Many methods have been developed to produce nanosized and uniformly shaped SiC powder, such as combustion synthesis , CVD , thermal evaporation , sol-gel carbothermal reduction , plasma technique  and microwave synthesis . The comparison of different methods is reported in Table 1.
Recently, SiC nano-powder is used for its unique mechanical , electrical , optical , photoluminescent , supercapacitative [20, 21], and thermal properties [22, 23]. However, the application of SiC as a support for catalytic materials is limited by its maximum attainable surface area . Consequently, there has been a lot of efforts for finding a process to produce nanosized SiC with considerable porosity and high surface area. To produce powders with smaller particle size, surfactants incorporation is required, surfactants are usually toxic and cause polluting the environment and damaging the ecosystem . Most of the materials used as carbon precursors, such as resins, are expensive petrochemicals . Also, among different approaches proposed for producing SiC nano-powder, the green synthesis approach might be a beneficial method . This concept includes the synthesis and manipulation of nanomaterials by an efficient, environment-friendly, and safe technique that could be used for human welfare [27-29]. Moreover, it has also received merit due to its simplicity, cost-effectiveness, and easy recoverability. This method includes the adoption of nontoxic raw materials and eco-friendly reactants and solvents as well as elimination of hazardous by-products, which are the main principles of green synthesis process [28, 30]. Green sol-gel is one of the best methods of chemical synthesis that is environmentally friendly and can synthesize high purity nanocrystals. This method is cost-effective and has simple preparation process that leads to nanostructured and homogeneous powder .
One of the best candidates for the green synthesis process is the extract of plants as the natural raw material. Due to their non-toxicity and the ability to synthesize at low temperatures and relatively high purity, these sources are of great interest. Stevia is a plant from the family of sweet herbs from South America that are grown in China, Southeast Asia  and Iran . Stevia glycosides are used for food and beverages in some countries [34-36]. This sweet substance contains different kinds of glucose, which is used to synthesize nanoparticles such as Au [33, 37], CdO [38, 39], FeO , ZnS  and Ag . On the other hand, Molasses are a dark substance from sugar beet . It is made in sugar beet factories and contains 50% glucose, minerals, and amino acids. It contains nutrients for the growth of yeasts, which is mainly used in the production of alcohol and animal feed . It was also used as a carbon source for the synthesis of Fe-C nano-reactors  and the synthesis of Ag nanoparticles . Accordingly, it can be used in the sol-gel process for preparing ceramic materials, due to the relatively simple procedure, low cost, good chemical uniformity and purity, and relatively low synthesis temperatures which is necessary for obtaining ultrafine powder [10, 47, 48].
The structure and reaction mechanism of a few materials have been studied based on density functional theory (DFT) [49,51]. DFT is to simplify the molecular system and perform quantum mechanical calculations using quasi-potentials for each atom in the system [52-54].
In this study, we have applied the green synthesis concept by the sol-gel process to prepare SiC nanocrystals with different green sources of carbon including, sugar, molasses and extract stevia. In this method, the lesser harmful chemical is used to produce SiC nanocrystals and finally the efficiency of these green sources for producing pure SiC is evaluated using DFT method. The reaction mechanisms and orbitals energy of the precursors of the SiC powder are analyzed via DFT method.
MATERIALS AND METHODS
Tetraethyl ortho-silicate (TEOS Merck) was used as silicon source and the carbon sources were sugar, molasses and stevia. Also ethanol was used as solvent and hydrochloride acid (HCl Merck) as catalyst. Extract stevia was produced with thirty grams of powder of dried leaf of stevia, that was boiled in 100 ml of distilled water for 5 minutes. The details of this method are explained in the previous work . Each carbon source (sugar, molasses, and stevia extract) with a constant ratio of 0.6 is mixed with other materials. In order to complete the hydrolysis reaction, the as-prepared solution was kept in an isolated place for 18 hours. Resultant gels were kept in the hot air oven for 4 hours in 60˚C to dry as shown schematically in
Fig. 1. During the pyrolysis process, the samples were kept in a furnace with an argon atmosphere at 800˚ C for a time period of 3 hours.
After sorting the samples, structural and composition analysis was performed using X-ray diffraction technique (Philips Xpert) with Cu-Kα as X-ray source. For the evaluation of agglomerates size and qualitative analysis of synthesized powders and whiskers, field-emission scanning electron microscope (FE-SEM) equipped with EDAX analyzer was used. In order to evaluate component’s bond type in the gel samples, Fourier-transform infrared spectroscopy (FTIR) device was used in the range of 400–4000 cm−1. Thermal analysis (TG/DTA) was also used to evaluate the nature of the reaction inside the gel samples, the rate of heating in this case, was 20 ˚C/min.
To obtain the molecular orbital energy of silicon and carbon precursors , the DFT method was used. The optimization of molecule’s structures by B3LYP theory using 6-31G +(d,p) bias set  were performed using Gaussian 09 . In addition, using GaussView , their highest occupied molecular orbitals (HOMOs) and the lowest unauthorized molecular orbitals (LUMOs) of silicon and carbon precursors are calculated.
RESULTS AND DISCUSSION
Samples synthesized using sugar, molasses, and stevia extract, are respectively named SiC1, SiC2 and SiC3. Fig. 2 presents the FTIR spectrum for the gel prepared by the reaction of TEOS and carbon sources. The peaks at 454-463, 522-585, 947-962 and 1071-1078 cm-1 correspond to siloxane bonds (Si-O-Si) [7, 58]. These bonds are related to condensation reactions and hydrolysis of silicon alkoxides. The peak at 3430-3437 cm-1 represents OH functional groups that seem to have formed (O-H) bonds with Si atoms . The peak in the range of 1642–1671 cm-1 belongs to doubled bonds of carbon (C=C) . Moreover, 791-797 cm-1 peak in this figure represents (Si-O-C) [47, 60] bonds that are originally formed between siloxane and carbon atoms from different carbon sources [61, 62]. As demonstrated in Fig. 2, the C=C peak at 1642 cm-1 for SiC3 sample is weaker than other two samples , which means lesser amount of available carbon content, and that could be the reason for the peaks of XRD patterns (Fig. 3) which shows a lot of silicon impurity in the product prepared using the gel of this sample. The lesser amount of available carbon will lead to incomplete reaction between the carbon source and silicon source in this gel. The down-fall at 791-797 cm-1,which is related to Si-C-O bond, is sharper in SiC2 sample as well (Fig. 2c) [47, 60], which means that SiC2 has the highest potential for SiC formation in comparison with other two carbon sources, since Si-C-O bond can easily change to Si-C bond by heating [63, 64]. This result is consistent with the XRD patterns shown in Fig. 3.
Fig. 3 shows XRD patterns, which were used to check the crystal structures of the prepared silicon carbide particles as the synthetic product. As you can see in this figure, all samples are fully crystalline which means that the pyrolysis temperature of 800 °C was high enough to generate a crystalline product. Indexed peaks in Fig. 3 illustrates the XRD analysis of SiC nano-powder as the SiC3 sample. The XRD peaks at 28, 47 and 71 degrees corresponding to silicon (Si)  and the 34, 38, 42 and 63 degrees regarding carbon (C), and the peaks of SiC are at 33, 46 and 56 degrees [65, 66]. XRD results indicate that the reaction between silicon source and the carbon source is not completed in the SiC3 sample since this sample has silicon and carbon contamination.
Fig. 3 also illustrates the XRD patterns of SiC1 and SiC2 samples. The XRD peaks are at 38, 44, 50, 65 and 78 degrees in this figure. And as it can be seen, all of those peaks belong to silicon carbide. This points out that the samples prepared by SiC1 and SiC2 precursors have no contamination and the SiC peaks in the SiC2 sample are sharper. According to Fig. 2, the SiC1 and SiC2 have stronger doubled bonds of (C=C) and (Si-O-C) compared to SiC3. These results support the idea that the reaction between molasses and TEOS is more efficient for the production of pure crystalline SiC. Since the calcination performed in argon atmosphere, there is no evidence of SiO2 impurity in the samples, which is due to the fact that direct carbon oxidation will not happen in the inert atmosphere .
The average crystallite size of the SiC nano-powder in the samples was calculated using the Scherrer equation as follows:
D = 0.9λ/(β cos θ) (1)
Where β is FWHM (full width at half maximum) of the peaks, θ is the Bragg’s angle and λ = 1.54. For crystallite size calculations, β is measured from (111) plane of SiC using the XRD pattern in Fig. 3. The calculated crystallite size of SiC, using the above equation was respectively 42.7 nm, 43.3 nm and 35.5 nm for the SiC1, SiC2, and SiC3.
Fig 4 illustrates the thermal analysis of the gel samples. From DTA analysis (Fig. 4a) it can be seen that all samples undergo similar reactions. The first endothermic peak at 160°C is due to dehydration of the sample (including the removal of the structural water) . This endothermic reaction is followed by two exothermic reactions at 370 °C and 731 °C, which leads to SiC nano-powder formation (Fig. 4a) . Thermogravimetric analysis of the samples (Fig. 4b) shows a secondary weight loss for the SiC3 sample, which could be due to the removal of carbon content or other volatile materials instead of the reaction of carbon with silicon. This carbon removal will reduce the amount of carbon available for SiC formation which consequently raises the amount of silicon contamination. This result confirms the XRD result, which shows huge amount of silicon contamination existed in the SiC3 sample. Among three samples, the SiC2 sample has the sharpest peaks related to exothermic reactions (Fig. 4a) and also undergoes lesser weight loss due to water removal. These phenomena facilitate the SiC nano-powder formation, as it can also be seen in XRD patterns.
DFT method was used to obtain the molecular orbital energy of silicon and carbon precursors. their highest occupied molecular orbitals (HOMOs) and the lowest unauthorized molecular orbitals (LUMOs) of silicon and carbon precursors are shown in Fig. 5.
In Fig. 6 we show the reactivity of each carbon precursor, which also illustrates the energy difference of HOMO and LUMO of silicon precursor and carbon sources. As it turns out, the least energy is related to the sugar precursor, which indicates among different carbon sources, sugar reacts more efficiently with Si precursor.
There are several proposed mechanism for SiC formation using TEOS and carboneous materials . Some of them assumed that the formation of SiC occurs through a direct solid state diffusion of C atoms towards the reacting interphase of SiO2 which is formed via decomposition of TEOS. The proposed reaction is as follows:
While other proposed mechanism are based on two step reaction system based on the following reactions:
The work carried out on carbon/silica powder mixtures suggest that the rate of SiC formation is controlled by the rate of SiO formation. Furthermore, when the free carbon on the surface has been consumed, SiO is produced via reaction between SiO2 and SiC as follows:
2SiO2+ SiC = 3SiO+CO (4)
When dealing with carbon and silica powders, at elevated temperatures, at the interface of C/SiO2, a SiC layer is produced, so the silicon monoxide cannot be produced by reaction (2) anymore. The binary interface of C/SiO2 is converted into the ternary interface of C/SiC/SiO2, and the silicon monoxide is produced by reaction (4), and this will slow down the rate of SiC formation.
However in the case of this study carbon and silica powders are not starting materials. Instead of that a homogenous gel like mixture of carboneous material and TEOS is used, and this will lead to closest possible distance between carbon and silicon, so there is no need for carbon diffusion and that’s why the reaction completes at lower temperatures in a short period of time.
The best results are obtained from sugar precursors, so FE-SEM and MAP element analysis were carried out for this sample. Fig. 7 shows the FE-SEM image of the SiC1 sample. Based on these Figs, powder prepared by this method has irregular morphology. As it can be seen from MAP elemental analysis (Fig. 7 c), SiC is formed in the entire sample surface and there are no other impurities. The size of the agglomerates is quite larger than the size of the nano-powder measured by XRD analysis. As it is expected and reported elsewhere [69, 70], it is due to the fact that XRD measures the crystallite size but SEM shows the agglomerate size. However, the resultant powder is brittle and could be easily crushed to fine particles by simple grinding. This easy crushing capability is due to low synthesis temperature of 800°C, and in comparison with the work of Wang et al. , and Nersisyan et al. , who used 1250°C to 2300°C for synthesis, it is reasonable to expect a spongier product, since higher synthesis temperature causes strong bonding between agglomerates.
This study shows that SiC nano-powder could be synthesized by the sol-gel process at a relatively low temperature with easily available chemicals and green carbon sources. Among carbon sources, sugar seems to be the best candidate, as it efficiently reacts with TEOS to produce nano-powder and pure SiC nano-powder. On the other hand, stevia did not react well with TEOS due to large amount of carbon source or other volatile material that had been removed from the sample before SiC formation reaction. This result is also supported by XRD results that show contamination exists in SiC nano-powder produced by SiC3. Comparing the XRD results with Thermal analysis (TG/DTA) and FTIR analysis also proves the efficiency of molasses as a green carbon source for the production SiC nano-powder. The SiC nano-powder formation mechanism was investigated using the DFT method. As HOMO and LUMO analysis shown in this research, sugar has better reactivity with Si precursor for the formation of SiC.
CONFLICT OF INTEREST
The authors declare that there are no conflicts of interest regarding the publication of this manuscript.