[Editor-in-Chief’s Circle]
Science and Technology Daily, Beijing, January 25 (Reporter Zhang Mengran) Chemists at the University of Amsterdam in the Netherlands have developed an autonomous chemical synthesis robot. Dubbed RoboChem, the desktop device features an integrated, artificial intelligence (AI)-driven machine learning unit that outperforms human chemists in speed and accuracy while also demonstrating a high degree of ingenuity sex. As the first of its kind, it dramatically accelerates the discovery of chemical molecules for use in pharmaceuticals and many other fields. The research paper was published in the magazine "Science" on the 25th.
The "Chembot" is a precise and reliable "chemist" that can perform a variety of reactions while producing minimal waste. The system works autonomously around the clock, delivering results quickly and tirelessly. In a week, it can optimize the synthesis of about 10-20 molecules, which would normally take a PhD student several months of work. The robot not only produces optimal reaction conditions but also provides amplification settings.
The research team took advantage of flow chemistry. This is a novel chemical method in which a system of small flexible tubes replaces beakers, flasks and other traditional chemical tools. In the "Chembot", robotic needles carefully collect the starting materials and mix them together in a volume of just over half a milliliter; they then flow through a system of pipes to the reactor; there, light from powerful LEDs activates A photocatalyst in the reaction mixture triggers the molecular conversion; finally, the liquid continues to flow to an automated NMR spectrometer, which identifies the converted molecules.
This data is fed back in real time to the computer controlling the "chemical robot." The "brain" of the "chemical robot" uses AI to process information, and machine learning algorithms autonomously determine which reactions to perform. It always aims for the best results and continually improves its understanding of chemistry.
The researchers used "chemical robots" to replicate previous research published in four randomly selected papers. They then determined whether the "chemical robot" produced the same or better results. The results showed that the system produced better results about 80% of the time; while the other 20% produced similar results. This shows that AI-assisted methods are expected to benefit chemical discovery in the broadest sense.
"Chemical robots" are not just about saving labor and time. In traditional chemical discovery, only a few molecules are studied thoroughly, and the results are then extrapolated to other similar molecules. But Chembot generates a complete and comprehensive data set, capturing all relevant parameters for each molecule. This provides people with more ways to interpret and a broader space for discovery of chemical molecules.