Synergistic effect of pistachio shell powder and nano-zerovalent copper for chromium remediation from aqueous solution

Pistachio shell powder supported nano-zerovalent copper (ZVC@PS) material prepared by borohydride reduction was characterized using SEM, FTIR, XRD, TGA/DTA, BET, and XPS. SEM, XRD, and XPS revealed the nano-zerovalent copper to consist of a core-shell structure with CuO shell and Cu(0) core with a particle size of 40–100 nm and spherical morphology aggregated on PS biomass. ZVC@PS was found to contain 39% (w/w %) Cu onto the pistachio shell biomass. Batch sorption of Cr(VI) from the aqueous using ZVC@PS was studied and was optimized for dose (0.1–0.5 g/L), initial Cr(VI) concentration(1–20 mg/L), and pH (2–12). Optimized conditions were 0.1 g/L doses of sorbent and pH=3 for Cr(VI) adsorption. Langmuir and Freundlich adsorption isotherm models fitted well to the adsorption behavior of ZVC@PS for Cr(VI) with a pseudo-second-order kinetic behavior. ZVC@PS (0.1g/L) exhibits qmax for Cr(VI) removal up to 110.9 mg/g. XPS and other spectroscopic evidence suggest the adsorption of Cr(VI) by pistachio shell powder, coupled with reductive conversion of Cr(VI) to Cr(III) by ZVC particles to produce a synergistic effect for the efficient remediation of Cr(VI) from aqueous medium.


Introduction
Consumer demands of expanding population on the planet are met by our industries with production, which is a threat to sustainable development. Herein, industries produce goods as per the demand of the consumer. Industries meet these demands through time-efficient processes, which demand the use of chemicals that could be fossil-oriented or miningoriented (Carvalho 2017). These chemicals then enter into the biosphere through various environmental pathways and food chains (Garvey 2019). Heavy metals are essential components of these chemicals due to their desired redox, coordination, and physical attributes (Aigberua et al. 2018;Vardhan et al. 2019). Chromium is considered as the 7th most abundant element on earth, existing in the core and mantle, and also ranked 21st in the earth crust with an average concentration of 185 mg kg −1 (Sperling 2005). Chromium finds anthropogenic use in leather tanning, manufacturing, electroplating, chemical refining, organic synthesis, etc. (Lunk 2015;Nigam et al. 2015;Tadesse et al. 2017). The majority of the method for the treatment of chromium in wastewater involves the use of chemical precipitation, coagulation, bioreduction, electrocoagulation, etc. (GracePavithra et al. 2019;Owlad et al. 2009;Peng and Guo 2020;Pradhan et al. 2017). In spite of these treatment techniques, chromium does find its way into our environment, wherein it exists in two oxidation states (i.e., Cr(VI) and Cr (III)). Trivalent chromium is relatively stable and less toxic but poses health concerns at higher concentrations. However, hexavalent chromium is very harmful even at lower concentrations due to its carcinogenicity, hemotoxicity, and genotoxicity (Barceloux and Barceloux 1999;Burrows 2019). Based on the toxicity associated with chromium, the World Health Organization (WHO 2003) and Bureau of Indian Standards (BIS 10500-2012) have set the permissible limit of chromium at 50 μg/L in wastewater.

Responsible Editor: Tito Roberto Cadaval Jr
The treatment of chromium in wastewater by conventional methods like coagulation, electrocoagulation, Mahringer et al. 2020;Peng and Guo 2020), and chemical precipitation (Minas et al. 2017) are irreversible processes, wherein extraction of chromium in the desired form may not be feasible from the waste/sludge generated. Apart from this, conventional methods have the drawbacks of being timeconsuming, require bulk wastewater, cost, and manpower intensive, poor reproducibility under variable wastewater physicochemical characteristics (Crini and Lichtfouse 2019;Hu et al. 2011;Sharma et al. 2009). More recently there have been studies on various emerging technologies for the treatment of wastewater including ion exchange , membrane technologies (Namdar et al. 2021), reverse osmosis (Zakmout et al. 2020), bioremediation (He et al. 2020a;Mohamed et al. 2020), and phytoremediation (Kullu et al. 2020). Among these techniques, adsorption is a simple and cost-effective technology. Adsorption is based upon the imbalance of intermolecular forces on the surface of solids leading the way to interaction and removal of atoms, ions, or molecules from solution onto the surface of the solid phase (Artioli 2008). Thus, adsorption is a surface phenomenon, requiring more and more active surfaces on the solid phase for efficient adsorption. Various surface-active adsorbents studied for chromium could be classified as lignocellulosic biomass (Çelebi 2020), biochar (Qu et al. 2021), zeolites (He et al. 2020b), chitosan (Lei et al. 2020), graphene oxide , resin (Vilayatkar et al. 2020), clay (Ashour and Tony 2020), activated charcoal (Mahmudi et al. 2020;Patel 2020), etc. Among lignocellulosic biomass, Picea smithiana cone (Mittal et al. 2016), pistachio shell (Banerjee et al. 2018), etc. have been studied for chromium adsorption. These biomass have poor adsorption and/or recovery for Cr(VI). On the other hand, zeolites and activated charcoal show poor surface adsorption for Cr(VI) oxoanions, due to the characteristic electrostatic surface negative charge (Jorfi et al. 2017;Puszkarewicz and Kaleta 2019;Vaid et al. 2014).
Further, new classes of surface-active materials in the form of nanomaterials have shown promising activity. Various nanomaterials studied for Cr(VI) remediation include nZVI (Sharma et al. 2009), MnO 2 (Dinh et al. 2020), ZnO (Le et al. 2019), TiO 2 (Gopinath et al. 2020), Fe 2 O 3 , and their variants with other metal/metal oxides and chalcogenides (Karimi-Maleh et al. 2020;Peng and Guo 2020;Zhao et al. 2019). Recently Cu/Cu 2 O nanoparticles have found application in electrochemical sensing (Devnani et al. 2019;Li et al. 2018) and adsorption of heavy metal (Iqbal et al. 2019). Cu/Fe bimetallic nanocomposite materials have been found suitable for remediation of Cr(VI) (Jia et al. 2019;Jiang et al. 2018;Qu et al. 2021;Zhu et al. 2016;Zhu et al. 2018), but very few literature reports are known for utilization of zerovalent copper as Cr(VI) adsorbent (Li et al. 2015a;Wu et al. 2009). However, CuNPs tend to agglomerate during synthesis, and thus, the surface area and the activity of the CuNPs is reduced to a large extent (Huang et al. 2012;Li et al. 2015a;Li et al. 2015b). Thus, to prevent agglomeration, stabilizers and immobilizers can be used in situ during the preparation of the nanoparticles, to ensure dispersion of the nanoparticles. The use of dispersing agents not only provides stable support to the nanoparticles, but also solves the problem pertaining to agglomeration of nanoparticles resulting in pressure drop in the reactor bed and also to the recyclability of nanoparticles (Wu et al. 2009). Supporting materials such as polymers (Saraswathi et al. 2019), zeolites (Xu et al. 2018a), clay (Pandey and Saini 2019), and graphene (Xu et al. 2018b) have been widely utilized for the synthesis of adhered nanoparticle scaffolds. Immobilizers have the added advantage of easy separation of the material from the water/wastewater system. The selection of material to be used as an immobilizer becomes very important. As the support material must not themselves act as secondary pollutants, after they have been employed for the removal of primary pollutants. Many reports support the use of agricultural byproducts and wastes for the removal of heavy metals from wastewater (Alalwan et al. 2020). These biodegradable adsorbent materials are significant in terms of their low cost, abundant availability, easy disposal, and use as green adsorbent without pre-treatments with chemicals (Afroze and Sen 2018). Therefore, the use of biodegradable materials as immobilizers becomes very important for this field of application (Guerra et al. 2018).
One of the authors (Sharma et al. 2018;Sharma et al. 2015) reported a unique reductive regeneration of the nZVI by cellulose as an immobilizer for enhanced adsorption of chromium. Thus, with the potential application of cellulosic materials in stabilizing and reviving the metal nanoparticles, lignocellulosic biomass holds the potential to substitute for cellulose. Thus, in an attempt to make the material greener and sustainable, pistachio shell biomass has been used for the immobilization of copper nanoparticles. As per the FAO data (http:// www.fao.org/faostat/en/#data/QC site visit 11.06.21), pistachio is produced up to 9×10 5 t/year over the world with the USA and Iran leading the production. Pistachio shell has cellulose as a major component at 60.62% (Movva and Kommineni 2017) with an extremely low ash content of 1. 6% (Komnitsas et al. 2015), which has been valorized for thermochemical energy. Due to its unique soft shell, pistachio shell as a value added waste finds use in medium density fiberboard, fillers, and 3D printing biomaterial (Du et al. 2021). Apart from this, the biochar obtained from Pistachio shell finds application as adsorbent for various heavy metals like Pb(II), Cu(II), or Cr(VI) (Komnitsas et al. 2015;Siddiqui and Ahmad 2017;Banerjee et al. 2019). However, the adsorption efficiency of pistachio shell was low (Banerjee et al. 2018) and is envisaged that upon introduction of nanozerovalent copper, the composite material will show high adsorption efficiency for Cr(VI) ions.
Therefore, this study envisages the potential use of pistachio shell powder as support for the synthesis of zerovalent copper nanoparticle dispersed nanocomposite material using appropriate chemical reduction methods and investigate the synergistic effect of these two adsorbents for their Cr(VI) removal efficacies.

Preparation of pistachio shell (PS) powder
Waste pistachio shells used for the studies were collected from the local market. The pistachio shells were washed twice with distilled water before subjecting to air drying in an electric oven for the duration of 24 h at 100 o C temperature. The dried shells were grounded using a common household grinder (Bajaj GX1) and sieved to get a homogenous mesh of opening size of 250-350 μm.

Reagents and material
The chemical reagents used for the studies including CuSO 4 .5H 2 O, NaBH 4 , K 2 Cr 2 O 7 , and diphenylcarbazide were analytical reagent grade and purchased from Loba Chemie Laboratory Reagents, India. All reagents were prepared in deionized water. Diphenylcarbazide was used for the determination of Cr(VI) concentrations (Lace et al. 2019).

Preparation of ZVC@PS composite
Pistachio shell powder (4.0 g) was suspended in 100 mL of distilled water at room temperature. CuSO 4 .5H 2 O (0.17 mol, 4.4 g) solution in 50 mL water was added to the reaction mixture with vigorous stirring. To this solution was added dropwise a 20% aqueous NaBH 4 solution (100 mL) at a constant rate of 1 mL/min. After the completion of the addition, the reaction mixture turned black and was subjected to stirring at room temperature for 30 min. The black material was filtered, washed with an excess amount of distilled water followed by methanol, and dried under inert conditions in vacuo.

Physical characterization of ZVC@PS
FTIR analysis (Bruker, Model: Tensor 27) was performed to determine the characteristic bonds in ZVC@PS. X-ray powder diffraction analysis (Bruker, Model: D8-Advance) was performed to evaluate the crystal structure and crystal phase of the ZVC@PS composite at 2θ angles ranging from 5 to 80°w ith a scanning rate of 3°(2θ) per minute at an accelerating voltage of 40 kV and emission current of 40 mA. The surface morphologies of the ZVC@PS and pistachio shell powder were characterized by field emission scanning electron microscopy (FESEM, Carl Zeiss, Model: Merlin Compact, Germany) with integrated energy dispersive X-ray (EDX) analyzer system. X-ray photoelectron spectroscopy (XPS, Physical Electronics, Model: PHI 5000 VersaProbe III, USA) was used to determine the main element composition and valence state changes of the ZVC@PS composite. Pistachio shell powder and ZVC@PS-specific surface area were investigated in BET (Micromeritics, Model: ASAP 2010, France) at −196°C. Adsorbent degassing was performed under a nitrogen environment at 150°C temperature for 4 h. Thermogravimetric measurements were performed using thermogravimetric/differential thermal analyzer (EXSTAR SII 6300, LabX, Canada) at a heating rate of 10°C /min. UV-Vis spectrophotometer (Shimadzu Japan, Model: UV-2450) was used for determination of Cr(VI) concentration using the diphenylcarbazide method.

Adsorption equilibrium study
A stock solution (1.0 mM) of K 2 Cr 2 O 7 (294.0 mg) was prepared by dissolving it into deionized water (1000 mL) and standard Cr(VI) concentrations of 2-20 mg/L obtained by serial dilution of the stock solution with deionized water and was used for the adsorption study. ZVC@PS (100 mg/L) was added to individual Erlenmeyer flasks containing 100 mL solutions of Cr(VI) (2-20 mg/L), and the resulting mixtures were incubated at 25 ± 2°C and 180 rpm for 24 h. The samples were subjected to filtration using Whatman 1 filter paper and analyzed for Cr(VI) concentration via diphenylcarbazide method using UV-Vis spectrophotometer at a wavelength of 540 nm (Lace et al. 2019). The absorption efficiency (i.e., % Cr removal) and equilibrium adsorption capacity, q e (mg Cr(VI) /g ZVC@PS ) were calculated using Equations 1 and 2.
where C o and C e (mg . L −1 ) are the initial and equilibrium Cr(VI) concentration in mg/L, respectively.
where V is the volume of the solution (in L) and W is the mass of adsorbent ZVC@PS taken. Studies were performed by varying parameters like pH, adsorbent concentration, and initial dichromate concentration to investigate the effect of different experimental parameters on the Cr(VI) adsorption and under optimized conditions. The Cr(VI) concentration of 20 mg/L and ZVC@PS dose concentration of 1.0 g/L was selected for pH optimization study. Cr(VI) solutions (100 mL) having different pH ranging from 2 to 10 were prepared by adjusting the pH with 1N HCl and or NaOH solutions. The optimum dose determination experiment was performed using 20 mg/L of Cr(VI) solution with varying amounts of ZVC@PS (0.05, 0.1, 0.15, and 0.2 g/L) at optimum pH 3. All experiments were carried out in duplicate and average results are reported.

Kinetic and isotherm experiments
Adsorption of Cr(VI) using ZVC@PS was analyzed for its kinetics of adsorption and for optimization of equilibration time. The procedure used to perform kinetic experiments was the same for the determination of adsorption efficiency of ZVC@PS. A 100 mL solution of Cr(VI) at a concentration of 20 mg/L added 10 mg of ZVC@PS and incubated at 25 ± 2°C and 180 rpm for 24 h. A fixed volume of the reaction mixture was withdrawn and diluted at regular intervals of time, i.e., 0.167, 0.33, 0.5, 1, 1.5, 2, 3, 6, 9, 10, 12, 17, and 22 h to analyze the residual Cr(VI) concentration. The data was thus obtained for Cr(VI) adsorption onto ZVC@PS as a function of time and was plotted to find the suitable fit to the kinetic model.
Adsorption kinetics depends upon various factors such as initial concentration, contact time, and temperature. The Cr(VI) adsorption kinetic studies are performed with 20 mg/ L of Cr(VI) aqueous solution and 100 mg/L of adsorbent ZVC@PS.
The pseudo-first-order (Eq. 3) and pseudo-second-order (Eq. 4) kinetic models are used for the investigation of the mechanism of reduction, mass transport, and chemical reaction process involved in adsorption of Cr(VI) by ZVC@PS.
where q 1 and q 2 are the theoretical values for chromium adsorbed at an equilibrium (mg/g); q t is the amount of chromium adsorbed at time t; and k 1 (h −1 ) and k 2 (h −1 g −1 mg −1 ) are the rate constant for pseudo-first-order and pseudo-second-order kinetic model, respectively.

Adsorption isotherm study
The adsorption equilibriums were determined using Langmuir and Freundlich isotherms between ZVC@PS and Cr. In Langmuir adsorption isotherm (Eq. 5), adsorption is assumed to be independent for each entity of adsorbate with the formation of a monolayer on the surface.
This non-linear regression model was used to draw a correlation between q e (mg/g), i.e., the amount of Cr(VI) adsorbed at equilibrium; and C e (mg/L), i.e., equilibrium solute concentration, where q max is the maximum saturation monolayer adsorption expressed in mg/g; and K L which is the Langmuir equilibrium constant.
The Freundlich adsorption isotherm (Eq. 6) is applied to non-ideal sorption on heterogeneous surfaces as well as multilayer sorption and adsorption capacity is related to the Cr(VI) concentration at equilibrium.
The adsorption capacity and adsorption intensity are related to Freundlich constant K F and 1 n F ; respectively.

Statistical evaluation of the fitted models
The kinetic and isotherm models were determined using nonlinear regression of the experimental data using the quasi-Newton estimation method. The fit quality along with accuracy of parameters were evaluated using coefficient of determination (R 2 ), adjusted determination coefficient (R 2 adj ), sum of squared errors (SSE), chi-square (χ 2 ), and average relative error (ARE) ( Table 1). Where, q i,model, q i,exp and q i,exp represents the predicted, experimentally measured and average of experimentally measured q values, respectively. n and p are the number of experimental points and parameters of fitted model, respectively (Dotto et al. 2013;Côrtes et al. 2019).

FTIR spectral analysis
Fourier transform infrared (FTIR) spectral analysis was employed for the determination of characteristic bonds in the lignocellulosic biomass containing pistachio shell powder and immobilized zerovalent copper on pistachio shell powder as a nanocomposite material. The pistachio shell powder with the presence of lignocellulose biomass was characterized by FTIR. The IR of PS showed absorption bands at 3390,2918,1632,1384,1178,1161,1139,1084, and 877 cm −1 (Fig. 2). The bands that appeared at 1178, 1161, 1084, and 877 cm −1 , for the PS, were the characteristic of the monomer pyranose ring structure of cellulose. The broad peak at 3390 cm −1 was assigned to -OH stretching vibrations of primary/secondary -OH groups of the glucopyranose of cellulose or phenolic -OH groups of lignin present as a basic framework of biomass. The peak at 2917 cm −1 is attributed to aliphatic C-H stretching, which confirmed the presence of CH 2 /CH 3 groups; however, the band at 1380 cm −1 is characteristic of the bending of the C-H group (Liu et al. 2006). The absorption band at 1632 cm −1 was attributed to the bending mode of the absorbed water. The peaks at 1084 and 877 cm −1 , respectively, correspond to the C-O-C group of primary hydroxyl stretching at C 6 and C 1 -H group deforming with ring vibration. This confirms the β-glycosidic linkage (Banerjee et al. 2018). The surface chemical interaction of zerovalent copper with the functional groups of the lignocellulose framework of pistachio biomass was evident from the significant shifts in all absorption peaks. The reductive adsorption of chromium on ZVC@PS changes the chemical nature of ZVC particles on the surface and hence results in its decreased interaction with the functional groups causing the shift of bands towards normal values as appeared in FTIR of pistachio shell powder.

FESEM and EDX analysis
The morphologies of the synthesized ZVC@PS nanocomposite material were analyzed using a field emission scanning electron microscope (FESEM). The pistachio shell powder exhibited undefined morphologies with particle sizes ranging from 10 to 100 μm as shown in Fig. 3a. FESEM of zerovalent copper (ZVC) particles (Fig. 3b), synthesized by borohydride reduction, were found primarily in an agglomerated state but well dispersed on the surface of the pistachio shell powder. The Cu 0 nanoparticles were spherical in shape (Fig. 3c) and were formed with an average size of 40 nm. The SEM-EDX spectrum analysis of PS contains peaks from carbon and oxygen (Fig. 3d).
The EDX spectrum of ZVC@PS exhibits peaks from copper along with carbon and oxygen, with no contamination from any other metal (Fig. 3e). The adsorption of Cr(VI) on ZVC@PS was confirmed from the appearance of the characteristic peak of Cr along with other elements (Fig. 3f).

BET analysis
Brunauer-Emmett-Teller (BET) analysis of ZVC@PS was performed and compared with PS to determine the surface area and contribution of immobilized nano-zerovalent copper (ZVC) towards the surface area of the material. The specific surface area of PS was found to be 0.483 m 2 /g with an average BJH (Barrett-Joyner-Halenda) pore size distribution of 5.7578 nm (Turan and Mesci 2011). The relatively lower surface area suggests the microporous nature of the PS. Further, it characterized the poor adsorption capacity of PS for Cr(VI) up to 28 mg/g (Banerjee et al. 2018). On the contrary, the specific surface area of the ZVC@PS was found to be 5.84 m 2 /g, which was significantly higher than that of PS. The average BJH pore diameter of ZVC@PS was observed at 911 nm with the satellite peak at 1374 nm (Fig. 4b). The substantial increase in specific surface area of ZVC@PS was attributed to additional pores created by immobilized nano-zerovalent copper (ZVC) on the surface of PS.

XPS analysis
X-ray photoelectron spectroscopy (XPS) was used to determine the elemental compositions of surfaces of ZVC@PS and  Fig.  5a. Further, the deconvolution of the data was carried out with curve smoothing for Cu 2p and O1s to find out the various source and oxidation states of these elements. The analysis of the survey scan of ZVC@PS showed characteristic peaks in the range 529-536, 823, and 930-960 eV, which are characteristic of elements O, C, and Cu, respectively, in the sample surface.
The binding energy peaks for Cu 2p of ZVC@PS were observed at 932.1, 934.3, 942.5, and 953.4 eV (Fig. 5b). These peaks were deconvoluted to find the contribution of various Cu species. The binding energy peak at 932.1 and 953.4 eV suggest the presence of Cu(0), whereas the binding energy peak at 934.3 and 942.5 eV are characteristic of Cu(II) present in ZVC@PS (Chang et al. 2019;Dong et al. 2014;Wu et al. 2009). The analysis of binding energy of Cu in ZVC@PS indicates the presence of Cu in Cu(0), as well as Cu(II) on the surface, wherein surface-Cu(II) could be attributed to the chemisorbed impurities of oxygen species. Fig. 5d is the O1s XPS spectrum of ZVC@PS that has three different peaks at 529.4, 531.1, and 532.4 eV, which are attributed to Cu(I) oxide, oxygen in biomass polymer, and chemisorbed oxygen species like O and -OH, respectively (Wang et al. 2015).
The Cr(VI)-adsorbed ZVC@PS showed peaks at 529-534, 570-590, 823, and 930-960 eV, which are characteristic of elements O, Cr, C, and Cu, respectively. Upon deconvolution of O1s spectral peak from 529 to 534 eV, it showed the presence of only two species in the sample characterized by the peak at 529.5 and 531.4 eV. The O1s peak at 532.4 is attributed to the C=O, C-O, of biomass as well as the Cu 2 O due to sorption of Cr(VI). The characteristic deconvoluted peak corresponding to O1s at 534.4 eV in ZVC@PS was missing in the Cr(VI)-adsorbed sample and could be attributed to the displacement of chemisorbed oxygen by the incoming Cr(III/VI) species on the surface of ZVC. The weak Cr2p 3/2 and Cr2p 1/2 binding energy peaks observed at 574 and 585 eV, respectively, were characteristic of Cr(III) oxide and Cr(OH) 3 , thus indicating reduction of Cr(VI) to Cr(III) under the adsorption conditions using ZVC@PS as adsorbent. The binding energy peak of Cu in Cr(VI)-adsorbed ZVC@PS showed four deconvoluted peaks at 931, 933, 942.2, and 951.8 eV. The presence of CuO in Cr(VI)-adsorbed ZVC@PS was characterized by the Cu 2p 3/2 and Cu 2p 1/2 at 933.6 and 951.8 eV, respectively (Chang et al. 2019;Ji et al. 2018;Park et al. 2006;Wang et al. 2015). The Cu 2p peaks after Cr(VI) adsorption showed a shift to lower binding energies as it was attributed to the lower concentration of Cu(II) species after Cr(VI) adsorption.

TGA analysis
Thermal stabilities of pistachio shell powder (PS), zerovalent copper loaded pistachio shell powder nanocomposite before (ZVC@PS), and after chromium adsorption (ZVC@PS-Cr) were studied by TGA and DTA techniques. In Fig. 6a, the PS decomposes completely in TGA analysis, as it consists of only organic matter. The TGA curve of ZVC@PS showed mass loss up to 60% at the temperature 410°C and retained the remaining mass up to 600°C. The TGA curve of ZVC@PS after Cr adsorption showed higher mass loss up to 80% until 410°C, and no further mass loss observed until 600°C . This trend of TGA curves indicates the presence of nonoxidizable inorganic substances in ZVC@PS and ZVC@PS-Cr materials that did not decompose until 600°C. The DTA curves of PS, ZVC@PS, and ZVC@PS-Cr are shown in Fig.  6b. All three materials exhibited thermal degradation in two stages. The first stage or primary decomposition having a temperature range from 220 to 320°C involves the weight loss of 65% with degradation of glycosyl units of hemicellulose and cellulose accompanied by the evolution of gases like CO, CO 2 , and H 2 O along with the formation of levoglucosan and charred residues (Shafizadeh 1982). The second stage or secondary decomposition with a temperature range from 340 to 500°C was assigned to degradation of levoglucosan,  before; and c Cu 2p after Cr(VI) adsorption on ZVC@PS; d O1s before; and e O1s after Cr(VI) adsorption on ZVC@PS; f Cr binding energy peaks decomposition of high-energy bonds of lignin, and burning of char, resulting in loss of 30% of total weight (Cheng et al. 2012). In ZVC@PS the thermal decomposition was observed in the temperature range of 200-290°C in the first stage and 310-400°C in the second stage. The shift in the degradation curve towards the lower temperature in ZVC@PS indicated the copper-mediated internal oxidative combustion of lignocellulosic biomass resulting in its lower thermal stability in comparison to PS. There was no significant change observed in the decomposition curve up to 600°C, which indicates the complete oxidation of Cu(0) to Cu(II). The TG curve of the chromium-adsorbed sample exhibited a shift towards higher temperature in comparison to ZVC@PS but with a narrow decomposition temperature range between 270 and 300°C in the first stage and 310 and 370°C in the second stage. The residual content was significantly higher than PS in before and after Cr(VI) adsorption samples of ZVC@PS, which accounted for oxidation of copper and adsorption of Cr(VI).
The two thermal decomposition peaks at 320°C and 480°C were observed in the first derivative (DTA) curve of PS pyrolysis. The two decomposition peaks were ascribed to the loss of two structural components of the lignocellulose framework, i.e., primary decomposition of relatively simpler cellulose structure and secondary decomposition of highly complex lignin structure with high-energy bonds. In the case of copper-loaded ZVC@PS, two endothermic peaks were observed at 285 and 380°C, with significant lower temperature shifts of 35 K and 100 K, in the primary and secondary decomposition curves, respectively, in comparison to PS. The agglomeration of copper on the surface of PS prevents the oxidation of cellulose and results in more heat gain in ZVC@PS. Further, the narrowing of heat gain peaks in the case of ZVC@PS suggested the heat-mediated conversion of copper to its oxide resulted in heat loss, which reinforced the overall degradation process. However, in Cr(VI)-adsorbed samples, the presence of oxides of copper and chromium further prevents cellulose oxidation and results in a high-intensity endothermic peak with a slight shift (15 K) of the primary decomposition curve towards the higher temperature in comparison to ZVC@PS.

Effect of pH on Cr (VI) adsorption
The pH of a solution plays a regulatory role in the adsorption of oxoanion like Cr(VI). Therefore, the pH of the solution was optimized for the adsorption of Cr(VI) onto ZVC@PC. The adsorption studies were performed in varying pH conditions from 2 to 10 with ZVC@PS dose of 100 mg/L and Cr(VI) concentration of 20 mg/L. As the pH increased from 2 to 10, the Cr(VI) uptake efficiency of ZVC@PC was decreased from 100 mg/g to 10 mg/g (Fig. 7). The adsorption under pH 2 and 3 showed similar results; however, under highly acidic conditions of pH 2, the solution turned yellowish due to dissolution of the oxidized Cu into the solution. Dissolution of Cu(I/II) from ZVC@PS was restricted upon increasing the pH to 3 and above. The higher removal efficiency for Cr(VI) at lower pH conditions is ascribed to (a) surface protonation of ZVC@PS resulting in positively charged surface favoring attractive interaction with Cr(VI) oxoanion and (b) protonation of Cr(VI) to reduce electrostatic repulsions with the surface of ZVC@PS and thus facilitate the overall reductive removal of Cr(VI) ions. At higher pH, the proton concentration is diminished, and the negative charge on Cr(VI) is electrostatically less favored for adsorption on the ZVC@PS surface.

Effect of ZVC@PS dose on Cr(VI) adsorption
Effect of adsorbent dose of ZVC@PS was investigated by changing the adsorbent dose from 50 to 200 mg/L with a 20 mg/L Cr(VI) solution maintained at pH 3. As the adsorbent dose of ZVC@PC increased from 50 to 200 mg/L, the removal of Cr(VI) increased from 20 to 70% (Fig. 8). This may account for the increase in the number of available adsorption sites on the nanocomposite adsorbent with the increase in adsorbent dose, resulting in the significant adsorptive removal of Cr(VI) from the solution. As the dose of the adsorbent was increased, the Cr(VI) adsorption (q e ) decreased from 73.22 mg/g to 58.13 mg/g. There was a marginal change in q e from 100 mg/L to 150 mg/g followed by a decrease in adsorption at 200 mg/g. Dose of adsorbent can be attributed to the decrease in the number of Cr(VI) ions available per unit mass of the adsorbent at higher adsorbent dose. Consequently, the active sites on the adsorbent surface remained unsaturated in higher doses. Thus for the optimized conditions used for further study is the adsorbent dose of 100 mg/L.

Effect of Cr(VI) concentration on adsorption by ZVC@PS and adsorption isotherms
Cr(VI) ion uptake equilibrium under varying initial concentration of Cr(VI) (1-20 mg/L) were studied. The Cr(VI) uptake increased from 9.42 to 99.3 mg/g, upon increasing the concentration of an aqueous solution from 1 to 20 mg/L. However, the percentage removal decreased from 100 to 53.17% upon increasing the initial concentration of Cr(VI). These results could be attributed to the increase in the number of Cr(VI) competing for the same amount of absorbent under increased initial concentration (Fig. 9).
Langmuir and Freundlich models were used to represent the adsorption of Cr(VI) on ZVC@PS based on the initial concentration behavior (Fig. 10). Table 2 presents the estimated parameters with fit quality indicators of the adjustment for the two models. The Langmuir model presented higher values of R 2 and R 2 adj and lower values of ARE in comparison to the Freundlich model. This confirmed the monolayer adsorption behavior of ZVC@PS for Cr(VI) adsorption from aqueous solution. The value of q max and adsorption energy (K L ) determined from a non-linear fit of Langmuir adsorption isotherm models were 110.9 mg/g and 1.36 L/mg, respectively (Table 2).

Kinetic equilibrium study
The rate of Cr(VI) adsorption was slow with the removal of 77.50% of Cr(VI) during the first 120 min. The Cr(VI) adsorption further increased from 77.50 up to 93.50% in 17 h. The initial increase in the adsorption could be attributed to the free availability of adsorption sites on the surface of the adsorbent. Further, the slow kinetics could be attributed to the diffusion of the anion to the surface of the ZVC@PS in the pores.
The pseudo-first-order and pseudo-second-order adsorption kinetic models were applied to the data. The non-linear fit diagram of pseudo-first-order and pseudo-second-order adsorption kinetic plots is presented in Fig. 11a-b. The more adequate model for Cr(VI) adsorption upon ZVC@PS was selected by evaluation of R 2 , R 2 adj , and ARE. According to the results in Table 2, the higher values of R 2 and R 2 adj and lower value of ARE were obtained using a pseudo-secondorder model. The kinetic plot shows a good fit with pseudosecond-order kinetics with R 2 equal to 0.9926, which was further verified with a significant chi-square value at a 95% confidence level. The q max and k 2 for the adsorption under pseudo-second-order kinetic fit is 90.03 mg . g −1 and 1.3 × 10 −3 g.mg −1 .min −1 .

Discussion
The adsorption experiments showed that the Cr(VI) adsorption capacity (110.9 mg/g) of nano-zerovalent copper immobilized onto pistachio shell powder (ZVC@PS) was significantly higher than pistachio shell powder (27.95 mg/g) (Banerjee et al. 2018 (Table 3). This higher Cr(VI) adsorption efficiency of ZVC@PS may be ascribed to the formation of well-dispersed ZVC nanoparticles on the surface of pistachio shell powder during synthesis, which otherwise gets agglomerated in the absence of support. A synergistic effect was observed between the Cr(VI) adsorption tendency of pistachio shell powder and reductive removal of Cr(VI) by zerovalent copper, which causes a significant increase in Cr(VI) removal efficiencies of ZVC@PS. The optimum pH for the Cr(VI) adsorption studies with ZVC@PS was Effect of initial Cr(VI) concentration (1 to 20 mg/L) on Cr(VI) adsorption (q e ) and % Cr(VI) removal efficiency of ZVC@PS in the acidic conditions, thus favoring its attractive interaction with Cr(VI) oxoanion. The removal of Cr(VI) was improved with increasing the dose of ZVC@PS, which was ascribed to an increase in the available adsorption sites in the nanocomposite.