Machine Learning in Polymer Research

Advanced Materials, Journal Year: 2025, Volume and Issue: unknown

Published: Feb. 9, 2025

Machine learning is increasingly being applied in polymer chemistry to link chemical structures macroscopic properties of polymers and identify patterns the that help improve specific properties. To facilitate this, a dataset needs be translated into machine readable descriptors. However, limited inadequately curated datasets, broad molecular weight distributions, irregular configurations pose significant challenges. Most off shelf mathematical models often need refinement for applications. Addressing these challenges demand close collaboration between chemists mathematicians as must formulate research questions terms while are required refine This review unites both disciplines address curation hurdles highlight advances synthesis modeling enhance data availability. It then surveys ML approaches used predict solid-state properties, solution behavior, composite performance, emerging applications such drug delivery polymer-biology interface. A perspective field concluded importance FAIR (findability, accessibility, interoperability, reusability) integration theory discussed, thoughts on machine-human interface shared.

Language: Английский

---

Property-guided generation of complex polymer topologies using variational autoencoders

npj Computational Materials, Journal Year: 2024, Volume and Issue: 10(1)

Published: June 29, 2024

Abstract The complexity and diversity of polymer topologies, or chain architectures, present substantial challenges in predicting engineering properties. Although machine learning is increasingly used science, applications to address architecturally complex polymers are nascent. Here, we use a generative model based on variational autoencoders data generated from molecular dynamics simulations design topologies that exhibit target Following the construction dataset featuring 1342 with linear, cyclic, branch, comb, star, dendritic structures, employ multi-task framework effectively reconstructs classifies while their dilute-solution radii gyration. This enables generation size, which subsequently validated through simulation. These capabilities then exploited contrast rheological properties topologically distinct otherwise similar behavior. research opens avenues for more intricate tailored learning.

Language: Английский

---

Tailoring Hierarchical Structure and Rare Earth Affinity of Compositionally Identical Polymers via Sequence Control

Journal of the American Chemical Society, Journal Year: 2024, Volume and Issue: 146(12), P. 8607 - 8617

Published: March 12, 2024

Macromolecule sequence, structure, and function are inherently intertwined. While well-established relationships exist in proteins, they more challenging to define for synthetic polymer nanoparticles due their molecular weight, conformational dispersities. To explore the impact of sequence on nanoparticle we synthesized a set 16 compositionally identical, sequence-controlled polymers with distinct monomer patterning dimethyl acrylamide bioinspired, structure-driving di(phenylalanine) (FF). Sequence control was achieved through multiblock polymerizations, yielding unique ensembles sequences which were simulated by kinetic Monte Carlo simulations. Systematic analysis global (tertiary- quaternary-like) structure this amphiphilic copolymer series revealed effect multiple descriptors: number domains, hydropathy terminal patchiness (density) FF within domain, each impacted both chain collapse distribution single- multichain assemblies. Furthermore, freedom segments local-scale, β-sheet-like interactions sensitive FF. connect target function, evaluated an additional nine copolymers as sequestrants rare earth elements (REEs) incorporating functional acrylic acid into select scaffolds. We identified key variables that influence binding affinity, capacity, selectivity REEs. Collectively, these results highlight potential boundaries via polymerizations drive primary hierarchical structures, ultimately functionality identical polymeric materials.

Language: Английский

---

Single Chain Nanoparticles in Catalysis

Angewandte Chemie International Edition, Journal Year: 2023, Volume and Issue: 63(7)

Published: Oct. 19, 2023

Abstract Over the last six decades folded polymer chains—so‐called Single Chain Nanoparticles (SCNPs)—have evolved from mere concept of intramolecularly crosslinked chains to tailored nanoreactors, underpinned by a plethora techniques and chemistries tailor analyze their morphology function. These monomolecular entities hold critical promise in wide range applications. Herein, we highlight exciting progress that has been made field catalytically active SCNPs recent years.

Language: Английский

---

Effects of Drug Conjugation on the Biological Activity of Single-Chain Nanoparticles

Biomacromolecules, Journal Year: 2024, Volume and Issue: 25(2), P. 675 - 689

Published: Jan. 24, 2024

The field of single-chain nanoparticles (SCNPs) continues to mature, and an increasing range reports have emerged that explore the application these small nanoparticles. A key for SCNPs is in drug delivery, recent work suggests can be readily internalized by cells. However, limited attention has been directed delivery small-molecule drugs using SCNPs. Moreover, studies on physicochemical effects loading SCNP performance so far missing, despite accepted view such should significantly affected content. To address this gap, we prepared a library bearing different amounts covalently conjugated therapeutic drug–sulfasalazine (SSZ). We evaluated impact both synthesis biological activity pancreatic cancer cells (AsPC-1). Our results reveal covalent conjugation side chains polymer precursor interferes with chain collapse cross-linking, which demands optimization reaction conditions reach high degrees cross-linking efficiencies. Small-angle neutron scattering diffusion-ordered spectroscopy nuclear magnetic resonance (DOSY NMR) analyses higher display larger sizes looser structures, as well increased hydrophobicity associated SSZ Increased led reduced cellular uptake when assessed vitro, whereby aggregation surface AsPC-1 toxicity. This highlights efficiency behavior

Language: Английский

---

Mapping Composition Evolution through Synthesis, Purification, and Depolymerization of Random Heteropolymers

Journal of the American Chemical Society, Journal Year: 2024, Volume and Issue: 146(9), P. 6178 - 6188

Published: Feb. 22, 2024

Random heteropolymers (RHPs) consisting of three or more comonomers have been routinely used to synthesize functional materials. While increasing the monomer variety diversifies side-chain chemistry, this substantially expands sequence space and leads ensemble-level heterogeneity. Most studies relied on composition simulated sequences design RHPs, but questions remain unanswered regarding heterogeneities within each RHP ensemble how closely these reflect experimental outcomes. Here, we quantitatively mapped out evolution compositions in four-monomer-based RHPs throughout a design-synthesis-purification-depolymerization process. By adopting Jaacks method, first determined 12 reactivity ratios directly from quaternary methacrylate RAFT copolymerization experiments account for influences competitive addition reversible activation/deactivation equilibria. The reliability silico analysis was affirmed by quantitative agreement (<4% difference) between results. Furthermore, conformation distribution different solvents as function composition, segmental characteristics via high-throughput computation based self-consistent field theory (SCFT). These comprehensive confirmed viable parameter engineer RHP-based materials long are accurately livingness synthesis is ensured.

Language: Английский

---

Unified Understanding of the Structure, Thermodynamics, and Diffusion of Single-Chain Nanoparticle Fluids

ACS Nano, Journal Year: 2024, Volume and Issue: 18(24), P. 15529 - 15544

Published: June 6, 2024

Single-chain nanoparticles (SCNPs) are a fascinating class of soft nano-objects with promising properties and relevance to protein condensates, polymer nanocomposites, nanomedicine, bioimaging, catalysis, drug delivery. We combine molecular dynamics simulations equilibrium time-dependent statistical mechanical theory construct unified understanding how the internal conformational structure SCNPs, both simple fractal globule-like form more complex objects multiple intermediate length scales, determines nm-scale intermolecular packing correlations, thermodynamic properties, center-of-mass diffusion over wide range concentrations up dense melts. The pair correlations generically exhibit distinctive deep correlation hole due SCNP connectivity repulsive interparticle interactions associated globular-like conformation on macromolecular scale, concentration-dependent deviations at small separations. Unanticipated exponential-like dependences equation-of-state, osmotic compressibility, constant fraction theoretically predicted confirmed via simulations. System-specific behaviors found structure, but overarching regularities identified understood based generalized effective globule scales. Diffusivity slows down by 2–3 decades increasing concentration is as consequence nonactivated excluded volume-driven weak-caging process space-time correlated forces experienced SCNP. Good agreement between established, testable predictions made, quantitative comparison viscosity measurements specific fluid carried out. basic theoretical approach can potentially be extended treat chemical physical consequences varying other classes nanoscale organization relevant in nanotechnology possible emergence kinetically arrested glasses.

Language: Английский

---

Designing single-polymer-chain nanoparticles to mimic biomolecular hydration frustration

Nature Chemistry, Journal Year: 2025, Volume and Issue: unknown

Published: March 12, 2025

Language: Английский

---

Physics-Guided Neural Networks for Transferable Property Prediction in Architecturally Diverse Copolymers

Macromolecules, Journal Year: 2025, Volume and Issue: unknown

Published: May 13, 2025

Language: Английский

---

Effect of the Protein Chain Conformation on the Collapse into Nanoparticles

Biomacromolecules, Journal Year: 2024, Volume and Issue: unknown

Published: Sept. 4, 2024

Protein single-chain nanoparticles can outperform synthetic in biomedical applications due to enhanced biocompatibility. Compared (co)polymers, the chemical complexity of proteins challenges chain conformation control. Here, we investigate impact precursor bovine serum albumin (BSA) on nanoparticle structure after intramolecular cross-linking. We explore urea concentration (denaturant), pH, salt, cross-linker length, and concentration. Small-angle neutron scattering dynamic light experiments reveal a shrinking upon However, ability collapse depends solvent conditions: more expanded chains more, whereas that are already compact barely change size Static measurements demonstrate binding is primarily intramolecular. The use shorter does not lead extended chains. Overall, BSA exhibits similar behavior polymer nanoparticles, which then allows harness for morphological

Language: Английский

---