While porous carbons are used in many applications including energy storage, gas storage, water treatment, and catalysis, they are still poorly characterised due to their disordered nature and complex porosity. We are seeking a PhD candidate to work on the development of a new method to determine accurately the structure of these disordered materials.

This PhD project will take place in the CIRIMAT (Toulouse, France) starting in October 2019.

A description of the PhD project is given below. More information on the work context and recruitment process is available at http://bit.ly/2GFMeAx

Informal inquiries are welcome, please contact Céline Merlet ([email protected]). All applications must be sent through the “portail emploi CNRS” and must include a CV and a cover letter.

Porous carbons are an important class of materials used in many applications including
energy storage, gas storage, water treatment, and catalysis. In all cases, the characterisation of the
porous materials, still challenging from an experimental point of view, is an essential step in order
to understand and optimise the performance of the systems. One of the most important
characteristics of a porous material is its pore size distribution (PSD). The recognized method to
determine the PSD of a material is the adsorption isotherm analysis. This method, relatively easy to
implement, is available in many laboratories. The determination of the PSD from the measured
adsorption isotherms relies on different adsorption models developed over the years. These models
go from the simple BJH method (Barrett, Joyner, Halenda) to more complex models based on
classical density functional theory which can include a parameter related to the roughness of the
surface. Nevertheless, even the latest models are still discussed regarding their ability to deal with
the smallest pores or the presence of functional groups. Another drawback of the adsorption
isotherm analysis is that it only provides the global pore size distribution. For microporous carbons,
there is to-date no method to determine the arrangement of the different pore sizes in space. On a
more local scale, atomistic structures are usually generated using input from experiments such as
pair distribution functions (PDF). These are weighted histograms of atom-to-atom distances
showing the likelihood of finding an atom pair separated by a certain distance. Unfortunately, the
refinement of an amorphous structure from PDF is a multiple-solution problem and different
methods (e.g. Hybrid Reverse Monte Carlo or Quench Molecular Dynamics) can end-up with
structures having different characteristics (e.g. proportions of 5-/6-/7-membered rings) but similar
PDF. It is then hard to assess precisely the quality of these structures.

The objective of this PhD project is to develop a new method to determine accurately the
structure of disordered porous carbons. A first step will consist in calculating properties relevant to
the determination of porous structures (adsorption isotherms, NMR spectra, PDF, …) for a number
of carbon structures and adsorbed species. The aim of this study is to bring to light the differences
in experimental results which could be observed for consistent changes in the system. For example,
it will be possible to predict NMR spectra and PDF for small as well as large ions, and examine the
evolution of the features with pore size variations. At this stage, it will also be possible to do
isotherm adsorption measurements on mesoporous carbons with well-controlled structures as model systems. This should allow us to extract key properties, measurable experimentally, which can be used as constraints for the refinement of the structures. Indeed, one flaw of the existing methods is to consider experimental results in a disconnected way. The idea here is to develop a new method which would integrate experimental information from various sources as a set of constraints to limit the number of solutions for the determination of the porous structure. On the mesoscopic scale, the association of constraints from different experiments (with different electrolytes or temperatures) might be sufficient to determine more accurately the pore size distribution. On a more local scale, the objective is to develop a new Reverse Monte Carlo based technique including additional constraints (from NMR experiments for example). In a basic Reverse Monte Carlo simulation, applied to porous carbons, only the PDF of the carbon is used. Starting from a given configuration, random displacements of carbon atoms are done and a new PDF is calculated and compared to the previous one in order to decide if the move is accepted or rejected. This approach can be rendered more accurate by adding constraints in the acceptance criterion. For example, a constraint on the average chemical shift could be set: for each move, a new chemical shift would be determined and the agreement between the new chemical shift and the experimental results would be used in the acceptance or rejection of the move. The developed method will be tested first on carbons for which an atomistic structure is known before being applied to a larger set of materials.