Geometry-Controlled Carbon Coils by SF6 Flow Injection Time with Reaction Temperature

Carbon nanocoils and/or microcoils were synthesized using C 2 H 2 as the source gas along with the injection of SF 6 as an incorporated additive gas under the thermal chemical vapor deposition (TCVD) system. To control the geometries of the carbon coils, we varied the SF 6 flow injection time at different reaction temperature ranges. At the lowest reaction temperature (550C), carbon microcoils were dominantly formed within a relatively short initial SF 6 flow injection time (less than 5min). By increasing the SF 6 flow injection time, carbon nanocoils could be well developed on the entire surface of the sample. At 750C, the formation of carbon microcoils dominated over the entire sample surface, irrespective of the SF 6 flow injection time. Based on these results, the growth mechanism for the dominantly formed carbon coils was suggested and discussed. In addition, the causes for the dominant formation of carbon nanocoils and/or microcoils according to the SF 6 flow injection times with the different reaction temperatures were analyzed.


Introduction
Owing to their unique shape, carbon coils have been noticed to be superior in specific material characteristics to carbon nanotubes.Compared with the linear type geometry of carbon nanotubes, carbon coils show a helical type springlike geometry.Therefore, they are supposed to give the promising materials characteristics such as superelasticity [1][2][3], good chirality [4], and high shielding effectiveness for electromagnetic wave radiation especially in high frequency regions [5][6][7][8].Their superior material characteristics allow them to be good potential candidates for micro/nanodevices [9,10], polymer reinforced constituents [11,12], hydrogen storage media [13][14][15], chiral catalysts [16], electromagnetic wave absorbers [17], and sensors [18][19][20][21][22][23].To extend the practical applications of carbon coils, their geometries should be controlled to enhance the electrical properties of helical coils, which might vary depending on their geometry, including the diameter, as in the case of the carbon nanotubes [24].To date, several methods have been attempted to obtain geometrycontrolled carbon coils [25][26][27][28][29][30][31][32][33].Among these methods, the in situ process during the carbon coils synthesis reaction has been noticed because the in situ process can be combined with the ex situ method.Consequently, these combined methods can provide an advantage for the formation of the geometry-controlled carbon coils.
The cycling on/off modulation of the flow of source gases has recently come into focus as an in situ method for the formation of geometry-controlled carbon coils [34,35].These carbon coils could be obtained by adjusting the number of cycles for the on/off modulation of the flow of source gases during the reaction [36].The manipulation of the cycling on/off time ratio of the flow of source gases during the reaction was also reported to produce geometry-controlled carbon coils [37].A modified cycling on/off method, that is, with the different injection sequences of the flow of source gases, had been reported to be effective in controlling the carbon coils geometries [38].In addition, several studies were carried out using C 2 H 2 /SF 6 gases to form the geometrycontrolled carbon coils [38][39][40][41].Despite these reports, the extension of the in situ method for geometry control of the carbon coils even to a low temperature (below 600 ∘ C) would be required to expand the application area of the carbon coils.In this work, we carried out the in situ cycling on/off modulation process of the flow of source gases at a low temperature (550 ∘ C).We compared the as-grown morphologies of the carbon coils produced at a low temperature with the ones from a high temperature (750 ∘ C).The characteristics of the carbon coils according to the reaction temperature were also investigated.Based on these results, the dominant growth modes for the formation of the geometry-controlled carbon coils at a low temperature (550 ∘ C) or high temperature (750 ∘ C) were discussed and proposed.

Experimental Details
Alumina (Al 2 O 3 ) plates with areas of 3.0 × 13.0 cm 2 and about 2 mm thickness were used as the substrate in this work.For the Ni catalyst preparation, approximately 0.1 g of bunchtype Ni powder (99.7%) with particle diameters ranging from 100 to 200 m was spread onto the substrate.Thermal chemical vapor deposition (TCVD) system was employed with C 2 H 2 as the source gas and SF 6 as the additive gas for the formation of the carbon coils.The total pressure in the reactor was maintained at 100 Torr.The flow rates for C 2 H 2 and SF 6 were fixed at 500 and 40 standard cm 3 per minute (sccm), respectively.The injection time of C 2 H 2 was fixed at 60 min.According to the different reaction processes, the SF 6 flow injection time was varied with the different reaction temperatures.Twelve samples prepared with different SF 6 flow injection times with different reaction temperatures were intensively studied.The reaction conditions of the twelve samples for this work are shown in Table 1.
Detailed morphologies and chemical compositions of the samples were examined using a field emission scanning electron microscope (FESEM, Hitachi S-4200).The crystalline phases were identified through X-ray diffraction (XRD, Shimadzu 6000) using Cu K radiation ( = 0.1541 nm) in the angular range of 10 ∘ < 2 < 60 ∘ .The quality of the carbon coils was investigated using a micro-Raman spectrometer (Renishaw inVia Reflex) with an Ar-ion laser (wavelength  = 514.5 nm) in the spectral range of 1,000-2,000 cm −1 .

Results and Discussion
At 550 ∘ C, four samples were prepared with SF 6 flow injection times of 5 min (sample A), 15 min (sample B), 30 min (sample C), and 60 min (sample D). Figure 1 shows FESEM images of the surface morphologies of samples A-D.
In the case of SF 6 flow injection time of 5 min (the shortest SF 6 flow injection time in this work), we observed the formation of coil morphology on the whole surface of sample A as shown in Figure 1(a).The inset of Figure 1(a) shows the magnified image of the area enclosed in the dashed square in Figure 1(a): the coexistence of well-developed carbon microcoils with untangled carbon microcoils is visible.So, this morphology was understood to be formed by a lot of the double-helix carbon microcoils.When the SF 6 flow injection time was increased from 5 to 15 min, various kinds of carbon coil-like geometries could be observed on the surface of sample B as shown in Figure 1(b).In this case, the single-helix carbon nanocoils occasionally appeared as shown in the magnified image (inset of Figure 1(b)) of the region enclosed in the dashed square in Figure 1(b).A further increase in the SF 6 flow injection time to 30 min gave rise to an increase in the density of the single-helix carbon nanocoils with linear carbon nanofilaments on sample C (see Figure 1(c)).Indeed, the continuous injection of SF 6 flow during the whole reaction time (60 min) resulted in the dominant formation of single-helix carbon nanocoils with linear carbon nanofilaments over the whole surface of sample D (see Figure 1(d)).
The results as shown in Figure 1 reveal that the doublehelix carbon microcoils were exclusively formed by the relatively short injection time (5 min) of the SF 6 flow.On the other hand, the single-helix carbon nanocoils could be dominantly formed on the whole surface of the sample by the continuous injection of SF 6 during the whole reaction time (60 min).
We carried out the same experiment at 650 ∘ C and observed a similar tendency of geometry change according to the SF 6 flow injection time, as shown in Figure 2. In this The causes for these results are understood as follows.After the carbon coils formation reaction, the following was calculated: product/catalyst mass ratio, that is, the ratio of the amount of the carbon coils-related product to the amount of the used catalyst.As shown in Figure 3 and Table 2, the ratio for sample A (SF 6 flow injection time = 5 min) was  In previous reports, the double-helix-type carbon microcoils were known to be formed from the various-shaped carbon nanostructures, namely, single-helix carbon nanocoils, wave-like carbon nanofilaments, and linear carbon nanofilaments [42,43].We could also observe the formation of the carbon microcoils from wave-like carbon nanofilaments (Figure 4(a)), from the carbon nanocoils (Figure 4(b)), and from untangled carbon microcoils (Figure 4(c)).At 550 ∘ C, the formation of the carbon microcoils was frequently observed from the carbon nanocoils (see Figure 4(b)).
In any case, the developed carbon microcoils are located in the upper region of the as-growing carbon coils.Because of their location, they may be the first to encounter the incoming fluorine element from the SF 6 flow.Consequently, they would readily be etched away by the fluorine due to fluorine's intrinsic characteristics to etch other materials.Therefore, the increase in the SF 6 flow injection time led to the decrease in the number of the as-grown carbon microcoils.In addition, the crystallinity of the carbon coils was higher than those of various types of carbon nanofilaments [44].Therefore, the carbon nanocoils, located in the lower part of the as-growing carbon coils, would be dominantly formed under the condition of continuous SF 6 flow injection.This way, the dominant formation of the carbon microcoils and nanocoils could be achieved by the manipulation of the SF 6 flow injection time at 550 ∘ C.
Figure 5 shows the mechanism, called the preferential etching and survival mechanism, for the formation of geometry-controlled carbon coils by the manipulation of SF 6 flow injection times at the lowest reaction temperature (550 ∘ C) in this work.
The preferential etching and survival mechanism is proposed to be as follows: (1) The carbon microcoils were usually formed from the carbon nanofilaments and/or the carbon nanocoils.
The developed carbon microcoils were mainly located in the upper position of the as-growing carbon coils.
(2) The developed carbon microcoils were preferentially etched away by the incoming fluorine-related elements due to their location on the upper position of the as-growing carbon coils.
(3) Owing to the higher crystallinities of the carbon coils than those of the various types of carbon nanofilaments, the carbon nanocoils would survive in the case of continuous SF 6 flow injection.
At 750 ∘ C, we could obtain different geometry change of the carbon coils according to the SF 6 flow injection time.As shown in Table 1, four samples were prepared according to the SF 6 flow injection times of 5 min (sample I), 15 min (sample J), 30 min (sample K), and 60 min (sample L) at 750 ∘ C.
Figure 6 shows FESEM images of the surface morphologies of samples I-L.In the case of SF 6 flow injection time of 5 min, we could see many linear structures on the surface of sample I as shown in Figure 6(a).The magnified image indicates the well-developed individual carbon microcoil for the line shape (see the inset in Figure 6(a)).The lengths of the individual carbon microcoils seemed to be much longer than those at 650 ∘ C or 550 ∘ C (compare Figure 6(a) with Figures 1(a) and 2(a)).Furthermore, most of the individual carbon microcoils on sample I had a well-developed shape, compared with those of the samples at 650 ∘ C or 550 ∘ C (compare Figure 6(a) with Figures 1(a) and 2(a)).These results reveal that the development of the individual carbon microcoil can be enhanced by increasing the reaction temperature from 550 ∘ C to 750 ∘ C.
In addition, the amount of the product/catalyst mass ratios for the samples at 750 ∘ C is much lower than those at 550 ∘ C (see Figure 3 and Table 2).This result strongly indicates that the etching effectiveness for the carbon nanostructures related to the carbon coils by fluorine's intrinsic characteristics would increase with increasing reaction temperature from 550 ∘ C to 750 ∘ C. The product/catalyst mass ratio for sample I (the SF 6 flow injection time = 5 min) was 55.58, less than double the 29.80 for sample L (the SF 6 flow injection time = 60 min).This result also reveals that the etching effectiveness of the carbon nanostructures related to the carbon coils was not significantly improved with increasing SF 6 flow injection time at 750 ∘ C, unlike in the case of 550 ∘ C.
With increasing SF 6 flow injection times from 5 to 30 min, the double-helix carbon microcoils seemed to deteriorate as shown in Figure 6.In the case of the continuous injection of the SF 6 flow during the whole reaction time (60 min), untangled carbon microcoils were mostly seen on the entire surface of sample L (see Figure 6(d)).The results in Figure 6 indicate that the well-developed carbon microcoils were formed at a relatively short injection time (5 min) of SF 6 flow.When the SF 6 flow injection time was increased from 5 to 30 min, the carbon microcoils were still present on the surface of the samples.However, the geometries of the carbon microcoils seemed to have deteriorated.Finally, untangled coils were realized by the continuous SF 6 flow injection during the whole reaction time (60 min).
The results in Figure 6 confirm that the carbon microcoils were predominantly formed over the whole surface of the sample at 750 ∘ C, irrespective of the SF 6 flow injection time.Indeed, the carbon microcoils seemed to be untangled by the continuous SF 6 flow injection.The cause for this result is understood as follows.After carbon coils formation reaction, the mass ratio of the amount of the product to the amount of the used catalyst was also calculated as shown in Figure 3 and Table 2.This ratio for sample I (the SF 6 flow injection time = 5 min, at 750 ∘ C) was 55.58, around one-third of that for sample A (the SF 6 flow injection time = 5 min, at 550 ∘ C).In addition, the mass ratio for sample L is slightly lower than that for sample I.These results indicate that the etching aspect of the carbon nanostructures was more effectively influenced by the reaction temperature rather than the injection time of the SF 6 flow.At 750 ∘ C, we observed the formation of the carbon microcoils from the untangled carbon microcoils rather than the carbon nanocoils as shown in Figure 4(c).Therefore, we propose the mechanism, called the survival mechanism, for the formation of the geometry-controlled carbon microcoils by the manipulation of SF 6 flow injection times at the high reaction temperature (750 ∘ C) as shown in Figure 7.
The survival mechanism is proposed to be as follows: (1) The carbon microcoils were usually formed from the carbon nanofilaments and/or the untangled carbon microcoils.Owing to the high reaction temperature, the carbon microcoils were readily formed and rapidly grown.
(2) The high reaction temperature enhanced the etching rate of the carbon nanostructures.Consequently, the amount of the as-grown carbon nanostructures is dramatically decreased even under the short SF 6 flow injection time.With increasing SF 6 flow injection time, the amount of the as-grown carbon nanostructures slightly decreased compared with that at the low reaction temperature.(3) Owing to the high crystallinities of the carbon microcoils being higher than that of any other carbon nanostructure, the double-helix carbon microcoils survived even under the condition of continuous SF 6 flow injection.
The crystallinity and crystal structures of samples A, D, I, and L were investigated by XRD using the Scherrer equation.Figure 8 shows two remarkable peaks around 24.0 ∘ and 43.0 ∘ assigned to the typical graphitic (002) and (100)/(101) planes, respectively [44].
Table 3 shows the intensities and the 2 values of the (002) positions for the different samples.Comparing the (002) peaks of the different samples, it is noticed that the intensities and the 2 value of sample A are higher than those of sample D. Consequently, the (002) peaks of the carbon microcoils have a higher intensity than those of the carbon nanocoils.In addition, sample A has a lower full width at half maximum (FWHM) than sample D, indicating a reduced line broadening.
Using the Scherrer equation,   = 0.91/( cos ) [45][46][47], the XRD data (the peak intensities and the 2 values of the maximum positions of the (002) plane) allow for the calculation of the average nanocrystalline sizes along the direction (  ), listed in Table 3.The   value of sample A is higher than that of sample D.
Based on the results of the XRD data, we can confirm that the samples with the predominant carbon microcoils formation (sample A) have a more regular crystal structure compared with the sample with the predominant carbon nanocoils formation (sample D).This result strongly confirms that the shortest SF 6 flow injection time gives rise to the most regular crystal structure of the carbon coils at 550 ∘ C. At 750 ∘ C, similarly, sample I having the shortest SF 6 flow injection time in this work had the most regular crystal structure of the carbon microcoils as shown in Figure 8 and Table 3.
Meanwhile, the quality of the carbon coils of the different samples was also investigated by micro-Raman spectroscopy as shown in Figure 9.The  and  peaks of all samples were observed around 1,350 cm −1 and 1,600 cm −1 , respectively [48,49].From the curve fitting (see the inset of Figure 9) of the Raman spectra, the corresponding   /  values were calculated [49,50].Table 4 shows the   /  values and the peak top positions of the  and  bands of the different samples.The  and  bands are known to be associated with the stretching vibrations of graphite and the disordered states of  2 -hybridized carbon, respectively [51,52].
As shown in Table 4, the   /  value of sample I was lower than those of the other samples, indicating the larger   3 and 4, we confirm that the shortest SF 6 flow injection time can promote the dominant formation of the carbon microcoils having a regularly developed crystal structure regardless of the reaction temperature.

Conclusion
At the low reaction temperature (550 ∘ C), carbon microcoils were exclusively obtained by the shortest SF 6 flow injection time (5 min), whereas the dominant formation of the carbon nanocoils was achieved by the continuous SF 6 flow injection.At the high reaction temperature (750 ∘ C), the dominant formation of the well-developed carbon microcoils over the whole surface of the sample was observed, irrespective of the SF 6 flow injection time.We suggest the preferential etching and survival mechanism at 550 ∘ C and, on the other hand, the survival mechanism at 750 ∘ C for the growth modes of the formation of the carbon nano-or microcoils according to the SF 6 flow injection time.The cause for the formation of the different geometries of the carbon coils according to the SF 6 flow injection time with the different reaction temperatures was discussed in detail.

Figure 1 :
Figure 1: FESEM images of (a) sample A, (b) sample B, (c) sample C, and (d) sample D. The insets of (a)-(d) show the magnified images for the dashed square areas in (a)-(d).CMCs, CNCs, and CNFs in the insets mean the carbon microcoils, the carbon nanocoils, and the carbon nanofilaments, respectively.

Figure 2 :
Figure 2: FESEM images of (a) sample E, (b) sample F, (c) sample G, and (d) sample H. Having the corresponding experimental parameters, except the reaction temperature, of samples A-D at 650 ∘ C.

Figure 3 :
Figure3: The product/catalyst mass ratio, namely, the ratio of the amount of the carbon coils-related product to the amount of the used catalyst, according to the samples.

Figure 4 :S＆ 6 Figure 5 :Figure 6 :
Figure 4: FESEM images showing the formation of the carbon microcoils (a) from the wave-like carbon nanofilaments in sample I, (b) from the carbon nanocoils in sample A, and (c) from the untangled carbon microcoils in sample I.

S＆ 6 Figure 7 :
Figure 7: The schematic diagram for the survival mechanism to form the geometry-controlled carbon coils at the highest reaction temperature (750 ∘ C) in this work.

Figure 8 :
Figure 8: XRD spectra of (a) sample A, (b) sample D, (c) sample I, and (d) sample L.

Figure 9 :
Figure 9: Micro-Raman spectra of (a) sample A, (b) sample D, (c) sample I, and (d) sample L. The inset shows the curve fitting of the spectrum of sample A.

Table 1 :
Experimental conditions for the deposition of samples A-L.

Table 2 :
The amount of the carbon coils-related product to the amount of the used catalyst, according to the samples.

Table 3 :
Intensities, 2 values, FWHM, and average size of the nanocrystals along the -direction (  ) obtained from the XRD (002) reflections of the samples.  = 0.91/ cos  (nm)

Table 4 :
The peak top positions of  and  bands of the Raman spectra and the ratios   /  calculated from the fitted Raman spectra.Samples-band peak-top (cm −1 ) -band peak-top (cm −1 )   /