A new “Interview around IT” in collaboration with Binarythe blog to understand digital issues.
Céline Merlet is a chemist and CNRS researcher at the Inter-University Center for Materials Research and Engineering (CIRIMAT) in Toulouse. She is a specialist in multi-scale models intended to describe energy storage materials. Energy storage (solar or wind for example) is becoming a major scientific challenge. Céline Merlet tells us about supercapacitors, a technology full of promise.
Binary: Could you briefly tell us about the career that led you to be a researcher in chemistry and a CNRS 2021 bronze medal
Celine Merlet: At first I didn’t go to do chemistry but biology. I did a prep and I wanted to become a veterinarian, but during the prep, I realized that I was more and more interested in chemistry. I also did a programming project and found it a lot of fun. I entered an engineering school, Chimie ParisTech. In 2e year, I did a three-week internship on modeling molten salts, salts that become liquid at very high temperatures. I discovered there the digital simulation of real world phenomena, I understood that I had found my way. After chemistry school, I went back to do a doctorate in the same lab where I had done my internship. A postdoc in England, and I was recruited at the CNRS in 2017.
B: Why didn’t you stay in England?
CM : With the difficulty of obtaining a position in France and the fact that I was well established in England, I also applied there. But there was Brexit and that confirmed my desire to return to France.
B: You are a chemist, specialist in electrochemical energy storage systems that involve complex materials. Could you explain to binary readers what this means?
CM : Electrochemical energy storage concerns the use of electrochemical reactions to store energy. The systems we know that do this are the batteries in phones and laptops, and cars. Batteries use complex materials with certain elements such as lithium, cobalt, and nickel. The device is charged and discharged by connecting it to an electrical circuit. Materials are changed during charging and discharging. This is what allows them to store energy.
Celine Merlet, Provided by the author
Celine Merlet, Provided by the author
My research focuses on supercapacitors. In these systems, we have two porous materials which are electrodes that we connect to each other via an external circuit. When we charge (or discharge), charged molecules will be placed in holes or on the contrary come out of them. A charge storage within the material results. But in a very different way than batteries. There is no chemical reaction. It is a simple adsorption of charged molecules.
B: You work on molecular modelling, what does it consist of?
CM : I spoke of the two electrodes which are in contact with this solution of charged ions. Often for supercapacitors, these are nanoporous carbons. The pores are about the size of a nanometer (1 millionth of a millimetre): this is something that cannot be observed with the naked eye. To understand how the ions go in and out of these carbon pores, instead of doing physical experiments, mixtures in a laboratory, I do numerical experiments, mixtures in the computer. I try to understand how ions move and what happens, on a scale that cannot be achieved experimentally.
B: Does it require a good understanding of the physical properties?
CM : Yes, to model the trajectory of ions, the way they move, you have to understand what is happening. When we throw a ball, if we give the forces that we apply at the start, we can deduce the trajectory. It’s the same for ions. We choose the starting point. We know which forces apply, the forces of attraction and repulsion. We have constraints such as the fact that one molecule cannot penetrate inside another. This allows us to calculate the evolution of the system of molecules over time. Sometimes you don’t even need to represent that very precisely. If even rough modeling is validated by experiments, we have the result we were looking for. In my lab, the CIRIMAT, there are mainly experimenters. We are just 4 or 5 theoreticians on permanent posts. In my team, researchers work directly on real chemical systems and we learn a lot from theory/experiment exchanges.
B: Typically, how many atoms are involved in these simulations?
CM : In these numerical simulations, a few hundred to a few thousand atoms are considered. In a real experiment, it’s at least 1024 atoms. (A milliliter of water already contains 1022 molecules.)
B: And despite that, you manage to understand what is happening for real…
CM : We use simulation tricks to find what is happening in reality. Part of my job is to develop models to make the link between the molecular scale and the experimental scale. When we change scale, it allows us to integrate certain elements but we lose other information on the molecular scale.
B: In these simulations of carbon electrodes within model supercapacitors in operation, what are the obstacles that you had to face?
CM : At the molecular level, there is still progress to be made, and more powerful computers could help. Materials conduct electricity, models consider carbons to be perfectly conductive, but in reality they are not. For a better representation, the semiconducting nature of these materials should be taken into account and some researchers are working on this aspect at the moment.
To obtain materials that would allow us to store more energy, we would have to better understand the microscopic properties that have an influence on what interests us, analyze molecular results to try to extract general trends. For example, if we have two liquids that have different ions, we make mixtures; we can brutally try lots of mixtures and carry out simulations for each one, or we can do only a few and try to understand from one mixture to another why the diffusion coefficient for example is different and thus predict what will happen for any mix. The better we understand what is happening, the less it is necessary to do molecular modeling on a massive number of examples.
Celine Merlet, Provided by the author
B: You received the “2021 Price Ada Lovelace” award for high performance computing (HPC). Do you present yourself more as a chemist, or as an HPC specialist?
CM : I do not present myself as a specialist in HPC computing but my activities require access to powerful computers and significant skills in this area. Part of my work consisted of improving certain programs to be able to use them on supercomputers. Making calculations possible on supercomputers opens up research perspectives, and it is a contribution to HPC computing.
B: What are the major applications in your field?
CM : Regarding supercapacitors, it is already used in car start-and-stop systems. It is also used in hybrid buses: we put supercapacitors on the roof of the bus, and each time it stops, we charge these supercapacitors and we use them to restart the bus. This can save up to 30% fuel. Questions that arise: Could we store more energy? Could other materials be used?
Toulouse Museum, Provided by the author
B: We know that the batteries of our phones weaken quite quickly. Could we replace them with supercapacitors?
CM : While batteries store more energy than supercapacitors, they degrade more over time. After a while the mobile phone no longer has the same autonomy as when you bought the phone. A supercapacitor can be charged and discharged very quickly a very large number of times without it being damaged. However, as the quantities of energy that they can store are much lower, we cannot imagine that standard supercapacitors can replace batteries. Rather, the two technologies are seen as complementary. And then, the line between supercapacitor and battery can be a little blurry.
B: You are active in “Women and Science”. Can you tell us what you do there and why you do it?
CM : I observe that we are still far from gender equality. In chemistry, we have a fairly good representation of women. In my laboratory, which corresponds well to national observations, there are 40% women. But in science in general, they are few.
One goal of “Women and Science” is to encourage young people, and particularly girls, to engage in scientific careers. I am on the board of directors, in charge of the website, and I coordinate with other people the activities in the Toulouse region. I am quite involved in interventions with schoolchildren, in high school or college classes: we talk about our backgrounds or we do workshops on stereotypes, small workshops to make young people aware of stereotypes, to understand what is is and what it can imply in career choices.
In 2019, we developed a game, Mendeleiev, for the celebration of 150 years of the periodic classification of the elements by Mendeleïev. We use it to highlight historical or contemporary women scientists: we have a table and we discover both the usefulness of the elements and the women scientists who have worked on these elements. We are digitizing this game.
The association also carries out many other actions such as exhibitions, booklets, etc.
B: Programming is a key part of your job; do you program yourself?
CM : I love programming. But since I spend a lot of time coaching, traveling, and attending meetings, I have less time to do it myself. I am the doctoral students who do this. Depending on their skill and appetite, I program more or less.
B: Where do the doctoral students who join your team come from? Are they chemists to begin with?
CM : Many come from all over the world: Morocco, Greece, India. They are physicists or chemists. I even have a student in computer science degree in L3 who is doing an internship with me.
B: Are there any research topics in computer science that are particularly important to you?
CM : At the moment, we are wondering what machine learning could bring to our field of research. For example, to model, we need to know the interactions between particles. Colleagues are trying to see if we could do automatic learning of force fields. We are not equipped to attack these problems, so we collaborate with computer scientists.