Role of High Fidelity Vs. Low Fidelity Experimental Data in Machine Learning Model Performance for Predicting Polymer Solubility.

Reliable classification of polymer-solvent compatibility is essential for solution formulation and materials discovery. Applying machine learning (ML) and artificial intelligence to this task is of growing interest in polymer science, but the effectiveness of such models depends on the quality/nature of the training data. This study evaluates how experimental data fidelity, as set by the experimental method, influences ML model performance by comparing classifiers trained on two experimental datasets: one generated from turbidity-based measurements using a Crystal16 parallel crystallizer as a high-fidelity source and another derived from visual solubility inspection as a low-fidelity dataset.

Solvent selection for polymers enabled by generalized chemical fingerprinting and machine learning.

We present machine learning models trained on experimental data to predict room-temperature solubility for any polymer-solvent pair. The new models are a significant advancement over past data-driven work, in terms of protocol, validity, and versatility. A generalizable fingerprinting method is used for the polymers and solvents, making it possible, in principle, to handle any polymer-solvent combination.

Stabilization of Metastable Indomethacin α in Cellulose Nanocrystal Aerogel Scaffolds.

Indomethacin (IM) is a small molecule active pharmaceutical ingredient (API) that exhibits polymorphism with the γ-form being the most thermodynamically stable form of the drug. The α-form is metastable, but it exhibits higher solubility, making it a more attractive form for drug delivery. As with other metastable polymorphs, α-IM undergoes interconversion to the stable form when subjected to certain stimuli, such as solvent, heat, pH, or exposure to seed crystals of the stable form.

Combining Surface Templating and Confinement for Controlling Pharmaceutical Crystallization.

Poor water solubility is one of the major challenges to the development of oral dosage forms containing active pharmaceutical ingredients (APIs). Polymorphism in APIs leads to crystals with different surface wettabilities and free energies, which can lead to different dissolution properties. Crystal size and habit further contribute to this variability.

Pharmaceutical crystallization in surface-modified nanocellulose organogels.

A significant research focus in the pharmaceutical industry is on methods to improve drug uptake into the body by increased dissolution of poorly water soluble active pharmaceutical ingredients (APIs) or sustained drug release behavior, which results in higher overall uptake. Production of higher energy, higher solubility polymorphs is one approach to address this problem. Here we utilize natural materials, cellulose nanocrystals (CNCs), that have a high surface area covered with readily-modified hydroxyl groups to form organogels that promote API crystallization into polymorphs that differ from the as-received materials.

In-Situ Platinum Deposition on Nitrogen-Doped Carbon Films as a Source of Catalytic Activity in a Hydrogen Evolution Reaction.

Copolymer-templated nitrogen-doped carbon (CTNC) films deposited on glassy carbon were used as electrodes to study electrochemically driven hydrogen evolution reaction (HER) in 0.5 M H2SO4. The activity of these materials was extremely enhanced when a platinum counter electrode was used instead of a graphite rod and reached the level of commercial Pt/C electrodes.