Design of carbon quantum dots via hydrothermal carbonization synthesis from renewable precursors

This paper proposes the synthesis characterization and factorial design for the development and composition of carbon quantum dots (CQDs). These are promising nanomaterials obtained from two renewable precursors which are biomass-derived materials namely chitin (CH), chitosan (CS), and graphite (G) as a third material used. These raw materials are low cost, non-toxic, and eco-friendly. The combination of these nanoparticles at quantum composition dots has shown promising properties of high thermal conductivity and high solar absorption. The determination of composition of (CQDs)n using 32 full factorial design with two factors at (CH/CS) ratio and mass (G) was first realized. The response surface methodology (RSM) was used to investigate the effect of band gap energy (Eband gap) engineering for semiconductor CQDs obtained via hydrothermal carbonization synthesis at 200 °C for 6 h and characterization of the composition (CQDs)9 to present Eband gap value of 3.16 eV and present the data with quantum yield of 17.1% thus promising for solar energy conversion application.


Introduction
Semiconductor of carbon quantum dots (CQDs) is a novel class of fluorescent carbon-based materials with crystalline graphitic structure, dimension of up to 10 nm, fluorescence emission property [1,2], spherical shape, and hydrophilic solubility for applications in the conversion of solar energy [3]. The luminescence properties of the CQDs are associated with band gap energy (E band gap ) which depends on the composition of the semiconductors as well as the size of the quantum dots [4]. Hydrothermal carbonization (HTC) is frequently used to synthesize CQDs because this method is considered as eco-friendly with great production viability and low cost. In previous literature studies, the use of biomass-derived materials for application as sensitizers in solar cells has been reported [5,6]. In literature, the CQDs present the quantum yield of luminescence of 12.1%, 31.6%, and 46.2% used hydrothermal method, for this study was used biomass from chitin/chitosan has been studied at literature quantum yield of luminescence approximately of the 13.1% used hydrothermal method [7]. Chitin and chitosan have attracted a lot of attention owing to features such as biocompatibility, non-toxicity, and being eco-friendly [8,9]. The graphite (G) form of these nanoparticles shows desired properties such as high thermal conductivity, low cost, and high solar absorptivity as compared with most of the other nanomaterials [10]. The 3 2 factorial designs were the experimental analysis procedure used for development and composition of carbon quantum dots (CQDs). This technique was useful to optimize the process variables used to obtain a response surface methodology (RSM) with polynomial equation with the use of optimized composition of the process at short time period and with a minimum number of trials [11].
Three renewable precursors were selected in this paper namely chitin, chitosan, and graphite for composition development of (CQDs) n using 3 2 factorial design for two factors which are biomass-derived chitin/chitosan (CH/CS) ratio and the mass graphite (G) using RSM band gap energy engineering of semiconductors via hydrothermal carbonization synthesis at 200°C for 6 h and characterization of CQDs obtained for application as sensitizers in solar cells.

Materials
The CQDs were synthesized from chitin and chitosan from shells industrial of white shrimp (Litopenaeus vannamei) located at northeastern region of Brazil, having a degree of the deacetylation DD ≥ 69.0% and DD ≥ 79.0%, corresponding to chitin and chitosan, respectively, and the graphite (G) and ethanol used in this experiment both have a purity of 97%.

Design of composition CQDs using 3 2 factorial design
A full factorial design (two factors at three levels) was investigated to investigate the influence of two factors, chitin/ chitosan (CH/CS) ratio (1:1, 1:3, 1:7) and the mass graphite (0.1, 0.08, 0.06) for the development and composition of (CQDs) using RSM of band gap energy which are associated with the luminescence properties of the CQDs. All statistical analyses were using Statistica software version 7.0 (StatSoft, USA) with 5% (p < 0.05) significance level, analysis of variance with (ANOVA), and which graph predicted the observed results. The effects of independent variables of the RSM were used as previously studied in literature [12]. The model polynomial equation used to illustrate the data is shown in Eq. (1): Where Z is the expected response; β 0 is a constant; β 1 , β 2 , β 11 , β 22 , and β 12 are the regression coefficients; and x 1 and x 2 are the levels of the independent variables. For the RSM, graphs of E band gap for each (CQDs) n obtained from the UV-Vis spectrophotometer Shimadzu UV-2600 was used using the BTauc relation^were plotted for obtaining the wavelength value (λ onset ) associated with the Babsorbance onset B [13], as shown in Eq. (2): Where α is the absorption coefficient, hν is the photon energy, and B is the band form parameter.

Synthesis and characterization of CQDs
The precursors chitin/chitosan (CH/CS) ratio (1:1, 1:3, 1:7) (0.7 g) and the mass graphite (0.1 g, 0.08 g, 0.06 g) were dissolved in ethanol (20 mL) and placed in Teflon-lined, hydrothermal reactor at 200°C for 6 h. The light brown solution obtained was centrifuged at 10,000 rpm for 10 min to remove the solution containing fluorescent CQDs from the solid black precipitate. The use of the synthesis of hydrothermal carbonization of biomass to produce CQDs, which are the nucleation clusters forming prior to the growth into the final micrometersized carbon particle materials.
The estimated average size of the nanoparticle is determined by the Henglein's empirical model, relationship of the nanoparticle diameter of CdS (2R) of the excitation absorption transition onset [14,15], as shown Eq. (3): Where 2R is related to the diameter of the CdS nanoparticle and λ exc is the excitation absorption transition onset. The CQDs that presented satisfactory values for E band gap designated were characterized by X-ray diffraction (XRD), fluorescence spectrophotometer, Fourier transform infrared spectroscopy (FTIR), and transmission electron microscopy (TEM).
The structure of the amorphous nature in the CQDs was investigated by X-ray diffraction (XRD) using a Rigaku MiniFlex II diffractometer with CuKα radiation (λ = 1.5418 Å). For the phase identification, the measurements were carried out in the range of 2θ within 10 to 60°and the step speed of 0.02°/min with fixed time of 1 s. The photoluminescence spectrometer (PL) was recorded measuring the intensity of emitted radiation as a function of either the excitation or the emission at a wavelength of 396 nm and 480 nm, respectively, at room temperature using a photoluminescence spectrometer model RF-5301PC. Photoluminescence quantum yield (F) of CQDs was determined following the procedure by using quinine sulfate as a reference [16]. FTIR measurements were recorded by using IRAffinity-1 instrument equipped with a HATR MIRACLE with ZnSe prism module in the range from 4000 to 400 cm −1 . The morphology of images

Results and discussion
The CQDs were prepared via hydrothermal carbonization synthesis from renewable precursors (chitin, chitosan, and graphite), as shown in Fig. 1. Table 1 shows the conditions of the complete experimental factorial design 3 2 applied for the composition of CQDs of the 12 experiments with three center points with the CH/CS ratio and mass G as the result of the experimental values obtained for RSM of E band gap calculated from UV-vis spectroscopy method. Table 2 shows the results obtained of the ANOVA for the studied response demonstrated that there was no lack of adjustment; the residues presented random distribution, normality, and homogeneity in the variance [17]. Figure 2(a) shows the UV-vis absorption spectrum behavior for the CQDs as formulated according to the factorial design 3 2 , similar to quantum dots synthesized from biomass derivatives, varying from 300 to 700 nm, with a strong peak at 344.1 nm attributed to the sp 2 transitions of the C〓C [18] and the graph of the Fig. 2(b) shows the BTauc relation for formulation (CQDs) 9 of the composition according to the design with lower E band gap value of 3.16 eV required for quantum fluorescence emission.
From the predicted vs. observed probability graph as shown in Fig. 3(a), it was noted that the data points were closely approximated to the diagonal line indicating a satisfactory result between the experimental data and the data predicted by the developed model with R 2 value of 0.974 confirming the normal distribution of observed data and the adequacy of the developed model [19]. The Pareto shown in Fig. 3(b) was observed and the main quadratic effect CH/CS is the most significant with respect to the G effect. The surface response was constructed for the (CH/CS) and (G) levels of the independent variables, as shown in Fig. 3(c). The surface response obtained and Pareto graph confirmed that the CH/CS have the greatest influence on the composition of CQDs and the determination of E band gap , where the formulation (CQDs) 4 had a higher value of 3.38 eV and a (CQDs) 9 to a lower value of 3.16 eV for of the formulations. The design of a secondorder polynomial equation was developed with adjusted interaction terms between the experimental results, shown in Eq. (4), which predicted the E band gap for CQD compositions: Where the variable (CH/CS) is chitin/chitosan ratio and graphite mass (G). XRD of the samples of the CQDs is shown in Fig. 4(a). All the patterns were the same of the main structure with an sp 2 set with stacking faults based on a wide 002 peak at  with the structures similar to graphite as studied previously [20]. FTIR of the CQDs confirmed the presence of different functional groups (such as carboxylic groups, hydroxyl groups, amines, amides), as shown in Fig. 4(b) the peaks at  3138, 3235, 3536 cm −1 , 1916, 1618, and 1312 cm −1 are attributed to the stretching vibration of -CH, -NH, and -OH, -CO, -CC, and -CH, respectively, for formulation (CQDs) 9 [21]. The investigated composition of quantum dots precursors derived from biomass (chitin, chitosan, and graphite) resulted in a more number of functional groups for better performance of the resulting solar cells. The fluorescence absorption spectra for (CQDs) 9 prepared under optimum conditions were shown in Fig. 4(c). The results show that the maximum excitation and emission wavelengths were 396 nm and 480 nm, respectively. Figure 4(c) showed that the CQDs, when subjected to ultraviolet irradiation and daylight, presented a light brown transparent liquid and emitted bright blue luminescence, respectively, without any further modification [21][22][23][24]. The fluorescent quantum yield was thus calculated to be 17.1% using quinine sulfate as a reference. TEM images of the (CQDs) 9 are shown in Fig. 4(d-e). The TEM image in Fig.  4(d) shows that they consisted of (CQDs) 9 that were monodisperse, spherical, well separated from each other, and partially crystalline structure. Figure 4(d) shows that the CQDs are composed of many nanoparticles having the diameter of about 2-3 nm, of the crystalline plane (100). Figure 4(e) shows a large number of well-dispersed CQDs within the red circles in details [23]. According to Eq. 2, with band value of λ exc = 396 nm, the (CQDs) 9 diameter value of 2R = 2.44 nm was obtain.

Conclusion
This study confirms that the hydrothermal carbonization synthesis method using biomass renewable precursors (chitin, chitosan, and graphite) and using a 3 2 factorial design was satisfactory in the composition of CQDs which demonstrated efficient band gap value and quantum yield calculated of 17.1% as thus promising for solar energy conversion application as thus the different combinations of functional groups present on the surface of CQDs are a key factor for the performance in solar cells which contain a high proportion of nitrogen in the form of amine groups with carbon, oxygen, and hydrogen.
Funding information The authors thank the financial support of the Brazilian research financing institutions: UFRN and CAPES.
Publisher's note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.