Unique Identification of Single-Walled Carbon Nanotubes in

Sep 16, 2014 - ... School of Materials, University of Manchester, Oxford Road, Manchester M13 9PL, U.K.. §. College of Engineering, Maths & Physical ...
1 downloads 0 Views 2MB Size
Download PDF
Article pubs.acs.org/JPCC

Unique Identification of Single-Walled Carbon Nanotubes in Electrospun Fibers Libo Deng,†,‡ Robert J. Young,*,‡ Rong Sun,*,† Guoping Zhang,† Daoqiang Daniel Lu,*,† Hui Li,† and Stephen J. Eichhorn§ †

Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences, Shenzhen 518055, China Materials Science Centre, School of Materials, University of Manchester, Oxford Road, Manchester M13 9PL, U.K. § College of Engineering, Maths & Physical Sciences, Physics Building, University of Exeter, Stocker Road, Exeter, Devon EX4 4QL, U.K. ‡

S Supporting Information *

ABSTRACT: Single-walled carbon nanotubes (SWNTs) have been exfoliated in a poly(vinyl alcohol) (PVA) matrix using electrospinning. Raman features of SWNTs with single chirality were studied systematically in terms of the band frequency, intensity, and full width at half-maximum (fwhm). Polarized Raman spectroscopy was used to investigate the orientation of SWNTs, and it was found they were highly aligned along the fiber axis. The response of the nanotube G′-band to external strain for SWNT has been found to be dependent on the nanotube chirality, which suggests nonuniform efficiency of mechanical reinforcement for different nanotube species. The preparation and characterization methods demonstrated in this study have led to a better understanding of the effects of aggregation state, chirality, and external strain on the properties of nanotubes that are incorporated in a polymer matrix.

1. INTRODUCTION

In this paper, SWNT bundles were separated and dispersed in a poly(vinyl alcohol) (PVA) matrix by ultrasonication followed by electrospinning of a PVA/SWNT solution. Raman spectra of nanotubes that were well dispersed in electrospun fibers were studied. The advantages of studying individual nanotubes using PVA/SWNT electrospun fibers compared to the nanotubes deposited on silicon or in dilute suspensions lie in that stress (or strain) can be applied to the matrix and deformation behavior of individual nanotubes can be studied in an electrospun fiber. The effect of the environment on the RBMs, G- and G′-bands in terms of their band position, intensity, and full width at half-maximum (fwhm, denoted as 2Γ) for SWNTs with single chirality were systematically investigated. Furthermore, the response of the G′-band of nanotubes to external strain was also investigated, which represents the first example of following the G′-band shifts during uniaxial deformation of single chirality nanotubes.

Understanding the behavior of single-walled carbon nanotubes (SWNTs) at the single nanotube level is crucial for a range of nanotube-based applications. Raman spectroscopy is a simple yet powerful technique to characterize SWNTs.1 It is interesting to note that Raman bands of individual nanotubes are significantly different from those of nanotube bundles due to the effect of bundling on the electronic transition energies of nanotubes.2 In addition to the effect of aggregation state, other environmental factors such as the sample substrate and whether the nanotube is wrapped by a surfactant or polymer can also influence the transition energies, which in turn influence the features of Raman spectra.3−6 For example, it is well established that the band position of the radial breathing modes (RBM), ωRBM, is dependent on the nanotube diameter through the relation ωRBM = A/dt + B. Various values for the parameters A and B have however been proposed in the literature for different samples.7−9 A systematic study of the effect of the environment is required if nanotubes are to be deposited on a substrate or embedded in a matrix for practical applications. Isolated individual nanotubes are obtained mostly by controlled chemical vapor deposition on a substrate or exfoliation of nanotube bundles in a solvent with the assistance of a dispersion agent such as surfactants, DNA and so on.10−14 To realize the full potential of nanotubes, it is necessary to incorporate them into polymer matrices. Their separation and dispersion in a polymer matrix, however, is a great challenge. © 2014 American Chemical Society

2. EXPERIMENTAL SECTION 2.1. Preparation of the PVA/SWNT Solution. To prepare PVA/SWNT solutions for electrospinning, 6 g of PVA (Sigma-Aldrich, molecular weight of 85 000−124 000) was dissolved in 48 g of deionized water in a glass vial in which the temperature was maintained at 90 °C. 2 mg of SWNTs Received: April 29, 2014 Revised: September 9, 2014 Published: September 16, 2014 24025

dx.doi.org/10.1021/jp504176b | J. Phys. Chem. C 2014, 118, 24025−24033

The Journal of Physical Chemistry C

Article

(Carbon Nanotechnology Inc., grown by a HiPco process) was dispersed in 6 g of deionized water using a sonic probe (ColeParmer Ultrasonic processor CPX 750) with a power of 350 W. The probe was used for 2 min and left to rest for 2 min before the power was applied again. The SWNT/water suspension was ultrasonicated for a total of 30 min, with 30 min of rest. The suspension was then added into the PVA solution and stirred intensively for 15 min. The PVA/SWNT blend solution was then ultrasonicated again using the sonic probe for another 60 min, with 60 min of rest. The concentration of nanotubes relative to the PVA polymer was approximately 0.03 wt %. 2.2. Electrospinning of PVA/SWNT Fibers. Electrospinning was carried out using the following conditions: electric voltage = 20 kV, flow rate = 0.02 mL min−1, and the needle-tip-to-collector distance = 8 cm. A paper frame was placed on top of a pair of grounded electrodes which were separated by 8 mm to collect the fibers. The collection time was limited to 5 s so that only a few isolated fibers were collected. 2.3. Deformation of the Fibers. The electrospun fibers deposited on the paper frames had diameters of approximately 1 μm. The two ends of the electrospun fiber were fixed using Scotch tape. The sample was then mounted on a tensile rig. Both sides of the paper frame were burnt away with caution to avoid breaking of the fiber prior to tensile deformation. The gauge length of the window was 2 mm, and a minimum strain step of 0.125% was applied by turning the micrometer on the tensile rig. 2.4. Raman Characterization. Raman spectra were obtained using a Renishaw 1000 Raman microscope system with a 633 nm laser. The laser beam was focused on the sample surface with a 50× objective lens to a spot size of about 2 μm. The aperture of the lens was 0.75, and the resolution of the spectrometer was 1 cm−1. To prevent heating induced by the laser, a minimum laser power of ∼1 mW was used. The exposure time was set at 10 s, and 30 accumulations were used. For deformation testing, a local reference point on the fiber was followed to return to the same location during deformation. Polarized Raman spectroscopy was used to characterize the orientation of nanotubes in the fibers. A rotary stage was employed to change the angle between the sample axis and the direction of laser polarization, and Raman spectra were recorded at different angles.

Figure 1. Raman spectra of the SWNT powder and a single PVA/ SWNT electrospun fiber.

electrospun fibers in our case. It is thought the ID/IG ratio for each nanotube chirality is different. The lower ID/IG ratio in the electrospun fiber could be due to the active nanotube having a lower D-band intensity than the nanotube powder for which the D-band results from all nanotube chiralities. Figure 2 shows a Raman spectrum in the radial breathing mode region taken for the HiPco SWNT powder. With the 633

3. RESULTS AND DISCUSSION 3.1. RBM of Individual SWNTs. The electrospun fibers had an average diameter of approximately 1 μm and could be manipulated under an optical microscope. Raman spectra of the HiPco SWNT powder and a single electrospun fiber are shown in Figure 1. Characteristic Raman bands of the SWNTs including the RBM peak in the range of 100−400 cm−1, the Dband at 1200−1400 cm−1, the G-band at 1500−1700 cm−1, and the G′-band at 2500−2800 cm−1 can be seen clearly in the fiber even with an extremely low loading of SWNTs. This is due to the resonance effect which enhances the signal greatly.4 It was noted the ID/IG intensity ratio for the electrospun fiber in which the nanotubes have been rigorously sonicated was even lower than that for the nanotube powders (0.03 vs 0.06). López-Lorente et al. demonstrated that for nanotubes dispersed in ethanol and deposited on a glass slide a lower aggregation level leads to a higher ID/IG ratio.15 However, the Raman spectra were taken from nanotube bundles (reaggregrated during evaporation of the solvent) in their work whereas spectra from single chirality nanotubes were obtained from the

Figure 2. Low-frequency Raman spectrum of the SWNT powder excited with the 633 nm laser. The black numbers denote the values of (2n + m) for the family of nanotubes contributing to the Raman peaks.

nm laser (Elaser = 1.96 eV), resonance occurs for an energy transition between the first valence band and conduction band (EM 11) for metallic nanotubes and between the second pair of sub-bands (ES22) for semiconducting nanotubes. The peaks below 225 cm−1 are from metallic nanotubes whereas the peaks above 250 cm−1 are from semiconducting nanotubes. The natural fwhm of the RBM peak of an individual nanotube is 3 cm−1;1 thus, each of the broad bands in Figure 2, centered at 195 cm−1 (2Γ = 14 cm−1), 219 cm−1 (2Γ = 11 cm−1), and 257 cm−1 (2Γ = 12 cm−1), have contributions from a group of nanotubes. The question arises, how many peaks should be used to fit the spectrum? This can be answered if the chiralities of the resonant nanotubes are known. We determined the 24026

dx.doi.org/10.1021/jp504176b | J. Phys. Chem. C 2014, 118, 24025−24033

The Journal of Physical Chemistry C

Article

Figure 3. RBM peaks and their corresponding G-bands and G′-bands for the PVA/SWNT electrospun fibers.

chirality by scanning the laser along the electrospun fibers, as described in the following sections. The nanotube chirality is characterized by a pair of indices (n, m), and those with the same value of (2n + m) can be classified into the same family.1 The family members are close in RBM frequency (sometimes overlapping each other), which usually makes the RBM bands appear broad in the spectra of SWNT bundles. The RBM family bands become separated as the nanotube diameter changes, such as for the peaks at 282 and 295 cm−1 in Figure 2. There are four nanotube families identified in the HiPco bundles when excited with the 633 nm laser, i.e., (2n + m) = 30, 27, 23, and 19, respectively. Within a single nanotube family, both the diameter and chiral angle decrease with increasing RBM frequency.1 In contrast to the SWNT powder which shows multiple RBM peaks and uniform spectra throughout the sample, the PVA/SWNT (0.03 wt %) electrospun fibers exhibit a single peak, double peaks, multiple peaks, or no peak at all in the lowfrequency region from different areas along the fibers (the frequency of the number of RBM peaks observed from the electrospun fibers is shown in Figure S1). The areas which exhibit single RBM peaks are of particular interest because their presence implies nanotubes of a single chirality are in resonance with the laser. Eleven different RBM peaks with a typical fwhm of 5 cm−1 were observed using the 633 nm laser. Figure 3 shows examples of single RBM peaks, along with their corresponding G-bands and G′-bands (see Figure S2 for a full list of the Raman bands for all nanotube chiralities). Considering the volume fraction of nanotubes in the fiber to be approximately 0.03% (since the SWNTs and PVA have about the same density ∼1.3 g/cm3) in our case, there are roughly 300 nanotubes on average in the cross section of the fiber (assuming an average diameter of the tubes dt ∼ 1 nm and that they are arranged end-to-end in the fibers so that nanotubes have the same length as the fibers). This means there are on average 300 nanotubes within the laser spot area, from which the Raman spectra were taken. Among the multiple nanotubes within this laser spot, only one nanotube is found to be in resonance with the energy of the excitation laser,16 which

allowed Raman characteristics of a single chirality nanotube to be investigated. To identify the chirality of the RBM peaks, the electronic transition energies Eii for different nanotubes (i.e., the Kataura plot) need to be known. A large number of values of transition energies for a large range of nanotube chiralities have been reported in the literature. It appears each specific sample has its own set of Eii values for the nanotubes,3,9,17,18 which is attributed to environmental effects. There are no Eii data for HiPco SWNTs dispersed in PVA available to date. It is assumed that PVA has an effect similar to sodium dodecyl sulfate (SDS)/water upon the Eii values of HiPco SWNTs. The data reported by Bachilo and Fantini et al.8,9 for aqueous solutions of SDS/SWNT have therefore been adapted here. The experimental Kataura plot adapted from the literature8,9 for isolated SWNTs is shown in Figure 4. The circles and

Figure 4. Experimental Kataura plot for isolated SWNTs.8,9

squares denote semiconducting and metallic nanotubes, respectively, the red line indicates the excitation energy of the 633 nm laser, and the orange box represents the resonance window. The black numbers denote the nanotube family by (2n + m), and the red numbers denote the nanotube chiralities of the RBMs observed from the electrospun fibers. As mentioned 24027

dx.doi.org/10.1021/jp504176b | J. Phys. Chem. C 2014, 118, 24025−24033

The Journal of Physical Chemistry C

Article

increasing RBM frequency,21 was also used to identify the chirality. For example, the branch 30 contains nanotubes from small to large diameter (15, 0), (14, 2), (13, 4), (12, 6), (11, 8), and (10, 10). Five of them are expected to be in resonance with the 633 nm laser, and we detected the first four. Therefore, the 201 cm−1 peak was assigned to (14, 2). The dispersion state characterized using a 785 nm laser provides strong evidence of isolated individual nanotubes in the electrospun fibers. It has been reported in the literature that when using the 785 nm excitation laser, the 267 cm−1 peak (the so-called roping in peak) which corresponds to a (10, 2) nanotube appears only in nanotube bundles, and it disappears when the nanotube is isolated.2,22 The Raman bands acquired using the 785 nm laser are shown in Figures S3−S5, and the roping in peak was absent when single RBM peaks were observed, which suggests that nanotubes are probably individualized in some areas in the fiber. The chirality assignments for the nanotubes in resonance with the 785 nm laser are listed in Table S1. The nanotubes identified in the electrospun fibers should also be present in the SWNT powder. The spectrum shown in Figure 2 has therefore been fitted with 11 peaks. As can be seen from Figure 4, other nanotube species such as the (11, 8) and (11, 5) nanotubes are expected to be in resonance with the 633 nm laser but have not been detected in the electrospun fibers. This may be due to their low populations in the fibers. It is also possible these nanotubes are no longer in resonance when they interact with the polymer matrix. The RBM frequencies are increased by ∼1−6 cm−1, and the fwhms are decreased by ∼5 cm−1 for nanotubes in the fibers compared to those in the SWNT powder. The upshift of the RBM frequency could be due to the interaction of the nanotubes with the surrounding environment. The similarity found in the electrospun fibers and in the SDS/SWNT/water solutions suggests similar environmental effects on nanotubes in these two samples. It also suggests the nanotubes are well separated in the electrospun fibers. 3.2. G-Band. The G-band also reveals the dispersion state of nanotubes as its line shape for individual nanotubes is different from that for nanotube bundles.23,24 Figure 5a shows the Gbands and their corresponding RBMs obtained from the SWNT powder, and Figures 5b and 5c show the spectra for PVA/SWNT electrospun fibers with different levels of dispersion. In Figure 5b, the nanotubes were ultrasonicated for a relatively short time (20 min) and were thus poorly dispersed in the fibers. While in Figure 5c, the nanotubes were ultrasonicated for a longer sonication time (90 min) and individualized as evident by the presence of a single and narrow RBM peak. The G-band from the SWNT powder shows a dominating semiconducting nature and can be fitted with four Lorentzian peaks, with their centers at 1541, 1553, 1591, and 1600 cm−1. The two higher-frequency peaks originate from the vibration along the nanotube axis, and the two lower-frequency peaks are related to the vibration in the circumferential direction. The Raman band located at 1600 cm−1 vanishes as the bundle size decreases, as can be seen from Figures 5b and 5c. When a single nanotube is in resonance with the laser, the corresponding G-band can be fitted with two peaks locating at ∼1588 and ∼1552 cm−1 which are denoted as the G+ and G− bands, respectively, as shown in Figure 5c.1 The shift of G+ frequency relative to the nanotube powder also suggests the separation of the nanotubes.25 The G+ fwhm decreases from 16

already, resonance occurs when the Eii transition energies of nanotubes matches the energy of the laser, Elaser; the RBM intensity becomes weaker when the two energies deviate from each other. The maximum deviation of the Eii from Elaser before a given nanotube becomes undetectable is described by the socalled resonance window. This parameter can be determined by measuring the RBM intensity at different Elaser values and is influenced by many factors such as the aggregation state and the chiral angle of the nanotubes. The resonance window is ±0.1 eV for nanotube bundles, while for isolated nanotubes dispersed in water and wrapped by SDS, it is around ±0.06 eV.1 A value of ±0.1 eV is used here as the resonance window for nanotubes dispersed in polymers can be broadened. This is because the polymer matrix might exhibit different environmental effects on the electronic properties of nanotubes, such as inhomogeneous residual stress and charge transfer. The resonance window (Elaser = ± 0.1 eV) around the red laser line (633 nm) is indicated by an orange box in Figure 4. Each of the 11 RBM peaks that are seen in Figure 3 and Figure S2 is then associated with a specific (n, m) species. The chirality is directly assigned for each RBM peak as we already know from the experimental Kataura plot which species are expected to be in resonance. The assignment procedure is also shown in Figure 4. It should be noted that the RBM frequency for a single chirality determined from different areas along the fibers was not a constant value, due possibly to the resolution of the Raman spectrometer. Slightly different RBMs were assigned to the same nanotube species when they met some basic criteria such as (1) the spectrum could be fitted by a single Lorentzian peak, (2) the difference between the band frequency from the average value was less than 2 cm−1, and (3) the fwhm was lower than 10 cm−1. The chiralities for the RBMs are assigned using the experimentally determined Eii energies are listed in Table 1. Table 1. Assignment of the RBMs Observed Using the 633 nm Laser experimental RBM in PVA/SWNT fiber (cm−1) 192 196 201 205 217 221 254 258 268 284 298

± 0.3 ± 0.6 ± 0.5 ± ± ± ± ±

1.0 0.8 0.9 1.3 0.9

RBM in HiPco powder (cm−1)

chirality

189 193 197 203 217 220 250 257 261 282 296

(12, 6) (13, 4) (14, 2) (15, 0) (12, 3) (13, 1) (10, 3) (11, 1) (7, 6) (7, 5) (8, 3)

EM 11 (eV)

ES22 (eV)

1.92 1.93 1.92 1.88 2.04 2.02

theoretical

EM 11 (eV)

ES22 (eV)

1.90 1.93 1.96 1.96 2.14 2.16 1.96 2.03 1.92 1.92 1.87

1.84 1.90 1.85 1.94 2.02

The Eii energies calculated using a theoretical model developed in our previously published work are also listed in Table 1 as a comparison.19,20 The RBM frequencies observed from the electrospun fibers are close to those reported for the SDS stabilized SWNTs, and thus these assignments were adopted here. For example, the peak at 254 cm−1 is assigned to a (10, 3) nanotube. A rule proposed by Telg et al., that for a given nanotube branch in which nanotubes have the same 2n + m value both the diameter and chiral angle decrease with 24028

dx.doi.org/10.1021/jp504176b | J. Phys. Chem. C 2014, 118, 24025−24033

The Journal of Physical Chemistry C

Article

Figure 5. G-bands and their corresponding RBMs at different dispersion states: (a) SWNT powder, (b) nanotubes poorly dispersed in an electrospun fiber (ultrasonicated for 20 min), and (c) an individual semiconducting nanotube in an electrospun fiber.

Figure 6. G′-band of (a) SWNT powder and (b) electrospun fibers with poorly dispersed nanotubes. Numbers in parentheses denote the fwhm of the band, and the insets show the corresponding RBM peaks.

cm−1 for the SWNT powder to 9 cm−1 for a single chirality. The G+ fwhm observed here, of 9 cm−1, is comparable to those found from individual nanotubes on a Si substrate, typically in the range of 5−10 cm−1.26 It is also noted the G− intensity decreases significantly (or disappears completely in some cases) as the dispersion improves. This suggests the PVA surrounding the nanotubes can suppress vibrations in the circumferential direction.23 The vanishing of the G− peak and narrowing of the G+ peak are important features of individual nanotubes, which have also been observed by other groups with individual nanotubes exfoliated by different dispersants.23,27 3.3. G′-Band. The G′-band scattering involves a double resonance Raman process and provides valuable information on the electronic structures of nanotubes. Figure 6 shows the G′bands of the nanotube powder and a PVA/SWNT electrospun fiber. It is noted when multiple nanotubes are in resonance with the excitation laser, the G′-band appears as a single peak (or a

broad peak containing a small shoulder) with a variation in the fwhm depending on the size of the nanotube bundle (see Figure S6 in the Supporting Information). The G′-band is broad when a large range of nanotubes with different diameters contribute to this band. In contrast to the single G′-peak observed for nanotube bundles, a double-peak structure for the G′-band was frequently observed when a single (n, m) nanotube was in resonance, as can be seen from Figure 3b. The double-peak structure of the G′-band can also be seen when two different nanotube species are involved in the double resonance, with one of them in incident resonance while another is in scattered resonance. Specifically, the lower-frequency component is associated with the scattered resonance while the higher-component is related to the incident resonance, according to the ωG′ = 2420 + 106Elaser relation.1 24029

dx.doi.org/10.1021/jp504176b | J. Phys. Chem. C 2014, 118, 24025−24033

The Journal of Physical Chemistry C

Article

nanotubes are thus not seen in the RBM region. The G′-band and RBM can be seen simultaneously only when incident resonance occurs.1 3.4. Orientation of the Nanotubes. The degree of alignment of SWNTs can be determined quantitatively using polarized Raman spectroscopy. This is based on the orientation-dependent Raman band intensity of nanotubes: the Raman band intensity is a maximum when the nanotube axis is parallel to both the incident and scatter light and is a minimum when the nanotube is perpendicular to the axis of the laser polarization. The nanotube G-bands obtained at different angles between the fiber axis and the axis of the laser polarization are shown in Figure 8a. Raman spectra were acquired with a VV configuration where both the incident light and scattered light are parallel to the principal axis of both the spectrometer. The intensities at different angles are normalized to that at 0° and are plotted as a function of the angle as shown in Figure 8b. A line generated for an I ∝ cos4 φ function (where φ is the angle between the fiber axis and the axis of laser polarization), which is expected to apply for perfect orientation of the nanotubes, is also shown in Figure 8b.28 The scatter of these data around the line due to the heterogeneity of the sample makes it difficult to perform quantitative analysis of the degree of alignment. The dramatic drop in the intensity of the Raman band at 90° is, however, an indication of a high degree of orientation. Our recent work on cellulose/nanotube composite electrospun fibers also indicated a high degree of orientation of nanotubes along the fiber axis.29 This is because in the electrospinning process, there are large extensional forces acting on tiny jets of the polymer solution, leading to the rigidrod-like nanotubes rotating around their center of gravity and thus orienting along the drawing direction to minimize the torque applied to them.30 3.5. Deformation Behavior. Following the deformation process of a nanotube using Raman spectroscopy allows the electronic properties and mechanical reinforcing efficiency (in composites) to be monitored.31 Most deformation tests have been carried out to date on nanotube bundles which are an ensemble of nanotube chiralities, making it difficult to analyze the properties of a single chirality. With the Raman features and orientation of nanotubes in the electrospun fibers well established, we are now in a position to study the deformation behavior of nanotubes with a single chirality. The deformation behavior of individual nanotubes for six different chiralities was followed using Raman spectroscopy.

Figure 7 shows a double-peak G′-band observed from an electrospun fiber along with the Kataura plot. The scattered

Figure 7. (a) Double-peak structure of the G′-band and the corresponding RBM peak observed from an electrospun fiber. (b) Experimental Kataura plot for isolated HiPco SWNTs showing two groups of nanotubes in incident resonance and scattered resonance.

photon has an energy of 1.63 eV when using the 633 nm incident laser (two phonons, each having an energy of 0.165 eV are emitted in a G′ scattering process). It can be seen from the Kataura plot that there is a group of nanotubes in the resonance window of the scattered photon (in the circled region). The double-peak G′-band observed from the electrospun fibers can therefore be explained as one nanotube in incident resonance and another group of nanotubes in scattered resonance. The nanotubes in resonance with the scattered G′-photon are not in resonance with the scattered RBM photon (the phonon emitted in an RBM scattering process is ∼0.02 eV); these

Figure 8. (a) Raman spectra of nanotubes at different angles between the fiber axis and the axis of laser polarization. (b) Normalized intensity of the G-band as a function of the angle. 24030

dx.doi.org/10.1021/jp504176b | J. Phys. Chem. C 2014, 118, 24025−24033

The Journal of Physical Chemistry C

Article

Figure 9. Raman G′-bands for (a) the (13, 4) and (b) (10, 3) nanotubes in PVA/SWNT electrospun fibers at different strains from bottom to top: 0%, 0.125%, 0.25% and 0.375%. (c) G′-band frequency as a function of the fiber strain for two nanotube chiralities.

The G′-bands for (13, 4) and (10, 3) nanotubes at different strain levels are shown in Figures 9a and 9b (see Figure S7 for the RBM and G′-bands at different strains for all the six nanotube chiralities). Some of the nanotubes that are listed in Tables 1 are absent in Figure S7 due to their low populations. It was found that for all nanotubes the RBM band frequency does not shift significantly during deformation, which is in agreement with the previous experimental findings and theoretical predictions.19,20 In addition, the RBM intensity does not change significantly within the experimental error. This is in contrast with the findings of Lucas et al.,19,20 where the RBM intensity was found to change monotonically with the strain, with the changing trend and magnitude depending on nanotube chirality as well as the excitation laser. The intensity change has been explained by Lucas et al. and well supported by theoretical calculations19,20 that deformation can bring the Eii closer to or further away from the Elaser. Similar to Lucas, the nanotubes were also deformed in this work through deformation of the composites. However, the steady change in RBM intensity was observed using a 785 nm laser whereas a 633 nm laser was used here. The apparent independence of the RBM intensity on the strain found in this case is possibly due to all of the nanotubes being close to maximum resonance or the resonance window of these nanotubes being broadened by some additional factors, such as the interaction between nanotubes and the polymer matrix, not taken into account in the theoretical calculation.32 The variation of the G′-band frequency for the (13, 4) and (10, 3) nanotubes under tensile strain is shown in Figure 9c. It can be seen in the strain range of 0%−0.6% the G′-bands for both nanotubes shift linearly with strain and large band shift rates are observed, suggesting good efficiency of stress transfer

in this range. When the strain exceeds 0.8%, the G′-band frequency deviates from this monotonic shift behavior, which could be due to interfacial slippage in the composites or yielding of the polymer matrix.16 The shift rate of −36.0 cm−1 %−1 observed for the (13, 4) nanotube is almost 3× that for the (10, 3) nanotube. It is concluded that the stress-induced G′-band shift rate for SWNTs is chirality-dependent. In fact, the chirality dependence of Raman bands’ responses to external perturbation has been observed by a number of groups in different situations.33−35 Gao et al. stretched isolated individual SWNTs and followed the G-band position during deformation. They demonstrated a clear chirality-dependent Raman response to uniaxial strain whereby the G-band shift rate for individual nanotube increases with the increase of diameter and decrease of the chiral angle. For example, the G+ shift rate for the (7, 5) and (18, 5) nanotubes is 3.95 and 14.59 cm−1/%, respectively.34 Chang et al. also stretched isolated individual SWNTs and observed that the G-band shift rates span from 6.2 to 23.6 cm−1/% for nanotubes with different chiralities.33 Different shifts of the Gband for different chiralities have been reported in a number of experimental and theoretical studies.33,34,36,37 This phenomenon can be explained by the effect of chiral angle on the projection of the strain vector onto the C−C bonds. No data on the G′-band shift as a function of strain for individual nanotubes, however, have been reported to the best of our knowledge, due possibly to that the G′-band is weaker than the G-band in most cases. The chirality-dependent G′-band shift rate can be understood since this band is related to the Eii transition, and the shift of the Eii energy under strain is dependent on the nanotube chirality.19 Further work is needed 24031

dx.doi.org/10.1021/jp504176b | J. Phys. Chem. C 2014, 118, 24025−24033

The Journal of Physical Chemistry C

Article

(5) Nascimento, G. M. d.; Hou, T.; Kim, Y. A.; Muramatsu, H.; Hayashi, T.; Endo, M.; Akuzawa, N.; Dresselhaus, M. S. Single-wall carbon nanotube interactions with copper-oxamato building block of molecule-based magnets probed by resonance Raman spectroscopy. J. Raman Spectrosc. 2011, 41, 1587−1593. (6) Shim, M.; Ozel, T.; Gaur, A.; Wang, C. J. Insights on charge transfer doping and intrinsic phonon line shape of carbon nanotubes by simple polymer adsorption. J. Am. Chem. Soc. 2006, 128, 7522− 7530. (7) Araujo, P. T.; Doorn, S. K.; Kilina, S.; Tretiak, S.; Einarsson, E.; Maruyama, S.; Chacham, H.; Pimenta, M. A.; Jorio, A. Third and fourth optical transitions in semiconducting carbon nanotubes. Phys. Rev. Lett. 2007, 98, 067401. (8) Bachilo, S. M.; Strano, M. S.; Kittrell, C.; Hauge, R. H.; Smalley, R. E.; Weisman, R. B. Structure-assigned optical spectra of singlewalled carbon nanotubes. Science 2002, 298, 2361−2366. (9) Fantini, C.; Jorio, A.; Souza, M.; Strano, M. S.; Dresselhaus, M. S.; Pimenta, M. A. Optical transition energies for carbon nanotubes from resonant Raman spectroscopy: Environment and temperature effects. Phys. Rev. Lett. 2004, 93, 147406 1−4. (10) Kwon, T.; Lee, G.; Choi, H.; Strano, M. S.; Kim, W. J. Highly efficient exfoliation of individual single-walled carbon nanotubes by biocompatible phenoxylated dextran. Nanoscale 2013, 5, 6773−6778. (11) Liu, Y.; Chipot, C.; Shao, X.; Cai, W. Edge effects control helical wrapping of carbon nanotubes by polysaccharides. Nanoscale 2012, 4, 2584−2589. (12) Zheng, M.; Jagota, A.; Semke, E. D.; Diner, B. A.; Mclean, R. S.; Lustig, S. R.; Richardson, R. E.; Tassi, N. G. DNA-assisted dispersion and separation of carbon nanotubes. Nat. Mater. 2003, 2, 338−342. (13) Liu, K.; Hong, X.; Zhou, Q.; Jin, C.; Li, J.; Zhou, W.; Liu, J.; Wang, E.; Zettl, A.; Wang, F. High-throughput optical imaging and spectroscopy of individual carbon nanotubes in devices. Nat. Nanotechnol. 2013, 8, 917−922. (14) Bonhommeau, S.; Deria, P.; Glesner, M. G.; Talaga, D.; Najjar, S.; Belin, C.; Auneau, L.; Trainini, S.; Therien, M. J.; Rodriguez, V. Raman spectroscopic investigation of individual single-walled carbon nanotubes helically wrapped by ionic, semiconducting polymers. J. Phys. Chem. C 2013, 117, 14840−14849. (15) López-Lorente, A. I.; Simoneta, B. M.; Valcárcel, M. Raman spectroscopic characterization of single walled carbon nanotubes: influence of the sample aggregation state. Analyst 2014, 139, 290−298. (16) Kannan, P.; Eichhorn, S. J.; Young, R. J. Deformation of isolated single-wall carbon nanotubes in electrospun polymer nanofibres. Nanotechnology 2007, 18, 235707−235713. (17) Strano, M. S. Probing chiral selective reactions using a revised Kataura plot for the interpretation of single-walled carbon nanotube spectroscopy. J. Am. Chem. Soc. 2003, 125, 16148−16153. (18) Weisman, R. B.; Bachilo, S. M. Dependence of optical transition energies on structure for single-walled carbon nanotubes in aqueous suspension: An empirical Kataura plot. Nano Lett. 2003, 3, 1235− 1238. (19) Lucas, M.; Young, R. J. Effect of uniaxial strain deformation upon the Raman radial breathing modes of single-wall carbon nanotubes in composites. Phys. Rev. B 2004, 69, 085405. (20) Lucas, M.; Young, R. J. Unique identification of single-walled carbon nanotubes in composites. Compos. Sci. Technol. 2007, 67, 2135−2149. (21) Telg, H.; Maultzsch, J.; Reich, S.; Hennrich, F.; Thomsen, C. Chirality distribution and transition energies of carbon nanotubes. Phys. Rev. Lett. 2004, 93, 177401. (22) Doorn, S. K.; Strano, M. S.; Haroz, E. H.; Rialon, K. L.; Hauge, R. H.; Smalley, R. E. Capillary electrophoresis separations of bundled and individual carbon nanotubes. J. Phys. Chem. B 2003, 107, 6063− 6069. (23) Kawamoto, H.; Uchida, H.; Kojima, T.; tachibana, M. G band Raman features of DNA-wrapped single-wall carbon nanotubes in aqueous solution and air. Chem. Phys. Lett. 2006, 432, 172−176.

to establish the detailed relationship between the G′-band shift rate and the chiral structure of SWNTs.

4. CONCLUSIONS The behavior of Raman bands of SWNTs that were dispersed well in PVA electrospun fibers has been studied systematically. Single RBM peaks were observed in the fibers, which allowed the properties of nanotubes with single chirality to be monitored. It has been found the features (including the frequency, the intensity, and the fwhm) of Raman spectra of nanotubes with single chirality were significantly different from those of multichiralities. Polarized Raman spectroscopy suggested alignment of the nanotubes along the longitudinal axis of the electrospun fibers. Tensile deformation was applied to the nanotubes by stretching the electrospun fibers, and the G′ shift per unit strain was found to vary with the variation of chirality, which suggests nonuniform efficiency of mechanical reinforcement for different nanotube species. The preparation and characterization methods demonstrated in this study have led to a better understanding of the effects of aggregation state, chirality, and external strain on the properties of nanotubes. This could foster the practical applications of nanotubes such as in strain sensors, optoelectronic devices, and as mechanical reinforcement.

■

ASSOCIATED CONTENT

S Supporting Information *

Raman bands of individual nanotubes obtained using a different laser, SEM images of nanotubes at different aggregation states, and the Raman bands at different strain levels. This material is available free of charge via the Internet at http://pubs.acs.org.

■

AUTHOR INFORMATION

Corresponding Authors

*E-mail [email protected], Tel +44-161-3063550 (R.J.Y.). *E-mail [email protected], Tel +86-755-86392158 (R.S.). *E-mail [email protected], Tel +86-755-86392158 (D.D.L.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was partially supported by Guangdong and Shenzhen Innovative Research Team (Program No. 2011D052, KYPT20121228160843692), National Natural Science Foundation of China (Grant No. 21201175), and R&D Funds for basic Research Program of Shenzhen (Grant No. JCYJ20120615140007998).

■

REFERENCES

(1) Dresselhaus, M. S.; Dresselhaus, G.; Saito, R.; Jorio, A. Raman spectroscopy of carbon nanotubes. Phys. Rep. 2005, 409, 47−99. (2) Heller, D. A.; Barone, P. W.; Swanson, J. P.; Mayrhofer, R. M.; Strano, M. S. Using Raman spectroscopy to elucidate the aggregation state of single-walled carbon nanotubes. J. Phys. Chem. B 2005, 108, 6905−6909. (3) Araujo, P. T.; Pesce, P. B.; Dresselhaus, M. S.; Sato, K.; Saito, R.; Jorio, A. Resonance Raman spectroscopy of the radial breathing modes in carbon nanotubes. Physica E 2010, 42, 1251−1261. (4) Dresselhaus, M. S.; Dresselhaus, G.; Jorio, A.; Filho, A. G. S.; Saito, R. Raman spectroscopy on isolated single wall carbon nanotubes. Carbon 2002, 40, 2043−2061. 24032

dx.doi.org/10.1021/jp504176b | J. Phys. Chem. C 2014, 118, 24025−24033

The Journal of Physical Chemistry C

Article

(24) Paillet, M.; Poncharal, P.; Zahab, A.; Sauvajol, J. L.; Meyer, J. C.; Roth, S. Vanishing of the Breit-Wigner-Fano component in individual single-wall carbon nanotubes. Phys. Rev. Lett. 2005, 94, 237401 1−4. (25) Yan, X.; Suzuki, T.; Kitahama, Y.; Sato, H.; Itoh, T.; Ozaki, Y. A study on the interaction of single-walled carbon nanotubes (SWCNTs) and polystyrene (PS) at the interface in SWCNT-PS nanocomposites using tip-enhanced Raman spectroscopy. Phys. Chem. Chem. Phys. 2013, 15, 20618−20624. (26) Jorio, A.; Filho, A. G. S.; Dresselhaus, G.; Dresselhaus, M. S.; Swan, A. K.; Unlu, M. S.; Goldberg, B. B.; Pimenta, M. A.; hafner, J. H.; Lieber, C. M.; Saito, R. G-band resonant Raman study of 62 isolated single-wall carbon nanotubes. Phys. Rev. B 2002, 65, 155412. (27) Karajanagi, S. S.; Yang, H.; Asuri, P.; Sellito, E.; Dordick, J. S.; Kane, R. S. Protein-assisted solubilization of single-walled carbon nanotubes. Langmuir 2006, 22, 1392−1395. (28) Hwang, J.; Gommans, H. H.; Ugawa, A.; Tashiro, H.; Haggenmueller, R.; Winey, K. I.; Fischer, J. E.; Tanner, D. B.; Rinzler, A. G. Polarized spectroscopy of aligned single-wall carbon nanotubes. Phys. Rev. B 2000, 62, R13310−R13313. (29) Deng, L.; Young, R. J.; Kinloch, I. A.; Abdelkader, A.; Holmes, S. M.; Rio, D. A. D. H.-D.; Eichhorn, S. J. Supercapacitance from cellulose and carbon nanotube nano-composite fibers. ACS Appl. Mater. Interfaces 2013, 5, 9983−9990. (30) Greiner, A.; Wendorff, J. H. Electrospinning: A fascinating method for the preparation of ultrathin fibres. Angew. Chem., Int. Ed. 2007, 46, 5670−5703. (31) Deng, L.; Eichhorn, S. J.; Kao, C. C.; Young, R. J. The effective Young’s modulus of carbon nanotubes in composites. ACS Appl. Mater. Interfaces 2011, 3, 433−440. (32) Cronin, S. B.; Swan, A. K.; Ü nlü, M. S.; Goldberg, B. B.; Dresselhaus, M. S.; Tinkham, M. Resonant Raman spectroscopy of individual metallic and semiconducting single-wall carbon nanotubes under uniaxial strain. Phys. Rev. B 2005, 72, 035425. (33) Chang, C. C.; Hsu, I. K.; Aykol, M.; Hung, W. H.; Chen, C. C.; Cronin, S. B. A new lower limit for the ultimate breaking strain of carbon nanotubes. ACS Nano 2010, 4, 5095−5100. (34) Gao, B.; Jiang, L.; Ling, X.; Zhang, J.; Liu, Z. Chiralitydependent Raman frequency variation of single-walled carbon nanotubes under uniaxial strain. J. Phys. Chem. C 2008, 112, 20123− 20125. (35) Ghandour, A. J.; Crowe, I. F.; Proctor, J. E.; Sun, Y. W.; Halsall, M. P.; Hernandez, I.; Sapelkin, A.; Dunstan, D. J. Pressure coefficients of Raman modes of carbon nanotubes resolved by chirality: Environmental effect on graphene sheet. Phys. Rev. B 2013, 87, 085416. (36) Li, L.; Chang, T.; Li, G. Strain dependent G-band mode frequency of single-walled carbon nanotubes. Carbon 2011, 49, 4412− 4419. (37) Yang, W.; Wang, R.; Yan, H. Strain-induced Raman-mode shift in single-wall carbon nanotubes: Calculation of force constants from molecular-dynamics simulations. Phys. Rev. B 2008, 77, 195440.

24033

dx.doi.org/10.1021/jp504176b | J. Phys. Chem. C 2014, 118, 24025−24033