| 8:45:31 PM PDT - Thu, May 23rd 2019 |
|
I don’t have a consistent understanding of how to define range-separated CAM functionals in NWChem. Take, for example, the CAM-B3LYP functional:
xc xcamb88 1.00 lyp 0.81 vwn_5 0.19 hfexch 1.00
cam 0.33 cam_alpha 0.19 cam_beta 0.46
My initial understanding was that the first line defined the “global” weights (so equal parts Becke 88 and HF exchange, 19% VWN5 correlation and 81% LYP correlation), while the second line defined the CAM range separation parameters to be applied on top of the global weights (so an RS parameter of .33, a short range HF fraction of .19, and a long range HF fraction of .65). This all seems consistent with the literature definition of CAM-B3LYP (although I do have two small questions - what is camb88 as opposed to b88? Is it just an implementation of B88 that enables CAM weights to be applied? And what if anything would be different if the “global exchange weights were .5 and .5 instead of 1.0 and 1.0?)
Unfortunately, I can’t really make sense of the definition of other RSHs using this understanding. For example, why is LC-wPBEh defined as
xc xwpbe 0.80 cpbe96 1.0 hfexch 1.00
cam 0.2 cam_alpha 0.20 cam_beta 0.80
? Why is xwpbe weighted at 0.80 instead 1.00? And what is wpbe as opposed to campbe96 or pbe96?
Also, why is HSE03 defined as
xc xpbe96 1.0 xcampbe96 -0.25 cpbe96 1.0 srhfexch 0.25
cam 0.33 cam_alpha 0.0 cam_beta 1.0
? I understand that srhfexch is short range HF, but why is that a separate exchange functional from hfexch? Why isn’t it
xc xcampbe96 1.00 cpbe96 1.00 hfexch 1.00
cam 0.33 cam_alpha 0.25 cam_beta -0.25
?
Any clarification would be much appreciated, thank you!
|