Industrial need of First-principle Design for manufacturing smaller, more complex devices is getting important. But it is hard to apply this scheme because calculation time and computational cost is still too high. For example, to observe the behavior of million atoms for 1ms, about 150 year calculation is needed. So suggestion of more effective calculation frame is needed.
There are two main theories in many-body quantum simulation: Ab-inito and DFT. Ab-initio is calculating energy of the system by using wave-function and DFT(Density Functional Theory) is using density-function. Ab-inito can exactly describe the natural behavior but it needs too much calculation time and expanse. On the other hand, DFT is more effective to calculate energy but it is not so exact than Ab-initio because we still dont know the exact Exchange and Correlation. So currently researchers focus on developing more exact exchange and correlation function.
Many DFT methods such as LDA, GGA, hybrid suggested nowadays. Well known DFT method is B3LYP which follows hybrid DFT method. This method is quite accurate but has problem with prediction of activation energy. So our laboratory suggested KMLYP DFT method to improve this problem and to implement better DFT scheme, we try to build up basis set free DFT frame to overcome limitation of existing DFT methods.