Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
The combination of techniques such as machine learning, artificial intelligence, robotics and automation can be used to accelerate chemical and materials synthesis. This Focus issue showcases developments in the automation and digitalization of synthesis, as well as highlights the challenges to be overcome in this area.
In this issue, we focus on the combination of techniques such as machine learning, artificial intelligence, robotics and automation for the synthesis of chemicals and materials.
King Kuok (Mimi) Hii, a professor at Imperial College London and director of the Centre for Rapid Online Analysis of Reactions (ROAR) and the Engineering and Physical Sciences Research Council (EPSRC) Centre of Doctoral Training in Next Generation Synthesis & Reaction Technology (rEaCt), talks to Nature Synthesis about reaction monitoring in automated workflows as well as the challenges to be overcome in automated synthesis.
Andrew Cooper, a professor at the University of Liverpool and Academic Director of the Materials Innovation Factory, talks to Nature Synthesis about the use of robotics and artificial intelligence for the synthesis and discovery of materials and chemicals.
Automated experiments with integrated characterization techniques greatly accelerate materials synthesis and provide data to be used by machine learning algorithms. We reflect on the current use of data-driven automated experimentation in materials synthesis and consider the future of this approach.
Automation and real-time reaction monitoring have enabled data-rich experimentation, which is critically important in navigating the complexities of chemical synthesis. Linking real-time analysis with machine learning and artificial intelligence tools provides the opportunity to accelerate the identification of optimal reaction conditions and facilitate error-free autonomous synthesis. This Comment provides a viewpoint underscoring the growing significance of data-rich experiments and interdisciplinary approaches in driving future progress in synthetic chemistry.
Self-driving labs (SDLs) combine machine learning with automated experimental platforms, enabling rapid exploration of the chemical space and accelerating the pace of materials and molecular discovery. In this Review, the application of SDLs, their limitations and future opportunities are discussed, and a roadmap is provided for their implementation by non-expert scientists.
Combinatorial synthesis has historically been the cornerstone of high-throughput experimentation. In this Review, we discuss the evolution of combinatorial synthesis and envision a future for accelerated materials science through its integration with artificial intelligence. We also evaluate the key aspects of combinatorial synthesis with respect to workflow design.
Nature evolves proteins by iterating through an untold number of mutations over time. Now, a method is reported to prepare and optimize synthetic polypeptides in an automated high-throughput fashion driven by artificial intelligence.
High-throughput synthesis of polypeptides through ring-opening polymerization of N-carboxyanhydride is challenging. Now a diversification approach is developed based on the post-polymerization modification of a selenium-containing polypeptide. With the assistance of automation and model-guided optimization, this approach enables the discovery of functional polypeptides from chemical space with little previous knowledge.
Trial-and-error synthesis and labour-intensive characterization procedures hinder the development of nanocrystals. Now, a data-driven robotic synthesis approach is used to prepare gold and double-perovskite nanocrystals. This approach combines data mining of synthesis parameters, robot-assisted synthesis and characterization, and machine-learning-facilitated inverse design of the nanocrystals.
Automated organic synthesis is often limited to making simple molecules, requiring a small number of synthetic steps, because of the complexity and variety of organic molecules. Now, a robotic platform has been instructed to build complex structures, such as the core fragment of (+)-kalkitoxin, in a stereochemically controlled and iterative manner.
The complexity of carbohydrate structures makes their synthesis challenging. Now, an automated glycan synthesizer is reported which is capable of preparing a library of bioactive oligosaccharides, including a fully protected fondaparinux pentasaccharide. Furthermore, the synthesizer can rapidly assemble arabinans up to 1,080-mer size, starting from monosaccharide building blocks.
Iterative synthesis can generalize, automate and democratize the molecule-making process. Now, by using a computer algorithm to scan the depths of chemical reactivity space, thousands of iterative ways to make small molecules are discovered.
Iterative sequences of organic reactions can be automated but are rare and challenging to identify. Now, a computer-driven strategy is reported for the systematic discovery and evaluation of such sequences. Several of the iterative sequences are validated experimentally and enable the syntheses of useful motifs in natural product targets.