. 2007).

The positive values of entropy changes for Ni(II) and Cr(III) present that randomness increased at solid-liquid interface during the adsorption mechanism. Therefore, it could be inferred that the adsorption was propelled by entropy (Sheng et al. 2013).

Conclusion

The raw clay, originally from the Jebel Romana, Tunisia, was investigated for the removal of Ni(II) and Cr(III) from aqueous solution in the present study. The (R5) was characterized by XRD, FTIR, TGA–DTA, and SEM that were identified as smectite with a high BET surface area (74.163 m2/g) and CEC (121.011 meq/100 mg). Furthermore, the ability of (R5) to remove Ni(II) and Cr(III) was evaluated using equilibrium and thermodynamics models.

The pseudo-second-order kinetic model shows the best correlation with the experimental kinetic data. The experimental qe and the calculated (by the second-order model) qe showed greater agreement for both heavy metal ions (qe(Ni) = 6.660, qe(Cr) = 15.479).

The adsorption results were confirmed through stable complexes established between cations and reactive groups disposed on the raw clay surface. The complex behavior was determined based on the thermodynamic constants obtained by Van’t Hoff correlation in the solid/liquid interface to give favorable results, and thermodynamic values showed that the process adsorption of Ni and Cr by R5 has endothermic enthalpy, negative Gibbs free energy, and positive entropy values. These thermodynamic values suggest the investigation of this worldwide available material to improve the heavy metals adsorption.

Currently, a large variety of adsorbent materials are used for heavy metal ion removal; the ideal adsorbent mandatorily should have interesting characteristics for industrial and environmental application. In this context, it is possible to conclude that the employed smectite (R5) is more economical than the commercially available adsorbents. Moreover, this raw clay is actually available and easily accessible in many geological outcrops in Tunisia, particularly in the Aleg Formation at Jebel Romana. Temperature, thermodynamic parameters, effect of concentration variation, and contact time are very important parameters that influenced the adsorption capacity in the real application of the adsorption process by these materials.