Rational synthesis, characterization, and application of environmentally friendly (polymer–carbon dot) hybrid composite film for fast and efficient UV-assisted Cd2+ removal from water

Carbon dots (CDs) are of particular interest in numerous applications. However, their efficiency for heavy metal removal from wastewater was not yet reported. Herein, we rationally synthesized CDs from petroleum coke waste via hydrothermal treatment in the presence of ammonia. This drove the formation of outstanding photoluminescent, water-soluble, biocompatible, and high yield of monodispersed sub-5 nm CDs. The CDs are co-doped with high 10% of N and 0.2% of S. The as-prepared CDs possess unprecedented photoluminescent properties over broad pH range making these dots unique efficient pH sensor. Chitosan (CH)–CDs hybrid hydrogel nanocomposite film was further prepared as a platform membrane for the removal Cd2+ metal from wastewater. The as-prepared CH–CDs membranes show relatively good mechanical properties, based on stress resistance and flexibility to facilitate handling. The equilibrium state was reached within 5 min. Intriguingly, the UV-light illuminations enhanced the Cd2+ removal efficiency of the photoluminescent CDs substantially by four times faster under. It was found that adsorption followed pseudo-second-order kinetic and Langmuir isotherm models. The maximum adsorption capacity at 25 °C was found to be 112.4 mg g−1 at pH 8. This work paves the way to new applications of CDs in water treatment.


Background
Water is the main crucial component of human life. Nevertheless, the worrying augmentation in pollution matters because of the inappropriate disposal of hazardous chemical wastes from industries [1,2] has led to severe negative impacts on human health and environment [3]. Thus, there is a massive demand for appropriate water purification methods to address water pollution issues [4]. The approach must usually be holistic and requires both science and technology on the one hand and environmental management, on the other hand. As part of the holistic problem solving, it is essential to design lowcost adsorbents for the effective removal of hazardous pollutants from wastewater [5]. In this regard, zeolites [6], clay minerals [7] and carbon-based materials [8] are among the most employed adsorbents for the removal of pollutants due to their high surface area. Recently, carbon dots (CDs) were recently used as efficient adsorbents for heavy metal removal from wastewater [9]. These quantum dots, with particle size smaller than 10 nm, have emerged among the most fluorescent nanomaterials [10]; actually, they are coined "nanolights" [11]. Their success story is due to their easy preparation, low cost, high photoluminescent properties (PL), thermal stability relatively non-toxic nature and facile functionalization [12]. These salient features of CDs led to their numerous applications in sensors [13], cell imaging [14], metal detection [15,16], organophosphate pesticide detection, light-emitting devices [17] and to name but a few [18][19][20]. Several protocols such as arc discharge [21], microwave digestion [22], ultrasonic oscillation [23], electrochemical method [24], hydrothermal synthesis [25] have been reported for the synthesis of CDs. Compared to previously reported protocols, hydrothermal method is the most used for its simplicity, soft conditions and great quantum yield [26].
Newly, hydrothermal treatment of many natural carbon sources [27,28] has been fruitfully used to synthesize fluorescent CDs. Using natural carbon resources, for low cost and ecofriendly synthesis of CDs, is becoming one of the trends of CDs research. Particularly, petroleum coke, which is produced in hundreds of tons as a byproduct during the oil refining process in many oil states (Additional file 1: Tables S1, S2), has a relatively low price [29]. Thus, one can take benefit of these waste products to recycle them to design "new carbon nanomaterials of interest" with high added value. These materials can be rich in benzene rings or aromatic domains and contain a graphene-like structure (Additional file 1: Table S3); therefore, herein, we used it to prepare highly fluorescent and water-soluble CDs. Several studies reported direct relationship between the fluorescence quenching, increase of absorbance and subsequent complexation at high concentration of heavy metal ions including Pb 2+ , Cu 2+ and Cd 2+ [30].
Previous studies described the feasibility of combining CDs with metallic nanomaterial or polymer matrix [31], without compromising their fluorescence [32]. There have been few reports describing the introduction of CDs in hydrogels for different emergent applications [33,34]. Among them, chitosan-based hydrogels could be used as a host platform for green carbon dots immobilization [35][36][37], due to their unique properties such as biocompatibility, non-toxicity, and biodegradability [35,38,39]. Chitosan solubilization, happens in acidic aqueous solution, through NH 2 functional group protonation. This makes chitosan water-soluble natural polymers of major interest for making thin films and hydrogels [40].
Despite the significant progress in the fabrication of various CDs for myriad applications [41], their activity toward UV-assisted heavy metal removal, from wastewater was not yet reported; moreover, the chemical doping of N-atom and S-atom into the conjugated carbon skeleton structure of CDs, modulated its electronic, surface chemical properties, and PL, which are greatly required merits for efficient and selective detection and removal of several metal ions [42].
In pursuit of this aim, herein, we present for the first time a simple and facile approach for controlled synthesis of CDs from petroleum coke waste via hydrothermal treatment followed by ammonia treatment. This led to the conversion of non-environmental benign petroleum coke into CDs with high benefits in many ecofriendly applications [43]. The chemical/physical properties of these novel N,S-doped CDs were studied using a series of surface and bulk characterization techniques including X-ray diffraction (XRD), scanning and transmission electron microscopy (SEM and TEM), atomic force microscopy (AFM), fluorescence, X-ray photoelectron (XPS), Raman, Fourier-transform infrared (FTIR) and UV/Vis spectroscopy (UV/Vis). Intriguingly, the presented method led to high-quantum yield of water-soluble, biocompatible, stable size dependent, pH-dependent photoluminescent (PL) and monodispersed sub-5 nm CDs enriched with N and S. The biocompatibility of CDs was proved via its integration with chitosan in the form of a membrane, which subsequently utilized for Cd 2+ removal from water for a current local project [44] over a wide ranging of pH. Furthermore, the effect of the UVlight irradiation on the removal efficiency of Cd 2+ via the fluorescent CDs was studied in addition to the mechanism as well.

Synthesis and purification of CDs and CH-CDs
A 4 g of petroleum coke was initially ground then dissolved in 90 mL of H 2 SO 4 and 30 mL of HNO 3 (30 mL) under stirring/refluxing at 120 °C for 12 h. Then, the mixture was ten times diluted with H 2 O then neutralized with ammonia followed by hydrothermal treatment in a Teflon-lined autoclave at 180 °C for 12 h. The resultant supernatant filtered and dialyzed with 3500 Da MWCO and kept for further analysis.
Then, the resulting hydrogel was gently placed on a glass substrate and then uniformly dispensed along one sidewall of the doctor blade with a spreader height of 3 mm. The membranes were dried at 80 °C for 24 h followed by neutralization using NaOH (3 M) and washing with H 2 O water several times before being dried and kept for further utilization.

Characterizations
Surface morphology of the prepared CDs was done using scanning electron microscopy (SEM) (ZEISS SUPRA), transmission electron microscope (TEM, JEM-2100 Plus, Jeol, USA) and atomic force microscopy (AFM) measurements were performed in tapping mode on a MFP 3D (AsylumResearch). Particle size was determined by DLS with 90° detection optics using a Zetasizer ZS90 (Malvern Instruments Ltd.,Westborough, MA). Instrument performance was first evaluated using particle standards, singly and as binary mixtures. The X-ray photoelectron spectra XPS measurements were conducted using a Kratos Shimadzu Axis Ultra DLD machine (Kratos, Manchester, UK) equipped with a monochromatic Al Kα radiation source (1486.6 eV). X-ray diffraction analysis was carried out on films using X'Pert PRO (PANalytical) diffractometer using CoKα (1.789 Å) radiation operation voltage and current was maintained at 40 kV and 200 mA, respectively, The theta interval (20-50 °C) and the analysis speed rate (0.05 °C s −1 ). The Fourier-transform infrared spectroscopic (FTIR) studies were carried out using the FTIR Frontier (Perkin Elmer) instrument. The Raman spectroscopy was measured using an Ar+ ion laser at 514.5 nm (Renishaw inVia 2000 Raman microscope, UK), with a spectral resolution < 1.5 cm −1 , all the spectra were initially baseline corrected with third-order polynomial and normalized to the max of the peak intensity. UV/Vis absorption spectra were determined using UV/ Vis spectrophotometer (Gold Spectrumlab 54, Shanghai Lengguang Technology Co. Ltd., China). The fluorescence spectra were performed for three samples of CDs prepared at different periods of time, using a HORIBA SPEX Fluoro-log-3-11 spectrofluorometer. No significant shifts in the fluorescence peaks were noted from batch to batch. Dynamic Mechanical Analysis: the prepared CH-CDs samples was tested using the conventional melt mixing technique were cut into rectangular shapes (50 mm/10 mm/1 mm) and were subjected to dynamic mechanical analysis (DMA) using a Perkin-Elmer DMA.

Selectivity toward heavy metal removal
The prepared CH-CDs hybrid films and pristine was tested for heavy metal removal application. To track the selectivity, the as-prepared CH and CH-CDs hybrid films were immersed in three selected metal picrate solution; Zn(NO 3 ) 2 , Pb(NO 3 ) 2 and Cd(NO 3 ) 2 (25 mL, 50 ppm) at room temperature, PH = 8, then washed several time with diazoned water, the supernatant are characterized using ICP to determine the metal extraction rate .

Adsorption experiment
The kinetics, and isotherm parameters were calculated. CD-based chitosan membranes (100 mg) and (50 ppm, 25 mL) of Cd 2+ solution were used, stirred (180 rpm, 30 min). The remaining cadmium was evaluated using the inductive coupled plasma (ICP) technique. All the adsorption experiments were repeated three times.
At equilibrium, the adsorption capacity, q e (mg g −1 ), was deduced using the Eq. (1): C i and C e (ppm) are the initial and equilibrium concentration of Cd 2+ . M (g) is the mass of adsorbent and V (L) is the volume of solution. Figure 1 is a schematic illustration of CDs and CH-CDs nanocomposite preparation. PC, as a green and costeffective material, is used as a carbon source for preparing of fluorescent carbon dots (see Additional file 1: Table S2).

Preparation of CDs and CH-CDs
The petroleum coke has been oxidized first using HNO 3 and H 2 SO 4 acids and then doped using ammonia via a hydrothermal process. The as-obtained CDs, functionalized with hydrophilic groups such as nitrogen and carboxylic groups [45], show a deep yellow color, exhibiting blue-green emission under UV light (360 nm) and excellent solubility and stability in water. No precipitation was noted even after 6 months. These prepared water-soluble CDs was impregnated with Chitosan polymeric matrix to form a membrane serving as solid and hybrid film for heavy metal complexation and exchange and to control the solubility of CDs in water solution [46]. The as-obtained film shows a blue-greenish color under UV as well. Indeed, the as-obtained CDs actuality negatively charged (zeta potential of − 16.05 mV, (Additional file 1: Figure S1) interact electrostatically with chitosan (positively charged, because its dispersion in acidic medium) with a zeta potential of + 34 mV which imparts uniform distribution and their uniform distribution in the obtained composite film. For complementary evaluation of the role of the prepared CDs in the chitosan biopolymer matrix, the mechanical properties of the chitosan/CDs nanocomposites were explored. The stress strain curves of the chitosan and chitosan/CDs nanocomposites at different concentrations of CDs are shown in Additional file 1: Figure S2a). The mechanical properties of the nanocomposites at the concentration of 20 wt% CDs were improved compared to the pure and 5 and 10 wt% composites. It can be concluded that the concentration of 20 wt% might be chosen as the optimal concentration. The mechanical properties of the developed membranes were investigated as well by DMA Additional file 1: Figure S2b. This latter indicates better high temperature resistance along with improved properties at high temperatures for the 20 wt% sample. Moreover, the chitosan-filled CD membranes display an increase in the storage modulus over the entire E′ spectrum as compared to the pristine. These features indicate a large decrease in the macromolecular motion ability of CH-CDs 20 wt% sample Additional file 1: Figure S2c. Figure 2 show the SEM, TEM, AFM image, DLS and XRD of CDs, Fig. 2a-c shows the data from SEM. The diameter of spherical CDs is around 4.5 nm, and the size distribution is consistent with the Gaussian distribution (Fig. 4g). The SEM images reveal that the PC has an irregular block-like shape. Meanwhile, the typically prepared CDs were formed in a high yield of regular size and shape nanoparticles. Figure 2c shows that the nanoparticles have a spherical-like shape, with an average diameter of 4.5 nm. The dynamic light scattering (DLS) histogram (Fig. 2g) displays the monodispersing of the as-formed nanoparticles with similar size obtained by the SEM image.

Physical structure
The XRD analysis of CDs (Fig. 2h) exhibited one dominant diffraction peak at 2θ of 27 o assigned to the facets of graphitic-like carbon. The resolved XRD peak was weak and broad, implies the small size and disordered stacking of CDs. This was emanated from the integration of both N and S into the C-skeleton structure, which results in the expansion of its lattice structure. This can be seen in the blue shift in the XRD peak relative to pure graphitic. The TEM images were carried out to further confirm the uniform structure of the as-prepared CDs. The lowmagnification TEM image displays that, the CDs have uniform spherical morphology with the absence of any kind of undesired macroscopic phases such as irregular or amorphous byproducts. These CDs were monodispersed with an average size of 4 ± 0.5 nm (Fig. 2c, g). The AFM image also confirmed the uniformity of the obtained CDs (Fig. 2f ) with a height of ranged from 0.5 to 5 nm (Additional file 1: Figure S3). The high AFM topographic height profile shows the average height of CDs was about 1.8 nm, demonstrating that, there are 2-4 layers of the graphitic-like nanostructure, in line with elsewhere reports [45]. The determined zeta potential of the prepared CDs (Additional file 1: Figure S1) found to be − 16.05 mV which is in line with the negatively charged CD surface. Encouraged by the uniformity and small particle size of CDs, its optical properties were benchmarked using both UV-Vis and photoluminescent spectroscopy.

Chemical structure
The XPS was employed to explore the chemical surface composition of the used and prepared materials. The XPS survey of PC resolved the presence of C1 s and O1s. Meanwhile, CDs showed C1s, O1s, S2p, and N1s, implying the formation of graphene CDs enriched with N and S (Fig. 3a). It is noteworthy that, the intensity of both C1s and O1s peak in CDs was significantly higher than their counterparts in PC, infers the high degree of surface carbon oxidation. The high-resolution XPS spectra of C 1s (Fig. 3b) exhibited C-C/C=C at 284.7 eV, and sp 2 carbon-nitrogen N-C=N at 286.5 eV, C-S 286.1 eV, and C=O at 288.3 eV. The determined percentage of sp 2 hybrid carbon atoms of CDs was found to be 89.8-93.8%, infers its high degree of graphitization. Likewise, the N1s spectra (Fig. 3c) (Fig. 3d) doublet is fitted with two peaks originating from spin-orbit coupling, attributed to S2p 3/2 and S2p 1/2 in the 2:1 ratio; S2p 3/2 is centered at 168.5 eV and assigned to C-SO 3 -H chemical environment [47].
The chemical compositions (in atomic ratios) are reported for all samples in (Additional file 1: Table S4). The atomic ratio C/N/O/S is determined to be 81/10/7/2, respectively, confirming the co-doping of CDs with both N and S. The chemical doping of N-atom and S-atom into the conjugated carbon skeleton structure of CDs, modulated its electronic, surface chemical properties, and PL, which are greatly required merits for efficient and selective detection and removal of various metal ions.
The Raman spectrum of the CDs (Fig. 4a) clearly warrants the typical characteristic G band centered at 1609 cm −1 corresponds to the sp 2 bonded C atoms and an insignificant peak for D band at 1330 cm −1 ascribes to the disordered C atoms at the edges of CDs [48,49]. The lower intensity of D band peak at 1330 cm −1 is owing to the high crystallinity and graphitization degree of thus obtained CDs high graphitization. This is clearly seen in the sharpness and higher intensity of G band as well as higher IG/ID ratio. The intensity of crystalline G band was substantially higher than the disordered D band, infers the formation of CDs with a great degree of graphitization and graphite-like stacking domains, in line with the XPS results. This could be further seen in the high ratio IG to ID (1.3), indicating the uniformity of the as-made CDs. Meanwhile, the slight broaden in both IG and ID, originated from the presence of both N-atom and S-atom in the C-backbone structure, which partially disrupts the π-extended framework and originates abundant defects to the basal planes and the edges as well. On the contrary, the IG/ID of PC (1.1) was significantly smaller than CDs, implying its successful conversion to CDs. The determined IG/ID of our newly synthesized CDs was higher than previously reported works [50,51].
The FTIR spectra of CDs (Fig. 4b) revealed obvious absorption bands centered at 1000-1400 cm −1 , which are attributed to the stretching vibrations of C-O, C-S and C-H, respectively [52]. Meanwhile, the bands at 2000-3500 cm −1 corresponded to the stretching vibrations of S-C-N, CH-, N-H, -OH, respectively [53]. This implies

Optical properties of CDs
The UV-Vis absorption of CDs (Fig. 5a), displays two absorption peak centered at 230 and 285 nm, assigned to π→π* transition of C=C and n→π* transition of C=O, which similar to previously reported carbon quantum dot-based materials [47]. The low-intensity absorption shoulders of n→π* is attributed to the lower content and weak electron transition of functional group such as C=O and N=O in CDs as usually noticed in the CDs prepared from industrial wastes [54]. The CDs excited at 340 nm UV light show strong greenish-blue color (Fig. 5a). Excitation-dependent PL behavior with optimized concentrations of CDs was observed (Fig. 5b), which is common for fluorescent carbon materials. The latter could be used in multi-color imaging purpose (Additional file 1: Figure S4). The emission spectra (335-500 nm range) of the CDs (Fig. 5b) show a variation in the PL intensity as the excitation wavelength varied from 360 to 500 nm. The maximum emission intensity reached 489 nm (blue emission) when it is excited at 440 nm. The size effect and the surface composition affecting the band gap of CDs are considered as reasons for the complexity of the PL behavior [55]. Intriguingly enough, the emission intensity was enhanced substantially upon inverting the excitation wavelengths. Additionally, the emissions were significantly red-shifted from 480 to 600 nm. The excitation-dependent PL behavior of the CDs is comparable to the CDs previously reported [47]. This latter could be allied with the aromatic C-C bonds and surface defects resulting from C-OH, C=O, C-S and C-N groups in the CDs. From UV data (Fig. 5a), the band gap energy of CDs was calculated using this equation [αυh = A (υh − Eg) 1/2 ], where α, υ, h, A, and Eg are the absorption coefficient, light frequency, blank constant, constant, and band gap energy, respectively. The band gap of N and S doped CDs was determined to be 3.3 eV. This lower band gap (compared to the CDs without doping 3.72 eV) is emanated from the co-doping effect with both N-atoms and S-atom, agrees with elsewhere reports [56]. The effect of pH on CDs and consequently on the quenching phenomenon was explored (Fig. 5d). The emission band is slightly shifted, shows the emission can be controlled by protons. It was noted as well, that at lower pH values, the photoluminescence intensity increased, and at higher pH values, the intensity decreased (Fig. 5c). Nevertheless, within the pH range 4-7, not a remarkable variation was noted. This pH-dependent photoluminescence behavior may be attributed to the extent of proton acceptance by the nitrogen/sulfur atoms followed by potential proton transfer from the protonated nitrogen to the conjugated carbon structure [57,58].

Heavy metal extraction test using CH-CDs hybrid films
Few recent studies have shown the ability of CDs based composites to adsorb heavy metals from aqueous solution [59]. In this work, to investigate the selectivity toward heavy metals removal, the as-prepared CH-CDs was tested in the presence of Cd 2+ , Zn 2+ and Pb 2+ at a (pH 8.0). The results showed that, the removal efficiency of CH-CDs towards Cd 2+ , Zn 2+ , and Pb 2+ were about 93, 55, and 40%, respectively, indicates the removal selectivity towards Cd 2+ ions (see Additional file 1: Figure S5).

Influence of pH and contact time on Cd 2+ removal
The adsorption of Cd 2+ on (CH-CDs) hybrid films as a function of solution pH and contact time was checked. Toward this end, the CH and CH-CDs films were immersed in a 50-ppm solution of Cd 2+ up to 30 min, under light and dark conditions and at different pH values and followed by the tracking of the remaining concentration of Cd 2+ ion in solution by the ICP Fig. 5 a UV-Vis absorption spectrum with inset photographs of fluorescent CDs under 365 nm UV irradiation and under dark, b excitation-dependent emission spectra of CDs, c excitation-dependent emission spectra of CDs at different pH analysis technique. More details about procedure were discussed in "Materials and methods" section. The extraction efficiency as a percentage (%E) was calculated via the following equation: %E =100(C 0 -C)/C 0 , where C 0 = 50 ppm. The effect of pH was studied as well, as shown in (Fig. 6a). The extraction efficiency percentage of Cd 2+ was found to increase in alkaline solution, the highest extraction efficiency percentages (E = 93%) were noted in alkaline solution a pH = 8 under illumination. It is important to note that the solubility product of Cd(OH) 2 = 7. ] * [10 −6 ] 2 = 2.1 × 10 −16 which is less than the Ksp of Cd(OH) 2 , based on this Cd(OH) 2 will never precipitate at This PH = 8 [60]. This was also confirmed by ICP where the concentrations at different pH values were the same. Figure 6b shows the trends of extraction efficiency at different periods of time for Cd 2+ using CH, and CH-CDs under dark and light at pH = 8. As shown after 6 min only, the CH-CDs sample reached the maximum extraction efficiency (93%) under UV-light irradiation, while the CH-CDs under dark and the pristine reached 80 and 61%, respectively. It is important to note that the kinetics of Cd 2+ removal was very fast; the maximum extraction efficiency percentages for the CH-CDs membranes was reached after 3 min only under UV irradiation at pH = 8 and it is five times faster than CH-CDs under dark conditions.

Adsorption kinetics and isotherm
To describe the adsorption of Cd 2+ onto CH-CDs, the first-and second-order kinetics models were applied as following; the first-order kinetics model is described by Eq. (2): The pseudo-second-order kinetic model can be described by the following equation: (2) ln (q e − q t ) = ln q e − k 1 × t.
(3) t/q t = 1/k 2 q 2 e + t/q e . Fig. 6 Trends of the extraction efficiency percentages a for Cd 2+ at different pH, b for Cd 2+ using CH, CH-CDs under dark and light at pH = 8, c pseudo-second-order mode for Cd 2+ adsorption onto CH-CDs, d the linear fitting results of Langmuir models q t and q e (mg g −1 ) are, respectively, the quantities of Cd 2+ adsorbed at time t (min) and equilibrium. k 1 (1/ min) present the rate constant of pseudo-first-order, and k 2 (g (mg min) −1 ) is the rate constant of pseudo-secondorder models.
The adsorption data can be calculated using Langmuir or Freundlich models, through the following equations, respectively; q max is the maximum sorption capacity (mg g −1 ), q e is the solid phase equilibrium concentration and C e is the equilibrium concentration of the metal ions (ppm) and K L is a constant related to the binding energy of the sorption system (L mg −1 ). The K L and q max can be deduced from the intercept and slope of the linear plots when C e /q e results in a properly line. Also, K f and 1/n are associated to the Freundlich equilibrium constant K f (mg g −1 ) (mg L −1 )1/n.
To evaluate the rate of Cd 2+ removal on CH-CDs, the kinetics of the sorption model was applied. First-order and pseudo-second-order equations were used to study the variations in adsorption with time. The better correlation coefficient values were noted with pseudo-secondorder model compared to the first-order kinetics (Fig. 6c, Table 1). From the above, the dominant mechanism could be chemical adsorption [61]. The adsorption isotherm is applied to estimate the adsorption capacity of an adsorbent. Moreover it gives a prediction, how the material interacts with adsorbents. Toward this end, the Langmuir and Freundlich isotherms [61] were used.
The adsorption isotherm, (Fig. 6d and Additional file 1: Figure S6) was conducted at constant pH = 8 and RT. As concluded from Table 1, Fig. 6d and Additional file 1: Figure S6), Langmuir model was better fitted than Freundlich model based on the correlation coefficients (R 2 ). According to Freundlich model study, n values are slightly higher than one; this confirms that the sorption process might happen through electrostatic interaction, ion exchange, or combined mechanism [62].
(4) C e /q e = 1/K L q max + C e /q max , (5) log q e = log K f + 1/n × log C e .

Mechanism
Efficiency toward heavy metal removal could be arisen from chitosan coating, as previously reported [63], it could be concluded that a decrease in Cd 2+ concentration in the presence of the pristine (pure chitosan) (Fig. 6b) happened because of the free NH 3+ groups were again deprotonated due to the addition of NaOH. The deprotonated NH 2 groups from chitosane.
The deprotonated NH 2 groups from chitosan chelates with Cd 2+ as per the already reported model [64]; however, for the CH-CDs membranes, some synergetic effect toward Cd 2+ removal was observed. In addition to the chelating properties of chitosan, the well-dispersed CDs are doped with oxygen, nitrogen and sulfur, which exacerbates their chelating properties. Different mechanisms have been suggested to explain the interactions of heavy metals with carbon materials, doped with nitrogen and sulfur. Recent studies show that N and S dopants on the carbon dots surface increase the negative charge density of the carbon surface, and then enhance the ability for the Cd 2+ removal compared to the non-doped carbon materials [65]. Based on Pearson theory, the affinity of heavy metals towards nitrogen/sulfur is explained by soft acid-soft base interactions, where, the CH-CDs act as soft base and Cd 2+ as soft acids [66][67][68]. Moreover, another probable mechanism can be added based on the principle of ion exchange, indeed, Na + ions present in the CH membrane after dipping in NaOH solution (Fig. 7) stay bounded to the negatively charged CDs surface (Fig. 7), which are present in the chitosan film as previously described. The Na + ions from the hybrid (CH-CDs) membrane are released into the solution, and ion exchange Cd 2+ ions are absorbed into the prepared membrane [46].
After the solution containing the (CH-CDs) and Cd 2+ (50 mg L −1 ) was irradiated by UV light (365 nm) for 30 min, removal of Cd 2+ was almost complete (ca. 93%). Figure 6b plots the E% of cd 2+ versus reaction time, from which we can observe that the process of photo adsorption is ultra fast and efficient. Such a high adsorption activity of CH-CDs may be attributed to N and S dopants which reducing the band gap energy, delay the electronhole recombination and enhance the light absorption of CDs results in increasing the electron density and enhancing the adsorption of positively charged Cd 2+ [69,70]. Nevertheless, the mechanism has not been fully explained and further study is needed. Results obviously demonstrated that N,S codoped CDs might enhance the UV-assisted adsorption efficiency towards Cd 2+ .

Comparison with other absorbents
Previous works were summarized in Table 2 to compare our prepared material with the previously reported adsorbent. The time of adsorption under UV illumination was faster than others (5 min incubation). The optimal pH of the solution was 8, which is close to neutral  conditions. The optimal pH of the solution was 8, which is close to neutral conditions.

Conclusions
Novel photoluminescent CDs were produced from petroleum coke waste by hydrothermal simple method. The as-made CDs are rich in oxygen, nitrogen, and sulfur functional groups and high yield of monodispersed sub-5 nm CDs, outstanding water solubility and sensitive to pH in a wide range of 1-14. The as-prepared CDs were successfully introduced into the chitosan polymer matrix and form well-dispersed CH-CD fluorescent films. The latter are promising platforms for the removal of Cd 2+ from industrial wastewater for a current project. The CH-CDs membranes show relatively good mechanical properties, based on stress-resistant. Interestingly, the UV-light illuminations, enhanced the Cd 2+ removal efficiency of the photoluminescent CDs substantially five times faster under and the adsorption process could be described using pseudo-second-order kinetics and Langmuir isotherm model. The equilibrium time of adsorption was ultra fast (only 5 min). In addition, the optimal pH of the solution was 8 which is close to neutral. The maximum adsorption capacity at RT was found to be 112.4 mg g −1 at pH 8. This work opens new avenues for producing low cost and rapid separation for practical adsorbent films based on green and low-cost fluorescent carbon quantum dots.

Additional file 1. Additional tables and figures.
Abbreviations CDs: carbon dots; CH: chitosan; AFM: atomic force microscopy; XPS: X-ray photoelectron; FTIR: Fourier-transform infrared; UV/Vis: ultraviolet/visible spectroscopy; (ICP) technique: inductive coupled plasma; q max : the maximum sorption capacity (mg g −1 ); q e : is the solid phase equilibrium concentration; C e : the equilibrium concentration of the metal ions (ppm); K L : constant related to the binding energy of the sorption system (L mg −1 ); C i and C e (ppm): initial and equilibrium concentration of Cd 2+ ; M (g): the mass of adsorbent; V (L): the volume of solution.