Chirality-dependent densities of carbon nanotubes by in situ 2D fluorescence-excitation and Raman characterisation in a density gradient after ultracentrifugation

Abstract

Density gradient ultracentrifugation (DGU) becomes increasingly important for the sorting of nanomaterials according to the particles' density, hence structure and dimensions, which determine their unique properties, but the further development of this separation technique is hindered by the limited precision with which the densities could be characterized. In this work, we determine these densities by position-dependent 2D wavelength-dependent IR fluorescence-excitation and resonant Raman spectroscopy measured directly in the density gradient after ultracentrifugation. We apply this method to study the diameter and chirality-dependent sorting of empty and water-filled single-walled carbon nanotubes coated with two different surfactants, sodium cholate (SC) and sodium deoxycholate (DOC). The results elucidate the long standing contradiction that SC would provide better diameter sorting, while DOC is the most efficient surfactant to solubilise the nanotubes. A more predictable separation is obtained for empty DOC-coated nanotubes since their density is found to vary very smoothly with diameter. The accurate and chirality-dependent densities furthermore provide information on the surfactant coating, which is also important for other separation techniques, and allow to determine the mass percentage of water encapsulated inside the nanotubes.

Ultrasonication-induced extraction of inner shells from double-wall carbon nanotubes characterized via in situ spectroscopy after density gradient ultracentrifugation

“…In this work, we guarantee a high-purity DWCNT parent solution through a multi-step DGU purification process in combination with in situ RRS and PL spectroscopy, measuring as a function of depth in the ultracentrifuge tube after DGU [37]. The resulting purified DWCNT sample shows, in line with Levshov et al [24], a negligible fluorescence from small-diameter SWCNTs, while on the contrary an unexpectedly strong DWCNT outer shell PL signal is detected.…”

“…To purify the solubilized DWCNTs from any SWCNTs remaining from the synthesis [31], the DWCNTs were ultracentrifuged in a gradient medium and subsequently probed using RRS and PL as a function of depth directly in the centrifuge tube, analogously to the procedure presented in reference [37]. The density gradient medium used is iohexol (5-(N-2,3dihydroxypropylacetamido)-2,4,6-triiodo-N,N'-bis(2,3-dihydroxypropyl)isophthalamide, tradename "NycoDenz"), obtained from Axis-Shield in powder form, and was dissolved in D2O at appropriate concentrations to obtain the required density ranges.…”

“…In total four ultracentrifuge tubes were centrifuged (Optima Max-XP tabletop ultracentrifuge, MLS-50 swing-out rotor) during each run at 200100 g and 20 °C for at least 22 hours to reach isopycnic equilibrium. Like in reference [37], one of the tubes is used to relate the depth in the tube to the density of the gradient medium via absorption spectroscopy, see Figure S1 in the Supporting Information (SI). This relation is identical for the other ultracentrifuge tubes as shown in Figure S2 in the SI.…”

“…This relation is identical for the other ultracentrifuge tubes as shown in Figure S2 in the SI. These other three ultracentrifuge tubes were thoroughly characterized as a function of depth using in situ RRS and PL spectroscopy by mounting the ultracentrifuge tubes on an automated translation stage (MTS50/M-Z8, Thorlabs) [37]. For these in situ measurements, the spectra were recorded with a 1 mm step size in height, while the spatial resolution in PL and RRS is 1 mm and 60 μm, respectively.…”

Quasiphase Transition in a Single File of Water Molecules Encapsulated in (6,5) Carbon Nanotubes Observed by Temperature-Dependent Photoluminescence Spectroscopy

“…[43,44] The resulting chirality-sorted nanotube suspension was first dialyzed to 1%wt/V DOC, using a stirred ultrafiltration cell, and subsequently density gradient ultracentrifugation was used to separate the empty and water-filled (6,5) SWCNTs. [48,49] Similar to previous work [13][14][15][16]48, 49] a combination of optical absorption, resonant Raman and PL-excitation spectroscopy was used to asign the composition of each of the fractions, see SM [46] indicating that the amount of filled SWCNTs are negligible in the empty fraction (< 3.7%), while in the PL spectra of the filled fraction a small contribution of empty SWCNTs can be observed, with an amplitude of ∼ 20% of the total PL intensity. For temperature-dependent optical experiments, SWCNTs were drop-cast on quartz substrates, dried and mounted in a Microstat He2 cryostat (Oxford Instruments).…”