About X3LYP functional

[Igor Ying Zhang]

X3LYP is considered to be one of the best hybrid functionals for
describing the hydrogon bond systems. Today I tried to implement X3LYP
functional in CP2K.

"X" exchange is indeed constructed by mixing the Becke88 and PW91
exchanges with the normalized scale factors of {0.765,0235}.
Obviously, we need a scalable Becke88 and PW91 exchanges. In fact,
CP2K offers us a flag of "SCALE_X" to scale Becke88 exchange. While we
could turn on PW91 exchange using the flag of "FUNCTIONAL = PW91X" in
the group of "XC%XC_FUNCTIONAL%XGGA", It does not offer us a scale
factor like "SCALE_X" to scale it.

If there are any misunderstandings, please let me know. Thanks:)

Thanks

Igor Ying Zhang