- A preview of this full-text is provided by Springer Nature.
- Learn more
Preview content only
Content available from Nature Nanotechnology
This content is subject to copyright. Terms and conditions apply.
Articles
https://doi.org/10.1038/s41565-019-0425-y
Laboratoire de Physique de l’École Normale Supérieure, ENS, Université PSL, CNRS, Sorbonne Université, Université Paris-Diderot, Sorbonne Paris Cité,
Paris, France. *e-mail: lyderic.bocquet@ens.fr
Ionic transport is key to numerous processes, from neurotransmis-
sion to ultrafiltration1–4. Over the past decade, it has been exten-
sively investigated in biological systems, evidencing advanced
functionalities such as high selectivity, ionic pumping and elec-
trical or mechanical gating5–7. However, it is only recently that
experimental progress in nanoscience has allowed the fabrication
of artificial pores with controlled material properties and channel-
or slit-like geometries. These new systems have reached unprec-
edented nanometre- and even angström-scale confinements8–11, yet
they are still far from exhibiting the same functions as biological
ionic machines.
At these scales, ion transport is usually described by the Poisson–
Nernst–Planck (PNP) framework2, which couples ion diffusive
dynamics to electric interactions. Although it may account for
nonlinear (for example, diode-type) effects2, the PNP framework is
intrinsically continuum and mean field. Building bio-inspired func-
tions, however, may require control of transport at the single ion
level, which is out of the reach of a mean-field description. Single
charge transport in fact echoes the canonical Coulomb blockade
(CB) phenomenon, which has been thoroughly explored in nano-
electronics. CB is typically observed in a single electron transistor:
under fixed bias, the current between source and drain peaks at
quantized values of the gating voltage12. The origins of this effect
stem from the many-body Coulomb interactions between electrons
and from the discreteness of the charge carriers13. Similar physi-
cal ingredients are at play in a nanoscale channel filled with a salt
solution (Fig. 1a): the ions also interact via Coulombic forces, and a
variable surface charge on the channel walls can play the role of the
gating voltage. One may therefore expect to observe an ‘ionic CB’,
namely peaks in the ionic conductance of a nanochannel at quan-
tized values of its surface charge. It is thus of interest that molecular
dynamics simulations14–17 and experiments18–20 have shown what
might be indirect signatures of ionic CB (although in the absence
of a gating voltage), and conductance gating by a surface charge
has been demonstrated in simulations of a biological ion channel
model21,22. These observations remain surprising, because ionic
systems in water at room temperature have specific features con-
trasting with electronic systems that may preclude the occurrence
of ionic CB. Beyond the absence of quantum effects, the fact that
ions are of both signs— while electrons are only negative—results
in Debye screening, which is expected to greatly weaken the many-
body interaction. It remains unclear under which conditions these
aspects may suppress ionic transport quantization.
Although pioneering analytical efforts have translated the results
established for electrons13 to the ionic case22–24, a general theory
for ionic CB, incorporating the unique features of ionic systems in
contrast to their electronic counterparts, is still lacking. We develop
such a theory in this Article.
Model definition and numerical results
Our theory is based on a simple but general model of a nanochan-
nel which confines ions in one dimension (Fig. 1a). The channel
has radius R and length
≫LR
, as opposed to nanopores which have
length L ≈ R. The nanochannel is filled with water, which under
confinement exhibits a priori an anisotropic dielectric permittivity
ϵ
(refs. 25,26), and it is embedded in a membrane with low permit-
tivity
ϵm
(whenever needed, we use
ϵ=2
m
). Under such conditions
(Fig. 1b), the electric field lines produced by an ion stay confined
inside the channel over a characteristic length ξ (ref. 27). This leads
to a stronger Coulomb interaction than in the bulk solution, which
is well described by the exponential potential
ξ
=ξ−∣ ∣∕
Vx kTxe() (1
)
x
B
T
This introduces a thermal length xT (ref. 28), which quantifies the
strength of the interaction. We detail in Supplementary Section 3
how the parameters ξ and xT are related to the channel geometry
and to the various dielectric constants. If the permittivity of con-
fined water is assumed to be the same as in the bulk, one has ξ ≈ 7R
and
=∕ℓx R 2
T2B
, where
ℓ= .07nm
B
is the Bjerrum length in bulk
water. We shall use these relations in the following, keeping in mind
that taking into account the anisotropic permittivity would result in
a stronger interaction for a given confinement.
A charge is imposed on the confining surface and acts as a gate
on the system; here, we reduce the surface charge to a point-like
charge Q. The ions interact between themselves and with the sur-
face charge through the potential given in equation (1); depending
on conditions, an electric field E may be applied along the channel.
Ionic Coulomb blockade as a fractional Wien effect
Nikita Kavokine , Sophie Marbach , Alessandro Siria and Lydéric Bocquet *
Recent advances in nanofluidics have allowed the exploration of ion transport down to molecular-scale confinement, yet arti-
ficial porins are still far from reaching the advanced functionalities of biological ion machinery. Achieving single ion transport
that is tunable by an external gate—the ionic analogue of electronic Coulomb blockade—would open new avenues in this quest.
However, an understanding of ionic Coulomb blockade beyond the electronic analogy is still lacking. Here, we show that the
many-body dynamics of ions in a charged nanochannel result in quantized and strongly nonlinear ionic transport, in full agree-
ment with molecular simulations. We find that ionic Coulomb blockade occurs when, upon sufficient confinement, oppositely
charged ions form ‘Bjerrum pairs’, and the conduction proceeds through a mechanism reminiscent of Onsager’s Wien effect.
Our findings open the way to novel nanofluidic functionalities, such as an ion pump based on ionic Coulomb blockade, inspired
by its electronic counterpart.
NATURE NANOTECHNOLOGY | VOL 14 | JUNE 2019 | 573–578 | www.nature.com/naturenanotechnology 573
The Nature trademark is a registered trademark of Springer Nature Limited.