Efficient CO2 capture is vital for mitigating climate change, with amine-based solvents being widely used due to their strong reactivity with CO2. However, optimizing key properties such as basicity, viscosity, and absorption capacity remains challenging, as traditional methods rely on labor-intensive experimentation and predefined chemical databases, limiting the exploration of novel solutions. Here, SAGE-Amine was introduced, a generative modeling approach that integrates Scoring-Assisted Generative Exploration (SAGE) with quantitative structure-property relationship models to design new amines tailored for CO2 capture. Unlike conventional virtual screening restricted to existing compounds, SAGE-Amine generates novel amines by leveraging autoregressive natural language processing models trained on amine datasets. SAGE-Amine identified known amines for CO2 capture from scratch and successfully performed single-property optimization, increasing basicity or reducing viscosity or vapor pressure. Furthermore, it facilitated multi-property optimization, simultaneously achieving high basicity with low viscosity and vapor pressure. The 10 top-ranked amines were suggested using SAGE-Amine and their thermodynamic properties were further assessed using COSMO-RS simulations, confirming their potential for CO2 capture. These results highlight the potential of generative modeling in accelerating the discovery of amine solvents and expanding the possibilities for industrial CO2 capture applications.

The increasing importance of carbon capture technologies for deployment in remediating CO2 emissions, and thus the necessity to improve capture materials to allow scalability and efficiency, faces the challenge of materials development, which can require substantial costs and time. Machine learning offers a promising method for reducing the time and resource burdens of materials development through efficient correlation of structure-property relationships to allow down-selection and focusing on promising candidates. Towards demonstrating this, we have developed an end-to-end "discovery cycle" to select new aqueous amines compatible with the commercially viable acid gas scrubbing carbon capture. We combine a simple, rapid laboratory assay for CO2 absorption with a machine learning based molecular fingerprinting model approach. The prediction process shows 60% accuracy against experiment for both material parameters and 80% for a single parameter on an external test set. The discovery cycle determined several promising amines that were verified experimentally, and which had not been applied to carbon capture previously. In the process we have compiled a large, single-source data set for carbon capture amines and produced an open source machine learning tool for the identification of amine molecule candidates (https://github.com/IBM/Carbon-capture-fingerprint-generation).

Proteins, the fundamental nanomachines of life, have provided scientists like me with many lessons in our own efforts to create nanomachinery. Proteins are large molecules containing hundreds to thousands of atoms and are typically a few nanometers (billionths of a meter) to tens of nanometers across. Our bodies contain at least 20,000 different proteins that, among other things, cause our muscles to contract, digest our food, build our bones, sense our environment and tirelessly recycle hundreds of small molecules within our cells. As a chemistry undergraduate in 1986, I dreamed of the possibility of designing and synthesizing macromolecules (molecules containing more than 100 atoms) that could do the amazing things that proteins do and more. I have programmed computers since the first TRS-80s came out in the late 1970s, and I thought it would be wonderful if I could build complex molecular machines as easily as I could write software. I wanted to create a programming language for matter--a combination of software and chemistry that would enable people to describe a nanomachines shape and would then determine the series of chemical processes that a chemist or a robot should carry out to build the nanodevice. Unfortunately, the idea of inventing nanomachines by designing new proteins runs into a severe obstacle.