Rapid and Quantitative Differentiation of Single-Walled and Double-Walled Carbon Nanotubes Using Analytical Ultracentrifugation

Over the last 15 years, single-walled carbon nanotubes (SWCNTs) have attracted increasing interest for use in fuel cells, semiconductors and biomedical applications. However, all of these fields face two major barriers.

Firstly, numerous carbonaceous impurities are formed as a result of SWCNT synthesis. This synthesis also results in the possible presence of multi-walled carbon nanotubes (MWCNTs). Furthermore, there is no reliable way of quantifying the percentage of non-SWCNT material formed.

Secondly, a heterogeneous variety of wrapping schemes, known as (n,m) chiralities, are used for the SWCNT synthesis (Figure 1) and different chiralities exhibit very different electronic and optical properties.

After synthesis, there needs to be large-scale separation of SWCNTs into a single, homogeneous chirality, particularly for imaging applications, in vivo drug delivery and semiconductors, where metallic SWCNTs significantly lower the on/off ratio.

Single-Walled Carbon Nanotube Schematic

Figure 1. Single-Walled Carbon Nanotube Schematic.

Efforts to determine SWCNT impurity in solution were initially focused on electron microscopy, but this lacks statistical significance, is subject to human bias and has difficulty in distinguishing between SWCNTs with a 1nm diameter and double-walled carbon nanotubes (DWCNTs) with 3nm diameter.

Researchers have also attempted to use Raman spectroscopy, near-infrared (NIR) fluorescence, and UV-Vis as solutions to the impurity problem, but all of these techniques have drawbacks, which make them unsuitable for quantitative analysis of SWCNT impurities.

Density gradient ultracentrifugation (DGU) has been one of the most successful solutions to date for chirality-enrichment of a single SWCNT species. DGU can achieve a highly desirable 99% purity of (6, 5) SWCNT.

However, scalability is a significant issue due to the handmade gradient typically used. This article discusses a workflow that rapidly and reproducibly separates heterogeneous, bulk SWCNT into a single (6,5) chirality, confirmed by UV-Vis. This workflow involves:

  • A fast, two-minute ultracentrifuge treatment using Beckman Coulter’s Optima MAX-XP for initial purification of SWCNT by eliminating large aggregated species
  • A density gradient setup using Beckman Coulter’s Biomek 4000 automated liquid handler
  • Use of the Optima X Series Preparative Ultracentrifuge to purify SWCNT and DWCNT using density gradient runs

Automation enables improved accuracy and reproducibly with the density gradient than can be achieved by hand. This level of accuracy is particularly relevant for the separation of SWCNT, where density steps only differ by approximately ±1%g/ml.

This article also emphasizes how analytical ultracentrifugation (AUC) is able to quantitatively differentiate between SWCNTs and length-fractionated DWCNTs. Traditionally, AUC has mainly been employed in protein analysis, but the analytical capabilities of AUC are also suitable for nanoparticle characterization.

AUC can fill in the gaps in nanomaterials analysis, where techniques such as optical spectroscopy and electron microscopy are lacking, by determining sedimentation coefficient, diffusion coefficient and frictional ratio of nanoparticles in various density solvents.

Experimental Procedures

Gradient Prep

The Biomek 4000 Workstation equipped with a P-1000SL Single-Tip Pipette Tool and P1000 wide bore tips was used to make the density gradient (Figure 2). The method has the flexibility to change volumes for each gradient and the number of tubes being prepared.

Centrifuge tubes were held in a 14mm 24-position tube rack and programmed in as new labware. Mixing of the gradients was minimized by using a slow pipetting technique with liquid level sensing (Figure 3). The gradient achieved was an overlaid gradient (Table 1).

Deck Layout of the Biomek 4000 Workstation Showing the Basic Tools Required for Gradient Prep

Figure 2. Deck Layout of the Biomek 4000 Workstation Showing the Basic Tools Required for Gradient Prep. (1) One 24-position tube rack for placing nanotubes: the centrifuge tubes fit the existing 24-position tube rack, but new labware type had to be created to accommodate the height of the tubes; (2) one P1000 tip box for P1000 Wide Bore tips; (3) one Biomek 4000 P1000SL Single-Tip Pipette Tool for liquid transfer; (4) one Modular Reservoir for gradient reagents.

Table 1. Density Gradient Architecture Single-Chirality Separation

Layer in Gradient

Density (g/mL)

%OP

1st Iteration Volume (μL)

SC

(% w/v)

SDS

(% w/v)

1

1.160

30

900

0.75

0.175

2

1.147

27.5

756

0.75

0.175

3

1.133

25

972

0.75

0.175

4

1.120

22.5

1,188

0.75

0.175

5

1.107

20

1,188

0.75

0.175

6

1.093

17.5

1,305

0.75

0.175

SWCNT

1.133

25

1,800

2

0

Nanotube Gradient Method. New tube transfer technique was created to minimize the mixing during gradient prep.

Figure 3. Nanotube Gradient Method. New tube transfer technique was created to minimize the mixing during gradient prep.

SWCNT Preparation

A Branson M1800H ultrasonic cleaner was used to bath-sonicate 20mgs of SWCNTs in 2% sodium cholate (SC) in deionized water in a 20mL glass vial for 1h. The next step was centrifuging the SWCNT solution with a TLA 120.2 rotor in an Optima MAX-XP ultracentrifuge in open-top, thick-wall polycarbonate centrifuge tubes at 55,000rpm (~131,000 x g) at 22°C for 2 min to crash out any large aggregates.

Then, 1100μL of the supernatant was carefully collected, to avoid disruption of the pelleted aggregates and used for density gradient run. 1.8mL of SWCNT solution balanced to a density of 1.13g/mL (25%OP) using 2% SC+OP was inserted between 27.5% and 25% Optiprep layers (Sigma-Aldrich) in pre-filled gradient tubes.

Using DI water with the same surfactant ratio as Optiprep, the centrifuge tubes were balanced and filled within 2 to 3mm of the top. This was followed by spinning the tubes using an SW 41 Ti rotor in an Optima XPN at 41,000rpm (~288,000 x g) at 22°C for 32h and using minimum acceleration and deceleration rates (Profile 9).

Once the centrifugation was complete, a large syringe was used to remove the top 2mL, with care being taken not to disturb the bands below. The final step was aliquoting the region with (6,5) SWCNT in 150μl fractions (Figure 5).

DWCNT Dispersion

A Branson M1800H ultrasonic cleaner was used to bath-sonicate 20mg of DWCNTs in 2% SC in deionized water in a 20mL glass vial for 1h. A TLA 120.2 rotor in an Optima MAX-XP ultracentrifuge was used to centrifuge the DWCNT solution in open-top polycarbonate centrifuge tubes at 30,000 x g, at 22°C for 2min to crash out any large aggregates.

1300μL of the supernatant was carefully collected, without disrupting the pelleted aggregates, and used for density gradient run. Polyallomer centrifuge tubes were used to manually prepare a density gradient, as presented in Table 2, to length-fractionate the DWCNT.

This was followed by overlaying 1.5mL of the DWCNT solution on the gradient. The centrifuge tubes were balanced and filled within 2 to 3mm of the top, using DI water with 2% SC.

An SW 41 Ti rotor in an Optima XPN was used to spin the tubes using a two-step program: first run at 15,000rpm (~38,500 x g) for 1h and second run at 30,500rpm (~159,500 x g) for 1h at 22°C with maximum acceleration and deceleration rates (Profile 0). Once the centrifugation was complete, the next step was collecting 600μL fractions from top to bottom and pooling fractions 4–6 (Figure 6).

Table 2. Density Gradient Architecture Double-Wall Separation

Layer in Gradient

Density (g/mL)

%OP

1st Iteration Volume (μL)

SC

(% w/v)

1

1.320

60

1,500

2

2

1.160

30

1,500

2

3

1.133

25

1,500

2

4

1.107

20

1,500

2

5

1.08

15

1,500

2

6

1.053

10

1,500

2

DWCNT

1

0

1,500

2

Fraction Analysis

UV-Vis-NIR absorption plots (Paradigm, Molecular Devices) were generated for each SWCNT fraction from 400–1,000nm and the fractions with the strongest absorption peaks at wavelengths of 570 and 990nm and with minimal absorption peaks at other wavelengths, were pooled.

Dialysis

Once fractionation was complete, the next step was dialyzing the separated, individual (6,5) chirality SWCNT with a 3.5kDa MWCO cellulose membrane against 1% SC in order to eliminate sodium dodecyl sulfate and iodixanol from the SWCNT solution and to re-establish the surfactant coating of the SWCNT (Figure 4).

The water was changed eight times, with at least 4h between changes. Following dialysis, the Amicon Ultra Centrifuge Filters (Millipore) were used on Beckman Coulter’s Microfuge 20 microcentrifuge to concentrate the resulting dispersion. The same steps were used for the pooled, length-fractionated DWCNT solution.

Test Gradient Preparation using Iodixanol with food coloring to show the distinct layering of each gradient, and comparison of manual versus Biomek 4000 workstation.

Figure 4. Test Gradient Preparation using Iodixanol with food coloring to show the distinct layering of each gradient, and comparison of manual versus Biomek 4000 workstation.

Sedimentation Range Determination

Beckman Coulter’s ProteomeLab XL-A Analytical Ultracentrifuge was used to run the SWCNT and DWCNT solutions. The next step was loading (6,5) chirality-enriched SWCNT with an absorption of O.D. 0.85 at 570nm into a two-sector 12mm charcoal-filled EPON cell with quartz windows.

The reference used was 1% SC in DI water, collected from the last dialysis water change. The sample and reference volumes were 370μL and 380μL, respectively. The length-fractionated DWCNT with an O.D. of 0.85 at 570nm was loaded into another cell with the same specifications.

50%/50% SWCNT/DWCNT dispersion was loaded into the third cell by absorption at 570nm. Initial run conditions were 27,000 rpm at 22°C for 4h. This experiment was performed again for absorption conditions with O.D. of 0.6 at 570nm to check for concentration-dependent effects.

Sedimentation Analysis

The analysis was performed in SEDFIT, fitting to models according to the Lamm equation. The best fit for the data is given from a fit of a single-component Lamm equation considering diffusion, according to the study by Arnold et al.

The next step was comparing the sedimentation coefficients for SWCNTs and DWCNTs and testing the capability of analytical ultracentrifugation to differentiate between both species in solution.

Size Distribution Analysis via Dynamic Light Scattering

A small volume of sample of the chirality-enriched (6,5) SWCNT and the length-fractionated DWCNT was analyzed using Beckman Coulter’s DelsaMax CORE Dynamic Light Scattering/Static Light Scattering instrument.

Approximately 10μL was put in the quartz sizing cuvette and was run at 25°C with 10 acquisitions, at an acquisition time of 5s. The runs were analyzed in Regularization (Multimodal) mode to generate the representative curves.

Results

Figure 5 illustrates the success of the DGU SWCNT separation. The SWCNT appeared as a black solution before ultracentrifugation (Figure 5a) due to the heterogeneous mixture of chiralities, which have absorption peaks across the visible range. Individual chiralities appeared as colored bands (Figure 5b) after ultracentrifugation.

The top purple band consisted of (6,5) SWCNTs, which were collected for subsequent AUC analysis. The Van Hoff singularity absorption peaks were observed for SWCNTs in the NIR and visible region. Theoretically, the peaks occur at 570–580nm and 980–990nm for (6,5) SWCNT coated with SC.

Figure 7 shows the absorption plot taken following dialysis and concentration of the (6,5) SWCNT in 1% SC solution. The purity of the chirality-enriched (6,5) SWCNT is indicated by the strong peaks observed at 571 and 990nm wavelengths, along with the lack of strong absorption peaks at other wavelengths.

A similar procedure for SWCNT is followed to length-fractionate DWCNT. The top-most fraction (Figure 6b) should mainly contain unbundled DWCNT. Dynamic light scattering data confirmed that the average length of the DWCNTs is near 200nm based on a diffusion coefficient of 2.1 * 10-8cm2/s.

SWCNT Separation Based on Chirality

Figure 5. (6,5) SWCNT Separation Based on Chirality. Pictures of centrifuge tube with SWCNT before (5a) and after (5b) Density Gradient Ultracentrifugation. 0.2mL fractions were collected from the region with the purple tint (indicated with the arrow) and were subjected to absorption analysis and pooled based on absorbance peak at 575nm.

DWCNT Length Separation

Figure 6. DWCNT Length Separation. Optical image of centrifuge tube with DWCNT (6a) before Density Gradient Ultracentrifugation and (6b) after Density Gradient Ultracentrifugation. 0.6mL fractions were aliquoted and fractions 4–6 were collected for further analysis. Approximate location of fractions 4–6 is indicated with a brace (}).

Dynamic Light Scattering data obtained using the DelsaMax CORE also highlights the challenges involved in differentiating between SWCNT and DWCNT (Figure 8). The optical properties of SWCNT and DWCNT are very different, including fluorescence and absorption (Figure 7), but the physical diameters and density are very similar.

The both have lengths in the range of 100–1,000nm and have closely related diameters, of approximately 1nm for SWCNT and approximately 2–3.5nm for DWCNT. This presents a challenge in terms of ensemble methods such as light scattering being able to differentiate between SWCNT and DWCNT.

Similarly, electron microscopy has difficulty having height sensitivity small enough to reliably distinguish between SWCNT and DWCNT. Also, counting of a few hundred nanotubes is not representative of a solution containing upwards of 1018 particles.

Absorption Plot of Concentrated Length-Separated Double-Walled Carbon Nanotubes

Figure 7. Absorption Plot of Concentrated Length-Separated Double-Walled Carbon Nanotubes (DWCNT, Red Curve) and Chirality-Enriched (6,5) Single-Walled Carbon Nanotubes (SWCNT, Black Curve). Inset are images of the AUC cells with reference buffer in the left chamber and sample solution in the right chamber. (a) contains DWCNT only; (b) contains primarily (6,5) SWCNT, indicated by the strong peak at 570 and 980nm.

Representative Dynamic Light Scattering Data on the DelsaMax CORE

Figure 8. Representative Dynamic Light Scattering Data on the DelsaMax CORE. The carbon nanotube species generate the peaks above 100 nm in diameter while the surfactant micelles are represented by the peaks near 10nm in diameter. Note that it would be impossible to distinguish between SWCNT and DWCNT based on dynamic light scattering.

AUC can easily differentiate between SWCNT and DWCNT (Figure 9c). Sedimentation of DWCNT is very rapid, with an average sedimentation coefficient of 80.4 ± 25.6 S (Figures 9b, 10b) during ultracentrifugation in SC buffer, because of the poor surfactant coating of the sodium cholate molecules.

By contrast, sedimentation of SWCNT is slow, with an average sedimentation coefficient of 11.3 S (Figures 9a, 10a). The power of AUC was further demonstrated by analyzing a mixture of SWCNT and DWCNT.

Both SWCNT and DWCNT solutions had absorptions of 0.894 O.D. at 570nm. The next step was mixing and running 175μL of each solution in the AUC (Figure 9c).

Observing pure (6,5) SWCNT showed that very few particles have sedimentation coefficients larger than 30 S (Figure 10a), whereas the length-fractionated DWCNT contains very few particles with sedimentation coefficients below 30 S.

The solution is quantitatively shown to have 50.4% SWCNT and 49.6% DWCNT by absorption, by applying 30 S as a cutoff point and integrating the sedimentation coefficient distribution plots (Figure 10c).

This quantitative measurement of SWCNT and DWCNT is not possible by any other analytical method. The quantitative power of AUC was confirmed by testing two other SWCNT/DWCNT mixtures.

A 29% SWCNT/71% DWCNT by absorption was found to have a ratio of 28.3%/71.7% by sedimentation coefficient distribution (Figure 10d). Likewise, a 71.4% SWCNT/28.6% DWCNT solution by absorption was found to have a ratio of 64.7%/35.3% by sedimentation coefficient distribution. The cutoff point between the (6,5) SWCNT and DWCNT used in both tests was 30 S.

AUC curves from SEDFIT.

Figure 9. AUC curves from SEDFIT. (9a) The raw absorbance data with fitting of a solution containing only (6,5) SWCNT. (9b) The raw absorbance data with fitting of a solution containing only length-fractionated DWCNT (9c) The raw absorbance data with fitting of a solution containing both (6,5) SWCNT and DWCNT.

Sedimentation Coefficient Distribution Plots

Figure 10. Sedimentation Coefficient Distribution Plots. (10a) Chirality-enriched (6,5) SWCNT. The average sedimentation coefficient for chirality-enriched (6,5) SWCNT is 11.3 S, which agrees well with literature. Note that essentially all particles in the (6,5) SWCNT solution have sedimentation coefficient less than 30 S. (10b) Length-Fractionated DWCNT. The average sedimentation coefficient for length-fractionated DWCNT is 80.4 ± 25.6 S; the large spread indicates that some DWCNT may exist as a bundled pair. Note that nearly all sedimenting particles in the DWCNT sample have sedimentation coefficient above 30 S. (10c) DWCNT and (6,5) SWCNT 50/50 mixture. DWCNT and (6,5) SWCNT solutions each had identical absorption at 570 nm and were mixed in equivolume amounts. Integrating the sedimentation distribution, 50.4% of the total signal has sedimentation coefficient between 5–30 S, with an average value of 11.2 ± 5.2 S, while 49.6% of the total signal has sedimentation coefficient between 30–140, with an average value of 70.2 ± 21.3 S. (10d) DWCNT and (6,5) SWCNT 71/29 mixture. DWCNT and (6,5) SWCNT solutions each had identical absorption at 570 nm and were mixed at a ratio of 71:29 DWCNT:(6,5) SWCNT. Integrating the sedimentation distribution, 28.3% of the total signal has sedimentation coefficient between 2–30 S, with an average value of 12.5 ± 3.9 S, while 71.7% of the total signal has sedimentation coefficient between 30–120, with an average value of 80.0 ± 21.0 S.

References

  1. Baughman R H, Zakhidov A A, and de Heer W A. Carbon nanotubes—the route toward applications. Science. 297.5582; 787–792: (2002).
  2. Che G et al. Carbon nanotubule membranes for electrochemical energy storage and production. Nature. 393.6683; 346–349: (1998).
  3. Liu Z et al. Carbon nanotubes in biology and medicine: in vitro and in vivo detection, imaging and drug delivery. Nano research. 2.2; 85–120: (2009).
  4. Lopez-Lorente A I, Simonet B M and Valcárcel M. Qualitative detection and quantitative determination of single-walled carbon nanotubes in mixtures of carbon nanotubes with a portable Raman spectrometer. Analyst. 138.8; 2378–2385: (2013).
  5. Saito R. Physical properties of carbon nanotubes: (1998).
  6. O’Connell M J et al. Band gap fluorescence from individual single-walled carbon nanotubes. Science. 297.5581; 593–596: (2002).
  7. Liu H et al. Large-scale single-chirality separation of single-wall carbon nanotubes by simple gel chromatography. Nature Communications. 2; 309: (2011).
  8. Zhang L et al. Assessment of chemically separated carbon nanotubes for nanoelectronics. Journal of the American Chemical Society. 130.8; 2686–2691: (2008).
  9. Welsher K et al. A route to brightly fluorescent carbon nanotubes for near-infrared imaging in mice. Nature nanotechnology. 4.11; 773–780: (2009).
  10. Liu Z et al. Supramolecular stacking of doxorubicin on carbon nanotubes for in vivo cancer therapy. Angewandte Chemie International Edition. 48.41; 7668–7672: (2009).
  11. Bhirde A A et al. Targeted killing of cancer cells in vivo and in vitro with EGF- directed carbon nanotube-based drug delivery. ACS nano. 3.2; 307–316: (2009).
  12. Green A A, Duch M C and Hersam M C. Isolation of single-walled carbon nanotube enantiomers by density differentiation. Nano Research. 2.1; 69–77: (2009).
  13. Komatsu N and Wang F. A comprehensive review on separation methods and techniques for single-walled carbon nanotubes. Materials. 3.7; 3818–3844: (2010).
  14. Carney R P et al. Determination of nanoparticle size distribution together with  density or molecular weight by 2D analytical ultracentrifugation. Nature  Communications. 2; 335: (2011).
  15. Falabella J B et al. Characterization of gold nanoparticles modified with single-stranded DNA using analytical ultracentrifugation and dynamic light scattering. Langmuir. 26.15; 12740–12747: (2010).
  16. Antaris A L et al. Ultra-Low Doses of Chirality Sorted (6, 5) Carbon Nanotubes for Simultaneous Tumor Imaging and Photothermal Therapy. ACS nano. 7.4; 3644–3652: (2013).
  17. Arnold M S et al. Hydrodynamic characterization of surfactant encapsulated carbon nanotubes using an analytical ultracentrifuge. ACS nano. 2.11; 2291–2300: (2008).
  18. Tabakman S M et al. Optical properties of single-walled carbon nanotubes separated in a density gradient: length, bundling, and aromatic stacking effects. The Journal of Physical Chemistry C. 114.46; 19569–19575: (2010).
  19. Hertel T et al. Spectroscopy of single-and double-wall carbon nanotubes in different environments. Nano letters. 5.3; 511–514: (2005).
  20. Fagan J A et al. Analyzing Surfactant Structures on Length and Chirality Resolved (6, 5) Single-Wall Carbon Nanotubes by Analytical Ultracentrifugation.  ACS nano. 7.4; 3373–3387: (2013).

About Beckman Coulter

Beckman Coulter develops, manufactures and markets products that simplify, automate and innovate complex biomedical tests. More than a quarter of a million Beckman Coulter instruments operate in laboratories around the world, supplying critical information for improving patient health and reducing the cost of care.


Sponsored Content Policy: News-Medical.net publishes articles and related content that may be derived from sources where we have existing commercial relationships, provided such content adds value to the core editorial ethos of News-Medical.Net which is to educate and inform site visitors interested in medical research, science, medical devices and treatments.

Last updated: Mar 1, 2019 at 5:24 AM

Citations

Please use one of the following formats to cite this article in your essay, paper or report:

  • APA

    Beckman Coulter Life Sciences - Centrifugation. (2019, March 01). Rapid and Quantitative Differentiation of Single-Walled and Double-Walled Carbon Nanotubes Using Analytical Ultracentrifugation. News-Medical. Retrieved on January 29, 2020 from https://www.news-medical.net/whitepaper/20160928/Rapid-and-Quantitative-Differentiation-of-Single-Walled-and-Double-Walled-Carbon-Nanotubes-using-Analytical-Ultracentrifugation.aspx.

  • MLA

    Beckman Coulter Life Sciences - Centrifugation. "Rapid and Quantitative Differentiation of Single-Walled and Double-Walled Carbon Nanotubes Using Analytical Ultracentrifugation". News-Medical. 29 January 2020. <https://www.news-medical.net/whitepaper/20160928/Rapid-and-Quantitative-Differentiation-of-Single-Walled-and-Double-Walled-Carbon-Nanotubes-using-Analytical-Ultracentrifugation.aspx>.

  • Chicago

    Beckman Coulter Life Sciences - Centrifugation. "Rapid and Quantitative Differentiation of Single-Walled and Double-Walled Carbon Nanotubes Using Analytical Ultracentrifugation". News-Medical. https://www.news-medical.net/whitepaper/20160928/Rapid-and-Quantitative-Differentiation-of-Single-Walled-and-Double-Walled-Carbon-Nanotubes-using-Analytical-Ultracentrifugation.aspx. (accessed January 29, 2020).

  • Harvard

    Beckman Coulter Life Sciences - Centrifugation. 2019. Rapid and Quantitative Differentiation of Single-Walled and Double-Walled Carbon Nanotubes Using Analytical Ultracentrifugation. News-Medical, viewed 29 January 2020, https://www.news-medical.net/whitepaper/20160928/Rapid-and-Quantitative-Differentiation-of-Single-Walled-and-Double-Walled-Carbon-Nanotubes-using-Analytical-Ultracentrifugation.aspx.

Other White Papers by this Supplier