Enhanced adsorptive composite foams for copper (II) removal utilising bio-renewable polyisoprene-functionalised carbon derived from coconut shell waste

A bio -renewable polyisoprene obtained from Hevea Brasiliensis was used to produce functionalised carbon composite foam as an adsorbent for heavy metal ions. Functionalised carbon materials (C-SO3H, C-COOH, or C-NH2) derived from coconut shell waste were prepared via a hydrothermal treatment. Scanning electron microscopy images showed that the functionalised carbon particles had spherical shapes with rough surfaces. X-ray photoelectron spectroscopy confirmed that the functional groups were successfully functionalised over the carbon surface. The foaming process allowed for the addition of carbon (up to seven parts per hundred of rubber) to the high ammonia natural rubber latex. The composite foams had open pore structures with good dispersion of the functionalised carbon. The foam performance on copper ion adsorption has been investigated with regard to their functional group and adsorption conditions. The carbon foams achieved maximum Cu(II) adsorption at 56.5 mg gfoam-1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{mg g}}_{\text{foam}}^{-1}$$\end{document} for C-SO3H, 55.7 mg gfoam-1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{mg g}}_{\text{foam}}^{-1}$$\end{document} for C-COOH, and 41.9 mg gfoam-1\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\text{mg g}}_{\text{foam}}^{-1}$$\end{document} for C-NH2, and the adsorption behaviour followed a pseudo-second order kinetics model.


Results and discussion
Characteristics of functionalised carbon and carbon foam. The untreated carbon particles have an irregular morphology and appear as partially flattened cylinders, as shown in the SEM images (Fig. 1a). After functionalisation via hydrothermal treatment, the particles become smaller and spherical with a rough surface (Fig. 1b). An example of the C-SO 3 H particles is shown in Fig. 1c, where clarity of elements distribution allows the elemental mapping. The foam surface primarily contains C, but S from the sulfonyl group is evenly distributed across the carbon surface. C-NH 2 and C-COOH could not be clearly characterised using SEM-EDX, and therefore, XPS was used to evaluate the binding energy of the functionalised carbon. From Fig. 1d, the (BET) surface areas (S BET ) of the carbon samples decrease after hydrothermal treatment, particularly for the C-COOH samples. This decrease in the surface area is attributed to the structural deformation or collapse of the carbon framework caused by the strong reaction conditions 28 . SEM-EDX was used to characterise the distribution of the functional groups across the carbon surface.
The FT-IR spectra of the sulfonated carbon (C-SO 3 H) exhibits the absorption peaks at 1702, 1588, and 1155 cm −1 corresponding to C=O, C=C, and C-O bonds, respectively (Fig. 2a). The peak at 1030 cm −1 is attributed to the -SO 3 H group. Aromatic C-H bonds and O-H bonds in the carboxylic acid and phenol groups absorb strongly from 3600 to 2400 cm −1 . The spectra of the carboxylated carbon (C-COOH) is characterised by an adsorption band at 1697 cm −1 , attributable to the carboxylate group. The signals at 1607 cm −1 and 1176 cm −1 correspond to C=C and C-OH stretching, respectively, and indicate that the residual hydroxyl groups are present at the hydrophilic surface. The spectra of the carbon modified with ammonia (C-NH 2 ) exhibit a broad peak at 3242-3400 cm −1 , and the peak at 1591 cm −1 is attributed to the stretching and bending vibrations of -NH 2 , indicating that the amino groups are present at the surface. The peaks in the 1000-1400 cm −1 range correspond to the oxygen-rich groups (C-O) in the amino-functionalised carbon.
XPS was used for the analysis of the surface composition by determining the elemental composition of the material and binding energy of the elements on the surface. The high-resolution XPS survey spectra shown in Fig. 2b    Similarly, the amine-functionalised carbon, C-NH 2 , has an atomic concentration of 3.51% for N in the -NH 2 group. Details of the functionalised carbon surface were determined from the C1s high-resolution XPS spectra (Fig. 2c-f). In the untreated carbon samples, the peaks corresponding to C are observed at 284.5, 284.9, and 286.8 eV, attributed specifically to sp 2 C (C=C), sp 3 C (C-C), and C-O (hydroxyl), respectively. Functionalisation leads to the appearance of new peaks. These include a peak at 289 eV corresponding to the -COO groups in C-COOH, 290 eV correlating to the C-S bond in C-SO 3 H (confirmed by the S2p core level spectrum), and 399 eV attributable to the N1s peak for C-N bonding in C-NH 2 . The C-COOH samples exhibit an increase in the carboxylic acid peak intensity, including O1s of O=C and O-C peaks at 533.3 and 531.9 eV. Similar XPS results have been previously reported in literatures [29][30][31][32][33] . Functionalised carbon slurries with concentrations ranging from 1 to 7 phr were added to the NRL. Curing agents were used to produce the carbon foam, which was subsequently steam vulcanised to yield a final biorenewable polyisoprene-functionalised carbon composite foam with an open-cell structure. Typically, a rubber composite with a high carbon content (up to 70 phr) is produced using a conventional two-roll mill 34 . However, NRL is a colloidal dispersion of cis-1,4-polyisoprene in an aqueous medium, and the amount of carbon that can be added to the latex is limited due to issues related to miscibility and phase separation of the carbon powder. Nanogrinding methods can be used to reduce the particle size in order to increase the carbon content ( Fig. 3a,b). The average particle size of carbon after nanogrinding is 4-31 µm, which is smaller than that of activated carbon (3 to 229 µm). The zeta potential as a function of pH of the functionalised carbon in water suspensions indicates that the points of zero charge (pH pzc ) of C-NH 2 and C-COOH are observed at a pH of 2 and 3, respectively, while C-SO 3 H is negatively charged in the pH range of 1-6 ( Fig. 3c). These observations are attributed to the effects of the surface functional groups on the suspension behaviour and heavy metal adsorption.
The colour of the carbon foam before and after the addition of carbon slurry is shown in Fig. 3d, which shows that the foam colour darkens with increasing carbon content. The SEM images reveal the open pore structure of the foam, with cell interconnection and fine particles distributed in the open cell wall (Fig. 3e). The open-cell wall of the unground carbon (Fig. 3f) shows agglomeration and non-uniform distribution, while the nanoground carbon slurry (Fig. 3g) has a better distribution in the rubber matrix (up to 7 phr). A pore diameter of below 200 µm is observed for the open pore structure, and the dispersion of carbon across the surface can also affect adsorption.
Effect of contact time, pH, and initial concentration on Cu(II) adsorption. All carbon foams were mixed with Cu(II) solutions with various pH values and initial concentrations and monitored over the full duration of the contact time to reveal that all samples reached their equilibrium adsorption within 60 min (Fig. 4). The quantitative measurement conducted via UV-Vis analysis depends on the amount of absorbed light, which varies with metal solution concentrations for bulk analysis. From UV-Vis analysis, the blank foam specimen (polyisoprene foam without functionalised carbon) showed no change in Cu(II) solution colour for all adsorption conditions (green line in Fig. 4), the calculation of the adsorption capacity (q) is based on the adsorption amount of Cu(II) ion per gram of carbon foam (mg g −1 foam ). The used-blank and used-carbon foams were characterised by XRF and SEM-EDX to observe the amount of Cu deposition at the surface and middle of the cubic foam specimen. From Fig. 5 and Table 2, the elemental www.nature.com/scientificreports/ mappings of XRF, SEM-EDX, and SRXTM show a similar trend of Cu deposition, wherein the deposition at outer surface was higher than that at the centre of the foam. Moreover, the functionalised carbon foam exhibited higher Cu amount than the blank foam, especially at the outer surface (approximately 6.9 times). This can be attributed to the efficiency of metal adsorption of the functionalised carbon foam. For bulk analysis, the adsorption depends on the diffusion of aqueous heavy metal ions into the open cell structure of the foam and the pores of the functionalised carbon. These ions attach to the active sites within these structures, and this mechanism requires ion-exchange and chelation. Ion-exchange is driven by electrostatic attraction between the anions of the functional groups and the cations of the heavy metal, whereas chelation is a result of the binding between the lone pairs of electrons of the functional groups and cations of the heavy metal 35 .
The C-SO 3 H and C-COOH carbon foams adsorb Cu(II) via ion-exchange in the pH range of 1-5, with a maximum adsorption of 57 mg g −1 foam and 56 mg g −1 foam , respectively, for the 1500 ppm Cu(II) solution. C-NH 2 carbon foam adsorbs Cu(II) in a pH range of 2-5, which is mainly driven by chelation between the lone electron pairs of the -NH 2 group and copper ions. A maximum adsorption of C-NH 2 carbon foam was 42 mg g −1 foam for the 1500 ppm Cu(II) solution. However, under highly acidic conditions of pH 1, the -NH 2 moieties are protonated to from ammonia ions (−NH 3 + ) and are unable to adsorb the heavy metal cations. The results are consistent with the zeta potential calculations (Fig. 3c). The mechanism for heavy metal ion adsorption by the foams is shown in Fig. 6. Adsorption is not studied at pH > 5 due to metal hydroxide precipitation, which requires additional separation techniques for heavy metal removal.
According to literature review, adsorption capacity is mostly examined in batch experiments at room temperature (approx. 30 °C). Some literatures reported the temperature effect (4-90 °C) on the metal ion adsorption where the temperature has a slight effect on the adsorption of Cu(II) ions 6,36,37 . Therefore, the adsorption temperature was kept constant at 30 °C in our study. from the models indicate that a pseudo-second order kinetic model best fits the adsorption mechanism (Table 3). This suggests that the adsorption mechanism involves adsorbate-adsorbent interaction and a rate-limiting step.
In most cases, rate limiting is due to the chemical reactions and diffusion mechanisms. The chemical reaction involves the sharing of valence forces or exchange of lone-pair electrons from the functional groups to the metal ions, while the diffusion mechanisms include film diffusion and intraparticle diffusion [38][39][40] . The metal ions initially diffuse from the aqueous media to the external foam surface, from which these are transported through the carbon foam network and retained in the pore structure, where ion adsorption occurs at the active sites of the functionalised carbon. The selective adsorption of mixed metal ions is a useful information for wastewater treatment application; however, the results could not observe by using a simple UV-spectrophotometer due to the detection limit of the instrument.    42 . Therefore, the high sample throughput analytical process by using AAS or ICP is necessary for the selective adsorption of mixed metal ions study.
Reusability of the composites. The removal efficiencies of all the functionalised carbon foams decrease with increasing reuse cycles (Fig. 7). Desorption requires a sufficient amount of H + from the desorbing agent to cover the adsorbent surface and replace the metal ions 43 ; decreased removal in the third and fourth cycles is attributed to an increase in the protonated sites on the adsorbent surface 38 . The protonated groups exhibit low levels of complexation with metal ions and ion-exchange, which reduce the adsorption capacity 44 . These adsorbents are reusable, but a decreased efficiency is observed after the several cycles. The functional groups on carbon plays an important role in Cu(II) adsorption where C-SO 3 H and C-COOH carbon foams adsorb Cu(II) via ion-exchange and C-NH 2 carbon foams adsorb Cu(II) via chelation between the lone electron pairs, a weaker interaction. In this work, the reusability of the functionalised carbon foams was studied via repeated adsorption/ desorption cycles by using 0.1 M HCl as the desorbing solution and 0.1 M NaOH as the regeneration solution.
An acid desorption agent is widely used as an acidic desorption agent for desorption of heavy metal ions, however, some researchers have found that acidic eluents also reduce the metal ion adsorption capabilities due to the saturation and occupation of adsorption sites with strongly adsorbed adsorbates which affect to functional groups in carbon foam by covering the active sites and reduced adsorption capacity 45,46 . Severely attenuated removal efficiencies of C-SO 3 H and C-COOH probably due to the deactivation of sulfonated carbon and carboxylated carbon via proton leaching induced by ion exchange process whilst the lesser types of interactions like physical force and complexation of C-NH 2 might take slightly effect on the removal performance 47 .

Conclusions
Three types of functionalised carbons (C-SO 3 H, C-COOH, C-NH 2 ) were successfully incorporated into a rubber foam matrix with good distribution. UV-Vis analysis was found efficient for bulk adsorption, and the amount of Cu on the foam surface could be characterised by XRF and SEM-EDX. Sulfonyl group (−SO 3 H) foam exhibited the maximum Cu(II) adsorption of 1672 mg g −1 active carbon , whereas the carbonyl group (−COOH) and amine group (−NH 2 ) foams showed Cu(II) adsorptions of 1705 mg g −1 active carbon and 1293 mg g −1 active carbon , respectively. Cu(II) adsorption was optimal in the pH range of 2-5 for all functionalised carbon foams. The adsorption mechanism revealed a pseudo-second order kinetics model. The functionalised carbon foams could be reused for up to four cycles, although a gradual decrease in removal efficiency was observed. This approach offers a bio-renewable polymer material capable of adsorbing heavy metal ions from the aqueous media and shows potential for effective wastewater treatment. The selective metal ion adsorption for mixed metal ion with proper measurements, such as atomic adsorption spectroscopy (AAS), and inductively coupled plasma spectrometer (ICP), would be beneficial to further study about the selective adsorption investigation of mixed metal ions, especially for wastewater treatment application.

Preparation of functionalised carbon and composite foam. Functionalisation was conducted in a
Teflon-lined acid digestion vessel. Sulfonation was achieved by a hydrothermally treating a mixture of 1.0 g of carbon with 20 mL conc. H 2 SO 4 at 170 °C for 12 h. The mixture was cooled, filtered, washed with DI water until neutralised, and further washed with ethanol. The functionalised carbon material (C-SO 3 H) was dried at 90 °C Table 3. Kinetic parameters of pseudo-first order and pseudo-second order kinetic models for Cu(II) adsorption onto functionalised carbon foam.
pH Initial concentration (ppm) q eq(exp) ( mg g −1 foam ) Pseudo-first-order kinetic Pseudo-second-order kinetic k 1 (min -1 ) q eq ( mg g −1 foam ) R 1 2 k 2 (g mg -1 min -1 ) q eq ( mg g −1 foam ) R 2 2 C-SO 3   were analysed by scanning electron microscope (SEM, FEI, Quanta 450) and energy dispersive X-ray spectrometer (EDX, Oxford, XMax) and XRF microscopy (Horiba, XGT-7200). X-ray photoelectron spectroscopy (XPS) data were recorded using an ULVAC-PHI, PHI 500 Versa Probe II (ULVAC-PHI Inc., Kanagawa, Japan) with Al Kα X-ray radiation as the excitation source. The atomic concentrations of the functionalised carbon were analysed from the C1s, O1s, N1s, and S2p spectra. The instrument sensitivity was 0.01-0.5% (atomic percentage). Tomographic volume of the samples were investigated by synchrotron radiation X-ray tomographic microscopy (SRXTM). The sample projections (10 keV, at a distance of 34 m from the source) were obtained from the detection system, which was equipped with 200-μm-thick YAG:Ce scintillator (Crytur, Czech Republic), the white-beam microscope (Optique Peter, France) and the pco.edge 5.5 sCMOS camera (2560 × 2160 pixels, 16 bits). After collecting data, it was normalised using flat-field correction algorithm and reconstructed by Octopus reconstruction, respectively. To provide the 3D representation of tomographic volume of the sample, the reconstructed images were rendered using Drishti software. where q is the adsorption amount ( mg g −1 foam ), C i and C t are the initial and final concentrations (mg L -1 ), respectively, m is the weight of active carbon (g), and V is the volume of the solution (L).

Cu(II) adsorption.
For the adsorption kinetics, the pseudo-first-order kinetic and pseudo-second-order kinetic equations are expressed as follows: (1) q = (C i −C t ) × V/m, www.nature.com/scientificreports/ where q t and q e are the amounts of adsorbed copper per weight of carbon foam ( mg g −1 foam ) at time t (min) and at equilibrium, respectively. Rate constant of pseudo-first-order adsorption is k 1 (min −1 ) and pseudo-second-order adsorption is k 2 (g −1 mg min −1 ).
The reusability of the functionalised carbon foams was studied via repeated adsorption/desorption cycles. The spent adsorbents can be recovered by acid washing adsorbents by 0.1 M HCl as the desorbing solution and 0.1 M NaOH as the regeneration solution. The regenerated carbon foams were washed with distilled water up to neutral pH and then reused in a new adsorption/desorption cycle. Four consecutive adsorption/desorption cycles were performed with three replications (n = 3). (2) log q e − q t = logq e − k 1 t 2.303 License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creat iveco mmons .org/licen ses/by/4.0/.