1. IntroductionSingle-walled carbon nanotubes (SWCNTs) have attracted considerable attention of researchers over the past decade because of their specific physical, chemical and mechanical properties [1-3]. However, because the carbon nanotubes are inorganic polymers of a highly polarized smooth surface there is a strong van der Waals force between the tube combined with its huge surface area and high aspect ratio, resulting in carbon nanotubes very easily wound reunion into a bundle, and the surface of carbon nanotubes lack of active groups, and insoluble in common organic solvents and water, which greatly limits the application of carbon nanotubes [4,5]. Therefore, the effective dispersion of carbon nanotubes has become the focus of the present research.
Now significant research efforts have already been devoted to improve the dispersion of carbon nanotubes. The main route available to enhance carbon nanotubes solubility in solvents and other matrix material in the dispersion of carbon nanotubes modified. By far, many kinds of surfactants (anionic, cationic, non-ionic) have been used to disperse carbon nanotubes. e.g. sodium deoxycholate(DOC) [6], sodium dodecyl benzenesulfate (SDBS) [7], sodium dodecyl sulphate (SDS) [8] and a family of Tween surfactants [9]. The dispersion of SWCNTs in solution can be studied by atomic force microscopy (AFM) [10,11], Raman spectros-copy [12-14], scanning electron microscopy [15,16], cryogenic transmission electron microscopy (cryo-TEM) [17]. However, there is no report about a reference to indicate the dispersion. In fact, we often need a simple technology in real producing. Moreover, UV-vis-NIR spectroscopy offers a more precise method for detecting dispersion of individual nanotubes [18]. In the UV-vis-NIR region, Individual SWNTs exhibit characteristic bands corresponding to additional absorption due to 1D van Hove singularities [19]. Bundled SCNTs are hardly active in the wavelength region between 200 - 1200 nm because of carrier are tunneling between the nanotubes [20]. Moreover, researchers have reported that the absorbance of SWCNTs solutions shows a maximum between 200 and 300 nm and gradually decreases from UV to near-IR. Therefore, it is possible to establish a relationship between the concentration of SWCNTs individually dispersed in solution and the intensity of the corresponding absorption spectrum.
In this paper, we will emphasis on the research of setting up an UV-vis-NIR absorbance to evaluate the dispersion effect, and also use it as reference to find the optimum conditions of SWCNTs dispersed in surfactant solutions.
2. ExperimentsSWCNTs (>80 wt%) were obtained from Sigma-Aldrich. (Hipco, Sigma-Aldrich, 0.7 - 1.1 nm). The surfactant for the dispersion of the SWCNTs was sodium dodecyl sulfate (SDS; 99%) provided by Sigma-Aldrich. All dispersion experiments were carried out with distilled water. All solutions were prepared by mixing a certain amount of SWCNTs with 10 mL of an aqueous SDS solution in a flask, after which the resulting mixture was sonicated for different times under mild conditions.Then the dispersed sample solution was centrifuged at 17,000 g for 15 min in a swing bucket rotor (S52ST, Hitachi KoKi) to remove the bundles, and the residue of catalytic metal particles. The upper 80% of the supernatant was collected and used for measurement.Then we take same amount (5 ml) of SWCNTs solution diluted to different times (0.5, 0.7, 0.9, 1.0, 1.3, 1.5, 1.7, 3, 5, 7, 9), and measurement the UVvis-NIR spectroscopy of these dilutied SWCNTs solutions.
UV-vis-NIR absorption spectra (Lambda900, PerkinElmer) were recorded with spectrometer operating between 200 and 1100 nm. Samples were taken regularly during the sonicating process and diluted by a certain factor, resulting in certain SWCNT contents that were suitable for UV-vis measurements. The blank used was the original SDS solution diluted by the same factor, under the same conditions as the samples themselves. SEM (Zeiss, Supra55) images were recorded by Signal A = InLens mode and D = 4.7 mm, EHT = 5.00 kV, Mag = 5.00 KX. Trans-mission electron microscopy (TEM) observations of SWCNT dispersions were performed using a Tecnai 20 (FEI Co.) operated at 200 kV.
3. Results and Discussions3.1. Relationship between UV-vis-NIR Spectra and Concentrations of Dispersed SWCNTsFigure 1(A) illustrates UV-vis-NIR spectra of SWCNTs dispersed in solutions with different dilutions. As show in Figure 1(A), increasing concentration of dispersed SWCNTs results in an increasing absorbance intensities. In order to understand the relationship between the intensities of the whole spectra range and the dispersion concentrations of SWCNTs, we calculate the SWCNTs concentration by subtracting the surfactant weight from the total dry weight of the supernatant, then we measured a series UV-vis-NIR spectra of different concentrations of carbon nanotubes, and measured the corresponding absorbance intensities at different wavelengths. Finally the plots of intensity and concentration are obtained as shown in Figure 1(B). Any wavelength within the UVvis-NIR range from 200 to 1100 nm displayed linear relationship between the absorbance intensities and dispersed SWCNTs concentrations. All of intensities at different wavelength increased with the increasing of the dispersed SWCNTs concentrations, but the slopes were
(A)
(B)
Figure 1. (A) Evolution of UV-vis-NIR spectra of an aqueous 0.3 mg/ml SWCNTs—1.5 wt% SDS solution as a function of diluted times. (a: diluted 9 times; b: diluted 7 times; c: diluted 5 times; d: diluted 3 times; e: diluted 1.7 times; f: diluted 1.5 times; g: diluted 1.3 times; h: diluted 1 times; i: diluted 0.9 times; j: diluted 0.7 times; k: diluted 0.5 times). (B) Calibration curves determining the true concentration of SWCNTs in aqueous dispersions with different wavelength.
different indicating that the SWCNTs concentration had different affections on the absorbance at different wavelengths. When the wavelength is 241 nm, the slope was the biggest, while the slope was the smallest when the wavelength is 744 nm, which indicated that the spectrum absorbance was sensitive to the dispersed SWCNTs concentrations in the UV wavelength range. So the UV-visNIR absorbance intensity at 241 nm in the following experiments was chosen as reference for the estimation of the dispersed SWCNTs concentration.
3.2. Optimal Original Concentrations of SWCNTs DispersedThe initial concentrations of the carbon nanotubes is the total amount of the carbon nanotubes added to the SDS solution, and the carbon nanotubes cannot be fully dispersed to the solution. The purpose of this experiment is by studying different dispersing effect of the initial concentration of carbon nanotubes, to determine an optimal initial amount of carbon nanotubes in the solution. Figure 2 shows the UV-visible spectra of dispersed different original concentrations of SWCNTs in aqueous (1.5 wt%) SDS solution. As show in Figure 2, the carbon nanotube concentration is higher, the absorption intensity is stronger. When the concentration of the carbon nanotubes is 0.4 mg/ml, the absorption intensity is the highest. The concentration of carbon nanotubes continues to increase, the absorption intensity began to decline. In other words, the best dispersion of the SWCNTs is 0.4 mg/ml. The results indicate that when the SDS concentrations of certain circumstances, increase the quantity of the carbon nanotubes can enhance the dispersing effect, but when the concentration of the carbon nanotubes reach a certain degree, continue to increase the number of the carbon nanotubes, carbon nanotubes has been dispersed in the solution will make reunited, reduced the dispersion effect.
3.3. Optimization of Surfactant Concentration for SWCNTs DispersionWe investigated the effects of different SDS concentrations on SWCNTs dispersion as show in Figure 3. As the results, increasing the concentration of SDS, the absorbance intensity reaches a plateau value, when the concentration of SDS is 1.5 wt%, the absorption intensity is the highest. In order to visualize the dispersion state of SWCNTs in different concentration of SDS solutions,
SEM investigations have been performed to support the interpretation of the UV-vis-NIR results. As show in Figure 4, dispersing SWCNTs in 0.5 wt% or 2.5 wt% SDS solutions, only a small amount of SWCNTs is exfoliated. This corresponds to the low absorption in the UVvis-NIR spectrum, as shown in Figure 3. When the SWCNTs are dispersed in 1.5 wt% SDS solutions, many SWCNTs are exfoliated. The utmost dispersion of SWCNTs in solution cannot be achieved at lower or higher SDS solutions. Because surfactant adsorption at the interface enriched monomolecular layer, when the surface adsorption reaches saturation, the surfactant molecules at the surface cannot continue enrichment, and the hydrophobic interaction of the hydrophobic group is still vigorously promote based molecules to escape from the water environment, so the surfactant molecules in the inside of the concentrated solution self-polymerization, forming the simple micelles. And begin to form micelles when the concentration of the surfactant called the critical micelle concentration. So at lower SDS solutions, the SDS is not enough to coat on the surface of the SWCNTs to prevent re-aggregation, at higher SDS solutions, when the surfactant concentration is too high, it exceeds the critical micelle concentration, so that the large number of surfactant to form micelles, cannot be sufficiently dispersed carbon nanotubes. Then sufficient surfactant is required to produce an efficient coating on the surface of the SWCNTs to prevent re-aggregation. In our work the optimal concentrations of SDS is 1.5 wt%.
3.4. Effects of Sonication Time on SWCNTs DispersionWe investigated the effects of sonication time on SWCNTs dispersion. UV-vis-NIR spectra of SWCNT-
SDS solutions with different sonication time as show in Figure 5. The result suggest that longer sonication time will helps the carbon nanotubes dispersion, which is consistent with reports in the literature relative [21]. From Figure 6, we can also found that the extension of ultrasonic time can break the bundles of carbon nanotubes into single. But from Figure 5, the curve of ultrasound six hours of the samples than ultrasound four hours more smooth. That because long time of ultrasound can provide more energy to weaken the interaction force between the carbon atoms in the carbon nanotubes, which makes the carbon nanotubes prone to breakage or other damage, make its characteristic peak disappeared. So an appropriate extension of ultrasonic time can effectively disperse carbon nanotubes, but Ultrasonic time is too long will damage the carbon nanotubes.
3.5. Effects of Sonication Temperature on SWCNTs DispersionThe process of dispersion of SWCNT into aqueous solution needs sonication. Many researchers have mentioned the need ice water bath in the process [22]. We also investigated the effects of sonication temperature on SWCNTs dispersion (Figures 7 and 8). As show in Figure 7, the absorbance intensities are higher at 24˚C than those at 30˚C. That means more SWCNTs are dispersed. However, when the sonication temperature is 30˚C, the peaks were disappeared indicating that SWCNTs regenerated to be bundles. Similarly, the TEM images confirmed it as show in Figure 8. So when the ultrasound temperature is 30˚C, the energy in the ultrasonic process makes the actual temperature exceed 30˚C, ultrasonic temperature too high makes SDS lapse, reunion together, cannot be wrapped in the surface of carbon nanotubes dispersion of carbon nanotubes. But when the ultrasound
temperature is 24˚C, although the ultrasonic actual temperature will rise in the process, SDS is not lapsed, the surface of carbon nanotubes can be covered by SDS, enhance the dispersion effect. The results indicating that low temperature contribute to the dispersion of carbon nanotubes.