Simon Gravelle

Postdoctoral researcher

Universidad Adolfo Ibáñez, Campus Viña del Mar, Chile

Fondecyt-Conicyt 2017 grant

Research interests:


Q1: Does the hourglass shape of aquaporins optimize water permeability?
Our answer is yes. Both numerical calculations and molecular dynamics simulations indicate that an aquaporin offers the minimal hydrodynamic resistance to flow thanks to its hourglass geometry which reduces the viscous entrance dissipation, see PNAS2013 and JCP2014 . See also a vulgarisation article (in french) MS2015 .

Q2: What makes the desert plant Tillandsia so efficient for water capture [In progress] ?
Combining micrographs of the plant epidermis and direct water flow measurement, our goal is to build a physical model explaining the water transport asymmetry that allow Tillandsia desert plant to survive in the Atacama. Our final goal is to mimick this property and fabricate artificial membranes with anomalous water transport properties.

The conical shape of an aquaporin / A trichome of Tillandsia and a Tillandsia.

Water-material interaction:

Q1: What equation rules the liquid transport through a nanopore below the continuum limit?
Liquid flow through a nanopore, which is a nanochannel with a low length over radius ratio, is controlled by the dissipation at the entrances. Using molecular dynamics simulations, we found that the classical continuum hydrodynamics is incredibly robust and predicts quantitatively transport below the continuum limit, even for pores with single file transport regime, see JCP2014 .

Q2: How does capillarity look like below the continuum limit?
We studied the capillary filling of sub-nano-metric carbon nanotubes (CNTs), and found strong deviations from the classical law of capillarity. Those deviations have been found to be the signature of the disjoining pressure. Strikingly, for some "magic" value of tube diameter, switching from hydrophilic to hydrophobic behaviour have been observed, see PRE2016. See also details here Capillary filling inside subnanometric carbon nanotube (CNT).

Q3: How to separate water from ethanol using membrane filtration?
Carbon based membrane with nanometric pores (short carbon nanotube membrane, nanoporous graphene and graphene oxyde-like membrane) can be used to separate very efficiently ethanol from water, as demonstrated by our molecular dynamics results. In a very counter-intuitive way, these membranes are shown to be permeable to both liquids when considered as pure components, but become semi-permeable to water for water-ethanol mixtures. More details in JCP2016 .

Q4: What equations rule the liquid/vapour transport of water through cellulose [In progress] ?

Water molecules inside a carbon nanotube (CNT) / the use of carbon membrane for the separation of water and ethanol.

Water flow measurement:

Q1: How can we build up a membrane that offers a low resistance to water flow, while offering a large resistance to vapour water [In progress]?

Q2: How can we detect picoliter-per-minute flows in a nanochannel using fluorescence correlation spectroscopy (FCS) [In progress]?


Q1: How does the solvent behaves inside a nanofluidic diodes?
We found that a nanofluidic diode rectifies solvent flow the same way it rectifies the ionic current, opening possibilities to control water flow at the nanoscale by the use of electric field, see PRL2013

Q2: What in the impact of interaction between pores in a network on the transport properties?
We found that interactions between pores in a cluster increase the overall electrical resistance of the membrane, see PoF2014.

Q3: What is the origin of the commonly observed pink noise in ionic current measurement through nanopore [In progress]?

Q4: What govern the division of cells [In progress]?
See PNAS2016.

The glandular trichome of Dionaea muscipula.




Simon Gravelle, Hiroaki Yoshida, Laurent Joly, Christophe Ybert and Lydéric Bocquet. Carbon membranes for efficient ethanol-water separation J. Chem. Phys. 145 (2016)

Adrien Guérin, Simon Gravelle and Jacques Dumais. Commentary - Forces Behind Plant Cell Division PNAS (2016)

Simon Gravelle, Christophe Ybert, Lydéric Bocquet and Laurent Joly. Anomalous capillary filling and wettability reversal in nanochannels PRE 93 (2016)

Simon Gravelle, Laurent Joly, François Detcheverry, Christophe Ybert, Cécile Cottin-Bizonne and Lydéric Bocquet. Perméabilité optimale des aquaporines : une histoire de forme ? médecine/sciences 31 (2015)

Simon Gravelle, Laurent Joly, Christophe Ybert and Lydéric Bocquet. Large permeabilities of hourglass nanopores: from hydrodynamics to single file transport J. Chem. Phys. 141 (2014)

Alessandro Gadaleta, Catherine Sempere, Simon Gravelle, Alessandro Siria, Rémy Fulcrand, Christophe Ybert and Lydéric Bocquet. Sub-additive ionic transport across networks of solid-state nanopores Phys. Fluids 26, 012005 (2014);

Clara B. Picallo, Simon Gravelle, Laurent Joly, Elisabeth Charlaix and Lydéric Bocquet. Nanofluidic Osmotic Diodes: Theory and Molecular Dynamics Simulations PRL 111, 244501 [5 pages] (2013) + Sup. Mat.

Simon Gravelle, Laurent Joly, François Detcheverry, Christophe Ybert, Cécile Cottin-Bizonne and Lydéric Bocquet. Optimizing water permeability through the hourglass shape of aquaporins PNAS, 110 (41), 16367–72 (2013) + Sup. Mat.


Simon Gravelle. Nanofluidics: a theoretical and numerical investigation of fluid transport in nanochannels Thesis manuscript (coming soon), Universite Claude Bernard (2015)
Simon Gravelle. A pedagogical introduction to nanofluidics, from my thesis manuscript.


Simon Gravelle
Universidad Adolfo Ibáñez
Campus Viña del Mar

Selected publications:

citations ~ 100, h-index=5, i10-index=3 (source Google Scholar)

Meetings (speaker):

Teaching background:

  • 2012-2015 : Supervisor in physics of material in the University Institute of Technology (Lyon1) (practical work and tutorial classes);
  • 2011-13 : Interrogator in CPGE (preparatory course for entrance examinations in Grandes Ecoles).



Xmgrace tips and tricks

Some usefull commands for the use of Xmgrace, an open-source plotting software.

Simple configuration

If you are new in molecular dynamic simulations, you can find here a starting configuration running with LAMMPS. This overly simplified simulation contains 3 water molecules only. Feel free to use it as a starting point for a much more complicated configuration. The output (dump file) can be visualized using VMD.
input_file and data_file

Water flowing through carbon nanotube (CNT)

This simulation contains two reservoirs filled with water separated by a membrane pierced with a single CNT. The flow is generated a pressure drop generated by two solid pistons. Run this simulation with LAMMPS and visualized the dump file with VMD.
input_file and data_file


Contact me by email:

simon.gravelle at

Find me on:

The universities I have learned in: