PhD position at CMM for a computational scientist

We are partners in a MSCA DN-JD project (Doctoral Network – Joint Doctorate) SENNET. The aim of SENNET is to create disruptive sensor technology for indoor air quality by incorporating porous materials and sensor technology. The project has pooled the interdisciplinary and intersectoral expertise of leading members located in Belgium, Germany, France, Ireland, Moldova, Spain, the UK and the Netherlands.

We are looking for an excellent PhD student to work on the “characterization of adsorption and dielectric and refractive index response of MOFs and zeolites”. The objective is to accurately predict (1) the adsorption isotherms and selectivity, and (2) the dielectric constants and refractive index response of nanoporous materials towards various volatile organic compounds (VOCs). For this purpose, accurate material specific force fields and/or ab initio trained machine learning potentials for a variety of VOCs need to be constructed to simulate single- and multicomponent isotherms. Additionally, dielectric constants and refractive index responses should be calculated for certain VOCs using ab initio techniques such as density functional theory.

We are looking for a computational scientist with a solid background in atomistic molecular simulations, quantum mechanics, statistical physics and thermodynamics applied to material science.

The candidate should have or will soon obtain a master’s degree of a university or international equivalent in the field of Physics or Physical Engineering. Depending on the curriculum also a master’s in chemistry, chemical engineering or another related field might have the desired background. Experience with molecular simulation software (LAMMPS, DLPOLY, RASPA, Gaussian, VASP, CP2K, …) and coding (Python, C, …) is an advantage. Additionally, the candidate has a pro-active working style, the willingness to look beyond the borders of his/her own discipline and a strong motivation to work in a multidisciplinary team. (S)he is highly motivated to become an independent researcher. We expect a researcher with excellent communication skills and a strong motivation to collaborate with other researchers, within the CMM, the SENNET consortium and our networks.

The selected candidate will not only receive state-of-the-art science/technology training but will also benefit from a unique soft-skills training program within the frame of the MSCA doctoral network, which will kick-start his/her career as a highly employable professional in the EU and beyond. We will host the potential PhD student at Ghent University in our research group, the Center for Molecular Modeling. However, this PhD position will lead to a joint-PhD between Ghent University (CMM) and Université de Montpellier (CNRS). As such a first secondment of six months is planned at CNRS Montpellier under the supervision of prof. Guillaume Maurin. Another secondment of 2-3 months is planned at one of the industrial partners SCM (Software for Chemistry & Materials) for which Stan van Gisbergen will be supervisor.

The successful candidates will receive an attractive salary in accordance with the MSCA regulations for Recruited Researchers. At Ghent University PhD funding is foreseen for 48 months. All applications proceed through the on-line recruitment portal on the website. Please carefully read the eligibility criteria on the website. The initial deadline for on-line registration has been extended for our research position. Ideally the selected PhD student starts prior to March 1st, 2023.

Potluck breakfast @CMM

After three years we could finally organize our yearly potluck breakfast again. On Thursday morning October 13th, 2022 we welcomed six new master students and a new PhD student.

At the beginning of October Alen T. Mathew joined CMM to work together with Reza Mehdipour. Alen received his master’s degree in pharmaceutical chemistry from the Indian Institute of Technology. During his master thesis he studied the folding patterns of Tau protein using replica exchange molecular dynamics. At CMM his PhD project is mainly focused on membrane transporters and receptors, which he will study using molecular dynamics simulations and docking.

Hereafter you can find an overview of our new master students and their topics. We wish them and Alen a good time here at CMM and a lot of inspiration upon their scientific journey.

Bjarne De Bruyn - Extracting high-dimensional free energy surfaces from molecular simulations to accurately estimate the rate of physical and chemical processes

Sander De Meyer - The mystery of high-temperature superconductivity: ab-initio calculations using density-functional theory, downfolding and tensor networks

Bram Mornie - From Schrödinger to Newton: ab initio derived classical force fields versus machine learning potentials for zeolite properties

Lisa Ronsyn - Artificial intelligence-driven modeling of protein complexes

Wim Temmerman - Direct CO2 valorization on zeolite catalysts: Identifying olefin formation pathways in a complex molecular environment

Daan Verraes - Increasing the accuracy of quantum-mechanical simulations for strongly correlated functional materials by designing effective Hamiltonians

Congratulations dr. Maarten Cools-Ceuppens!

On September 15th our colleague Maarten Cools-Ceuppens successfully defended his PhD. During his defence Maarten presented his work entitled ‘Incorporating long-range interactions and polarization in machine learning potentials with explicit electrons’. He was supervised by prof. dr. ir. Toon Verstraelen and prof. dr. ir.  Joni Dambre.

Congratulations Maarten!

Summary of the PhD in laymen's terms

With molecular modeling, one can simulate the behavior of all kinds of materials at the level of individual atoms. At this length scale, systems composed of electrons and nuclei adhere to the laws of quantum mechanics, meaning that they can be simulated by solving the Schrödinger equation. In realistic applications, this approach requires an enormous amount of computing power, which remains challenging, even with today's supercomputers. In practice, one must resort to approximate methods instead. A force field is such a popular approximate method, in which electrons are no longer described explicitly, enabling simulations of larger atomistic systems over longer time scales.

For advanced applications, e.g. those involving charge transfer or polarization, conventional force fields do not capture all the relevant physics. Therefore, in this PhD, we have developed a new force field, in which electrons are re-introduced as semi-classical particles. This force field does capture the physics of charge transfer and polarization, yet without the computational burden of quantum-mechanical models. A machine-learning model was trained to incorporate the complex short-range quantum-mechanical interactions between semi-classical electrons. This new model can indeed learn these interactions automatically and truthfully, which was confirmed by accurate predictions of various electronic properties of organic molecules and periodic atomistic models of solids.

CMM @MOF2022 in Dresden

From September 4th until 7th, 2022, the 8th International Conference on Metal-Organic Frameworks and Open Framework Compounds (MOF2022) took place in Dresden. It was an ideal platform for scientists, developers and users to connect and present their novel generation of porous framework materials, discover new functions and meet experts from all disciplines. Of course, the CMM also attended this scientific meeting to show our latest results on metal-organic frameworks (MOFs).

Veronique Van Speybroeck was invited to talk about our challenging work on modeling spatiotemporal processes in realistic MOFs at length and time scales comparable to experimental observations. To bridge this gap towards larger length and time scales, one needs fundamentally new methods which combine the accuracy of density functional theory with the computational efficiency of classical force fields. The rapidly developing field of machine learning potentials may offer such a hybrid alternative.

Sven Rogge introduced the concept of strain engineering to quantify the impact building block alterations have on the local and overall flexibility in MOFs as well as to design MOFs for specific applications. Strain fields are local and time-dependent tensor quantities that describe a material’s deviation from its equilibrium structure under external triggers, thereby constituting an important fingerprint for (local) flexibility.

Sander Borgmans presented our protocol to explore the phase stability of covalent organic frameworks (COFs) exhibiting the flexible dia topology. With a case study he demonstrated how we successfully described the observed flexibility in COF-300, using an umbrella sampling protocol relying on judiciously chosen collective variables that describe the transitions.

Alexander Hoffman talked about the work we have done together with the group of Stefan Kaskel (Institute of Inorganic Chemistry - Dresden University of Technology) and prof. Alexander Krylov (Kirensky Institute of Physics – Federal Research Center KSC SB RAS). In this contribution, he presented our new theoretical approach, supported by experimental Raman measurements, to identify rigid unit modes in a set of MIL-53-type materials i.e. soft porous crystals with a winerack topology.

During the poster session Aran Lamaire introduced our work on getting Atomic insight in the flexibility and heat transport properties of MIL-53(Al) for water-adsorption applications.

CMM attends DFT2022 Conference in Brussels

Last week the CMM attended the 19th International Conference on Density Functional Theory and its Applications in Brussels ( The conference covers a broad range of topics in the field, from the latest theoretical developments to cutting-edge applications in chemistry and physics, bringing together scientists from all over the world.

Stefaan Cottenier was invited to talk about his efforts on comparing different DFT methods and codes. If you use DFT to predict a property of a crystal, how confident can you be that the prediction is computed in a bug-free way? And if your DFT-code uses pseudopotentials, can you trust the pseudopotential does not modify your predictions? If you want an answer to these questions, feel free to (re)watch his talk here.

Tom Braeckevelt presented our recent work on metal halide perovskites (MHPs). In the past decade, MHPs have shown great potential for various optoelectronic applications. However, their spontaneous transition to an inactive yellow phase impedes the widespread adoption of, e.g., CsPbI3 and FAPbI3. Using RPA calculations and phase-transferable machine learning potentials we determined the transition temperature of MHPs.

YingXing Cheng elaborated on a new ACKS2ω model, which offers a solid connection between the quantum-mechanical description of frequency-dependent response and computationally efficient force-field models. With this methodology we want to propose an alternative for today’s quantum-mechanics-based methods, which are still computationally expensive for extended systems.

Finally the CMM was also represented with two posters. Liesbeth De Bruecker presented our computational study of the electronic structure of Co2+ Aqua-Complexes and Sander Vandenhaute introduced our work on Machine Learning Potentials for Activated Processes using Active Learning.

Together with CNRS Montpellier we are looking for two computational scientists

Together with CNRS Montpellier we are looking for two Joint-PhD students within the frame of the Marie-Skłodowska-Curie Doctoral Network SENNET within the Horizon Europe Programme of the European Commission. SENNET is the “Porous Networks for Gas Sensing” project and will create disruptive sensor technology for indoor air quality by incorporating porous materials and sensor technology. The project has pooled the interdisciplinary and intersectoral expertise of leading members located in Belgium, Germany, France, Ireland, Moldova, Spain, UK and the Netherlands (Figure 1). The 12 SENNET researchers will not only receive state-of-the-art science/technology training but will also benefit from a unique soft-skills training programme. This will kick-start their careers as highly employable professionals in the EU and beyond.

Figure 1 - SENNET Consortium

About the project  

Air pollution is one of the most pressing environmental challenges worldwide. While outdoor air pollution appears often in the media, the effects of indoor air pollution are not to be underestimated since the average person spends about 22 h per day indoors. Many chemicals that decrease the quality of indoor air are emitted by carpets, paints, furniture, etc. The majority of these pollutants are volatile organic compounds (VOCs). Since VOCs can cause not only discomfort but also debilitating or even fatal conditions, it is desirable to measure VOC concentrations with high spatial and temporal resolution, via low-cost but reliable miniature sensors. However, selectively measuring harmful VOCs is particularly challenging because of the low concentration of the analyte and the multitude of interfering compounds present in indoor air. Currently available miniature sensors (e.g., metal oxide semiconductor sensors) typically cannot distinguish a VOC of interest from, for instance, an air freshener.

SENNET will develop novel sensors for the selective detection of priority VOCs, based on leveraging the adsorption properties of tunable porous materials, namely metal-organic frameworks (MOFs) and zeolites. To do so, we will bridge the gap between two fields, namely porous crystalline materials and sensor technology, that have thus far been separated by traditional subject boundaries. SENNET is the first training network that will tackle this challenge, and will do so by combining expertise in chemistry, physics, materials engineering, and sensors. A coordinated effort by 9 beneficiaries and 7 associated partner organizations from 8 countries guarantees a pan-European approach in a multi-environment context (universities, research centers, SMEs and large companies). The proposed ‘follow-the-challenge’ strategy ensures that young researchers are exposed to a variety of research environments and get involved in each step of the materials & sensors value chain.

Breakthrough materials for developing these novel sensors are MOFs and zeolites, crystalline, porous network materials. Whereas zeolites are inorganic aluminosilicates containing extra-framework cations, MOFs are hybrid solids that consist of inorganic nodes connected by multitopic organic molecules. Both classes of material have uniform pores with dimensions comparable to the size of the VOCs to be detected and have high surface areas (up to 6000 m2 g-1). Because of their chemical and structural characteristics, MOFs and zeolites can capture VOCs even at trace concentrations. Moreover, the adsorption preference (or selectivity) can be tuned by changing the nature of the framework. Although these adsorption properties are potentially very promising for VOC sensors, the integration of MOFs or zeolites into real devices has been largely overlooked. Firstly, due to a lack of systematic knowledge about which of the many different MOFs or zeolites would be best suited to adsorb a particular VOC from the air, and, secondly, the lack of suitable methods to integrate these materials with a sensor technology, e.g., coating deposition and signal transduction.

SENNET will address these challenges by combining expertise from fundamental chemistry all the way through to sensor engineering, resulting in the scientific objectives (SOs) outlined below . The overall approach is to combine multiple sensor elements coated by a porous material that displays cross-selective but different adsorption behaviour. The nature and concentration of the target VOC will be determined from the combined response of all the sensor elements through a multivariate calibration approach borrowed from chemometrics. The focus will be on the priority risk VOCs determined by the World Health Organization, i.e., tetrachloroethylene, formaldehyde and benzene.

The project’s five Scientific Objectives (SOs) are defined below:

  • Because of the large numbers of MOFs (> 10,000) and zeolites (> 240), and their numerous multicomponent adsorption behaviours, an efficient screening approach is needed. High-throughput computational approaches will be developed to tackle this problem and identify structure-adsorption property relationships.
  • The predicted behaviour must be validated by experimental adsorption tests for both single components and complex mixtures to fully understand the behaviour of the materials in indoor air.
  • Synthesis of MOFs and zeolites and fine-tuning of their properties to obtain the required mixed-component adsorption behaviour, based on the identified structure-adsorption property relationships. The materials will be prepared in a form that facilitates their integration in sensors.
  • Sensor fabrication and testing based on the most promising MOFs and zeolites. New fabrication approaches will be explored and the testing will involve real-world conditions. Different strategies to transduce the VOCs’ adsorption characteristics into a measurable signal will be benchmarked.
  • Signal-processing methods will be developed to calibrate the sensors and sensor arrays and to prevent non-selective or drifting signals as a result of temperature, relative humidity, or interfering compounds.

About Ghent University and CMM  

Ghent University is a top 100 university and one of the major universities in Belgium. Five Doctoral Schools support PhD students during their research training at Ghent University. The Center for Molecular Modeling (CMM,, led by prof. Veronique Van Speybroeck, is an interfaculty research unit at the Ghent University grouping about 40 scientists from the Faculties of Science and Engineering and Architecture. The CMM performs interdisciplinary research at the crossroads between physics, chemistry and materials engineering with the aim to design molecules, materials, and processes at the nanoscale. Excellence is pursued by stimulating interactions in the research team consisting of chemists, chemical engineers, physicists, physical engineers and bioengineers as well as with the vast network of national and international partners. The CMM has a proven track record in simulating nanoporous materials, with among others 2 ERC grants for V. Van Speybroeck in the topic. Within the CMM, prof. Louis Vanduyfhuys launched a new research track which devotes to a thermodynamic characterization of the properties of MOFs and related nanoporous materials. The group published 134 peer-reviewed papers in the last five years. Several papers were published in the highest ranked journals such as Science, Chem. Soc. Rev., Ang. Chem., JACS, Nature Mat., Nature Chem., Nat. Comm., Phys. Rev. Lett. The group has a vast experience in supervising PhD students and postdocs.

What can be your role as a computational scientist?

PhD student 1

We are looking for a PhD student to work on the “characterisation of adsorption and dielectric and refractive index response of MOFs and zeolites”.


  • Accurate prediction of adsorption isotherms and selectivity of nanoporous materials towards various VOCs
  • Accurate prediction of the dielectric constants and refractive index of nanoporous materials loaded with VOCs

Short Description of Work & Expected Results:

  • Construction of accurate material-specific force fields (FFs) and/or machine learning potentials (MLPs) from ab initio training data for 50 of the most promising materials for each VOC as identified by CNRS-Montpellier
  • Use these FFs/MLPs for the simulation of single- and multicomponent isotherms for the most promising 50 materials for each VOC identified by CNRS-Montpellier
  • Calculation of the dielectric constant and refractive index response for the 10 most promising materials for each VOC identified as function of their guest (VOC/water) loading using ab initio techniques such as DFT

The CMM will be your host institution with prof. Van Speybroeck and prof. Vanduyfhuys as your supervisors. Two secondments are planned, one at CNRS Montpellier under the supervision of prof. Guillaume Maurin and one at SCM (Software for Chemistry & Materials) under the supervision of Stan van Gisbergen.  

PhD student 2

CNRS-Montpellier is looking for a PhD student to work on the “high-throughput (HT) screening & rationalization of porous materials for selective VOC adsorption”.


  • HT screening of the IZA (zeolites) and CoreMOF 2019 (MOFs) databases using Monte Carlo simulations to identify the porous materials.
  • QSPR analysis of the so-created database using advanced statistical tools (ANN, etc.)
  • In silico design of novel MOFs with improved VOC adsorption performances assembling the key features identified in the previous step using an automated assembly of structure building units (AASBU) approach (case of MOF).

Short Description of Work & Expected Results:

  • Creation of an unprecedented database listing the adsorption performances of porous materials with respect to VOCs
  • Establishment of structure-adsorption property relationships using advanced statistical tools
  • Prediction of novel porous materials with improved adsorption performances

CNRS Montpellier will be your host institution with prof. Guillaume Maurin as your supervisor. Two secondments are planned, one at the CMM under the supervision of prof. Van Speybroeck and prof. Vanduyfhuys  and one at SCM (Software for Chemistry & Materials) under the supervision of Stan van Gisbergen.

Candidate requirements 

  • We are looking for a computational scientist with a good background in atomistic molecular simulations, quantum mechanics, statistical physics and thermodynamics applied to material science.
  • Experience with molecular simulation software (LAMMPS, DLPOLY, RASPA, Gaussian, VASP, CP2K, …) and coding (Python, C, …) is an advantage.
  • You have a pro-active working style, the willingness to look beyond the borders of your own discipline and a strong motivation to work in a multidisciplinary team. You are highly motivated to become an independent researcher.
  • You have excellent communication skills and have a strong motivation to collaborate with other researchers, within the CMM/CNRS Montpellier, the SENNET consortium and our networks.
  • You have or will soon obtain a master’s degree of a university or international equivalent in the field of Chemistry, Chemical Engineering, Physics, Physical Engineering or a related field.

Benefits and salary

The successful candidates will receive an attractive salary in accordance with the MSCA regulations for Recruited Researchers. The exact (net) salary will be confirmed upon appointment and is dependent on local tax regulations and on the country correction factor (to allow for the difference in cost of living in different EU Member States). The salary includes a living allowance, a mobility allowance and a family allowance (if applicable). The guaranteed PhD funding is for 36 months (i.e. EC funding, additional funding is possible, depending on the local Supervisor, and in accordance with the regular PhD time in the country of origin). In addition to their individual scientific projects, all fellows will benefit from further continuing education, which includes internships and secondments, a variety of training modules as well as transferable skills courses and active participation in workshops and conferences.

On-line Recruitment Procedure 

All applications proceed through the on-line recruitment portal on the website. Candidates apply electronically for one to maximum three positions and indicate their preference. Candidates provide all requested information including a detailed CV (Europass format obligatory) and motivation letter. During the registration, applicants will need to prove that they are eligible (cf. Recruited Researchers definition in Horizon Europe MSCA work programme 2021-2022, mobility criteria, and English language proficiency):

  • Supported researchers must be doctoral candidates, i.e. not already in possession of a doctoral degree at the date of the recruitment.
  • Researchers must be enrolled in a doctoral programme leading to the award of a doctoral degree in at least one EU Member State or Horizon Europe Associated Country, and for Joint Doctorates in at least two.
  • Recruited researchers can be of any nationality and must comply with the following mobility rule: they must not have resided or carried out their main activity (work, studies, etc.) in the country of the recruiting beneficiary for more than 12 months in the 36 months immediately before their recruitment date.

The deadline for the on-line registration is 15 September 2022. Prior to the recruitment, videoconferencing (or in person, when possible) interviews between the Supervisors and the candidates will be organized. The final decision on who to recruit is communicated no later than October 2022. The selected researchers are to start their research as quickly as possible (ideally prior to 1 March 2023).

Applicants need to fully respect three eligibility criteria (to be demonstrated in the Europass CV):

  • Conditions of international mobility of researchers:
    • Researchers are required to undertake trans-national mobility (i.e. move from one country to another) when taking up the appointment.
    • At the time of selection by the host organisation, researchers must not have resided or carried out their main activity (work, studies, etc.) in the country of their host organisation for more than 12 months in the 3 years immediately prior to their recruitment. Short stays, such as holidays, are not taken into account.
  • English language proficiency:
    • Network fellows must demonstrate that their ability to understand and express themselves in both written and spoken English is sufficiently high for them to derive the full benefit from the network training.

Three talks by prof. Van Speybroeck in June

At the university of OSLO during the NordCO2 Monthly Seminar

The Nordic Consortium for CO2 Conversion (NordCO2) is a network for researchers working on chemical CO2 conversion in the Nordic countries. NordCO2 promotes knowledge exchange, initiates novel scientific collaborations, trains Nordic students and organizes outreach activities. For their monthly seminars, they invite experts in the fields relevant to the NordCO2 consortium. This month prof. Van Speybroeck was invited for an online guest lecture entitled “A molecular modeling perspective on the C1 chemistry for light olefin and aromatics formation over zeolites”. Conversion of C1 feedstocks such as CO2 or methanol is an important technology in our ambition to produce chemical building blocks from non-fossil feedstocks. For this, new catalysts need to be developed which are selective, active and have long lifetimes under changing operating conditions. Modelling may play a detrimental role in the development of these future catalysts.

For the Royal Society in London during the scientific meeting on supercomputer modelling of advanced materials

Prof. Van Speybroeck gave an invited lecture on the discussion meeting on supercomputer modelling of advanced materials organised by the Royal Society in London. The Royal Society is a Fellowship of many of the world’s most eminent scientists and is the oldest scientific academy in continuous existence. The scientific meeting was organised to discuss the development of advanced materials in key scientific and industrial areas, including energy, catalysis and quantum technologies. High end computing and data science offer unprecedented opportunities for predictive modelling of complex materials. The meeting explored the scientific and methodological challenges in the field, focusing on structure prediction, nucleation and crystal growth, biomaterials and catalysis. While there, prof. Van Speybroeck presented how we model realistic nanoporous materials at operating conditions.

Nanoporous materials used in catalysis, sorption, separation and other applications are far from perfect. They possess a broad range of heterogeneities in space and time extending over several orders of magnitude. Furthermore, their functional behaviour is largely determined by the conditions in which they do the work. Not only for the experimentalist, but also for us as computational researchers this forms a tremendous challenge. How can we model realistic materials having defects at length and time scales comparable to experiment? How should we model active sites in true operating conditions of temperature and pressure? A further reading on how we deal with these challenges in the case of metal-organic frameworks (MOFs) can be found in our publication in Trends in Chemistry.

During the fifth annual UK Porous Materials Conference at the University of Strathclyde in Glasgow

The UK Porous Materials meetings are organised by the Porous Materials Interest Group of the Royal Society of Chemistry (RSC). This group acts as a focal point for researchers working on all aspects related to porous materials, including synthesis and design, characterisation, applications and modelling. They cover research on a wide variety of materials, including zeolites and zeotypes, Metal-Organic Frameworks (MOFs), Covalent Organic Frameworks (COFs), porous silicas, porous carbons, porous polymers, porous organic cages i.a.. This meeting is thus not only an ideal moment to meet our computational colleagues, but also the perfect opportunity to change ideas with experimentalists. In Glasgow, Van Speybroeck will highlight the challenges of modelling realistic nanostructured materials at longer length and time scales.  For this we use enhanced sampling techniques to construct the free energy surfaces.  Those methods were to a large extent developed within an ERC CoG grant DYNPOR.  In our endeavour to model in a more realistic way materials with quantum accuracy, we are currently developing machine learning potential.  Some of our proof-of-concept results will be shared with the audience.

Guest lecture by dr. Sanggyu Chong

Last week on June 9th and 10th, dr. Sanggyu Chong visited the CMM to exchange ideas in small-group and one-on-one discussions with us. Dr. Chong recently started working as a postdoctoral researcher with prof. Michele Ceriotti at the Laboratory of Computational Science and Modeling (COSMO) at the École Polytechnique Fédérale de Lausanne (EPFL), focusing on machine learning potentials for MOFs and other materials. Before, he completed his PhD at KAIST (Korea Advanced Institute of Science and Technology), working on the electronic structure modelling of MOFs under the supervision of prof. Jihan Kim.

During his research visit to the CMM Sanggyu gave a talk about his research on the ‘Computational Design of Electrically Conductive Metal-Organic Frameworks by Exploring Strategies to Construct Charge Transport Pathways’. Metal–organic frameworks, or MOFs, which are widely recognized for their ultrahigh porosity and chemical tunability, have shown great promise in a variety of adsorption-based applications. More recently, it has been shown that electrical conductivity can be induced in MOFs, which has led to their successful utilization in electronic or electrochemical applications. One could expect that chemical tunability and high porosity of MOFs would allow these materials to achieve outstanding and unprecedented performances in these new application fields. Nonetheless, progress is stunted by the limited number of electrically conductive MOFs discovered to date, and hence continued discovery of new and improved electrically conductive MOFs is crucial. In his talk, Sanggyu discussed the computational design approaches for the discovery of new electrically conductive MOFs by constructing long-range charge transport (CT) pathways in the materials. More specifically, he presented on the following: (1) rational modifications of previously insulating MOFs to newly induce electrical conductivity, (2) installation of electroactive moieties in MOFs with significant framework flexibility, (3) topologically guided construction of 3D π-d conjugated MOFs, and (4) development of a deep learning model to predict the electronic structures of MOFs. We are eager to witness how the design approaches and computational methods proposed and developed by Sanggyu will significantly expedite the discovery of new electrically conductive MOFs!

Two new FWO postdoctoral fellowships

We are proud to announce that two of our postdoctoral researchers obtained a personal fellowship from the Research Foundation – Flanders (FWO) for three years to support their independent international research career. Jelle Vekeman started at CMM in October 2020 with the guidance of prof. Toon Verstraelen. Jenna Mancuso arrived at CMM in September 2021 to join the research group of prof. Veronique Van Speybroeck. We are pleased to continue our collaboration with them in the frame of their FWO postdoctoral fellowship. Congratulations!

About Jelle Vekeman

From the start Jelle was convinced that international experience and collaboration is a key to success. Already during his master year he went to the University of Girona in Spain for an Erasmus stay of 10 months. For his master thesis entitled ‘The role of electron correlation and atomic partition on bond Fukul functions’ he worked in the lab of prof. dr. Solà under supervision of dr. Matito and prof. dr. Bultinck at his home institute, Ghent University.

For his PhD he worked in two different institutes abroad, namely the University of Valencia, also in Spain, and the university of Perugia in Italy. Furthermore he performed secondments at Alya Technology & Innovation S.L., an industrial partner of the Marie-Sklodowska-Curie Innovative Training Network of which he was one of the early stage researchers (ESR). In Valencia he performed static CCSD(T) and DFT calculations and developed force fields which he later used for molecular dynamics simulations (in Perugia) and grand canonical Monte Carlo simulations (at Alya Technology & Innovation S.L.).

Back in Belgium he performed two postdoctoral projects at the Vrije Universiteit Brussel, the first on spectroscopic characterization of oxides using periodic DFT and the second on molecular dynamics simulations of polymer solubility. Since October 2020 he is a postdoctoral researcher at CMM and working on non-reactive molecular dynamics simulations of HSIL systems as precursors for zeolite formation.

About Jelle’s project – A Reactive Molecular Model for Aluminosilicate Chemistry to Study Zeolite Formation

Despite their large commercial importance, zeolite formation is poorly understood due to the complex, heterogeneous nature of traditional synthesis. COK-KAT (KU Leuven) recently reported a novel synthesis path via hydrated silicate ionic liquids (HSILs), completely homogeneous, inorganic liquids which yield zeolites at moderate conditions. HSILs are severely subhydrated, room temperature alkali-silicate melts consisting of small oligomers. Water is not present as bulk, but as a ligand to the ionic species. HSILs are very stable, until addition of aluminate triggers nucleation and zeolite growth even at room temperature (~6 months) or within minutes at 180°C. The unique properties of HSILs allow for development of a reactive molecular model for aluminosilicate chemistry at the Center for Molecular Modeling (CMM, Ghent University), which can be carefully tested against detailed experimental results obtained at COK-KAT. As zeolite formation involves successive condensation reactions, reactive neural network potentials will be trained on high-level DFT-D data to be used in large-scale molecular dynamics simulations. To minimize the amount of expensive DFT-D calculations, an active learning scheme will be employed. Enhanced sampling methods will be used to efficiently explore the free energy surface. This, in combination with detailed experimental insight, will lead to a better understanding of the relationship between HSIL composition and the experimentally observed topology.

About Jenna Mancuso

Jenna worked at the Hendon Materials Simulation group at the University of Oregon during her PhD. She focused on the application of DFT to metal-organic framework development for the purposes of catalysis and renewable energy (e.g. charge storage, supercapacitors, fuel cells, photomaterials). Additionally, she has extensive experimental experience relating to the mechanical, electronic and optical properties of functional polymeric components (e.g. in photovoltaic modules) as well as the synthesis, characterization, and purification of inorganic phosphors and optically active organic molecules. With expertise ranging across fields of polymer engineering, organic and inorganic synthesis, and theoretical chemistry she possesses a unique scientific perspective and ability to connect theory and experiment.

Jenna joined the CMM in September 2021 to provide atomic scale insights into the spatiotemporal evolution of chemical systems in industrial processes using a variety of computational methods, focusing on methanol-to-hydrocarbon conversions in zeolites. She mainly applies enhanced sampling methods of molecular dynamics to establish robust kinetic and thermodynamic models of realistic materials under operating conditions, utilizing rigorous spectroscopic benchmarking to experiment.

About Jenna’s project - Realistic molecular simulations of diffusion and reactivity in hierarchical zeolite catalysts

Methanol-to-hydrocarbon (MTH) conversion from renewable feedstocks is a viable source of light olefins or fuels, provided catalyst selectivity and lifetime can be enhanced. Hierarchical zeolite structures, imbued with a secondary mesopore system within the innate microporous crystal are promising architectures for improved MTH conversion. However, the origins of increased lifetime and short-olefin selectivity in these catalysts is poorly understood. To guide future catalyst development, this project will establish the first quantum mechanical models of hierarchical zeolites to clarify mesopore surface chemistry and elucidate the impact on diffusion and reactivity of key MTH intermediates and products. Rigorous structural validation with experiment will help define accurate structural models, to which state-of-the-art operando modeling techniques will be applied. To accelerate simulation times with first-principles accuracy and facilitate the use of larger models, machine learning potentials will be developed with the generated data. Thus, this work will provide crucial insights into mesopore chemistry and stimulate future reaction modeling in more accurate hierarchical models.

Two brand new PhD’s at CMM

© UGent, foto Anneke D'Hollander

Last week two of our PhD fellows successfully defended their work. Congratulations to both!

Elias Van Den Broeck presented his work entitled ‘A Multiscale Modeling Approach to Understand Reactivity and Interactions in Complex Molecular Environments with Applications in Polymer Chemistry’ on Wednesday April 20th, 2022. He was supervised by prof. dr. ir. Veronique Van Speybroeck.


In this doctoral thesis, molecular modeling is employed to gain a fundamental understanding of reactivity and interactions in complex molecular environments. More specifically a broad range of modeling approaches is used to solve various scientific questions related to polymer chemistry, ranging from short to long timescales and from small to large scale systems. We show that molecular modeling is indeed an indispensible tool to understand polymerization features, governing molecular interactions, reaction mechanisms, reaction kinetics and material properties. Given the complexity of the studied systems, a complementary set of techniques is necessary to answer the scientific question at hand combining both static and dynamic approaches with classical and/or quantum mechanical models.

Additionally, to obtain an accurate description of the molecular system under investigation and its molecular environment, models need to account for the operating conditions such as realistic temperatures and a proper solvent environment. In this thesis we adopted, to a large extent, molecular dynamics (MD) simulations while explicitly considering the solvent environment. The work was performed in close collaboration with various experimental partners and to answer the scientific questions at hand, we had to apply a multiscale modeling approach. To this end, we have set up different protocols and workflows throughout this thesis in order to construct and equilibrate the molecular systems realistically and describe the corresponding chemistry in their complex molecular environments. Typically a trade-off is made between accuracy and computational cost when setting up the molecular model which inherently depends on the system under investigation and the scientific problem which has to be solved.

Leonid Komissarov worked under the supervision of prof. dr. ir. Toon Verstraelen. His dissertation entitled ‘Optimizing Potential Energy Surface Models’ was presented on Friday April 22nd, 2022.


The digital revolution has undoubtedly been a major contributor to the shaping of modern society. Nowadays computational simulations play an integral part in science and industry, enabling novel discoveries at an unprecedented pace. Such reliance on simulations means that there is a constant demand for hard- and software that produces results faster, more accurately and at a lower cost.

This thesis highlights a strategy that can deliver fast and accurate computational models: optimization; specifically in the context of empirical models of the potential energy surface, as used in physics and chemistry. Here, the problem of inaccurate predictions can be addressed by fitting model parameters to reference data. In doing so, a previously poor model can be trained to perform significantly better. Although this seems like a simple and viable approach on paper, the implementation is not straightforward: To date, there exists a plethora of models for molecular simulation, various optimization algorithms, and a number reference data sources. Until now the process of interfacing the above components has been tedious and prone to produce workflows that are of little comprehensive use to the scientific community.

We have developed a tool that alleviates these issues. It facilitates parameter optimization, allowing researchers to focus more on their science than the time-consuming technical details. In addition, the tool is highly flexible as users can mix and match various optimizers with different models and include any computable physicochemical property in the fitting procedure. The following pages provide an overview of how parametric molecular models can be optimized. We will discuss various types of models and optimization strategies before introducing our software tool. The advantages of our software are underlined with multiple successful parametrization examples. First applications of it resulted in improved performance of the ReaxFF and GFN1-xTB models. Both parametrizations have been made available to the scientific community and are discussed in the included papers.


Subscribe to News