Novel synthesis of Ni/Fe layered double hydroxides using urea and glycerol and their enhanced adsorption behavior for Cr(VI) removal

Novel modified Ni/Fe layered double hydroxides with different morphology of spherical – like shape were fabricated via using urea as a ligand and glycerol (Ni/Fe LDH/GL) with Ni:Fe molar ratios of 2:1 by the simplest co -precipitation method. Also, for comparison purposes, Ni/Fe LDH was synthesized to be used as a control one. A suggested interpretation for the morphology change was also given. The materials were characterized by X-ray diffraction (XRD), The Fourier transform infrared (FT - IR) spectroscopy, field emission scanning electron microscopy (FESEM), EDX for elemental analysis, high resolution transmission electron microscopy (HRTEM), Brunauer, Emmett, and Teller (BET) equation, particle size distributions and Zeta potential measurements. In addition, the synthesized materials were used as adsorbents for removal of potassium dichromate from aqueous solutions under various experimental conditions. The adsorption of Cr (VI) was strongly pH dependant and the pHPZC was studied. Kinetic studies were evaluated through different models including, pseudo first and second orders, mixed 1, 2 orders, intra particle diffusion and Avrami models. For adsorption isotherms, two-parameter models (Langmuir, Freundlich and Temkin) and three parameter models (Sips, Langmuir-Freundlich and Tooth) were investigated showing maximum adsorption capacity of 50.43 mg/g and 136.05 mg/g for Ni/Fe LDH and Ni/Fe LDH/GL, respectively. Also, the effect of temperature was investigated at (23, 35, 45, 55 °C) and the thermodynamic parameters (∆H°, ∆S° and ∆G°) were calculated showing exothermic and spontaneous adsorption process. The effect of coexisting anions (Cl−, SO42− and HPO42−) and humic acid at different concentrations on the removal efficiency of dichromate ions was investigated. Chemical stability and recyclability of these adsorbents were also studied. The intermolecular hydrogen bonds formation between dichromate ion, urea, glycerol, LDH was explored by Monte Carlo simulation This study suggested that the modified Ni/Fe LDH/GL materials were promising nanoadsorbents for efficient potassium dichromate removal.


Results and Discussions
Characterization of Ni/Fe LDH and Ni/Fe LDH/GL Nanocomposite. Crystal structure. Figure 1(a,b) displays the powder X-ray patterns of the prepared Ni/Fe LDH and Ni/Fe LDH/GL Nanocomposite. Both samples show reflection peaks typical of the layered double hydroxide structure. The diffraction peaks matched those in the reported in the reference data (ICDD card no. 01-082-8040). The reflections peaks at 11.6°, 23.2°, 34.5°, 39.1°, 46.5°, 59.9°, and 60.9° could be indexed to the plane families (003), (006), (012), (015), (018), (110), and (0015) respectively. The samples possessed a Rhombohedral crystal system with R-3 space group (group number 148). The (003) interlayer distance was calculated to be (7.82 Å) for the Ni/Fe LDH which is close to the literature value 37,45 . After Glycerol loading the (003) interlayer distance did not show any significant change suggesting that the Glycerol was adsorbed on the surface of the LDH layers rather than being intercalated between the layers.
The crystallite size for Ni/Fe LDH and Ni/Fe LDH/GL was calculated to be 10.92 nm and 12.75 nm respectively and this confirm the loading of glycerol. As shown in Fig. 1(c,d), after adsorption the samples still show the same characteristic diffraction peaks compared to before adsorption and this illustrates the stability of the Ni/Fe LDH and Ni/Fe LDH/GL adsorbents used after the adsorption tests. Figure S1 represents the XRD of Ni/Fe LDH prepared by urea and NaOH.
The FT-IR spectra of the Ni/Fe LDH, Glycerol and Ni/Fe LDH/GL Nanocomposite are shown in Fig. 2(a-c) The chemical bonds in the Ni/Fe-NO 3 LDH were identified by the band that is pointed at 787 cm −1 and attributed to M-O-M vibration 46,47 . This band like the M-O-H bending 47 , involved the oxygen metal ions translational motion in the brucite -like layers 48 . The strong broad band at 3398 cm −1 was related to the stretching vibrations of the H-bond of the OH group (ν O-H) in the brucite-like layers 49,50 . The bending vibration (δ H 2 O) of the H 2 O molecules in the interlayers 49,51 appeared at 1629 cm −1 . The peak located at 1357 cm −1 is related to the ν stretching www.nature.com/scientificreports www.nature.com/scientificreports/ vibration of the NO 3 groups in the LDH interlayer. As For glycerol, The peak at 3279 cm −1 can be attributed to the O-H bond stretching while the bimodal peak around 2900 cm −1 is due to the C-H stretching [52][53][54] . The peak at 1423 cm −1 is due to the bending of C-O-H group. The peak at 1035 cm −1 and the other peaks in the low wave number region (1100-900 cm −1 ) are attributed to C-O stretching [52][53][54] while that at 1220 cm −1 is due to CH2 bending 54 . As Shown in Fig. 2(c) it is interesting to note the presence of important bands of GL in the spectra of Ni/Fe LDH/GL Nano Composites indicated a successful loading of Gl in the Ni/Fe LDH host 55,56 . Besides Also, the intensity of some peak at 1357 cm −1 was decreased indicating the exchange of nitrate anions by the GL 56 . The main absorption peak at 3398 cm −1 in Ni/Fe LDH has shifted to higher wavelength 3420 cm −1 in Ni/Fe LDH/ GL indicating the promotion of hydrogen bonding interactions between Ni/Fe LDH and glycerol 53 . The overlapping of peak at 2900 cm −1 with the peak at 3398 cm −1 and the presence of the two absorption peaks at 1112 and 1057 cm −1 (Fig. 2) in the Ni/Fe LDH/GL nanocomposite related to C-O stretching indicates the successful loading of the glycerol onto the LDH phase.  Morphology of Ni/Fe LDH and Ni/Fe LDH/GL. The morphologies of the synthesized materials were mainly affected by the experimental conditions, such as the ratio of H 2 O to GL, amount of urea used, ratio of Ni/Fe and the reaction temperature 57 . Firstly, the effect of the GL on the morphology of the prepared materials was shown in Fig. 3. It is clearly appear that without the addition of GL ( Fig. 3(a)), two kinds of morphology, sphere-like and plate-like structure can be obtained. The formation of Ni/Fe-LDH microsphere was mainly through the using of urea which forming binary complex with Ni and Fe metal ions 58 which confirmed by the appearance of a peak at 1488 cm −1 and this peak is related to N-C-N of urea 59 as shown in Fig. S2 which represents the FTIR of Ni/Fe LDHs prepared by urea and by NaOH. after that, they converted to Ni/Fe LDH by olation reactions and then crystallization 60,61 . Meanwhile, the interaction between urea and (Ni and Fe) metal ions may be the reason of gradual morphology change from layers into the microspheres structure to decrease the surface energy 61 . Furthermore, this like morphology can expose the active site channel for the prepared adsorbate, which was beneficial to the improvement of the adsorption capacity. By addition of GL (H 2 O: GL 1:4.5), the plate shape like morphology decreased and only sphere-like samples were observed ( Fig. 3(b)) due to the aggregation with smooth layer by-layer nanoplates structure. Slow precipitation can be occurred due to the interaction through complex formation of urea with metal ions causing gradual LDH formation which give a chance for glycerol molecules to form further coordination bonds with nickel and iron metal ions and these glycerol molecules form vesicles that act as a template for LDH so sphere like structure appear. Finally, both urea and GL have essential roles in the morphological conversion 58 . The EDX analysis of Ni/Fe LDH and Ni/Fe LDH/GL is demonstrated in Fig. 3(f,g). The corresponding HRTEM image of Ni/Fe LDH (Fig. 3(d)) reveals that there are plenty of thin nanoflakes aggregates on each other, this phenomenon related to the coprecipitation method of preparation 62 . The image of Ni/Fe LDH/GL was also given in Fig. 3(e) which also showed strong evidences of loading of GL on LDH surface. www.nature.com/scientificreports www.nature.com/scientificreports/ Surface area and pore size distribution. The surface area and porosity of the Ni/Fe LDH and Ni/Fe LDH/GL were studied using nitrogen adsorption/desorption isotherms. The measured BET surface area of Ni/Fe LDH/GL was 74.13 m 2 /g. While of Ni/Fe LDH was 34.21 m 2 /g. It can be seen that introducing glycerol to the Ni/Fe LDH result in an obvious increase in surface area and porosity. It is apparent from the adsorption and desorption curves ( Fig. 4(a)) that Ni/Fe LDH exhibits a isotherm type IV and H2 hysteresis shape according IUPAC classification 63 , the desorption branch is likely to be dependent on network and/or 'ink bottle' effects 64 which is suggested to be due to mesopores and capillary condensation 65 . After modification, in case of Ni/Fe LDH/GL, It was observed that there is a narrow adsorption-desorption hysteresis loop with relative pressure above 0.4, it turned to type-H3 hysteresis loop with slit shape pores ascribed to open end pores and ink-bottle pores 66 . Table 1 summarized the results of surface texture. In order to design a specific nanomaterial, controlling pore size, area and distribution will achieve our target 67 . Figure 4(b) present micro-and mesoporous distribution with the peak pore size of 1.6, 2.63 nm (Ni/Fe LDH/GL) and 2.52 nm (Ni/Fe LDH), respectively. The SSA of Ni/Fe LDH is attributed to the mesopores while the increase in the SSA of Ni/Fe LDH/GL resulted from accumulation of the recently developed micropores and the existing mesopores. As shown in Table 1, the high SSA of Ni/Fe LDH/GL attributed mainly to surface area of micropores and mespores while SSA of Ni/Fe LDH attributed only to surface area of mespores.
Zeta potential of materials is an important property that explains many water treatment mechanisms. It is a measure of the magnitude of the charge repulsion/attraction between particles in solution. It is one of the main factors known to affect suspension stability, and provides details about the causes of dispersion or aggregation. The zeta potentials of aqueous dispersions of the Ni/Fe LDH and Ni/Fe LDH/GL were 43.3 and 32.8 mV, respectively, at pH 6 ( Fig. S3(c,d)). The results show that the low positive zeta potential value of the Ni/Fe LDH/GL nanocomposite can be due to the interactions between the inorganic material and organic molecules (GL). This decrease is expected to be based on electrostatic considerations. The size distribution intensities of the Ni/Fe LDH and Ni/Fe LDH/GL were 270 and 469 nm, respectively as shown in Fig. S3(a,b).
Adsorption study. Effect of adsorption pH. The optimum pH value for dichromate uptake from water can be obtained through studying the adsorption process over a range of pHs. The pH of the solution has a great effect on the adsorption process of the dichromate anions because an hexavalent chromium solution has different species at different pH values according to literature as chromate (CrO 4 2− ), dichromate (Cr 2 O 7 2− ), hydrogen chromate (HCrO 4 − ), chromic acid (H 2 CrO 4 ) and others 68 . Also, LDH characteristics and structure are greatly affected by pH value where, at lower pHs (<3) LDH disordering may be occurred 69 . As shown in Fig. 5(a), the effect of pH on adsorption process was studied over pH range from 4 to 9. The Cr(VI) uptake efficiency decreased with increasing of pH value and pH 5 was found to be the optimum value for adsorption of dichromate anions.   68 and there is high concentration of H + ions which neutralize the adsorbent surface and hence decrease the hindrance to chromium penetration 70 . At higher pH level the significant species is (CrO 4 2− ) however, the absorption percentage decrease this may be due to high concentration of OH − ions (basic medium) which compete with negatively charged Cr(VI) species for the active sites of the adsorbent surface 71 . The charges of the material surfaces and the ionization of the adsorbates can be greatly affected by the pH of the solution 72 potassium dichromate has 2 values of pKa (strongest acidic pKa = −2.3) and (strongest basic pKa = −6.1) and this means that when pH values are lower or higher than pKa values of dichromate it becomes positively or negatively charged, respectively. The pH PZC is the pH at which the material is neutral and has equal positive and negative charges 73 and the material surface will be positive at pHs lower than the pH PZC and negative at pHs higher than the pH PZC here, in this study the pH PZC was 6.8 ( Fig. 6) so the LDH surfaces were occupied by positive charges at pHs less than 6.8. From the experimental data, as shown in Fig. 5(a), the maximum adsorption capacity of chromium was found at pH in the range from 5 to 7, which may be due to the electrostatic attraction between the opposite charges of dichromate and LDH materials. Adsorption of amount of chromium ions at pH higher than pH PZC (LDH were negatively charged) indicate that adsorption not only due to electrostatic attraction effect but also may be due to hydrogen bond formation and ion exchange [74][75][76] .
Effect of adsorbent dose. The uptake of Cr(VI) increased with increasing Ni/Fe LDH and Ni/Fe LDH/GL concentration up to 0.1 gm and this may be accounted for increasing available active sits for adsorption ( Fig. 5(b)). The absorptivity (%) in case of Ni/Fe LDH/GL was higher than that of Ni/Fe LDH where, at 0.01 gm of both catalysts the adsorption efficiency reached 88.87% and 34.44% and at 0.1 gm it reached 93% and 84.89% for Ni/Fe LDH/GL and Ni/Fe LDH, respectively and this may be related to introducing glycerol molecules during preparation which causes an obvious increase in surface area as shown in BET measurements, and also provide more active sites including oxygen-containing functional groups such as -OH, C-O, and C-O-H as demonstrated in FT-IR analysis so Cr (VI) ions can be removed by hydrogen bond formation and complexation with these oxygen-containing groups 77 .  www.nature.com/scientificreports www.nature.com/scientificreports/ Adsorption isotherm studies. Adsorption isotherm represents the amount of adsorbate per unit mass of the catalyst as a function of its concentration at constant temperature and pH 78 . It can be used to calculate the maximum adsorption capacity of dichromate ions and describe how substances are adsorbed onto the surface of the catalyst 79 . as shown in Fig. 7, for both catalysts the amount of adsorbed dichromate ions increase by increasing its concentration because high initial dichromate concentration give a chance for each active site in the catalyst to be surrounded by more dichromate ions so more adsorption would be achieved, but the adsorbed amount in case of Ni/Fe LDH/GL is more than that of Ni/Fe LDH indicating that more active sites were provided by introducing glycerol during the preparation of LDH. In this study the most popular models (Langmuir, freundlich, Temkin, Sips, Toth and Langmuir-Freundlich) were be applied for the explanation of adsorption isotherms. For the two parameter models, the Langmuir model assumes that adsorption takes place at specific identical adsorption sites localized on the surface of the catalyst covering it with a monolayer of the adsorbate 3 . Langmuir adsorption isotherm is given by Eq. (1) where, C e (mg/l) is the concentration at equilibrium, q e (mg/g) is the amount of the adsorbed molecules per unit mass of the catalyst, q max (mg/g) is the maximum adsorption capacity and K L is Langmuir constant related to the adsorption rate. The separation factor R L that determine the affinity between adsorbent and adsorbate can be calculated by applying Eq. (2) 80,81 .
where, K L (l/mg) is the Langmuir constant and C o (mg/l) is the initial concentration of dichromate ions. by knowing R L parameter value we can predict whether the shape of the isotherm is favorable (0 < R L < 1), unfavorable (R L > 1), linear (R L = 1) or irreversible (R L = 0) 82 where, Q e (mg/g) is the amount of adsorbed substances, C e (mg/l) is the concentration at equilibrium, K F (mg/g) is freundlich constant that gives information about adsorption capacity of adsorbed molecules onto catalyst www.nature.com/scientificreports www.nature.com/scientificreports/ surface 15 and n is freundlich constant. 1/n give indication about adsorption intensity and surface heterogeneity 83 . 1/n usually < 1, if favorable adsorption 84 . Temkin model suggest uniform distribution of binding energies (up to maximum binding energy) and linearly decreasing of adsorption heat of particles in the layer regardless very low and very high concentrations 85 . Temkin adsorption isotherm is given by using Eqs. (4-7).
where, A T is Temkin isotherm equilibrium binding constant (l.g −1 ), b T is Temkin isotherm constant, R is universal gas constant (8.314 J.mol −1 .K −1 ), T is temperature at 298 K and B is constant related to heat of sorption (J.mol −1 ). Three parameter models (Sips, Langmuir-freundlich and Toth) were also applied. Sips isotherm model represents a mix of Langmuir and freundlich isotherm models. The parameter of this model are governed by experimental conditions (pH, temperature and concentration) 86 at high concentration of adsorbate, it indicates a monolayer adsorption related to Langmuir isotherm while, at low concentration it reduces to freundlich isotherm 86  where, q max is the adsorption capacity (mg/g), K s is sips constant related to adsorption energy and 1/n is sips isotherm exponent. Toth isotherm model is a modified form of Langmuir model and it is principally used for predicting the systems which cover both high and low adsorbate concentration 87,88 . Toth model can be presented by using Eq. (9) 87 .
e e e L n n 1/ where, K l and n are Toth isotherm constants (mg/g) the parameter n give an indication of heterogeneity of the system where being more than unity the system is considered to be heterogeneous 88,89 . Langmuir-Freundlich isotherm model give an indication about heterogeneous surfaces and the distribution of adsorption energy on these surfaces 90 . At low concentration of adsorbate the system follows freundlich model while at high concentration of adsorbate it follows Langmuir model 91 . This model is expressed by Eq. (10). Adsorption kinetics. The effect of time on the removal efficiency was evaluated through applying pseudo first 100,101 and second 102-104 order, mixed 1,2 order 105 , intraparticle diffusion 106,107 and Avrami 108-110 models, as described by Eqs. (11)(12)(13)(14)(15), respectively. All parameters related to these kinetic models are reported in Table 4. From the experimental data demonstrated in Fig. 8, it is obvious that through the first 10 min there was a fast increasing in the up take efficiency which may be accounted for the more available vacant active sites then, it showed less and gradually adsorption behavior until 30 min after that, saturation and equilibrium is reached with no change in removal percent up to 360 min. for both Ni/Fe LDH and Ni/Fe LDH/GL, the adsorption process was best fitted with pseudo-first order, Avrami model, pseudo second order and 1,2 mixed order models as presented in R 2 shown in Table 4 also it was modest fitted with intraparticle diffusion model with R 2 (0.715 and 0.699), respectively.   where, q t is the adsorption capacity (mg g −1 ) at time t (min), q e is the adsorption capacity (mg g −1 ) at equilibrium, k 1 is the pseudo first order rate constant (min −1 ), k 2 is the pseudo second order rate constant (min −1 ), f 2 is the mixed 1,2 order coefficient (dimensionless), k is the adsorption rate constant (mg g −1 min −1 ), k ip is the measure of diffusion coefficient (mg g −1 min −1(1/2) ), c ip is the interaparticle diffusion constant (mg g −1 ), k av is the Avrami rate constant (min −1 ) and n av is the Avrami component (dimensionless).
Thermodynamic study. The effect of temperature on the Cr(VI) ions adsorption on Ni/Fe LDH and Ni/Fe LDH/ GL was evaluated at 23 °C, 35 °C, 45 °C and 55 °C. Figure 9 shows the effect of temperature on the chromium ions removal efficiency. The experimental results and k d (q e /c e ) values at different temperature were employed to calculate the thermodynamic parameters (∆H°, ∆S° and ∆G°) by using Eq. (16) to predict the adsorption process nature 111,112 .  Table 4. Represent the kinetic models parameters. The thermodynamic parameters of this system are reported in Table 5. Obtaining ∆H° in negative value predict an exothermic adsorption process so increasing the temperature of the system does not usually favor the process 15 . Entropy (∆S°) was reported in a negative value and this shows a decrease in randomness at the interface between solid and liquid in the adsorption of dichromate ions on LDH materials 113 . The negative values of Gibbs free energy (∆G°) show a spontaneous adsorption process. From the experimental data, increasing the temperature causes increasing in ∆G° values (decreasing in negativity) i.e. decreasing the degree of spontaneity and this may be related to the exothermic behavior of the reaction between Cr (VI) ions and both Ni/Fe LDH and Ni/Fe LDH/GL 114 .
Effect of coexisting anions and HA. Figure 10 represents the effect of interfering anions such as Cl − , SO 4 2− , and HPO 4 2− with various concentrations of 25, 50 and 100 ppm on the adsorpitivity of dichromate ions. As shown, increasing the concentration of chloride anions caused no significant difference in the adsorpitivity of Cr (VI) ions onto Ni/Fe LDH while, it caused a great decreasing in removal efficiency in case of Ni/Fe LDH/GL. This may be related to competitive behavior between chloride anions and dichromate to be adsorbed onto the active sites of LDH materials. sulphate anions had a more pronounced impact on the adsorption capacity where it made the adsorpitivity of dichromate ions to be lowered from 90.48% to 78.06% and from 88.88% to 22.9% onto Ni/Fe LDH and Ni/Fe LDH/Gl, respectively when it was added with concentration of 100 ppm and this impact can be attributed to greatly competitive effect of high charged anions through electrostatic attraction to the catalyst 115 . Also, difference between the anions effect on the adsorption efficiency may related to the value of Z/r (charge/ radius) for SO 4 2− is bigger, the tendency of SO 4 2− uptake on active sites of used adsorbents is higher and less Cr(VI) adsorption. On the other hand, in basic environment, at optimum pH phosphate anion will be transformed to HPO 4 2− and PO 4 3− ions which being have a charge density higher than nitrate anions resulting in a lower adsorption efficiency for Cr(VI). The same results are reported in previous studies 116 .

T (K) ∆G°(kJ/mol) ∆H°(kJ/mol) ∆S°(J/K mol)
Ni/Fe LDH  www.nature.com/scientificreports www.nature.com/scientificreports/ Organic humic acid can be found together with Cr (VI) ions in wastewater. As shown in Fig. 11 increasing the concentration of humic acid in the solutions negatively affected the adsorption capacity of dichromate ions and this trend can be accounted for two reasons: firstly, the available reactive sites of LDH materials were consumed by humic acid molecules i.e. HA competed with dichromate ions to be adsorbed onto LDH material. Secondly, the adsorbed humic acid particles promoted the electrostatic repulsion and hence prevented the mass transfer of negatively charged dichromate ions 117 .
Adsorbents chemical stability. The chemical stability of the prepared LDH materials was tested by analyzing XRD patterns of these adsorbents at different pHs (3)(4)(5)(6)(7)(8)(9). For Ni/Fe LDH (Fig. 12(a)) the intensities of bands were greatly weaker than that shown in Fig. 1(a) especially that at pH = 3 which possessed no crystal structure. However, for Ni/Fe LDH/GL nanocomposite ( Fig. 12(b)), all band reflections were not affected at different pHs showing the same band intensities compared to XRD patterns Fig. 1(b) and exhibited intact layer-by-layer ordered structures even at low pHs. Finally, the addition of glycerol to LDH adsorbents obviously improved their stability.
Recyclability of the Nano-adsorbents. It is of great important to determine the reusability performance of the exhausted adsorbents to evaluate their economic situation. As shown in Fig. 13, after 7 cycles of regeneration, the adsorption capacity was not greatly changed where, Ni/Fe LDH maintain about 88.2% of its removal efficiency after 5 cycles and about 85.2% after 7 cycles while, Ni/Fe LDH/GL maintain about 90% of its removal capacity after 5 cycles and about 89.3% after 7 cycles and this means that these materials have a good regeneration behavior.
Adsorption mechanisms. From FT-IR analysis after adsorption (Fig. 2) one can conclude the mechanisms of adsorption of dichromate anions onto LDH materials. The O-H absorption bands of Ni/Fe LDH and Ni/Fe LDH/ GL were shifted from 3398 to 3387 cm −1 and from 3420 to 3408 cm −1 , respectively after adsorption of Cr(VI)  www.nature.com/scientificreports www.nature.com/scientificreports/ ions indicating the formation of hydrogen bonds between OH groups on the surface of LDHs and the oxygen of dichromate ions 56 so Cr (VI) ions can be removed by complexation with hydroxyl groups. the metal oxide band at 787 cm −1 was shifted to 777 cm −1 in Ni/Fe LDH and disappeared in Ni/Fe LDH/GL because of the electrostatic attraction between negatively charged dichromate ions and higher valence metal ions (Ni +2 , Fe +3 ) 56,118 and this can be confirmed by zeta potential measurements. the intensity of nitrate absorption bands located at 1357 in Ni/ Fe LDH and Ni/Fe LDH/GL after adsorption were decreased and this is attributed to the intercalated of dichromate anions by the ion exchange of NO 3 − groups 119 The results suggested that Cr (VI) ions were absorbed onto LDH through electrostatic attraction, ion exchange, complexation and hydrogen bond formation. Figure 14(a-f) displays the interactions of urea and glycerol molecules with the Ni-Fe LDH surface. It can be seen form Fig. 14(a,b) that the nitrogen or oxygen atoms of urea molecules formed intermolecular hydrogen bonds with the hydroxyl hydrogen atoms of the LDH surface. A net of hydrogen bonds among the urea molecules were formed, in which each carbonyl oxygen atom formed two hydrogen bonds with both (-NH 2 ) groups. Similarly, the hydroxyl oxygen atoms of glycerol molecules formed hydrogen bonds with LDH, Fig. 14(b,c). Intermolecular hydrogen bonds were formed among glycerol hydroxyl groups. Additionally, intramolecular hydrogen bonds were formed between the glycerol hydroxyl groups. Figure 14(e,f) shows the interactions of 4:1 (glycerol: urea) molar ratio with Ni-Fe LDH surface. Beside the hydrogen bonds between glycerol or urea with LDH surface, intermolecular hydrogen bonds were also formed between urea and glycerol molecules.  www.nature.com/scientificreports www.nature.com/scientificreports/ The adsorption of a dichromate ion on the LDH, urea-LDH, and 4:1 (glycerol/ urea)-LDH surfaces are shown in Fig. 15(a-c). The dichromate oxygen atoms were found to form five hydrogen bonds with the hydroxyl hydrogen atoms of LDH (Fig. 15a). In case of dichromate adsorption on urea-LDH, the dichromate oxygen atoms formed three hydrogen bonds with the LDH and another three hydrogen bonds with urea hydrogen atoms (Fig. 15b). While in case of dichromate adsorption on glycerol-urea LDH, the dichromate ion formed three hydrogen bonds with the LDH surface, two hydrogen bonds with two glycerol molecules, and a hydrogen bond with urea. Form the-above discussion, one can conclude that the presence of urea/glycerol molecules on the LDH surface enhances the dichromate adsorption.

experimental Section
Materials. All used chemicals (Table S1)  Ni/Fe LDH/GL. Ni/Fe LDH/GL with molar ratio of 2:1 was synthesized by the same co-precipitation method. A 100 ml mixed solvent of glycerol and H 2 O with volume ratio of (4.5:1) was stirred for 10 minute at room temperature to form homogeneous solution. Nickel nitrate and ferric nitrate (molar ratio of 2:1) were dissolved in 50 ml of this mixed solvent and urea was dissolved in the 50 ml remain to form urea solution of 1.5 M. urea solution was added dropwise on metal ions solution under vigorous stirring at 40 °C then, a solution of NH 4 OH (2 M) was added dropwise with very slow rate until pH reached 8. After 24 h stirring, the prepared LDH was collected and washed many times using bidistilled water and ethanol until pH 7 and then dried at 50 °C overnight.  www.nature.com/scientificreports www.nature.com/scientificreports/ concentration of Cr(VI) ions after adsorption; UV-visible spectrophotometer (Shimadzu UV-3600, Japan) was used at 350 nm 120 after filtration the adsorbent by 0.22 µm pore size syringe filter (Millipore Millex-G, Hydrophilic PVDF). To compare the adsorption behavior of Cr(VI) ions between Ni/Fe LDH and Ni/Fe LDH/GL; many factors were studied such as pH ranging from 4 to 9, catalyst dose (0.20-2 g/l), dichromate ions concentrations ranging from 20 to 200 ppm, contact time from 5 min to 6 hr and finally temperature (23, 35, 45, 55 °C). The adsorptivity percent of Cr(VI) ions on Ni/Fe LDH and modified Ni/Fe LDH was calculated through Eq. (18).
where Q is adsorptivity percent, C O is Cr(VI) ions initial concentration and C t represents Cr(VI) ions concentration after adsorption at time t. equilibrium adsorption capacity was obtained by applying Eq. (19). The chemical stability of LDH materials was carried out by adding 0.05 g of adsorbent into 50 ml of H 2 O and pHs were adjusted at (3.0, 4.0, 5.0, 7.0, and 9.0) using 0.1 M HCl and 0.1 M NaOH and were set for 1 week. Then, XRD analysis was performed on the tested LDH materials after drying them at 50 °C to examine their chemical stability 122 .
The recycling experiments are significant important factor in the suitability of prepared adsorbent for using and reduce the overall cost of treatment process. Adsorption desorption behaviour of Cr(VI) from Ni/Fe LDH and Ni/Fe LDH/GL was conducted using 0.05 N NaOH solution 95 and stirred for 1 hr then, the concentration of Cr(VI) was determined by UV-visible spectrophotometer and LDH materials were washed by distilled water until pH reached 7 and dried at 50 o C for 2 hr then, they were injected into another adsorption cycle to determine their regeneration efficiency. computation details. Two layers of [Ni 12 Fe 6 (OH) 36 ]6NO 3 •6H 2 O were constructed as suggested by Fan et al. 123 . A vacuum slab with 15 Å was build above the two layers. This surface was optimized by density functional theory (DFT) with DNP basis set and GGA/RPBE functional. The core electrons of the Ni-Fe LDH were treated by the effective core potential. The DMol 3 code 124 was used in the DFT calculations. A large LDH surface, a (4 × 1 × 1) times of the optimized LDH, was built to be used in the Monte Carlo (MC) simulation. The Adsorption Locator module in BIOVIA Materials Studio (BIOVIA, 2017) package was used to carry out the MC simulation, which identifies the lowest-energy configurations. We have simulated six systems: (1) 10 molecules of urea + LDH; (2) 10 molecules of glycerol molecules + LDH; (3) 8 molecules of glycerol + 2 molecules of urea + LDH; (4) dichromate ion + LDH; (5) dichromate ion + system (1); and (6) dichromate ion + system (3). The forcefield COMPASS II 125 was used in the MC simulation. The summation method was Ewald and group based for the electrostatic and van der Waals forces, respectively. The quality of the MC simulation was set to fine. conclusion Briefly, Ni/Fe LDH and modified Ni/Fe LDH/GL nanocomposite with spherical -like shape were synthesized by co precipitation method and were used for the removal of potassium dichromate from aqueous solutions by applying batch adsorption experiments and the results indicated that the adsorption process is strongly pH dependent with optimum pH = 5 and the removal precent increase by increasing the concentration of Cr(VI). The maximum adsorption capacity of Cr(VI) on Ni/Fe LDH/GL (136.05 mg/g) was higher than that on Ni/Fe LDH (50.43 mg/g). In addition, the adsorption processes on Ni/Fe LDH and Ni/Fe LDH/GL were simulated by pseudo-first order, Avrami model, pseudo second order and 1,2 mixed order models and they were spontaneous exothermic adsorption process. The investigation of coexisting anions and humic acid effect on the adsorption process of dichromate ions showed generally negative influence on the removal capacity. The modification of LDH materials by using glycerol improved their chemical stability at different pHs. these materials have a good regeneration behaviour even after exposed them to 7 cycles. The results suggested that Cr (VI) ions were absorbed onto LDH through electrostatic attraction, ion exchange, complexation and hydrogen bond formation. Finally, the fabricated materials are highly efficient adsorbents for removal of Cr(VI) from natural environments.