Microwave-assisted synthesis of amphoteric fluorescence carbon quantum dots and their chromium adsorption from aqueous solution

The chromium adsorption behavior from aqueous solution by the amphoteric Janus nitrogen-doped carbon quantum dots (AJ–N–CQDs) was investigated. The pseudo-first-order and the second-order adsorption kinetics models were employed to analyze the experimental data; the second-order adsorption kinetics model presented a better correlation to the experimental data, suggesting a chemisorptions process. The values obtained in the pseudo-first-order are still suitable for describing the Kinetics of Cr(VI) sorption. These values elucidate the surface processes involving chemisorption and physisorption in the adsorption of Cr(VI) by AJ–N–CQDs. The R2 of the Boyd model gave a better fit to the adsorption data of AJ–N–CQDs (i.e., external diffusion), which means the surface processes involving external Cr(VI) adsorption by AJ–N–CQDs. The higher value of α may be due to the greater surface area of the AJ–N–CQDs for the immediate adsorption of Cr(VI) from the aqueous solution. AJ–N–CQDs have fluorescence spectra before and after Cr(VI) adsorption, indicating they are promising for chemical sensor applications.


Characterization and analysis. Fourier-transform infrared spectroscopy (FT-IR).
Fourier-transform infrared spectra were collected employing Mattson 5000 spectrometer (Unicam, United Kingdom) using the KBr disk method.
Transmission electron microscope (TEM) analysis. TEM images were taken with a JEOL JEM-2100 electron microscopy at an acceleration voltage of 120 kV.
Scanning electron microscopy with energy-dispersive electron spectroscopy (EDX-SEM). The elemental distribution of GQDs and AJ-N-CQDs was investigated using the non-destructive energy dispersive X-ray (EDX) unit attached to scanning electron microscopy (JSM 6360LV, JEOL/Noran). For surface morphology, imaging was recorded using an accelerating voltage of 10-15 kV.
X-ray diffraction (XRD). The crystallinity was studied by X-ray powder diffraction. The diffraction patterns were measured by Bruker D-8 Advance X-ray diffractometer (Germany), applying a 40 kV voltage and a 40 mA current employing copper (Kα) radiation (1.5406 Å).
where A 1 = area of the crystalline domain, At = area of the total domain 9 .
Nitrogen gas adsorption measurements (BET). The BET was evaluated by Brunauer-Emmett-Teller (BET) technique using St 2 on NOVA touch 4LX, Quantachrome Instruments. The samples were pre-heated at a high temperature (120 °C). These samples were measured at liquid nitrogen after existence gassed at 200 °C under the flow of N 2 for 3 h. The Barrett-Joyner-Halenda (BJH) method was used to measure the pore size.
Thermogravimetric analysis (TGA/DTG). TGA analysis was performed on the prepared polymer powder employing Perkin Elmer thermogravimetric analyzer by heating the sample to 1000 °C of 10 °C/min under a nitrogen atmosphere. Thermal decomposition kinetics was explored for both N-CQDs and AJ-N-CQDs samples. The thermal analysis data were recorded to determine the thermal degradation's activation energy (Ea). Equations (3) and (4) were applied under the Coats-Redfern approach.
Plotting the relationship between {log 10 [1-(1-α)1-n] / T 2 (1-n)} and 1/T using various suitable n values should show a straight-line correlation following Eq. (3). For n = 1, the relationship between log {-log(1-α)] / T 2 } and 1/T was plotted following Eq. (4). As a result, the least square method was used by selecting several n values (ranging from 0 to 3), calculating the correlation coefficient (r) for each value of n, and estimating the standard error (SE). The frequency factor A was determined from the intercept (log AR/ßE) of the Coats-Redfern www.nature.com/scientificreports/ equation by the most suitable value of n, while the activation energy (E) was calculated from the slope (E/2.303R). Equation (5) was used to estimate the other kinetic parameters, such as enthalpy (∆H), entropy (∆S), and free energy change (∆G) 10 .
where (h) and (k) are Planck and Boltzmann constants, respectively.
Fluorescence spectroscopy. The Spectrofluorometer model Jasco FP -6500, Japan, evaluated fluorescence spectra with a light source: a Xenon arc lamp 150 Watt.
where C o and C t are the Cr(VI) concentrations (mg/L) in the solution before and after adsorption, respectively. V is the volume of solution (L), m is the amount (g) of the sorbent employed in the adsorption experiment 10 . Zero order, Pseudo-first-order, and Pseudo-second-order can be determined from the Eqs. (8), (9) and ((10).
where q e and q t are the amounts of Cr (VI) adsorbed (mg/g) at equilibrium sorption capacity and time t, respectively. C e is the final concentration of the AJ-N-CQDs with t (contact time). K 1 (min −1 ) is the Pseudo-first-order rate constant of adsorption. K 2 is the rate constant of Pseudo second-order adsorption. Values of q e2 and K 2 were calculated from the slope and intercept plot of t/qt against t, respectively 6 . The Weber-Morris intra-particle diffusion can be determined from Eq. (11).
where K 3 is the intra-particle diffusion rate constant and C is the slope which represents the thickness of the boundary layer 5 . The Boyd, and Elovich models can be determined from Eqs. (12) and (13), respectively 11 .
where B t is a mathematical function of F, equivalent to qt/qe, representing the fraction of adsorbate adsorbed at different times. Also, α is the initial rate of adsorption (mg/g/min), and b is related to the extent of surface coverage and activation energy for chemisorption (g/mg). A plot of qt against ln t yields a straight line with α and b determined using the slope (1/b) and intercept (ln αb/b), respectively 12 . The Elovich kinetic model can study the adsorption rate based on absorption capacity on heterogeneous surfaces 13 . This model is used further to describe the pseudo-second-order kinetic (i.e., chemisorptions), assuming that the sorbent surface is energetically heterogeneous 14 .  Morphological analysis and elemental composition. From TEM analysis (Fig. 2), graphene (G) sheets were observed for GQDs with small amounts of dots, while AJ-N-CQDs depicted pure nanoparticles with spherical regularity without graphene sheets. The average diameters of nanoparticles were 13.34 and 8.42 nm for GQDs and AJ-N-CQDs, respectively. In addition, the particle size distribution is wide and narrow for GQDs (8-15 nm) and AJ-N-CQDs (between 8 and 9 nm), respectively. Using a domestic microwave, TEM images revealed the presence of GQDs 5 . While by using a lab microwave with changing the conditions produced pure CQDs. Figure 2 shows that the number of dots in AJ-N-CQDs is higher than in GQDs, confirming the lab's efficiency over the domestic microwave for preparing pure AJ-N-CQDs.

Results and discussion
SEM images showed fluffy sheets of GO for GQDs. The dots didn't appear here due to their low number. In addition, AJ-N-CQDs showed crumpled precipitated structures, which didn't show any dots due to the small particle size of the AJ-N-CQDs and the agglomeration existing because of the storage of the sample. EDX results showed that the AJ-N-CQDs are nitrogenized with nitrogen content of ~ 2.90% while, GQDs are nonnitrogenized (Fig. 2). BET surface area determination. The nitrogen (N 2 ) adsorption isotherm was analyzed using the Brunauer-Emmett-Teller (BET) method to calculate the specific area. Figure 4 illustrates the N 2 adsorption isotherms and pore size distribution. The pore size and distribution provide the necessary information about the chemical and physical interaction of the adsorbed metal with the adsorbent surface. The N 2 adsorption isotherm was used to measure the BET surface and volume of the pores.   At the same time, the first ML of AJ-N-CQDs was in the range of 68.38-189.32 °C with a maximum temperature of 163.90 °C and an average ML of 61.79% due to the moisture content. As it is seen, the AJ-N-CQDs need more temperature above 1000 °C to complete their degradation due to their high thermal stability 6 .
According to Table 1, all samples under investigation have values for ΔH for the GQDs decrease from about 9.40 kJ mol -1 at the first stage of the reaction to about 3.94 kJ mol -1 at the second stage, indicating a shallow energy requirement as the reaction progresses (Ea decreased from 12.31 to 8.90 kJ mol -1 ). With lowering n (from 1.5 to 0.5) values, the ΔH was reduced (3.94 kJ mol -1 ). After that, with an increasing n value (2.5), the ΔH was increased (16.89 kJ mol -1 ), indicating a high energy requirement for the final stage (Ea ≈ 25.26 kJ mol -1 ) 1,6 . According to the ΔG observations, the AJ-N-CQDs are more non-spontaneous and require external heat input than the GQDs, which improves their high thermal stability 4 .  www.nature.com/scientificreports/ Fluorescence spectra. Firstly, GQDs did not produce any fluorescence spectra. This may be due to the presence of a large number of graphene, which blocks the fluorescent dots in the GQDs. So, in this study, we will use AJ-N-CQDs for Cr(VI) adsorption. Figure 6 shows the fluorescence emission spectra of the AJ-N-CQDs and AJ-N-CQDs/Cr(VI). They were excited at 350 nm and showed a maximum emission wavelength of 442.0 and 459.0 nm, respectively. This blue shift emission could theoretically prove using of these materials as sensors.  www.nature.com/scientificreports/ where F 0 and F refer to the F.I. of AJ-N-CQDs and AJ-N-CQDs/Cr(VI) Cr(VI), respectively. The calculated FQE was 49.57%, indicating relatively high sensitivity. The interaction between nitrogenized and oxygenated surface functionalities (-COOH, -OH, and -NH 2 ) of AJ-N-CQDs/Cr(VI) was found to be responsible for the FQE 18 . The reduction in fluorescence intensity of AJ-N-CQDs/Cr(VI) compared to AJ-N-CQDs is due to IFE. The Cr(VI) ions absorb light radiation in the same range of wavelength of AJ-N-CQDs. As a result, the intensity is reduced, reducing the fluorescence intensity of AJ-N-CQDs/Cr(VI) 19 . In addition, because of high FQE, these findings validated the efficiency of AJ-N-CQDs synthesized from bagasse as an excellent material for further utilization in chemical sensing applications.
Adsorption study. In this study, we choose AJ-N-CQDs with fluorescent properties to remove Cr(VI). The effect of contact time on the adsorption efficiency of AJ-N-CQDs was studied at different times, namely 15, 30, 45, 60, 120, 240, and 360 min. As shown in Fig. 7, the affinity of AJ-N-CQDs towards Cr(VI) was not the same, and the removal of Cr(VI) by AJ-N-CQDs was found to be fast at first due to the presence of more free functional groups then became slow. There was no remarkable increase in the adsorption rate observed after 45 min. So, the optimum time for Cr(VI) adsorption on AJ-N-CQDs at 25 °C is 45 min. Cr(VI) adsorption to AJ-N-CQDs was 83.85% and stable until 45 min. The adsorption may be due to high nitrogen and oxygen content. After 45 min, the adsorption rate decreased due to the leaching process 1,6 . The Zero order reaction did not make it suitable for describing the Kinetics of Cr (VI) sorption on AJ-N-CQDs which assumes that increasing the concentration of reactants does not affect the magnitude of the reaction rate 20 . Therefore, pseudo-first-order and pseudo-second-order equations are utilized to model the Cr (VI) Kinetics on AJ-N-CQDs. Concerning the values of R 2 presented in Table 2, it is seen that the pseudosecond-order model gave a better fit to the adsorption data of AJ-N-CQDs (chemical bonds in the adsorption). However, the values obtained in the pseudo-first-order are still suitable for describing the Kinetics of Cr (VI) sorption. These values elucidate the surface processes involving chemisorption and physisorption in the adsorption of Cr (VI) by AJ-N-CQDs 5 .
The Boyd model shows that the linear adjustment does not cross the axis at the origin, indicating that external diffusion was not only the rate-controlling step. Therefore, the lowest slope for the Boyd model is relatively associated with a strong effect of intra-particle diffusion. At the same time, R 2 of the Boyd model gave a better fit to the adsorption data of AJ-N-CQDs (external diffusion). These values elucidate the surface processes involving external and intra-particle diffusion in Cr(VI) adsorption by AJ-N-CQDs. The higher value of α may be due to the greater surface area of the AJ-N-CQDs for the immediate Cr(VI) adsorption from an aqueous solution 5,21 .
Comparative study. In our study we used a Cr(VI) of 15 mg/L to stimulate the pollutant water, so at 100% removal of Cr(VI) the adsorbent capacity will be 15 mg/g. The maximum Cr(VI) removal % of AJ-N-CQDs is 83.85% by AJ-N-CQDs which is higher than the adsorption of other adsorbents stated in the literature (Table 3). For example, R % was 48, 45, and 37% for carbon dots modified mesoporous organosilica as an adsorbent for the removal of Hg(II), Cu(II), and Pb(II), respectively 22 . In addition, the maximum R % of Cr(VI) by TiO 2 /CQDs was 43.39% 23 .

Conclusions
This research reported a simple microwave method for preparing nitrogen-doped CQDs (AJ-N-CQDs) from bagasse, citric acid, and DMF. The AJ-N-CQDs have a high adsorption capacity and high sensing to Cr(VI). It is shown that Cr(VI) adsorption on AJ-N-CQDs is rapid in the first 45 min due to the presence of high nitrogen and oxygen functional group. FTIR-Spectra, XRD, TEM, and SEM/EDX confirmed the formation of AJ-N-CQDs. The morphology and BET studies confirmed that AJ-N-CQDs had a porous structure, significantly affecting their absorption properties. The kinetic absorption studies of as-prepared AJ-N-CQDs for