Carbon nanotubes (CNTs) possess extraordinary mechanical, thermal, and electrical properties due to their atomically perfect structure and sp2-hybridization. Recently, Single-wall CNTs (SWCNTs) have become an interesting host for the nanoscale confinement of fluids, with a wealth of surprising phenomena appearing: spontaneous filling, frictionless mass transport, unusual phase diagram, etc[1]. Most of these phenomena are still under debate and call for experimental confirmation. But the field is struggling with finding experimental approaches sensitive enough to carry out measurements at the level of individual nanotubes. SWCNT field effect transistors (SWCNT-FETs) showed that the electronic properties of SWCNTs are sensitive to their diameter, defects, doping, adsorbates, and environment [2], [3].In this contribution, we demonstrate that individual carbon nanotube field effect transistors (CNTFET) are excellent tools for this aim, for the first time allowing to precisely identify water confined inside the nanotube. By studying the electrical performances of several unopened and opened CNTFETs submitted to various atmosphere and temperature treatments, i.e. dry air, humidity, secondary vacuum, and current annealing, we show that it is possible to distinguish water being outside and inside the nanotube, just outside, or the nanotube free from water. We thus observed that for opened SWCNT both secondary vacuum and current annealing move threshold gate voltage towards more negative values, while for closed SWCNT secondary vacuum had no effect compared to current annealing. To sum up, the current annealing treatment is essential to distinguish water adsorbed outside from water confined inside. We show that this behavior is universal, as all devices' metallicities behave similarly, provided that the surface of the nanotube is pre-cleaned by current annealing. We will also discuss the mechanisms behind the coupling of electronic transport and the presence of water.Our results open up the possibility to use CNTFET for instance reliable, selective, and sensitive chemical and biological sensors, and also, to resolve the long-standing questions in the nanofluidic community about the behavior of water under nanoscale confinement.Reference:[1] T. A. Pascal, W. A. Goddard, and Y. Jung, “Entropy and the driving force for the filling of carbon nanotubes with water,” Proc. Natl. Acad. Sci., vol. 108, no. 29, pp. 11794–11798, 2011, doi: 10.1073/pnas.1108073108.[2] D. Cao et al., “Electronic sensitivity of carbon nanotubes to internal water wetting,” ACS Nano, vol. 5, no. 4, pp. 3113–3119, 2011, DOI: 10.1021/nn200251z.[3] I. Heller, A. M. Janssens, J. Männik, E. D. Minot, S. G. Lemay, and C. Dekker, “Identifying the mechanism of biosensing with carbon nanotube transistors,” Nano Lett., vol. 8, no. 2, pp. 591–595, 2008, DOI: 10.1021/nl072996i.

Density functional tight-binding calculations of the adsorption of OH − and other related diatomic anions on carbon nanotubes and graphene show that two adsorption states coexist: a physisorption state caused by Debye interactions, and a chemisorption state at shorter distances dominated by ionic-covalent interactions. For all anions, both adsorption energies scale linearly with the curvature of the carbon surface, and increase with decreasing anion polarizability but also, in metallic nanotubes, with chiral angle. At short distances, charge transfer modifies the attractive Debye forces into repulsive forces of the dipole-dipole type, tuning the strength of the chemisorption state, and may determine its existence and stability.

The field of ion transport through carbon nanotubes (CNTs) is marked by a large variability of the ionic conductance values reported by different groups. There is also a large uncertainty concerning the relative contributions of channel and access resistances in the experimentally measured currents, both depending on experimental parameters (nanotube length and diameter). In this perspective article, we discuss the ionic conductance values reported so far in the case of 2 individual CNTs and compare them with standard nano-fluidic models considering both the access and channel resistances. With a view toward guiding experimentalists, we thus show in which conditions the access or the channel resistance can predominate in CNTs. We explain in particular that it is not justified to use phenomenological models neglecting the channel resistance in the case of micrometer-long CNTs. This comparison reveals that most experimental conductance values can be explained in the framework of current nanofluidic models by considering experimental variations of slip length and surface charge density and that just a few extraordinarily high values cannot be accounted for even using extreme parameter values. Finally, we discuss how to complete existing models and how to improve the statistical reliability of experimental data in the field.