Accelerated Chemical Reaction Optimization Using Multi-Task Learning.

Functionalization of C-H bonds is a key challenge in medicinal chemistry, particularly for fragment-based drug discovery (FBDD) where such transformations require execution in the presence of polar functionality necessary for protein binding. Recent work has shown the effectiveness of Bayesian optimization (BO) for the self-optimization of chemical reactions; however, in all previous cases these algorithmic procedures have started with no prior information about the reaction of interest. In this work, we explore the use of multitask Bayesian optimization (MTBO) in several in silico case studies by leveraging reaction data collected from historical optimization campaigns to accelerate the optimization of new reactions. This methodology was then translated to real-world, medicinal chemistry applications in the yield optimization of several pharmaceutical intermediates using an autonomous flow-based reactor platform. The use of the MTBO algorithm was shown to be successful in determining optimal conditions of unseen experimental C-H activation reactions with differing substrates, demonstrating an efficient optimization strategy with large potential cost reductions when compared to industry-standard process optimization techniques. Our findings highlight the effectiveness of the methodology as an enabling tool in medicinal chemistry workflows, representing a step-change in the utilization of data and machine learning with the goal of accelerated reaction optimization.

Competitive Adsorption of Interfacially Active Nanoparticles and Anionic Surfactant at the Crude Oil-Water Interface.

The interfacial activity of poly(N-isopropylacrylamide) (pNIPAM) nanoparticles in the absence and presence of an anionic surfactant (sodium dodecyl sulfate, SDS) was studied at a crude oil-water interface. Both species are interfacially active and can lower the interfacial tension, but when mixed together, the interfacial composition was found to depend on the aging time and total component concentration. With the total component concentration less than 0.005 wt %, the reduced interfacial tension by pNIPAM was greater than SDS; thus, pNIPAM has a greater affinity to partition at the crude oil-water interface. However, the lower molecular weight (smaller molecule) of SDS compared to pNIPAM meant that it rapidly partitioned at the oil-water interface. When mixed, the interfacial composition was more SDS-like for low total component concentrations (≤ 0.001 wt %), while above, the interfacial composition was more pNIPAM-like, similar to the single component response. Applying a weighted arithmetic mean approach, the surface-active contribution (%) could be approximated for each component, pNIPAM and SDS. Even though SDS rapidly partitioned at the oil-water interface, it was shown to be displaced by the pNIPAM nanoparticles, and for the highest total component concentration, pNIPAM nanoparticles were predominantly contributing to the reduced oil-water interfacial tension. These findings have implications for the design and performance of fluids that are used to enhance crude oil production from reservoirs, particularly highlighting the aging time and component concentration effects to modify interfacial tensions.