In the study of subverting material design, researchers have proved through computer simulation that a crystal can be designed and reverse-processed into a self-assembled particle shape. This may lead to the emergence of a new material, such as a crystal coating that never fades. Sharon Glotzer, director of Anthony C. Lembke, Department of Chemical Engineering, University of Michigan, and senior author of Scientific Progress, said: These results have changed the design of materials and our understanding of entropy. Materials with truly new properties are usually discovered by accident. For example, in 2004, an interesting experiment was made with scotch tape and a piece of graphite, and graphene was discovered.

Graphene has become a magical material that won the Nobel Prize because of its strength, flexibility, transparency and conductivity. Materials scientists hope to create a magical material and then find out how to make it, instead of waiting for an unexpected discovery. What this team calls "digital alchemy" is the "reverse" method of designing materials-working backwards from expected characteristics. Greg van Anders, the correspondent of the research paper and an assistant professor of physics at Queen's University in Kingston, Ontario, Canada, said: It really allows us to focus on the research results and use what we know to find the starting point for building this material.

If the particles are organized into a crystal structure, the particle system is the freest. This is the law of physics, and particles have to obey it. According to the shape, the researchers showed how to obtain all kinds of interesting crystals: some are similar to salt crystals or atomic lattices in metals, and some are obviously new (such as "quasi-crystals" without repeated patterns). In the past, this goal was usually achieved by choosing the shape and simulating the crystal it formed. It took several years to study and found the design rules for making particles with specific shapes form specific crystals. Now, it is turned over so that a crystal structure can be inserted into the new program.

This is achieved by changing the question from "What kind of crystal will this shape produce?" Change to give the particle shape to be constructed and recombine it into "Will this shape produce my crystal?" The research team explored more than 6,543.8 billion different shapes in their research. On ordinary computers, there are more different kinds of particles that can be studied in one day than reported in the past ten years. The software is used to identify particle shapes to construct four common crystal lattices (simple cube, body-centered cube, face-centered cube and diamond) and two more complex crystal lattices (-manganese and-tungsten). In doing this, we tried a kind of crystal lattice unknown in nature, which was designed by researchers: a crystal variant called "hexagonal close arrangement".

The team expects that experimental nano-scientists will be able to make these crystals by making a batch of correctly shaped particles and adding them to the liquid. In liquid, nanoparticles will self-assemble. As long as they continue to be restricted, they will maintain this structure. This may lead to the improvement of artificial structure color, just as butterfly wings produce bright colors by interacting with light. Unlike pigments, structural colors do not fade, and this color can be turned on and off by a mechanism, which either restricts the particles from forming crystals or leaves room for them to split crystals.