Anda di halaman 1dari 6


Sorting carbon nanotubes by electronic

structure using density differentiation
Department of Materials Science and Engineering, Northwestern University, Evanston, Illinois 60208-3108, USA

Published online: 4 October 2006; doi:10.1038/nnano.2006.52

The heterogeneity of as-synthesized single-walled carbon nanotubes (SWNTs) precludes their widespread application in
electronics, optics and sensing. We report on the sorting of carbon nanotubes by diameter, bandgap and electronic type using
structure-discriminating surfactants to engineer subtle differences in their buoyant densities. Using the scalable technique of
density-gradient ultracentrifugation, we have isolated narrow distributions of SWNTs in which >97% are within a 0.02-nm-
diameter range. Furthermore, using competing mixtures of surfactants, we have produced bulk quantities of SWNTs of
predominantly a single electronic type. These materials were used to fabricate thin-film electrical devices of networked SWNTs
characterized by either metallic or semiconducting behaviour.

Carbon nanotubes have recently received extensive attention1 due to Other methods8,10–18 for sorting SWNTs have been reported
their nanoscale dimensions and outstanding materials properties recently; however, none of these techniques has demonstrated
such as ballistic electronic conduction2, immunity from the simultaneous sensitivity, scalability and effectiveness that we
electromigration effects at high current densities1, and transparent report here. In addition, we have previously exploited density-
conduction3. However, as-synthesized carbon nanotubes vary in gradient ultracentrifugation to enrich DNA-wrapped SWNTs by
their diameter and chiral angle, and these physical variations result diameter and bandgap8. However, there are critical drawbacks
in striking changes in their electronic and optical behaviours1,4. For to using DNA for carbon nanotube functionalization. First,
example, about one-third of all possible SWNTs exhibit metallic DNA-wrapped SWNTs have limited stability in aqueous density
properties and the remaining two-thirds act as semiconductors. gradients and thus are not amenable to the refinements in
Moreover, the bandgap of semiconducting SWNTs scales inversely enrichment gained from repeated centrifugation. Furthermore,
with tube diameter. For instance, semiconducting SWNTs produced complete removal of the DNA wrapping after enrichment has
by the laser-ablation method range from 11 to 16 Å in diameter and not been demonstrated, and sensitivity to electronic type has
have optical bandgaps that vary from 0.65 to 0.95 eV (ref. 5). The not been observed. Finally, the availability and cost of specific,
currently unavoidable structural heterogeneity of as-synthesized custom oligomers of single-stranded DNA are prohibitive.
SWNTs prevents their widespread application as high-performance To overcome these obstacles, we have recently explored
field-effect transistors, optoelectronic near-infrared emitters/ surfactant encapsulating agents in place of DNA and have
detectors, chemical sensors, materials for interconnects in integrated discovered that bile salts and their mixtures with other surfactants
circuits and conductive additives in composites1. Accordingly, their enable the separation of SWNTs by diameter, bandgap and/or
use will be limited until large quantities of these nanomaterials can electronic type. Using surfactants, the isolation of specific chiralities
be produced that are monodisperse in their structure and properties. of SWNTs can be significantly refined by separating in multiple
To address the problem of heterogeneity and enable future successive density gradients. Furthermore, the adsorption of
SWNT-based technologies, we have developed a general approach surfactants to SWNTs is reversible and compatible with a wide range
for sorting carbon nanotubes by diameter, bandgap and electronic of tube diameters. For example, we demonstrate here the sorting of
type (metallic versus semiconducting), using the technique of SWNTs over the diameter range 7–16 Å. Most importantly, the
density-gradient ultracentrifugation. This scalable approach structure–density relationship for SWNTs can be easily controlled
exploits differences in the buoyant densities (mass per volume) by varying the surfactant itself. For instance, by using mixtures of
among SWNTs of different structures and has been adapted from two surfactants that competitively adsorb to the SWNT surface, we
similar techniques used in biochemistry6,7 for the purification of have achieved optimal metal–semiconductor separation.
biological macromolecules, such as nucleic acids and proteins. In
this technique, purification is induced by ultracentrifugation in a
density gradient. In response to the resulting centripetal force, RESULTS
particles sediment toward their respective buoyant densities and COMPARISON OF ENCAPSULATING AGENTS
spatially separate in the gradient. This approach differs from The buoyant density of SWNTs in aqueous solution will subtly
previously reported work on the ultracentrifugation of SWNTs9,10, depend on multiple factors, including the mass and volume of the
in which density gradients were not used. carbon nanotube itself, its surface functionalization and

60 nature nanotechnology | VOL 1 | OCTOBER 2006 |

©2006 Nature Publishing Group

a b c





ρ 0.1

900 1,100 1,300
Wavelength (nm)

d e f g







S33 M11 S22

0.0 0.00
900 1,100 1,300 400 600 800 1,000 1,200
Wavelength (nm) Wavelength (nm)

Figure 1 Sorting of SWNTs by diameter, bandgap and electronic type using density gradient ultracentrifugation. a, Schematic of surfactant encapsulation and
sorting, where r is density. b – g, Photographs and optical absorbance (1 cm path length) spectra after separation using density gradient ultracentrifugation. A rich
structure – density relationship is observed for SC-encapsulated SWNTs, enabling their separation by diameter, bandgap and electronic type. In contrast, no separation
is observed for SDBS-encapsulated SWNTs. b,c, SC encapsulated, CoMoCAT-grown SWNTs (7 – 11 Å). Visually, the separation is made evident by the formation of
coloured bands (b) of isolated SWNTs sorted by diameter and bandgap. Bundles, aggregates and insoluble material sediment to lower in the gradient. The spectra
indicate SWNTs of increasing diameter are more concentrated at larger densities. Three diameter ranges of semiconducting SWNTs are maximized in the third, sixth
and seventh fractions (highlighted by the pink, green and light brown bands). These have chiralities of (6,5), (7,5) and (9,5)/(8,7), and diameters of 7.6, 8.3 and
9.8/10.3 Å respectively. d,e, SDBS-encapsulated CoMoCAT-grown SWNTs (7 – 11 Å). In contrast, all of the SWNTs have converged to a narrow black band (d) and
diameter or bandgap separation is not indicated (e). f,g, SC-encapsulated, laser-ablation-grown SWNTs (11 – 16 Å). Both enrichment by diameter and electronic type
are observed. Visually, coloured bands of SWNTs (f) are apparent, suggesting separation by electronic structure. In the optical absorbance spectra, the second- and
third-order semiconducting (highlighted pink) and first-order metallic (highlighted blue) optical transitions are labelled S22, S33 and M11, respectively5,22. The purple
highlighted regions show where the semiconducting and metallic transitions overlap. The diameter separation is indicated by a red shift in the S22 band for fractions
of increasing density. Additionally, the metallic SWNTs (M11) are depleted in the most buoyant fractions. Dr from top to bottom fraction, and r for the top fraction
for c, e and g are 0.022, 0.096 and 0.026 g cm23 and 1.08, 1.11 and 1.08 + 0.02 g cm23, respectively. pH ¼ 7 for all parts. SWNTs before sorting are depicted as
a dashed grey line in c and g.

electrostatically bound hydration layers. To gain insight into the were sodium cholate (SC), sodium deoxycholate and sodium
structure–density relationship for SWNTs and the role of taurodeoxycholate. The bile salts are more molecularly rigid and
encapsulating agents, we first compared two different families planar amphiphiles with a charged face opposing a hydrophobic
of surfactants—anionic-alkyl amphiphiles19 and bile salts20. one21, which is expected to interact with the SWNT surface (Fig. 1a).
Specifically, we used two amphiphiles with anionic head groups and Initially, we explored the sorting of SWNTs in the 7– 11 Å
flexible alkyl tails: sodium dodecyl sulphate (SDS) and sodium diameter range synthesized by the CoMoCAT method, using SC
dodecylbenzene sulphonate (SDBS). The three bile salts used and SDBS encapsulations, as depicted in Fig. 1. For the case of

nature nanotechnology | VOL 1 | OCTOBER 2006 | 61

©2006 Nature Publishing Group


Photoluminescence intensity (a.u.) (6, 5)

0.0 0.5 1.0

Excitation wavelength (nm)


650 (7, 5)



900 1,000 1,100 1,200

Emission wavelength (nm)

Figure 2 Refinement by repeated centrifugation in density gradients. By successively separating SC-encapsulated SWNTs, the isolation of specific, targeted
chiralities improves. Plotted are photoluminescence intensities as a function of excitation and emission wavelengths. Here, the isolation of the (6,5) and (7,5)
chiralities (circled red and green in the left-most plot) of SWNTs grown by the CoMoCAT-method before sorting, is improved (in the top and bottom panels,
respectively) by successively repeating density gradient centrifugation for three iterations (from left to right). After three iterations of enriching the (6,5) chirality
(7.6 Å), a narrow diameter distribution is achieved in which .97% of semiconducting SWNTs are within 0.2 Å of the mean diameter. Alternatively, refined isolation of
the (7,5) chirality can be realized (bottom). In this case, after three iterations of sorting, the (7,5) chirality (8.3 Å), initially substantially less concentrated than the
(6,5) chirality, becomes dominant. Further improvements may be possible with additional centrifugation cycles.

SC encapsulation, multiple regions of separated SWNTs are REPEATED, REFINED SORTING

visible throughout the density gradient (Fig. 1b). The most Using density-gradient centrifugation with bile salts such as SC,
buoyant region is characterized by SWNTs that have been sorted it is clear that we can enrich SWNTs by both their structure and
into bands of various colours, corresponding to the different electronic properties. However, the degree of isolation achieved
bandgaps of the semiconducting tubes. In contrast, for the case after a single step of the technique is limited by the diffusion of
of SDBS-encapsulated SWNTs, all of the SWNTs are compressed SWNTs during ultracentrifugation, mixing during fractionation,
into a narrow black band (Fig. 1d). and statistical fluctuations in surfactant encapsulation. To
After centrifugation, the separated SWNTs can be removed overcome these limitations and improve the sorting process, the
from the centrifuge tubes, layer by layer, using established centrifugation process can be repeated for multiple cycles. For
techniques for fractionation, and each layer can be optically example, an enriched fraction of SWNTs sorted in a density
characterized to determine quantitatively the mode and quality gradient can be further enriched in a second density gradient.
of separation (see Supplementary Information, Figs S1 and S2, This enables the optimal isolation of a targeted electronic type or
and Methods). For the case of SC-encapsulated SWNTs, the a specific chirality of SWNT. To demonstrate the approach, we
amplitudes of optical absorbance for different transitions in the targeted the enrichment of the (6,5) and (7,5) chiralities of
900–1,340 nm range (first-order semiconducting transitions) semiconducting SWNTs (7.6 and 8.3 Å in diameter, respectively).
indicate separation by diameter and bandgap. More specifically, In Fig. 2, photoluminescence emission –excitation matrices depict
the spectra illustrate that SWNTs of increasingly larger diameters the photoluminescence intensity of semiconducting SWNTs as a
are enhanced at increasingly larger densities. A similar correlation function of excitation and emission wavelengths for SWNTs
between diameter and density was also observed for the cases of before and after each of three iterations of density-gradient
sodium deoxycholate and sodium taurodeoxycholate. However, centrifugation (see Supplementary Information, Fig. S4, for plots
for the case of SDBS (Fig. 1e) and SDS (see Supplementary of the corresponding absorbance spectra). After each iteration,
Information, Fig. S3) encapsulations, separation as a function of the relative concentrations of the (6,5) and (7,5) chiralities of
diameter was absent. semiconducting SWNTs increase (Fig. 2). After enriching the
This trend of increasing density with increasing diameter also (6,5) chirality (7.6 Å) three times, we achieved bulk solutions of
extends to SC-encapsulated SWNTs in the 11– 16 Å diameter the SWNTs in which .97% are within 0.2 Å of the mean
range that were synthesized by laser ablation (Fig. 1f ). In the diameter (see Supplementary Information, Fig. S5, and
optical spectra, separation by diameter is observed as a red shift Methods). Further improvements in the isolation of individual
in the second-order optical absorbance transitions for chiralities of SWNTs may be possible with additional cycles.
semiconducting SWNTs, 800 –1,075 nm (ref. 5), with increasing
density (Fig. 1g). Moreover, an enrichment of these SWNTs by TUNING OF THE STRUCTURE –DENSITY RELATIONSHIP
electronic type is also detected. In the most buoyant fractions, we Although the separation of SWNTs can be significantly enhanced
observe an enhancement in concentration of semiconducting via multiple cycles of ultracentrifugation, further improvements
SWNTs with respect to metallic SWNTs, which have first-order can be realized by optimizing the effectiveness of a single cycle
optical transitions ranging from 525 to 750 nm (ref. 22). through tuning of the structure– density relationship for SWNTs.

62 nature nanotechnology | VOL 1 | OCTOBER 2006 |

©2006 Nature Publishing Group

a 1.0 d e
Relative concentration (a.u.)
1.0 Semiconducting

Normalized absorbance (a.u.)

–4 0 4 8 12 16 20 0.6

b 1.0
Relative concentration (a.u.)

0.8 0.4
0.4 Semiconducting
S33 M11 S22
–4 0 4 8 12 16 20 400 600 800 1,000
Metallic Wavelength (nm)
c 1.0
Relative concentration (a.u.)

–10 0 10 20 30 40 50
∆ρ (mg cm–3)

Figure 3 Tuning the structure–density relationship for optimal separation by diameter and bandgap or electronic type (metal–semiconductor). a–c, Optimization
of separation by diameter and bandgap. The concentration of the (6,5), (7,5) and (9,5)/(8,7) chiralities of CoMoCAT-grown SWNTs (coloured red, green and blue; diameters
(ref. 5) of 7.6, 8.3 and 9.8/10.3 Å, respectively) are plotted against Dr. Concentrations were determined from absorbance spectra (Fig. 1c and Supplementary Fig. S1). The
encapsulation agents and conditions were SC, no buffer, pH 7.4 (a), SC, 20 mM Tris buffer, pH 8.5, enhanced isolation of the larger diameter SWNTs, (9,5)/(8,7) (b), SC with
the addition of SDS as a co-surfactant (1:4 ratio by weight, SDS/SC), enhanced isolation of the smaller diameter SWNTs, (6,5), pH 7.4 (c). r for the fractions with the
highest (6,5) chirality relative concentration in a–c are all 1.08 + 0.02 g cm23. Arrows mark shifts with respect to a. d,e, Optimization of separation by electronic type. d,
Photograph of laser-ablation-grown SWNTs separated in a co-surfactant solution (1:4 SDS/SC). The top band (orange) corresponds to predominantly semiconducting SWNTs
(absorbance spectra plotted in red in e) and the band just below it (green) is highly enriched in metallic SWNTs, although some semiconducting SWNTs remain (absorbance
spectra plotted in Supplementary Fig. S6). Dr between the two bands and r for the top band are 0.006 g cm23 and 1.12+ 0.02 g cm23, respectively. Further tuning of the
structure–density relationship (3:2 ratio by weight SDS/SC) results in the isolation of predominantly metallic SWNTs (absorbance spectra plotted in blue in e; heterogeneous
mixture before sorting plotted with a dashed grey line). (S33, M11, S22 highlighted as in Fig. 1g.)

For example, by adjusting pH or by adding competing co- Co-surfactant populations have an even greater effect on the
surfactants to a gradient, the isolation of a specific diameter optimization of metal –semiconductor separation for SWNTs in
range or electronic type can be targeted. the 11– 16 Å diameter regime. By adding one part SDS for every
In Fig. 3a–c we demonstrate diameter tunability. The relative four parts SC (by weight, 2% by weight overall) to a gradient,
concentration of several different diameters (7.6, 8.3 and much more distinct metal– semiconductor separation is made
9.8/10.3 Å) of SWNTs is plotted against density for the cases of evident. For SWNTs separated in 1:4 SDS/SC mixtures, only
SC-encapsulated SWNTs at pH 7.4, SC-encapsulated SWNTs at three bands are observed (Fig. 3d). We can deduce from
pH 8.5, and for a mixture of 1:4 SDS/SC (by weight) at pH 7.4. measured optical absorbance spectra that the top band (orange
By tuning the structure–density relationship, the differences in hue) consists of predominantly semiconducting SWNTs (Fig. 3d)
density among SWNTs of these diameters can be modified. and that the band just below the top band (green hue) is highly
For example, by increasing the pH to 8.5, the SWNTs near 8.3 Å enriched in metallic SWNTs (see Supplementary Information,
in diameter shift to more buoyant densities, enabling optimal Fig. S6, for a plot of the absorbance spectrum). Further tuning
separation of SWNTs in the 9.8/10.3 Å range (Fig. 3b). of the co-surfactant mixture to a 3:2 SDS/SC ratio permits
Alternatively, by adding SDS to compete with the SC for non- significantly improved isolation of metallic SWNTs. In Fig. 3e,
covalent binding to the nanotube surface, the SWNTs in the 8.3 spectra corresponding to primarily metallic (3:2 SDS/SC) and
and 9.8/10.3 Å diameter regime shift to significantly larger primarily semiconducting (1:4 SDS/SC) SWNTs are shown.
buoyant densities, enabling optimal separation of SWNTs
near 7.6 Å in diameter (Fig. 3c). At the highest densities, the FIELD-EFFECT TRANSISTORS
relative concentration of SWNTs can appear anomalously high in To demonstrate the applicability of SWNTs sorted in density
some cases (for example, the blue curve of Fig. 3a) due to gradients and to confirm their separation by electronic type, field-
contributions to the absorbance spectrum from high-density effect transistors were fabricated (see Supplementary Information,
SWNT aggregates. Methods) consisting of percolating networks of thousands of

nature nanotechnology | VOL 1 | OCTOBER 2006 | 63

©2006 Nature Publishing Group

a b characterized by on/off ratios of less than two. The two distinct
behaviours of the semiconducting and metallic films independently
confirm the separation by electronic type initially observed by
optical absorption spectroscopy of the sorted materials presented in
Fig. 3e. Additionally, the two films establish the applicability of this
method in producing usable quantities of sorted, functional
material. For example, a single fraction of semiconducting SWNTs
(150 ml) contains enough SWNTs for 20 cm2 of a thin-film
c network similar to that demonstrated in Fig. 4, corresponding to
.1011 SWNTs (see Supplementary Information, Fig. S7). Such
thin-film networks have possible applications as flexible and
transparent semiconductors and conductors.
We believe that surfactant-based separation using density-
gradient ultracentrifugation is largely driven by how surfactants
10–5 organize around SWNTs of different structures and electronic
types. The energetic balance among nanotube –surfactant, water–
surfactant and surfactant –surfactant interactions as well as the
(Sheet resistance)–1 (Ω per square)–1

10–6 packing density, orientation, ionization and the resulting

hydration of the surfactants should all be critical parameters
affecting buoyant density and the quality of sorting. Additionally,
the capacity of a surfactant to disperse SWNTs in aqueous
10–7 solution should also play a role in determining the degree of
separation. However, this characteristic alone is not a good
predictor. For example, sorting of SDBS-encapsulated SWNTs
15 was not observed here, despite the fact that Wenseleers and
10 co-workers20 have demonstrated that both SDBS and SC disperse
5 SWNTs equally well in aqueous solution.
0 Differences in the organization of anionic-alkyl surfactants and
10–9 –50 0 50 bile salts around carbon nanotubes are expected based on previous
studies of these amphiphiles in other systems. On graphene, which
–60 –40 –20 0 20 40 60 80 is the closest analogue to an SWNT, anionic-alkyl surfactants
Gate bias (V) organize into hemicylindrical micelles with liquid-like
hydrophobic cores19,23, whereas bile salts form well-structured
monolayers with their less-polar sides facing the hydrophobic
Figure 4 Electrical devices of semiconducting and metallic SWNTs. surface24. Bile salts also order to form well-defined host–guest
a, Periodic array of source and drain electrodes (single device highlighted in red, structures around small hydrophobic molecules21,25. Accordingly,
scale bar 40 mm, gap 20 mm). b, Representative AFM image of thin-film, the rigidity and planarity of bile salts, in contrast with anionic-
percolating SWNT network (scale bar ¼ 1 mm). The density of SWNTs per unit alkyl surfactants, are expected to result in encapsulation layers that
area is .10 times the percolation limit (Supplementary Fig. S7). c, Field-effect are sensitive to subtle changes in the underlying SWNT.
transistor geometry (s ¼ source; g ¼ gate; d ¼ drain). The SWNT networks were Furthermore, the observed metal– semiconductor selectivity
formed on a 100-nm, thermally grown SiO2 layer, which served as the gate indicates a coupling of the surfactant and/or its hydration with the
dielectric. d, Inverse of sheet resistance as a function of gate bias for electronic nature of the underlying SWNT. Lu and co-workers have
semiconducting (red, triangles) and metallic (blue, squares) SWNTs sorted in suggested that metallic SWNTs interact more strongly with
co-surfactant density gradients (characterized in Fig. 3e). The metallic SWNTs adsorbates via p interactions than semiconducting SWNTs, due to
did not significantly switch with gate bias (,2), in contrast with the their larger electronic polarizability26. Additionally, the packing
semiconducting SWNTs, which switched by a factor of .2  104. (Error bars density of the surfactants and their hydration are likely to be
described in Supplementary Information, Methods.) The inset shows a sensitive to electrostatic screening by the underlying SWNT. Other
semiconducting device plotted on a linear scale (red curve, same units). A lower effects, such as partial charge transfer between metallic SWNTs
bound for mobility in the semiconducting SWNTs is estimated (from the grey fit) and surfactants induced by CH– p interactions could also
to be 20 cm2 V21 s21 (see Supplementary Information, Methods), comparable to be important.
previously reported mobilities for thin films of as-synthesized mixtures of Density-gradient ultracentrifugation provides a scalable
metallic and semiconducting SWNTs near their percolation threshold29. approach for sorting carbon nanotubes by diameter, bandgap and
electronic type. This strategy has been demonstrated for SWNTs
encapsulated by bile salts and mixtures of bile salts with anionic-
alkyl surfactants for SWNTs between 7 and 16 Å in diameter. By
metallic or semiconducting SWNTs (Fig. 4). At negative gate successive iterations of ultracentrifugation, sharp diameter
biases, both networks exhibited similar sheet resistances of about distributions have been achieved in which more than 97% of
500 kV per square. However, by varying the voltage applied across semiconducting SWNTs are within 0.2 Å of the mean diameter.
the gate dielectric capacitor (100 nm SiO2), the resistivity of the Furthermore, the structure –density relationship for SWNTs has
semiconducting network was increased by over four orders of been engineered to achieve exceptional metal– semiconductor
magnitude (on/off ratio .20,000). In contrast, the metallic separation by using mixtures of competing co-surfactants, thus
networks were significantly less sensitive to the applied gate bias enabling the isolation of bulk quantities of SWNTs that are

64 nature nanotechnology | VOL 1 | OCTOBER 2006 |

©2006 Nature Publishing Group

predominantly a single electronic type (typical yields are shown in 6. Graham, J. M. Biological Centrifugation (BIOS Scientific, Milton Park, 2001).
7. Anderson, N. G. et al. K-series centrifuges. I. Development of K-2 continuous-sample-flow-with-
the Supplementary Information, Fig. S8, and Methods). Because banding centrifuge system for vaccine purification. Anal. Biochem. 32, 460– 494 (1969).
SWNTs sorted by this method are highly compatible with 8. Arnold, M. S., Stupp, S. I. & Hersam, M. C. Enrichment of single-walled carbon nanotubes by
diameter in density gradients. Nano Lett. 5, 713 –718 (2005).
subsequent processing techniques and can be integrated into 9. O’Connell, M. J. et al. Band gap fluorescence from individual single-walled carbon nanotubes.
devices, it is expected that density-gradient ultracentrifugation will Science 297, 593 –596 (2002).
10. Chen, Z. H. et al. Bulk separative enrichment in metallic or semiconducting single-walled carbon
directly impact the large number of technological applications that nanotubes. Nano Lett. 3, 1245–1249 (2003).
require SWNTs with monodisperse structure and properties. 11. Haddon, R. C., Sippel, J., Rinzler, A. G. & Papadimitrakopoulos, F. Purification and separation of
carbon nanotubes. Mater. Res. Soc. Bull. 29, 252– 259 (2004).
12. Krupke, R., Hennrich, F., Kappes, M. M. & Lohneysen, H. V. Surface conductance induced
METHODS dielectrophoresis of semiconducting single-walled carbon nanotubes. Nano Lett. 4,
1395 –1399 (2004).
13. Maeda, Y. et al. Large-scale separation of metallic and semiconducting single-walled carbon
nanotubes. J. Am. Chem. Soc. 127, 10287–10290 (2005).
To disperse SWNTs in solutions of bile salts or other surfactants, 1 mg ml21 14. Collins, P. G., Arnold, M. S. & Avouris, P. Engineering carbon nanotubes and nanotube circuits
SWNTs were dispersed in solutions of 2% weight per volume (w/v) surfactant using electrical breakdown. Science 292, 706 –709 (2001).
via ultrasonication (see Supplementary Information, Methods, for more 15. Samsonidze, G. G. et al. Quantitative evaluation of the octadecylamine-assisted bulk separation of
semiconducting and metallic single-wall carbon nanotubes by resonance Raman spectroscopy.
details). In co-surfactant experiments, SWNTs were initially dispersed in a Appl. Phys. Lett. 85, 1006 –1008 (2004).
2% w/v SC surfactant solution and then diluted into 2% w/v co-surfactant 16. Strano, M. S. et al. Electronic structure control of single-walled carbon nanotube
solution. For example, in a 1:4 SDS/SC co-surfactant experiment, one part functionalization. Science 301, 1519–1522 (2003).
17. Zheng, M. et al. Structure-based carbon nanotube sorting by sequence-dependent DNA assembly.
2% w/v co-surfactant was added to four parts of a 2% w/v SC solution Science 302, 1545–1548 (2003).
containing dispersed SWNTs. Furthermore, density gradients were loaded 18. Peng, H. Q., Alvarez, N. T., Kittrell, C., Hauge, R. H. & Schmidt, H. K. Dielectrophoresis field
homogeneously from top to bottom with these surfactants at the same ratios. flow fractionation of single-walled carbon nanotubes. J. Am. Chem. Soc. 128, 8396– 8397 (2006).
19. Islam, M. F., Rojas, E., Bergey, D. M., Johnson, A. T. & Yodh, A. G. High weight fraction
Thus, no gradients in surfactant concentrations were present. surfactant solubilization of single-wall carbon nanotubes in water. Nano Lett. 3, 269 –273 (2003).
20. Wenseleers, W. et al. Efficient isolation and solubilization of pristine single-walled nanotubes in
bile salt micelles. Adv. Funct. Mater. 14, 1105 –1112 (2004).
Density gradients were formed from aqueous solutions of a non-ionic density 21. Mukhopadhyay, S. & Maitra, U. Chemistry and biology of bile acids. Curr. Sci. 87,
gradient medium, iodixanol6,8,27,28, purchased as OptiPrep 60% w/v iodixanol, 1666 –1683 (2004).
22. Strano, M. S. Probing chiral selective reactions using a revised Kataura plot for the interpretation
1.32 g cm23 (Sigma-Aldrich). Gradients were created directly in centrifuge
of single-walled carbon nanotube spectroscopy. J. Am. Chem. Soc. 125, 16148–16153 (2003).
tubes by one of two methods: by layering of discrete steps and subsequent 23. Wanless, E. J. & Ducker, W. A. Organization of sodium dodecyl sulfate at the graphite–solution
diffusion into linear gradients or by using a linear gradient maker (see interface. J. Phys. Chem. 100, 3207– 3214 (1996).
Supplementary Information, Figs. S9 –S11, and Methods). 24. Sasaki, Y. et al. The adsorption behavior of four bile salt species on graphite in water—evaluation
of effective hydrophobicity of bile acids. Colloids Surf. B 5, 241– 247 (1995).
CENTRIFUGATION AND FRACTIONATION 25. Tamminen, J. & Kolehmainen, E. Bile acids as building blocks of supramolecular hosts. Molecules
6, 21 – 46 (2001).
Centrifugation was carried out in two different rotors, a fixed angle TLA100.3 26. Lu, J. et al. Selective interaction of large or charge-transfer aromatic molecules with metallic
rotor and a swing bucket SW41 rotor (Beckman-Coulter), at 22 8C and at single-wall carbon nanotubes: Critical role of the molecular size and orientation. J. Am. Chem.
64,000 r.p.m. and 41,000 r.p.m., respectively, for 9–24 h, depending on the Soc. 128, 5114–5118 (2006).
27. Ford, T., Graham, J. & Rickwood, D. Iodixanol—a nonionic isosmotic centrifugation medium for
spatial extent and initial slope of the gradient. At the average radii (37.9 mm the formation of self-generated gradients. Anal. Biochem. 220, 360 –366 (1994).
and 110 mm, respectively), these rotational velocities result in centripetal 28. Davies, I. & Graham, J. M. The use of self-generated gradients of iodixanol for the purification of
accelerations of 174,000 g and 207,000 g, respectively. To fractionate TLA100.3 macromolecules and macromolecular complexes. FASEB J. 11, A908 –A908 (1997).
29. Snow, E. S., Novak, J. P., Campbell, P. M. & Park, D. Random networks of carbon nanotubes as
tubes, a modified Beckman Fractionation System (Beckman-Coulter) was an electronic material. Appl. Phys. Lett. 82, 2145–2147 (2003).
utilized in an upward displacement mode using Fluorinert FC-40 (Sigma-
Aldrich) as a dense chase media. 25 ml fractions were collected. To fractionate
SW41 centrifuge tubes, a Piston Gradient Fractionator system was utilized This work was supported by the US Army Telemedicine and Advanced Technology Research Center,
(Biocomp Instruments, Canada). Fractions of 0.5–3.0 mm were collected the National Science Foundation and the Department of Energy. A National Science Foundation
(70 –420 ml in volume). In both cases, fractions were diluted to 1 ml in Graduate Student Fellowship (M.S.A.), a Natural Sciences and Engineering Research Council of
Canada Postgraduate Scholarship (A.A.G.), and an Alfred P. Sloan Research Fellowship (M.C.H.) are
2% w/v surfactant solution for optical characterization. also acknowledged. Furthermore, J. Suntivich, X. Du and M. Disabb are gratefully recognized for
measurement of optical absorbance spectra (J.S., X.D.) and evaporation of Au electrodes (M.D.). We
thank J. Widom and the Keck Biophysics Facility for use of their ultracentrifuges, J. Chen for
Received 24 July 2006; accepted 25 August 2006; published 4 October 2006. providing laser-ablation-grown SWNTs, and L. Palmer and Ph. Avouris for useful discussions.
Correspondence and requests for materials should be addressed to M.C.H.
Supplementary Information accompanies this paper on
1. Baughman, R. H., Zakhidov, A. A. & de Heer, W. A. Carbon nanotubes—the route toward
applications. Science 297, 787 –792 (2002). Author contributions
2. Javey, A., Guo, J., Wang, Q., Lundstrom, M. & Dai, H. J. Ballistic carbon nanotube field-effect All authors conceived and designed the experiments; M.S.A. and A.A.G. performed the experiments;
transistors. Nature 424, 654– 657 (2003). J.F.H. measured and analysed the optical spectra; and all authors co-wrote the paper.
3. Wu, Z. C. et al. Transparent, conductive carbon nanotube films. Science 305, 1273– 1276 (2004).
4. Charlier, J. C. & Issi, J. P. Electronic structure and quantum transport in carbon nanotubes. Appl.
Phys. A 67, 79 – 87 (1998). Competing financial interests
5. Weisman, R. B. & Bachilo, S. M. Dependence of optical transition energies on structure for The authors declare that they have no competing financial interests.
single-walled carbon nanotubes in aqueous suspension: An empirical Kataura plot. Nano Lett. 3,
1235– 1238 (2003). Reprints and permission information is available online at

nature nanotechnology | VOL 1 | OCTOBER 2006 | 65

©2006 Nature Publishing Group