Automated parallel synthesis and high-throughput experimentation have been successfully utilized in the last two decades for the preparation of a broad range of functional polymer libraries featuring different properties and offering promising candidates for a wide range of applications in several fields of science and technology. For instance, we utilize commercially available robotized equipment for enabling the automated parallel synthesis and optimization of polymer libraries as well as the unattended sample preparation for a rapid primary screening of polymers and their synthesis conditions.
Nevertheless, after synthesis, polymers (or polymer libraries) contain unreacted monomers, residual solvents, and other impurities after synthesis. These unintended compounds can be problematic in further synthetic steps, characterization methods, processing, or applications including, for example, the manufacture of functionalized (co)polymers or drug delivery systems for biomedical usages. To solve this problem, we have developed new and cost-effective purification methods for polymer libraries that are suitable for automation and/or parallelization.
For instance, the implementation of a dialysis method for the simultaneous purification of different polymer materials in a commercially available automated parallel synthesizer (APS) was achieved.[1] The efficiency of this “unattended” automated parallel dialysis (APD) method was investigated by means of proton nuclear magnetic resonance (1H-NMR) measurements, which confirmed that the method enables the removal of up to 99% of the unreacted monomer from the synthesis of the corresponding polymers in the APS. Size-exclusion chromatography (SEC) revealed that the molar mass and molar mass distribution of the investigated polymers remained largely unchanged after using the APD method. The method presented herein offers a good alternative to the “unattended” and reliable purification of polymer libraries prepared in APS.
An additional method in this direction, recently developed at our laboratories, is based on physisorption (i.e., polymer purification by physisorption (3P)) onto suitable substrates.[2] For this case, model polymers were used to evaluate the capabilities and advantages of this method over conventional purification techniques in terms of separation of volatile compounds, solvent-to-polymer recovery efficiency, and final properties of the resulting polymers. Compared to dialysis and precipitation, the 3P method could also separate up to 99% of the residual monomer from a polymerization reaction using a significantly lower amount of solvent. Unlike the benchmark methods, the molar mass and dispersity of polymers purified by the 3P process remained practically unaffected; polymers were quantitatively recovered for subsequent applications (e.g., precursors for block copolymer synthesis or characterization purposes). Beyond serving as a straightforward technique for polymer purification, the 3P method is readily adaptable for automation within APS, which may help to overcome the unsolved and labor-intensive task of purifying (co)polymer libraries derived from high-throughput synthesis.
All in all, these two purification methods offer promising solutions to the current limitations of high-throughput/-output synthesis of polymer libraries, where purification currently represents a significant constraint for establishing more automated experimental workflows.[3] These strategies are valuable to advance towards the generation of reliable experimental data for artificial intelligence models and, therefore, support a more integrated digitalization of the research and development loop of new polymers for diverse applications.
References
[1] I. Terzioglu, C. Ventura-Hunter, J. Ulbrich, E. Saldívar-Guerra, U.S. Schubert, C. Guerrero-Sanchez, Polymers 2022, 14, 4835.
[2] V.D. Lechuga-Islas, M. Trejo-Maldonado, S. Stumpf, R. Guerrero-Santos, L. Elizalde-Herrera, U.S. Schubert, C. Guerrero-Sanchez, Eur. Polym. J. 2021, 159, 110748.
[3] D.C. Struble, B.G. Lamb, B. Ma, MRS Commun. 2024, 14, 752.