"MIT chemists have developed a computational model that can rapidly predict the structure of the transition state of a reaction (left structure), if it is given the structure of a reactant (middle) and product (right). Credit: David W. Kastner" (ScitechDaily, Machine Learning Meets Chemistry: New MIT Model Predicts Transition States With Unprecedented Speed)
Machine learning means. The system makes memos about things that it does. A learning machine is like a man, who writes in a notebook. And then that machine can escalate those notes over the entire system. The learning machine is like some laboratory assistant, who writes everything that is done in a test environment with chemical tests into notebooks. In machine learning, computers make those memos automatically.
In chemistry, that means that the system observes some reactions, and then it puts all the details in the memory. This thing helps the AI-based systems to multiply the chemical and physical environment for the chemical reactions in other laboratories. The AI-based system can transform the test environment conditions straight into full-scale systems and reactions. That thing makes chemical research and development faster than ever before.
Along with things like nano printers and attosecond lasers that thing can make the new type of chemical compounds. The nano printers allow creation the of catalytic layers with accurately adjusted surface area. The attosecond lasers can observe chemical reactions with ultimate accuracy. The attosecond lasers can adjust the energy levels in molecules and even turn the molecular bonds out from the environment. That allows to connect the ions and atoms into the molecule's certain point.
In those systems gravity is a problem. In complex 3D structures, all kinds of disturbances and artifacts are problematic. If the gas mixture or some other things in the reaction chamber is wrong, that is catastrophic. That thing makes automatized orbital laboratories important. Those laboratories are small-sized satellites there remote control system makes molecular structures in the zero-gravity environment. And that thing makes the revolution in chemistry.
"Scientific visualization of the AI-guided assembly of a novel metal-organic framework with high carbon dioxide adsorption capacity and synthesizable linkers. Building blocks, predicted by generative AI, are shown on the left, while the final AI-predicted structure is shown on the right. Credit: Xiaoli Yan/University of Illinois Chicago and the ALCF Visualization & Data Analytics Team" (ScitechDaily, Supercomputers and AI Unlock Secret Materials for Next-Gen Carbon Capture)
The same methodology. That used in complex chemical structures can also used in complex material structures.
The ability to make complex 3D structures makes it possible to create new types of composite materials. The researchers can make a material with an extremely large surface area to clean toxic chemicals and carbon from the air. The box-like structure below the graphene surface can turn material very hard, and resistant to impacts.
The 3D structures that can make the soundwave jump across it can used to make rooms and materials without echoes. That kind of material allows researchers to create a pure acoustic test environment. The graphene layer that is connected with a box structure using nanotubes can used for the new acoustic materials. The nanotubes transport wave movement to a nano-acoustic layer that conducts energy from soundwaves into itself.
In some models, nano springs connect those nano-boxes. Nano springs are the DNA bites. The idea is taken from nuclear-protecting bunkers. They are like boxes that hover in the artificial caves. When energy impulse hits those points the bunker that is hanging on hydraulic pistons in the water layer would survive.
There are tubes around the water layer. They allow the water to expand in those tubes if the pressure or seismic strike hits the ground. The water layer covers the bunker against the seismic impulse. The hydraulic pistons also minimize energy transport to the box. In nanostructures, that thing makes it possible for the material can maximize energy absorption from the pressure impulses. And that minimizes echo from the structure.
https://scitechdaily.com/machine-learning-meets-chemistry-new-mit-model-predicts-transition-states-with-unprecedented-speed/
https://scitechdaily.com/supercomputers-and-ai-unlock-secret-materials-for-next-gen-carbon-capture/
https://learningmachines9.wordpress.com/2024/02/19/machine-learning-meets-chemistry/
Comments
Post a Comment