Issue with designing functionals

Questions
#1

Hi,
I wanted to design user-defined functional, so I thought to start with cam-b3lyp, so I had two input files, which are below

$molecule
0 1
H 6.073950 0.701360 -0.000000
H 7.015820 1.245150 -0.000000
$end

$rem
METHOD CAM-B3LYP
SCF_CONVERGENCE 7
MAX_SCF_CYCLES 100
BASIS TZVP
XC_GRID 000099000590
MEM_TOTAL 2000000
THRESH 14
SYM_IGNORE TRUE
$end

$molecule
0 1
H 6.073950 0.701360 -0.000000
H 7.015820 1.245150 -0.000000
$end

$rem
EXCHANGE gen
LRC_DFT true
OMEGA 330
SCF_CONVERGENCE 7
MAX_SCF_CYCLES 100
BASIS TZVP
XC_GRID 000099000590
MEM_TOTAL 2000000
THRESH 14
SYM_IGNORE TRUE
$end

$xc_functional
C LYP 0.81
C VWN5 0.19
X B88 0.35
X muB88 0.46
K 0.19
$end

the output which confirmed me i am on the right track

Exchange: 0.1900 Hartree-Fock + 0.3500 B88 + 0.4600 muB88 + LR-HF
Correlation: 0.1900 VWN5 + 0.8100 LYP
Using Euler-Maclaurin-Lebedev (99,590) quadrature formula
2) -----------------------------------------------------------------------
Exchange: 0.1900 Hartree-Fock + 0.3500 B88 + 0.4600 muB88 + LR-HF
Correlation: 0.1900 VWN5 + 0.8100 LYP
Using Euler-Maclaurin-Lebedev (99,590) quadrature formula
but then the final results are different, did i miss anything

  1.             Orbital Energies (a.u.)

Alpha MOs
– Occupied –
-0.4398
– Virtual –
0.0331 0.2668 0.4225 1.2687 1.2687 1.4420 1.7779 1.7779
2.2673 2.4329 2.7451

  1.             Orbital Energies (a.u.)

Alpha MOs
– Occupied –
-0.5422
– Virtual –
0.0292 0.2669 0.4242 1.2711 1.2711 1.4468 1.7825 1.7825
2.2730 2.4387 2.7499

#2

CAM-B3LYP is not a simple LRC functional due to its use of two parameters in the range-separation scheme. This is discussed, e.g., here:
https://doi.org/10.1021/acs.jpcc.0c07932

You may be able to recreate CAM-B3LYP using the HF_LR and HF_SR parameters, as described here:
https://manual.q-chem.com/latest/subsec_TDDFT-Xray.html

#3

Thanks, you are right Camb3lyp is differently designed. I am unable to recreate camb3lyp, even with HF_SR and HF_LR.

#4

See Eq. (10) in the JPC paper cited above.

#5

Thanks, this is clear. I will try again.