Stephan Mohr

Nature has only provided us with a limited number of bio-based and biodegradable building blocks. Therefore, the fine tuning of the sustainable polymer properties is expected to be achieved through the control of the composition of bio-based copolymers for targeted applications such as cosmetics. Until now, the main approaches to alleviate the experimental efforts and accelerate the discovery of new polymers have relied on machine learning models trained on experimental data, which implies an enormous and difficult work in the compilation of data from heterogeneous sources. On the other hand, molecular dynamics simulations of polymers have shown that they can accurately capture the experimental trends for a series of properties. However, the combination of different ratios of monomers in copolymers can rapidly lead to a combinatorial explosion, preventing the investigation of all possibilities via molecular dynamics simulations. In this work, we show that the combination of machine learning approaches and high-throughput molecular dynamics simulations permits to quickly and efficiently sample and characterize the relevant chemical design space for specific applications. Reliable simulation protocols have been implemented to evaluate the glass transition temperature of a series of 58 homopolymers, which exhibit a good agreement with experiments, and 488 copolymers. Overall, 2,184 simulations (4 replicas per polymer) were performed, for a total simulation time of 143.052 µs. These results, constituting a dataset of 546 polymers, have been used to train a machine learning model for the prediction of the MD-calculated glass transition temperature with a mean absolute error of 19.34 K and a R2 score of 0.83. Overall, within its applicability domain, this machine learning model provides an impressive acceleration over molecular dynamics simulations: the glass transition temperature of thousands of polymers can be obtained within seconds, whereas it would have taken node-years to simulate them. This type of approach can be tuned to address different design spaces or different polymer properties and thus have the potential to accelerate the discovery of new polymers.

Molecular dynamics simulations have been employed to investigate the effect of molecular polydispersity on the aggregation of asphaltene. To make the large combinatorial space of possible asphaltene blends accessible to a systematic study via simulation, an upfront unsupervised machine-learning approach (clustering) was employed to identify a reduced set of model molecules representative of the diversity of asphaltene. For these molecules, single asphaltene model simulations have shown a broad range of aggregation behaviors, driven by their structural features: size of the aromatic core, length of the aliphatic chains, and presence of heteroatoms. Then, the combination of these model molecules in a series of mixtures have highlighted the complex and diverse effects of molecular polydispersity on the aggregation process of asphaltene. Simulations yielded both antagonistic and synergistic effects mediated by the trigger or facilitator action of specific asphaltene model molecules. These findings illustrate the necessity of accounting for molecular polydispersity when studying the asphaltene aggregation process and have permitted establishing a robust protocol for the in silico evaluation of the performance of asphaltene inhibitors, as illustrated for the case of a nonylphenol resin.

We have performed molecular dynamics simulations to study the adsorption of ten hydrate anti-agglomerants onto a mixed methane–propane sII hydrate surface covered by layers of liquid water of various thickness. As a general trend, we found that the more liquid water that is present on the hydrate surface, the less favorable the adsorption becomes even though there are considerable differences between the individual molecules, indicating that the presence and thickness of this liquid water layer are crucial parameters for anti-agglomerant adsorption studies. Additionally, we found that there exists an optimal thickness of the liquid water layer favoring hydrate growth due to the presence of both liquid water and hydrate-forming guest molecules. For all other cases of liquid water layer thickness, hydrate growth is slower due to the limited availability of hydrate-forming guests close to the hydrate formation front. Finally, we investigated the connection between the thickness of the liquid water layer and the degree of subcooling and found a very good agreement between our molecular dynamics simulations and theoretical predictions.

The performance of organic friction modifiers (OFMs) depends on their ability to adsorb onto surfaces and form protective monolayers. Understanding the relationship between OFM concentration in the base oil and the resulting surface coverage is important for improving lubricant formulations. Here, we use molecular dynamics (MD) simulations to study the adsorption of three OFMs─stearic acid (SA), glycerol monoostearate (GMS), and glycerol monooleate (GMO)─onto a hematite surface from two hydrocarbon solvents─n-hexadecane and poly(α-olefin) (PAO). We calculate the potential of mean force of the adsorption process using the adaptive biasing force algorithm, and the adsorption strength increases in the order SA < GMS < GMO. We estimate the minimum area occupied by OFM molecules on the surface using annealing MD simulations and obtained a similar hard-disk area for GMS and GMO but a lower value for SA. Using the MD results, we determine the adsorption isotherms using the molecular thermodynamic theory (MTT), which agree well with one previous experimental data set for SA on hematite. For two other experimental data sets for SA, lateral interactions between surfactant molecules need to be accounted for within the MTT framework. SA forms monolayers with lower surface coverage than GMO and GMS at low concentrations but also has the highest plateau coverage. We validate the adsorption energies from the MD simulations using high-frequency reciprocating rig friction experiments with different concentrations of the OFMs in PAO. For OFMs with saturated tailgroups (SA and GMS), we obtain good agreement between the simulations and the experiments. The results deviate for OFMs containing Z-unsaturated tailgroups (GMO) due to the additional steric hindrance, which is not accounted for in the current simulation framework. This study demonstrates that MD simulations, alongside MTT, are an accurate and efficient tool to predict adsorption isotherms at solid–liquid interfaces.

The dissociation process of spherical sII mixed methane–propane hydrate particles in liquid hydrocarbon was investigated via microsecond-long molecular dynamics simulations. A strong dependence of the melting temperature on the particle size was found. Analysis in the context of the Gibbs–Thomson effect provided insights into the fundamental properties of gas hydrates.