The role of kaolin and kaolin/ZnO nanoadsorbents in adsorption studies for tannery wastewater treatment

In the present study, comparative studies of kaolin and kaolin/ZnO nanocomposites for the adsorption of Cr(VI), Fe(III), COD, BOD, and chloride from tannery wastewater were investigated. ZnO nanoparticles and kaolin/ZnO nanocomposites were prepared by sol–gel followed by wet-impregnation methods. The prepared adsorbents were characterized using different analytical tools such as X-ray diffraction, Fourier transforms infrared, high-resolution transmission electron microscopy, energy dispersive spectroscopy, selective area electron diffraction and Brunauer Emmett–Teller (BET) and X-ray Photoelectron Spectroscopy (XPS). The HRSEM/EDS/XPS analysis confirmed successful immobilization of clay structural network on the lattice layers of zincite hexagonal structure of ZnO nanoparticles. BET measurement showed an increase in the surface area of kaolin/ZnO nanocomposites (31.8 m2/g) when compared to kaolin (17 m2/g). Batch adsorption studies were carried out by varying the parameters such as contact time, adsorbent dosage and temperature. The maximum removal of Cr(VI) (100%), Fe(III) (98%), COD (95%), BOD (94%) and Chloride (78%) was obtained at 15 min by kaolin/ZnO composites. While 78% Cr(VI), 91% Fe(III), 91% COD, 89% BOD and 73% Chloride were removed by kaolin under the same conditions. The kaolin/ZnO nanocomposites exhibited better adsorption performance than kaolin due to higher surface area of the former than the latter. It was found that the Jovanovic isotherm model fitted the adsorption experimental data most with the highest correlation (R2 > 0.99) for both nanoadsorbents and indicate the occurrence of adsorption on monolayer and heterogeneous surfaces. The mechanism for the adsorption of metal ions in tannery wastewater onto the nano-adsorbents was examined using Weber Morris intra-particle diffusion model and Boyd plot which showed that the adsorption process was both intra-particle and film diffusion controlled. The thermodynamic parameters such as enthalpy change showed that that adsorption of metal ions and other parameters was feasible, spontaneous and endothermic. The ZnO/clay nanocomposites exhibited excellent recyclable and re-useable properties even after six repeated applications and can, therefore, be applied in wastewater treatment for removal of heavy metals and other physicochemical parameters.

treatment of raw kaolin. The kaolin sample was suspended in a 1,000 cm 3 beaker filled with deionized water and stirred for 30 min for particle size screening. Afterwards, the suspension was allowed to settle for 24 h. Supernatant from the beaker was collected by dispersion, decantation and filtration techniques. The filtrate was oven-dried at 100 °C for 24 h until there was complete dryness. Further treatment was done on the kaolin as described by Mustapha et al. 24 . The dried sample was ground and then sieved prior to further analysis.
Synthesis of ZnO nanoparticles. The zinc oxide nanoparticles were synthesized using the sol-gel method. 50 cm 3 of 0.5 M Zn(C 2 H 3 O 2 ) 2 ·2H 2 O solution was measured into a 250 cm 3 beaker and about 100.0 cm 3 of de-ionized water was added. The solution was stirred using a magnetic stirrer at 150 rpm for 30 min. A solution of 0.5 M NaOH was added drop-wise to the Zn(C 2 H 3 O 2 ) 2 ·2H 2 O solution to obtain the desired pH. The mixture was stirred for 5 min and the precipitates obtained were filtered using a Whatmann No. 42 filter paper. The precipitates were washed with deionized water and ethanol to eliminate traces of the unreacted precursors. The final product was oven-dried at 105 °C for 24 h and finally calcined in the furnace at a temperature of 450 °C for 3 h to obtain ZnO nanoparticles.

Synthesis of kaolin supported ZnO nanoparticles.
Combination of sol-gel and wet impregnation method was employed for the preparation of kaolin/ZnO as shown in Fig. 1. Zn(C 2 H 3 O 2 ) 2 ·2H 2 O (5.0 g) was first dissolved in de-ionized water (50.0 cm 3 ) and stirred for 30 min, then sodium hydroxide solution (0.5 M) was added in drops into the precursor solution to adjust the pH to 6   www.nature.com/scientificreports/ 0.5 g of the beneficiated clay was dispersed under vigorous stirring at 150 rpm for 1 h. A kaolin/ZnO homogeneous gel was formed and allowed to age overnight. The product was washed with de-ionized water and oven-dried at 80 °C overnight and finally calcined in the furnace at 450 °C for 3 h.

Characterization of kaolin, ZnO nanoparticles and kaolin/ZnO nanocomposites. The identifi-
cation of phases and the crystallite sizes of the synthesized nanoparticles and nanocomposites were determined by a Bruker AXS D8 Advance X-ray diffractometer with Cu Kα radiation. The powder samples were sprinkled on a de-greased glass slide and their diffraction pattern were recorded between diffraction angles of 20°-90°. The phase identifications were done by comparison with available d-spacing information and peaks from the Joint Committee on Powder Diffraction Standard (JCPDS). The morphologies of the synthesized samples were examined using a Zeiss Auriga HRSEM. About 0.05 mg of each sample was sprinkled onto carbon adhesives tape and sputter-coated with Au-Pd using a Quorum T15OT for 5 min. The microscope was operated with electron high tension at 5 kV for imaging. High resolution scanning electron microscopy (HRSEM) equipped with energy dispersive spectroscopy (EDS) was further used to determine the elemental composition of the synthesized nanoparticles and nanocomposites. The particle size and distribution pattern were analyzed by Zeiss Auriga High-Resolution Transmission Electron Microscopy (HRTEM) coupled with energy dispersive spectroscopy (EDS). The BET surface area and average pore volume distributions were obtained from the plot of the volume adsorbed (cm 3 /g STP) against relative pressure. The Brunauer-Emmett-Teller (BET) surface area, total pore volume and pore size were determined using a NovawinQuantachrome instrument. Fourier transform infrared spectra (FTIR) of the synthesized samples were recorded using Thermo Scientific Nicolet iS5 instrument with iD5 ATR spectrometer at a wavenumber range of 4,000-500 cm −1 . X-ray photoelectron spectroscopy (XPS) model with monochromated Mg Kα (1,253.6 eV) operated at a power of 300 W and 15 eV were used to examine the surface elemental oxidation states of ZnO nanoparticles and kaolin/ZnO nanocomposites.
Batch adsorption experiment. The adsorption experiments were performed by shaking a known weight of kaolin and kaolin/ZnO nanocomposites with 40 cm 3 tannery wastewater solution in a 250 cm 3 Erlenmeyer flask. Contact times ranging from 0 to 30 min were used to study the effect of time on adsorption of the pollutants in the solution and to attain the equilibrium time at 150 rpm. Furthermore, the effects of dosage (between 0.4 and 1.4 g) and temperature (30 °C and 80 °C) on the removal of pollutants from wastewater at their optimum time using speed rotation of 150 rpm were investigated. Supernatants were analyzed to determine the final concentration of some physicochemical parameters (COD, BOD and chloride) using APHA, manual method, 2005 and Atomic Absorption Spectrophotometer (AAS), (Perkin Elmer 200) respectively for Cr 6+ and Fe 3+ . The various operating conditions of batch adsorption studies were carried out in duplicate. The percentage removal and adsorption capacity of the parameter by the adsorbents were calculated using Eqs. (1) and (2), respectively.
where C 0 (mg/dm 3 ) is the initial concentration of a pollutant in aqueous solution, C e (mg/dm 3 ) is the concentration of pollutant aqueous solution at equilibrium, V (dm 3 ) is the volume of the wastewater used in the experiment, and M (g) is the mass of the adsorbent.

Results and discussion
Characterization of tannery wastewater. The physical and chemical properties of the tannery wastewater were reported in our previous studies by Mustapha et al. 24 . The results showed that the indicator parameters in the wastewater exceeded the standard permissible limits established by the World Health Organization 25 for industrial wastewater. This informed the decision to treat the tannery wastewater prior to discharge into the environment using kaolin and ZnO/kaolin nanocomposites as adsorbents. Figure 2 depicts the X-ray diffraction (XRD) pattern of the ZnO/kaolin nanocomposites prepared at pH (a) 6 (b) 8 (c) 10 and (d) 12     www.nature.com/scientificreports/ concentric ring pattern corresponding to each peak observed in the XRD pattern. shape ZnO nanoparticles homogeneously dispersed on the surface of the kaolin. HRTEM micrograph shown in Fig. 4c indicates successful coverage and deposition of spherical nanoparticles of ZnO on the silica-alumina interlayers spaces of kaolinites. The average size distribution of ZnO nanoparticles was within 21.34 nm. It was also noticed that the hexagonal and plate-like morphology of kaolin disappeared and transformed to spherical shapes in the composites. This shows that Zn with smaller ionic radius (139 pm) was responsible for the transformation of Si-Al dominant kaolinite network with an ionic radius of 210 pm and 184 pm respectively. Thus suggesting that the diffusion mechanism of Zn onto the kaolinite structures was responsible for the observed morphological changes. Figure 4c also indicates the presence of pores which aids the diffusion of metal ions and other pollutants during the adsorption process. The increase in porosity of kaolin/ZnO compared to kaolin alone as observed in HRTEM images may suggest higher removal efficiency of the pollutants by the former than the latter. The selective area electron diffraction (SAED) pattern of kaolin/ZnO nanocomposites (Fig. 4d) demonstrated the presence of a bright spot surrounded by sharp and intense concentric rings corresponding to the diffraction peaks noticed in the XRD pattern. Energy dispersive spectroscopy (EDX) analysis was carried out to confirm the presence and spatial distribution of Zn in the composites material. The EDX result shows the presence of Na, Al, O, Si, Fe and Zn elements (as seen in Fig. 5) in different amount. It can be seen that the dominant elements were Zn, O, Si and Al. Si and Al originated from kaolin while Zn and O were from the zinc salt precursors and other oxygen-containing species used for the synthesis. High percentage proportion of Zn 2+ as shown in the EDX result implies evenly distribution of Zn in the composite matrix. While the EDX result suggests that other elements such as Si, Al were randomly impregnated into ZnO layers. A similar observation was reported by Singh et al. 28 who embedded Fe 3 O 4 onto lattice layers of ZnO nanoparticles. Thus, immobilization of ZnO nanoparticles on kaolin matrix prevents agglomeration of nanoparticles, leading to large and active surface sites of the nanocomposite for the adsorption process. BET analysis. The N 2 adsorption-desorption isotherms of kaolin and kaolin/ZnO nanocomposites are shown in Figs. 6 and 7, respectively. The specific surface areas and pore diameter calculated by BET and Barret-Joyner Halenda methods for kaolin was 17 m 2 /g and 3.59 nm respectively. While that of kaolin/ZnO nanocomposites was 31.8 m 2 /g and 4.81 nm. It can be noticed that the surface area of kaolin/ZnO nanocomposites is much higher than that of kaolin alone, evidence of more binding sites in the former than the latter. One of the most important properties of adsorbent is surface area, the higher the surface area of kaolin/ZnO, the larger its adsorptive capacity. This suggests that the addition of porous and polar ZnO nanoparticles which further open the pores on kaolin was responsible for the increased surface area of the composites material. Also, Figs. 6 and 7, displayed a typical Type IV isotherm accompanied by Type H3 hysteresis loop according to IUPAC classification and suggest that the prepared adsorbents were predominantly mesoporous in nature 29 . However, the mesoporous dominant property was higher in kaolin than kaolin/ZnO nanocomposites due to wide pores opening in the former compared to the latter with closed pore opening till P/P O of 0.7 (see Fig. 7). The differences in the extent of mesoporosity can be linked to the entrapment of ZnO nanoparticles by the kaolinite in kaolin/ZnO www.nature.com/scientificreports/ which block the pores compared to kaolin alone without ZnO nanoparticles. The prevalence of large mesopores in the two materials indicate that adsorption of N 2 gas by these samples proceeds is multilayer in nature. It is also obvious that the capillary condensation of gases vis-a-viz the amount of adsorbed gases gradually increases with increasing relative pressure (P/P o ) until approach unity. A closer look shows that the adsorption-desorption curve for both materials was stepwise and steeply slopy and more visible in kaolin than kaolin modified with ZnO composites. The pore size value strongly supports the finding that the kaolin/ZnO nanocomposites are mesoporous in nature. The pore volume vs pore diameter inserted in Fig. 7 revealed evenly distribution of ZnO nanoparticles onto the intercalated layers of kaolinite sample. The homogeneous distribution of ZnO nanoparticles may be linked to the slight decrease in the intensity of peaks observed in the range 5-100 nm for kaolin alone to 5-90 nm for kaolin/ZnO composites. The reduction in the pore size distribution of kaolin/ZnO composites compared to kaolin alone according to BJH curve may be due to entrapment of ZnO nanoparticles by kaolin and uncontrollable pore collapse of the kaolinite structure during calcination at 450 °C. It can also be observed the intensity of peak at 5 nm for kaolin in the BJH curve ( Fig. 6) reduced after doping with ZnO nanoparticles (Fig. 7) is also an indication of more active sites in the latter than the former. Thus, modification of kaolin with ZnO nanoparticles causes a significant increase in the surface area compared to kaolin alone. A similar trend was reported by Singh et al. 28 .   www.nature.com/scientificreports/ should be mentioned that below the binding energy of 200 eV, Zn exists in the following orbital types (Zn 3s, 3p and 3d). While two peaks of Zn found in the binding energies of 1,021 and 1,045 eV ascribed to spin-orbit coupling corresponds to Zn 2p 3/2 and Zn 2p 1/2 and further confirmed the existence of Zn in ZnO environment in the valence states of + 2. The energy differences between the doublet peaks (Zn 2p 3/2 and Zn 2p 1/2 orbitals) is 24 eV, which is similar to that of typical ZnO nanoparticles. This is in good agreement with previous studies 27, 30 . The XPS general survey of kaolin/ZnO nanocomposites displays in Fig. 10 revealed the following dominant elements and orbital types (Na 1s, Zn 2p, Zn 3s, Zn 3p, Zn 3d, Na 2s, Si 2s, C 1s, Al 2s and Al 3p) at different binding energies. The elements detected are key constituted in the samples except for Na which originated  www.nature.com/scientificreports/ from NaOH used as a precipitating agent. The detected carbon (C 1s) at the binding energy of 286.4 eV in the deconvoluted XPS spectra shown in Fig. 11a can be from two sources either (1) carbon adsorbed on the surface during the exposure of the samples to ambient temperature or (2) zinc acetate dihydrate precursor used for the synthesis of ZnO nanoparticles 31 . Besides, deconvoluted XPS spectra of Zn core electrons presented in Fig. 11b revealed the presence of two strong peaks at binding energies of 1,027.9 eV and 1,050.9 eV respectively compared to one peak observed for Zn in ZnO (Fig. 9a) at 1,021.3 eV. The strong doublet peaks found binding energies of at 1,027.9 eV and 1,050.9 eV were assigned to Zn 2p 3/2 and Zn 2p 1/2 earlier observed in Fig. 9a. These values are in agreement with the binding energies of Zn 2+ ion previously reported 32,33 . The shift in the binding energy of Zn to higher values in kaolin/ZnO relative to Zn in ZnO alone further suggest substitutional effects of Si or Al in kaolin with oxygen in ZnO, thereby creating surface defect and oxygen vacancies. Very importantly, despite the differences in the binding energies of Zn in kaolin/ZnO composites and Zn in ZnO nanoparticles, Zn (Zn 2p 3/2 and Zn 2p 1/2 ) still exists in the chemical states of + 2. This is because the energy differences between 1,050.9 and 1,027.9 eV, for Zn in kaolin/ZnO composites is 23 eV, which means immobilization of kaolin onto ZnO lattice layer did not change its oxidation state as well ss zincite phase of ZnO nanoparticles. The deconvoluted XPS spectra of Al 2p and Si 2p shown in displayed in Fig. 11c, d respectively. The Al and Si originated from the kaolin and base on the difference in the ionic radius, Zn (139 pm) and O (152 pm) could diffuse into the Si-Al framework with the ionic radius of 210 pm and 184 pm respectively. This diffusion mechanism of Zn onto the kaolin was responsible for the morphological transformation of hexagonal-shaped kaolin to spherical particles. The XPS result further corroborates the HRTEM result earlier shown in Fig. 4. As such, Al and Si will capture oxygen in competition with Zn and result in more oxygen vacancies.    (Fig. 13c), after 10 min, the amount of chloride sorbed by the two materials reduced. Thus, 15 min was considered as the optimum equilibrium time for the two adsorbents because the additional increase in the time did not increase the uptake rates of the pollutants due to the deposition of the pollutants on the surface of the adsorbents. The fast removal rate during the first 10 min was ascribed to the availability of binding sites on the surfaces of kaolin and kaolin/ZnO nanocomposites. The maximum COD, BOD and chlorides removed by kaolin/ZnO composites under the applied conditions were 93%, 96% and 91%. While the removal efficiency of the indicator parameters by kaolin alone were: 65%, 67% and 62%, respectively. Based on the results obtained, the removal of pollutants was rapid in the early stages and the reduction in scavenging of the metal ions at the later stage of the adsorption process was due to desorption and over-saturation or reduction of available active adsorption sites. It was noticed that at every contact time, kaolin/ZnO composites removed more of COD, BOD and chlorides than kaolin alone due to higher surface area (31.8 m 2 /g) of the former than the latter (kaolin 17 m 2 /g). A similar trend was observed in Fig. 14a, b on the removal of Fe and Cr ions from tannery wastewater by the two materials. It was found that the adsorbents (kaolin/ZnO composites, kaolin) removed more of Cr (75%, 60%) than Fe (69%, 50%) ions under the applied conditions. The mesoporosity nature of the kaolin also contributed to the improved and better performance of kaolin/ZnO composites than kaolin alone. The faster rate of adsorption of Cr ion, when compared to Fe ion, may www.nature.com/scientificreports/ be attributed to the fact that metal ion with smaller ionic radii diffuses faster than ion with larger ionic radii. According to literature, the ionic radii of Cr 6+ ion (0.052 nm) is smaller than Fe 3+ ion (0.0645 nm) which makes it to diffuse faster to the pores of the adsorbents as a result of its smaller size. A similar trend was observed by Ogbu et al. 34 and El-Naggar et al. 35 who independently utilized 50%:50% mixture of kaolinite and sawdust and kaolinite/smectite composites to remove (Cd and Pb ions) and Pb ion from simulated wastewater after 45 min and 120 min respectively. Singh et al. 28 reported and achieved 100% removal efficiency for Pb 2+ , Cd 2+ , As 3+ , and Hg 2+ using Fe 3 O 4 /ZnO composites from simulated wastewater after 24 h. Zolfaghari et al. 18 achieved 87% Pb removal rate within 120 min using ZnO/coated nanoporous carbon. The differences in the adsorptive performance of the materials and contact time can be linked to their surface areas and functional groups. While Singh et al. 28 reported surface area of 43 m 2 /g for Fe 3 O 4 /ZnO composites, in this study surface area of 31.8 m 2 /g was obtained for kaolin/ZnO composites. It is important to mention that 100% removal efficiency of the target pollutants was not achieved under the influence of contact time and can be due to the following. (1) presence of several scavengers or other interfering metal ions in the tannery wastewater which also competes for active surface sites compared to the work of Singh et al. 28 who reported 100% removal rate. Singh et al. used simulated wastewater that did not contain scavengers such as sulphates, carbonates, nitrates, phosphates and others. (2) the quantity of adsorbent used was small and did not cover the entire surface of wastewater and as such no proper interaction with the adsorbate. From the results, the efficiency of an adsorbent depends on its adsorption capacity and how fast metal ions can be removed from wastewater. In this regard, kaolin/ZnO nanocomposite was found to be more effective in removing pollutants compared to pure kaolin. Thus, the potential of immobilized ZnO on kaolin as nanocomposites adsorbent for the removal of metal ions from wastewater is demonstrated.

XPS analysis.
Effect of adsorbent dosage. The scavenging of metal ions depends on the nature and amount of the adsorbent. The effect of adsorbent dosage on removal of COD, BOD, chloride and heavy metals (Cr 6+ and Fe 3+ ) are shown in Figs. 14 and 15, respectively. The removal efficiency of metal ions, COD, BOD and chloride increased with increase in adsorbent dosage from 0.2 to 1.4 g which implies that there was an increase in adsorption of pollutants for kaolin and kaolin/ZnO nanocomposites. The simultaneous increase in metal ion uptake and other indicator parameters (COD, BOD and chloride) with increase dosage can be explained in terms of surface functionality, greater availability of more active or exchangeable sites and increased surface area of the adsorbents. The mechanism of removal of Cr(VI) and Fe(III) ions can be explained as follow: kaolin and Kaolin/ZnO composites possess negative charges and through electrostatic interactions, the positively charged species like Cr(VI) and Fe(III) ions, can be removed from tannery wastewater. Similar explanation hold for COD, BOD and chloride. The polar and porous network structure of ZnO nanoparticles which provides more active sites for the adsorption of metal ions may also be responsible for greater removal efficiency by kaolin/ZnO composites than kaolin alone. On the other hand, the high electronegativity values of Fe(III) (1.83) than that of Cr(VI) (1.66) did not suggest removal of more Fe(III) than Cr(VI). Instead more of Cr(VI) was removed by the two adsorbents than Fe(III). This shows that the mechanism of heavy metal ions adsorption from aqueous solution by the adsorbent did not depend on the electronegativity parameter but rather on surface areas and surface functionality 17 . The excellent 100% and 98% removal efficiency achieved for Cr(VI) and (Fe(II) at a higher dose of kaolin/ZnO composites compared to kaolin can be linked-to existence of synergetic effects between the two materials as evidence in the number of functional groups shown in Fig. 3. Also, at every dose of the adsorbents, kaolin/ZnO    www.nature.com/scientificreports/ nanocomposites adsorbed metal ions, COD, BOD and chloride at a higher rate than kaolin due to an increase in adsorption surface area and a decrease in diffusion path length leading to less aggregation of adsorption sites. The adsorption trend of metal ions uptake onto the adsorbents was in the order kaolin/ZnO nanocomposites > kaolin due to ZnO network structures which provides extra active binding sites on the former than the latter. The trend in this study corroborates the previous studies on the removal of Cr(VI) and other pollutants 9 .
The maximum uptake of some pollutants by the adsorbents was compared with other adsorbents in literature as shown in Table 1. Taking into account that is cumbersome to compare the adsorption capacities of the tannery wastewater with different adsorbents, since the tannery wastewater depends on its constituents. The prepared and synthesized kaolin/ZnO nanocomposites in this study show its suitability for use as an adsorbent for the removal of the metal ions and other pollutants from wastewater. The trend obtained is evidence of a direct relationship between percentage removal efficiency and temperature. This implies that the sorption of the pollutants onto kaolin and kaolin/ZnO adsorbents is endothermic, and can equally be referred to as heterogeneous and reversible process. Comparatively, it was found that the rate of adsorption of metal ions and other pollutants followed the order kaolin/ZnO composites > kaolin. This result clearly illustrated that high surface coverage, high surface area, creation of more active and reactive sites as a result of anchoring of kaolin onto ZnO nanoparticles improved the adsorption of the nanocomposites in this study. The effect of temperature on the adsorption of COD, BOD, chloride, Cr(VI) and Fe(III) leading to estimation of Gibbs free energy, entropy and enthalpy of the reaction is shown in Table 3. where K j is a Jovanovic constant and q e(max) is the maximum adsorption capacity pollutants uptake. The relationship between lnq e and C e for the removal of COD, BOD, chloride and metal ions give straight lines. The results of the slope ( K j ), q e , and correlation coefficient (R 2 ) are given in Table 2. The calculated constants for the Jovanovic isotherm model presented in Table 2 show a slight increase in correlation coefficient values (R 2 > 0.99). This indicates that monolayer adsorption is the major mechanism for the removal of pollutants using kaolin and kaolin/ZnO nanocomposites.

Adsorption isotherm studies.
Halsey isotherm. Halsey adsorption isotherm models can be used for the discussion of the multilayer adsorption at a relatively large distance from the surface. The fitting of the results is responsible for the porosity nature of the nanoadsorbent. Halsey adsorption isotherm model is given in Eq. (4).
where n H and I n K H Halsey isotherm constants. The plots of q e against lnC e for the removal of COD, BOD, chloride and metal ions onto kaolin and kaolin/ZnO nanocomposites describe Halsey model isotherm. The   Table 2. It can be seen from Table 2 that the R 2 values are > 0.98 but less than R 2 values of Jovanovic isotherm. The experimental data showed that there is a lower likelihood of multilayer adsorption on the nanoadsorbents. The low correlation coefficient values of the Halsey adsorption isotherm indicate that it is not a good fit for the experimental data for pollutant removal using these adsorbents.   where K FH is Flory-Huggins constant and n is the numerical size parameter of pollutants occupying sorption sites. The K FH positive values obtained from the intercepts represent Flory-Huggins isotherm for these adsorbents (as seen in Table 2), indicating that the Flory-Huggins isotherm model is applicable for modelling the adsorption of pollutants onto the adsorbents. This shows that the formation of multilayer of adsorbate molecules on the adsorbent surfaces can be predicted. Thus, the value of n less than 1 shows that the adsorbate molecule will occupy more than one active site.
Redlich-Peterson isotherm. The Redlich-Peterson isotherm model incorporated the hybrid features of Langmuir and Freundlich isotherms. In this present study, this model describes the mono and multilayer adsorption of pollutants from aqueous system using kaolin and kaolin/ZnO nanocomposites. The linear R-P isotherm is given in Eq. (7).
where β and A are the Redlich-Peterson constants. The plots of ln C e q e against lnC e for the removal of COD, BOD, chloride and metal ions from tannery wastewater describe the Redlich-Peterson isotherm. From the Redlich-Peterson plot, the slope and the intercept were determined. The values of R 2 , β and A are presented in Table 2 with  Table 3. Thermodynamic parameters of some parameters in wastewater adsorption on kaolin and kaolin/ZnO at different temperature.

Parameter Sample
�G(kJ/mol) Thermodynamic studies. The thermodynamic parameters values were estimated to evaluate the feasibility of adsorption of metal ions onto the adsorbents. The distribution coefficient, K D , was calculated from Eq. (8) The Gibbs free energy, enthalpy and entropy at different temperatures were calculated from the Vant Hoff 's plot using Eq. (9).
where R is the universal gas constant (8.314 J/molK), T is the absolute temperature (K), S is the change in entropy (J/molK) and H is the change in enthalpy (J/mol). The change in free Gibbs energy ( G ) (kJ/mol) values were calculated from Eq. (10).
The thermodynamic parameters are listed in Table 3 and it can be observed that the H values were positive and further confirmed that the adsorption is an endothermic process. It can be noticed that the calculated H was less than 40 kJ/mol for the removal of the pollutants by the two adsorbents. This is evidence of physisorption dominance over chemisorptions process of the adsorbates onto the adsorbents 42 .
The positive values of S suggested an increased in the degree of randomness or disorderliness at the adsorbate-adsorbent interface. The disorderliness may be ascribed to the increment in the translational entropy of the displaced water molecules relative to that lost by the pollutants during adsorption process 42 .
The negative and positive values of G in the temperature ranges confirmed that the adsorption of metal ions onto adsorbents was spontaneous and non-spontaneous in nature. However, it was further confirmed that as the temperature increase, G reduces (becoming negative), indicating that the adsorption processes became spontaneous and highly favoured at high temperature.
Adsorption kinetics and mechanism. The data from the effect of contact time was used to examine the adsorption mechanism and kinetics of metal ions onto the kaolin and kaolin/ZnO nanocomposites by fitting the experimental data to the intraparticle diffusion equation 43 presented as follows: where q t is the amount of metal ions loaded onto the adsorbent at time t (mg/g), k i is the intra-particle diffusion (mg/g), t is the contact time (min) and I is the intercept. The intra-particle diffusion model parameters were calculated from the plot of q t against t at various contact time. The linear profile of the plots of the COD, BOD, chloride and metal ions did not pass through the origin. This is an indication that some degree of boundary layer controlled the removal of metal ions, COD, BOD and chloride onto the adsorbents. The deviation from the origin may be due to mass transfer and adsorption at the initial and final stages, respectively. The intercept, I, values are greater than 0 as presented in Table 4, indicating that the intra-particle diffusion or internal diffusion was not only the rate-controlling step, but the film diffusion may also play an important role in the adsorption  Figure 18. Steps in the adsorption mechanism. where q t and q e in (mg/g) are the amount of metal ion uptake at the time and maximum equilibrium uptake. F is the fraction of metal ion adsorbed at the time, t.
B t is the time constant and is a function of F from the linear plot of B t versus time, t. The Boyd plots proved that the linear plots did not pass through the origin and show that the adsorption process was governed by film or external diffusion. Generally, during the adsorption of pollutants onto the adsorbents, three adsorption mechanistic steps occur and are depicted in Fig. 18.
Evaluation of the nanoadsorbent. Table 5 presents the comparison of the maximum adsorption capacities of different nanoadsorbents for the removal of Cr(VI) from wastewater solution. The high adsorption capacity (117.25 mg/g) showed that ZnO/kaolin nanocomposites was more effective nanoadsorbent for the removal of Cr(VI) from wastewater than previously reported nanoadsorbents. The lower contact time, high percentage removal, cost-effectiveness of the natural material used as support (kaolin), and non-toxic nature of the nanocomposites make this candidate promising and useful adsorbent. The demonstrated results from the study show the efficacy of ZnO/kaolin nanocomposites on industrial tannery wastewater and it can be employed to treat other industrial effluents. Therefore, it could be enlisted among effective nanoadsorbents for pollutants industrial wastewater treatment.   www.nature.com/scientificreports/ Reusability study. The reusability and stability of the nanoadsorbents are imperative regarding the commercial application. The behaviour of kaolin and kaolin/ZnO nanocomposites in tannery wastewater in repeated adsorption-desorption cycles was carried out for six successive runs. In this research, the adsorbents were allowed to settle and collected via filtration technique and then reused by washing with 0.1 M NaOH after the adsorption of Cr(VI) and Fe(III) from tannery wastewater. The resultant sample was first oven-dried at 105 °C for 2 h and then calcined at 450 °C in a furnace for 3 h prior use in the following runs. The experiment was repeated using the optimum conditions established in batch adsorption studies. Figures 19 and 20 show the regeneration potential of kaolin and kaolin/ZnO nanocomposites on the percentage removal efficiency of Fe and Cr ions after six successive runs. It can be noticed that the amount of Fe and Cr removed decreased slightly after the second cycle (90%, 95%) compared to the first cycle (98%, 100%) and then decreased significantly after the second run for kaolin/ZnO composites. In the repeated adsorption-desorption cycles, the removal efficiencies of the heavy metals ions slightly decreased compared to the removal of metal ions in the first and second cycles. This shows that the regenerated nanocomposites showed remarkably high adsorption efficiencies for both metal ions in the first, second cycles and subsequently compared to the used of kaolin alone Generally, regeneration of nanoadsorbents is partially inhibited as a result of the loss of nanoparticles and weakness of the supporting material in the case of kaolin/ZnO mixtures. Hence, the reduction in nanoadsorbent shelf life is proportional to the adsorption removal efficiency of nanoadsorbents. The results show the good recyclability potentials of kaolin/ZnO nanocomposites for the removal of Cr and Fe ions in tannery wastewater via adsorption than kaolin alone. Yu et al. 22 also reported excellent reusability and stability potential of 30% ZnO/biochar composite for the removal of Cr(VI) from aqueous solution after 5 cycles.

Conclusions
The present study proves that kaolin and kaolin/ZnO nanocomposites could serve as effective adsorbents for the sequestration of Cr, Fe, COD, BOD and chloride from tannery wastewater. The adsorbents were analysed by XRD, FTIR, BET, HRTEM, SAED, EDX and XPS. The adsorption process was found to be dependent on the contact time, adsorbent dosage and temperature. Kaolin/ZnO composites exhibited excellent adsorptive behaviour under the applied conditions than kaolin alone based on the differences in surface area. The optimum conditions for the removal of Cr, Fe, COD, BOD and chloride varies and the differences were linked to the nature of the pollutants in tannery wastewater. The equilibrium adsorption data could be best described by Jovanovic adsorption isotherm model. Adsorption kinetic results indicated that both intra-particle and bulk diffusion were the rate-determining steps. Thermodynamic parameters revealed that the adsorption processes were endothermic in nature and non-spontaneous. The results show the practical applicability of kaolin/ZnO nanocomposites for the adsorption of pollutants from tannery wastewater thus making the adsorbent a promising candidate in the field of water and wastewater management.