Technical Support Information
Last update: 22 June 2007

Overlay 3

IOp(3/5)

TYPE OF BASIS SET. The same numbers are used for all basis sets, whether intended for use in expanding AOs (IOp(5)) or in expanding the density (IOp(82)).

0        MINIMAL STO-2G TO STO-6G
1        EXTENDED  4-31G,5-31G,6-31G
2        MINIMAL STO-NG   (VALENCE FUNCTIONS ONLY)
3        EXTENDED LP-N1G (VALENCE BASIS FOR CORELESS
          HARTREE-FOCK PSEUDOPOTENTIALS)
4        EXTENDED 6-311G (UMP2 FROZEN CORE OPTIMIZED)
          BASIS for first row, MacLean-Chandler (12s,9p)-->(631111,52111) for second row.
          USE IOp(8) TO SELECT 5D/6D.
5        SPLIT VALENCE N-21G (OR NN-21G) BASIS FOR FIRST OR SECOND ROW
          ATOMS. (VARIOUS IMPLEMENTATIONS MAY OMIT SECOND ROW ATOMS.)
          SEE IOp(6) FOR DETERMINATION OF THE NUMBER OF GAUSSIANS IN THE INNER SHELL.
6        LANL ECP basis sets. IOp(6) selects options.
7        GENERAL--SEE ROUTINE GenBas FOR INPUT INSTRUCTIONS.
8        Dunning/Caltech basis sets. Type selected by IOp(6).
9        Stevens/Basch/Krauss/Jasien/Cundari ECP basis sets for H-Lu. Type selected by IOp(6)
          for H-Ar. Literature citations in CEPPot.
10      CBS basis #1 -- 6-31+g(d,p) on H, He 6-311+G(2df) on Li - Ne 6-311+g(3d2f) on Na - Ar
11      CBS basis #2 -- 6-31G, use daggers if any polarization
12      CBS basis #3 -- 6-311++G(2df,2p) on H - Ne 6-311++g(3d2f) on Na - Ar
13      CBS basis #4 -- 6-31+G(d,p) on H - Si 6-31+G(df,p) on P, S, Cl
14      CBS basis #5 -- Large APNO basis set
15      CBS basis #6 -- Core correlation basis set
16      Dunning cc basis sets, type selected by IOp(6) (=0-4 for V{D,T,Q,5,6}Z) and augmented
          if IOp(7)=10.

IOp(3/6)=5 for MTsmall basis set.

17      Stuttgart/Dresden ECP basis sets. IOp(6) specifies type. Literature citations in SDDPot.
18      Ahlrichs SV basis sets.
19      Ahlrichs TZV basis sets.
20      MIDI! basis sets.
21      EPR-II basis sets.
22      EPR-III basis sets.
23      UGBS basis set.
24      G3large basis set.
25      G3MP2large basis set.
26      Coreless: Li,Be 2SDF, B-Ne 2MWB, rest LANL1MB.
27      DGauss basis sets, selected by IOp(6)
28      Auto-generated, useful only for density basis sets.
30      One s-gaussian per atom; dummy basis for MM.

IOp(3/6)

NUMBER OF GAUSSIAN FUNCTIONS

N        STO-NG,N-31G,LP-N1G,STO-NG-VALENCE, N-21G.
NOTE IF IOp(5)=3 AND IOp(6)=8 ; LP-31G FOR LI,BE,B,NA,MG,AL LP-41G FOR OTHER
ROW1 AND TWO ATOMS.

DEFAULT OPTIONS IOp(6)=0
IF IOp(5)=0: N=3 STO-3G
IF IOp(5)=1: N=4 4-31G
IF IOp(5)=2: N=3 STO-3G (VALENCE)
IF IOp(5)=3: N=3
IF IOp(5)=5: N=3

WHEN IOp(5)=7 (GENERAL BASES), THIS OPTION IS USED TO CONTROL WHERE THE BASIS IS TAKEN FROM:
0        READ GENERAL BASIS FROM THE INPUT STREAM.
1        READ THE GENERAL BASIS FROM THE RW-FILES AND MERGE WITH THE
          COORDINATES IN BLANK COMMON TO PRODUCE THE CURRENT BASIS.
2        Read the general basis from the checkpoint file.
3        Same as 1, for density basis (generated here from 1)
4        Same as 2, for density basis (generated here from 2)
1x      Read from the alternate file and remove functions/ECPs for inactive atoms. Used for
          counterpoise calculations, where one wants to modify the basis differently during different
          steps. This option is useful when doing general basis geometry optimizations or properties
          using a wavefunction on the checkpoint file. If non-standard ECPs are in use, they are
          read along with the basis set information.     
2x      This option is useful when doing general basis geometry optimizations or properties using a wavefunction on the checkpoint           file. If non-standard ECPs are in use, they are read along with the basis set information.

When IOp(5)=6 (LANL basis and potentials) this selects the type:

0        LANL1 ECP, MBS.

1        LANL1 ECP, DZ.

2        LANL2 ECP (where available, otherwise LANL1), MBS.

3        LANL2 ECP (where available, otherwise LANL1), DZ.

When IOp(5)=8 (Dunning bases) this option selects the type:

0        Dunning full double-zeta.

1        Dunning valence double-zeta.

2        WAG basis (Dunning VDZ on first row, SHC ECP on second row). See Rappe,
          Smedley, and Goddard, J. Phys. Chem. 85, 1662 (1981) and J. Phys. Chem. 85, 3546 (1981).        

When IOp(5)=9 (CEP basis) this option selects the type (H-Ar only):

0        CEP-4G.

1        CEP-31G.

2        CEP-121G.

When IOp(5)=17 (Stuttgart/Dresden ECP bases) this option selects the type according to:

6        SDD

7        SDD for Z > 18, D95 and no ECP otherwise.

When IOp(5)=26 (Coreless basis) this selects the choice of basis (the same ECPs are used regardless):

0        Default (3)

1        Primitives which match the ECPs.

2        Functions from extended Huckel theory.

3        VSTO-4G basis for 1st row, along with LP-31G potential.

>3      Huckel basis for method N-1

When IOp(5)=7 (DGauss basis sets):

1        DGDZVP

2        DZVP2

3        DGTZVP

4        DGA1 (fitting basis)

5        DGA2 (fitting basis)

IOp(3/7)

DIFFUSE AND POLARIZATION FUNCTIONS.

0        NONE.

1        D-FUNCTIONS ON HEAVY ATOMS (2ND ROW ONLY FOR 3-21G).

2        2 D-funcs. on heavy atoms (scaled up/down by a factor of 2 from the standard single D value).

3        ONE SET OF D-FUNCTIONS AND ONE SET OF F-FUNCTIONS ON HEAVY
         ATOMS (indicates an extra tight 2df with ccp basis sets.

4        TWO SETS OF D-FUNCTIONS AND ONE SET OF F-FUNCTIONS ON HEAVY ATOMS.

5        Three sets of d functions.

6        Three sets of d functions and one set of f-functions.

7        Three sets of d functions and two sets of f-functions.

8        CBS-Q d(f),d,p polarization basis

9        Tight d for VnZ+1 (W1 theory)

10        A SET OF DIFFUSE SP FUNCTIONS ON HEAVY ATOMS.

20        Augment non-hydrogens only (cc basis sets only).

100      P-FUNCTIONS ON HYDROGENS.

200      2 SETS OF P-FUNCTIONS ON HYDROGENS.

300      ONE SET OF P-FUNCTIONS AND ONE SET OF D-FUNCTIONS ON HYDROGENS.

400      TWO SETS OF P-FUNCTIONS AND ONE SET OF D-FUNCTIONS ON HYDROGENS.

500      Three sets of p-functions.

600      Three sets of p-functions and one set of d-functions.

700       2d,d,p) -- 2d on 2nd and later atoms, 1d on 1st row atoms.

1000    A DIFFUSE S FUNCTION ON HYDROGENS.

IOp(3/8)

SELECTION OF PURE/CARTESIAN FUNCTIONS.

0        SELECTION DETERMINED BY THE BASIS
          N-31G            6D/7F
          N-311G          5D/7F
          N-21G*          5D
          STO-NG*      5D
          LP-N1G*       5D
          LP-N1G**     5D
          GENERAL BASIS        5D/7F
1        FORCE 5D.
2        FORCE 6D.
10        FORCE 7F.
20        FORCE 10F.

IOp(3/9)

Where 308 should store dipole velocity integrals.

0        Usual place (572).

-1        Write over the dipole length integrals (518).

N        Store in RWF N.

IOp(3/10)

Modification of internally stored bases (default 12000):

0           None.

1           Read in general basis data in addition to setting up a standard basis.

10         Massage the data in Common /B/ and Common /Mol/.

100       Add ghost atoms to /B/ so that every shell is on a separate center.

1000     Split S=P AO basis shells into separate S and P shells.

2000     Do not split S=P AO shells.

10000   Split S=P=D=        AO shells into S=P, D, F,        

20000   Do not split AO S=P=D        shells.

100000 Uncontract the AO basis.

200000  Uncontract the density basis

300000  Uncontract both basis sets.

1000000  Modification 1 for Fermi-contact spin-spin coupling.

2000000  Modification 2 for Fermi-contact spin-spin coupling.

DEFAULTS

STO-NG STANDARD SCALE-FACTORS.

For VSTO-nG, the values for H-Ar can be determined by Slater's rules: H=1.2, He=1.7, Li-Ne=0.325*(IA-1), Na-Ar=(0.65*I-4.95)/3

ATOM        1S        2SP        3SP

H              1.24

HE           1.69

LI             2.69        0.80

BE            3.68        1.15

B              4.68        1.50

C              5.67        1.72

N              6.67        1.95

O              7.66        2.25

F               8.65        2.55

NE            9.64        2.88

NA          10.61        3.48        1.75

MG          11.59        3.90        1.70

AL           12.56        4.36        1.70

SI             13.53        4.83        1.75

P              14.50        5.31        1.90

S              15.47        5.79        2.05

CL           16.43        6.26        2.10

A              17.40        6.74        2.33

INNER SHELLS ARE BEST ATOM VALUES  J.CHEM.PHYS. 38, 2686 (1963) OUTER SHELL HAS BEEN SELECTED ON THE BASIS OF NUMEROUS OPTIMIZATION STUDIES ON VARIED SMALL MOLECULES.

N-31G (ALSO N-31G* AND N-31G**) STANDARD SCALE-FACTORS

HYDROGEN    1S         1S*

H                    1.20        1.15

FIRST ROW ATOMS

ATOM   1S       2SP        2SP*

  B        1.00        1.03        1.12

  C        1.00        1.00        1.04

  N        1.00        0.99        0.98

  O        1.00        0.99        0.98

  F         1.00        1.00        1.00

SECOND ROW ATOMS

ATOM        1S         2SP        3SP        3SP*

     P           1.00        1.00        0.98        1.02

    S            1.00        1.00        0.98        1.01

   CL          1.00        1.00        1.00        1.01

LP-N1G SCALE=1.0 FOR LI-AR (INNER AND OUTER)

STANDARD POLARIZATION EXPONENTS FOR N-31G* AND N-31G** BASES

ATOM        VALUE

H                 1.1

LI                0.2

BE               0.4

B                 0.6

C-NE          0.8

STANDARD POLARIZATION EXPONENTS FOR STO-NG* BASIS.

ATOM        VALUE

NA, MG        0.09

AL-CL           0.39

IOp(3/11)

CONTROL OF TWO-ELECTRON INTEGRAL STORAGE FORMAT.

0        REGULAR INTEGRAL FORMAT IS USED.

1        RAFFENETTI '1' INTEGRAL FORMAT IS USED. CAN ONLY BE USED WITH THE CLOSED SHELL SCF.

2        RAFFENETTI '2' INTEGRAL FORMAT. SUITABLE FOR USE WITH THE OPEN SHELL (UHF) SCF.

3        RAFFENETTI '3' INTEGRAL FORMAT. SUITABLE FOR USE WITH OPEN SHELL RHF SCF AND THE POST-SCF PROCEDURES, but not yet accepted by them.

9        USE ILSW TO DECIDE BETWEEN RAFFENETTI 1 AND 2.

IOp(3/12)

Flag for semi-empirical runs, to account for sparkles, translation vectors and d functions properly:

1        MNDO/AM1.

2        CNDO/2, INDO/2.

3        ZINDO/1, ZINDO/S.

IOp(3/13)

Nuclear center whose Fermi contact terms are to be added to the core hamiltonian. The magnitude is specified by IOp(3/15).

IOp(3/14)

Addition of electrostatic integrals to core hamiltonian.

0        No.

-1x        SCRF calculation -- multiply moments by fudge factor for charged species.

-6        Read coefficients of field, starting with electric field, up through 34 elements (hexadecapoles) in free format, blank terminated.

-5        Read components of electric field only from /Gen/ on checkpoint file.

-4        Read components of moments off rwf 521 on chk file.

-3        Read components of electric field only from /Gen/.

-2        Read components of moments off rwf 521.

-1        Yes, read 12 cards with x,y,z components of electric field, followed by xx,yy,zz,xy,xz,yz electric field gradient, xxx,yyy,zzz,xyy, xxy,xxz,xzz, yzz,yyz,xyz field second derivatives, and xxxx,yyyy,zzzz,xxxy, xxxz, yyyx,yyyz,zzzx,zzzy,xxyy,xxzz,yyzz,xxyz,yyxz, zzxy field third derivatives in format (3D20.10).

(These correspond to dipole, quadrupole, octopole, and hexadecapole perturbations).

1-34        Just component number n in the above order with magnitude given by IOp(3/15).

The nuclear repulsion energy is also modified appropriately, and the electric field is stored in Gen(2-4).

IOp(3/15)

Magnitude of electric field.

N        N * 0.0001.

IOp(3/16)

Pseudopotential option

0        Default. ECPs if defined with the basis set.

1        Yes, read if general basis.

2        No.

IOp(3/17)

SPECIFICATION OF PSEUDOPOTENTIALS

-1        Read potential in old format.

0        Default, based on IOp(3/5).

1        USE INTERNALLY STORED 'CORELESS HARTREE-FOCK'

2        Goddard/Smedley SECE/SHC potentials.

3        Stevens/Basch/Krauss CEP potentials.

4        LANL1 potentials.

5        LANL2 potentials.

6-7     unused

8        READ IN FROM CARDS (SEE PINPUT FOR DETAILS)

9        Dresden/Stuttgart potentials - SDD combination

10      Dresden/Stuttgart potentials - SDD for Z > 18, D95V, no ECP otherwise.

11      Dresden/Stuttgart potentials - SDF

12      Dresden/Stuttgart potentials - SHF

13      Dresden/Stuttgart potentials - MDF

14      Dresden/Stuttgart potentials - MHF (first set)

15      Dresden/Stuttgart potentials - MHF (second set)

16      Dresden/Stuttgart potentials - MWB (first set)

17      Dresden/Stuttgart potentials - MWB (second set)

18      Dresden/Stuttgart potentials - MWB (third set)

19      Pseudopotentials for all coreless basis.

20      Alternative potentials for coreless basis.

IOp(3/18)

PRINTING OF PSEUDOPOTENTIALS

0        PRINT ONLY WHEN INPUT IS FROM CARDS or if GFPrint was specified.

1        PRINT

2        DON'T PRINT

IOp(3/19)

SPECIFICATION OF SUBSTITUTION POTENTIAL TYPE

0        DONT USE ANY SUBSTITUTION POTENTIALS

N       REPLACE THE STANDARD POTENTIAL OF THIS RUN (EG.CHF), WITH A SUBSTITUTION POTENTIAL OF TYPE N WHEREVER SUCH A SUBSTITUTION POTENTIAL EXISTS.

IOp(3/20)

Size of buffers for integral file.

0        Default (Machine dependant; 16384 integer words on VAX, 55296 words on Cray).

N       N integer words.

IOp(3/21)

Size of buffers for integral derivative file.

No longer used.

0        Default (3200 integer words).

N       N integer words.

IOp(3/22)

CONTROL OF THE PRE-CUTOFF IN THE TWO-ELECTRON D-INTEGRAL PROGRAM. Used only in L312.

0        NO PRE-CUTOFF.

1        PRE-CUTOFFS DESIGNED FOR THE 6-31G* BASIS.

IOp(3/23)

Disable use of certain basis functions.

0        Use all basis functions.

1        Read in a list of basis function numbers in Format (10I5), terminated by a blank line, and set their dialgonal core Hamiltonian elements to +100.0.

IOp(3/24)

Printing of gaussian function table.

0        Default (don't print).

1        Print old-fashioned table.

10      Print as GenBas input.

100    Print in more readable format.

1000  Print shell coordinates.

IOp(3/25)

NUMBER OF LAST TWO-ELECTRON INTEGRAL LINK.

-2        Use integrals from a previous job read /IBF/ from the checkpoint file.

-1        We are re-using integrals produced earlier in the current calculation; use the /IBF/ already on the RWF.

0         WE ARE NOT USING TWO-ELECTRON INTEGRALS.

1         Direct SCF.

>0       LINK NUMBER.

IOp(3/26)

ACCURACY OPTION.

0        DEFAULT. INTEGRALS ARE COMPUTED TO 10**-10 ACCURACY.

1        TEST. DO ALL INTEGRALS AS WELL AS POSSIBLE in L311.

2        STO-3G. USE OLD very inaccurate CUTOFFS IN LINK 311.

10      TEST. DO ALL INTEGRALS AS WELL AS POSSIBLE in L314.

20      Sleazy. Use looser cutoffs in L314.

IOp(3/27)

HANDLING OF SMALL TWO-ELECTRON INTEGRALS.

0        DISCARD INTEGRALS WITH MAGNITUDE LESS THAN 10**-10.

N       DISCARD INTEGRALS WITH MAGNITUDE LESS THAN 10**-N.

IOp(3/28)

Special SP code control.

0        Default, use IsAlg.

1        All integrals with d's -- L311 does nothing.

2        SP integrals in link 311, d and higher elsewhere.

IOp(3/29)

Accuracy in L302:

0        Default (10**-12).

N       10**-N.

IOp(3/30)

CONTROL OF TWO-ELECTRON INTEGRAL SYMMETRY.

0        TWO-ELECTRON INTEGRAL SYMMETRY IS TURNED OFF.

1        TWO-ELECTRON INTEGRAL SYMMETRY IS TURNED ON. NOTE, HOWEVER, THE SET2E WILL INTERROGATE ILSW TO SEE IF THE SYMMETRY RW-FILES EXIST. IF THEY DON'T, SYMMETRY HAS BEEN TURNED OFF ELSEWHERE, AND SET2E WILL ALSO TURN IT OFF HERE.

IOp(3/31)

USE OF SYMMETRY IN COMPUTING GRADIENT (Obsolete).

IOp(3/32)

Whether to check the eigenvalues of the overlap matrix:

0        Default (4).

1        Yes.

2        No.

3        Yes, and reduce expansion space if linear dependence is found (NYI).

4        Yes, and use Schmidt orthogonalization to reduce expansion space.

5        Yes, using SVD to reduce expansion space.

IOp(3/33)

INTEGRAL PACKAGE PRINTING.

0        NO INTEGRALS ARE PRINTED.

1        PRINT ONE-ELECTRON INTEGRALS.

3        PRINT TWO-ELECTRON INTEGRALS IN STANDARD FORMAT.

4        PRINT TWO-ELECTRON INTEGRALS IN DEBUG FORMAT.

5        COMBINATION OF 1 AND 3.

6        COMBINATION OF 1 AND 4.

IOp(3/34)

DUMP OPTION.

0        NO DUMP.

1        CONTROL WORDS PRINTED (AS USUAL).

2        ADDITIONALLY, COMMON/B/ IS DUMPED AT THE BEGINNING OF EACH INTEGRAL LINK.

3        ADDITIONALLY, THE INTEGRALS ARE PRINTED (STANDARD FORMAT).

IOp(3/35)

LAST INTEGRAL DERIVATIVE LINK (No longer used in overlay 3).

0        WHATEVER LINK STARTS WRITING THE INTEGRAL DERIVATIVE FILE SHOULD ALSO CLOSE IT.

N        IS THE NUMBER OF THE LAST TWO-ELECTRON INTEGRAL DERIVATIVE PROGRAM.

IOp(3/36)

Maximum order of multipoles to compute in L303:

-1       None

0        Default (dipole).

1        Dipole.

2        Quadrupole.

3        Octopole.

4        Hexadecapole.

00      Default (same as 20).

10      Do not compute absolute overlaps.

20      Compute absolute overlap over contracted functions.

30      Compute absolute overlap over both contracted and over primitive functions.

IOp(3/37)

Whether to sort integrals in L320.

0        Default (No).

1        Yes.

2        No.

IOp(3/38)

Algorithm for 1e integrals:

0        Default in 302, same as 1.

1        PRISM.

2        Rys.

00      Default in 308, same as 1.

10      PRISM.

20      Explicit spdf code.

IOp(3/39)

Initialization of force and force constant rwfs.

0        Initialize.

1        Leave alone.

IOp(3/40)

Neglect of integrals:

0        keep all integrals.

1        neglect four center integrals.

2        neglect three center two-electron integrals as well.

3        neglect 2e integrals with diatomic differential overlap.

10      neglect three center one-electron integrals.

20      neglect 1e integrals with diatomic differential overlap.

30      Do only overlap and not other 1e integrals.

IOp(3/41)

Various semi-empirical methods.

0        No NDDO

1        NDDO

00      Default use of NDDO beta parameters (arithmetic mean for indo parameters, geometric mean for NDDO/1 or read-in parameters).

10        Arithmetic mean in NDDO.

20        Geometric mean in NDDO.

000       Default parameters (same as 5).

100        Read parameters for atomic numbers 1-18 in the order Scale (D20.12), followed by
             ((HDiag(J,I),J=1,3),I=1,18) (Format 3D20.12), followed by ((Beta(J,I),J=1,3),I=1,18)

200        Read parameters from rwf.

300        Read parameters from chk.

400        Original INDO/2 Beta and HDiag Parameters.

500        GNDDO/1 parametrization.

0000       Use STO-3G scale factors.

1000       Use Slater's rules scale factors.

00000     Default (unit overlap matrix).

10000     Use the unit matrix for the overlap.

20000     Use the real overlap matrix.

100000   Do CNDO/2.

200000   Do INDO/2.

300000   Do ZINDO/1.

400000   Do ZINDO/S.

1000000 Do Harris functional.

1100000 Do Harris functional scaling atomic densities for current charge and multiplicity.

1200000 First-order XC.

1300000 Second-order XC (NYI).

1400000 Regular SCF with separate K, for testing.

1500000 J as usual but NDDO for K.

IOp(3/42)

How to form NDDO core hamiltonian in L317:

0        Default (same as 1).

1        Read the integrals sequentially.

2        Load all the integrals into memory.

IOp(3/43)

Solvent charge distribution to add to Hamiltonian:

0        None

1        Read charges and DBFs from input stream in input orientation

2        Read from RWF.

3        Read from Chk.

4        Same as 1.

5        Read charges and DBFs from input stream in standard orientation

10      Force units of Angstroms for coordinates.

20      Force units of Bohr for coordinates.

If negative, the perturbation is computed separately and stored in the third and fourth matrices in the core Hamiltonian rwf.

IOp(3/44)

integral rejection using L318.

0        keep all integrals.

1        neglect four center transformed integrals.

2        neglect four center and 3 center (ab|ac) integrals.

3        neglect four center and three center (0,0) integrals.

4        NDDO approximation -- no (ab|xx) and no <a|X|b>

5        NDDO on 2e and V ints only -- T and S unchanged.

6        Do not transform 2e integrals, only 1e.

IOp(3/45)

Transformation matrix in L318.

0        use S**-1/2.

1        just orthogonalize functions on the same center.

2        Use unit matrix (for debugging).

Order of multipoles in SCRF for L303.

IOp(3/46)

Whether to abort the job if badbas detects an error:

0        Default (yes).

1        no.

2        yes

IOp(3/47)

Flags for use in PRISM and CalDFT throughout the program.

-1        Force use of only the OS path for all calculations.

Bit flags:

0        If bit 0 is set (use AllowP array) then read in a list of allowed paths.

1        Use expanded matrix logic for PBC exact exchange.

2        Reverse choice of whether to precompute distance matrix during numerical quadrature.

3        Skip consistency checks for XC quadrature

4        Do not do extra work to use cutoffs better, currently only affects CalDFT.

5        Reverse normal choice of diagonal/canonical sampling in Prism and PrmRaf. The default is
         diagonal only on vector machines.

6        Trace input and output using Linda/subprocess.

7        Force single matrix code in CPKS.

8        Force all near field in FMM.

9        Turn off dynamic allocation of parallel work in CalDSu and CoulSu.

10       Force square loops, currently only in PrismC.

11       Force use of FoFCou, even if not doing FMM.

12       Reverse normal choice of Scat20 vs. replicated Fock matrices. Default is to use replicated matrices only on Fujitsu and NEC.

13        Turn off Schwartz during FMM/NFx calculations.

14        Turn off MP-based cutoffs in FF part of NFx.

15        Forbid use of gather/scatter digestion even for small numbers of density matrices.

16        Reserved for more control of scatter/gather.

17        Turn on angular offsets in XC grid generation.

18        Use Mura radial grid instead of Euler-2 grid.

19        Do nuclear contribution in FoFCou even for non-PBC

20        Do not use special Coulomb algorithm in FoFCou.

21        Forbid use of FoFCou.

22        Turn off use of Sqrt(P) in density-based cutoffs.

23        No longer used.

24        Do allocation for parallel 2e integrals but run sequentially.

25        Do allocation for parallel XC but run sequentially.

26        Make all atoms large in XC quadrature.

27        Make all shells large in XC quadrature.

28        Do not symmetry reduce grid points on unique atoms.

29        Turn on use of precomputed XC weights.

IOp(3/48)

Options for FMM:

RRLLNNTTWW        

RR:        Range (default 2)

LL:        LMax (default from tolerance)

NN:       Number of levels (default 8)

TT:         Tolerance (default 18)

WW:      IWS (default 2).

IOp(3/49)

More options for FMM:

1        Indicates whether FMM can be used by FoFCou.

2        Uncontract all shell pairs.

4        Apply symmetry to derivative distributions (NYI).

8        Do not save as many multipole expansions as possible in memory.

16      Turn on FMM print.

32      Convert to sparse storage under FoFCou for testing.

64      Split primitives for better boxification.

128    Default UseUAB/Use 256.

256    UseUAB, if 128 set.

512    Turn off parallelism in FMM (does not use parallel logic).

1024  Set up for parallel FMM but run loops sequentially.

2048  Do not default to FMM.

4096  Force FMM on.

8192  Set by PsmSet to indicate whether the NAtoms test for defaulting FMM was passed.

IOp(3/51)

Parameters for NF exchange and box length (MMMMNN):

00                no NFx

NN              NFx range NN (R+n with n=NN-1).

0000xx        Default box length, based on geometry (but minimum for molecules 3.0 Bohr).

MMMMxx   Box length MMMM/1000 Bohr.

IOp(3/52)

Turn off normal evaluation of ECP integrals.

0        Default: if needed, ECP integrals are evaulated in L302.

1        Old routines will be used, so L302 does not do ECP ints.

IOp(3/53)

Accuracy in ECP integral evaluation:

0          Default.

-1        No Cutoffs

N        10**-N.

IOp(3/54)

Type of core density to use with ECPs:

-1      None

0        Default (None)

1        Non-relativistic

2        Relativistic

IOp(3/55)

Use of sparse storage:

N<-100        Yes, cutoff 5 x 10 ** (N+100)

-3        Yes, intermediate accuracy (5x10**-7)

-2        Yes, crude accuracy (5x10**-5)

-1        Yes, default accuracy (10**-10).

0          No

N        Yes, cutoff 10**(-N)

IOp(3/56)

Cutoff for intermediate matrices during sparse operations:

0         100 times smaller than storage cutoff.

N        10**(-N).

IOp(3/57)

No. of core electrons for Stuttgart/Dresden ECP's.

IOp(3/58)

Cholesky control options.

IOp(3/59)

Threshold for throwing avay eigenvectors of S:

0         Default (10**-6)

N        10**-N.

IOp(3/60)

Control of orthogonalization and simplification of ccp basis sets.

0        Default (1).

1        Orthogonalize and remove primitives with 0 coefficients.

2        Orthogonalize and remove primitives with 0 or small coefficients.

IOp(3/61)

Sparse Semiempirical Hamiltonian Cutoffs in L302:

XX        F(Mu,Lambda) atom--atom cutoff criterion (angstroms). Mu, Lambda are basis functions
             on different atoms. (Defaults to 15 angstroms).

XX00    F(Mu,Nu) atom--atom cutoff criterion (angstroms) Mu, Nu are basis functions on the
             same atom. (Defaults to no F(Mu,Nu) cutoff).

IOp(3/62)

Maximum allowed error in S over orthogonalized basis functions:

0          Default (10**-9).

N        10**-(N).

IOp(3/63)

Debug option to test point charge FMM.

0        No.

1        Yes.

IOp(3/64)

Set value for ILSW derivative flag. Only active if IOp(3/39)=0.

-2        Set to zero

-1        Set to -1.

0         Leave alone.

N        set to N.

IOp(3/65)

Number of k-points:

-1        Just Gamma point.

N        About N points.

-N       Old logic for NRecip=N.

IOp(3/66)

Over-ride setting of NThInc in lineary dependence cutoff:

-1       0

0        Don't change.

N       Set to N.

IOp(3/67)

Electric-field dependent functions:

0        Default (on if already present in basis read from rwf or chk, otherwise off).

1        No.

2        Yes, with standard values.

3        Yes, with read-in values.

IOp(3/70)

SCRF flag.

0           Default (1)

1           Use defaults.

2           Read setting from checkpoint.

3           Read setting from the input stream.

4           Read setting from checkpoint and modify them by reading from the input stream.

5           Read from rwf.

0100     Flag for macroiterations (IPCM).

1000     SCI-PCM.

2000     D-PCM.

2100     C-PCM.

2200     IEF-PCM.

2300     IVC-PCM

4000     Onsager.

IOp(3/71)

IDeriv level flag (for SCRF setup).

IOp(3/72)

Solvent type flag (for SCRF setup).

IOp(3/73)

ONIOM system flag (for SCRF setup).

IOp(3/74)

Type of exchange and correlation potentials:

-34 BMK.
-33 X3LYP
-32 t-HCTH hybrid
-31 t-HCTH
-30 OmPW3PBE.
-29 OmPW1PBE.
-28 OmPW1LYP.
-27 OmPW1PW91.
-24        O3LYP.
-23        HCTH407.
-22        HCTH147.
-21        B97-2.
-20        B97-1.
-19        HCTH93.
-18        B98.
-17        B1B95.
-16        BA3PBE.
-15        BA1PBE.
-14        PBE3PBE.
-13        PBE1PBE
-12        mPW3PBE.
-11        mPW1PBE.
-10        mPW1LYP.
-9          LG1LYP.
-8          B1LYP.
-7          mPW91PW91.
-6          Becke3 with Perdew 91 correlation.
-5          Becke3 using VWN/LYP for correlation.
-4          Becke 3 with Perdew 86 correlation.
-3          Becke "Half and Half" with LYP/VWN correlation.
-2          Becke "Half and Half": 0.5 HF + 0.5 LSD
-1          Do only coulomb part; skip exchange-correlation.
00         Default, same as 100.
00         No correlation.
01         Vosko-Wilk-Nusair method 5 correlation.
02         Lee-Yang-Parr correlation.
03         Perdew 81 correlation.
04         Perdew 81 + Perdew 86 correlation.
05         VWN 80 (LSD) correlation
06         VWN 80 (LSD) + Perdew 86 correlation
07         OS1 correlation
08         PW91
09         PBE
10         VSXC
11         Bc96
18         VWN5+P86
19         LYP+VWN5 for scaling
20         KCIS
100       Hartree-Fock exchange.
200       Hartree-Fock-Slater exchange (Alpha = 2/3).
300       X-alpha exchange (alpha= 0.7)
400       Becke 1988 exchange.
500       LG exchange.
600       PW91 exchange
700       Gill 96 exchange
800       PW86 exchange
900       mPW exchange
1000     PBE exchange
1100     BA exchange
1200     VSXC exchange
1400 B98 (JCP 108,9624(1998) eq.2c ) exchange
1500 HCTH (JCP 109,6264 (1998) exchange
1600 B97-1 (CPL 316,160(2000)) exchange
1700 B97-2 (JCP 115,9233(2001)) exchange
1800 HCTH147 exchange
1900 HCTH407 exchange
2000 OPTX exchange
2100 OPTX exchange as in O3LYP
2800 Old mPW exchange (local scaling in non-local term)

So 100 is Hartree-Fock, 200 is Hartree-Fock-Slater, 205 is Local Spin Density, and 402 is BLYP.

IOp(3/75)

Number of radial and angular points in numerical integration for DFT:

0              Use a special grid designed for efficiency (default).

IIIJJJ        III radial points, JJJ angular points.

IOp(3/76)

Mixing of HF and DFT.

-13      B1B95 coefficients.

-10      O3LYP coefficients.

-9        B97-2 coefficients.

-8        B97-1 coefficients.

-7        HCTH coefficints.

-6        B98 coefficients.

-5        mPW91PW91 coefficients.

-4        Becke 3 coefficients: aLSD + (1-a)HF + b(dBx) + VWN + c(LYP-VWN),
           with a=0.8 b=0.72 c=0.81. Note that Becke actually used Perdew correlation rather than LYP.

-3        Becke "Half and Half" 0.5 HF + 0.5 Xc + Corr

-2        Coefficients of 0 and 0 (no exchange).

-1        Coefficients of 0.0 and 1.0 for DFT and HF, respectively.

0          Default: pure HF, DFT or mixed in accord with IOp(3/76)

MMMMMNNNNN         Mixture of MMMMM/10000 DFT exchange and NNNNN/10000 HF exchange.

IOp(3/77)

Mixing of local and non-local exchange:

-1       0 for both.

0        Default (coefficients of 1 and zero as determined by IOp(3/42)

MMMMMNNNNN         MMMMM/10000 non-local plus NNNNN/10000 local. Sign is applied to the local term.

IOp(3/78)

Mixing of local and non-local correlation:

-1        0 for both.

0         Default (coefficients of 1 and zero as determined by IOp(3/42)

MMMMMNNNNN         MMMMM/10000 non-local plus NNNNN/10000 local. Sign is applied to the local term.

In L510, 1 to set up for CAS-MP2 or 2 to do spin-orbit calculation.

IOp(3/79)

Range cutoff in Becke weights.

0        Default (SS weights)

-1       Use SS weights.

-2       Use Becke weights with default cutoff of 30 au.

-3       Use Savin weights.

-M<-3  Use SS weights with XCal = M/1000.

N        Use Becke weights with cutoff N Bohr.

IOp(3/80)

Range for microbatching in DFT. Negative to turn off screening of basis functions and grid points. 1000000000 turns of microbatching logic.

IOp(3/81)

Frequency-dependence (if any) for XC functional.

0        Default (same as 1).

1        None (static limit).

2        Also static limit, but returning zero for imaginary response contributions, for debugging.

3        Gross-Kohn form.

IOp(3/82)

Fitting density basis set for Coulomb in DFT.

-1        None.

0         Default (-1).

N        Same numbering of basis sets as for AO basis, including

7=General basis. See comments for IOp(3/5) and IOp(3/6)

28=Generate automatically from AO basis.

IOp(3/83)

Equivalent of IOp(3/6) for density basis. For auto-generated basis sets:

-1       Keep all generated functions.

0         Keep all functions with angular momentum up to MaxTyp+1, where MaxTyp is the highest AO angular momentum.

N        Discard functions with L>=N.

IOp(3/84)

Equivalent of IOp(3/7) for density basis. For auto-generated basis sets:

0        Default (22)

1        Use all products of AOs.

2        Use only AO primitives squared in fititing basis.

10      Do not split shells

20      Split F and higher shells away from S=P=D.

IOp(3/85)

Pure vs. Cartesian functions in density basis.

0        Default (pure for read-in basis).

1        Pure.

2        Cartesian.

IOp(3/86)

Discard basis functions based on angular momentum:

0        No.

N       Discard basis functions with angular momentum >= N.

IOp(3/87)

Discard density basis functions based on angular momentum:

0         No.

N        Discard density basis functions with angular momentum >= N.

IOp(3/88)

Modification of internally stored density basis.

0          None.

1          Read in general basis data in addition to setting up a standard basis.

10        Massage the data in Common /B/ and Common /Mol/.

100      Add ghost atoms to /B/ so that every shell is on a separate center. This is also done if requested in IOp(3/10).

1000    Split S=P density basis shells into separate S and P shells.

2000    Do not split S=P density shells.

10000  Split S=P=D=... density shells into S=P, D, F, ...

20000  Do not split density S=P=D... shells.

IOp(3/89)

Set up for density fitting.

0        Default (102 if a fitting set has been included and pure DFT is being used, 1 otherwise).

1        Do not use density fits.

2        Use fits, forming Z = modified A^-1.

3        Use fits, solving iterative with stored A.

4        Use fits, solving iterative with direct products, with A formed to generate preconditioning.

5        Iterative, no formation of A.

6        Form A' over neutral distributions via multiplies by A.

7        Form A' over neutral distributions via direct products.

1xx     Form inverse matrix once.

2xx     Solve iteratively with no preconditioning

3xx     Solve iteratively with diagonal preconditioning.

4xx     Solve iteratively with symmetric block-diagonal preconditioning.

5xx     Solve iteratively with non-symmetric block-diagonal preconditioning.

6xx     Solve non-iterative using precomputed A'^-1.

1xxx    Put all functions into a single block in forming the preconditioning matrix.

IOp(3/90)

Thresholds for density fitting

MMNN        

10**(-MM) on iterative solution, default MM=09.

10**(-NN) on generalized inverse, default NN=06.

IOp(3/91)

Scalar relativistic core Hamiltonian:

0        Default (1)

1        Non-relativistic.

2        RESC.

3        Douglass-Kroll-Hess 0th order.

4        Douglass-Kroll-Hess 2nd order.

IOp(3/92)

Whether read-in basis sets are in terms of normalized primitives?

0        Default (12).

1        AO coefficients are for raw primitives.

2        AOs have normalization as for AOs.

3        AOs have J-normalization.

10      DBF coefficients are for raw primitives.

20      DBFs have normalization as for AOs.

30      DBFs have J-normalization.

IOp(3/93)

Nuclear charge distribution:

0        Default (1, unless scalar relativistic)

1        Point nuclei

2        Single s Gaussians using formula of Quiney et. al

3        Very tight single s Gaussians, for debugging.

4        Same as 2 but exponents are 100x smaller, for debugging.

10x    Include nuclear charge distributions in DBF set.

Mxxx Use method M to handle nuclear charges during density fitting.

IOp(3/94)

Range of PBC cells in Bohr.

0        default (100).

N        N Bohr.

-M      Multiply usual range by M.

IOp(3/95)

Minimum number of PBC cells.

-N       At least N cells in each direction.

0          Based on range estimate (IOp(3/94)).

N        At least N cells total.

IOp(3/96)

Number of PBC cells for DFT:

0         As many as look significant.

N        At least N.

IOp(3/97)

Number of PBC cells for exact exchange:

0         As many as look significant.

N        At least N.

IOp(3/98)

Maximum number of density matrices in PBC.

0        Default, based on number of cells having overlap with cell 0.

N       No more than N matrices.

IOp(3/99)

Whether to set up precomputed quadrature grid in L302:

0         Default (2 if doing DFT, -1 otherwise).

-1       No

1        Yes, storing only grid parameter

2        Yes, storing grid parameters and weights.

3        Yes, storing grid parameters, weights, and point coordinates.

IOp(3/100)

Minimum Number of PBC cells for PBC-MP2

0         Same as for HF exchange.

N        N.

IOp(3/101)

Maximum range of cells

-N      No more than N in each direction

0        No limit.

N       No more than N total.

IOp(3/102)

Number of density fittings solutions to save from previous SCF iterations. Default is 6 (using 5 previous solutions plus the current right-hand side to generate the initial guess). Negative to use projected equations rather than least-squares.

IOp(3/103)

Maximum number of vectors allowed in expansion space during iterative density fitting. Default is Max(NDBF/2,1000).

IOp(3/104)

Maximum number of iterations during iterative density fitting. Default is Max((1000,NDBF+100).

IOp(3/105)

Re-use of PBC cell data.

0        Default (re-use if present).

1        Reuse.

2        Do not reuse.

3        Read from chk file.

IOp(3/106)

Override default number of atoms threshold for turning on FMM (for debugging). This number is scaled up appropriately if symmetry is in use, to compensate for the loss of some symmetry with FMM.

0         Default (80)

N        N atoms for the C1 case.

IOp(3/113)

Generate SABF data.

00      Default (12).
1        Generate AO basis function SABF data if symmetry is on.
2        Make AO SABF data C1 regardless.
10      Generate density basis function SABF data if symmetry is on.
20      Make density basis SABF data C1 regardless.