Incorrect nelec in quantum chemistry-formatted "ccECP-soft" pseudopotentials

[djstaros]

Dear all,

While attempting to use the ccECP-soft pseudopotentials recently reported on the arXiv by @bkincaid256, @aannabe and I found that all of the quantum chemistry formats (gamess, dirac, etc.) list "nelec" as the atomic number instead of the number of core electrons. With chromium as an example,

ecp
Cr nelec 24
Cr ul
1 3.497383 14.000000
3 3.611831 48.963362
2 3.449201 -56.466431
2 2.009794 0.968440
Cr s
2 9.800322 89.846846
2 8.010010 18.997257
Cr p
2 8.785958 44.926062
2 7.014726 14.003861
end

should instead be

ecp
Cr nelec 10
Cr ul
1 3.497383 14.000000
3 3.611831 48.963362
2 3.449201 -56.466431
2 2.009794 0.968440
Cr s
2 9.800322 89.846846
2 8.010010 18.997257
Cr p
2 8.785958 44.926062
2 7.014726 14.003861
end

Best regards,
Dan Staros

[aannabe]

Also, the *.man files should be deleted for these cases, as they are latex-type files only.

[bkincaid256]

@djstaros I have just finished a pull request that should correct the issues with the files as well as remove the unnecessary .man files as suggested by @aannabe.