A research team from Cornell University in the United States managed to set a new world record by capturing the most accurate picture yet of atoms of crystals of a substance called "praseodymium orthoscandite".
Which opens the door to massive futuristic imaging technologies.
Artificial intelligence
According to the study, which was published in the journal Science and announced in an official press release from the university on May 21, this team has used complex techniques that include the capabilities of modified electron microscopes on the one hand, and the capabilities of artificial intelligence on the other.
Cornell's world record setting electron microscope resolution is getting so fine, the only blurring that remains is the thermal jiggling of the atoms themselves.
https://t.co/cau32b8NPT pic.twitter.com/rFEPx8um6C
— Cornell Engineering (@CornellEng) June 2, 2021
The team was able to develop a smart mechanism called "ptychography" that can assemble the parts of a three-dimensional image and join them together to build an accurate image that contains very little noise, which represents only the thermal vibration of the atoms themselves.
About a century ago, the problems of scattering rays - emanating from imaging devices - on atoms prevented researchers from obtaining accurate images, but this technique used for artificial intelligence is able to track and study a huge amount of scattered rays, and then be able to extract the image of atoms with the best accuracy. possible.
In addition - according to the study - the researchers used a material consisting of heavy atoms, and therefore fluctuates at lower rates, and by cooling the sample, this fluctuation decreases more, but it is known that even at zero temperature the atoms still suffer from quantum fluctuations, and therefore the improvement will not be significant. very.
The problems of scattering rays on the atoms prevented researchers from obtaining accurate images (Pixabay)
Tech race
This is not the first time that the Cornell University team has achieved such a feat;
In 2018, they were able to achieve the first record in this range, but - within 3 years - they were able to develop their techniques to reach double accuracy in 2021, and their magnification capacity reached 100 million times.
According to the study, this achievement may be useful in several areas, such as imaging semiconductor atoms, chemical catalysts, and properties of quantum materials, for the purpose of studying them, and many of these materials are used in quantum computing.
On the other hand, the application of the new imaging method - to biological cells or tissues - can help to achieve a better understanding of several areas in the molecular biology sciences.