13.5.4 Job Control for the NEO methods

NEO
       Enable a NEO-SCF calculation.
TYPE:
       BOOLEAN
DEFAULT:
       FALSE
OPTIONS:
       TRUE Enable a NEO-SCF calculation. FALSE Disable a NEO-SCF calculation.
RECOMMENDATION:
       Set to TRUE if desired.

JOBTYPE
       Specifies the type of calculation.
TYPE:
       STRING
DEFAULT:
       SP
OPTIONS:
       SP Single point energy. FORCE Analytical force calculation. OPT Geometry minimization. TS Transition structure search. FREQ Frequency calculation.
RECOMMENDATION:
       Application-dependent. Always use POINT_GROUP_SYMMETRY = FALSE with geometry optimization.

METHOD
       Specifies the exchange-correlation functional.
TYPE:
       STRING
DEFAULT:
       No default
OPTIONS:
       NAME Use METHOD = NAME, where NAME is one of the following: HF for Hartree-Fock theory; one of the DFT methods listed in Section 5.3.5.;
RECOMMENDATION:
       In general, consult the literature to guide your selection. Our recommendations for DFT are indicated in bold in Section 5.3.5.

BASIS
       Specifies the electronic basis sets to be used.
TYPE:
       STRING
DEFAULT:
       No default basis set
OPTIONS:
       General, Gen User defined ($basis keyword required). Symbol Use standard basis sets as per Chapter 8. Mixed Use a mixture of basis sets (see Chapter 8).
RECOMMENDATION:
       Consult literature and reviews to aid your selection.

BASIS_LIN_DEP_THRESH
       Sets the threshold for determining linear dependence in the basis set
TYPE:
       INTEGER
DEFAULT:
       6 Corresponding to a threshold of ${10}^{-6}$
OPTIONS:
       $n$ Sets the threshold to ${10}^{-n}$
RECOMMENDATION:
       Set to 5 or smaller if you have a poorly behaved SCF and you suspect linear dependence in you basis set. Lower values (larger thresholds) may affect the accuracy of the calculation.

NEO_BASIS_LIN_DEP_THRESH
       This keyword is used to set the liner dependency threshold for nuclear basis sets. It is defined as ${10}^{-\mathrm{NEO}\mathrm{\_}\mathrm{BASIS}\mathrm{\_}\mathrm{LIN}\mathrm{\_}\mathrm{DEP}\mathrm{\_}\mathrm{THRESH}}$.
TYPE:
       DOUBLE
DEFAULT:
       5.0
OPTIONS:
       User-defined
RECOMMENDATION:
       No recommendation.

PURECART
       INTEGER
TYPE:
       Controls the use of pure (spherical harmonic) or Cartesian angular forms
DEFAULT:
       1111 Pure $h,g,f,d$ functions
OPTIONS:
       $hgfd$ Use 1 for pure and 2 for Cartesian.
RECOMMENDATION:
       This is pre-defined for all standard basis sets

NEO_PURECART
       This keyword is used to specify Cartesian or spherical Gaussians for nuclear basis functions.
TYPE:
       INTEGER
DEFAULT:
       2222
OPTIONS:
       User-defined
RECOMMENDATION:
       The default value corresponds to the use of Cartesian Gaussians for all angular momentum classes. The value NEO_PURECART = 1111 would use spherical Gaussians instead, similar to the use of PURECART.

NEO_ISOTOPE
       Enable calculations of different types of isotopes. Only one type of isotope is allowed at present.
TYPE:
       INTEGER
DEFAULT:
       1 Default is the proton isotope.
OPTIONS:
       1 This NEO calculation is using proton isotope. 2 This NEO calculation is using deuterium isotope. 3 This NEO calculation is using tritium isotope.
RECOMMENDATION:
       Refer to the NEO literature for the best performance on the isotope effects calculations.

NEO_VPP
       This keyword is used to control whether to remove $J-K$ terms from the nuclear Fock matrix and the corresponding kernel terms for NEO excited state methods. The purpose for the removal of these terms in the case of one quantum proton is to save on computational cost. Note that the ground state NEO-HF or NEO-DFT energy is the same with or without removal of these nuclear $J-K$ terms. The nuclear virtual orbitals will however be different, which may affect the results for perturbative methods (i.e., multicomponent excited-state methods and post-NEO-HF wavefunction methods).
TYPE:
       LOGICAL/INTEGER
DEFAULT:
       TRUE
OPTIONS:
       TRUE (or 1) Enable this option (include nuclear $J-K$ terms). FALSE (or 0) Disable this option (remove nuclear $J-K$ terms).
RECOMMENDATION:
       Set NEO_VPP = 0 only in the case of one quantum hydrogen.

XC_GRID
       Specifies the type of grid to use for DFT calculations.
TYPE:
       INTEGER
DEFAULT:
       Functional-dependent; see Table 5.3.
OPTIONS:
       0 Use SG-0 for H, C, N, and O; SG-1 for all other atoms. $n$ Use SG-$n$ for all atoms, $n=1,2$, or 3 $XY$ A string of two six-digit integers $X$ and $Y$, where $X$ is the number of radial points and $Y$ is the number of angular points where possible numbers of Lebedev angular points, which must be an allowed value from Table 5.2 in Section 5.5. $-XY$ Similar format for Gauss-Legendre grids, with the six-digit integer $X$ corresponding to the number of radial points and the six-digit integer $Y$ providing the number of Gauss-Legendre angular points, $Y=2N^2$.
RECOMMENDATION:
       We strongly recommend using an unpruned, Euler-Maclaurin-Lebedev grid type for NEO-DFT calculations. Larger grids may be required for optimization and frequency calculations.

UNRESTRICTED
       Controls the use of restricted or unrestricted orbitals.
TYPE:
       LOGICAL
DEFAULT:
       FALSE Closed-shell systems. TRUE Open-shell systems.
OPTIONS:
       FALSE Constrain the spatial part of the alpha and beta orbitals to be the same. TRUE Do not Constrain the spatial part of the alpha and beta orbitals.
RECOMMENDATION:
       The ROHF method is not available. Note that for unrestricted calculations on systems with an even number of electrons it is usually necessary to break $\alpha$/$\beta$ symmetry in the initial guess, by using SCF_GUESS_MIX or providing $occupied information (see Section 4.4 on initial guesses).

S2THRESH
       Cutoff for neglect of overlap integrals, defined via a two-electron shell-pair threshold of ${10}^{-\mathrm{S2THRESH}}$ (S2THRESH $\le 14$).
TYPE:
       INTEGER
DEFAULT:
       Same as THRESH.
OPTIONS:
       $n$ for a threshold of ${10}^{-n}$.
RECOMMENDATION:
       Increase the value of S2THRESH if the program finds negative eigenvalues for the overlap matrix.

THRESH
       Cutoff for neglect of two electron integrals. ${10}^{-\mathrm{THRESH}}$ (THRESH $\le 14$).
TYPE:
       INTEGER
DEFAULT:
       8 For single point energies. 10 For optimizations and frequency calculations. 14 For coupled-cluster calculations.
OPTIONS:
       $n$ for a threshold of ${10}^{-n}$.
RECOMMENDATION:
       The value should satisfy THRESH $\ge$ 3 + SCF_CONVERGENCE, although THRESH = 4 + SCF_CONVERGENCE is the default (in most cases) since Q-Chem v. 6.0. See Ref.  459 Gray M., Bowling P. E., Herbert J. M.
J. Phys. Chem. A
(2024), 128, pp. 7739. Link for recommended values of THRESH in the presence of diffuse basis functions, where tighter thresholds are often required and too-loose thresholds may lead to slower convergence or convergence failure.

MEM_STATIC
       Sets the memory for AO-integral evaluations and their transformations in Q-Chem 4.1 or older versions.
TYPE:
       INTEGER
DEFAULT:
       192 corresponding to 192 MB.
OPTIONS:
       $n$ User-defined number of megabytes.
RECOMMENDATION:
       Use the default.

MEM_TOTAL
       Sets the total memory available to Q-Chem, in megabytes.
TYPE:
       INTEGER
DEFAULT:
       2000 2 GB
OPTIONS:
       $n$ User-defined number of megabytes.
RECOMMENDATION:
       Use the default, or set to the physical memory of your machine. The minimum requirement is $3X^2$.

SCF_CONVERGENCE
       NEO-SCF is considered converged when the electronic wave function error is less that ${10}^{-\mathrm{SCF}\mathrm{\_}\mathrm{CONVERGENCE}}$. Adjust the value of THRESH at the same time. (Starting with Q-Chem 3.0, the DIIS error is measured by the maximum error rather than the RMS error as in earlier versions.)
TYPE:
       INTEGER
DEFAULT:
       5 For single point energy calculations. 8 For geometry optimizations.
OPTIONS:
       User-defined
RECOMMENDATION:
       None.

NEO_N_SCF_CONVERGENCE
       NEO-SCF is considered converged when the nuclear wave function error is less that ${10}^{-\mathrm{NEO}\mathrm{\_}\mathrm{N}\mathrm{\_}\mathrm{SCF}\mathrm{\_}\mathrm{CONVERGENCE}}$.
TYPE:
       INTEGER
DEFAULT:
       7
OPTIONS:
       User-defined
RECOMMENDATION:
       None.

NEO_E_CONV
       Energy convergence criteria in the NEO-SCF calculations so that the difference in energy between consecutive iterations is less than ${10}^{-\mathrm{NEO}\mathrm{\_}\mathrm{E}\mathrm{\_}\mathrm{CONV}}$.
TYPE:
       INTEGER
DEFAULT:
       8
OPTIONS:
       User-defined
RECOMMENDATION:
       Tighter criteria for geometry optimization are recommended. Note that for stepwise calculations, this convergence criteria checks the difference in energy between electronic and protonic microiterations blocks.

MAX_SCF_CYCLES
       Controls the maximum number of SCF iterations permitted.
TYPE:
       INTEGER
DEFAULT:
       50
OPTIONS:
       $n$ $n>0$ User-selected.
RECOMMENDATION:
       Increase for slowly converging systems such as those containing transition metals.

NEO_E_MAX_SCF_CYCLES
       Controls the maximum number of electronic SCF micoiterations permitted. This variable is only used for stepwise NEO-SCF optimizations.
TYPE:
       INTEGER
DEFAULT:
       150
OPTIONS:
       $n$ $n>0$ User-selected.
RECOMMENDATION:
       None.

NEO_N_MAX_SCF_CYCLES
       Controls the maximum number of nuclear SCF micoiterations permitted. This variable is only used for stepwise NEO-SCF optimizations.
TYPE:
       INTEGER
DEFAULT:
       150
OPTIONS:
       $n$ $n>0$ User-selected.
RECOMMENDATION:
       None.

NEO_SIMULTANEOUS_SCF
       Enables simultaneous optimization algorithm.
TYPE:
       LOGICAL
DEFAULT:
       TRUE
OPTIONS:
       TRUE FALSE
RECOMMENDATION:
       Use the default unless issues with convergence are observed, in which cause set to FALSE to utilize the traditional stepwise approach.

SCF_ALGORITHM
       Algorithm used for converging the SCF.
TYPE:
       STRING
DEFAULT:
       DIIS Pulay DIIS.
OPTIONS:
       DIIS Pulay DIIS. GDM Geometric Direct Minimization. DIIS_GDM Use DIIS and then later switch to geometric direct minimization.
RECOMMENDATION:
       In the NEO methods, the GDM procedure is recommended. Note that DIIS_GDM is only supported for simultaneous optimizations.

NEO_SCF_GUESS_N
       Specifies the initial guess procedure for the nuclear orbitals in a NEO-SCF calculation.
TYPE:
       STRING
DEFAULT:
       TIGHT Default unless ghost atoms are present. CORE Default when there are ghost atoms.
OPTIONS:
       CORE Diagonalize the nuclear core Hamiltonian TIGHT Occupies the least diffuse nuclear MO AUTOSAD On-the-fly superposition of nuclear atomic densities from an atomic NEO calculation READ Read previous nuclear MOs from disk
RECOMMENDATION:
       Use the default. Note that if NEO_SCF_GUESS_N is set to READ, only the nuclear MOs from a previous NEO-SCF job are read from disk. The starting guess for the electronic MOs in a NEO calculation remains the converged result from the conventional electronic calculation of the current job. To read in both the electronic and nuclear MOs from a previous NEO-SCF job, NEO_SCF_GUESS_N must be set to READ in conjunction with SKIP_SCFMAN set to TRUE.

NEO_STEPWISE_SCF_STEPS
       Specifies the number of NEO-SCF stepwise/alternating macro-iterations to perform before switching to simultaneous algorithm.
TYPE:
       INTEGER
DEFAULT:
       0
OPTIONS:
       User-defined
RECOMMENDATION:
       The rate of convergence of the NEO-SCF procedure is dependent on the initial guess for the electronic and protonic orbitals. For especially difficult systems, it may be helpful to do stepwise optimization before switching to simultaneous.

DIIS_SUBSPACE_SIZE
       Controls the size of the DIIS subspace during the SCF.
TYPE:
       INTEGER
DEFAULT:
       10
OPTIONS:
       User-defined
RECOMMENDATION:
       None.

DIIS_ERR_RMS
       Changes the DIIS convergence metric from the maximum to the RMS error.
TYPE:
       LOGICAL
DEFAULT:
       FALSE
OPTIONS:
       TRUE FALSE
RECOMMENDATION:
       Use the default, the maximum error provides a more reliable criterion.

MAX_DIIS_CYCLES
       The maximum number of DIIS iterations before switching to (geometric) direct minimization.
TYPE:
       INTEGER
DEFAULT:
       50
OPTIONS:
       1 Only a single Roothaan step before switching to (G)DM $n$ $n$ DIIS iterations before switching to (G)DM.
RECOMMENDATION:
       None

THRESH_DIIS_SWITCH
       The threshold for switching between DIIS extrapolation and direct minimization of the SCF energy is ${10}^{-\text{THRESH\_DIIS\_SWITCH}}$.
TYPE:
       INTEGER
DEFAULT:
       2
OPTIONS:
       User-defined.
RECOMMENDATION:
       None